United States
Environmental Protection
Agency
Office of Health and
Environmental Assessment
Washington DC 20460
Research and Development
&EPA
Air Quality
Criteria for
Carbon Monoxide
EPA/600/8-90/045 A
March 1990
External Review Draft
Draft
(Do Not
Cite or Quote)
NOTICE
This document is an external review draft. It has not been
formally released by EPA and should not at this stage be
construed to represent Agency policy. It is being circulated for
comment on its technical accuracy and policy implications
-------
DRAFT:
DO NOT CITE
OR QUOTE
EPA 600-8/90/045A
March 1990
External Review Draft
Air Quality Criteria
for
Carbon Monoxide
NOTICE
This document is a preliminary draft It has not been formally
released by EPA and should not at this stage be construed to
represent Agency policy. It 1« being circulated for comment on
tts technical accuracy and policy Implication*.
Environmental Criteria and Assessment Office
Office of Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
U.S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard, 12th Floor
Chicago, IL 60604-3590
-------
DISCLAIMER
This document is an external draft for review purposes only and does not constitute
Agency policy. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
.' "\
-------
CONTENTS
Page
TABLES xiii
FIGURES xix
AUTHORS, CONTRIBUTORS, AND REVIEWERS xxv
1. SUMMARY AND CONCLUSIONS 1-1
2. INTRODUCTION 2-1
2.1 ORGANIZATION AND CONTENT OF THIS DOCUMENT ... 2-1
2.2 LEGISLATIVE HISTORY OF NAAQS 2-3
2.3 REGULATORY BACKGROUND FOR CARBON MONOXIDE
NAAQS 2-4
2.4 SCIENTIFIC BACKGROUND FOR THE CURRENT
CARBON MONOXIDE NAAQS 2-6
2.4.1 Mechanisms of Action 2-6
2.4.2 Carbon Monoxide Exposure Levels 2-8
2.4.3 Health Effects of Low-Level Carbon Monoxide
Exposures 2-8
2.4.3.1 Cardiovascular Effects 2-8
2.4.3.2 Neurobehavioral Effects 2-10
2.4.3.3 Other Health Effects 2-11
2.5 CRITICAL ISSUES IN REVIEW OF THE NAAQS
FOR CARBON MONOXIDE 2-11
2.5.1 Exposure Assessment in the Population 2-11
2.5.2 Mechanisms of Action of Carbon Monoxide 2-14
2.5.3 Health Effects from Exposure to Carbon Monoxide .... 2-16
2.5.3.1 Effects on the Cardiovascular System 2-16
2.5.3.2 Neurobehavioral Effects 2-18
2.5.3.3 Perinatal Effects 2-19
2.5.4 Population Groups at Greatest Risk for Ambient
CO Exposure Effects 2-19
2.6 CARBON MONOXIDE POISONING 2-20
2.7 REFERENCES 2-23
3. PROPERTIES AND PRINCIPLES OF FORMATION OF
CARBON MONOXIDE 3-1
3.1 INTRODUCTION 3-1
3.2 PHYSICAL PROPERTIES 3-2
3.3 GASEOUS CHEMICAL REACTIONS OF
CARBON MONOXIDE 3-2
3.4 PRINCIPLES OF FORMATION BY SOURCE CATEGORY ... 3-7
3.4.1 General Combustion Processes 3-8
in
-------
CONTENTS (cont'd)
Page
3.4.2 Combustion Engines 3-11
3.4.3 Other Sources 3-14
3.5 REFERENCES 3-15
4. THE GLOBAL CYCLE OF CARBON MONOXIDE:
TRENDS AND MASS BALANCE 4-1
4.1 INTRODUCTION 4-1
4.2 GLOBAL SOURCES, SINKS, AND LIFETIME 4-1
4.2.1 Sources 4-2
4.2.2 Sinks 4-3
4.2.3 Atmospheric Lifetime 4-5
4.2.4 Latitudinal Distribution of Sources 4-5
4.2.5 Uncertainties and Consistencies 4-6
4.3 GLOBAL DISTRIBUTIONS 4-9
4.3.1 Seasonal Variations 4-9
4.3.2 Latitudinal Variation 4-10
4.3.3 Variations with Altitude 4-10
4.3.4 Other Variations 4-12
4.4 GLOBAL TRENDS 4-12
4.5 SUMMARY 4-15
4.6 REFERENCES 4-16
5. MEASUREMENT METHODS FOR CARBON MONOXIDE 5-1
5.1 INTRODUCTION 5-1
5.1.1 Overview of Techniques for Measurement
of Ambient Carbon Monoxide 5-2
5.1.2 Calibration Requirements 5-5
5.2 PREPARATION OF STANDARD REFERENCE MATERIALS . 5-5
5.2.1 Gas Standards 5-5
5.2.2 Gravimetric Method 5-6
5.2.3 Volumetric Gas Dilution Methods 5-8
5.2.4 Other Methods 5-8
5.3 MEASUREMENT IN AMBIENT AIR 5-9
5.3.1 Sampling System Components 5-9
5.3.2 Quality Assurance Procedures for Sampling 5-11
5.3.3 Sampling Schedules 5-13
5.3.4 Continuous Analysis 5-14
5.3.4.1 Nondispersive Infrared Photometry 5-14
5.3.4.2 Gas Chromatography - Flame lonization ... 5-18
5.3.4.3 Other Analyzers 5-19
5.3.5 Intermittent Analysis 5-23
5.3.5.1 Colorimetric Analysis 5-23
IV
-------
CONTENTS (cont'd)
5.4 MEASUREMENT USING PERSONAL MONITORS 5-25
5.5 REFERENCES 5-27
6. AMBIENT CARBON MONOXIDE SOURCES, EMISSIONS, AND
CONCENTRATIONS 6-1
6.1 ESTIMATING NATIONAL EMISSION FACTORS 6-1
6.2 EMISSION SOURCES AND EMISSION FACTORS BY
SOURCE CATEGORY 6-1
6.2.1 Transportation Sources 6-3
6.2.1.1 Motor Vehicles 6-3
6.2.1.2 Aircraft 6-4
6.2.1.3 Railroads 6-4
6.2.1.4 Vessels 6-4
6.2.1.5 Nonhighway Use of Motor Fuels 6-4
6.2.2 Stationary Source Fuel Combustion 6-5
6.2.3 Industrial Processes 6-5
6.2.4 Solid Waste Disposal 6-6
6.2.5 Miscellaneous Combustion Sources 6-6
6.2.5.1 Forest Fires 6-6
6.2.5.2 Agricultural Burning 6-6
6.2.5.3 Coal Refuse 6-7
6.2.5.4 Structural Fires 6-7
6.3 NATIONAL CO EMISSIONS ESTIMATES 1970-1988 6-7
6.4 OUTDOOR AIR CONCENTRATIONS 6-11
6.4.1 Introduction 6-11
6.4.2 Site Selection 6-11
6.4.3 United States Data Base 6-14
6.4.4 Techniques of Data Analysis 6-15
6.4.4.1 Frequency Analysis , 6-16
6.4.4.2 Trend Analyses 6-17
6.4.4.3 Special Analyses 6-17
6.4.5 Urban Levels of Carbon Monoxide 6-19
6.4.5.1 Ten-year CO Trends 1979-1988 6-19
6.4.5.2 Five-year CO Trends 1984-1988 6-22
6.4.5.3 Air Quality Levels in Metropolitan Statistical
Areas 6-24
6.4.6 Effects of Meteorology and Topography 6-24
6.5 CARBON MONOXIDE DISPERSION MODELS 6-32
6.5.1 Line Source Modeling 6-32
6.5.1.1 CALINE3 6-33
6.5.1.2 GMLINE 6-33
6.5.1.3 HIWAY-2 6-33
6.5.1.4 PAL 6-34
-------
CONTENTS (cont'd)
Page
6.5.1.5 Model Evaluation 6-34
6.5.2 Intersection Modeling 6-35
6.5.2.1 "Volume 9" 6-35
6.5.2.2 Intersection Midblock Model 6-36
6.5.2.3 Georgia Intersection Model 6-37
6.5.2.4 TEXIN2 6-37
6.5.2.5 CAL3Q 6-39
6.5.2.6 CALINE4 6-39
6.5.2.7 Comparison of Intersection Models 6-40
6.5.3 Urban Area Modeling 6-43
6.5.3.1 APRAC-3 6-44
6.5.3.2 Urban Airshed Model 6-45
6.5.3.3 RAM 6-45
6.6 REFERENCES 6-46
7. INDOOR CARBON MONOXIDE SOURCES, EMISSIONS, AND
CONCENTRATIONS 7-1
7.1 INTRODUCTION 7-1
7.2 EMISSIONS FROM INDOOR SOURCES 7-4
7.2.1 Emissions from Gas Cooking Ranges, Gas
Ovens, and Gas Appliances ; . 7-5
7.2.2 Emissions from Unvented Space Heaters 7-12
7.2.3 Emissions from Wood Stoves and Tobacco Combustion . 7-15
7.2.4 Summary of Emission Data 7-17
7.3 CONCENTRATIONS IN INDOOR ENVIRONMENTS 7-20
7.3.1 Indoor Concentrations Recorded in Personal Exposure
Studies 7-20
7.3.2 Targeted Microenvironmental Studies 7-25
7.3.2.1 Indoor Microenvironmental Concentrations .. 7-25
7.3.2.2 Concentrations Associated with Indoor
Sources 7-29
7.3.3 Spatial Concentration Variations 7-39
7.3.4 Summary of Indoor Concentrations 7-42
7.4 REFERENCES 7-44
8. POPULATION EXPOSURE TO CARBON MONOXIDE 8-1
8.1 INTRODUCTION 8-1
8.2 EXPOSURE MONITORING IN THE POPULATION 8-3
8.2.1 Personal Monitoring 8-4
8.2.2 Carbon Monoxide Exposures Indoors 8-8
8.2.3 Carbon Monoxide Exposures Inside Vehicles 8-12
8.2.4 Carbon Monoxide Exposures Outdoors 8-14
VI
-------
CONTENTS (cont'd)
Page
8.3 ESTIMATING POPULATION EXPOSURE TO
CARBON MONOXIDE 8-15
8.3.1 Defining Concentration, Exposure, and Dose 8-16
8.3.2 Components of Exposure 8-17
8.3.3 Relationship to Fixed-Site Monitors 8-20
8.3.4 Alternative Approaches to Exposure Estimation 8-21
8.3.5 Statistical Models Based on Personal Monitoring Data . . 8-23
8.3.6 Physical and Physical-Stochastic Models 8-28
8.4 OCCUPATIONAL EXPOSURE TO CARBON MONOXIDE . . . 8-44
8.4.1 Historical Perspective 8-45
8.4.2 Exposure Monitoring Techniques 8-46
8.4.3 Occupational Exposures 8-51
8.5 BIOLOGICAL MONITORING 8-61
8.5.1 Blood Carboxyhemoglobin Measurement 8-61
8.5.1.1 Measurement Methods 8-61
8.5.1.2 Carboxyhemoglobin Measurements in
Populations 8-75
8.5.2 Carbon Monoxide in Expired Breath 8-80
8.5.2.1 Measurement Methods 8-82
8.5.2.2 Breath Measurements in Populations 8-89
8.5.3 Potential Limitations 8-98
8.5.3.1 Pulmonary Disease 8-98
8.5.3.2 Subject Age 8-99
8.5.3.3 Effects of Smoking 8-99
8.6 SUMMARY AND CONCLUSIONS 8-100
8.7 REFERENCES 8-103
9. PHARMACOKINETICS AND MECHANISMS OF
ACTION OF CARBON MONOXIDE 9-1
9.1 ABSORPTION, DISTRIBUTION, AND
PULMONARY ELIMINATION 9-1
9.1.1 Introduction 9-1
9.1.2 Pulmonary Uptake 9-1
9.1.2.1 Mass Transfer of Carbon Monoxide 9-1
9.1.2.2 Effects of Dead Space and Uneven Distribution
of Ventilation and Perfusion 9-2
9.1.2.3 Alveolo-Capillary Membrane and
Blood-Phase Diffusion 9-4
9.1.3 Tissue Uptake 9-5
9.1.3.1 The Blood 9-6
9.1.3.2 The Lung 9-8
9.1.3.3 Heart and Skeletal Muscles 9-8
9.1.3.4 Brain and Other Tissues 9-9
9.1.4 Pulmonary and Tissue Elimination 9-10
vii
-------
CONTENTS (cont'd)
Page
9.2 TISSUE PRODUCTION AND METABOLISM OF
CARBON MONOXIDE 9-11
9.3 MODELING CARBOXYHEMOGLOBIN FORMATION 9-12
9.3.1 Introduction 9-12
9.3.2 Regression Models 9-12
9.3.3 The Coburn-Forster-Kane Differential Equations 9-14
9.3.3.1 Linear and Nonlinear CFK Differential
Equations 9-14
9.3.3.2 Confirmation Studies of the CFK Model .... 9-16
9.3.3.3 Modified CFK Models 9-18
9.3.3.4 Application of the CFK Model 9-20
9.3.4 Summary 9-21
9.4 INTRACELLULAR EFFECTS OF CARBON MONOXIDE .... 9-22
9.4.1 Introduction 9-22
9.4.2 Carbon Monoxide Binding to Myoglobin 9-26
9.4.3 Carbon Monoxide Uptake by Cytochrome P-450 9-27
9.4.4 Carbon Monoxide and Cytochrome c Oxidase 9-28
9.5 REFERENCES 9-32
10. HEALTH EFFECTS OF CARBON MONOXIDE 10-1
10.1 INTRODUCTION 10-1
10.2 ACUTE PULMONARY EFFECTS OF CARBON MONOXIDE . 10-4
10.2.1 Introduction 10-4
10.2.2 Effects on Lung Morphology 10-4
10.2.2.1 Studies in Laboratory Animals 10-5
10.2.2.2 Studies in Humans 10-7
10.2.3 Effects on Lung Function 10-7
10.2.3.1 Lung Function in Laboratory Animals 10-7
10.2.3.2 Lung Function in Humans 10-9
10.2.4 Summary 10-13
10.3 CARDIOVASCULAR EFFECTS OF CARBON MONOXIDE . . 10-14
10.3.1 Introduction 10-14
10.3.2 Experimental Studies in Humans 10-15
10.3.2.1 Cardiorespiratory Response to Exercise .... 10-15
10.3.2.2 Arrhythmogenic Effects 10-35
10.3.2.3 Effects on Coronary Blood Flow 10-37
10.3.3 Relationship between CO Exposure and Risk of
Cardiovascular Disease in Man 10-38
10.3.3.1 Risk of Ischemic Heart Disease 10-38
10.3.3.2 Risk of Hypertension 10-42
10.3.4 Studies in Laboratory Animals 10-42
10.3.4.1 Introduction 10-42
10.3.4.2 Ventricular Fibrillation Studies 10-43
viii
-------
CONTENTS (cont'd)
Page
10.3.4.3 Hemodynamic Studies 10-47
10.3.4.4 Cardiomegaly 10-52
10.3.4.5 Hematology Studies 10-58
10.3.4.6 Atherosclerosis and Thrombosis 10-60
10.3.5 Summary and Conclusions 10-69
10.4 CEREBROVASCULAR AND BEHAVIORAL EFFECTS OF
CARBON MONOXIDE 10-71
10.4.1 Control of Cerebral Blood Flow and Tissue PO2
with Carbon Monoxide and Hypoxic Hypoxia 10-71
10.4.1.1 Introduction 10-71
10.4.1.2 Effects on Global Cerebral Blood Flow .... 10-72
10.4.1.3 Effects on Regional Cerebral Blood Flow ... 10-83
10.4.1.4 Effect of Low Levels of Carbon Monoxide on
Cerebral Blood Flow 10-87
10.4.1.5 Synergistic Effects of Carbon Monoxide .... 10-93
10.4.1.6 Mechanism of Regulation of Cerebral Blood
Flow in Hypoxia 10-99
10.4.1.7 Summary 10-102
10.4.2 Behavioral Effects of Carbon Monoxide 10-103
10.4.2.1 Introduction 10-103
10.4.2.2 Sensory Effects 10-105
10.4.2.3 Motor and Sensorimotor Performance 10-113
10.4.2.4 Vigilance 10-122
10.4.2.5 Miscellaneous Measures of Performance .... 10-124
10.4.2.6 Automobile Driving 10-131
10.4.2.7 Brain Electrical Activity 10-132
10.4.2.8 Schedule-Controlled Behavior 10-137
10.4.2.9 Summary and Discussion of Behavioral
Literature 10-139
10.4.2.10 Hypotheses 10-146
10.4.2.11 Conclusions 10-147
10.5 DEVELOPMENTAL TOXICITY OF CARBON MONOXIDE . . 10-148
10.5.1 Introduction 10-148
10.5.2 Theoretical Basis for Fetal Exposure to Excessive
Carbon Monoxide and for Excess Fetal Toxicity 10-151
10.5.2.1 Evidence for Elevated Fetal
Carboxyhemoglobin Relative to
Maternal Hemoglobin 10-151
10.5.2.2 Effect of Maternal Carboxyhemoglobin on
Placental O2 Transport 10-152
10.5.3 Measurement of Carboxyhemoglobin Content in
Fetal Blood 10-153
10.5.4 Consequences of Carbon Monoxide in Development ... 10-154
ix
-------
CONTENTS (cont'd)
Page
10.5.4.1 Fetotoxic and Teratogenic Consequence of
Prenatal Carbon Monoxide Exposure 10-155
10.5.4.2 Carbon Monoxide and Body Weight 10-159
10.5.4.3 Alteration in Cardiovascular Development
following Early Carbon Monoxide Exposure . 10-161
10.5.4.4 Neurobehavioral Consequences of Perinatal
Carbon Monoxide Exposure 10-167
10.5.4.5 Neurochemical Consequences of Prenatal and
Perinatal Carbon Monoxide Exposure 10-174
10.5.4.6 Morphological Consequences of Acute Prenatal
Carbon Monoxide 10-177
10.5.5 Summary 10-178
10.6 OTHER SYSTEMIC EFFECTS OF CARBON MONOXIDE . . . 10-178
10.7 ADAPTATION, HABITUATION, AND COMPENSATORY
RESPONSES TO CARBON MONOXIDE
EXPOSURE 10-182
10.7.1 Short-Term Habituation 10-183
10.7.2 Long-Term Adaptation 10-184
10.7.3 Summary 10-187
10.8 REFERENCES 10-188
11. COMBINED EXPOSURE OF CARBON MONOXIDE WITH
OTHER POLLUTANTS, DRUGS, AND ENVIRONMENTAL
FACTORS 11-1
11.1 HIGH ALTITUDE EFFECTS OF CARBON MONOXIDE 11-1
11.1.1 Introduction 11-1
11.1.2 Carboxyhemoglobin Formation 11-3
11.1.3 Cardiovascular Effects 11-4
11.1.4 Chronic Studies 11-10
11.1.5 Neurobehavioral Effects 11-16
11.1.6 Compartmental Shifts 11-17
11.1.7 Conclusions 11-18
11.2 CARBON MONOXIDE INTERACTIONS WITH DRUGS .... 11-18
11.2.1 Introduction 11-18
11.2.2 Alcohol 11-20
11.2.3 Barbiturates 11-23
11.2.4 Other Psychoactive Drugs 11-24
11.3 COMBINED EXPOSURE TO CARBON MONOXIDE AND
OTHER AIR POLLUTANTS AND ENVIRONMENTAL
FACTORS 11-24
11.3.1 Exposure in Ambient Air 11-25
11.3.2 Exposure to Combustion Products 11-30
11.3.3 Exposure to Other Environmental Factors 11-35
-------
CONTENTS (cont'd)
11.3.3.1 Environmental Heat 11-35
11.3.3.2 Environmental Noise 11-36
11.3.4 Summary 11-37
11.4 ENVIRONMENTAL TOBACCO SMOKE 11-38
11.5 REFERENCES 11-40
12. EVALUATION OF SUBPOPULATIONS POTENTIALLY AT RISK
TO CARBON MONOXIDE EXPOSURE 12-1
12.1 INTRODUCTION 12-1
12.2 AGE AND GENDER AS RISK FACTORS 12-2
12.3 RISK OF CARBON MONOXIDE EXPOSURE IN
INDIVIDUALS WITH PREEXISTING DISEASE 12-4
12.3.1 Subjects with Coronary Artery Disease 12-4
12.3.2 Subjects with Congestive Heart Failure 12-5
12.3.3 Subjects with Other Vascular Diseases 12-5
12.3.4 Subjects with Anemia and Other Hematologic
Disorders 12-6
12.3.5 Subjects with Obstructive Lung Disease 12-7
12.4 SUBPOPULATIONS AT RISK FROM COMBINED
EXPOSURE TO CARBON MONOXIDE AND OTHER
CHEMICAL SUBSTANCES 12-8
12.4.1 Interactions with Psychoactive Drugs 12-8
12.4.2 Interactions with Cardiovascular Drugs 12-9
12.4.3 Mechanisms of Carbon Monoxide Interactions with
Drugs: Need for Further Research 12-10
12.4.3.1 Metabolic Effects , 12-10
12.4.3.2 Central Nervous System Depression 12-10
12.4.3.3 Alteration in Cerebral Blood Flow 12-11
12.4.4 Interactions with Other Chemical Substances
in the Environment 12-13
12.5 SUBPOPULATIONS EXPOSED TO CARBON MONOXIDE
AT HIGH ALTITUDES 12-14
12.6 REFERENCES 12-17
APPENDIX A: GLOSSARY OF TERMS AND SYMBOLS A-l
XI
-------
TABLES
Number Page
2-1 National Ambient Air Quality Standards for Carbon Monoxide .... 2-6
2-2 Lowest Observed Effect Levels for Human Health Effects
Associated with Low-Level Carbon Monoxide Exposure 2-9
3-1 Physical Properties of Carbon Monoxide 3-3
3-2 Reported Room Temperature Rate Constants for the Reaction
of OH* Radicals with CO 3-5
3-3 Summary of Light-Duty Vehicle (LDV) Emissions Standards .... 3-9
4-1 Sources of Carbon Monoxide 4-4
5-1 Performance Specifications for Automated Analytical Methods
for Carbon Monoxide (Code of Federal Regulations, 1977a) .... 5-3
6-1 Carbon Monoxide National Emission Estimates (teragrams/year) . . 6-2
6-2 Carbon Monoxide Emissions from Transportation
(gigagrams/year) 6-8
6-3 Specific Probe Exposure Criteria 6-13
6-4 National Carbon Monoxide Emission Estimates, 1979-1988 6-22
6-5 Distribution of Population in Metropolitan Statistical Areas 6-25
6-6 Parameters for Intersection Scenarios 6-41
6-7 Results of Model Comparisons for the Undercapacity
Intersection Scenario 6-42
6-8 Results of Model Comparisons for the Near Capacity
Intersection Scenario 6-42
6-9 Results of Model Comparisons for the Overcapacity Intersection
Scenario 6-43
7-1 Carbon Monoxide-Emission Rates for 12 Range-Top Burners
Operating with Blue and Yellow-Tipping Flames by the Direct
Sampling Method 7-7
xiii
-------
TABLES (cont'd)
Number Page
7-2 Carbon Monoxide-Emission Rates for Gas Range Ovens,
Gas Range Pilot Lights, and Gas Dryers 7-8
7-3 Carbon Monoxide-Emission Rates from 18 Gas Ranges,
Gas Ovens, and Gas Pilot Lights for Blue Flame and
Yellow-Tipping Flame by the Direct-Sampling Method 7-9
7-4 Carbon Monoxide Emissions from Gas Ranges for Studies
of Small Sample Size 7-12
7-5 Carbon Monoxide Emissions from Unvented Gas Space Heaters . . 7-14
7-6 Carbon Monoxide Emissions from Unvented Kerosene Space
Heaters 7-16
7-7 Summary of CO Exposure Levels and Time Spent Per Day in
Selected Microenvironments 7-22
7-8 Indoor Microenvironments Listed in Descending Order of
Weighted Mean CO Concentration 7-23
7-9 Weighted Means of Residential Exposure Grouped According to
the Presence or Absence of Selected Indoor Carbon
Monoxide Sources 7-24
7-10 Average Residential CO Exposures (ppm): Impact of
Combustion Appliance Use and Tobacco Smoking 7-25
7-11 Carbon Monoxide Concentrations Measured in Various
Indoor Environments as a Function of Microenvironments 7-26
7-12 Weighted Summary Statistics for CO Concentrations (ppm)
in the Main Living Area by Use for Selected Sources
by County 7-31
7-13 Summary of Continuous CO Monitoring Results by
Heating Equipment 7-34
7-14 Peak CO Concentrations by Indoor Source Measured in
Field Studies 7-35
xiv
-------
TABLES (cont'd)
Number Page
7-15 Measured Concentrations of Carbon Monoxide in Environmental
Tobacco Smoke 7-40
8-1 Carbon Monoxide Concentrations in In-Transit
Microenvironments - Denver, Colorado 8-8
8-2 Carbon Monoxide Concentrations in Outdoor
Microenvironments - Denver, Colorado 8-9
8-3 Carbon Monoxide Concentrations in Indoor
Microenvironments - Denver Colorado 8-10
8-4 Comparison of Different Approaches to Air Pollution Exposure
Modeling 8-24
8-5 Models Which Have Been Used to Estimate CO Exposure
by Model Type 8-25
8-6 Results of Weighted Linear Regression Analysis with Nontransit
PEM Value as Dependent Variable and Simultaneous Value
at Nearest Denver Fixed-Site as Independent Variable 8-27
8-7 Results of Weighted Linear Regression Analyses with In-Transit
PEM Value as Dependent Variable and Simultaneous Value
from Denver Composite Data Set as Independent Variable 8-29
8-8 Results of Weighted Linear Regression Analyses with Nontransit
PEM Value as Dependent Variable and Simultaneous Value at
Nearest Fixed-Site in Washington, DC as Independent Variable . . . 8-30
8-9 Results of Weighted Linear Regression Analyses with In-Transit
PEM Value as Dependent Variable and Simultaneous Value from
Composite Washington, DC Data Set as Independent Variable .... 8-31
8-10 Diagnostic Criteria for CO Intoxication 8-50
8-11 Comparison of Representative Methods for Analysis of
Carbon Monoxide in Blood 8-63
8-12 Evaluation of the Ability of Co-Oximeters to Measure Low Levels
of COHB as Compared to Proposed Reference Methods 8-73
xv
-------
TABLES (cont'd)
Number Page
8-13 Regression Parameters for the Relationship Between COHb
and Eight-Hour CO Averages for 20 Cities 8-79
8-14 Summary of Studies Comparing End-Expired Breath CO with
COHB Levels 8-83
9-1 In Vitro Inhibition Ratios for Hemoproteins that Bind
Carbon Monoxide 9-24
10-1 Summary of Effects of Carbon Monoxide on Maximal and
Submaximal Exercise Performance 10-16
10-2 Summary of Effects of Carbon Monoxide Exposure in Patients
with Angina 10-22
10-3 Comparison of Subjects in Studies of the Effect of
Carbon Monoxide Exposure on Occurrence of Angina
During Exercise 10-30
10-4 Ventricular Fibrillation and Hemodynamic Studies in
Laboratory Animals 10-44
10-5 Cardiac Hypertrophy Studies in Laboratory Animals 10-53
10-6 Hematology Studies in Laboratory Animals 10-59
10-7 Atherosclerotic Studies in Laboratory Animals 10-62
10-8 Brain Regions Ranked from Greatest to Least in Response
to Hypoxia 10-84
10-9 Effects of COHb on Absolute Visual Threshold 10-106
10-10 Effects of COHb on Critical Flicker Fusion 10-108
10-11 Effects of COHb on Miscellaneous Visual Functions 10-111
xvi
-------
TABLES (cont'd)
Number
10-12
10-13
10-14
10-15
10-16
10-17
10-18
10-19
10-20
10-21
10-22
10-23
10-24
10-25
10-26
10-27
10-28
10-29
Effects of COHb on Miscellaneous Auditory Functions
Effects of COHb on Fine Motor Skills
Effects of COHb on Reaction Time
Effects of COHb on Tracking
Effects of COHb on Vigilance
Effects of COHb on Continuous Performance ,
Effects of COHb on Time Estimation
Effects of COHb on Miscellaneous Cognitive Tasks ,
Effects of COHb on Automobile Driving Tasks
Effects of COHb on Brain Electrical Activity
Effects of COHb on Schedule-Controlled Behavior
Effect of Blind Conditions
Effect of Statistical Methodology
Probability of Effects of COHb
Effect of Single vs. Multiple Task Performance ,
Effect of Rate of COHb Formation
Teratogenic Consequences of Prenatal Carbon Monoxide
Exposure in Laboratory Animals
Consequences of Prenatal Carbon Monoxide Exposure on
Cardiovascular Development in Laboratory Rats
10-114
10-115
10-117
10-120
10-123
10-125
10-128
10-129
10-133
10-135
10-138
10-140
10-141
10-142
10-145
10-145
10-156
10-163
XVll
-------
TABLES (cont'd)
Number Page
10-30 Neurobehavioral Consequences of Prenatal
Carbon Monoxide Exposure in Laboratory Animals 10-169
10-31 Consequences of Human Carbon Monoxide Intoxication During
Early Development 10-171
10-32 Other Systemic Effects of Carbon Monoxide 10-179
11-1 Calculated Equilibrium Values of Percent COHb and Percent
02Hb in Humans Exposed to Ambient CO at Various
Altitudes 11-4
11-2 Summary of Effects of Carbon Monoxide at Altitude 11-5
11-3 Chronic Effects of Altitude and Carbon Monoxide Exposure 11-15
11-4 Combined Exposure to Carbon Monoxide and Other Pollutants .... 11-26
11-5 Combined Exposure to Carbon Monoxide and Combustion
Products 11-31
XVlll
-------
FIGURES
Number
2-1 Relationship between carbon monoxide exposure and
carboxyhemoglobin levels in the blood 2-7
2-2 Currently accepted or proposed mechanisms of action of
carbon monoxide resulting from external exposure sources
can interfere with cellular respiration and cause
tissue hypoxia 2-15
3-1 Effect of air-fuel ratio on exhaust gas carbon monoxide
concentrations from three gasoline-fueled test engines 3-13
4-1 The estimated sources of CO as a function of latitude 4-7
4-2 The global seasonal variations of CO 4-11
4-3 The global concentrations and trends of CO 4-13
5-1 Loss of carbon monoxide with time in mild steel cylinders .... 5-7
5-2 Carbon monoxide monitoring system 5-10
5-3 Schematic diagram of gas filter correlation (GFC) monitor
for CO 5-17
6-1 Estimated emissions of carbon monoxide from highway
vehicles, 1970-1988 6-9
6-2 CO pollution rose for St. Louis, MO 6-18
6-3 National trend in the composite average of the second
highest nonoverlapping 8-hour average carbon monoxide
concentration 1979-1988 6-20
6-4 Boxplot comparisons of trends in second highest
nonoverlapping 8-hour average carbon monoxide
concentrations at 248 sites, 1979-1988 6-20
xix
-------
FIGURES (cont'd)
Number Page
6-5 National trend in the composite average of the estimated
number of exceedances of the 8-hour carbon monoxide
NAAQS, 1979-1988 6-21
6-6 Boxplot comparisons of trends in second highest
nonoverlapping 8-hour average carbon monoxide
concentrations at 359 sites, 1984-1988 6-23
6-7 Regional comparisons of the 1986, 1987, 1988 composite
averages of the second highest nonoverlapping 8-hour
average carbon monoxide concentration 6-24
6-8 United States map of the highest second maximum
nonoverlapping 8-hour average carbon monoxide
concentration by Metropolitan Statistical Area
for 1988 6-26
6-9 Effect of terrain roughness on the wind speed profile 6-28
6-10 Schematic representation of an elevated inversion 6-30
6-11 Hourly variations in inversion height and wind speed
for Los Angeles in summer 6-31
7-1 Cumulative frequency distributions and summary statistics
for indoor CO concentrations in three groups of
monitored homes 7-33
7-2 A time history of CO concentrations, 2-hour averages,
winter of 1974 7-38
8-1 Frequency distributions of maximum eight-hour carbon monoxide
population exposures and fixed-site monitor values in
Denver, CO and Washington, DC; November 1982 -
February 1983 8-6
8-2 Typical individual exposure as a function of time 8-18
8-3 Logarithmic-probability plot of cumulative frequency
distribution of maximum one-hour average exposure of
CO predicted by SHAPE, plus an observed frequency
distribution for Day 2 in Denver 8-37
xx
-------
FIGURES (cont'd)
Number
8-4 Logarithmic-probability plot of cumulative frequency
distribution of maximum moving average eight-hour
exposure of CO predicted by SHAPE, plus an observed
frequency distribution for Day 2 in Denver 8-38
8-5 Frequency distributions of carboxyhemoglobin levels
in the U.S. population, by smoking habit 8-78
8-6 Changes in alveolar CO of nonsmoking basement office
workers compared to nonsmoking workers in other offices
between Friday afternoon, Monday morning, and
Monday afternoon 8-91
8-7 Eight-hour average CO concentrations in basement office
before and after corrective action 8-91
8-8 Distributions of CO in breath of adult nonsmokers in
Denver and Washington 8-94
8-9 Percent of Washington sample population with eight-hour
average CO exposures exceeding the concentrations shown .... 8-95
9-1 Oxyhemoglobin dissociation curves of normal human blood,
of blood containing 50% carboxyhemoglobin, and of
blood with a 50% normal hemoglobin concentration
due to anemia 9-7
9-2 Measured and predicted COHb concentrations from six
intermittently exercising subjects 9-19
10-1 Relationship between carboxyhemoglobin level (COHb) and
decrement in maximal oxygen uptake (VO2 max) for
healthy nonsmokers 10-19
10-2 The effect of CO exposure on time to onset of angina 10-32
10-3 Effect of hypoxic hypoxia and CO hypoxia on cerebral
blood flow in 13 control and 9 chemodenervated dogs 10-74
xxi
-------
FIGURES (cont'd)
Number
10-4 Effects of hypoxic and carbon monoxide hypoxia on cerebral
blood flow, mean arterial blood pressure, and cerebral
vascular resistance in control, carotid baroreceptor-,
and chemoreceptor-denervated animals
10-5 Effects of hypoxic and carbon monoxide hypoxia on cerebral
blood flow, mean arterial blood pressure, and cerebral
vascular resistance in control and vagotomized animals ....
10-6 Cerebral blood flow as function of fractional arterial O2
saturation
10-7 Comparison of newborn and adult responses of the reciprocal
of the cerebral arteriovenous O2 content difference
(CaO2 - QOj)-! to a reduction in arterial O2 content
(C.O2) during hypoxic hypoxia (HH)
10-8 Profiles of slopes of regional blood flow responses to
hypoxic hypoxia (HH, solid lines) and CO hypoxia
(COH, dashed lines) in adults (top) and newborns (bottom) . .
10-9 Effect of hypoxic hypoxia and carbon monoxide (CO) hypoxia
on neurohypophyseal and regional cerebral blood flow (rCBF)
10-10 Effect of complete chemoreceptor denervation on total
cerebral and neurohypophyseal blood flow
10-11 Effect of increasing carboxyhemoglobin levels on cerebral
blood flow, with special reference to low-level
administration (below 20% COHb)
10-12 Effect of cyanide (CN) and CO hypoxia, alone and in
combination, on cerebral blood flow
10-13 Effect of CN and CO hypoxia, alone and in combination,
on cerebral oxygen consumption
10-14 Relationship of CBF to cerebral O2 consumption
during CN and CO hypoxia
10-76
10-77
10-79
10-81
10-85
10-88
10-89
10-91
10-96
10-97
10-98
xxu
-------
FIGURES (cont'd)
Number Pa;
11-1 Increment in percent carboxyhemoglobin (A% HbCO) over basal
(control) levels at the end of a maximum aerobic
capacity test and at the 5th min of recovery from a test
in a typical (A) male and (B) female subject 11-9
11-2 Change in carboxyhemoglobin concentration (% COHb) during
eight-hour exposures to 0 to 9 ppm CO for (A) resting and
(B) exercising subjects 11-11
11-3 The effects of altitude and ambient CO exposure on COHb
in Fischer 344 rats 11-13
11-4 Higher concentrations of COHb observed at the end of
a five-minute recovery period after attainment of the
subject's maximum aerobic capacity indicate that
liberation of CO from tissue stores is linearly
related to COHb concentration present at exhaustion 11-19
XXlll
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
CHAPTER 3. PROPERTIES AND PRINCIPLES OF FORMATION OF CARBON
MONOXIDE
Principal Authors
Dr. Marcia C. Dodge
Atmospheric Research and Exposure Assessment Laboratory (MD-84)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Harold G. Richter
Environmental Sciences
NSI Technology Services Corporation
P.O. Box 12313
Research Triangle Park, NC 27709
Contributors and Reviewers
Dr. Aubrey P. Altshuller
Atmospheric Research and Exposure Assessment Laboratory (MD-59)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Joseph J. Bufalini
Atmospheric Research and Exposure Assessment Laboratory (MD-84)
U.S Environmental Protection Agency
Research Triangle Park, NC 27711
CHAPTER 4. THE GLOBAL CYCLE OF CARBON MONOXIDE: TRENDS AND MASS
BALANCE
Principal Author
Dr. Aslam K. Khalil
Institute of Atmospheric Sciences
Oregon Graduate Center
19600 N.W. Von Neumann Drive
Beaverton, OR 97006
xxv
-------
Contributors and Reviewers
Dr. Aubrey P. Altshuller
Atmospheric Research and Exposure Assessment Laboratory (MD-59)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Joseph J. Bufalini
Atmospheric Research and Exposure Assessment Laboratory (MD-84)
U.S Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Jack Fishman
Atmospheric Sciences Division
NASA - Langley Research Center
Mail Stop 401A
Hampton, VA 23665
CHAPTER 5. MEASUREMENT METHODS FOR CARBON MONOXIDE
Principal Author
Mr. Gerald G. Akland
Atmospheric Research and Exposure Assessment Laboratory (MD-75)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Contributors and Reviewers
Dr. William McClenny
Atmospheric Research and Exposure Assessment Laboratory (MD-44)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Joseph R. Stetter
Transducer Research, Inc.
1228 Olympus Drive
Naperville, IL 60540
xxvi
-------
CHAPTER 6. AMBIENT SOURCES, EMISSIONS, AND CONCENTRATIONS
Principal Authors
Dr. James N. Braddock
Atmospheric Research and Exposure Assessment Laboratory (MD-46)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Thomas N. Braverman
Office of Air Quality Planning and Standards (MD-14)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. William B. Petersen
Atmospheric Research and Exposure Assessment Laboratory (MD-80)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Thomas B. McMullen
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Contributors and Reviewers
Mr. Paul Benson
Transportation Laboratory
5900 Folsom Boulevard
Sacramento, CA 95819
Dr. Thomas C. Curran
Office of Air Quality Planning and Standards (MD-14)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Tom Hansen
U.S. Environmental Protection Agency
(4AT Air Program Branch)
345 Courtland Street, NE
Atlanta, GA 30365
XXVll
-------
Mr. George Schewe
PEI Associates, Inc.
11499 Chester Road
Cincinnati, OH 45246-0100
CHAPTER 7. INDOOR SOURCES, EMISSIONS, AND CONCENTRATIONS
Principal Author
Dr. Brian Leaderer
John B. Pierce Foundation Laboratory
290 Congress Avenue
New Haven, CT 06519
Contributors and Reviewers
Dr. Irwin H. Billick
Gas Research Institute
8600 West Bryn Mawr Avenue
Chicago, IL 60631
Mr. Tom Hansen
U.S. Environmental Protection Agency
(4AT Air Program Branch)
345 Courtland Street, NE
Atlanta, GA 30365
Dr. P. Barry Ryan
Department of Environmental Sciences and Physics
Harvard University School of Public Health
665 Huntington Avenue
Boston, MA 02115
CHAPTER 8. POPULATION EXPOSURE TO CARBON MONOXIDE
Principal Authors
Mr. Gerald G. Akland
Atmospheric Research and Exposure Assessment Laboratory (MD-75)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
XXVlll
-------
Dr. Steven D. Colome
Integrated Environmental Services
University Tower, Suite 1090
4199 Campus Drive
Irvine, CA 92715
Dr. Thomas E. Dahms
Department of Anesthesiology
St. Louis University School of Medicine
3635 Vista Avenue
St. Louis, MO 63110
Mr. Ted Johnson
PEI Associates, Inc.
505 S. Duke Street
Durham, NC 27701
Dr. Brian Leaderer
John B. Pierce Foundation Laboratory
290 Congress Avenue
New Haven, CT 06519
Dr. Wayne Ott
Office of Research and Development (RD-680)
U.S. Environmental Protection Agency
Washington, DC 20460
Dr. Lance Wallace
U.S. Environmental Protection Agency
Building 166
Vint Hill Farms Station
Bicher Road
Warrenton, VA 22186
Contributors and Reviewers
Dr. William F. Biller
68 Yorktown Road
East Brunswick, NJ 08816
Mr. N. O. Gerald
Office of Air Quality Planning and Standards (MD-14)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
XXIX
-------
Dr. Steven M. Horvath
Environmental Stress Laboratory
Neuroscience Research Institute
University of California
Santa Barbara, CA 93106
Mr. Thomas R. McCurdy
Office of Air Quality Planning and Standards (MD-12)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Harvey M. Richmond
Office of Air Quality Planning and Standards (MD-12)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. P. Barry Ryan
Department of Environmental Sciences and Physics
Harvard University School of Public Health
665 Huntington Avenue
Boston, MA 02115
Dr. Carr J. Smith
Bowman Gray Technical Center
R.J. Reynolds Tobacco Company
Winston-Salem, NC 27102
Dr. John Spengler
Harvard School of Public Health
665 Huntington Avenue
Boston, MA 02115
Dr. David K. Stevenson
Department of Pediatrics
Laboratory for Neonatal Metabolism
Stanford University School of Medicine
Stanford, CA 94305
Dr. HenkJ. Vreman
Laboratory for Neonatal Metabolism
Department of Pediatrics (S214)
Stanford University School of Medicine
Stanford, CA 94305
XXX
-------
CHAPTER 9. PHARMACOKINETICS AND MECHANISMS OF ACTION OF CARBON
MONOXIDE
Principal Authors
Dr. Milan J. Hazucha
Pulmonary Division
Department of Medicine
Center for Environmental Medicine and Lung Biology
The University of North Carolina
Trailer #4, Medical Building C 224H
Chapel Hill, NC 27599
Dr. Marjolein V. Smith
2501 Anne Carol Court
Raleigh, NC 27603
Dr. Claude A. Piantadosi
Division of Allergy, Critical Care and Respiratory Medicine
Department of Medicine, Box 3315
Duke University Medical Center
Durham, NC 27710
Contributors and Reviewers
Dr. Clyde H. Barlow
Laboratory I
Evergreen State College
Olympia, WA 98505
Dr. William F. Biller
68 Yorktown Road
East Brunswick, NJ 08816
Dr. Henry J. Forman
Department of Pediatrics
Cell Biology Group (Box 83)
University of Southern California
Childrens Hospital of Los Angeles
4650 Sunset Boulevard
Los Angeles, CA 90054
xxxi
-------
Dr. Steven M, Horvath
Environmental Stress Laboratory
Neuroscience Research Institute
University of California
Santa Barbara, CA 93106
Dr. Robert Jensen
LDS Hospital
University of Utah
Salt Lake City, UT 84143
Maj. David Farmer
U.S. Army
Biomedical Research and Development Lab
Fort Detrick, Building 568
Frederick, MD 21701
Dr. Peter Tikuisis
Defence and Civil Institute of Environmental Medicine
1133 Sheppard Avenue, W
Downsview, Ontario CANADA M3M3B9
CHAPTER 10. HEALTH EFFECTS OF CARBON MONOXIDE
Principal Authors
Dr. Vernon A. Benignus
Health Effects Research Laboratory (MD-58)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Lars-Goran Ekelund
Department of Biostatistics
School of Public Health
The University of North Carolina
Suite 203, NCNB Plaza 322A
Chapel Hill, NC 27514
XXXll
-------
Dr. Laurence D. Fechter
Department of Environmental Health Sciences
The Johns Hopkins University School of Hygiene
and Public Health
615 N. Wolfe Street
Baltimore, MD 21205
Dr. Thomas R. Griggs
Division of Cardiology
School of Medicine
The University of North Carolina
349 Clinical Sciences Building 229H
Chapel Hill, NC 27599
Dr. Steven M. Horvath
Environmental Stress Laboratory
Neuroscience Research Institute
University of California
Santa Barbara, CA 93106
Dr. James J. McGrath
Department of Physiology
School of Medicine
Texas Tech University Health Sciences Center
Lubbock, TX 79430
Mr. James A. Raub
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. David S. Sheps
Division of Cardiology
School of Medicine
The University of North Carolina
338 Clinical Sciences Building 229H
Chapel Hill, NC 27599
Dr. Richard J. Traystman
Department of Anesthesiology and Critical Care Medicine
The Johns Hopkins Medical Institutions
600 N. Wolfe Street
Baltimore, MD 21205
XXXlll
-------
Contributors and Reviewers
Dr. Zoltan Annau
Department of Environmental Health Sciences
School of Hygiene and Public Health
The Johns Hopkins University
615 N. Wolfe Street
Baltimore, MD 21205
Dr. Steven M. Ayres
Medical College of Virginia, Box 565
Virginia Commonwealth University
Richmond, VA 23298
Dr. Robert L. Balster
Department of Pharmacology and Toxicology
Medical College of Virginia
Virginia Commonwealth University
Richmond, VA 23298
Dr. Thomas Clarkson
Department of Comparative Medicine
Bowman Gray School of Medicine
300 South Hawthorne Road
Winston-Salem, NC 27103
Dr. Steven D. Colome
Integrated Environmental Services
University Tower, Suite 1090
4199 Campus Drive
Irvine, CA 92715
Dr. Thomas E. Dahms
Department of Anesthesiology
St. Louis University School of Medicine
3635 Vista Avenue
St. Louis, MO 63110
Dr. Samuel Fox
Georgetown University
3800 Reservoir Road, NW
Washington, DC 20007
XXXIV
-------
Dr. Donald Heistad
Cardiology Division
Department of Internal Medicine
College of Medicine
University of Iowa
Iowa City, IA 52242
Dr. Gregory L. Hirsch
Escondido Pulmonary Medical Group
215 South Hickory
Escondido, CA 92025
Dr. John R. Holmes
State of California
Air Resources Board
1102 Q Street
P.O. Box 2815
Sacramento, CA 95812
Dr. Steven M. Horvath
Environmental Stress Laboratory
Neuroscience Research Institute
University of California
Santa Barbara, CA 93106
Dr. Michael T. Kleinman
Department of Community and Environmental Medicine
California College of Medicine
University of California
Irvine, CA 92717
Dr. Victor G. Laties
Department of Radiation Biology and Biophysics
School of Medicine and Dentistry
University of Rochester
Rochester, NY 14642
Dr. Lawrence D. Longo
School of Medicine
Department of Physiology
Division of Perinatal Biology
Loma Linda University
Loma Linda, CA 92350
xxxv
-------
Dr. George Malindzak
NIEHS North Campus (MD-3-03)
104 Alexander Drive
Research Triangle Park, NC 27709
Dr. Steve McFaul
Letterman Army Institute of Research
Presidio in San Francisco
Blood Research Division
San Francisco, CA 94129
Dr. James J. McGrath
Department of Physiology
School of Medicine
Texas Tech University Health Sciences Center
Lubbock, TX 79430
Dr. Fathy Messiha
Department of Pharmacology
University of North Dakota
Medical Science North 501
North Columbia Road
Grand Forks, ND 58201
Maj. David Farmer
U.S. Army
Biomedical Research and Development Lab
Fort Detrick, Building 568
Frederick, MD 21701
Dr. David G. Penney
Department of Physiology
Wayne State University School of Medicine
Scott Hall, Room 5374
540 East Canfield
Detroit, MI 48201
Mr. Harvey M. Richmond
Office of Air Quality Planning and Standards (MD-12)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
XXXVI
-------
Dr. Carr J. Smith
Bowman Gray Technical Center
RJ. Reynolds Tobacco Company
Winston-Salem, NC 27102
Dr. David K. Stevenson
Department of Pediatrics
Laboratory for Neonatal Metabolism
Stanford University School of Medicine
Stanford, CA 94305
Dr. Jane Warren
Health Effects Institute
215 First Street
Cambridge, MA 02142
Dr. Robert Winslow
Combat Trauma Management Directorate
Blood Research Division
Letterman Army Institute of Research
Presidio in San Francisco
San Francisco, CA 94129-6800
Dr. Ronald W. Wood
Department of Environmental Medicine
New York University Medical Center
Lanza Laboratory
Long Meadow Road
Tuxedo, NY 10987
Dr. Henk J. Vreman
Laboratory for Neonatal Metabolism
Department of Pediatrics (S214)
Stanford University School of Medicine
Stanford, CA 94305
XXXVll
-------
CHAPTER 11. COMBINED EXPOSURE OF CARBON MONOXIDE WITH OTHER
POLLUTANTS, DRUGS, AND ENVIRONMENTAL FACTORS
Principal Authors
Dr. Robert L. Balster
Department of Pharmacology and Toxicology
Medical College of Virginia
Virginia Commonwealth University
Richmond, VA 23298
Dr. Steven M. Horvath
Environmental Stress Laboratory
Neuroscience Research Institute
University of California
Santa Barbara, CA 93106
Dr. James J. McGrath
Department of Physiology
School of Medicine
Texas Tech University Health Sciences Center
Lubbock, TX 79430
Mr. James A. Raub
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Contributors and Reviewers
Dr. Zoltan Annau
Department of Environmental Health Sciences
School of Hygiene and Public Health
The Johns Hopkins University
615 N. Wolfe Street
Baltimore, MD 21205
Dr. Thomas E. Dahms
Department of Anesthesiology
St. Louis University School of Medicine
3635 Vista Avenue
St. Louis, MO 63110
Dr. Robert F. Grover
216 Mariposa Circle
Arroyo Grande, CA 93420
xxxviii
-------
Dr. Donald H. Horstman
Health Effects Research Laboratory (MD-58)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Victor G. Laties
Department of Radiation Biology and Biophysics
School of Medicine and Dentistry
University of Rochester
Rochester, NY 14642
Dr. Barbara C. Levin
Center for Fire Research
National Institute for Standards and Technology
Building 224, Room A363
Gaithersburg, MD 20899
Dr. Lawrence D. Longo
School of Medicine
Department of Physiology
Division of Perinatal Biology
Loma Linda University
Loma Linda, CA 92350
Dr. Robert Winslow
Combat Trauma Management Directorate
Blood Research Division
Letterman Army Institute of Research
Presidio in San Francisco
San Francisco, CA 94129-6800
Dr. Ronald W. Wood
Department of Environmental Medicine
New York University Medical Center
Lanza Laboratory
Long Meadow Road
Tuxedo, NY 10987
XXXIX
-------
CHAPTER 12. EVALUATION OF SUBPOPULATIONS POTENTIALLY AT RISK
Principal Authors
Dr. Robert L. Balster
Department of Pharmacology and Toxicology
Medical College of Virginia
Virginia Commonwealth University
Richmond, VA 23298
Dr. Lars-Goran Ekelund
Department of Biostatistics
School of Public Health
The University of North Carolina
Suite 203, NCNB Plaza 322A
Chapel Hill, NC 27514
Dr. Robert F. Grover
216 Mariposa Circle
Arroyo Grande, CA 93420
Dr. Steven M. Horvath
Environmental Stress Laboratory
Neuroscience Research Institute
University of California
Santa Barbara, CA 93106
Dr. David S. Sheps
Division of Cardiology
School of Medicine
The University of North Carolina
338 Clinical Sciences Building 229H
Chapel Hill, NC 27599
Contributors and Reviewers
Dr. Zoltan Annau
Department of Environmental Health Sciences
School of Hygiene and Public Health
The Johns Hopkins University
615 N. Wolfe Street
Baltimore, MD 21205
xl
-------
Dr. Thomas Clarkson
Department of Comparative Medicine
Bowman Gray School of Medicine
300 South Hawthorne Road
Winston-Salem, NC 27103
Dr. Thomas E. Dahms
Department of Anesthesiology
St. Louis University School of Medicine
3635 Vista Avenue
St. Louis, MO 63110
Dr. Samuel Fox
Georgetown University
3800 Reservoir Road, NW
Washington, DC 20007
Dr. Carr J. Smith
Bowman Gray Technical Center
RJ. Reynolds Tobacco Company
Winston-Salem, NC 27102
Dr. Ronald W. Wood
Department of Environmental Medicine
New York University Medical Center
Lanza Laboratory
Long Meadow Road
Tuxedo, NY 10987
xli
-------
PROJECT TEAM FOR DEVELOPMENT OF
AIR QUALITY CRITERIA FOR CARBON MONOXIDE
Scientific Staff
Mr. James A. Raub, Project Manager and Health Effects Coordinator
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Thomas B. McMullen, Air Quality Coordinator
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Ms. Ellie Speh, Secretary
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Technical Support Staff
Ms. Frances P. Bradow, Support and Operations Management
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Douglas B. Fennell, Technical Information Retrieval
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Allen G. Hoyt, Technical Editing and Graphics
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Ms. Diane H. Ray, Docket Information (Public Comments)
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
xliii
-------
1. SUMMARY AND CONCLUSIONS
Carbon monoxide (CO) is a colorless, odorless gas. It is a trace constituent of the
5 troposphere, produced by both natural processes and human activities. The major source of
CO in urban areas is incompletely combusted fuels; two-thirds of total emissions are
attributed to vehicle exhaust. The principal effect on humans inhaling CO-contaminated air is
reduced transport of oxygen by the blood stream, a consequence of CO displacing oxygen in
hemoglobin.
10
The Global Cycle of Carbon Monoxide
Limited data on global trends in tropospheric CO concentrations indicate a 1 to 2%
annual increase over the last several decades. Global background concentrations fall in the
range of 50 to 120 ppb; higher levels are found in the northern hemisphere, lower levels in
15 the southern hemisphere. Average background concentrations fluctuate seasonally; higher
levels occur in the winter months, lower levels in the summer months. About 60% of the CO
in the non-urban troposphere is attributed to human activities, both directly from combustion
processes, and indirectly through the oxidation of hydrocarbons and ammonia that, in turn,
arise from agricultural activities, landfills, etc. Atmospheric reactions involving CO can
20 produce O3 in the troposphere. Other reactions may deplete concentrations of the hydroxyl
radical (OH*), a key participant in the global removal cycles of many other natural and
anthropogenic trace gases, thus possibly contributing to changes in atmospheric chemistry
and, ultimately, global climate change.
25 Measurement Methods for Carbon Monoxide
The non-dispersive infrared (NDIR) optical transmission technique, the technique on
which the EPA designated reference analytical method is based, is the only technique being
used for compliance monitoring of CO. One category of NDIR monitor, the gas filter
correlation (GFC) monitor, is currently the single most widely used NDIR-type analyzer for
30 fixed-site monitoring stations. In general, NDIR monitors have significant advantages
including small size, good sensitivity and specificity for CO, and reliability of operation under
March 14, 1990 1-1 DRAFT - DO NOT QUOTE OR CITE
-------
typical network monitoring conditions. An associated recorder compiles and stores hourly
averages for subsequent computer storage and analysis.
The recent development of small, portable electrochemical monitors has made possible
the measurement of CO concentrations incurred by individuals as they move through
5 numerous diverse indoor and outdoor microenvironments that cannot be monitored by
fixed-site ambient stations. Such measurements show that some individuals can receive
exposures significantly higher than would be inferred from a simple interpretation of data
from local fixed-site stations.
10 Ambient Sources. Emissions and Concentrations of Carbon Monoxide
Current air quality standards define 1-hour and 8-hour average concentrations that
should not be exceeded more than once per year. The 1-hour standard of 35 ppm is almost
never exceeded in data reported from fixed-site monitoring stations. Concentrations of CO
exceeding the 8-hour standard of 9 ppm have declined over the 10-year period 1979 to 1988
15 from an average of about ten per station per year to about two per year (U.S. Environmental
Protection Agency, 1990). This decline reflects the efficacy of emission control systems on
newer vehicles. In 1979, vehicular emissions of CO accounted for about 72% of total U. S.
emissions; in 1988, it was 67%. During this same period, there was a 33% increase in
highway vehicle miles traveled. The other categories of CO emissions are other fuel
20 combustion sources, such as steam boilers (12%), industrial processes (8%), Solid waste
disposal (3%), and miscellaneous other sources (10%).
Indoor Sources. Emissions, and Concentrations of Carbon Monoxide
EPA's mandate is to monitor and regulate pollutants in the ambient, i.e. outdoor, air;
25 however, the great majority of people spend most of their time indoors, one place or another.
A realistic assessment of ambient exposure to CO, therefore, must be set in the context of
total exposure, a major component of which is exposure while indoors.
Indoor concentrations of CO are a function of outdoor concentrations, indoor sources,
infiltration, ventilation and air mixing between and within rooms. In residences without
30 sources, average CO concentrations are approximately equal to average outdoor levels.
March 14, 1990 1-2 DRAFT - DO NOT QUOTE OR CITE
-------
Proximity to outdoor sources (e.g., heavily traveled roadways, attached garages, or parking
garages) can have a major impact on indoor CO concentrations.
The extensive total personal CO exposure studies conducted by EPA in Washington, DC
and Denver, CO (Akland et al., 1985; Whitmore et al., 1984; Hartwell et al., 1984; Johnson,
5 1984) have shown that the highest CO concentrations occur in indoor microenvironments
associated with transportation sources. Concentrations in these environments frequently were
found to exceed 9 ppm. Studies targeted toward specific indoor microenvironments have also
identified the indoor commuting microenvironment as one in which CO concentrations
frequently exceed 9 ppm and occasionally exceed 35 ppm. Similar concentrations have been
10 reported in special environments or accompanying unusual occurrences (indoor ice skating
rinks, offices where emissions from parking garages migrate indoors, etc.).
Unvented gas and kerosene space heaters that are used for substantial periods of time
appear to be the major contributors to residential CO concentrations. Peak concentrations of
CO in such residences often exceed an 8-hour average of 9 ppm and a 1-hour average of
15 35 ppm. One extensive study (Koontz and Nagda, 1988) of unvented gas space heaters
indicated that 12 % of the homes had 15-hour average CO concentrations greater than 8 ppm;
the highest recorded concentration was 36.6 ppm.
Intermittent sources such as gas cooking ranges can result in high peak CO
concentrations (in excess of 9 ppm), while long-term average concentrations (i.e., 24 hours)
20 associated with gas ranges are on the order of 1 ppm. The contribution of tobacco
combustion to indoor CO levels is variable. One study suggested that its contribution to
residential CO concentrations was on the order of 1 ppm; another study showed no significant
increase.
Very limited data on CO levels in residences with wood burning stoves or fireplaces
25 indicate non-airtight stoves can contribute substantially to residential CO concentrations;
airtight stoves can contribute small increases. The available data indicate that fireplaces do
not contribute measurably to average indoor concentrations. No information is available on
residences with leaky flues or on the impact of attached garages.
The available data on the spatial and temporal variability of indoor CO concentrations as
30 a function of microenvironments and associated sources are not adequate to assess exposures
March 14, 1990 1-3 DRAFT - DO NOT QUOTE OR CITE
-------
in these environments. These indoor microenvironments represent the most important CO
exposures for the majority of individuals and as such need to be better characterized.
Population Exposure to Carbon Monoxide
5 The current NAAQS for CO (9 ppm for 8 hours, 35 ppm for 1 hour) are designed to
protect against actual and potential human exposures in outdoor air that would cause adverse
health effects. Compliance with the NAAQS is determined by measurements taken at
fixed-site ambient monitors, the use of which is intended to provide some measure of the
general level of exposure of the population represented by the CO monitors. Results of both
10 exposure monitoring in the field, and modeling studies, summarized in this document indicate
that individual personal exposure does not directly correlate with CO concentrations
determined by using fixed-site monitors alone. This observation is due to the mobility of
people and to the spatial and temporal variability of CO concentrations. While failing to
show a correlation between individual personal monitor exposures and simultaneous nearest
15 fixed-site monitor concentrations, studies do suggest that aggregate personal exposures are
lower on days of lower ambient CO levels as determined by the fixed-site monitors and
higher on days of higher ambient levels.
A unique feature of carbon monoxide exposure is that there is a biological marker of the
dose that the individual has received: the blood level of CO. This level may be calculated by
20 measuring blood carboxyhemoglobin (COHb) or by measuring CO in exhaled breath.
The use of CO-Oximeters to measure low levels of COHb can provide useful
information regarding mean values, provided a reference technique is used to properly
calibrate the instrument. It has been shown, however, that the range of values obtained with
this optical method will be greater than that obtained with a reference method. For example,
25 in a group of subjects with cardiovascular disease, the standard deviation of the percent
COHb values for non-smoking, resting subjects was 2 to 2.5 times greater for the CO-
Oximeter values than for the gas chromatograph values on paired samples (Allred et al.,
1989b). Therefore, the potential exists with the CO-Oximeter for having an incorrect
absolute value for COHb as well as an incorrectly broadened range of values.
30 In addition, it is not clear exactly how sensitive the CO-Oximeter techniques are to
small changes in COHb at the low end of the CO dissociation curve. Allred et al. (1989b)
March 14, 1990 1-4 DRAFT - DO NOT QUOTE OR CITE
-------
have noted that the interference from changing O2 saturation can have a very significant
influence on the apparent COHb reading in a sample. The interaction between hemoglobin
species was also reported by Dennis and Valeri (1980). This suggests nonlinearity or a
disproportionality in the absorption spectrum of these two species of hemoglobin. It is also a
5 potential source of considerable error in the estimation of COHb by optical methods.
The measurement of exhaled breath has the advantages of ease, speed, precision, and
greater subject acceptance than measurement of blood COHb. However, the accuracy of the
breath measurement procedure and the validity of the Haldane relationship between breath and
blood at low environmental CO concentrations remains in question. There appears to be a
10 clear research need to validate the breath method at low CO exposures. In view of the
possible problems with the CO-Oximeter, such validation should be done using gas
chromatography for the blood COHb measurements.
Cigarette consumption represents a special case of CO exposure; for the smoker it
almost always dominates over personal exposure from other sources. Studies by Radford and
15 Drizd (1982) show that COHb levels of cigarette smokers average 4% while those of
nonsmokers average 1 %. Therefore, this document focuses on environmental exposure of
nonsmokers to CO.
People encounter CO in a variety of environments that include travelling in motor
vehicles, working at their jobs, visiting urban locations associated with combustion sources,
20 or cooking over a gas range. Studies of human exposure have shown that among these
settings the motor vehicle is the most important for regularly encountered elevations of CO.
Studies by Flachsbart et al. (1987) indicated that CO exposures while commuting in
Washington, D.C. average 9 to 14 ppm at the same time that fixed station monitors record
concentrations of 2.7 to 3.1 ppm. Similar studies conducted by EPA in Denver and
25 Washington, D.C., have demonstrated that the motor vehicle interior has the highest average
CO concentrations (averaging 7 to 10 ppm) of all microenvironments (Johnson, 1984). In
these studies, 8% of all commuters experienced 8-hour exposures greater than 9 ppm while
only 1 % of noncommuters received exposures over that level. Furthermore, commuting
exposures have been shown to be highly variable with some commuters breathing CO in
30 excess of 25 ppm.
March 14, 1990 1-5 DRAFT - DO NOT QUOTE OR CITE
-------
Another important setting for CO exposure is the workplace. In general, exposures at
work exceed CO exposures during nonwork periods, apart from commuting to and from
work. Average concentrations may be elevated during this period since workplaces are often
located in congested areas that have higher background CO concentrations than do many
5 residential neighborhoods. Occupational and nonoccupational exposures may overlay one
another and result in a higher concentration of CO in the blood. Certain occupations also
increase the risk of high CO exposure (e.g., those occupations involved directly with vehicle
driving, maintenance, and parking). Occupational groups exposed to CO by vehicle exhaust
include auto mechanics; parking garage and gas station attendants; bus, truck or taxi drivers;
10 police; and warehouse workers. Other industrial processes produce CO directly or as a by-
product, including steel production, coke ovens, carbon black production, and petroleum
refining. Firefighters, cooks, and construction workers also may be exposed at work to
higher CO levels. Occupational exposure in industries or setting with CO production also
represent some of the highest individual exposures observed in field monitoring studies. For
15 example, in EPA's CO exposure study in Washington, DC, of the approximately 4% (29 of
712) of subjects working in jobs classified as having a high potential for CO exposure, seven
subjects (or approximately 25%) experienced 8-hour CO exposures in excess of 9 ppm.
The highest indoor nonoccupational CO exposures are associated with combustion
sources and include enclosed parking garages, service stations, restaurants and stores. The
20 lowest indoor CO concentrations are found in homes, churches, and health care facilities.
EPA's Denver Study showed that passive cigarette smoke is associated with increasing a
nonsmoker's exposure by an average of about 1.5 ppm and that use of a gas range is
associated with about 2.5 ppm increase at home. Other sources which may contribute to CO
in the home include combustion space heaters and wood burning stoves.
25 As noted above, people encounter different and often higher exposures than predicted
from fixed-site monitoring data because of the highly localized nature of CO sources. For
example, during the winter sampling period, 10% of Denver volunteers and 4% of
Washington, DC volunteers recorded personal exposures in excess of 9 ppm for 8 hours.
Breath measurements from the Washington volunteers indicated that as much as 9% of the
30 population could have experienced a 9 ppm, 8-hour average. In contrast, during the entire
winter period of 1982-1983, the two ambient CO monitors in Washington reported only one
March 14, 1990 1-6 DRAFT - DO NOT QUOTE OR CITE
-------
exceedance of the 9-ppm level. In another study, using data from analyses of COHb in
blood, Wallace and Ziegenfus (1985) report that CO in blood is uncorrelated with CO
measured by ambient monitors. These findings point out the necessity of having personal CO
measurements to augment fixed-site ambient monitoring data when total human exposure is to
5 be evaluated. Data from these field studies can be used to construct and test models of
human exposure that account for time and activity patterns known to affect exposure to CO.
Models developed to date tend to underpredict the variability of CO exposures observed in
field studies and have not been able to successfully predict individual exposures. The models
may be modified and adjusted using information from field monitoring studies in order to
10 capture the observed distribution of CO exposures, including the higher exposures found in
the tail of the exposure distribution. The models also are useful for evaluating alternative
pollutant control strategies.
Pharmacokinetics and Mechanisms of Action of Carbon Monoxide
15 This section of the document reviews the basic relationships of O2 and CO with Hb and
other O2 binding proteins, outlines the fundamentals of CO uptake and elimination from blood
and extravascular space, examines the mechanisms of CO toxicity, and evaluates the models
of COHb formation.
The exchange of CO between the atmosphere (airway opening) and RBC is controlled
20 by physical (mass transport, diffusion) and physiological (alveolar ventilation, cardiac output,
etc.) processes. The final step of the competitive binding between CO and O2 to Hb to form
COHb and O2Hb, respectively, is a complex sequence of reversible reactions, the kinetics of
which are still not fully understood. The toxic effects of CO are due to its high affinity for
Hb. The presence of CO reduces O2-carrying capacity of blood and impairs release of O2
25 from O2Hb to extravascular tissues. Brain and heart tissues are particularly sensitive to the
resultant drop in PO2 and CO hypoxia. Because of tight binding of CO to Hb, the elimination
half-time is quite long. This might lead to accumulation of COHb and even relatively low
concentrations of CO might produce substantial blood levels of COHb.
The rate of rise of blood COHb levels as well as the time to equilibrium saturation
30 depends on the inhaled concentration of CO, alveolar ventilation, lung diffusion, cardiac
March 14, 1990 1-7 DRAFT - DO NOT QUOTE OR CITE
-------
output, and duration of exposure. These factors have been integrated into many empirical and
mathematical models of COHb formation under static and dynamic conditions.
The best all around model for COHb prediction is still the equation developed by
Coburn, Forster, and Kane (1965). The linear solution is useful for examining air pollution
5 data leading to relatively low COHb levels, whereas the nonlinear solution shows good
predictive power even for high CO exposures. The two regression models might be useful
only when the conditions of application closely approximate those under which the parameters
were estimated.
It is important to remember that almost all of the available modeling studies assumed a
10 constant rate of CO uptake and elimination, which is rarely true. A number of physiological
factors, particularly changes in ventilation, will affect both rates. The predicted COHb values
also will differ from individual to individual due to smoking, age, or lung disease. There
does not appear to be a single optimal averaging time period for ambient CO; however, the
shorter the period the greater the precision. In general, the averaging time period should be
15 well within the [COHb] half-life, which decreases with increased activity.
Although the principal cause of CO toxicity is tissue hypoxia due to CO binding to Hb,
certain physiological aspects of CO exposure are not explained well by decreases in
intracellular PO2 related to the presence of COHb. For many years, it has been known that
CO is distributed to extravascular sites such as skeletal muscle (Coburn et al., 1971; Coburn
20 et al., 1973) and that 10 to 50% of the total body store of CO may be extravascular
(Luomanmaki and Coburn, 1969). Furthermore, extravascular CO is metabolized slowly to
CO2 in vivo (Fenn, 1970). Consequently, secondary mechanisms of CO toxicity related to
intracellular uptake of CO have been the focus of a great deal of research interest. CO
binding to many intracellular compounds has been well documented both in vitro and in vivo;
25 however, it is still uncertain whether or not intracellular uptake of CO in the presence of Hb
is sufficient to cause either acute organ system dysfunction or long-term health effects. The
virtual absence of sensitive techniques capable of assessing intracellular CO binding under
physiological conditions has resulted in a variety of indirect approaches to the problem as well
as many negative studies.
30 Current knowledge pertaining to intracellular CO-binding proteins suggests that the most
likely ones to be inhibited functionally at relevant levels of COHb are myoglobin (Mb), found
March 14, 1990 1-8 DRAFT - DO NOT QUOTE OR CITE
-------
predominantly in heart and skeletal muscle, and cytochrome oxidase. The physiological
significance of CO uptake by Mb is uncertain at this time but sufficient concentrations of
carboxymyoglobin (COMb) could potentially limit maximal O2 uptake of exercising muscle.
Although there is suggestive evidence for significant binding of CO to cytochrome oxidase in
5 heart and brain tissue, it is unlikely that any significant CO binding would occur at low
COHb levels. Therefore, further research is needed to determine if secondary, intracellular
mechanisms will occur at exposure concentrations found in ambient air.
Health Effects of Carbon Monoxide
10 Concerns about the potential health effects of exposure to carbon monoxide have been
addressed in extensive studies with various animal species as subjects. Under varied
experimental protocols, considerable information has been obtained on the toxicity of CO, its
direct effects on the blood and other tissues, and the manifestations of these effects in the
form of changes in organ function. Many of these studies, however, have been conducted at
15 extremely high levels of CO (i.e., levels not found in ambient air). Although severe effects
from exposure to these high levels of CO are not directly germane to the problems from
exposure to current ambient levels of CO, they can provide valuable information about
potential effects of accidental exposure to CO, particularly those exposures occurring indoors.
20 Acute Pulmonary Effects
Currently available studies on the effects of CO exposures producing COHb
concentrations of up to 39% fail to find any consistent effects on lung parenchyma and
vasculature (Hugod, 1980; Fisher et al., 1969) or on alveolar macrophages (Chen et al.,
1982; Weissbecker et al., 1969). The lack of significant changes in lung tissue is consistent
25 with the lack of histologic changes in the pulmonary and coronary arteries. Alveolar
epithelial permeability to 51Cr-EDTA increased in rabbits (Fein et al., 1980) exposed to high
concentrations of CO (63% COHb), and increased capillary endothelial permeability to 131I-
labeled human serum albumin was reported in early human studies (Parving, 1972) following
acute, high-level CO exposure (23% COHb); however, no accumulation of lung water was
30 found in dogs (Halebian et al., 1984a,b) with COHb levels of 59% and no edema was found
in the lungs of rats chronically exposed to CO concentrations as high as 1300 ppm (Penney
March 14, 1990 1-9 DRAFT - DO NOT QUOTE OR CITE
-------
et al., 1988). In addition, no changes in diffusing capacity of the lung were found in dogs
with COHb levels up to 18% (Fisher et al., 1969). It is unlikely, therefore, that CO has any
direct effect on lung tissue except at extremely high concentrations. The capillary endothelial
and alveolar epithelial edema found with high levels of CO exposure in victims of CO
5 poisoning may be secondary to cardiac failure produced by myocardial hypoxia (Fisher et al.,
1969) or may be due to acute cerebral anoxia (Naeije et al., 1980).
Ventilatory responses to CO are related to the CO concentration as well as to the
experimental conditions and the animal species being studied. In conscious goats (Chapman
et al., 1980; Doblar et al., 1977; Santiago and Edelman, 1976) and cats (Gautier and Bonora,
10 1983), after an initial depression, ventilation suddenly increases, particularly at high CO
concentrations (>2000 ppm). This response may result from the direct effects of hypoxia
(and possibly central acidosis) and/or a specific CNS effect of CO. No effects on ventilation
and perfusion distribution were found, however, in dogs exposed to 1 % CO for 10 min,
resulting in COHb levels of 59% (Robinson et al., 1985). At very high concentrations of CO
15 (COHb >60%) total pulmonary resistance, measured indirectly by trachea! pressure, was
reported to increase (Mordelet-Dambrine et al., 1978; Mordelet-Dambrine and Stupfel, 1979).
Human studies on the pulmonary function effects of CO are complicated by the lack of
adequate exposure information, the small number of subjects studied, and the short exposures
explored. Occupational or accidental exposure to the products of combustion and pyrolysis,
20 particularly indoors, may lead to acute decrements in lung function if the COHb levels are
> 17% (Sheppard et al., 1986) but not at concentrations <2% (Evans et al., 1988; Hagberg
et al., 1985; Cooper and Alberti, 1984). It is difficult, however, to separate the potential
effects of CO from those due to other respiratory irritants in the smoke and exhaust.
Community population studies on CO in ambient air have not found any relationships with
25 pulmonary function, symptomatology, and disease (Lebowitz et al., 1987; Robertson and
Lebowitz, 1984; Lutz, 1983).
Cardiovascular Effects
The 1984 Addendum to the 1979 Air Quality Criteria Document for Carbon Monoxide
30 (U.S. Environmental Protection Agency, 1984) reported what appears to be a linear
relationship between level of COHb and decrements in human exercise performance,
March 14, 1990 1-10 DRAFT - DO NOT QUOTE OR CITE
-------
measured as maximal O2 uptake. Exercise performance consistently decreases at a blood level
of about 5.0% COHb in young, healthy, nonsmoking individuals (Klein et al., 1980; Stewart
et al., 1978; Weiser et al., 1978). Some studies have even observed a decrease in
performance at levels as low as 2.3 to 4.3% COHb (Horvath et al., 1975; Drinkwater et al.,
5 1974; Raven et al., 1974a); however, this decrease is so small as to be of concern mainly for
competing athletes rather than for ordinary people conducting the activities of daily life.
Cigarette smoking has a similar effect on cardiorespiratory response to exercise in nonathletic
human subjects indicating a reduced ability for sustained work (Hirsch et al., 1985; Klausen
etal., 1983).
10 Since the 1979 Air Quality Criteria Document (U.S. Environmental Protection Agency,
1979), several important studies appearing in the literature have expanded the cardiovascular
data base. Adverse effects in patients with reproducible exercise-induced angina (Allred
et al., 1989a,b) have been noted with postexposure COHb levels (CO-Oximeter measurement)
as low as 3.2% (corresponding to an increase of 2.0% from baseline). Sheps et al. (1987)
15 also found a similar effect in a group of patients with angina at COHb levels of 3.8%
(representing an increase of 2.2% from baseline). Kleinman et al. (1989) studied subjects
with angina and found an effect at 3% COHb representing an increase of 1.5% from baseline.
Thus, the lowest observed adverse effect level in patients with stable angina is somewhere
between 3 and 4% COHb (CO-Oximeter measurement), representing an increase from
20 baseline of from 1.5 to 2.2%. Effects on silent ischemia episodes, which represent the
majority of episodes in these patients, have not been studied.
Exposure sufficient to achieve 6% COHb recently has been shown to adversely affect
exercise-related arrhythmia in patients with coronary artery disease (Sheps et al., 1989). This
finding combined with the epidemiologic work of Stern et al. (1988) in tunnel workers is
25 suggestive but not conclusive that CO exposure may provide an increased risk of sudden death
from arrhythmia in patients with coronary artery disease.
There is also strong evidence from both theoretical considerations and experimental
studies in animals that carbon monoxide can adversely affect the cardiovascular system.
Results from animal studies suggest that inhaled CO can cause disturbances in cardiac rhythm
30 and conduction in healthy as well as cardiac-impaired animals. The lowest observed effect
level varies, depending upon the exposure regime used and species tested. Values reported in
March 14, 1990 1-11 DRAFT - DO NOT QUOTE OR CITE
-------
the literature for 6 to 24 week exposures range from 2,6 to 12.0% COHb (50 and 100 ppm
CO) in dogs to 12.4% (100 ppm CO) in monkeys. For shorter duration exposures of 0.6 to
16 h, disturbances in cardiac rhythm were found at 4.9 to 17.0% COHb (500 ppm CO) in
dogs and 9.3% COHb (100 ppm CO) in monkeys. Results from animal studies also indicate
5 that inhaled CO can increase hemoglobin concentration and hematocrit ratio at COHb levels
of 9.26% (100 ppm CO). Small increases in hemoglobin and hematocrit probably represent a
compensation for the reduction in oxygen transport caused by CO. At higher CO
concentrations, excessive increases in hemoglobin and hematocrit may impose an additional
workload on the heart and compromise blood flow to the tissues. For example, cardiomegaly
10 has been reported in adult animals at COHb levels of 12% (200 ppm CO). The oxygen
transport system of the fetus, however, is especially sensitive to CO inhaled by the mother,
and may be affected by CO at concentrations as low as 60 ppm.
There is conflicting evidence that CO exposure will enhance development of
atherosclerosis in laboratory animals; and most studies show no measurable effect. Similarly,
15 the possibility that CO will promote significant changes in lipid metabolism that might
accelerate atherosclerosis is suggested in only a few studies. Any such effect must be subtle
at most. Finally, CO probably inhibits rather than promotes platelet aggregation. Except for
the studies by Rogers et al. (1980, 1988) on baboons, the CO exposures used in the studies
on atherosclerosis created COHb levels of 7% or higher; sometimes much higher. While
20 occupational exposures in some workplace situations might regularly lead to COHb levels of
10% or more, such high exposure levels are almost never encountered in the
nonoccupationally exposed general public. In this general population, exposures are rarely as
much as 25 to 50 ppm, and COHb levels typically are below 3% in nonsmokers. When
examined in this context, this document therefore, provides little data to indicate that an
25 atherogenic effect of exposure would be likely to occur in human populations at commonly
encountered levels of ambient CO.
Cerebrovascular and Behavioral Effects
The data reviewed in this document indicate that carbon monoxide hypoxia increases
30 cerebral blood flow, even at very low exposure levels. Cerebral O2 consumption is well
maintained until levels of COHb reach upwards of 30%. The overall responses of the
March 14, 1990 1-12 DRAFT - DO NOT QUOTE OR CITE
-------
cerebrovasculature are similar in the fetus, newborn, and adult animal; however, the
mechanism of the increase in cerebral blood flow is still unclear. In fact, several mechanisms
working simultaneously to increase cerebral blood flow appear likely and these may involve
metabolic and neural aspects as well as the oxyhemoglobin dissociation curve, tissue O2
5 levels, and even a histotoxic effect of CO. These potential mechanisms of CO-induced
alterations in the cerebral circulation need to be investigated further. The interaction of CO
with cyanide (additive and synergistic) on the cerebral vasculature is clear, however the
interaction of CO with other agents and their combined effects on brain blood vessels is
unknown. This also is true for the long term (chronic) effects of CO alone, or in
10 combination with other agents in low- or high-dose levels on the cerebral vasculature.
Finally, under normal circumstances the brain can increase its blood flow or its O2 extraction
in order to compensate for a reduced O2 environment. Whether these compensatory
mechanisms continue to operate successfully in a variety of conditions where the brain, or its
vasculature are compromised (i.e., stroke, head injury, atherosclerosis, hypertension) is
15 unknown and requires further investigation.
Behaviors that require sustained attention and/or sustained performance are most
sensitive to disruption by COHb. Therefore, the group of studies of tracking, vigilance, and
continuous performance offer the most consistent and defensible evidence of COHb effects on
behavior. The results across studies is, however, far from consistent. Further examination of
20 the three areas seems appropriate.
Compensatory tracking was studied by two groups of investigators using virtually
identical task parameters and equipment (Putz et al., 1976, 1979; Benignus et al., 1987,
1989). Both of the studies by Putz et al. (1976, 1979) found significant and moderately large
effects of 5% COHb. Benignus et al. (1987) reported similar but smaller significant effects in
25 a nearly identical experiment to Putz et al. (1976). However, in a dose-effects study
including another direct replication group, Benignus et al. (1989a) found no significant
effects, even for COHb levels of 17%. In the latter study, the means were nearly doseordinal
but too small to be statistically significant. It is particularly puzzling why the latter study,
using a large number of subjects on an identical task, should find no significant effects for
30 even 17% COHb when three other studies found effects at lower levels. Three other double-
blind tracking studies of various methods found no effects of COHb levels of 12% or greater.
March 14, 1990 1-13 DRAFT - DO NOT QUOTE OR CITE
-------
There is a similar disunity among studies on the effects of COHb on vigilance. Because of
the many failed attempts at direct replication, the conclusions seem weaker than for tracking.
Of the five double-blind experiments in which continuous performance was measured,
three were mentioned earlier in the discussion of tracking. In these studies (direct
5 replications), continuous performance was measured simultaneously with tracking (Putz et al.,
1976, 1979; Benignus et al., 1987). The latter of the three found no effects. A small study
reported continuous performance effects that were disordinal in COHb (O'Donnell et al.,
1971). The remaining study (Benignus et al., 1977) used a different task and obtained no
COHb effects.
10 As discussed above, a proportional vasodilation occurs in the brain in response to COHb
elevation. This vasodilation is sufficient, on the average, to keep the cerebral O2 consumption
from being reduced even though the COHb has reduced the blood's O2-carrying capacity 20 to
30% and the presence of COHb has shifted the oxyhemoglobin dissociation curve to the left.
The cerebral vasodilation may be viewed, Ideologically, as a closed-loop compensatory
15 mechanism to assure adequate oxygenation of the brain in the presence of elevated COHb. If
the cerebral vasodilation is adequate in any individual and if the vasodilation is homogeneous
for all cerebral tissue, then that individual should not be behaviorally impaired by COHb
elevation. This statement assumes that the sole mechanism for CO toxicity is the hypoxic
effect of COHb.
20 The agreement between the behavioral literature and the compensatory mechanism
hypothesis is noteworthy. According to the compensatory mechanism data, O2 consumption
in the brain does not begin to decrease until COHb exceeds 20 to 30%. Data from behavioral
studies in laboratory animals demonstrate that significant effects in schedule-controlled
behavior do not occur below 20 to 30% COHb. Behavioral effects in healthy humans have
25 not been clearly demonstrated below 20 to 30% COHb. It seems unwise, however, to totally
ignore the evidence suggesting effects below these levels. Even if effects are small or
occasional, they might be important to the performance of critical tasks in some individuals.
Some of the differences among studies of the effect of COHb on the behavior of humans
are due apparently to technical problems in the execution of experiments, because single-
30 blind or nonblind experiments tend to yield a much higher rate of significant effects than do
double-blind studies. Even when non-double-blind experiments are eliminated from
March 14, 1990 1-14 DRAFT - DO NOT QUOTE OR CITE
-------
consideration, however, a substantial amount of disparity remains among results of studies. It
is possible that such residual disagreement is due to the action of an unsuspected variable that
is not being controlled across experiments.
If the compensatory CNS blood flow hypothesis has validity, it is possible that there
5 exist groups that are at higher risk to COHb elevation than the usual subjects who were
studied in the behavioral experiments. Disease or injury might either impair the
compensatory mechanism or reduce the non-exposed O2 delivery. Aging increases the
probability of such injury and disease. It also is possible that there exist individual
differences with regard to COHb sensitivity and/or compensatory mechanisms. Too little is
10 known about the compensatory process to make conjectures, but the matters seem important
to investigate.
The literature on the behavioral effects of COHb elevation has grown considerably since
the last Criteria Document was written (U.S. Environmental Protection Agency, 1979). It
seems safe to state that the effect of the new information did not increase the certainty about
15 COHb effects. Unless some key piece of information is uncovered by new research, there
does not seem to be much hope of gaining clarification in the conflicting findings. The
solution to the puzzle would seem to lie in the conduct of more research into mechanisms of
action of CO rather than in further attempts to show reliable behavioral effects. The latter
approach, which has not been successful in the past, should be resumed only when
20 mechanisms of toxicity are understood better. More findings of behavioral effects of COHb
would not appreciably alter the conclusions of this document unless future studies were to
show an unusual unanimity.
Developmental Toxicity
25 The data reviewed in this document provide strong evidence that CO exposures of 150
to 200 ppm produce reductions in birthweight, cardiomegaly, delays in behavioral
development, and disruption in cognitive function in laboratory animals of several species.
Isolated experiments suggest that some of these effects may be present at doses as low as 60
to 65 ppm maintained throughout gestation. The current data from human children suggesting
30 a link between environmental CO exposures and sudden infant death syndrome (SIDS) are
weak, but further study should be encouraged. Human data from cases of accidental high
March 14, 1990 1-15 DRAFT - DO NOT QUOTE OR CITE
-------
dose CO exposures are difficult to use in identifying a lowest observed effect level (LOEL) or
a no observed effects level (NOEL) for this agent because of the small numbers of cases
reviewed and problems in documenting levels of exposure. However, such data if
systematically gathered and reported could be useful in identifying possible ages of special
5 sensitivity to CO and cofactors or other risk factors that might identify sensitive
subpopulations.
Other Systemic Effects of Carbon Monoxide
Laboratory animal studies reviewed in the previous criteria document (U.S.
10 Environmental Protection Agency, 1979) and again in this document suggest that enzyme
metabolism and the P-450-mediated metabolism of xenobiotic compounds may be affected by
CO exposure (Montgomery and Rubin, 1971; Kustov et al., 1972; Pankow and Ponsold,
1972, 1974; Martynjuk and Dacenko, 1973; Swiecicki, 1973; Pankow et al., 1974; Roth and
Rubin, 1976a,b). Most of the authors have concluded, however, that effects on metabolism
15 at low COHb levels (<15%) are attributable entirely to tissue hypoxia produced by increased
levels of COHb because they are no greater than the effects produced by comparable levels of
hypoxic hypoxia. At higher levels of exposure, where COHb concentrations exceed 15 to
20%, there may be direct inhibitory effects of CO on the activity of mixed-function oxidases
but more basic research is needed. The decreases in xenobiotic metabolism shown with CO
20 exposure might be important to individuals receiving treatment with drugs.
The effects of CO on tissue metabolism noted above may partially explain the body
weight changes associated with exposure to high concentrations of CO. Short-term exposure
to 250-1000 ppm for 24 h was reported previously to cause weight loss in laboratory rats
(Koob et al., 1974) but no significant body weight effects were reported in long-term
25 exposure studies in laboratory animals at CO concentrations ranging from 50 ppm for 3 mo to
3000 ppm for 300 days (Theodore et al., 1971; Musselman et al., 1959; Campbell, 1934;
Stupfel and Bouley, 1970).
Inhalation of high levels of CO, leading to COHb concentrations greater than 10 to
15%, have been reported to cause a number of systemic effects in laboratory animals as well
30 as effects in humans suffering from acute CO poisoning. Tissues of highly active oxygen
metabolism, such as heart, brain, liver, kidney, and muscle, may be particularly sensitive to
March 14, 1990 1-16 DRAFT - DO NOT QUOTE OR CITE
-------
CO poisoning. The impairment of function in the heart and brain caused by CO exposure is
well known and has been described above. Other systemic effects of CO poisoning are not as
well known and are, therefore, less certain. There are reports in the literature of effects on
liver (Katsumata et al., 1980), kidney (Kuska et al., 1980), and bone (Zebro et al., 1983).
5 Results from one additional study in adult guinea pigs suggest that immune capacity in the
lung and spleen was affected by intermittent exposure to high levels of CO for 3 to 4 weeks
(Snella and Rylander, 1979). It generally is agreed that these systemic effects are caused by
the severe tissue damage occurring during acute CO poisoning due to (1) ischemia resulting
from the formation of COHb, (2) inhibition of O2 release from HbO2, (3) inhibition of cellular
10 cytochrome function (e.g., cytochrome oxidases), and (4) metabolic acidosis.
Adaptation
The only evidence for short- or long-term COHb compensation in man is indirect.
Experimental animal data indicate that COHb levels produce physiological responses that tend
15 to offset other deleterious effects of CO exposure. Such responses are (1) increased coronary
blood flow, (2) increased cerebral blood flow, (3) increased hemoglobin through increased
hemopoiesis, and (4) increased O2 consumption in muscle.
Short-term compensatory responses in blood flow or O2 consumption may not be
complete or might even be lacking in certain persons. For example, from laboratory animal
20 studies it is known that coronary blood flow is increased with COHb, and from human
clinical studies it is known that subjects with ischemic heart disease respond to the lowest
levels of COHb (5%, or less). The implication is that in some cases of cardiac impairment,
the short-term compensatory mechanism is impaired.
From neurobehavioral studies, it is apparent that decrements due to CO have not
25 occurred consistently in all subjects, or even in the same studies, and have not demonstrated a
dose-response relationship with increasing COHb levels. The implication from this data
suggests that there might be some threshold or time lag in a compensatory mechanism such as
increased cerebral blood flow. Without direct physiological evidence in either laboratory
animals or, preferably humans, this concept only can be hypothesized. The observed results
30 from the neurobehavioral studies could be explained by differences or problems in
experimental protocols or due to possible nonrandom sampling.
March 14, 1990 1-17 DRAFT - DO NOT QUOTE OR CITE
-------
The idea of a threshold or a time lag in compensatory mechanisms should not be
rejected entirely, however. There simply is no direct evidence. Studies need to be performed
to (1) measure cerebral blood flow and tissue PO2 with low COHb levels at various ambient
concentrations of CO to determine early and low level effects accurately, and (2) design
5 behavioral studies where threshold effects or time lags are factors in the experimental
protocols that can be explicitly studied.
The mechanism by which long-term adaptation would occur, if it could be demonstrated
in humans, is assumed to be an increased Hb concentration via a several-day increase in
hemopoiesis. This alteration in Hb production has been demonstrated repeatedly in animal
10 studies but no recent studies have been conducted indicating or suggesting that some
adaptational benefit has or would occur. Furthermore, even if the Hb increase is a signature
of adaptation, it has not been demonstrated to occur at low ambient concentrations of CO.
The human studies of the 1940s have not been replicated, so the question of adaptation
remains unresolved.
15
Combined Exposure of Carbon Monoxide with Other Pollutants. Drugs, and Environmental
Factors
High Altitude Effects
20 While there are many studies comparing and contrasting inhaling CO with exposure to
altitude, there are relatively few reports on the effects of inhaling CO at altitude. There are
data to support the possibility that the effects of these two hypoxia episodes are at least
additive. These data were obtained at CO concentrations that are too high to have much
meaning for regulatory concerns. There also are data that indicate decrements in visual
25 sensitivity and flicker-fusion frequency in subjects exposed to CO (COHb = 5 to 10%) at
higher altitudes. These data, however, are somewhat controversial.
There are even fewer studies of the long-term effects of CO at high altitude. These
studies generally indicate few changes at CO concentrations below 100 ppm and altitudes
below 4572 m (15,000 ft). A provocative study by McDonagh et al. (1986) suggests that the
30 increase in ventricular capillarity seen with altitude exposure may be blocked by CO. The
fetus may be particularly sensitive to the effects of CO at altitude; this is especially true with
the high levels of CO associated with maternal smoking.
March 14, 1990 1-18 DRAFT - DO NOT QUOTE OR CITE
-------
Carbon Monoxide Interaction with Drugs
There remains little direct information on the possible enhancement of CO toxicity by
concomitant drug use or abuse; however, there are some data suggesting cause for concern.
There is some evidence that interactions of drug effects with CO exposure can occur in both
5 directions, that is, CO toxicity may be enhanced by drug use and the toxic or other effects of
drugs may be altered by CO exposure. Nearly all the published data that are available on CO
combinations with drugs concern psychoactive drugs.
The use and abuse of psychoactive drugs and alcohol is ubiquitous in society. Because
of CO's effects on brain functioning, interactions between CO and psychoactive drugs could
10 be anticipated. Unfortunately, very little systematic research has addressed this question. In
addition, very little of the research that has been done has utilized models for expected effects
for treatment combinations. Thus, often it is not possible to assess whether the combined
effects of drugs and CO exposure are additive or differ from additivity. It is important to
recognize that even additive effects of combinations can be of clinical significance, especially
15 when the individual is unaware of the combined hazard.
Combined Exposure of Carbon Monoxide with Other Air Pollutants
and Environmental Factors
Much of the data concerning the combined effects of CO and other pollutants found in
20 the ambient air are based on animal experiments. Only a few human studies are available.
Early studies in healthy human subjects by Hackney et al. (1975a,b), Raven et al. (1974a,b),
Gliner et al. (1975), and Drinkwater et al. (1974) on common air pollutants such as NO2, O3,
or PAN and more recent work on CO + O3 by DeLucia et al. (1983) failed to show any
interaction from combined exposure.
25 In animal studies, no interaction was observed following combined exposure of CO and
pollutants such as HCN, NO2, SO2, or PbClBr (Hugod, 1979; Busey, 1972; Murray et al.,
1978). However, an additive effect was observed following combined exposure of high levels
of CO + NO (Groll-Knapp et al., 1988), and a synergistic effect was observed after
combined exposure to CO and O3 (Murphy, 1964).
30 Toxicological interactions of combustion products, primarily CO, CO2, and HCN, from
indoor and outdoor fires, have shown a synergistic effect following CO + CO2 exposure
(Rodkey and Collison, 1979; Levin et al., 1987a) and an additive effect with CO + HCN
March 14, 1990 1-19 DRAFT - DO NOT QUOTE OR CITE
-------
(Levin et al., 19875). Additional studies are needed, however, to evaluate the effects of CO
under conditions of hypoxic hypoxia.
Finally, laboratory animal studies (Yang et al., 1988; Fechter et al., 1988; Young
et al., 1987) suggest that the combination of environmental factors such as heat stress and
5 noise may be important determinants of health effects occurring in combination with exposure
to CO. Of the effects described, the one potentially most relevant to typical human exposures
is a greater decrement in exercise performance seen when heat stress is combined with 50
ppm CO (Drinkwater et al., 1974; Raven et al., 1974a,b; Gliner et, al., 1975).
10 Environmental Tobacco Smoke
Although tobacco smoke is another source of CO for smokers as well as nonsmokers, it
is also a source of other chemicals with which environmental CO levels could interact.
Available data strongly suggest that acute and chronic CO exposure attributed to tobacco
smoke can affect the cardiopulmonary system, but the potential interaction of CO with other
15 products of tobacco smoke confounds the results. In addition, it is not clear if incremental
increases in COHb caused by environmental exposure would actually be additive to
chronically elevated COHb levels due to tobacco smoke, because some physiological
adaptation may take place. There is, therefore, a need for further research to describe these
relationships better.
20
Evaluation of Subpopulations Potentially At Risk to Carbon Monoxide
Exposure
Most of the information on the human health effects of carbon monoxide discussed in
this document has concentrated on two carefully defined population groups - young healthy,
25 predominantly male adults and patients with diagnosed coronary artery disease. On the basis
of the known effects described, patients with reproducible exercise-induced angina appear to
be best established as a sensitive group within the general population that is at increased risk
for experiencing health effects (i.e., decreased exercise duration due to exacerbation of
cardiovascular symptoms) of concern at ambient or near-ambient CO-exposure concentrations
30 that result in COHb levels of <5%. A smaller sensitive group of healthy individuals
experience decreased exercise duration at similar levels of CO exposure, but only during
short-term maximal exercise. Decrements in exercise duration in the healthy population,
March 14, 1990 1-20 DRAFT - DO NOT QUOTE OR CITE
-------
therefore, would be mainly of concern to competing athletes rather than for nonathletic people
carrying out the common activities of daily life.
It is known, however, from both theoretical work and from experimental research in
laboratory animals that certain other groups in the population are at potential risk to exposure
5 from CO. Another purpose of this document is to explore the potential effects of CO in
population groups that have not been studied adequately, but which could be expected to be
susceptible to CO because of underlying physiological status either due to gender differences,
aging, preexisting disease, or because of the use of medications or alterations in their
environment. These probable risk groups include (1) fetuses and young infants; (2) pregnant
10 women; (3) the elderly, especially those with compromised cardiopulmonary or
cerebrovascular functions; (4) individuals with obstructed coronary arteries, but not yet
manifesting overt symptomatology of coronary artery disease; (5) individuals with congestive
heart failure; (6) individuals with peripheral vascular or cerebrovascular disease;
(7) individuals with hematological diseases (e.g., anemia) that affect oxygen-carrying capacity
15 or transport in the blood; (8) individuals with genetically unusual forms of hemoglobin
associated with reduced oxygen-carrying capacity; (9) individuals with chronic obstructive
lung diseases; (10) individuals using medicinal or recreational drugs having CNS depressant
properties; (11) individuals exposed to other pollutants (e.g., methylene chloride) that
increase endogenous formation of CO; and (12) individuals who have not been adapted to
20 high altitude and are exposed to a combination of high altitude and CO. Unfortunately, little
empirical evidence currently is available by which to specify health effects associated with
ambient or near-ambient CO exposures in these probable risk groups.
25 References
Akland, G. G.; Hartwell, T. D.; Johnson, T. R.; Whitmore, R. W. (1985) Measuring human exposure to carbon
monoxide in Washington, D.C., and Denver, Colorado, during the winter of 1982-1983. Environ. Sci.
Technol. 19: 911-918.
30
Allred, E. N.; Bleecker, E. R.; Chaitman, B. R.; Dahms, T. E.; Gottlieb, S. O.; Hackney, J. D.; Pagano, M.;
Selvester, R. H.; Walden, S. M.; Warren, J. (1989a) Short-term effects of carbon monoxide exposure on
the exercise performance of subjects with coronary artery disease. N. Engl. J. Med. 321: 1426-1432.
35
March 14, 1990 1-21 DRAFT - DO NOT QUOTE OR CITE
-------
Allred, E. N.; Bleecker, E. R.; Chairman, B. R.; Dahms, T. E.; Gottlieb, S. O.; Hackney, J. D.; Hayes, D.;
Pagano, M. Selvester, R. H.; Walden, S. M.; Warren, J. (1989b) Acute effects of carbon monoxide
exposure on individuals with coronary artery disease. Cambridge, MA: Health Effects Institute; research
report no. 25.
5
Benignus, V. A.; Otto, D. A.; Prah, J. D.; Benignus, G. (1977) Lack of effects of carbon monoxide on human
vigilance. Percept. Mot. Skills 45: 1007-1014.
Benignus, V. A.; Muller, K. E.; Barton, C. N.; Prah, J. D. (1987) Effect of low level carbon monoxide on
10 compensatory tracking and event monitoring. Neurotoxicol. Teratol. 9: 227-234.
Benignus, V. A.; Muller, K. E.; Pieper, K. S.; Prah, J. D. (1989) Compensatory tracking in humans with
elevated carboxyhemoglobin. Neurotoxicol. Teratol.: in press.
15 Busey, W. M. (1972) Chronic exposure of albino rats to certain airborne pollutants [unpublished material].
Vienna, VA: Hazleton Laboratories, Inc.
Campbell, J. A. (1934) Growth, fertility, etc. in animals during attempted acclimatization to carbon monoxide.
Exp. Physiol. 24: 271-281.
20
Chapman, R. W.; Santiago, T. V.; Edelman, N. H. (1980) Brain hypoxia and control of breathing:
neuromechanical control. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 49: 497-505.
Chen, S.; Weller, M. A.; Penney, D. G. (1982) A study of free lung cells from young rats chronically exposed
25 to carbon monoxide from birth. Scanning Electron Microsc. 2: 859-867.
Coburn, R. F.; Forster, R. E.; Kane, P. B. (1965) Considerations of the physiological variables that determine
the blood carboxyhemoglobin concentration in man. J. Clin. Invest. 44: 1899-1910.
30 Cobum, R. F.; Wallace, H. W.; Abboud, R. (1971) Redistribution of body carbon monoxide after hemorrhage.
Am. J. Physiol. 220: 868-873.
Coburn, R. F.; Ploegmakers, F.; Gondrie, P.; Abboud, R. (1973) Myocardial myoglobin oxygen tension. Am. J.
Physiol. 224: 870-876.
35
Cooper, K. R.; Alberti, R. R. (1984) Effect of kerosene heater emissions on indoor air quality and pulmonary
function. Am. Rev. Respir. Dis. 129: 629-631.
DeLucia, A. J.; Whitaker, J. H.; Bryant, L. R. (1983) Effects of combined exposure to ozone and carbon
40 monoxide (CO) in humans. In: Lee, S. D.; Mustafa, M. G.; Mehlman, M. A., eds. International
symposium on the biomedical effects of ozone and related photochemical oxidants; March; Pinehurst,
NC. Princeton, NJ: Princeton Scientific Publishers, Inc.; pp. 145-159. (Advances in modern
environmental toxicology: v. 5).
45 Dennis, R. C.; Valeri, C. R. (1980) Measuring percent oxygen saturation of hemoglobin, percent
carboxyhemoglobin and methemoglobin, and concentrations of total hemoglobin and oxygen in blood of
man, dog, and baboon. Clin. Chem. (Winston-Salem, NC) 26: 1304-1308.
Doblar, D. D.; Santiago, T. V.; Edelman, N. H. (1977) Correlation between ventilatory and cerebrovascular
50 responses to inhalation of CO. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 43: 455-462.
Drinkwater, B. L.; Raven, P. B.; Horvath, S. M.; Gliner, J. A.; Ruhling, R. O.; Bolduan, N. W.; Taguchi, S.
(1974) Air pollution, exercise, and heat stress. Arch. Environ. Health 28: 177-181.
March 14, 1990 1-22 DRAFT - DO NOT QUOTE OR CITE
-------
Evans, R. G.; Webb, K.; Homan, S.; Ayres, S. M. (1988) Cross-sectional and longitudinal changes in
pulmonary function associated with automobile pollution among bridge and tunnel officers. Am. J. Ind.
Med. 14: 25-36.
5 Fechter, L. D.; Young, J. S.; Carlisle, L. (1988) Potentiation of noise induced threshold shifts and hair cell loss
by carbon monoxide. Hear. Res. 34: 39-47.
Fein, A.; Grossman, R. F.; Jones, J. G.; Hoeffel, J.; McKay, D. (1980) Carbon monoxide effect on alveolar
epithelial permeability. Chest 78: 726-731.
10
Fenn, W. O. (1970) The burning of CO in the tissues. In: Cobum, R. F., ed. Biological effects of carbon
monoxide. Ann. N. Y. Acad. Sci. 174: 64-71.
Fisher, A. B.; Hyde, R. W.; Baue, A. E.; Reif, J. S.; Kelly, D. F. (1969) Effect of carbon monoxide on
15 function and structure of the lung. J. Appl. Physiol. 26: 4-12.
Flachsbart, P. G.; Mack, G. A.; Howes, J. E.; Rodes, C. E. (1987) Carbon monoxide exposures of Washington
commuters. J. Air Pollut. Control Assoc. 37: 135-142.
20 Gautier, H.; Bonora, M. (1983) Ventilatory response of intact cats to carbon monoxide hypoxia. J. Appl.
Physiol.: Respir. Environ. Exercise Physiol. 55: 1064-1071.
Gliner, J. A.; Raven, P. B.; Horvath, S. M.; Drinkwater, B. L.; Sutton, J. C. (1975) Man's physiologic
response to long-term work during thermal and pollutant stress. J. Appl. Physiol. 39: 628-632.
25
Groll-Knapp, E.; Haider, M.; Kienzl, K.; Handler, A.; Trimmel, M. (1988) Changes in discrimination learning
and brain activity (ERP's) due to combined exposure to NO and CO in rats. Toxicology 49: 441-447.
Hackney, J. D.; Linn, W. S.; Mohler, J. G.; Pedersen, E. E.; Breisacher, P.; Russo, A. (1975a) Experimental
30 studies on human health effects of air pollutants: II. four-hour exposure to ozone alone and in
combination with other pollutant gases. Arch. Environ. Health 30: 379-384.
Hackney, J. D.; Linn, W. S.; Law, D. C.; Karuza, S. K.; Greenberg, H.; Buckley, R. D.; Pedersen, E. E.
(1975b) Experimental studies on human health effects of air pollutants: III. two-hour exposure to ozone
35 alone and in combination with other pollutant gases. Arch. Environ. Health 30: 385-390.
Hagberg, M.; Kolmodin-Hedman, B.; Lindahl, R.; Nilsson, C.-A.; Norstrom, A. (1985) Irritative complaints,
carboxyhemoglobin increase and minor ventilatory function changes due to exposure to chain-saw
exhaust. Eur. J. Respir. Dis. 66: 240-247.
40
Halebian, P. H.; Barie, P. S.; Robinson, N.; Shires, G. T. (1984a) Effects of carbon monoxide on pulmonary
fluid accumulation. Curr. Surg. 41: 369-371.
Halebian, P.; Sicilia, C.; Hariri, R.; Inamdar, R.; Shires, G. T. (1984b) A safe and reproducible model of
45 carbon monoxide poisoning. Ann. N. Y. Acad. Sci. 435: 425-428.
Hartwell, T. D.; Clayton, C. A.; Ritchie, R. M.; Whitmore, R. W.; Zelon, H. S.; Jones, S. M.; Whitehurst, D.
A. (1984) Study of carbon monoxide exposure of residents of Washington, DC and Denver, Colorado.
Research Triangle Park, NC: U. S. Environmental Protection Agency, Environmental Monitoring
50 Systems Laboratory; EPA report no. 600/4-84-031. Available from: NTIS, Springfield, VA;
PB84-183516.
Hirsch, G. L.; Sue, D. Y.; Wassennan, K.; Robinson, T. E.; Hansen, J. E. (1985) Immediate effects of cigarette
smoking on cardiorespiratory responses to exercise. J. Appl. Physiol, 58: 1975-1981.
March 14, 1990 1-23 DRAFT - DO NOT QUOTE OR CITE
-------
Horvath, S. M.; Raven, P. B.; Dahms, T. E.; Gray, D. J. (1975) Maximal aerobic capacity at different levels of
carboxyhemoglobin. J. Appl. Physiol. 38: 300-303.
Hugod, C. (1979) Effect of exposure to 0.5 ppm hydrogen cyanide singly or combined with 200 ppm carbon
S monoxide and/or 5 ppm nitric oxide on coronary arteries, aorta, pulmonary artery, and lungs in the
rabbit. Int. Arch. Occup. Environ. Health 44: 13-23.
Hugod, C. (1980) The effect of carbon monoxide exposure on morphology of lungs and pulmonary arteries in
rabbits: a light- and electron-microscopic study. Arch. Toxicol. 43: 273-281.
10
Johnson, T. (1984) A study of personal exposure to carbon monoxide in Denver, Colorado. Research Triangle
Park, NC: U. S. Environmental Protection Agency, Environmental Monitoring Systems Laboratory; EPA
report no. EPA-600/4-84-014. Available from: NTIS, Springfield, VA; PB84-146125.
15 Katsumata, Y.; Aoki, M.; Oya, M.; Yada, S.; Suzuki, O. (1980) Liver damage in rats during acute carbon
monoxide poisoning. Forensic Sci. Int. 16: 119-123.
Klausen, K.; Andersen, C.; Nandrup, S. (1983) Acute effects of cigarette smoking and inhalation of carbon
monoxide during maximal exercise. Eur. J. Appl. Physiol. Occup. Physiol. 51: 371-379.
20
Klein, J. P.; Forster, H. V.; Stewart, R. D.; Wu, A. (1980) Hemoglobin affinity for oxygen during short-term
exhaustive exercise. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 48: 236-242.
Kleinman, M. T.; Davidson, D. M.; Vandagriff, R. B.; Caiozzo, V. J.; Whittenberger, J. L. (1989) Effects of
25 short-term exposure to carbon monoxide in subjects with coronary artery disease. Arch. Environ. Health
44: 361-369.
Koob, G. F.; Annau, Z.; Rubin, R. J.; Montgomery, M. R. (1974) Effect of hypoxic hypoxia and carbon
monoxide on food intake, water intake, and body weight in two strains of rats. Life Sci. 14: 1511-1520.
30
Koontz, M. D.; Nagda, N. L. (1987) Survey effectors affecting NO2 concentrations. In: Seifert, B.; Esdorn, H.;
Fischer, M.; Rueden, H.; Wegner, J., eds. Indoor air '87: proceedings of the 4th international conference
on indoor air quality and climate, v. 1, volatile organic compounds, combustion gases, particles and
fibres, microbiological agents; August; Berlin, Federal Republic of Germany. Berlin, Federal Republic of
35 Germany: Institute for Water, Soil, and Air Hygiene; pp. 430-434.
Koontz, M. D.; Nagda, N. L. (1988) A topical report on a field monitoring study of homes with unvented gas
space heaters: v. III. methodology and results. Gas Research Institute final report; contract no.
5083-251-0941.
40
Kuska, J.; Kokot, F.; Wnuk, R. (1980) Acute renal failure after exposure to carbon monoxide. Mater. Med. Pol.
(Engl. Ed.) 12: 236-238.
Kustov, V. V.; Belkin, V. I.; Abidin, B. L; Ostapenko, O. F.; Malkuta, A. N.; Poddubnaja, L. T. (1972)
45 Aspects of chronic carbon monoxide poisoning in young animals. Gig. Tr. Prof. Zabol. 5: 50-52.
Lebowitz, M. D.; Collins, L.; Holberg, C. J. (1987) Time series analyses of respiratory responses to indoor and
outdoor environmental phenomena. Environ. Res. 43: 332-341.
50 Levin, B. C.; Paabo, M.; Gurman, J. L.; Harris, S. E.; Braun, E. (1987a) Toxicological interactions between
carbon monoxide and carbon dioxide. Toxicology 47: 135-164.
March 14, 1990 1-24 DRAFT - DO NOT QUOTE OR CITE
-------
Levin, B. C.; Paabo, M.; Gurman, J. L.; Harris, S. E. (1987b) Effects of exposure to single or multiple
combinations of the predominant toxic gases and low oxygen atmospheres produced in fires. Fundam.
Appl. Toxicol. 9: 236-250.
5 Luomanmaki, K.; Coburn, R. F. (1969) Effects of metabolism and distribution of carbon monoxide on blood and
body stores. Am. J. Physiol. 217: 354-363.
Lutz, L. J. (1983) Health effects of air pollution measured by outpatient visits. J. Fam. Pract. 16: 307-313.
10 Martynjuk, V. C.; Dacenko, I. I. (1973) Aktivnost' transaminaz v usloviyakh khronicheskoi intoksikatsii okis'yu
ugleroda [Aminotransferase activity in chronic carbon monoxide poisoning]. Gig. Naselennykh. Mest. 12:
53-56.
McDonagh, P. F.; Reynolds, J. M.; McGrath, J. J. (1986) Chronic altitude plus carbon monoxide exposure
15 causes left ventricular hypertrophy but an attenuation of coronary capillarity. Presented at: 70th annual
meeting of the Federation of American Societies for Experimental Biology; April; St. Louis, MO. Fed.
Proc. Fed. Am. Soc. Exp. Biol. 45: 883.
Montgomery, M. R.; Rubin, R. J. (1971) The effect of carbon monoxide inhalation on in vivo drug metabolism
20 in the rat. J. Pharmacol. Exp. Ther. 179: 465-473.
Mordelet-Dambrine, M.; Stupfel, M. (1979) Comparison in guinea-pigs and in rats of the effects of vagotomy
and of atropine on respiratory resistance modifications induced by an acute carbon monoxide or nitrogen
hypoxia. Comp. Biochem. Physiol. A 63: 555-559.
25
Mordelet-Dambrine, M.; Stupfel, M.; Duriez, M. (1978) Comparison of tracheal pressure and circulatory
modifications induced in guinea pigs and in rats by carbon monoxide inhalation. Comp. Biochem.
Physiol. A 59: 65-68.
30 Murphy, S. D. (1964) A review of effects on animals of exposure to auto exhaust and some of its components. J.
Air Pollut. Control Assoc. 14: 303-308.
Murray, F. J.; Schwetz, B. A.; Crawford, A. A.; Henck, J. W.; Staples, R. E. (1978) Teratogenic potential of
sulfur dioxide and carbon monoxide in mice and rabbits. In: Mahlum, D. D.; Sikov, M. R.; Hackett, P.
35 L.; Andrew, F. D., eds. Developmental toxicology of energy-related pollutants: proceedings of the
seventeenth annual Hanford biology symposium; October 1977; Richland, WA. Oak Ridge, TN: U. S.
Department of Energy, Technical Information Center; pp. 469-478. Available from: NTIS, Springfield,
VA; CONF-771017.
40 Mussehnan, N. P.; Groff, W. A.; Yevich, P. P.; Wilinski, F. T.; Weeks, M. H.; Oberst, F. W. (1959)
Continuous exposure of laboratory animals to low concentration of carbon monoxide. Aerosp. Med. 30:
524-529.
Naeije, R.; Peretz, A.; Cornil, A. (1980) Acute pulmonary edema following carbon monoxide poisoning.
45 Intensive Care Med. 6: 189-191.
O'Donnell, R. D.; Mikulka, P.; Heinig, P.; Theodore, J. (1971) Low level carbon monoxide exposure and
human psychomotor performance. Toxicol. Appl. Pharmacol. 18: 593-602.
50 Pankow, D.; Ponsold, W. (1972) Leucine aminopeptidase activity in plasma of normal and carbon monoxide
poisoned rats. Arch. Toxicol. 29: 279-285.
March 14, 1990 1-25 DRAFT - DO NOT QUOTE OR CITE
-------
Pankow, D.; Ponsold, W. (1974) Kombinationswirkungen von Kohlenmonoxid mit anderen biologischaktiven
Schadfaktoren auf den Organismus [The combined effects of carbon monoxide and other biologically
active detrimental factors on the organism]. Z. Gesamte Hyg. Ihre Grenzgeb. 20: 561-571.
5 Pankow, D.; Ponsold, W.; Fritz, H. (1974) Combined effects of carbon monoxide and ethanol on the activities of
leucine aminopeptidase and glutamic-pyruvic transaminase in the plasma of rats. Arch. Toxicol. 32:
331-340.
Parving, H.-H. (1972) The effect of hypoxia and carbon monoxide exposure on plasma volume and capillary
10 permeability to albumin. Scand. J. Clin. Lab. Invest. 30: 49-56.
Penney, D. G.; Davidson, S. B.; Gargulinski, R. B.; Caldwell-Ayre, T. M. (1988) Heart and lung hypertrophy,
changes in blood volume, hematocrit and plasma renin activity in rats chronically exposed in increasing
carbon monoxide concentrations. JAT J. Appl. Toxicol. 8: 171-178.
15
Penney, D. G.; Gargulinski, R. B.; Hawkins, B. J.; Santini, R.; Caldwell-Ayre, T. M.; Davidson, S. B. (1988)
The effects of carbon monoxide on persistent changes in young rat heart: cardiomegaly, tachycardia and
altered DNA content. JAT J. Appl. Toxicol. 8: 275-283.
20 Putz, V. R.; Johnson, B. L.; Setzer, J. V. (1976) Effects of CO on vigilance performance: effects of low level
carbon monoxide on divided attention, pitch discrimination, and the auditory evoked potential.
Cincinnati, OH: U. S. Department of Health, Education, and Welfare, National Institute for Occupational
Safety and Health. Available from: NTIS, Springfield, VA; PB-274219.
25 Putz, V. R.; Johnson, B. L.; Setzer, J. V. (1979) A comparative study of the effects of carbon monoxide and
methylene chloride on human performance. J. Environ. Pathol. Toxicol. 2: 97-112.
Radford, E. P.; Drizd, T. A. (1982) Blood carbon monoxide levels in persons 3-74 years of age: United States,
1976-80. Hyattsville, Md: U. S. Department of Health and Human Services, National Center for Health
30 Statistics; DHHS publication no. (PHS) 82-1250. (Advance data from vital and health statistics, no. 76).
Raven, P. B.; Drinkwater, B. L.; Horvath, S. M.; Ruhling, R. O.; Gliner, J. A.; Sutton, J. C.; Bolduan, N. W.
(1974a) Age, smoking habits, heat stress, and their interactive effects with carbon monoxide and
peroxyacetylnitrate on man's aerobic power. Int. J. Biometeorol. 18: 222-232.
35
Raven, P. B.; Drinkwater, B. L.; Ruhling, R. O.; Bolduan, N.; Taguchi, S.; Gliner, J.; Horvath, S. M. (1974b)
Effect of carbon monoxide and peroxyacetyl nitrate on man's maximal aerobic capacity. J. Appl. Physiol.
36: 288-293.
40 Robertson, G.; Lebowitz, M. D. (1984) Analysis of relationships between symptoms and environmental factors
over time. Environ. Res. 33: 130-143.
Robinson, N. B.; Barie, P. S.; Halebian, P. H.; Shires, G. T. (1985) Distribution of ventilation and perfusion
following acute carbon monoxide poisoning. In: 41st annual forum on fundamental surgical problems held
45 at the 71st annual clinical congress of the American College of Surgeons; October; Chicago, IL. Surg.
Forum 36: 115-118.
Rodkey, F. L.; Collison, H. A. (1979) Effects of oxygen and carbon dioxide on carbon monoxide toxicity. J.
Combust. Toxicol. 6: 208-212.
50
Rogers, W. R.; Bass, R. L., Ill; Johnson, D. E.; Kruski, A. W.; McMahan, C. A.; Montiel, M. M.; Mott, G.
E.; Wilbur, R. L.; McGill, H. C., Jr. (1980) Atherosclerosis-related responses to cigarette smoking in
the baboon. Circulation 61: 1188-1193.
March 14, 1990 1-26 DRAFT - DO NOT QUOTE OR CITE
-------
Rogers, W. R.; Carey, K. D.; McMahan, C. A.; Montiel, M, M.; Mott, G. E.; Wigodsky, H. S.; McGill, H.
C., Jr. (1988) Cigarette smoking, dietary hyperlipidemia, and experimental atherosclerosis in the baboon.
Exp. Mol. Pathol. 48: 135-151.
5 Roth, R. A., Jr.; Rubin, R. J. (1976a) Role of blood flow in carbon monoxide- and hypoxic hypoxia-induced
alterations in hexobarbital metabolism in rats. Drug Metab. Dispos. 4: 460-467.
Roth, R. A., Jr.; Rubin, R. J. (1976b) Comparison of the effect of carbon monoxide and of hypoxic hypoxia. II.
Hexobarbital metabolism in the isolated, perfused rat liver. J. Pharmacol. Exp. Ther. 199: 61-66.
10
Santiago, T. V.; Edelman, N. H. (1976) Mechanism of the ventilatory response to carbon monoxide. J. Clin.
Invest. 57: 977-986.
Sheppard, D.; Distefano, S.; Morse, L.; Becker, C. (1986) Acute effects of routine firefighting on lung function.
15 Am. J. Ind. Med. 9: 333-340.
Sheps, D. S.; Adams, K. F., Jr.; Bromberg, P. A.; Goldstein, G. M.; O'Neil, J. J.; Horstman, D.; Koch, G.
(1987) Lack of effect of low levels of carboxyhemoglobin on cardiovascular function in patients with
ischemic heart disease. Arch. Environ. Health 42: 108-116.
20
Sheps, D. S.; Herbst, M. C.; Hinderliter, A. L.; Adams, K. F.; Ekelund, L. G.; O'Neil, J. J.; Goldstein, G.
M.; Bromberg, P. A.; Herdt, J.; Ballenger, M.; Davis, S. M.; Koch, G. (1989) Effects of 4% and 6%
carboxyhemoglobin on arrhythmia production in patients with coronary artery disease. Submitted for
publication.
25
Snella, M.-C.; Rylander, R. (1979) Alteration in local and systemic immune capacity after exposure to bursts of
CO. Environ. Res. 20: 74-79.
Stern, F. B.; Halperin, W. E.; Hornung, R. W.; Ringenburg, V. L.; McCammon, C. S. (1988) Heart disease
30 mortality among bridge and tunnel officers exposed to carbon monoxide. Am. J. Epidemiol. 128:
1276-1288.
Stewart, R. D.; Newton, P. E.; Kaufman, J.; Forster, H. V.; Klein, J. P.; Keelen, M. H., Jr.; Stewart, D. J.;
Wu, A.; Hake, C. L. (1978) The effect of a rapid 4% carboxyhemoglobin saturation increase on maximal
35 treadmill exercise. New York, NY: Coordinating Research Council, Inc.; report no.
CRC-APRAC-CAPM-22-75. Available from: NTIS, Springfield, VA; PB-296627.
Stupfel, M.; Bouley, G. (1970) Physiological and biochemical effects on rats and mice exposed to small
concentrations of carbon monoxide for long periods. Ann. N. Y. Acad. Sci. 174: 342-368.
40
Swiecicki, W. (1973) Wplyw wibracji i treningu fizycznego na przemiane weglowodanowa u szczurow zatrutych
tlenkiem wegla [The effect of vibration and physical training on carbohydrate metabolism in rats
intoxicated with carbon monoxide]. Med. Pr. 34: 399-405.
45 Theodore, J.; O'Donnell, R. D.; Back, K. C. (1971) Toxicological evaluation of carbon monoxide in humans and
other mammalian species. JOM J. Occup. Med. 13: 242-255.
U. S. Environmental Protection Agency. (1979) Air quality criteria for carbon monoxide. Research Triangle
Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and Assessment
50 Office; EPA report no. EPA-600/8-79-022. Available from: NTIS, Springfield, VA; PB81-244840.
March 14, 1990 1-27 DRAFT - DO NOT QUOTE OR CITE
-------
U. S. Environmental Protection Agency. (1984) Revised evaluation of health effects associated with carbon
monoxide exposure: an addendum to the 1979 EPA air quality criteria document for carbon monoxide.
Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria
and Assessment Office; EPA report no. EPA-600/9-83-033F. Available from: NTIS, Springfield, VA;
5 PB85-103471.
U. S. Environmental Protection Agency. (1990) National air quality and emissions trends report, 1988. Research
Triangle Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA-450/4-90-002.
10 Wallace, L. A.; Ziegenfus, R. C. (1985) Comparison of carboxyhemoglobin concentrations in adult nonsmokers
with ambient carbon monoxide levels. J. Air Pollut. Control Assoc. 35: 944-949.
Weiser, P. C.; Morrill, C. G.; Dickey, D. W.; Kurt, T. L.; Cropp, G. J. A. (1978) Effects of low-level carbon
monoxide exposure on the adaptation of healthy young men to aerobic work at an altitude of 1,610
15 meters. In: Folinsbee, L. J.; Wagner, J. A.; Borgia, J. F.; Drinkwater, B. L.; Gliner, J. A.; Bedi, J. F.,
eds. Environmental stress: individual human adaptations. New York, NY: Academic Press, Inc.; pp.
101-110.
Weissbecker, L.; Carpenter, R. D.; Luchsinger, P. C.; Osdene, T. S. (1969) In vitro alveolar macrophage
20 viability: effect of gases. Arch. Environ. Health 18: 756-759.
Whitmore, R. W.; Jones, S. M.; Rosenzweig, M. S. (1984) Final sampling report for the study of personal CO
(carbon monoxide) exposure. Research Triangle park, NC: U. S. Environmnetal Protection Agency,
Environmental Monitoring Systems Laboratory; EPA report no. EPA-600/4-84-034. Available from:
25 NTIS, Springfield, VA; PB84-181957.
Yang, L.; Zhang, W.; He, H.; Zhang, G. (1988) Experimental studies on combined effects of high temperature
and carbon monoxide. J. Tongji Med. Univ. 8: 60-65.
30 Young, J. S.; Upchurch, M. B.; Kaufman, M. J.; Fechter, L. D. (1987) Carbon monoxide exposure potentiates
high-frequency auditory threshold shifts induced by noise. Hear. Res. 26: 37-43.
Zebro, T.; Wright, E. A.; Littleton, R. J.; Prentice, A. I. D. (1983) Bone changes in mice after prolonged
continuous exposure to a high concentration of carbon monoxide. Exp. Pathol. 24: 51-67.
35
March 14, 1990 1-28 DRAFT - DO NOT QUOTE OR CITE
-------
Most of the scientific information selected for review and comment in this document
comes from the more recent literature published since completion of the previous criteria
document (U.S. Environmental Protection Agency, 1979). Some of the these newer studies
were reviewed briefly in the addendum to that document (U.S. Environmental Protection
5 Agency, 1984a). Emphasis has been placed on studies conducted at or near CO
concentrations found in ambient air. Other studies, however, were included if they contained
unique data, such as the documentation of a previously unreported effect or a mechanism of
an effect; or if they were multiple-concentration studies designed to provide exposure-
response relationships relevant to total human exposure to CO. Studies that were presented in
10 the previous criteria document and whose data are still considered relevant are summarized in
tables or reviewed briefly in the text. Older studies were considered for discussion in the
document if they were (1) judged to be significant because of their usefulness in deriving the
current NAAQS, (2) open to reinterpretation because of newer data, or (3) potentially useful
in deriving revised standards for CO. Generally, only published information that has
15 undergone scientific peer review is included in this criteria document. Some newer studies
not published in the open literature but meeting high standards of scientific reporting also are
included.
20 2.2 LEGISLATIVE HISTORY OF NAAQS
Two sections of the Clean Air Act (CAA) govern the establishment, review, and
revision of NAAQS. Section 108 (U.S. Code, 1982) directs the Administrator of the U.S.
EPA to identify pollutants that reasonably may be anticipated to endanger public health or
welfare and to issue air quality criteria for them. These air quality criteria are to reflect the
25 latest scientific information useful in indicating the kind and extent of all identifiable effects
on public health or welfare that may be expected from the presence of the pollutant in
ambient air.
Section 109(a) of the CAA (U.S. Code, 1982) directs the Administrator of EPA to
propose and promulgate primary and secondary NAAQS for pollutants identified under
30 Section 108. Section 109(b)(l) defines a primary standard as one the attainment and
maintenance of which in the judgment of the Administrator, based on the criteria and
March 12, 1990 2-3 DRAFT - DO NOT QUOTE OR CITE
-------
allowing for an adequate margin of safety, is requisite to protect the public health. The
secondary standard, as defined in Section 109(b)(2), must specify a level of air quality the
attainment and maintenance of which in the judgment of the Administrator, based on the
criteria, is requisite to protect the public welfare from any known or anticipated adverse
5 effects associated with the presence of the pollutant in ambient air. Section 109(d) of the
CAA (U.S. Code, 1982) requires periodic review and, if appropriate, revision of existing
criteria and standards. If, in the Administrator's judgment, the Agency's review and revision
of criteria make appropriate the proposal of new or revised standards, such standards are to be
revised and promulgated in accordance with Section 109(b). Alternately, the Administrator
10 may find that revision of the standards is inappropriate and may conclude the review by
leaving the existing standards unchanged.
In keeping with the requirements of the CAA, the Environmental Criteria and
Assessment Office of EPA's Office of Health and Environmental Assessment has started to
review and revise once again the criteria for CO. New data on the health and air quality
15 aspects of CO exposure have become available since completion of the previous Air Quality
Criteria Document (U.S. Environmental Protection Agency, 1979) and an addendum to that
document (U.S. Environmental Protection Agency, 1984a).
20 2.3 REGULATORY BACKGROUND FOR CARBON MONOXIDE
NAAQS*
On April 30, 1971, EPA promulgated identical primary and secondary NAAQS for CO
at levels of 9 ppm for an 8-h average and 35 ppm for a 1-h average, not to be exceeded more
than once per year. The scientific basis for the primary standard, as described in the first
25 criteria document (National Air Pollution Control Administration, 1970), was a study
suggesting that low levels of CO exposure resulting in carboxyhemoglobin (COHb)
concentrations of 2 to 3% were associated with neurobehavioral effects in exposed subjects
(Beard and Wertheim, 1967).
"This text is excerpted and adapted from "Review of the National Ambient Air Quality Standards for Carbon
30 Monoxide; Final Rule" (Federal Register, 1985).
March 12, 1990 2-4 DRAFT - DO NOT QUOTE OR CITE
-------
In accordance with Sections 108 and 109 of the CAA, EPA has reviewed and revised
the criteria upon which the existing NAAQS for CO (Table 2-1) are based. On August 18,
1980, EPA proposed certain changes in the standards (Federal Register, 1980) based on
scientific evidence reported in the revised criteria document for CO (U.S. Environmental
5 Protection Agency, 1979). Such evidence indicated that the Beard and Wertheim (1967)
study was no longer considered to be a sound scientific basis for the standard. Additional
medical evidence accumulated since 1970, however, indicated that aggravation of angina
pectoris and other cardiovascular diseases would occur at COHb levels as low as 2.7 to 2.9%.
The proposed changes included (1) retaining the 8-h primary standard level of 9 ppm,
10 (2) revising the 1-h primary standard level from 35 ppm to 25 ppm, (3) revoking the existing
secondary CO standards (because no adverse welfare effects have been reported at or near
ambient CO levels), (4) changing the form of the primary standards from deterministic to
statistical, and (5) adopting a daily interpretation for exceedances of the primary standards, so
that exceedances would be determined on the basis of the number of days on which the 8- or
15 1-h average concentrations are above the standard levels.
The 1980 proposal was based in part on health studies conducted by Dr. Wilbert
Aronow. In March of 1983 EPA learned that the Food and Drug Administration (FDA) had
raised serious questions regarding the technical adequacy of several studies conducted by Dr.
Aronow on experimental drugs, leading FDA to reject use of the Aronow drug study data.
20 Therefore, EPA convened an expert committee to examine the Aronow CO studies before any
final decisions were made on the NAAQS for CO. The committee concluded that EPA
should not rely on Dr. Aronow's data due to concerns regarding the research, which
substantially limited the validity and usefulness of the results.
An addendum to the 1979 criteria document for CO (U.S. Environmental Protection
25 Agency, 1984a) reevaluated the scientific data concerning health effects associated with
exposure to CO at or near ambient exposure levels in light of the committee recommendations
and taking into account new findings reported beyond those previously reviewed. (These data
are summarized in the following section.) On September 13, 1985, EPA issued a final notice
(Federal Register, 1985) announcing retention of the existing primary NAAQS for CO and
30 rescinding the secondary NAAQS for CO.
March 12, 1990 2-5 DRAFT - DO NOT QUOTE OR CITE
-------
TABLE 2-1. NATIONAL AMBIENT AIR QUALITY STANDARDS
FOR CARBON MONOXIDE
5 Date of Promulgation Primary NAAQS Averaging Time
September 13, 1985 9 ppm' (10 mg/m3) ITh1
35 ppm' (40 mg/m3) l-hb
10
* 1 ppm = 1.145 mg/m3, 1 mg/m3 = 0.873 ppm @ 25°C, 760 mm Hg.
b Not to be exceeded more than once per year.
See glossary of terms and symbols for abbreviations and acronyms.
15
2.4 SCIENTIFIC BACKGROUND FOR THE CURRENT CARBON
20 MONOXIDE NAAQS
The following is a summary of the scientific basis for the current CO NAAQS. These
key points were derived from a revised evaluation of the health effects of CO that was
released as an addendum (U.S. Environmental Protection Agency, 1984a) to the previous air
quality criteria document for CO (U.S. Environmental Protection Agency, 1979).
25
2.4.1 Mechanisms of Action
The binding of CO to hemoglobin, producing COHb and decreasing the oxygen-
carrying capacity of blood, appears to be the principal mechanism of action underlying the
induction of toxic effects of low-level CO exposures. The precise mechanisms by which toxic
30 effects are induced via COHb formation are not understood fully, but likely include the
induction of a hypoxic state in many tissues of diverse organ systems. Alternative or
secondary mechanisms of CO-induced toxicity (besides COHb) have been hypothesized, but
none have been demonstrated to operate at relatively low (near-ambient) CO-exposure levels.
Blood COHb levels, then, currently are accepted as representing a useful physiological marker
35 by which to estimate internal CO burdens due to the combined contribution of
(1) endogenously derived CO and (2) exogenously derived CO resulting from exposure to
external sources of CO. COHb levels likely to result from particular patterns (concentrations,
durations, etc.) of external CO exposure can be estimated reasonably well from equations
developed by Coburn et al. (1965) as demonstrated in Figure 2-1.
March 12, 1990 2-6 DRAFT - DO NOT QUOTE OR CITE
-------
0)
V
a
.a
O
o
14
13-
12-
11 -
10-
9-
8-
7-
6-
5-
4-
3-
2-
1 -
0
1 h, 10 t/min
20
40 60
CARBON MONOXIDE, ppm
80
100
Figure 2-1. Relationship between carbon monoxide exposure and carboxyhemoglobin levels
in the blood. Predicted COHb levels resulting from 1- and 8-h exposures to carbon
monoxide at rest (10 L/min) and with light exercise (20 L/min) are based on the Coburn-
Forster-Kane equation (Coburn et al., 1965) using the following assumed parameters for
nonsmoking adults: altitude = 0 ft; initial COHb level = 0.5%; Haldane constant = 218;
blood volume = 5.5 L; hemoglobin level = 15 g/100 ml; lung diffusivity = 30 ml/torr/min;
endogenous rate = 0.007 ml/min. See glossary of terms and symbols for abbreviations and
acronyms.
March 12, 1990
2-7 DRAFT - DO NOT QUOTE OR CITE
-------
2.4.2 Carbon Monoxide Exposure Levels
Evaluation of human CO-exposure situations indicates that occupational exposures in
some workplaces or exposures in homes with faulty combustion appliances can exceed
100 ppm CO, often leading to COHb levels of 10% or more with continued exposure. In
5 contrast, such high exposure levels are encountered much less commonly by the general
/
public exposed under ambient conditions. More frequently, exposures to less than 25 to
V
50 ppm CO for any extended period of time occur among the general population and, at the
low exercise levels usually engaged in under such circumstances, the resulting COHb levels
most typically remain 2 to 3 % among nonsmokers. Those levels can be compared to the
10 physiologic norm for nonsmokers, which is estimated to be in the range of 0.3 to 0.7%
COHb. Baseline COHb concentrations in smokers, however, average 4% with a usual range
of 3 to 8%, reflecting absorption of CO from inhaled smoke.
2.4.3 Health Effects of Low-Level Carbon Monoxide Exposures
15 Four types of health effects reported or hypothesized to be associated with CO
exposures (especially those producing COHb levels below 10%) were evaluated in the last
review of the CO NAAQS (U.S. Environmental Protection Agency, 1984a): (1) cardio-
vascular effects, (2) neurobehavioral effects, (3) fibrinolysis effects, and (4) perinatal effects.
Data available at that time (Table 2-2) demonstrated an association between cardiovascular
20 and neurobehavioral effects at relatively low-level CO exposures. Much less clear evidence
existed to indicate that other types of health effects were associated with low-level CO
exposures.
2.4.3.1 Cardiovascular Effects
25 In regard to cardiovascular effects, decreased oxygen uptake and resultant decreased
work capacity under maximal exercise conditions clearly have been shown to occur in healthy
young adults starting at 5.0% COHb; and several studies observed small decreases in work
capacity at COHb levels as low as 2.3 to 4.3%. These cardiovascular effects may have health
implications for the general population in terms of potential curtailment of certain physically
30 demanding occupational or recreational activities under circumstances of sufficiently high CO-
exposure. However, of greater concern at more typical ambient CO-exposure levels were
March 12, 1990 2-8 DRAFT - DO NOT QUOTE OR CITE
-------
TABLE 2-2. LOWEST OBSERVED EFFECT LEVELS FOR HUMAN HEALTH
EFFECTS ASSOCIATED WITH LOW-LEVEL CARBON MONOXIDE EXPOSURE
Effects
COHb
concentration,
percent*
References
10
15
20
25
30
35
40
45
50
55
Statistically significant
decreased (3-7%) work time
to exhaustion in exercising,
young, healthy men
Statistically significant
decreased exercise capacity
(i.e., shortened duration of
exercise before onset of
pain) in patients with angina
pectoris and increased
duration of angina attacks
No statistically significant
vigilance decrements after
exposure to CO
Statistically significant
decreased maximal oxygen
consumption and exercise time
during strenuous exercise in
young, healthy men
Statistically significant
diminution of visual percep-
tion, manual dexterity,
ability to learn, or
performance in complex
sensorimotor tasks (such
as driving)
Statistically significant
decreased maximal oxygen
consumption during strenuous
exercise in young, healthy men
2.3-4.3
2.9-4.5
Below 5
5-5.5
5-17
7-20
Horvath et al. (1975)
Drinkwater et al. (1974)
Anderson et al. (1973)
Haider et al. (1976)
Winneke (1974)
Christensen et al. (1977)
Benignus et al. (1977)
Putz et al. (1976)
Klein et al. (1980)
Stewart et al. (1978)
Weiser et al. (1978)
Bender et al. (1971)
Schulte (1973)
O'Donnell et al. (1971)
McFarland et al. (1944)
McFarland (1973)
Putz et al. (1976)
Salvatore (1974)
Wright et al. (1973)
Rockwell and Weir (1975)
Rummo and Sarlanis (1974)
Putz et al. (1979)
Putz (1979)
Ekblom and Huot (1972)
Pirnay et al. (1971)
Vogel and Gleser (1972)
'The physiologic norm (i.e., COHb levels resulting from the normal catabolism of hemoglobin and other heme-
containing materials) has been estimated to be in the range of 0.3 to 0.7 % (Coburn et al., 1963). See glossary
of terms and symbols for abbreviations and acronyms.
Source: U.S. Environmental Protection Agency (1984b).
March 12, 1990
2-9 DRAFT - DO NOT QUOTE OR CITE
-------
certain cardiovascular effects (i.e., aggravation of angina symptoms during exercise) likely to
occur in a smaller, but sizeable, segment of the general population. This group, chronic
angina patients, is presently viewed as the most sensitive risk group for CO-exposure effects,
based on evidence for aggravation of angina occurring in patients at COHb levels of 2.9 to
5 4.5%. Such aggravation of angina is thought to represent an adverse health effect for several
reasons articulated in the 1980 proposal preamble (Federal Register, 1980), and the Clean Air
Scientific Advisory Committee (CASAC) concurred with EPA's judgment on this matter.
Dose-response relationships for cardiovascular effects in coronary artery disease patients
remain to be defined more conclusively, and the possibility cannot be ruled out at this time
10 that such effects may occur at levels below 2.9% COHb (as hinted at by the results of the
now-questioned Aronow studies). Therefore, new studies published since the last review
cycle are evaluated in this revised criteria document to determine the effects of CO on
aggravation of angina at levels in the range of 2 to 6% COHb.
15 2.4.3.2 Neurobehavioral Effects
No reliable evidence demonstrating decrements in neurobehavioral function in healthy,
young adults has been reported at COHb levels below 5%. Results of studies conducted at or
above 5% COHb are equivocal. Much of the research at 5% COHb did not show any effect
even when behaviors similar to those affected in other studies at higher COHb levels were
20 involved. However, investigators failing to find CO decrements at 5% or higher COHb
levels may have utilized tests not sufficiently sensitive to reliably detect small effects of CO.
From the empirical evidence, then, it can be said that COHb levels >5% do produce
decrements in neurobehavioral function. It cannot be said confidently, however, that COHb
levels lower than 5% would be without effect. One important point made in the 1979 criteria
25 document should be reiterated here. Only young, healthy adults have been studied using
demonstrably sensitive tests and COHb levels at 5% or greater. The question of groups at
special risk for neurobehavioral effects of CO, therefore, has not been explored. Of special
note are those individuals who are taking drugs that have primary or secondary depressant
effects which would be expected to exacerbate CO-related neurobehavioral decrements. Other
30 groups at possibly increased risk for CO-induced neurobehavioral effects are the aged and ill,
but these groups have not been evaluated for such risk.
March 12, 1990 2-10 DRAFT - DO NOT QUOTE OR CITE
-------
2.4.3.3 Other Health Effects
Only relatively weak evidence points toward possible CO effects on fibrinolytic activity,
generally only at rather high CO-exposure levels. Similarly, whereas certain data also suggest
that perinatal effects (e.g., reduced birth weight, slowed postnatal development, Sudden
Infant Death Syndrome) are associated with CO exposure, insufficient evidence presently
exists by which to either qualitatively confirm such an association in humans or to establish
any pertinent exposure-effect relationships.
10 2.5 CRITICAL ISSUES IN REVIEW OF THE NAAQS FOR CARBON
MONOXIDE
Based on the scientific evidence currently evaluated in air quality criteria documents
(U.S. Environmental Protection Agency, 1979, 1984a), potentially adverse health effects of
CO have been demonstrated to occur at COHb levels in the range of 2.3 to 20% (see
15 Table 2-2). However, several critical issues have developed during the current review of the
scientific criteria for CO air quality standards that will need to be resolved in order to
determine the extent to which adverse effects are occurring in the population, particularly at
the lower COHb levels of greatest interest to standard-setting (<5 percent). The following
section will focus on specific issues pertaining to (1) exposure assessment in the general
20 population, including the measurement of CO in ambient air and in blood; (2) mechanisms of
action of CO; (3) health effects from exposure to CO; and (4) groups of individuals
considered to be at greatest risk to CO at ambient or near-ambient exposure levels.
2.5.1 Exposure Assessment in the Population
25 The 1986 and 1987 trends in ambient air quality reported by EPA (U.S. Environmental
Protection Agency, 1988, 1989a) summarize fixed-site monitoring data for carbon monoxide
but only focus on 8-h averages. The rationale for this approach is that the 8-h standard
(9 ppm) is typically the controlling standard and the 1-h standard (35 ppm) rarely is exceeded.
For example, in 1987 there were only four exceedances of the 1-h standard for the entire
30 United States and in each case the 8-h standard was exceeded by a greater percentage on that
day. In 1988, only two areas (Denver, CO, and Steubenville, OH) exceeded the 1-h CO
March 12, 1990 2-11 DRAFT - DO NOT QUOTE OR CITE
-------
standard, while 44 areas failed to meet the 8-h standard (U.S. Environmental Protection
Agency, 1989b).
Ambient CO-concentration data from fixed-site monitors alone will not necessarily give
a good estimate of potential total exposure to the population, based on experience from the
5 Denver and Washington, DC human-exposure field studies using personal monitors. It is
estimated that over 10% of the residents in Denver and 4% of the residents in Washington,
DC, were exposed to CO levels above 9 ppm for 8 h during the winter of 1982-83 (Akland
et al., 1985). The effects of personal activity, indoor sources, and time spent commuting
contribute greatly to an individual's total exposure to CO. Available 1-h CO concentrations
10 taken at fixed-site monitors in these field studies did not correlate well (0.14 < r < 0.27) with
measurements made by personal monitors.
The best available study for determining relevant exposure to the most susceptible target
population, that is, individuals with ischemic heart disease (IHD), is the work of Lambert and
Colome (1988). A total of 36 nonsmoking men with IHD were followed during personal-
15 exposure monitoring. A wide range of peak exposures to CO were measured. The highest
CO exposures were found while the subjects were commuting and when the subjects were
near internal combustion engines. For example, CO exposures on freeways in Los Angeles
averaged 10 to 12 ppm. The average personal exposure for all time spent in automobiles was
8.6 ppm with a maximum 1-min average of 239 ppm. Concentrations of CO, averaging 7.9
20 ppm, also were found in parking lots, parking structures, service stations, and motor repair
facilities. Residential CO exposure was much lower, averaging 2.0 ppm. In typical outdoor
residential activities, transient peaks as high as 134 ppm were observed for woodcutting with
a gas-powered chain saw and 226 ppm for gardening activity where a two-stroke, gasoline-
powered engine was utilized. Exposures under these conditions would be expected, based on
25 equations developed by Coburn et al. (1965), to cause COHb levels in excess of 2.5%.
The best indicator of exposure to CO continues to be the direct measurement of COHb
in blood. There are, however, several issues regarding measurement techniques for COHb
that have been raised during the current review of the CO air quality criteria. For many
years, routine clinical laboratory measurements of COHb commonly have been made using
30 the IL 182 and it successor, the IL 282 CO-Oximeter (Instrumentation Laboratory, Inc.,
Lexington, MA) which is a spectrophotometric instrument. Optical methods of COHb
March 12, 1990 2-12 DRAFT - DO NOT QUOTE OR CITE
-------
measurement, however, are limited in sensitivity, particularly in the range of 0 to 5% where
the lowest observed health effects associated with CO exposure have been described. Other,
more sensitive techniques require the release of CO from hemoglobin into a gas phase that
can be detected directly. One method for COHb measurement that has become more widely
5 used in laboratory settings is gas chromatography. Recent efforts to compare COHb
measurement by spectrophotometry versus gas chromatography have indicated that the high
correlation over a wide range of concentrations (0 to >20%) becomes much worse at COHb
levels <5% because of an apparent instrument offset or potential error. Thus, there has been
concern about the relative accuracy and precision of the COHb measurements at levels that
10 are of particular concern to the CO NAAQS review. Further, ongoing work is needed in
order to determine (1) which method should be used to accurately quantify low levels of
COHb; (2) if there is a scientifically acceptable way to compare COHb measurements made
by different instruments across different laboratories; and (3) the relationship of measured
COHb values to those derived from modeling efforts based on actual CO exposures in the
15 general population.
Most "real-life" exposures to CO are to concentrations that vary with time and those
exposures are experienced by people with differing physiological attributes and at varying
exercise levels. Direct measurements of COHb are not readily available in the general
population exposed to CO under these conditions. Mathematical models, therefore, have been
20 developed to predict COHb levels from known CO exposures under a variety of
circumstances. The most used model for COHb formation is still the Coburn-Forster-Kane
equation (CFKE) developed by Coburn et al. (1965). The COHb levels predicted by this
equation generally have been accepted as the best available estimates of COHb levels likely to
result from varying CO concentrations, exposure durations, and exercise levels. Further
25 research, however, is needed to evaluate the predictive capabilities of the CFKE in individuals
exposed to low concentrations of CO leading to COHb levels of less than 10%. Of particular
interest is the variation of predicted COHb in a population whose pattern of CO exposure
involves frequent concentration variations. In addition, the CFKE needs to be evaluated for
applicability to CO-susceptible subjects, such as patients with cardiovascular or pulmonary
30 disease. Clinical evaluation of CO uptake by these individuals should be considered.
March 12, 1990 2-13 DRAFT - DO NOT QUOTE OR CITE
-------
Epidemiology studies have suggested the possibility that increased mortality from heart
attacks and increased cardiovascular complaints may be associated with elevated ambient
concentrations of CO. Unfortunately, due to inadequate characterization of exposure as well
as other limitations, inconclusive results have been obtained from existing studies. The
5 availability of both personal-exposure monitors for CO and ambulatory EKG monitoring
techniques have made it possible to design epidemiology studies to determine whether ambient
CO exposures are related to serious or irreversible cardiovascular effects. It would be
desirable, therefore, to obtain CO exposure data on CO-susceptible individuals in order to
characterize their risk from elevated levels of COHb. Potentially susceptible individuals
10 include infants, the elderly, and patients with known cardiovascular diseases.
2.5.2 Mechanisms of Action of Carbon Monoxide
The accepted mechanisms of action underlying the potentially toxic effects of low-level
CO exposure continue to be the decreased oxygen-carrying capacity of blood and subsequent
15 interference of oxygen release at the tissue level that is caused by the binding of CO with
hemoglobin, producing COHb (Figure 2-2). The resulting impaired delivery of oxygen can
interfere with cellular respiration and cause tissue hypoxia.
Review of the newer information on mechanisms of action of CO has focused on the
possibility that secondary mechanisms that also can impair cellular respiration may be
20 occurring at relatively low (near-ambient) CO-exposure levels. Approximately 10 to 50% of
the total-body burden of CO can be distributed to extravascular sites, suggesting that intra-
cellular uptake of CO may contribute to CO-induced toxicity. It is uncertain, however, if
intracellular uptake of CO occurs at low levels of COHb or if it would be likely to contribute
to the physiological effects of CO.
25 Carbon monoxide will bind to intracellular hemoproteins such as myoglobin (Mb),
cytochrome oxidase, mixed function oxidases (e.g., cytochrome P-450), tryptophan
oxygenase, and dopamine nydroxylase. Binding to CO would be favorable under conditions
of low intracellular partial pressure of oxygen (PO2), particularly in brain and myocardial
tissue where intracellular PO2 decreases with increasing COHb levels. The most likely
30 hemoproteins to be inhibited functionally at relevant levels of COHb are Mb, found
predominantly in heart and skeletal muscle, and cytochrome oxidase. The physiological
March 12, 1990 2-14 DRAFT - DO NOT QUOTE OR CITE
-------
MECHANISMS OF ACTION OF CARBON MONOXIDE
Source
Internal
storage
compartment
Target
organ
External Exposure
Carbon
Monoxide
Ambient
Indoor
Occupational
Halogenated
Hydrocarbons
e.g.
Internal
Endogenous
Production
of CO
1
Extravascular
CO + Mb
CO + Cyt
Decreased 02 Delivery
Decreased Og-carrying capacity
L-shift Hb02 dissociation curve
I
Compensatory
vasodilation
and increased
blood flow
to maintain
Og consumption
Decreased cellular respiration
i
Tissue hypoxia
Ischemia cascade
Figure 2-2. Currently accepted or proposed mechanisms of action of carbon monoxide
resulting from external exposure sources can interfere with cellular respiration and cause
tissue hypoxia (see text for details). See glossary of terms and symbols for abbreviations and
acronyms.
March 12, 1990
2-15 DRAFT - DO NOT QUOTE OR CITE
-------
significance of CO uptake by Mb is uncertain at this time but sufficient concentrations of
carboxymyoglobin (COMb) could potentially limit maximal oxygen uptake of exercising
muscle. Although there is suggestive evidence for significant binding of CO to cytochrome
oxidase in heart and brain tissue, it is unlikely that any significant CO binding would occur at
5 low COHb levels. Therefore, further research still is needed to determine if secondary,
intracellular mechanisms will occur at exposure concentrations found in ambient air.
2.5.3 Health Effects from Exposure to Carbon Monoxide
2.5.3.1 Effects on the Cardiovascular System
10 Scientific support for the current NAAQS for CO is based primarily on studies of
patients with stable angina pectoris (chest pain) from coronary artery disease. Although it is
assumed that the development of angina reflects adverse effects of CO on myocardial
metabolism, more specific research supporting the validity of this assumption is needed. For
example, little is known about the reproducibility or reoccurrence of this disease. Time to
15 onset of angina and the duration of angina are measurable outcomes that need to be defined
more precisely. Research also is needed on more objective measures of myocardial ischemia,
such as continuous EKG tracing for ST depression and arrhythmias, and on measurement of
ventricular function using a gamma camera or thallium scan.
In view of questions concerning the validity of angina studies by Aronow et al. reviewed
20 in the previous criteria document, additional data were clearly needed in order to (1) provide
more reliable dose-response information in individuals with stable angina, (2) allow a better
determination of the level of COHb necessary to cause adverse effects in the sensitive
population, and (3) ultimately set an appropriate level for the CO NAAQS. In response to
this need, additional studies recently have been completed by a number of independent
25 laboratories to identify the relationship between COHb and aggravation of preexisting chronic
heart disease. Four of these studies now have been published (Sheps et al., 1987; Adams
et al., 1988; Kleinman et al., 1989; Allred et al., 1989). Collectively, all four studies
provide new information on the likelihood that patients exposed to CO will experience angina
earlier during exercise when compared to clean air exposure. Levels of COHb across the
30 studies range from 2.9 to 5.9%, as measured by the spectrophotometric method (CO-
Oximeter). An evaluation of these data is provided in this document as part of the overall
March 12, 1990 2-16 DRAFT - DO NOT QUOTE OR CITE
-------
review of the scientific basis for the CO NAAQS. Any potential differences in the results
between these studies primarily will be due to either the patient population studied or to the
experimental design of the study itself.
Heart attack is the leading cause of death in the United States. In 1985 alone, over
5 540,000 deaths were attributed to coronary artery disease and over 4.8 million people alive at
that time were estimated to have a history of heart attack, angina, or both (American Heart
Association, 1988). Today that estimate may be as high as 6 to 7 million individuals inflicted
with coronary artery disease. A major question that will become important in the evaluation
of all the clinical studies involving subjects with coronary heart disease is whether the study
10 population is representative of this broad group of patients with angina and, therefore, is
applicable to the subpopulation of potentially susceptible individuals that are exposed routinely
to ambient levels of CO. The possibility of studying the effects of CO in a more
representative group of patients with coronary heart disease should be investigated. Recent
changes in the treatment of coronary artery disease indicate that the sensitive subpopulation of
15 angina patients may be changing from one of untreated patients to one of angina patients who
have had coronary artery bypass or balloon angioplasty. The susceptibility of this new
population to CO may not be the same. In addition, there is a greater likelihood of increased
risk to CO exposure in a virtually unknown group of individuals who have silent ischemia (no
symptomatic episodes of chest pain).
20 Additional research is needed to determine dose-response relationships for the acute
effects of CO in other potentially susceptible groups. Patients with arteriosclerosis of the
arteries of the lower limbs who develop intermittent claudication are analogous to patients
with angina and could be studied in a similar manner. Research is needed to determine dose-
response relationships for cardiovascular effects in individuals with ventricular arrhythmias.
25 Patients with anemia may be susceptible to increased levels of COHb, because CO would
further reduce the already compromised arterial oxygen content of the blood. Patients with
chronic obstructive pulmonary disease and those with congestive heart failure also should be
studied to determine if they are at increased risk to low levels of CO exposure.
Other cardiovascular effects of low-level CO exposure, particularly with prolonged or
30 chronic exposure, have not been demonstrated. Previous studies on laboratory animals that
were reviewed in the last criteria document failed to clearly link CO exposure with
March 12, 1990 2-17 DRAFT - DO NOT QUOTE OR CITE
-------
atherogenesis and the development of arteriosclerosis. Newer data published since then still
fail to prove conclusively an atherogenic effect of exposure to low concentrations of CO
despite strong evidence from epidemiology studies showing an association between cigarette
smoke and increased risk for arteriosclerosis. Other components of cigarette smoke (e.g.,
5 nicotine) as well as other risk factors (e.g., diet) also may promote atherogenesis, making it
difficult to attribute the atherogenic effects of cigarette smoke to CO alone.
2.5.3.2 Neurobehavioral Effects
Neurobehavioral effects of CO exposure, such as changes in (1) hand-eye coordination
10 (compensatory tracking), (2) detection of infrequent events (vigilance), and (3) visual system
sensitivity have been reported in healthy young adults at COHb levels as low as 5%. These
effects at low CO-exposure concentrations, however, have been very small and somewhat
controversial. The newer data on neurobehavioral effects of CO discussed in this document
apparently have provided little help in resolving this controversy. Nevertheless, the potential
15 consequences of a lapse of coordination, vigilance, and visual sensitivity in the performance
of critical tasks by operators of machinery such as public transportation vehicles could be
serious. Therefore, additional research is necessary to provide a better understanding of the
mechanisms of action of CO and compensatory changes in the vascular bed that may act to
maintain an adequate oxygen supply to the brain.
20 Certain subgroups of the population are at increased risk from the neural and behavioral
effects of elevated COHb. For example, any condition that would reduce oxygen supply to
the brain also would potentially exacerbate the effects of CO exposure. A very large
subgroup that is known to have a reduced oxygen supply to the brain is the aged. Therefore,
it is important to determine COHb dose-response functions for neurobehavioral variables in
25 older subjects. Other conditions that might reduce oxygen supply to the brain include certain
cerebrovascular, cardiovascular, and pulmonary disease states mentioned above.
Another large subgroup that may be at increased risk from neurobehavioral effects of
CO exposure are those people who take prescription or over-the- counter medications such as
antihistamines, sedatives, antipsychotics, antiseizure drugs, antiemetics, and analgesics, that
30 reduce alertness or motor abilities. The effects of ethanol, caffeine, nicotine, and other
nonprescription drugs should not be overlooked. Such individuals already would be affected
March 12, 1990 2-18 DRAFT - DO NOT QUOTE OR CITE
-------
behaviorally so that any further impairment due to elevated COHb might have serious
consequences.
2.5.3.3 Perinatal Effects
5 The fetus and newborn infant are theoretically susceptible to CO exposure for several
reasons. Fetal circulation is likely to have a higher COHb level than the maternal circulation
due to differences in uptake and elimination of CO from fetal hemoglobin. Since the fetus
also has a lower oxygen tension in the blood than adults, any further drop in fetal oxygen
tension due to the presence of COHb could have a potentially serious effect. The newborn
10 infant with a comparatively high rate of oxygen consumption and lower hemoglobin blood
oxygen transport capacity than most adults also would be potentially susceptible to the
hypoxic effects of increased COHb. Newer data from laboratory animal studies on the
developmental toxicity of CO suggest that prolonged exposure to high levels (> 100 ppm) of
CO during gestation may produce a reduction in birthweight, cardiomegaly, and delayed
15 behavioral development. Human data are scant and more difficult to evaluate, but further
research is warranted. Therefore, additional studies are needed in order to determine if
chronic exposure to CO, particularly at low, near-ambient levels, can compromise the already
marginal conditions existing in the fetus and newborn infant.
The effects of CO on maternal-fetal relationships are not understood well. In addition to
20 fetuses and newborn infants, pregnant women also represent a susceptible group because
pregnancy is associated with increased alveolar ventilation and an increased rate of oxygen
consumption that serves to increase the rate of CO uptake from inspired air. Perhaps a more
important factor is that pregnant women experience hemodilution due to the disproportionate
increase in plasma volume as compared to erythrocyte volume. This group, therefore, should
25 be studied to evaluate the effects of CO exposure and elevated COHb levels.
2.5.4 Population Groups at Greatest Risk for Ambient CO Exposure
Effects
Angina patients or others with obstructed coronary arteries, but not yet manifesting overt
30 symptomatology of coronary artery disease, appear to be best established as a sensitive group
within the general population that is at increased risk for experiencing health effects (i.e.,
March 12, 1990 2-19 DRAFT - DO NOT QUOTE OR CITE
-------
exacerbation of cardiovascular symptoms) of concern at ambient or near-ambient CO-
exposure levels. Several other probable risk groups were identified: (1) fetuses and young
infants; (2) pregnant women; (3) the elderly, especially those with compromised
cardiopulmonary or cerebrovascular functions; (4) individuals with obstructed coronary
5 arteries, but not yet manifesting overt symptomatology of coronary artery disease;
(5) individuals with congestive heart failure; (6) individuals with peripheral vascular or
cerebrovascular disease; (7) individuals with hematological diseases (e.g., anemia) that affect
oxygen-carrying capacity or transport in the blood; (8) individuals with genetically unusual
forms of hemoglobin associated with reduced oxygen-carrying capacity; (9) individuals with
10 chronic obstructive lung diseases; (10) individuals using medicinal or recreational drugs
having central nervous system (CNS) depressant properties; (11) individuals exposed to other
pollutants (e.g., methylene chloride) that increase endogenous formation of CO; and
(12) individuals who have not been adapted to high altitude and are exposed to a combination
of high altitude and CO. However, little empirical evidence currently is available by which
15 to specify health effects associated with ambient or near-ambient CO exposures in these
probable risk groups.
2.6 CARBON MONOXIDE POISONING
20 The majority of this document deals with the relatively low concentrations of CO that
induce effects in humans at or near the lower margin of detection by current medical
technology. Yet, the health effects associated with exposure to this pollutant range from the
more subtle cardiovascular and neurobehavioral effects at low-ambient concentrations, as
identified in the preceding sections, to unconsciousness and death after prolonged chronic
25 exposure or after acute exposure to high concentrations of carbon monoxide. The morbidity
and mortality resulting from the latter exposures are described briefly here to complete the
picture of CO exposure in present-day society.
Carbon monoxide is responsible for more than half of the fatal poisonings that are
reported in the United States each year (National Safety Council, 1982). At sublethal levels,
30 CO poisoning occurs in a small but important fraction of the population. Certain conditions
exist in both the indoor and outdoor ambient environments that cause a small percentage of
March 12, 1990 2-20 DRAFT - DO NOT QUOTE OR CITE
-------
the population to become exposed to dangerous levels of CO. Outdoors, concentrations of
CO are highest near intersections, in congested traffic, near exhaust gases from internal
combustion engines and from industrial combustion sources, and in poorly ventilated areas
such as parking garages and tunnels. Indoors, CO concentrations in the workplace or in
5 homes that have faulty appliances or downdrafts and backdrafts have been measured in excess
of 100 ppm, resulting in COHb levels of greater than 10% for 8 h of exposure. In addition,
CO is found in the smoke produced by all types of fires. Of the 6000 deaths from burns in
the United States each year, more than half are related to inhalation injuries where victims die
from CO poisoning, hypoxia, and smoke inhalation (Heimbach and Waeckerle, 1988).
10 Carbon monoxide poisoning is not new, although more attention to this problem has
been addressed recently in the scientific literature as well as in the popular media. The first
scientific studies of the hypoxic effects of CO were described by Claude Bernard (1865). The
attachment of CO to hemoglobin, producing carboxyhemoglobin, was evaluated by Douglas
et al. (1912), providing the necessary tools for studying man's response to CO. During the
15 next half century, numerous studies were conducted with the principal emphasis being on high
concentrations of COHb. Carbon monoxide poisoning as an occupational hazard (Grut, 1949)
received the greatest attention due to the increased use of natural gas and the potential for
leakage of exhaust fumes in homes and industry. Other sources of CO have become more
important and more insidious. The clinical picture of CO poisoning, as described by Grut
20 (1949), relate primarily to the alterations in cardiac and central nervous system function due
to the extreme hypoxia induced.
Mortality from carbon monoxide exposure is high. In 1985, 1365 deaths due to CO
exposure were reported in England and Wales (Meredith and Vale, 1988). In the United
States, more than 3800 people die annually from CO (accidental and intentional), and more
25 than 10,000 individuals seek medical attention or miss at least one day of work because of a
sublethal exposure (U.S. Centers for Disease Control, 1982). The per capita mortality and
morbidity statistics for CO are surprisingly similar for the Scandinavian countries and for
Canada, as well. However, not all instances of CO poisoning are reported and complete up-
to-date data are difficult to obtain. Often the individuals suffering from CO poisoning are
30 unaware of their exposure because symptoms are similar to those associated with the flu or
with clinical depression. This may result in a significant number of misdiagnoses by medical
March 12, 1990 2-21 DRAFT - DO NOT QUOTE OR CITE
-------
professionals (Heckerling et al., 1988, 1987; Kirkpatrick, 1987; Dolan et al., 1987; Barrett
et al., 1985; Fisher and Rubin, 1982; Grace and Platt, 1981). The precise number of
individuals who have suffered from CO intoxication, therefore, is not known but it is
certainly larger than the mortality figures indicate. Nonetheless, the reported literature
5 available for review indicates the seriousness of this problem.
The symptoms, signs, and prognosis of acute poisoning correlates poorly with the level
of COHb measured at the time of arrival at the hospital (Meredith and Vale, 1988).
Carboxyhemoglobin levels below 10% usually are not associated with symptoms. At the
higher COHb saturations of 10 to 30%, neurological symptoms of CO poisoning can occur,
10 such as headache, dizziness, weakness, nausea, confusion, disorientation, and visual
disturbances. Exertional dyspnea, increases in pulse and respiratory rates, and syncope are
observed with continuous exposure producing COHb levels in excess of 30 to 50%. When
COHb levels are higher than 50%, coma, convulsions, and cardiorespiratory arrest may
occur.
15 Different individuals experience very different clinical manifestations of CO poisoning
and, therefore, have different outcomes even under similar exposure conditions. Norkool and
Kirkpatrick (1985) found that COHb levels in individuals who had never lost consciousness
ranged from 5 to 47%. In individuals who were found unconscious but regained
consciousness at hospital arrival, the range was 10 to 64%; for those remaining unconscious,
20 COHb levels varied from 1 to 53%. These data clearly indicate that COHb saturations
correlate poorly with clinical status and, furthermore, have little prognostic significance.
The level of CO in the tissues may have an equal or greater impact on the clinical
status of the patient than the blood level of CO (Broome et al., 1988). For example a short
exposure to CO at high ambient concentrations may allow insufficient time for significant
25 increases in tissue levels of CO to occur. The syncope observed in these individuals may be
the result of simple hypoxia with rapid recovery despite high COHb levels. A prolonged
exposure to low concentrations of CO prior to hospital arrival may allow sufficient uptake of
CO by tissues to inhibit the function of intracellular compounds such as myoglobin. This
effect, in combination with hypoxia, may cause irreversible central nervous system or cardiac
30 damage.
March 12, 1990 2-22 DRAFT - DO NOT QUOTE OR CITE
-------
Patients with CO poisoning respond to treatment with 100% oxygen (Pace et al., 1950).
If available, treatment with hyperbaric oxygen (HBO) at 2.5 to 3 times atmospheric pressure
for 90 min is preferable (Myers, 1986), but the precise conditions requiring treatment have
been a topic of debate in the literature (Thorn and Keim, 1989; Roy et al., 1989; Raphael
5 et al., 1989; Brown et al., 1989; James, 1989; Van Hoesen et al., 1989; Broome et al.,
1988; Norkool and Kirkpatrick, 1985; Mathieu et al., 1985). It has been suggested that if
COHb is above 25%, HBO treatment should be initiated (Norkool and Kirkpatrick, 1985),
although treatment plans based on specific COHb saturations is not well founded (Thorn and
Keim, 1989). Most hyperbaric centers treat patients with CO intoxication when they manifest
10 loss of consciousness or other neurological signs and symptoms (excluding headace)
regardless of the COHb saturation at presentation (Piantadosi, 1990). The halftime
elimination of CO while breathing air is approximately 320 min; when breathing 100%
oxygen it is 80 min, and when breathing oxygen at 3 atmospheres it is 23 min (Penney et al.,
1983; Myers etal., 1985).
15 Successful removal of CO from the blood does not ensure an uneventful recovery with
no further clinical signs or symptoms. Neurological problems may develop insidiously weeks
after recovery from the acute episode of CO poisoning (Meredith and Vale, 1988). These
problems include intellectual deterioration; memory impairment; and cerebral, cerebellar, and
mid-brain damage. Up to two-fifths of patients develop memory impairment and a third
20 suffer late deterioration of personality. Arrhythmias are a common complication of CO
poisoning. Conduction defects also are found, possibly from cardiomyopathies, but the
precise mechanisms by which these occur are not understood. Other systemic complications,
such as skeletal muscle necrosis, renal failure, blood dyscrasias, pulmonary edema, and
hemorrhage in various tissues also can occur as a result of CO poisoning.
25
2.7 REFERENCES
30 Adams, K. F.; Koch, G.; Chatterjee, B.; Goldstein, G. M.; O'Neil, J. J.; Bromberg, P. A.; Sheps, D. S.;
McAllister, S.; Price, C. J.; Bissette, J. (1988) Acute elevation of blood carboxyhemoglobin to 6%
impairs exercise performance and aggravates symptoms in patients with ischemic heart disease. J. Am.
Coll. Cardiol. 12: 900-909.
March 12, 1990 2-23 DRAFT - DO NOT QUOTE OR CITE
-------
Akland, G. G.; Hartwell, T. D.; Johnson, T. R.; Whitmore, R. W. (1985) Measuring human exposure to carbon
monoxide in Washington, D.C., and Denver, Colorado, during the winter of 1982-1983. Environ. Sci.
Technol. 19: 911-918.
5 Allred, E. N.; Bleecker, E. R.; Chaitman, B. R.; Dahms, T. E.; Gottlieb, S. O.; Hackney, J. D.; Hayes, D.;
Pagano, M. Selvester, R. H.; Walden, S. M.; Warren, J. (1989) Short-term effects of carbon monoxide
exposure on the exercise performance of subjects with coronary artery disease. N. Engl. J. Med.
321: 1426-1432.
10 American Heart Association. (1988) 1988 heart facts. Dallas, TX: American Heart Association.
Anderson, E. W.; Andelman, R. J.; Strauch, J. M.; Fortuin, N. J.; Knelson, J. H. (1973) Effect of low-level
carbon monoxide exposure on onset and duration of angina pectoris: a study in ten patients with ischemic.
heart disease. Ann. Intern. Med. 79: 46-50.
15
Barret, L.; Danel, V.; Faure, J. (1985) Carbon monoxide poisoning, a diagnosis frequently overlooked. J.
Toxicol. Clin. Toxicol. 23: 309-313.
Beard, R. R.; Wertheim, G. A. (1967) Behavioral impairment associated with small doses of carbon monoxide.
20 Am. J. Public Health 57: 2012-2022.
Bender, E.; Cothert, M.; Malomy, G.; Sebbesse, P. (1971) The effects of low concentrations of carbon
monoxide in man. Arch. Toxicol. 27: 142-158.
25 Benignus, V. A.; Otto, D. A.; Prah, J. D.; Benignus, G. (1977) Lack of effects of carbon monoxide on human
vigilance. Percept. Mot. Skills 45: 1007-1014.
Bernard, C. (1865) An introduction to the study of experimental medicine. 1957 reprint. Greene, H. C., trans.
New York, NY: Dover.
30
Broome, J. R.; Skrine, H.; Pearson, R. R. (1988) Carbon monoxide poisoning: forgotten not gone! Br. J. Hosp.
Med. 39: 298, 300, 302, 304-305.
Brown, S. D.; Piantadosi, C. A.; Gorman, D. F.; Gilligan, J. E. F.; Clayton, D. G.; Neubauer, R. A.;
35 Gottlieb, S. F.; Raphael, J.-C.; Elkharrat, D.; Jars-Guincestre, M.-C.; Chastang, C.; Chasles, V.;
Vercken, J.-B.; Gajdos, P. (1989) Hyperbaric oxygen for carbon monoxide poisoning [letters]. Lancet
(8670): 1032-1033.
Christensen, C. L.; Gliner, J. A.; Horvath, S. M.; Wagner, J. A. (1977) Effects of three kinds of hypoxias on
40 vigilance performance. Aviat. Space Environ. Med. 48: 491-496.
Coburn, R. F.; Blakemore, W. S.; Forster, R. E. (1963) Endogenous carbon monoxide production in man. J.
Clin. Invest. 42: 1172-1178.
45 Coburn, R. F.; Forster, R. E.; Kane, P. B. (1965) Considerations of the physiological variables that determine
the blood carboxyhemoglobin concentration in man. J. Clin. Invest. 44: 1899-1910.
Dolan, M. C.; Haltom, T. L.; Barrows, G. H.; Short, C. S.; Ferriell, K. M. (1987) Carboxyhemoglobin levels
in patients with flu-like symptoms. Ann. Emerg. Med. 16: 782-786.
50
Douglas, C. G.; Haldane, J. S.; Haldane, J. B. S. (1912) The laws of combination of haemoglobin with carbon
monoxide and oxygen. J. Physiol. (London) 44: 275-304.
March 12, 1990 2-24 DRAFT - DO NOT QUOTE OR CITE
-------
Drinkwater, B. L.; Raven, P. B.; Horvath, S. M.; Gliner, J. A.; Ruhling, R. O.; Bolduan, N. W.; Taguchi, S.
(1974) Air pollution, exercise, and heat stress. Arch. Environ. Health 28: 177-181.
Ekblom, B.; Huot, R. (1972) Response to submaximal and maximal exercise at different levels of
5 carboxyhemoglobin. Acta Physiol. Scand. 86: 474-482.
Federal Register. (1980) Carbon monoxide; proposed revisions to the national ambient air quality standards:
proposed rule. F. R. (August 18) 45: 55066-55084.
10 Federal Register. (1985) Review of the national ambient air quality standards for carbon monoxide; final rule.
F. R. ( September 13) 50: 37484-37501.
Fisher, J.; Rubin, K. P. (1982) Occult carbon monoxide poisoning [editorial]. Arch. Int. Med. 142: 1270-1271.
15 Grace, T. W.; Platt, F. W. (1981) Subacute carbon monoxide poisoning: another great imitator. JAMA J. Am.
Med. Assoc. 246: 1698-1700.
Grut, A. (1949) Chronic carbon monoxide poisoning: a study in occupational medicine. Copenhagen, Denmark:
Ejnar Munksgaard.
20
Haider, M.; Groll-Knapp, E.; Hoeller, H.; Neuberger, M.; Stidl, H. (1976) Effects of moderate carbon
monoxide dose on the central nervous system—electrophysiological and behaviour data and clinical
relevance. In: Finkel, A. J.; Duel, W. C., eds. Clinical implications of air pollution research: air
pollution medical research conference; December 1974; San Francisco, CA. Acton, MA: Publishing
25 Sciences Group, Inc.; pp. 217-232.
Heckerling, P. S.; Leikin, J. B.; Maturen, A.; Perkins, J. T. (1987) Predictors of occult carbon monoxide
poisoning in patients with headache and dizziness. Ann. Intern. Med. 107: 174-176.
30 Heckerling, P. S.; Leikin, J. B.; Maturen, A. (1988) Occult carbon monoxide poisoning: validation of a
prediction model. Am. J. Med. 84: 251-256.
Heimbach, D. M.; Waeckerle, J. F. (1988) Inhalation Injuries. Presented at: ACEP winter symposium clinical
advances track; March; San Diego, CA. Ann. Emerg. Med. 17: 1316-1320.
35
Horvath, S. M.; Raven, P. B.; Dahms, T. E.; Gray, D. J. (1975) Maximal aerobic capacity at different levels of
carboxyhemoglobin. J. Appl. Physiol. 38: 300-303.
James, P. B. (1989) Hyperbaric and normobaric oxygen in acute carbon monoxide poisoning [letter]. Lancet
40 (8666): 799-780.
Kirkpatrick, J. N. (1987) Occult carbon monoxide poisoning. West. J. Med. 146: 52-56.
Klein, J. P.; Forster, H. V.; Stewart, R. D.; Wu, A. (1980) Hemoglobin affinity for oxygen during short-term
45 exhaustive exercise. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 48: 236-242.
Kleinman, M. T.; Davidson, D. M.; Vandagriff, R. B.; Caiozzo, V. J.; Whittenberger, J. L. (1989) Effects of
short-term exposure to carbon monoxide in subjects with coronary artery disease. Arch. Environ. Health
44: 361-369.
50
Lambert, W. E.; Colome, S. D. (1988) Distribution of carbon monoxide exposures within indoor residential
locations. Presented at: 81st annual meeting of the Air Pollution Control Association; June; Dallas, TX.
Pittsburgh, PA: Air Pollution Control Association; paper no. 88-90.1.
March 12, 1990 2-25 DRAFT - DO NOT QUOTE OR CITE
-------
Mathieu, D.; Nolf, M.; Durocher, A.; Saulnier, F.; Frimat, P.; Furon, D.; Wattel, F. (1985) Acute carbon
monoxide poisoning risk of late sequelae and treatment by hyperbaric oxygen. J. Toxicol. Clin. Toxicol.
24: 315-324.
5 McFarland, R. A. (1973) Low level exposure to carbon monoxide and driving performance. Arch. Environ.
Health 27: 355-359.
McFarland, R. A.; Roughton, F. J. W.; Halperin, M. H.; Niven, J. I. (1944) The effects of carbon monoxide
and altitude on visual thresholds. J. Aviat. Med. 15: 381-394.
10
Meredith, T.; Vale, A. (1988) Carbon monoxide poisoning. Br. Med. J. 296: 77-79.
Myers, R. A. M., chairman. (1986) Hyperbaric oxygen therapy: a committee report. Bethesda, MD: Undersea
and Hyperbaric Medical Society; pp. 33-36.
15
Myers, R. A. M.; Snyder, S. K.; Emhoff, T. A. (1985) Subacute sequelae of carbon monoxide poisoning. Ann.
Emerg. Med. 14: 1163-1167.
National Air Pollution Control Administration. (1970) Air quality criteria for carbon monoxide. Washington,
20 DC: U. S. Department of Health, Education, and Welfare, Public Health Service; report no.
NAPCA-PUB-AP-62. Available from: NTIS, Springfield, VA; PB-190261.
National Safety Council. (1982) Accident facts. National Safety Council; pp. 80-84.
25 Norkool, D. M.; Kirkpatrick, J. N. (1985) Treatment of acute carbon monoxide poisoning with hyperbaric
oxygen: a review of 115 cases. Ann. Emerg. Med. 14: 1168-1171.
O'Donnell, R. D.; Mikulka, P.; Heinig, P.; Theodore, J. (1971) Low level carbon monoxide exposure and
human psychomotor performance. Toxicol. Appl. Pharmacol. 18: 593-602.
30
Pace, N.; Strajman, E.; Walker, E. L. (1950) Acceleration of carbon monoxide elimination in man by high
pressure oxygen. Science (Washington, DC) 111: 652-654.
Penney, D. G.; Zak, R.; Aschenbrenner, V. (1983) Carbon monoxide inhalation: effect on heart cytochrome c in
35 the neonatal and adult rat. J. Toxicol. Environ. Health 12: 395-406.
Piantadosi, C. A. (1990) Carbon monoxide intoxication. In: Update in intensive care and emergency medicine,
Vol. 10 (in press).
40 Pirnay, J.; DuJardin, J.; DeRoanne, R.; Petit, J. M. (1971) Muscular exercise during intoxication by carbon
monoxide. J. Appl. Physiol. 31: 573.
Putz, V. R. (1979) The effects of carbon monoxide on dual-task performance. Human Factors 21: 13-24.
45 Putz, V. R.; Johnson, B. L.; Setzer, J. V. (1976) Effects of CO on vigilance performance: effects of low level
carbon monoxide on divided attention, pitch discrimination, and the auditory evoked potential.
Cincinnati, OH: U. S. Department of Health, Education, and Welfare, National Institute for Occupational
Safety and Health. Available from: NTIS, Springfield, VA; PB-274219.
50 Putz, V. R.; Johnson, B. L.; Setzer, J. V. (1979) A comparative study of the effects of carbon monoxide and
methylene chloride on human performance. J. Environ. Pathol. Toxicol. 2: 97-112.
March 12, 1990 2-26 DRAFT - DO NOT QUOTE OR CITE
-------
Raphael, J. C.; Elkharrat, D.; Jars-Guincestre, M.-C.; Chastang, C.; Chasles, V.; Vercken, J. B.; Gajdos, P.
(1989) Trial of nonnobaric and hyperbaric oxygen for acute carbon monoxide intoxication. Lancet
(8660): 414-419.
5 Rockwell, T. J.; Weir, F. W. (1975) The interactive effects of carbon monoxide and alcohol on driving skills.
Columbus, OH: The Ohio State University Research Foundation; CRC-APRAC project CAPM-9-69.
Available from: NTIS, Springfield, VA; PB-242266.
Roy, T. M.; Mendieta, J. M.; Ossorio, M. A.; Walker, J. F. (1989) Perceptions and utilization of hyperbaric
10 oxygen therapy for carbon monoxide poison-ing in an academic setting. J. Ky. Med. Assoc. 87: 223-
226.
Rummo, N.; Sarlanis, K. (1974) The effect of carbon monoxide on several measures of vigilance in a simulated
driving task. J. Saf. Res. 6: 126-130.
15
Salvatore, S. (1974) Performance decrement caused by mild carbon monoxide levels on two visual functions. J.
Saf. Res. 6: 131-134.
Schulte, J. H. (1973) Effects of mild carbon monoxide intoxication. Arch. Environ. Health 7: 524-530.
20
Sheps, D. S.; Adams, K. F., Jr.; Bromberg, P. A.; Goldstein, G. M.; O'Neil, J. J.; Horstman, D.; Koch, G.
(1987) Lack of effect of low levels of carboxyhemoglobin on cardiovascular function in patients with
ischemic heart disease. Arch. Environ. Health 42: 108-116.
25 Statutes-at-Large. (1986) Superfund Amendments and Reauthorization Act of 1986, PL 99-499, October 17,
1986, title IV, §403: radon gas and indoor air quality research program. Stat. 100: 1758.
Stewart, R. D.; Newton, P. E.; Kaufman, J.; Forster, H. V.; Klein, J. P.; Keelen, M. H., Jr.; Stewart, D. J.;
Wu, A.; Hake, C. L. (1978) The effect of a rapid 4% carboxyhemoglobin saturation increase on maximal
30 treadmill exercise. New York, NY: Coordinating Research Council, Inc.; report no.
CRC-APRAC-CAPM-22-75. Available from: NTIS, Springfield, VA; PB-296627.
Thorn, S. R.; Keim, L. W. (1989) Carbon monoxide poisoning: a review epidemiology, pathophysiology, clinical
findings, and treatment options including hyperbaric oxygen therapy. J. Toxicol. Clin. Toxicol.
35 27: 141-156.
U. S. Centers for Disease Control. (1982) Carbon monoxide intoxication—a preventable environmental health
hazard. Morb. Mortal. Wkly. Rep. 31: 529-531.
40 U. S. Code. (1982) Clean Air Act: air quality criteria and control techniques; national primary and secondary
ambient air quality standards. U. S. C. 42: §7408-7409.
U. S. Environmental Protection Agency. (1979) Air quality criteria for carbon monoxide. Research Triangle
Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and Assessment
45 Office; EPA report no. EPA-600/8-79-022. Available from: NTIS, Springfield, VA; PB81-244840.
U. S. Environmental Protection Agency. (1984a) Revised evaluation of health effects associated with carbon
monoxide exposure: an addendum to the 1979 EPA air quality criteria document for carbon monoxide.
Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria
50 and Assessment Office; EPA report no. EPA-600/9-83-033F. Available from: NTIS, Springfield, VA;
PB85-103471.
March 12, 1990 2-27 DRAFT - DO NOT QUOTE OR CITE
-------
U. S. Environmental Protection Agency. (1984b) Review of the NAAQS for carbon monoxide: reassessment of
scientific and technical information. Research Triangle Park, NC: Office of Air Quality Planning and
Standards; EPA report no. EPA-450/5-84-004. Available from: NTIS, Springfield, VA; PB84-231315.
5 U. S. Environmental Protection Agency. (1988) National air quality and emissions trends report, 1986. Research
Triangle Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA/450/4-88/001.
Available from NTIS, Springfield, VA; PB88-172473/XAB.
U. S. Environmental Protection Agency. (1989a) National air quality and emissions trends report, 1987. Research
10 Triangle Park, NC: Office of Air Quality Planning and Standards; report no. EPA/450/4-89/001.
U. S. Environmental Protection Agency. (1989b) Ozone and carbon monoxide 1986-1988 air quality data.
Environmental News July 27.
15 Van Hoesen, K. B.; Camporesi, E. M.; Moon, R. E.; Hage, M. L.; Piantadosi, C. A. (1989) Should hyperbaric
oxygen be used to treat the pregnant patient for acute carbon monoxide poisoning? A case report and
literature review. JAMA J. Am. Med. Assoc. 261: 1039-1043.
Vogel, J. A.; Gleser, M. A. (1972) Effect of carbon monoxide on oxygen transport during exercise. J. Appl.
20 Physiol. 32: 234-239.
Weiser, P. C.; Morrill, C. G.; Dickey, D. W.; Kurt, T. L.; Cropp, G. J. A. (1978) Effects of low-level carbon
monoxide exposure on the adaptation of healthy young men to aerobic work at an altitude of 1,610
meters. In: Folinsbee, L. J.; Wagner, J. A.; Borgia, J. F.; Drinkwater, B. L.; Gliner, J. A.; Bedi, J. F.,
25 eds. Environmental stress: individual human adaptations. New York, NY: Academic Press, Inc.;
pp. 101-110.
Winneke, G. (1974) Behavioral effects of methylene chloride and carbon monoxide as assessed by sensory and
psychomotor performance. In: Xintaras, C.; Johnson, B. L.; de Groot, I., eds. Behavioral toxicology:
30 early detection of occupational hazards [proceedings of a workshop; June 1973; Cincinnati, OH].
Cincinnati, OH: U. S. Department of Health, Education, and Welfare, National Institute for Occupational
Safety and Health; pp. 130-144; DHEW publication no. (NIOSH) 74-126. Available from: NTIS,
Springfield, VA; PB-259322.
35 Wright, G.; Randell, P.; Shephard, R. J. (1973) Carbon monoxide and driving skills. Arch. Environ. Health
27: 349-354.
March 12, 1990 2-28 DRAFT - DO NOT QUOTE OR CITE
-------
3. PROPERTIES AND PRINCIPLES OF FORMATION
OF CARBON MONOXIDE
3.1 INTRODUCTION
5 Carbon monoxide was first discovered to be a minor constituent of the earth's
atmosphere by Migeotte (1949) in 1948. While taking measurements of the solar spectrum,
he observed a strong absorption band in the infrared region at 4.7 ^m, which he attributed to
CO (Lagemann et al., 1947). On the twin bases of the belief that the solar contribution to
that band was negligible and his observation of a strong day-to-day variability in absorption,
10 Migeotte concluded that an appreciable amount of CO was present in the terrestrial
atmosphere of Columbus, Ohio. In the 1950s many more observations (Benesch et al., 1953;
Faith et al., 1959; Locke and Herzberg, 1953; Migeotte and Neven, 1952; Robbins et al.,
1968; Sie et al., 1976) of CO were made, with measured concentrations ranging from 0.08 to
100 ppm. On the basis of these and other measurements available in 1963, Junge (1963)
15 stated that CO appeared to be the most abundant trace gas, other than carbon dioxide, in the
atmosphere. The studies of Sie et al. (1976) indicated higher mixing ratios near the ground
than in the upper atmosphere, implying a source in the biosphere, but Junge emphasized that
knowledge of the sources and sinks of atmospheric CO was extremely poor. It was not until
the late 1960s that concerted efforts were made to determine the various production and
20 destruction mechanisms for CO in the atmosphere.
Even far from human habitation, carbon monoxide occurs in air at an average
background concentration of 0.05 mg/m3, primarily as a result of natural processes such as
forest fires and the oxidation of methane. Much higher concentrations occur in cities from
technological sources such as automobiles and the production of heat and power. Carbon
25 monoxide emissions are increased when the fuel is burned in an incomplete or inefficient
way. The physical and chemical properties of CO suggest that its atmospheric removal occurs
primarily by reaction of CO with hydroxyl (OH*) radicals.
The remainder of this chapter focuses on the physical properties and formation
principles of CO that contribute to its release into the atmosphere. In Chapter 4, other source
30 and sink estimates as well as the global cycle of CO are described; in Chapter 6, the various
factors that determine the technological emission source strengths are discussed.
February 8, 1990 3-1 DRAFT-DO NOT QUOTE OR CITE
-------
3.2 PHYSICAL PROPERTIES
Carbon monoxide is a tasteless, odorless, colorless diatomic molecule that exists as a
gas in the earth's atmosphere. Radiation in the visible and near ultraviolet regions of the
electromagnetic spectrum is not absorbed by CO, although the molecule does have weak
5 absorption bands between 125 and 155 nm. It absorbs radiation in the infrared region
corresponding to the vibrational excitation of its electronic ground state. Carbon monoxide
has a low electric dipole moment (0.10 debye), short interatomic distance (1.23A), and high
heat of formation from atoms, or bond strength (2072 kJ/mol). These observations suggest
that the molecule is a resonance hybrid of three structures (Perry et al., 1977), all of which
10 contribute nearly equally to the normal ground state. General physical properties of CO are
given in Table 3-1.
3.3 GASEOUS CHEMICAL REACTIONS OF CARBON MONOXIDE
15 In the atmosphere, carbon monoxide reacts with OH* radicals to produce carbon
dioxide and (CO2) hydrogen (H*) atoms.
OH* + CO -> CO2 + H* (3-1)
20 The H' atoms formed in this process react very rapidly with oxygen (OJ to produce
hydroperoxyl radicals (HOj ).
30
H* + O2 (+M) -> HO; (+M) (3-2)
25 The liberated HO; radicals can react with nitric oxide (NO) to form nitrogen dioxide (NO2)
and regenerate OH* radicals.
; + NO -> NO2 + OH* (3-3)
February 8, 1990 3-2 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-1. PHYSICAL PROPERTIES OF CARBON MONOXIDE"
Molecular weight
Critical point
Melting point
Boiling point
Density
at 0°C, 1 atm
at 25°C, 1 atm
Specific gravity relative to air
Solubility in water1"
atO°C
at 20°C
at 25 °C
Explosive limits in air
Fundamental vibration transition
Conversion factors
at 0°C, 1 atm
at25°C, 1 atm
28.01
-140°C at 34.5 atm
-199°C
-191.5°C
1.250g/L
1.145 g/L
0.967
3.54 mL/100 mL (44.3 ppmm)c
2.32 mL/100 mL (29.0 ppmm)
2.14 mL/100 mL (26.8 ppmm)
12.5-74.2%
2,143.3cm-1
CO(X'Zg+, v1 = 1 Ev"0)(4.67
1 mg/m3 = 0.800 ppmd
1 ppm = 1.250 mg/m3
1 mg/m3 = 0.873 ppm
1 ppm = 1.145 mg/m3
" National Research Council (1977).
Volume of carbon monoxide is at 0°C, 1 atm (atmospheric pressure at sea
level = 760 ton).
c Parts per million by mass (ppmm = ftg/g).
d Parts per million by volume (ppm = mg/L).
See glossary of terms and symbols for abbreviations and acronyms.
February 8, 1990
3-3 DRAFT - DO NOT QUOTE OR CITE
-------
The photolysis of NO2 leads to the formation of ozone (O3); hence, CO can contribute to the
production of photochemical smog in the lower troposphere. Other radicals besides OH* also
can react with CO.
5 CO + HO; •* CO2 + OH* (3-4)
CO + NO; -> CO2 + NO2 (3-5)
CO + CH3O; •* product (3-6)
The rates of these reactions, however, are so slow that their contribution to the overall
10 chemistry occurring in the atmosphere is expected to be very slight. Hampson and Garvin
(1978), in their review of chemical kinetics data, recommended a rate constant of < 10"" cm3
molecule"1 s"1 for reaction (3-4). DeMore et al. (1987), based on their analysis of rate data,
suggested a rate constant of <4.0 x 10~19 cm3 molecule"1 s"1 for reaction (3-5) and Heicklen
(1973) recommended a value of 4 x 10~17 for reaction (3-6). In contrast, the rate constant for
15 the CO + OH* reaction is of the order of 10"13 cm3 molecule"1 s"1, a factor of at least 104 to
10* greater than other known reactions between CO and atmospheric constituents. Thus, the
reaction with OH* is the only reaction involving CO that is expected to be of any consequence
in the atmosphere.
The reaction of CO with OH* is one of the most studied of all atmospheric reactions.
20 Table 3-2 summarizes the results obtained in a few of these studies. More complete reviews
of the kinetics of this reaction can be found in Hampson and Garvin (1978), Baulch et al.
(1980), and DeMore et al. (1987). As seen in Table 3-2, the rate constants obtained in the
late 1960s and early 1970s agreed fairly well and led the National Bureau of Standards
(Hampson and Garvin, 1975) to recommend a value of 1.4 x 10~13 cm3 molecule"1 s"1 for the
25 rate constant. At that time there did not appear to be either a substantial temperature or
pressure dependency for this reaction. In the mid 1970s, however, Cox et al. (1976) studied
the reaction at 700 torr using a mixture of N2 and O2 as the diluent gas. They obtained a rate
constant of 2.7 x 10"13 cm3 molecule"1 s"1 and suggested that the reaction might be subject to
a pressure effect. At approximately the same time, Sie et al. (1976) studied the CO + OH"
30 reaction as a function of pressure. When molecular hydrogen was used as the diluent gas,
they found that the rate constant increased from 0.9 x 10"13 cm3 molecule"1 s~' at a pressure
February 8, 1990 3-4 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-2. REPORTED ROOM TEMPERATURE RATE CONSTANTS FOR THE
REACTION OF OH* RADICALS WITH CO
Reference
No observed pressure dependence
Greiner (1969)
Stuhl and Niki (1972)
Westenberg and deHaas (1973)
Smith and Zellner (1973)
Howard and Evenson (1974)
Davis et al. (1974)
Gordon and Mulac (1975)
Atkinson et al. (1976)
Pressure dependence observed*
Cox et al. (1976)
Sie et al. (1976)
Perry et al. (1977)
Chan et al. (1977)
Biermann et al. (1978)
Paraskevopoulos and Irwin (1984)
DeMore (1984)
Hofzumahaus and Stuhl (1984)
Hynes et al. (1986)
Niki et al. (1984)
Hynes et al. (1986)
Pressure
(torr)
100
20
1-3
10-20
0.3-6
20
730
25-650
700
20-700
25-600
100-700
25-750
20-700
200-730
20-700
50-700
700
50-700
Diluent
He
He
He or Ar
He or N2O + H2
He, Ar, or N2
He or N2
Ar
Ar
N2 + O2
H2
SF6
Air
N2
N2
N2
N2
N2
Air
Air
Rate constant x 10 "
(cm3 molecule'1 s ')
1.4 + 0.2
1.3 ± 0.2
1.3
1.4
1.6 ± 0.2
1.6
1.5 + 0.1
1.5 ± 0.2
2.7 ± 0.2
3.3 + 0.2
3.4 ± 0.3
3.0 + 0.2
2.8 + 0.3
2.2 + 0.1
2.1 ± 0.4
2.3 + 0.2
2.1 ±0.2
2.4 ± 0.1
2.3 + 0.2
* Rate constants listed are the values obtained at the highest pressure used in each study.
See glossary of terms and symbols for abbreviations and acronyms.
February 8, 1990
3-5 DRAFT - DO NOT QUOTE OR CITE
-------
of 20 torr to 3.3 X 10~13 cm3 molecule"1 s"1 at 700 torr. When argon was used as the diluent
gas, however, the rate of the reaction was found to be insensitive to pressure changes.
Subsequent research has supported this finding. In general, it appears that there is no
pressure effect if noble gases (for example, helium or argon) are used as the carrier gas, but
5 when other gases are used, such as N2 or O2, which are more representative of the
atmosphere, the CO + OH* reaction rate exhibits a strong pressure dependency. Table 3-2
summarizes the more recent pressure dependency studies. In all cases, the rate constant listed
for the pressure dependent studies is the value obtained at the highest pressure used in each
study. Excellent agreement is noted for the studies conducted in 1984 and later years.
10 The National Aeronautics and Space Administration (NASA) Data Evaluation Panel
(DeMore, 1987) recently examined the kinetics data for the CO + OH* reaction. They first
analyzed all of the direct, low-pressure determinations to derive a zero pressure value for the
rate constant. They then performed a weighted squares analysis of all the pressure-dependent
data obtained since 1984 and fitted it to the expression
15
K = K° x (1 + CPalm) (3-7)
where
K° = zero pressure value for the rate constant,
20 c = constant,
P = pressure in atmospheres.
They found that the data were best fit using the expression
25 K = (1.50 ± 0.45) X 10'13 (1 + 0.6Patm ) (3-8)
At a pressure of 760 torr, this corresponds to a rate constant of 2.4 + 0.7 x 10~13 cm3
molecule"1 s"1, independent of temperature. Thus, the rate constant at atmospheric pressure is
substantially larger than the value that previously had been assumed for this reaction. The
30 larger value has led to important changes in our understanding of the global CO cycle. This
point is discussed in more detail in Chapter 4.
February 8, 1990 3-6 DRAFT-DO NOT QUOTE OR CITE
-------
3.4 PRINCIPLES OF FORMATION BY SOURCE CATEGORY
Carbon monoxide is produced at the earth's surface during the combustion of fuels and
in the atmosphere during the oxidation of anthropogenic and biogenic hydrocarbons. The role
of man-made and natural hydrocarbons in CO production is discussed in Chapter 6 and only
5 production of CO from combustion sources is addressed here.
The burning of any carbonaceous fuel produces two primary products: CO2 and CO.
The production of CO2 predominates when the air or oxygen supply is in excess of the
stoichiometric needs for complete combustion. If burning occurs under fuel-rich conditions,
with less air or oxygen than is needed, CO will be produced in abundance. Most of the CO
10 and CO2 formed in past years simply was emitted into the atmosphere.
In recent years, concerted efforts have been made to reduce concentrations of potentially
harmful materials in ambient air. Today, the CO in urban air originates almost entirely from
local combustion processes. The background concentration of CO contributes less than
0.23 mg/m3 (0.20 ppm) to the ambient air concentration at a given urban location. As a
15 result of natural processes such as forest fires, oxidation of methane, and biological activity.
the background level of CO is estimated to be about 0.05 mg/m3 (0.04 ppm) (Seiler and
Junge, 1970). See Chapter 4 for a discussion of global background concentrations.
Considerable effort has been made to reduce emissions of CO and other pollutants to the
atmosphere. Since the automobile engine is recognized to be the major source of CO in most
20 urban areas, special attention is given to the control of automotive emissions. Generally the
approach has been technological: reduction of CO emissions to the atmosphere either by
improving the efficiency of the combustion processes, thereby increasing the yield of CO2 and
decreasing the yield of CO; or by applying secondary catalytic combustion reactors to the
waste gas stream to convert CO to CO2.
25 The development and application of control technology to reduce emissions of CO from
combustion processes generally have been successful and are continuing to receive deserved
attention. The reduction of CO emissions from 7.0 to 3.4 g/mi, scheduled for the 1981
model year, was delayed two years, reflecting in part the apparent difficulty encountered by
the automobile industry in developing and supplying the required control technology. The
30 CO emission limit for light-duty vehicles (LDVs) at low altitude has been 3.4 g/mi since
1983; since 1984, this limit applied to LDVs at all altitudes.
February 8, 1990 3-7 DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-3 shows the automobile emissions control schedules that have resulted from the
1970 Clean Air Act and subsequent amendments, notably the 1977 and 1981 CAA
Amendments.
The problems encountered in mass producing and marketing effective control technology
5 for automobile engines are complex because a number of simultaneous requirements are
involved (i.e., control of multiple air pollutants, fuel economy and efficiency, durability and
quality control of components, and maintenance). Emission factor program testing conducted
by EPA during the 1980s indicates that the durability of emission control systems continues to
present a problem for in-use vehicles intended to comply with the 1983 and later requirement
10 that CO emissions be limited to not more than 3.4 g/mi through the 50,000-mile, useful-life
compliance period. (CO emissions are less than 3.4 g/mi for new, low-mileage automobiles.)
The following subsections present a brief discussion of the general principles and
mechanisms of CO formation and control of emissions associated with the many combustion
processes. The processes commonly are classified in two broad types, mobile sources and
15 stationary sources, because this division generally does separate distinct types of major
combustion devices. Control techniques for CO emissions from mobile and stationary sources
are detailed in two National Air Pollution Control Administration publications (National Air
Pollution Control Administration, 1970a,b).
20 3.4.1 General Combustion Processes
Incomplete combustion of carbon-containing compounds creates varying amounts of CO.
The chemical and physical processes that occur during combustion are complex because they
depend not only on the type of carbon compound reacting with oxygen but also on the
conditions existing in the combustion chamber (Mellor, 1972; Pauling, 1960). Despite the
25 complexity of the combustion process, certain general principles regarding the formation of
CO from the combustion of hydrocarbon fuels are accepted widely.
Gaseous or liquid hydrocarbon fuel reacts with molecular oxygen in a chain of reactions
that result in CO. Carbon monoxide then reacts with OH* radicals to form CO2. The second
reaction is approximately ten times slower than the first. In coal combustion, too, the
30 reaction of carbon and oxygen to form CO is one of the primary reactions, and a large
February 8, 1990 3-8 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-3. SUMMARY OF UGHT-DUTY VEHICLE (LDV) EMISSIONS STANDARDS1
Year
Prior to
controls
1968-69
1970
1971
1972
1973-74
1975-76
1977*
1978-79
1980
1981
198210
198310
1984-8612
1987 &
later12
Teat
Procedure2
7-raode
7-mode
CVS-75
7-mode
50-100 CID
101-140 CID
> 140 OD
7-mode
7-mode
CVS-72
CVS-72
CVS-75
CVS-75
CVS-75
CVS-75
CVS-75
CVS-75
CVS-75
CVS-75
CVS-75
Hydro-
carbom
850 ppm
11 g/mi
8.8 g/mi
410 ppm
350 ppm
275 ppm
2.2 g/mi
2.2 g/mi
3.4 g/mi
3.4 g/mi
1.5 g/mi
1.5 g/mi
1.5 g/mi
0.41 g/mi
0.41 g/mi
0.41 g/mi
(0.57)
0.41 g/mi
(0.57)
0.41 g/mi
0.42 g/mi
(0.41)
Carbon Oxides of
Monoxide Nitrogen
Partial- Evaporative
Isles3 Hydrocarbons4
Gtsofine-fociedLDVs
3.4 % 1000 ppm
80 g/mi 4 g/mi
87.0 g/mi 3. 6 g/mi
2.3%
2.0%
1.5*
23 g/mi
23 g/mi
39 g/mi
39 g/mi
Ganfioe-fiiekd and dioei LDV.
15 g/mi
15 g/mi
15 g/mi
7.0 g/mi
3.4 g/mi7
3.4 g/mi7
(7.8)"
3.4 g/mi
(7.8)
3.4 g/mi
3.4 g/mi
(3.4)
-
-
-
-
-
3.1 g/mi
2.0 g/mi
2.0 g/mi
2.0 g/mi
1.0 g/mi8-9
1.0 g/mi8-9
(1.0)8-9
1.0 g/mi8
(l.O)8
1.0 g/mi8
1.0 g/mi
(1.0)
-
-
6.0 g/test3
2.0 g/test
2.0 g/test
2.0 g/test
2.0 g/test
6.0 g/test
6.0 g/test
2.0 g/test
0.60 g/mi 2.0 g/test
(-) (2.6)
0.60 g/mi 2.0 g/test
(0.60) (2-6)
0.60 g/mi 2.0 g/test
0.20 g/mi13 2.0 g/test
(0.20)13 (2-0)
February 8, 1990
3-9
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-3. (cont'd) SUMMARY OF LIGHT-DUTY VEHICLE (LDV) EMISSIONS STANDARDS
1 Standards do not apply to LDVs with engines less than SO CID from 1968 through 1974.
2 Different test procedures, which vary in stringency, have been used since the early years of emission control. The appearance that the standards were relaxed
from 1971 to 1972 is incorrect; the 1972 standards actually are more stringent because of the 1972 test procedure.
3 Applies only to diesel LDVs.
4 Evaporative emissions determined by carbon-trap method through 1977, SHED procedure beginning in 1978. Applies only to gasoline-fueled LDVs.
5 Evaporative standard does not apply to off-road utility LDVs for 1971.
* LDVs sold in specified high-altitude counties are required to meet these standards at high altitude.
7 Carbon monoxide standard is waived to 7.0 g/mi for 1981-82 for certain LDVs.
8 Oxides of nitrogen standard is waived to 7.0 g/mi for 1981-82 for certain LDVs.
9 Oxides of nitrogen standard for 1981-82 is 2.0 g/mi for American Motors Corporation LDVs.
10 Standards in parentheses apply to LDVs sold in specified high-altitude counties.
11 LDVs eligible for a carbon monoxide waiver to 7.0 g/mi at low altitude are eligible for a waiver to 11 g/mi at high altitude.
12 The same numerical standards apply to LDVs sold in high-altitude areas. Exemptions from compliance at high-altitude are provided for qualifying low-
performance vehicles.
13 Emissions averaging may be used to meet this standard, provided that emissions from LDVs produced for sale in California or in designated high-altitude areas
may be averaged only within each of those areas.
CID - cubic inch displacement
CVS-72 - constant volume sample cold start test
CVS-75 - constant volume sample test including cold and hot starts
g/mi - grams per mile
ppm - parts per million
7-mode - 137 second driving cycle test
SHED - sealed housing for evaporative determination
Source: AP-42 Supplement (1989); (in press)
February 8, 1990 3-10 DRAFT-DO NOT QUOTE OR CITE
-------
fraction of carbon atoms go through the monoxide form. Again, the reaction of monoxide to
dioxide is much slower.
Four basic variables control the concentration of CO in the combustion of all
hydrocarbon gases. These are (1) O2 concentration, (2) flame temperature, (3) gas residence
5 time at high temperatures, and (4) combustion chamber turbulence. Oxygen concentration
affects the formation of both CO and CO2 because O2 is required in the initial reactions with
the fuel molecule and in the formation of the OH* radical. As the availability of O2 increases,
more complete conversion of monoxide to dioxide results. Flame and gas temperatures affect
both the formation of monoxide and the conversion of monoxide to dioxide because both
10 reaction rates increase exponentially with increasing temperature. Also, the OH* radical
concentration in the combustion chamber is very temperature-dependent. The conversion of
CO to CO2 also is enhanced by longer residence time, because this is a relatively slow
reaction in comparison with CO formation. Increased gas turbulence in the combustion zones
increases the actual reaction rates by increasing the mixing of the reactants and assisting the
15 relatively slower gaseous diffusion process, thereby resulting in more complete combustion.
3.4.2 Combustion Engines
Mobile Combustion Engines—Most mobile sources of CO are internal combustion engines of
two types: (1) gasoline-fueled, spark-ignition, reciprocating engines (carbureted or fuel-
20 injected); and (2) diesel-fueled reciprocating engines. The CO emitted from any given engine
is the product of the following factors: (1) the concentration of CO in the exhaust gases,
(2) the flow rate of exhaust gases, and (3) the duration of operation.
InternaLCQrnbustiQn Engines (Gasoline-Fueled. Spark-Ignition EnginesV-Exhaust
concentrations of CO increase with lower (richer) air-to-fuel (A/F) ratios, and decrease with
25 higher (leaner) A/F ratios, but remain relatively constant with ratios above the stoichiometric
ratio of about 15:1 (Hagen and Holiday, 1964). The behavior of gasoline automobile engines
before and after the installation of pollutant control devices differs considerably. Depending
on the mode of driving, the average uncontrolled engine operates at A/F ratios ranging from
about 11:1 to a point slightly above the stoichiometric ratio. During the idling mode, at low
30 speeds with light load (such as low-speed cruise), during the full-open throttle mode until
speed picks up, and during deceleration, the A/F ratio is low in uncontrolled cars and CO
February 8, 1990 3-11 DRAFT-DO NOT QUOTE OR CITE
-------
emissions are high. At higher speed cruise and during moderate acceleration, the reverse is
true. Cars with exhaust controls generally remain much closer to stoichiometric A/F ratios in
all modes, and thus the CO emissions are kept lower. The relationship between CO
concentrations in engine exhaust and A/F ratios is shown in Figure 3-1. The exhaust flow
5 rate increases with increasing engine power output.
The decrease in available oxygen with increasing altitude has the effect of enriching the
A/F mixture and increasing CO emissions from carbureted engines. Fuel-injected gasoline
engines, which predominate in the vehicle fleet today, have more closely controlled A/F
ratios and are designed and certified to comply with applicable emission standards regardless
10 of elevation (U.S. Environmental Protection Agency, 1983).
Correlations between total emissions of CO in grams per vehicle mile and average route
speed show a decrease in emissions with increasing average speed (Simonaitis and Heicklen,
1972; Stuhl and Niki, 1972; U.S. Environmental Protection Agency, 1985). During low
speed conditions (below 32 km/h or 20 mi/h average route speed), the greater emissions per
15 unit of distance traveled are attributable to (1) an increased frequency of acceleration,
deceleration, and idling encountered in heavy traffic; and (2) the consequent increase in the
operating time per mile driven.
The CO and the unburned hydrocarbon exhaust emissions from an uncontrolled engine
result from incomplete combustion of the fuel-air mixture. Emission control on new vehicles
20 is being achieved by engine modifications, improvements in engine design, and changes in
engine operating conditions. Substantial reductions in CO and other pollutant emissions result
from consideration of design and operating factors such as leaner, uniform mixing of fuel and
air during carburetion, controlled heating of intake air, increased idle speed, retarded spark
timing, improved cylinder head design, exhaust thermal reactors, oxidizing and reducing
25 catalysts, secondary air systems, exhaust recycle systems, electronic fuel injection, A/F ratio
feedback controls, and modified ignition systems (National Research Council, 1973).
Internal Combustion Engines (Diesel Engines)—Diesel engines in use are primarily the heavy-
duty type that power trucks and buses. Diesel engines allow more complete combustion and
30 use less volatile fuels than do spark-ignition engines. The operating principles are
February 8, 1990 3-12 DRAFT-DO NOT QUOTE OR CITE
-------
£ <
^ e
j
UJ
Ik
I
a
5
o> eo
% »(oiu '3QIXONOW N09HVO
Figure 3-1. Effect of air-fuel ratio on exhaust gas carbon monoxide concentrations from three
gasoline-fueled test engines.
Source: Hagen and Holiday, 1964.
February 8, 1990
3-13 DRAFT-DO NOT QUOTE OR CITE
-------
significantly different from those of the gasoline engine. In diesel combustion, CO
concentrations in the exhaust are relatively low because high temperature and large excesses
of oxygen are involved in normal operation. The exhaust emissions from diesel engines have
the same general composition as gasoline engine emissions, though the concentrations of
5 different pollutants vary considerably. For example, the diesel emits larger quantities of
nitrogen oxides (NOJ and polycyclic organic particulates than gasoline engines; it emits less
CO.
Stationary Combustion Sources (Steam Boilers)--This section refers to fuel-burning
10 installations such as coal-, gas-, or oil-fired heating or power generating plants (external
combustion boilers).
In these combustion systems, the formation of CO is lowest at a ratio near or slightly
above the stoichiometric ratio of air to fuel. At lower than stoichiometric A/F ratios, high
CO concentrations reflect the relatively low O2 concentration and the possibility of poor
15 reactant mixing from low turbulence. These two factors can increase emissions even though
flame temperatures and residence time are high. At higher than stoichiometric A/F ratios,
increased CO emissions result from decreased flame temperatures and shorter residence time.
These two factors remain predominant even when O2 concentrations and turbulence increase.
Minimal CO emissions and maximum thermal efficiency, therefore, require combustor
20 designs that provide high turbulence, sufficient residence time, high temperatures, and near
stoichiometric A/F ratios. Combustor design dictates the actual approach to that minimum.
The measurement of CO in effluent gas is used as an indication of improper and
inefficient operating practice for any given combustor, or of inefficient combustion.
25 3.4.3 Other Sources
There are numerous industrial activities that result in the emission of CO at one or more
stages of the process (Walsh and Nussbaum, 1978). Manufacturing pig iron can produce as
much as 700 to 1,050 kg CO/metric ton of pig iron. Other methods of producing iron and
steel can produce CO at a rate of 9 to 118.5 kg/metric ton. However, most of the CO
30 generated is normally recovered and used as fuel. Conditions such as "slips", abrupt
collapses of cavities in the coke-ore mixture, can cause instantaneous emissions of CO that
February 8, 1990 3-14 DRAFT-DO NOT QUOTE OR CITE
-------
temporarily exceed the capacity of the control equipment. Slips have been reduced greatly
with modern equipment. Grey-iron foundries can produce 72.5 kg CO/metric ton of product,
but an efficient afterburner can reduce the CO emission to 4.5 kg/metric ton.
Charcoal production results in CO emissions of 160 kg/metric ton with or without the
5 installation of chemical recovery equipment. Emissions from carbon black manufacture can
range from 5 to 3,200 kg CO/metric ton depending on the efficiency and quality of the
emission control systems.
Some chemical processes such as phthalic anhydride production give off as little as 6 kg
CO/metric ton with proper controls or as much as 200 kg CO/metric ton if no controls are
10 installed. There are numerous other chemical processes that produce relatively small CO
emissions per metric ton of product, such as sulfate pulping for paper at 1 to 30 kg CO/metric
ton; lime manufacturing normally runs 1 to 4 kg CO/metric ton; and CO from adipic acid
production is zero or slight with proper controls. Other industrial chemical processes that
cause CO emissions are the manufacture of terephthalic acid and the synthesis of methanol
15 and higher alcohols. As a rule, most industries find it economically desirable to install
suitable controls to reduce CO emissions.
Even though some of these CO emission rates seem excessively high, they are, in fact,
only a small part of the total pollutant load. Mention of these industries is made to emphasize
the concern for localized pollution problems when accidents occur or proper controls are not
20 used.
While the estimated CO emissions resulting from forest wildfires in the United States for
1971 was 4 X 106 metric tons, the estimated total industrial process CO emission of the U.S.
for 1971 was 10.3 X 10* metric tons.
25
3.5 REFERENCES
Atkinson, R.; Perry, R. A.; Pitts, J. N., Jr. (1976) Kinetics of the reactions of OH radicals with CO and N2O.
Chem. Phys. Lett. 44: 204-208.
30
Baulch, D. L.; Cox, R. A.; Hampson, R. F., Jr.; Kerr, J. A.; Troe, J.; Watson, R. T., eds. (1980) Evaluated
inetic and photochemical data for atmospheric chemistry. J. Phys. Chem. Ref. Data 9: 295-471.
February 8, 1990 3-15 DRAFT-DO NOT QUOTE OR CITE
-------
Benesch, W.; Migeotte, M.; Neven, L. (1953) Investigations of atmospheric CO at the Jungfraujoch. J. Opt. Soc.
Am. 43: 1119-1123.
Biermann, H. W.; Zetsch, C.; Stuhl, F. (1978) On the pressure dependence of the reaction of HO with CO. Ber.
5 Bunsenges. Phys. Chem. 82: 633-639.
Chan, W. H.; Uselman, W. M.; Calvert, J. G.; Shaw, J. H. (1977) The pressure dependence of the rate constant
for the reaction: OH + CO — > H + CO2. Chem. Phys. Lett. 45: 240-244.
10 Cox, R. A.; Derwent, R. G.; Holt, P. M. (1976) Relative rate constants for the reactions of OH radicals with
H2, CH4, CO, NO, and HONO at atmospheric pressure and 296 K. J. Chem. Soc. Faraday Trans. 1
72: 2031-2043.
Davis, D. D.; Fischer, S.; Schiff, R. (1974) Flash photolysis-resonance fluorescence kinetics study: temperature
15 dependence of the reactions OH + CO = CO2 + H and OH + CH4 = H2O + CH3. J. Chem. Phys.
61: 2213-2219.
DeMore, W. B. (1984) Rate constant for the OH + CO reaction: pressure dependence and the effect of oxygen.
Int. J. Chem. Kinet. 16: 1187-1900.
20
DeMore, W. B.; Molina, M. J.; Sander, S. P.; Golden, D. M.; Hampson, R. F.; Kurylo, M. J.; Howard, C. J.;
Ravishankara, A. R. (1987) Chemical kinetics and photochemical data for use in stratospheric modeling:
evaluation number 8. Washington, DC: National Aeronautics and Space Administration;; report no.
NASA-CR-182919. Available from: NTIS, Springfield, VA; N88-24012.
25
Faith, W. L.; Renzetti, N. A.; Rogers, L. H. (1959) Fifth technical progress report. San Marino, CA: Air
Pollution Foundation.
Gordon, S.; Mulac, W. A. (1975) Reaction of the OH(^ II) radical produced by the pulse radiolysis of water
30 vapor. In: Benson, S. W.; Golden, D. M.; Barker, J. R., eds. Chemical kinetics data for the upper and
lower atmosphere: proceedings of the symposium; September 1974; Warrenton, VA. Int. J. Chem. Kinet.
Symp. (1): 289-299.
Greiner, N. R. (1969) Hydroxyl radical kinetics by kinetic spectroscopy: V. reactions with H2 and CO in the
35 range 300-500° K. J. Chem. Phys. 51: 5049-5051.
Hagen, D. F.; Holiday, G. W. (1964) The effects of engine operating and design variables on exhaust emissions.
In: Vehicle emissions (selected SAE papers). New York, NY: Society of Automotive Engineers;
pp. 206-223. (Technical progress series volume 6).
40
Hampson, R. F., Jr.; Garvin, D., eds. (1975) Chemical kinetic and photochemical data for modelling
atmospheric chemistry. Washington, DC: National Bureau of Standards; NBS technical note 866.
Hampson, R. F., Jr.; Garvin, D., eds. (1978) Reaction rate and pyhotochemical data for atmospheric chemistry -
45 1977. Washington, DC: National Bureau of Standards; NBS special publication 513.
Heicklen, J. (1973) Photochemical and rate data for methyl nitrite, methoxy and methylperoxy. In: Garvin, D.,
ed. Chemical kinetics data survey V. Washington, DC: National Bureau of Standards; NBSIR 73-206.
50 Hofzumahaus, A.; Stuhl, F. (1984) Rate constant of the reaction HO + CO in the presence of N2 and O2. Ber.
Bunsenges. Phys. Chem. 88: 557-561.
Howard, C. J.; Evenson, K. M. (1974) Laser magnetic resonance study of the gas phase reactions of OH with
CO, NO, and NO2. J. Chem. Phys. 61: 1943-1952.
February 8, 1990 3-16 DRAFT-DO NOT QUOTE OR CITE
-------
Hynes, A. J.; Wine, P. H.; Ravishankara, A. R. (1986) Kinetics of the OH + CO reaction under atmospheric
conditions. JGR J. Geophys. Res. 91: 11,815-11,820.
Junge, C. E. (1963) Air chemistry and radioactivity: v. 4. New York, NY: Academic Press. (Mieghem, J.;
5 Hales, A. L., eds. International geophysics series).
Lagemann, R. T.; Nielsen, A. H.; Dickey, F. P. (1947) The infra-red spectrum and molecular constants of
C12O16 and C13O16. Phys. Rev. 72: 284-289.
10 Locke, J. L.; Herzberg, L. (1953) The absorption due to carbon monoxide in the infrared solar spectrum. Can. J.
Phys. 31: 504-516.
Mellor, A. (1972) Current kinetic modeling techniques for continuous flow combustors. In: Cornelius, W.;
Agnew, W. G., eds. Emissions from continuous combustion systems: proceedings of the symposium;
15 September 1971; Warren, MI. New York, NY: Plenum Press; pp. 23-53.
Migeotte, M. V. (1949) The fundamental band of carbon monoxide at 4.7 jw, in the solar spectrum. Phys. Rev.
75: 1108-1109.
20 Migeotte, M.; Neven, L. (1952) Recents progres dans 1'observation du spectre infra-rouge du soleil a la station
scientifique du Jungfraujoch (Suisse) [Recent progress in observing the infrared solar spectrum at the
scientific station at Jungfraujoch, Switzerland]. Mem. Soc. R. Sci. Liege 12: 165-178.
National Air Pollution Control Administration. (1970a) Control techniques for carbon monoxide emissions from
25 stationary sources. Washington, DC: U. S. Department of Health, Education, and Welfare, Public Health
Service; National Air Pollution Control Administration publication no. AP-65. Available from: NTIS,
Springfield, VA; PB-190263.
National Air Pollution Control Administration. (1970b) Control techniques for carbon monoxide, nitrogen oxide,
30 and hydrocarbon emissions from mobile sources. Washington, DC: U. S. Department of Health,
Education, and Welfare; publication no. AP-66.
National Research Council. (1973) Automotive spark ignition engine emission control systems to meet the
requirements of the 1970 clean air amendments. Washington, DC: National Academy of Sciences.
35 Available from: NTIS, Springfield, VA; PB-224862.
National Research Council. (1977) Carbon monoxide. Washington, DC: National Academy of Sciences. (Medical
and biologic effects of environmental pollutants).
40 Niki, H.; Maker, P. D.; Savage, C. M.; Breitenbach, L. P. (1984) Fourier transform infrared spectroscopic
study of the kinetics for the HO radical reaction of 13C"O and 12C18O. J. Phys. Chem. 88: 2116-2119.
Paraskevopoulos, G.; Invin, R. S. (1984) The pressure dependence of the rate constant of the reaction of OH
radicals with CO. J. Chem. Phys. 80: 259-266.
45
Pauling, L. (1960) The nature of the chemical bond and the structure of molecules and crystals: an introduction
to modem structural chemistry. 3rd ed. Ithaca, NY: Cornell University Press; pp. 194-195.
Perry, R. A.; Atkinson, R.; Pitts, J. N., Jr. (1977) Kinetics of the reactions of OH radicals with C2H2 and CO.
50 J. Chem. Phys. 67: 5577-5584.
Robbins, R. C.; Borg, K. M.; Robinson, E. (1968) Carbon monoxide in the atmosphere. J. Air Pollut. Control
Assoc. 18: 106-110.
February 8, 1990 3-17 DRAFT-DO NOT QUOTE OR CITE
-------
Seiler, W.; Junge, C. (1970) Carbon monoxide in the atmosphere. JGR J. Geophys. Res. 75: 2217-2226.
Sie, B. K. T.; Simonaitis, R.; Heicklen, J. (1976) The reaction of OH with CO. Int. J. Chem. Kinet. 8: 85-98.
5 Simonaitis, R.; Heicklen, J. (1972) Kinetics and mechanism of the reaction of O(jP) with carbon monoxide. J.
Chem. Phys. 56: 2004-2011.
Smith, I. W. M.; Zellner, R. (1973) Rate measurements of reactions of OH by resonance absorption: part 2.
reactions of OH with CO, C2H4, and C2H2. J. Chem. Soc. Faraday Trans. 2 69: 1617-1627.
10
Stuhl, F.; Niki, H. (1972) Pulsed vacuum-uv photochemical study of reactions of OH with H2, D2, and CO using
a resonance-fluorescent detection method. J. Chem. Phys. 57: 3671-3677.
U. S. Code. (1970-1981) Clean Air Act as amended by PL 91-604, 95-95, 97-23. U. S. C. 42: §7401-7626.
15
U. S. Environmental Protection Agency. (1983) Controlling emissions from light-duty vehicles at higher
elevations. Ann Arbor, MI: Office of Mobile Sources; EPA report no. EPA-460/3-83-001. Available
from: NTIS, Springfield, VA; PB83-204883.
20 U. S. Environmental Protection Agency. (1985) Compilation of air pollutant emission factors: v. 1, stationary
point and the area sources, v. 2, mobile sources. 4th ed. Research Triangle Park, NC: Office of Air
Quality Planning and Standards; EPA report nos. AP-42-ED-4-VOL-1 and AP-42-ED-4-VOL-2.
Available from: NTIS, Springfield, VA; PB86-124906 and PB87-205266.
25 U. S. Environmental Protection Agency. (1989) Supplement to compilation of air pollutant emission factors.
Research Triangle Park, NC: Office of Air Quality Planning Standards; EPA report no. AP-42 (in press).
Walsh, M. P.; Nussbaum, B. D. (1978) Who's responsible for emissions after 50,000 miles? Automot. Eng.
86: 32-35.
30
Westenberg, A. A.; deHaas, N. (1973) Rates of CO + OH and H2 + OH over an extended temperature range. J.
Chem. Phys. 58: 4061-4065.
February 8, 1990 3-18 DRAFT-DO NOT QUOTE OR CITE
-------
4. THE GLOBAL CYCLE OF CARBON MONOXIDE:
TRENDS AND MASS BALANCE
5 4.1 INTRODUCTION
In the troposphere CO may control the removal, and therefore the concentrations, of
OH" radicals (Crutzen, 1974; Khalil and Rasmussen, 1985; Levine et al., 1985; Sze, 1977;
Thompson and Cicerone, 1986). The chemical reactions of CO also may produce substantial
amounts of 03 in the troposphere (Conrad and Seiler, 1982; Fishman and Crutzen, 1978;
10 Fishman et al., 1980; Fishman and Seiler, 1983; Seiler and Fishman, 1981). If the
concentrations of CO increase, O3 may increase; concomitantly, OH* may be depleted thus
affecting the global cycles of many natural and anthropogenic trace gases that are removed
from the atmosphere by reacting with OH*. Therefore increasing CO may indirectly affect
the global climate and contribute to widespread changes in atmospheric chemistry.
15 The purpose of this chapter is to review the present understanding of the sources and
sinks of CO and the resulting global distributions and trends. The first section is a review of
the global sources and sinks and the estimated atmospheric lifetime of CO. The next section
deals with the global distribution of CO resulting from the sources and sinks, including
variations with seasons, altitude, and latitude. Next, there is an analysis of the current
20 evidence for global trends that reflect an imbalance of the sources and sinks probably caused
by steadily increasing emissions from anthropogenic sources. The chapter is concluded with a
summary.
25 4.2 GLOBAL SOURCES, SINKS, AND LIFETIME
The mass balance of a trace gas in the atmosphere can be described as a balance
between the rate of change of the global burden added to the annual rate of loss on the one
side and balanced by the global emissions on the other side (d concentration/dt + loss rate =
source emissions). In steady state the atmospheric lifetime (T) is the ratio of the global
30 burden to the loss rate. The global burden is the total number of molecules of a trace gas in
the atmosphere or its total mass. The concentration of a trace gas can vary (dC/dt is not 0)
March 5, 1990 4-1 DRAFT-DO NOT QUOTE OR CITE
-------
when either the loss rate or the emissions vary cyclically in time representing seasonal
variations or vary over a long time often representing trends of human industrial activities and
increasing population. For CO both types of trends exist. There are large seasonal cycles
mostly driven by seasonal variations in the loss rate but also affected by seasonal variations of
5 emissions, and there are indications of long-term trends probably caused by increasing
anthropogenic emissions.
It appears that the largest sources of CO in the global atmosphere are combustion
processes and the oxidation of hydrocarbons. CO is produced in the atmosphere by reactions
of OH* with methane (CH4) and other hydrocarbons, both man-made and natural, and also
10 from the reactions of alkenes with O3 and of isoprene and terpenes with both OH* and O3.
Most of the CO is removed from the atmosphere by reacting with tropospheric OH* radicals.
4.2.1 Sources
Carbon monoxide comes from both natural and anthropogenic processes. About half of
15 the CO is released at the earth's surface while the rest is produced in the atmosphere. Many
papers on the global sources of CO have been published over the la^t 15 years; whether most
of the CO in the atmosphere is from human activities or from natural processes has been
debated for nearly as long. Before 1970 it was believed that CO in the troposphere was
almost all man-made (Jaffe, 1968, 1973). Later, based on the theory that oxidation of CH4
20 produces large amounts of CO, it was suggested that much of the CO in the non-urban
troposphere was of natural origin (Levy, 1971, 1973; McConnell et al., 1971; National
Research Council, 1977; Weinstock and Niki, 1972; Wofsy, 1976; Wofsy et al., 1972).
However, this view was controversial (Newell, 1977; Stevens et al., 1972). At that time, it
commonly was believed that CH4 came from natural processes, although the existing
25 tabulation of the sources suggested otherwise (Ehhalt and Schmidt, 1978). Even now the
source of CO from the oxidation of CH4 often is regarded as natural as opposed to the direct
emissions of CO from combustion processes. However, there is good evidence that about
half the CH4 in the atmosphere is from human activities, particularly rice paddies, cattle,
urban areas, landfills, and other sources (Khalil and Rasmussen, 1983). Therefore, regardless
30 of how much CO is estimated to come from the oxidation of CH4 about half of it could be
considered to come indirectly from anthropogenic activities. In recent years the estimates of
March 5, 1990 4-2 DRAFT-DO NOT QUOTE OR CITE
-------
the average level of OH* radicals have been revised downward so that the production of CO
from CH4 and other hydrocarbons no longer is thought to be the dominant source (Hameed
and Stewart, 1979; Logan et al., 1981; Pinto et al., 1983).
The recent budgets that take into account previously published data, suggest that human
5 activities are responsible for about 60% of the CO in the non-urban troposphere while natural
processes account for the remaining 40%. It also appears that combustion processes directly
produce about 40% of the annual emissions of CO (Jaffe, 1968, 1973; Robinson and Robbins,
1969, 1970; Swinnerton et al., 1971), while oxidation of hydrocarbons make up most of the
remainder (Greenberg et al., 1985; Hanst et al., 1980; Rasmussen and Went, 1965; Went,
10 1960, 1966; Zimmerman et al., 1978) (about 50%) along with other sources such as the
oceans (Bauer et al., 1980; DeMore et al., 1985; Lamontagne et al., 1971; Logan et al.,
1981; Linnenbom et al., 1973; Liss and Slater, 1974; National Research Council, 1977;
Seiler, 1974; Seiler and Junge, 1970; Seller and Schmidt, 1974; Swinnerton et al., 1969,
1974) and vegetation (Bauer et al., 1980; Bidwell and Fraser, 1972; DeMore et al., 1985;
15 Krall and Tolbert, 1957; Logan et al., 1981; National Research Council, 1977; Seiler, 1974;
Seiler and Giehl, 1977; Seiler et al., 1978; Seiler and Junge, 1970; Siegel et al., 1962;
Wilks, 1959). Some of the hydrocarbons that eventually end up as CO also are produced by
combustion processes constituting an indirect source of CO from combustion. These
conclusions are summarized in Table 4-1 which is adapted from the 1981 budget of Logan
20 et al. (1981) in which most of the previous work was incorporated (Logan et al., 1981; World
Meteorological Organization, 1986), including the CO budget of Seiler (1974). The total
emissions of CO are about 2,600 Tg/year. Other budgets by Volz et al. (1981) and by Seiler
and Conrad (1987) are reviewed by Warneck (1988). Global emissions between 2,000 and
3,000 Tg/year are consistent with these budgets.
25
4.2.2 Sinks
It is believed that reaction with OH* radicals is the major sink for removing CO from
the atmosphere. The cycle of OH* itself cannot be uncoupled from the cycles of CO, CH4,
water (H2O), and O3. In the troposphere OH* radicals are produced by the photolysis of O3
30 (hu + O3 —> O('D) + O2) followed by the reaction of the excited oxygen atoms with H2O
March 5, 1990 4-3 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 4-1. SOURCES OF CARBON MONOXIDE
(Teragrams per year)
10
15
20
25
30
35
40
45
50
I.
II.
in.
Directly from Combustion
Fossil Fuels
Forest Clearing
Savanna Burning
Wood Burning
Forest Fires
Oxidation of Hydrocarbons
Methane
Non-methane HCs
Other Sources
Plants
Oceans
TOTALS (Rounded)
Notes:
1.
Table adapted from Logan el
Anthropogenic
500
400
200
50
300
90
-
1,500
t al. (1981) and revisions reporte
Natural Global
500
400
200
50
30 30
300 600
600 690
100 100
40 40
1,100 2,600
d by World Meteorological Organization (198
Range
400 -
200 -
100 -
25 -
10 -
400 -
300 -
50 -
20 -
2,000 -
6). All estimates
1,000
800
400
150
50
1,000
1,400
200
80
3,000
are in
Tg/year of CO. Tg/year = megatons/year = 10 g/year.
2. All estimates are expressed to one significant figure. The sums are rounded to two significant digits.
3. Half the production of CO from the oxidation of CH4 is attributed to anthropogenic sources and the other half to natural sources
based on the budget of CH4 from Khalil and Rasmussen (1984c).
vapor to produce two hydroxyl radicals (O(1D) + H2O —> OH* + OH"). The production
of OH* radicals is balanced by their removal principally by reactions with CO and CH4. On a
global scale CO removes many more OH* radicals than CH4, however CH4 becomes more
important in the southern hemisphere where there is much less CO than in the northern
hemisphere but the amount of CH4 is only slightly less.
The amount of CO that is removed by reactions with OH* radicals can be estimated by
calculating the loss as = KJ^OH'LJCO]™. The reaction rate constant of CO + OH* is K =
(1.5 x 10'13)(1 + 0.6 PaJ cmVmolecules-s (DeMore et al., 1987). K^ describes the effective
reaction rate constant taking into account the decreasing atmospheric pressure and decreasing
CO concentrations with height. Estimating K^ to be 2 X 10'13 cmVmolecules-s and taking
[OH*]ave to be 8 x 105 molecules/cm3 and [CO].W to be 90 ppbv, the annual loss of CO from
reactions with OH* is about 2,200 Tg/year. The values adopted for [OH*]>ve and [CO]m are
discussed in more detail later in this chapter.
March 5, 1990
4-4 DRAFT - DO NOT QUOTE OR CITE
-------
Uptake of CO by soils has been documented and may amount to about 250 Tg/year or
about 10% of the total emitted into the atmosphere (Bartholomew and Alexander, 1981;
Ingersoll et al., 1974; Inman et al., 1971; Seiler and Schmidt, 1974), although arid soils may
release CO into the atmosphere (Conrad and Seiler, 1982). Another 100 Tg (5%) or so are
5 probably removed annually in the stratosphere (Seiler, 1974).
4.2.3 Atmospheric Lifetime
Based on the global sources and sinks described above, the average atmospheric lifetime
of CO can be calculated to be about 2 months with a range of between 1 and 4 months which
10 reflects the uncertainty in the annual emissions of CO (T = C/S). The lifetime, however, can
vary enormously with latitude and season compared to its global average value. During
winters at high and middle latitudes CO has a lifetime of more than a year but during
summers at mid latitudes the lifetime may be closer to the average global lifetime of about
2 months. Moreover, in the tropics the average lifetime of CO is probably about 1 month.
15 These calculated variations reflect the seasonal cycles of OH* at various latitudes.
4.2.4 Latitudinal Distribution of Sources
When the sources, sinks, transport, and observed concentrations are combined into a
mass balance model it is possible to calculate any one of these four components if the others
20 are known. In the case of CO, the sources can be estimated assuming that the sinks (OH*
reaction and soils), transport, and concentrations are known. The latitudinal distribution of
sources can be described in a one-dimensional model as follows:
25 " ^ —•— - — ™* H(/i)7}-C(M,t) (4-1)
30 where
C = tropospheric mixing ratio
K = zonally and height averaged transport coefficient
T = the lifetime
35 S = emissions
£ = a factor to account for the lower concentrations of CO in the stratosphere
H = a factor to account for the variation of the tropopause height with latitude
H = the sine of latitude
R = the radius of the earth
March 5, 1990 4-5 DRAFT-DO NOT QUOTE OR CITE
-------
This model is similar to that described by Czeplak and Junge (1974) and Fink and Klais
(1978). A time-averaged version was applied to the CO budget by Hameed and Stewart
(1979) and a somewhat modified and time-dependant version shown above was applied by
Khalil and Rasmussen (1988c) to derive the latitudinal distribution of CO shown in
5 Figure 4-1. Calculations by Khalil and Rasmussen (1988c) also suggest that emissions are
higher in spring and summer compared with the other seasons, particularly in the middle
northern latitudes. This is expected for two reasons: (1) oxidation of CH4 and other
hydrocarbons is faster during the summer because of the seasonal variation of OH", and
(2) other direct emissions are also greater during spring and summer.
10 From Figure 4-1 the emissions from the northern and southern tropical latitudes sum up
to 480 Tg/year and 330 Tg/year, respectively; from the northern and southern middle
latitudes the emissions are 960 Tg/year and 210 Tg/year; from the Arctic some 50 Tg are
emitted each year and some 10 Tg/year come from the Antarctic. The largest fluxes of CO
are from the industrial band of latitudes between 30° to 50° North. From this region some
15 620 Tg/year are emitted representing about 30% of the total emissions of 2050 Tg/year. The
model does not distinguish between the anthropogenic or natural sources nor does it
distinguish between direct emissions and photochemical production of CO from the oxidation
of hydrocarbons. Much of estimated fluxes from the mid-northern latitudes and from tropical
regions are likely to be of anthropogenic origin. The latitudinal distribution in Figure 4-1 is
20 compatible with the estimate (Table 4-1) that about 60% of the total emissions are from
anthropogenic activities.
4.2.5 Uncertainties and Consistencies
The first consistency one notes is that the total emissions of CO estimated from the
25 various sources are balanced by the estimated removal of CO. The approximate balance
between sources and sinks is expected because the trends, to be discussed later, are only about
4 to 8 Tg/year compared to the total global emission rate of more than 2000 Tg/year. On the
other hand, there are many uncertainties in the sources and sinks.
March 5, 1990 4-6 DRAFT-DO NOT QUOTE OR CITE
-------
GO
o
o
IH
2
o
CO
O
O
Sine of Latitude
-1.0
-0.6
-0.2
0.2
0.6
1.0
50
40 -
30 -
20 -
10 -
0
CO
to
I
00
I
CO
CM
I
- Eq •:
I
Latitude
CO
CM
10
CO
CO
N
Figure 4-1. The estimated sources of CO as a function of latitude. The sources are, in
teragrams per year in each latitude band 0.02 units in sine of latitude. The solid lines are
estimates of uncertainties as OH* concentrations and the rate of dispersion are varied
simultaneously so that the maximum values of each of these parameters are twice the
minimum values.
Source: Khalil and Rasmussen (1988c).
March 5, 1990
4-7
DRAFT-DO NOT QUOTE OR CITE
-------
There are large uncertainties in the estimates of emissions from individual sources as
expressed in Table 4-1. In most cases the stated uncertainty is a qualitative expression of the
likely range of emissions and it cannot be interpreted statistically. Therefore the resulting
uncertainty in the total emissions, obtained by adding up the uncertainties in individual
sources, appears to be large.
There are two difficulties in improving the estimates of CO from individual sources.
First, although many critical experiments to determine the production and emissions of CO
from individual sources are yet to be done, there is a limit to the accuracy with which
laboratory data can be extrapolated to the global scale. Second, the cycle of CO may be so
intimately tied up with the cycles of hydrocarbons that accurate global estimates of CO
emissions may not be possible until the cycles of the hydrocarbons are better understood.
Whereas the global distribution and seasonal variations of OH* can be calculated, there
are no direct measurements of OH" that can be used to estimate the removal of CO. The
effective average concentration of OH* that acts on trace gases can be estimated indirectly
from the cycles of other trace gases with known global emissions. Therefore the total
emissions of CO are constrained by the budgets of other trace gases even though the estimates
of emissions from individual sources may remain uncertain. The most notable constraint may
be the budget of methylchloroform (CH3CC13). Methylchloroform is a degreasing solvent that
has been emitted into the atmosphere in substantial quantities for more than 20 years. It is
thought to be removed principally by reacting with OH* radicals and to a lesser extent by
photodissociation in the stratosphere. Because industry records on CH3CC13 production and
sales have been kept for a long time, it can be used to estimate the average amount of OH*
radicals needed to explain the observed concentrations compared to the emissions. The
accuracy of the source estimates of CH3CC13 is improved by the patterns of its uses; most of it
tends to be released shortly after purchase, so that large unknown or unqualified reservoirs
probably do not exist. The recent budgets of CH3CC13 suggest that on an average there are
about 8 X 10s molecules of OH* per cubic centimeter although significant uncertainties remain
(see for example Khalil and Rasmussen, 1984c; Prinn et al., 1987). This is the value used
earlier in estimating the loss of CO from reaction with OH*. The same average value of OH*
also explains the CH4 concentrations compared to estimated sources, lending more support to
the accuracy of the estimated OH* concentrations. Neither of these constraints is very
March 5, 1990 4-8 DRAFT-DO NOT QUOTE OR CITE
-------
stringent; however, if the total global emissions of CO from all sources are much different
from the estimated 2600 Tg/year then revisions of the budgets of both CH4 and CH3CC13 may
be required.
There are other sources and sinks of CO, believed to be of lesser importance on a global
scale, which are reviewed in the previous EPA criteria document on CO (Chan et al., 1977;
Swinnerton et al., 1971).
4.3 GLOBAL DISTRIBUTIONS
Atmospheric concentrations and thus the global distribution generally are the most
accurately known components of a global mass balance of a trace gas because direct
atmospheric measurements can be taken (Dianov-Klokov and Yurganov, 1981; Ehhalt and
Schmidt, 1978; Fraser et al., 1986; Heidt et al., 1980; Hoell et al., 1984; Khalil and
Rasmussen, 1988a,b; Pratt and Falconer, 1979; Rasmussen and Khalil, 1982; Reichle et al.,
1982; Seiler, 1974; Seiler and Fishman, 1981; Wilkness et al., 1973). Much has been
learned about the global distribution of CO over the last decade. The experiments leading to
the present understanding range from systematic global observations at ground level for the
last 8 to 10 years reported by Khalil and Rasmussen (Khalil and Rasmussen, 1988a,b) and
Seiler (Seiler, 1974; Seiler and Junge, 1970), to finding the instantaneous global distribution
of CO from remote-sensing instruments on board NASA's space shuttle as reported by
Reichle et al. (1982, 1989a).
4.3.1 Seasonal Variations
The seasonal variations of CO are well established (Dianov-Klokov and Yurganov,
1981; Fraser et al., 1986; Khalil and Rasmussen, 1988b; Seiler et al., 1984). High
concentrations are observed during the winters in each hemisphere and lowest concentrations
are seen in late summer. The amplitude of the cycle is largest at high northern latitudes and
diminishes as one moves towards the equator until it is reversed in the southern hemisphere
reflecting the reversal of the seasons. The seasonal variations are small in the equatorial
region. These patterns are expected from the seasonal variations of OH* concentrations. At
mid and high latitudes, diminished solar radiation, water vapor, and O3 during winters cause
March 5, 1990 4-9 DRAFT-DO NOT QUOTE OR CITE
-------
the concentrations of OH* to be much lower than during summer. The removal of CO is
slowed down and its concentrations build up. In summer the opposite effect exists causing
the large seasonal variations of CO. These variations are apparent in the observed global
seasonal cycles shown in Figure 4-2a.
On the hemispherical scale the seasonal variation of CO is approximately proportional to
the concentration. Therefore, because there is much more CO in the northern hemisphere
than in the southern hemisphere, the decline of concentrations during the summer of the
northern hemisphere is not balanced by the rise of concentrations in the southern hemisphere.
This causes a global seasonal variation. The total amount of CO in the earth's atmosphere
undergoes a remarkably large seasonal variation with highest global burden during northern
winters and lowest during northern summers. This feature is shown in Figure 4-2b.
4.3.2 Latitudinal Variation
The global seasonal variation of CO in the earth's atmosphere also creates a seasonal
variation in the latitudinal distribution (Khalil and Rasmussen, 1988a,b; Newell et al., 1974;
Reichle et al., 1982, 1986; Seiler, 1974). During northern winters, CO levels are at their
highest in the northern hemisphere whereas southern hemisphere concentrations are at a
minimum. The interhemispheric gradient, defined as the ratio of the amount of CO in the
northern and southern hemisphere, is at its maximum of about 3.2 during northern
hemisphere winters and falls to about 1.8 during northern hemisphere summers, which is
about half of the winter value. The average latitudinal gradient is about 2.5, which means
that on an average there is about 2.5 times as much CO in the northern hemisphere as in the
southern hemisphere. Earlier data on the latitudinal variations did not account for the
seasonal variations.
4.3.3 Variations with Altitude
In the northern hemisphere troposphere, the concentrations of CO generally decline with
altitude but in the southern hemisphere the vertical gradient may be reversed due to the
transport of CO from the northern hemisphere into the southern hemisphere. Above the
tropopause concentrations decline rapidly so that there is very little CO between 20 km and
March 5, 1990 4-10 DRAFT - DO NOT QUOTE OR CITE
-------
a
P.
O
U
S -20
-40 -
ft
o,
O
U
o
•O
Time of year
-10
-20
o AK
A OR
a HA
O SA
V TA
+ SP
GL
Figure 4-2. The global seasonal variations of CO. Figure 4-2a shows the seasonal cycle at
6 sites in polar, middle, and tropical latitudes of both hemispheres (AK = Alaska, OR =
Oregon, HA = Hawaii, SA = Samoa, TA = Tasmania, SP = South Pole). Figure 4-2b
shows the seasonal variation of the global burden of atmospheric CO. The atmospheric
content of CO is much higher during northern hemisphere winters compared to summers.
Sources: Khalil and Rasmussen (1988b).
March 5, 1990
4-11 DRAFT-DO NOT QUOTE OR CITE
-------
40 km; at still higher altitudes the mixing ratio may increase again (Fabian et al., 1981; Seller
and Junge, 1969; Seiler and Warneck, 1972).
4.3.4 Other Variations
5 The concentration of CO generally is higher over populated continental areas compared
to the air over oceans, even though oceans release CO into the atmosphere. Other regions,
such as tropical forests also may be a source of isoprene and other hydrocarbons that may
form CO in the atmosphere. Such sources produce shifting patterns of high CO
concentrations over regional and perhaps even larger spatial scales. Concentrations are
10 representative of the middle troposphere and were measured during the 1984 flights of the
space shuttle and reported by Reichle et al. (1989a). Eventually, CO in the lower troposphere
may be measured from space using the techniques described by Reichle et al. (1989b). The
new method uses gas correlation filter radiometry at 2.3 /Ltm in addition to the 4.67 pun line
used earlier to obtain mid-tropospheric concentrations of CO.
15 Occasionally in some locations significant diurnal variations of CO also may occur. For
instance, diurnal variations have been observed over some parts of the oceans with high
concentrations during the day and low concentrations at night. Because similar patterns also
exist in the surface sea water, the diurnal variations in the air can be explained by emissions
from the oceans.
20 Finally, after the repeating cycles and other trends are subtracted, considerable random
fluctuations still remain in time series of measurements. These fluctuations reflect the short
lifetime of CO and the vicinity of the sources and complicate the detection of long-term trends
(see Figure 4-3).
25
4.4 GLOBAL TRENDS
Because some 60% of the global emissions of CO are believed to come from
anthropogenic sources with increasing emissions, it stands to reason that the global
March 5, 1990 4-12 DRAFT-DO NOT QUOTE OR CITE
-------
.0
ex
p.
O
U
170
160
150
140
130
120
110
100
90
SO
70
60
50
40
30
20
10
1979
1982 1985
Time (months)
1988
•—-Alaska ±—* Oregon •—• Hawaii
T—»Samoa *—• Tasmania * "Antarctic
Figure 4-3. The global concentrations and trends of CO. The seasonal cycles have been
subtracted from the time series of measurements at various latitudes ranging from inside the
Arctic Circle to the South Pole. The latitudinal variation of CO also is apparent in the figure.
Sources: Khalil and Rasmussen (1988a).
March 5, 1990
4-13
DRAFT-DO NOT QUOTE OR CITE
-------
concentration of CO should be increasing. At present, there are several independent pieces of
evidence for an increasing trend although none are definitive. First, direct atmospheric
observations reported by Khalil and Rasmussen (1984a) showed a detectable trend at Cape
Meares in Oregon between 1979 and 1982. Over these 3 years the rate of increase was about
5 5%/year. Subsequent data from the same site showed that the rate was not sustained for long
and a much smaller trend of a somewhat less than 2%/year emerged over the longer period of
1970 to 1987 (Khalil and Rasmussen, 1988a). Similar data from other sites distributed
worldwide now show that there is evidence for a global increase of about 1 % per year as
shown in Figure 4-3 (Khalil and Rasmussen, 1988a). This is the only study in which trends
10 from different parts of the world are evaluated. It shows that the trends are strongest in the
mid-northern latitudes where most of the sources are located and become smaller and weakei
in the southern hemisphere. At the mid-southern latitude site the trends persist but are not
statistically significant (Khalil and Rasmussen, 1988a). Second, Rinsland and Levine (1985)
have reported estimates of CO concentrations from spectroscopic plates from Europe that
15 show that between 1950 and 1984 CO increased at about 2% per year. Finally, spectroscopic
measurements of CO taken by Dvoryashina et al. and Dianov-Klokov et al. in the Soviet
Union also suggest an increase of about 2%/year between 1974 and 1982 (Dianov-Klokov
et al., 1978; Dianov-Klokov and Yurganov, 1981; Dvoryashina et al., 1982, 1984; Khalil and
Rasmussen, 1988a; Khalil and Rasmussen, 1984b).
20 There is good evidence that the concentrations of CO are increasing in the non-urban
troposphere, however the rate of increase still is not well known and may vary considerably
over time. The random variability is so large and the trends so small that there are just
enough data to detect the increase but not enough to estimate the long-term rate of increase
with confidence. Such increases of tropospheric CO can cause a reduction of OH*
25 concentrations and thus reduce the oxidizing capacity of the atmosphere causing other trace
gases, including CH4, to build up more rapidly in the atmosphere and reach higher levels.
This occurrence could add to the greenhouse effect and deplete the stratospheric O3 layer. In
the troposphere, increased CO in the presence of NOX also could result in an increase of O3
concentrations.
30 All the studies show increases of 1 to 2% per year over the last several decades. These
trends and the sources shown in Table 4-1 suggest that the anthropogenic sources, both direct
March 5, 1990 4-14 DRAFT-DO NOT QUOTE OR CITE
-------
and indirect, probably were very small until this century. Therefore, the average CO
concentration may have doubled over the last 50 years or so. This change could be a
significant contributing factor to increasing levels of O3 in the non-urban troposphere.
Because the influence of CO on tropospheric O3 is not understood fully, the role of increasing
5 CO on tropospheric O3 also remains uncertain.
4.5 SUMMARY
The annual global emissions of CO are estimated to be about 2600 ± 600 Tg, of which
10 about 60% are from human activities including combustion of fossil fuels and oxidation of
hydrocarbons including CH4. The remaining 40% of the emissions are from natural
processes, mostly from the oxidation of hydrocarbons, but also from plants and the oceans.
Almost all the CO emitted into the atmosphere each year is removed by reactions with OH*
radicals (85%), by soils (10%), and by diffusion into the stratosphere. There is a small
15 imbalance between annual emissions and removal causing an increase of about 1 % per year.
It is very likely that the imbalance is due to increasing emissions from anthropogenic
activities. The average concentration of CO is about 90 ppbv, which amounts to about
400 Tg in the atmosphere and the average lifetime is about 2 mo. This view of the global
cycle of CO is consistent with the present estimates of average OH* concentrations and the
20 budgets of other trace gases including CH4 and CH3CC13.
There are large remaining uncertainties that in the future may upset the apparently
cohesive present budget of CO. Although the patterns of the global distribution are becoming
established, there still are uncertainties about the absolute concentrations. Estimates of
emissions from individual sources are very uncertain, however the total annual emissions are
25 likely to be more accurate.
There are just sufficient data on the trends to suggest that CO is increasing, but the rates
are not certain. However, if the present view of the global cycle of CO is correct, then it is
likely that, in time, increasing levels of CO will contribute to widespread changes in
atmospheric chemistry and the global climate.
30
March 5, 1990 4-15 DRAFT-DO NOT QUOTE OR CITE
-------
4.6 REFERENCES
5 Bartholomew, G. W.; Alexander, M. (1981) Soil as a sink for atmospheric carbon monoxide. Science
(Washington, DC) 212: 1389-1391.
Bauer, K.; Conrad, R.; Seiler, W. (1980) Photooxidative production of carbon monoxide by phototrophic
microorganisms. Biochim. Biophys. Acta 589: 46-55.
10
Bidwell, R. G. S.; Fraser, D. E. (1972) Carbon monoxide uptake and metabolism by leaves. Can. J. Bot. 50:
1435-1439.
Chan, W. H.; Uselman, W. M.; Calvert, J. G.; Shaw, J. H. (1977) The pressure dependence of the rate constant
15 for the reaction: OH* + CO — > H' + CO2. Chem. Phys. Lett. 45: 240-244.
Conrad, R.; Seiler, W. (1982) Arid soils as a source of atmospheric carbon monoxide. Geophys. Res. Lett.
9: 1353-1356.
20 Crutzen, P. J. (1974) Photochemical reactions initiated by and influencing O3 in unpolluted tropospheric air.
Tellus 26: 47-56.
Czeplak, G.; Junge, C. (1974) Studies of interhemispheric exchange in the troposphere by a diffusion model. In:
Frenkiel, F. N.; Munn, R. E., eds. Turbulent diffusion in environmental pollution: proceedings of a
25 symposium; April 1973; Charlottesville, VA. New York, NY: Academic Press; pp. 57-72.
DeMore, W. B.; Margitan, J. J.; Molina, M. J.; Watson, R. T.; Golden, D. M.; Hampson, R. F.; Kurylo,
M. J.; Howard, C. J.; Ravishankara, A. R. (1985) Chemical kinetics and photochemical data for use in
stratospheric modeling: evaluation number 7. Washington, DC: National Aeronautics and Space
30 Administration; report no. NASA-CR-176198. Available from: NTIS, Springfield, VA; N86-10707.
DeMore, W. B.; Molina, M. J.; Sander, S. P.; Golden, D. M.; Hampson, R. F.; Kurylo, M. J.; Howard, C. J.;
Ravishankara, A. R. (1987) Chemical kinetics and photochemical data for use in stratospheric modeling:
evaluation number 8. Washington, DC: National Aeronautics and Space Administration;; report
35 no. NASA-CR-182919. Available from: NTIS, Springfield, VA; N88-24012.
Dianov-Klokov, V. I.; Yurganov, L. N. (1981) A spectroscopic study of the global space-time distribution of
atmospheric CO. Tellus 33: 262-273.
40 Dianov-Klokov, V. I.; Fokeyeva, YE. V.; Yurganov, L. N. (1978) A study of the carbon monoxide content of
the atmosphere. Izv. Acad. Sci. USSR Atmos. Oceanic Phys. (Engl. Transl.) 14: 263-269.
Dianov-Klokov, V. I.; Yurganov, L. N. (1982) Izmereniya integral'nogo soderzhaiya prime sei CO, CH4 i N2O v
atmosfere [The measurements of integral atmospheric CO, CH4 and N2O abundances in the atmosphere].
45 Izv. Akad. Nauk SSSR Fiz. Atmos. Okeana 18: 738-743.
Dvoryashina, E. V.; Dianov-Klokov, V. I.; Yurganov, Y. L. (1982) Results of carbon monoxide abundance
measurements at Zvenigorod, 1970-1982. Moscow, USSR: USSR Academy of Sciences (preprint).
50 Dvoryashina, E. V.; Dianvo-Klokov, V. I.; Yurganov, Y. L. (1984) Variations of the carbon monoxide content
of the atmosphere for 1970-1982. Izv. 20: Akad. Nauk SSSR Fiz. Atmos. Okeana 40-47.
Ehhalt, D. H.; Schmidt, U. (1978) Sources and sinks of atmospheric methane. Pure Appl. Geophys.
116: 452-464.
March 5, 1990 4-16 DRAFT-DO NOT QUOTE OR CITE
-------
Fabian, P.; Borchers, R.; Flentje, G.; Matthews, W. A.; Seiler, W.; Giehl, H.; Bunse, K.; Mueller, F.;
Schmidt, U.; Volz, A.; Khedim, A.; Johnen, F. J. (1981) The vertical distribution of stable trace gases at
mid-latitudes. JGR J. Geophys. Res. 86: 5179-5184.
5 Fink, H. J.; Klais, O. (1978) Global distribution of fluorocarbons. Ber. Bunsen Ges. Phys. Chem.
82: 1147-1150.
Fishman, J.; Crutzen, P. J. (1978) The origin of O3 in the troposphere. Nature (London) 274: 855-858.
10 Fishman, J.; Seiler, W. (1983) Correlative nature of O3 and carbon monoxide in the troposphere: implications for
the tropospheric O3 budget. JGR J. Geophys. Res. 88: 3662-3670.
Fishman, J.; Seiler, W.; Haagenson, P. (1980) Simultaneous presence of O3 and CO bands in the troposphere.
Tellus 32: 456-463.
15
Fraser, P. J.; Hyson, P.; Rasmussen, R. A.; Crawford, A. J.; Khalil, M. A. K. (1986) Methane, carbon
monoxide and methylchloroform in the Southern Hemisphere. J. Atmos. Chem. 4: 3-42.
Greenberg, J. P.; Zimmerman, P. R.; Chatfield, R. B. (1985) Hydrocarbons and carbon monoxide in African
20 savannah air. Geophys. Res. Lett. 12: 113-116.
Hameed, S.; Stewart, R. W. (1979) Latitudinal distribution of the sources of carbon monoxide in the
troposphere. Geophys. Res. Lett. 6: 841-844.
25 Hanst, P. L.; Spence, J. W.; Edney, E. O. (1980) Carbon monoxide production in photooxidation of organic
molecules in the air. Atmos. Environ. 14: 1077-1088.
Heidt, L. E.; Krasnec, J. P.; Lueb, R. A.; Pollock, W. H.; Henry, B. E.; Crutzen, P. J. (1980) Latitudinal
distributions of CO and CH4 over the Pacific. JGR J. Geophys. Res. 85: 7329-7336.
30
Hoell, J. M.; Gregory, G. L.; Carroll, M. A.; McFarland, M.; Ridley, B. A.; Davis, D. D.; Bradshaw, J.;
Rodgers, M. O.; Torres, A. L.; Sachse, G. W.; Hill, G. F.; Condon, E. P.; Rasmussen, R. A.;
Campbell, M. C.; Farmer, J. C.; Sheppard, J. C.; Wang, C. C. ; Davis, L. I. (1984) An inter-
comparison of carbon monoxide, nitric oxide, and hydroxyl measurement techniques: overview of results.
35 JGRJ. Geophys. Res. 89: 11,819-11,825.
Ingersoll, R. B.; Inman, R. E.; Fisher, W. R. (1974) Soil's potential as a sink for atmospheric carbon monoxide.
Tellus 26: 151-159.
40 Inman, R. E.; Ingersoll, R. B.; Levy, E. A. (1971) Soil: a natural sink for carbon monoxide. Science
(Washington, DC) 172: 1229-1231.
Jaffe, L. S. (1968) Ambient carbon monoxide and its fate in the atmosphere. J. Air Pollut. Control Assoc.
18: 534-540.
45
Jaffe, L. S. (1973) Carbon monoxide in the biosphere: sources, distribution, and concentrations. JGR J. Geophys.
Res. 78: 5293-5305.
Khalil, M. A. K.; Rasmussen, R. A. (1983) Sources, sinks, and seasonal cycles of atmospheric methane. JGR J.
50 Geophys. Res. 88: 5131-5144.
Khalil, M. A. K.; Rasmussen, R. A. (1984a) Carbon monoxide in the earth's atmosphere: increasing trend.
Science (Washington, DC) 224: 54-56.
March 5, 1990 4-17 DRAFT-DO NOT QUOTE OR CITE
-------
Khalil, M. A. K.; Rasmussen, R. A. (1984b) The global increase of carbon monoxide. In: Aneja, V. P., ed.
Environmental impact of natural emissions: proceedings of an Air Pollution Control Association specialty
conference; March. Pittsburgh, PA: Air Pollution Control Association; pp. 403-414.
5 Khalil, M. A. K.; Rasmussen, R. A. (1984c) The atmospheric lifetime of methylchloroform (CH3CC13). Tellus
Ser. B 36: 317-332.
Khalil, M. A. K.; Rasmussen, R. A. (1985) Causes of increasing atmospheric methane: depletion of hydroxyl
radicals and the rise of emissions. Atmos. Environ. 19: 397-407.
10
Khalil, M. A. K.; Rasmussen, R. A. (1988a) Carbon monoxide in the earth's atmosphere: indications of a global
increase. Nature (London) 332: 242-245.
Khalil, M. A. K.; Rasmussen, R. A. (1988b) The global distribution of carbon monoxide. Oregon Graduate
15 Center preprint.
Khalil, M. A. K.; Rasmussen, R. A. (1989) Atmospheric carbon monoxide: latitudinal distribution of sources.
Geophys. Res. Lett.: submitted.
20 Krall, A. R.; Tolbert, N. E. (1957) A comparison of the light dependent metabolism of carbon monoxide by
barley leaves with that of formaldehyde, formate and carbon dioxide. Plant Physiol. 32: 321-326.
Lamontagne, R. A.; Swinnerton, J. W.; Linnenbom, V. J. (1971) Nonequilibrium of carbon monoxide and
methane at the air-sea interface. JGR J. Geophys. Res. 76: 5117-5121.
25
Levine, J. S.; Rinsland, C. P.; Tennille, G. M. (1985) The photochemistry of methane and carbon monoxide in
the troposphere in 1950 and 1985. Nature (London) 318: 254-257.
Levy, H., II. (1971) Normal atmosphere: large radical and formaldehyde concentrations predicted. Science
30 (Washington, DC) 173: 141-143.
Levy, H., II. (1973) Tropospheric budgets for methane, carbon monoxide, and related species. JGR J. Geophys.
Res. 78: 5325-5332.
35 Linnenbom, V. J.; Swinnerton, J. W.; Lamontagne, R. A. (1973) The ocean as a source for atmospheric carbon
monoxide. JGR J. Geophys. Res. 78: 5333-5340.
Liss, P. S.; Slater, P. G. (1974) Flux of gases across the air-sea interface. Nature (London) 247: 181-184.
40 Logan, J. A.; Prather, M. J.; Wofsy, S. C.; McElroy, M. B. (1981) Tropospheric chemistry: a global
perspective. JGR J. Geophys. Res. C: Oceans 86: 7210-7254.
McConnell, J. C.; McElroy, M. B.; Wofsy, S. C. (1971) Natural sources of atmospheric CO. Nature (London)
233: 187-188.
45
National Research Council. (1977) Carbon monoxide. Washington, DC: National Academy of Sciences. (Medical
and biologic effects of environmental pollutants).
Newell, R. E. (1977) One-dimensional models: a critical comment, and their application to carbon monoxide.
50 JGR J. Geophys. Res. 82: 1449-1450.
Newell, R. E.; Boer, G. J., Jr.; Kidson, J. W. (1974) An estimate of the interhemispheric transfer of carbon
monoxide from tropical general circulation data. Tellus 26: 103-107.
March 5, 1990 4-18 DRAFT-DO NOT QUOTE OR CITE
-------
Pinto, J. P.; Yung, Y. L.; Rind, D.; Russell, G. L.; Lerner, J. A.; Hansen, J. E.; Hameed, S. (1983) A general
circulation model study of atmospheric carbon monoxide. JGR J. Geophys. Res. 88: 3691-3702.
Pratt, R.; Falconer, P. (1979) Circumpolar measurements of O3, particles, and carbon monoxide from a
5 commercial airliner. JGR J. Geophys. Res. 84: 7876-7882.
Prinn, R.; Cunnold, D.; Rasmussen, R.; Simmonds, P.; Alyea, F.; Crawford, A.; Fraser, P.; Rosen, R. (1987)
Atmospheric trends in methylchloroform and the global average for the hydroxyl radical. Science
(Washington, DC) 238: 945-950.
10
Rasmussen, R. A.; Khalil, M. A. K. (1982) Latitudinal distributions of trace gases in and above the boundary
layer. Chemosphere 11: 227-235.
Rasmussen, R. A.; Went, F. W. (1965) Volatile organic material of plant origin in the atmosphere. Proc. Natl.
15 Acad. Sci. U. S. A. 53: 215-220.
Reichle, H. G., Jr.; Beck, S. M.; Haynes, R. E.; Hesketh, W. D.; Holland, J. A.; Hypes, W. D.; Orr, H. D.,
Ill; Sherrill, R. T.; Wallio, H. A.; Casas, J. C.; Saylor, M. S.; Gormsen, B. B. (1982) Carbon
monoxide measurements in the troposphere. Science (Washington, DC) 218: 1024-1026.
20
Reichle, H. G., Jr.; Connors, V. S.; Holland, J. A.; Hypes, W. D.; Wallio, H. A.; Casas, J. C.; Gormsen,
B. B.; Saylor, M. S.; Hesketh, W. D. (1986) Middle and upper tropospheric carbon monoxide mixing
ratios as measured by a satellite-borne remote sensor during November 1981. JGR J. Geophys. Res.
91: 10,865-10,887.
25
Reichle, H. G., Jr.; Connors, V. S.; Wallio, H. A.; Holland, J. A.; Sherrill, R. T.; Casas, J. C.; Gormsen,
B. B. (1989a) The distribution of middle tropospheric carbon monoxide during early October 1984. JGR
J. Geophys. Res.: submitted.
30 Reichle, H. G., Jr.; Wallio, H. A.; Gormsen, B. B. (1989b) Feasibility of determining the vertical profile of
carbon monoxide from a space platform. Appl. Opt. 28: 2104-2110.
Rinsland, C. P.; Levine, J. S. (1985) Free tropospheric carbon monoxide concentrations in 1950 and 1951
deduced from infrared total column amount measurements. Nature (London) 318: 250-254.
35
Robinson, E.; Robbins, R. C. (1969) Sources, abundance, and fate of gaseous atmospheric pollutants:
supplement. Menlo Park, CA: Stanford Research Institute.
Robinson, E.; Robbins, R. C. (1970) Atmospheric background concentrations of carbon monoxide. In: Coburn,
40 R. F., ed. Biological effects of carbon monoxide. Ann. N. Y. Acad. Sci. 174: 89-95.
Seiler, W. (1974) The cycle of atmospheric CO. Tellus 26: 116-135.
Seiler, W.; Conrad, R. (1987) Contribution of tropical ecosystems to the global budget of trace gases especially
45 CH4, H2, CO and N2O. In: Dickenson, R., ed. Geophysiology of Amazonia. New York, NY: John
Wiley.
Seiler, W.; Fishman, J. (1981) The distribution of carbon monoxide and O3 in the free troposphere. JGR J.
Geophys. Res. 86: 7255-7265.
50
Seiler, W.; Giehl, H. (1977) Influence of plants on the atmospheric carbon monoxide. Geophys. Res. Lett.
4: 329-332.
March 5, 1990 4-19 DRAFT-DO NOT QUOTE OR CITE
-------
Seller, W.; Junge, C. (1969) Decrease of carbon monoxide mixing ratio above the polar tropopause. Tellus
21: 447-449.
Seller, W.; Junge, C. (1970) Carbon monoxide in the atmosphere. JGR J. Geophys. Res. 75: 2217-2226.
5
Seller, W.; Schmidt, U. (1974) Dissolved nonconservative gases in seawater. In: Goldberg, E. D., ed. The sea:
volume 5, marine chemistry. New York, NY: John Wiley & Sons, Inc.; pp. 219-243.
Seiler, W.; Warneck, P. (1972) Decrease of the carbon monoxide mixing ratio at the tropopause. JGR J.
10 Geophys. Res. 77: 3204-3214.
Seiler, W.; Giehl, H.; Bunse, G. (1978) The influence of plants on atmospheric carbon monoxide and dinitrogen
oxide. Pure Appl. Geophys. 116: 439-451.
15 Seiler, W.; Giehl, H.; Brunke, E.-G.; Halliday, E. (1984) The seasonally of CO abundance in the Southern
Hemisphere. Tellus Ser. B 36: 219-231.
Siegel, S. M.; Renwick, G.; Rosen, L. A. (1962) Formation of carbon monoxide during seed germination and
seedling growth. Science (Washington, DC) 137: 683-684.
20
Stevens, C. M.; Krout, L.; Walling, D.; Venters, A.; Engelkemeir, A.; Ross, L. E. (1972) The isotopic
composition of atmospheric carbon monoxide. Earth Planet. Sci. Lett. 16: 147-165.
Swinnerton, J. W.; Lamontagne, R. A.; Linnenbom, V. J. (1971) Carbon monoxide in rainwater. Science
25 (Washington, DC) 172: 943-945.
Swinnerton, J. W.; Lamontagne, R. A. (1974) Carbon monoxide in the South Pacific Ocean. Tellus
26: 136-142.
30 Swinnerton, J. W.; Linnenbom, V. J.; Cheek, C. H. (1969) Distribution of methane and carbon monoxide
between the atmosphere and natural waters. Environ. Sci. Technol. 3: 836-838.
Swinnerton, J. W.; Lamontagne, R. A.; Linnenbom, V. J. (1971) Carbon monoxide in rainwater. Science
(Washington, DC) 172: 943-945.
35
Sze, N. D. (1977) Anthropogenic CO emissions: implications for the atmospheric CO-OH*-CH4 cycle. Science
(Washington, DC) 195: 673-675.
Thompson, A. M.; Cicerone, R. J. (1986) Possible perturbations to atmospheric CO, CH4, and OH*. JGR J.
40 Geophys. Res. D: Atmos. 91: 10853-10864.
Volz, A.; Ehhalt, D. H.; Derwent, R. G. (1981) Seasonal and latitudinal variation of HCO and the tropospheric
concentration of OH* radicals. JGR J. Geophys. Res. 86: 5163-5171.
45 Warneck, P. (1988) Chemistry of the natural atmosphere. New York, NY: Academic Press, Inc.
Weinstock, B.; Niki, H. (1972) Carbon monoxide balance in nature. Science (Washington, DC) 176: 290-292.
Went, F. W. (1960) Organic matter in the atmosphere, and its possible relation to petroleum formation. Proc.
50 Natl. Acad. Sci. U. S. A. 46: 212-221.
Went, F. W. (1966) On the nature of the Aitken condensation nuclei. Tellus 18: 549-556.
March 5, 1990 4-20 DRAFT-DO NOT QUOTE OR CITE
-------
Wilkness, P. E.; Lamontagne, R. A.; Larson, R. E.; Swinnerton, J. W.; Dickson, C. R.; Thompson, T. (1973)
Atmospheric trace gases in the Southern Hemisphere. Nature (London) 245: 45-47.
Wilks, S. S. (1959) Carbon monoxide in green plants. Science (Washington, DC) 129: 964-966.
5
Wofsy, S. C. (1976) Interactions of CH4 and CO in the earth's atmosphere. Annu. Rev. Earth Planet. Sci. 4:
441-469.
Wofsy, S. C.; McConnell, J. C.; McElroy, M. B. (1972) Atmospheric CH4, CO, and C02. JGR J. Geophys.
10 Res. 77: 4477-4493.
World Meteorological Organization. (1986) Atmospheric O3 1985: assessment of our understanding of the
processes controlling its present distribution and change. Geneva, Switzerland: World Meteorological
Organization; Global O3 Research and Monitoring Project report no. 16. 3v.
15
Zimmerman, P. R.; Chatfield, R. B.; Fishman, J.; Crutzen, P. J.; Hanst, P. L. (1978) Estimates on the
production of CO and H2 from the oxidation of hydrocarbon emissions from vegetation. Geophys. Res.
Lett. 5: 679-682.
March 5, 1990 4-21 DRAFT-DO NOT QUOTE OR CITE
-------
5. MEASUREMENT METHODS FOR CARBON MONOXIDE
5.1 INTRODUCTION
To promote uniform enforcement of the air quality standards set forth under the Clean
5 Air Act as amended (U.S. Senate Committee, 1977), the EPA has established provisions
under which analytical methods can be designated as "reference" or "equivalent" methods
(Code of Federal Regulations, 1977a). A reference method or equivalent method for air
quality measurements is required for acceptance of measurement data. An "equivalent
method" for monitoring CO can be so designated when the method is shown to produce
10 results equivalent to the approved reference monitoring method based on absorption of
infrared radiation from a nondispersed beam.
EPA-designated reference methods are automated, continuous methods utilizing the
nondispersive infrared (NDIR) technique, which generally is accepted as being the most
reliable method for the measurement of CO in ambient air. The official EPA reference
15 methods are described in Code of Federal Regulations, 1988. Eleven reference methods for
CO have been designated for use in determining compliance and all methods employ the
NDIR technique (Code of Federal Regulations, 1988). Before a particular NDIR instrument
can be used in a reference method, it must be designated by the EPA as approved in terms of
manufacturer, model number, components, operating range, etc. Several NDIR instruments
20 have been so designated (Code of Federal Regulations, 1988), including the gas filter
correlation (GFC) technique which was developed through EPA-sponsored research (Burch
et al., 1976). No equivalent methods that use a principle other than NDIR have as of January
1988 been designated for measuring CO in ambient air. The performance specifications for
automated CO analyzers are shown in Table 5-1 (Code of Federal Regulations, 1977a).
25 The normal full scale operating range for reference methods is 0 to 50 ppm (0 to
58 mg/m3). Some instruments offer higher ranges, typically 0 to 100 ppm (0 to 116 mg/m3)
or lower ranges such as 0 to 20 ppm (0 to 23 mg/m3). A narrower range up to 1 ppm may be
needed to measure background levels in unpolluted atmospheres. Higher ranges up to
1000 ppm (1150 mg/m3) are used to measure CO concentrations in vehicular tunnels and
30 parking garages.
March 12, 1990 5-1 DRAFT - DO NOT QUOTE OR CITE
-------
5.1.1 Overview of Techniques for Measurement of Ambient Carbon
Monoxide
There have been several excellent reviews on the measurement of CO in the atmosphere
(National Research Council, 1977; Driscoll and Berger, 1971; Harrison, 1975; American
5 Industrial Hygiene Association, 1972; Leithe, 1971; Repp, 1977; Schnakenberg, 1976;
Stevens and Herget, 1974; National Air Pollution Control Association, 1970; National
Institute for Occupational Safety and Health, 1972; Verdin, 1973). The nondispersive
infrared (NDIR) method is discussed widely in the literature (Dailey and Fertig, 1977;
Houben, 1976; McKee and Childers, 1972; McKee et al., 1973; Perez et al., 1975; Pierce
10 and Collins, 1971; Schunck, 1976; Scott, 1975; Smith and Nelson, 1973; Smith, 1969, and
Luft, 1975). Currently, the most commonly-used measurement technique is the type of NDIR
method referred to as gas filter correlation (Acton et al., 1973; Bartie and Hall, 1977; Burch
and Gryvnak, 1974; Burch et al., 1976; Chaney and McClenny, 1977; Goldstein et al., 1976;
Gryvanak and Burch, 1976a,b; Herget et al., 1976; Ward and Zwick, 1975). This technique
15 was developed to a commercial prototype stage through EPA sponsored research (Burch
etal., 1976).
The NDIR method is an automated, continuous method that generally is accepted as
being the most reliable method for the measurement of CO in ambient air. NDIR analyzers
are based on the specific absorption of infrared radiation by the CO molecule (Feldstein,
20 1967). Most commercially available analyzers incorporate a gas filter to minimize
interferences from other gases; they operate at atmospheric pressure and the most sensitive
analyzers are able to detect minimum CO concentrations of about 0.05 mg/m3. Interferences
due to CO2 and water vapor can be dealt with so as not to affect the data quality. NDIR
analyzers with Luft type detectors are relatively insensitive to flow rate, require no wet
25 chemicals, are sensitive over wide concentration ranges, and have short response times.
NDIR analyzers of the newer GFC type have overcome zero and span problems and minor
problems due to vibrations.
A more sensitive method for measuring low, background levels is gas chromatography
(Bergman et al., 1975; Bruner et al., 1973; Dagnall et al., 1973; Porter and Volman, 1962;
30 Feldstein, 1967; Smith et al., 1975a; Swinnerton et al., 1968; Tesarik and Krejci, 1974).
This technique is an automated, semicontinuous method where CO is separated from water,
March 12, 1990 5-2 DRAFT - DO NOT QUOTE OR CITE
-------
TABLE 5-1. PERFORMANCE SPECIFICATIONS FOR AUTOMATED
ANALYTICAL METHODS FOR CARBON MONOXIDE
(CODE OF FEDERAL REGULATIONS, 1977a)
5 Range 0 to 57 mg/m3 (0 to 50 ppm)
Noise 0.6 mg/m3 (0.50 ppm)
Lower detectable limit 1.2 mg/m3 (1.0 ppm)
Interference equivalent
Each interfering substance ±1.2 mg/m3 (±1.0 ppm)
10 Total interfering substances 1.7 mg/m3 (1.5 ppm)
Zero drift
12 h ±1.2 mg/m3 (±1.0 ppm)
24 h ±1.2 mg/m3 (±1.0 ppm)
Span drift, 24 h
15 20% of upper range limit ± 10.0%
80% of upper range limit ±2.5%
Lag time 10 min
Rise time 5 min
Fall time 5 min
20 Precision
20% of upper range limit 0.6 mg/m3 (0.5 ppm)
80% of upper range limit 0.6 mg/m3 (0.5 ppm)
Definitions:
25
Range: Nominal minimum and maximum concentrations that a method is capable of
measuring.
Noise: The standard deviation about the mean of short duration deviations in output that are
not caused by input concentration changes.
30 Lower detectable limit: The minimum pollutant concentration that produces a signal of
twice the noise level.
Interference equivalent: Positive or negative response caused by a substance other than the
one being measured.
Zero drift: The change in response to zero pollutant concentration during continuous
35 unadjusted operation.
Span drift: The percent change in response to an upscale pollutant concentration during
continuous unadjusted operation.
Lag time: The time interval between a step change in input concentration and the first
observable corresponding change in response.
40 Rise time: The time interval between initial response and 95% of final response.
Fall time: The time interval between initial response to a step decrease in concentration and
95% of final response.
Precision: Variation about the mean of repeated measurements of the same pollutant
concentration expressed as one standard deviation about the mean.
45
March 12, 1990 5-3 DRAFT - DO NOT QUOTE OR CITE
-------
CO2, and hydrocarbons other than methane by a stripper column. Carbon monoxide and CH4
then are separated on an analytical column and the CO is passed through a catalytic reduction
tube where it is converted to CH4. The CO (converted to CH4) passes through a flame
ionization detector (FID), and the resulting signal is proportional to the concentration of CO
5 in the air. This method has been used throughout the world. It has no known interferences
and can be used to measure levels from 0.03 to 50 mg/m3. These analyzers are expensive
and require continuous attendance by a highly trained operator to produce valid results. For
high levels, a useful technique is catalytic oxidation of the CO by Hopcalite or other catalysts
(Stetter and Blurton, 1976), either with temperature-rise sensors (Naumann, 1975; Poli et al.,
10 1976; Schnakenberg, 1976) or with electrochemical sensors (Bay et al., 1974, 1972; Bergman
et al., 1975; Dempsey et al., 1975; Repp, 1977; Schnakenberg, 1975).
Other analytical schemes used for CO in air include dual-isotope infrared fluorescence,
another technique derived from NDIR (Link et al., 1971; McClatchie, 1972; McClatchie
et al., 1972); reaction with hot mercuric oxide to give elemental mercury vapor (Beckman
15 et al., 1948; McCullough et al., 1947; Mueller, 1954; Palanos, 1972; Robbins et al., 1968);
reaction with heated iodine pentoxide to give elemental iodine (Adams and Simmons, 1951;
Moore et al., 1973; Newton and Morss, 1974; van Dijk and Falkenburg, 1976; Vol'berg and
Pochina, 1974); and color reactions (Allen and Root, 1955; Bell et al., 1975; Feldstein,
1965; Jones, 1977; Lambert and Wiens, 1974; Levaggi and Feldstein, 1964; Ray et al.,
20 1975; Simonescu et al., 1975; Smith et al., 1975b), as with palladium salts or the silver salt
of/?-sulfamoylbenzoate. Many of these methods are described in Section 5.3, Measurement
in Ambient Air. A classical procedure for many decades was to use gasometric apparatus
such as the Orsat or Haldane (Cormack, 1972), in which the CO present in a gas sample is
absorbed by cuprous chloride solution and the decrease in volume or pressure is measured.
25 This method, however, is not sensitive enough for trace amounts.
Microwave rotational spectroscopy is an analytical technique with high specificity
(Hrubesh, 1973; Morgan and Morris, 1977). Other possible ways to determine CO include
chemiluminescent reaction with ozone (v. Heusden and Hoogeveen, 1976), X-ray excited
optical fluorescence (Goldstein et al., 1974), radiorelease of 85Kr from the kryptonates of
30 mercuric oxide or iodine pentoxide (I2O5) (Goodman, 1972; Naoum et al., 1974), and
March 12, 1990 5-4 DRAFT - DO NOT QUOTE OR CITE
-------
utilization of narrow-band infrared laser sources (Chancy et al., 1979; Optical Society of
America, 1975; Golden and Yeung, 1975).
5.1.2 Calibration Requirements
5 Whichever method or instrument is used, it is essential that the results be validated by
frequent calibration with samples of known composition similar to the unknowns (Commins
etal., 1977; Goldstein, 1977; National Bureau of Standards, 1975). Chemical analyses can
be relied on only after the analyst has achieved acceptable accuracy in the analysis of such
standard samples through an audit program.
10
5.2 PREPARATION OF STANDARD REFERENCE MATERIALS
5.2.1 Gas Standards
A set of reliable gas standards for CO in air, certified at levels of approximately 12, 23,
15 and 46 mg/cm3 (10, 20, and 40 ppm) is obtainable from the National Institute of Standards
and Technology (formerly National Bureau of Standards), Washington, DC (National Bureau
of Standards, 1975). These Standard Reference Materials (SRMs) are supplied as compressed
gas (at about 1700 psi) in high-strength aluminum cylinders containing 31 ft3 of gas at dry
standard temperature and pressure and are accurate to better than 1 % of the stated values.
20 Because of the time and effort required in their preparation, SRMs are not intended for use as
daily working standards, but rather as primary standards against which transfer standards can
be calibrated.
5.2.2 Gravimetric Method
25 The gravimetric method used by NBS for preparing primary standards of CO (Hughes,
1975, 1976) is as follows. An empty gas cylinder is tared on an analytical balance; then 2 g
of pure CO, weighed accurately to ±2 mg, is added from a high-pressure tank. Next, 100 g
of pure air (accurately weighed) is added from a pressure tank, and the concentration of CO is
calculated from the sum of the respective weights added to the molecular weights of the two
30 gases. Not only the average "molecular weight" of the air, but also the requisite careful
check of purity, is obtained by mass spectrometry and gas chromatography analyses of the air
March 12, 1990 5-5 DRAFT - DO NOT QUOTE OR CITE
-------
and the CO. Lower-concentration primary standards are prepared by serial dilutions (not
more than a factor of 100 for each step) by the same technique.
The commercial suppliers of compressed gases are another source of air samples
containing CO in the milligram per cubic meter or parts-per-million range. However, the
5 nominal values for CO concentration supplied by the vendor should be verified by
intercomparison with an SRM or other validated standard sample. A three-way
intercomparison has been made among the NBS SRMs, commercial gas blends, and an
extensive set of standard gas mixtures prepared by gravimetric blending at the Environmental
Protection Agency (Paulsell, 1976). Results of the comparison showed that commercial gas
10 blends are within +2% of the true value represented by a primary standard. Another study
on commercial blends (Elwood, 1976) found poorer accuracy. To achieve compatible results
in sample analyses, different laboratories should interchange and compare their respective
working standards frequently.
In making and using standards, many precautions are needed (Hughes, 1975): One
15 deserves special mention. Large but unpredictable decreases in CO concentration occur
within a few months in mixtures prepared in ordinary mild steel gas cylinders, as shown in
Figure 5-1. This may be due to carbonyl formation or oxidation and CO to CO2. The
difficulty can be avoided by the use of gas cylinders made of stainless steel or aluminum. A
special treatment for aluminum, which includes enhancement of the aluminum oxide surface
20 layer, has been recommended (Wechter, 1976).
In addition to the set of SRMs for CO in air, another set of SRMs is available from
NBS for CO in nitrogen. This second set covers concentrations from 10 to 957 ppm.
5.2.3 Volumetric Gas Dilution Methods
25 Standard samples of CO in air also can be prepared by volumetric gas dilution
techniques. In a versatile system designed for this purpose (Hughes et al., 1973), air at a
pressure of 10 to 100 psi is first purified and dried by passage through cartridges of charcoal
and silica gel, then passed through a sintered metal filter into a flow control and flowmeter
system. The CO (or a mixture of CO in air that is to be diluted further), also under pressure,
is passed through a similar flow control and flowmeter system.
March 12, 1990 5-6 DRAFT - DO NOT QUOTE OR CITE
-------
I
o
a
cw
VO
CONCENTRATION OF CARBON MONOXIDE , ppm
O
o
n
-------
Both gas streams are fed into a mixing chamber, which is designed to mix the gas
streams rapidly and completely before passage into the sampling manifold from which the
standard samples will be withdrawn. From the air flow rate, FA, and the CO flow rate, Fco,
the concentration of CO in the sample, Cco, is readily calculated by the expression
Cco= Fc° (5-1)
F
co
For samples prepared by dilution of a more concentrated bulk mixture, the concentration is
given by
F"
(Q (5-2)
where Fb and Cb are the values of flow rate and concentration of CO, respectively, for the
bulk mixture.
5.2.4 Other Methods
Permeation tubes have been used for preparing standard mixtures of such pollutant gases
as SO2 and NO2 (O'Keeffe and Ortman, 1966; Scaringelli et al., 1970). Permeation tubes are
not used routinely in the United States for making CO standard samples. In the permeation
tube techniques, a sample of the pure gas under pressure is allowed to diffuse through a
calibrated partition at a defined rate into a diluent gas stream to give a standard sample of
known composition.
Another possible way to liberate known amounts of CO into a diluent gas is by thermal
decomposition of nickel tetracarbonyl. However, an attempt to use this as a gravimetric
calibration source showed that the relation between CO output and weight loss of the Ni(CO)4
is nonstoichiometric (Stedman et al., 1976).
5.3 MEASUREMENT IN AMBIENT AIR
Ambient CO monitoring is an expensive and time-consuming task, requiring skilled
personnel and sophisticated analytical equipment. This section discusses several important
March 12, 1990 5-8 DRAFT - DO NOT QUOTE OR CITE
-------
aspects of the continuous and intermittent measurement of CO in the atmosphere, including
sampling techniques, sampling schedules, and recommended analytical methods for CO
measurement.
5.3.1 Sampling System Components
Carbon monoxide monitoring requires a sample introduction system, an analyzer system,
and a data recording system, as illustrated in Figure 5-2 (Smith and Nelson, 1973). While
the "heart" of any air pollution monitoring system is the air pollution analyzer, Figure 5-2
shows that there is a considerable amount of supportive equipment necessary for continuous
air monitoring.
A sample introduction system consists of a sampling probe, an intake manifold, tubing,
and air movers. This system is needed to collect the air sample from the atmosphere and to
transport it to the analyzer without altering the original concentration. It also may be used to
introduce known gas concentrations in order to check periodically the reliability of the
analyzer output. Construction materials for the sampling probe, intake manifold, and tubing
should be tested to demonstrate that the test atmosphere composition or concentration is not
altered significantly. It is recommended that sample introduction systems be fabricated from
borosilicate glass or PEP Teflon® (Code of Federal Regulations, 1977b) if several pollutants
are to be monitored. However, in monitoring for CO only, it has been reported (Wohlers
et al., 1967) that no measurable pollutant losses were observed at the high (> 1 L/min)
sampling flow rates when sampling systems were constructed of tygon, polypropylene,
polyvinylchloride, aluminum, or stainless steel piping. The sample introduction system
should be constructed so that it presents no pressure drop to the analyzer. At low flow and
low concentrations, such operation may require validation.
The analyzer system consists of the analyzer itself and any sample preconditioning
components that may be necessary. Sample preconditioning might require a moisture control
system to help minimize the false positive response of the analyzer (e.g., the NDIR analyzer)
to water vapor, and a particulate filter to help protect the analyzer from clogging and possible
chemical interference due to particulate buildup in the sample lines or analyzer inlet. The
sample preconditioning system also may include a flow metering and flow control device to
control the sampling rate to the analyzer. As for the analyzer, there are several analytical
March 12, 1990 5-9 DRAFT - DO NOT QUOTE OR CITE
-------
ar
>—»
K)
I
§
I
I
a
§
n
SAMPLE INTRODUCTION SYSTEM
SAMPLE INTAKE PORT
BLOWER
FIRST STAGE
PRESSURE V_
GAUGE Y?)
CYLINDER
PRESSURE
VALVE
SECOND STAGE
\ PRESSURE GAUGE
SECOND STAGE
PRESSURE VALVE
ZERO GAS
SPAN GAS
DATA RECORDING
AND
DISPLAY SYSTEM
CARBON MONOXIDE ANALYZER
Figure 5-2. Carbon monoxide monitoring system (Smith and Nelson, 1973).
-------
methods for the continuous measurement of CO. These are described in Section 5.3.4,
Continuous Analysis.
A data recording system is needed to record the output of the analyzer. Data recording
systems range from simple strip chart recorders to digital magnetic tape recorders to
5 computerized telemetry systems that transfer data from remote stations to a central location
via telephone lines or radio waves.
5.3.2 Quality Assurance Procedures for Sampling
The accuracy and validity of data collected from a CO monitoring system must be
10 ensured through a quality assurance program. Such a program consists of procedures for
calibration, operational and preventive maintenance, data handling, and auditing; and the
procedures are documented fully in a quality assurance program manual maintained by the
monitoring organization.
Calibration procedures consist of periodic multipoint primary calibration and secondary
15 calibration, both of which are prescribed to minimize systematic error. Primary calibration
involves the introduction of test atmospheres of known concentration to an instrument in its
normal mode of operation for the purpose of producing a calibration curve.
A calibration curve is derived from the analyzer response obtained by introducing
several successive test atmospheres of different known concentrations. One recommended
20 method for generating CO test atmospheres is to use zero air (containing no CO) along with
several known concentrations of CO in air or nitrogen contained in high-pressure gas
cylinders and verified by NBS-certified SRMs wherever possible (Code of Federal
Regulations, 1977a). The number of standard gas mixtures (cylinders) necessary to establish
a calibration curve depends on the nature of the analyzer output. A multipoint calibration at
25 five or six different CO concentrations covering the operating range of the analyzer is
recommended by EPA (Code of Federal Regulations, 1977b; Federal Register, 1978).
Alternatively, the multipoint calibration is accomplished by diluting a known high-
concentration CO standard gas with zero gas in a calibrated flow dilution system.
Primary calibrations should be performed when the analyzer is first purchased and every
30 30 days thereafter (Smith and Nelson, 1973). Primary calibration also is recommended after
the analyzer has had maintenance that could affect its response characteristics or when results
March 12, 1990 5-11 DRAFT - DO NOT QUOTE OR CITE
-------
from auditing show that the desired performance standards are not being met (Smith and
Nelson, 1973).
Secondary calibration consists of a zero and upscale span of the analyzer. This is
recommended to be performed daily (Federal Register, 1978). If the analyzer response
5 differs by more than 2% from the certified concentrations, then the analyzer is adjusted
accordingly. Complete records of secondary calibrations should be kept to aid in data
reduction and for use in auditing.
Operational and preventive maintenance procedures consist of operational checks to
ensure proper operation of the analyzer and a preventive maintenance schedule necessary to
10 prevent unexpected analyzer failure and the associated loss of data (PEDCo Environmental
Specialists, Inc., 1971). Operational checks include checks of zero and span control settings,
sample flow rate, gas cylinder pressures, sample cell pressure, shelter temperature, water
vapor control, the paniculate filter, the sample introduction system, the recording system, and
the strip chart record. These checks may indicate the need for corrective/remedial action.
15 They usually are performed in conjunction with secondary calibrations. In addition to
operational checks, a routine schedule of preventive maintenance should be developed.
Maintenance requirements for the analyzer usually are specified in the manufacturer's
instrument manual. Routine maintenance of supportive equipment (i.e., the sample
introduction system and the data recording system) also is required. This may include sample
20 line filter changes, water vapor control changes, sample line cleaning, leak checks, and chart
paper supply changes.
Data handling procedures consist of data generation, reduction, validation, recording,
and analysis and interpretation. Data generation is the process of generating raw,
unprocessed, and unvalidated observations as recorded on a strip chart record. Data reduction
25 is the conversion, by use of calibration records, of raw data to concentration units. Data
validation involves final screening of data before recording. Then, questionable data
"flagged" by the monitoring technician are reviewed with the aid of daily calibration and
operation records to assess their validity. Specific criteria for data selection and several
instrument checks are available (Smith and Nelson, 1973). Data recording involves recording
30 in a standard format for data storage, interchange of data with other agencies, and/or data
analysis. Data analysis and interpretation usually include a mathematical or statistical analysis
March 12, 1990 5-12 DRAFT - DO NOT QUOTE OR CITE
-------
of air quality data and a subsequent effort to interpret results in terms of exposure patterns,
meteorological conditions, characteristics of emission sources, and geographic and
topographic conditions.
Auditing procedures consist of several quality control checks and subsequent error
5 analyses to estimate the accuracy and precision of air quality measurements. The quality
control checks for CO include a data processing check, a control sample check, and a water
vapor interference check, which should be performed by a qualified individual independent of
the regular operator. The error analysis is a statistical evaluation of the accuracy and
precision of air quality data. Guidelines have been published by EPA (Smith and Nelson,
10 1973) for calculating an overall bias and standard deviation of errors associated with data
processing, measurement of control samples, and water vapor interference, from which the
accuracy and precision of CO measurements can be determined. Since January 1, 1983, all
state and local agencies submitting data to EPA must provide estimates of accuracy and
precision of the CO measurements based on primary and secondary calibration records
15 (Federal Register, 1978). The precision and accuracy audit results through 1985 indicate that
the 95% national probability limits for precision are +9% and the 95% national probability
limits for accuracy are within +1.5% for all audit levels from 3 to 8 ppm to 80 to 90 ppm.
The results for CO are better than comparable results for the other pollutants with national air
quality standards (Rhodes and Evans, 1987).
20
5.3.3 Sampling Schedules
Carbon monoxide concentrations in the atmosphere exhibit large temporal variations due
to changes in the time and rate that CO is emitted by different sources and due to changes in
meteorological conditions that govern the amounts of transport and dilution that take place.
25 During a one-year period an urban CO station may monitor hourly concentrations of CO
ranging from 0 to as high as 50 mg/m3 (45 ppm). The NAAQS for CO are based on the
second highest one- and eight-hour average concentrations; violations represent extreme
events when compared to the 8760 hours that constitute a year. In order to measure the
highest two values from the distribution of 8760 hourly values, the "best" sampling schedule
30 to employ is continuous monitoring 24 hours per day, 365 days per year. Even so,
continuous monitors rarely operate for long periods without data losses due to malfunctions,
March 12, 1990 5-13 DRAFT - DO NOT QUOTE OR CITE
-------
upsets, and routine maintenance. Data losses of 5 to 10% (438 to 876 hours per year) are not
uncommon. Consequently, the data must be interpreted in terms of the "likelihood" that the
NAAQS were attained or violated. Statistical methods can be employed to interpret the
results (Garbarz et al., 1977; Larsen, 1971).
5 Compliance with one- and eight-hour NAAQS requires continuous monitoring.
Statistically valid sampling could be performed on random or systematic schedules, however,
if annual averages or relative concentration levels were of importance. Most investigations of
various sampling schedules have been conducted for paniculate air pollution data (Hunt,
1972; Ott and Mage, 1975; Phinney and Newman, 1972), but the same schedules also could
10 be used for CO monitoring. However, most instruments do not perform reliably in
intermittent sampling.
5.3.4 Continuous Analysis
5.3.4.1 Nondispersive Infrared Photometry
15 Carbon monoxide has a characteristic infrared absorption near 4.6 urn: The absorption
of infrared radiation by the CO molecule therefore can be used to measure CO concentration
in the presence of other gases. The NDIR method is based on this principle.
Nondispersive infrared systems have several advantages. They are not sensitive to flow
rate, they require no wet chemicals, they are reasonably independent of ambient air
20 temperature changes, they are sensitive over wide concentration ranges, and they have short
response times. Further, NDIR systems may be operated by nontechnical personnel. NDIR
analyzers using Luft-type detectors were widely used in the 1970s while GFC analyzers are
most commonly used now in documenting compliance with ambient air standards.
25 NDIR Using Luft-Type Detectors
The Luft-type detector is the primary distinguishing feature for the type of NDIR
monitor that was widely used in the 1970s. This type of analyzer contains a hot filament
source of infrared radiation, a rotating sector (chopper), a sample cell, and reference cell, and
a detector. The reference cell contains a non-infrared-absorbing gas, while the sample cell is
30 continuously flushed with the sample atmosphere. The detector consists of a two-
compartment gas cell (both filled with CO under pressure) separated by a diaphragm whose
March 12, 1990 5-14 DRAFT - DO NOT QUOTE OR CITE
-------
movement causes a change of electrical capacitance in an external circuit and ultimately an
amplified electrical signal suitable for input to a servo-type recorder.
During analyzer operation a mechanical chopper alternately exposes the reference and
sample cells to the infrared sources. At the frequency imposed by the chopper, infrared
5 energy passes unattended through the reference cell to one compartment of the detector cell.
Transmission through the sample cell is adjusted with no CO present so that the two beams
are matched. Subsequently, when sample is introduced into the sample cell, infrared energy
is attenuated by CO absorption, causing an imbalance in the energy reaching the two
compartments of the detector cell. These unequal amounts of infrared energy differentially
10 heat the absorbing gas in the detector cell and the resulting pressure difference inside the cells
causes movement of the diaphragm that forms their common wall. A signal is generated at
the chopping frequency with an amplitude related to the concentration of CO in the sample.
This in turn produces the electrical signal previously discussed.
Because water vapor is the principal interfering substance in determining CO by NDIR
15 techniques, a moisture control or compensation system is particularly important. Water vapor
can be removed by absorption using in-line drying agents or by removal of condensate in a
cooled inlet line. Alternatively, the water vapor concentration can be measured independently
and its contribution subtracted from the total signal.
20 Gas-Filter Correlation Spectroscopy
A GFC monitor (Burch et al., 1976) is in essence a modern NDIR monitor. It has all
the advantages of an NDIR instrument and the additional advantages of smaller size, no
interference from CO2, and very small interference from water vapor. It is not sensitive to
flow rate, requires no wet chemicals, has a very fast response, and is relatively independent
25 of normal ambient temperature changes.
A top view of the GFC monitor is presented schematically in Figure 5-3A, showing the
components of the optical path for CO detection. During operation, sample air is
continuously pushed through the sample cell. Radiation from the source is directed by optical
transfer elements through the two main optical subsystems: the rotating gas filter (designated
30 as correlation cell in Figure 5-3A) and the optical multipass (sample) cell. The beam exits the
sample cell through interference filter FC, which limits the spectral passband to a few of the
March 12, 1990 5-15 DRAFT - DO NOT QUOTE OR CITE
-------
strongest CO absorption lines in the 4.6-fj.m region. Detection of the transmitted radiation
occurs at the infrared detector, C.
Although the passbound of filter FC is chosen to minimize interference from other
gases, some residual H2O interference occurs. This residual interference is not significant at
5 criteria pollutant levels, but can be corrected by independent measurement of H2O in the same
cell.
The gas correlation cell is constructed with two compartments (Figure 5-3B): one
compartment (gas cell 1) is filled with one-half atmosphere of CO, and the other compartment
(gas cell 2) is filled with pure N2. Radiation transmitted through cell 1 is completely
10 attenuated at spectral positions where CO absorbs strongly. The radiation transmitted by cell
2 is reduced by coating the exit window of the cell with a neutral attenuator. In this way, the
amounts of radiation transmitted by the two cells are made approximately equal in the spectral
passband that reached detector C through filter FC.
In operation, radiation passes alternately through the two cells as they are rotated by a
15 synchronous motor drive. This establishes a signal modulation frequency. Transmission to
the detector is constant if no absorption by the ambient sample occurs. If CO is present in
the sample, the radiation transmitted through cell 1 is not appreciably changed, while that
through cell 2 is changed. This imbalance is linearly related to CO concentration for small
concentrations. Other gas species absorb the radiation transmitted by cells 1 and 2 in
20 approximately equal amounts since their absorption structure does not correlate with that of
CO.
Superimposed on the entrance window of the cell is a typical light chopper pattern
(Figure 5-3B) that creates a carrier frequency 12 times the signal modulation frequency, i.e.,
a carrier frequency of 400 Hz. The detector output from the CO channel is fed to two phase-
25 sensitive amplifiers that separate the detector response at the signal frequency from the
detector response at the reference (carrier) frequency. The signal due to CO is divided by the
reference signal to substantially reduce many of the causes of sensitivity change, such as
accumulation of material on optical components and variation in detector sensitivity.
30
March 12, 1990 5-16 DRAFT - DO NOT QUOTE OR CITE
-------
DETC
SOURCE
SAMPLE CELL
ELECTRONICS
Ms
PLATED PATTERN
CHOPPER
CELL CONTAINING
NEUTRAL
ATTENUATOR
CELL CONTAINING
CO
Figure 5-3. Schematic diagram of gas filter correlation (GFC) monitor for CO. A:
Optical layout (M denotes mirror reflector; L denotes lens); B: Detail of correlation
cell.
Source: Chancy and McClenny (1977).
March 12, 1990
5-17 DRAFT - DO NOT QUOTE OR CITE
-------
5.3.4.2 Gas Chromatography - Flame lonization
In this type of system, CO is separated from other trace gases by gas chromatography
and catalytically converted to CH4 prior to detection. A gas sampling valve, a back flush
valve, a precolumn, a gas chromatographic column, a catalytic reactor, and an FID comprise
5 the gas chromatography-flame ionization system. In operation, measured volumes of air are
delivered 4 to 12 times per hour to a hydrogen FID that measures the total hydrocarbon
content (THC). A portion of the same air sample, injected into a hydrogen carrier gas
stream, is passed through the precolumn where it is separated from water, CO2, and
hydrocarbons other than CH4. Methane then is separated from CO on a second gas
10 chromatographic column. The CH4, which is eluted first, is unchanged after passing through
a catalytic reduction tube into the FID. The CO eluted into a catalytic reduction tube is
reduced to CH4 before passing through the FID (Porter and Volman, 1962). Between
analyses the precolumn is flushed out. Nonmethane hydrocarbon concentrations also can be
determined by subtracting the CH4 value from the total hydrocarbon (TH) value.
15 There are two possible modes of operation. One of these is a complete chromatographic
analysis showing the continuous output from the detector for each sample injection. In the
other, the system is programmed for both automatic zero and span settings to display selected
elution peaks as bar graphs. The peak height is then the measure of the concentration. The
first operation is referred to as the chromatographic or "spectro" mode and the second as the
20 barographic or "normal" mode.
Because measuring CO entails only small increases in cost, instrument complexity, and
analysis time, these instruments customarily are used to measure three pollutants: CH4, THs,
and CO.
The instrumental sensitivity for each of these three components is 0.023 mg/m3
25 (0.02 ppm). The lowest full-scale range available is usually 2.3 mg/m3 (2 ppm) to 5.7 mg/m3
(5 ppm), although at least one instrument has a 1.2 mg/m3 (1 ppm) range. Because of the
complexity of these instruments, continuous maintenance by skilled technicians is required to
minimize downtime. This maintenance requirement may be considered a possible
disadvantage of the system. Depending on the frequency of analysis and the temporal
30 variability of CO, the representativeness over short averaging times may not be accurate
(Chancy and McClenny, 1977).
March 12, 1990 5-18 DRAFT - DO NOT QUOTE OR CITE
-------
5.3.4.3 Other Analyzers
Controlled-Potential Electrochemical Analysis
Carbon monoxide is measured by means of the current produced in aqueous solution by
its electro-oxidation by an electro-catalytically active noble metal. The concentration of CO
5 reaching the electrode is controlled by its rate of diffusion through a membrane. This is
dependent on its concentration in the sampled atmosphere (Bay et al., 1974; 1972). Proper
selection of both the membrane and such cell characteristics as the nature of the electrodes,
the electrode potential, and the solution make the technique selective for various pollutants.
A similar technique has been reported by Yamate and Inoue (1973).
10 The generated current is linearly proportional to the CO concentration from 0 to
115 mg/m3 (0 to 100 ppm). A sensitivity of 1.2 mg/m3 (1 ppm) and a 10-s response time (to
reach 90% of full scale) are claimed for currently available commercial instruments.
Acetylene and ethylene are the chief interfering substances: one part acetylene responds
as 11 parts CO, and one part ethylene as 0.25 part CO. For hydrogen, ammonia, hydrogen
15 sulfide, NO, NO2, SO2, natural gas, and gasoline vapor, interference is less than 0.03 part CO
per one part interfering substance.
Galvanic Analyzer
Galvanic cells employed in the manner described by Hersch (1966; 1964) can be used to
20 measure atmospheric CO continuously. When an air stream containing CO is passed into a
chamber packed with I2O5 and heated to 150°C, the following reaction takes place:
SCO + I2O5 * 5CO2 + I2 (5-3)
25 The liberated iodine is absorbed by an electrolyte and transferred to the cathode of a galvanic
cell. At the cathode, the iodine is reduced and the resulting current is measured by a
galvanometer. Instruments with this detection system have been used successfully to measure
CO levels in traffic along freeways (Haagen-Smit, 1966).
Mercaptans, hydrogen sulfide, hydrogen, olefins, acetylenes, and water vapor interfere.
30 Water may be removed by sampling through a drying column; hydrogen, hydrogen sulfide,
March 12, 1990 5-19 DRAFT - DO NOT QUOTE OR CITE
-------
acetylene, and olefin interferences can be minimized by sampling through an absorption tube
containing mercuric sulfate on silica gel.
Coulometric Analyzer
5 A coulometric method employing a modified Hersch-type cell has been used for
continuous measurement of CO in ambient air (Dubois et al., 1966). The reaction of I2O3
with CO liberates iodine, which then is passed into a Ditte cell, and the current generated is
measured by an electrometer-recorder combination. Interferences are the same as those
discussed above for the galvanic analyzer.
10 This technique may be used for a minimum detectable concentration of 1.2 mg/m3
(1 ppm) with good reproducibility and accuracy if flow rates and temperatures are controlled
well. This method requires careful column preparation and use of filters to remove
interferences. Its relatively slow response time may be an added disadvantage in some work.
15 Mercury Replacement
Mercury vapor formed by the reduction of mercuric oxide (HgO) by CO is detected
photometrically by its absorption of ultraviolet light at 253.7 nm. The reaction involved is as
follows:
20 (2io° c>
CO 4- HgO > CO2 + Hg (5-4)
This is potentially a much more sensitive method than infrared absorption because the
25 oscillator strength of Hg at 253.7 nm is 2000 times greater than that of CO at 4.6 jwm.
Hydrogen and hydrocarbons also reduce HgO to Hg, and there is some thermal decomposition
of the oxide. Operation of the detector at constant temperature results in a regular
background concentration of Hg from thermal decomposition. The instrument is portable and
can analyze CO concentrations of 0.025 to 12 mg/m3 (0.020 to 10.0 ppm). Changes of
30 0.002 mg/m3 (0.002 ppm) are detectable. For this reason, this instrument has been used to
determine global CO levels. McCullough et al. (1947) and Beckman et al. (1948)
recommended a detector operating temperature of 175° C to minimize hydrogen interference.
A commercial instrument employing these principles was made and used during the middle
March 12, 1990 5-20 DRAFT - DO NOT QUOTE OR CITE
-------
1950s (Mueller, 1954). The technique recently has been used for measuring background CO
concentrations. Robbins et al. (1968) have described an instrument in which the HgO
chamber is operated at 210°C, and the amount of hydrogen interference is assessed by
periodic introduction of a tube of silver oxide into the intake air stream. At room temperature
5 silver oxide quantitatively oxidizes CO, but not hydrogen. Thus, the baseline hydrogen
concentration can be determined. Additional minor improvements are discussed by Seiler and
Junge (1970), who gave the detection limit for CO as 0.003 mg/m3 (0.003 ppm).
More recently, Palanos (1972) described a less sensitive model of this instrument
intended for use in urban monitoring. It has a range of 0 to 23 mg/m3 (20 ppm), a sensitivity
10 of about 0.58 mg/m3 (0.5 ppm), and a span and zero drift of less than 2% per day. As in
other similar instruments, specificity is achieved by removal of the potentially interfering
substances (which is less than 10%) other than hydrogen.
With all of these instruments, a constant geophysical hydrogen concentration is assumed.
In unpolluted atmospheres, the hydrogen concentration is roughly 46.5 /xg/m3 (0.56 ppm).
15 However, the automobile not only is a source of CO but also of hydrogen. Therefore, if this
technique is used in polluted areas, it will be necessary to measure the hydrogen concentration
frequently.
Dual Isotope Fluorescence
20 This instrumental method utilizes the slight difference in the infrared spectra of isotopes.
The sample is alternately illuminated with the characteristic infrared wavelengths of carbon
monoxide-16 (12C16O) and carbon monoxide-18 (12C18O). The CO in the sample that has the
normal isotope ratio, nearly 100% 12C16O, absorbs only the 12C16O wavelengths. Therefore,
there is a cyclic variation in the intensity of the fluorescent light that is dependent on the
25 12C16O content of the sample (Link et al., 1971; McClatchie, 1972; McClatchie et al., 1972).
Full-scale ranges of 0 to 23 mg/m3 (0 to 20 ppm) and up to 0 to 230 mg/m3 (0 to
200 ppm) with a claimed sensitivity of 0.23 mg/m3 (0.2 ppm) are available in this instrument.
The response time (to reach 90% of full scale) is 25 s, but a 1-s response time also is
available. An advantage of this technique is that it minimizes the effects of interfering
30 substances.
March 12, 1990 5-21 DRAFT - DO NOT QUOTE OR CITE
-------
Catalytic Combustion - Thermal Detection
Determination of CO by this method is based on measuring the temperature rise
resulting from catalytic oxidation of the CO in the sample air.
The sample air is pumped first into a furnace that brings it to a preset, regulated
5 temperature and then over the catalyst bed in the furnace. A thermopile assembly measures
the temperature difference between the air leaving the catalyst bed and the air entering it.
The output of the thermopile, which is calibrated with known concentrations of CO in air, is
read on a strip chart recorder as parts of CO per million parts of air. The sensitivity is about
1.2 mg/m3 (1 ppm). Most HCs are oxidized by the same catalyst, and will interfere unless
10 removed. These systems are widely used in enclosed spaces; their applicability for ambient
air monitoring is limited because they function best at high CO concentrations.
Second-Derivative Spectrometry
A second-derivative spectrometer processes the transmission-versus-wavelength function
15 of an ordinary spectrometer to produce an output signal proportional to the second derivative
of this function. Ultraviolet light of continuous wavelength is collected and focused onto an
oscillating entrance slit of a grating spectrometer. When the grating orientation is changed
slowly, a slowly scanning center wavelength with sinusoidal wavelength modulation is created
in the existing light by the oscillating entrance slit. This radiation passes through a gas
20 sample and is detected with a photomultiplier tube. The signal then is electronically
processed to produce a second-derivative spectrum (Lawrence Berkeley Laboratory, 1973).
This method has the advantage that it can be used to measure other pollutants as well as CO.
Fourier-Transform Spectroscopy
25 Fourier-transform spectroscopy is an extremely powerful infrared spectroscopic
technique (Bell, 1972) that has developed in the past 20 years and has been applied in the last
10 years to air pollution measurement problems (Hanst et al., 1973; Lawrence Berkeley
Laboratory, 1973). The advantage of this technique over a standard grating or prism
spectrometer is that it has a higher throughput, which means that the available energy is used
30 more effectively and that a much higher resolving power is obtainable. In air pollution
measurements individual absorption lines can be resolved.
March 12, 1990 5-22 DRAFT - DO NOT QUOTE OR CITE
-------
A special advantage for air pollution measurements is that all the data required to
reconstruct the entire absorption spectrum are acquired at the same time. The spectrum as a
function of wavelength is generated by a built-in computer. This means that several gases can
be measured simultaneously. Several commercial instruments now are available with
5 resolutions of 0.06/cm or better. These instruments are capable of clearly defining the
spectrum of any gaseous pollutant, including CO and currently are being used for special air
pollution studies.
5.3.5 Intermittent Analysis
10 Intermittent samples may be collected in the field and later analyzed in the laboratory by
the continuous analyzing techniques described above. Sample containers may be rigid (glass
cylinders or stainless steel tanks) or they may be nonrigid (plastic bags). Because of location
or cost, intermittent sampling at times may be the only practical method for air monitoring.
Samples can be taken over a few minutes or accumulated intermittently to obtain, after
15 analysis, either "spot" or "integrated" results. Additional techniques for analyzing
intermittent samples are described below.
5.3.5.1 Colorimetric Analysis
Colored Silver Sol Method
20 Carbon monoxide reacts in an alkaline solution with the silver salt of p-sulfamoyl-
benzoate to form a colored silver sol. Concentrations of 12 to 23,000 mg/m3 (10 to
20,000 ppm) CO may be measured by this method (Ciuhandu, 1958, 1957, 1955; Ciuhandu
and Krall, 1960; Ciuhandu et al., 1965; Levaggi and Feldstein, 1964). The method has been
modified to determine CO concentrations in incinerator effluents. Samples are collected in an
25 evacuated flask and reacted. The absorbance of the resulting colloidal solution is measured
spectrophotometrically. Acetylene and formaldehyde interfere, but can be removed by
passing the sample through mercuric sulfate on silica gel. Carbon monoxide concentrations of
5.8 to 20,700 mg/m3 (5 to 18,000 ppm) may be measured with an accuracy of 90 to 100% of
the true value.
30
March 12, 1990 5-23 DRAFT - DO NOT QUOTE OR CITE
-------
National Bureau of Standards Colorimetric Indicating Gel
An NBS colorimetric-indicating gel (incorporating palladium and molybdenum salts) has
been devised to measure CO in the laboratory and in the field (Shepherd, 1947; Shepherd
et al., 1955). The laboratory method involves colorimetric comparison with freshly prepared
5 indicating gels exposed to known concentrations of CO. The method has an accuracy range
of 5 to 10% of the amount of CO involved, and the minimum detectable concentration is
1.2 mg/m3 (1 ppm). This technique requires relatively simple and inexpensive equipment;
however, oxidizing and reducing gases interfere, and the preparation of the indicator tube is a
tedious and time-consuming task.
10
Length-of-Stain Indicator Tube
An indicator tube that uses potassium palladosulfite is a commonly employed manual
method (Silverman and Gardner, 1965). Carbon monoxide reacts with the contents of the
tube and produces a discoloration.
15 The length of discoloration is an exponential function of the CO concentration. This
method and other indicator tube manual methods are estimated to be accurate to within ±25%
of the amount present, particularly at CO concentrations of about 115 mg/m3 (100 ppm).
Such indicator tube manual methods have been used frequently in air pollution studies.
Ramsey (1966) used the technique to measure CO at traffic intersections, and Brice and
20 Roesler (1966) estimated CO concentrations with an accuracy of ± 15% by means of color-
shade detector tubes.
Colorimetric techniques and length-of-stain discoloration methods are recommended for
use only when other physicochemical monitoring systems are not available. They may be
used in the field for gross mapping where accuracy is not required and may possibly be of
25 great value during emergencies.
Frontal Analysis
Air is passed over an adsorbent until equilibrium is established between the
concentration of CO in the air and the concentration of CO on the adsorbent. The CO then is
30 eluted with hydrogen, reduced to CH4 on a nickel catalyst at 250°C, and determined by flame
ionization as CH4.
March 12, 1990 5-24 DRAFT - DO NOT QUOTE OR CITE
-------
Concentrations of CO as low as 0.12 mg/m3 (0.10 ppm) can be measured. This method
does not give instantaneous concentrations, but does give averages over a six-minute or longer
sampling period (Dubois and Monkman, 1972; 1970).
5.4 MEASUREMENT USING PERSONAL MONITORS
Until the 1960s, most of the data available on ambient CO concentrations came from
fixed monitoring stations operated routinely in urban areas. The accepted measurement
technique was by NDIR spectrometry, but the instruments were large and cumbersome, often
10 requiring vibration-free, air-conditioned enclosures. Without a portable, convenient monitor
for CO, it was extremely difficult to measure CO concentrations accurately in the
microenvironments that people usually visited. In the late 1960s, studies were initiated to
investigate the CO concentrations within vehicles (Brice and Roesler, 1966; Lynn et al.,
1967). In 1971, an investigator walked on congested downtown streets alongside pedestrians
15 to measure their exposures (Ott, 1971). With a portable pump, the investigator filled
sampling bags in various locations, then transported them to the laboratory where the contents
were analyzed by NDIR spectrometry.
In the early 1970s, portable electrochemical monitors about the size of a shoe box
became available. Using the Ecolyzer monitor, CO concentrations were measured in traffic
20 in Boston, MA. (Cortese and Spengler, 1976). In the late 1970s, smaller personal monitors
using electrochemical sensing systems became available and were deployed in specialized field
surveys involving a few people (Jabara et al., 1980).
As CO monitors continued to evolve, they were used in studies of indoor
microenvironments. Many of the microenvironmental CO data on indoor concentrations were
25 collected as an integral part of multipollutant indoor health or dosage studies in homes,
offices, or rooms (Berglund et al., 1982; Hoffman et al., 1984; Hugod, 1984), or as more
narrowly focused multipollutant exposure field studies in homes (Quackenboss et al., 1984;
Koontz and Nagda, 1984; Traynor et al., 1984) and in buildings (Konopinski, 1984;
Malaspinaetal., 1984; Clarkson, 1984).
30 Although the CO personal monitors evolved rapidly, they were not used in large-scale
field surveys of indoor microenvironments until the early 1980s. Personal monitors have
March 12, 1990 5-25 DRAFT - DO NOT QUOTE OR CITE
-------
been used in studies of CO concentrations in sustained-use vehicles (Ziskind et al., 1981) and
in passenger compartments of vehicles traveling on highways (Ott and Willits, 1981;
Flachsbart and Ah Yo, 1986).
Ultimately, small personal exposure monitors were developed that could measure CO
5 concentrations continuously over time and store the readings automatically on internal digital
memories (Ott et al., 1986). These small personal exposure monitors (PEMs) made possible
the large-scale CO human exposure field studies in Denver, CO, and Washington, DC, in the
winter of 1982-1983 (Akland et al., 1985). The PEM employed in these studies uses an
aqueous solid polymer ion exchange material as the electrolyte in which CO is converted to
10 CO2 by an electrochemical reaction at a noble metal electrode, thereby generating an electrical
current. The signal (current) is proportional to the quantity of CO present in the gas stream,
and the continuous electrical signal is recorded in internal memory. A small pump operates
continuously to send air into the sensing cell, and chemical filters in the intake stream remove
interfering chemicals, such as ethanol. The pump operates on batteries for up to 40 h with a
15 precision of 2 ppm, with zero and span checks required before and after field service. Other
studies have employed the CO detector and combined it with small computers such as the HP-
41CV to enhance the utility of the detector for studying factors that affect CO concentration
variability (Fitz-Simons and Sauls, 1984). These monitors proved effective in generating
24-h CO exposure profiles on more than 1600 persons. By breaking up the profiles into the
20 microenvironments visited by these people, it was possible to develop CO concentration
readings on more than 40 indoor and in-transit microenvironments (see Chapter 8).
Detector tubes also can be used in studies where high concentrations occur (above
5 mg/m3) or long exposure times are possible and only cumulative exposures, are required.
Air is drawn through specifically manufactured tubes containing an absorbent impregnated
25 with a chemical reagent that changes color if CO is present (Jacobs, 1949). The length of the
stain produced in the tube after exposure is read on a chart to give the concentration of CO.
Unfortunately, interferences also may produce color changes, unless additional precautions are
taken to filter out particles and to absorb interfering gases such as oxides of nitrogen, SO2,
HCs, ammonia, hydrogen sulfide, and water vapor. Techniques such as the detector tube,
30 may have the greatest utility to the researcher by providing inexpensive approximate value for
March 12, 1990 5-26 DRAFT - DO NOT QUOTE OR CITE
-------
screening purposes, which would require confirmation found about some predetermined
"action" level.
5.5 REFERENCES
5
Acton, L. L.; Griggs, M.; Hall, G. D.; Ludwig, C. B.; Malkmus, W.; Hesketh, W. D.; Reichle, H. (1973)
Remote measurement of carbon monoxide by a gas filter correlation instrument. AIAA J. 11: 899-900.
10
Adams, E. G.; Simmons, N. T. (1951) The determination of carbon monoxide by means of iodine pentoxide. J.
Appl. Chem. l(suppl. 1): S20-S40.
Akland, G. G. (1972) Design of sampling schedules. J. Air Pollut. Control Assoc. 22: 264-266.
15
Akland, G. G.; Hartwell, T. D.; Johnson, T. R.; Whitmore, R. W. (1985) Measuring human exposure to carbon
monoxide in Washington, D.C., and Denver, Colorado, during the winter of 1982-1983. Environ. Sci.
Technol. 19: 911-918.
20 Allen, T. H.; Root, W. S. (1955) An improved palladium chloride method for the determination of carbon
monoxide in blood. J. Biol. Chem. 216: 319-323.
American Industrial Hygiene Association. (1972) Intersociety Committee methods for ambient air sampling and
analysis: report II. Am. Ind. Hyg. Assoc. J. 33: 353-359.
25
Bartle, E. R.; Hall, G. (1977) Airborne HC1 - CO sensing system: final report. La Jolla, CA: Science
Applications, Inc.; report no. SAI-76-717-LJ. Available from: NTIS, Springfield, VA; N77-21733.
Bay, H. W.; Blurton, K. F.; Lieb, H. C.; Oswin, H. G. (1972) Electrochemical measurement of carbon
30 monoxide. Am. Lab. 4: 57-58, 60-61.
Bay, H. W.; Blurton, K. F.; Sedlak, J. M.; Valentine, A. M. (1974) Electrochemical technique for the
measurement of carbon monoxide. Anal. Chem. 46: 1837-1839.
35 Beckman, A. O.; McCullough, J. D.; Crane, R. A. (1948) Microdetermination of carbon monoxide in air: a
portable instrument. Anal. Chem. 20: 674-677.
Bell, D. R.; Reiszner, K. D.; West, P. W. (1975) A permeation method for the determination of average
concentrations of carbon monoxide in the atmosphere. Anal. Chim. Acta 77: 245-254.
40
Bell, R. J. (1972) Introductory Fourier transform spectroscopy. New York, NY: Academic Press, Inc.
Berglund, B.; Johansson, L; Lindvall, T. (1982) A longitudinal study of air contaminants in a newly built
preschool. Environ. Int. 8: 111-115.
45
Bergman, I.; Coleman, J. E.; Evans, D. (1975) A simple gas chromatograph with an electrochemical detector for
the measurement of hydrogen and carbon monoxide in the parts per million range, applied to exhaled air.
Chromatographia 8: 581-583.
50 Brice, R. M.; Roesler, J. F. (1966) The exposure to carbon monoxide of occupants of vehicles moving in heavy
traffic. J. Air Pollut. Control Assoc. 16: 597-600.
March 12, 1990 5-27 DRAFT - DO NOT QUOTE OR CITE
-------
Bruner, F.; Ciccioli, P.; Rastelli, R. (1973) The determination of carbon monoxide in air in the parts per billion
range by means of a helium detector. J. Chromatogr. 77: 125-129.
Burch, D. E.; Gates, F. J.; Pembrook, J. D. (1976) Ambient carbon monoxide monitor. Research Triangle Park,
5 NC: U. S. Environmental Protection Agency, Environmental Sciences Research Laboratory; EPA report
no. EPA-600/2-76-210. Available from: NTIS, Springfield, VA; PB-259577.
Burch, D. E.; Gryvnak, D. A. (1974) Cross-stack measurement of pollutant concentrations using gas-cell
correlation spectroscopy. In: Stevens, R. K.; Herget, W. F., eds. Analytical methods applied to air
10 pollution measurements. Ann Arbor, MI: Ann Arbor Science Publishers, Inc.; pp. 193-231.
Chaney, L. W.; McClenny, W. A. (1977) Unique ambient carbon monoxide monitor based on gas filter
correlation: performance and application. Environ. Sci. Technol. 11: 1186-1190.
15 Chaney, L. W.; Rickel, D. G.; Russwurm, G. M.; McClenny, W. A. (1979) Long-path laser monitor of carbon
monoxide: system improvements. Appl. Opt. 18: 3004-3009.
Ciuhandu, Gh. (1955) O noua metoda de determinare a oxidului de carbon in aer [New method for the
determination of carbon monoxide in air]. Acad. Repub. Pop. Rom. Baza Cercet. Stiint. Timisoara Stud.
20 Cercet. Stiint. Ser. 1: 133-142.
Ciuhandu, G. (1957) Photometrische Bestimmung von Kohlenmonoxyd in der Luft [Photometric determination of
carbon monoxide in air]. Fresenius' Z. Anal. Chem. 155: 321-327.
25 Ciuhandu, G. (1958) Mikromethode fuer die photometrische Kohlenmonoxydbestimmung in der Luft
[Micromethod for the photometric determination of carbon monoxide in air]. Fresenius' Z. Anal. Chem.
161: 345-348.
Ciuhandu, G.; Krall, G. (1960) Photometrische Bestimmung von Kohlenmonoxydspuren in Wasserstoff
30 [Photometric determination of traces of carbon monoxide in hydrogen]. Fresenius' Z. Anal. Chem.
172: 81-87.
Ciuhandu, G.; Rusu, V.; Diaconovici, M. (1965) Bestimmung von Kohlenmonoxidspuren in Sauerstoff
[Determination of traces of carbon monoxide in oxygen]. Fresenius' Z. Anal. Chem. 208: 81-86.
35
Clarkson, M. (1984) Indoor air quality as a part of total building performance. In: Berglund, B.; Lindvall, T.;
Sundell, J., eds. Indoor air: proceedings of the 3rd international conference on indoor air quality and
climate, v. 5, buildings, ventilation and thermal climate; August; Stockholm, Sweden. Stockholm,
Sweden: Swedish Council for Building Research; pp. 493-498. Available from: NTIS, Springfield, VA;
40 PB85-104222.
Code of Federal Regulations. (1977a) Ambient air monitoring reference and equivalent methods. C. F. R.
40: 812-839.
45 Code of Federal Regulations. (1977b) National primary and secondary ambient air quality standards. C. F. R.
40(50) (July 1): 3-33.
Code of Federal Regulations. (1988) List of designated reference and equivalent methods. C. F. R. 53: 812-839.
50 Commins, B. T.; Berlin, A.; Langevin, M.; Peal, J. A. (1977) Intercomparison of measurement of
carboxyhaemoglobin in different European laboratories and establishment of the methodology for the
assessment of COHb levels in exposed populations. Luxembourg, Sweden: Commission of the European
Communities, Health and Safety Directorate; doc. V/F/1315/77e.
March 12, 1990 5-28 DRAFT - DO NOT QUOTE OR CITE
-------
Cormack, R. S. (1972) Eliminating two sources of error in the Lloyd-Haldane apparatus. Respir. Physiol.
14: 382-390.
Cortese, A. D.; Spengler, J. D. (1976) Ability of fixed monitoring stations to represent personal carbon
5 monoxide exposure. J. Air Pollut. Control Assoc. 26: 1144-1150.
Dagnall, R. M.; Johnson, D. J.; West, T. S. (1973) A method for the determination of carbon monoxide, carbon
dioxide, nitrous oxide and sulphur dioxide in air by gas chromatography using an emissive helium plasma
detector. Spectrosc. Lett. 6: 87-95.
10
Dailey, W. V.; Fertig, G. H. (1977) A novel NDIR analyzer for NO, SO2 and CO analysis. Anal. Instrum.
15: 79-82.
Dempsey, R. M.; LaConti, A. B.; Nolan, M. E.; Torkildsen, R. A.; Schnakenberg, G.; Chilton, E. (1975)
15 Development of fuel cell CO detection instruments for use in a mine atmosphere. Washington, DC: U.S.
Department of the Interior, Bureau of Mines; report no. 77-76. Available from: NTIS, Springfield, VA;
PB-254823.
Driscoll, J. N.; Berger, A. W. (1971) Improved chemical methods for sampling and analysis of gaseous
20 pollutants from the combustion of fossil fuels. Volume III. Carbon monoxide. Cincinnati, OH: U.S.
Environmental Protection Agency, Office of Air Programs; report no. APTD-1109. Available from:
NTIS, Springfield, VA; PB-209269.
Dubois, L.; Monkman, J. L. (1970) L'emploi de Panalyse frontale pour 1'echantillonage et le dosage de 1'oxyde
25 de carbone dans 1'air [Sampling by frontal analysis and determination of carbon monoxide]. Mikrochim.
Acta27: 313-320.
Dubois, L.; Monkman, J. L. (1972) Continuous determination of carbon monoxide by frontal analysis. Anal.
Chem. 44: 74-76.
30
Dubois, L.; Zdrojewski, A.; Monkman, J. L. (1966) The analysis of carbon monoxide in urban air at the ppm
level, and the normal carbon monoxide value. J. Air Pollut. Control Assoc. 16: 135-139.
Elwood, J. H. (1976) A calibration procedural system and facility for gas turbine engine exhaust emission
35 analysis. In: Calibration in air monitoring, a symposium; August 1975; Boulder, CO. Philadelphia, PA:
American Society for Testing and Materials; ASTM special technical publication no. 598; pp. 255-272.
Federal Register. (1978) Air quality surveillance and data reporting: proposed regulatory revisions. F. R.
(August 7) 43: 34892-34934.
40
Feldstein, M. (1965) The colorimetric determination of blood and breath carbon monoxide. J. Forensic Sci.
10: 35-42.
Feldstein, M. (1967) Methods for the determination of carbon monoxide. Prog. Chem. Toxicol. 3: 99-119.
45
Fitz-Simons, T.; Sauls, H. B. (1984) Using the HP-41CV calculator as a data acquisition system for personal
carbon monoxide exposure monitors. J. Air Pollut. Control Assoc. 34: 954-956.
Flachsbart, P. G.; Ah Yo, C. (1986) Test of a theoretical commuter exposure model to vehicle exhaust in traffic.
50 Presented at: 79th annual meeting of the Air Pollution Control Association; June; Minneapolis, MN.
Pittsburgh, PA: Air Pollution Control Association; paper no. 86-79.4.
Garbarz, J.-J.; Sperling, R. B.; Peache, M. A. (1977) Time-integrated urban carbon monoxide measurements.
San Francisco, CA: Environmental Measurements, Inc.
March 12, 1990 5-29 DRAFT - DO NOT QUOTE OR CITE
-------
Golden, B. M.; Yeung, E. S. (1975) Analytical lines for long-path infrared absorption spectrometry of air
pollutants. Anal. Chem. 47: 2132-2135.
Goldstein, G. M. (1977) [Letter to Mr. Jack Thompson concerning carboxyhemoglobin measurements]. Research
5 Triangle Park, NC: U.S. Environmental Protection Agency; September 23, 1977.
Goldstein, S. A.; D'Silva, A. P.; Fassel, V. A. (1974) X-ray excited optical fluorescence of gaseous atmospheric
pollutants: analytical feasibility study. Radial. Res. 59: 422-437.
10 Goldstein, H. W.; Bortner, M. H.; Grenda, R. N.; Dick, R.; Barringer, A. R. (1976) Correlation interferometric
measurement of carbon monoxide and methane from the Canada Centre for Remote Sensing Falcon
Fan-Jet Aircraft. Can. J. Remote Sens. 2: 30-41.
Goodman, P. (1972) Measurement of automobile exhaust pollutant concentrations by use of solid kryptonates.
15 Highw. Res. Rec. (412): 9-12.
Gryvnak, D. A.; Burch, D. E. (1976a) Monitoring of pollutant gases in aircraft exhausts by gas- filter correlation
methods. In: Proceedings of the AIAA 14th aerospace sciences meeting; January; Washington, DC.
New York, NY: American Institute of Aeronautics and Astronautics; AIAA paper no. 76-110.
20
Gryvnak, D. A.; Burch, D. E. (1976b) Monitoring NO and CO in aircraft jet exhaust by a gas-filter correlation
technique. Wright-Patterson AFB, OH: U. S. Air Force, Aero-propulsion Laboratory; report
no. AFAPL-TR-75-101. Available from: NTIS, Springfield, VA; AD-A022353.
25 Haagen-Smit, A. J. (1966) Carbon monoxide levels in city driving. Arch. Environ. Health 12: 548-551.
Hanst, P. L.; Lefohn, A. S.; Gay, B. W., Jr. (1973) Detection of atmospheric pollutants at parts-per-billion
levels by infrared spectroscopy. Appl. Spectrosc. 27: 188-198.
30 Harrison, N. (1975) A review of techniques for the measurement of carbon monoxide in the atmosphere. Ann.
Occup. Hyg. 18: 37-44.
Herget, W. F.; Jahnke, J. A.; Burch, D. E.; Gryvnak, D. A. (1976) Infrared gas-filter correlation instrument for
in situ measurement of gaseous pollutant concentrations. Appl. Opt. 15: 1222-1228.
35
Hersch, P. (1964) Galvanic analysis. In: Reilley, C. N., ed. Advances in analytical chemistry and
instrumentation, v. 3. New York, NY: Interscience Publishers; pp. 183-249.
Hersch, P. A. (1966) Method and apparatus for measuring the carbon monoxide content of a gas stream.
40 Washington, DC: U. S. Patent Office; patent no. 3,258,411; June 28, 1966.
Hoffmann, D.; Brunnemann, K. D.; Adams, J. D.; Haley, N. J. (1984) Indoor air pollution by tobacco smoke:
model studies on the uptake by nonsmokers. In: Berglund, B.; Lindvall, T.; Sundell, J., eds. Indoor air:
proceedings of the 3rd international conference on indoor air quality and climate, v. 2, radon, passive
45 smoking, particulates and housing epidemiology; August; Stockholm, Sweden. Stockholm, Sweden:
Swedish Council for Building Research; pp. 313-318.
Houben, W. P. (1976) Continuous monitoring of carbon monoxide in the atmosphere by an improved NDIR
method. In: APCA specialty conference on air pollution measurement accuracy as it relates to regulation
50 compliance; October 1975; New Orleans, LA. Pittsburgh, PA: Air Pollution Control Association;
pp. 55-64.
Hrubesh, L. W. (1973) Microwave rotational spectroscopy: a technique for specific pollutant monitoring. Radio
Sci. 8: 167-175.
March 12, 1990 5-30 DRAFT - DO NOT QUOTE OR CITE
-------
Hughes, E. E. (1975) Development of standard reference materials for air quality measurement. ISA Trans.
14: 281-291.
Hughes, E. E. (1976) Role of the National Bureau of Standards in calibration problems associated with air
5 pollution measurements. In: Calibration in air monitoring, a symposium; August 1975; Boulder, CO.
Philadelphia, PA: American Society for Testing and Materials; ASTM special technical publication
no. 598; pp. 223-231.
Hughes, E. E.; Dorko, W. D.; Scheide, E. P.; Hall, L. C.; Beilby, A. L.; Taylor, J. K. (1973) Gas generation
10 systems for the evaluation of gas detecting devices. Washington, DC: U. S. Department of Commerce,
National Bureau of Standards; report no. NBSIR 73-292.
Hugod, C. (1984) Passive smoking - a source of indoor air pollution. In: Berglund, B.; Lindvall, T.; Sundell, J.,
eds. Indoor air: proceedings of the 3rd international conference on indoor air quality and climate, v. 2,
15 radon, passive smoking, particulates and housing epidemiology; August; Stockholm, Sweden. Stockholm,
Sweden: Swedish Council for Building Research; pp. 319-325.
Hunt, W. F., Jr. (1972) The precision associated with the sampling frequency of log-normally distributed air
pollutant measurements. J. Air Pollut. Control Assoc. 22: 687-691.
20
Jabara, J. W.; Beaulieu, H. J.; Buchan, R. M.; Keefe, T. J. (1980) Carbon monoxide: dosimetry in occupational
exposures in Denver, Colorado. Arch. Environ. Health 35: 198-204.
Jacobs, M. B. (1949) The analytical chemistry of industrial poisons, hazards, and solvents. 2nd ed. New York,
25 NY: Wiley Interscience.
Jones, J. K. (1977) Determination of carbon monoxide by means of a palladium-promazine complex. J. Anal.
Toxicol. 1: 54-56.
30 Konopinski, V. J. (1984) Residential formaldehyde and carbon dioxide. In: Berglund, B.; Lindvall, T.; Sundell,
J., eds. Indoor air: proceedings of the 3rd international conference on indoor air quality and climate,
v. 3, sensory and hyperreactivity reactions to sick buildings; August; Stockholm, Sweden. Stockholm,
Sweden: Swedish Council for Building Research; pp. 329-334. Available from: NTIS, Springfield, VA;
PB85-104206.
35
Koontz, M. D.; Nagda, N. L. (1984) Infiltration and air quality in well-insulated homes: 3. measurement and
modeling of pollutant levels. In: Berglund, B.; Lindvall, T.; Sundell, J., eds. Indoor air: proceedings of
the 3rd international conference on indoor air quality and climate, v. 5, buildings, ventilation and thermal
climate; August; Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building Research;
40 pp. 511-516. Available from: NTIS, Springfield, VA; PB85-104222.
Lambert, J. L.; Wiens, R. E. (1974) Induced colorimetric method for carbon monoxide. Anal. Chem.
46: 929-930.
45 Larsen, R. I. (1971) A mathematical model for relating air quality measurements to air quality standards.
Research Triangle Park, NC: U. S. Environmental Protection Agency, Office of Air Programs;
report AP-89.
Lawrence Berkeley Laboratory. (1973) CO monitoring methods and instrumentation. In: Instrumentation for
50 environmental monitoring air: v. 1. Berkeley, CA: University of California, Lawrence Berkeley
Laboratory.
March 12, 1990 5-31 DRAFT - DO NOT QUOTE OR CITE
-------
Leithe, W. (1971) The analysis of air pollutants. Ann Arbor, MI: Ann Arbor Science Publishers, Inc. Levaggi,
D. A.; Feldstein, M. (1964) The colorimetric determination of low concentrations of carbon monoxide.
Am. Ind. Hyg. Assoc. J. 25: 64-66.
5 Link, W. T.; McClatchie, E. A.; Watson, D. A.; Compher, A. B. (1971) A fluorescent source NDIR carbon
monoxide analyzer. Presented at: the joint conference on sensing of environmental pollutants; November;
Palo Alto, CA. New York, NY: American Institute of Aeronautics and Astronautics; AIAA paper
no. 71-1047.
10 Luft, K. F. (1975) Infrared techniques for the measurement of carbon monoxide. Ann. Occup. Hyg. 18: 45-51.
Lynn, D. A.; Ott, W.; Tabor, E. C.; Smith, R. (1967) Present and future commuter exposure to carbon
monoxide. Presented at: 60th annual meeting of the Air Pollution Control Association; June; Cleveland,
OH. Pittsburgh, PA: Air Pollution Control Association; paper no. 67-5.
15
Malaspina, J.-P.; Bodilis, H.; Giacomoni, L.; Marble, G. (1984) Indoor air pollution: study of two buildings in
the Paris area. In: Berglund, B.; Lindvall, T.; Sundell, J., eds. Indoor air: proceedings of the 3rd
international conference on indoor air quality and climate, v. 5, buildings, ventilation and thermal
climate; August; Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building Research;
20 pp. 499-504. Available from: NTIS, Springfield, VA; PB85-104222.
McClatchie, E. A. (1972) Development of an infrared fluorescent gas analyzer. Research Triangle Park, NC: U.
S. Environmental Protection Agency, Office of Air Quality Planning and Standards; EPA report
no. EPA-R2-72-121. Available from: NTIS, Springfield, VA; PB-213846.
25
McClatchie, E. A.; Compher, A. B.; Williams, K. G. (1972) A high specificity carbon monoxide analyzer. Anal.
Instrum. 10: 67-69.
McCullough, J. D.; Crane, R. A.; Beckman, A. O. (1947) Determination of carbon monoxide in air by use of
30 red mercuric oxide. Anal. Chem. 19: 999-1002.
McKee, H. C.; Childers, R. E. (1972) Collaborative study of reference method for the continuous measurement
of carbon monoxide in the atmosphere (nondispersive infrared spectrometry). Houston, TX: Southwest
Research Institute. Available from: NTIS, Springfield, VA; PB-211265.
35
McKee, H. C.; Margeson, J. H.; Stanley, T. W. (1973) Collaborative testing of methods to measure air
pollutants. II. The non-dispersive infrared method for carbon monoxide. J. Air Pollut. Control Assoc.
23: 870-875.
40 Moore, J.; West, T. S.; Dagnall, R. M. (1973) A microwave plasma detector system for the measurement of
trace levels of carbon monoxide in air. Proc. Soc. Anal. Chem. 10: 197.
Morgan, M. G.; Morris, S. C. (1977) Needed: a national R&D effort to develop individual air pollution monitor
instrumentation. J. Air Pollut. Control Assoc. 27: 670-673.
45
Mueller, R. H. (1954) A supersensitive gas detector permits accurate detection of toxic or combustible gases in
extremely low concentrations. Anal. Chem. 26: 39A-42A.
Naoum, M. M.; Pruzinec, J.; Toelgyessy, J.; Klehr, E. H. (1974) Contribution to the determination of carbon
50 monoxide in air by the radio-release method. Radiochem. Radioanal. Lett. 17: 87-93.
National Air Pollution Control Administration. (1970) Air quality criteria for carbon monoxide. Washington,
DC: U. S. Department of Health, Education, and Welfare, Public Health Service; report
no. NAPCA-PUB-AP-62. Available from: NTIS, Springfield, VA; PB-190261.
March 12, 1990 5-32 DRAFT - DO NOT QUOTE OR CITE
-------
National Bureau of Standards. (1975) Catalog of NBS standard reference materials, 1975-76 edition. Washington,
DC: U. S. Department of Commerce, National Bureau of Standards; NBS special publication no. 260.
National Institute for Occupational Safety and Health. (1972) Criteria for a recommended standard....occupational
5 exposure to carbon monoxide. Washington, DC: U. S. Department of Health, Education, and Welfare;
report no. NIOSH-TR-007-72. Available from: NTIS, Springfield, VA; PB-212629.
National Research Council. (1977) Carbon monoxide. Washington, DC: National Academy of Sciences. (Medical
and biologic effects of environmental pollutants).
10
Naumann, R. J. (1975) Carbon monoxide monitor. Washington, DC: U. S. Patent Office; patent no. 3,895,912;
July 22, 1975.
Newton, C.; Morss, L. R. (1974) Portable apparatus for determining atmospheric carbon monoxide. Chemistry
15 47: 27-29.
O'Keeffe, A. E.; Ortman, G. C. (1966) Primary standards for trace gas analysis. Anal. Chem. 38: 760-763.
Optical Society of America. (1975) Applications of laser spectroscopy. In: Proceedings of the Spring conference;
20 March; Anaheim, CA. Washington, DC: Optical Society of America.
Ott, W. R. (1971) An urban survey technique for measuring the spatial variation of carbon monoxide
concentrations in cities. Stanford, CA: Stanford University, Dept. of Civil Engineering.
25 Ott, W. R.; Mage, D. T. (1975) Random sampling as an inexpensive means for measuring average annual air
pollutant concentrations in urban areas. Presented at: 68th annual meeting of the Air Pollution Control
Association; June; Boston, MA. Pittsburgh, PA: Air Pollution Control Association; paper no. 75-14.3.
Ott, W. R.; Willits, N. H. (1981) CO exposures of occupants of motor vehicles: modeling the dynamic response
30 of the vehicle. Stanford, CA: Stanford University, Dept. of Statistics; SIMS technical report no. 48.
Ott, W. R.; Rodes, C. E.; Drago, R. J.; Williams, C.; Burmann, F. J. (1986) Automated data-logging personal
exposure monitors for carbon monoxide. J. Air Pollut. Control Assoc. 36: 883-887.
35 Palanos, P. N. (1972) A practical design for an ambient carbon monoxide mercury replacement analyzer. Anal.
Instrum. 10: 117-125.
Paulsell, C. D. (1976) Use of the National Bureau of Standards standard reference gases in mobile source
emissions testing. In: Calibration in air monitoring, a symposium; August 1975; Boulder, CO.
40 Philadelphia, PA: American Society for Testing and Materials; ASTM special technical publication 598;
pp. 232-245.
PEDCo Environmental Specialists, Inc. (1971) Field operations guide for automatic air monitoring equipment.
Research Triangle Park, NC: U. S. Environmental Protection Agency; report no. APTD-0736. Available
45 from: NTIS, Springfield, VA; PB-202249.
Perez, J. M.; Broering, L. C.; Johnson, J. H. (1975) Cooperative evaluation of techniques for measuring nitric
oxide and carbon monoxide (Phase IV tests). Presented at: the Automotive Engineering Congress and
Exposition; February; Detroit, MI. Warrendale, PA: Society of Automotive Engineers, Inc.; SAE
50 technical paper no. 750204.
Phinney, D. E.; Newman, J. E. (1972) The precision associated with the sampling frequencies of total particulate
at Indianapolis, Indiana. J. Air Pollut. Control Assoc. 22: 692-695.
March 12, 1990 5-33 DRAFT - DO NOT QUOTE OR CITE
-------
Pierce, J. O.; Collins, R. J. (1971) Calibration of an infrared analyzer for continuous measurement of carbon
monoxide. Am. Ind. Hyg. Assoc. J. 32: 457-462.
Poli, A. A.; Dessert, C. J.; Benzie, T. P. (1976) Carbon monoxide detection apparatus and method. Washington,
5 DC: U. S. Patent Office; patent no. 4,030,887; June 18, 1976.
Porter, K.; Volman, D. H. (1962) Flame ionization detection of carbon monoxide for gas chromatographic
analysis. Anal. Chem. 34: 748-749.
10 Quackenboss, J. J.; Kanarek, M. S.; Kaarakka, P.; Duffy, C. P.; Flickinger, J.; Turner, W. A. (1984)
Residential indoor air quality, structural leakage and occupant activities for 50 Wisconsin homes. In:
Berglund, B.; Lindvall, T.; Sundell, J., eds. Indoor air: proceedings of the 3rd international conference
on indoor air quality and climate, v. 5, buildings, ventilation and thermal climate; August; Stockholm,
Sweden. Stockholm, Sweden: Swedish Council for Building Research; pp. 411-420. Available from:
15 NTIS, Springfield, VA; PB85-104222.
Ramsey, J. M. (1966) Concentrations of carbon monoxide at traffic intersections in Dayton, Ohio. Arch.
Environ. Health 13: 44-46.
20 Ray, R. M.; Carroll, H. B.; Armstrong, F. E. (1975) Evaluation of small, color-changing carbon monoxide
dosimeters. In: Bureau of Mines report of investigations 8051. Washington, DC: U. S. Department of the
Interior, Bureau of Mines.
Repp, M. (1977) Evaluation of continuous monitors for carbon monoxide in stationary sources. Research Triangle
25 Park, NC: U. S. Environmental Protection Agency, Environmental Science Research Laboratory; EPA
report no. EPA-600/2-77-063. Available from: NTIS, Springfield, VA; PB-268861.
Rhodes, R. C.; Evans, E. G. (1987) Precision and accuracy assessments for state and local air monitoring
networks, 1985. Research Triangle Park, NC: U. S. Environmental Protection Agency, Environmental
30 Monitoring Systems Laboratory; EPA report no. EPA-600/4-87-003. Available from: NTIS, Springfield,
VA; PB87-145447.
Robbins, R. C.; Borg, K. M.; Robinson, E. (1968) Carbon monoxide in the atmosphere. J. Air Pollut. Control
Assoc. 18: 106-110.
35
Saltzman, B. E. (1972) Simplified methods for statistical interpretation of monitoring data. J. Air Pollut. Control
Assoc. 22: 90-95.
Scaringelli, F. P.; Rosenberg, E.; Rehme, K. A. (1970) Comparison of permeation devices and nitrite ion as
40 standards for the colorimetric determination of nitrogen dioxide. Environ. Sci. Technol. 4: 924-929.
Schnakenberg, G. H. (1975) Gas detection instrumentation...what's new and what's to come. Coal Age
80: 84-92.
45 Schnakenberg, G. H. (1976) Improvements in coal mine gas detection instrumentation. Pap. Symp. Underground
Min. 2: 206-216.
Schunck, G. (1976) Nichtdispersive Infrarot-Gasanalysatoren fuer Industrieprozesse und Umweltschutz
[Nondispersive infrared gas analyzer for industrial processes and protection of the environment].
50 DECHEMA Monogr. 80: 753-761.
Scott, B. (1975) Development of an optical carbon monoxide detector. Washington, DC: U. S. Department of
the Interior, Bureau of Mines; report no. 93-75. Available from: NTIS, Springfield, VA; PB-246570.
March 12, 1990 5-34 DRAFT - DO NOT QUOTE OR CITE
-------
Seller, W.; Junge, C. (1970) Carbon monoxide in the atmosphere. JGR J. Geophys. Res. 75: 2217-2226.
Shepherd, M. (1947) Rapid determination of small amounts of carbon monoxide: preliminary report on the NBS
colorimetric indicating gel. Anal. Chem. 19: 77-81.
5
Shepherd, M.; Schuhmann, S.; Kilday, M. V. (1955) Determination of carbon monoxide in air pollution studies.
Anal. Chem. 27: 380-383.
Silverman, L.; Gardner, G. R. (1965) Potassium pallado sulfite method for carbon monoxide detection. Am. Ind.
10 Hyg. Assoc. J. 26: 97-105.
Simonescu, T.; Rusu, V.; Kiss, L. (1975) Noi aplicatii analitice ale unor compusi sulfonamidici: metoda cinetica
de determinare a oxidului de carbon din aer [New analytical applications of some sulfonamide
compounds: kinetic method for the determination of carbon monoxide in air]. Rev. Chim. (Bucharest)
15 26: 75-78.
Smith, F.; Nelson, A. C., Jr. (1973) Guidelines for development of a quality assurance program: reference
method for the continuous measurement of carbon monoxide in the atmosphere. Research Triangle Park,
NC: U. S. Environmental Protection Agency, Quality Assurance and Environmental Monitoring
20 Laboratory; EPA report no. EPA-R4-73-028a. Available from: NTIS, Springfield, VA; PB-222512.
Smith, R. G. (1969) Air quality standards for carbon monoxide. New York, NY: American Petroleum Institute,
Division of Environmental Affairs; air quality monograph no. 69-9.
25 Smith, R. G.; Bryan, R. J.; Feldstein, M.; Locke, D. C.; Warner, P. O. (1975a) Tentative method for constant
pressure volumetric gas analysis for O2, CO2, CO, N2, hydrocarbons (ORSAT). Health Lab. Sci.
12: 177-181.
Smith, R. G.; Bryan, R. J.; Feldstein, M.; Locke, D. C.; Warner, P. O. (1975b) Tentative method for gas
30 chromatographic analysis of O2, N2, CO, CO2, and CH4. Health Lab. Sci. 12: 173-176.
Stedman, D. H.; Kok, G.; Delumyea, R.; Alvord, H. H. (1976) Redundant calibration of nitric oxide, carbon
monoxide, nitrogen dioxide, and ozone air pollution monitors by chemical and gravimetric techniques. In:
Calibration in air monitoring, a symposium; August 1975; Boulder, CO. Philadelphia, PA: American
35 Society for Testing and Materials; ASTM special technical publication 598; pp. 337-344.
Stetter, J. R.; Blurton, K. F. (1976) Portable high-temperature catalytic reactor: application to air pollution
monitoring instrumentation. Rev. Sci. Instrum. 47: 691-694.
40 Stevens, R. K.; Herget, W. F. (1974) Analytical methods applied to air pollution measurements. Ann Arbor, MI:
Ann Arbor Science Publishers, Inc.
Swinnerton, J. W.; Linnenbom, V. J.; Cheek, C. H. (1968) A sensitive gas chromatographic method for
determining carbon monoxide in seawater. Limnol. Oceanogr. 13: 193-195.
45
Tesarik, K.; Krejci, M. (1974) Chromatographic determination of carbon monoxide below the 1 ppm level. J.
Chromatogr. 91: 539-544.
Traynor, G. W.; Apte, M. G.; Carruthers, A. R.; Dillworth, J. F.; Grimsrud, D. T.; Thompson, W. T. (1984)
50 Indoor air pollution and inter-room transport due to unvented kerosene-fired space heaters. Berkeley, CA:
Lawrence Berkeley Laboratory; report no. LBL-17600. Available from: NTIS, Springfield, VA;
DE84015949.
March 12, 1990 5-35 DRAFT - DO NOT QUOTE OR CITE
-------
U. S. Senate Committee on Environment and Public Works. (1977) The Clean Air Act as amended August 1977.
Washington, DC: U.S. Government Printing Office; serial no. 95-11.
v. Heusden, S.; Hoogeveen, L. P. J. (1976) Chemiluminescent determination of reactive hydrocarbons. Z. Anal.
5 Chem. 282: 307-313.
van Dijk, J. F. M.; Falkenburg, R. A. (1976) A high sensitivity carbon monoxide monitor for ambient air. In:
International conference on environmental sensing and assessment. Volume 2. A joint conference
comprising the International Symposium on Environmental Monitoring and Third Joint Conference on
10 Sensing of Environmental Pollutants; September 1975; Las Vegas, NV. New York, NY: Institute of
Electrical & Electronics Engineers, Inc.; paper no. 35-5.
Verdin, A. (1973) Gas analysis instrumentation. New York, NY: John Wiley & Sons, Inc.
15 Vol'berg, N. Sh.; Pochina, I. I. (1974) Continuous determination of the concentration of carbon monoxide in the
atmosphere by the coulometric method. In: Voprosy atmosfernoy diffuzii i zagryazneniya vozdukha
[Atmospheric diffusion and air pollution problems]. Leningrad, USSR: Gidrometeoizdat Press;
pp. 146-157.
20 Ward, T. V.; Zwick, H. H. (1975) Gas cell correlation spectrometer: GASPEC. Appl. Opt. 14: 2896-2904.
Wechter, S. G. (1976) Preparation of stable pollution gas standards using treated aluminum cylinders. In:
Calibration in air monitoring, a symposium; August 1975; Boulder, CO. Philadelphia, PA: American
Society for Testing and Materials; ASTM special technical publication 598; pp. 40-54.
25
Wohlers, H. C.; Newstein, H.; Daunis, D. (1967) Carbon monoxide and sulfur dioxide adsorption on- and
desorption from glass, plastic, and metal tubings. J. Air Pollut. Control Assoc. 17: 753-756.
Yamate, N.; Inoue, A. (1973) Continuous analyzer of carbon monoxide in ambient air using electrochemical
30 analysis. Kogai to Taisaku 9: 292-296.
Ziskind, R. A.; Rogozen, M. B.; Carlin, T.; Drago, R. (1981) Carbon monoxide intrusion into sustained-use
vehicles. Environ. Int. 5: 109-123.
March 12, 1990 5-36 DRAFT - DO NOT QUOTE OR CITE
-------
6. AMBIENT CARBON MONOXIDE SOURCES,
EMISSIONS, AND CONCENTRATIONS
5 6.1 ESTIMATING NATIONAL EMISSION FACTORS
The national carbon monoxide emission estimates presented herein are taken from two
U.S. Environmental Protection Agency (EPA) reports: National Air Quality and Emissions
Trends Report, 1988 (U.S. Environmental Protection Agency, 1990a) and National Air
Pollutant Emission Estimates, 1940-1988 (U.S. Environmental Protection Agency, 1990b).
10 These data are most useful as indicators of overall emission trends, since national totals or
averages are not the best guide for estimating or predicting specific trends in local areas. The
emission data represent calculated estimates based on standard emission inventory procedures
developed by the Office of Air Quality Planning and Standards of the U.S. Environmental
Protection Agency (1990b). These procedures either estimate the emissions directly or
15 estimate the magnitude of other variables that can then be related to emissions. For CO,
these indicators include fuel consumption, vehicle population, vehicle miles traveled (VMT),
sales of new vehicles, tons of refuse burned, raw materials processed, etc., which are then
multiplied by the appropriate CO emission factor(s) to obtain the CO emission estimate(s). It
should be noted that emission factors have specific limitations and applicability. Emission
20 factors, in general, are not precise indicators of emissions from a single source; rather, they
are quantitative estimates of the average rate of pollutant released as a result of some activity.
They are most valid when applied to a large number of sources and processes. Emission
factors thus relate quantity of pollutants emitted to indicators such as those noted above, and
are EPA's approach for determining national estimates of emissions from various source
25 categories.
6.2 EMISSION SOURCES AND EMISSION FACTORS BY SOURCE
CATEGORY
30 Emission source categories, as presented in Table 6-1, are divided into five individual
categories: transportation, stationary source fuel combustion, industrial processes, solid waste
March 12, 1990 6-1 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 6-1. CARBON MONOXIDE NATIONAL EMISSION ESTIMATES (TERAGRAMS/YEAR)
K>
VO
£
to
Source Category
Transportation
Highway Vehicles
Aircraft
Railroads
Vessels
Other-Off Highway
Transportation Total
Stationary Source Fuel Combustion
Electric Utilities
Industrial
Commercial-Institutional
Residential
Fuel Combustion Total
Industrial Processes
Solid Waste Disposal
Incineration
Open Burning
Solid Waste Total
Miscellaneous
Forest Fires
Other Burning
Miscellaneous Organic Solvent
Miscellaneous Total
Total of All Sources
1970
65.3
0.9
0.3
1.2
6.8
74.4
0.2
0.7
0.1
3.5
4.5
8.9
2.7
3.7
6.4
5.1
2.1
0.0
7.2
101.4
1975
57.2
0.9
0.2
1.4
5.4
65.0
0.3
0.7
0.1
3.3
4.3
6.9
1.8
1.3
3.1
4.0
0.8
0.0
4.8
84.1
1978
55.6
1.0
0.3
1.5
4.8
63.1
0.3
0.8
0.1
4.8
5.9
7.2
1.4
1.1
2.5
5.0
0.7
0.0
5.7
84.4
1979
51.9
1.0
0.3
1.4
4.5
59.1
0.3
0.7
0.1
5.7
6.7
7.1
1.3
1.0
2.3
5.8
0.7
^Q
6.5
81.7
1980
48.7
1.0
0.3
1.4
4.7
56.1
0.3
0.7
0.1
6.4
7.4
6.3
1.2
1.0
2.2
6.9
0.7
0.0
7.6
79.6
1981
48.0
1.0
0.3
1.4
4.7
55.4
0.3
0.6
0.1
6.7
7.7
5.9
1.2
0.9
2.1
5.8
0.6
0.0
6.4
77.4
1982
45.9
1.0
0.2
1.4
4.4
52.9
0.3
0.6
0.1
7.3
8.2
4.3
1.1
0.9
2.0
4.3
0.6
0.0
4.9
72.4
1983
45.9
1.0
0.2
1.4
3.9
52.4
0.3-
0.6
0.1
i2
8.2
4.3
1.0
.09
1.9
7.1
0.6
0.0
7.7
74.5
1984
43.5
1.0
0.2
1.7
4.2
50.6
0.3
0.6
0.1
JL3
8.3
4.7
1.0
.02
1.9
5.7
0.6
0.0
6.3
71.8
1985
40.7
1.1
0.2
1.4
4.5
47.9
0.3
0.6
0.1
^5
7.4
4.4
1.1
.09
2.0
4.7
0.6
0.0
5.3
67.0
1986
37.5
1.1
0.2
1.5
4.4
44.6
0.3
0.6
0.1
A6
7.5
4.3
0.9
.08
1.7
4.4
0.6
0.0
5.0
63.1
1987
36.0
1.1
0.2
1.6
4.4
43.2
0.3
0.6
0.1
^6
7.6
4.5
0.9
0.8
1.7
6.5
0.6
0.0
7.1
64.1
1988
34.1
1.1
0.2
1.6
4.2
41.2
0.3
0.6
0.1
^6
7.6
4.7
0.9
0.8
1.7
5.4
0.6
_OQ
6.0
61.2
Source: U.S. Environmental Protection Agency (1990b).
-------
disposal, and miscellaneous. The methodology used in the generation of emission estimates
for the individual source categories is summarized below.
6.2.1 Transportation Sources
5 Transportation sources include emissions from all mobile sources including highway and
other off-highway motor vehicles. Highway motor vehicles include passenger cars, trucks,
buses and motorcycles. Off-highway vehicles include aircraft, railroads, vessels, and
miscellaneous engines such as farm equipment, industrial and construction machinery,
lawnmowers, and snowmobiles.
10
6.2.1.1 Motor Vehicles
Emission estimates from gasoline- and diesel-powered motor vehicles are based upon
vehicle-mile tabulations and emission factors. Eight vehicle categories are considered; light
duty gasoline (mostly passenger cars), light duty diesel passenger cars, light duty gasoline
15 trucks (weighing less than 6000 pounds) light duty gasoline trucks (weighing 6001 to
8500 pounds), light duty diesel trucks, heavy duty gasoline trucks and buses, and heavy duty
diesel trucks and buses, and motorcycles. The emission factors used are based on EPA's
MOBILE4 mobile source emission factor model, which uses the latest available data to
estimate average in-use emissions from highway vehicles (U.S. Environmental Protection
20 Agency, 1989a,b). The MOBILE4 model, developed by the EPA Office of Mobile Sources,
was used to calculate emission factors for each year. The emission factors are weighted to
consider the approximate amount of motor vehicle travel in low- and high-altitude areas to
obtain overall national average emission factors. For each area a representative average
annual temperature (low altitude average annual temperature = 57°F, high altitude = 54°F,
25 California = 65 °F), together with national averages for motor vehicle registration
distributions and annual mileage accumulation rates by age and hot/cold start vehicle
operation percentages were used to calculate the emission factors. Average speed is taken
into account according to the published distribution of vehicle-miles travelled (VMT) (U.S.
Department of Transportation, 1988). The published VMT are divided into three road
30 categories corresponding to roads with assumed average speeds of 55 miles per hour for
March 12, 1990 6-3 DRAFT-DO NOT QUOTE OR CITE
-------
interstates and other primary highways, 45 miles per hour for rural roads, and 19.6 miles per
hour for urban streets.
6.2.1.2 Aircraft
5 Aircraft emissions are based on emission factors and aircraft activity statistics reported
by the Federal Aviation Administration (1988). Emissions are based on the number of
landing-takeoff (LTO) cycles. Any emissions in cruise mode, which is defined to be above
3000 feet (1000 meters) are ignored. Average emission factors for each year, which take into
account the national mix of aircraft types for general aviation, military, and commercial
10 aircraft, are used to compute the emissions.
6.2.1.3 Railroads
The Department of Energy reports consumption of diesel fuel and residual fuel oil by
railroads (U.S. Department of Energy, 1988a). Average emission factors applicable to diesel
15 fuel consumption were used to calculate emissions.
6.2.1.4 Vessels
Vessel use of diesel fuel, residual oil, and coal is reported by the Department of Energy
(U.S. Department of Energy, 1988a,b). Gasoline use is based on national boat and motor
20 registrations, coupled with a use factor (gallons/motor/year) (U.S. Environmental Protection
Agency, 1973) and marine gasoline sales (U.S. Department of Transportation, 1988).
Emission factors from AP-42 are used to compute emissions (U.S. Environmental Protection
Agency, 1985).
25 6.2.1.5 Nonhighway Use of Motor Fuels
Gasoline and diesel fuel are consumed by off-highway vehicles in substantial quantities.
The fuel consumption is divided into several categories including farm tractors, other farm
machinery, construction equipment, industrial machinery, snowmobiles, and small general
utility engines such as lawnmowers and snowthrowers. Fuel use is estimated for each
30 category from estimated equipment population and an annual use factor of gallons per unit per
year (Hare and Springer, 1974), together with reported off-highway diesel fuel deliveries
March 12, 1990 6-4 DRAFT-DO NOT QUOTE OR CITE
-------
(U.S. Department of Energy, 1988a) and off-highway gasoline sales (U.S. Department of
Transportation, 1988).
6.2.2 Stationary Source Fuel Combustion
5 Stationary combustion equipment, such as coal-, gas-, or oil-fired heating or power
generating plants, generate CO as a result of improper or inefficient operating practices or of
inefficient combustion techniques. The specific emission factors for stationary fuel
combustors vary according to the type and size of the installation and the fuel used, as well as
the mode of operation. The EPA compilation of air pollutant emission factors provides
10 emission data obtained from source tests, material balance studies, engineering estimates, etc.,
for the various common emission categories. For example, coal-fired electricity-generating
plants report coal use to the Department of Energy (U.S. Department of Energy, 1988b,c).
Distillate oil, residual oil, kerosene, and natural gas consumed by stationary combustors are
also reported by user category to the U.S. Department of Energy (1988a, 1989a,b). Average
15 emission factors from AP-42 (U.S. Environmental Protection Agency, 1985) were used to
calculate the emission estimates. The consumption of wood in residential wood stoves has
likewise been estimated by the U.S. Department of Energy (1982, 1984).
6.2.3 Industrial Processes
20 In addition to fuel combustion, certain other industrial processes generate and emit
varying quantities of CO into the air. The lack of published national data on production, type
of equipment, and controls, as well as an absence of emission factors, makes it impossible to
include estimates of emissions from all industrial process sources.
Production data for industries that produce the great majority of emissions were derived
25 from literature data. Generally, the Minerals Yearbook, (U.S. Department of the Interior,
1987) published by the Bureau of Mines, and Current Industrial Reports, (U.S. Department of
Commerce, annual) published by the Bureau of the Census, provide adequate data for most
industries. Average emission factors were applied to production data to obtain emissions.
Control efficiencies applicable to various processes were estimated on the basis of published
30 reports and from NEDS data (U.S. Environmental Protection Agency, 1971).
March 12, 1990 6-5 DRAFT-DO NOT QUOTE OR CITE
-------
6.2.4 Solid Waste Disposal
Solid waste CO emissions result from the combustion of wastes in municipal and other
incinerators, and also from the open burning of domestic and municipal refuse. Specific
emission estimates for the various waste combustion procedures in use were taken from a
5 study conducted in 1968 concerning solid waste collection and disposal practices (U.S.
Department of Health, Education, and Welfare, 1968). Results of this study indicate that the
average collection rate of solid waste is about 5.5 pounds per capita per day in the United
States. It has been stated that a conservative estimate of the total generation rate is 10 pounds
per capita per day. The results of this survey were updated based on data reported in NEDS
10 and used to estimate, by disposal method, the quantities of solid waste generated (NEDS,
National Emissions Data System, n.d.). Average emission factors were applied to these totals
to obtain estimates of total emissions from the disposal of solid wastes.
6.2.5 Miscellaneous Combustion Sources
15 Miscellaneous CO emissions results from the burning of forest and agricultural
materials, smoldering coal refuse materials, and structural fires.
6.2.5.1 Forest Fires
The Forest Service of the Department of Agriculture publishes information on the
20 number of forest fires and the acreage burned (U.S. Forest Service, 1988). Estimates of the
amount of material burned per acre are made to estimate the total amount of material burned.
Similar estimates are made to account for managed burning of forest areas. Average emission
factors were applied to the quantities of materials burned to calculate emissions.
25 6.2.5.2 Agricultural Burning
A study was conducted by EPA (Yamate, 1974) to obtain from local agricultural and
pollution control agencies estimates of the number of acres and estimated quantity of material
burned per acre in agricultural burning operations. These data have been updated and used to
estimate agricultural burning emissions, based on average emission factors.
30
March 12, 1990 6-6 DRAFT-DO NOT QUOTE OR CITE
-------
6.2.5.3 Coal Refuse Burning
Estimates of the number of burning coal-refuse piles existing in the United States are
made in reports by the Bureau of Mines (U.S. Department of the Interior, 1971). This
publication presents a detailed discussion of the nature, origin, and extent of this source of
5 pollution. Rough estimates of the quantity of emissions were obtained using this information
by applying average emission factors for coal combustion. It was assumed that the number of
burning refuse piles decreased to a negligible amount by 1975.
6.2.5.4 Structural Fires
10 The United States Department of Commerce publishes, in their statistical abstracts,
information on the number and types of structures damaged by fire (U.S. Department of
Commerce, 1988). Emissions were estimated by applying average emission factors for wood
combustion to these totals.
15
6.3 NATIONAL CO EMISSIONS ESTIMATES 1970-1988
Table 6-1 displays the total annual 1970-1988 CO emissions from the various source
categories. The CO estimations cited herein are the result of current methodology and refined
emission factors and should not be compared with data reported earlier. These data indicate
20 that CO from all man-made sources in the U.S. declined from 101.4 teragrams in 1970 (one
teragram equals 1012 grams, 103 gigagrams, 106 metric tons, or approximately 1.1 x 10* short
tons) to 61.2 teragrams in 1988. The majority, about 70 percent, of the CO emissions total
comes from transportation sources, 12 percent comes from fuel combustion processes,
7 percent comes from industrial processes, and 11 percent comes from miscellaneous sources.
25 Table 6-2 contains a more detailed listing of CO emissions from the dominant category,
transportation sources.
The single largest contributing source of CO emissions is highway vehicles, which emit
60 percent of the total national CO estimate. Because of the implementation of the Federal
Motor Vehicle Control Program (FMVCP), CO emissions from highway vehicles have
30 declined 48 percent, from 65.3 teragrams in 1970 to 34.1 teragrams in 1988. Figure 6-1
displays how CO emissions from the major highway vehicle categories have changed from
March 12, 1990 6-7 DRAFT-DO NOT QUOTE OR CITE
-------
1
Ki
t — i
VO
'O
o
TABLE 6-2.
Source Category
Highway vehicles
Gasoline-powered
Passenger cars
Light trucks - 1
Light trucks - 2
Heavy duty vehicles
Motorcycles
1970
49,090
5,800
2,070
7,810
260
CARBON MONOXIDE EMISSIONS FROM TRANSPORTATION (GIGAGRAMS/YEAR)
1975
41,430
5,730
2,450
6,610
540
1978
37,840
6,100
4,120
6,500
550
1979
34,450
5,960
4,340
6,170
490
1980
31,850
5,810
4,210
5,870
370
1981
30,160
6,370
4,700
5,780
280
1982
30,150
5,760
4,220
4,910
200
1983
29,510
6,190
4,610
4,720
190
1984
27,790
6,050
4,450
4,380
170
1985
25,410
6,280
4,330
3,750
130
1986
23,650
6,060
4,040
2,920
120
1987
22,530
6,100
3,740
2,800
130
1988
21,220
5,770
3,540
2,730
120
Total - Gasoline
65,030 56,760 55,110 51,410 48,110 47,290 45,240 45,220 42,840 39,900 36,790 35,300 33,380
oo
Diesel-powered
Passenger cars
Light trucks
Heavy duty vehicles
Total - Diesel
0
0
300
300
0
0
390
390
2
1
500
503
5
1
530
536
8
3
610
621
10
6
700
716
10
6
680
696
20
5
650
675
20
3
650
673
20
4
770
794
20
4
670
694
20
3
680
703
10
4
720
734
Highway Vehicle Total 65,330
55,613 51,
48,006
43,513
37,484 36,003 34,114
O
o
25
3
o
d
3
w
o
5*3
O
3
Aircraft
Railroads
Vessels
Farm Machinery
Construction Machinery
Industrial Machinery
Other Off-highway Vehicles
Transportation Total
900
250
1,150
3,570
580
1,780
840
74,400
880
240
1,360
2,930
370
1,060
990
64,980
960
260
1,470
2,370
340
1,070
1.050
63,133
990
270
1,420
2,240
370
820
1.080
59,136
990
270
1,380
2,040
460
1,110
1.100
56,081
960
250
1,440
1,880
370
1,330
1.150
55,386
950
240
1,390
1,780
320
1,190
1.130
52,936
980
190
1,410
1,470
260
1,040
1.140
52,385
1,010
200
1,700
1,900
250
900
1.130
50,603
1,090
190
1,400
2,120
410
850
1.150
47,904
1,080
180
1,500
1,910
450
840
1.170
44,614
1,060
190
1,560
1,830
520
880
1.190
43,233
1,050
190
1,620
1,630
530
880
1.200
41,214
Source: U.S. Environmental Protection Agency (1990b).
-------
Heavy Duty Gasoline
Vehicles
Light Duty Trucks
1&2
Light Duty Gasoline
Vehicles
1975 1980 1981
1982 1983
Year
1984 1985 1986 1987 1988
Figure 6-1. Estimated emissions of carbon monoxide from highway vehicles, 1970-1988.
Source: U.S. Environmental Protection Agency (1990b).
March 12, 1990
6-9
DRAFT-DO NOT QUOTE OR CITE
-------
1970-1988 as a result of the FMVCP. Although VMT increased 38 percent from 1970 to
1978, total CO emissions from highway vehicles decreased 15 percent, because of the
installation of FMVCP-mandated air pollution control devices on new vehicles. Petroleum
crises contributed to a 1.7 percent decrease in VMT from 1978 to 1980. This lack of growth
5 in vehicle travel together with an increased degree of pollution control because of stricter
emission standards for new vehicles, coupled with the gradual disappearance of older
uncontrolled vehicles from the vehicle fleet, produced an estimated 12 percent drop in
highway vehicle CO emissions in this two year period from 1978 to 1980. Since 1980, VMT
have grown each year. From 1980 to 1988, VMT increased by 33 percent. However, due to
10 the FMVCP controls, CO emissions from highway vehicles actually decreased 30 percent
during this period. Overall from 1970 to 1988, without the implementation of FMVCP,
highway vehicle emissions would have increased 61 percent; with FMVCP implementation,
emissions are estimated to have decreased 48 percent.
CO emissions from other sources have also generally decreased. In 1970, emissions
15 from burning of agricultural crop residues were greater than in more recent years. Solid
waste disposal emissions have also decreased as the result of implementation of regulations
limiting or prohibiting burning of solid waste in many areas. Emissions of CO from
stationary source fuel combustion occur mainly from the residential sector. These emissions
were reduced somewhat through the mid-1970's as residential consumers converted to natural
20 gas, oil, or electric heating equipment. Recent growth in the use of residential wood stoves
has reversed this trend, but increased CO emissions from residential sources continue to be
small compared to highway vehicle emissions. Nevertheless, in 1988 residential wood
combustion accounted for about 10 percent of national CO emissions, more than any source
category except highway vehicles. CO emissions from industrial processes have generally
25 been declining since 1970 as the result of the obsolescence of a few high-polluting processes
such as manufacture of carbon black by the channel process and installation of controls on
other processes. However, industrial process emissions increased slightly (4 percent) from
1987 to 1988 due to increased industrial activity.
30
March 12, 1990 6-10 DRAFT-DO NOT QUOTE OR CITE
-------
6.4 OUTDOOR AIR CONCENTRATIONS
6.4.1 Introduction
Ambient concentrations of carbon monoxide (CO) in urban communities vary widely
with time and space. Actual human exposure to CO in various indoor and outdoor activities
5 is affected by highly localized microenvironments which are influenced by nearness to
sources, including vehicles, occupations, and by personal activities, such as smoking. Indoor
sources and concentrations are summarized in Chapter 7. Exposure information is presented
in Chapter 8. This section presents information about ambient concentrations. It will
describe observed diurnal, seasonal, and annual patterns of ambient urban CO levels and will
10 explain the importance of air monitoring site selection, of meteorological and geographic
effects on CO exposures, techniques of CO trend analyses, and special CO exposure
situations. An overview of meteorological diffusion models is also provided.
6.4.2 Site Selection
15 Site selection is one of the most complex and critical elements in the design of CO air
monitoring programs. This is especially important for CO monitoring because the proximity
of the monitor to traffic will influence the magnitude of CO concentrations. Naturally, the
choice of monitoring sites depends greatly on the objective of the monitoring to be performed.
The U.S. Environmental Protection Agency (1977a) recognizes the following as general
20 objectives for monitoring:
1. To judge compliance with and/or progress made toward meeting ambient air
quality standards.
25 2. To activate emergency control procedures to ameliorate air pollution episodes.
3. To observe pollution trends throughout the region, including the nonurban areas.
(Information from nonurban areas is needed to evaluate whether air quality in the
cleaner portions of a region is deteriorating significantly and to gain knowledge
30 about background pollutant levels.)
March 12, 1990 6-11 DRAFT-DO NOT QUOTE OR CITE
-------
4. To provide a data base for application in the evaluation of effects; in urban, land
use, and transportation planning; in development and evaluation of abatement
strategies; and in development and validation of diffusion models.
In addition to these general objectives, site selection is also based on the scale of
representativeness that will meet the objectives. Data representativeness, like measures of
concentration at a site, is dependent on the proximity of the monitor to the CO source but,
10 further, is influenced by the intended use to which the data will be put. Ground level
concentrations of CO within an urban area vary widely because the principal source of CO in
cities is automobiles which, obviously, move and are more concentrated in some areas at
some times than at other times. Monitoring sites at the edge of a highway will measure CO
concentrations representative of a fairly small area. Sites well removed from highways can be
15 representative of a fairly large-scale area. The EPA has defined six scales of spatial
representativeness for CO monitoring sites: microscale, middle scale, neighborhood scale,
urban scale, regional scale, and national and global scale (Federal Register, 1979).
Most CO monitoring conducted in the United States is for the purpose of determining
attainment or nonattainment of air quality standards. Since monitoring resources have been
20 and continue to be severely limited, monitoring sites are usually selected by a "worst case"
principle; that is, they are set up where maximum CO concentrations are expected because the
NAAQSs focus on peaks. As a result, many CO sites are located in close proximity to major
highways, arterials, and downtown street canyons. This means that they are situated where
maximum CO levels occur, but that their scale of representativeness is small. Monitoring
25 results may thus relate primarily to pedestrian exposure near the monitor. Sites located away
from the major roadways, but within highly populated neighborhoods with high traffic
densities, may be more representative of the maximum CO concentrations to which a large
portion of the population of a city may be exposed.
The EPA has published guidelines (Federal Register, 1979) for CO monitor siting
30 (Table 6-3). EPA guidelines (U.S. Environmental Protection Agency, n.d.) give the highest
priority to microscale sites within street canyons and to neighborhood sites where maximum
concentrations are expected.
March 12, 1990 6-12 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 6-3. SPECIFIC PROBE EXPOSURE CRITERIA
I
I—>
K)
vo
O
O
O
3
9
a
O
X)
n
Site Type
Height
Above
Ground
Expected
Concentration
Gradient
with Height
(1-hr Average)
Separation
of Monitor from
Influencing Sources
General Remarks
Street Canyon
Peak Concentration 3 ± 1/2 m
Average Concentration 3 ± 112m
~.5 ppm/m
~3 ppni/m
Mid-sidewalk or 2 m
from side of building.
On leeward side of
street.
Mid-sidewalk or 2 m
from side of building.
Central Business District.
High density, slow-moving
traffic.
Dense multiple-story buildings
lining both sides of street.
Neighborhood
Peak Concentration 3 ± 1/2 m 5%/m
Average Concentration 3 ± 1/2 m 5%/m
<.3 ppm/m
Background 3 to 10 m .2%m
New Source Review
Preconstruction 3 ± 1/2 m 5%/m
Postconstruction 3 ± 1/2 m > 5%/m to
< .Sppm
Setback VPD
3.5 km 100,000
1.5 km 50,000
200 m 10,000
100 m 5,000
35 m 1,000
25 m any
Dependent on traffic
volume, road config-
uration and setback
distance of commercial
or residential activity.
5 to 6 km; > 3,000 VPH
maximum.
400 m; > 100 VPD.
Usually the same as
neighborhood.
Usually the same as
corridor or street canyon.
Commericial or residential
neighborhood. This separation
criteria limits the effect of
these streets to =1 ppm.
Stop and go or limited access
traffic > 50,000 VPD or greatest
in area.
35 km downwind in least
frequent wind direction from
city, limit effects to .2 ppm.
Area of lowest concentration
in proposed indirect source
location for background.
Area of maximum concentration
in area of complete area source.
Source: Federal Register (1979).
-------
The variability of CO concentration with height in the vicinity of a highway is
sufficiently large that the representativeness of measurements will be strongly affected by
variability of the inlet probe height. It is, therefore, necessary to standardize the height of the
inlet probe so that data collected at one air monitoring station is comparable to data collected
5 at others. In an effort to characterize typical human exposure, the sample inlet probe height
should ideally be at breathing level. However, as a compromise between representation of
breathing height and practical considerations, such as prevention of vandalism, it is
recommended that inlets for most kinds of sampling be at a height of 3 ± 0.5 m (Altshuller
et al., 1966; Bach et al., 1973). A one-meter minimum separation of the probe from adjacent
10 structures is also recommended to avoid the frictional effects of surfaces on the movement of
air (Anonymous, 1976).
Site selection for monitors used for purposes other than trend analysis and determination
of compliance with air quality standards may not follow the specific criteria that apply to
continuous monitoring sites. In fact, special purpose studies in which CO concentrations are
15 measured at many locations provide information about the spatial variations of ambient CO
that form the basis for setting site-selection criteria. Among the principal types of special
purpose monitoring are research studies for diffusion model development and improvement
and for source surveillance studies.
20 6.4.3 United States Data Base
Monitoring stations reporting data to EPA's Aerometric Information Retrieval System
(AIRS) fall into two major categories: the National Air Monitoring Stations (NAMS) and the
State and Local Air Monitoring Stations (SLAMS). The NAMS were established through
monitoring regulations promulgated in May 1979 (Federal Register, 1979) to provide EPA
25 with accurate and timely data on a national scale. The NAMS are located at sites expected to
incur high pollutant concentrations and to typify areas with the potential for high population
exposure. These stations meet uniform criteria for site location, quality assurance, and
equivalent analytical methodology, sampling intervals, and instrument selection to assure
consistent data reporting nationwide. The SLAMS, in general, meet the same rigid criteria
30 but, in addition to the above siting criteria for highest concentrations and population exposure
potential, they may be located to monitor a greater diversity of urban neighborhoods.
March 12, 1990 6-14 DRAFT-DO NOT QUOTE OR CITE
-------
In accordance with requirements of the Clean Air Act and EPA regulations for State
Implementation Plans (SIPs) (Code of Federal Regulations, 1977) ambient CO data from
Federal networks must be reported each calendar quarter to AIRS. State and local agencies
report most of the data from their SLAMS stations as well. As a result, continuous
5 measurements of ambient CO concentrations from numerous cities throughout the United
States are available from the U.S. Environmental Protection Agency.
Computer retrievals of raw data submitted to the EPA's AIRS data bank and published
data summaries such as the National Air Quality and Emission Trends Report (U.S.
Environmental Protection Agency, 1988) and Air Quality Data - Annual Statistics are
10 available (U.S. Environmental Protection Agency, 1974b,c, 1976d,e,f,g,h, 1977b).
However, state and local air pollution control agencies are not required to submit all CO data
collected from their monitoring network. These agencies may also conduct special studies for
certain "in-house" purposes. State departments of transportation and local metropolitan
planning commissions are sources of CO data for the preparation of environmental impact
15 statements for proposed transportation projects and/or in the preparation of SIP revisions. Air
quality impact research sponsored by the EPA, the Federal Highway Administration,
universities, and private industries also are sources of CO data.
6.4.4 Techniques of Data Analysis
20 Air quality surveys inherently involve taking a limited number of samples from a highly
variable and uncontrolled population (i.e., the environment). For this reason, air quality data
should be analyzed through statistical methods, which can be used to describe the behavior of
the total population on the basis of a finite number of samples. In particular, statistical
parameters can be calculated to describe the typical values observed, the maximum or peak
25 values observed, and the range of values observed.
Although intermittent sampling is an important research tool for conducting special
studies, the majority of CO monitoring instruments in use today are intended to operate
continuously and to yield successive hourly averages. These data are applied for two
principal uses: (1) characterizing environmental conditions by describing short-term (hourly,
30 daily, seasonal) and long-term (year-to-year) urban CO concentration patterns, and
March 12, 1990 6-15 DRAFT-DO NOT QUOTE OR CITE
-------
(2) evaluating, for statutory purposes, an area's status with respect to the 1-hour and 8-hour
average NAAQS for CO.
At a minimum, an analysis of CO air quality data should include a comparison of the
highest (or second highest) observed pollution concentration to established air quality
5 standards. In addition, an analysis of CO data may include calculation of population
statistics, frequency analyses, averaging time analyses, trend analyses, and case analysis.
6.4.4.1 Frequency Analysis
In most areas of air pollution monitoring modeling, we do not have enough knowledge
10 about the generation, dispersion, and transport of air pollutants to formulate a convincing
theoretical model for air quality data. Air pollutant concentrations are often generated by
autocorrelated stochastic processes. In most situations we never know which of several
hypothetical models are "correct." Fortunately, it is usually possible to identify time periods
and pollutant averaging times when observations are approximately stationary and independent
15 sequences of concentrations. Horowitz and Barakat (1979) showed that autocorrelation does
not significantly affect the validity of the usual methods for estimating the parameters of the
maximum pollutant concentration distribution. The most widely used model has been the
two-parameter lognormal distribution which has played a major role in the formulation of air
quality standards for many pollutants Georgopoulos and Seinfeld (1982). However, there are
20 many data sets for which some other distribution fits better. These candidate models include
the three-parameter lognormal, Weibull, exponential, and gamma distributions (Bencala and
Seinfeld, 1976; Ott et al., 1979; Pollack, 1975; Simpson et al., 1984). This variety of
distribution types probably reflects the phenomenon that an air pollution concentration is the
superposition of a random number of point, line, and area sources of different emission
25 strengths.
The NAAQSs for CO are currently based on a 1-hour and an 8-hour averaging time.
Carbon monoxide data are most frequently collected in time averages of one hour. Evaluating
compliance with the 1-hour standard simply requires rank-ordering 1-hour values for a year
and comparing the second highest value with the 1-hour standard, which is currently
30 40 mg/m3 (35 ppm), not to be exceeded more than once per year. If the second highest
1-hour value is less than 40 mg/m3, the standard has been met.
March 12, 1990 6-16 DRAFT-DO NOT QUOTE OR CITE
-------
Evaluating compliance with the 8-hour standard involves the calculation of moving
8-hour averages from the 1-hour data set. These 8-hour averages are also rank-ordered to
obtain the second highest nonoverlapping value for comparison with the 8-hour standard,
which is currently 10 mg/m3 (9 ppm). For enforcement purposes, only nonoverlapping
5 8-hour intervals are counted as violations, as discussed in the Guidelines for the Interpretation
of Air Quality Standards (U.S. Environmental Protection Agency, 1977c). It has been
shown, however, that the full set of moving 8-hour averages should be examined for
excessive values. Proposed simplifications, such as calculating only three consecutive
nonoverlapping 8-hour averages per day, can easily result in missing peak 8-hour intervals
10 and may not afford equitable comparisons among stations with differing diurnal patterns.
6.4.4.2 Trend Analyses
Carbon monoxide ambient concentrations vary considerably from hour to hour, day to
day, season to season, and year to year. These variations are usually not random but follow
15 fairly predictable temporal patterns according to season of the year, day of the week, and
hour of the day. Long-term, statistical patterns in CO concentrations are referred to as
trends. Carbon monoxide trends are best illustrated by graphs that can show diurnal, daily,
seasonal, or yearly CO concentration comparisons. Examples of the different ways trends can
be shown are provided later in this section. Carbon monoxide concentrations also follow
20 fairly predictable spatial patterns. Spatial distributions of CO concentrations can be illustrated
by the use of isopleth maps.
6.4.4.3 Special Analyses
A useful analysis technique is the "pollution rose," as illustrated in Figure 6-2. The
25 pollution rose presents the joint frequency distribution of wind direction versus ambient CO
concentration. The pollution rose is very helpful in determining the wind direction associated
with the highest ambient CO concentrations and, intuitively, the location of high CO
emissions sources.
Another analysis technique is case analysis, which can be used to characterize the
30 meteorological and/or emission conditions associated with observed CO concentrations. For
example, in order to characterize the meteorological conditions associated with the occurrence
March 12, 1990 6-17 DRAFT-DO NOT QUOTE OR CITE
-------
NORTH
10%
LEGEND
CO CONC, mg/m3
0-6.8 6.8-9.1 >91
Figure 6-2. CO pollution rose for St. Louis, MO.
March 12, 1990
6-18 DRAFT-DO NOT QUOTE OR CITE
-------
of high CO levels, meteorological records can be evaluated for the days when highest CO
concentrations were observed concurrently at several monitoring sites throughout an urban
area. The results of the analysis can then be used to develop a meteorological scenario for
input to a mathematical model for the purpose of modeling "worst-case" CO concentrations.
5
6.4.5 Urban Levels of Carbon Monoxide
The ambient CO data cited in this document were obtained from EPA's Air Quality and
Emissions Trend Report (1990) and directly from the Aerometric Information Retrieval
System (AIRS, no date). To be included in the 10-year trend analyses, a given station had to
10 report data for at least 8 of the 10 years in the period 1979-1988; 248 stations qualified. For
the 5-year (1984-1988) trend and urbanized area trend analyses, a station had to report data
for at least 4 of those five years; 359 stations qualified. The shorter time period was used in
the urbanized area analyses to expand the number of stations available for analysis.
15 6.4.5.1 Ten-year CO Trends 1979-1988
Figure 6-3 illustrates the national 1979-1988 composite average trend for the second
highest nonoverlapping 8-hour CO value for the 248 long-term sites and the subset of
72 NAMS sites (U.S. Environmental Protection Agency, 1990). The national average for all
248 stations and for the NAMS subset of 72 stations decreased both by 28 percent.
20 A Box plot of the data for all stations (Figure 6-4) provides a measure of the distribution
changes (U.S. Environmental Protection Agency, 1990). Each horizontal line of a Box plot
represents a percentile value. Starting at the top, each line represents the 95th, 90th, 75th,
50th (median), 25th, 10th, and 5th percentile values. The composite average is represented
by an x near the median value. Although certain percentiles fluctuate from year to year, the
25 general long-term improvement is clear.
The 10-year trend of the composite average of the estimated number of nonoverlapping
8-hr CO average concentrations that exceed the 8-hr NAAQS across all stations is shown in
Figure 6-5 (U.S. Environmental Protection Agency, 1990). The trend is clearly decreasing
with an 88 percent improvement for the 248 long-term stations. Note that these percentage
30 improvements for exceedances are typically much larger than those found for the second
March 12, 1990 6-19 DRAFT-DO NOT QUOTE OR CITE
-------
CONCENTRATION, PPM
I 1 l 1 1 1 1 1 1 1
1979 1980 1981 1982 1983 1984 1985 1986 1987 1988
Figure 6-3. National trend in the composite average of the second highest nonoverlapping
8-hour average carbon monoxide concentration 1979-1988. Bars show 95 percent confidence
intervals.
Source: (U.S. Environmental Protection Agency, 1990).
20
CONCENTRATION, PPM
15-
10-
5-
248 SITES
MSi
1979 1980 1981 1982 1983 1984 1985 1986 1987 1988
Figure 6-4. Boxplot comparisons of trends in second highest nonoverlapping 8-hour average
carbon monoxide concentrations at 248 sites, 1979-1988.
Source: U.S. Environmental Protection Agency (1990).
March 12, 1990
6-20 DRAFT-DO NOT QUOTE OR CITE
-------
20
EST. 8-HR EXCEEDANCES
15-
10-
5-
• NAMS SfTES (72) ° AU-_SITESJ[248l
1979 1980 1981 1982 1983 1984 1985 1986 1987 1988
Figure 6-5. National trend in the composite average of the estimated number of exceedances
of the 8-hour carbon monoxide NAAQS, 1979-1988. Bars show 95 percent confidence
intervals.
Source: U.S. Environmental Protection Agency (1990).
maximum 8-hour concentrations depicted in Figure 6-4. The concentrations data are more
likely to reflect percentage change in emission levels.
National CO emission estimates (Table 6-4) show a 25 percent decrease over the
10-year period (U.S. Environmental Protection Agency, 1990). The predominant CO
emission source, transportation, accounted for about 72 percent of total CO emissions in
March 12, 1990
6-21
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 64. NATIONAL CARBON MONOXIDE EMISSION ESTIMATES, 1979-1988
(million metric tons/year)
10
1979
Source Category
Transportation
Fuel Combustion
Industrial
Processes
Solid Waste
Miscellaneous
Total
59
6.
7.
2.
6.
81
.1
7
1
3
5
.7
1980
56.1
7.4
6.3
2.2
7.6
79.6
1981
55.4
7.7
5.9
2.1
6.4
77.4
1982
52.9
8.2
4.3
2.0
4.9
72.4
1983
52.4
8.2
4.3
1.9
7.7
74.5
1984
50.6
8.3
4.7
1.9
6.3
71.8
1985
47.9
7.4
4.4
2.0
5.3
67.0
1986
44.3
7.5
4.3
1.7
5.0
63.1
1987
43.2
7.6
4.5
1.7
7.1
64.1
1988
41.2
7.6
4.7
1.7
6.0
61.2
15
Note: The sums of sub-category may not equal total due to rounding.
20 Source: U.S. Environmental Protection Agency (1990).
1979, but had decreased to about 67 percent in 1988. This result provides further evidence
25 that the Federal Motor Vehicle Control Program has been effective on a national scale with
controls more than offsetting the growth during the period. It should be noted that CO
monitors are typically located to identify potential problems and are placed in areas of high
traffic densities that may not experience significant increases in traffic. Thus, CO levels at
these locations may improve at a faster rate than the nationwide reduction in emissions.
30
6.4.5.2 Five-year CO Trends 1984-1988
Evaluation of five-year trends allows use of the expanded data base (359 stations). The
composite averages (Figure 6-6) indicate 16 percent improvement between 1984 and 1988.
Total estimated CO emissions (Table 6-4) decreased 15 percent across the five-year period;
35 emissions from transportation sources decreased 19 percent.
Composite Regional averages for 1986 through 1988 of the second highest
nonoverlapping 8-hour CO averages (Figure 6-7) show the largest declines occurring in
Regions I, II, VI, VII, and VIII. Smaller, sometimes fluctuating declines occurred in
Regions III, IV, V, and IX; Region X ended 1988 on a level with 1986.
March 12, 1990 6-22 DRAFT-DO NOT QUOTE OR CITE
-------
CONCENTRATION, PPM
1984 1985 1986 1987
1988
Figure 6-6. Boxplot comparisons of trends in second highest nonoverlapping 8-hour average
carbon monoxide concentrations at 359 sites, 1984-1988.
Source: U.S. Environmental Protection Agency (1990).
March 12, 1990
6-23 DRAFT-DO NOT QUOTE OR CITE
-------
15
U
CONCENTRATION, PPM
12-
10-
8-
6-
4-
o _
0
COMPOSITE AVERAGE
E3 1986 • 1987 O 1988
EPA REGION I
NO. OF SITES 15
II III IV V VI VII VIII IX X
27 48 50 49 29 18 17 83 23
Figure 6-7. Regional comparisons of the 1986, 1987, 1988 composite averages of the second
highest nonoverlapping 8-hour average carbon monoxide concentration.
Source: U.S. Environmental Protection Agency (1990).
20
6.4.5.3 Air Quality Levels in Metropolitan Statistical Areas
Metropolitan Statistical Areas (MSAs) consist of a central urban county or counties and
any adjacent counties with at least 50% of their population within the urban perimeter. The
nation's 339 MSAs, grouped by population range in Table 6-5, include 77% of the U.S.
5 population. Figure 6-8 compares the highest second-high nonoverlapping 8-hour value
recorded during 1988 for the 90 largest MSAs in the continental U.S. (not shown: Honolulu,
HI, and San Juan, PR), containing approximately 55% of the U.S. population. Nineteen of
these MSAs exceeded the current 8-hour standard of 9 ppm in 1988.
10 6.4.6 Effects of Meteorology and Topography
Meteorology governs the transport and dispersion of CO emissions in the atmosphere
and thus has a strong influence on the ground level CO concentrations detected at receptor
points downwind of emission sources. Meteorological parameters that determine CO
March 12, 1990
6-24
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 6-5. DISTRIBUTION OF POPULATION IN METROPOLITAN STATISTICAL
AREAS (Based on 1987 estimates)
10
Population Range
< 100,000
100,000 < 250,000
250,000 < 500,000
500,000 < 1,000,000
1,000,001 < 2,000,000
> 2,000,000
Total
No. of MSAs
27
147
73
48
26
18
339
Total Population
2,274,000
23^,372,000
25,218,000
34,367,000
38,685,000
65,747,000
189,663,000
15
Source: U.S. Environmental Protection Agency (1990).
20
transport and dispersion patterns include wind speed, wind direction, atmospheric stability,
and mixing depth. The relative importance of each parameter depends upon the scale of the
analysis. For example, concentration patterns around an intersection would not be greatly
25 influenced by mixing depth. However, concentration patterns over the whole urban area
would be.
Wind direction determines the direction of horizontal transport of CO emissions and
consequently the impact that CO emissions from one are will have on air quality in another
area. If emissions are uniform across the urban area, as air flows across the whole urban
30 area, the additive effect of the CO emissions being transported downwind will result in higher
CO concentrations at the downwind edge of the urban area. However, concentrations at a
particular location may be dominated by local emissions. Concentrations adjacent to a
highway will be higher than urban levels away from large highways. Wind directions nearly
parallel to a highway will allow for an accumulation of CO emissions in the downwind
35 direction, resulting in CO concentrations higher than would be expected for winds
perpendicular to the highway under the same conditions.
Low wind speeds provide little atmospheric dilution, allowing CO emissions to build up,
resulting in higher CO concentrations. Conversely, high wind speeds aid in the dispersion of
March 12, 1990 6-25 DRAFT-DO NOT QUOTE OR CITE
-------
-------
CO emissions by increasing the amount of dilution that takes place, thus decreasing CO
concentrations. The effect of surface roughness (i.e., mountains, buildings, etc.) on the wind
speed profile over several types of topographic features is illustrated in Figure 6-9 (Benson,
1979). With increased surface roughness, either natural or man-made, the depth of the
5 affected layer is increased. The winds affected by frictional drag are reduced, but the
turbulence induced by mechanical effects in increased. The net effect of increased surface
roughness over an urban area is to mix the CO emissions through a larger depth in the
atmosphere, which aids in the dispersion of CO emissions. Thermal forces in the atmosphere
either enhance or suppress the production of turbulent motion in the atmosphere. The
10 dispersive properties of the atmosphere are correlated with atmospheric stability, which is
generally easier to characterized.
Radiation and thermal properties of topographic features influence the heating and
cooling of the atmosphere near the ground surface. The most notable of these effects is the
urban "heat island" effect. Heat sources, including the asphalt and concrete associated with
15 an urban area, tend to radiate heat, causing a "heat island" compared to the cooler
surrounding terrain. The buoyant effect of warmer air over the city tends to induce thermal
turbulence (i.e., more unstable atmospheric conditions), which tend to aid in the dispersion of
CO emissions, thus lowering ambient CO concentrations.
Local wind circulations, such as sea-land breezes, lake-land breezes, or mountain-valley
20 winds, are caused by the differential heating of topographic forms. These circulations
generally flow in one direction during the day, and in the opposite direction at night. As a
result, an urban area can experience "blow-back" of CO emissions emitted during the day;
these will be experienced as higher CO concentrations at night. The boundary region between
the local circulation winds and the prevailing synoptic flow sometimes remains nearly
25 stationary, or slowly oscillates back and forth for periods up to several hours, and can be the
site of nearly calmwind conditions. These characteristics result in slow net transport of CO
emissions which then accumulate and result in higher ambient CO concentrations.
The depth through which pollutants are routinely mixed affects the total ventilation
capacity of the atmosphere. When the potential temperature lapse rate is positive, the
30 resulting increase in temperature with increase in height procedures an inversion or inversion
lid that limits vertical mixing, and thereby limits the dilution capacity of the atmosphere.
March 12, 1990 6-27 DRAFT-DO NOT QUOTE OR CITE
-------
to
QO
8
O
600
500 -
400
t 300 -
o
Ul
X
200 -
100 -
to
0 6
WIND SPEED, m/uc
Figure 6-9. Effect of terrain roughness on the wind speed profile.
Source: Benson (1979).
-------
An important form of inversion for CO dispersion is the surface or radiation inversion.
This usually occurs at night with light winds and clear skies, when the loss of heat by
longwave radiation from the ground surface cools the surface and subsequently the air
adjacent to it. With the proper relative humidity, these same conditions will lead to the
5 formation of radiation fog. The presence of early morning fog is often associated with a
surface based temperature inversion. The surface inversion usually persists for hours, and
because it typifies stable atmospheric conditions, it tends to result in high microscale and
mesoscale CO concentrations.
Another type of inversion is the subsidence inversion. It is caused by a gradual descent
10 of air aloft, that results in adiabatic wanning of the descending layer. The resulting
subsidence inversion is illustrated in Figure 6-10, which shows the temperature decreased with
height and the capping by a subsidence inversion layer, above which there is a normal
decrease of temperature with height. The subsidence inversion usually persists for days and
tends to contribute to high urban background CO concentrations. The subsidence inversion is
15 usually more persistent during summer and fall than in winter or spring.
The shape of typical plots of hourly CO concentrations can be attributed in large part to
the effect of changing wind speeds, atmospheric stability, and inversion height during the
course of a day. Figure 6-11 shows average hourly wind speeds and inversion heights
occurring in Los Angeles during summer (Tiao et al., 1975). The higher wind speeds and
20 inversion height during early afternoon are typical throughout the continental United States
and play a significant role in lowering urban CO concentrations at midday. Traffic volumes,
and subsequently CO emissions from cars, would still be expected to be high at this time of
day. Around midnight, when traffic volumes are relatively low, the effects of low wind
speeds and low inversion heights tend to cause increases in CO concentrations. Many
25 monitoring stations in the United Sates observe these relatively high CO concentrations late at
night.
Ambient surface temperature also has a unique effect on the production rate of CO
emissions from automobiles. Using a variety of automobiles tested at artificially controlled
ambient temperatures of 20°, 50°, 75°, and 110°F, the EPA (Bullin et al., 1986) found that
30 lowest CO emissions were produced at 75 °F and tend to increase with colder temperatures.
Colder temperatures coupled with a strong surface based radiative inversion are generally
March 12, 1990 6-29 DRAFT-DO NOT QUOTE OR CITE
-------
1000
9)
**
a
H
I
o
500
2nd MIXING LAYER —
INVERSION "LID"
lit MIXING LAYER —
I I I I I \ I " I I
0 10 20
TEMPERATURE, °C
Figure 6-10. Schematic representation of an elevated inversion.
March 12, 1990
6-30 DRAFT-DO NOT QUOTE OR CITE
-------
10
£
a
a"
UJ
Ul
8s
a
2 5
i
I I I I I I I
riii
WIND SPEED
J I \ L
J L
12
18
24
TtME.hourt
20
o
iij
X
15
Figure 6-11. Hourly variations in inversion height and wind speed for Los Angeles in
summer.
Source: Tiao et al. (1975).
March 12, 1990
6-31 DRAFT-DO NOT QUOTE OR CITE
-------
30
associated with poor dispersion in the atmosphere. The combined effect of higher emission
rates and poor dispersive conditions results in higher ambient CO concentrations than would
be expected for warmer temperatures.
6.5 CARBON MONOXIDE DISPERSION MODELS
A dispersion model relates pollutant emissions to ambient air quality by providing a
mathematical description of the transport, dispersion, and chemical transformations that occur
in the atmosphere. This ability to relate source emissions to receptor air quality is very
10 important to air quality maintenance planning and environmental impact assessment.
Dispersion models vary in complexity from simple empirical or statistical relationships
to sophisticated multi-source models that describe the transport and dispersion of CO
throughout an urban area. For estimates of ambient CO concentrations, a line source model
is needed to estimate the CO levels near a highway, an intersection model is needed to
15 estimate CO levels near an intersection, and an urban model is needed to estimate CO levels
that result from the cumulative effects of urban sources such as roadways and wood stoves.
The types of models used will depend mainly on the source configuration to be modeled (i.e.,
highway intersection, or urban area).
20 6.5.1 Line Source Modeling
Several models have been used to estimate CO concentrations from line sources. The
guideline on Air Quality Models (Revised) (U.S. Environmental Protection Agency, 1986)
makes specific recommendations on procedures to utilize for line source modeling. Refer to
the latest version of this document for these recommendations. Available line source models
25 are CALINE-3 (Benson, 1979), GMLINE (Chock, 1978), HIWAY-2 (Petersen, 1980), and
PAL (Petersen, 1978). A brief description of each of the line source models excerpted from
"Evaluation of Mobile Source Air Quality Simulation Models" (Wackter and Bodner, 1986)
follows:
March 12, 1990 6-32 DRAFT-DO NOT QUOTE OR CITE
-------
6.5.1.1 CALINE3
The CALINE3 model was developed by the California Department of Transportation. It
simulates dispersion of highway emissions by dividing individual roadway links into a series
of elements from which incremental concentrations are computed using a finite line source
5 equation. The incremental concentrations are summed to obtain a total concentration estimate
at a particular receptor location.
CALINE3 simulates the region directly over the roadway as a zone of uniform
emissions and turbulence called the "mixing zone." This zone experiences increased
dispersion due to mechanical turbulence created by moving vehicles as well as thermal
10 turbulence created by hot vehicle exhaust. CALINE3 adjusts the level of turbulence as a
function of wind speed. At low wind speeds residence time of an air parcel within the mixing
zone is increased, resulting in turbulence enhancement through the use of a larger initial
vertical sigma value.
The CALINE3 model includes options for simulating dispersion from four types of
15 roadways: at grade, elevated filled sections, elevated bridges, and cut or depressed sections.
Multiple lanes, links and orientations can be simulated.
6.5.1.2 GMLINE
GMLINE (Chock, 1978) was developed by General Motors Research Laboratories to
20 describe dispersion near straight-line, at-grade highways. Multiple parallel or crossing
roadway links can be simulated and the model allows for a variable emissions height. The
model was not designed to treat cut-sections.
GMLINE simulates dispersion of vehicle emissions by dividing the roadway into
separate, straight-line sources, each with a uniform emission rate. Downwind concentrations
25 at a receptor are calculated for each infinite line source, then summed to obtain a total
concentration. The model accounts for plume rise due to heated exhaust and includes a wind
speed correction to account for increased turbulence created by traffic wakes.
6.5.1.3 HIWAY-2
30 HIWAY-2 (Petersen, 1980) was developed by EPA to replace the HIWAY model
(Zimmerman and Thompson, 1975) for estimating roadway pollutant impacts. The model
March 12, 1990 6-33 DRAFT-DO NOT QUOTE OR CITE
-------
was designed to determine concentrations at receptors downwind of at-grade roadways and cut
sections (outside of the cut only).
HIWAY-2 simulates dispersion by treating highway emissions as a series of finite line
sources, each with a uniform emission rate. Concentrations downwind are calculated by
5 numerically integrating a Gaussian point-source plume along each line segment. The primary
differences between HIWAY-2 and HIWAY are that HIWAY-2 includes a new set of
dispersion curves and an aerodynamic drag factor to account for dispersion due to vehicle
motion under low wind speed conditions.
10 6.5.1.4 PAL
The PAL model (Petersen, 1978) was developed by EPA to estimate pollutant dispersion
from point, area and line sources. It was designed to simulate dispersion from several types
of roadway geometries including straight or curved horizontal lines and straight or curved
elevated lines with variable emissions along each line segment. Model documentation
15 specifies that treatment of elevated line sources is appropriate for open bridge type road
segments but not for elevated filled roadways. Cut or depressed roadway sections are not
treated by PAL.
PAL determines concentrations at a receptor due to a line source by numerically
integrating the Gaussian point source equation. Calculations are made for a number of points
20 along the finite line, assuming a linear change in concentration between these points.
Subsequent estimates of concentrations are made by including additional points along the line.
When the difference between succeeding estimates becomes smaller than a prescribed value,
the calculations are considered complete.
25 6.5.1.5 Model Evaluation
A comprehensive evaluation of CALINE3, GMLINE, HIWAY-2, and PAL was
undertaken using 5 field measurement programs and is described in "Evaluation of Mobile
Source Air Quality Simulation Models" (Wackier and Bodner, 1986). This report contains
numerous tabulations of each model's performance in terms of statistical measures
30 recommended by the American Meteorological Society. The results indicate that the
GMLINE model performed the best most often, while the PAL model ranked lowest most
March 12, 1990 6-34 DRAFT-DO NOT QUOTE OR CITE
-------
frequently. All the models tended to overpredict for light wind speeds and near parallel
wind/road angles, while underpredictions occurred for high wind speeds.
6.5.2 Intersection Modeling
5 Several models have been used to estimate CO concentrations from intersections. The
Guideline on Air Quality Models (Revised) (U.S. Environmental Protection Agency, 1986)
makes specific recommendations on procedures to utilize for intersection modeling. Refer to
the latest version of this document for these recommendations. Available intersection models
are: "Volume 9" (U.S. Environmental Protection Agency, 1978), CAL3Q (Smith, 1985),
10 CALINE4 (Benson, 1984), GIM (EMI Consultants, 1985), IMM (New York State
Department of Transportation, 1980), and TEXIN2 (Bullin et al., 1986) a brief description of
each of these models excerpted from the above references follows.
6.5.2.1 "Volume 9"
15 Carbon monoxide concentrations are calculated in a three-step process. In the first step,
the network description and traffic demand volume are used to estimate the traffic flow
characteristics. Emissions are then computed as the sum of two parts: cruise emissions
produced by non-stopping vehicles and excess emissions emitted by stopping/starting vehicles.
Lastly, the effect of atmospheric dispersion on actual concentrations at the specified receptor
20 locations is estimated.
Excess emissions consist of deceleration, idle, and acceleration emissions due to vehicles
stopping and starting at intersections. Idle emissions rates are determined using MOBILES.
Acceleration, deceleration, and cruise emission rates are determined using modal emission
factors based on the updated (December 1977) version of the Modal Emissions Model
25 (Kunselman et al., 1974). MOBILES correction factors (Wolcott, 1986) to the modal
emission factors can then be utilized to adjust for calendar year, cold starts, hot starts, speed,
temperature, and vehicle mix.
The traffic model contained in "Volume 9" calculates the length over which excess
emissions apply. This calculation is based on the proportion of vehicles that stop and the
30 number of vehicles subject to queueing delay. It should be noted that the traffic model
March 12, 1990 6-35 DRAFT-DO NOT QUOTE OR CITE
-------
contained in "Volume 9" is not applicable for overcapacity intersections thus, "Volume 9"
cannot be utilized for such intersections scenarios.
6.5.2.2 Intersection Midblock Model
5 The Intersection Midblock Model (IMM) is a combination of signalization and vehicle
queueing estimation procedures using accepted traffic engineering principles. It also predicts
emissions using the Modal Analysis Model and the MOBILE-2 program, and models
dispersion with the HIWAY-2 model.
The IMM first calculates various traffic parameters. Once the traffic calculations have
10 been performed, the estimation of emission rates is carried out. Using the input parameters
of speed into the queue, speed out of the queue, deceleration into the queue and acceleration
out of the queue, the IMM utilizes the Modal Analysis Model as a subroutine to calculate
cruise and acceleration/deceleration emissions for all approaches. Idle emissions are
calculated by use of the MOBILE-2 program. Based on the previously calculated queue
15 lengths, a set of pseudolinks is constructed. These pseudolinks lie along the actual links with
the same termination points and center lines as the actual links, but each has a length equal to
the calculated queue length for that approach. The only emissions assigned to the actual links
are the cruise emissions (calculated with the Modal Analysis Model). The emissions assigned
to the pseudolinks are the excess emissions due to accelerating, decelerating and idling.
20 A correction factor is applied to the emissions calculated from the Modal Analysis
Model since these apply only for 1977 emission rates from stabilized light-duty vehicles. The
correction factor used is the ratio of the MOBILE-2 composite emission estimate for the
specified scenario to the MOBILE-2 composite emission estimate for 1977 stabilized light-
duty vehicles.
25 Once the traffic calculations have been performed and emission rates assigned to each
lane, the HIWAY-2 model is employed as a subroutine to calculate carbon monoxide
concentrations at selected receptors. For the special case of a "street canyon" intersection
between tall buildings in a highly urban area, a special dispersion routine is used.
30
March 12, 1990 6-36 DRAFT-DO NOT QUOTE OR CITE
-------
6.5.2.3 Georgia Intersection Model
The Georgia Intersection Model (GIM) uses a computer program to calculate the
average vehicle delay, the average route speed, and the emission rate of carbon monoxide of
vehicles traveling through the intersection over a distance called the "effective length" where
5 speeds are lower due to the effect of vehicles slowing and stopping during red light cycles. It
eliminates the need for using modal emission factors and allows for the analysis of
overcapacity intersections.
GIM uses many of the same assumptions and equations as in the "Volume 9" approach,
with few modifications. The procedure can be summarized briefly as follows. GIM
10 calculates the effected length of roadway upstream of the intersection where vehicle speeds
are reduced due to delays caused by vehicle slowing and stopping. It calculates the CO
emissions for vehicles traversing the effected length, based on average speed over the length
and MOBILES emission factors. Using this approach, modal emission factors (i.e., accelera-
tion mode emissions, deceleration mode emissions, idle emissions, and cruise emissions) are
15 not utilized. The output of GM defines finite line source segments with their associated CO
emission rates which can be input to the CALENE3 line source dispersion model.
6.5.2.4 TEXIN2
The TEXIN2 Model follows a general three-step process:
20
(1) Estimation of traffic parameters.
(2) Estimation and distribution of vehicle emissions.
25 (3) Modeling downwind dispersion of pollutants.
Traffic parameters are calculated using either the modified Planning or Operations and
30 Design procedures of the Critical Movement Analysis (CMA) (National Cooperative Highway
Research Program, 1979) for signalized intersections. Basically, the difference between the
two traffic algorithms concerns the different adjustment factors present in the CMA
Operations and Design algorithm. These adjustment factors tend to decreased the capacity of
a given intersection. Therefore, the Operations and Design technique will occasionally
March 12, 1990 6-37 DRAFT-DO NOT QUOTE OR CITE
-------
calculate that an intersection is over capacity while the Planning procedure indicates that the
intersection is below full capacity.
Research has provided adjustment factors for a number of elements that affect traffic
flow and hence modify critical volumes. These elements are: (1) left turns; (2) bus and
5 truck volume; (3) peaking characteristics; (4) lane width; (5) bus stop operations; (6) right
turns with pedestrian activity; and (7) parking activity. In the TEXIN2 Model, the CMA
Planning procedure utilizes only the left turn adjustment factor, while the CMA Operations
and Design procedure uses the first four adjustment factors listed above with no additional
user input. In both algorithms, left turns are treated in detail for the simple reason that left
10 turns have a large impact on intersection capacity. This effect is created using passenger car
equivalency (PCE) values. PCE values are multiplicative adjustment factors applied to the
left turning traffic volumes.
The second function performed by TEXIN2 is the estimation of vehicle emissions. The
emissions are modeled as the sum of two components: cruise and excess emissions. Cruise
15 emissions and excess emissions are released by free-flowing and delayed vehicles,
respectively. Initially, cruise emissions are assumed to be released along the entire length of
each intersection leg. The emissions are subsequently redistributed to better reflect actual
traffic movement. A modified version of the MOBILES program is used to estimate cruise
emissions and an idle emission factor, while acceleration and deceleration emissions are
20 calculated using modal emission factors as suggested by Ismart (1981). As an alternative, a
shortcut method combining the MOBILE3 estimation of the idle emission factor with values
for individual vehicle emission rates based on speed, temperature, percent hot/cold starts, and
the vehicle scenario is available to the user (Federal Highway Administration, 19 ).
As used in TEXIN2, the MOBILES program provides inspection/maintenance (I/M) and
25 antitampering program (ATP) options. To conserve computer time, several sizable portions
of the extremely large MOBILES program were deleted, namely the nitrogen oxide and
hydrocarbon emission factors modeling and user- supplied corrections to the emission rates.
Since the MOBILE-2 program does not allow for California scenarios, the California data and
options from the MOBILE-2 program were added to the emission routine.
30
March 12, 1990 6-38 DRAFT-DO NOT QUOTE OR CITE
-------
6.5.2.5 CAL3Q
The CAL3Q model utilizes the Connecticut Department of Transportation queueing
model to calculate traffic parameters including queue length. The average speed of vehicles
through the intersection is estimated so a composite MOBILES emission factor can be applied
5 over the length of the queue. In addition, the MOBILES idle emission rate is applied over
the queue length. No modal emission factors are utilized in CAL3Q and the model cannot
handle overcapacity intersections. The emissions, and queue length are input to the
CALINE3 dispersion model to calculate carbon monoxide concentrations at selected receptors.
10 6.5.2.6 CALINE4
The CALINE4 intersection model focuses on a rather complex concept of spatially
resolved modal emissions over links. A CALINE4 intersection link encompasses the
acceleration and deceleration zones created by the presence of the intersection. Each link can
treat only one direction of traffic flow, so that four links are required to model a full
15 intersection.
Four cumulative modal emission profiles representing the deceleration, idle, acceleration
and cruise modes of operation are constructed for each intersection link. These profiles are
determined using the following input variables:
20 SP = Cruise speed (mph)
ACCT = Acceleration time (seconds)
DCLT = Deceleration time (seconds)
IDT1 = Maximum idle time (seconds)
IDT2 = Minimum idle time (seconds)
25 NCYC = Total number of vehicles per cycle per lane
NDLA = Number of vehicles delayed per cycle per lane.
30 NCYC and NDLA are chosen to represent the dominant movement for the link. The
model assumes a uniform vehicle arrival rate, constant acceleration and deceleration rates, and
full stops for all delayed vehicles. Acceleration and deceleration rates (ACCR, DCLR) and
acceleration and deceleration lengths (LACC, LDCL) are determined using the input values
for SPD, ACCT and DCLT. By assuming an "at rest" vehicle spacing (VSP) of 7 meters,
35 the average queue length (LQU) is also determined. IDT1 represents the delay at full stop
March 12, 1990 6-39 DRAFT-DO NOT QUOTE OR CITE
-------
experienced by the first vehicle in the queue. Similarly, IDT2 represents this same measure
for the last vehicle. IDT2 is used to model a platooned arrival and should be assigned a value
of zero for non-platooned applications.
The time rate modal emission factors over the link are computed by a rather complex
5 method. To develop these factors, the model must be provided with composite emission rates
for average route speeds of 0 (idle) and 16 mph. The resulting time rate factors are denoted
as EFA (acceleration), EFD (deceleration), EFC (cruise) and EFI (idle).
The cumulative emission profile for a given mode is developed by determining the time
in mode for each vehicle as a function of distance from link endpoint 1 (ZD), multiplying the
10 time by the respective modal emission rate and summing the results over the number of
vehicles per cycle per lane (NCYC). The elementary equations of motion are used to relate
time to ZD for each mode. The assumed vehicle spacing (VSP) is used to specify the
positional distribution of the vehicles. The total cumulative emissions per cycle per lane at
distance ZD from XLl, YLl are denoted as ECUMk(ZD) in the CALINE4 coding, where the
15 subscript signifies the mode (1 =accel., 2=decel., 3=cruise, 4=idle).
The CALINE4 model handles atmospheric dispersion, somewhat similarly to that in the
CALINE3 model. The most significant difference is that CALINE4, unlike CALINE3,
requires the input of site specific wind direction, fluctuation, and sigma-theta data.
20 6.5.2.7 Comparison of Intersection Models
A study was conducted by Braverman (1987) to compare the emission rates and
distances over which acceleration, cruise, deceleration, and idle emissions occur as generated
by different intersection models. The models included in the study were CAL3Q, GIM,
IMM, TEXIN2, and "Volume 9". CALINE4 was not included in this study because it does
25 not contain an explicit traffic model and does not print out emission rates.
In this study, scenarios for a simple undercapacity, near capacity, and overcapacity
intersection were modeled. All of the parameters for each of the intersection scenarios
modeled were held constant except for the approach volumes which were 408, 608, and 768
vehicles/hour for the under capacity, near capacity, and overcapacity scenarios, respectively.
30 The parameters for the intersection scenarios modeled are given in Table 6-6.
March 12, 1990 6-40 DRAFT-DO NOT QUOTE OR CITE
-------
The results of the emission rates and distances generated by the models are given in
Tables 6-7, 6-8, and 6-9 for the undercapacity, near capacity, and overcapacity intersection
scenarios, respectively. The traffic models contained in "Volume 9" and CAL3Q cannot
handle overcapacity intersections, so no results are reported for these models for the
overcapacity scenario. In addition, the traffic model contained in CAL3Q calculates the near
capacity scenario as slightly overcapacity, so no results are reported for CAL3Q for the near
capacity scenario.
TABLE 6-6. PARAMETERS FOR INTERSECTION SCENARIOS
10
Approach Volumes (Vehicles/Hour) 480*, 608*, and 768*
Approach Capacity (Vehicles/Hour) 640
Capacity Service Volume (Vehicles/Hour of Green Time) 1600
15 Cycle Length (Seconds) 90
Percent Green Time 40
Uninterrupted Speed (Miles/Hour) 30
Number of Lanes 1
Temperature (°F) 40
20 Year 1985
Percent Cold Non-Catalyst 20.6
Percent Hot Catalyst 27.3
Percent Cold Catalyst 20.6
Altitude Low
25 Vehicle Mix Fractions
Light Duty Gas Vehicle 0.652
Light Duty Gas Truck 1 0.128
Light Duty Gas Truck 2 0.087
30 Heavy Duty Gas Vehicle 0.040
Light Duty Diesel Vehicle 0.023
Light Duty Diesel Truck 0.008
Heavy Duty Diesel Vehicle 0.054
Motorcycle 0.007
35
40
*Undercapacity, near capacity, and overcapacity, respectively.
Source: Braverman (1987).
March 12, 1990 6-41 DRAFT-DO NOT QUOTE OR CITE
-------
10
TABLE 6-7. RESULTS OF MODEL COMPARISONS FOR THE UNDERCAPACITY
INTERSECTION SCENARIO
Emission Rate (g/m-s)
Model Unit Distance Distance (m) Emission Rate (g/s)
Volume 9
IMM
TEXIN2
GIM
CAL3Q
0.0327
0.0119*
0.0236*
0.0102
0.0218
106.3
220.1
81.7
162.1
54.0
3.476
2.615
1.930
1.653
1.177
15
"'Emission Rate was obtained by multiplying each component (acceleration, United Distance deceleration, and
idle) of emissions times the length over which they occur individually, summing the products, and dividing by
the entire length of emissions.
20 Source: Braverman (1987).
TABLE 6-8. RESULTS OF MODEL COMPARISONS FOR THE NEAR CAPACITY
INTERSECTION SCENARIO
25 ;
Emission Rate (g/m-s)
Model Unit Distance Distance (m) Emission Rate (g/s)
30
Volume 9
IMM
TEXIN2
GIM
0.0452
0.0229*
0.0413*
0.0124
269.7
294
141.5
193.4
12.19
6.74
5.85
2.40
35
*Emission Rate was obtained by multiplying each component (acceleration, United Distance deceleration, and
idle) of emissions times the length over which they occur individually, summing the products, and dividing by
the entire length of emissions.
40
Source: Braverman (1987).
March 12, 1990 6-42 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 6-9. RESULTS OF MODEL COMPARISONS FOR THE OVERCAPACITY
INTERSECTION SCENARIO
Emission Rate (g/m-s)
Model Unit Distance Distance (m) Emission Rate (g/s)
IMM
GIM
TEXIN2
0.0413*
0.0228
0.0426*
611
733.3
176.6
25.24
16.7
7.53
10
*Emission Rate was obtained by multiplying each component (acceleration, United Distance deceleration, and
15 idle) of emissions times the length over which they occur individually, summing the products, and dividing by
the entire length of emissions.
Source: Braverman (1987).
20
The results indicate that under- to near-capacity situations the order of model predicted
emission rates from highest to lowest is "Volume 9", IMM, TEXIN2, GIM, CAL3Q. For
overcapacity situations, the order of model predicted emission rates from highest to lowest is
IMM, GIM, TEXIN2. Note that "Volume 9", IMM, and TEXIN2 utilize modal emission
25 factors. GIM and CAL3Q do not utilize modal emission factors; instead they utilize
composite MOBILES emission factors. Since "Volume 9", IMM, and TEXIN2 give higher
emission rates for the under to near capacity situations than GIM and CAL3Q, it can be
inferred that models that utilize modal emission factors generate higher emission rates than
models that utilize composite MOBILES emission factors. The reason that GIM gives higher
30 emissions than TEXIN2 for the overcapacity scenario is that TEXIN2 simply extrapolates
from the undercapacity situation to generate overcapacity emission rates, whereas GIM treats
overcapacity as an entirely different situation.
6.5.3 Urban Area Modeling
35 Several models have been used to model the cumulative effects of urban CO sources
such as roadways and wood stoves. The Guideline on Air Quality Models (Revised) (U.S.
Environmental Protection Agency, 1986) does not recommend one specific model for
March 12, 1990 6-43 DRAFT-DO NOT QUOTE OR CITE
-------
urbanwide CO analysis; instead, it recommends these analyses be considered on a case-by-
case basis.
Urban exceedances of the 8-hour CO NAAQS result primarily from the cumulative
effects of motor vehicle emissions throughout the urban area. The APRAC-3 model (Simmon
5 et al., 1981), which is briefly described below, was developed to handle this situation.
6.5,3.1 APRAC-3
APRAC-3 is a Gaussian-plume diffusion model which computes hourly average carbon
monoxide concentrations for any urban location. The model calculates contributions from
10 dispersion on various scales: extraurban, mainly from sources upwind of the city of interest;
intraurban, from freeway, arterial, and feeder street sources; and local, from dispersion
within a street canyon. APRAC-3 requires an extensive traffic inventory for the city of
interest.
Traffic links may have arbitrary length and orientation. Off-link traffic is allocated to
15 two-mile square grids. Link traffic emissions are aggregated into a receptor oriented area
source array. The boundaries of the area sources actually treated are (1) arcs at radial
distances from the receptor which increase in geometric progression, (2) the sides of a 22.5°
sector oriented upwind for distances greater than 1,000 m, and (3) the sides of a 45° sector
oriented upwind for distances less than 1,000 m. A similar area source array is established
20 for each receptor. Up to 625 receptors are accepted for a single-hour.
Meteorological data requirements are hourly wind direction (nearest 10 degrees), hourly
wind speed, and hourly cloud cover for stability calculations. Constant, uniform (steady-
state) wind is assumed within each hour. The model can interpolate winds at receptors if
more than one wind is provided. Mixing height is ignored until the concentration equals that
25 calculated using a box model. A box model the (uniform vertical distribution) is used beyond
that distance.
A secondary contributor to some urban exceedances of the 8-hour CO NAAQS is the
cumulative effect of wood stove emissions throughout the urban area. These emissions are
trapped under the nighttime radiation inversion along with evening traffic emission on cold
30 clear nights with light and variable winds. This situation can be handled best by either a
Gaussian or a numerical model. A numerical model provides a better treatment than a
March 12, 1990 6-44 DRAFT-DO NOT QUOTE OR CITE
-------
Gaussian model of the time dependent changes in meteorology under these conditions.
However, use of numerical models is extremely data and resource intensive. Thus, numerical
models have only been applied in very large cities. One numerical model that has been used
in a few cases is the Urban Airshed Model (Ames et ah, 1985). An acceptable Gaussian
5 model for urban area applications is RAM (Catalano et ah, 1987). A brief description of the
Urban Airshed Model and RAM follows.
6.5.3.2 Urban Airshed Model
The Urban Airshed Model (Ames et ah, 1985) simulates the major physical and
10 chemical processes in the polluted troposphere. These include gas phase chemistry, advective
transport, and turbulent diffusion. The modeling domain is divided into a large array of grid
cells. Horizontally the cells are uniformly sized squares 3 to 5 kilometers on a side.
Typically, four or five layers of cells represent the vertical domain. The depth of the layers
is scaled by the height of the mixed layer and the height of the top of the modeling domain
15 (region top). The latter typically ranges from 500 meters in the morning hours to
1,000 meters or more in the afternoon. Emissions are injected into individual cells depending
on the location of the sources, their height of release, and the buoyant rise of individual stack
gas plumes.
The theoretical basis for the Urban Airshed Model rests of the conservation of mass
20 equation for atmosphere diffusion. Primary inputs to the Urban Airshed Model are point and
area source emissions, initial and boundary concentrations both at the surface and aloft, and a
variety of meteorological data. These include a three-dimensional wind field, mixing depths,
surface temperature, and exposure class, the latter an indicator of thermal instability.
25 6.5.3.3 RAM
RAM (Catalano et ah, 1987) provides a readily available computer program based on
the assumptions of steady-state Gaussian dispersion for short-term (one-hour to one-day)
determination of urban air quality resulting from pollutants released from point and/or area
sources.
30 RAM is applicable for locations with level or gently rolling terrain where a single wind
vector for each hour is a reasonable approximation of the flow over the source area
March 12, 1990 6-45 DRAFT-DO NOT QUOTE OR CITE
-------
considered. Calculations are performed for each hour. Hourly meteorological data required
are wind direction, wind speed, temperature, stability class, and mixing height.
Computations are performed hour by hour as if the atmosphere had achieved a steady-
state condition. Therefore, errors will occur where there is a gradual buildup (or decrease) in
5 concentrations from hour to hour, such as with light wind conditions. Also under light wind
conditions the definition of wind direction is likely to be inaccurate, and variations in the
wind flow from location to location in the area are quite probable.
Considerable time is saved in calculating concentrations from area sources by using a
narrow plume simplification which considers sources at various distances on a line directly
10 upwind from the receptor to be representative in the crosswind direction of the sources at
those distances affecting the receptor. Area source sizes are used as given in the emission
inventory in lieu of creating an internal inventory of uniform elements.
15 6.6 REFERENCES
AIRS, Aerometric Information Retrieval System [data base], (n. d.) [Standard computer retrievals]. Unpublished
20 computer reports available from: U. S. Environmnetal Protection Agency, Office of Air Quality Planning
and Standards, Research Triangle Park, NC 27711.
Altshuller, A. P.; Ortman, G. C.; Saltzman, B. E.; Neligan, R. E. (1966) Continuous monitoring of methane and
other hydrocarbons in urban atmospheres. J. Air Pollut. Control Assoc. 16: 87-91.
25
Ames, J.; Myers, T. C.; Reid, L. E.; Whitney, D. C.; Golding, S. H. (1985) SAI Airshed Model operations
manuals. Volume I-user's manual. Research Triangle Park, NC: U. S. Environmental Protection
Agency, Atmospheric Sciences Research Laboratory; EPA report no. EPA-600/8-85-007A. Available
from: NTIS, Springfield, VA; PB85-191567.
30
Anonymous. (1976) Report of the air monitoring siting workshop. Las Vegas, NV.
Bach, W. D.; Crissman, B. W.; Decker, C. E.; Minear, J. W.; Rasberry, P. P.; Tommerdahl, J. B. (1973)
Carbon monoxide measurements in the vicinity of sports stadiums. Research Triangle Park, NC: U. S.
35 Environmental Protection Agency, Office of Air Quality Planning and Standards; EPA report no.
EPA-450/3-74-049. Available from: NTIS, Springfield, VA; PB-250850.
Bencala, K. E.; Seinfeld, J. H. (1976) On frequency distributions of air pollutant concentrations. Atmos.
Environ. 10: 941-950.
40
Benson, P. E. (1979) CALINE3 - a versatile dispersion model for predicting air pollutant levels near highways
and arterial streets. Washington, DC: Federal Highway Administration; report no. FHWA/CA/TL-79/23.
March 12, 1990 6-46 DRAFT-DO NOT QUOTE OR CITE
-------
Benson, P. E. (1984) CALJNE4 - a dispersion model for predicting air pollutant concentrations near roadways.
Washington, DC: Federal Highway Administration; report no. FHWA/CA/TL-84/15.
Braverman, T. N. (1987) Intersection model comparison. Research Triangle Park, NC: U. S. Environmental
5 Protection Agency.
Bullin, J. A.; Korpics, J. J.; Hlavinka, M. W. (1986) User's guide to TEXIN2 Model~a model for predicting
carbon monoxide concentrations near intersections. College Station, TX: Texas Transportation Institute.
10 Catalano, J. A.; Turner, D. B.; Novak, J. H. (1987) User's guide for RAM. 2nd ed. Research Triangle Park,
NC: U. S. Environmental Protection Agency; EPA report no. EPA-600/8-87-046.
Chock, D. P. (1978) A simple line-source model for dispersion near roadways. Atmos. Environ. 12: 823-829.
15 Code of Federal Regulations. (1977) Requirements for preparation, adoption, and submission of implementation
plans. C. F. R. 40: sect. 51.
EMI Consultants. (1985) The Georgia intersection model for air quality analysis. Knoxville, TN: EMI
Consultants.
20
Federal Aviation Administration. (1988) FAA air traffic activity. Washington, DC: U. S. Department of
Transportation, Office of Management Systems.
Federal Highway Administration, (n.d.) Factors in highway-project analysis. Washington, DC: Federal Highway
25 Administration; FHWA technical advisory T6640.3.
Federal Register. (1979) Ambient air quality monitoring, data reporting and surveillance provisions. F. R. (May
10) 44: 27558-27604.
30 Georgopoulos, P. G.; Seinfeld, J. H. (1982) Statistical distributions of air pollutant concentrations. Environ. Sci.
Technol. 16: 401A-416A.
Hare, C. T.; Springer, K. J. (1974) Exhaust emissions from uncontrolled vehicles and related equipment using
internal combustion engines: part 4, small air-cooled spark ignition utility engines; part 5, heavy-duty
35 farm, construction, and industrial engines; part 7, snowmobiles. Research Triangle Park, NC: U. S.
Environmental Protection Agency; EPA contract no. EHS 70-108. Available from: NTIS, Springfield,
VA; PB-224885, PB-232507, and PB-238295.
Horowitz, J.; Barakat, S. (1979) Statistical analysis of the maximum concentration of an air pollutant: effects of
40 autocorrelation and nonstationarity. Atmos. Environ. 13: 811-812.
Ismart, D. (1981) Mobile source emissions and energy analysis at an isolated intersection. Washington, DC:
Federal Highway Administration.
45 Kunselman, P.; McAdams, H. T.; Domke, C. J.; Williams, M. (1974) Automobile exhaust emission modal
analysis model. Ann Arbor, MI: U. S. Environmental Protection Agency, Office of Mobile Source Air
Pollution Control; EPA report no. EPA-460/3-74-005. Available from: NTIS, Springfield, VA;
PB-229635.
50 National Cooperative Highway Research Program. (1979) Development of an improved highway capacity manual.
National Cooperative Highway Research Program report 3-28.
March 12, 1990 6-47 DRAFT-DO NOT QUOTE OR CITE
-------
NEDS, National Emissions Data System [data base], (n.d.) [Standard computer retrievals]. Unpublished computer
report available from: U. S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, Research Triangle Park, NC.
5 New York State Department of Transportation. (1980) Intersection Midblock Model user's guide. Albany, NY:
New York State Department of Transportation.
Ott, W. R.; Mage, D. T.; Randecker, V. W. (1979) Testing the validity of the lognormal probability model:
computer analysis of carbon monoxide data from U. S. cities. Washington, DC: U. S. Environmental
10 Protection Agency, Office of Monitoring and Technical Support; EPA report no. EPA-600/4-79-040.
Petersen, W. B. (1978) User's guide for PAL - a Gaussian-plume algorithm for point, area, and line sources.
Research Triangle Park, NC: U. S. Environmental Protection Agency, Environmental Sciences Research
Laboratory; EPA report no. EPA-600/4-78-013. Available from: NTIS, Springfield, VA; PB80-227564.
15
Petersen, W. B. (1980) User's guide for HIWAY-2: a highway air pollution model. Research Triangle Park, NC:
U. S. Environmental Protection Agency, Environmental Sciences Research Laboratory; EPA report no.
EPA-600/8-80-018. Available from: NTIS, Springfield, VA; PB80-227556.
20 Pollack, R. I. (1975) Studies of pollutant concentration frequency distributions. Research Triangle Park, NC:
U. S. Environmental Protection Agency, National Environmental Research Center; EPA report no.
EPA-650/4-75-004. Available from: NTIS, Springfield, VA; PB-242549.
Simmon, P. B.; Patterson, R. M.; Ludwig, F. L.; Jones, L. B. (1981) The APRAC-3/Mobile 1 emissions and
25 diffusion modeling package. San Francisco, CA: U. S. Environmental Protection Agency, Region IX;
EPA report no. EPA-909/9-81-002. Available from: NTIS, Springfield, VA; PB82-103763.
Simpson, R. W.; Butt, J.; Takeman, A. J. (1984) An averaging time model of SO2 frequency distributions from
a single point source. Atmos. Environ. 18: 115-118.
30
Smith, W. A. (1985) Updated methodology to assess carbon monoxide (CO) impact at intersections
[memorandum to Richard G. Rhoads]. Atlanta, GA: U. S. Environmental Protection Agency; January 23.
Tiao, G. C.; Box, G. E. P.; Hamming, W. J. (1975) A statistical analysis of the Los Angeles ambient carbon
35 monoxide data 1955-1972. J. Air Pollut. Control Assoc. 25: 1129-1136.
U. S. Department of Commerce. (1988) Statistical abstract of the United States. 108th ed. Washington, DC:
Bureau of the Census.
40 U. S. Department of Commerce, (n.d.) Current industrial reports. Washington, DC: Bureau of the Census.
U. S. Department of Energy. (1982) Estimates of U.S. wood energy consumption from 1949 to 1981.
Washington, DC: Energy Information Adminstration; publication no. DOE/EIA-0341.
45 U. S. Department of Energy. (1984) Estimates of U.S. wood energy consumption 1980-1983. Washington, DC:
U. S. Department of Energy, Office of Coal, Nuclear, Electric and Alternate Fuels; publication no.
DOE/EIA-0341(83).
U. S. Department of Energy. (1988a) Petroleum marketing monthly. Washington, DC: Energy Information
50 Administration; publication no. DOE/EIA-0380(88/06).
U. S. Department of Energy. (1988b) Coal distribution January-December. Washington, DC: Energy Information
Administration; publication no. DOE/EIA-25(88/4Q).
March 12, 1990 6-48 DRAFT-DO NOT QUOTE OR CITE
-------
U. S. Department of Energy. (1988c) Electric power annual. Washington, DC: Energy Information
Administration; publication no. DOE/EIA-0348(87).
U. S. Department of Energy. (1989a) Natural gas annual. Washington, DC: Energy Information Administration;
5 publication no. DOE/EIA-0131(88)/1.
U. S. Department of Energy. (1989b) Cost and quality of fuels for electric utility plants 1988. Washington, DC:
Office of Coal, Nuclear, Electric and Alternative Fuels; publication no. DOE/EIA-0191(88).
10 U. S. Department of Health, Education, and Welfare. (1968) 1968 national survey of community solid waste
practices. Cincinnati, OH: Public Health Service; PHS publication no. 1867.
U. S. Department of the Interior. (1971) Coal refuse fires: an environmental hazard. Washington, DC: Bureau of
Mines; information circular 8515.
15
U. S. Department of the Interior. (1987) Minerals yearbook. Washington, DC: Bureau of Mines.
U. S. Department of Transportation. (1988) Highway statistics. Washington, DC: Federal Highway
Administration.
20
U. S. Environmental Protection Agency. (1971) Particulate pollutant system study. Research Triangle Park, NC:
National Air Pollution Control Administration; contract no. CPA 22-69-104. Available from: NTIS,
Springfield, VA; PB-203522, PB-203521 and PB-203128.
25 U. S. Environmental Protection Agency. (1974a) Air quality data - 1972 annual statistics. Research Triangle
Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA-450/2-74-001. Available
from: NTIS, Springfield, VA; PB-232588.
U. S. Environmental Protection Agency. (1974b) Air quality data - 1973 annual statistics. Research Triangle
30 Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA-450/2-74-015.
U. S. Environmental Protection Agency. (1976a) Air quality data - 1969 annual statistics. Research Triangle
Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA-450/2-76-018.
35 U. S. Environmental Protection Agency. (1976b) Air quality data - 1970 annual statistics. Research Triangle
Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA-450/2-76-019.
U. S. Environmental Protection Agency. (1976c) Air quality data - 1971 annual statistics. Research Triangle
Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA-450/2-76-020.
40
U. S. Environmental Protection Agency. (1976d) Air quality data - 1974 annual statistics. Research Triangle
Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA-450/2-76-011.
U. S. Environmental Protection Agency. (1976e) National air quality and emissions trends report, 1975. Research
45 Triangle Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA-450/1-76-002.
Available from: NTIS, Springfield, VA; PB-263922.
U. S. Environmental Protection Agency. (1977a) Guidelines: air quality surveillance networks. Research Triangle
Park, NC: Office of Air Programs; publication no. AP-98.
50
U. S. Environmental Protection Agency. (1977b) Air quality data - 1975 annual statistics including summaries
with reference to standards. Research Triangle Park, NC: Office of Air Quality Planning and Standards;
EPA report no. EPA-450/2-77-002. Available from: NTIS, Springfield, VA; PB83-127712.
March 12, 1990 6-49 DRAFT-DO NOT QUOTE OR CITE
-------
U. S. Environmental Protection Agency. (1977c) Guidelines for the interpretation of air quality standards.
Research Triangle Park, NC: Office of Air Quality Planning and Standards; report no 1.2-008. (Guideline
series).
5 U. S. Environmental Protection Agency. (1978) Guidelines for air quality maintenance planning and analysis.
Volume 9 (revised): evaluating indirect sources. Research Triangle Park, NC: Office of Air Quality
Planning and Standards; EPA report no. EPA-450/4-78-001. Available from: NTIS, Springfield, VA;
PB-288206.
10 U. S. Environmental Protection Agency. (1985) Compilation of air pollutant emission factors: v. 1, stationary
point and the area sources, v. 2, mobile sources. 4th ed. Research Triangle Park, NC: Office of Air
Quality Planning and Standards; EPA report nos. AP-42-ED-4-VOL-1 and AP-42-ED-4-VOL-2.
Available from: NTIS, Springfield, VA; PB86-124906 and PB87-205266.
15 U. S. Environmental Protection Agency. (1986) Guideline on air quality models (revised). Research Triangle
Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA-450/2-78-027R. Available
from: NTIS, Springfield, VA; PB86-245248.
U. S. Environmental Protection Agency. (1987a) National air quality and emissions trends report, 1985. Research
20 Triangle Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA-450/4-87-001.
Available from: NTIS, Springfield, VA; PB87-180352.
U. S. Environmental Protection Agency. (1987b) National air pollutant estimates, 1940-1985. Research Triangle
Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA-450/4-86-018. Available
25 from: NTIS, Springfield, VA; PB87-181236.
U. S. Environmental Protection Agency. (1988) National air quality and emissions trends report, 1986. Research
Triangle Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA/450/4-88/001.
Available from NTIS, Springfield, VA; PB88-172473/XAB.
30
U. S. Environmental Protection Agency. (1989a) Supplement to compilation of air pollutant emission factors.
Research Triangle Park, NC: Office of Air Quality Planning Standards; EPA report no. AP-42 (in press).
U. S. Environmental Protection Agency. (1989b) User's guide to MOBILE4 (Mobile Source Emissions Factor
35 Model). Ann Arbor, MI: Office of Mobile Sources; EPA report no. EPA-AA-TEB-89-01. Available
from: NTIS, Springfield, VA; PB89-164271.
U. S. Environmental Protection Agency. (1990a) National air quality and emissions trends report, 1988. Research
Triangle Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA-450/4-90-002.
40
U. S. Environmental Protection Agency. (1990b) National air pollutant emissions estimates, 1940-1988. Research
Triangle Park, NC: Office of Air Quality Planning and Standards; EPA report no. EPA-450/4-90-001.
U. S. Forest Service. (1988) Wildfire statistics. Washington, DC: U. S. Department of Agriculture, State and
45 Private Forestry.
Wackter, D. J.; Bodner, P. (1986) Evaluation of mobile source air quality simulation models. Research Triangle
Park, NC: U. S. Environmental Protection Agency, Office of Air Quality Planning and Standards; EPA
report no. EPA-450/4-86-002. Available from: NTIS, Springfield, VA; PB86-167293.
50
Wolcott, M. (1986) Volume 9 update [memorandum to Raymond Vogel]. Ann Arbor, MI: U. S. Environmental
Protection Agency, Office of Mobile Sources; January 14.
March 12, 1990 6-50 DRAFT-DO NOT QUOTE OR CITE
-------
Yamate, G. (1974) Emissions inventory from forest wildfires, forest managed burns, and agricultural bums.
Research Triangle Park, NC: U. S. Environmental Protection Agency; EPA report no.
EPA-450/3-74-062. Available from: NTIS, Springfield, VA; PB-238766.
Zimmerman, J. R.; Thompson, R. S. (1975) User's guide for HIWAY: a highway air pollution model. Research
Triangle Park, NC: U. S. Environmental Protection Agency, National Environmental Research Center;
EPA report no. EPA-650/4-74-008. Available from: NTIS, Springfield, VA; PB-239944.
March 12, 1990 6-51 DRAFT-DO NOT QUOTE OR CITE
-------
7. INDOOR CARBON MONOXIDE SOURCES,
EMISSIONS, AND CONCENTRATIONS
5 7.1 INTRODUCTION
The activities of individuals are the most important determinants of their exposure to air
borne contaminants. In the course of a day individuals spend varying amounts of time in a
variety of microenvironments (residences, industrial and nonindustrial workplaces,
automobiles, public access buildings, outdoors, etc.). Exposures across all
10 microenvironments need to be assessed in evaluating adverse health or comfort effects and in
formulating cost-effective mitigation measures to reduce or minimize the risks associated with
exposure. Exposure can be assessed by direct methods (personal monitoring or by measuring
biomarkers of exposure) or by indirect methods (microenvironmental monitoring and
questionnaires combined with an appropriate human exposure model).
15 In recent years there has been a growing recognition of the importance of nonindustrial
indoor microenvironments in assessing exposures to a wide range of air contaminants
(National Research Council, 1981; World Health Organization, 1985). This recognition
reflects the fact that concentrations for many important air contaminants are higher in many
indoor microenvironments than outdoors and that most individuals spend little time outdoors.
20 Carbon monoxide is introduced to indoor environments through emissions from a variety
of combustion sources and in the infiltration or ventilation air from outdoors. The resulting
indoor concentration, both average and peak, is dependent on a complex interaction of several
interrelated factors affecting the introduction, dispersion, and removal of CO. These factors
include, for example, such variables as: (1) the type, nature (factors affecting the generating
25 rate of CO), and number of sources; (2) source-use characteristics; (3) building
characteristics; (4) infiltration or ventilation rates; (5) air mixing between and within
compartments in an indoor space; (6) removal rates and potential remission or generation by
indoor surfaces and chemical transformations; (7) existence and effectiveness of air
contaminant removal systems; and (8) outdoor concentrations; etc. The interaction of these
30 factors to produce the resulting indoor concentrations usually is considered within the
framework of the mass balance principle.
March 12, 1990 7-1 DRAFT - DO NOT QUOTE OR CITE
-------
In its simplest form, where steady-state or equilibrium conditions are assumed for a
single compartment with complete mixing and no air cleaner, the mass-balance model can be
represented by the following equation.
5 Q = C, + C2 (7-1)
where:
C, = PAQ/A + K = outdoor air contribution (7-2)
10
C2 = S/V/A + K = indoor source contribution (7-3)
15 and where:
Q = steady-state indoor concentration of CO (jig/m3)
Cj = contribution to indoor CO from outdoor air 0*g/m3)
20
C2 = contribution to indoor CO from indoor sources (/*g/m3)
C0 = outdoor CO (jtg/m3)
25 P = fraction of outdoor CO that penetrates the building shell
A = air-exchange rate in air changes per hour - ACH per hour
S = generation rate or source strength of CO (jwg/hr)
30
V = volume of the indoor space (m3)
K = removal rate of CO by indoor surfaces or chemical transformations -
equivalent ACH
35
This simplified form of the model generally is used to evaluate CO levels indoors. In
actuality, however, indoor spaces often are multicompartments with incomplete mixing where
40 the source-generation and contaminant-removal rates and air-contaminant concentrations vary
considerably in time. Equation 7-1 is useful particularly for determining the impact on indoor
air-contaminant concentrations from sources that are used over relatively long periods of time
March 12, 1990 7-2 DRAFT - DO NOT QUOTE OR CITE
-------
(e.g., unvented kerosene or gas space heaters) where steady-state or equilibrium conditions
are reached. When applied to sources that are intermittent in their use (e.g., gas range or
tobacco combustion), Equation 7-1 averages over the off/on periods of the sources to
determine average input parameters for the model. Short-term indoor concentrations of air
5 contaminants associated with sources whose use varies considerably with time can be modeled
with the differential version of Equation 7-1, when detailed information on the time
variability of the source use, mixing, and removal terms are available. Field data on short-
term variability of contaminant concentrations and associated variables, however, are not
available.
10 Carbon monoxide generally is assumed have low reactivity indoors (Yamanaka, 1984;
Leaderer et al., 1986; Traynor et al., 1982; Borrazzo et al., 1987; Caceres et al., 1983), that
is, CO removal by indoor surfaces or chemical transformations is approximately equal to zero
for typical indoor CO residence times (K=0 in Equations 7-2 and 7-3). There are no
chamber or field studies that have measured the penetration factor for CO. Given the low CO
15 reactivity, P generally is assumed equal to one (P= 1) for conditions where outdoor levels of
CO do not vary rapidly. Under these typical assumptions (K=0 and P= 1), indoor
concentrations of CO can be represented by the following simplified form of Equations 7-1,
7-2, and 7-3.
20 Q = C0 + S/V/A (7-4)
In the case where there are no indoor sources, then the average indoor concentration is equal
25 to the average outdoor concentration. With short variations in outdoor concentrations,
however, indoor CO concentrations will lag outdoor concentrations and will be dependent on
the air-exchange rate in a space. When an indoor source exists, the indoor CO concentration
will be equal to the outdoor concentration plus the contribution of the indoor source, which is
a function of the source strength (CO emission rate), volume of the indoor environment and
30 air-exchange rate.
This chapter will summarize the data currently available on emissions of CO from
sources commonly found indoors and on levels of CO measured in a variety of indoor
microenvironments.
March 12, 1990 7-3 DRAFT - DO NOT QUOTE OR CITE
-------
7.2 EMISSIONS FROM INDOOR SOURCES
Carbon monoxide emitted directly into the indoor environment is one of several air
contaminants resulting from combustion sources. Such emissions into occupied spaces can be
unintentional or the result of accepted use of unvented or partially vented combustion sources.
5 Faulty or leaky flue pipes, backdrafting and spillage from combustion appliances that draw
their air from indoors, (i.e., Moffatt, 1986), improper use of combustion sources (i.e., use of
a poorly maintained kerosene heater), and air intake into a building from attached parking
garages are all examples of unintentional or accidental indoor sources of CO. The U.S.
Departmental of Health and Human Services (1986) estimates between 700 and 1,000 deaths
10 per year in the U.S. alone are due to accidental carbon monoxide poisoning. The number of
individuals experiencing severe adverse health effects at sublethal carbon monoxide
concentrations from accidental indoor sources is no doubt many times the estimated deaths.
While the unintentional or accidental indoor sources of CO represent a serious health hazard,
little is known about the extent of the problem. Such sources cannot be characterized for CO
15 emissions in any standard way that would make the results extendable to the general
population. Unintentional or accidental sources are not considered in this review of emissions
of CO from indoor sources.
The major indoor sources of CO emissions that result from the accepted use of unvented
or partially vented combustion sources include gas cooking ranges and ovens, gas appliances,
20 unvented gas space heaters, unvented kerosene space heaters, cigarette combustion, and wood
burning stoves. This section of the chapter will summarize the available CO-emission data
for the major indoor-sources of CO. The experimental design, measurement methods, and
results of studies of indoor-source CO emissions will be discussed.
The characterization of CO emissions from indoor sources is essential in providing
25 source strength input data for indoor modeling, understanding fundamental processes
influencing emissions, guiding field study designs aimed at assessing indoor CO exposures,
identifying and ranking important sources, and in developing cost-effective mitigation
measures that will minimize exposures.
30
March 12, 1990 7-4 DRAFT - DO NOT QUOTE OR CITE
-------
7.2.1 Emissions from Gas Cooking Ranges, Gas Ovens, and Gas
Appliances
Estimates indicate that gas (natural gas and liquid propane) is used for cooking, heating
water, and drying clothes in approximately 45.1% of all homes in the United States (U.S.
5 Bureau of the Census, 1982) and near 100% of the homes in other countries (e.g., The
Netherlands). Unvented, partially vented, and improperly vented gas appliances, particularly
the gas cooking range and oven, represent an important source category of CO emissions into
the indoor residential environment. Emissions of CO from these gas appliances (the source
term, S, in Equation 7-4) are a function of a number of variables relating the source type
10 (range top or oven, water heater, dryer, number of pilot lights, burner design, etc.), source
condition (age, maintenance, combustion efficiency, etc.), source use (number of burners
used, frequency of use, fuel consumption rate, length of use, improper use, etc.) and venting
of emissions (existence and use of outside vents over ranges, efficiency of vents, venting of
gas dryers, etc.).
15 A number of chamber studies have investigated CO emissions from gas cooking ranges,
ovens, and appliances. The studies have used two basic approaches. The first is called the
direct or sampling-hood approach (Moschandreas et al., 1985; Himmel and DeWerth, 1974)
and the second is the mass-balance or chamber approach (Moschandreas et al., 1985; Traynor
et al., 1982; Leaderer, 1982). In the direct approach, emissions are sampled using a quartz
20 hood through which the combustion emissions pass and are sampled. Because the effluent
gases measured by this method contain substantial amounts of excess air that can vary
considerably from run to run, the measured concentrations are converted to a hypothetical
undiluted or "air-free" basis for calculating CO emission rates. This method is used in both
chamber and field evaluations of emission rates (Moschandreas et al., 1985; Borrazzo et al.,
25 1987). The mass-balance approach utilizes a well-mixed environmental chamber where the
relationship between changes in concentration of CO over time in relation to outdoor
concentration, source-emission rate, air-exchange rate, and removal rate are evaluated (mass-
balance equation). Both approaches yield source emission rates for CO, usually expressed as
micrograms per kilojoule (/xg/kJ).
30 Emissions from gas range-top burners typically are evaluated using a standardized water
load in a cooking pot (American National Standards Institute Z21.1-1974, 1982) or a
March 12, 1990 7-5 DRAFT - DO NOT QUOTE OR CITE
-------
modification of it (Borrazzo et al., 1987). The cooking pot has a top that is sealed, except
for a 3/4 in. pipe that extends from the center of the top to allow steam to escape.
Using both the direct-sampling hood and mass-balance approach and a standardized
water load, Moschandreas et al. (1985) evaluated CO emissions from three new gas ranges
5 (including pilot and non-pilot light and self-cleaning oven and conventional oven) with six
hours of conditioning before use in actual testing. The data of Cole and Zawacki (1985) are
incorporated into the Moschandreas et al. (1985) report. The range CO emissions were
evaluated for appliance type, the conditions of blue-flame operation (burner air shutter set at
the manufacturer's recommended level), and yellow-tipping (air shutters are closed - worst
10 condition). Two natural gas mixes were evaluated: lean mix (983 Btu/scf) and rich mix
(1022 Btu/scf).
A total of 116 direct-sampling experiments were conducted in which CO emissions were
measured for all 12 burners (four burners per stove) using rich and lean fuels. Thirty direct-
sampling experiments were conducted with the burners operating with yellow tipping. No
15 significant differences (criterion: p<0.05) were found by fuel type. The results of the tests
using blue and yellow flame setting are shown in Table 7-1. The improper operation of the
burners (yellow-tipping flame) resulted in an approximately two- to fivefold increase in CO
emissions. Considerable variation in CO emissions from burner to burner within and between
ranges were noted by the authors for blue-flame operation although overall CO emission
20 averages for burners by each range were within a factor of two.
A total of 144 mass-balance experiments on the three ranges were conducted in a 33-m3
chamber to evaluate the impact of fuel composition, range type, primary aeration level (blue
or yellow flame), fuel-consumption rate (high-9149 Btu/h, medium-7673 Btu/h,
low-1492 Btu/h, and warm-807 Btu/h), test chamber air-exchange rate, temperature and
25 humidity (range of 15 to 50%), and temporal effects on CO-emission rates. CO emissions
showed little variation by fuel composition (lean vs. rich), range type, test chamber air-
exchange rate, and time. Yellow-tipping flame conditions resulted in a two- to tenfold
increase in CO emissions over blue-flame operation conditions. No changes in emissions
were noted as a function of changes in chamber temperature and relative humidity. Little
30 change in CO emissions were noted for the high, medium, and low fuel-consumption rates
while a sevenfold increase in emissions were observed for the warm setting. In a comparison
March 12, 1990 7-6 DRAFT - DO NOT QUOTE OR CITE
-------
TABLE 7-1. CARBON MONOXIDE-EMISSION RATES' FOR 12 RANGE-TOP BURNERS
OPERATING WITH BLUE AND YELLOW-TIPPING FLAMES BY THE DIRECT
SAMPLING METHOD (calculated from Moschandreas et al., 1985)
10
Number of
Tests
25
33
58
11
9
12
Gas
Range
1
2
3
1
2
3
Flame
Condition
Blue"
Blue
Blue
Yellow0
Yellow
Yellow
Emission
Rate 0*g/kJ)
Average
50.7
34.3
70.9
190.0
196.9
108.4
SD
9.46
4.93
12.9
5.8
3.8
5.6
Range
17.2 - 107.5
15.1 - 71.4
15.1-215.0
53.8 - 344.0
60.2 - 344.0
94.6 - 227.9
15
'Lean and rich fuel mixtures combined.
20 bBlue-flame condition - well tuned.
"Yellow-tipping flame - improperly tuned.
25 of CO emission results obtained from both the direct-sampling and mass-balance experiments
(Moschandreas et al., 1985), for blue-flame and yellow-tipping flame conditions, differences
were observed. No clear trend emerged, however, because CO emissions varied among
experimental runs by as much as an order of magnitude.
As part of the above study CO emissions from gas ovens, gas range pilot lights, and a
30 gas dryer were evaluated for blue-flame operating conditions. The results of these
experiments are shown in Table 7-2. Pilot light emissions are comparable to those of gas
range burners (Table 7-1). Variability by oven use was observed but the limited data is not
sufficient to draw any conclusions. Gas dryer CO emissions appear to be comparable to those
from gas range burners.
35 Eighteen different gas ranges, representing greater than 90% of the gas stoves in use at
the time, were tested for CO emissions by the American Gas Association Laboratories
(Himmel and Dewerth, 1974). CO emissions were evaluated for top burners, ovens and
broilers, burner pilot lights, and oven pilot lights. The protocol utilized the direct
measurement method for both blue-flame (well-adjusted flame) and yellow-tipping flame
40 (poorly adjusted flame) operating conditions. Range-top burner evaluations, a total of 72,
March 12, 1990 7-7 DRAFT - DO NOT QUOTE OR CITE
-------
10
15
20
25
TABLE 7-2. CARBON MONOXIDE-EMISSION RATES' FOR GAS RANGE OVENS,
GAS RANGE PILOT LIGHTS, AND GAS DRYERS
(calculated from Moschandreas et al., 1985)
Gas
Range
1
1
2
3
3
3
1
1
2
Gas dryer"
Burner
Bake
Broil
Broil
Bake
Broil
Self-clean
Oven door openb
Oven door closedb
Pilot light0
Number
of
Tests
4
4
9
2
7
3
3
3
20
4/3
Average Emission Rates
G*g/kJ)
Direct Method
19.1 (6.7)
13.8 (3.9)
13.2(2.1)
68.5(6.1)
21.5 (1.0)
16.0 (2.2)
40.4 (3.0)
(SD)
Mass Balance
54.6 (2.6)
127.3 (3.5)
40.4 (5.2)
68.8 (14.2)
"Combined and rich and lean fuel and no oven load.
bBroiler operated.
"Results are on a per pilot light basis, experiments covered various pilot light combinations.
dUsing Association of Home Appliance Manufacturers Standard HLD-2EC, four tests are by the
direct method and three tests by the mass-balance method.
30 were made with a standard water pot load centered on the grate (American National Standards
Institute, 1974). Oven tests, 27 in all, were conducted without a cooking load because the
substantial thermal mass of the cavity itself makes a load unnecessary.
A summary of the CO-emission rates measured in the Himmel and Dewerth (1974)
study are shown in Table 7-3. Yellow-tipping flame operating conditions resulted in higher
35 CO emissions for burners and ovens than the blue-flame conditions. Considerable variability
existed from burner to burner within and between gas ranges and among ovens, yet the
average emission rates for blue-flame operation among top burners, ovens, and burner pilot
lights were generally within a factor of four or five of each other. The authors noted that the
CO emission-rate distributions were skewed, leading them to average the emission data using
40 a log-normal transform and to present the average concentration and an interval representing
66% of the measurements. The infrared burner, pyrolytic self-cleaning oven, and oven pilot
light emission rates were considerably higher than the typical range top burner.
March 12, 1990 7-8 DRAFT - DO NOT QUOTE OR CITE
-------
TABLE 7-3. CARBON MONOXIDE-EMISSION RATES FROM 18 GAS RANGES,
GAS OVENS, AND GAS PILOT LIGHTS FOR BLUE FLAME AND
YELLOW-TIPPING FLAME BY THE DmECT-S AMPLING METHOD
Burner Type
Average and Range of Emissions'
Blue Flame
Yellow-Tipping
Flame
10
15
20
25
30
35
Top burners
Ovens and broilers
Top burners with thermostat
Top burners (142 kJ/min)
Top burners (190 kJ/min)
Infrared burners
Ovens and broilers with
catalytic clean
Pyrolytic self-clean oven
Pilot lights-burner
a. Free standing
b. Baffle around flame
c. Baffle around flame and
shield above flame
Pilot lights-oven
a. Constant horizontal
b. Constant horizontal
operates in two modes
22.6 (8.0-64.2)'
15.7 (6.3-38.8)
51.8 (11.9-228.8)
15.3 (6.6-35.4)
26.1 (9.4-72.0)
77.4
11.9 (4.9-29.1)
87.7
44.0 (30.4-63.8)
35.7
28.3
56.1 (39.6-69.7)b
248.3 (146.5-420.0)
322.2 (158.5-491.2)"
208.8 (135.8-281.8)"
156.6 (58.0-421.0)
62.0 (11.1-349.2)
53.5 (11.8-243.4)
40 "Values are the range of emission rates that contain two-thirds of the measured values.
"Values at the low and high measured level.
Source: Himmel and Dewerth (1974).
March 12, 1990
7-9 DRAFT - DO NOT QUOTE OR CITE
-------
Himmel and Dewerth (1974) noted that significant differences (p^O.05) for the range
averages for all four burners existed for three of the eighteen ranges tested. Emissions from
front burners were found to average 13% higher than back burners. Significant CO emission
differences (p>0.05) were noted between oven burners. Type of cooking pot (material), size
5 of cooking pot, physical properties of the individual pots (density, thermal conductivity, etc.)
did not have a pronounced impact on emissions. Burner cap design was found to influence
CO emissions.
Using the indirect measurement method and the standard water pot load (American
National Standards Institute, 1982), Tikalsky et al. (1987) reported the results of CO-
10 emission measurements made on gas ranges in 10 homes with each home having a different
range make, spanning a use age of from 7 to 30 years. The sample of homes was drawn
from a sample of 50 homes from which house NO2 measurements (Dames and Moore, 1986)
were available. This is the only study reported where field measurements were made on gas
cooking ranges actually in use. Five of the 10 homes were resampled after routine service
15 adjustments. Emissions were measured for top burners and ovens (without a cooking load) by
two different research groups using similar protocols but different air-sampling equipment.
Emissions measurements were made for two top burners and for bake and broil oven use on
each range. The top-burner emissions were measured for high, medium, and low settings.
Results of the Tikalsky et al. (1987) study showed CO-emission rates for the pre-
20 serviced top burners (across all burners, burner settings, and both measurement groups) to
range from 9.5 to 1746 /ig/kJ. CO emissions for baking ranged from 6.9 to 413 ftg/kJ,
whereas emissions for broiling ranged from 4 to 310 ^g/kJ. Low top-burner settings resulted
in significantly higher CO emissions than the medium or high settings (p>0.05). Comparison
of the emission-rate measurements between the two measurement teams indicated that one
25 team measured higher rates. CO emission rates showed a significant reduction after routine
service adjustments (p^O.05). The authors noted that the field measurements of CO emissions
in their study sometimes exceeded those previously measured by other investigators by a
factor of four, while oven emissions for both the bake and broil were within the range of
those reported by others.
30 A number of studies of CO emission rates from gas cooking ranges have been conducted
in which a limited number of samples (ranges, burners, and number of experiments) have
March 12, 1990 7-10 DRAFT - DO NOT QUOTE OR CITE
-------
been collected. These studies have utilized the direct and mass-balance methods. The results
of these studies for both top burners and ovens are shown in Table 7-4. As seen in the
previous studies the CO-emission rates for both the top burners and ovens showed
considerable variation with the yellow-tipping flame condition resulting in higher emissions
5 than the blue-flame condition. Borrazzo et al. (1987) measured CO-emission rates for the
pilot lights of the gas stove in their test house as well as for fugitive emissions from other
vented combustion appliances (gas dryer, water heater, furnace, etc.). A CO-emission rate
for a pilot light flow of 7.6 kJ/min was measured at 91 + 16 pig/U, whereas emissions from
other vented sources were negligible.
10 Natural gas is used for domestic water heating in approximately 55 million residences in
the United States. Cole and Zawacki (1985) summarized the available data on CO emissions
from domestic hot water heaters. They reported an average CO-emission rate for a total of
18 gas water heaters tested in three studies (Belles et al., 1979; Thrasher and DeWerth, 1977;
A.G.A.L., 1983) of 12.0 /*g/kJ. Thrasher and DeWerth (1977). In comparing emissions
15 from 13 water heaters for blue-flame and yellow-tipping flame conditions, they found that
yellow-tipping flame operation conditions of the heaters resulted in a fivefold increase in
emissions.
The available literature for gas appliances indicates that CO emissions are (1) highly
variable for range-top burners on a single range and between ranges and for ovens for blue-
20 flame conditions (properly adjusted), varying by as much as an order of magnitude or more;
(2) much higher for range top and oven burners operated under yellow-tipping flame
conditions (improperly adjusted) than for blue-flame conditions; (3) not different for rich or
lean fuels; (4) higher for top burners when they are operated under very low fuel consumption
rates; (5) comparable for top burners, ovens, pilot lights, and unvented gas dryers; and
25 (6) roughly comparable when obtained by either the direct or mass-balance method.
The data base on CO emissions from gas cooking ranges is largely based upon
laboratory studies in which a relatively few ranges were tested. The one field study in which
CO emissions were measured, for a small number of gas cooking ranges used in private
residences, indicates that for top burners, the laboratory data may underestimate actual CO
30 emission rates. More extensive field CO emission data for cooking ranges is needed to
determine how representative the laboratory derived data is.
March 12, 1990 7-11 DRAFT - DO NOT QUOTE OR CITE
-------
10
15
20
25
30
TABLE 7-4. CARBON MONOXIDE EMISSIONS FROM GAS RANGES
FOR STUDIES OF SMALL SAMPLE SIZE
Study
MIT/AGA
(1976)
Fortman et al.
(1984)
Borrazzo et al.
(1987)
Cote et al.
(1974)
Traynor et al.
(1982)
Goto&
Tammura
(1984)
Burner
1 top burner
4 top burners
2 ovens
4 top burners
1 oven-300 U/min
-150 kJ/min
-160 kJ/min
2 ranges/top burner
2 ranges/top burner
2 ovens
2 top burners
1 oven
1 top burner
Test"
Method
Db
D
D
D
D
D
D
C
C
C
C
C
C
Flame
Condition
Blue
Blue
Blue
Blue
Blue
Blue
Blue
Blue
Yellow
Blue
Blue
Blue
Blue
Number
of
Tests
1
11
4
16
2
6
2
2
2
2
5
2
1
Emission
Rate
(Mg/kJ)
11.6
110 ± 40
25.8 ± 4
98 ± 18
150 ± 18
33 + 1.3
33 ± 6.5
56
92.5
257
200 + 34
226 ± 17
86.9
35
*D = direct method, C = mass-balance chamber method.
bDid not use standard pot water load.
7.2.2 Emissions from Unvented Space Heaters
40 Unvented kerosene and gas space heaters are used in the colder climates to supplement
central heating systems or in more moderate climates as the primary source of heat. During
the heating season, space heaters generally will be used for a number of hours during the day
resulting in emissions over relatively long periods of time.
March 12, 1990
7-12 DRAFT - DO NOT QUOTE OR CITE
-------
Over the last several years there has been a dramatic increase in the use of unvented
kerosene space heaters in residential and commercial establishments primarily as a
supplemental heat source. The U.S. Consumer Product Safety Commission estimates that a
total of 16.1 million such heaters have been sold through 1986 (Womble, 1988). A
5 residential energy survey conducted by the U.S. Department of Housing and Urban
Development (1980) estimated that three million residences use unvented gas space heaters
(fueled by natural gas or propane), with their use more prevalent in the South Census region
of the United States. The large number of unvented space heaters sold in the United States
and the potential for their high use, particularly during periods when energy costs rise
10 quickly, make them an important source of CO indoors.
CO emissions from unvented kerosene and gas space heaters can vary considerably and
are a function of heater design (convective, radiant, combination, etc.), condition of heater,
and manor of operation (e.g., flame setting).
Unvented gas space heaters (UVGSHs) range in size from 7000 to 40,000 Btu/h and
15 vary in design and operation. Design characteristics of different heaters include the burners
(cast iron, steel, ceramic tile, catalytic surfaces, screen, etc.), ignition, heat exchanger, and
auxiliary equipment (i.e., oxygen depletion sensor). Operation characteristics include type of
fuel (natural or LPG), input modulation, primary air-shutter, flame type (blue, yellow-
tipping, infrared, etc.), and flame discharge temperature (blue flame or convective - 2800°F,
20 infrared - 1800°F, and catalytic - 1200°F). UVGSHs often are distinguished by their flame
discharge temperature and type of fuel used, when characterizing CO emissions. Both the
direct (hood over the heater from which gases are sampled) and the mass balance or chamber
methods have been used to evaluate emissions from invented gas space heaters. Many of the
studies were parametric in nature, seeking to evaluate the impact of some of the design and
25 operational features on emissions. Table 7-5 presents a general summary of the CO emission
data from UVGSHs that were obtained by operating the heaters in a well-adjusted and
typically full-input mode.
The summary data in Table 7-5 indicates that there is considerable variability of CO
emissions from UVGSHs. Infrared heaters produce higher emissions than the convective or
30 catalytic heaters. The mass-balance or chamber method seems to result in somewhat higher
emissions than the direct method. The Traynor et al. (1984, 1985) data indicate, that for the
March 12, 1990 7-13 DRAFT - DO NOT QUOTE OR CITE
-------
TABLE 7-5. CARBON MONOXIDE EMISSIONS FROM UNVENTED GAS SPACE HEATERS
VO
71
Study
Tray nor et al.
(1984, 1985)
Moschandreas
et al. (1985)
Thrasher and
DeWerth (1979)
Zawacki et al.
(1984)
Private11
Communication
Number
Type of of
Heater Heaters
Convective 9
3
12
Infrared 4
1
5
Convective 1
Infrared 1
Catalytic 1
Convective 2
3
Convective 1
Convective 1
Infrared 1
Number of
Tests/
Heaters
1
1
1
1
1
1
16
4
7
3
6
3
2
3
Fuel"
Type
NG
P
Both
NG
P
Both
NG
NG
NG
NG
NG
Testb
Method
C
C
C
C
C
C
D
C
D
C
D
C
D
D
D
U
U
Fuel Consumption
Rate (kJ/min)
177 - 784
353 - 660
277 - 368
258
186
186
260
260
207
207
131
381
263
211
ND
Flame
Blue
Blue
Infrared
Infrared
Blue
Blue
Infrared
Infrared
Blue
Blue
Blue
Blue
Blue
Type(s)c
of
Burner
SP, DP, R
SP, DP, R
CT
CT
R
R
SP
SP
CT, R
CT, R
DP
DP
R
RS
RS
Suspended
Radiating
Tiles
Yes
Yes
No
No
No
No
No
No
No
Yes
Yes
No
Yes
Emission Rate
(us/kJ)
Average
33
16
29
47
45
47
9.7
16.8
69
54.6
9.0
14.2
3.1
5.2
16.3
32.7
0.22
SD
26
1.9
2.5
1.6
1.5
0.4
1.7
0.9
8.2
0.9
2.6
"NG = natural gas, P = propane.
bC = chamber mass balance method, D = direct method, U = unknown.
CSP = slotted port burner, DP = drilled port burner, R = pressed metal ribbon burner, CT = ceramic tile, RS = retention screen.
dData obtained and reported by Moschandreas et al. (1985).
-------
subsample of convective heaters tested for the impact of fuel (natural gas vs. propane),
natural gas use results in higher emissions.
In a series of tests (mass-balance methods applied in a chamber and test house) on
subsamples of heaters (Traynor et al., 1984, 1986) CO emissions were found to (1) be lower
5 for partial input heater operation, (2) increase at lower chamber oxygen concentrations,
(3) be comparable between chamber and test house studies and show some decrease in time in
test house studies, and (4) be increased substantially for some maltuned heaters.
Moschandreas et al. (1985) found no difference in emissions for lean versus rich fuel.
Unvented kerosene space heaters fall into four basic design categories: convective,
10 radiant, two-stage, and wickless. The convective heaters operate at a relative high
combustion temperature and, depending upon burner design, can be a blue or white flame.
The radiant or infrared heaters utilize perforated ceramic or metal cylinders which become red
hot. The two-stage heaters (newer design kerosene heaters) are very similar to the radiant
heaters in design except they have a second combustion chamber above the first. The
15 wickless heaters have a chamber where the fuel and air are mixed and combustion occurs with
the resultant heat distributed via a fan. CO emissions from unvented kerosene heaters of the
convective, radiant, and two-stage design have been evaluated by both the direct and mass-
balance method for conditions of wick height (fuel consumption rate) for well-tuned heaters.
Few emission data are available for the wickless heaters.
20 A summary of CO-emission data for unvented kerosene space heaters is shown in
Table 7-6. There is considerable variability in emissions between and among heater types
and heater wick settings. Radiant heaters, when operated under normal wick settings,
produce considerably more CO than normal wick settings for the convective and two-stage
heaters. Two-stage heater emissions, however, jump considerably at lower wick settings.
25 Low-wick settings can increase CO emissions for convective heaters.
7.2.3 Emissions from Wood Stoves and Tobacco Combustion
Use of wood-burning stoves has been a popular cost savings alternative to conventional
central heating systems using gas or oil. CO and other combustion by-products enter the
30 indoor environment during fire start-up, fire-tending functions, or through leaks in the stove
March 12, 1990 7-15 DRAFT - DO NOT QUOTE OR CITE
-------
TABLE 7-6. CARBON MONOXIDE EMISSIONS FROM UNVENTED KEROSENE SPACE HEATERS
o
^^
^3
»— *
^^5
^CJ
^
H-L
o
5
5
i
o
o
^-\
3
*§
g
3
o
5*
O
HH
a
Study
Leaderer (1982)
Tray nor et al.
(1983)
Jones et al.
(1983)c
Moschandreas et al.
(1985)
Number
Type of of
Heater Heaters
C 1
R 1
C 2
R 3
2S 2
C
R
C 1
R 1
Test"
Method
C
C
C
C
C
D
D
C
C
Fuel Consumption
Rate (kJ/min)
37.3
97.9
158.0
84.4
113
144
130
193
113
148
132
182
202
168d
100
129
Number
of
Tests
3
3
3
3
3
3
4
4
Emission Rate f^g/kJ)
Average
25.8
22.3
10.1
72.9
58.2
42.6
60
12
173
68
54
9
4.7
27.5
35.3
64.1
SD
4.7
1.5
4.1
2.6
5.0
2.5
1.1
1.1
1.4
1.5
2.5
1.2
8.2
5.2
"C = convective, R = radiant, 2S = two-stage.
bC = chamber/mass balance
°As reported by Michandreas
dManufacturer's rating.
method, D = direct/load method.
et al. (1985).
-------
or venting system. Hence it is difficult to evaluate indoor CO-emission rates for wood-
burning stoves. Traynor et al. (1987) evaluated indoor CO levels from four wood-burning
stoves (three airtight and one non-airtight stove) in a residence. The non-airtight stove
emitted substantial amounts of CO to the residence, particularly when operated with a large
5 fire. The airtight stoves contributed considerably less. The average CO source strengths
during stove operation reported for the airtight stoves ranged from 10 to 140 cm3/h, whereas
levels for the non-airtight stove source strengths ranged from 220 to 1800 cmVh.
In 1980, 32% of the U.S. adult population was reported to be smokers (U.S. Depart-
ment of Commerce, 1984). The combustion of tobacco represents an important source of
10 indoor air contaminants. CO is emitted indoors from tobacco combustion through the exhaled
mainstream smoke (MS) and from the smoldering end of the cigarette (sidestream smoke -
SS). MS and SS CO-emission rates have been evaluated extensively in small chambers (less
than a liter in volume) using a standardized smoking machine protocol. The results of these
studies have been summarized and evaluated in the Surgeon General's reports (e.g., 1986)
15 and the National Academy of Science Report on environmental tobacco smoke (1986). These
results indicate considerable variability in total (MS+SS) CO emissions, with a typical range
of from 40 to 67 mg/cig. A small chamber study of 15 brands of Canadian cigarettes
(Rickert et al., 1984) found the average CO-emission rate (MS+SS) to be 65 mg/cig. A
more limited number of studies have been done using large chambers with the occupants
20 smoking or using smoking machines. Girman et al. (1982) reported a CO-emission rate of
94.6 mg/cig for a large chamber study in which one cigarette was evaluated. A CO-emission
factor of 88.3 mg/cig was reported by Moschandreas et al. (1985) for a large chamber study
of one reference cigarette.
25 7.2.4 Summary of Emission Data
Indoor sources of CO can be considered as unintentional or accidental (leaky flue pipes,
backdrafting, etc.) and intentional (emissions from unvented combustion sources). Emissions
from unintentional sources can result in indoor concentrations associated with serious acute
health effects and result in several hundred deaths per year in the U.S. CO emissions from
30 these unintentional sources, despite their importance, cannot be characterized in any
standardized way. Unvented or partially vented gas cooking ranges and ovens, gas
March 12, 1990 7-17 DRAFT - DO NOT QUOTE OR CITE
-------
appliances, unvented gas space heaters, unvented kerosene space heaters, cigarette
combustion, and wood-burning stoves are all notable "intentional" indoor sources of CO
emissions.
Unvented or partially vented sources of CO have been evaluated for CO emissions by
5 either the direct method or the mass-balance approach. The direct method samples the
emitted combustion gases as they pass through a sampling hood above the source. The mass-
balance approach measures the changes in CO over time in an environmental chamber or test
house in relation to changes in outdoor CO concentration, source emission rates, and CO
removal rates. For gas range top burners, emissions typically are evaluated using a
10 standardized water load in a cooking pot.
Emissions from unvented, partially vented, or improperly vented gas cooking ranges and
ovens and gas appliances represent an important source of CO in the residential environment
in the United States due to the high percentage of homes (approximately 45%) using gas to
cook. CO emissions from these sources are a function of a number of variables relating to
15 the source type (range top or oven, burner design, pilot light, etc.), source condition (age,
maintenance, etc.), source use (number of burners used, fuel consumption rate, etc.) and use
of outside vents. The source emission studies typically have been conducted in the laboratory
setting and involved relatively few gas ranges and gas appliances. The reported studies
indicate that CO emissions are highly variable among burners on a single gas cooking range
20 and between gas cooking ranges and ovens, varying by as much as an order of magnitude.
Operating a gas cooking range or oven under improperly adjusted flame conditions (yellow-
tipping) can result in greater than a fivefold increase in emissions when compared to properly
operating flame conditions (blue flame). Use of a rich or lean fuel appeared to have little
effect on CO emissions. In general, CO emissions were roughly, on an average, comparable
25 for top burners, ovens, pilot lights, and unvented gas dryers when corrected for fuel
consumption rate. The emissions rates gathered by either the direct or mass-balance method
were comparable. Only one study attempted to evaluate gas stove emissions in the field for a
small number (10) of residences. This study found CO emissions to be as much as a factor of
four higher than chamber studies. Given the prevalence of the source, limited field
30 measurements and poor agreement between existing laboratory and field derived CO emission
March 12, 1990 7-18 DRAFT - DO NOT QUOTE OR CITE
-------
data, there is a need to establish a better CO emission data base for gas cooking ranges in
residential settings.
CO emissions from unvented gas space heaters were found to be variable from heater to
heater but roughly comparable to those for gas cooking ranges. Infrared gas space heaters
5 produced higher emissions than the convective or catalytic heaters. Emissions of CO for
these heaters were higher for maltuned heaters and for the mass-balance versus direct method
of testing. No differences for rich or lean fuel were found, but use of natural gas resulted in
higher emissions than propane. Lower fuel consumption settings resulted in lower CO
emissions. Emissions were observed to vary in time during a heater run and increase when
10 room or chamber oxygen levels decreased.
Among the three principal unvented kerosene space heater designs (radiant, convective
and two-stage burners), radiant heaters produced the highest CO emissions with the
convective heaters producing the lowest emissions. Wick setting (low, normal, or high) had a
major impact on emissions with the low-wick setting resulting in the highest CO emissions.
15 Data from different laboratories are in good agreement for this source.
CO emissions into indoor spaces from wood-burning stoves occur during fire start-up,
fire-tending, or through leaks in the stove or venting system. Few data are available
characterizing CO emissions for normal wood stove or fireplace operation. The available data
indicate that the non-airtight, wood-burning stoves can contribute substantial amounts of CO
20 directly to the indoor environment while the airtight stoves contribute little or none.
Tobacco combustion represents an important indoor source of CO based upon the
number of cigarettes smoked. In comparison to other unvented combustion sources, CO
emissions into indoor spaces from tobacco combustion are relatively low and show little
variation from brand to brand.
25 The available data on CO emissions from unvented combustion sources are based largely
upon chamber or test house studies using the mass-balance or direct-measurement method for
a small sample of sources (i.e., a small number of gas cooking ranges). Given the high
variability of CO emissions observed from these sources in the available studies, additional
data are needed to better understand the factors impacting those emissions. Little or nothing
30 is known about CO emissions from unvented combustion sources actually in use in residences.
March 12, 1990 7-19 DRAFT - DO NOT QUOTE OR CITE
-------
The few data available indicate that CO emissions from sources in the field are considerably
more variable and typically higher than those observed in the chamber or test house studies.
5 7.3 CONCENTRATIONS IN INDOOR ENVIRONMENTS
CO concentrations in an enclosed environment are affected by a number of factors in
addition to the source factors discussed in the previous sections. These factors include
outdoor concentrations, proximity to outdoor sources (i.e., parking garages or traffic),
volume of the space, and mixing within and between indoor spaces.
10 Outdoor CO concentrations have been measured in a number of locations across the
United States utilizing continuous CO monitoring based upon NDIR spectroscopic detection.
The NDIR instruments, however, are too bulky and complicated for either indoor or personal
monitoring. Over the past decade small, lightweight, and portable CO monitors have been
developed. These monitors are based primarily on electrochemical detection (see Chapter 5).
15 These highly versatile CO monitors, when equipped with internal or external data loggers,
have permitted the measurement of personal exposures to CO as well as CO concentrations in
a number of indoor environments.
CO measurements in enclosed spaces have been made either in support of total personal
exposure studies or in targeted indoor studies. In the personal exposure studies, individuals
20 wear the monitors in the course of their daily activities, taking them through a number of
different microenvironments. In targeted studies CO measurements are taken in indoor spaces
independent of the activities of occupants of those spaces.
7.3.1 Indoor Concentrations Recorded in Personal Exposure Studies
25 Three studies have reported CO concentrations in various microenvironments as part of
an effort to measure total human exposure to CO and to assess the accuracy of exposure
estimates calculated from fixed-site monitoring data. In each study, subjects wore personal
CO exposure monitors for one or more 24-h periods. CO concentrations were recorded on
data loggers at varying time intervals as a function of time spent in various
30 microenvironments. A activity diary was kept by participants where they were asked to
provide information such as time, activity (i.e., cooking), location (microenvironment type),
March 12, 1990 7-20 DRAFT - DO NOT QUOTE OR CITE
-------
presence and use of sources (smokers present, gas stove, etc.), etc. CO concentrations by
microenvironment were extracted from the measured concentrations by use of the activity
diaries. This section will discuss the results of those studies as they relate to the
concentrations measured in different microenvironments. A discussion of the results as they
5 relate to total exposure to CO are discussed in Chapter 8.
Two of the studies, conducted in Denver, CO, and Washington, DC, by EPA (Akland
et al., 1985; Whitmore et al., 1984; Hartwell et al., 1984; Johnson, 1984), measured the
frequency distribution of CO exposure in a representative sample of the urban population.
The study populations were selected using a multistage sampling strategy. The third study,
10 also conducted in Washington, DC, (Nagda and Koontz, 1985), utilized a convenience
sample.
The first-mentioned Washington study obtained a total of 814 person-day samples for
1161 participants while the Denver study obtained 899 person-day samples for
485 participants. The Denver study obtained consecutive 24-h samples for each participant
15 while the Washington study obtained one 24-h sample for each participant. Both studies were
conducted during the winter of 1982-1983.
A comparison of CO concentrations measured in the Washington and Denver studies is
shown in Table 7-7 (from Akland et al., 1985). Concentrations measured in all
microenvironments for the Denver study were higher than those for the Washington study.
20 This is consistent with the finding that daily maximum one- and eight-hour CO concentrations
at outdoor fixed monitoring sites were about a factor of two higher in the Denver area than
the Washington area during the course of the studies (Akland et al., 1985). The highest
concentrations in both studies were associated with commuting while the lowest levels were
measured in indoor environments. Concentrations associated with commuting are no doubt
25 higher due to the proximity to and density of outside CO sources (cars, buses, and trucks)
particularly during commuting hours when traffic is heaviest. Indoor levels, especially
residential levels in the absence of indoor sources, are lower primarily due to the time of day
of sampling (noncommuting hours with lower outdoor levels). A more detailed breakdown of
CO concentrations by microenvironments for the Denver study is shown in Table 7-8
30 (Johnson, 1984). Microenvironments associated with motor vehicles result in the highest
March 12, 1990 7-21 DRAFT - DO NOT QUOTE OR CITE
-------
TABLE 7-7. SUMMARY OF CO EXPOSURE LEVELS AND TIME SPENT PER DAY
KT
i— '
JO
^O
&
o
IN SELECTED MICROENVIRONMENTS
Location
Concentration
§
g
1
i
O
o
55
O
^^
1
0
Microenvironment
Indoors, parking garage
In transit, car
In transit, other (bus, truck, etc.)
Outdoors, near roadway
In transit, walking
Indoors, restaurant
Indoors, office
Indoors, store/shopping mall
Indoors, residence
Indoor, total
"n = number of person-days with nonzero
n
31
643
107
188
171
205
283
243
776
776
durations, x =
X
18.8
8.0
7.9
3.9
4.2
4.2
3.0
3.0
1.7
2.1
mean, SE
Denver. CO
,' ppm
SE
4.96
0.32
0.61
0.36
0.45
0.29
0.20
0.22
0.10
0.09
= standard error.
Median
time, min
14
71
66
33
28
58
478
50
975
1,243
Washington. DC
Concentration,' ppm
n
59
592
130
164
226
170
349
225
705
705
X
10.4
5.0
3.6
2.6
2.4
2.1
1.9
2.5
1.2
1.4
SE
4.43
0.14
0.30
0.20
0.29
0.32
0.27
0.49
0.10
0.08
Median
time, min
11
79
49
20
32
45
428
36
1,048
1,332
^ Source: Akland et al. (1985).
-------
10
15
20
25
30
TABLE 7-8. INDOOR MICROENVIRONMENTS LISTED IN DESCENDING ORDER
OF WEIGHTED MEAN CO CONCENTRATION
Category
Public garage
Service station or motor vehicle
repair facility
Other location
Other repair shop
Shopping mall
Residential garage
Restaurant
Office
Auditorium, sports arena,
concert hall, etc.
Store
Health care facility
Other public buildings
Manufacturing facility
Residence
School
Church
Number
of Subjects
116
125
427
55
58
66
524
2,287
100
734
351
115
42
21,543
426
179
CO Concentration
Mean
13.46
9.17
7.40
5.64
4.90
4.35
3.71
3.59
3.37
3.23
2.22
2.15
2.04
2.04
1.64
1.56
T ppm
SD
18.14
9.33
17.97
7.67
6.50
7.06
4.35
4.18
4.76
5.56
4.25
3.26
2.55
4.06
2.76
3.35
Source: Johnson et al. (1984).
concentrations, with concentrations reaching or exceeding the 9 ppm reference level
(NAAQS).
35 No statistical difference (p>0.05) in CO concentrations were found for residences with
and without gas ranges in the Washington study. The results of a similar analysis on the
Denver data, according to the presence or absence of selected indoor sources, is shown in
Table 7-9 (Johnson, 1984).
Attached garages, use of gas ranges, and presence of smokers were all shown to result
40 in higher indoor CO concentrations. Concentrations were well below the 9 ppm reference
level (NAAQS), but substantially above concentrations in residences without the sources.
March 12, 1990
7-23 DRAFT - DO NOT QUOTE OR CITE
-------
10
15
20
TABLE 7-9. WEIGHTED MEANS OF RESIDENTIAL EXPOSURE GROUPED
ACCORDING TO THE PRESENCE OR ABSENCE OF SELECTED INDOOR
CARBON MONOXIDE SOURCES
CO Source
Attached garage
Operating gas
stove
Smokers
Carbon
Source
Mean
2.29
4.52
3.48
Monoxide Concentration, oom
Present
SD
5.34
6.10
6.58
Source
Mean
1.88
1.93
1.89
Absent
SD
3.00
3.92
3.69
Difference
in Means
0.41
2.59
1.59
Significance
Level of
ttesf
p< 0.0005
p< 0.0005
p< 0.0005
"Student t test was performed on logarithms of PEM values.
Source: Johnson (1984).
25
30
35
In the second Washington study (Nagda and Koontz, 1985), a total of 197 person-days
of samples were collected from 58 subjects, representing three population subgroups
(housewives, office workers, and construction workers). A comparison of residential CO
concentrations from that study as a function of combustion sources and whether smoking was
reported is shown in Table 7-10. Use of gas ranges and kerosene space heaters were found to
result in higher indoor CO concentrations. The statistical significance of the differences was
not given. Concentrations were highest in microenvironments associated with commuting.
The data collected in this study were consistent with the data collected in the EPA
Washington study discussed above.
It is difficult for all three studies to assess the contribution to indoor CO concentrations
from either outdoor or indoor sources because concentrations outside each indoor
microenvironment were not measured.
40
March 12, 1990
7-24 DRAFT - DO NOT QUOTE OR CITE
-------
10
TABLE 7-10. AVERAGE RESIDENTIAL CO EXPOSURES (ppm):
IMPACT OF COMBUSTION APPLIANCE USE AND TOBACCO SMOKING1
Reported Tobacco Smoking
Appliances
None
Gas stove
Kerosene space heater
Wood burning
Multiple appliances
All cases
No
1.2 (66)
2.2 (15)
5.1 (3)
0.7 (2)
1.0 (1)
1.5 (87)
Yes
1.5 (12)
1.3 (1)
NDb
NDb
NDb
1.5 (13)
All
Cases
1.2
2.2
5.1
0.7
1.0
1.5
(78)
(16)
(3)
(2)
(1)
(100)
15
"Percentage of subjects' time in their own residences indicated in parentheses for each category of appliance use
and tobacco smoking.
20 bNo data available.
Source: Nagda and Koontz (1985).
25
7.3.2 Targeted Microenvironmental Studies
As demonstrated from the personal exposure studies discussed above, individuals, in the
course of their daily activities, can encounter a wide range of CO concentrations as a function
30 of the microenvironments in which they spend time. A number of studies have been
conducted over the last decade to investigate concentrations of CO in indoor
microenvironments. These "targeted" studies have either focused on indoor CO
concentrations as a function of the microenvironment or sources in specific
microenvironments.
35
7.3.2.1 Indoor Microenvironmental Concentrations
A summary of the results of the larger studies that have investigated CO levels in
various indoor environments, independent of the existence of specific indoor sources, is
shown in Table 7-11. Major foci of these studies are microenvironments associated with
40 commuting. A wide range of concentrations were recorded in these studies with the highest
March 12, 1990 7-25 DRAFT - DO NOT QUOTE OR CITE
-------
TABLE 7-11. CARBON MONOXIDE CONCENTRATIONS' MEASURED IN VARIOUS
INDOOR ENVIRONMENTS AS A FUNCTION OF MICROENVIRONMENTS
g
ON
Study
Cortege and
Spengler
(1976)
Spengler
et al. (1978)
Colwill and
Hickman
(1980)
Ziskind et al.
(1981)
Wallace (1983)
Holland (1983)
Locations Microenvironment
Boston, MA Autos
Transit
Split
All
Outside*
Boston, MA Seven skating
rinks
London, Autos
England Outside0
Denver, CO Buses
and Taxis
Boston, MA Police cars
One office
Stamford, CT Commercial
Commuting
Residential
Los Angeles, Commercial
CA Commuting
Residential
Average
Time-
Frame of
Sampling
(min)
40-70
40-70
40-70
40-70
40-70
40-160
65-90
-540
-540
-540
Hourly
10-30
10-30
10-30
10-30
10-30
10-30
Number of
Observations
248
28
70
346
1,076
17
11
75
38
19
80
659
1,341
577
1,938
96
807
Inside
Mean
1.34
7.4
8.3
11.9
53.6
25.2
4-36d
lO-lT*
0-46d
19.0
5.8
6.2
2.9
3.3
16.1
7.6
SD
5.4
3.7
2.8
5.5
18.0
7.0
5.9
8.0
4.7
3.9
2.5
5.8
5.0
CO - ppm
Max
>35
192
40
84
48
59
50
61
38
39
61
42
38
Outside Source
Mean SD Identified
Traffic
Traffic
Traffic
Traffic
6.0 4.0 Ambient
fee
Traffic
47.0 13.1
Traffic
Traffic
Traffic
Leakage
from
garage
4.2" 3.0
5.5e 4.1 Traffic
4.3° 3.1
4.0 3.1
5.8 4.2 Traffic
3.9 2.8
Comments
66 volunteers used-some
levels (4%) related to
faulty exhaust
Ventilation measures-
cleaning from CO decay
1 1 driven over a 35 km
route
Only data gathered by
electrochemical
monitors presented.
Passive dosimeter
data not included;
58% of values for
rides >8 h were
>9ppm
65 employees affected;
corrective action taken
-------
£
8*
TABLE 7-11 (cont'd). CARBON MONOXIDE CONCENTRATIONS- MEASURED IN VARIOUS
INDOOR ENVIRONMENTS AS A FUNCTION OF MICROENVIRONMENTS
i
^
to
"^
V
^
3
I
o
*-/
o
H
O
Q
w
0
»
n
Study
Flachsbart
al. (1987)
Yocom
et al. (1987)
Peterson and
Sabersky (1975)
Chancy
(1978)
Ziskind
et. a. (1982)
Amendola and
Hanes (1984)
Flachsbart and
Ott (1984)
Locations
Phoenix, AZ
Denver, CO
DC
Hartford
CT
Los Angeles
CA
Several
U.S. Cities
Los Angeles
CA
New England
5 California
cities
Average
Time-
Frame of
Sampling
Microenvironment (min)
Commercial 10-30
Commuting 10-30
Residential 10-30
Commercial 10-30
Commuting 10-30
Residential 10-30
Autos 34-69
Bus 82-115
Rail 27-48
2 Garages 3
Public Building
Office Building
Private Home
Autos 3
Autos
Home
Work
Commute
Service Station 480
Car dealership
Enclosed parking 2-5
Bldg. attached to 2-5
enclosed parking
Commercial settings 2-5
Number of
Inside CO - ppm
Observations Mean SD Max
380
839
48
1,949
3,634
528
213
35
8
47
-
564
557
461
81
10
7
202
2.2 2.2 17
6.8 4.9 50
5.8 3.6 17
5.9 4.3 30
11.0 7.7 54
5.6 4.4 45
9 i_22 3 2-9
3.7-10.2 1-7
2.2-5.2 0.5-5
21-94 10-56
1.8-22.7 -
2.1-22.9
1.8-21.9
<2.5 - 45
2-50
4-4.6
2.2-4.3
6.7-10.0 -
2.2-110.8 - 110.8
27.7 12.5
6.1 2.9
2.1 1.6
Outside Source
Mean SD Identified
2.8 2.5
3.9 3.3 Traffic
2.4 2.1
5.0 3.3
5.8 3.7 Traffic
3.1 23
Traffic
— — Traffic
- - Traffic
- - Traffic
Very similar Traffic
to indoor
concentrations
Similar to Traffic
auto levels
Traffic
Autos
3.0-2.6 Autos
Autos
Autos
Comments
Measurements made et
during commuting hours
Two week avenges day
and night over a summer,
fall, and winter period
Slower the traffic
the higher the CO
Higher in Winter
than Summer
Indoor Values have
outdoor concentrations
subtracted
-------
I
I—«
N>
\»
SO
00
H
I
o
o
TABLE 7-11 (cont'd). CARBON MONOXIDE CONCENTRATIONS' MEASURED IN VARIOUS
INDOOR ENVIRONMENTS AS A FUNCTION OF MICROENVIRONMENTS
Study
Sisoric and
Fugas (1985)
Average
Time-
Frame of
* Sampling Number of Inside CO - ppm Outside Source
Locations Microenvironment (min) Observations Mean SD Max Mean SD Identified Comments
Zagreb 8 institutions Winter - 1.1-6.0 0.6-13.7 - - Traffic
Yugoslavia and
summer
periods
"All measurements made with electrochemical devices.
bFixed central station sites.
'Measurement made outside auto.
^9S% confidence limits.
"Average of two fixed sites.
8
n
-------
CO concentrations found in the indoor commuting microenvironments. These concentrations
frequently are higher than concentrations recorded at fixed-site monitors but lower than
concentrations measured immediately outside the vehicles. Concentrations generally are
higher in automobiles than in public transportation microenvironments. A number of the
5 studies noted that CO concentrations in commuting vehicles can exceed both the eight-hour,
9-ppm level and the one-hour, 35-ppm level specified in the NAAQS (1970). Flachsbart
et al. (1987) noted that the most important factors influencing CO concentrations inside
automobiles were such factors as link-to-link variability (a proxy for traffic density, vehicle
mix, and roadway setting), day-to-day variability (proxy for variations in meteorological
10 factors and ambient CO concentrations), and time of day. This study noted that with
increased automobile speed, interior CO concentrations decreased.
Service stations, car dealerships, parking garages and office space that have attached
garages can exhibit high concentrations of CO due to automobile exhaust. In one case
(Wallace, 1983), corrective measures reduced office space CO concentrations originating from
15 an attached parking garage from 19 ppm to approximately 4 ppm. In an investigation of
seven ice skating rinks in the Boston area, one study (Spengler et al., 1978) reported
exceptionally high average CO concentrations (53.6 ppm) with a high reading of 192 ppm.
Ice-cleaning machines and poor ventilation were found to be responsible.
Residential and commercial levels generally were found to have low concentrations, but
20 no information was provided on the presence of indoor sources or outdoor levels.
7.3.2.2 Concentrations Associated with Indoor Sources
As noted earlier, the major indoor sources of CO in residences are gas ranges and
unvented kerosene and gas space heaters, with properly operating wood-burning stoves and
25 fireplaces (non-leaky venting system) and tobacco combustion of secondary importance.
Properly used gas ranges (ranges used for cooking and not space heating) are used
intermittently and thus would contribute to short-term peak CO levels indoors but likely
would not result in substantial increases in longer-term average concentrations. Unvented
kerosene and gas space heaters typically are used for several hours at a time and thus are
30 likely to result in sustained higher levels of CO. The improper operation of gas ranges or
unvented gas or kerosene space heaters (e.g., low-wick setting for kerosene heaters or yellow-
March 12, 1990 7-29 DRAFT - DO NOT QUOTE OR CITE
-------
tipping operation of gas ranges) could result in substantial increases in indoor CO levels. CO
levels indoors associated with tobacco combustion are, based upon source emission data,
expected to be low unless there is a very high smoking density and low ventilation. In the
absence of a leaky flue or leaky fire box, indoor CO levels from fire places or stoves should
5 be low with short peaks associated with charging the fire when some back draft might occur.
The majority of studies investigating CO concentrations in residences, as a function of
the presence or absence of a known CO source, typically have measured CO concentrations
associated with the source's use over short time periods (on the order of a few minutes to a
few hours). These studies typically have involved fewer than 10 residences and have reported
10 peak CO levels (on the order of minutes). Only two studies (Hartwell et al., 1988; Koontz
and Nagda, 1987) have reported long-term average CO concentrations (over several hours) as
a function of the presence of a CO source for large residential sample sizes, while one study
(McCarthy et al., 1987) reported longer term average indoor CO concentrations for a small
sample.
15
Average Indoor Source Related Concentrations
As part of a study to determine the impact of combustion sources on indoor air quality,
a sample of 382 homes in New York State (172 in Onondaga County and 174 in Suffolk
County) were monitored for CO concentrations during the winter of 1986 (Hartwell et al.,
20 1988). In this study four combustion sources were examined: gas cooking appliances,
unvented kerosene space heaters, wood-burning stoves and fireplaces, and tobacco products.
A factorial sample that included all sixteen combinations of combustion sources was utilized.
CO concentrations were monitored in the main living area (e.g., family room) and source
area (e.g., in the kitchen for homes with gas ranges) for each home over a three-day period
25 using an electrochemical monitor with the data stored on a data logger. Outdoor CO levels
were not recorded for these homes. CO concentrations were reported as averages for the full
three-day period of measurement.
Average CO concentrations measured in the main living area as a function of county and
the presence or absence of a combustion source are shown in Table 7-12. Gas ranges and
30 kerosene heaters were found to result in small increases in average CO levels. Use of a
wood-burning stove or fireplace resulted in lower average CO levels, presumably due to
March 12, 1990 7-30 DRAFT - DO NOT QUOTE OR CITE
-------
10
15
20
25
30
35
40
TABLE 7-12. WEIGHTED SUMMARY STATISTICS FOR CO
CONCENTRATIONS (ppm) IN THE MAIN LIVING AREA BY USE FOR
SELECTED SOURCES BY COUNTY
Source
and
County
Source
Present
Sample Percent
Size Detected
Arith.
Mean
SE
Geo.
Mean
Geo.
SE
KEROSENE HEATER
Onondaga
Suffolk
Yes
No
Yes
No
10
198
16
158
89.3*
60.0
100.0'
72.1
3.33
1.72
3.86*
2.03
1.34
0.15
0.73
0.15
2.20
1.29
3.35*
1.62
1.33
1.06
1.22
1.07
WOOD-BURNING STOVE/FIREPLACE
Onondaga
Suffolk
GAS STOVE
Onondaga
Suffolk
Yes
No
Yes
No
Yes
No
Yes
No
39
169
33
141
90
118
86
88
44.5
62.9
82.7
72.7
77.4*
47.0
82.8
68.2
1.04
1.86*
1.93
2.24
2.29*
1.33
2.55*
1.91
0.09
0.16
0.23
0.17
0.24
0.17
0.21
0.19
0.93
1.37*
1.72
1.72
1.74*
1.04
2.04*
1.51
1.09
1.06
1.14
1.07
1.08
1.07
1.10
1.09
45
"Significantly different at .05 level.
Source: Hartwell et al. (1988).
increased air-exchange rates associated with use. The study found no effect on average CO
levels with tobacco combustion and no difference by location in the residence. The data base
has not yet been analyzed for differences in short-term CO concentrations (eight-hour, one-
hour, or less than one-hour concentrations) as a function of sources and source use. When
March 12, 1990
7-31 DRAFT - DO NOT QUOTE OR CITE
-------
such an analysis is made available the impact of the combustion sources on residential CO
levels will be much more pronounced.
Koontz and Nagda (1987 and 1988) utilizing Census data for sample selection,
monitored 157 homes in 16 neighborhoods in North Central Texas over a nine-week period
5 between January and March 1985. Unvented gas space heaters were used as the primary
means of heating in 82 residences (13 had one UVGSH, 36 had two UVGSHs, and 33 had
three or more UVGSHs) and as a secondary heat source in 29 (17 had one UVGSH and 12
had two or more UVGSHs). There was no gas space heater present or used in 41 of the
homes and 5 of the homes were not included for various reasons (e.g., air sample lost). A
10 majority of all the homes in each UVGSH use category had gas ranges and gas water heaters
(typically greater than 80%). Air samples were collected for all residences on two separate
occasions over integrating periods of approximately 15 h using Collectaire samplers. CO
concentrations in each sampler then were measured using a electrochemical monitor. In 30%
of the residences (46 residences) CO was monitored continuously, consisting of sequential 15-
15 min averages over an average monitoring period of about five days with an electrochemical
monitor. Measurements were made close to the geometric center of the house.
The cumulative frequency distributions for the first integrated CO measurements by
source category are shown in Figure 7-1. Residences where UVGSHs are the primary heat
source exhibited the highest CO concentrations. CO concentrations were greater than or
20 equal to 9 ppm in 12% of the homes with the highest concentration measured at 36.6 ppm.
No values were measured above 9 ppm for residences where a UVGSH was not used at all or
was used as a secondary heat source. The second CO sample produced summary statistics
virtually identical to the first. Table 7-13 presents a comparison of the CO concentrations
measured in the continuously monitored residences (15-min average concentrations summed
25 by one- and eight-hour time periods) with the one-hour, 35 ppm and eight-hour, 9 ppm CO
standard by source category. The table also presents the mean concentrations measured in
these home over the full five-day periods. Five of the residences exceeded the one-hour,
35 ppm level while seven of the residences exceeded the eight-hour, 9 ppm level. Higher CO
levels were associated with maltuned unvented gas appliances and the use of multiple
30 unvented gas appliances.
March 12, 1990 7-32 DRAFT - DO NOT QUOTE OR CITE
-------
0
o
z
o
a.
100-
90-
80-
70-
60-
50-
40-
30-
20-
10-
0
Meant Standard Deviation:
O Non-UVQ8H—2.211.7
O Secondary UVQSH—2.9±1.6
A Primary UVQSH—5.5± 8.0
T-
8
i
10
CO.ppm
—r-
12
—r—
14
—r~
16
—T~
18
20
Figure 7-1. Cumulative frequency distributions and summary statistics for indoor CO
concentrations in three groups of monitored homes.
Source: Koontz and Nagda (1988).
March 12, 1990
7-33 DRAFT - DO NOT QUOTE OR CITE
-------
TABLE 7-13. SUMMARY OF CONTINUOUS CO MONITORING RESULTS
BY HEATING EQUIPMENT
5
10
15
Heating
Equipment
Primary UVGSH
Secondary UVGSH
Non-UVGSH
Number
of
Homes
26
11
9
Number
1-h, 34 ppm
4
1
0
Exceeding
8-h, 9 ppm
5
0
2
CO
Concentration.
Mean
6.2
2.3
2.2
ppm
SD
7.6
1.1
1.2
Source: Koontz and Nagda (1988).
20
In a study of 14 homes with one or more unvented gas space heaters (primary source of
heat) in the Atlanta, GA, area, McCarthy et al. (1987) measured CO levels by continuous
NDIR monitors in two locations in the homes (room with the heater and a remote room in the
house) and outdoors. Measurements were taken over five-minute periods in turn from each of
25 the three sampling points for each house over 96-h sampling periods. The authors reported
only the summary statistics for CO (average 96-h concentrations) based on the continuously
collected data in the room with the heater and outdoors. One out of the 14 UVGSH homes
exceeded 9 ppm during the sampling period. Mean indoor values ranged 0.26 ppm to
9.49 ppm and varied as a function of the use pattern of the heater. Only one of the homes
30 used more than one heater during the air sampling. Outdoor concentrations varied from
0.3 ppm to 1.6 ppm.
Peak Indoor Source-Related Concentrations
Short-term or peak CO concentrations indoors associated with specific sources were
35 obtained for a few field studies. The peak CO concentrations measured in these studies, by
location in the house and presence of specific sources, are shown in Table 7-14. A wide
range of peak CO concentrations were observed in these studies between and among
residences with different indoor CO sources. The highest concentrations measured
March 12, 1990 7-34 DRAFT - DO NOT QUOTE OR CITE
-------
sr
to
H- »
&
u
§>
H
I
O
O
§
H
O
C
o
a
g
1AJ
Reference
Hartwell
etal.
(1988)
Koontz
etal.
(1987)
Leaderer
etal.
(1984)
BLJ1 /-14. FilAJk Ut
Indoor
Source"
GR
K
UVGSHP, GR
UVGSHS, GR
GR
GR
CK
RK
) UUINUtUNlK
Number of
Residents
12
1
26
11
9
1
8
5
A11UINS JSI
Location1"
K
LR, D, B
K
D
C
C
C
A
B
LR
B
LR
B
JUNUUUKJS
Averaging
Time
(min)
30
30
30
30
15
15
15
15
15
5
5
5
5
iUUKUtl JVUVASUKJtJJ JLN 1*11
CO Concentration (ppm)
Peaks Outside
1.8 -> 100 0.7-10
1.8- 17
5.7 5.0
9.7 5.0
?-69
?-69
?-26
3.5
3.0
0-3.2
0-3.4
2.1 -21.1
4.8-8.2
1LJJ MULUca
Comments
Excluding one house with
> 100 range in kitchen is
1.8 - 15, wood stoves
and smokers were present
in same houses but no
effect was seen
Houses may contain more
than one heater
Outdoor levels subtracted
Outdoor levels subtracted
Outdoor levels subtracted
-------
TABLE 7-14 (cont'd). PEAK CO CONCENTRATIONS BY INDOOR SOURCE MEASURED IN FIELD STUDIES
»
o
o
1
o
§
w
g
Indoor
Reference Source"
Lebret GA
etal.
(1987)
Brunekreef GA
etal.
(1982)
Moschandreas GR
and
Zabransky
(1982)
Sterling and GR
Sterling
(1979)
Number of
Residents Location1"
12 K
LR
B
254 K
8 K
LR
B
9 K
Averaging
Time CO Concentration (ppm)
(min)
1
1
1
15
60
60
60
2
Peaks Outside
4.0-
3.3-
3.3-
<10
7.2-
1.0-
1.0-
29-
90
23
40
- >600
11.3
12.6
13.0
120 3.0 - 8.5
Comments
Sample of Dutch homes
Sample of Dutch homes,
breathing zone samples,
levels related to geisers
Measurements were taken
under a variety of gas
range operating and
ventilation conditions
"GR = gas range, K = unvented kerosene space heater, CK = unvented convective kerosene space heater, RK = radiant unvented kerosene space heater,
UVGSHP = unvented gas space heater used as primary heat source, UVGSHS = UVGSH as a secondary heat source, GA = gas appliances includes geisers
(water heaters).
bK = kitchen, LR = living room, D = den, B = bedroom.
-------
(>600 ppm) were associated with emissions from geisers (water heaters), found in a large
study conducted in The Netherlands (Brunekreef et al., 1982). Peak levels of CO associated
with gas ranges were from 1.0 to over 100 ppm. This broad range is somewhat consistent
with the range of CO emissions observed in studies evaluating CO emissions from gas ranges
5 (i.e., Table 7-1). The variability is in part due to number of burners used, flame condition,
condition of the burners, etc. As might be expected radiant kerosene heaters produced higher
CO concentrations than convective heaters. UVGSHs generally were associated with higher
CO peaks than gas ranges or kerosene heaters. As noted earlier the peaks associated with gas
or kerosene heaters are likely to be sustained over longer periods of time because of the long
10 source use times.
Test houses have been used by investigators to evaluate the impact of specific sources,
modifications to sources, and variations in their use on residential peak CO concentrations.
In one of the earliest investigations of indoor air quality, Wade et al. (1975) measured
indoor and outdoor CO levels in four houses that had gas stoves. Using an NDIR analyzer,
15 indoor concentrations were found to range from 1.7 to 3.8 times higher than the outdoor
levels. CO levels in one house exceeded 9 ppm, the NAAQS reference level. A time history
of CO measured in one house is shown in Figure 7-2. For this house and for the time
averaging period used, CO was well mixed through the house. As part of a modeling study of
emissions from a gas range, Davidson et al. (1987) measured CO concentrations in three
20 residences. Peak CO levels in excess of 5 ppm were measured in one town house.
Indoor CO levels associated with wood-burning stoves were measured in two test house
studies. In one study (Humphreys et al. 1986) indoor CO levels associated with the use of
both airtight (conventional and catalytic) and non-airtight wood heaters were evaluated in a
337 m3 weatherized home. Indoor CO concentrations were higher than outdoor levels for all
25 tests. Conventional airtight stoves produced indoor CO levels typically about 1 to 2 ppm
above background level with a peak concentration of 9.1 ppm. Use of non-airtight stoves
resulted in average indoor CO concentrations 2 to 3 ppm above outdoors with peak
concentrations as high as 29.6 ppm. In a 236 m3 house (Traynor et al., 1984), four wood-
burning stoves (three airtight and one non-airtight) were tested. The airtight stoves generally
30 resulted in small contributions to both average and peak indoor CO levels (0.1 to 1 ppm for
March 12, 1990 7-37 DRAFT - DO NOT QUOTE OR CITE
-------
Kitchen over stove
Kitchen i meter from stove
Living room
Outside
on i
14 rmnj
1 burner on |
3 mm -
1 burner on 1
7mm.— J
jurrer on
3 mm. —^
t b
6 mm
1 burner on 1
5 mm — .
1 burner onj
3 mm -*-4
1 burner on I _J L2«n OP
13000V- 11 mm.—1 ' ' 55 mm.
12000 |
11000^
.20000
M 9000
JJ 8000
| 7000
2 6000
§ 5000-
§ 4000 -
8 3000-
2000
1000-
0400 0800 1200 1600 2000 2400 0400 0800 1200
J/31 Time, hrs 2/01
Figure 7-2. A time history of CO concentrations, 2-hour averages, winter of 1974.
Source: Wade et al. (1975).
March 12, 1990
7-38 DRAFT - DO NOT QUOTE OR CITE
-------
the average and 0.2 to 2.7 ppm for the peak). The non-airtight stove contributed as much as
9.1 ppm to the average indoor level and 43 ppm to the peak.
Indoor Concentrations Related to Environmental Tobacco Smoke
5 Carbon monoxide has been measured extensively in chamber studies as a surrogate for
environmental tobacco smoke (e.g., Bridge and Corn, 1972; Hoegg, 1972; Penkala and
De Oliveira, 1975; Weber et al., 1976, 1979a,b; Weber, 1984; Leaderer et al., 1984;
Clausen et al., 1985). Under steady-state conditions in chamber studies, where outdoor CO
levels are monitored and the tobacco brands and smoking rates are controlled, CO can be a
10 reasonably good indicator of environmental tobacco smoke (ETS) and is used as such. Under
such chamber conditions CO concentrations typically range from less than 1 to over 10 ppm.
A number of field studies have monitored CO in different indoor environments with and
without smoking occupants. A summary of the results of these studies is shown in
Table 7-15 (National Research Council, 1986, Table 2-4). Although CO concentrations
15 generally were higher in indoor spaces when smoking occurred, the concentrations were
highly variable. The variability of CO production from tobacco combustion, number of
cigarettes smoked, and differences in ventilation and variability of outdoor concentrations
make it difficult to assess the contribution of tobacco combustion in indoor CO
concentrations. The chamber studies and field studies conducted do indicate that under
20 typical smoking conditions encountered in residences or offices, CO concentrations can be
expected to be above background outdoor levels, but lower than the levels resulting from
other unvented combustion sources. In indoor spaces where heavy smoking occurs and in
small indoor spaces CO emissions from tobacco combustion will be an important contributor
to CO concentrations.
25
7.3.3 Spatial Concentration Variations
Spatial variations of CO concentrations within a space are a function of mixing within
and between spaces. Spatial variations of CO in a space are likely to be minimized if a
continuous or nearly continuous source of CO exists (i.e., unvented kerosene or gas space
30 heater) due to the strong convective currents which enhance rapid mixing. Intermittent
sources (i.e., gas burner use or tobacco combustion) are likely to produce a more pronounced
March 12, 1990 7-39 DRAFT - DO NOT QUOTE OR CITE
-------
TABLE 7-15. MEASURED CONCENTRATIONS OF CARBON MONOXIDE
IN ENVIRONMENTAL TOBACCO SMOKE"
i_« Indoor
•O
°
^j
j^.
O
o
^m*t
5*
3
i
i
*Q
Q
H
tn
2
Jo
O
Location
Rooms
Train
Submarines
(66m3)
18 military
aircraft
8 commercial
aircraft
Rooms
14 public places
Ferry boat
Theater foyer
Intercity bus
2 conference rooms
Office
Automobile
9 night clubs
14 restaurants
45 restaurants
33 stores
3 hospital lobbies
6 coffee houses
Room
Hospital lobby
2 train
compartments
Automobile
10 offices
15 restaurants
14 night clubs
and taverns
Tobacco Burned
-
1-18 smokers
157 cigarettes/day
94 - 103 cigarettes/day
-
—
—
-
-
-
23 cigarettes
3 cigarettes
-
-
-
2 smokers
(4 cigarettes)
-
-
-
-
-
Varied
18 smokers
12-30 smokers
2-3 smokers
3 smokers
2 smokers
—
—
-
Ventilation
-
Natural
Yes
Yes
Yes
Yes
-
-
-
-
15 changes/h
15 changes/h
8 changes/h
236m3/h
Natural
Natural
Mechanical
Varied
-
-
-
-
-
-
-
-
Natural, open
Natural, closed
-
-
-
Mean
—
-
<40
<40
<2-5
<2
-
<10
18.4 ± 8.7
3.4 ± 0.8
32
18
-
-
-
-
-
13.4
9.9 ± 5.5
8.2 ± 2.2
10.0 ± 4.2
-
2-23
50
5
-
14
20
2.5 ± 10
4.0 ± 2.5
13.0 ± 7.0
Range
4.3 -9
0-40
-
-
-
-
5-25
-
-
—
-
—
8 (peak)
<2.5 - 4.6
<2.9 - 9.0
42 (peak)
32 (peak)
6.5-41.9
-
7.1 ± 1.7
11.5 ± 6.5
4.8
-
—
-
4-5
-
-
1.5 - 1.0
1.0-9.5
3.0 - 29.0
Outdoor
Mean
2.2 ± 0.98
-
-
-
-
-
-
-
3.0 ± 2.4
1.4 ± 0.8
-
—
1 -2
-
—
—
-
-
9.2 (outdoor)
—
11.5 ±6.5
-
-
_
-
-
-
-
2.5 ± 1.0
2.5 ± 1.5
3.0 ± 2.0
Range References
0.4 - 4.5 Coburn et al. (1965)
— Harmsenand Eflenberger
(1957)
Cano et al. (1970)
—
U.S. Department of
Transportation (1971)
- U.S. Department of
Transportation (1971)
Porthein (1971)
Perry (1973)
- Godin et al. (1972)
Godin et al. (1972)
Seiff(1973)
_
Slavin and Hertz (1975)
Harke (1974)
— _
13.5 (peak) HarkeandPeters(1974)
15.0 (peak)
Sebben et al. (1977)
3.0-35 Sebben et al. (1977)
- Sebben et al. (1977)
Sebben et al. (1977)
Sebben et al. (1977)
Badre et al. (1978)
Badre et al. (1978)
Badre et al. (1978)
Badre et al. (1978)
Badre et al. (1978)
- Badre et al. (1978)
1 .5 ± 4.5 Chappell and Parker
(1977)
1 .0 - 5 .0 Chappell and Parker
(1977)
1 .0 - 5 .0 Chappell and Parker
(1977)
-------
I
i
O
O
1
TABLE 7-15 (cont'd). MEASURED CONCENTRATIONS OF CARBON MONOXIDE
IN ENVIRONMENTAL TOBACCO SMOKE'
Indoor
Location
Tavern
Office
Restaurant
Restaurant
Bar
Cafeteria
44 offices
25 offices
Tavem
Tavern
Tobacco Burned Ventilation
- Artifical
— None
— Natural, open
— Mechanical
- Natural
— Natural, open
— 11 changes/h
— —
_ —
v- 6 changes/h
— 1-2 changes/h
Mean
8.5
—
1.0
5.1
2.6
4.8
1.2
1.1
2.78 ± 1.42
11.5
12.0
Range
—
35 (peak)
10.0 (peak)
2.1-9.9
1.4-3.4
2.4 - 9.6
0.7 - 1.7
6.5 (max)
—
10- 12
3 -22
Outdoor
Mean Range
_ _
- -
- -
4.8 (outdoors) -
1.5 (outdoors) —
1.7 (outdoors) -
0.4 (outdoors) -
- -
2.59 ± 2.33
2 (outdoors) -
— —
References
Chappell and Parker
(1977)
Chappell and Parker
(1977)
Chappell and Parker
(1977)
Fischer et al. (1978)
Fischer et al. (1978)
Weber et al. (1976)
Weber et al. (1976)
Weber (1984)
Szadkowski and Harke
(1976)
Cuddebacketal. (1976)
Cuddebacketal.(1976)
Time-weighted average (TWA) of carbon monoxide, 50 ppm (55 mg/m3). TWA = average concentration to which worker may be exposed continuously for 8 hr without damage to health (National
Institute for Occupational Safety and Health, 1971).
Source: NRC (1986) Table 2-4.
I
r
5
-------
spatial gradient. The within-home spatial variations are related to such variables as air-
exchange rates among rooms, air mixing within a room, volume of a house, and location and
use of the source. The question of the spatial variability of CO indoors as a function of
different indoor sources has not been evaluated in any detail in any field study.
5
7.3.4 Summary of Indoor Concentrations
Indoor concentrations of CO are a function of outdoor concentrations, indoor sources
(source type, source condition, source use, etc.), infiltration/ventilation, and air mixing
between and within rooms. In residences without sources, average CO concentrations are
10 approximately equal to average outdoor levels. Proximity to outdoor sources (i.e., structures
near heavily traveled roadways or with attached garages or parking garages) can have a major
impact on indoor CO concentrations.
The development of small lightweight and portable electrochemical CO monitors over
the past decade has permitted the measurement of personal CO exposures and CO
15 concentrations in a number of indoor environments. The available data on indoor CO
concentrations have been obtained from total personal exposure studies or studies where
various indoor environments have been targeted for measurements.
The extensive total personal CO exposure studies conducted by EPA in Washington,
DC, and Denver, CO, have shown that the highest CO concentrations occur in indoor
20 microenvironments associated with transportation sources (parking garages, cars, buses, etc.).
Concentrations in these environments were found to frequently exceed 9 ppm. Studies
targeted toward specific indoor microenvironments also have identified the indoor commuting
microenvironment as an environment in which CO concentrations frequently exceed 9 ppm
and occasionally exceed 35 ppm. Special environments or occurrences (indoor ice skating
25 rinks, offices where emissions from parking garages migrate indoors, etc.) have been reported
where indoor CO levels can exceed the current ambient one- and eight-hour standards (9 and
35 ppm, respectively).
A majority of the targeted field studies monitored indoor CO levels as a function of the
presence or absence of combustion sources (gas ranges, unvented gas and kerosene space
30 heaters, wood burning stoves and fireplaces, and tobacco combustion). The results of these
studies indicate that the presence and use of a unvented combustion source results in indoor
March 12, 1990 7-42 DRAFT - DO NOT QUOTE OR CITE
-------
CO levels above those found outdoors. The associated increase in CO concentrations can
vary considerably as a function of the source, source use, condition of the source, and
averaging time of the measurement. Intermittent sources such as gas cooking ranges can
result in high peak CO concentrations (in excess of 9 ppm), while long-term average
5 concentrations (i.e., 24-h) associated with gas ranges are considerably lower (on the order of
1 ppm). The contribution of tobacco combustion to indoor CO levels is variable. Under
conditions of high smoking and low ventilation the contribution can be on the order of a few
parts per million. One study suggested that the contribution to residential CO concentrations
of tobacco combustion is on the order of 1 ppm while another study showed no significant
10 increase in residential CO levels.
Unvented combustion sources that are used for substantial periods of time (i.e.,
unvented gas and kerosene space heaters) appear to be the major contributors to residential
CO concentrations. One extensive study of unvented gas space heaters indicated that 12% of
the homes had 15-h average CO concentrations greater than 8 ppm with the highest
15 concentration at 36.6 ppm. Only very limited data are available on the contribution of
kerosene heaters to the average CO concentrations in residences and these data indicate a
much lower contribution than gas heaters. Peak CO concentrations associated with both
unvented gas and kerosene space heaters often exceed the current ambient one- and eight-
hour, standards (9 and 35 ppm, respectively) in residences, and due to the nature of the
20 source (continuous) those peaks tend to be sustained for several hours.
Very limited data on CO levels in residences with wood-burning stoves or fireplaces is
available. Non-airtight stoves can contribute substantially to residential CO concentrations,
while airtight stoves can result in small increases. The available data indicate that fireplaces
do not contribute measurably to average indoor concentrations. No information is available
25 for samples of residences with leaky flues. In addition, there is no information available on
short-term indoor CO levels associated with these sources nor are there studies that examine
the impact of attached garages on residential CO concentrations.
The available data on short-term (one-hour) and long-term (eight-hour) indoor CO
concentrations as a function of microenvironments and sources in those microenvironments
30 are not adequate to assess exposures in those environments. In addition, little is known about
March 12, 1990 7-43 DRAFT - DO NOT QUOTE OR CITE
-------
the spatial variability of CO indoors. These indoor microenvironments represent the most
important CO exposures for individuals and as such need to be characterized better.
7.4 REFERENCES
Akland, G. G.; Hartwell, T. D.; Johnson, T. R.; Whitmore, R. W. (1985) Measuring human exposure to carbon
10 monoxide in Washington, D.C., and Denver, Colorado, during the winter of 1982-1983. Environ. Sci.
Technol. 19: 911-918.
Amendola, A. A.; Hanes, N. B. (1984) Characterization of indoor carbon monoxide levels produced by the
automobile. In: Berglund, B.; Lindvall, T.; Sundell, J., eds. Indoor air: proceedings of the 3rd
15 international conference on indoor air quality and climate, v. 4, chemical characterization and personal
exposure; August; Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building Research;
pp. 97-102. Available from: NTIS, Springfield, VA; PB85-104214.
American National Standards Institute. (1982) American national standard for household cooking gas appliances.
20 ANSI Z21-1982.
Badre, R.; Guillerme, R.; Abram, N.; Bourdin, M.; Dumas, C. (1978) Pollution atmospherique par la fumee de
-tabac [Atmospheric pollution by smoking]. Ann. Pharm. Fr. 36: 443-452.
25 Belles, F. E.; DeWerth, D. W.; Himmel, R. L. (1979) Emissions of furnace and water heaters predominant to
the California market. Presented at: 72nd annual meeting and exhibition of the Air Pollution Control
Association; June; Cincinnati, OH. A.G.A.L.
Borrazzo, J. E.; Osborn, J. F.; Fortmann, R. C.; Keefer, R. L.; Davidson, C. I. (1987) Modeling and
30 monitoring of CO, NO and NO2 in a modern townhouse. Atmos. Environ. 21: 299-311.
Bridge, D. P.; Corn, M. (1972) Contribution to the assessment of exposure of nonsmokers to air pollution from
cigarette and cigar smoke in occupied spaces. Environ. Res. 5: 192-209.
35 Brunekreef, B.; Smit, H. A.; Biersteker, K.; Boleij, J. S. M.; Lebret, E. (1982) Indoor carbon monoxide
pollution in The Netherlands. Environ. Int. 8: 193-196.
Caceres, T.; Soto, H.; Lissi, E.; Cisternas, R. (1983) Indoor house pollution: appliance emissions and indoor
ambient concentrations. Atmos. Environ. 17: 1009-1014.
40
Cano, J. P.; Catalin, J.; Badre, R.; Dumas, C.; Viala, A.; Guillerme, R. (1970) Determination de la nicotine par
chromatographic en phase gazeuse. II. Applications. Ann. Pharm. Fr. 28: 633-640.
Chaney, L. W. (1978) Carbon monoxide automobile emissions measured from the interior of a traveling
45 automobile. Science (Washington, DC) 199: 1203-1204.
50
Chappell, S. B.; Parker, R. J. (1977) Smoking and carbon monoxide levels in enclosed public places in New
Brunswick. Can. J. Public Health 68: 159-161.
March 12, 1990 7-44 DRAFT - DO NOT QUOTE OR CITE
-------
Clausen, G. H.; Fanger, P. O.; Cain, W. S.; Leaderer, B. P. (1985) The influence of aging, particle filtration
and humidity on tobacco smoke odor. In: Fanger, P. O., ed. CLIMA 2000: proceedings of the world
congress on heating, ventilating and air-conditioning, v. 4, indoor climate; August, Copenhagen,
Denmark. Copenhagen, Denmark: VVS Kongres; pp. 345-350.
5
Cobum, R. F.; Forster, R. E.; Kane, P. B. (1965) Considerations of the physiological variables that determine
the blood carboxyhemoglobin concentration in man. J. Clin. Invest. 44: 1899-1910.
Cole, J. T.; Zawacki, T. S. (1985) Emissions from residential gas-fired appliances. Chicago, IL: Institute of Gas
10 Technology; report no. 84/0164.
Colwill, D. M.; Hickman, A. J. (1980) Exposure of drivers to carbon monoxide. J. Air Pollut. Control Assoc.
30: 1316-1319.
15 Cortese, A. D.; Spengler, J. D. (1976) Ability of fixed monitoring stations to represent personal carbon
monoxide exposures. J. Air Pollut. Control Assoc. 26: 1144-1150.
Cote, W. A.; Wade, W. A., HI; Yocom, J. E. (1974) A study of indoor air quality. Washington, DC: U. S.
Environmental Protection Agency, Office of Research and Development; EPA report no.
20 EPA-650/4-74-042. Available from: NTIS, Springfield, VA; PB-238556.
Cuddeback, J. E.; Donovan, J. R.; Burg, W. R. (1976) Occupational aspects of passive smoking. Am. Ind. Hyg.
Assoc. J. 35: 263-267.
25 Dames; Moore. (1986) 50-home weatherization and indoor air quality study - summary report. Final report to
Wisconsin Power and Light Company.
Fischer, T.; Weber, A.; Grandjean, E. (1978) Luftverunreinigung durch Tabakrauch in Gaststaetten [Air
pollution due to tobacco smoke in restaurants]. Int. Arch. Occup. Environ. Health 41: 267-280.
30
Flachsbart, P. G.; Ott, W. R. (1984) Field surveys of carbon monoxide in commercial settings using personal
exposure monitors. Washington, DC: U. S. Environmental Protection Agency, Office of Monitoring
Systems and Quality Assurance; EPA report no. EPA-600/4-84-019. Available from: NTIS, Springfield,
VA; PB84-211291.
35
Flachsbart, P. G.; Mack, G. A.; Howes, J. E.; Rodes, C. E. (1987) Carbon monoxide exposures of Washington
commuters. J. Air Pollut. Control Assoc. 37: 135-142.
Fortmann, R. C.; Borrazzo, J. E.; Davidson, C. I. (1984) Characterization of parameters influencing indoor
40 pollutant concentrations. In: Berglund, B.; Lindvall, T.; Sundell, J., eds. Indoor air: proceedings of the
3rd international conference on indoor air quality and climate, v. 4, chemical characterization and
personal exposure; August; Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building
Research; pp. 259-264. Available from: NTIS, Springfield, VA; PB85-104214.
45 Girman, J. R.; Apte, M. G.; Traynor, G. W.; Allen, J. R.; Hollowell, C. D. (1982) Pollutant emission rates
from indoor combustion appliances and sidestream cigarette smoke. Environ. Int. 8: 213-221.
Godin, G.; Wright, G.; Shephard, R. J. (1972) Urban exposure to carbon monoxide. Arch. Environ. Health
25: 305-313.
50
Goto, Y.; Tamura, G. T. (1984) Measurement of combustion products from a gas cooking stove in a two-storey
house. Presented at: 77th annual meeting of the Air Pollution Control Association; San Francisco, CA.
Pittsburgh, PA: Air Pollution Control Association; paper no. 84-32.5.
March 12, 1990 7-45 DRAFT - DO NOT QUOTE OR CITE
-------
Harke, H.-P. (1974) Zum Problem des Passivrauchens: I. ueber den Einfluss des Rauchens auf die
CO-Konzentration in Buroraumen [The problem of passive smoking: I. the influence of smoking on the
CO concentration in office rooms]. Int. Arch. Arbeitsmed. 33: 199-206.
5 Harke, H.-P.; Peters, H. (1974) Zum Problem des Passivrauchens: IE. ueber den Einfluss des Rauchens auf die
CO-Konzentration im Kraftfahrzeug bei Fahrten im Stadtgebiet [The problem of passive smoking: III. the
influence of smoking on the CO concentration in driving automobiles]. Int. Arch. Arbeitsmed.
33: 221-229.
10 Harmsen, H.; Effenberger, E. (1957) Tabakrauch in Verkehrsmitteln, Wohn- und Arbeitsraumen. Arch. Hyg.
Bakteriol. 141: 383-400.
Hartwell, T. D.; Clayton, C. A.; Ritchie, R. M.; Whitmore, R. W.; Zelon, H. S.; Jones, S. M.; Whitehurst,
D. A. (1984) Study of carbon monoxide exposure of residents of Washington, DC and Denver, Colorado.
15 Research Triangle Park, NC: U. S. Environmental Protection Agency, Environmental Monitoring
Systems Laboratory; EPA report no. 600/4-84-031. Available from: NTIS, Springfield, VA;
PB84-183516.
Hartwell, T., et al. (1988) An investigation of infiltration and indoor air quality: determination of the impact of
20 indoor combustion sources on indoor air quality - final report. Submitted to the New York State Energy
Research and Development Authority.
Himmel, R. L.; DeWerth, D. W. (1974) Evaluation of the pollutant emissions from gas-fired ranges. American
Gas Association Laboratories; report no. 1492.
25
Hoegg, U. R. (1972) Cigarette smoke in closed spaces. EHP Environ. Health Perspect. 2: 117-128.
Holland, D. M. (1983) Carbon monoxide levels in microenvironment types of four U.S. cities. Environ. Int.
9: 369-378.
30
Humphreys, M. P.; Knight, C. V.; Pinnix, J. C. (1986) Residential wood combustion impacts on indoor carbon
monoxide and suspended participates. In: Proceedings from the 1986 EPA/APCA symposium on the
measurement of toxic air pollutants; April; Raleigh, NC. Research Triangle Park, NC: U. S.
Environmental Protection Agency, Environment Monitoring Systems Laboratory; pp. 736-747; EPA
35 report no. EPA-600/9-86-013. Available from: NTIS, Springfield, VA; PB87-182713.
Johnson, T. (1984) A study of personal exposure to carbon monoxide in Denver, Colorado. Research Triangle
Park, NC: U. S. Environmental Protection Agency, Environmental Monitoring Systems Laboratory; EPA
report no. EPA-600/4-84-014. Available from: NTIS, Springfield, VA; PB84-146125.
40
Johnson, D.; Billick, I.; Moschandreas, D.; Relwani, S. (1984) Emission rates from unvented gas appliances. In:
Berglund, B.; Lindvall, T.; Sundell, J., eds. Indoor air: proceedings of the 3rd international conference
on indoor air quality and climate, v. 4, chemical characterization and personal exposure; August;
Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building Research; pp. 367-373. Available
45 from: NTIS, Springfield, VA; PB85-104214.
Jones, K. H., et al. (1983) Kerosene heater indoor air quality study - monitoring protocol and preliminary results.
Weston report prepared for Kerosun, Inc.
50 Koontz, M. D.; Nagda, N. L. (1987) Survey of factors affecting NO2 concentrations. In: Seifert, B.; Esdorn, H.;
Fischer, M.; Rueden, H.; Wegner, J., eds. Indoor air '87: proceedings of the 4th international conference
on indoor air quality and climate, v. 1, volatile organic compounds, combustion gases, particles and
fibres, microbiological agents; August; Berlin, Federal Republic of Germany. Berlin, Federal Republic of
Germany: Institute for Water, Soil, and Air Hygiene; pp. 430-434.
March 12, 1990 7-46 DRAFT - DO NOT QUOTE OR CITE
-------
Koontz, M. D.; Nagda, N. L. (1988) A topical report on a field monitoring study of homes with unvented gas
space heaters: v. III. methodology and results. Gas Research Institute final report; contract
no. 5083-251-0941.
5 Leaderer, B. P. (1982) Air pollutant emissions from kerosene space heaters. Science (Washington, DC)
218: 1113-1115.
Leaderer, B. P.; Cain, W. S.; Isseroff, R.; Berglund, L. G. (1984) Ventilation requirements in buildings-n.
particulate matter and carbon monoxide from cigarette smoking. Atmos. Environ. 18: 99-106.
10
Leaderer, B. P.; Stolwijk, J. A. J.; Zagraniski, R. T.; Qing-Shan, M. (1984) A field study of indoor air
contaminant levels associated with unvented combustion sources. Presented at: 77th annual meeting of the
Air Pollution Control Association; June; San Francisco, CA. Pittsburgh, PA: Air Pollution Control
Association; paper no. 84-33.3.
15
Leaderer, B. P.; Zagraniski, R. T.; Berwick, M.; Stolwijk, J. A. J.; Qing-Shan, M. (1984) Residential exposures
to NO2, SO2 and HCHO associated with unvented kerosene space heaters, gas appliances and sidestream
tobacco smoke. In: Berglund, B.; Lindvall, T.; Sundell, J., eds. Indoor air: proceedings of the 3rd
international conference on indoor air quality and climate, v. 4, chemical characterization and personal
20 exposure; August; Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building Research;
pp. 151-156. Available from: NTIS, Springfield, VA; PB85-104214.
Leaderer, B. P.; Zagraniski, R. T.; Holford, T. R.; Berwick, M.; Stolwijk, J. A. J. (1984) Multivariate model
for predicting NO2 levels in residences based upon sources and source use. In: Berglund, B.; Lindvall,
25 T.; Sundell, J., eds. Indoor air: proceedings of the 3rd international conference on indoor air quality and
climate, v. 4, chemical characterization and personal exposure; August; Stockholm, Sweden. Stockholm,
Sweden: Swedish Council for Building Research; pp. 43-48. Available from: NTIS, Springfield, VA;
PB85-104214.
30 Leaderer, B. P.; Zagraniski, R. T.; Berwick, M.; Stolwijk, J. A. J. (1986) Assessment of exposure to indoor air
contaminants from combustion sources: methodology and application. Am. J. Epidemiol. 124: 275-289.
Lebret, E.; Noy, D.; Boley, J.; Brunekreef, B. (1987) Real-time concentration measurements of CO and NO2 in
twelve homes. In: Seifert, B.; Esdorn, H.; Fischer, M.; Rueden, H.; Wegner, J., eds. Indoor air '87:
35 proceedings of the 4th international conference on indoor air quality and climate, v. 1, volatile organic
compounds, combustion gases, particles and fibres, microbiological agents; August; Berlin, Federal
Republic of Germany. Berlin, Federal Republic of Germany: Institute for Water, Soil, and Air Hygiene;
pp. 435-439.
40 Massachusetts Institute of Technology. (1976) Experimental evaluation of range-top burner modification to reduce
NOX formation. American Gas Association; no. M40677.
McCarthy, J.; Spengler, J.; Chang, B.-H.; Coultas, D.; Samet, J. (1987) A personal monitoring study to assess
exposure to environmental tobacco smoke. In: Seifert, B.; Esdorn, H.; Fischer, M.; Rueden, H.;
45 Wegner, J., eds. Indoor air '87: proceedings of the 4th international conference on indoor air quality and
climate, v. 2, environmental tobacco smoke, multicomponent studies, radon, sick buildings, odours and
irritants, hyperreactivities and allergies; August; Berlin, Federal Republic of Germany. Berlin, Federal
Republic of Germany: Institute for Water, Soil and Air Hygiene; pp. 142-146.
50
March 12, 1990 7-47 DRAFT - DO NOT QUOTE OR CITE
-------
McCarthy, S. M.; Yarmac, R. F.; Yocom, J. E. (1987) Indoor nitrogen dioxide exposure: the contribution from
unvented gas space heaters. In: Seifert, B.; Esdorn, H.; Fischer, M.; Rueden, H.; Wegner, J., eds.
Indoor air '87: proceedings of the 4th international conference on indoor air quality and climate, v. 1,
volatile organic compounds, combustion gases, particles and fibres, microbiological agents; August;
5 Berlin, Federal Republic of Germany. Berlin, Federal Republic of Germany: Institute for Water, Soil,
and Air Hygiene; pp. 478-482.
Moffatt, S. (1986) Backdrafting woes. Prog. Builder (Dec.): 25-36.
10 Moschandreas, D. J.; Zabransky, J., Jr. (1982) Spatial variation of carbon monoxide and oxides of nitrogen
concentrations inside residences. Environ. Int. 8: 177-183.
Moschandreas, D. J.; Relwani, S. M.; O'Neill, H. J.; Cole, J. T.; Elkins, R. H.; Macriss, R. A. (1985)
Characterization of emission rates from indoor combustion sources. Chicago, IL: Gas Research Institute;
15 report no. GRI 85/0075. Available from: NTIS, Springfield, VA; PB86-103900.
Nagda, N. L.; Koontz, M. D. (1985) Microenvironmental and total exposures to carbon monoxide for three
population subgroups. J. Air Pollut. Control Assoc. 35: 134-137.
20 National Center for Health Statistics. (1986) Vital statistics of the United States 1982: v. n - mortality, part A.
Hyattsville, MD: U. S. Department of Health and Human Services, Public Health Service.
National Institute for Occupational Safety and Health/National Institute of Environmental Health Sciences. (1971)
Radon daughter exposure and respiratory cancer - quantitative and temporal aspects. Washington, DC:
25 National Institute for Occupational Safety and Health/National Institute of Environmental Health Sciences;
joint monograph no. 1. Available from: NTIS, Springfield, VA.
National Research Council. (1981) Indoor pollutants. Washington, DC: National Academy Press.
30 National Research Council. (1986) Environmental tobacco smoke: measuring exposures and assessing health
effects. Washington, DC: National Academy Press.
Perry, J. (1973) Fasten your seatbelts: non-smoking. B. C. Med. J. 15: 304-305.
35 Petersen, G. A.; Sabersky, R. H. (1975) Measurements of pollutants inside an automobile. J. Air. Pollut.
Control Assoc. 25: 1028-1032.
Porthein, F. (1971) Zum Problem des "Passiv-rauchens." Munch. Med. Wochenschr. 118: 707-709.
40 Rickert, W. S.; Robinson, J. C.; Collishaw, N. (1984) Yields of tar, nicotine, and carbon monoxide in the
sidestream smoke for 15 brands of Canadian cigarettes. Am. J. Public Health 74: 228-231.
Sebben, J.; Pimm, P.; Shephard, R. J. (1977) Cigarette smoke in enclosed public facilities. Arch. Environ.
Health 32: 53-58.
45
Seiff, H. E. (1973) Carbon monoxide as an indicator of cigarette-caused pollution levels in intercity buses.
Washington, DC: U. S. Department of Transportation, Bureau of Motor Carrier Safety; publication no.
BMCS-IHS-73-1.
50 Sisovic, A.; Fugas, M. (1985) Indoor concentrations of carbon monoxide in selected urban microenvironments.
Environ. Monit. Assess. 5: 199-204.
March 12, 1990 7-48 DRAFT - DO NOT QUOTE OR CITE
-------
Slavin, R. G.; Hertz, M. (1975) Indoor air pollution: a study of the thirtieth annual meeting of the American
Academy of Allergy. 30th annual meeting of the American Academy of Allergy; February; San Diego,
CA.
5 Spengler, J. D.; Stone, K. R.; Lilley, F. W. (1978) High carbon monoxide levels measured in enclosed skating
rinks. J. Air Pollut. Control Assoc. 28: 776-779.
Sterling, T. D.; Sterling, E. (1979) Carbon monoxide levels in kitchens and homes with gas cookers. J. Air
Pollut. Control Assoc. 29: 238-241.
10
Surgeon General of the United States. (1986) The health consequences of involuntary smoking: a report of the
Surgeon General. Rockville, MD: U. S. Department of Health and Human Services, Office on Smoking
and Health.
15 Szadkowski, D.; Harke, J. (1976) Body burden of carbon monoxide from passive smoking in offices. Inn. Med.
3: 310-313.
Thrasher, W. H.; Dewerth, D. W. (1977) Evaluation of the pollutant emissions from gas-fired water heaters.
Cleveland, OH: American Gas Association Laboratories; research report no. 1507.
20
Tikalsky, S.; Reisdorf, K.; Flickinger, J.; Totzke, D.; Haywood, J. (1987) Gas range/oven emisssions impact
analysis. Final report July 1985-December 1987. Chicago, IL: Gas Research Institute; report no.
D/M-7683-062. Available from: NTIS, Springfield, VA; PB88-232756/XAB.
25 Traynor, G. W.; Anthon, D. W.; Hollowell, C. D. (1982) Technique for determining pollutant emissions from a
gas-fired range. Atmos. Environ. 16: 2979-2987.
Traynor, G. W.; Allen, J. R.; Apte, M. G.; Girman, J. R.; Hollowell, C. D. (1983) Pollutant emissions from
portable kerosene-fired space heaters. Environ. Sci. Technol. 17: 369-371.
30
Traynor, G. W.; Apte, M. G.; Carruthers, A. R.; Dillworth, J. F.; Grimsrud, D. T. (1984) Pollutant emission
rates from unvented infrared and convective gas-fired space heaters. Berkeley, CA: U. S. Department of
Energy, Lawrence Berkeley Laboratory; report no. LBL-18258; Available from: NTIS, Springfield, VA;
DE85010647/XAB.
35
Traynor, G. W.; Girman, J. R.; Apte, M. G.; Dillworth, J. F.; White, P. D. (1985) Indoor air pollution due to
emissions from unvented gas-fired space heaters. J. Air Pollut. Control Assoc. 35: 231-237.
Traynor, G. W.; Apte, M. G.; Sokol, H. A.; Chuang, J. C.; Mumfbrd, J. L. (1986) Selected organic pollutant
40 emissions from unvented kerosene heaters. Presented at: 79th annual meeting of the Air Pollution Control
Association; June; Minneapolis, MN. Pittsburgh, PA: Air Pollution Control Association; paper
no. 86-52.5.
Traynor, G. W.; Apte, M. G.; Carruthers, A. R.; Dillworth, J. F.; Grimsrud, D. T.; Gundel, L. A. (1986)
45 Indoor air pollution due to emission from woodburning stoves. Berkeley, CA: Lawrence Berkeley
Laboratories; report no. LBL-17854.
Traynor, G. W.; Apte, M. G.; Carruthers, A. R.; Dillworth, J. F.; Grimsrud, D. T.; Gundel, L. A. (1987)
Indoor air pollution due to emissions from wood-burning stoves. Environ. Sci. Technol. 21: 691-697.
50
U. S. Bureau of the Census. (1982) Census of population and housing. Supplementary report: provisional
estimates of social, economic and housing characteristics - states and selected standard metropolitan
statistical areas. Washington, DC: U. S. Government Printing Office.
March 12, 1990 7-49 DRAFT - DO NOT QUOTE OR CITE
-------
U. S. Department of Transportation; U. S. Department of Health, Education, and Welfare. (1971) Health aspects
of smoking in transport aircraft. Washington, DC: U. S. Department of Transportation, Federal Aviation
Administration, U. S. Department of Health, Education, and Welfare, NIOSH.
5 Wade, W. A., Ill; Cote, W. A.; Yocom, J. E. (1975) A study of indoor air quality. J. Air Pollut. Control
Assoc. 25: 933-939.
Wallace, L. A. (1983) Carbon monoxide in air and breath of employees in an underground office. J. Air Pollut.
Control Assoc. 33: 678-682.
10
Weber, A. (1984) Annoyance and irritation by passive smoking. Prev. Med. 13: 618-625.
Weber, A.; Jermini, C.; Grandjean, E. (1976) Irritating effects on man of air pollution due to cigarette smoke.
Am. J. Public Health 66: 672-676.
15
Weber, A.; Fischer, T.; Grandjean, E. (1979a) Passive smoking in experimental and field conditions. Environ.
Res. 20: 205-216.
Weber, A.; Fischer, T.; Grandjean, E. (1979b) Passive smoking: irritating effects of the total smoke and the gas
20 phase. Int. Arch. Occup. Environ. Health 43: 183-193.
Whitmore, R. W.; Jones, S. M.; Rosenzweig, M. S. (1984) Final sampling report for the study of personal CO
(carbon monoxide) exposure. Research Triangle park, NC: U. S. Environmnetal Protection Agency,
Environmental Monitoring Systems Laboratory; EPA report no. EPA-600/4-84-034. Available from:
25 NTIS, Springfield, VA; PB84-181957.
Womble, S. E. (1988) Personal communication. U. S. Consumer Product Safety Commission.
World Health Organization. (1985) Air quality guidelines: indoor air pollutants. Geneva, Switzerland: World
30 Health Organization, Regional Office for Europe.
Yamanaka, S. (1984) Decay rates of nitrogen oxides in a typical Japanese living room. Environ. Sci. Technol.
18: 566-570.
35 Yocom, J. E.; Clink, W. L.; Cote, W. A. (1971) Indoor/outdoor air quality relationships. J. Air Pollut. Control
Assoc. 21: 251-259.
Zawacki, T. S.; Cole, J. T.; Huang, V.; Banasiuk, H.; Macriss, R. A. (1984) Efficiency and emissions
improvement of gas-fired space heaters. Task 2. Unvented space heater emission reduction. Chicago, IL:
40 Gas Research Institute; report no. GRI-84/0021. Available from: NTIS, Springfield, VA; PB84-237734.
Ziskind, R. A.; Rogozen, M. B.; Carlin, T.; Drago, R. (1981) Carbon monoxide intrusion into sustained-use
vehicles. Environ. Int. 5: 109-123.
45 Ziskind, R. A.; Fite, K.; Mage, D. T. (1982) Pilot field study: carbon monoxide exposure monitoring in the
general population. Environ. Int. 8: 283-293.
March 12, 1990 7-50 DRAFT - DO NOT QUOTE OR CITE
-------
8. POPULATION EXPOSURE TO CARBON
MONOXIDE
5 8.1 INTRODUCTION
A fundamental purpose of the Clean Air Act is to protect public health. The NAAQS
are set at levels which attempt to provide a margin of safety to protect the health of the
populace from adverse effects of air pollutants. In setting a pollutant standard, population
exposure is an important consideration, because public health can be affected adversely by an
10 air pollutant only if the following two conditions coincide.
(1) Persons actually are or potentially would be exposed in daily living to levels of
the pollutant occurring in ambient air locations at or above undesirable
concentrations.
15
(2) The air pollutant causes adverse effects on human health in sensitive population
groups at these concentrations.
20
This chapter focuses on the degree to which the population actually is exposed to
outdoor, in-transit, or indoor concentrations of CO that might produce adverse health effects.
Chapter 10 deals with the second topic, the physiological and other health effects associated
with a person's exposure to various concentrations of CO.
25 Carbon monoxide is emitted from sources of incomplete combustion of HC fuels.1
Consequently, ambient concentrations of CO can reach high levels close to emission sources.
The strong source-dependence of CO leads to highly variable spatial and temporal
concentrations in urban environments (see Chapter 6). Because of the variable concentration
patterns exhibited by CO, it is necessary to address the relationships between ambient
30 concentrations and human exposures to evaluate the potential health risk associated with actual
population exposures.
'Hydrocarbon fuels are burned incompletely by internal combustion engines (e.g., automobiles, trucks, and
small utility engines), and by sources such as cigarettes, forest fires, and poorly adjusted gas burners.
March 12, 1990 8-1 DRAFT-DO NOT QUOTE OR CITE
-------
Compliance with the NAAQS currently is judged by comparing the standards to data
from fixed, ambient-air monitoring stations. The sites for such stations are chosen, however,
to monitor ambient conditions in a number of neighborhood types (Ott, 1977) rather than
actual human exposure. In fact, a number of studies have shown that ambient-air monitoring
5 stations do not necessarily reflect the concentrations to which people are actually exposed.
This difference between fixed stations and exposures occurs because of the spatial and -
temporal variability of carbon monoxide. In contrast to the stationary location of an ambient
monitor, people are usually moving through a succession of microenvironments (e.g., homes,
sidewalks, buses, automobiles, shopping malls, downtown street canyons, restaurants, offices,
10 factories, and garages) where they may spend time in closer proximity to CO sources and in
more enclosed spaces than the outdoors. The result is that existing ambient monitoring
stations often do not reflect individual exposure patterns, nor do they necessarily reflect the
highest concentrations to which those people are exposed. However, fixed monitors do give
some general information on the overall level of exposure of a population to CO and are
15 useful for a variety of other purposes (see Chapter 6).
Among all major air pollutants, CO has one of the clearest measures available of
biological dose (see Section 8.5). The concentration of CO circulating in the blood,
expressed as the percentage of Hb bound with CO, or COHb, is a useful measure of dose for
relating this pollutant to deleterious health effects (see Chapter 10). Blood COHb is in turn a
20 function of inhaled CO, breathing rate, blood volume, and other physiological factors (see
Chapter 9: Pharmacokinetic modeling). Because the relationship between ambient CO and
blood COHb is dynamic, it is necessary to know the timing and duration of an exposure series
in order to predict resulting COHb.
Because of the spatial and temporal variability of CO as well as the known functional
25 relationships between concentration and a measure of dose, CO is a model pollutant for
development and evaluation of improved approaches for assessing human exposure. A
number of field studies now have been undertaken that provide a quantitative assessment of
the disparity between fixed monitoring stations and actual exposures. In addition, field
measurements of body burden (for example, blood COHb, and breath CO) are available for
30 comparison with fixed station monitoring data. Several personal-monitoring CO field studies
have employed representative statistical sampling procedures, allowing inferences to be made
March 12, 1990 8-2 DRAFT-DO NOT QUOTE OR CITE
-------
about the CO exposures (or COHb levels) of an entire population of a city or a region.
Finally, models of population exposure and activity patterns have been developed to bridge
the gap between ambient, fixed-station measurements, and actual personal exposures.
Additional data need to be developed from personal monitoring field studies for use in
5 validating these models. The models are important for improving our understanding of
human exposure to CO and for evaluating different control strategies. The models can
indicate locations and sources of most significant exposure and therefore may suggest control
strategies to reduce human exposure and the resulting deleterious health effects.
10
8.2 EXPOSURE MONITORING IN THE POPULATION
In recent years, researchers have focused on the problem of determining actual
population exposures to CO. There are two alternative approaches for estimating the
exposures of a population to air pollution: the "direct approach" using field measurement of a
15 representative population carrying PEMs; and the "indirect approach" involving computation
from field data of activity patterns and measured concentration levels within
microenvironments (Ott, 1982).
In the direct approach, as study participants engage in regular daily activities, they are
responsible for recording their exposures to the pollutant of interest. Subjects can record their
20 exposures in a diary, the method used in a study in Los Angeles (Ziskind et al., 1982), or
they can automatically store exposure data in a data logger, the method used in studies in
Denver (Johnson, 1984) and Washington, DC (Hartwell et al., 1984), which are summarized
by Akland et al. (1985). In all of these studies, subjects recorded the time and nature of their
activities while they monitored personal exposures to CO. The direct approach - sometimes
25 called the Total Exposure Assessment Methodology (TEAM) - is useful to obtain an exposure
inventory of a representative sample from either the general population, or from a specific
subpopulation, which can be defined by many demographic, occupational, and health factors.
The inventory can cover a range of microenvironments encountered over a period of interest
(e.g., a day), or it can focus on one particular microenvironment. With this flexibility,
30 policy analysts can assess the problem that emission sources pose to a particular subgroup
(e.g., commuters) active in a specific microenvironment (e.g., automobiles).
March 12, 1990 8-3 DRAFT-DO NOT QUOTE OR CITE
-------
The indirect approach to estimating personal exposure is to use PEMs or
microenvironmental monitors (MEMs) to monitor microenvironments rather than individuals.
Combined with additional data on human activities that occur in these microenvironments,
data from the indirect approach can be used to estimate the percentage of a subpopulation that
5 is at risk to pollutant concentrations that exceed national or state air quality standards.
Flachsbart and Brown (1989) conducted this type of study to estimate merchant exposure to
CO from motor vehicle exhaust at Honolulu's Ala Moana Shopping Center.
8.2.1 Personal Monitoring
10 The development of small PEM, as discussed in Chapter 5, made possible the large-
scale CO human exposure field studies in Denver, CO, and Washington, DC, in the winter of
1982-1983 (Akland et al., 1985). These monitors proved effective in generating 24-h CO
exposure profiles on 450 persons in Denver and 800 in Washington. Because personal
monitoring techniques are new, and few field studies have been done, the science of
15 measuring the exposures to chemicals in human populations is in an early stage of
development. The use of PEMs and concurrent diaries in large-scale population studies
requires rigorous quality control and introduces many new problems not present in ambient
monitoring studies. The PEMs must be rugged, self-powered, lightweight, and free of drift
while being carried and exposed to temperature variations; associated data loggers are
20 required to store the continuous PEM readings. The diary format must be clear, easy for the
subject to complete, and easy for the researcher to interpret. With good calibration practices,
the CO PEMs can provide a precision of less than +1 ppm. The Denver-Washington, DC
study is the only large-scale population exposure field study that yet has been undertaken.
Despite the complexity of such a study, the large probability sample and high time resolution
25 of the PEMs yielded a rich data base for characterizing the exposures of the population to CO
in two major U.S. cities. The findings have greatly increased the understanding of the
causes, severity, and variability of the exposures of human populations to CO.
Results from the Denver-Washington DC study (Akland et al., 1985) show that over
10% of the residents in Denver and 4% of the Washington DC residents were exposed to
30 eight-hour average CO levels above 9 ppm during the winter study period. This degree of
population exposure could not accurately be deduced from simultaneous data collected by the
March 12, 1990 8-4 DRAFT-DO NOT QUOTE OR CITE
-------
fixed-site monitors without taking into account other factors such as contributions from indoor
sources, elevated levels within vehicles, and individuals' activity patterns. In Denver, for
example, the fixed-site monitors exceeded the 9 ppm level only 3.1 % of the time. These
results indicate that the effects of personal activity, indoor sources, and especially time spent
5 commuting, greatly contribute to a person's CO exposure.
This study emphasizes that additional strategies are required to augment data from fixed
site monitoring networks in order to evaluate actual human CO exposures and health risks
within a community. The cumulative frequency distributions of CO data for both Denver and
Washington DC in Figure 8-1 show that personal monitors often measure higher concentration
10 than do fixed stations. As part of this study, comparisons were made of exposure to one-
hour CO concentrations as determined by personal monitors and of measured ambient
concentrations at fixed monitor sites. Correlations between personal monitor data and fixed-
site data were consistently poor; the fixed-site data usually explained less than 10% of the
observed variation in personal exposure. For example, one-hour CO measurements taken at
15 the nearest fixed stations only were weakly correlated (0.14 < r < 0.27) with office or
residential measurements taken with personal monitors (Akland et al., 1985).
The conclusion that exposure of persons to ambient CO and other pollutants does not
directly correlate with concentrations determined at fixed-site monitors is supported by the
work of others (Ott and Eliassen, 1973; Cortese and Spengler, 1976; Dockery and Spengler,
20 1981; Wallace and Ott, 1982).
In view of the high degree of variability of ambient CO concentrations over both space
and time, (see Chapter 6) the reported results are not surprising. A given fixed monitor is
unable to track the exposure of individuals to ambient CO as they go about their daily
activities, moving from one location to another, and seldom in the immediate vicinity of the
25 monitor. This does not necessarily mean, however, that fixed monitors do not give some
general information on the overall level of exposure of a population to CO. The Akland data,
although failing to show a correlation between individual personal monitor exposures and
simultaneous nearest fixed-site monitor concentrations did suggest that, in Denver, aggregate
personal exposures were lower on days of lower ambient CO levels as determined by fixed
30 site monitors and higher on days of higher ambient levels. Also, both fixed-site and personal
exposures were higher in Denver than in Washington.
March 12, 1990 8-5 DRAFT-DO NOT QUOTE OR CITE
-------
99.99
100
50-
E 20
0.
a.
c
.2 10.
c
o
g 5
O
O
o
•O
'x
i 2-
O
c
o
n
CO
O
1 -
0.5-
0.2-
0.1
Population Above Concentration Shown (%)
99 90 50 10 1
I I ! I I I I I I I I
9 ppm NAAQS
Denver:
Personal Exposure
Fixed Stations
Washington. DC:
_ Personal Exposure
Fixed Stations
0.01
10
I
50
M I i
90 99
0.01
-100
-50
-20 E
Q.
c
Mo .2
-5
-2
-1
0)
o
o
O
o
p
I
o
O
n
- o
-0.5
-0.2
99
0.1
99
Population Below Concentration Shown (%)
Figure 8-1. Frequency distributions of maximum eight-hour carbon monoxide population
exposures and fixed-ale monitor values in Denver, CO and Washington, DC;
November 1982 - February 1983.
Source: Based on Akland, et al. (1984).
March 12, 1990
8-6 DRAFT-DO NOT QUOTE OR CITE
-------
The PEMs have shown themselves to be powerful tools for quantifying air quality levels
in in-transit, outdoor, and indoor microenvironments. A great number of microenvironments
can be compared in one study. For example, Table 8-1 shows in-transit microenvironments
in Denver, ranked from highest to lowest concentration by arithmetic mean. The in-transit
5 microenvironment with the highest estimated CO concentration is the motor vehicle, whereas
walking and bicycling have the lowest CO concentrations. Outdoor microenvironments also
can be ranked (Table 8-2) for these data. Outdoor public garages and outdoor residential
garages and carports had the highest CO concentrations; outdoor service stations, vehicle
repair facilities, and parking lots had intermediate concentrations. In contrast, school grounds
10 and residential grounds had relatively low concentrations, whereas extremely low CO
concentrations were found in outdoor sports arenas, amphitheaters, parks, and golf courses.
Finally, a wide range of concentrations was found in Denver within indoor micro-
environments (Table 8-3). The highest indoor CO concentrations occurred in service stations,
vehicle repair facilities and public parking garages; intermediate concentrations were found in
15 shopping malls, residential garages, restaurants, offices, auditoriums, sports arenas, concert
halls, and stores; the lowest concentrations were found in health care facilities, public
buildings, manufacturing facilities, homes, schools, and churches.
One activity that influences personal exposure is commuting. An estimated 1 % of the
noncommuters in Washington were exposed to concentrations above 9 ppm for eight hours.
20 By comparison, an estimated 8% of persons reporting that they commuted more than 16 h per
week had CO exposures above the 9-ppm, eight-hour level. Finally, certain occupational
groups whose work brings them in close proximity to the internal combustion engine had a
potential for elevated CO exposures. These include automobile mechanics, parking garage or
gas station attendants, crane deck operators, cooks, taxi, bus, and truck drivers, firemen,
25 policemen, and warehouse and construction workers. Of the 712 CO exposure profiles
obtained in Washington 29 persons fell into this "high- exposure" category. Of these, 25%
had eight-hour CO exposures above the 9-ppm level.
Several field studies also have been conducted by the U.S. Environmental Protection
Agency to determine the feasibility and effectiveness of monitoring selected micro-
30 environments for use in estimating exposure profiles indirectly. One study (Flachsbart et al.,
March 12, 1990 8-7 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 8-1. CARBON MONOXIDE CONCENTRATIONS IN IN-TRANSIT
MICROENVIRONMENTS - DENVER, COLORADO
(Listed in descending order of mean CO concentration)
10
15
20
Microenvironment
Motorcycle
Bus
Car
Truck
Walking
Bicycling
n
22
76
3,632
405
619
9
Mean'
(ppm)
9.79
8.52
8.10
7.03
3.88
1.34
Standard
Error
(ppm)
1.74
0.81
0.16
0.49
0.27
1.20
"An observation was recorded whenever a person changed a microenvironment, and on every clock hour; thus,
each observation had an averaging time of 60 min or less.
25
Source: Johnson (1984).
30 1987) conducted in Washington in 1982-1983 concentrated on the commuting micro-
environment, because earlier studies identified this microenvironment type as the single most
important nonoccupational microenvironment relative to total CO population exposure. It was
observed that for the typical automobile commuter the time-weighted average CO exposure
while commuting ranged from 9 to 14 ppm. The corresponding rush-hour (7 to 9 am, 4 to
35 6 pm) averages at fixed-site monitors were 2.7 to 3.1 ppm.
8.2.2 Carbon Monoxide Exposures Indoors
The majority of people in the United States spend a majority of their time indoors;
therefore, a comprehensive depiction of exposure to CO must include this setting. The indoor
40 sources, emissions, and concentrations are sufficiently diverse, however, that only a few
example studies can be cited here; a thorough discussion of CO in homes, offices, and similar
environments, is presented in Chapter 7. Although a number of these studies report on
March 12, 1990 8-8 DRAFT-DO NOT QUOTE OR CITE
-------
10
15
20
25
30
TABLE 8-2. CARBON MONOXIDE CONCENTRATIONS IN OUTDOOR
MICROENVIRONMENrS - DENVER, COLORADO
(Listed in descending order of mean CO concentration)
Microenvironment
Public garages
Residential garages
or carports
Service stations or
vehicle repair facilities
Parking lots
Other locations
School grounds
Residential grounds
Sports arenas,
amphitheaters
Parks, golf courses
n
29
22
12
61
126
16
74
29
21
Mean*
(ppm)
8.20
7.53
3.68
3.45
3.17
1.99
1.36
0.97
0.69
Standard
Error
(ppm)
0.99
1.90
1.10
0.54
0.49
0.85
0.26
0.52
0.24
"An observation was recorded whenever a person changed a microenvironment, and on every clock hour; thus,
each observation had an averaging time of 60 min or less.
35 Source: Johnson (1984).
microenvironmental concentrations, they do not specifically address human exposure while
40 indoors.
Early studies date back to before 1970 when it was found that indoor and outdoor levels
do not necessarily agree. For example, one study determined indoor-outdoor relationships for
CO over two-week periods during summer, winter, and fall in 1969-1970 in buildings in
Hartford, CT (Yocom et al., 1971). With the exceptions of the private homes, which were
45 essentially equal, there was a day to night effect in the fall and winter seasons; days were
March 12, 1990 8-9 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 8-3. CARBON MONOXIDE CONCENTRATIONS IN INDOOR
MICROENVIRONMENTS - DENVER, COLORADO
(Listed in descending order of mean CO concentration)
10
15
20
25
30
35
40
Microenvironment
Public garages
Service stations or
vehicle repair facilities
Other locations
Other repair shops
Shopping malls
Residential garages
Restaurants
Offices
Auditoriums, sports arenas,
concert halls
Stores
Health care facilities
Other public buildings
Manufacturing facilities
Homes
Schools
Churches
n
116
125
427
55
58
66
524
2,287
100
734
351
115
42
21,543
426
179
Mean"
(ppm)
13.46
9.17
7.40
5.64
4.90
4.35
3.71
3.59
3.37
3.23
2.22
2.15
2.04
2.04
1.64
1.56
Standard
Error
(ppm)
1.68
0.83
0.87
1.03
0.85
0.87
0.19
0.002
0.48
0.21
0.23
0.30
0.39
0.02
0.13
0.25
"An observation was recorded whenever a person changed a microenvironment, and on every clock hour; thus,
45 each observation had an averaging time of 60 min or less.
Source: Johnson (1984).
March 12, 1990 8-10 DRAFT-DO NOT QUOTE OR CITE
-------
higher by about a factor of 2. These differences are consistent with higher traffic-related CO
levels outdoors in the daytime.
Indoor and outdoor CO concentrations were measured in four homes also in the
Hartford, CT area in 1973-1974 (Wade et al., 1975). All used gas-fired cooking stoves.
5 Concentrations were measured in the kitchen, living room, and bedroom. Stove use, as
determined by activity diaries, correlated directly with CO concentrations. Peak CO
concentrations in several of the kitchens exceeded 9 ppm, but average concentrations ranged
from 2 to 3 ppm to about 8 ppm. These results are in general agreement with results
obtained in Boston, MA (Moschandreas and Zabransky, 1982). In this study, they found
10 significant differences between rooms in homes where there were gas appliances.
Effects of portable kerosene-fired space heaters on indoor air quality were measured in
an environmental chamber and a house (Traynor et al., 1982). CO emissions from white
flame (WF) and blue flame (BF) heaters were compared. The WF convective heater emitted
less CO than the BF radiant heater. Concentrations in the residence were < 2 ppm and 2 to
15 7 ppm, respectively. The authors conclude that high levels might occur when kerosene
heaters are used in small spaces and/or when air exchange rates are low.
A rapid method using an electrochemical PEM to survey CO was applied in nine high-
rise buildings in the San Francisco and Los Angeles areas during 1980 and 1984 (Flachsbart
and Ott, 1986). One building had exceptionally high CO levels compared to the other
20 buildings; average concentrations on various floors ranged from 5 to 36 ppm. The highest
levels were in the underground parking garage, which was found to be the source of elevated
CO within the building.
The effect of residential wood combustion and specific heater type on indoor CO has
been investigated (Humphreys et al., 1986). Airtight and non-airtight heaters were compared
25 in a research home in Tennessee. CO emissions from the non-airtight heaters was generally
higher than from airtight heaters. Peak indoor CO concentration (ranging from 1.3 to
29.6 ppm, depending on heater type) was related to fuel reloadings.
CO levels in 254 Netherland homes with unvented gas-fired water heaters were
investigated during the winter of 1980 (Brunekreef et al., 1982). Concentrations (ppm) at
30 breathing height fell into the following categories: < 10 (n = 154), 11-50 (n=50), 51-100
March 12, 1990 8-11 DRAFT-DO NOT QUOTE OR CITE
-------
(n=25), > 100 (n= 17). They found that a heater vent reduced indoor CO concentrations,
and the type of burner affected CO levels.
Air pollution in Dutch homes was investigated by Lebret (1985). CO concentrations
(ppm) in various locations were kitchen, 0 to 17.5; living room, 0 to 8.7; and bedroom, 0 to
5 3.5. CO levels were elevated in homes with gas cookers and unvented geysers (water
heaters). Kitchen CO levels were higher than those in other locations due to peaks from the
use of gas appliances. Living room CO values were slightly higher in houses with smokers.
The overall mean CO level indoors was 0 to 2.7 ppm above outdoor levels.
In Zagreb, Yugoslavia, CO was measured in eight urban institutions housing sensitive
10 populations, including kindergartens, a children's hospital, and homes for the elderly (Sisovic
and Fugas, 1985). Winter CO concentrations ranged from 1.1 to 13.7 ppm, and summer
concentrations ranged from 0.6 to 6.9 ppm. The authors attributed indoor CO concentrations
to nearby traffic density, general urban pollution, seasonal differences, and day-to-day
weather conditions. Indoor sources were not reported.
15 Toxic levels of CO also were found in measurements at six ice skating rinks (Johnson
et al., 1975b). This study was prompted by the reporting of symptoms of headache and
nausea among 15 children who patronized one of the rinks. CO concentrations were found to
be as high as 304 ppm during operation of a propane-powered, ice-resurfacing machine.
Depending on skating activity levels, the ice-resurfacing operation was performed for 10 min
20 every one to two hours. As this machine was found to be the main source of CO, use of
catalytic converters and properly tuning the engine greatly reduced emissions of CO. Similar
findings have been reported by Spengler et al. (1978).
8.2.3 Carbon Monoxide Exposures Inside Vehicles
25 Studies of CO concentrations inside automobiles also have been reported over the past
decade. Petersen and Sabersky (1975) measured pollutants inside an automobile under typical
driving conditions. CO concentrations were generally less than 25 ppm, with one three-
minute peak of 45 ppm. Average concentrations inside the vehicle were similar to those
outside. No in-vehicle CO sources were noted.
30 Drowsiness, headache, and nausea were reported by eight children who had ridden in
school buses for about two hours while traveling on a ski trip (Johnson et al., 1975a). The
March 12, 1990 8-12 DRAFT-DO NOT QUOTE OR CITE
-------
students reporting symptoms were seated in the rear of the bus, which had a rear-mounted
engine and a leaky exhaust. The exhaust system subsequently was repaired. During a later
ski outing for students, CO concentrations also were monitored for a group of 66 school buses
in the parking lot. The investigators found five buses with CO concentrations of 5 to 25 ppm
5 (mean 15 ppm), 24 buses showing concentrations in excess of 9 ppm for short periods, and
two buses showing up to 3 times the 9-ppm level for short periods. Drivers were advised to
park so that exhausts from one bus would not be adjacent to the fresh air intake for another
bus.
During a cross-country trip in the spring of 1977, Chaney (1978) measured in-vehicle
10 CO concentrations. The CO levels varied depending on traffic speed. On expressways in
Chicago, San Diego, and Los Angeles when traffic speed was less than 10 mph, CO exceeded
15 ppm. Levels increased to 45 ppm when traffic stopped. In addition, it was observed that
heavily loaded vehicles (e.g., trucks) produced high CO concentrations inside nearby
vehicles, especially when the trucks were ascending a grade.
15 Colwill and Hickman (1980) measured CO concentrations in 11 new cars as they were
driven on a heavily trafficked route in and around London. The inside mean CO level for the
11 cars was 25.2 ppm vs. 47.0 ppm for the outside mean.
In a study mandated by Congress in the 1977 Clean Air Act Amendments, the EPA
studied CO intrusion into vehicles (Ziskind et al., 1981). The objective was to determine
20 whether CO was leaking into the passenger compartments of school buses, police cars, and
taxis, and, if so, how prevalent the situation was. The study involved 1164 vehicles in
Boston, and Denver. All vehicles were in use in a working fleet at the time of testing. The
results indicated that all three types of vehicles often have multiple (an average of 4 to 5)
points of CO intrusion - worn gaskets, accelerator pedals, rust spots in the trunk, and such.
25 In 58% of the rides lasting longer than eight hours, CO levels exceeded 9 ppm. Thus the
study provided evidence that maintenance and possibly design of vehicles may be an
important factor in human exposure to CO.
Flachsbart (1989) investigated the effectiveness of priority lanes on a Honolulu arterial
in reducing commuter travel time and exposure to CO. The CO concentrations and exposure
30 of commuters in these lanes was substantially lower than in the nonpriority lanes. CO
exposure was reduced approximately 61% for express buses, 28% for high-occupancy
March 12, 1990 8-13 DRAFT-DO NOT QUOTE OR CITE
-------
vehicles, and 18% for carpools when compared to that for regular automobiles. The higher
speed associated with priority lanes helped reduce CO exposure. These observations
demonstrate that CO concentrations have a high degree of spatial variability on roadways.
Additional findings on CO levels inside vehicles are summarized in a literature review
5 by Flachsbart and Ah Yo (1989). In general, a wide variation in CO exposures has been
observed in in-transit microenvironments.
8.2.4 Carbon Monoxide Exposures Outdoors
Carbon monoxide concentrations in outdoor settings (besides those measured at fixed
10 monitoring stations) also show considerable variability, as is evident from the eight Denver
microenvironmental groupings listed in Table 8-2. Ott (1971) made 1128 CO measurements
at outdoor locations in San Jose at breathing height over a six-month period and compared
these results with the official fixed monitoring station data. This study included the
measurements of the outdoor CO exposures of pedestrians in downtown San Jose by requiring
15 them to carry personal monitoring pumps and bags while walking standardized routes on
congested sidewalks. If an outdoor measurement was made more than 100 m away from any
major street, its CO concentration was similar, suggesting the existence of a generalized urban
background concentration in San Jose that was spatially uniform over the city (within a 33-
km2 grid) when one is sufficiently far away from mobile sources. Because the San Jose
20 monitoring station then was located near a street with heavy traf^.c, it recorded concentrations
approximately 100% higher than this background value. In contrast, outdoor CO levels from
personal monitoring studies of downtown pedestrians were 60% above the corresponding
monitoring station values and the correlation coefficient was low (r = 0.20). By collecting
the pedestrian personal exposures over eight-hour periods, it was possible to compare the
25 levels with the NAAQS concentration level. On two of seven days for which data were
available, the pedestrian concentrations were particularly high (13 and 14.2 ppm) and were
two to three times the corresponding levels recorded at the same time (4.4 and 6.2 ppm) at
the air monitoring station (Ott and Eliassen, 1973; Ott and Mage, 1974). These results show
that concentrations to which pedestrians are exposed on downtown streets can exceed a
30 9 ppm, 8-hour average while the official air monitoring station records values significantly
less than that. It can be argued, however, that not many pedestrians spend eight hours
March 12, 1990 8-14 DRAFT-DO NOT QUOTE OR CITE
-------
outdoors walking along downtown sidewalks, and that is one of the important reasons for
including realistic human activity patterns in exposure assessments, as indicated in
Section 8.3.
Godin et al. (1972) conducted similar studies in downtown Toronto using 100-mL glass
5 syringes in conjunction with nondispersive spectrometry. They measured CO concentrations
along streets, inside passenger vehicles, and at a variety of other locations. Like other
investigators, they found that CO concentrations were determined by very localized
phenomena. In general, CO concentrations in traffic and along streets were much higher than
those observed at conventional fixed air monitoring stations. In a subsequent study in
10 Toronto, Wright et al. (1975) used Ecolyzers to measure four to six minute average CO
concentrations encountered by pedestrians and street workers and obtained similar results.
Levels ranged from 10 to 50 ppm, varying with wind speed and direction, atmospheric
stability, traffic density, and height of buildings. He also measured CO concentrations on the
sidewalks of a street that subsequently was closed to traffic to become a pedestrian mall.
15 Before the street was closed, the average concentrations at two intersections were 9.4 +
4.0 ppm (st. dev.) and 7.9 ± 1.9 ppm; after the street was closed, the averages dropped to
3.7 ± 0.5 ppm and 4.0 + 1.0 ppm, respectively, which were equivalent to the background
level.
A large-scale field investigation was undertaken of CO concentrations in indoor and
20 outdoor locations in five California cities using personal monitors (Ott and Flachsbart, 1982).
For outdoor commercial settings, the average CO concentration was 4 ppm. This CO level
was statistically, but not substantially, greater than the average CO concentration of 1.98 ppm
recorded simultaneously at nearby fixed-monitoring stations. The final report of this field
study (Flachsbart and Ott, 1984) contains an extensive literature review of CO exposures
25 found in indoor, outdoor, and in-transit microenvironments.
8.3 ESTIMATING POPULATION EXPOSURE TO CARBON
MONOXIDE
30 Accurate estimates of human exposure to CO are a prerequisite for a realistic appraisal
of both the risks posed by the pollutant and the design and implementation of effective control
March 12, 1990 8-15 DRAFT-DO NOT QUOTE OR CITE
-------
strategies. This section discusses the general concepts on which exposure assessment is
based, the limitations of using ambient fixed-site monitoring data alone for estimating
exposure, alternative approaches which have been proposed for estimating population
exposure to air pollution, and specific applications of these approaches to estimating CO.
5 Because of problems in estimating population exposure solely from fixed station data, several
formal human exposure models have been developed. Some of these models include
information on human activity patterns: the microenvironments people visit and the times
they spend there. These models also contain submodels depicting the sources and
concentrations likely to be found in each microenvironment, including indoor, outdoor, and
10 in-transit settings.
8.3.1 Defining Concentration, Exposure, and Dose
In evaluating models for estimating CO, it is important to understand the basic concepts
of concentration, exposure, and dose. Sexton and Ryan (1988) provide the following
15 definitions.
The "concentration" of a specific air pollutant is the amount of that material per unit
volume of air. Air pollution monitors measure pollutant concentrations, which may or may
not provide accurate exposure estimates.
The term "exposure" is defined as any contact between air contaminant and the outer
20 (e.g., skin) or inner (e.g., respiratory tract epithelium) surface of the human body. Exposure
implies the simultaneous occurrence of two events (Ott, 1982):
(1) A pollutant concentration, C, is present at location x,y,z at time t.
25 (2) A person, i, is present at location x,y,z at time t.
A key distinction is apparent between a concentration and an exposure. The
30 concentration of an airborne contaminant measured in an empty room is just that, a
concentration. A concentration measured in a room with people present is a measurement of
exposure. A measured concentration is a surrogate for exposure only to the degree to which
it represents concentrations actually experienced by individuals.
March 12, 1990 8-16 DRAFT-DO NOT QUOTE OR CITE
-------
A more important distinction exists between "exposure" and "dose." Whereas exposure
is the pollutant concentration at the point of contact between the body and the external
environment, dose is defined as the amount of pollutant that actually crosses one of the body's
boundaries and reaches the target tissue. Among the factors that affect the magnitude of the
5 dose received are respiration rate, respiration mode (e.g., mouth breathing versus nose
breathing), uptake, metabolism, and clearance.
8.3.2 Components of Exposure
Two aspects of exposure bear directly on the related health consequences.
10
(1) Magnitude: What is the pollutant concentration?
(2) Duration: How long does the exposure last?
15
The magnitude is an important exposure parameter, since concentration typically is
assumed to be directly proportional to dose, and ultimately, to the health outcome. But
exposure implies a time component, and it is essential to specify the duration of an exposure.
20 The health risks of exposure to a specific concentration for five minutes are likely to be
different, all other factors being equal, than exposure to the same concentration for an hour.
The magnitude and duration of exposure can be determined by plotting an individual's
air pollution exposure over time (Figure 8-2). The function Q(t) describes the air pollutant
concentration to which an individual is exposed at any point in time t. The shaded area under
25 the graph represents the accumulation of instantaneous exposures over some period of time (tr
O. This area also is equal to the integral of the air pollutant concentration function, C;(t),
between to and t,. Ott (1982) defines the quantity represented by this area as the integrated
exposure.
By dividing the integrated exposure by the period of integration (t,-to), the average
30 exposure represents the average air pollutant concentration that an individual was exposed to
over the defined period of exposure. To facilitate comparison with established air quality
standards, an averaging period is chosen to equal the averaging period of the standard (t9). In
this case, the average exposure is referred to as a standardized exposure.
March 12, 1990 8-17 DRAFT-DO NOT QUOTE OR CITE
-------
o
z
<
cr
UJ
u
o
u
TWECO
Figure 8-2. Typical individual exposure as a function of time.
Source: Ott (1982).
March 12, 1990
8-18 DRAFT-DO NOT QUOTE OR CITE
-------
As previously discussed, exposure represents the joint occurrence of an individual being
located at point (x,y,z) during time t, with the simultaneous presence of an air pollutant at
concentration C^t). Consequently, an individual's exposure to an air pollutant is a function
of location as well as time. If a volume at a location can be defined such that air pollutant
5 concentrations within it are homogeneous yet potentially different from other locations, the
volume may be considered a "microenvironment" (Duan, 1982). Microenvironments may be
aggregated by location (i.e., indoor or outdoor) or activity performed at a location (i.e.,
residential, commercial) to form microenvironment types.
It is important to distinguish between individual exposures and population exposures.
10 Sexton and Ryan (1988) define the pollutant concentrations experienced by a specific
individual during normal daily activities as "personal" or "individual" exposures. A personal
exposure depends on the air pollutant concentrations that are present in the locations through
which the person moves, as well as on the time spent at each location. Because time-
activity patterns can vary substantially from person to person, individual exposures exhibit
15 wide variability (Dockery and Spengler, 1981; Quackenboss et al., 1982; Sexton et al., 1984;
Spengler et al., 1985; Stock et al., 1985; Wallace et al., 1985). Thus, although it
is a relatively straightforward procedure to measure any one person's exposure, many such
measurements may be needed to quantify exposures for a defined group. The daily activities
of a person in time and space define his or her activity pattern. Accurate estimates of air
20 pollution exposure generally require that an exposure model account for the activity patterns
of the population of interest. The activity patterns may be determined through "time budget"
studies of the population. Studies of this type have been performed by Szalai (1972), Chapin
(1974), Robinson (1977), Michelson and Reed (1975), Johnson (1987) and Schwab et al.
(1989). The earlier studies may now be dated and were not designed to investigate human
25 exposure questions. Ongoing exposure studies have adopted the diary methods that were
developed for sociological investigations and applied them to current exposure and time
budget investigations. A few of these studies have been reported (e.g., Schwab et al., 1989;
Johnson, 1987).
From a public health perspective, it is important to determine the "population
30 exposure," which is the aggregate exposure for a specified group of people (e.g., a
community, an identified occupational cohort). Because exposures are likely to vary
March 12, 1990 8-19 DRAFT-DO NOT QUOTE OR CITE
-------
substantially between individuals, specification of the distribution of personal exposures
within a population, including the average value and the associated variance, is often the
focus of exposure assessment studies. The upper tail of the distribution, which represents
those individuals exposed to the highest concentrations, is frequently of special interest
5 because the determination of the number of individuals who experience elevated pollutant
levels can be critical for health risk assessments. This is especially true for pollutants for
which the relationship between dose and response is highly nonlinear.
8.3.3 Relationship to Fixed-Site Monitors
10 Many early attempts to estimate exposure of human population used ambient air quality
from fixed stations. An example of such an analysis can be found in the 1980 Annual Report
of the President's Council on Environmental Quality (CEQ, 1980). In this analysis, a
county's exposure to an air pollutant was estimated as the product of the number of days that
violations of the primary NAAQS were observed at county monitoring sites multiplied by the
15 county's population. Exposure was expressed in units of person-days. National exposure to
an air pollutant was estimated by the sum of all county exposures.
The methodology employed by CEQ provides a relatively crude estimate of exposure
and is limited by four assumptions.
20 (1) The exposed populations do not travel outside areas represented by fixed-site
monitors.
(2) The air pollutant concentrations measured with the network of fixed-site monitors
are representative of the concentrations breathed by the population throughout the
25 area.
(3) The air quality in any one area was only as good as that at the location that had
the worst air quality.
30 (4) There were no violations in areas of the county not monitored.
Many studies cast doubt on the validity of this assumption for CO. Reviews of these
35 studies are provided by Ott (1982) and by Spengler and Soczek (1984). Doubts over the
March 12, 1990 8-20 DRAFT-DO NOT QUOTE OR CITE
-------
10
ability of fixed-site monitors alone to accurately depict air pollutant exposures are based on
two major findings on fixed-site monitor representativeness.
(1) Indoor and in-transit concentrations of CO may be significantly different from
ambient CO concentrations.
(2) Ambient outdoor concentrations of CO that people come in contact with may vary
significantly from CO concentrations measured at fixed-site monitors.
In estimating exposure, CEQ also assumed that each person in the population spends
24 h at home. This assumption permitted the use of readily available demographic data from
the U.S. Census Bureau. Data collected 20 years ago indicate that people spend a substantial
15 portion of their time away from home. In a study of metropolitan Washington residents
during 1968, Chapin (1974) found that people spent an average 6.3 h away from home on
Sunday and 10.6 h away from home on Friday. This translates to between 26.4 and 44.3%
of the day spent away from home. More recent personal exposure and time budget studies
(e.g., Schwab et al., 1989; Johnson, 1987) also indicate that a substantial portion of time is
20 spent away from home.
Fixed-site monitors measure concentrations of pollutants in ambient air. Ambient air
has been defined by EPA in the Code of Federal Regulations (1977) as air that is "external to
buildings, to which the general public has access." But the nature of modern urban life-
styles in many countries, including the United States, indicates that people spend an average
25 of over 20 h per day indoors (Meyer, 1983). Reviews of studies on this subject by Yocom
(1982), Meyer (1983) and Spengler and Soczek (1984) show that indoor CO concentration
measurements vary significantly from simultaneous measurements in ambient air. The
difference between indoor and outdoor air quality and the amount of time people spend
indoors reinforces the conclusion that using ambient air quality measurements alone will not
30 provide accurate estimates of population exposure.
8.3.4 Alternative Approaches to Exposure Estimation
In recent years, the limitations of using fixed-site monitors alone to estimate public
exposure to air pollutants have stimulated interest in using portable monitors to measure
March 12, 1990 8-21 DRAFT-DO NOT QUOTE OR CITE
-------
personal exposure. These instruments, which were developed for CO in the late 1970s by
Energetics Science Incorporated and by General Electric, are called PEMs. Wallace and Ott
(1982) surveyed PEMs available then for CO and other air pollutants. (See Chapter 5,
Section 5.4, for a more complete description of PEMs.)
5 The availability of these monitors has facilitated use of the direct and indirect
approaches to assessing personal exposure (see Section 8.2). Whether the direct or indirect
approach is followed, the estimation of population exposure requires a "model"; that is, a
mathematical or computerized approach of some kind. Sexton and Ryan (1988) suggest that
most exposure models can be classified as one of three types: statistical, physical, or
10 physical-stochastic.
The statistical approach requires the collection of data on human exposures and the
factors thought to be determinants of exposure. These data are combined in a statistical
model, normally a regression equation or an analysis of variance (ANOVA), to investigate the
relationship between air pollution exposure (dependent variable) and the factors contributing
15 to the measured exposure (independent variables). An example of a statistical model is the
regression model developed by Johnson et al. (1986) for estimating CO exposures in Denver
based on data obtained from the Denver Personal Monitoring Study.
If the study group constitutes a representative sample, the derived statistical model may
be extrapolated to the population defined by the sampling frame. It also should be noted that
20 selection of factors thought to influence exposure has a substantial effect on the outcome of
the analysis. Spurious conclusions can be drawn, for example, from statistical models that
include parameters that are correlated with, but not causally related to, air pollution exposure.
In the physical modeling approach, the investigator makes an a priori assumption about
the underlying physical processes that determine air pollution exposure and then attempts to
25 approximate these processes through a mathematical formulation. Because the model is
chosen by the investigator, it may produce biased results because of the inadvertent inclusion
of inappropriate parameters or the improper exclusion of critical components. The NAAQS
Exposure Model (NEM) as originally applied to CO by Johnson and Paul (1983) is an
example of a physical model.
30 The physical-stochastic approach combines elements of both the physical and statistical
modeling approaches. The investigator begins by constructing a mathematical model that
March 12, 1990 8-22 DRAFT-DO NOT QUOTE OR CITE
-------
describes the physical basis for air pollution exposure. Then a random or stochastic
component that takes into account the imperfect knowledge of the physical parameters that
determine exposure is introduced into the model. The physical-stochastic approach limits the
effect of investigator-induced bias by the inclusion of the random component, and allows for
5 estimates of population distributions for air pollution exposure. Misleading results still may
be produced, however, because of poor selection of model parameters. In addition, the
required knowledge about distributional characteristics may be difficult to obtain. Examples
of models based on this approach which have been applied to CO include the simulation of
human activity and pollutant exposure (SHAPE) model (Ott, 1984; Ott et al., 1988) and
10 two NEM-derived models developed by Johnson (1988) and by Johnson and Wijnberg (1988).
Table 8-4 provides a summary of the three model types. Table 8-5 lists exposure
models which have been applied to CO by model type. These models are described in the
following sections. General reviews of the exposure modeling literature have been provided
by Repace et al. (1980), Ott (1985), Fugas (1986), Ott et al. (1986), Sexton and Ryan
15 (1988), and Pandian (1987). EPA has developed a computerized bibliographic literature
information system (BLIS) to facilitate access to literature concerned with total human
exposure monitoring. Included in the BLIS data base is an extensive bibliography on human
exposure modeling (Dellarco et al., 1988; Shackelford et al., 1988).
20 8.3.5 Statistical Models Based on Personal Monitoring Data
As discussed above, fixed-site monitoring data may not provide an accurate indication of
personal exposure within an urban population, which is a function of both geographic location
(e.g., downtown vs. suburbia) and immediate physical surroundings (e.g., indoors vs.
outdoors). Better estimates of personal exposure can be developed by equipping a large
25 number of subjects with portable monitors and activity diaries. If the subjects are
properly selected, their exposures can be extrapolated to a larger "target" population.
Two such studies were conducted during the winter of 1982-1983 in Denver and
Washington. In the Denver study, each of 454 subjects carried a PEM and completed an
activity diary for two consecutive 24-h sampling periods and provided a breath sample at the
30 end of each sampling period (Johnson, 1984). Each participant also was requested to
complete a detailed background questionnaire. The questionnaire results and approximately
March 12, 1990 8-23 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 8-4. COMPARISON OF DIFFERENT APPROACHES TO
AIR POLLUTION EXPOSURE MODELING
Parameter
Statistical
Physical
Physical-stochastic
10
15
20
Method of
formulation
Required
input
Advantages
25 Disadvantages
Hypothesis
testing
Collected data
on human expo-
posure
Makes use of
real data in the
model building
process
30
Requires data on
hand for model
building;
extrapolation
beyond data base
is difficult
Physical
laws
Knowledge of
important parameters
and their values in
the system to be
modeled
True model
developed from
a priori consid-
erations
Includes
researcher's
biases; must
be validated
Physical Laws and
Statistics
Knowledge of
important parameters
and their distributions
in the systems to be
modeled.
Model developed
from a priori
considerations.
Stochastic part allows
uncertainty to contribute,
which reduces
importance of research
biases.
Requires much know-
ledge of system.
Must be validated.
35
40
Source: Sexton and Ryan (1988).
900 subject-days of PEM and activity diary data collected between 1 November 1982, and
28 February 1983, were analyzed to determine if factors such as microenvironment and the
presence of indoor CO sources significantly affect personal CO exposure. In addition, the
exposure of a defined target population was extrapolated from exposures recorded by the
study participants. Detailed descriptions of the Denver study design and data collection
procedures, together with results of initial data analyses, are available in a report by Johnson
(1984).
March 12, 1990
8-24
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 8-5. MODELS WHICH HAVE BEEN USED TO ESTIMATE
CO EXPOSURE BY MODEL TYPE
Model type
Model
References
10
20
25
30
35
Statistical
15 Physical
Physical/
Stochastic
Regression models based on
statistical analyses of data
obtained from Denver and Washington
Personal Monitoring Studies
Results of ANOVA of data obtained
from Washington Commuter Study
NAAQS Exposure Model (NEM)
Ott-Willits Commuter Model
Simmon-Patterson Commuter Model
Davidson Indoor Mass-Balance
Models
Pierce Integrated Exposure Model
Duan Convolution Model
Duan Hybrid Model
Flachsbart Prototypical Commuter
Models
Flachsbart-Ah Yo Commuter Model
SHAPE
Probabilistic NEM
REHEX
Johnson et al. (1986)
Flachsbart et al. (1987)
Johnson and Paul (1983)
Ott and Willits (1981)
Simmon and Patterson (1983)
Davidson et al. (1984)
Pierce et al. (1984)
Duan (1985)
Duan (1985)
Flachsbart (1985)
Flachsbart and Ah Yo (1989)
Ott (1984)
Johnson and Wijnberg (1988)
Lurmann et al. (1989)
40
Source: Adapted from Sexton and Ryan (1988).
March 12, 1990
8-25
DRAFT-DO NOT QUOTE OR CITE
-------
The Washington study has been described in detail by Hartwell et al. (1984). It differs
from the Denver study in that (1) twice as many subjects were used in the Washington study,
and (2) each subject carried a PEM and a diary for a single 24-h period. Results of analyses
of the Washington data base are provided by Settergren et al. (1984), Clayton et al. (1985),
5 and Johnson et al. (1986).
A primary goal of the Denver and Washington personal monitoring studies was to
investigate whether personal exposures could be predicted by fixed-site ambient monitoring
data. This investigation was conducted by performing linear regression analyses that used
PEM values grouped by microenvironment as the dependent variable and simultaneously-
10 recorded fixed-site values as the independent variable.
To perform these analyses, each PEM value had to be paired with a value reported by a
single fixed-site monitor. Because the census tract of each nontransit PEM value was known,
it was possible to use census tracts as a means of linking PEM and fixed-site values.
Whenever a PEM value was reported for a given census tract, it was paired with the
15 simultaneous value of the fixed-site monitor assigned to that census tract.
This analysis suggested that a linear regression analysis that pairs each PEM value
reported for a nontransit microenvironment with the simultaneous value reported at the nearest
fixed-site might be appropriate for the Denver study data. Weighted linear regression
analyses were performed with the data grouped by selected codes related to
20 microenvironment. Results for nontransit microenvironments are listed in Table 8-6. Values
of R2 range from 0.00 to 0.46. As might be expected, many of the microenvironments with
small R2 values are associated with local CO sources that tend to reduce the correlation
between PEM value and nearest fixed-site value; however, other microenvironments not
associated with local CO sources have relatively larger R2 values (e.g., park or golf course, or
25 "other locations").
Table 8-6 does not list any in-transit microenvironments because of the difficulty in
pairing in-transit PEM values with a " nearest" fixed-site monitor value. In the Denver data
base, each in-transit PEM value has two census tract listings, one associated with the start
address and the other with the end address. Neither was considered a good indicator of the
30 CO conditions encountered during the trip. An alternative procedure consisted of pairing
in-transit PEM values with simultaneous values from a composite data set created by
March 12, 1990 8-26 DRAFT-DO NOT QUOTE OR CITE
-------
rO
TABLE 8-6. RESULTS OF WEIGHTED LINEAR REGRESSION ANALYSIS WITH NONTRANSIT
PEM VALUE AS DEPENDENT VARIABLE AND SIMULTANEOUS VALUE AT NEAREST
DENVER FIXED-SITE AS INDEPENDENT VARIABLE
Microenvironment
Category
Outdoors
Outdoors
Outdoors
Indoors
Indoors
Outdoors
Outdoors
Indoors
Outdoors
Indoors
Outdoors
Indoors
Indoors
Indoors
Indoors
Indoors
Indoors
Indoors
Outdoors
Not specified
Outdoors
Indoors
Indoors
Indoors
Indoors
Subcategory
Other location
Park or golf course
School grounds
Service station or motor
vehicle repair facility
Restaurant
Service station or
motor vehicle repair
facility
Within 10 yards of road
Church
Parking lot
Other repair shop
Sports arena, amphitheater,
etc.
Other public building
Shopping mall
Store
Health care facility
Residence
School
Office
Residential garage or carport
Not specified
Residential grounds
Public garage
Auditorium, sports arena,
concert hall, etc.
Manufacturing facility
Residential garage
Other location
n
115
18
15
112
486
11
468
178
51
46
16
111
55
675
333
20,969
342
2,090
22
583
70
139
94
41
66
381
Intercept
0.35
-0.09
-0.37
4.18
1.69
1.61
1.58
0.09
2.26
3.69
3.05
0.74
1.24
1.67
0.97
1.00
0.97
2.53
5.67
2.07
0.84
8.44
2.25
1.41
4.98
7.94
Linear regression*
Slope
1.11
0.39
1.15
1.68
0.76
1.21
0.89
0.70
0.60
0.88
-1.76
0.42
1.43
0.56
0.45
0.43
0.32
0.34
0.61
0.63
0.30
0.72
0.38
0.18
0.14
0.07
R2
0.46
0.44
0.27
0.27
0.25
0.23
0.21
0.21
0.21
0.18
0.15
0.14
0.14
0.09
0.09
0.07
0.07
0.05
0.05
0.05
0.04
0.04
0.04
0.03
0.00
0.00
Pb
0.000
0.003
0.049
0.000
0.000
0.134
0.000
0.000
0.000
0.003
0.128
0.000
0.005
0.000
0.000
0.000
0.000
0.000
0.304
0.000
0.099
0.019
0.060
0.246
0.662
0.791
Listed in order of R value.
b Probability that slope = 0.
Source: Johnson et al. (1986).
-------
averaging the data from the 15 fixed-site monitors. The composite data set was found to
exhibit relatively high correlations with most of the fixed-site data sets. Consequently, the
composite site was assumed to provide an indication of the average ambient CO level in the
study area. Table 8-7 lists the results of linear regression analyses pairing in-transit PEM
5 values with simultaneous values from the composite data set. Values of R2 range from 0.04
(car) to 0.58 (motorcycle).
The linear regression analyses described above suggested that the correlation between
PEM values and fixed-site CO values is weak for most microenvironments. A statistical
analysis was subsequently performed to investigate whether the one-hour CO values reported
10 by a particular fixed-site monitor or groups of fixed-site monitors were better correlated with
PEM values. Again, the correlations were low with R2 values ranging from approximately
0.01 to 0.05 (Johnson et al., 1986).
Similar regression analyses were performed on the Washington CO data, and are shown
in Tables 8-8 and 8-9. Values of R2 range from 0.00 to 0.66. Several of the
15 microenvironments with small R2 values are associated with local CO sources that tend to
reduce the correlation between PEM value and nearest fixed-site value.
Only two nontransit microenvironments have R2 values exceeding 0.20: hospital (R2 =
0.66) and church (R2 = 0.60). The R2 value for office is 0.06; the R2 value for residence is
0.02. The in-transit microenvironments also tend to have low R2 values (e.g., the R2 value
20 for car is 0.08).
The analyses discussed above suggest that individual PEM readings are not highly
correlated with simultaneous fixed-site readings. Also it was learned that composite fixed-
site daily maximum values are poor predictors of daily maximum exposures. However, the
magnitude of daily maximum eight-hour exposures among the Denver study participants on
25 days when violations of the eight-hour NAAQS occurred (median exposure of 5.6 ppm)
versus exposures on days when violations did not occur (median exposure of 3.2 ppm) was
statistically significant at the p< 0.001.
8.3.6 Physical and Physical-Stochastic Models
30 In applying physical and physical-stochastic models, the analyst constructs a
mathematical model that describes the physical basis for air pollution exposure. As discussed
March 12, 1990 8-28 DRAFT-DO NOT QUOTE OR CITE
-------
10
TABLE 8-7. RESULTS OF WEIGHTED LINEAR REGRESSION ANALYSES WITH
IN-TRANSIT PEM VALUE AS DEPENDENT VARIABLE AND
SIMULTANEOUS VALUE FROM DENVER COMPOSITE DATA SET
AS INDEPENDENT VARIABLE
In-Transit
Subcategory
Motorcycle
Bus
Walking
Track
Car
All
Linear Regression
n
22
76
619
405
3,632
4,763
Intercept
4.50
3.17
0.06
3.27
6.01
5.15
Slope
2.14
2.02
1.47
1.54
0.78
0.92
R2
0.58
0.36
0.23
0.11
0.04
0.05
P'
0.000
0.010
0.000
0.000
0.000
0.000
15
'Probability that slope = 0.
20 Source: Johnson et al. (1986).
above, physical-stochastic models differ from physical models in that the former include a
25 random component that reflects the analyst's imperfect knowledge concerning the physical
parameters in the model.
The Convolution and Hybrid Models - Duan (1985) evaluated two methods for
estimating CO exposures which combine activity pattern data obtained from one source with
data on CO levels measured in microenvironments obtained from another source. Duan used
30 the Washington Personal Monitoring Study (Hartwell et al., 1984) as the source of activity
pattern data and the Washington Commuter Study (Flachsbart et al., 1987) as the source of
the CO data. Each of 705 subjects in the former study completed a 24-h activity diary from
which could be determined the sequence of microenvironments occupied by the subject. The
latter study measured CO levels in a variety of microenvironments on each of 43 days. In the
35 first method - referred to as the Convolution Method - each of the 43 sets of
microenvironmental CO data was paired with each of the 705 person-days of activity diary
data to yield (43) (705) = 30,315 "convoluted" person-days of CO exposure. The CO levels
for all microenvironments occupied by a subject on a given convoluted person-day are
obtained from a single day of microenvironmental monitoring data.
40 In the second method - referred to as the hybrid approach - the average CO level across
all 43 days was determined for each microenvironment and was used as the estimate of CO
March 12, 1990 8-29 DRAFT-DO NOT QUOTE OR CITE
-------
oo
u>
o
TABLE 8-8. RESULTS OF WEIGHTED LINEAR REGRESSION ANALYSES WITH NONTRANSIT PEM VALUE
AS DEPENDENT VARIABLE AND SIMULTANEOUS VALUE AT NEAREST FIXED-SITE
IN WASHINGTON, DC AS INDEPENDENT VARIABLE
Category
Indoors
Indoors
Indoors
Outdoors
Indoors
Outdoors
Indoors
Outdoors
Indoors
Indoors
Outdoors
Indoors
Indoors
Indoors
Indoors
Microenvironment
Subcategory"
Hospital
Church
Garage
Park, sports arena
Laboratories
Residential area
Office
Within 10 yards of road or street
Store
Residence
Garage, parking lot
Not specified
School, school gym
Restaurant
Other indoor
n
46
44
70
11
23
82
1,741
224
178
14,962
38
57
239
120
129
Intercept
-0.05
-0.04
4.02
0.06
0.30
0.53
0.94
1.33
1.25
1.21
5.05
3.52
1.01
2.88
5.07
Linear regression
Slope
0.63
0.58
3.43
-0.01
0.26
0.52
0.45
0.50
0.33
0.18
-0.42
-0.16
0.06
-0.03
0.09
R2
0.66
0.60
0.19
0.15
0.22
0.10
0.10
0.04
0.02
0.02
0.00
0.00
0.00
0.00
0.00
Pb
0.000
0.000
0.000
0.239
0.132
0.003
0.003
0.002
0.047
0.000
0.709
0.751
0.555
0.848
0.900
"Listed in order of R2 value.
•"Probability that slope = 0.
Source: Johnson et al. (1986).
-------
10
15
20
25
30
35
40
45
TABLE 8-9. RESULTS OF WEIGHTED LINEAR REGRESSION ANALYSES WITH
IN-TRANSIT PEM VALUE AS DEPENDENT VARIABLE AND
SIMULTANEOUS VALUE FROM COMPOSITE WASHINGTON, DC DATA SET
AS INDEPENDENT VARIABLE
In-Transit
Subcategory*
Train/subway
Jogging
Multiple response
Missing
Car
Truck
Bus
Walking
Van
Bicycle
Linear Regression
n
38
11
20
22
2,646
85
67
510
21
16
Intercept
0.05
0.43
-0.98
-0.21
1.51
2.16
1.01
1.21
1.91
3.62
Slope
1.09
0.67
2.58
1.83
1.74
2.00
2.45
0.94
0.33
-0.08
R2
0.61
0.25
0.20
0.13
0.08
0.07
0.05
0.03
0.03
0.01
P'
0.000
0.118
0.050
0.100
0.000
0.014
0.066
0.000
0.478
0.721
"Listed in order of R2 value.
""Probability that slope = 0.
Source: Johnson et al. (1986).
exposure whenever a diary-derived activity pattern indicated a subject was in the
microenvironment. This method yielded 705 person-days of CO exposure. The exposures
estimated by each of the two methods were compared to exposures indicated by the PEMs
carried by the Washington subjects. The convolution and hybrid methods produced exposure
estimates that were, on average, approximately 40% higher than the PEM-derived exposure
estimates. Despite this discrepancy, Duan (1985) found that the two methods were powerful
predictors of PEM-derived exposure estimates, in that the correlations between model
estimates and PEM-derived estimates were relatively high.
NAAQS Exposure Model (NEM) - In assessing the health risks associated with alternative
forms of NAAQS, EPA routinely uses the NEM to estimate the pollutant exposures of
March 12, 1990
8-31
DRAFT-DO NOT QUOTE OR CITE
-------
sensitive population groups. NEM itself is a general modeling framework that can be applied
to estimate the exposures of the population to individual criteria air pollutants (Biller et al.,
1981). The general NEM framework, which continues to evolve over time, can be tailored to
reflect the characteristics of particular air pollutants. NEM is designed to estimate population
5 exposures under alternative values of the NAAQS.
NEM divides time into finite intervals (e.g., 1 min, 10 min, 1 h) over which pollutant
concentrations are assumed to be constant. Geographic locations can be as small as tiny
microenvironments (e.g., homes, automobiles) or they can be aggregated into larger physical
areas (e.g., neighborhoods). The population is divided into cohorts and their activity patterns
10 (movement into successive microenvironments over time) are based on census data and
information from transporation agencies (e.g., commuter travel times). Human activities can
be represented either as deterministic or stochastic variables, as can pollutant concentrations.
In the initial application of NEM to CO (Johnson and Paul, 1983), four cities (Chicago,
Los Angeles, Philadelphia, and St. Louis) were selected as representative study areas. In
15 applications of NEM to CO, the assumption was made that the CO concentration reflects
(1) ambient CO levels as reported by outdoor fixed-site monitors and (2) sources and sinks
specific to a microenvironment. In the initial version of CO NEM, the CO exposure
associated with an event occurring at time t in microenvironment m was estimated by a first
order approximation which can be stated in general terms as:
20
CO(m,t) = MULT(m) * MON(t) + ADD(m) (8-1)
where MON(t) is the CO concentration expected to occur at a fixed-site monitor at time t,
MULT(m) is a multiplicative constant specific to m, and ADD(m) is an additive constant
25 specific to m (Johnson and Paul, 1983). This deterministic approximation does not capture
the findings of PEM studies which point to relatively low correlations between
microenvironment exposure concentrations and fixed-site monitor concentrations. Further, it
captures neither the stochastic nature of any relationship that might exist between exposure
concentrations and fixed-site monitor concentrations nor the stochastic nature of source/sink
30 contributions within a microenvironment.
March 12, 1990 8-32 DRAFT-DO NOT QUOTE OR CITE
-------
A modified version treated the term ADD(m) as an independent, identically-distributed
stochastic variable which could be characterized by the Box-Cox distribution (Johnson et al.,
1988). This change resulted in reduced levels of correlation between CO(m,t) and MON(t)
that were in agreement with correlations observed in a personal monitoring study conducted in
5 Denver, Colorado (Akland et al., 1985). A further refinement incorporates serial correlation
(Johnson and Wijnberg, 1988).
Simulation of Human Activity and Pollutant Exposure (SHAPE) - SHAPE simulates the
activity patterns and CO exposures of a sample of urban commuters during their daily
routines (Ott, 1984). The simulation is over a fixed period for all individuals in the sample,
10 usually a 24-h period. The model uses the following equation (Duan, 1981, 1982):
J
E, = Z
-------
A fundamental assumption about microenvironmental pollutant concentrations for inert
pollutants such as CO in the SHAPE model is the "superposition hypothesis." According to
this hypothesis, the total concentration Cj(t) as a function of time encountered in
microenvironment j is treated as the sum of two concentration components: (1) a
microenvironmental component concentration cm(t) resulting from the sources of CO within
the microenvironment, and, (2) an ambient (background) component concentration cu(t)
assumed to be free of any microenvironmental source influences; that is,
c,(t) = [cm(t) + c.(t)],, (8-4)
10
The basis for this hypothesis is the interpretation of the spatial variability of CO
concentrations from field studies (Ott, 1971; Ott and Eliassen, 1973).
In the SHAPE model, the microenvironmental component depends only on the sources
of CO within the microenvironment and is independent of location in the urban area or of
15 conditions in the metropolitan area. An example is the CO concentrations contributed by
motor vehicles inside an indoor parking garage. In contrast, the background concentration
component is the CO concentration that would be present if there were no specific sources of
CO. For example, in a house or building, the background component would be the CO in the
outdoor air entering through the ventilation system or the windows, which depends primarily
20 on seasonal and daily changes in meteorological conditions.
Because data on true ambient background concentrations of CO are generally unavailable
for the many microenvironments that an urban population regularly visits on a daily basis, Ott
et al. (1988) investigated the use an "overall surrogate" ambient CO concentration thought to
be associated with all the microenvironments of an urban area. Usually, the only data
25 available to serve as an overall surrogate measurement are CO concentrations measured by
fixed monitoring stations located in metropolitan areas. These data may yield unrealistically
high estimates of ambient CO levels as most air monitoring stations are placed near streets
with heavy traffic. Ott recommended using the ambient component given by the hourly CO
readings from fixed-site monitors located away from streets, but the hourly average of all
30 fixed stations in Denver performed satisfactorily.
March 12, 1990 8-34 DRAFT-DO NOT QUOTE OR CITE
-------
The original version of SHAPE (Ott, 1981, 1984) assumed that pollutant concentrations
in microenvironments behave stochastically. This assumption was based partly on a study by
Ott and Willits (1981) in which CO concentrations inside an automobile passenger
compartment were found to show considerable random fluctuation from minute to minute.
5 The CO data were collected on drives during a one-year study of an urban arterial highway,
El Camino Real in California. Statistical analysis indicated that the one-minute average CO
concentration [Cj(t)] could be treated as independent, lognormally distributed random variables
during the length of a car trip (one hour or less). Ott and Willits (1981) developed these
conclusions for the exposures incurred by the occupants of vehicles free of CO intrusion from
10 the vehicle's own exhaust system.
The SHAPE model (Ott, 1981, 1984) was designed with these findings in mind. All
microenvironmental CO component concentrations were represented by stationary two-
parameter lognormal distributions with cu(t) held constant. Thus, the computer treated the
microenvironmental component as the random variable [cjj whose mean and variance for
15 each microenvironment j where specified by the user and were held constant. The values of
the mean and variance usually were based on CO field studies in various microenvironments
reported in the literature (inside moving automobiles, buses, trucks; on bicycles in traffic; in
indoor parking garages, houses, and similar environments) and on the judgment of the user.
The SHAPE model (Ott, 1981, 1984) originally sampled microenvironmental CO
20 concentrations on a minute-by-minute basis. Fourteen microenvironments were defined for
this purpose. Associated with each was a lognormal distribution of one-minute CO values
from which one-minute CO exposures were drawn.
The original SHAPE model simulated activity patterns for each individual by sampling
from probability distributions representing the chance of entry, the time of entry, and time
25 spent in specific activities or microenvironments (Ott, 1981). For example, the probability
distributions for the starting times of home-to-work trips, trip times, and travel modes (the
proportion of commuters traveling to work by car, bus, truck, and such) were based on data
provided by Svercl and Asin (1973). Reliable data were not available for some activities; in
these cases the probability distributions were assumed by the user.
30 In an attempt to validate SHAPE, Ott et al. (1988) compared measured personal CO
exposures obtained from the Denver personal monitoring study to CO exposures estimated by
March 12, 1990 8-35 DRAFT-DO NOT QUOTE OR CITE
-------
SHAPE. Microenvironmental CO concentrations for the model were generated by Monte
Carlo simulation based on Denver PEM data reported for 22 microenvironments. The
activity simulation portions of the model were modified to accommodate actual activity data
obtained from the diaries carried by Denver subjects.
5 A total of 899 24-h responses from the Denver study yielded 772 usable profiles after
invalid responses were eliminated, giving 33 paired days of observations (CO exposure
profiles from two successive days for the same respondent). From these data,
22 microenvironments were identified with at least 10 measurements on each of the two days.
Microenvironmental CO concentrations were calculated by subtracting hourly ambient
10 background CO concentrations. Ambient background CO concentrations were estimated by
three different approaches. All three yielded similar results, with the average value from all
fixed monitoring sites performing slightly better than the nearest fixed monitoring site. For
nearly every microenvironment, the study found negligible differences between the
microenvironmental CO frequency distributions on the two days, showing the statistical
15 stability of the microenvironmental concentrations.
In the SHAPE validation project (Ott et al., 1988), the microenvironmental CO
frequency distributions for Day 1 provided the basis for SHAPE model estimates of Day 2
exposure profiles, and the activity patterns were based on the Denver diaries for Day 2 (the
observed times at which people entered and left each microenvironment). The CO exposure
20 profiles were calculated using Monte Carlo sampling from the Day 1 microenvironmental CO
concentration distributions and adding the estimated ambient background components.
The arithmetic means of the predicted one- and eight-hour maximum average CO
exposures agreed well with the corresponding observed arithmetic means. The variability of
the observed values, however, exceeded the variability of the predicted values by a significant
25 amount (Figures 8-3 and 8-4). Ott et al. (1988) suggested that the lack of agreement may be
caused by use of a histogram rather than a continuous distribution in implementing the Monte
Carlo simulation or the model's implicit assumption that the successive exposures of a subject
are uncorrelated. Ott et al. (1988) suggested that better estimates would result if an
autoregressive process was used to model successive exposures.
30 Commuter Exposure Models - Ott and Willits (1981) conducted a study in which the CO
exposures of occupants of a motor vehicle were measured by weekly drives on an urban
March 12, 1990 8-36 DRAFT-DO NOT QUOTE OR CITE
-------
2
LU
o
o
o
o
u
00' 0 OS 01 0 2 o.S 1 2
100
90
8(1
70
60
bO
40
30
20
CUMULATIVE FREQUENCY, X
10 20 30 40 SO SO 70 10 10
M 99 99.S99J99J 99.99
_ I I
1
09
08
07
06
0.5
04
0.3
02
01
I I
I I
I I
I I I
I
I III
OBSERVED
Tl _
OBSERVED:
(n-336)
PREDICTED:
Composite
(n-336)
Mean:
S.D.:
Max:
Mean:
S.D.:
Max:
111 III
PREDICTED
10.2 ppm
8.9 ppm
70.7 ppm
10.6 ppm
6.0 ppm
42.7 ppm
I
I
I
I I
J I
I
0.01 O.OS 0.1 0.2 0.5 1
10 20 30 40 SO 60 70 10 90
CUMJLATIVE FREQUENCY, %
100
99
10
70
(0
SO
40
30
20
10
9
7 —
6 §
1
0.9
OJ
0.7
0.6
O.S
0.4
0.3
0.2
0.1
91 H 99i S9J 91.} 99.99
O
Figure 8-3. Logarithmic-probability plot of cumulative frequency distribution of maximum
one-hour average exposure of CO predicted by SHAPE, plus an observed frequency
distribution for Day 2 in Denver.
Source: Ott et al. (1988).
March 12, 1990
8-37
DRAFT-DO NOT QUOTE OR CITE
-------
2
cc
t-
2
UJ
u
O
u
o
o
100
001 005 01 0.2 0.5 I 2
CUMULATIVE FREQUENCY, %
10 20 30 40 50 60 70 10 90
95
M 99 99.5 99J 99.9 99.99
90
80 -
70 -
60 —
59 —
40 —
30 -
20 —
1
09
0.8
o;
06
05
02
0 1
I
r
r i
i
OBSERVED
100
90
10
—1 70
10
58
40
Mean:
S.D.:
Max.:
PREDICTED:
(n=336)
1. Composite of fixed stations
Mean:
S.D.:
Max:
2. Nearest fixed station
Mean:
S.D.:
Max:
3. No-Source microenvironment
Mean: 3.8ppm
S.D.: 1.9ppm
Max.: 11 3 ppm
4.9 ppm
4.2 ppm
38.7 ppm
4.8 ppm
2.4 ppm
12.4 ppm
4.4 ppm
2.7 ppm
15.4 ppm
I
I
I
I
I
I
I I
OCt 0.050.102 05 1 2
10 20 30 40 50 SO 70 10 90
CUMULATIVE FREQUENCY, %
20
99 99.5 99.1 99.9 99.99
a
4 S
1
0.9
0.1
0.7
0.6
0.5
0.4
0.3
0.2
z
UJ
-------
artery in California. The study consisted of 93 repeated drives over exactly the same route -
5.9 miles in each direction for an 11.8 mile total distance - on El Camino Real with a 1974
VW test vehicle. Data were collected on CO levels inside the passenger compartment of the
motor vehicle, traffic counts and time spent waiting at each traffic light, meteorological
5 factors, as well as other variables. Measurements in the vehicle showed that the passenger
compartment was free of self-generated CO intrusion. Ott and Willits developed a theoretical
model for estimating diffusion of CO into a motor vehicle and then applied the model to data
collected during the study. The model incorporates a time constant that was found to vary
according to the position of the windows (closed, partially open, completely open).
10 Simmon and Patterson (1983) developed a model for simulating commuter exposures
individually and collectively based on traffic flow, emissions, and atmospheric dispersion.
This model consists of two programs that are run separately on the computer. The first
program is an emissions preprocessor, which has been separated from the main model to
facilitate updating of the model package when emission factors are revised by EPA. The
15 second program is the main portion of the commuter exposure model, which simulates traffic
flow, computes the emission rates resulting from the traffic (using the emission factors
calculated by the preprocessor), simulates the dispersive effects of the atmosphere, and
computes statistics describing commuter exposure. Because the model treats the spatial
variation of exposure, regions of the city in which commuters experience high exposures can
20 be identified from model output. If a single commute pathway is of interest, that pathway
can be examined in detail. Dispersion modeling is performed by the CALINE 3 model. In-
vehicle CO concentrations are assumed to equal roadway CO concentrations. To date, the
Simmon-Patterson model has not been used in a modeling analysis.
Petersen and Sabersky (1975) conducted experiments to measure the CO concentrations
25 inside a vehicle under typical driving conditions during the summer in Los Angeles, CA.
They observed that the average CO concentrations inside the vehicle were about equal to the
outside air concentrations.
Petersen and Allen (1982) conducted a similar experiment in Los Angeles over 5 days in
October 1979. They found that the average ratio of interior to exterior CO concentrations
30 was 0.92. However, the hourly average interior CO concentrations were 3.9 times higher
than the fixed-site measurements. In their analysis of the factors that influence interior CO
March 12, 1990 8-39 DRAFT-DO NOT QUOTE OR CITE
-------
levels, they observed that traffic flow, and traffic congestion (stop-and-go) are important, but
"comfort state" (i.e., car windows open/closed, fan on/off, etc.) and meteorological
parameters (i.e., wind speed, wind direction) have little influence on incremental exposures.
Flachsbart (1985) developed three empirical models for predicting commuter exposure
5 inside a well-ventilated vehicle on a congested Honolulu artery during morning rush hour
under neutral atmospheric stability. PEMs were used to collect exposure data for commuting
trips on 12 days between November 1981 and April 1982. Model A assumed that commuter
CO exposure was a function of the roadway's source strength and the ambient CO level, as
expressed in the following equation.
10
Commuter CO = (0.00012728)(CO emissions/mi)1-06 + ambient CO (8-5)
Model B assumed that the roadway's source strength was diluted by windspeed, as expressed
in the following equation.
15
Commuter CO = (0.0001972) (CO emissions/mil1-039 + ambient CO (8-6)
(windspeed)0083
20
Model C assumed that commuter CO exposure was simply a function of the emission factor
and ambient CO level. This model was developed for situations for which the analyst does
not have access to traffic counts.
25 Commuter CO = (0.0713358)(CO emission factor)1-289 + ambient CO (8-7)
Flachsbart considered these models to be prototypes because they were the first such
30 models to link commuter exposure, inside a vehicle, directly to automotive emission factors.
Each model assumed that exposure is an additive function of a background CO level and a
roadway CO contribution, as affected by meteorological and traffic characteristics.
March 12, 1990 8-40 DRAFT-DO NOT QUOTE OR CITE
-------
Flachsbart compared the observed exposure values with exposures predicted by each of
the three models. Correlations between observed and predicted values, expressed as R2, were
0.78 for Models A and B, and 0.64 for Model C.
Flachsbart's prototypical models (1985) served as model templates in subsequent efforts
5 by Flachsbart and Ah Yo (1989) to develop a general model of commuter exposure based on
data obtained from a study of commuter exposures in Washington. Their approach described
commuter exposure on a specified commuting link with an expression that superimposes a
microenvironment component upon a background concentration:
10 E, = B; + M, (8-8)
where:
E, = commuter exposure on link i,
15 B; = background concentration on link i,
M = microenvironment concentration on link i.
20 Ideally, the background concentration should be measured near the link and should
reflect concentrations that would exist on the roadway if there were no traffic. Flachsbart
(1985) approximated this value with CO ambient air quality readings from the nearest fixed-
site station away from heavy traffic. The microenvironment component mathematically
describes the air pollutant emission and dispersion processes over the roadway. This
25 component also considers how the air pollutant infiltrates the vehicle's interior. An air
pollutant infiltration factor, however, was not included in Flachsbart's prototypal models,
since there was a free exchange of air between the vehicle and the ambient environment.
Using the format of the Honolulu prototypal models, Flachsbart and Ah Yo (1989)
developed 33 commuter exposure models from the Washington survey data base. Of these
30 models, only five were considered unsatisfactory based on the statistical significance of the
model or an illogical sign for the emission coefficient. However, the explanatory power of
the best of these models (R2 = 0.12) did not approach that of the worst Honolulu model (R2
= 0.63).
March 12, 1990 8-41 DRAFT-DO NOT QUOTE OR CITE
-------
Flachsbart and Ah Yo (1989) found use of the Honolulu prototypal models for
characterizing the Washington data to be overly simplistic. For morning trips originating in
low density suburbs, the Washington data showed that a commuter's average exposure to CO
was less than the ambient concentration measured at the fixed-site station providing the
5 background concentration. In addition, commuters, who began their homeward evening trips
from highly polluted parking garages, had unusually high concentrations in their cars as they
traveled along downtown streets. Tests of each vehicle at the beginning, middle, and end of
the study indicated that the high CO levels were not caused by leaks from the exhaust system
into the passenger compartment.
10 These observations suggested that vehicle occupants were, to some degree,
"encapsulated" from the ambient environment such that their exposure on the early links of a
trip had more to do with the concentration inside the vehicle (prior to the trip) than with any
traffic or meteorological factors on these links. Statistical analysis supported this hypothesis.
For the evening commute from downtown Washington on Route 1, the average CO exposure
15 on the link was significantly correlated with pretrip interior CO concentrations. For the
morning commute into downtown Washington on Route 2, the average link exposure was well
correlated with pretrip interior CO concentrations.
Given winter temperatures and closed windows and vents on the test vehicles, Flachsbart
and Ah Yo (1989) decided to treat the roadway setting and the vehicular passenger
20 compartment as separate microenvironments. Each microenvironment was modeled separately
and then combined into a two-stage model.
The data base available for development of a roadway CO empirical model was limited
to 150 measurements of roadway CO. Of 43 different models applied to this data set, the
best model was a loglinear relationship between predicted roadway CO concentrations and the
25 density of CO emissions. This density was the product of the CO emission factor and the
average 15-min traffic count divided by the test vehicle's average link speed. This model had
an R2 value equal to 0.26; the F statistic was significant at p< 0.0001. The final step of the
regression left two independent variables in the equation: the CO emission factor and the
average 15-min traffic count.
30
March 12, 1990 8-42 DRAFT-DO NOT QUOTE OR CITE
-------
The equation for this model was:
= (0.7906252[(Fe)(QT)/Uv]036fiW2 (8-9)
where:
= predicted CO concentration within the roadway microenvironment (ppm),
Fe = MOBILE3 emission factor estimated using observed traffic speeds, ambient
temperatures, the percentages of five vehicle types, and default values for other
10 required inputs (g/veh-mi),
QT = observed average 15-min traffic count (veh/15 min),
Uv = test vehicle's average link speed.
15
Flachsbart and Ah Yo (1989) assumed that commuters are exposed to CO from three
major sources within the passenger compartment: passenger smoking, vehicle exhaust system
leaks, and emissions from traffic. Flachsbart and Ah Yo (1989) further assumed that the in-
20 vehicle CO concentrations created by these three sources can be described by box or cell
models. Such models are based on the principle of conservation of mass: The total mass of
an air pollutant within a volume is equal to the balance of the mass entered, exited, emitted,
and reacted within that volume. Using this principle, Flachsbart and Yo (1989) derived a
theoretical commuter exposure model for the passenger compartment:
25
E = COR + (T/tR)[COv - COJtl - e(lR>/T] (8-10)
where:
30 E = average CO exposure of the commuter (ppm),
COR = observed CO concentration within the roadway microenvironment (ppm),
T = time constant for the vehicle (s),
tR = time vehicle spends within the roadway microenvironment (s),
COV = CO concentration within the vehicle when it enters the microenvironment (ppm),
35 e = base of a natural logarithm (2.72).
March 12, 1990 8-43 DRAFT-DO NOT QUOTE OR CITE
-------
This model predicts commuter exposure to CO inside a vehicle by exponentially
diffusing observed roadway concentrations and by exponentially decaying initial
compartmental concentrations that exist when the vehicle enters a new link on the roadway.
The plot of the observed average CO exposure with average CO exposures estimated with
5 Equation 8-10 suggested a linear relationship. These data had an R2 = 0.75 and the
significance of the F statistic was p< 0.001.
Predicted values CO^ generated by the roadway microenvironmental model
(Equation 8-9) were substituted for the observed roadway concentrations COR in the passenger
compartment model (Equation 8-10). The values estimated by the resulting two-stage model
10 correlated well with the observed values (R = 0.737); the coefficient of determination (R2)
indicated that the estimates explained approximately 54% of the variation in observed average
exposures. Although the two-stage model did not have the predictive power of the passenger
compartment model which used observed roadway CO concentrations, Flachsbart and Ah Yo
(1989) considered the performance of the two-stage model to be respectable and far better
15 than any of the 33 models initially developed.
Other Exposure Models - Davidson et al. (1984) developed one- and two-compartment
mass-balance models for estimating indoor pollutant concentrations. They compared
measured levels of NO, NO2, and CO in a new townhouse residence with estimates provided
by the one-compartment model. The townhouse was constructed according to rigid energy-
20 conservation guidelines. Reasonable agreement between estimated and measured
concentrations was observed, although the measured CO levels decayed somewhat faster than
predicted.
Pierce et al. (1984) presented a model for estimating integrated (i.e., cumulative) and
average exposures based on an activity pattern listing a sequence of indoor and outdoor
25 locations and estimates of the pollutant concentration at each location. The model was used to
estimate CO exposures for a hypothetical 24-h activity pattern. The CO level assigned to
each location was derived from microenvironmental monitoring data obtained from other
researchers.
30
March 12, 1990 8-44 DRAFT-DO NOT QUOTE OR CITE
-------
8.4 OCCUPATIONAL EXPOSURE TO CARBON MONOXIDE
Carbon monoxide is a ubiquitous contaminant occurring in a variety of settings.
Exposures, both acute and chronic, that occur in the occupational environment represent only
one of several sources that may contribute to a potential body burden for carbon monoxide.
5 Two main sources for background exposures in both occupational and nonoccupational
settings appear to be smoking and the internal combustion engine (National Academy of
Sciences, 1969). Smoking is a personal habit that must be considered in evaluating exposures
in general, as well as those occurring in work places.
In addition, work environments are often located in densely populated areas, and such
10 areas frequently have a higher background concentration of CO compared to less densely
populated residential areas. Thus, background exposures during work hours may be greater
than during nonwork hours. There are several sources other than smoking and the internal
combustion engine that contribute to exposure during work hours. These include
contributions to background by combustion of organic materials in the geographic area of the
15 work place, work in specific industrial processes that produce CO, and work in environments
that result in accumulations of CO, such as garages, toll booths, and confined spaces.
8.4.1 Historical Perspective
Production of CO results from incomplete combustion of organic substances such as
20 natural gas, coal, wood, petroleum, coal, coke, vegetation, explosives, and manufactured gas.
A rich fuel mixture favors generation of CO. Carbon monoxide also can be produced when
rapid cooling or submersion of the flame is used to quench the combustion process.
Dangerous concentrations of carbon monoxide can occur in numerous settings, including
environmental background, the home, and the street - at work or play. Sources include
25 exhaust gases from internal combustion engines, gas manufacturing plants, blast furnaces in
iron and steel manufacturing, coke ovens, coal mines, incinerators, and numerous other
processes that involve combustion of organics. CO is used in the manufacture of metal
carbonyls, and CO is produced in industrial quantities by the partial oxidation of
hydrocarbons or natural gas, and by gasification of coal or coke (Lundgren, 1971). (See
30 Chapter 6 for a more complete discussion of sources and emissions of CO.)
March 12, 1990 8-45 DRAFT-DO NOT QUOTE OR CITE
-------
Both chronic and acute CO intoxication in a variety of occupations and processes is
discussed by Grut (1949). Acute effects related to production of anoxia from exposures to
CO historically have been a basis for concern. In recent years, however, this concern has
grown to include concerns for potential effects from chronic exposures as well (Rosenstock
5 and Cullen, 1986a, 1986b; Sammons and Coleman, 1974).
With regard to the occupational environment, the National Institute for Occupational
Safety and Health (1972) published "Criteria for a Recommended Standard...Occupational
Exposure to Carbon Monoxide." NIOSH observed that"... the potential for exposure to
carbon monoxide for employees in the work place is greater than for any other chemical or
10 physical agent." NIOSH recommended that exposure to CO be limited to a concentration no
greater than 35 ppm, expressed as TWA for a normal eight-hour workday, 40 hours per
week. A ceiling concentration was also recommended at a limit of 200 ppm, not to exceed an
exposure time greater than 30 min. Occupational exposures at the proposed concentrations
and conditions underlying the basis of the standard were considered to maintain COHb in
15 blood below 5 %. The Occupational Safety and Health Administration has recently adopted
these exposure limits in order to substantially reduce the risk of deleterious health effects
among American workers (Federal Register, 1989).
Although it was not stated, the basis of the recommended NIOSH standard (i.e.,
maintaining COHb below 5% in blood), assumes that contributions from other
20 nonoccupational sources would also be less than a TWA concentration of 35 ppm. It was
recognized that such a standard may not provide the same degree of protection to smokers,
for example. Although recognizing that biologic changes might occur at the low level of
exposure recommended in the proposed standard, NIOSH concluded that subtle aberrations in
the nervous system with exposures producing COHb concentrations in blood at or below 5%
25 did not demonstrate significant impairments which would cause concern for the health and
safety of workers. In addition, NIOSH observed that individuals with impairments that
interfere with normal O2 delivery to tissues (e.g., emphysema, anemia, coronary heart
disease) may not have the same degree of protection as for less impaired individuals. It also
was recognized that work at higher altitudes (e.g., 5000 to 8000 feet above sea level) would
30 necessitate decreasing the exposure limit below 35 ppm, to compensate for a decrease in the
oxygen partial pressure as a result of high altitude environments and a corresponding decrease
March 12, 1990 8-46 DRAFT-DO NOT QUOTE OR CITE
-------
in oxygenation of the blood. High altitude environments of concern include airline cabins at a
pressure altitude of 5000 feet or greater (National Research Council, 1986) or work in high
mountain tunnels (Miranda et al., 1967).
5 8.4.2 Exposure Monitoring Techniques
Exposures to CO in air can be manifested in a variety of ways. At low levels,
manifestations include development and reporting of symptoms. In the work place,
environmental monitoring and inventory of sources for the presence of CO may occur.
Additionally, biologic tests, medical surveillance, diagnosis and treatment may be conducted
10 on individuals who show signs and symptoms of exposure. Finally, mathematical models
may be used to predict exposures, doses, and responses to CO inhalation.
Acute and chronic CO intoxication (Grut, 1949) may be indicated by a range of signs
and symptoms from headache, dizziness, weakness, and nausea at low levels and short
durations of exposure to unconsciousness, coma, and death at high levels and durations of
15 exposure. Headache and nausea resulting from CO intoxication has been described in a study
of tollbooth collectors (Johnson et al., 1974) exposed at low concentrations of CO from
exhaust gases.
A medical study of the occupational hazards of fire fighting demonstrates the signs and
symptoms of CO, as well as other associated exposures (Gordon and Rogers, 1969). A group
20 of 35 fire fighters were evaluated in a medical study for heart, lung, liver, and kidney
diseases, and were also provided neurologic examinations. Half of the study group were
smokers. Baseline tests including enzyme tests, EKG, COHb, and other measurements were
conducted at the start of the study. The fire fighters were in normal ranges for COHb and the
enzyme tests performed. They were followed through 31 fires of less than five minutes,
25 4 fires of more than five minutes, and 6 staged fires; they were also subjected to exercise
tests. Occasionally, there were substantial exposures to CO, and changes in blood enzyme
levels were greater when fighting longer fires. These changes were not associated with
exercise, and they were reversible when not fighting fires. EKG tests did not reflect changes
related to enzyme levels. Masks were found to provide substantial protection.
30 Occupational exposure and associated signs and symptoms for fire fighters also have
been described in a study using age-matched controls (Sammons and Coleman, 1974). Blood
March 12, 1990 8-47 DRAFT-DO NOT QUOTE OR CITE
-------
samples were collected from a group of 27 fire fighters and a group of 27 control subjects
every 28 days for five months. Differences between the cardiac enzyme levels found in fire
fighters versus those of the matched control suggested that chronic low level exposures to CO
have a deleterious effect on the body and myocardium.
5 Environmental monitoring for CO is often carried out in studies that are primarily
concerned with potential exposures to other substances, such as exhaust gases, environmental
tobacco smoke and combustion processes. CO analyses also are used to screen for the
presence of other gaseous pollutants. Exposures to CO therefore are often associated with
exposures to other substances as well, including lead, paniculate matter containing
10 polyaromatic hydrocarbons (PAHs), NOs, and SO2.
Monitoring for exposures to CO has included peak and TWA sampling of ambient or
breathing zone air, collection and analysis of expired air, analysis of blood gases by gas
chromatographic methods, and use of empirical relationships to estimate CO in air from
determinations of percent COHb in blood. Measurements techniques for CO include infrared,
15 volumetric, colorimetric tubes, electrolytic detection, and gas chromatographic methods.
Samples are collected to represent the breathing zone or environmental air; these may be grab
samples or periodic or continuous samples.
Several investigators have proposed approaches to medical surveillance of workers who
are potentially exposed to carbon monoxide. Medical surveillance activity is usually
20 precipitated by complaints that are associated with a source of potential exposure to CO. A
recent study of stevedores who loaded and unloaded cars and diesel trucks in a ferrying
operation (Purdham et al., 1987), assessed medical conditions by administering a
questionnaire and conducting pulmonary function tests. The questionnaire included questions
on work history; smoking history; respiratory symptoms; and nose, eye, and skin complaints.
25 Questions on respiratory symptoms included details on cough, sputum, wheeze, chest
tightness, and shortness of breath. Pulmonary function tests were conducted for forced vital
capacity (FVC) and forced expiratory volume at one minute (FEV,). The subjects were
seated and their noses were clipped closed for the tests. A minimum of three and as many as
six efforts were required for each subject.
30 The focus of the study was on characterizing adverse responses to exhaust fumes,
primarily from diesel trucks, and secondarily from gasoline-powered vehicles transported in
March 12, 1990 8-48 DRAFT-DO NOT QUOTE OR CITE
-------
garages on the ferry. The medical findings were that the stevedores had significantly lower
values for all lung function tests except for FVC, as compared to unexposed controls, and to
normal values for a general population. Environmental sampling for CO and other substances
in exhaust fumes then were conducted. There was no direct correlation offered by the authors
5 between exposure to CO and the differences found in the medical assessments of the
stevedores versus the controls. The authors suggested use of percent COHb to assess CO
exposure, and they recommended that a larger group of longshoreman should be assessed for
chronic obstructive pulmonary disease.
Rosenstock and Cullen (1986a) have linked cardiovascular diseases occasioned by angina
10 at the end of a workday with high percent COHb when this phenomenon is associated with
exposure to CO in the work place. Consequences of chronic low level exposure are not well
established; however, in workers with underlying coronary artery heart disease, a level of
3 to 5% COHb has been associated with increasing frequency of angina and decrease in
exercise tolerance (See Chapter 10). When levels approaching 25% COHb are reached, there
15 are manifestations of ischemia, dysrhythmias and EKG abnormalities in otherwise healthy
workers.
Miranda et al. (1967) feel that medical surveillance for high altitude work should
include screening for cardiopulmonary abnormalities and blood dyscrasias (sickle-cell
anemia). They also recommend acclimatization before the start of work.
20 This study classified the onset of CO intoxication into three groups: fulminating (a
decrease in O2 to tissues within seconds), acute (a decrease in O2 occurring in minutes), and
chronic (a decrease in O2 oxygen to tissues over days, months, or years).
Miranda et al. (1967) listed the concerns for evaluation of CO exposures at high
altitudes as: decreased oxygen in the air, percent COHb due to smoking, and accumulation of
25 fumes, particularly in vehicular tunnels. Altitude tolerance is lowered by about 335 ft for
each percentage point increase in COHb. The average percent COHb for smokers who smoke
20 to 30 cigarettes per day is 5%, with a range of 3 to 10%. To decrease from 20 to
5% COHb requires breathing fresh air at sea level for three to five hours. Inhalation of CO
at a concentration of 100 ppm for two hours at 11,000 feet results in 18 ± 5% COHb. This
30 level does not threaten survival, but it may impair visual threshold (see Chapter 10).
March 12, 1990 8-49 DRAFT-DO NOT QUOTE OR CITE
-------
Empirical relationships have been proposed for use as diagnostic criteria for CO
intoxication (Castellino, 1984). The criteria proposed are shown in Table 8-10.
Blankart et al. (1986) found hyperbaric O2 to be the best form of treatment for
decreasing the percent COHb in blood when it was administered to traffic policemen in a
5 clinical study. The study compared cycling ergometry, administration of pure O2 at
atmospheric pressure, and administration of pure O2 under hyperbaric conditions (2.8 atm).
The authors recommended the hyperbaric treatment approach for both acute and chronic CO
poisoning.
A study of toll bridge authority workers investigated normal red cell adaptation to
10 anemia as a measure of CO effects on tollbooth collectors and maintenance personnel
(Goldstein et al., 1975). Diphosphoglycerides (DPG) increase the release of oxygen to tissues
as an adaptive mechanism in anemia. Results of the studies were inconclusive, in that they
considered increased DPG to be a response to hypoxia from increased percent COHb.
However, formation of methoxyhemoglobin from exposures to NO was independent of the
15 COHb reaction, and the hypoxic effects of CO and NO exposures were considered to be
additive.
Use of percent COHb in blood has been proposed for use as a biological exposure index
(BEI), as a supplement to the threshold limit value (TLV) value for CO exposure
recommended by the American Conference of Governmental Industrial Hygienists (ACGIH)
20 (Lowry, 1986). The proposed BEI is intended to be an index of exposure. It is not
necessarily an indication of an adverse response.
Finally, Hickey et al. (1975) expressed the need to consider genetic and other factors
resulting in differences in hemoglobin and other individual characteristics that could influence
the extent of COHb formation on exposures to CO in air or cigarette smoke.
25
8.4.3 Occupational Exposures
The number of persons potentially exposed to CO in the work environment is greater
than that for any other physical or chemical agent (Hosey, 1970), with estimates as high as
975,000 occupationally exposed at high levels (National Institute for Occupational Safety and
30 Health, 1972).
March 12, 1990 8-50 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 8-10. DIAGNOSTIC CRITERIA FOR CO INTOXICATION
Normal Abnormal
5
Nonsmokers Smokers Increased Surveillance
COHb <3% COHb <8% COHb 8 to 12%
10 SCN/blood <40mg/L SCN/blood <200 mg/L
Increased Risk
COHb 12 to 15%
15 Medical Treatment
COHb > 15%
Source: Castellino (1984).
20
The contribution of occupational exposures can be separated from other sources of
exposure, but there are at least two conditions to consider.
25-
(1) When CO concentrations at work are higher than the CO equilibrium
concentration associated with the percent COHb at the start of the work shift,
there will be a net absorption of CO and an increase in percent COHb.
Nonsmokers will show an increase that is greater than that for smokers because
30 they start from a lower baseline COHb level. In some cases, nonsmokers may
show an increase, and smokers a decrease in percent COHb.
(2) When CO concentrations at work are lower than the equilibrium concentration
necessary to produce the worker's current level of COHb, then the percent COHb
35 will show a decrease. There will be a net loss of CO at work.
Occupational exposures can stem from two sources: (1) through background
concentrations of CO obtained by working in a densely populated area (as compared to the
40 residential environment), or (2) through work in industrial processes that produce CO as a
product or by-product. In addition, work in environments that tend to accumulate CO
concentrations may result in occupational exposures. Rosenman (1984) lists a number of
occupations where the workers may be exposed to high CO concentrations. This list includes
acetylene workers, blast furnace workers, coke oven workers, diesel engine operators, garage
March 12, 1990 8-51 DRAFT-DO NOT QUOTE OR CITE
-------
mechanics, steel workers, metal oxide reducers, miners, mond process (nickel refining)
workers, organic chemical synthesizers, petroleum refinery workers, pulp and paper workers,
and water gas workers. In addition, because methylene chloride is metabolized to CO in the
body, aerosol packagers, anesthetic makers, bitumen makers, degreasers, fat extractors,
5 flavoring makers, leather finish workers, oil processors, paint remover makers, resin makers,
solvent workers, and stain removers also can have high COHb levels.
Background sources are generally a result of combustion of organic materials. With rich
fuel mixtures, decreased amounts of O2 are available, and therefore production of CO as a
product of incomplete combustion is favored. There are numerous sources for CO
10 background exposures, and there is considerable variation and uncertainty in identifying the
CO exposure resulting from specific sources. Traffic patterns and emissions from mobile
sources, as well as an overlay of emissions from stationary sources along with wind and
weather conditions, make predictions difficult. (See Chapters 6 and 7 for a complete
discussion of the mobile, stationary, and indoor sources and emissions of CO.)
15 Investigations and analyses of exhaust gas in Paris (Chovin, 1967) showed that CO air
concentrations were correlated with the activity of and distribution pattern for traffic in Paris.
The average CO concentrations for the years 1965 and 1966 were 16.0 and 16.6 ppm, based
on 15,187 samples for 1965 and 15,203 samples for 1966, respectively. The maps of
pollutant distribution indicate that the areas of high and low concentrations were similar for
20 each year and were closely associated with the volumes and patterns of vehicular traffic.
When the measured concentration of CO in the air exceeded 100 ppm, the sample was
automatically diluted 10-fold for analysis, thereby introducing dilution and scale factors as
possible sources of error at high concentrations of CO. The variations in the measurements of
CO were closely linked to variations in the volume of traffic at each sampling location.
25 Carbon Monoxide concentrations in the blood were determined for 331 traffic policemen
during 5 hours of duty. Blood samples were collected at the beginning and end of the five-
hour shifts. Carbon monoxide in blood was determined by heating samples and measuring the
evolved gas by an infrared method. The values obtained were compared with the average
concentrations of CO found for the air breathed. The correlation was good between CO in
30 blood and CO in the air breathed for nonsmokers, but with smokers, initial concentrations of
CO in blood were high, and there was often a decrease in the CO in blood over an exposure
March 12, 1990 8-52 DRAFT-DO NOT QUOTE OR CITE
-------
period. This was observed with sampling of smokers and nonsmokers at the same locations
with similar concentrations and durations of exposure. Car drivers showed increases in CO
concentrations in blood, as did traffic policemen who were nonsmokers. Carbon Monoxide
concentrations in blood for smoking and nonsmoking drivers involved in traffic accidents
5 were greater than those for traffic policemen and others in the population considered to be
accidentally exposed to CO.
Aircraft accidents involving 113 aircraft, 184 crew members, and 207 passengers were
investigated to characterize accident toxicology and to aid in search for causation of a crash
(Blackmore, 1974). Determinations of percent COHb in blood samples obtained from victims
10 enabled differentiation of a variety of accident sequences involving fires. For example,
percent COHb determinations combined with passenger seating information and crew
assignments can allow differentiating between fire in flight or after the crash, survivability of
crash with death due to smoke inhalation, specific equipment malfunctions in equipment
operated by a particular crew member, or defects in space heating in the crew cabin or
15 passenger compartment. One accident in the series was associated with a defective space
heater in the crew compartment. Another accident also was suspicious with regard to a space
heater.
Contributions to background CO concentrations from industrial processes may be
determined by an inventory of sources and locations for the processes, as well as by emission
20 rates for CO as a function of production, and the air pollution distribution pattern for the
region (see Chapter 6). The types and distributions of industrial and community activities
contributing to CO concentrations in air depend on identification of the various sources and
volumes of production involved. Production schedules are dynamic; it is therefore difficult to
model sources and predict levels.
25 Carbon monoxide concentrations measured in the air were used to classify workers from
20 foundries into three groups: those with definite occupational exposure, those with slight
exposure, and controls (Hernberg et al., 1976). Angina pectoris, EKG findings, and blood
pressures of foundry workers were evaluated in terms of CO exposure for the 1000 workers
who had the longest occupational exposures for the 20 foundries. Angina showed a clear dose
30 response with exposure to CO either from occupational sources or from smoking, but there
was no such trend in EKG findings. The systolic and diastolic pressures of CO-exposed
March 12, 1990 8-53 DRAFT-DO NOT QUOTE OR CITE
-------
workers were higher than those for other workers, when age and smoking habits were
considered.
Carboxyhemoglobin and smoking habits were studied for a population of steel workers
and compared to blast furnace workers, as well as to employees not exposed at work (Jones
5 and Walters, 1962). Carbon monoxide is produced in coke ovens, blast furnaces, and in
sintering operations. Exhaust gases from these operations are often used for heating and as
fuels for other processes. Fifty-seven volunteers working in the blast furnace area were
studied for smoking habits, symptoms of CO exposure, and estimations of COHb levels by an
expired air technique. The main increase in COHb for blast furnace personnel was 2.0% for
10 both smokers and nonsmokers in the group. For smokers in the unexposed control group,
there was a decrease in percent COHb. A follow-up study found similar results (Butt et al.,
1974). Virtamo and Tossavainen (1976) report a study of CO measurements in air of 67
iron, steel, or copper alloy foundries. Blood COHb of ironworkers exceeded 6% in 26% of
the nonsmokers and 71 % of the smokers studied.
15 Poulton (1987) found that a medical helicopter with engine running in a narrowed or
enclosed helipad was found to be a source of potential exposure to CO, JP-4 fuel and possibly
other combustion products for flight crews, medical personnel, bystanders and patients being
evacuated. Measurements were made by means of a portable infrared analyzer. Carbon
monoxide concentrations were found to be greatest near the heated exhaust. Concentrations
20 ranged from 8 ppm to 43 ppm.
Exhaust from seven most commonly used chain saws (Nilsson et al., 1987) were
analyzed under laboratory conditions to characterize emissions. The investigators conducted
field studies on exposures of loggers using chain saws in felling operations, and also in
limbing and bucking into lengths. In response to an inquiry, 34% of the loggers responded
25 that they often experienced discomfort from the exhaust fumes of chainsaws, and another 50%
complained of occasional problems. Sampling for CO exposures was carried out for five days
during a two-week work period in a sparse pine stand at an average wind speed of 0 to
3 m/s, a temperature range of 16 to 1°C, and a snow depth of 50 to 90 cm. Carbon
monoxide concentrations ranged from 10 to 23 mg/m3 (9 to 20 ppm) with a mean value of
30 20.0 mg/m3. Carbon monoxide concentrations measured under similar but snow free
conditions ranged from 24 to 44 mg/m3 with a mean value of 34.0. In another study, CO
March 12, 1990 8-54 DRAFT-DO NOT QUOTE OR CITE
-------
exposures were monitored for nonsmoking chain saw operators with average exposures
recorded from 20 to 55 ppm with COHb levels ranging from 1.5 to 3.0% (Van Netten et al.,
1987).
Fork lift operators, stevedores and winch operators were monitored for CO in expired
5 air to calculate percent COHb, using an MSA analyzer (Breysse and Bovee, 1969). Periodic
blood samples were collected to validate the calculations. Bull operators and stevedores work
in the holds of ships; winch operators do not work in the holds. The ships to be evaluated
were selected on the basis of their use of gasoline-powered fork lifts for operations. To
evaluate seasonal variations in percent COHb, analyses were performed for one five-day
10 period per month for a full year. Efforts were made to select a variety of ships for
evaluation. A total of 689 determinations of percent COHb were made from blood samples to
compare with values from expired air samples. The samples were collected on 51 separate
days involving 26 different ships. Two hundred men were available before work, while only
147 were available at the end of the work day. Men lost to follow-up either left before the
15 end of shift or were transferred to other work. Smoking was found to be a major
contributing factor to percent COHb levels found. Carboxyhemoglobin values for
nonsmokers indicated that the use of gasoline powered lifts in the holds of the ships did not
produce a CO concentration in excess of 50 ppm for up to eight hours as a TWA under the
work rules and operating conditions in practice during the study. Smoking behavior
20 confounded exposure evaluations. The exposure conditions may not provide the same degree
of protection for smokers as they do for nonsmokers.
Carbon monoxide concentrations have been measured in a variety of work places where
potential exists for accumulation from outside sources. Exposure conditions in work places,
however, are substantially different. There is no standard approach that applies in all
25 situations requiring evaluation and study. The methods to be applied, group characteristics,
jobs being performed, smoking habits, and physical characteristics of the facilities themselves
introduce considerable variety in the approaches used. Typical studies are discussed below.
Wallace (1983) investigated CO in air and breath of employees working in an office
constructed in an underground parking garage at various times over a one-month period.
30 Carbon monoxide levels were determined by use of a device containing a proprietary solid
polymer electrolyte to detect electrons emitted in the oxidation of CO to CO2. The device
March 12, 1990 8-55 DRAFT-DO NOT QUOTE OR CITE
-------
was certified by the Mine Safety and Health Administration (MSHA) to be accurate within
15%. A data logger was attached to provide readings each second, and to provide one-hour
averages from the CO monitors placed on desks. Variation in CO measurements in ambient
air showed a strong correlation with traffic activity in the parking garage. Initially, the office
5 CO levels were found to be at an average of 18 ppm per day with the average from 12:00 to
4:00 p.m. at 22 ppm and from 4:00 to 5:00 p.m. at 36 ppm. Analyses of expired air
collected from a group of 20 nonsmokers working in the office showed a strong correlation
with ambient air concentrations for CO and traffic activity. For example, the average CO in
expired air for one series of measurements was 23.4 ppm, as compared to simultaneous
10 measurements of air concentrations of CO at 22 to 26 ppm. After a weekend, CO
concentrations in breath on Monday morning were substantially decreased (around 7 ppm) but
rose again on Monday afternoon to equal the air levels of 12 ppm. Closing fire doors, and
the use of existing garage fans decreased CO concentrations in the garage offices to 2 ppm or
less, concentrations similar to those for other offices in the complex that were located away
15 from the garage area.
Carboxyhemoglobin levels (Ramsey, 1967) were determined over a three-month period
during winter months for 38 parking garage attendants, and the values for COHb were
compared with values from a group of 27 control subjects. Blood samples were collected by
finger stick on Monday mornings at the start of the work week, at the end of the work shift
20 on Mondays and at the end of the work week on Friday afternoons. Hourly analyses were
carried out on three different weekdays using potassium palado sulfite indicator tubes for the
concentrations of CO at three of the six garages in the study. Hourly values ranged from 7 to
240 ppm, and the composite mean of the 18 daily averages was 58.9 + 24.9 ppm. While the
Monday versus Friday afternoon values for COHb were not significantly different, there were
25 significant differences between Monday morning and Monday afternoon values. Smokers
showed higher starting baseline values, but there was no apparent difference in net increase in
COHb body burden between smokers and nonsmokers. COHb values for nonsmokers ranged
from a mean of 1.5 + 0.83% for the a.m. samples to 7.3 ± 3.46% for the p.m. samples.
For smokers these values were 2.9 + 1.88% for the a.m. and 9.3 + 3.16% for the p.m.
30 The authors observed a crude correlation between daily average for CO in air and COHb
values observed for a two-day sampling period.
March 12, 1990 8-56 DRAFT-DO NOT QUOTE OR CITE
-------
In a study of motor vehicle examiners conducted by NIOSH (Stern et al., 1981) CO
levels were recorded in six outdoor motor vehicle inspection stations with TWA levels of 4 to
21 ppm. In contrast, the semi-open and enclosed stations had levels of 10 to 40 ppm TWA.
The levels exceeded the recommended NIOSH standard of 35 ppm TWA on 10% of the days
5 sampled. In addition, all stations experienced peak short-term levels above 200 ppm.
Carboxyhemoglobin levels were measured for 22 employees of an automobile dealership
during the winter months when garage doors were closed and ceiling exhaust fans were turned
off (Andrecs et al., 1979). Employees subjected to testing included garage mechanics,
secretaries, and sales personnel. This included 17 males aged 21 to 37 and five females aged
10 19 to 36. Blood samples were collected on a Monday morning before start of work, and on
Friday at the end of the work week. Analysis for COHb was by addition of sodium dithionite
and tris aminomethane, and COHb was measured in duplicate samples using a
spectrophotometer. Smokers working in the garage area showed a Monday mean value for
COHb of 4.87 + 3.64% and a Friday mean value of 12.9 + 0.83%. Nonsmokers in the
15 garage showed a corresponding increase in COHb, with a Monday mean value of
1.50 ± 1.37% and a Friday afternoon mean value of 8.71 +2.95%. Nonsmokers working
in areas other than the garage had a Friday mean value of 2.38 + 2.32%, which was
significantly lower than the mean values for smokers and nonsmokers in the garage area.
Environmental concentrations or breathing zone samples for CO were not collected. The
20 authors concluded that smokers have a higher baseline level of COHb than do nonsmokers,
but both groups show similar increases in COHb during the work week while working in the
garage area. The authors observed that the concentrations of COHb found in garage workers
were at levels reported to produce neurologic impairment. These results are consistent with
those reported by Amendola and Hanes (1984). They reported some of the highest indoor
25 levels collected at automobile service stations and dealerships. Concentrations ranged from
16.2 to 110.8 ppm on cold weather to 2.2 to 21.6 ppm in warm weather.
A group of 34 employees, 30 men and 4 women, working multi-story garages, were
evaluated for exposures to exhaust fumes (Fristedt and Akesson, 1971). Thirteen were
service employees working at street level, and 21 were shop employees working either one
30 story above or one story below street level. Six facilities were included in the study. Blood
samples were collected on a Friday at four facilities, on Thursday and Friday at another, and
March 12, 1990 8-57 DRAFT-DO NOT QUOTE OR CITE
-------
on a Thursday only at a sixth facility. The blood samples were evaluated for red blood cell
(RBC), and white blood cell (WBC) counts, COHb, lead and delta-ALA. Work histories,
medical case histories, and smoking habits were recorded. Among the employees evaluated,
11 of 24 smokers and 3 of 10 nonsmokers complained of discomfort from exhaust fumes.
5 Smokers complaining of discomfort averaged 6.6% COHb and nonsmokers complaining
averaged 2.2% COHb. The corresponding values for non-complaining workers averaged
4.2% and 1.1%, respectively.
Air pollution by CO in underground garages was investigated as part of a larger study of
traffic pollutants in Paris (Chovin, 1967). Work conducted between the hours of 8:00 a.m.
10 and 10:00 p.m. resulted in exposures in excess of 50 ppm and up to 75 ppm, on a TWA
basis.
As part of a larger study of CO concentrations and traffic patterns in Paris (Chovin,
1967), samples were taken in road tunnels. There was good correlation between traffic
volumes combined with the lengths of the tunnels and CO concentrations found. None of the
15 tunnels studied had mechanical ventilation. The average CO concentrations in the tunnels
were 27 and 30 ppm for 1965 and 1966, respectively, as compared to an average of 24 ppm
CO in the streets for both years. The "real average risk" for a man working or walking in a
street or tunnel was considered by the authors to be 3 to 4 times less than the maximal risk
indicated by values for CO from instantaneous air sample measurements. In the United
20 States, Evans et al. (1988) studied bridge and tunnel workers in metropolitan New York City.
The average COHb concentration over the 11 years of study averaged 1.73% for nonsmoking
bridge workers and 1.96% for tunnel workers.
In a discussion of factors to consider in CO control of high altitude highway tunnels,
Miranda et al. (1967) reviewed the histories of several tunnels. Motor vehicles were
25 estimated to emit about 0.1 Ib of CO/mi at sea level. At 11,000 feet and a grade of 1.64%,
emissions were estimated at 0.4 Ib/mi (for vehicles moving upgrade). Tunnels with
ventilation are generally designed to control CO concentrations at or below 100 ppm. The
Holland Tunnel in New York was reported to average 65 ppm, with a recorded maximum of
365 ppm due to a fire. For the Sumner Tunnel in Boston, ventilation is started at CO
30 concentrations of 100 ppm and additional fans are turned on with the sounding of an alarm at
250 ppm. The average value for CO concentration is 50 ppm. The Mont Blanc Tunnel is
March 12, 1990 8-58 DRAFT-DO NOT QUOTE OR CITE
-------
7.2 mi long at an average elevation of 4179 ft. This tunnel is designed to maintain CO
concentrations at or below 100 ppm. The Grand Saint Bernard tunnel is 3.5 mi long at an
average elevation of 6000 feet. The tunnel is designed to maintain CO concentrations at or
below 200 ppm. For the tunnel at 11,000 feet, the authors recommended maintaining CO
5 concentrations at or below 25 ppm for long-term exposures, and no greater than 50 ppm for
peaks of one-hour exposure. The recommendations are based on considerations of a
combination of hypoxia from lack of O2 due to the altitude and stress of CO exposures of
workers and motorists. The authors recommend that warning signs and notices be posted to
warn susceptible individuals to take another route.
10 Carbon monoxide exposures of tollbooth operators were studied along the New Jersey
Turnpike. The results reported by Heinold et al. (1987), indicated peak exposures for one-
hour ranged from 12 to 24 ppm with peak eight-hour exposures of 6 to 15 ppm.
Carboxyhemoglobin levels were determined for 15 nonsmokers at the start, middle, and
end of a 40-day submarine patrol (Bondi et al., 1978). Values found were 2.1 %, 1.7%, and
15 1.7%, respectively. The average ambient air concentration for CO was 7 ppm. The authors
observed that the levels of percent COHb found would not cause significant impairment of the
submariners.
In contrast, Iglewicz et. al. (1984) found in a 1981 study that CO concentrations inside
ambulances in New Jersey were often above the EPA eight-hour standard of 9 ppm. For
20 example, measurements made at the head of the stretcher exceeded 9 ppm on nearly 27% of
the 690 vehicles tested, with 4.2% (29 vehicles) exceeding 35 ppm.
Environmental tobacco smoke (ETS) has been reviewed (National Research Council,
1986) for contributions to air contaminants in airliner cabins, and to potential exposures for
passengers and flight crew members. ETS is described as a complex mixture containing
25 many components. Analyses of CO content and paniculate matter in cabin air were used as
surrogates for the vapor phases and solid components of ETS, respectively. A mathematical
model was developed and used to calculate the dilution of contaminants by outside make-up
air. Total emissions for CO in mainstream smoke range from 10,000 to 23,000 mg per
cigarette. More CO is emitted in sidestream smoke; the ratio of sidestream smoke to
30 mainstream smoke ranges from 2.5:1 to 4.7:1. This ratio depends on the length of time a
cigarette is held without active smoking compared to the total inhalation and smoking time.
March 12, 1990 8-59 DRAFT-DO NOT QUOTE OR CITE
-------
The amount of CO in the cabin environment depends on the rate and number of cigarettes
smoked, and on the rate of dilution by outside make-up air. An additional factor to consider
is the influence of pressure altitude on the absorption of CO and other gases. The legal limit
for pressure altitude is 8000 feet. The partial pressure of O2 is 120 mm Hg assuming
5 20% O2 in the cabin air, compared to 152 mmHg at sea level. It is possible that the
absorption rate for CO would be increased under hypobaric conditions.
An examination of CO hazard in city traffic for policemen in three Swedish towns
(Gome et al., 1969) showed that the increases observed in the percent COHb in blood for a
group of 28 policemen were associated with exposures to exhaust fumes from heavy traffic.
10 Conversely, results from studies of 28 traffic policemen who were smokers and had relatively
high percent COHb in blood at the beginning of a work period either showed no change, or
showed a decrease in percent COHb while exposed to exhaust fumes in directing traffic.
Exposures were higher in a larger, more congested city, as compared to two smaller cities in
the study.
15 Carbon monoxide levels in city driving in Los Angeles were measured using a prototype
CO measuring device mounted in the passenger seat (Haagen-Smit, 1966). Carbon monoxide
concentrations were continuously monitored and were sampled by means of a glass tube
projecting through the window. Typical commuting trips were made throughout the
downtown Los Angeles area during commuting hours. The distance traveled was about
20 30 mi. The shortest time was 40 min and the longest time was one-hour and 55 min.
Concentrations of CO averaged 37 ppm for the best trips, with an average of 54 ppm in
heavy traffic moving at 20 mph; peak CO concentrations reached as high as 120 ppm.
A study of municipal bus drivers in the San Francisco Bay area by Quinlan et al. (1985)
showed a TWA of 1 to 23 ppm, with mean TWA = 5.5 ppm and standard deviation of
25 4.9 ppm. The peak exposures ranged from 7 to 47 ppm with mean 25.3 ppm and standard
deviation of 12.5 ppm.
Cooke (1986) reports finding no significant increases outside normal ranges, as
compared to the general population, for levels of blood lead and COHb in a group of
13 roadside workers. Samples were collected in the afternoon of a workday. Among the
30 subjects, 7 to 13 were smokers and showed percent COHb in blood ranging from 3.0 to 8.8
(mean of 5.5%). Each nonsmoker percent COHb ranged from 0.5 to 1.4 (mean of 1.2%).
March 12, 1990 8-60 DRAFT-DO NOT QUOTE OR CITE
-------
Each smoker had smoked at least one cigarette in the four hours preceding collection of blood
samples. No samples were collected before start of work, and no measurements of CO in air
at the work sites were presented.
8.5 BIOLOGICAL MONITORING
A unique feature of carbon monoxide exposure is that there is a biological marker of the
dose that the individual has received: the blood level of CO. This level may be calculated by
measuring blood COHb or by measuring CO in exhaled breath.
10
8.5.1 Blood Carboxyhemoglobin Measurement
Carbon monoxide in the inspired air is rapidly transferred to the blood in the alveoli at a
rate that is dependent upon several physiological variables. The blood level of CO is
conventionally represented as a percentage of the total Hb available (i.e., the percentage of
15 Hb that is in the form of percent COHb or simply COHb). The high affinity of CO for Hb
has the effect of retaining the bulk (90 to 95%) of the absorbed gas in the vascular space and
at the same time amplifying the exposure. This latter phenomenon occurs because the affinity
of Hb for CO is 200 to 250 times that for O2 (Douglas et al., 1912) resulting in a relatively
high COHb level at very low partial pressures of CO in the alveolar gas phase. The primary
20 physical and physiological variables that determine the relationship between ambient exposure
and blood levels of CO are presented in detail in Chapter 9.
8.5.1.1 Measurement Methods
Any technique for the measurement of CO in blood must be specific for CO and have
25 sufficient sensitivity and accuracy for the purpose of the values obtained. The majority of
technical methods that have been published on measurement of CO in blood have been for
forensic purposes. These methods are less accurate than generally required for the
measurement of low levels of COHb (<5% COHb). Blood levels of CO resulting from
exposure to existing NAAQS levels of CO would not be expected to exceed 5% COHb in
30 nonsmoking subjects. The focus of the forensic methods has been the reliability of
measurements over the entire range of possible values: from less than 1 % to 100% COHb.
March 12, 1990 8-61 DRAFT-DO NOT QUOTE OR CITE
-------
These forensically oriented methods are adequate for the intended use of the values and the
nonideal storage conditions of the samples being analyzed.
In the areas of exposure assessment and low level health effects of CO, it is more
important to know the accuracy of any method in the low level range of 0 to 5% COHb.
5 There is little agreement upon acceptable reference methods in this range nor are there
accurate reference standards available in this range. The use of techniques that have
unsubstantiated accuracy in the low range of COHb levels can lead to considerable differences
in estimations of exposure conditions. Measurement of low levels of CO in blood demands
careful evaluation because of the implications based upon this data for the setting of air
10 quality standards. Therefore this section will focus on the methods that have been evaluated
at levels below 10% COHb and methods that have been extensively used in assessing
exposure to CO.
The measurement of CO in blood can be accomplished by a variety of techniques that
have been divided into destructive and nondestructive methods (U.S. Environmental
15 Protection Agency, 1979). Carboxyhemoglobin can be determined nondestructively by
observing the change in the absorption spectrum in either the Soret or visible region brought
about the combination of CO with Hb. With present optical sensing techniques, however, all
optical methods are limited in sensitivity to approximately 1% of the range of expected
values. If attempts are made to expand the lower range of absorbances, sensitivity is lost on
20 the upper end where, in the case of COHb, total Hb is measured. For example, in the
spectrophotometric method described by Small et al. (1971), a change in absorbance equal to
the limits of resolution of 0.01 units can result in a difference in 0.6% COHb. Therefore,
optical techniques can not be expected to obtain the resolution that is possible with other
means of detection of CO (Table 8-11). The more sensitive (higher resolution) techniques
25 require the release of the CO from the Hb into a gas phase that can be detected directly by:
(1) infrared absorption (Coburn et al., 1964; Maas et al., 1970) following separation using
gas chromatography; (2) the difference in thermal conductivity between CO and the carrier
gas (Allred et al., 1989; Ayres et al., 1966; Dahms and Horvath, 1974; Goldbaum et al.,
1986; Horvath et al., 1988; McCredie and Jose, 1967); (3) the amount of ionization
30 following quantitative conversion of CO to CH4 and ionization of the CH4 (Clerbaux et al.,
1984; Collison et al., 1968; Constantino et al., 1986; Dennis and Valeri, 1980; Guillot et al.,
March 12, 1990 8-62 DRAFT-DO NOT QUOTE OR CITE
-------
10
15
20
25
30
TABLE 8-11. COMPARISON OF REPRESENTATIVE METHODS FOR ANALYSIS
OF CARBON MONOXIDE IN BLOOD
Source
Gasometric Detection
Scholander and Roughton
(1943)
Horvath and Roughton
(1942)
Spectrophotometric Detection
Coburn et al.
(1964)
Small et al.
(1971)
Maas et al.
(1970)
Brown (1980)
Gas Chromatographv
Ayres et al.
(1966)
Goldbaum et al.
(1986)
McCredie and Jose
(1967)
Dahms and Horvath
(1974)
Collison et al.
(1968)
Kane (1985)
Vreman et al. (1984)
Method
Syringe-
capillary
Van Slyke
Infared
Spectro-
photometry
CO-Oximeter
(IL 182)
CO-Oximeter
(IL-282)
Thermal
conductivity
Thermal
conductivity
Thermal
conductivity
Thermal
conductivity
Flame
ionization
Flame
ionization
Mercury vapor
Resolution*
ml/dl
0.02
0.03
0.006
0.12
0.21
0.2
0.001
ND
0.005
0.006
0.002
ND
0.002
CV%b
2 to 4%
6%
1.8%
ND
5%
5%
2.0%
1.35%
1.8%
1.7%
1.8%
6.2%
2.2%
Reference Method
Van Slyke
Van Slyke-Neill
Van Slyke-Syringe
Flame Ionization
Spectrophotometric
Flame Ionization
ND
Flame Ionization
ND
Van Slyke
Van Slyke
CO-Oximeter
ND
r"
ND"
ND
ND
ND
ND
0.999
ND
0.996
ND
0.983
ND
1.00
ND
35
40
45
'The resolution is the smallest detectable amount of CO or the smallest detectable difference between
samples.
50 ""Coefficient of variation was computed on samples containing less than 15 % COHb, where possible.
55
The r value is the correlation coefficient between the technique reported and the reference method used to
verify its accuracy.
dlndicates no data were available.
March 12, 1990
8-63 DRAFT-DO NOT QUOTE OR CITE
-------
1981; Kane, 1985; Katsumata et al., 1985); or (4) the release of Hg vapor due to the
combination of CO with mercuric oxide (Vreman et al., 1984).
Sample Handling
5 Carbon monoxide bound to Hb is a relatively stable compound which can be dissociated
by exposure to O2 or ultraviolet (UV) radiation (Chace et al., 1986; Horvath and Roughton,
1942). Ultraviolet light has not been shown by Goldstein et al. (1986) to affect COHb levels
in glass vials exposed to room lighting conditions, however the ability of UV light to
penetrate these tubes was not demonstrated. If the blood sample is maintained in the dark
10 under cool, sterile conditions, the CO content will remain stable for a long period of time.
Various investigators have reported no loss of percent COHb over 10 days (Collison et al.,
1968), three weeks (Dahms and Horvath, 1974), four months (Ocak et al., 1985) and six
months (Vreman et al., 1984). The anticoagulant system used appears to influence the CO
level since some EDTA vacutainer tube stoppers contain CO (Vreman et al., 1984). The
15 increased levels of COHb due to this amount of CO have been demonstrated (Goldstein et al.,
1986; Vreman et al., 1984). The stability of the CO content in properly stored samples does
not indicate that constant values will be obtained by all techniques of analysis. The
spectrophotometric methods are particularly susceptible to changes in optical qualities of the
sample which results in small changes in COHb with storage (Allred et al., 1989).
20 Carboxyhemoglobin values obtained with the IL 282 CO-Oximeter have been shown to
decrease over the first three days following collection (Allred et al., 1989; Goldstein et al.,
1986). This decrease occurs within the first 24 h (Allred et al., 1989) and does not fall
further over the next 14 days (Goldstein et al., 1986). Storage of blood samples can result in
the formation of methemoglobin (Goldstein et al., 1986) and under some conditions
25 sulfhemoglobin (Rai and Minty, 1987). Both species of Hb can result in the optical methods
of COHb detection being incorrect depending upon the specific wavelengths utilized.
Therefore the care needed to make a COHb determination depends upon the technique
that is being utilized. It appears as though measurement of low levels of COHb with optical
techniques should be conducted out as soon as possible following collection of the samples.
30
March 12, 1990 8-64 DRAFT-DO NOT QUOTE OR CITE
-------
Potential Reference Methods
Exposure to CO at equilibrium conditions results in COHb levels of between 0.1 and
0.2% COHb for each part per million of CO in the atmosphere. A reference technique for
the measurement of COHb should be able to discriminate between two blood samples with a
5 difference of 0.1 % COHb (approximately 0.02 mL/dL). To accomplish this task the
coefficient of variation (standard deviation of repeated measures on any given sample divided
by the mean of the values times 100) of the method should be less than 5% so that the two
values that are 0.1 % COHb different can be statistically proven to be distinct. In practical
terms a reference method should have the sensitivity to detect approximately 0.025% COHb
10 to provide this level of confidence in the values obtained.
The accurate measurement of CO in a blood sample requires the quantitation of the
content of CO in blood. Optically based techniques have limitations of resolution and
specificity due to the potential interference from many sources. The techniques that can be
used as reference methods involve the quantitative release of CO from the Hb followed by the
15 measurement of the amount of CO released. Classically this quantitation was measured
manometrically with a Van Slyke apparatus (Horvath and Roughton, 1942) or a Roughton-
Scholander syringe (Roughton and Root, 1945). These techniques have served as the "Gold
Standard" in this field for almost 50 years. However, there are limitations of resolution with
these techniques at the lower ranges of COHb. The gasometric standard methodology has
20 been replaced with head-space extraction followed by the use of solid phase gas
chromatographic separation with several different types of detection: thermal conductivity,
flame ionization, and mercury vapor reduction. With the use of National Institute of
Standards and Technology (NIST) standard gas mixtures of CO, the gas chromatographic
techniques can be standardized when proper consideration is given to potential sources of loss
25 of standard. The CO in the headspace can also be quantitated by infrared detection which can
be calibrated with gas standards.
Flame Ionization Detection. This technique requires the separation of CO from the
other headspace gases and the reduction of the CO to CH4 by catalytic reduction. Collison
30 et al. (1968) reported that the results from their method correlated with the Van Slyke
gasometric method at high levels of CO (8 to 13 mL/dL) where the error in the gasometric
March 12, 1990 8-65 DRAFT-DO NOT QUOTE OR CITE
-------
method was minimal. The values obtained from the two independent techniques were highly
significantly correlated (p< 0.0001) with a linear regression r = 0.992. The limit of
detection was reported to be 0.01% COHb using 100 pL of blood. The coefficient of
variation was 1.08% on a sample containing approximately 50% COHb and 1.80% on a
5 sample containing approximately 0.8% COHb. The basic technique of Collison et al. (1968)
using headspace analysis and flame ionization detection is the most sensitive method that has
been compared with the gasometric methods. Modifications of this method have been widely
used by other investigators for evaluating technically simpler methods of CO analysis
(Clerbaux et al., 1984; Collison et al., 1968; Dennis and Valeri, 1980; Guillot et al., 1981;
10 Kane, 1985; Katsumata et al., 1985). This method conforms to all the requirements of a
reference method.
Thermal Conductivity Detection. Ayres et al. (1966) reported a method for using
vacuum extraction of CO from blood in a Van Slyke apparatus for gas chromatographic
15 separation with thermal conductivity analysis of the CO. The gas phase of the reaction
chamber was eluted onto a 5A molecular-sieve column for separation of the components.
This technique was reported to have a lower limit of detectability of 0.001 mL/dL or
approximately 0.005% COHb. The coefficient of variation was reported to be 1.95% on a
sample of unspecified percent COHb. The analysis system was calibrated using a gas sample
20 of known CO content injected directly onto the column. No comparisons were performed
with other standard techniques. McCredie and Jose (1967) also reported results from
chromatographic separation of vacuum extracted gas. Thermal conductivity detection enabled
the limit of detection to be 0.005 mL/dL or approximately 0.025% COHb. This system was
also calibrated with standard gas mixtures injected directly onto the column. A coefficient of
25 variation was not presented but a standard deviation of 0.004 mL/dL on a series of repeat
analyses on an average blood sample indicates that this method is sufficiently reproducible.
This would represent a coefficient of variation on the blood CO content measured from the
average nonsmoker of 2.5 %. Dahms and Horvath (1974) described a technique of headspace
analysis of CO using thermal conductivity detection. The CO was released from the blood
30 while the mixture was stirred to produce a vortex, using Van Slyke reagents in a sealed
reaction vial. The extraction occurred into the headspace of a sealed vial pressurized to the
March 12, 1990 8-66 DRAFT-DO NOT QUOTE OR CITE
-------
head pressure on the column. The limit of detection with this technique was 0.006 mL/dL or
0.03% COHb with a coefficient of variation of 1.7% on a sample containing 6.5% COHb.
This method used standard gases injected into the reaction vial to calibrate the system. The
results of this technique were compared to the standard Van Slyke method (Horvath and
5 Roughton, 1942) over the whole range of values and more specifically on 90 blood samples
containing less than 10% COHb. The correlation coefficient between the gas chromatography
and the Van Slyke method was 0.984. Linear regression analysis demonstrated essentially a
zero intercept (0.009 mL/dL) between the two techniques. This close agreement between
values obtained with these independent methods provides a basis for the use of standard gases
10 to calibrate gas chromatographic techniques. All of the above mentioned gas chromatographic
methods for determination of CO in blood are acceptable as reference methods.
Infrared Detection. Coburn et al. (1964) described a method for extracting CO from
blood under normal atmospheric conditions and then injecting the headspace gas into an
15 infrared analyzer. This technique has a reported limit of detectability of 0.007 mL/dL or
0.035% COHb. The coefficient of variation was 1.8% on an average COHb of 1.67%. The
results of this technique were compared with the gasometric technique of Roughton and Root
(1945) on five samples; there was no difference between the two techniques. This method is
acceptable as a reference method for the measurement of CO in blood.
20
Hemoglobin Measurement. The conventional means of representing the quantity of CO
in a blood sample is the percent COHb: the percentage of the total CO combining capacity
that is in the form of COHb. This is conventionally determined by the use of the following
formula:
25
%COHb = [CO content/(hemoglobin x 1.389)] X 100 (8-11)
March 12, 1990 8-67 DRAFT-DO NOT QUOTE OR CITE
-------
where:
CO content = cc/dL blood STPD,
hemoglobin = g/dL blood, and
1.389 = the stoichiometric combining capacity of hemoglobin for CO in units of
mL/g STPD.
10 The analytical methods that quantify the CO content in blood require the conversion of these
quantities to percent COHb. The product of the Hb and the theoretical combining capacity
(1.389 according to International Committee for Standardization in Haemotology, 1978)
yields the CO capacity. With the use of capacity and the measured content, the percent of
CO capacity (percent COHb) is calculated. To be absolutely certain of the accuracy of the
15 Hb measurement, the theoretical value should be routinely substantiated by direct
measurement (internal validation) of the Hb CO combining capacity. The total CO combining
capacity should be determined as accurately as the content of CO. The error of the
techniques that measure CO content are dependent on the error in Hb analysis for the final
form of the data, percent COHb. Therefore the actual CO combining capacity should be
20 measured and compared with the calculated value based upon the reference method for Hb
measurement. The measurement of CO combining capacity can be routinely carried out by
equilibration of a blood sample with CO (Allred et al., 1989).
The standard methods for Hb determination involve the conversion of all species of Hb
to cyanmethemoglobin with the use of a mixture of potassium ferricyanide, potassium
25 cyanide, and sodium bicarbonate. Three combinations of similar reagents have been routinely
used for the quantification of Hb. Drabkin's solution contains 0.6 mM K4Fe(CN)6, 0.8 mM
KCN, and 12 mM NaHCO3 (Drabkin and Austin, 1932). Van Kampen and Zijlstra (1961)
substituted 0.7 mM K2HPO4 for the bicarbonate in the reagent mixture. A third reagent for
producing cyanmethemoglobin is that of Taylor and Miller (1965) who increased the
30 concentration of potassium ferricyanide to 3 mM in Van Kampen and Zijlstra's mixture to
decrease the reaction time with COHb. The presence of high levels of COHb slows the rate
of conversion to cyanmethemoglobin so that the use of the conventional Drabkin's reagent
requires a reaction time of at least 180 min (Allred et al., 1989; Kane, 1985) as opposed to
the recommended time of 20 to 30 min. This increased reaction time is essential for the
March 12, 1990 8-68 DRAFT-DO NOT QUOTE OR CITE
-------
accurate comparison of cyanmethemoglobin values with CO combining capacity
measurements. If the reaction is not permitted to go to completion the spectrophotometric
method will underestimate the amount of Hb present in the sample.
5 Other Methods of Measurement
There are a wide variety of techniques that have been described for the analysis of CO
in blood. These methods have been reviewed previously (U.S. Environmental Protection
Agency, 1979) and include UV-visible spectrophotometry (Brown, 1980; Small et al., 1971;
Zwart et al., 1984; 1986), magnetic circular dichroism spectroscopy (Wigfield et al., 1981),
10 photochemistry (Sawicki and Gibson, 1979), gasometric methods (Horvath and Roughton,
1942; Roughton and Root, 1945), and a calorimetric method (Sjostrand, 1948). Not all of
these methods have been as well characterized for the measurement of low levels of COHb as
those listed above as potential reference methods.
15 Spectrophotometric Methods. The majority of the techniques are based upon optical
detection of COHb which is more rapid than the reference techniques because these methods
do not involve extraction of the CO from the blood sample. These direct measurements also
enable the simultaneous measurement of several species of hemoglobin including reduced Hb,
O2Hb, and COHb. The limitations of the spectrophotometric techniques have been reviewed
20 by Kane (1985). The optical methods utilizing ultraviolet wave lengths require dilution of the
blood sample which can lead to the loss of CO due to the competition with the dissolved O2in
solution. Removing the dissolved O2 with dithionite can lead to the formation of
sulfhemoglobin which interferes with the measurement of COHb (Rai and Minty, 1987).
Another limitation is that the absorption maxima (and spectral curves) are not precisely
25 consistent between individuals. This may be due to slight variations in types of hemoglobin
in subjects. For these reasons the techniques using fixed wave length measurement points
would not be expected to be as precise, accurate, or specific as the proposed reference
methods mentioned above.
Rodkey et al. (1979) reported a modification of the spectrophotometric technique for
30 measuring COHb. This method converts all the hemoglobin species in a blood sample to
either COHb or Hb by the quantitative addition of the reducing agent sodium hydrosulfite.
March 12, 1990 8-69 DRAFT-DO NOT QUOTE OR CITE
-------
The absorbance at 420 nm was used for the determination of COHb and 432 nm for Hb. The
optically based values were compared with those obtained by gas chromatography on the same
28 samples. Twenty-five of these values were below 6% COHb and linear regression analysis
demonstrated a slope of 1.038 with an intercept of -0.154 and an r = 0.994. The number of
5 samples studied was relatively small and no error term was presented for the relationship.
Visual inspection of the data, however, indicates a wide scatter of optical values for any given
gas chromatograph value when the levels were at or below 1 % COHb (normal range of values
for unexposed, nonsmoking individuals).
A multicomponent spectrophotometric technique for the measurement of hemoglobin
10 derivatives was reported by Zwart et al. (1984). This technique employs a multiwavelength
spectrophotometer that uses reversed optics to enable the rapid collection of the absorbance
spectrum from an array of photomultiplier tubes that detect transmission of light at intervals
of 2 nm. This method offers the possibility of instantaneous absorption data over the entire
spectrum rather than the collection of data at a few selected wavelengths. This optical system
15 offers the potential for correcting for individual variability in absorption characteristics of
hemoglobin. The COHb data produced with this technique has not been compared with any
of the proposed reference methods but has been compared with that obtained from the
mercury vapor detector. The correlation coefficient of the optical data with the gas
chromatographic-mercury vapor technique was only 0.87; linear regression analysis resulted
20 in the following relationship: GC = 0.65 (MCA) + 0.24 (Vreman et al., 1987).
Mercury Vapor Detection. The most sensitive detector for the measurement of CO is the
UV-photometer that senses mercury vapor produced by the reaction of CO with HgO (Trace
Analytical). This unit has the reported ability to resolve 1 ppb. The use of such a sensitive
25 detector for blood determinations requires that measurements be carried out on only 1 to
10 n\ quantities of blood. Vreman et al. (1984) reported the use of this detector following
gas chromatographic separation of the CO from other gases in blood. Mercuric oxide will
react with other gases so the chromatographic separation is an essential step in the use of this
detector. Values for COHb obtained with this technique have not been compared those
30 obtained with any of the proposed reference methods. The COHb analysis method of Vreman
et al. (1984) was used in a parallel with a gas chromatography method using thermal
March 12, 1990 8-70 DRAFT-DO NOT QUOTE OR CITE
-------
conductivity detection for the routine measurement of COHb in a series of samples from
subjects exposed to CO to produce levels of COHb up to 6% (Allred et al., 1989). Paired
data from samples analyzed by both techniques were obtained on 108 samples. The values
were significantly correlated r = 0.987; however, the reduction gas analyzer results were not
5 corrected to STPD conditions, so an absolute comparison was not possible. This technique
needs further validation by comparison with other methods to assure that the levels measured
are accurate.
CO-Oximeter Measurements of Carboxyhemoglobin
10 The speed of measurement and relative accuracy of spectrophotometric measurements
over the entire range of expected values led to the development of CO-Oximeters. These
instruments utilize from two to seven wavelengths in the visible region for the determination
of proportions of oxyhemoglobin, carboxyhemoglobin, reduced hemoglobin and
methemoglobin. The proportion of each species of Hb is determined from the absorbance and
15 molar extinction coefficients at present wavelengths. All of the commercially available
instruments provide rapid results for all the species of Hb being measured. In general, the
manufacturers' listed limit of accuracy for COHb for all of the instruments is 1% COHb.
However, this level of accuracy is not suitable for measurements associated with background
CO levels (<2% COHb) because it corresponds to errors exceeding 50%. The precision of
20 measurement for these instruments is excellent and has misled users regarding the accuracy of
the instrument. The relatively modest level of accuracy is adequate for the design purposes of
the instrument; however, at low levels of COHb the ability of the instrument to measure the
percent COHb accurately is limited.
The commercially available instruments for the measurement of COHb all utilize the
25 same basic principles: hemolysis, constant temperature, and the measurement of absorbance
at several wave lengths. These instruments have been designed to provide information
regarding COHb measurement that is ±1% COHb. However, these instruments are all very
precise so that the coefficient of error between repeat measurements (standard deviation of
repetitions/mean of the repetitions) is very low. Unfortunately, very few studies have
30 evaluated the accuracy of the measurements made with these instruments as a routine aspect
of quality control. The concern regarding the accuracy of any optical measurement on a
March 12, 1990 8-71 DRAFT-DO NOT QUOTE OR CITE
-------
diluted blood sample should be of greatest concern due to the wide variety of substances that
can subtly alter the absorption spectrum of Hb and the optical quality of the blood sample
itself. Because of the widespread use of these instruments, the evaluations of this instrument
will be carefully reviewed. These instruments consist of the Instrumentation Laboratories
5 CO-Oximeters known as the IL 182 and the IL 282, the Radiometer Oximeter OSM-3, and
the Coming Oximeter 2500. There is very little information regarding the accuracy of the
OSM-3 in the low range of COHb values compared to the data obtained from paired analysis
with a reference method.
The instruments that have been used to the greatest extent in the health effects of CO
10 have been the IL 182 and IL 282 (Instrumentation Laboratory, Inc., Lexington, MA). The
IL 282 instrument uses absorbances at four wavelengths in the visible region and a matrix of
molar extinction coefficients to calculate each species of hemoglobin. This method is
susceptible to interference from high concentrations of methemoglobin and sulfhemoglobin.
The IL 282 CO-Oximeter has been shown to provide accurate data when the range of 0 to
15 100% COHb is considered (Brown, 1980). However, comparison of the results from this
method with the proposed reference methods indicates that at low levels of COHb the results
from this instrument are not sufficiently accurate to warrant their use alone for low level
COHb investigation. Resting levels of percent COHb have been shown to be below 0.9% for
non-smokers by all the proposed reference methods (Ayres et al., 1966; Coburn et al., 1964;
20 Collison et al., 1968; Dahms and Horvath, 1974; McCredie and Jose, 1967). The limit of
accuracy for the IL 282 CO-oximeter for percent COHb is 1%, which has raised concern over
the capability of all CO-Oximeters in the low range of COHb levels. Therefore the accuracy
of these instruments has been determined by paired observations on blood samples with the
CO-Oximeter and a reference method. The results are shown in Table 8-12 below.
25 The results from the linear regression analyses of all these comparisons indicate that
there is considerable difference between instruments of the same model type. The slope of
the relationship between the optical methods are sufficiently close to unity that there is no
difference between instruments in the linearity of the measurements. Confidence intervals for
the regressions are not given so this comparison can not be made. The intercept values vary
30 widely relative to the purpose of accurately measuring low levels of COHb. These
differences probably reflect the difference between instruments. In order to use these
March 12, 1990 8-72 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 8-12. EVALUAITON OF THE ABILITY OF CO-OXIMETERS TO MEASURE LOW LEVELS
OF COHb AS COMPARED TO PROPOSED REFERENCE METHODS
t— »
i
oo
o
£
>
HH
3
6
o
1
H
O
CJ
B
a
o
5*
n
^^
Instrument
IL 182
IL 182
IL182
IL282
IL282
IL282
IL282
IL282
Corning
2500
Corning
2500
"Abbreviations:
Reference
Method"
GC-FID
GC-FID
Infrared
GC-FID
GC-FID
GC-TCD
GC-FID
CG-TCD
GC-FID
GC-FID
GC-TCD is gas c
Slope
0.690*GC
1.049*GC
0.977*IR
0.990*GC
1.122*GC
0.919*GC
0.895*GC
1.0069*GC
1.05*GC
1.05*GC
1.05*GC
0.92*GC
1.013*GC
:hromatography thermal
Intercept
+3.59
-0.54
+3.33
+0.45
-0.907
-0.068
+0.66
-0.01
+0.79
+0.55
+0.47
+ 1.17
-1.279
conductivity
R
0.59
NDb
ND
0.997
0.993
0.961
0.856
0.99
0.99
0.979
0989
r detection; GC-FID
n
16
275
12
39
13
20
16
ND
203
192
162
50
286
is gas chrc
COHb
Range Reference
< 15% Constantino et al. (1986)
<15% Guillot et al. (1981)
< 100% Maas et al. (1970)
< 100% Dennis and Valeri (1980)
< 17% Dennis and Valeri (1980)
< 8 % Goldbaum et al. (1986)
< 15% Constantino et al. (1986)
<9% Horvath et al. (1988)
<6% Allred et al. (1989)
<6% Allred et al. (1989)
<6% Allred et al. (1989)
<20% Kane (1985)
< 15% Tikuisis et al. (1987)
imatography with flame ionization detection.
blndicates no data were available.
-------
instruments for the measurement of low levels of COHb they must be individually and
routinely calibrated with a reference method. The linearity of the response of the instruments
implies that a standard correction can be applied to the value for COHb with the result that
the average value of COHb obtained with these instruments will be correct.
5 The interaction of Hb species with the measurement of COHb below 10% has been
evaluated by Allred et al. (1989). In freshly drawn blood samples, levels of COHb were
maintained constant, as measured by GC, while levels of metHb, total Hb and O2Hb were
varied. Only the level of O2Hb interacted significantly with the COHb value. In a series of
46 subjects the effect of O2Hb was measured to determine its role in routine measurements of
10 COHb. The effect of O2Hb varied considerably between individuals, with the average change
being approximately 0.1% COHb for every 10% change in O2Hb. Almost all COHb
measurements were made on venous blood which can vary considerably in O2Hb
concentration and consequently affect the measurement of low levels of COHb.
Hydrogen ion concentration was shown to have an effect on the measured percent COHb
15 in blood stored in acid citrate dextrose (ACD) solution for two days. However the effect of
pH on percent COHb in freshly drawn samples has not been clearly demonstrated (Allred
et al., 1989). Hydrogen ion concentration has been demonstrated to change the absorbance
spectrum of oxyhemoglobin and therefore may be expected to have an effect on the ability of
CO-Oximeters to measure COHb. Plasma lipid, triglyceride and cholesterol levels were
20 found to not have any effect on the ability of the IL282 CO-Oximeter to measure COHb as
determined by the difference between in instrument value and the reference value obtained by
gas chromatography.
The content of CO in blood stored in a tightly capped syringe at 4°C in the dark has
been shown to remain stable for up to four months. Measurement of COHb by IL 282 CO-
25 Oximeter on blood samples (COHb range of 4.3 to 1.3%) within 15 min of collection,
followed by storage for 24 and 48 h as described above, resulted in a decrease in the detected
percent COHb. The apparent loss of COHb occurred in the first 24 h and averaged 16%
(Allred et al., 1989). There was no change in percent COHb as determined by GS. It is not
clear when in the first 24-h period this change occurred.
30 The use of CO-Oximeters to measure low levels of COHb can provide useful
information regarding mean values, provided a reference technique is used to properly
March 12, 1990 8-74 DRAFT-DO NOT QUOTE OR CITE
-------
calibrate the instrument. It has been shown, however, that the range of values obtained with
the optical method will be greater than that obtained with a reference method. In a group of
subjects with cardiovascular disease the standard deviation of the percent COHb values for
non-smoking, resting subjects was 2 to 2.5 times greater for the CO-Oximeter values than for
5 the GC values on paired samples (Allred et al., 1989). Therefore, the potential exists with
the CO-Oximeter for having an incorrect absolute value for COHb as well as an incorrectly
broadened range of values.
In addition, it is not clear exactly how sensitive the CO-Oximeter techniques are to
small changes in COHb at the low end of the CO dissociation curve. Allred et al. (1989)
10 have noted that the interference from changing O2 saturation can have a very significant
influence on the apparent COHb reading in a sample. The interaction between Hb species
was also reported by Dennis and Valeri (1980). This suggests nonlinearity or a
disproportionality in the absorption spectrum of these two species of Hb. It is also a potential
source of considerable error in the estimation of COHb by optical methods.
15
8.5.1.2 Carboxyhemoglobin Measurements in Populations
Numerous studies have used the above described methodologies to characterize the
levels of COHb for the general population. These studies have been designed to determine
frequency distributions of COHb levels in the populations being studied. In general the
20 higher the frequency of COHb levels above baseline in nonsmoking subjects the greater the
incidence of significant CO exposure.
Carboxyhemoglobin levels in blood donors have been studied for various urban
populations in the United States. Included have been studies of blood donors and sources of
CO in the metropolitan St. Louis population (Kahn et al., 1974), evaluation of smoking and
25 COHb in the St. Louis metropolitan population (Wallace et al., 1974), analyses of
16,649 blood samples for COHb provided by the Red Cross Missouri-Illinois blood donor
program (Davis and Gantner, 1974), a survey of blood donors for percent COHb in Chicago,
Milwaukee, New York and Los Angeles (Stewart et al., 1976), a national survey for COHb
in American blood donors from urban, suburban, and rural communities across the United
30 States (Stewart et al., 1974), and the trend for percent COHb associated with vehicular traffic
in Chicago blood donors (Stewart et al., 1976). These extensive studies of volunteer blood
March 12, 1990 8-75 DRAFT-DO NOT QUOTE OR CITE
-------
donor populations show three main sources of exposure to CO in urban environments. These
are smoking, general activities (usually associated with internal combustion engines), and
occupational exposures. For comparisons of sources, the populations are divided into two
main groups - smokers and nonsmokers. The main groups often are divided further into
5 subgroups consisting of industrial workers, drivers, pedestrians, and others, for example.
Among the two main groups, smokers show an average of 4% COHb with a usual range of 3
to 8%; nonsmokers average about 1% COHb (Radford and Drizd, 1982). Smoking behavior
generally occurs as an intermittent diurnal pattern, but in some individuals who chain smoke,
COHb levels can rise to a maximum of about 15%.
10 In addition to tobacco smoke, the most significant sources of other potential exposure to
CO in the population are community air pollution, occupational exposures, and household
exposures (Goldsmith, 1970). Community air pollution comes mainly from auto exhaust and
has a typical intermittent diurnal pattern (see Chapter 6). Occupational exposures occur for
up to eight hours for 5 days a week, producing COHb levels generally less than 10%.
15 However, exposures to high concentrations of CO in occupational settings have caused death
from CO intoxication. Household exposures usually result in less than 2% COHb, but high
concentrations, occurring particularly during nighttime hours, have been known to cause
death. For example, during the winter, a number of people die as a result of using a variety
of space heating devices in poorly ventilated spaces (Goldsmith, 1970). Poorly vented floor
20 heaters are also a source of CO intoxications, with many such exposures occurring at night.
More recent studies characterizing COHb levels in the population have appeared in the
literature. Turner et al. (1986) used an IL 182 CO-Oximeter to determine percent COHb in
venous blood of a study group consisting of both smoking and nonsmoking hospital staff,
inpatients and outpatients. Blood samples were collected for 3487 subjects
25 (1255 nonsmokers) during morning hours over a five-year period. A detailed smoking history
was obtained at the time of blood collection. Secondary pipe or secondary cigar smokers
were considered to be those who were initially cigarette smokers but subsequently switched to
cigars or pipes. Primary cigar or pipe smokers were those who had never smoked cigarettes
and were not in the habit of inhaling large amounts of tobacco smoke as is the observed
30 custom with cigarette smoking. Using 1.7% COHb as a normal cutoff value, the distribution
for the population studied showed above normal results for 94.7% of cigarette smokers,
March 12, 1990 8-76 DRAFT-DO NOT QUOTE OR CITE
-------
10.3% of primary cigar smokers, 97.4% of secondary cigar smokers, and 94.7% of
secondary pipe smokers.
Zwart and Van Kampen (1985) tested a blood supply using a routine spectrophotometric
method for total Hb and for COHb in 3022 samples of blood for transfusion in hospital
5 patients in the Netherlands. For surgery patients over a one-year period, the distribution of
percent COHb in samples collected as a part of the surgical protocol showed 65% below
1.5% COHb, 26.5% between 1.5 and 5% COHb, 6.7% between 5 and 10% COHb, and
0.3% in excess of 10% COHb. This distribution of percent COHb was homogeneous across
the entire blood supply, resulting in 1 in 12 patients having blood transfusions at 75%
10 available Hb capacity.
Radford and Drizd (1982) have analyzed blood COHb in approximately 8400 samples
obtained from respondents in the 65 geographic areas of the nationwide Health and Nutrition
Examination Survey (HANES) during the period 1976-1980. When the frequency
distributions of blood COHb levels are plotted on logarithmic-probability paper (Figure 8-5)
15 to facilitate comparison of the results for different age groups and smoking habits, it is
evident that adult smokers in the U.S. have COHb levels considerably higher than those of
nonsmokers, with 79% of the smokers' blood samples above 2% COHb and 27% of the
observations above 5% COHb. The nationwide distributions of persons aged 12 to 74 who
have never smoked and ex-smokers were similar, with 5.8% of the ex-smokers and 6.4% of
20 the never-smokers above 2% COHb. It is evident that a significant proportion of the
nonsmoking United States population had blood levels above 2% COHb. For these two
nonsmoking groups, blood levels above 5% were found in 0.7% of the never-smokers and
1.5% of the ex-smokers. It is possible that these high blood levels could be due, in part, to
misclassification of some smokers as either ex- or nonsmokers. Children aged 2 to 11 had
25 lower COHb levels than the other groups, with only 2.3% of the children's samples above
2% COHb and 0.2% above 5% COHb.
Wallace and Ziegenfus (1985) utilized the Radford HANES data to analyze the
relationship between the measured COHb levels and the associated eight-hour CO
concentration at nearby fixed monitors. COHb data were available for a total of 1658
30 nonsmokers in 20 cities. The day and hour the blood samples were drawn for each individual
were obtained from the NHANES II data, and the preceding one-hour and eight-hour running
March 12, 1990 8-77 DRAFT-DO NOT QUOTE OR CITE
-------
% 'PHOQ
Figure 8-5. Frequency distributions of carboxyhemoglobin levels in the U.S. population, by
smoking habits.
Source: Adapted from Radford and Drizd (1982); data for NHANES II.
March 12, 1990 8-78 DRAFT-DO NOT QUOTE OR CITE
-------
average ambient CO levels at each fixed station in the city were calculated using the U.S.
EPA centralized data base (SAROAD). For each of the 20 cities the station with the highest
Spearman correlation between COHb concentrations and the preceding eight-hour CO
averages was selected for a linear regression. The results (Table 8-13) show that 17 of the
5 20 stations had /f2 values ranging from 0.00 (6 cities) to 0.10.
Finally, because the participants were part of a nationwide probability sample, all COHb
data were merged with the CO data from the station within each city that showed the
strongest correlation with the COHb values and a linear regression was run. The /J2 value for
the 1528 paired measurements was 0.031 (i.e., only 3% of the variance in the COHb
10 concentrations was explained by the ambient CO data). The authors concluded that fixed
outdoor CO monitors alone are, in general, not providing useful estimates of carbon
monoxide exposure of urban residents.
8.5.2 Carbon Monoxide in Expired Breath
15 Carbon monoxide levels in expired breath can be used to estimate the levels of carbon
monoxide in the subject's blood. The basic determinants of CO levels in the alveolar air have
been described by Douglas et al. (1912), indicating that there are predictable equilibrium
conditions that exist between CO bound to the Hb and the partial pressure of the CO in the
blood. The equilibrium relationship for CO between blood and the gas phase to which the
20 blood is exposed can be described as follows:
PCO/P02 = M (%COHb/%O2Hb) (8-12)
where:
25 Pco = partial pressure of CO in the blood,
P02 = partial pressure of 02 in the blood,
M = Haldane constant (reflecting the relative affinity of the hemoglobin for O2
and CO),
%COHb = percent of total Hb combining capacity bound with CO, and
30 %O2Hb = percent of total Hb combining capacity bound with O2.
35
March 12, 1990 8-79 DRAFT-DO NOT QUOTE OR CITE
-------
10
15
20
25
30
35
TABLE 8-13. REGRESSION PARAMETERS FOR THE RELATIONSHIP BETWEEN
COHb AND EIGHT-HOUR CO AVERAGES FOR 20 CITIES'
City
Atlanta
Bronx
Cincinnati
Chicago
Dayton
Des Moines
Washington
Hampton
Honolulu
Houston
Indianapolis
Los Angeles
Manhattan
Pittsburgh
Racine
Rock Hill
San Diego
San Jose
Tacoma
Washington
Wichita
All cities
n
63
65
93
78
91
90
73
89
65
71
93
66
71
55
91
85
67
59
82
73
81
1528
Slopeb
0.12(±0.03)
0.18(±0.10)
-0.02(±0.21)
-0.02(+0.06)
-0.03(+0.08)
0.003(±0.03)
0.06(±0.04)
0.16(+0.10)
0.39(+0.14)
0.24(+0.12)
0.0005(±0.019)
0.12(±0.03)
0.09(±0.03)
0.03(±0.02)
-0.12(±0.13)
0.23(±0.11)
0.01(±0.08)
0.08(±0.04)
0.04(±0.06)
0.06(±0.04)
-0.11(+0.28)
0.066(±0.009)
Intercept"
0.41(±0.09)
0.60(±0.29)
0.94(±0.22)
1.21(±0.18)
0.93(+0.15)
0.52(±0.12)
1.38(±0.18)
0.50(±0.09)
0.44(±0.18)
0.68(±0.15)
0.79(±0.07)
0.99(±0.18)
0.84(+0.08)
0.77(±0.13)
0.75(±0.14)
0.61(±0.24)
0.84(±0.13)
0.87(±0.11)
0.76(±0.14)
1.38(±0.18)
0.84(±0.35)
0.77(±0.03)
I?
0.27
0.05
0.00
0.00
0.00
0.00
0.03
0.03
0.11
0.06
0.00
0.16
0.10
0.05
0.01
0.05
0.00
0.06
0.01
0.03
0.00
0.03
"For cities with multiple CO stations, the station with the strongest Spearman correlation was chosen for the
regression.
Percent COHb per mg/m3.
"Percent COHb.
Source: Wallace and Ziegenfus (1985).
March 12, 1990
8-80
DRAFT-DO NOT QUOTE OR CITE
-------
The partial pressure of CO in the arterial blood will reach a steady state value relative to
the partial pressure of CO in the alveolar gas. Therefore, by measuring the end-expired
breath from a subject's lungs, one can measure the end-expired CO partial pressure and, with
the use of the Haldane relationship, estimate the blood level of COHb. This measurement
5 will always be an estimate because the Haldane relationship is based upon attainment of an
equilibrium which does not occur under physiological conditions.
The measurement of CO levels in expired breath to estimate blood levels is based upon
application of the Haldane relationship to gas transfer hi the lung (Equation 8-12). For
example, when the O2 partial pressure is increased in the alveolar gas, it is possible to predict
10 the extent to which the partial pressure of CO will increase in the alveolar gas. This
approach is limited, however, because of the uncertainty associated with variables that are
known to influence gas transfer in the lung and that mediate the direct relationship between
liquid phase gas partial pressures and air phase partial pressures.
The basic mechanisms that are known to influence CO transfer in the lung have been
15 identified through the establishment of the techniques to measure pulmonary diffusion capacity
for CO (Forster, 1964). Some of the factors that can result in decreased diffusion capacity
for CO (altering the relationship between expired CO pressures and COHb levels) are
increased membrane resistance, intravascular resistance, age, alveolar volume, pulmonary
vascular blood volume, pulmonary blood flow, and ventilation/perfusion inequality (Forster,
20 1964). The extent to which each of these variables actually contributes to the variability in
the relationship has not been experimentally demonstrated. There are very few experiments
that focus on the factors leading to variability in the relationship between alveolar CO and
percent COHb at the levels of COHb currently deemed to be of regulatory importance. This
may be due in part to the difficulties in working with analytical techniques, particularly the
25 blood techniques, that are very close to their limits of reproducibility. For example, a change
of approximately 6 ppm of CO in the alveolar gas occurs for every change of 1 % COHb
(Jones et al., 1958). Therefore, in order to reliably measure COHb levels to better than 0.1 %
COHb the analytical method must be able to resolve at least 0.5 ppm CO. This is well within
the range of precision of the electrochemical methods (Lambert et al., 1988; Wallace et al.,
30 1988). Without the use of a well-established method for the measurement of CO levels in
March 12, 1990 8-81 DRAFT-DO NOT QUOTE OR CITE
-------
blood, the influence of all the physiological variables on the accuracy of this method remain
undetermined.
The expired breath method for obtaining estimates of blood levels of CO has a distinct
advantage for monitoring large numbers of subjects because of the non-invasive nature of the
5 method. Other advantages include the ability to obtain an instantaneous reading and the
ability to take an immediate replicate sample for internal standardization. The breathholding
technique for enhancing the normal CO concentration in exhaled breath has been widely used;
however, it should be noted that the absolute relationship between breathhold CO pressures
and blood CO pressures has not been thoroughly established for percent COHb levels below
10 5%. The breathholding method allows time (20 seconds) for diffusion of CO into the
alveolar air so that CO levels are higher than following normal tidal breathing.
Partial pressures of CO in expired breath are highly correlated with percent COHb
levels over a wide range of COHb levels (Table 8-14). The accuracy of the breathhold
method is unknown due to the lack of paired sample analyses of CO partial pressures in
15 exhaled breath and concurrent COHb levels in blood utilizing a sensitive reference method
(see Section 8.5.1). No one has attempted to determine the error of estimate involved in
applying group average regression relationships to the accurate determination of COHb.
Therefore, the extrapolation of breathhold CO partial pressures to actual COHb levels must be
made with reservation until the accuracy of this method is better understood.
20
8.5.2.1 Measurement Methods
Ventilation in healthy individuals involves air movement through areas in the pulmonary
system that are either primarily involved in conduction of gas or in gas exchange in the
alveoli. In a normal breath (tidal volume) the proportion of the volume in the non-gas
25 exchanging area is termed the dead space. In the measurement of CO in the exhaled air, the
dead space gas volume serves to dilute the alveolar CO concentration. Several methods have
been developed to account for the dead space dilution.
March 12, 1990 8-82 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 8-14. SUMMARY OF STUDIES COMPARING END-EXPIRED BREATH CO WITH COHb LEVELS
JO
to
Thesis
Methods
Sample
population
(n)
% COHb
range
Expired
CO range
(ppm)
Blood-breath
relationship
Reference
Developed rebrealhing
method to estimate
COHb from alveolar air
CO concentration
Relationship between
alveolar breath CO and
blood COHb levels
Rebreathing into Douglas bag
Rebreathing method of
Sjostrand (1948)
Blood: venous; van Slyke
23 (sex not 5-35
reported)
55 (men and women; 0-6
smokers and nonsmokers)
COHb
O2Hb
No regression equation
reported; line of fit
as predicted by Haldane
equation
Sjostrand (1948)
Carlsten et al. (19S4)
CO
1
oo
G
?0
P**
H
i
6
o
H
O
H
a
o
«
1
Using lungs as aero-
tonometers, sampling
of alveolar air allows
estimation of COHb
Verify method of
Jones et al. (1958);
apply to community
exposure survey
End-expired breath
measurements can be
used as an indicator
of exposure to ciga-
rette smoking and
community air pollu-
tion
Experimental exposure
study correlating
alveolar breath CO
with venous blood
COHb
20-s breath-hold; save end-
expired sample
Breath: NDIR corrected for CO2
Blood: Venous; NDIR
20-s breath-hold; first few
hundred ml volume discarded;
save end-expired sample
Breath: IR (CO2 scrubbed by
Ascarite)
Blood: Venous; NDIR
Not described
20-s breath-hold; discard first
half expired; save end-
expired
Breath: GC and long path IR
Blood: Venous; GC
13 (men and
women)
4 (men; 2
smokers, 2
nonsmokers)
209 (men, long
shoremen, smokers
and nonsmokers)
14 (men, white,
ages 24 to 42
year)
0.7-26.0 2-185 Line of fit as predicted
by Haldane equation:
«COHb= °-206fCOPpJ
1.2-20.0 3-100 %COHb = O.aiCO^J+O.S
0.2-19.0 0-82 Forrespondents(N=130)
with cardiorespiratory
conditions:
%COHb = l.W+O.HICOppJ
r2 = 0.56
0-32 4-250 %COHb = 109.08
+7.60[COp|DI-11.89
SE = 1.06% COHb
r = 0.976
Jones et al. (1958)
Ringold et al. (1962)
Goldsmith (1965)
Peterson (1970)
-------
s
ar
H-k
K)
TABLE 8-14 (cont'd). SUMMARY OF STUDIES COMPARING END-EXPIRED BREATH CO WITH COHb LEVELS
00
oo
Thesis
Epidemiologic research
investigating tobacco
smoking behavior and
blood COHb levels
Developed practical
method to rapidly
estimate COHb from
breath samples in
field firefighting
situation
Methods
20-s breath-hold; discard first
300 ml; save next 500 ml
expired air
Breath: IR (CO2 scrubbed by
soda lime)
Blood: Venous; Spcctrophotomclric
20-s breath-hold; discard first
portion; save remainder
expired breath
Breath: Electrochemical
(Ecolyzer 2100) and GC
Blood: not described
Sample
population
(n)
59 (men and
women, smokers
and nonsmokers)
56 (men, fire-
fighters)
Expired
% COHb CO range Blood-breath
range (ppm) relationship
0.3-8.1 2-41 No regression equation
reported; estimated
regression from
bivariate plot:
«COHb = 0.21 [COJ
0.8-33 1-239 Line of fit as predicted
by Haldane equation
(without correction for
water vapor pressure):
*COHh -•197*COfpnl
Reference
Rea et al. (1973)
Stewart et al. (1976)
End-expired air
analysis may be used
to distinguish
between populations of
smokers and non-smokers
Breath: IR (CO2 scrubbed by
soda lime)
Blood: Venous; spcclropholo-
inetric (Tietz and Fiereck,
(1973)
14 (sex not
reported)
0.3-8.0
4-46
%COHb = O.lSICO^J-0.26
r2 = 0.92;
95% confidence limits
= ±1% COHb
Rawbone et al. (1976)
O
o
2
0
H
O
a
o
H
W
0
Ambient CO levels
during time of breath-
holding maneuver bias
% COHb estimate
20-s breath-hold; discard first 46 (sex not
portion; save end-expired reported)
Breath: Electrochemical
(Ecolyzer 2000)
Blood: Venous; IL 192 (veri-
fied by unspecified spectro-
photometric technique)
0.4-11.5 2-64 For constant, low
ambient CO environment:
%COHb = 0.18[CO ]
r2 = 0.94
For fluctuating, high
ambient CO environment:
«COHb = O.HfCO^J
r2 = 0.48
Smith (1977)
-------
TABLE 8-14 (cont'd). SUMMARY OF STUDIES COMPARING END-EXPIRED BREATH CO WITH COHb LEVELS
M3
00
oo
Tl
H
6
O
n
i—i
a
Thesis
Mixed-expired air
samples are equiva-
lent to 30-s end-
expired air sample
collection method
Methods
30-s breath-hold and
rebreathing methods
Breath: K (CO2 scrubbed by
soda lime)
Blood: Venous, IL282; verified
by spcctropholometric method
of Tietz and Fiereck (1973)
Sample Expired
population % COHb CO range Blood-breath
(n) range (ppm) relationship Reference
29 (sex not 0.8-10.4 8-62 %COHb = 0.395(00^] Rees et al. (1980)
reported)
(4 non -0.0032([COBJ)2-2.4
smokers, 25
smokers)
End-expired breath
analysis is useful
for estimating %COHb
in traffic control
personnel
Breath: Electrochemical,
Ecolyzer 2000
Blood: Venous; IL282
ND
1.1 -12.5 5-60 Cites Stewart and (1980)
Stewart (1978):
%COHb = 0.202[CO_]
+0.0365
Jabara et al. (1980)
In subjects with
emphysema, decreased
end-expired [CO] is
attributed to impaired
diffusion
End-expired breath
analysis may be used
to distinguish between
smokers and non-smokers
20-8 breath-hold; expire to bag
Breath: Electrochemical,
Ecolyzer 2000
Blood: Venous; IL282
20-s breath-hold; expire to
collection tube
Breath: Electrochemical,
Ecolyzer 2000
Blood: Venous, 1L182
182 smokers
35 emphysema
patients
(sex not reported)
187 (men; 162
smokers, 25 non
smokers)
0.3-14.5 4-90 For normal smokers:
%COHb =
-0.28 + 0. 175 [COppJ
r2 = 0.98;SE = 0.76%COHb
For emphysema patients:
«COHb=
-0.12+0.21 UCO^J
r2 = 0.92
Slopes of two regres-
sion lines were signi-
ficantly different
0.4-13 3-65 *COHb = O.ISICO^J
-0.14
r = 0.97
larvis et al. (1980)
Wald et al. (1981)
-------
i
TABLE 8-14 (cont'd). SUMMARY OF STUDIES COMPARING END-EXPIRED BREATH CO WITH COHb LEVELS
OO
oo
Thesis
To most accurately
estimate % COHb, end-
expired breath samples
require a correction
for inspired ambient
CO at time of sampling
The correction for
inspired air may vary
between persons
Cigarette smoking interferes
with alveolar sampling
Methods
20-s breath-hold; discard first
portion; save end-expired
Breath: Electrochemical,
COED-1 (GE)
Blood: Not sampled
20-s breath-hold at room air
CO level, and al 10, 30,
and 50 ppm CO
20-s breath-hold
Breath: Dt (CO2 scrubbed
before analysis)
Blood: Venous; OSM2
spectrophotomctcr
Sample
population
(n)
1 (male, non
smoker)
7 (sex not
reported)
101 smokers
(42 men;
59 women)
Expired
% COHb CO range Blood-breath
range (ppm) relationship Reference
tCOBJnMMM
-------
25
Mixed Expired Gas Using the Bohr Equation
This technique involves the measurement of the mixed expired CO concentration from
which the alveolar CO concentration is calculated. The Bohr equation used to determine the
physiological dead space is:
Ex
* VE = F^ * VA + Fh * VD (8-13)
10 where:
F& = the fractional concentration of a gas in the mixed expired air,
VE = the minute volume of ventilation or a tidal volume,
F^ = the fractional concentration of the gas in the alveolar space,
15 VA = the volume of alveolar gas,
Ffc = the fractional concentration of gas in the inspired air,
VD = the volume of dead space gas.
20
Solving this equation for CO concentration in the alveolar gas results in:
= (VE* FE - VD* F, )/(VE - VD) (8-14)
CO CO CO
This equation has been used by Rawbone et al. (1976) to describe the relationship
between alveolar CO concentrations and COHb levels. These investigators measured inspired
30 ventilatory volume, inspired CO concentration and estimated dead space from anatomical
correlations. Carbon monoxide concentration must be converted to partial pressure in order
to relate alveolar gas tension to percent COHb. However, the transfer of CO from blood to
the alveolar gas phase is not in equilibrium so the alveolar gas is a reflection of the PO, in the
capillary blood. This is demonstrated by the increase in alveolar CO with breathholding.
35 The relationship between alveolar levels determined from mixed expired CO concentrations
and percent COHb is comparable to that of other methods (Table 8-14).
March 12, 1990 8-87 DRAFT-DO NOT QUOTE OR CITE
-------
Breathhold
Early methods of measurement of CO concentration in air samples by non-
chromatographic techniques required a relatively large sample of gas, usually larger than 1 L.
Therefore, end-tidal (alveolar) gas samples from normal respiration would not provide a
5 sample of sufficient volume for analysis. Jones et al. (1958) developed a method of
inspiration to total lung capacity followed by a breathhold period of various durations. A
breathhold times of 20 s was found to provide near maximal values for CO pressures. The
breathhold period allows more time for diffusion of CO from the blood into the alveolar
space.
10 The precision of the method has been found to be of the order of 0.1 to 0.2 ppm by
several investigators (Hartwell et al., 1984; Wallace et al., 1988; Lambert et al., 1988). This
is the theoretical equivalent of 0.02 to 0.04% COHb. Physiologically, however, the breath-
hold gas is not normal alveolar gas, since this breathhold maneuver results in the CO2
concentration being below normal, with presumably an elevated O2 tension (Jones et al.,
15 1958; Guyattetal., 1988).
The blood-breathhold alveolar air CO relationship is influenced by the inspired pressure
level of CO. Several investigators (Smith, 1977; Wallace, 1983; Wallace et al., 1988) have
found that a correction is required in the CO pressure found in the breathhold sample. This is
an important consideration when this method is used to assess the exposure of subjects in their
20 normal environment (see Section 8.5.2.2).
Rebreathing
The earliest approach to obtaining a sufficient volume of exhaled air was rebreathing
5 L of O2 for two to three minutes while removing the CO2 (Hackney et al., 1962; Carlsten
25 et al., 1954). Hackney et al. (1962) reported that the O2 content in the rebreathing system
fell due to dilution over the first minute after which time the decline in O2 was related to the
O2 consumption of the subject. The CO concentration in the system reached its peak value at
one minute of rebreathing in healthy subjects. Hackney et al. also reported that the CO
concentration in the system was related to the O2 tension in the system. The advantage to
30 using a rebreathing system is that the ratio of change in percent COHb to change in expired
CO is between 27 (Hackney et al., 1962) and 30 ppm/percent COHb (Carlsten et al., 1954).
March 12, 1990 8-88 DRAFT-DO NOT QUOTE OR CITE
-------
This a gain of fivefold over the breathhold method of Jones et al. (1958). The disadvantage
is the time required for the measurement and the need to measure O2 in the system.
Summary of the Methods
5 Kirkham et al. (1988) compared all three techniques for measuring expired CO to
predict percent COHb. The rebreathing and breathhold methods both yield approximately
20% high levels of "alveolar" CO than does the Bohr computation from mixed expired gas.
Subjects rebreathed from a system that contained 20% O2 for the three minutes of the
rebreathing. Kirkham et al. (1988) also carried out an experiment to determine if these
10 techniques had reached a steady state between alveolar gas and blood levels. If a steady state
existed, then changes in ventilation/perfusion and capillary blood volume would not effect the
relationship. Ventilation/perfusion was altered by changing body position from lying to
standing. Both the mixed expired and breathholding techniques showed a significant decline
in the alveolar CO tension when standing. Therefore, measurements of expired CO must be
15 made in the same body position relative to control measurements or reference measurements.
The conventional relationship between blood and expired CO is assumed to be linear
(Table 8-14). Data collected by Rees et al. (1980), however, indicates that the relationship is
not linear. A second order polynomial equation proved to be the best fit of the data where:
percent COHb = 3.95 (CO) - 0.32 (CO)2 - 2.4. Guyatt et al. (1988) also reported a
20 nonlinear relationship where: %COHb = -0.47 + 0.217 (CO) - 0.0006 (CO)2. Peterson
(1970) also found that a quadratic equation described the relationship between FA(CO) and
percent COHb over the range of 0 to 30% COHb. Without more precise data, the
relationship between FA(CO) and COHb for under 5% COHb appears to be sufficiently linear to
justify the use of a linear expression to predict percent COHb from FA(CO) measurements.
25
8.5.2.2 Breath Measurements in Populations
There are numerous approaches described in the literature utilizing the above methods
for the collection and analysis of CO in expired air. In addition, many of the investigators
have also provided data demonstrating a relationship between the concentration of CO in
30 ambient or expired air samples and the percent COHb in blood. All of these approaches show
internally consistent results and are based on the assumption that the air collection
March 12, 1990 8-89 DRAFT-DO NOT QUOTE OR CITE
-------
methodology represents expired alveolar air. In making comparisons, differences in
collection methods, analytical techniques, smoking history, and types of subjects being studied
must be considered. For example, possible subjects include hospital patients with certain
types of medical histories, joggers, and the general population. Sampling locations vary as
5 well, ranging from outdoors to indoors, and from clinics to living rooms.
A study in which breath measurements of CO were used to detect an indoor air problem
has been reported (Wallace, 1983). Sixty-five workers in an office had been complaining for
some months of late-afternoon sleepiness and other symptoms, which they attributed to the
new carpet. About 40 of the workers had their breath tested for CO on a Friday afternoon
10 and again on a Monday morning. The average breath CO levels decreased from 23 ppm on
Friday to 7 ppm on Monday morning (Figure 8-6), indicating a work-related condition. Non-
working fans in the parking garage and broken fire doors were identified as the cause of the
problem. In this case, the ease with which the breath measurements were taken contributed to
the swiftness with which the problem was identified and rectified (Figure 8-7). All
15 measurements were taken in a period of less than two hours, without the necessity for
drawing blood, sterilizing needles, or using a trained phlebotomist.
Wald et al. (1981) obtained measurements of percent COHb for 11,749 men, ages 35 to
64, who attended a medical center in London for comprehensive health screening
examinations between 11am and 5pm. The time of smoking for each cigarette, cigar, or pipe
20 since waking was recorded at the time of collection of a venous blood sample. COHb was
determined using an IL 181 CO-Oximeter. Using 2% COHb as a normal cutoff value, 81%
of cigarette smokers, 35 % of cigar and pipe smokers, and 1 % of nonsmokers were found to
be above normal. An investigation of COHb and alveolar CO was conducted on a subgroup
of 187 men (162 smokers and 25 nonsmokers). Three samples of alveolar air were collected
25 at two-minute intervals within five minutes of collecting venous blood for COHb estimation.
Alveolar air was collected by having the subject hold his breath for 20 s and then
exhale through a one-meter glass tube with an internal diameter of 17 mm and fitted with a
three-liter anesthetic bag at the distal end. Air at the proximal end of the tube was considered
to be alveolar air, and a sample was removed by a small side tube located at 5 mm from the
30 mouthpiece. CO content was measured using an Ecolyzer. The instrumental measurement is
based on detection of the oxidation of CO to CO2 by a catalytically active electrode in an
March 12, 1990 8-90 DRAFT-DO NOT QUOTE OR CITE
-------
15
8
10
1
i
11
Basement office I
I
P
I
•1
f
1
m
Legend
Tfift Alveolar
T3 Ambient
l
n
Control office
J,
I
n
*<*
'x'X
Figure 8-6. Changes in alveolar CO of nonsmoking basement office workers compared to
nonsmoking workers in other offices between Friday afternoon, Monday morning, and
Monday afternoon.
Source: Wallace (1983).
JU
25-
20-
10-
5-
0
Before
After
i
I
EPA 8-hour
CO standard
Illll ,
Period of record, February 8 to March 15
Figure 8-7. Eight-hour average CO concentrations in basement office before and after
corrective action.
Source: Wallace (1983).
March 12, 1990
8-91 DRAFT-DO NOT QUOTE OR CITE
-------
aqueous electrolyte. The mean of the last two readings to the nearest 0.25 ppm was recorded
as the alveolar CO. Subjects reporting recent alcohol consumption were excluded because
ethanol in the breath affects the response of the Ecolyzer. A linear regression equation of
percent COHb on alveolar CO (see Table 8-14) had a correlation coefficient of 0.97,
5 indicating that a COHb level could be estimated reliably from an alveolar CO level.
Honigman et al. (1982) determined alveolar CO concentrations by end-expired breath
analysis for athletes (joggers). The group included 36 nonsmoking males and 7 nonsmoking
females, all conditioned joggers, covering at least 21 mi per week for the previous
six months in the Denver area. The participants exercised for a 40-min period each day over
10 one of three defined courses in the Denver urban environment (elevation 1610 meters).
Expired air samples were collected and analyzed before start of exercise, after 20 min and
again at the end of the 40-min exercise period. Heart rate measurements at 20 min and
40 min were 84 and 82% of mean age predicted maxima, respectively, indicating exercise in
the aerobic range. Relative changes in expired air CO concentrations were plotted and
15 compared to ambient air concentrations for CO measured at the time of collecting breath
samples. Air and breath samples were analyzed using an MSA model 70. Relative changes
in expired end air CO based on the concentration of CO in breath before the start of exercise
were plotted in terms of the ambient air concentrations measured during the exercise period,
at both 20 and 40 min of exercise. For ambient concentrations of CO below 6 ppm, the
20 aerobic exercise served to decrease the relative amount of end air expired CO as compared to
the concentration measured before the start of exercise. For ambient concentrations in the
range of 6 to 7 ppm, there was no net change in the CO concentrations in the expired air.
For ambient air concentrations in excess of 7 ppm, the aerobic exercise resulted in relative
increases of expired CO, with the increases after 40 min being greater than similar increases
25 observed at the 20 min measurements. Sedentary controls at the measurement stations
showed no relative changes. Thus, aerobic exercise, as predicted by the physiologic models
of uptake and elimination, is shown to enhance transport of CO, thereby decreasing the time
to reach equilibrium conditions.
Verhoeff et al. (1983) surveyed fifteen identical residences which used natural gas for
30 cooking and geyser units for water heating. CO concentrations in the flue gases were
measured using an Ecolyzer (2000 series). The flue gases were diluted to the dynamic range
March 12, 1990 8-92 DRAFT-DO NOT QUOTE OR CITE
-------
of the instrument for CO (determined by Draeger tube analyses for CO2 dilution to 2-2.5%).
Theoretical concentrations for CO2 in the flue gasses is 11.70% CO2 under conditions of zero
excess air for the natural gas to air mixture used. Breath samples were collected from
29 inhabitants by having each participant hold a deep breath for 20 seconds and exhale
5 completely through a glass sampling tube (225 mL volume). The sampling tube was
stoppered and taken to a laboratory for analysis of CO content using a gas-liquid
chromatograph (Hewlett Packard, 5880A). The overall coefficient of variation for sampling
and analysis was 7%, based on results of previous measurements. No significant differences
were observed for nonsmokers as a result of their cooking or dishwashing activities using the
10 natural gas fixtures. There was a slight increase in expired air CO for smokers, but this may
be due to the possibility of increased smoking during the dinner hour.
Wallace et al. (1984) report data on measurements of end expired air CO and
comparisons with predicted values based on personal CO measurements for populations in
Denver and Washington, D.C. Correlations between breath CO and preceding eight-hour
15 average CO exposures were high (0.6 to 0.7) in both cities. However, breath CO levels
showed no relationship with ambient CO measurements at the nearest fixed-station monitor.
Correlation coefficients were calculated for one-hour, two-hour, ... up to 10-h average
personal CO exposures; the highest correlations occurred at seven- to nine-hours, providing
support for the EPA choice of 8 h as an averaging time for the NAAQS.
20 The major large-scale study employing breath measurements of CO was carried out by
EPA in Washington and Denver in the winter of 1982-1983 (Johnson 1984; Hartwell et al.,
1984; Akland et al., 1985; Wallace et al., 1984, 1988). In Washington, 870 breath samples
were collected from 812 participants; 895 breath samples were collected from 454 Denver
participants (two breath samples on two consecutive days in Denver). All participants also
25 carried personal monitors to measure their exposures over a 24-h period in Washington or a
48-h period in Denver. The subjects in each city formed a probability sample representing
1.2 million adult nonsmokers in Washington and 500,000 adult nonsmokers in Denver. The
distribution of breath levels in the two cities is shown for the subjects themselves
("unweighted" curves) and the larger populations they represented ("weighted" curves) in
30 Figure 8-8.
March 12, 1990 8-93 DRAFT-DO NOT QUOTE OR CITE
-------
10*
» **** ****?'
•
t
f
1
•
*
7
4
1
a
i
. 'T
- ' i
1-rt
:!'
: i yK-
! ': i
• *(-
. , . .
• t ' •
1 ( •"
1 . .,.
1 ' '"
'•
•4 "i- r
\ I 1
•+ :|
': i :i
DENVER
NASHIN
. 1 ..
• i •.
• • ••
001 008
1
:
• - ' 1 :
! •
i •
•ii i:
: 1 ! ,1
; -j- -7— fr
', ! I
• ' ::
.
i : i
i—trJi
'I i "
'. rr •-
l :
(N-M9)
QTON. DC (
. 1
W W
r irfr
! : 1
44 TTf
li-W
• 1
i: : :
:: ::
i. .1
T — r-
I* • :
•r — "+-
:. r
•*ili
: j |
i:
N"«2
* > • i
• i -i
... !"
»4 . 90 W TO
| 1 ! 1 ! 1 I'M. I"1 1 . : 1 ' 1. '
«> •
•
— r~
—r-]
:
.
6)
r^
t 1
y
01 02 OJ 1 2
\\i
':
!
Ill'
: t
. ^^
£ !
|~T
^tr
:: i
....
f4
: l :
• i :
*X^
/•
TT*
ii
'is*
: 1 1 ;
4
<-?
&'
*:
ili*
; "
|....
.'. cu
MUU
1 : . .
tO 50 40 30 70
.l.i; Ti. — 1. 1 I1 . ! . I !l" !
fr
"ITT
n
-rr
;;;•
kTIVI
5 10 MX
i i
•1::
r^
d
i: :
•rrr
• ;;
EFR
•"-r
T*
^
i
. i .
EQU
lir
T^*~
^
4-\
i
: ::
ENC
i i
-tt~
-r^-
j^K
H
: f :
= ;.
: i:
:::.
1:1 i
V
'.^
w
10
^
yf*
7
.r «
[/
1
: L' i
EIOI
•~rrj
i 2 S
• i '
...*-'
; :
k
Jrt
• • : .
-U-
•fj-r
: 4 ;. .
4TEO
I
^
• y.
V
-4-i—
i i
UNWEIOHTED :
....
.::
1"'
'. 1 .
..
40 90 w » n
. :
A
..
«0 '
U5
i ' :
. . I
~- T^^"
02010
:':< rt/i
,-n
fj
El
:::; :
f. . . {*•&
«
^ '••
....',....
•^TTI
• i
-t— i
M i
— - - T
'
'
I
I
:
'::
.L.
-
'
• i
fi M t*
r •
^
r4-
/-y
y
/
/
,
; ;
yJ
;;l
'::
»
001
T]
— i — i —
: i :
: i :
; ! j
l :
' !
* ) |
1 '
• 1 •
**l*9»
1111
*
/
t
»
4
J
2
1
to«
»•
M
7*
I
8
II
8
Figure 8-8. Distributions of CO in breath of adult nonsmokers in Denver and Washington.
Source: Wallace et al. (1988).
March 12, 1990
8-94 DRAFT-DO NOT QUOTE OR CITE
-------
These distributions appear to be roughly lognormal, with geometric means of 5.2 ppm
CO for Denver and 4.4 ppm for Washington. Geometric standard deviations were about 1.6
for each city. Arithmetic means were 7.1 ppm for Denver and 5.2 ppm for Washington.
Of greater regulatory significance is the number of people whose COHb levels exceeded
5 the value of 2.1 %, because EPA has determined that the current 9 ppm, eight-hour average
standard would keep more than 99.9% of the most sensitive nonsmoking adult population
below this level of protection (Federal Register, 1985). An alveolar CO value of about
10 ppm would correspond to a COHb level of 2%. The percent of people with measured
breath values exceeding this level was about 6% in Washington. This percentage was
10 increased to 10% when the correction for the effect of room air was applied (Figure 8-9). Of
course, since the breath samples were taken on days and at times when they were not
necessarily at their highest level during the year, these percentages are lower limits of the
estimated number of people who may have incurred COHb levels above 2%. Yet the two
central stations in Washington recorded a total of one exceedance of the 9 ppm standard
15 during the winter of 1982-1983. Models based on fixed-station readings would have
predicted that an exceedingly tiny proportion of the Washington population received exposures
exceeding the standard. Therefore, the results from the breath measurements indicated that a
much larger portion of both Denver and Washington residents were exceeding 2% COHb than
was predicted by models based on fixed-station measurements.
20 It also should be noted that the number of people with measured maximum eight-hour
exposures exceeding the EPA outdoor standard of 9 ppm was only about 3.5% of the
Washington subjects. This value appears to disagree with the value of 10% obtained from the
corrected breath samples. However, the personal monitors used in the study were shown to
experience several different problems, including a loss of response associated with battery
25 discharge toward the end of the 24-h monitoring period, which caused them to read low.
Therefore, Wallace et al. (1988) concluded that the breath measurements were correct and the
personal air measurements were biased low. The importance of including breath
measurements in future exposure and epidemiology studies is indicated by this study.
Hwang et al. (1984) describe the use of expired air analysis for CO in an emergency
30 clinical setting to diagnose the presence and extent of CO intoxication. The subjects were
47 Korean patients brought for emergency treatment showing various levels of
March 12, 1990 8-95 DRAFT-DO NOT QUOTE OR CITE
-------
PERCENT OF SAMPLE POPULATION (N?625) EXCEEDING
CONCENTRATION SHOWN
95 90 80 70 8050 40 30 20 10 5 21
50
40
30
20
10
? 8
£ 6
_ EPA 8-h STANDARD
I- 3
H 2h
~Z.
UJ
8 °-8
0.6
0.3
0.2
ai
CORRECTED-
OBSERVED
12 5 10 20 3040506070 80 90 95 9899
CUMULATIVE FREQUENCY (%)
Figure 8-9. Percent of Washington sample population with eight-hour average CO exposures
exceeding the concentrations shown. The eight-hour period ended at the time the breath
sample was taken. The curve marked "OBSERVED" contains the actual readings of the
personal monitors; these readings were corrected using the measured bream values.
Source: Wallace et al. (1988).
March 12, 1990
8-96 DRAFT-DO NOT QUOTE OR CITE
-------
consciousness ranging from alertness (11), drowsiness (21), stupor (7), semicoma (5), coma
(1), and unknown (2). The study group included 16 males, ages 16 to 57, and 31 females,
ages 11 to 62. Exposure durations ranged from two to ten hours, with all exposures
occurring in the evening and nighttime hours. The source of CO was mainly from use of
5 charcoal fires for cooking and heating. In order to estimate expired air CO concentrations, a
detector tube (Gastec ILa containing potassium palado sulfite as both a reactant and color
change indicator for the presence of CO on silica gel) was fitted to a Gastec manual sampling
pump. One stroke of the sampling plunger represents 100 cc of air. A 100-cc expired air
sample was collected by inserting a detector tube at a nostril and slowly pulling back the
10 plunger for one full stroke for expired air. A 10-cc sample of venous blood also was
collected at this time for determination of percent COHb using a CO-Oximeter. The subjects
showed signs of acute intoxication, and two relationships were found: a low CO (less the
100 ppm) and a high CO, (greater than 100 ppm) between expired air CO (CO^) and percent
COHb.
15 Cox and Whichelow (1985) analyzed end-exhaled air (collected over approximately the
last half of the exhalation cycle) for CO concentrations for a random population of 168 adults
- 69 smokers and 99 nonsmokers. The results were used to evaluate the influence of home
heating systems on exposures to and adsorption of CO. Ambient indoor concentrations of CO
were measured in the homes of study subjects. The subjects included 86 men and 82 women,
20 ranging in age from 18 to 74. Interviews were conducted usually in the living room of the
subject's home. The type of heating system in use was noted and indoor air concentration of
CO was measured using an Ecolyzer. After the ambient indoor CO was determined, a breath
sample was collected from the subject. The subject was asked to hold a deep breath for 20 s,
and then to exhale completely into a trilaminate plastic bag. The bag was fitted to the port of
25 the Ecolyzer and the CO content of the exhaled air was measured. For smokers, the time
since smoking their last cigarette and the number of cigarettes per day were noted. For
nonsmokers there was a strong correlation between ambient CO and expired air CO. With
smokers, the correlation was strongest with the number of cigarettes per day. The data also
supported the supposition that smokers are a further source of ambient CO in the indoor
30 environment.
March 12, 1990 8-97 DRAFT-DO NOT QUOTE OR CITE
-------
Lambert et al. (1988) compared breath CO levels to blood COHb levels in 28 subjects
(including two smokers). Breath CO was collected using the standard technique developed by
Jones et al. (1958): maximal inspiration followed by a 20-s breath hold and discarding the
first portion of the expired breath. One-liter bags were used to collect the breath samples,
5 which were measured on an Ecolyzer 2000 monitor equipped with Purafil® and activated
charcoal filters to scrub interferences such as alcohol. Excellent precision (±0.2 ppm) was
obtained in 35 duplicate samples. Blood samples were collected within 15 min of the breath
samples using a gas-tight plastic syringe rinsed with sodium heparin. Carboxyhemoglobin
was measured using an IL 282 CO-Oximeter. Some samples also were measured using a GC.
10 The CO-Oximeter appeared to be reading high, particularly in the <2% COHb range of
interest. A reading of 0.5 %COHb on the particular CO-Oximeter used in this study would
be only 0.3% using the GC and a reading of 1% COHb on the CO-Oximeter would be only
0.7% on the GC.
The results showed poor correlation between the pooled nonsmokers' breath CO and
15 blood COHb levels (n = 104 measurements, r2 = 0.19). However, better correlation was
observed for three individual nonsmoking subjects, who appeared to have roughly parallel
slopes (0.13 to 0.27) but widely differing blood COHb intercepts (0.1, 0.4, and
1.0 %COHb). The authors interpreted these findings as suggesting that the CO-Oximeter
may be sensitive to an unidentified factor in the blood of individuals. Possible factors
20 suggested by the authors include triglycerides and hemochromagens (a group of compounds
formed when heme combines with organic nitrogen species), which are known to absorb light
in the 550 to 555 nm wavelength used by the CO-Oximeter. Another concern regarding the
CO-Oximeter is the calibration method, which uses saturated (98%COHb) bovine serum as
the only span calibration point. This is far above the 0.5 to 3 %COHb range of interest for
25 nonsmoking subjects.
In view of the great dependence in laboratory studies on the CO-Oximeter, the authors
concluded that there was "an important and immediate need to further investigate the
instrument's performance at COHb levels resulting from typical ambient exposures." Such
studies should include a comprehensive side-by-side study with other reference methods,
30 including gas chromatography, manometry, and other spectrophotometric methods. Full
spectral scans should be performed to quantify light absorbance and scattering effects on
March 12, 1990 8-98 DRAFT-DO NOT QUOTE OR CITE
-------
COHb measurement. Also, an improved calibration method should be developed, including
whole blood and dye standards and the use of multiple calibration points in the 0 to
3% COHb range of interest.
5 8.5.3 Potential Limitations
8.5.3.1 Pulmonary Disease
A major potential influence on the relationship between blood and alveolar partial
pressures of CO is the presence of significant lung disease. Hackney et al. (1962)
demonstrated the slow increase in exhaled CO concentration in a rebreathing system peaked
10 after 1.5 minutes in healthy subjects but required 4 minutes in a subject with lung disease.
These findings have been substantiated by Guyatt et al. (1988) who reported that patients with
pulmonary disease did not have the same relationship between percent COHb and breathhold
CO concentrations. The group with pulmonary disease had a FEV/FVC percentage of
<71.5% compared to the healthy subjects with a FEV/FVC percentage of > 86%. The
15 linear regression for the healthy group was COHb = 0.629 + 0.158(ppm CO); and for the
pulmonary disease group was COHb = 0.369 + 0.185(ppm CO). This means that at low
CO levels, individuals with obstructive pulmonary disease would have a lower "alveolar" CO
level for any given percent COHb level than would the healthy subjects.
20 8.5.3.2 Subject Age
The relationship between age and COHb level is not well established. Kahn et al.
(1974) reported that nonsmoking subjects under the age of 19 years had a significantly lower
percent COHb than older subjects but there was no difference in COHb between the ages of
20 and 59 years. Kahn et al. also reported that there was a slight decrease in the COHb
25 levels in nonsmoking subjects over the age of 60 years. Radford and Drizd (1982) also
reported that younger subjects, 3 to 11 years' old, had lower levels of COHb than did the
older age group of 12 to 74 years. Goldsmith (1970) reported that expired CO levels were
unchanged with age in nonsmokers; however, there was a steady decline in the expired CO
levels with age in smokers. The decrease in expired CO is disproportionately large for the
30 decrease in COHb levels measured by Kahn et al. (1974) in older subjects. Therefore, by
comparison of the data from these two studies, it would appear that older subjects have higher
March 12, 1990 8-99 DRAFT-DO NOT QUOTE OR CITE
-------
levels of COHb than predicted from the expired CO levels. It is not known how much of this
effect is due to aging of the pulmonary system, resulting in a condition similar to the subjects
with obstructive pulmonary disease.
5 8.5.3.3 Effects of Smoking
Studies evaluating the effect of cigarette smoking on end-expired CO have found a
phasic response that depends on smoking behavior (Woodman et al., 1987; Henningfield
et al., 1980). There is an initial rapid increase in the CO concentration of expired air as a
result of smoking. This is followed by a rapid (five-minute) decrease after cessation of
10 smoking and a slow decrease over the 5- to 60-min period after smoking. A comparison of
the results from one study (Tsukamoto and Matsuda, 1985) showed that the CO concentration
in expired air increases by approximately 5 ppm by smoking one cigarette. This corresponds
to an increase of 0.67% COHb based on blood-breath relationships developed by the authors.
Use of cigarettes with different tar and nicotine yields or the use of filter tip cigarettes showed
15 no apparent effect on end-expired CO concentrations (Castelli et al.. 1982). However,
knowledge of the breath sample results does. King et al. (1983) were able to show that
immediate feedback on CO concentrations promoted behavioral changes in cigarette smokers
which subsequently resulted in lower CO concentrations in expired air for return visits.
Furthermore, reported rates of smoking were lower for the second visit than those reported
20 for the first visit.
The relationship between breathhold CO and blood CO is apparently altered due to
smoking making the detection of small changes difficult. Guyatt et al. (1988) have shown
that smoking one cigarette results in a variable response in the relationship between breath-
hold CO and COHb levels. The range of FA(CO) values for a 1 % increase in COHb was from -
25 5 ppm to +5 ppm. The correlation between the change in FA(CO) and the change in COHb in
500 subjects was only 0.705. This r value indicates that only 50% of the change in FA(CO) was
due to changes in COHb. It is not known how much of this residual error is due to subject
compliance or to error in the method. Therefore, the results obtained with breathholding in
smoking subjects should be viewed with caution unless large differences in FA(CO) are reported
30 (i.e., considerable cigarette consumption is being evaluated).
March 12, 1990 8-100 DRAFT-DO NOT QUOTE OR CITE
-------
In summary, the measurement of exhaled breath has the advantages of ease, speed,
precision, and greater subject acceptance than measurement of blood COHb. However, the
accuracy of the breath measurement procedure and the validity of the Haldane relationship
between breath and blood at low environmental CO concentrations remains in question.
There appears to be a clear research need to validate the breath method at low CO exposures.
In view of the possible problems with the CO-Oximeter, such validation should be done using
gas chromatography for the blood COHb measurements.
10 8.6 SUMMARY AND CONCLUSIONS
The current NAAQS for CO (9 ppm for eight hours, 35 ppm for one hour) are designed
to protect against actual and potential human exposures in outdoor air that would cause
adverse health effects. Compliance with the NAAQS is determined by measurements taken at
fixed-site ambient monitors, the use of which is intended to provide some measure of the
15 general level of exposure of the population represented by the CO monitors. Results of both
exposure monitoring in the field, and modeling studies, summarized in this Chapter indicate
that individual personal exposure does not directly correlate with CO concentrations
determined by using fixed-site monitors alone. This observation is due to the mobility of
people and to the spatial and temporal variability of CO concentrations. While failing to
20 show a correlation between individual personal monitor exposures and simultaneous nearest
fixed-site monitor concentrations, studies do suggest that aggregate personal exposures are
lower on days of lower ambient CO levels as determined by the fixed-site monitors and
higher on days of higher ambient levels.
Cigarette consumption represents a special case of CO exposure; for the smoker it
25 almost always dominates over personal exposure from other sources. Studies by Radford and
Dridz (1982) show that COHb levels of cigarette smokers average 4% while those of
nonsmokers average 1 %. Therefore, this summary focuses on environmental exposure of
nonsmokers to CO.
People encounter CO in a variety of environments that include travelling in motor
30 vehicles, working at their jobs, visiting urban locations associated with combustion sources,
or cooking over a gas range. Studies of human exposure have shown that among these
March 12, 1990 8-101 DRAFT-DO NOT QUOTE OR CITE
-------
settings the motor vehicle is the most important for regularly encountered elevations of CO.
Studies by Flachsbart et al. (1987) indicated that CO exposures while commuting in
Washington, D.C. average 9 to 14 ppm at the same time that fixed station monitors record
concentrations of 2.7 to 3.1 ppm. Similar studies conducted by EPA in Denver and
5 Washington have demonstrated that the motor vehicle interior has the highest average CO
concentrations (averaging 7 to 10 ppm) of all microenvironments (Johnson, 1984). In these
studies, 8% of all commuters experienced eight-hour exposures greater than 9 ppm while only
1 % of noncommuters received exposures over that level. Furthermore, commuting exposures
have been shown to be highly variable with some commuters breathing CO in excess of
10 25 ppm.
Another important setting for CO exposure is the workplace. In general, exposures at
work exceed CO exposures during nonwork periods, apart from commuting to and from
work. Average concentrations may be elevated during this period since workplaces are often
located in congested areas that have higher background CO concentrations than do many
15 residential neighborhoods. Occupational and nonoccupational exposures may overlay one
another and result in a higher concentration of CO in the blood. Certain occupations also
increase the risk of high CO exposure (e.g., those occupations involved directly with vehicle
driving, maintenance, and parking). Occupational groups exposed to CO by vehicle exhaust
include auto mechanics; parking garage and gas station attendants; bus, truck or taxi drivers;
20 police and warehouse workers. Other industrial processes produce CO directly or as a by-
product, including steel production, coke ovens, carbon black production, and petroleum
refining. Firefighters, cooks, and construction workers also may be exposed at work to
higher CO levels. Occupational exposure in industries or setting with CO production also
represent some of the highest individual exposures observed in field monitoring studies. For
25 example, in EPA's CO exposure study in Washington, of the approximately 4% (29 of 712)
of subjects working in jobs classified as having a high potential for CO exposure, seven
subjects (or approximately 25%) experienced eight-hour CO exposures in excess of 9 ppm.
The highest indoor nonoccupational CO exposures are associated with combustion
sources and include enclosed parking garages, service stations, restaurants and stores. The
30 lowest indoor CO concentrations are found in homes, churches, and health care facilities.
EPA's Denver Study showed that passive cigarette smoke is associated with increasing a
March 12, 1990 8-102 DRAFT-DO NOT QUOTE OR CITE
-------
nonsmoker's exposure by an average of about 1.5 ppm and that use of a gas range is
associated with about 2.5 ppm increase at home. Other sources which may contribute to CO
in the home include combustion space heaters and wood burning stoves.
As noted above, people encounter different and often higher exposures than predicted
5 from fixed-site monitoring data, because of the highly localized nature of CO sources. For
example, during the winter sampling period, 10% of Denver volunteers and 4% of
Washington volunteers recorded personal exposures in excess of 9 ppm for eight hours.
Breath measurements from the Washington volunteers indicated that as much as 9% of the
population could have experienced a 9 ppm, eight-hour average. In contrast, during the
10 entire winter period of 1982-1983, the two ambient CO monitors in Washington reported only
one exceedance of the 9-ppm level. In another study, using data from analyses of COHb in
blood, Wallace and Ziegenfus (1985) report that CO in blood is uncorrelated with CO
measured by ambient monitors. These findings point out the necessity of having personal CO
measurements augment fixed-site ambient monitoring data when total human exposure is to be
15 evaluated. Data from these field studies can be used to construct and test models of human
exposure that account for time and activity patterns known to affect exposure to CO. Models
developed to date tend to underpredict the variability of CO exposures observed in field
studies and have not been able to successfully predict individual exposures. The models may
be modified and adjusted using information from field monitoring studies in order to capture
20 the observed distribution of CO exposures, including the higher exposures found in the tail of
the exposure distribution. The models also are useful for evaluating alternative pollutant
control strategies.
25 8.7 REFERENCES
Akland, G. G.; Ott, W. R.; Wallace, L. A. (1984) Human exposure assessment: background concepts, purpose,
30 and overview of the Washington, DC - Denver, Colorado field studies. Presented at: 77th annual meeting
of the Air Pollution Control Association; June; San Francisco, CA. Pittsburgh, PA: Air Pollution Control
Association; paper no. 84-121.1.
Akland, G. G.; Hartwell, T. D.; Johnson, T. R.; Whitmore, R. W. (1985) Measuring human exposure to carbon
35 monoxide in Washington, D.C., and Denver, Colorado, during the winter of 1982-1983. Environ. Sci.
Technol. 19: 911-918.
March 12, 1990 8-103 DRAFT-DO NOT QUOTE OR CITE
-------
Allred, E. N.; Bleecker, E. R.; Chairman, B. R.; Dahms, T. E.; Gottlieb, S. O.; Hackney, J. D.; Hayes, D.;
Pagano, M. Selvester, R. H.; Walden, S. M.; Warren, J. (1989) Acute effects of carbon monoxide
exposure on individuals with coronary artery disease. Cambridge, MA: Health Effects Institute; research
report no. 25.
5
Amendola, A. A.; Hanes, N. B. (1984) Characterization of indoor carbon monoxide levels produced by the
automobile. In: Berglund, B.; Lindvall, T.; Sundell, J., eds. Indoor air: proceedings of the 3rd
international conference on indoor air quality and climate, v. 4, chemical characterization and personal
exposure; August; Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building Research;
10 pp. 97-102. Available from: NTIS, Springfield, VA; PB85-104214.
Andrecs, L.; Stenzler, A.; Steinberg, H.; Johnson, T, (1979) Carboxyhemoglobin levels in garage workers. Am.
Rev. Respir. Dis. 119: 199.
15 Ayres, S. M.; Criscitiello, A.; Giannelli, S., Jr. (1966) Determination of blood carbon monoxide content by gas
chromatography. J. Appl. Physiol. 21: 1368-1370.
Biller, W. F.; Johnson, T. R.; Paul, R.; Feagans, T. B.; Duggan, G. M.; McCurdy, T. R.; Thomas, H. C.
(1981) A general model for estimating exposure associated with alternative NAAQS. Presented at: 74th
20 annual meeting of the Air Pollution Control Association; June. Pittsburgh, PA: Air Pollution Control
Association; paper no. 81-18.4.
Blackmore, D. J. (1974) Aircraft addicent toxicology: U.K. experience 1967-1972. Aerosp. Med. 45: 987-994.
25 Blankart, R.; Koller, E.; Habegger, S. (1986) Occupational exposure to carbon monoxide and treatment with
hyperbaric oxygenation. J. Clin. Chem. Clin. Biochem. 24: 806-807.
Bondi, K. R.; Very, K. R.; Schaefer, K. E. (1978) Carboxyhemoglobin levels during a submarine patrol.
Undersea Biomed. Res. 5: 17-18.
30
Breysse, P. A.; Bovee, H. H. (1969) Use of expired air-carbon monoxide for Carboxyhemoglobin determinations
in evaluating carbon monoxide exposures resulting from the operation of gasoline fork lift trucks in holds
of ships. Am. Ind. Hyg. Assoc. J. 30: 477-483.
35 Brown, L. J. (1980) A new instrument for the simultaneous measurement of total hemoglobin, %
oxyhemoglobin, % Carboxyhemoglobin, % methemoglobin, and oxygen content in whole blood. IEEE
Trans. Biomed. Eng. 27: 132-138.
Brunekreef, B.; Smit, H. A.; Biersteker, K.; Boleij, J. S. M.; Lebret, E. (1982) Indoor carbon monoxide
40 pollution in The Netherlands. Environ. Int. 8: 193-196.
Butt, J.; Davies, G. M.; Jones, J. G.; Sinclair, A. (1974) Carboxyhaemoglobin levels in blast furnace workers.
Ann. Occup. Hyg. 17: 57-63.
45 Carlsten, A.; Holmgren, A.; Linroth, K.; Sjostrand, T.; Strom, G. (1954) Relationship between low values of
alveolar carbon monoxide concentration and Carboxyhemoglobin percentage in human blood. Acta
Physiol. Scand. 31: 62-74.
Castelli, W. P.; Garrison, R. J.; McNamara, P. M.; Feinleib, M.; Faden, E. (1982) The relationship of carbon
50 monoxide in expired air of cigarette smokers to the tar nicotine content of the cigarette they smoke: the
Framingham study. In: 55th scientific sessions of the American Heart Association: November; Dallas TX.
Circulation 66: 11-315. (American Heart Association monograph no. 91).
March 12, 1990 8-104 DRAFT-DO NOT QUOTE OR CITE
-------
Castellino, N. (1984) Evaluation des tests de dose et d'effet dans 1'exposition professionnelle au monoxyde de
carbone [Evaluation of dose and effect tests for occupational exposure to carbon monoxide]. In:
Hommage Professeur Rene Truhaut; pp. 191-193.
5 Chace, D. H.; Goldbaum, L. R.; Lappas, N. T. (1986) Factors affecting the loss of carbon monoxide from
stored blood samples. J. Anal. Toxicol. 10: 181-189.
Chaney, L. W. (1978) Carbon monoxide automobile emissions measured from the interior of a traveling
automobile. Science (Washington, DC) 199: 1203-1204.
10
Chapin, F. S., Jr. (1974) Human activity patterns in the city. New York, NY: Wiley-Interscience Publishers.
Chovin, P. (1967) Carbon monoxide: analysis of exhaust gas investigations in Paris. Environ. Res. 1: 198-216.
15 Clayton, A. C.; White, S. B.; Settergren, S. K. (1985) Carbon monoxide exposure of residents of Washington,
D. C.: comparative analyses. Research Triangle Park, NC: U. S. Environmental Protection Agency,
Environmental Monitoring Systems Laboratory; contract no. 68-02-3679.
Clerbaux, T.; Willems, E.; Brasseur, L. (1984) Evaluation d'une methode chromatographique de reference pour
20 la determination du taux sanguin de carboxyhemoglobine [Assessment of a reference chromatographic
method for the determination of the carboxyhemoglobin level in blood]. Pathol. Biol. 32: 813-816.
Coburn, R. F.; Danielson, G. K.; Blakemore, W. S.; Forster, R. E., II. (1964) Carbon monoxide in blood:
analytical method and sources of error. J. Appl. Physiol. 19: 510-515.
25
Code of Federal Regulations. (1977) National primary and secondary ambient air quality standards. C. F. R.
40(50) (July 1): 3-33.
Collison, H. A.; Rodkey, F. L.; O'Neal, J. D. (1968) Determination of carbon monoxide in blood by gas
30 chromatography. Clin. Chem. (Winston-Salem, NC) 14: 162-171.
Colwill, D. M.; Hickman, A. J. (1980) Exposure of drivers to carbon monoxide. J. Air Pollut. Control Assoc.
30: 1316-1319.
35 Constantino, A. G.; Park, J.; Caplan, Y. H. (1986) Carbon monoxide analysis: a comparison of two
co-oximeters and headspace gas chromatography. J. Anal. Toxicol. 10: 190-193.
Cooke, R. A. (1986) Blood lead and carboxyhaemoglobin levels in roadside workers. J. Soc. Occup. Med.
36: 102-103.
40
Cortese, A. D.; Spengler, J, D. (1976) Ability of fixed monitoring stations to represent personal carbon
monoxide exposures. J. Air Pollut. Control Assoc. 26: 1144-1150.
Council on Environmental Quality. (1980) The eleventh annual report of the Council on Environmental Quality.
45 Washington, DC: Council on Environmental Quality.
Cox, B. D.; Whichelow, M. J. (1985) Carbon monoxide levels in the breath of smokers and nonsmokers: effect
of domestic heating systems. J. Epidemiol. Commun. Health 39: 75-78.
50 Dahms, T. E.; Horvath, S. M. (1974) Rapid, accurate technique for determination of carbon monoxide in blood.
Clin. Chem. (Winston-Salem, NC) 20: 533-537.
March 12, 1990 8-105 DRAFT-DO NOT QUOTE OR CITE
-------
Davidson, C. I.; Osborn, J. F.; Fortmann, R. C. (1984) Modeling and measurement of pollutants inside houses
in Pittsburgh, Pennsylvania. In: Berglund, B.; Lindvall, T.; Sundell, J., eds. Indoor air: proceedings of
the 3rd international conference on indoor air quality and climate, v. 4, chemical characterization and
personal exposure; August; Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building
5 Research; pp. 69-74. Available from: NTIS, Springfield, VA; PB85-104214.
Davis, G. L.; Gantner, G. E., Jr. (1974) Carboxyhemoglobin in volunteer blood donors. JAMA J. Am. Med.
Assoc. 230: 996-997.
10 Dellarco, M. J.; Wallace, L. A.; Ott, W. R. (1988) A computerized bibliographic literature information system
for total human exposure monitoring research. Presented at: 81st annual meeting of the Air Pollution
Control Association; June; Dallas, TX. Pittsburgh, PA: Air Pollution Control Association; paper
no. 88-115.2.
15 Dennis, R. C.; Valeri, C. R. (1980) Measuring percent oxygen saturation of hemoglobin, percent
carboxyhemoglobin and methemoglobin, and concentrations of total hemoglobin and oxygen in blood of
man, dog, and baboon. Clin. Chem. (Winston-Salem, NC) 26: 1304-1308.
Dockery, D. W.; Spengler, J. D. (1981) Personal exposure to respirable particulates and sulfates. J. Air Pollut.
20 Control Assoc. 31: 153-159.
Douglas, C. G.; Haldane, J. S.; Haldane, J. B. S. (1912) The laws of combination of haemoglobin with carbon
monoxide and oxygen. J. Physiol. (London) 44: 275-304.
25 Drabkin, D. L.; Austin, J. H. (1932) Spectrophotometric constants for common haemoglobin derivatives in
human, dog and rabbit blood. J. Biol. Chem. 98: 719.
Duan, N. (1981) Microenvironment types: a model for human exposure to air pollution. Stanford, CA: Stanford
University, Dept. of Statistics; SIMS technical report no. 47.
30
Duan, N. (1982) Models for human exposure to air pollution. Environ. Int. 8: 305-309.
Duan, N. (1985) Application of the microenvironment monitoring approach to assess human exposure to carbon
monoxide. Research Triangle Park, NC: U. S. Environmental Protection Agency, Environmental
35 Monitoring Systems Laboratory; EPA report no. EPA-600/4-85-046. Available from: NTIS, Springfield,
VA; PB85-228955.
Evans, R. G.; Webb, K.; Homan, S.; Ayres, S. M. (1988) Cross-sectional and longitudinal changes in
pulmonary function associated with automobile pollution among bridge and tunnel officers. Am. J. Ind.
40 Med. 14: 25-36.
Federal Register. (1985) Review of the national ambient air quality standards for carbon monoxide; final rule.
F. R. (September 13) 50: 37484-37501.
45 Federal Register. (1989) Air contaminants; final rule, carbon monoxide. F. R. (January 19) 54: 2651-2652.
Flachsbart, P. G. (1985) Prototypal models of commuter exposure to CO from motor vehicle exhaust. Presented
at: 78th annual meeting of the Air Pollution Control Association; June; Detroit, MI. Pittsburgh, PA: Air
Pollution Control Association; paper no. 85-39.6.
50
Flachsbart, P. G. (1989) Effectiveness of priority lanes in reducing travel time and carbon monoxide exposure.
Inst. Transport. Engin. J. 59: 41-45.
March 12, 1990 8-106 DRAFT-DO NOT QUOTE OR CITE
-------
Flachsbart, P. G.; Ah Yo, C. (1989) Microenvironmental models of commuter exposure to carbon monoxide
from motor vehicle exhaust. Research Triangle Park, NC: U. S. Environmental Protection Agency,
Atmospheric Research and Exposure Assessment Laboratory; in press.
5 Flachsbart, P. G.; Brown, D. E. (1989) Employee exposure to motor vehicle exhaust at a Honolulu shopping
center. J. Architect. Plan. Res. 6: 19-33.
Flachsbart, P. G.; Ott, W. R. (1984) Field surveys of carbon monoxide in commercial settings using personal
exposure monitors. Washington, DC: U. S. Environmental Protection Agency, Office of Monitoring
10 Systems and Quality Assurance; EPA report no. EPA-600/4-84-019. Available from: NTIS, Springfield,
VA;PB84-211291.
Flachsbart, P. G.; Ott, W. R. (1986) A rapid method for surveying CO concentrations in high-rise buildings. In:
Berglund, B.; Berglund, U.; Lindvall, T.; Spengler, J.; Sundell, J., eds. Indoor air quality: papers from
15 the third international conference on indoor air quality and climate; August 1984; Stockholm, Sweden.
Environ. Int. 12: 255-264.
Flachsbart, P. G.; Mack, G. A.; Howes, J. E.; Rodes, C. E. (1987) Carbon monoxide exposures of Washington
commuters. J. Air Pollut. Control Assoc. 37: 135-142.
20
Forster, R. E. (1964) Diffusion of gases. In: Fenn, W. O.; Rahn, H., eds. Handbook of physiology: a critical,
comprehensive presentation of physiological knowledge and concepts. Section 3: Respiration, volume I.
Washington, DC: American Physiological Society; pp. 839-872.
25 Fristedt, B.; Akesson, B. (1971) Haelsorisker av bilavgaser i parkeringshusens serviceanlaeggningar [Health
hazards from automobile exhausts at service facilities of multistory garages]. Hyg. Revy 60: 112-118.
Fugas, M. (1986) Assessment of true human exposure to air pollution. In: Berglund, B.; Berglund, U.; Lindvall,
T.; Spengler, J.; Sundell, J., eds. Indoor air quality: papers from the third international conference on
30 indoor air quality and climate; August 1984; Stockholm, Sweden. Environ. Int. 12: 363-367.
Godin, G.; Wright, G.; Shephard, R. J. (1972) Urban exposure to carbon monoxide. Arch. Environ. Health
25: 305-313.
35 Goldbaum, L. R.; Chace, D. H.; Lappas, N. T. (1986) Determination of carbon monoxide in blood by gas
chromatography using a thermal conductivity detector. J. Forensic Sci. 31: 133-142.
Goldsmith, J. R. (1965) Discussion: epidemiologic studies of chronic respiratory diseases. In: Seventh annual air
pollution medical research conference; February 1964; Los Angeles, CA. Arch. Environ. Health
40 10: 386-388.
Goldsmith, J. R. (1970) Contribution of motor vehicle exhaust, industry, and cigarette smoking to community
carbon monoxide exposures. In: Coburn, R. F., ed. Biological effects of carbon monoxide. Ann. N. Y.
Acad. Sci. 174: 122-134.
45
Goldstein, B. D.; Goldring, R. M.; Amorosi, E. L. (1975) Investigation of mechanisms of oxygen delivery to
the tissues in individuals with chronic low-grade carbon monoxide exposure. New York, NY:
Coordinating Research Council; report no. CRC-APRAC-CAPM-8-68-4. Available from: NTIS,
Springfield, VA; PB-244153.
50
Goldstein, G. M.; Raggio, L.; House, D. (1986) Factors influencing carboxyhemoglobin stability. Research
Triangle Park, NC: U. S. Environmental Protection Agency, Health Effects Research Laboratory; EPA
technical report no. TR-1811. Available from: NTIS, Springfield, VA; AD-A165032.
March 12, 1990 8-107 DRAFT-DO NOT QUOTE OR CITE
-------
Gordon, G. S.; Rogers, R. L. (1969) Project monoxide: a medical study of an occupational hazard of fire
fighters. Washington, DC: International Association of Fire Fighters.
Gothe, C.-J.; Fristedt, B.; Sundell, L.; Kolmodin, B.; Ehrner-Samuel, H.; Gothe, K. (1969) Carbon monoxide
5 hazard in city traffic: an examination of traffic policemen in three Swedish towns. Arch. Environ. Health
19: 310-314.
Grut, A. (1949) Chronic carbon monoxide poisoning: a study in occupational medicine. Copenhagen, Denmark:
Ejnar Munksgaard.
10
Guillot, J. G.; Weber, J. P.; Savoie, J. Y. (1981) Quantitative determination of carbon monoxide in blood by
head-space gas chromatography. J. Anal. Toxicol. 5: 264-266.
Guyatt, A. R.; Kirkham, A. J. T.; Mariner, D. C.; Gumming, G. (1988) Is alveolar carbon monoxide an
15 unreliable index of carboxyhaemoglobin changes during smoking in man? Clin. Sci. 74: 29-36.
Haagen-Smit, A. J. (1966) Carbon monoxide levels in city driving. Arch. Environ. Health 12: 548-551.
Hackney, J. D.; Kaufman, G. A.; Lashier, H.; Lynn, K. (1962) Rebreathing estimate of carbon monoxide
20 hemoglobin. Arch. Environ. Health 5: 300-307.
Hartwell, T. D.; Clayton, C. A.; Ritchie, R. M.; Whitmore, R. W.; Zelon, H. S.; Jones, S. M.; Whitehurst,
D. A. (1984) Study of carbon monoxide exposure of residents of Washington, DC and Denver, Colorado.
Research Triangle Park, NC: U. S. Environmental Protection Agency, Environmental Monitoring
25 Systems Laboratory; EPA report no. 600/4-84-031. Available from: NTIS, Springfield, VA;
PB84-183516.
Heinold, D. W.; Sacco, A. M.; Insley, E. M»(1987) Toolbooth operator exposure on the New Jersey Turnpike.
Presented at: 80th annual meeting of the Air Pollution Control Association; June; New York, NY.
30 Pittsburgh, PA: Air Pollution Control Association; paper no. 87-84A. 12.
Henningfield, J. E.; Stitzer, M. L.; Griffiths, R. R. (1980) Expired air carbon monoxide accumulation and
elimination as a function of number of cigarettes smoked. Addict. Behav. 5: 265-272.
35 Hernberg, S.; Karava, R.; Koskela, R.-S.; Luoma, K. (1976) Angina pectoris, ECG findings and blood pressure
of foundry workers in relation to carbon monoxide exposure. Scand. J. Work Environ. Health 2(suppl.
1): 54-63.
Hickey, R. J.; Clelland, R. C.; Boyce, D. E.; Bowers, E. J. (1975) Carboxyhemoglobin levels. JAMA J. Med.
40 Assoc. 232: 486-488.
Honigman, B.; Cromer, R.; Kurt, T. L. (1982) Carbon monoxide levels in athletes during exercise in an urban
environment. J. Air Pollut. Control Assoc. 32: 77-79.
45 Horvath, S. M.; Roughton, F. J. W. (1942) Improvements in the gasometric estimation of carbon monoxide in
blood. J. Biol. Chem. 144: 747-755.
Horvath, S. M.; Agnew, J. W.; Wagner, J. A.; Bedi, J. F. (1988) Maximal aerobic capacity at several ambient
concentrations of carbon monoxide at several altitudes. Health Effects Institute research report no. 21.
50
Hosey, A. D. (1970) Priorities in developing criteria for "breathing air" standards. J. Occup. Med. 12: 43-46.
March 12, 1990 8-108 DRAFT-DO NOT QUOTE OR CITE
-------
Humphreys, M. P.; Knight, C. V.; Pinnix, J. C. (1986) Residential wood combustion impacts on indoor carbon
monoxide and suspended particulates. In: Proceedings from the 1986 EPA/APCA symposium on the
measurement of toxic air pollutants; April; Raleigh, NC. Research Triangle Park, NC: U. S.
Environmental Protection Agency, Environment Monitoring Systems Laboratory; pp. 736-747; EPA
5 report no. EPA-600/9-86-013. Available from: NTIS, Springfield, VA; PB87-182713.
Hwang, S. J.; Cho, S. H.; Yun, D. R. (1984) [Postexposure relationship between carboxyhemoglobin in blood
and carbon monoxide in expired air]. Seoul J. Med. 25: 511-516.
10 Iglewicz, R.; Rosenman, K. D.; Iglewicz, B.; O'Leary, K.; Hockemeier, R. (1984) Elevated levels of carbon
monoxide in the patient compartment of ambulances. Am. J. Public Health 74: 511-512.
International Committee for Standardization in Haemotology. (1978) Recommendations for reference method for
haemoglobinometry in human blood (ICSH standard EP 6/2: 1977) and specifications for international
15 haemiglobincyanide reference preparation (ICSH standard EP 6/3: 1977). J. Clin. Pathol. 31: 139-143.
Jabara, J. W.; Keefe, T. J.; Beaulieu, H. J.; Buchan, R. M. (1980) Carbon monoxide: dosimetry in occupational
exposures in Denver, Colorado. Arch. Environ. Health 35: 198-204.
20 Jarvis, M. J.; Russell, M. A. H.; Saloojee, Y. (1980) Expired air carbon monoxide: a simple breath test of
tobacco smoke intake. Br. Med. J. 281: 484-485.
Johnson, T. (1984) A study of personal exposure to carbon monoxide in Denver, Colorado. Research Triangle
Park, NC: U. S. Environmental Protection Agency, Environmental Monitoring Systems Laboratory; EPA
25 report no. EPA-600/4-84-014. Available from: NTIS, Springfield, VA; PB84-146125.
Johnson, T. (1987) A study of human activity patterns in Cincinnati, OH. Palo Alto, CA: Electric Power
Research Institute; report no. RP940-06.
30 Johnson, T. (1988) A methodology for estimating carbon monoxide exposure and resulting carboxyhemoglobin
levels in Denver, Colorado. Presented at: Research planning conference on human activity patterns; May;
Las Vegas, NV. Las Vegas, NV: University of Nevada, Las Vegas, Environmental Research Center.
Johnson, T.; Paul, R. A. (1983) The NAAQS exposure model (NEM) applied to carbon monoxide. Research
35 Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards; report no. EPA-450/5-83-003. Available from: NTIS, Springfield, VA; PB84-242551.
Johnson, T.; Wijnberg, L. (1988) A model for simulating personal exposure to carbon monoxide which
incorporates serial correlation. PEI Associates, Inc; report no. 68-02-4606 WA6, PN3766-6.
40
Johnson, B. L.; Cohen, H. H.; Struble, R.; Setzer, J. V.; Anger, W. K.; Gutnik, B. D.; McDonough, T.;
Hauser, P. (1974) Field evaluation of carbon monoxide exposed toll collectors. In: Xintaras, C.; Johnson,
B. L.; de Groot, I., eds. Behavioral toxicology: early detection of occupational hazards: proceedings of a
workshop; National Institute for Occupational Safety and Health and University of Cincinnati; June 1973;
45 Cincinnati, OH. Cincinnati, OH: U. S. Department of Health, Education, and Welfare, Robert A. Taft
Laboratories; National Institute for Occupational Safety and Health report no. 74-126; pp. 306-328.
Available from: NTIS, Springfield, VA; PB-259322.
Johnson, C. J.; Moran, J.; Pekich, R. (1975a) Carbon monoxide in school buses. Am. J. Public Health
50 65: 1327-1329.
Johnson, C. J.; Moran, J. C.; Paine, S. C.; Anderson, H. W.; Breysse, P. A. (1975b) Abatement of toxic levels
of carbon monoxide in Seattle ice -skating rinks. Am. J. Public Health 65: 1087-1090.
March 12, 1990 8-109 DRAFT-DO NOT QUOTE OR CITE
-------
Johnson, T.; Capel, J.; Wijnberg, L. (1986) Selected data analyses relating to studies of personal carbon
monoxide exposure in Denver and Washington, D.C. Research Triangle Park, NC: U. S. Environmental
Protection Agency, Environmental Monitoring Systems Laboratory; contract no. 68-02-3496.
5 Johnson, T.; Paul, R. A.; McCurdy, T. (1988) Advancements in estimating urban population exposure. Presented
at: 81st annual meeting of the Air Pollution Control Association; June; Dallas, TX. Pittsburgh, PA: Air
Pollution Control Association; paper no. 88-127.1.
Jones, J. G.; Walters, D. H. (1962) A study of carboxyhaemoglobin levels in employees at an integrated
10 steelworks. Ann. Occup. Hyg. 5: 221-230.
Jones, R. H.; Ellicott, M. F.; Cadigan, J. B.; Gaensler, E. A. (1958) The relationship between alveolar and
blood carbon monoxide concentrations during breathholding. J. Lab. Clin. Med. 51: 553-564.
15 Kahn, A.; Rutledge, R. B.; Davis, G. L.; Altes, J. A.; Gantner, G. E.; Thornton, C. A.; Wallace, N. D. (1974)
Carboxyhemoglobin sources in the metropolitan St. Louis population. Arch. Environ. Health
29: 127-135.
Kane, D. M. (1985) Investigation of the method to determine carboxyhaemoglobin in blood. Ontario, Canada:
20 Department of National Defence - Canada, Defence and Civil Institute of Environmental Medicine;
DCIEM no. 85-R-32.
Katsumata, Y.; Sato, K.; Yada, S. (1985) A simple and high-sensitive method for determination of carbon
monoxide in blood by gas chromatography. Hanzaigaku Zasshi 51: 139-144.
25
King, A. C.; Scott, R. R.; Prue, D. M. (1983) Reactive effects of assessing reported rates and alveolar carbon
monoxide levels on smoking behavior. Addict. Behav. 8: 323-327.
Kirkham, A. J. T.; Guyatt, A. R.; Gumming, G. (1988) Alveolar carbon monoxide: a comparison of methods of
30 measurement and a study of the effect of change in body posture. Clin. Sci. 74: 23-28.
Lambert, W. E.; Colome, S. D. (1988) Distribution of carbon monoxide exposures within indoor residential
locations. Presented at: 81st annual meeting of the Air Pollution Control Association; June; Dallas, TX.
Pittsburgh, PA: Air Pollution Control Association; paper no. 88-90.1.
35
Lambert, W. E.; Colome, S. D.; Wojciechowski, S. L. (1988) Application of end-expired breath sampling to
estimate carboxyhemoglobin levels in community air pollution exposure assessments. Atmos. Environ.
22: 2171-2181.
40 Lebret, E. (1985) Air pollution in Dutch homes: an exploratory study in environmental epidemiology.
Wageningen, The Netherlands: Department of Air Pollution, Department of Environmental and Tropical
Health; report R-138; report 1985-221.
Lowry, L. K. (1986) Biological exposure index as a complement to the TLV. JOM J. Occup. Med. 28: 578-582.
45
Lundgren, G. R. (1971) Carbon monoxide. In: Encyclopedia of occupational health and safety, v. 1, A-K. New
York, NY: McGraw-Hill Book Company; pp. 253-256.
Lurmann, F. W.; Coyner, L.; Winer, A. M.; Colome, S. (1989) Development of a new regional human exposure
50 (REHEX) model and its application to the California south coast air basin. Presented at: 60th annual
meeting of the Air and Waste Management Association; June; Anaheim, CA. Pittsburgh, PA: Air and
Waste Management Association; paper no. 89-27.5.
March 12, 1990 8-110 DRAFT-DO NOT QUOTE OR CITE
-------
Maas, A. H. J.; Hamelink, M. L.; De Leeuw, R. J. M. (1970) An evaluation of the spectrophotometric
determination of HbO2, HbCO and Hb in blood with the CO-Oximeter IL 182. Clin. Chim. Acta
29: 303-309.
5 McCredie, R. M.; Jose, A. D. (1967) Analysis of blood carbon monoxide and oxygen by gas chromatography. J.
Appl. Physiol. 22: 863-866.
Meyer, B. (1983) Indoor air quality. Reading, MA: Addison-Wesley Publishing Co., Inc.
10 Michelson, W.; Reed, P. (1975) The time budget. In: Michelson, W., ed. Behavioral research methods in
environmental design. Stroudsburg, PA: Dowden, Hutchinson, and Ross; pp. 180-234.
Miranda, J. M.; Konopinski, V. J.; Larsen, R. I. (1967) Carbon monoxide control in a high highway tunnel.
Arch. Environ. Health 15: 16-25.
15
Moschandreas, D. J.; Zabransky, J., Jr. (1982) Spatial variation of carbon monoxide and oxides of nitrogen
concentrations inside residences. Environ. Int. 8: 177-183.
National Academy of Sciences. (1969) Effects of chronic exposure to low levels of carbon monoxide on human
20 health, behavior, and performance. Washington, DC: National Academy of Sciences and National
Academy of Engineering.
National Institute for Occupational Safety and Health. (1972) Criteria for a recommended standard....occupational
exposure to carbon monoxide. Washington, DC: U. S. Department of Health, Education, and Welfare;
25 report no. NIOSH-TR-007-72. Available from: NTIS, Springfield, VA; PB-212629.
National Research Council. (1986) The airliner cabin environment: air quality and safety. Washington, DC:
National Academy Press.
30 Nilsson, C.-A.; Lindahl, R.; Norstrom, A. (1987) Occupational exposure to chain saw exhausts in logging
operations. Am. Ind. Hyg. Assoc. J. 48: 99-105.
Ocak, A.; Valentour, J. C.; Blanke, R. V. (1985) The effects of storage conditions on the stability of carbon
monoxide in postmortem blood. J. Anal. Toxicol. 9: 202-206.
35
Ott, W. R. (1971) An urban survey technique for measuring the spatial variation of carbon monoxide
concentrations in cities. Stanford, CA: Stanford University, Dept. of Civil Engineering.
Ott, W. R. (1977) Development of criteria for siting air monitoring stations. J. Air Pollut. Control Assoc.
40 27: 543-547.
Ott, W. R. (1981) Exposure estimates based on computer generated activity patterns. Presented at: 74th annual
meeting of the Air Pollution Control Association; June; Philadelphia, PA. Pittsburgh, PA: Air Pollution
Control Association; paper no. 81-57.6.
45
Ott, W. R. (1982) Concepts of human exposure to air pollution. Environ. Int. 7: 179-196.
Ott, W. R. (1984) Exposure estimates based on computer generated activity patterns. J. Toxicol. Clin. Toxicol.
21: 97-128.
50
Ott, W. R. (1985) Total human exposure: an emerging science focuses on humans as receptors of environmental
pollution. Environ. Sci. Technol. 19: 880-886.
March 12, 1990 8-111 DRAFT-DO NOT QUOTE OR CITE
-------
Ott, W.; Eliassen, R. (1973) A survey technique for determining the representativeness of urban air monitoring
stations with respect to carbon monoxide. J. Air Pollut. Control Assoc. 23: 685-690.
Ott, W.; Flachsbart, P. (1982) Measurement of carbon monoxide concentrations in indoor and outdoor locations
5 using personal exposure monitors. Environ. Int. 8: 295-304.
Ott, W.; Mage, D. (1974) A method for simulating the true human exposure of critical population groups to air
pollutants. In: proceedings [of the] international symposium, recent advances in assessing the health
effects of environmental pollution, v. IV; June; Paris, France. Luxembourg; Commission of the European
10 Communities; pp. 2097-2107.
Ott, W. R.; Willits, N. H. (1981) CO exposures of occupants of motor vehicles: modeling the dynamic response
of the vehicle. Stanford, CA: Stanford University, Dept. of Statistics; SIMS technical report no. 48.
15 Ott, W. R.; Rodes, C. E.; Drago, R. J.; Williams, C.; Burmann, F. J. (1986) Automated data-logging personal
exposure monitors for carbon monoxide. J. Air Pollut. Control Assoc. 36: 883-887.
Ott, W.; Thomas, J.; Mage, D.; Wallace, L. (1988) Validation of the simulation of human activity and pollutant
exposure (SHAPE) model using paired days from the Denver, CO, carbon monoxide field study. Atmos.
20 Environ. 22: 2101-2113.
Pandian, M. D. (1987) Evaluation of existing total human exposure models. Las Vegas, NV: U. S.
Environmental Protection Agency, Environmental Monitoring Systems Laboratory; EPA report no.
EPA/600/4-87/044. Available from: NTIS, Springfield, VA; PB88-146840.
25
Petersen, W. B.; Allen, R. (1982) Carbon monoxide exposures to Los Angeles area commuters. J. Air Pollut.
Control Assoc. 32: 826-833.
Petersen, G. A.; Sabersky, R. H. (1975) Measurements of pollutants inside an automobile. J. Air Pollut. Control
30 Assoc. 25: 1028-1032.
Peterson, J. E. (1970) Postexposure relationship of carbon monoxide in blood and expired air. Arch. Environ.
Health 21: 172-173.
35 Pierce, R. C.; Louie, A. H.; Sheffer, M. G.; Woodbury, N. L. (1984) The estimation of total human exposure
to pollutants: integrated models for indoor and outdoor exposure to air pollutants. In: Berglund, B.;
Lindvall, T.; Sundell, J., eds. Indoor air: proceedings of the 3rd international conference on indoor air
quality and climate, v. 4, chemical characterization and personal exposure; August; Stockholm, Sweden.
Stockholm, Sweden: Swedish Council for Building Research; pp. 397-402. Available from: NTIS,
40 Springfield, VA; PB85-104214.
Poulton, T. J. (1987) Medical helicopters: carbon monoxide risk? Aviat. Space Environ. Med. 58: 166-168.
Purdham, J. T.; Holness, D. L.; Pilger, C. W. (1987) Environmental and medical assessment of stevedores
45 employed in ferry operations. Appl. Ind. Hyg. 2: 133-139.
Quackenboss, J. J.; Kanarek, M. S.; Spengler, J. D.; Letz, R. (1982) Personal monitoring for nitrogen dioxide
exposure: methodological considerations for a community study. Environ. Int. 8: 249-258.
50 Quinlan, P.; Connor, M.; Waters, M. (1986) Environmental evaluation of stress and hypertension in municipal
bus drivers. Cincinnati, OH: National Institute for Occupational Safety and Health. Available from:
NTIS, Springfield, VA; PB86-144763.
March 12, 1990 8-112 DRAFT-DO NOT QUOTE OR CITE
-------
Radfoid, E. P. (1981) Sensitivity of health endpoints: effect on conclusions of studies. EHP Environ. Health
Perspect. 42: 45-50.
Radford, E. P.; Drizd, T. A. (1982) Blood carbon monoxide levels in persons 3-74 years of age: United States,
5 1976-80. Hyattsville, Md: U. S. Department of Health and Human Services, National Center for Health
Statistics; DHHS publication no. (PHS) 82-1250. (Advance data from vital and health statistics, no. 76).
Rai, V. S.; Minty, P. S. B. (1987) The determination of carboxyhaemoglobin in the presence of
sulfhaemoglobin. Forensic Sci. Int. 33: 1-6.
10
Ramsey, J. M. (1967) Carboxyhemoglobinemia in parking garage employees. Arch. Environ. Health
15: 580-583.
Rawbone, R. G.; Coppin, C. A.; Guz, A. (1976) Carbon monoxide in alveolar air as an index of exposure to
15 cigarette smoke. Clin. Sci. Mol. Med. 51: 495-501.
Rea, J. N.; Tyrer, P. J.; Kasap, H. S.; Beresford, S. A. A. (1973) Expired air, carbon monoxide, smoking, and
other variables: a community study. Br. J. Prev. Soc. Med. 27: 114-120.
20 Rees, P. J.; Chilvers, C.; Clark, T. J. H. (1980) Evaluation of methods used to estimate inhaled dose of carbon
monoxide. Thorax 35: 47-51.
Repace, J. L.; Ott, W. R.; Wallace, L. A. (1980) Total human exposure to air pollution. Presented at: 73rd
annual meeting of the Air Pollution Control Association; June; Montreal, QB, Canada. Pittsburgh, PA:
25 Air Pollution Control Association; paper no. 80-61.6.
Ringold, A.; Goldsmith, J. R.; Helwig, H. L.; Finn, R.; Schuette, F. (1962) Estimating recent carbon monoxide
exposures: a rapid method. Arch. Environ. Health 5: 308-318.
30 Robinson, J. P. (1977) How Americans use their time: a social psychological analysis of everyday behavior. New
York, NY: Praeger Publishers.
Rodkey, F. L.; Hill, T. A.; Pitts, L. L.; Robertson, R. F. (1979) Spectrophotometric measurement of
caroxyhemoglobin and methemoglobin in blood. Clin. Chem. (Winston-Salem, NC) 25: 1388-1393.
35
Rosenman, K. D. (1984) Cardiovascular disease and work place exposures. Arch. Environ. Health 39: 218-224.
Rosenstock, L.; Cullen, M. R. (1986a) Cardiovascular disease. In: Clinical occupational medicine. Philadelphia,
PA: W. B. Saunders; pp. 71-80.
40
Rosenstock, L.; Cullen, M. R. (1986b) Neurologic disease. In: Clinical occupational medicine. Philadelphia, PA:
W. B. Saunders; pp. 118-134.
Roughton, F. J. W.; Root, W. S. (1945) The estimation of small amounts of carbon monoxide in air. J. Biol.
45 Chem. 160: 135-148.
Sammons, J. H.; Coleman, R. L. (1974) Firefighters' occupational exposure to carbon monoxide. J. Occup.
Med. 16: 543-546.
50 Sawicki, C. A.; Gibson, Q. H. (1979) A photochemical method for rapid and precise determination of carbon
monoxide levels in blood. Anal. Biochem. 94: 440-449.
Scholander, P. F.; Roughton, F. J. W. (1943) Micro gasometric estimation of the blood gases: II. carbon
monoxide. J. Biol. Chem. 148: 551-563.
March 12, 1990 8-113 DRAFT-DO NOT QUOTE OR CITE
-------
Schwab, M.; Colome, S. D.; Spengler, J. D.; Ryan, P. B.; Billick, I. H. (1989) J. Ind. Toxicol.: submitted.
Settergren, S. K.; Hartwell, T. D.; Clayton, C. A. (1984) Study of carbon monoxide exposure of residents of
Washington, D. C. - additional analysis. Research Triangle Park, NC: U. S. Environmental Protection
5 Agency, Environmental Monitoring Systems Laboratory; contract no. 68-02-3679.
Sexton, K.; Ryan, P. B. (1988) Assessment of human exposure to air pollution: methods, measurements, and
models. In: Watson, A. Y.; Bates, R. R.; Kennedy, D., eds. Air pollution, the automobile, and public
health. Washington, DC: National Academy Press; pp. 207-238.
10
Sexton, K.; Spengler, J. D.; Treitman, R. D. (1984) Personal exposure to respirable particles: a case study in
Waterbury, Vermont. Atmos. Environ. 18: 1385-1398.
Shackelford, J.; Ott, W.; Wallace, L. (1988) Total human exposure and indoor air quality: an automated
15 bibliography (BLIS) with summary abstracts. Research Triangle Park, NC: U. S. Environmental
Protection Agency, Office of Research and Development; EPA report no. EPA-600/9-88-011.
Simmon, P. B.; Patterson, R. M. (1983) Commuter exposure model: description of model methodology and
code. Research Triangle Park, NC: U. S. Environmental Protection Agency, Environmental Monitoring
20 Systems Laboratory; EPA report no. EPA-600/8-83-022. Available from: NTIS, Springfield, VA;
PB83-215566.
Sisovic, A.; Fugas, M. (1985) Indoor concentrations of carbon monoxide in selected urban microenvironments.
Environ. Monit. Assess. 5: 199-204.
25
Sjostrand, T. (1948) A method for the determination of carboxyhaemoglobin concentrations by analysis of the
alveolar air. Acta Physiol. Scand. 16: 201-210.
Small, K. A.; Radford, E. P.; Frazier, J. M.; Rodkey, F. L.; Collison, H. A. (1971) A rapid method for
30 simultaneous measurement of carboxy- and methemoglobin in blood. J. Appl. Physiol. 31: 154-160.
Smith, N. J. (1977) End-expired air technic for determining occupational carbon monoxide exposure. J. Occup.
Med. 19: 766-769.
35 Spengler, J. D.; Soczek, M. L. (1984) Evidence for improved ambient air quality and the need for personal
exposure research. Environ. Sci. Technol. 18: 268A-280A.
Spengler, J. D.; Stone, K. R.; Lilley, F. W. (1978) High carbon monoxide levels measured in enclosed skating
rinks. J. Air Pollut. Control Assoc. 28: 776-779.
40
Spengler, J. D.; Treitman, R. D.; Tosteson, T. D.; Mage, D. T.; Soczek, M. L. (1985) Personal exposures to
respirable particulates and implications for air pollution epidemiology. Environ. Sci. Technol.
19: 700-707.
45 Stem, F. B.; Lemen, R. A.; Curtis, R. A. (1981) Exposure of motor vehicle examiners to carbon monoxide: a
historical prospective mortality study. Arch. Environ. Health 36: 59-66.
Stewart, R. D.; Baretta, E. D.; Platte, L. R.; Stewart, E. B.; Kalbfleisch, J. H.; Van Yserloo, B.; Rimm, A. A.
(1974) Carboxyhemoglobin levels in American blood donors. JAMA J. Am. Med. Assoc. 229:
50 1187-1195.
Stewart, R. D.; Hake, C. L.; Wu, A.; Stewart, T. A.; Kalbfleisch, J. H. (1976) Carboxyhemoglobin trend in
Chicago blood donors, 1970-1974. Arch. Environ. Health 31: 280-286.
March 12, 1990 8-114 DRAFT-DO NOT QUOTE OR CITE
-------
Stock, T. H.; Kotchmar, D. J.; Contant, C. F.; Buffler, P. A.; Holguin, A. H.; Gehan, B. M.; Noel, L. M.
(1985) The estimation of personal exposures to air pollutants for a community-based study of health
effects in asthmatics - design and results of air monitoring. J. Air Pollut. Control Assoc. 35: 1266-1273.
5 Svercl, P. V.; Asin, R. H. (1973) Home-to-work trips and travel: report no. 8 prepared for the Nationwide
Personal Transportation Study. Washington, DC: U. S. Department of Transportation, Federal Highway
Administration.
Szalai, A., ed. (1972) The use of time: daily activities of urban and suburban populations in 12 countries. The
10 Hague, The Netherlands: Mouton and Co.
Taylor, J. D.; Miller, J. D. M. (1965) A source of error in the cyanmethaemoglobin method of determination of
haemoglobin concentration in blood containing carbon monoxide. Am. J. Clin. Pathol. 43: 265.
15 Tietz, N. W.; Fiereck, E. A. (1973) The spectrophotometric measurement of carboxyhemoglobin. Ann. Clin.
Lab. Sci. 3: 36-42.
Tikuisis, P.; Buick, F.; Kane, D. M. (1987) Percent carboxyhemoglobin in resting humans exposed repeatedly to
1,500 and 7,500 ppm CO. J. Appl. Physiol. 63: 820-827.
20
Traynor, G. W.; Allen, J. R.; Apte, M. G.; Dillworth, J. F.; Girman, J. R.; Hollowell, C. D.; Koonce, J. F.,
Jr. (1982) Indoor air pollution from portable kerosene-fired space heaters, wood-burning stoves, and
wood-burning furnaces. In: Proceedings of the Air Pollution Control Association specialty conference on
residential wood and coal combustion; March; Louisville, KY. Pittsburgh, PA: Air Pollution Control
25 Association; pp. 253-263.
Tsukamoto, H.; Matsuda, Y. (1985) [Relationship between smoking and both the carbon monoxide concentration
in expired air and the carboxyhemoglobin content]. Kotsu Igaku 39: 367-376.
30 Turner, J. A. M.; McNicol, M. W.; Sillett, R. W. (1986) Distribution of carboxyhaemoglobin concentrations in
smokers and non-smokers. Thorax 41: 25-27.
U. S. Environmental Protection Agency. (1979) Air quality criteria for carbon monoxide. Research Triangle
Park, NC: Environmental Criteria and Assessment Office; EPA report no. EPA-600/8-79-022. Available
35 from: NTIS, Springfield, VA; PB81-244840.
van Kampen, E. J.; Zijlstra, W. G. (1961) Standardization of hemoglobinometry. II. The hemiglobincyanide
method. Clin. Chim. Acta 6: 538-544.
40 Van Netten, C.; Brubaker, R. L.; Mackenzie, C. J. G.; Godolphin, W. J. (1987) Blood lead and
carboxyhemoglobin levels in chainsaw operators. Environ. Res. 43: 244-250.
Verhoeff, A. P.; van der Velde, H. C. M.; Boleij, J. S. M.; Lebret, E.; Brunekreef, B. (1983) Detecting indoor
CO exposure by measuring CO in exhaled breath. Int. Arch. Occup. Environ. Health 53: 167-174
45
Virtamo, M.; Tossavainen, A. (1976) Carbon monoxide in foundry air. Scand. J. Work Environ. Health 2(suppl.
1): 37-41.
Vreman, H. J.; Kwong, L. K.; Stevenson, D. K. (1984) Carbon monoxide in blood: an improved microliter
50 blood-sample collection system, with rapid analysis by gas chromatography. Clin. Chem.
(Winston-Salem, NC) 30: 1382-1386.
Vreman, H. J.; Stevenson, D. K.; Zwart, A. (1987) Analysis for carboxyhemoglobin by gas chromatography and
multicomponent spectrophotometry compared. Clin. Chem. (Winston-Salem, NC) 33: 694-697.
March 12, 1990 8-115 DRAFT-DO NOT QUOTE OR CITE
-------
Wade, W. A., Ill; Cote, W. A.; Yocom, J. E. (1975) A study of indoor air quality. J. Air Pollut. Control
Assoc. 25: 933-939.
Wald, N. J.; Idle, M.; Boreham, J.; Bailey, A. (1981) Carbon monoxide in breath in relation to smoking and
5 carboxyhaemoglobin levels. Thorax 36: 366-369.
Wallace, L. A. (1983) Carbon monoxide in air and breath of employees in an underground office. J. Air Pollut.
Control Assoc. 33: 678-682.
10 Wallace, L. A.; Ott, W. R. (1982) Personal monitors: a state-of-the-art survey. J. Air Pollut. Control Assoc.
32: 601-610.
Wallace, L. A.; Ziegenfus, R. C. (1985) Comparison of carboxyhemoglobin concentrations in adult nonsmokers
with ambient carbon monoxide levels. J. Air Pollut. Control Assoc. 35: 944-949.
15
Wallace, N. D.; Davis, G. L.; Rutledge, R. B.; Kahn. A. (1974) Smoking and carboxyhemoglobin in the St.
Louis metropolitan population: theoretical and empirical considerations. Arch. Environ. Health
29: 136-142.
20 Wallace, L. A.; Thomas, J.; Mage, D. T. (1984) Comparison of end-tidal breath CO estimates of COHb with
estimates based on exposure profiles of individuals in the Denver, Colorado and Washington, DC area.
Research Triangle Park, NC: U. S. Environmental Protection Agency, Environmnetal Monitoring
Systems Laboratory; EPA report no. EPA-600/D-84-194. Available from: NTIS, Springfield, VA;
PB84-229822.
25
Wallace, L. A.; Pellizzari, E. D.; Hartwell, T. D.; Sparacino, C. M.; Sheldon, L. S.; Zelon, H. S. (1985)
Results from the first three seasons of the TEAM Study: personal exposures, indoor-outdoor
relationships, and breath levels of toxic air pollutants measured for 355 persons in New Jersey. Presented
at: 78th annual meeting of the Air Pollution Control Association; June; Detroit, MI. Pittsburgh, PA: Air
30 Pollution Control Association; paper no. 85-31.6.
Wallace, L.; Thomas, J.; Mage, D.; Ott, W. (1988) Comparison of breath CO, CO exposure, and Coburn model
predictions in the U.S. EPA Washington-Denver (CO) study. Atmos. Environ. 22: 2183-2193.
35 Wigfield, D. C.; Hollebone, B. R.; Mackeen, J. E.; Selwin, J. C. (1981) Assessment of the methods available
for the determination of carbon monoxide in blood. J. Anal. Toxicol. 5: 122-125.
Woodman, G.; Wintoniuk, D. M.; Taylor, R. G.; Clarke, S. W. (1987) Time course of end-expired carbon
monoxide concentration is important hi studies of cigarette smoking. Clin. Sci. 73: 553-555.
40
Wright, G. R.; Jewczyk, S.; Onrot, J.; Tomlinson, P.; Shephard, R. J. (1975) Carbon monoxide in the urban
atmosphere: hazards to the pedistrian and street worker. Environ. Health 30: 123-129.
Yocom, J. E. (1982) Indoor-outdoor air quality relationships: a critical review. J. Air Pollut. Control Assoc.
45 32: 500-520.
Yocom, J. E.; Clink, W. L.; Cote, W. A. (1971) Indoor/outdoor air quality relationships. J. Air Pollut. Control
Assoc. 21: 251-259.
50 Ziskind, R. A.; Rogozen, M. B.; Carlin, T.; Drago, R. (1981) Carbon monoxide intrusion into sustained-use
vehicles. Environ. Int. 5: 109-123.
Ziskind, R. A.; File, K.; Mage, D. T. (1982) Pilot field study: carbon monoxide exposure monitoring in the
general population. Environ. Int. 8: 283-293.
March 12, 1990 8-116 DRAFT-DO NOT QUOTE OR CITE
-------
Zwart, A.; van Kampen, E. J. (1985) Dyshaemologlobins, especially carboxyhaemoglobin, levels in hospitalized
patients. Clin. Chem. (Winston-Salem, NC) 31: 945.
Zwart, A.; Buursma, A.; Zijlstra, W. G. (1984) Multicomponent analysis of hemoglobin derivatives by means of
5 a reversed optics spectrophotometer. Clin. Chem. (Winston-Salem, NC) 30: 373-379.
Zwart, A.; Van Kampen, E. J.; Zijlstra, W. G. (1986) Results of routine determination of clinically significant
hemoglobin derivatives by multicomponent analysis. Clin. Chem. (Winston-Salem, NC) 32: 972-978.
10
March 12, 1990 8-117 DRAFT-DO NOT QUOTE OR CITE
-------
9. PHARMACOKINETICS AND MECHANISMS OF ACTION
OF CARBON MONOXIDE
Pharmacokinetics in the classical sense has been concerned primarily with the
5 determination of blood levels for various dosage regimens of pharmacological agents. More
modern approaches tend to extend the definition to include other aspects of substance kinetics.
For example, it may include membrane diffusion, substance binding and release
characteristics, modeling, metabolic pathways, and other processes. The general tendency is
to depart from anatomically or physiologically defined region(s) to a more encompassing and
10 unifying concept of compartment(s) comprised of real as well as abstract constructs (Bischoff,
1986). It will be in this sense that the CO pharmacokinetics will be approached and presented
in this chapter.
15 9.1 ABSORPTION, DISTRIBUTION, AND PULMONARY
ELIMINATION
9.1.1 Introduction
The scope of this chapter and generally of this criteria document does not allow for an
extensive review of the mechanisms and factors involved in CO uptake and elimination. The
20 review will concentrate on fundamental processes and the key factors affecting CO
metabolism and resultant effects. For a more in-depth explanation of certain facets of CO
toxicity the reader is referred to other chapters of this document and to other review material
(Fishmanetal., 1987).
25 9.1.2 Pulmonary Uptake
9.1.2.1 Mass Transfer of Carbon Monoxide
Although CO is not one of the respiratory gases, the similarity of the physicochemical
properties of CO and O2 permit an extension of the findings of studies on the kinetics of
transport of O2 to that of CO.
30
March 15, 1990 9-1 DRAFT - DO NOT QUOTE OR CITE
-------
The rate of formation and elimination of COHb, its concentration in blood, as well as its
catabolism is controlled by numerous physical and physiological mechanisms. The relative
contribution of these mechanisms to the overall COHb kinetics will depend on the
environmental conditions (ambient CO concentration, altitude, etc.), physical activity of an
5 individual, and many other physiological processes, some of which are complex and still
poorly understood.
The mass transport of CO between the airway opening (mouth and nose) and red blood
cell (hemoglobin) is predominantly controlled by physical processes. The CO transfer to the
Hb-binding sites is accomplished in two sequential steps: (1) transfer of CO in a gas phase,
10 between the airway opening and the alveoli, and (2) transfer in a "liquid" phase, across air-
blood interface including the red blood cell (RBC). While the mechanical action of the
respiratory system and the molecular diffusion within the alveoli are the key mechanisms of
transport in the gas phase, the diffusion of CO across the alveolo-capillary barrier, plasma,
and RBC is the virtual mechanism of the liquid phase.
15
9.1.2.2 Effects of Dead Space and Uneven Distribution of Ventilation and Perfusion
Ideally, the optimal transfer of gases across alveolo-capillary membrane can be achieved
only if regional distribution of ventilation is uniform and matches regional blood flow.
Numerous studies have shown that in the upright subject ventilation is preferentially
20 distributed to the lower lung zones (Milic-Emili et al., 1966). Besides posture (Clarke et al.,
1969), changes in resting lung volume (Sutherland et al., 1968), airway resistance (Hughes
et al., 1972), and lung compliance (Glaister et al., 1973) by either exogenous factors or
pathophysiological conditions will aggravate maldistribution of ventilation. The unevenness is
further affected by inspiratory (Anthonisen et al., 1970) and expiratory flow rates (Millette
25 et al., 1969), which influence sequential filling and emptying of the lung regions.
Even in perfectly healthy subjects the homogeneity of ventilation, perfusion, and
consequent VA/Q ratio of unity is unattainable because of a right-to-left shunt. Normally, only
a small amount of mixed venous blood (2 to 4%) bypasses the alveoli and reaches systemic
circulation without oxygenation. Any increase in the alveolo-arterial O2 gradient (A-a DOj)
30 will contribute to hypoxemia, thus enhancing CO loading (Riley and Permutt, 1973). It
follows that any imbalance in the distribution patterns of these two compartments must result
March 15, 1990 9-2 DRAFT - DO NOT QUOTE OR CITE
-------
in a decrease in the efficiency of gas exchange (Scrimshire, 1977) including CO. The
average ventilation to perfusion ratio of about 0.9 reported in the upright subjects indicates
that overall perfusion (Q) exceeds ventilation (V) ; regional nommiformity, however, is
considerably greater (the VA/Q ratios range from 0.6 to 3.0; Inkley and Maclntyre, 1973).
5 Consequently, in the underventilated but overperfused regions of the lung the amount of CO
available for diffusion will be less than if the ventilation and perfusion were matched, while
in the overventilated but underperfused regions the amount of CO that could diffuse would be
the same as if the distributions were matched. On exercise, when the distribution of both
ventilation and perfusion becomes more uniform, the ratios approach unity and the rate of
10 COHb formation will accelerate (Harf et al., 1978).
Besides regional inhomogeneity of distribution, the bulk movement of inhaled air will be
influenced by factors related to inspiratory flow and subsequent mixing with residual air. At
rest, mixing of gases is almost complete and no discernible stratification of concentration
between the large airways and the alveoli occurs. However, any changes in ventilation or
15 pattern of breathing (e.g., during exercise) will aggravate stratified inhomogeneity and
increase a concentration gradient between central and peripheral airways. The relative effects
of ventilation and perfusion inhomogeneities on convectional and diffusional transport of CO
will very much depend on the rate of change and concentration of CO in inspired air. The
higher the concentration and the shorter the rise time of CO in the inspired air, the greater the
20 effects these factors will have on the CO uptake and ultimately COHb concentration in blood.
The ventilation-perfusion unevenness will not only contribute to hypoxemia, but the
mismatch will influence the size of the physiological dead space (VD) (Standfuss, 1970) and
ultimately alveolar ventilation, which is one of the principal, but seldom-measured
determinants of the rate of uptake of CO (see Section 9.3 on Coburn-Forster-Kane [CFK]
25 modeling). Any increase in a dead space to tidal volume ratio (VD /VT) will decrease VA and
vice versa. In normal healthy subjects at rest VD comprises about 25 to 45% of tidal volume;
in older subjects or in patients with pulmonary disease the percentage might be as high as
70% (Martin etal., 1979).
30
March 15, 1990 9.3 DRAFT - DO NOT QUOTE OR CITE
-------
9.1.2.3 Alveolo-Capillary Membrane and Blood-Phase Diffusion
While the above mechanisms controlling the rate of formation of blood COHb are
predominantly active processes, the second key mechanism, a diffusion of gases across the
alveolar air-hemoglobin barrier, is an entirely passive process. In order to reach the
5 Hb-binding sites, the CO and other gas molecules have to pass across the alveolo-capillary
membrane, diffuse through the plasma, pass across the RBC membrane, and finally the RBC
stroma before reaction between CO and Hb can take place. The molecular transfer across the
membrane and the blood phase is governed by general physicochemical laws, particularly by
the Pick's first law of diffusion. The exchange and equilibration of gases between the two
10 compartments (air and blood) is very rapid. The dominant driving force is a partial pressure
differential of CO across this membrane. For example, inhalation of a bolus of air containing
high levels of CO will rapidly increase blood COHb; by immediate and tight binding of CO
to Hb the partial pressure of CO within the RBC is kept low, thus maintaining a high pressure
differential between air and blood, and consequent diffusion of CO into blood. Subsequent
15 inhalation of CO-free air progressively decreases the gradient to the point of its reversal
(higher CO pressure on the blood side than alveolar air) and CO will be released into alveolar
air. Because binding of CO to Hb is a much stronger and considerably faster reaction (half-
time <0.07) than clearance of CO by ventilation, the air-blood pressure gradient is usually
higher than the blood-air gradient, and the CO uptake will be a proportionally faster process
20 than CO elimination. The rate of CO release will be further affected by the products of tissue
metabolism. Under pathologic conditions, where one or several components of the air-blood
interface might be severely affected, as in emphysema, fibrosis, or edema, both the uptake
and elimination of CO will be affected.
The rate of diffusion of gases might be altered considerably by many physiological
25 factors acting concomitantly. Diurnal variations in CO diffusion related to variations in Hb
have been reported in normal healthy subjects (Frey et al., 1987). Others found the changes
to be related also to physiological factors such as O2Hb, COHb, partial pressure of alveolar
CO2, ventilatory pattern, oxygen consumption (VOJ, blood flow, functional residual capacity,
etc. (Forster, 1987). It has been confirmed repeatedly that diffusion is body-position and
30 ventilation dependent. In a supine position at rest, CO diffusion has been significantly higher
than that at rest in a sitting position. In both positions CO diffusion during exercise has been
March 15, 1990 9-4 DRAFT - DO NOT QUOTE OR CITE
-------
greater than at rest (McClean et al., 1981). CO diffusion will increase with exercise, and at
maximum work rates the diffusion will be maximal regardless of position. This increase is
attained by increases in both the membrane-diffusing component and the pulmonary capillary
blood flow (Stokes et al., 1981). Diffusion seems to be relatively independent of lung
5 volume within the midrange of vital capacity. However, at extreme volumes the differences
in diffusion rates could be significant; at total lung capacity the diffusion is higher, while at
residual volume it is lower than the average (McClean et al., 1981). Smokers showed on the
average lower diffusion rates than nonsmokers (Knudson et al., 1989).
The above physiological processes will affect minimally COHb formation in healthy
10 individuals exposed to low and relatively uniform levels of CO. Under such ambient
conditions these factors will be the most influential during the initial period of CO distribution
and exchange. If sufficient time is allowed for equilibration, the sole determinant of COHb
concentration in blood will be the ratio of CO to O2. However, the shorter the half-time for
equilibration (e.g., due to hyperventilation, high concentration of CO, increased cardiac
15 output, etc.) the more involved these mechanisms will become in modulating the rate of CO
uptake (Pace et al., 1950; Coburn et al., 1965). At high transient CO exposures of resting
individuals both the cardiac and the lung function mechanisms will control the rate of CO
uptake. Incomplete mixing of blood might result in a substantial difference between the
arterial and venous COHb concentrations (Godin and Shephard, 1972). In chronic
20 bronchitics, asthmatics, and other subpopulations at risk (pregnant women, the elderly, etc.)
the kinetics of COHb formation will be even more complex, because any abnormalities of
ventilation and perfusion and gas diffusion will aggravate CO exchange (see Chapter 12 for
details on subpopulations at risk).
25 9.1.3 Tissue Uptake
Distribution of CO within the tissue(s) will be determined primarily by exchange and
chemical reaction kinetics. In order to facilitate understanding of these well integrated
processes it would be helpful to consider CO uptake by the most involved physiological
compartments/organs.
30
March 15, 1990 9-5 DRAFT - DO NOT QUOTE OR CITE
-------
9.1.3.1 The Blood
Although the rate of CO binding with Hb is about 1/5 slower and the rate of dissociation
from Hb is an order of magnitude slower than the respective rates for O2, the CO chemical
affinity (M) for Hb is about 245 (240 to 250) times greater than that of O2 (Roughton, 1970).
5 One part of CO and 245 parts of O2 would form equal parts of O2Hb and COHb (50% of
each) which would be achieved by breathing air containing 21% oxygen and 570 ppm CO.
Moreover, under steady-state conditions (gas exchange between blood and atmosphere
remains constant), the ratio of COHb to OzHb is proportional to the ratio of their respective
partial pressures. The relationship between the affinity constant M and PO2 and PCO first
10 expressed by Haldane (1898), has the following form.
COHb/O2Hb = M * (PCO/PO,) (9-1)
At equilibrium, when Hb is maximally saturated by O2 and CO at their respective gas
15 tensions, the M value for all practical purposes is independent of pH and 2,3-DPG over a
wide range of PCO/PO2 ratios. The M, however, is temperature dependent (Wyman et al.,
1982).
Under dynamic conditions competitive binding of O2 and CO to Hb is complex; simply
said, the greater the number of hemes bound to CO, the greater is the affinity of free hemes
20 for O2. Any decrease in the amount of available Hb for O2 transport (CO poisoning,
bleeding, anemia, blood diseases, etc.) will reduce the quantity of O2 carried by blood to the
tissue. However, CO not only occupies O2-binding sites, molecule for molecule, thus
reducing the amount of available O2, but also alters characteristic relationships between O2Hb
and PO2 which in normal blood is S-shaped. With increasing concentration of COHb in
25 blood, the dissociation curve is shifted gradually to the left and its shape is transformed into
that of a rectangular hyperbola (Figure 9-1). Because the shift occurs over a critical
saturation range for release of O2 to tissues, a reduction in O2Hb by CO poisoning will have
more severe effects on the release of O2 than the equivalent reduction in Hb due to anemia.
Thus, in an anemic patient (50%) at the tissue PO2 of 26 torr (v',), 5 vol % of O2 (50%
30 desaturation) might be extracted from blood, the amount sufficient to sustain tissue
metabolism. In contrast, in a person poisoned with CO (50% COHb), the tissue PO2 will
March 15, 1990 9-6 DRAFT - DO NOT QUOTE OR CITE
-------
VOLUME PERCENT OXYGEN. mL/100 ml blood
v
PERCENT Hb SATURATION
-------
have to drop to 16 ton (v'2; severe hypoxia) to release the same, 5 vol % O2 (Figure 9-1).
Any higher demand on oxygen under these conditions (e.g., by exercise) might result in coma
of the CO-poisoned individual.
5 9.1.3.2 The Lung
Although the lung in its function as a transport system for gases is exposed continuously
to CO, very little CO actually diffuses and is stored in the lung tissue itself, except for the
alveolar region. The epithelium of the conductive zone (nasopharynx and large airways)
presents a significant barrier to diffusion of CO (Guyatt et al., 1981). Therefore, diffusion
10 and gas uptake by the tissue, even at very high CO concentrations, will be very slow; most of
this small amount of CO will be dissolved in the mucosa of the airways. Diffusion into the
submucosal layers and interstitium will depend very much on the concentration of CO and
duration of exposure. Experimental exposures of the oronasal cavity of monkeys to very high
concentrations of CO for a very short period of time increased their blood COHb level to only
15 1.5%. Comparative exposures of the whole lung, however, elevated COHb to almost 60%
(Schoenfisch et al., 1980). Thus diffusion of CO across the airway mucosa will contribute
very little if at all to overall COHb concentration. In the transitional zone (<20th generation)
where both conductive and diffusive transport take place, diffusion of CO into lung
interstitium will be much easier, and at times more complete. In the respiratory zone
20 (alveoli), which is the most effective interface for CO transfer, diffusion into lung interstitium
will be complete. Because the total lung tissue mass is rather small compared to other CO
compartments, relatively small amount of CO (primarily as dissolved CO) will be distributed
within the lung structures.
25 9.1.3.3 Heart and Skeletal Muscles
The role of myoglobin in O2 transport is not yet fully understood. Myoglobin (Mb) as a
respiratory hemoprotein of muscular tissue will undergo a reversible reaction with CO in a
manner similar to O2. The greater affinity of O2 for myoglobin than hemoglobin (hyperbolic
versus S-shaped dissociation curve) is in this instance physiologically beneficial because a
30 small drop in tissue PO2 will release a large amount of O2 from oxymyoglobin (O2Mb). The
March 15, 1990 9-8 DRAFT - DO NOT QUOTE OR CITE
-------
main function of Mb is thought to serve as a temporary store of O2 and act as a diffusion
facilitator between hemoglobin and the tissues (for details see Section 9.4.2).
Myoglobin has an affinity constant approximately eight times lower than hemoglobin
(M=20 to 40 vs. 245, respectively). As with Hb, the combination velocity constant between
5 CO and Mb is only slightly lower than for O2, but the dissociation velocity constant is much
lower than for O2. The combination of greater affinity (Mb is 90% saturated at PO2 of
20 mmHg) and lower dissociation velocity constant for CO favors retention of CO in the
muscular tissue. Thus, a considerable amount of CO potentially can be stored in the skeletal
muscle. The ratio of carboxymyoglobin (COMb) to COHb saturation for skeletal muscle of a
10 resting dog and cat has been determined to be 0.4 to 0.9; for cardiac muscle the ratio is
slightly higher (0.8 to 1.2) (Coburn et al., 1973; Sokal et al., 1986). Prolonged exposures
did not change this ratio in either muscle indicating that certain, not yet identified,
mechanisms prevented equilibrium between the vascular and extravascular compartments
(Sokal et al., 1984). During exercise the relative rate of CO binding increases more for Mb
15 than for Hb and CO will diffuse from blood to skeletal muscle (Werner and Lindahl, 1980);
consequently, the COMb/COHb will increase for both skeletal and cardiac muscles (Sokal
et al., 1986). A similar shift in CO has been observed under hypoxic conditions because a
fall in intracellular PO2 below a critical level will increase the relative affinity of Mb to CO
(Coburn et al., 1971). Consequent reduction in Mb-carrying capacity of O2 might have a
20 profound effect on the supply of O2 to the tissue (see Section 11.1).
9.1.3.4 Brain and Other Tissues
Apart from Hb and Mb, which are the largest stores of CO, other hemoproteins will
react with CO. However, the exact role of such compounds on O2-CO kinetics still needs to
25 be ascertained (see Section 9.4). Concentration of CO in brain tissue has been found to be
about 30 to 50 times lower than that in blood. During the elimination of CO from brain the
above ratio of concentrations was still maintained (Sokal et al., 1984). (For a more in-depth
discussion see Chapter 10, Section 10.4.)
30
March 15, 1990 9-9 DRAFT - DO NOT QUOTE OR CITE
-------
9.1.4 Pulmonary and Tissue Elimination
An extensive amount of data available on the rate of CO uptake and the formation of
COHb contrast sharply with the limited information available on the dynamics of CO washout
from body stores and blood. Although the same factors that govern CO uptake will affect CO
5 elimination, the relative importance of these factors might not be the same (Landaw, 1973;
Petersen and Stewart, 1970). Both the formation as well as the decline of COHb fit a second-
order function best, increasing during the uptake period and decreasing during the elimination
period. Hence, an initial rapid decay will gradually slow down (Landaw, 1973; Wagner
et al., 1975; Stewart et al., 1970). The elimination rate of CO from an equilibrium state will
10 follow a monotonically decreasing second-order (logarithmic or exponential) function (Pace
et al., 1950). The rate, however, might not be constant following transient exposures to CO,
whereas at the end of exposure the steady-state conditions were not reached yet. In this
situation, particularly after very short and high CO exposures, it is possible that COHb
decline could be biphasic and it can be approximated best by a double-exponential function:
15 The initial rate of decline or "distribution" might be considerably faster than the later
"elimination" phase (Wagner et al., 1975). Reported divergence of COHb decline rate in
blood and in exhaled air suggests that CO elimination rate(s) from extravascular pool(s) is
(are) slower than that reported for blood (Landaw, 1973). Although the absolute elimination
rates are associated positively with the initial concentration of COHb, the relative elimination
20 rates appear to be independent of the initial concentration of COHb (Wagner et al., 1975).
The half-time of CO disappearance from blood under normal recovery conditions while
breathing air showed considerable between-individual variance. For COHb concentration of 2
to 10%, the half-time ranged from 3 to 5 h (Landaw, 1973); others reported the range to be 2
to 6.5 h for slightly higher initial concentrations of COHb (Petersen and Stewart, 1970).
25 Increased inhaled concentration of oxygen accelerated elimination of CO; by breathing 100%
oxygen the half-time was shortened by almost 75% (Petersen and Stewart, 1970). The
elevation of partial pressure of oxygen to three atmospheres reduced the half-time to about
20 min, which is approximately a 14-fold decrease over that seen when breathing room air
(Britten and Myers, 1985; Landaw, 1973). Although the washout of CO can be somewhat
30 accelerated by an admixture of 5% CO2 in O2, hyperbaric O2 treatment is more effective in
facilitating displacement of CO.
March 15, 1990 9-10 DRAFT - DO NOT QUOTE OR CITE
-------
9.2 TISSUE PRODUCTION AND METABOLISM OF CARBON
MONOXIDE
In the process of natural degradation of hemoglobin to bile pigments, a carbon atom (a-
bridge C) is separated from the porphyrin nucleus and subsequently is catabolized by
5 microsomal heme oxygenase into CO. The major site of heme breakdown and therefore the
major production organ of endogenous CO is the liver (Berk et al., 1976). The spleen and
the erythropoietic system are other important catabolic generators of CO. Because the amount
of porphyrin breakdown is stoichiometrically related to the amount of endogenously formed
CO, the blood level of COHb or the concentration of CO in the alveolar air has been used
10 with mixed success as quantitative indices of the rate of heme catabolism (Landaw and
Callahan, 1970; Solanki et al., 1988). Not all of endogenous CO comes from RBC
degradation. Other hemoproteins, such as myoglobin, cytochromes, peroxidases, and catalase
contribute approximately 20 to 25% to the total amount of generated CO (Berk et al., 1976).
Approximately 0.4 mL/h of CO is formed by hemoglobin catabolism and about 0.1 mL/h
15 originates from nonhemoglobin sources (Coburn et al., 1964). Metabolic processes other than
heme catabolism contribute only a very small amount of CO (Miyahara and Takahashi, 1971).
In both males and females, week-to-week variations of CO production are greater than day-
to-day or within-day variations. Moreover, in females COHb levels fluctuated with the
menstrual cycle; the mean rate of CO production in the premenstrual, progesterone phase
20 almost doubled (Lynch and Moede, 1972; Delivoria-Papadopoulos et al., 1970). Neonates
and pregnant women also showed a significant increase in endogenous CO production related
to increased breakdown of RBC.
Any disturbance leading to increased destruction of RBC and accelerated breakdown of
other hemoproteins would lead to increased production of CO. Hematomas, intravascular
25 hemolysis of RBC, blood transfusion, and ineffective erythropoiesis all will elevate CO
concentration in blood. Degradation of RBC under pathologic conditions such as anemias
(hemolytic, sideroblastic, sickle cell), thalassemia, Gilbert's syndrome with hemolysis, and
other hematological diseases also will accelerate CO production (Berk et al., 1974; Solanki
et al., 1988). In patients with hemolytic anemia the CO production rate was 2 to 8 times
30 higher, and blood COHb concentration 2 to 3 times higher than in normals (Coburn et al.,
1966). Increased CO-production rates have been reported after administration of
March 15, 1990 9-11 DRAFT - DO NOT QUOTE OR CITE
-------
phenobarbital, diphenylhydantoin (Coburn, 1970), and progesterone (Delivoria-Papadopoulos
etal., 1970).
5 9.3 MODELING CARBOXYHEMOGLOBIN FORMATION
9.3.1 Introduction
The NAAQS for CO were designed to establish ambient levels of CO which would
protect sensitive individuals from experiencing adverse health effects. In retaining the current
CO primary standards, both EPA and the Clean Air Scientific Advisory Committee concluded
10 that the critical effects level for NAAQS-setting purposes was approximately 3% COHb
without including a margin of safety (Federal Register, 1985). Using exposure modeling and
available monitoring data, EPA estimated that the current 9 ppm, eight-hour average standard
would keep more than 99.9% of the adult population with cardiovascular disease below 2.1%
COHb. Considering uncertainties regarding the lowest level at which adverse health effects
15 may occur, as well as uncertainties about the exposure estimates, EPA concluded that this
level of protection would provide an adequate margin of safety for this sensitive group.
Because of the variability of ambient CO concentration profiles, and other exogenous and
endogenous factors affecting formation of COHb in an individual, it is obvious that the only
practical approach to evaluate the protection provided by these standards is to continue to use
20 mathematical models. The COHb formation modeling, however, has much wider application
because the quantification of the relationship between exogenous CO and blood COHb is also
of clinical and occupational interest.
9.3.2 Regression Models
25 The most direct approach to establishing a prediction equation for COHb is to regress
observed COHb values against the level and duration of exogenous CO exposure. Inclusion
of other predictor variables such as initial COHb level and alveolar ventilation generally will
improve the precision of the predictions. All regression models are purely empirical and have
no physiological basis. Their applicability therefore is limited to the exact conditions that
30 were used to collect the data on which they are based. So far, the most viable models have
been tested and used to estimate COHb levels for healthy subjects only. No validation studies
March 15, 1990 9-12 DRAFT - DO NOT QUOTE OR CITE
-------
have been reported on potentially at-risk subpopulations (see Chapter 12), such as patients
with cardiovascular or hematologic dysfunction.
Peterson and Stewart (1970) developed regression Equation 9-2 for percent COHb after
exposure to moderate CO levels, where CO refers to the concentration of CO in ambient air
5 inhaled parts per million, t is the exposure duration in minutes, and t' is the postexposure
time in minutes. The final term (-.000941') reflects CO elimination and was computed using
Log10 %COHb = .85753 Log,0 CO + .62995 Log,01 - 2.29519 -.000941' (9-2)
10 the average COHb half-life found in the study. The percent COHb in the blood samples was
determined twice, using an IL CO-Oximeter and a gas chromatograph. The percent COHb
values that were used to estimate the equation were themselves averages over observations on
2 to 10 subjects (r = 0.985). The range of CO concentrations used was 25 to 523 ppm CO,
and the exposures lasted from 15 min to eight hours. The subjects were 18 healthy males that
15 did not smoke during the duration of the study. More recently Equation 9-2, without its final
term, was modified by Zankl (1981) to correct the time, t, in the equation for altitude and
subject activity level. No justification, however, nor reference was cited for these changes.
Another regression equation (9-3) developed by Stewart et al. (1973) applies to briefer
exposures of considerably higher levels of CO. In this study the exposures ranged from
20 1000 ppm (for 10 min) to 35,600 ppm (for 45 sec). The regression equation was based on
Log10[%COHb(t)] = Log10[%COHb(t)] + Log10[%COHb(0] (9-3)
+ 1.036 Log10 CO-4.4793
25 + Log10 (liters inhaled)
13 experimental exposures but only on six different subjects (r = 0.995). The subjects
remained sedentary throughout the study. Possible correlations between readings on the same
30 subject were not taken into account. The predicted quantity is the logarithm of the "increase
in percent COHb saturation in venous blood per liter of CO mixture inhaled." The percent
COHb in the blood samples was determined twice, using automated blood analyzing system
and a gas chromatograph. The increase in COHb saturation was computed using the peak
March 15, 1990 9-13 DRAFT - DO NOT QUOTE OR CITE
-------
COHb concentration occurring approximately two minutes after CO exposure stopped.
However, the immediate postexposure inhalation of pure O2 almost certainly lowered the peak
COHb values and influenced subsequent estimates.
5 9.3.3 The Coburn-Forster-Kane Differential Equations
In 1965, Coburn, Forster, and Kane developed a differential equation to describe the
major physiological variables that determine blood [COHb] for the examination of the
endogenous production of CO. The equation, referred to as the CFK model, is still much in
use today for the prediction of [COHb] consequent to inhalation of CO for two reasons.
10 First, the model is quite robust to challenges to the original assumptions. Second, the model
can be relatively easily adapted to more specialized applications.
9.3.3.1 Linear and Nonlinear CFK Differential Equations
Equation 9-4 represents the linear CFK model, with constant O2Hb level:
15
VB d[COHb]/dt = Vco - [COHb]PcO2 / MB[02Hb] + P.CO/B (9-4)
where:
20 [COHb] = milliliters of CO per milliliter of blood, maximum O2 capacity of
blood
[O2Hb] = milliliters of O2 per milliliter of blood (=0.2)
25 Hb = grams hemoglobin per milliliter of blood, hemoglobin concentration in
blood (=1.38)
B = 1/DLCO + PL/VA
30 DLCO = milliliters per minute per millimeter of Hg, pulmonary diffusing
capacity for CO (=30)
PL = millilmeters of Hg, pressure dry gases in the lungs (=713)
35 VA = milliliters per minute alveolar ventilation rate (=6000)
M = Haldane affinity ratio (=218)
VB = millimeters of blood volume (=5500)
March 15, 1990 9-14 DRAFT - DO NOT QUOTE OR CITE
-------
v
co = milliliters per minute, endogenous CO production (=0.007)
PIcO = millimeters of Hg, partial pressure CO in air inhaled
PCO2 = millimeters of Hg, average partial pressure of O2 in lung capillaries
(=100)
10 The values in parentheses are the values given in Peterson and Stewart (1970), although it is
not clear whether a consistent set of conditions (i.e., BTPS or STPD) was used. Restricting
the conditions to low CO exposures allows the mathematical assumption of instant equili-
bration of (1) the gases in the lungs, (2) COHb concentrations between venous and arterial
blood, and (3) COHb concentrations between the blood and CO stores in nonvascular tissues.
15 In addition, the washout time becomes unimportant, and the inhaled and exhaled volumes
could be presumed equal. In solving Equation 9-4, Coburn, Forster, and Kane (1965) further
assumed that [O2Hb] is constant and not dependent on [COHb]. The resulting linear
differential equation is restricted to relatively low COHb levels. For high ambient CO levels,
it may erroneously predict equilibrium values greater than 100% COHb.
20 The advantage to using the linear differential equation (where applicable) is that the
solution can be written explicitly as.
[COHb] (t) = [COHb]0e'At + C/A (1 - e'*) (9-5)
where
25
A = PCO2 /V, MB [O, Hb]
C = VCO/VB + P, CO /V, B
From this solution, we see that for small t, the formation of COHb proceeds linearly, as
30
A [COHb] * P, CO (t) /V, B (9-6)
March 15, 1990 9-15 DRAFT - DO NOT QUOTE OR CITE
-------
Since O2 and CO combine with Hb from the same pool, higher COHb values do affect
the amount of Hb available for bonding with O2. Such interdependence can be modeled by
substituting (1.38 Hb - [COHb]) for [OjHb]) (see e.g., Tikuisis et al., 1987b). The CFK
differential equation then becomes nonlinear and iterative methods or numerical integration
5 must be used to solve the equation (Muller and Barton, 1987). Solutions of either CFK
equation require that the volumes of all gases be adjusted to the same conditions. Coburn,
Forster, and Kane (1965) use STPD conditions, but the equation can be solved under any
conditions if consistently used (Tikuisis et al., 1987a,b).
The equilibrium value predicted by the nonlinear differential equation will always be
10 less than 100% COHb, and is given by the following expression.
[COHb] = 1.38 Hb M(P,CO + BVo,) / PCO2 + M (P,CO +BVCO (9-7)
A sensitivity analysis has been done on the parameters of both the linear and nonlinear
15 CFK equation at five different work levels (McCartney, 1990). The author shows that a 1%
error in any one of the parameters produces no more than a 1 % error in COHb prediction by
the nonlinear model.
The nonlinear CFK model is more accurate physiologically, but has no explicit solution.
It is reasonable, therefore, to ask under what conditions the solutions to the linear and
20 nonlinear equations are "close" together. Because both solutions are generated by known
differential equations, the question is a purely mathematical one. The precise answer is
complex and depends on the ambient CO level. In general, the linear CFK differential
equation is a better approximation to the nonlinear equation during the uptake of CO than
during the elimination of CO. The approximation also is better for COHb levels further from
25 the equilibrium predicted by the nonlinear model. In particular, it can be shown that as long
as the linear CFK equation predicts COHb levels at or below 6% COHb, the solution to the
nonlinear CFK model will be no more than 0.5% COHb away (Smith et al., 1990).
9.3.3.2 Confirmation Studies of the CFK Model
30 Since the publication of the original paper (Coburn et al., 1965), other investigators
have tested the fit of the CFK model to experimental data by using different exposure profiles
March 15, 1990 9-16 DRAFT - DO NOT QUOTE OR CITE
-------
and different approaches to evaluating the parameters of the model. Stewart et al. (1970)
and Peterson and Stewart (1970) tested the CFK linear differential equation on 18 resting
subjects exposed to 25 CO exposure profiles for periods of one-half to 24 hours and to CO
concentrations ranging from 1 to 1000 ppm. All physiological coefficients were assumed (see
5 p. 9-14). The percent COHb in the blood samples was determined twice, using an IL CO-
Oximeter and a gas chromatograph. It is important to note that in this experiment the
predictions were compared to individual observations instead of averages. The predicted
values yielded COHb values quite close to the measured values. The greatest discrepancy
(4.9%) was observed in the experiment with steadily rising inhaled CO concentration over a
10 two-hour period, which is not surprising because the authors assumed a constant inhaled
concentration of CO in solving the CFK equation.
In 1975, Peterson and Stewart presented a second series of experiments testing the
nonlinear CFK model. Three women were included among the 22 subjects, and three
different levels of exercise were used. The parameter values of PCO2, DLCO, VB, and VA
15 were estimated for each subject. The percent COHb in the blood samples was determined by
a CO-Oximeter that was continually compared to a gas chromatograph. Based on summary
data, they concluded that the predicted and measured values were very close for both males
and females under conditions at rest and exercise as well.
In 1981 Joumard et al. tested both the linear and nonlinear CFK models for CO uptake
20 and elimination in pedestrians and car passengers exposed to ambient CO levels in the city of
Lyon, France. The cohort, consisting of 37 male and 36 female nonsmoking subjects who
were 18 to 60 years old, was divided into two groups. One group was driven around the city
in cars while the second group walked on the street at a nearly uniform pace. Each journey
lasted about two hours. Blood COHb readings were taken at the beginning and end of each
25 journey. The percent COHb in the blood samples was determined by infrared spectroscopy.
All other physiological parameters were estimated. As might be expected at these COHb
levels (-2.3%), the authors found no significant difference between the linear and nonlinear
CFK equations. No significant difference (a = 0.05) was found between the final predicted
and observed COHb values except for male pedestrians. The unspecified difference for that
30 group was attributed to an underestimate of the alveolar ventilation.
March 15, 1990 9-17 DRAFT - DO NOT QUOTE OR CITE
-------
In 1984, Hauck and Neuberger ran a series of experiments testing the predictive ability
of the CFK model on four subjects exposed to a total of 10 different CO exposure profiles
combined with a variety of exercise (bicycle ergometer) patterns so that each exposure was a
unique combination of CO concentration and exercise pattern. The group, all nonsmokers,
5 included three adult males and one ten-year-old female. The COHb values were calculated at
1-min intervals using a numerical solution of the CFK model; all but ventilation-derived
parameters, which were updated every minute, were kept constant. The percent COHb in the
blood was determined by an improved van Slyke method. The maximal differences within
each experimental run (expressed as percent of a maximal predicted value) ranged from 4.2 to
10 11.1%.
The most recent validation of the nonlinear CFK model was reported by Tikuisis et al.
(1987a,b). Experiments were completed on 6 to 11 nonsmoking middle-aged males. All of
the CFK parameters but DLCO and VA were estimated; DLCO and VA were measured for each
subject. The percent COHb in the blood samples was determined by gas chromatography.
15 Several transient intermittent CO exposure profiles were tested: 1500 ppm for 5 min, and
7500 ppm for 1 min at rest along with stepwise symmetric profiles of 500 to 4000 ppm for
4.5 min and 4000 ppm for 75 during rest and intermittent exercise (VA « 30L/min;
Figure 9-2). On an average, the predicted and measured values at rest were very close, with
the CFK model slightly overpredictive (<0.5% COHb). This overprediction was greater
20 during exercise, reaching almost 3% COHb in one of the subjects (Figure 9-2). It is of
interest to note that predicted values based on a current National Institute for Occupational
Safety and Health (NIOSH) solution of the CFK model are even higher, overpredicting by as
much as 6% COHb. The model appeared to be most sensitive to VA; thus errors in
conversion of gas volumes (e.g., from ATPS to BTPS) will affect the predicted values.
25
9.3.3.3 Modified CFK Models
Bernard and Duker (1981) simplified the linear form of the CFK model in a unique
way. Using regression equations from the literature, they were able to relate physiological
parameters to the O2 uptake by the body (VOj), which in turn related to an activity level. A
30 linear relationship was assumed between the rate of O2 uptake and the maximum COHb level
March 15, 1990 9-18 DRAFT - DO NOT QUOTE OR CITE
-------
%COHb
20^
15-
10-
5
0
20-
15-
10-
5-
0
20-
15-
10
5
0
Subject RV
Subject DH
Subject JG
Subject RP
4487
Subject MB
5036 4956
J
Subject RE
4945 4923
-4000
-3000
-2000
-1000
0
-4000
-3000
ppmCO
-2000
-1000
0
-4000
-3000
2000
1-1000
0 5 10 15 20 25 30 35 40 45 50 0 5 10 15 20 25 30 35 40 45 50
Time, mm.
Figure 9-2. Measured and predicted COHb concentrations from six intermittently exercising
subjects. The solid lines represent the measured percent COHb; the short-dashed lines are the
solutions to the nonlinear CFK equation; and the long-dashed lines are predicted values based
on the CFK model adapted by NIOSH.
Source: Tikuisis et al. (1987b).
March 15, 1990
9-19 DRAFT - DO NOT QUOTE OR CITE
-------
under which that rate could be sustained. A summary of predictive relationships between
pairs of variables were developed, but none were experimentally tested.
A more fundamental modification of the CFK model was made by Hill et al. (1977) to
study the effect of CO inspired by the mother on the level of fetal COHb. The Hill
5 equation (9-8) combines the CFK equation (for maternal COHb), with a term denoting COHb
transfer from the placenta into the fetus (the subscripts m and f denote maternal and fetal
quantities, respectively).
Vta d[COHbJ/dt = Vco* - [COHbJ PcO2/(Mm[O2Hb]B) (9-8)
10 + P,CO/B - DPCO (PmCO - P£0)
Thus, Equation 9-8 is the same as Equation 9-4, except for the final term on the right.
15 DPCO is the CO diffusion coefficient across the placenta. PmCO and P£O are the partial
pressures of CO in the maternal and fetal placenta! capillaries, respectively. The latter two
quantities are estimated using the Haldane relationship and separate models for the lungs and
placenta. The level of fetal COHb is predicted from a similar equation. Comparative
evaluation of predicted and measured fetal COHb concentrations under time-varying and
20 steady-state conditions in both men and animals showed acceptable agreement only under
steady-state conditions (Hill et al., 1977; Longo and Hill, 1977).
9.3.3.4 Application of the CFK Model
Ott and Mage (1978), using a linear differential equation model that was patterned after
25 the linear CFK differential equation, examined the dynamics of blood COHb concentration
fluctuation as a function of ambient CO concentration for a one-year period. Other
parameters of the model were estimated and kept constant The calculated COHb levels
exceeded 2% on 25 occasions without violating the one-hour standard, whereas the eight-
hour standard was violated 23 times. During the same year 29 violations of the CO standard
30 occurred but the 2% COHb level was exceeded in only 23 instances. Besides evaluation of
the averaged CO concentrations, the authors examined the effects of peak, transient CO
concentrations on the target COHb. They showed that the presence of such spikes in CO data
averaged over hourly intervals may lead to underestimating the COHb level (due to exogenous
March 15, 1990 9-20 DRAFT - DO NOT QUOTE OR CITE
-------
CO) by as much as 21%. Consequently, they recommended that monitored CO be averaged
over shorter periods, such as 10 to 15 min. (See Chapter 8 for a more complete description
of population exposure to CO.)
Venkatram and Louch (1979) extended the above application to more dynamic
5 conditions by fitting interpolated values of the ambient one-hour CO averages from Toronto,
Canada into the CFK model. In addition, they reexpressed the solution of the model from
units of percent COHb to parts per million of CO. Such a transformation allows the
examination of a variety of CO concentration profiles, while keeping a simple preselected
target COHb as a constant. They calculated that a 2% COHb level in blood very likely
10 would be exceeded on numerous occasions without ever violating the standard. By including
transients their approach appears to predict COHb more accurately, particularly in response to
eight-hour running averages.
Biller and Richmond (1982) investigated the effects of inhaling various patterns of
hourly-averaged CO concentrations that just attained alternative 1-hour and 8-hour CO
15 NAAQS using the CFK equation. Their analysis also estimated the distributions of various
physiological parameters that are inputs to the CFK equation for individuals with
cardiovascular disease. The authors found that depending on which air quality pattern was
used, the percentage of the population exceeding 2.1% COHb ranged from less than 0.01 % to
10%.
20 More recently, Saltzman and Fox (1986) investigated the effect of inhaling oscillating
levels of CO on the COHb level of rabbits using the linear CFK equation simplified by
combining the original parameters. They concluded that ambient CO values could be
averaged safely over any time period less than or equal to the half-life of blood COHb.
25 9.3.4 Summary
The best all around model for COHb prediction is still the equation developed by
Coburn, Forster, and Kane (1965). The linear solution is useful for examining air pollution
data leading to relatively low COHb levels, whereas the nonlinear solution shows good
predictive power even for high CO exposures. The two regression models might be useful
30 only when the conditions of application closely approximate those under which the parameters
were estimated.
March 15, 1990 9-21 DRAFT - DO NOT QUOTE OR CITE
-------
It is important to remember that almost all of the above studies assumed a constant rate
of CO uptake and elimination, which is rarely true. A number of physiological factors,
particularly changes in ventilation, will affect both rates. The predicted COHb values also
will differ from individual to individual due to smoking, age, or lung disease. There does not
appear to be a single optimal averaging time period for ambient CO; however, the shorter the
period the greater the precision. In general, the averaging time period should be well within
the [COHb] half-life, which decreases with increased activity.
10 9.4 INTRACELLULAR EFFECTS OF CARBON MONOXIDE
9.4.1 Introduction
The principal cause of CO toxicity is tissue hypoxia due to CO binding to Hb, yet
certain physiological aspects of CO exposure are not explained well by decreases in
intracellular PO2 related to the presence of COHb. For many years, it has been known that
15 CO is distributed to extravascular sites such as skeletal muscle (Coburn et al., 1971; Coburn
et al., 1973) and that 10 to 50% of the total body store of CO may be extravascular
(Luomanmaki and Coburn, 1969). Furthermore, extravascular CO is metabolized slowly to
CO2 in vivo (Fenn, 1970). Consequently, secondary mechanisms of CO toxicity related to
intracellular uptake of CO have been the focus of a great deal of research interest. CO
20 binding to many intracellular compounds has been well documented both in vitro and in vivo,
however, it is still uncertain whether or not intracellular uptake of CO in the presence of Hb
is sufficient to cause either acute organ system dysfunction or long-term health effects. The
virtual absence of sensitive techniques capable of assessing intracellular CO binding under
physiological conditions has resulted in a variety of indirect approaches to the problem as well
25 as many negative studies. The purposes of this section of the document are to summarize
current knowledge pertaining to intracellular CO-binding proteins and to evaluate the potential
contribution of intracellular CO uptake to the overall physiological effects of CO exposure.
Selected aspects of this topic have been reviewed previously (Forster, 1970; Coburn et al.,
1977; Coburn, 1979; Piantadosi, 1987; Coburn and Forman, 1987).
30 CO is known to react with a variety of metal-containing proteins found in nature. CO-
binding metalloproteins present in mammalian tissues include O2-carrier proteins such as
March 15, 1990 9-22 DRAFT - DO NOT QUOTE OR CITE
-------
hemoglobin (Douglas et al., 1912) and myoglobin (Antonini and Brunori, 1971) and
metalloenzymes (oxidoreductases) such as cytochrome c oxidase (Keilin and Hartree, 1939),
cytochromes of the P-450 type (Omura and Sato, 1964), tryptophan oxygenase (Tanaka and
Knox, 1959), and dopamine hydroxylase (Kaufman, 1966). These metalloproteins contain
5 iron and/or copper centers at their active sites that form metal-ligand complexes with CO in
competition with molecular oxygen. CO and O2 form complexes with metalloenzymes only
when the iron and copper are in their reduced forms (Fe II, Cu I). Caughey (1970) has
reviewed the similarities and differences in the physicochemical characteristics of CO and O2
binding to these transition metal ions. The competitive relationship between CO and O2 for
10 the active site of intracellular hemoproteins usually is described by the Warburg partition
coefficient (K), which is the CO/O2 ratio that produces 50% inhibition of the O2 uptake of the
enzyme or, in the case of myoglobin, a 50% decrease in the number of available O2-binding
sites.
The measured Warburg coefficients of various mammalian CO-binding proteins have
15 been tabulated recently by Coburn and Forman (1987) (see Table 9-1). These K values range
from approximately 0.025 for myoglobin to 0.1 to 12 for cytochromes P-450. K values of 2
to 28 have been reported for cytochrome c oxidase (Keilin and Hartree, 1939; Wohlrab and
Ogunmola, 1971; Wharton and Gibson 1976). By comparison, the K value for human
hemoglobin of 0.005 is some three orders of magnitude less than that of cytochrome c
20 oxidase. This means, for example, that CO would bind to cytochrome oxidase in vivo only if
O2 gradients from RBCs in the capillary to the mitochondria were quite steep. Application of
K values for intracellular hemoproteins in this way, however, needs to be used with caution
because most measurements of CO binding have not been made at physiological temperatures
or at relevant rates of electron transport.
25 Apart from questions about the relevance of extrapolating in vitro partition coefficients
to physiological conditions, experimental problems arise that are related to determining actual
CO/O2 ratios in intact tissues. Reasonably good estimates of tissue Pco may be obtained by
calculating the value in mean capillary blood from the Haldane relationship (Coburn et al.,
1977), neglecting the low rate of CO metabolism by the tissue. Experimental estimates of the
30 Pa, in animal tissues have been found to be in close agreement with these calculations and
average slightly less than alveolar P^ (Goethert et al., 1970; Goethert, 1972). In general,
March 15, 1990 9-23 DRAFT - DO NOT QUOTE OR CITE
-------
TABLE 9-1. IN VTTRO INHIBITION RATIOS FOR HEMOPROTEINS THAT BIND CARBON MONOXIDE
Hemoprotein
Hemoglobin
Myoglobin
Cytochrome c oxidase
VO
K> Cytochrome P-450
Dopamine ft hydroxylase
O
j> Tryptophan oxygenase
S'R = CO/O2 at 50% inhibition
bM = 1/R
1
-~ Source: Adapted from Cobum and
•§
s
o
Source R' M" Temperature (°C)
Human RBC 0.0045 218 37
Sperm whale 0.025-0.040 25-40 25
Bovine heart 5-15 0.1 - 0.2 25
Rat liver 0.1-12 10-0.1 30-37
Bovine adrenal 2 0.5 —
Pseudomonas 0.55 1.8 25
Fonnan (1987).
-------
steady state estimates for tissue PCQ range from 0.02 to 0.5 torr at COHb concentrations of 5
to 50%. Therefore, at 50% COHb, a CO/O2 ratio of 5 may be achieved at sites of
intracellular O2 uptake only if tissue PO2 in the vicinity of the CO-binding proteins is
approximately 0.1 torr.
5 Whether such low intracellular PO2 values exist in target tissues such as brain and heart
during CO exposure is difficult to determine from the existing scientific literature.
Experimental measurements of tissue PO2 using polarographic microelectrodes indicate
significant differences in PO2 in different tissues and regional differences in PO2 within a
given tissue. This normal variability in tissue PO2 is related to differences in capillary
10 perfusion, red blood cell spacing, velocity and path length, and local requirements for
O2. Normal PO2 values obtained from such recordings are generally in the range of 0 to
30 torr (Leniger-Follert et al., 1975). These PO2 values usually represent average interstitial
values, although it is often difficult to determine the exact location of the electrode and the
effect of O2 consumption by the electrode on the PO2 measurement. Furthermore, the
15 gradient between the capillary and the intracellular sites of O2 utilization are thought to be
quite steep (Sies, 1977). A major component of the gradient arises between the red blood cell
and interstitium (Heliums, 1977) but the PO2 gradient between the cell membrane and
respiring mitochondria and other O2-requiring organelles remains undetermined in intact
normal tissues. Even less is known about intracellular PO2 in the presence of COHb. It has
20 been determined, however, that both PO2 in brain tissue (Zorn, 1972) and cerebrovenous PO2
(Koehler et al., 1984) decrease linearly as a function of COHb concentration. Presumably
then, intracellular PO2 declines with increasing COHb concentration, and at certain locations,
CO forms ligands with the O2-dependent, intracellular hemoproteins. As the intracellular PO2
decreases, the CO/O2 ratio in the tissue increases at constant Pco and an increasing fraction of
25 the available intracellular O2-binding sites become occupied by CO.
The intracellular uptake of CO behaves according to the preceding general principles;
most of the experimental evidence for this line of reasoning was derived from in vivo studies
of COHb formation by Coburn and colleagues (1965) at the University of Pennsylvania. For
all intracellular hemoproteins, however, two crucial quantitative unknowns remain. These are
30 (1) the fraction of intracellular-binding sites in discrete tissues inhibited by CO at any level of
COHb saturation, and (2) the critical fraction of inhibited sites necessary to amplify or initiate
March 15, 1990 9-25 DRAFT - DO NOT QUOTE OR CITE
-------
a deleterious physiological effect, or trigger biochemical responses with long-term health
effects. In general then, the activities of certain intracelliilar hemoproteins may be altered at
physiologically tolerable levels of carboxyhemoglobin. The problem is in determining what
level of intracellular reserve is available during CO hypoxia. In view of this general
5 conclusion, recent literature for the candidate hemoproteins has been evaluated to obtain
positive evidence for intracellular CO binding and corroboration of functional consequences of
the intracellular CO effects at specific COHb levels.
9.4.2 Carbon Monoxide Binding to Myoglobin
10 The red protein myoglobin is involved in the transport of O2 from capillaries to
mitochondria in red muscles. The binding of CO to Mb in heart and skeletal muscle in vivo
has been demonstrated at levels of COHb below 2% in heart and 1% in skeletal muscle
(Coburn and Mayers, 1971; Coburn et al., 1973). The ratio of COMb/COHb saturation has
been found to be approximately one in cardiac muscle and less than one in skeletal muscle.
15 These ratios did not increase with increases in COHb up to 50% saturation. In the presence
of hypoxemia and hypoperfusion, the amount of CO uptake by Mb has been measured and
was shown to increase (Coburn et al., 1973; Coburn et al., 1971). A similar conclusion has
been reached during maximal exercise in humans, where CO shifts from Hb to the
intracellular compartment (i.e., Mb, at COHb levels of 2 to 2.5%) (Clark and Coburn,
20 1975). The significance of CO uptake by Mb is uncertain because our understanding of the
functional role of Mb in working muscle is incomplete. Myoglobin undoubtedly enhances the
uptake of O2 by muscle cells so that the continuous O2 demand of working muscle is satisfied
(Wittenberg et al., 1975). Myoglobin may contribute to muscle function by serving as an O2
store, by enhancing intracellular diffusion of O2, or by acting as an O2 buffer to maintain a
25 constant mitochondria! PO2 during changes in O2 supply. Functional Mb has been found to be
necessary for maintenance of maximum O2 uptake and mechanical tension in exercising
skeletal muscle (Cole, 1982). The binding of CO to Mb would therefore be expected to limit
O2 availability to mitochondria in working muscle. This possibility has been verified theoreti-
cally by computer simulations of Hoofd and Kreuzer (1978) and Agostoni et al. (1980). The
30 three-compartment (arterial and venous capillary blood, and Mb) computer model of Agostoni
et al. (1980) predicted that COMb formation in low PO2 regions of the heart (e.g.,
March 15, 1990 9-26 DRAFT - DO NOT QUOTE OR CITE
-------
subendocardium) could be sufficient to impair intracellular O2 transport to mitochondria at
COHb saturations of 5 to 10%. The [COMb] also was predicted to increase during conditions
of hypoxia, ischemia, and increased O2 demand.
The direct effects of CO on cardiac function also have been evaluated in the absence of
5 Hb in fluorocarbon-perfused rabbits (Takano et al., 1981). Exposure of these animals to high
concentrations of CO (CO/02 = 0.05-0.25) significantly decreased the heart rate-systolic
pressure product in the absence of COHb formation. Cardiac output and [COMb], however,
were not determined. Increases in cardiac [COMb] have been measured after heavy work
loads in CO-exposed rats, independent, of changes in [COHb] (Sokal et al., 1986). These
10 investigators reported that exercise significantly increased cardiac [COMb] at COHb
saturations of approximately 10, 20, and 50%, although metabolic acidosis worsened only at
50% COHb. It remains unknown, however, whether or not low [MbCO] could be
responsible for decreases in maximal O2 uptake during exercise reported at COHb levels of 4
to 5% (see Chapter 10, Section 10.3).
15
9.4.3 Carbon Monoxide Uptake by Cytochrome P-450
Mixed-function oxidases (cytochrome P-450) are involved in the detoxification of a
number of drugs and steroids by "oxidation." These enzymes are distributed widely through-
out mammalian tissues; the highest concentrations are found in the microsomes of liver,
20 adrenal gland, and the lungs of some species (Estabrook et al., 1970). These oxidases also
are present in low concentrations in kidney and brain tissues. Mixed-function oxidases
catalyze a variety of hydroxylation reactions involving the uptake of a pair of electrons from
NADPH with reduction of one atom of O2 to H2O and incorporation of the other into
substrates (White and Coon, 1980). These enzymes bind CO, and their Warburg binding
25 coefficients range from 0.1 to 12 in vitro (see Coburn and Forman, 1987). The sensitivity of
cytochrome P-450 to CO is increased under conditions of rapid electron transport (Estabrook
et al., 1970), however, previous calculations have indicated that tissue Pco is too low to
inhibit the function of these hemoproteins in vivo at less than 15 to 20% COHb (Coburn and
Forman, 1987). There have been few attempts to measure CO-binding coefficients in intact
30 tissues. In isolated rabbit lung, the effects of CO on mixed-function oxidase are consistent
with a Warburg coefficient of approximately 0.5 (Fisher et al., 1979). CO exposure
March 15, 1990 9-27 DRAFT - DO NOT QUOTE OR CITE
-------
decreases the rate of hepatic metabolism of hexobarbital and other drugs in experimental
animals (Montgomery and Rubin, 1973; Roth and Rubin, 1976a,b). Most authors have
concluded that these effects of CO on xenobiotic metabolism are attributable entirely to
COHb-related tissue hypoxia because they are no greater than the effects of "equivalent"
5 levels of hypoxic hypoxia. Three optical studies of rat liver perfused in situ with Hb-free
buffers have demonstrated uptake of CO by cytochrome P-450 at CO/O2 ratios of 0.03 to
0.10 (Sies and Brauser, 1970; lyanagi et al., 1981; Takano et al., 1985). In the study by
Takano et al. (1985) significant inhibition of hexobarbital metabolism was found at a CO/O2
of about 0.1. This CO/O2 ratio, if translated directly to [COHb], would produce [COHb] that
10 are incompatible with survival (-95%). At present, there is no scientific evidence that CO
significantly inhibits the activity of mixed-function oxidases at COHb saturations below 15 to
20%. Although most studies do not indicate effects of CO on cytochrome P-450 activity at
physiologically relevant CO concentrations, specific P-450 isoenzymes may have higher
affinities for CO. Also, the rate of substrate metabolism and substrate type may increase CO
15 binding by P-450 enzymes. More basic research is needed in this area because of the
important role of these enzymes in living organisms.
9.4.4 Carbon Monoxide and Cytochrome c Oxidase
Cytochrome c oxidase, a.k.a. cytochrome a a3, is the terminal enzyme in the
20 mitochondrial electron transport chain that catalyzes the reduction of molecular O2 to water.
Although the enzyme complex binds CO, three reasons are often cited for why this should
occur only under conditions of severe hypoxia. First, the Warburg binding constant for
cytochrome oxidase is unfavorable for CO uptake relative to the other candidate
hemoproteins. Second, the enzyme has an in vitro Michaelis-Menten constant (KJ for O2 of
25 less than 1 torr (Chance and Williams, 1955). Because intracellular PO2 is probably higher
than this, the oxidase should remain oxidized until severe tissue hypoxia is present. The
above arguments, however rational, are not supported well by in vivo observations and may
not be valid for the conditions encountered in living systems. The reasons for this difficulty
center around differences in the redox behavior of cytochrome oxidase in vivo relative to its
30 in vitro behavior. The enzyme has a high resting reduction level at normal PO2 in brain
(Jobsis et al., 1977) and other tissues, and its oxidation state varies directly with PO2 in vitro
March 15, 1990 9-28 DRAFT - DO NOT QUOTE OR CITE
-------
(Kreisman et al., 1981). These findings may indicate that the oxidase operates near its
effective K^ratios in vivo or that the availability of O2 to each mitochondrion or respiratory
chain is not continuous under most physiological circumstances. There also may be
differences in or regulation of the K,,, for O2 of the enzyme according to regional metabolic
5 conditions. For example, the apparent B^ for O2 of cytochrome oxidase increases several
times during rapid respiration (Oshino et al., 1974), and in isolated cells it varies as a
function of the cytosolic phosphorylation potential (Erecinska and Wilson, 1982). Conditions
of high respiration and/or high cytosolic phosphorylation potential in vitro increase the
concentration of CO-cytochrome oxidase at any CO/O2 ratio. This concept is particularly
10 relevant for tissues like the heart and brain.
Enhanced sensitivity of cytochrome oxidase to CO has been demonstrated in uncoupled
mitochondria, where CO/O2 ratios as low as 0.2 delay the oxidation of reduced cytochrome
oxidase in transit from anoxia to normoxia (Chance et al., 1970). Several studies of respiring
tissues, however, have found CO/O2 ratios of 12 to 20 to be necessary for 50% inhibition of
15 O2 uptake (Coburn et al., 1979; Fisher and Dodia, 1981; Kidder, 1980). In this context, it is
important to note that in a given tissue, the CO/O2 ratio necessary to inhibit one half of the
O2 uptake does not necessarily correspond to CO binding to one half of the oxidase
molecules. This is because unblocked cytochrome oxidase molecules may oxidize respiratory
complexes of blocked chains, thus causing the O2 consumption to fall more slowly than
20 predicted for strictly linear systems. The capacity of tissues to compensate for electron
transport inhibition by branching has not been investigated systematically as a function of
PO2, CO/O2 ratio, cytosolic phosphorylation potential, or rate of electron transport in vivo.
The contention that intracellular CO uptake by cytochrome oxidase occurs is supported
by a few experiments. It has been known for many years, primarily through the work of
25 Fenn (Fenn and Cobb, 1932; Fenn, 1970), that CO is slowly oxidized in the body to CO2.
This oxidation occurs normally at a much lower rate than the endogenous rate of CO
production, however, the rate of oxidation of CO increases in proportion to the CO body
store (Luomanmaki and Coburn, 1969). The oxidation of CO to CO2 was shown in 1965 by
Tzagoloff and Wharton to be catalyzed by reduced cytochrome oxidase. More recently,
30 Young et al. (1979) demonstrated that oxidized cytochrome oxidase promotes CO oxidation,
and subsequently, that cytochrome oxidase in intact heart and brain mitochondria was capable
March 15, 1990 9-29 DRAFT - DO NOT QUOTE OR CITE
-------
of catalyzing the reaction at a CO/O2 ratio of approximately 4 (Young and Caughey, 1986).
The physiological significance of this reaction is unknown.
Other studies indicating possible direct effects of CO on cytochrome oxidase include a
photoreversible effect of 500 to 1000 ppm CO on spontaneous electrical activity of cerebellar
5 Purkinje cells in tissue culture (Raybourn et al., 1978). These CO concentrations would be
expected to produce [COHb] in the range of 33 to 50%. At 7.5% COHb, inhibition of the fa-
wave of the electroretinogram has been reported in the cat (Ingenito and Durlacher, 1979).
Persistent changes in the retinogram were reminiscent of the "remnant effect" of CO on visual
thresholds in humans reported by Halperin et al. (1959). Other optical evidence suggesting
10 that cytochrome oxidase is sensitive to CO in vivo comes from studies of the effects of CO on
cerebrocortical cytochromes in fluorocarbon-perfused rats (Piantadosi et al., 1985, 1987). In
these studies, CO/O2 ratios of 0.006 to 0.06 were associated with spectral evidence of CO
binding to reduced cytochrome oxidase. The spectral data also indicated that the intracellular
uptake of CO produced increases in the reduction level of b-type cytochromes in the brain
15 cortex. At CO/O2 ratios of 0.06, most (> 80%) of the cytochrome b became reduced in the
cerebral cortex. The cytochrome b response is not understood well; it is thought to represent
an indirect (e.g., energy-dependent) response of mitochondrial fr-cytochromes to CO because
these cytochromes are not known to bind CO in situ. The CO/O2 ratios used in the studies of
Piantadosi et al. (1985, 1987) would produce [COHb] in the range of 50 to 90%. The
20 venous PO2 in those experiments, however, was about 100 torr; thus at tissue PO2s that are
significantly lower, this effect should occur at lower COHb saturations. It is unlikely,
however, that cerebral uptake of CO is significant at COHb below 5 % because tissue Pco is so
low in the presence of Hb. The physiological significance of these effects of CO have not yet
been determined.
25 Direct effects of CO on mitochondrial function have been suggested by several recent
studies which indicate decreases in cytochrome oxidase activity by histochemistry in brain and
heart after severe CO intoxication in experimental animals (Pankow and Ponsold, 1984;
Savolainen et al., 1980; Somogyi et al., 1981). The magnitude of the decrease in cytochrome
oxidase activity may exceed that associated with severe hypoxia, although problems of deter-
30 mining "equivalent" levels of CO hypoxia and hypoxic hypoxia have not been addressed
adequately by these studies. The effects of passive cigarette smoking on oxidative
March 15, 1990 9-30 DRAFT - DO NOT QUOTE OR CITE
-------
phosphorylation in myocardial mitochondria have been studied in rabbits (Gvozdjakova et al.,
1984). Mitochondrial respiratory rate (State 3 and State 4) and rates of oxidative
phosphorylation were found to be decreased significantly by [COHb] of 6 to 7%. These data,
however, are not definitive with respect to CO because they include effects of nicotine, which
5 reached concentrations of 5.7 jig/L in blood. A recent study by Snow et al. (1988) in dogs
with prior experimental myocardial infarction indicated that a COHb of 9.4% increased the
resting reduction level of cytochrome oxidase in the heart. The CO exposures also were
accompanied by more rapid cytochrome oxidase reductions after coronary artery occlusion
and less rapid reoxidation of the enzyme after release of the occlusion. The authors
10 concluded that CO trapped the oxidase in the reduced state during transient cardiac ischemia.
There is also evidence that formation of the CO-cytochrome oxidase ligand occurs in the brain
of the rat at COHb saturations of 40 to 50% (Brown and Piantadosi, 1990). This binding
appears to be related to hypotension and probable cerebral hypoperfusion during CO
exposure. This effect is in concert with experimental evidence that CO produces direct
15 vasorelaxation of smooth muscle. This vasodilation occurs in rabbit aorta (Coburn et al.,
1979), in the coronary circulation of isolated perfused rat heart (McFaul and McGrath, 1987),
and in the cerebral circulation of the fluorocarbon-perfused rat (Piantadosi et al., 1987). The
mechanism of this vasodilator effect is unclear, although it appears to be related to decreased
calcium concentrations in vascular smooth muscle (Lin and McGrath, 1988) and elevation of
20 cellular cyclic guanosine monophosphate (GMP) levels (Ramos et al., 1989). The stimulus
does not require hypoxia, adenosine or prostaglandins and it is possible that it represents a
direct toxic effect of CO on the cytochrome system in vascular smooth muscle. The
physiological significance of this phenomenon is undetermined.
In summary, there is evidence to suggest that CO binds to cytochrome oxidase in
25 mammalian heart and brain tissues at a range of systemic PO2 values. The only experimental
evidence at present that this effect occurs at COHb levels less than 10% is the slow oxidation
of CO to CO2, which has been shown to occur in vivo and in isolated mitochondria in vitro.
Experimental evidence indicates that CO binding to cytochrome oxidase does occur during
tissue hypoxia produced by overtly toxic COHb concentrations. The physiological
30 significance of these effects beyond those of tissue hypoxia remains unknown. Once CO
binding to cytochrome oxidase occurs, however, the small rate constant for CO dissociation
March 15, 1990 9-31 DRAFT - DO NOT QUOTE OR CITE
-------
from the enzyme yields an apparent rate-dependent inhibition constant for CO under
nonequilibrium conditions. This means that at high rates of respiration and low O2
concentrations, recovery of enzymatic function by the oxidase is relatively slow in comparison
to simple O2 deprivation.
5
9.5 REFERENCES
10
Agostoni, A.; Stabilini, R.; Viggiano, G.; Luzzana, M.; Samaja, M. (1980) influence of capillary and tissue PQ
on carbon monoxide binding to myoglobin: a theoretical evaluation. Microvasc. Res. 20: 81-87.
Anthonisen, N. R.; Robertson, P. C.; Ross, W. R. D. (1970) Gravity-dependent sequential emptying of lung
15 regions. J. Appl. Physiol. 28: 589-595.
Antonini, E.; Brunori, M. (1971) The partition constant between two ligands. In: Hemoglobin and myoglobin in
their reactions with ligands. Amsterdam, The Netherlands: North-Holland Publishing Co.; pp. 174-175.
20 Berk, P. D.; Rodkey, F. L.; Blaschke, T. F.; Collison, H. A.; Waggoner, J. G. (1974) Comparison of plasma
bilirubin turnover and carbon monoxide production in man. J. Lab. Clin. Med. 83: 29-37.
Berk, P. D.; Blaschke, T. F.; Scharschmidt, B. F.; Waggoner, J. G.; Berlin, N. I. (1976) A new approach to
quantitation of the various sources of bilirubin in man. J. Lab. Clin. Med. 87: 767-780.
25
Bernard, T. E.; Duker, J. (1981) Modeling carbon monoxide uptake during work. Am. Ind. Hyg. Assoc. J.
42: 361-364.
Biller, W. F.; Richmond, H. M. (1982) Sensitivity analysis on Coburn model predictions of COHb levels
30 associated with alternative CO standards. Research Triangle Park, NC: U. S. Environmental Protection
Agency, Office of Air Quality Planning and Standards; EPA contract no. 68-02-3600.
Bischoff, K. B. (1986) Physiological pharmacokinetics. Bull. Math. Biol. 48: 309-322.
35 Bosisio, E.; Grisetti, G. C.; Panzuti, F.; Sergi, M. (1980) Pulmonary diffusing capacity and its components (DM
and Vc) in young, healthy smokers. Respiration 40: 307-310.
Britten, J. S.; Myers, R. A. M. (1985) Effects of hyperbaric treatment on carbon monoxide elimination in
humans. Undersea Biomed. Res. 12: 431-438.
40
Brown, S. D.; Piantadosi, C. A. (1990) In vivo binding of carbon monoxide to cytochrome c oxidase hi rat
brain. J. Appl. Physiol.: in press.
Buerk, D. G.; Bridges, E. W. (1986) A simplified algorith for computing the variation in oxyhemoglobin
45 saturation with pH, PCO2, T and DPG. Chem. Eng. Commun. 47: 113-124.
Caughey, W. S. (1970) Carbon monoxide bonding in hemeproteins. In: Coburn, R. F., ed. Biological effects of
carbon monoxide. Ann. N. Y. Acad. Sci. 174: 148-153.
March 15, 1990 9-32 DRAFT - DO NOT QUOTE OR CITE
-------
Chance, B.; Williams, G. R. (1955) Respiratory enzymes in oxidative phosphorylation: HI. the steady state J.
Biol. Chem. 217: 409-427. Clark, B. J.; Cobum, R. F. (1975) Mean myoglobin oxygen tension during
exercise at maximal oxygen uptake. J. Appl. Physiol. 39: 135-144.
5 Chance, B.; Erecinska, M.; Wagner, M. (1970) Mitochondrial responses to carbon monoxide toxicity. In:
Coburn, R. F., ed. Biological effects of carbon monoxide. Ann. N. Y. Acad. Sci. 174: 193-204.
Clarke, S. W.; Jones, J. G.; Glaister, D. H. (1969) Change in pulmonary ventilation in different postures. Gin.
Sci. 37: 357-369.
10
Coburn, R. F. (1970) Enhancement by phenobarbital and diphenylhydantoin of carbon monoxide production in
normal man. N. Engl. J. Med. 283: 512-515.
Coburn, R. F. (1979) Mechanisms of carbon monoxide toxicity. Prev. Med. 8: 310-322.
15
Coburn, R. F.; Forman, H. J. (1987) Carbon monoxide toxicity. In: Geiger, S. R.; Farhi, L. E.; Tenney, S. M.;
Fishman, A. P., eds. Handbook of physiology: a critical, comprehensive presentation of physiological
knowledge and concepts. Section 3: the respiratory system, volume IV. gas exchange. Bethesda, MD:
American Physiological Society; pp. 439-456.
20
Coburn, R. F.; Mayers, L. B. (1971) Myoglobin O2 tension determined from measurements of carboxymyoglobin
in skeletal muscle. Am. J. Physiol. 220: 66-74.
Cobum, R. F.; Williams, W. J.; Forster, R. E. (1964) Effect of erythrocyte destruction on carbon monoxide
25 production in man. J. Clin. Invest. 43: 1098-1103.
Coburn, R. F.; Forster, R. E.; Kane, P. B. (1965) Considerations of the physiological variables that determine
the blood carboxyhemoglobin concentration in man. J. Clin. Invest. 44: 1899-1910.
30 Coburn, R. F.; Williams, W. J.; Kahn, S. B. (1966) Endogenous carbon monoxide production in patients with
hemolytic anemia. J. Clin. Invest. 45: 460-468.
Cobum, R. F.; Wallace, H. W.; Abboud, R. (1971) Redistribution of body carbon monoxide after hemorrhage.
Am. J. Physiol. 220: 868-873.
35
Coburn, R. F.; Ploegmakers, F.; Gondrie, P.; Abboud, R. (1973) Myocardial myoglobin oxygen tension. Am. J.
Physiol. 224: 870-876.
Coburn, R. F.; Allen, E. R.; Ayres, S.; Bartlett, D., Jr.; Horvath, S. M.; Kuller, L. H.; Laties, V.; Longo,
40 L. D.; Radford, E. P., Jr. (1977) Carbon monoxide. Washington, DC: National Academy of Sciences.
Coburn, R. F.; Grubb, B.; Aronson, R. D. (1979) Effect of cyanide on oxygen tension-dependent mechanical
tension in rabbit aorta. Circ. Res. 44: 368-378.
45 Cole, R. P. (1982) Myoglobin function in exercising skeletal muscle. Science (Washington, DC) 216: 523-525.
Delivoria-Papadopoulos, M.; Coburn, R. F.; Forster, R. E. (1970) Cyclical variation of rate of heme destruction
and carbon monoxide production (Vco) in normal women. Physiologist 13: 178.
50 Douglas, C. G.; Haldane, J. S.; Haldane, J. B. S. (1912) The laws of combination of haemoglobin with carbon
monoxide and oxygen. J. Physiol. (London) 44: 275-304.
Erecinska, M.; Wilson, D. F. (1982) Regulation of cellular energy metabolism. J. Membr. Biol. 70: 1-14.
March 15, 1990 9-33 DRAFT - DO NOT QUOTE OR CITE
-------
Estebrook, R. W.; Franklin, M. R.; Hildebrandt, A. G. (1970) Factors influencing the inhibitory effect of
carbon monoxide on cytochrome P-450-catalyzed mixed function oxidation reactions. In: Coburn, R. F.,
ed. Biological effects of carbon monoxide. Ann. N. Y. Acad. Sci. 174: 218-232.
5 Federal Register. (1985) Review of the national ambient air quality standards for carbon monoxide; final rule.
F. R. (September 13) 50: 37484-37501.
Fenn, W. O. (1970) The burning of CO in the tissues. In: Coburn, R. F., ed. Biological effects of carbon
monoxide. Ann. N. Y. Acad. Sci. 174: 64-71.
10
Fenn, W. O.; Cobb, D. M. (1932) The burning of carbon monoxide by heart and skeletal muscle. Am. J.
Physiol. 102: 393-401.
Fisher, A. B.; Dodia, C. (1981) Lung as a model for evaluation of critical intracellular PQ and P^. Am. J.
15 Physiol. 241: E47-E50. 2
Fisher, G. L.; Chrisp, C. E.; Raabe, O. G. (1979) Physical factors affecting the mutagenicity of fly ash from a
coal-fired power plant. Science (Washington, DC) 204: 879-881.
20 Fishman, A. P.; Farhi, L. E.; Tenney, S. M., eds. (1987) Handbook of physiology section 3: the respiratory
system, v. 4, gas exchange. Bethesda, MD: American Physiological Society.
Forster, R. E. (1970) Carbon monoxide and the partial pressure of oxygen in tissue. In: Coburn, R. F., ed.
Biological effects of carbon monoxide. Ann. N. Y. Acad. Sci. 174: 233-241.
25
Forster, R. E. (1987) Diffusion of gases across the alveolar membrane. In: Fishman, A. P.; Farhi, L. E.;
Tenney, S. M., eds. Handbook of physiology section 3: the respiratory system, v. 4, gas exchange.
Bethesda, MD: American Physiological Society; pp. 71-88.
30 Frey, T. M.; Crapo, R. O.; Jensen, R. L.; Elliott, C. G. (1987) Diurnal variation of the diffusing capacity of the
lung: is it real? Am. Rev. Respir. Dis. 136: 1381-1384.
Glaister, D. H.; Schroter, R. C.; Sudlow, M. F.; Milic-Emili, J. (1973) Transpulmonary pressure gradient and
ventilation distribution in excised lungs. Respir. Physiol. 17: 365-385.
35
Godin, G.; Shephard, R. J. (1972) On the course of carbon monoxide uptake and release. Respiration
29: 317-329.
Goethert, M. (1972) Factors influencing the CO content of tissues. Staub Reinhalt. Luft 32: 15-20.
40
Goethert, M.; Lutz, F.; Malorny, G. (1970) Carbon monoxide partial pressure in tissue of different animals.
Environ. Res. 3: 303-309.
Guyatt, A. R.; Holmes, M. A.; Gumming, G. (1981) Can carbon monoxide be absorbed from the upper
45 respiratory tract in man? Eur. J. Respir. Dis. 62: 383-390.
Gvozdjakova, A.; Bada, V.; Sany, L.; Kucharska, J.; Kruty, F.; Bozek, P.; Trstansky, L.; Gvozdjak, J. (1984)
Smoke cardiomyopathy: disturbance of oxidative processes in myocardial mitochondria. Cardiovasc. Res.
18: 229-232.
50
Haldane, J. (1898) Some improved methods of gas analysis. J. Physiol. (London) 22: 465-480.
Halperin, M. H.; McFarland, R. A.; Niven, J. I.; Roughton, F. J. W. (1959) The time course of the effects of
carbon monoxide on visual thresholds. J. Physiol. (London) 146: 583-593.
March 15, 1990 9-34 DRAFT - DO NOT QUOTE OR CITE
-------
Harf, A.; Pratt, T.; Hughes, J. M. B. (1978) Regional distribution ofV •
e 6 a/Q in man at rest and with exercise
measured with krypton-Sim. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 44: 115-123.
Hauck, H.; Neuberger, M. (1984) Carbon monoxide uptake and the resulting carboxyhemoglobin in man. Eur. J.
5 Appl. Physiol. 53: 186-190.
Heliums, J. D. (1977) The resistance to oxygen transport in the capillaries relative to that in the surrounding
tissue. Microvasc. Res. 13: 131-136.
10 Hill, E. P.; Hill, J. R.; Power, G. G.; Longo, L. D. (1977) Carbon monoxide exchanges between the human
fetus and mother: a mathematical model. Am. J. Physiol. 232: H311-H323.
Hoofd, L.; Kreuzer, F. (1978) Calculation of the facilitation of O2 or CO transport by Hb or Mb by means of a
new method for solving the carrier-diffusion problem. In: Silver, I. A.; Erecinska, M.; Bicher, H. L,
15 eds. Oxygen transport to tissue - III. New York, NY: Plenum Press; pp. 163-168. (Advances in
experimental medicine and biology: v. 94).
Hughes, J. M. B.; Grant, B. J. B.; Greene, R. E.; Iliff, L. D.; Milic-Emili, J. (1972) Inspiratory flow rate and
ventilation distribution in normal subjects and in patients with simple chronic bronchitis. Clin. Sci.
20 43: 583-595.
Ingenito, A. J.; Durlacher, L. (1979) Effects of carbon monoxide on the b-wave of the cat electroretinogram:
comparisons with nitrogen hypoxia, epinephrine, vasodilator drugs and changes in respiratory tidal
volume. J. Pharmacol. Exp. Ther. 211: 638-646.
25
Inkley, S. R.; Maclntyre, W. J. (1973) Dynamic measurement of ventilation-perfusion with xenon-133 at resting
lung volumes. Am. Rev. Respir. Dis. 107: 429-441.
lyanagi, T.; Suzaki, T.; Kobayashi, S. (1981) Oxidation-reduction states of pyridine nucleotide and cytochrome
30 P-450 during mixed-function oxidation in perfused rat liver. J. Biol. Chem. 256: 12933-12939.
Jobsis, F. F.; Keizer, J. H.; LaManna, J. C.; Rosenthal, M. (1977) Reflectance spectrophotometry of
cytochrome aa3 in vivo. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 43: 858-872.
35 Joumard, R.; Chiron, M.; Vidon, R.; Maurin, M.; Rouzioux, J.-M. (1981) Mathematical models of the uptake
of carbon monoxide on hemoglobin at low carbon monoxide levels. EHP Environ. Health Perspect.
41: 277-289.
Kaufman, S. (1966) D. Coenzymes and hydroxylases: ascorbate and dopamine-/3-hydroxylase;
40 tetrahydropteridines and phenylalaniue and tyrosine hydroxylases. Pharmacol. Rev. 18: 61-69.
Keilin, D.; Hartree, E. F. (1939) Cytochrome and cytochrome oxidase. Proc. R. Soc. London B 127: 167-191.
Kidder, G. W., III. (1980) Carbon monoxide insensitivity of gastric acid secretion. Am. J. Physiol.
45 238: G197-G202.
Knudson, R. J.; Kaltenborn, W. T.; Burrows, B. (1989) The effects of cigarette smoking and smoking cessation
on the carbon monoxide diffusing capacity of the lung in asymptomatic subjects. Am. Rev. Respir. Dis.
140: 645-651.
50
Koehler, R. C.; Traystman, R. J.; Zeger, S.; Rogers, M. C.; Jones, M. D., Jr. (1984) Comparison of
cerebrovascular responses to hypoxic and carbon monoxide hypoxia in newborn and adult sheep. J.
Cereb. Blood Flow Metab. 4: 115-122.
March 15, 1990 9-35 DRAFT - DO NOT QUOTE OR CITE
-------
Kreisman, N. R.; Sick, T. J.; LaManna, J. C.; Rosenthal, M. (1981) Local tissue oxygen tension - cytochrome
8,% redox relationships in rat cerebral cortex in vivo. Brain Res. 218: 161-174.
Landaw, S. A. (1973) The effects of cigarette smoking on total body burden and excretion rates of carbon
5 monoxide. JOM J. Occup. Med. 15: 231-235.
Landaw, S. A.; Callahan, E. W., Jr.; Schmid, R. (1970) Catabolism of heme in vivo: comparison of the
simultaneous production of bilirubin and carbon monoxide. J. Clin. Invest. 49: 914-925.
10 Leniger-Follert, E.; Luebbers, D. W.; Wrabetz, W. (1975) Regulation of local tissue Po2 of the brain cortex at
different arterial O2 pressures. Pfluegers Arch. 359: 81-95.
Lin, H.; McGrath, J. J. (1988) Carbon monoxide effects on calcium levels in vascular smooth muscle. Life Sci.
43: 1813-1816.
15
Longo, L. D.; Hill, E. P. (1977) Carbon monoxide uptake and elimination in fetal and maternal sheep. Am. J.
Physiol. 232: H324-H330.
Luomanmaki, K.; Coburn, R. F. (1969) Effects of metabolism and distribution of carbon monoxide on blood and
20 body stores. Am. J. Physiol. 217: 354-363.
Lynch, S. R.; Moede, A. L. (1972) Variation in the rate of endogenous carbon monoxide production in normal
human beings. J. Lab. Clin. Med. 79: 85-95.
25 Marcus, A. H. (1980a) Mathematical models for carboxyhemoglobin. Atmos. Environ. 14: 841-844.
Marcus, A. H. (1980b) Author's reply. Atmos. Environ. 14: 1777-1781.
Martin, C. J.; Das, S.; Young, A. C. (1979) Measurements of the dead space volume. J. Appl. Physiol.: Respir.
30 Environ. Exercise Physiol. 47: 319-324.
McCartney, M. L. (1990) Sensitivity analysis applied to Coburn-Forster-Kane models of carboxyhemoglobin
formation. Am. Ind. Hyg. Assoc. J.: in press.
35 McClean, P. A.; Duguid, N. J.; Griffin, P. M.; Newth, C. J. L.; Zamel, N. (1981) Changes in exhaled
pulmonary diffusing capacity at rest and exercise in individuals with impaired positional diffusion. Bull.
Eur. Physiopathol. Respir. 17: 179-186.
McFaul, S. J.; McGrath, J. J. (1987) Studies on the mechanism of carbon monoxide-induced vasodilation in the
40 isolated perfused rat heart. Toxicol. Appl. Pharmacol. 87: 464-473.
Milic-Emili, J.; Henderson, J. A. M.; Dolovich, M. B.; Trop, D.; Kaneko, K. (1966) Regional distribution of
inspired gas in the lung J. Appl. Physiol. 21: 749-759.
45 Millette, B.; Robertson, P. C.; Ross, W. R. D.; Anthonisen, N. R. (1969) Effect of expiratory flow rate on
emptying of lung regions. J. Appl. Physiol. 27: 587-591.
Miyahara, S.; Takahashi, H. (1971) Biological CO evolution: carbon monoxide evolution during auto- and
enzymatic oxidation of phenols. J. Biochem. (Tokyo) 69: 231-233.
50
Montgomery, M. R.; Rubin, R. J. (1973) Oxygenation during inhibition of drug metabolism by carbon monoxide
or hypoxic hypoxia. J. Appl. Physiol. 35: 505-509.
March 15, 1990 9-36 DRAFT - DO NOT QUOTE OR CITE
-------
Muller, K. E.; Barton, C. N. (1987) A nonlinear version of the Coburn, Forster and Kane model of blood
carboxyhemoglobin. Atmos. Environ. 21: 1963-1968.
Omura, T.; Sato, R. (1964) The carbon monoxide-binding pigment of liver microsomes: I. Evidence for its
5 hemoprotein nature. J. Biol. Chem. 239: 2370-2378.
Oshino, N.; Sugano, T.; Oshino, R.; Chance, B. (1974) Mitochondrial function under hypoxic conditions: the
steady states of cytochrome a+a3 and their relation to mitochondria! energy states. Biochim. Biophys.
Acta 368: 298-310.
10
Ott, W. R.; Mage, D. T. (1978) Interpreting urban carbon monoxide concentrations by means of a computerized
blood COHb model. J. Air Pollut. Control Assoc. 28: 911-916.
Ott, W.; Mage, D. (1980) Evaluation of CO air quality criteria using a COHb model. Atmos. Environ.
15 14: 979-980.
Pace, N.; Strajman, E.; Walker, E. L. (1950) Acceleration of carbon monoxide elimination in man by high
pressure oxygen. Science (Washington, DC) 111: 652-654.
20 Pankow, D.; Ponsold, W. (1984) Effect of carbon monoxide exposure on heart cytochrome c oxidase activity of
rats. Biomed. Biochim. Acta 43: 1185-1189.
Peterson, J. E.; Stewart, R. D. (1970) Absorption and elimination of carbon monoxide by inactive young men.
Arch. Environ. Health 21: 165-171.
25
Peterson, J. E.; Stewart, R. D. (1975) Predicting the carboxyhemoglobin levels resulting from carbon monoxide
exposures. J. Appl. Physiol. 39: 633-638.
Piantadosi, C. A. (1987) Carbon monoxide, oxygen transport, and oxygen metabolism. J. Hyperbaric Med.
30 2: 27-44.
Piantadosi, C. A.; Sylvia, A. L.; Saltzman, H. A.; Jobsis-Vandervliet, F. F. (1985) Carbon
monoxide-cytochrome interactions in the brain of the fluorocarbon-perfused rat. J. Appl. Physiol.
58: 665-672.
35
Piantadosi, C. A.; Sylvia, A. L.; Jobsis-Vandervliet, F. F. (1987) Differences in brain cytochrome responses to
carbon monoxide and cyanide in vivo. J. Appl. Physiol. 62: 1277-1284.
Rahn, H.; Fenn, W. O. (1955) A graphical analysis of the respiratory gas exchange: the O2-CO2 diagram.
40 Washington, DC: American Physiological Society; p. 37.
Raybourn, M. S.; Cork, C.; Schimmerling, W.; Tobias, C. A. (1978) An in vitro electrophysiological assessment
of the direct cellular toxicity of carbon monoxide. Toxicol. Appl. Pharmacol. 46: 769-779.
45 Riley, R. L.; Permutt, S. (1973) Venous admixture component of the AaPg gradient. J. Appl. Physiol.
35: 430-431. 2
Roth, R. A., Jr.; Rubin, R. J. (1976a) Role of blood flow in carbon monoxide- and hypoxic hypoxia-induced
alterations in hexobarbital metabolism in rats. Drug Metab. Dispos. 4: 460-467.
50
Roth, R. A., Jr.; Rubin, R. J. (1976b) Comparison of the effect of carbon monoxide and of hypoxic hypoxia. I.
In vivo metabolism, distribution and action of hexobarbital. J. Pharmacol. Exp. Ther. 199: 53-60.
March 15, 1990 9-37 DRAFT - DO NOT QUOTE OR CITE
-------
Roughton, F. J. W. (1970) The equilibrium of carbon monoxide with human hemoglobin in whole blood. In:
Biological effects of carbon monoxide, proceedings of a conference; January; New York, NY. Ann.
N. Y. Acad. Sci. 174(art. 1): 177-188.
5 Roughton, F. J. W.; Darling, R. C. (1944) The effect of carbon monoxide on the oxyhemoglobin dissociation
curve. Am. J. Physiol. 141: 17-31.
Saltzman, B. E.; Fox, S. H. (1986) Biological significance of fluctuating concentrations of carbon monoxide.
Environ. Sci. Technol. 20: 916-923.
10
Savolainen, H.; Kurppa, K.; Tenhunen, R.; Kivisto, H. (1980) Biochemical effects of carbon monoxide
poisoning in rat brain with special reference to blood carboxyhemoglobin and cerebral cytochrome
oxidase activity. Neurosci. Lett. 19: 319-323.
15 Schoenfisch, W. H.; Hoop, K. A.; Struelens, B. S. (1980) Carbon monoxide absorption through the oral and
nasal mucosae of cynomolgus monkeys. Arch. Environ. Health 35: 152-154.
• •
Scrimshire, D. A. (1977) Theoretical analysis of independent V. and Q inequalities upon pulmonary gas exchange.
Respir. Physiol. 29: 163-178.
20
Severinghaus, J. W.; Stupfel, M. (1957) Alveolar dead space as an index of distribution of blood flow in
pulmonary capillaries. J. Appl. Physiol. 10: 335-348.
Sies, H. (1977) Oxygen gradients during hypoxic steady states in liver: urate oxidase and cytochrome oxidase as
25 intracellular O2 indicators. Hoppe Seylers Z. Physiol. Chem. 358: 1021-1032.
Sies, H.; Brauser, B. (1970) Interaction of mixed function oxidase with its substrates and associated redox
transitions of cytochrome P-450 and pyridine nucleotides hi perfused rat liver. Eur. J. Biochem.
15: 531-540.
30
Smith, M. V. (1990) Comparing solutions to the linear and nonlinear CFK equations for predicting COHb
formation. Math. Biosci.: in press.
Smith, R. G. (1968) Discussion of paper by Dr. Dinman. JOM J. Occup. Med. 10: 456-462.
35
Snow, T. R.; Vanoli, E.; De Ferrari, G.; Stramba-Badiale, M.; Dickey, D. T. (1988) Response of cytochrome
a.aj to carbon monoxide in canine hearts with prior infarcts. Life Sci. 42: 927-931.
Sokal, J. A.; Majka, J.; Palus, J. (1984) The content of carbon monoxide in the tissues of rats intoxicated with
40 carbon monoxide in various conditions of acute exposure. Arch. Toxicol. 56: 106-108.
Sokal, J.; Majka, J.; Palus, J. (1986) Effect of work load on the content of carboxymyoglobin in the heart and
skeletal muscles of rats exposed to carbon monoxide. J. Hyg. Epidemiol. Microbiol. Immunol.
30: 57-62.
45
Solanki, D. L.; McCurdy, P. R.; Cuttitta, F. F.; Schechter, G. P. (1988) Hemolysis in sickle cell disease as
measured by endogenous carbon monoxide production: a preliminary report. Am. J. Clin. Pathol.
89: 221-225.
50 Somogyi, E.; Balogh, L; Rubanyi, G.; Sotonyi, P.; Szegedi, L. (1981) New findings concerning the
pathogenesis of acute carbon monoxide (CO) poisoning. Am. J. Forensic Med. Pathol. 2: 31-39.
March 15, 1990 9-38 DRAFT - DO NOT QUOTE OR CITE
-------
Standfuss, K. (1970) Die Auswirkung der physiologischen Aenderungen des Ventilations-Perfusionsverhaeltnisses
in der Zeit auf den funktionellen Totraum [Variations of the ventilation-perfusion ratio during the
respiratory cycle and their effect upon physiologic dead space]. Pfluegers Arch. 317: 198-227.
5 Stewart, R. D.; Peterson, J. E.; Baretta, E. D.; Bachand, R. T.; Hosko, M. J.; Herrmann, A. A. (1970)
Experimental human exposure to carbon monoxide. Arch. Environ. Health 21: 154-164.
Stewart, R. D.; Peterson, J. E.; Fisher, T. N.; Hosko, M. J.; Baretta, E. D.; Dodd, H. C.; Herrmann, A. A.
(1973) Experimental human exposure to high concentrations of carbon monoxide. Arch. Environ. Health
10 26: 1-7.
Stokes, D. L.; Maclntyre, N. R.; Nadel, J. A. (1981) Nonlinear increases in diffusing capacity during exercise
by seated and supine subjects. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 51: 858-863.
15 Sutherland, P. W.; Katsura, T.; Milic-Emili, J. (1968) Previous volume history of the lung and regional
distribution of gas. J. Appl. Physiol. 25: 566-574.
Takano, T.; Miyazaki, Y.; Shimoyama, H.; Maeda, H.; Okeda, R.; Funata, N. (1981) Direct effects of carbon
monoxide on cardiac function. Int. Arch. Occup. Environ. Health 49: 35-40.
20
Takano, T.; Motohashi, Y.; Miyazaki, Y.; Okeda, R. (1985) Direct effect of carbon monoxide on hexobarbital
metabolism in the isolated perfused liver in the absence of hemoglobin. J. Toxicol. Environ. Health
15: 847-854.
25 Tanaka, T.; Knox, W. E. (1959) The nature and mechanism of the tryptophan pyrrolase (peroxidase-oxidase)
reaction of Pseudomonas and of rat liver. J. Biol. Chem. 234: 1162-1170.
Tikuisis, P.; Buick, F.; Kane, D. M. (1987a) Percent carboxyhemoglobin in resting humans exposed repeatedly
to 1,500 and 7,500 ppm CO. J. Appl. Physiol. 63: 820-827.
30
Tikuisis, P.; Madill, H. D.; Gill, B. J.; Lewis, W. F.; Cox, K. M.; Kane, D. M. (1987b) A critical analysis of
the use of the CFK equation in predicting COHb formation. Am. Ind. Hyg. Assoc. J. 48: 208-213.
Tyuma, I.; Ueda, Y.; Imaizumi, K.; Kosaka, H. (1981) Prediction of the carbonmonoxyhemoglobin levels during
35 and after carbon monoxide exposures in various animal species. Jpn. J. Physiol. 31: 131-143.
Tzagoloff, A.; Wharton, D. C. (1965) Studies on the electron transfer system: LXII. the reaction of cytochrome
oxidase with carbon monoxide. J. Biol. Chem. 240: 2628-2633.
40 Venkatram, A.; Louch, R. (1979) Evaluation of CO air quality criteria using a COHb model. Atmos. Environ.
13: 869-872.
Wagner, J. A.; Horvath, S. M.; Dahms, T. E. (1975) Carbon monoxide elimination. Respir. Physiol. 23: 41-47.
45 Werner, B.; Lindahl, J. (1980) Endogeneous carbon monoxide production after bicycle exercise in healthy
subjects and in patients with hereditary spherocytosis. J. Clin. Lab. Invest. 38: 319-325.
Wharton, D. C.; Gibson, Q. H. (1976) Cytochrome oxidase from Pseudomonas aeruginosa: IV. reaction with
oxygen and carbon monoxide. Biochim. Biophys. Acta 430: 445-453.
50
White, R. E.; Coon, M. J. (1980) Oxygen activation by cytochrome P-450. Annu. Rev. Biochem. 49: 315-356.
Wittenberg, B. A.; Wittenberg, J. B.; Caldwell, P. R. B. (1975) Role of myoglobin in the oxygen supply to red
skeletal muscle. J. Biol. Chem. 250: 9038-9043.
March 15, 1990 9-39 DRAFT - DO NOT QUOTE OR CITE
-------
Wohlrab, H.; Ogunmola, B. G. (1971) Carbon monoxide binding studies of cytochrome a^ hemes in intact rat
liver mitochondria. Biochemistry 10: 1103-1106.
Wyman, J.; Bishop, G.; Richey, B.; Spokane, R.; Gill, S. (1982) Examination of Haldane's first law for the
5 partition of CO and O2 to hemoglobin A,,. Biopolymers 21: 1735-1747.
Young, L. J.; Caughey, W. S. (1986) Mitochondrial oxygenation of carbon monoxide. Biochem. J.
239: 225-227.
10 Young, L. J.; Choc, M. G.; Caughey, W. S. (1979) Role of oxygen and cytochrome c oxidase in the
detoxification of CO by oxidation to CO2. In: Caughey, W. S.; Caughey, H., eds. Biochemical and
clinical aspects of oxygen: proceedings of a symposium; September 1975; Fort Collins, CO. New York,
NY: Academic; pp. 355-361.
15 Zankl, J. G. (1981) Berechnung der CO-Aufnahme des menschlichen Blutes [Calculation of the uptake of CO into
human blood]. In: Mitteilungen des Institutes fuer Verbrennungskraftmaschinen und Thermodynamik.
Graz, V/est Germany: Technische University. Available from: NTIS, Springfield, VA; DE83-901060.
Zorn, H. (1972) The partial oxygen pressure in the brain and liver at subtoxic concentrations of carbon
20 monoxide. Staub Reinhalt. Luft 32: 24-29.
March 15, 1990 9-40 DRAFT - DO NOT QUOTE OR CITE
-------
10. HEALTH EFFECTS OF CARBON MONOXIDE
10.1 INTRODUCTION
5 Concerns about the potential health effects of exposure to CO have been addressed in
extensive studies with various animal species as subjects. Under varied experimental
protocols, considerable information has been obtained on the toxicity of CO, its direct effects
on the blood and other tissues, and the manifestations of these effects in the form of changes
in organ function. Many of these studies, however, have been conducted at extremely high
10 levels of CO (i.e., levels not found in ambient air). Although severe effects from exposure to
these high levels of CO are not directly germane to the problems from exposure to current
ambient levels of CO, they can provide valuable information about potential effects of
accidental exposure to CO, particularly those exposures occurring indoors. These higher level
studies, therefore, are being considered in this chapter only if they extend dose-response
15 information or if they provide clues to other potential health effects of CO that have not been
identified already. Emphasis has been placed on studies conducted at ambient or near-
ambient concentrations of CO that have been published in the more recent peer-reviewed
literature since completion of the previous criteria document (U.S. Environmental Protection
Agency, 1979) and an addendum to that document (U.S. Environmental Protection Agency,
20 1984). Where appropriate, information available from older studies either has been
summarized in the text or placed in tables.
The effects observed from nonhuman experimental studies have provided some insight
into the role CO plays in cellular metabolism. Caution must be exercised, however, in
extrapolating the results obtained from these data to man. Not only are there questions
25 related to species differences, but exposure conditions differ markedly in the studies
conducted by different investigators. Although these studies must be interpreted with caution,
they do serve the valuable ends of (1) suggesting studies to be verified in man, (2) exploring
the properties and principles of an effect much more thoroughly and extensively than is
possible in man, (3) protecting human subjects from unwarranted exposure, (4) permitting a
30 compression of exposure duration in relation to aging as a result of the shorter life
March 12, 1990 10-1 DRAFT-DO NOT QUOTE OR CITE
-------
expectancies of laboratory animals, and (5) providing tissues, organs, and cellular material
more readily, allowing more precise observation of specific functions.
Fortunately, our knowledge of the influence of CO on biological systems is not limited
to studies on nonhuman animals. Many direct experiments on humans have been conducted
5 during the last century. Although many reports describe inadvertent exposures to various
levels of CO, there are a considerable number of precise and delineated studies utilizing
human subjects. Most of these have been conducted by exposing young adult males to
concentrations of CO equivalent to those frequently or occasionally detected during ambient
monitoring. Research on human subjects, however, also can be limited by methodological
10 problems. As with the literature on experimental laboratory animals, many methodological
and reporting problems make the data difficult to interpret. These problems include
(1) failure to measure blood COHb levels; (2) failure to distinguish between the physiological
effects from a CO dose of high concentration (i.e., bolus effect) and the slow, insidious
increment in COHb over time from lower inhaled CO concentrations; (3) failure to
15 distinguish between normal blood flow and blood flow increased in response to hypoxia
(compensatory responses); and (4) the use of small numbers of experimental subjects. Other
factors involve failure to provide control measures (e.g., double-blind conditions) for
experimenter bias and experimenter effects; control periods so that task-learning effects do not
mask negative results; homogeneity in the subject pool, particularly in groups labeled
20 "smokers"; control of possible boredom and fatigue effects; and poor or inadequate statistical
treatment of the data. In this chapter, an effort will be made to account for such
methodological and reporting problems whenever possible by making appropriate comments
in the text. Contributors to this chapter are limited, however, by the data provided in the
reports published in the peer-reviewed literature. For example, information on the COHb
25 levels achieved and the duration of exposure utilized in the studies will be provided in the text
or tables if they were available in the original manuscript. Where this information is lacking,
only the CO levels (parts per million) will be reported.
An almost universal problem in research on both humans and laboratory animals is the
use of inappropriate statistical techniques for data analysis. Experimenters commonly use
30 tests designed for simple two-group designs when analysis of variance is required, or use of
several univariate tests when more than one dependent variable is measured and multivariate
March 12, 1990 10-2 DRAFT-DO NOT QUOTE OR CITE
-------
tests are inappropriate. Such statistical problems usually yield results in which the p- value is
too small, so that possibility exists that too many results were falsely declared to be
statistically significant. Possible consequences of such errors will be discussed in the text or
appropriate corrections will be made. Unless actual p values are given, all statements of
5 effects reported in the text or tables are statistically significant at p<0.05.
One problem that is particularly unique to human research is that only low levels of CO
exposure are commonly used. In such instances of low-level exposure, research findings
necessarily deal with near-threshold effects. When research, by necessity, is restricted to
such barely noticeable effects it may be expected that (1) results will be more variable because
10 of statistical sampling fluctuations, and (2) other uncontrolled variables that also affect the
dependent variable in question will be of major importance and will increase the variability of
results. For these reasons, data on human subjects, although being of prime interest, also will
be of highest variability. Such high variability must be resolved with (1) large groups of
subjects, (2) theoretical interpretation of results relying on knowledge gained from
15 experimental laboratory animal data, and (3) consideration of consistency of the data within
and across experiments.
This chapter is intended to review available data from published studies in which both
humans and laboratory animals have been exposed to low levels of CO. The chapter is
4
divided according to specific health effects starting with pulmonary and cardiovascular effects.
20 The neurobehavioral effects of CO are described next, followed by developmental toxicity and
other systemic effects of CO. Finally, adaptation to CO exposure is discussed. An
introduction and summary is provided for each major section of the chapter in order to set the
tone for a clearer understanding of the health effects of CO. Although human and laboratory
animal data may be presented separately under each effect category, the summary and
25 conclusion of these sections makes an attempt to integrate the relevant material from each of
these types of studies.
March 12, 1990 10-3 DRAFT-DO NOT QUOTE OR CITE
-------
10.2 ACUTE PULMONARY EFFECTS OF CARBON MONOXIDE
10.2.1 Introduction
The binding of carbon monoxide to hemoglobin, producing COHb, decreases the
O2-carrying capacity of blood and interferes with O2 release at the tissue level; these two main
5 mechanisms of action underly the potentially toxic effects of low-level CO exposure (see
Chapter 9). Impaired delivery of O2 can interfere with cellular respiration and result in tissue
hypoxia. Hypoxia of sensitive tissues, in turn, can affect the function of many organs
including the lungs. The effects would be expected to be more pronounced under conditions
of stress, as with exercise, for example. Although the physiological mechanism by which
10 adverse effects of COHb formation are well known, CO-induced toxicity at the cellular level
and its related biochemical effects still are not fully understood. Other mechanisms of CO-
induced toxicity have been hypothesized, but none have been demonstrated to operate at
relatively low (near-ambient) CO exposure levels. The effect of CO on cytochromes involved
in cellular oxidative pathways is just one of the possible mechanisms of action of CO.
15 Mitochrondia, the principal site of oxygen utilization, are present in parenchymal lung cells
and the highest concentrations are found in the type 2 epithelial cell. Prolonged exposure to
low levels of CO, therefore, may potentially interfere with cell function and cause loss of
alveolar epithelial integrity.
This section will review the available literature on morphological effects of CO and
20 determine if it is likely that CO can cause direct toxicity to cells lining the respiratory tract
through an effect on O2 transport or cellular metabolism. In addition, this section will review
a predominately newer data base on the effects of CO on pulmonary function.
10.2.2 Effects on Lung Morphology
25 Reports appearing in the published literature have investigated the histotoxic effects of
CO on lung parenchyma and vasculature, an area not reviewed in the previous criteria
document (U.S. Environmental Protection Agency, 1979). Results from human autopsies
have indicated that severe pulmonary congestion and edema was produced in the lungs of
individuals who died from acute smoke inhalation resulting from fires (Burns et al., 1986;
30 Fein et al., 1980). These individuals, however, were exposed to relatively high
concentrations of CO as well as other combustion components of smoke, such as carbon
March 12, 1990 10-4 DRAFT-DO NOT QUOTE OR CITE
-------
dioxide, hydrogen cyanide, various aldehydes (e.g., acrolein), hydrochloric acid, phosgene,
and ammonia (see Section 11.3.2). If CO, contained in relatively high concentrations in the
inhaled smoke, was responsible for the pathological sequelae described in fire victims, then to
what extent can edema be attributed to the primary injury of capillary endothelial or alveolar
5 epithelial cells?
10.2.2.1 Studies in Laboratory Animals
Laboratory animal studies by Niden and Schulz (1965) and Fein et al. (1980) found that
very high levels of CO (5000 to 10,000 ppm) for 15 to 45 min were capable of producing
10 capillary endothelial and alveolar epithelial edema in rats and rabbits, respectively. Evidence
of increased capillary permeability to protein also was reported in early studies on human
subjects by Siggaard-Andersen et al. (1968) and Parving (1972) following acute, high-level
CO exposure. These effects of CO have not been reported, however, at lower levels of CO
exposure.
15 In a small number (n = 5) of New Zealand white rabbits, Fein et al. (1980) reported a
significant increase in the permeability of 51Cr-EDTA from alveoli to arterial blood within
15 min after the start of exposure to 0.8% (8000 ppm) CO. Passage of this labeled marker
persisted and increased throughout the remaining 30 min of the study. The mean COHb level
after exposure was 63 + 4 (SEM) percent. Although morphometric examination was not
20 performed, transmission electron microscopy (TEM) showed evidence of capillary endothelial
and alveolar epithelial swelling and edema along with detachment of the endothelium from the
basement membrane. Mitochondria were disintegrated and alveolar type 2 cells were depleted
of lamellar bodies. None of these effects were found in four control animals exposed to air.
Despite an increase in gross lung weight, Penney et al. (1988a) were unable to
25 demonstrate any evidence of edema in the lungs of male albino rats after 7.5 weeks of
exposure to incrementally increasing concentrations of CO ranging from 250 to 1300 ppm.
The authors also reported that this effect was not due to increased blood volume in the lung
nor due to fibrosis, as measured by lung hydroxyproline content. There was, therefore, no
obvious explanation for the lung hypertrophy reported in this study after chronic exposure to
30 high concentrations of CO.
March 12, 1990 10-5 DRAFT-DO NOT QUOTE OR CITE
-------
Fisher et al. (1969) failed to find any histologic changes in the lungs of mongrel dogs
exposed to CO concentrations of 8000 to 14,000 ppm for 14 to 20 min (up to 18% COHb).
Similarly, no morphological changes were found by Hugod (1980) in the lungs of adult
rabbits continuously exposed to 200 ppm CO for up to six weeks (range of 11.9 to 19%
5 COHb) or to 1900 ppm CO for five hours (range of 31 to 39% COHb).
Niden (1971) speculated about possible effects of low levels of CO on cellular oxidative
pathways when he reported that exposure of mice to concentrations of CO from 50 to 90 ppm
for one to five days, resulting in COHb levels of < 10%, produced increased cristae in the
mitochondria and dilation of the smooth endoplasmic reticulum in the nonciliated bronchiolar
10 (Clara) cell. Minimal changes, consisting of fragmentation of lamellar bodies, were found in
the type 2 epithelial cell. Morphological appearance of the remaining cells of the terminal
airways was normal. The results of this study were not presented in detail, however, and
have not been confirmed at low concentrations of CO. Thus, the significance, if any, of
changes in the structure of cells lining the terminal airways is unknown.
15 Weissbecker et al. (1969) found no significant changes in the viability of alveolar
macrophages exposed in vitro to high concentrations of CO (up to 190,000 ppm). These
results were later confirmed in more extensive in vivo exposure studies by Chen et al. (1982).
They obtained alveolar macrophages by bronchoalveolar lavage from rats exposed to
500 ppm (41 to 42% COHb) from birth through 33 days of age. Morphological and
20 functional changes in the exposed cells were minimal. There were no statistically significant
differences in cell number, viability, maximal diameter, surface area, or acid phosphatase
activity. The phagocytic ability of alveolar macrophages was enhanced by CO exposure, as
determined by a statistically significant (p<0.05) increase in the percentage of spread forms
and cells containing increased numbers of retained latex particles. The biological
25 significance of this finding is questionable, however, because very few (n = 5) animals were
evaluated and no follow-up studies have been performed.
March 12, 1990 10-6 DRAFT-DO NOT QUOTE OR CITE
-------
10.2.2.2 Studies in Humans
In a study by Parving (1972) on 16 human subjects, transcapillary permeability to
131I-labeled human serum albumin increased from an average 5.6% per hour in controls to
7.5% per hour following exposure to CO. The subjects were exposed for three to five hours
5 to 0.43% (4300 ppm) CO, producing approximately 23% COHb. There were no associated
changes in plasma volume, hematocrit, or total protein concentration.
The only other relevant permeability studies were conducted on cigarette smoke. Mason
et al. (1983) showed rapidly reversible alterations in pulmonary epithelial permeability
induced by smoking using "TcDTPA as a marker. This increased permeability reverted to
10 normal fairly rapidly when subjects stopped smoking (Minty et al., 1981). Using a rat
model, the permeability changes associated with cigarette smoke were demonstrated later by
Minty and Royston (1985) to be due to the paniculate matter contained in the smoke. The
increase in ""TcDTPA clearance observed after exposure to dilute whole smoke did not occur
when the particles were removed, suggesting that the CO contained in the gaseous phase does
15 not alter permeability of the alveolar-capillary membrane.
10.2.3 Effects on Lung Function
10.2.3.1 Lung Function in Laboratory Animals
Laboratory animal studies of lung function changes associated with CO exposure parallel
20 the morphology studies previously described (see Section 10.2.2) because high concentrations
(1500-10,000 ppm) of CO were utilized.
Fisher et al. (1969) ventilated the left lung of seven dogs with 8 to 14% CO for 14 to
20 min. Femoral artery blood COHb levels ranged from 8 to 18% at the end of CO
breathing. No changes in the diffusing capacity or pressure-volume characteristics of the lung
25 were found.
Fein et al. (1980) measured lung function in the same study discussed in
Section 10.2.2. Nine New Zealand white rabbits were exposed for 45 min to either 0.8% CO
or air. After CO exposure, COHb levels reached 63%. Dynamic lung compliance
significantly decreased and airway resistance significantly increased at 15 and 30 min after the
30 start of CO exposure, respectively. Mean blood pressure fell to 62% of the baseline value by
March 12, 1990 10-7 DRAFT-DO NOT QUOTE OR CITE
-------
the end of exposure; heart rate was not changed. Arterial pH decreased progressively
throughout exposure although there were no changes in the alveolar-arterial PO2 difference.
Robinson et al. (1985), also interested in the effects of acute CO poisoning in humans,
used mongrel dogs to examine ventilation (VJ and perfusion (Q) distribution during and
5 following CO exposure. A small number (n = 5) of mongrel dogs were exposed to 1 % CO
(10,000 ppm) for 10 min, resulting in peak COHb levels of 59 ± 5.4%. Inert gas
distributions were measured at peak exposure and 2,4, and 24 h after exposure. No changes
in VA/Q were found. Previous studies were unable to show accumulation of lung water in
the same model (Halebian et al., 1984a,b). The authors concluded that other constituents of
10 smoke, besides CO, were responsible for the pulmonary edema and VA/Q mismatching
found in victims exposed to smoke in closed-space fires.
Very little is known about the effects of CO on ventilation in laboratory animals and the
few studies available are contradictory. No effects of CO on ventilation were found in
unanesthetized rabbits (Korner, 1965) or cats (Neubauer et al., 1981), while large increases
15 were reported in conscious goats (Chapman et al., 1980; Doblar et al., 1977; Santiago and
Edelman, 1976). In anesthetized cats, high concentrations of CO (10,000 ppm) increased
ventilation (Lahiri and Delaney, 1976). Gautier and Bonora (1983) used cats to compare the
central effects of hypoxia on control of ventilation under conscious and anesthetized
conditions. The cats were exposed for 60 min to either low inspired O2 fraction
20 (FIO2 = O.l 15) or CO diluted in air. In conscious cats, 1500 ppm CO caused a decreased
ventilation, while higher concentrations (2000 ppm) induced first a small decrease followed
by tachypnea that is typical of hypoxic hypoxia in carotid-denervated conscious animals. In
anesthetized cats, however, CO caused only mild changes in ventilation.
Other respiratory effects of CO hypoxia, such as the increased total pulmonary
25 resistance estimated by trachea! pressure, have been reported in anesthetized laboratory rats
and guinea pigs (Mordelet-Dambrine et al., 1978; Mordelet-Dambrine and Stupfel, 1979).
The significance of this effect is unknown, however, particularly under the extremely high
CO exposure conditions utilized in these studies (4 min inhalation of 2.84% CO) that
produced COHb concentrations >60% (Stupfel et al., 1981). Similar increases in tracheal
30 pressure also were seen with hypoxic hypoxia (FIO2 = 0.89), suggesting a possible general
mechanism associated with severe tissue hypoxia.
March 12, 1990 10-8 DRAFT-DO NOT QUOTE OR CITE
-------
10.2.3.2 Lung Function in Humans
Human studies of pulmonary function mostly are devoted to the identification of effects
occurring in the lungs of individuals exposed to relatively high concentrations of CO. Older
studies in the literature describe the effects of brief, controlled experiments with high CO-air
5 mixtures. Chevalier et al. (1966) exposed 10 subjects to 5000 ppm CO for 2 to 3 min until
COHb levels reached 4%. Measurements of pulmonary function and exercise studies were
performed before and after exposure. Inspiratory capacity and total lung capacity decreased
7.5 (p<0.05) and 2.1% (p<0.02), respectively, while maximum breathing capacity increased
5.7% (p<0.05) following exposure. Mean resting diffusing capacity of the lungs decreased
10 7.6% (p<0.05) compared to air-exposed controls. Fisher et al. (1969) exposed a small
number (n = 4) of male subjects, aged 23 to 36 years, to 6000 ppm CO for 6 s, resulting in
estimated COHb concentrations of 17 to 19%. There were no significant changes in lung
volume, mechanics, or diffusing capacity. Neither of these studies was definitive, however,
and no follow-up studies were reported.
15 More recent studies in the literature describing effects of CO on pulmonary function
have been concerned with exposure to the products of combustion and pyrolysis from such
sources as tobacco, fires, or gas- and kerosene-fueled appliances and engines. One group of
individuals, representing the largest proportion of the population exposed to CO, is tobacco
smokers. The reader is referred to Section 11.4 for a discussion on environmental tobacco
20 smoke and to other reviews on the direct effects of smoking.
A second group evaluated for potential changes in acute ventilatory function includes
occupations where individuals are exposed to variable, and often unknown, concentrations of
CO in both indoor and outdoor environments (see Section 8.4 for a more complete discussion
of occupational exposure to CO). Firefighters, tunnel workers, and loggers are typical
25 examples of individuals at possible risk. Unfortunately, as described above in
Section 10.2.2, these individuals also are exposed to high concentrations of other combustion
components of smoke and exhaust. It is very difficult to separate the potential effects of CO
from those due to other respiratory irritants (see Section 11.3.2 for more complete discussion
of exposure to combustion products).
30 Firefighters previously have been shown to have a greater loss of lung function
associated with acute and chronic exposure to smoke inhalation (as reviewed by Sparrow
March 12, 1990 10-9 DRAFT-DO NOT QUOTE OR CITE
-------
et al., 1982). None of these earlier studies, however, characterized the exposure variables,
particularly the concentrations of CO found in smoke, nor did they report the COHb levels
found in firefighters after exposure. Most reports of lung function loss associated with other
occupational exposures also fail to characterize exposure to CO. The following studies have
5 attempted to monitor, or at least estimate, the CO and COHb levels found in occupational
settings where lung function also was measured.
Sheppard et al. (1986) reported that acute decrements in lung function were associated
with routine firefighting. Baseline airway responsiveness to methacholine was measured in
29 firefighters from one fire station in San Francisco, CA, who were monitored over an
10 eight-week period. Spirometry was measured before and after each 24-h workshift and after
each fire. Exhaled gas was sampled 55 times from 21 firefighters immediately after each fire
and analyzed for CO. Despite the use of personal respiratory protection, exhaled CO levels
exceeded 100 ppm on four occasions, with a maximum of 132 ppm, corresponding to
calculated COHb values of 17 to 22%. Of the 76 spirometry measurements obtained within
15 two hours after a fire, 18 showed a greater fall in forced expiratory volume (FEV,) and/or
forced vital capacity (FVC) compared to routine workshifts without fires. Decrements in lung
function persisted for as long as 18 h in some of the individuals, but they did not appear to
occur selectively in those individuals with preexisting airway hyperresponsiveness.
Evans et al. (1988) reported on changes in lung function and respiratory symptoms
20 associated with exposure to automobile exhaust among bridge and tunnel officers. Spirometry
and symptom questionnaires were administered on a voluntary basis to 944 officers of the
Triborough Bridge and Tunnel Authority in New York City over an 11-year period between
1970 and 1981. Regression analyses were performed on 466 individuals (49%) who had been
tested at least three times during that period. Carboxyhemoglobin levels were calculated from
25 expired-air breath samples. Small, but significant differences were found between the bridge
and tunnel officers. Estimated levels of COHb were consistently higher in tunnel workers
compared to bridge workers for both nonsmoking individuals (1.96 and 1.73%, respectively)
and smoking individuals (4.47 and 4.25%, respectively). Lung function measures of FEV,
and FVC were reduced, on an average, in tunnel versus bridge workers. There were no
30 reported differences in respiratory symptoms except for a slightly higher symptom prevalence
in tunnel workers who smoked. Because differences in lung function between the two groups
March 12, 1990 10-10 DRAFT-DO NOT QUOTE OR CITE
-------
were small, it is questionable if the results are clinically significant or if they were even
related to CO exposure.
Hagberg et al. (1985) evaluated the complaints of 211 loggers reporting dyspnea and
irritative symptoms in their eyes, nose, and throat after chain-saw use. Measurements of lung
5 spirometry, COHb, and exposure to CO, HCs, and aldehydes were conducted on 23 loggers
over 36 work periods lasting two hour each. Ventilation levels during tree felling averaged
41 L/min. Carboxyhemoglobin levels increased after chain-saw use (p<0.05) but were
weakly correlated (r = 0.63) with mean CO concentrations of 17 ppm (4 to 73 ppm range) in
nonsmokers. Corresponding COHb levels were apparently <2%; unfortunately, the absolute
10 values before and after exposure were not reported. Peripheral bronchoconstriction, measured
by a decreased FEV/FVC (p<0.03) and forced expiratory flow (FEF)^^ (p< 0.005), was
found after the work periods but no correlations were obtained between lung function, COHb
levels, and exposure variables. There were no reported changes in FEV, or FVC.
High CO concentrations also can be found indoors near unvented space heaters (see
15 Section 7.2). The potential effects on lung function by indoor combustion products of
kerosene space heaters was evaluated by Cooper and Alberti (1984). Carbon monoxide and
SO2 concentrations were monitored in 14 suburban homes in Richmond, VA, during January
and February of 1983 while modern kerosene heaters were in operation. Spirometry was
measured in 29 subjects over a two-day period, randomizing exposures between days with and
20 without the heater on. During heater operation, CO concentration was 6.8 ± 5.9 ppm (0 to
14 ppm range), and SO2 concentration was 0.4 ± 0.4 ppm (0 to 1 ppm range). On control
days, indoor CO concentration was 0.14 + 0.53 ppm, whereas SO2 was undetectable. Six of
the homes had CO concentrations exceeding the NAAQS primary eight-hour standard of
9 ppm. Corresponding outdoor CO concentrations were 0 to 3 ppm. Carboxyhemoglobin
25 levels significantly increased from 0.82 ± 0.43% on control days to 1.11 ± 0.52% on days
when kerosene heaters were used. Exposure to heater emissions, however, had no effect on
FVC, FEV,, or maximum mid-expiratory flow rate (MMFR).
Most of the published community population studies on CO have investigated the
relationship between ambient CO levels and hospital admissions, deaths, or symptoms
30 attributed to cardiovascular diseases (see Section 10.3). Little epidemiological information is
March 12, 1990 10-11 DRAFT-DO NOT QUOTE OR CITE
-------
available on the relationship between CO and pulmonary function, symptomatology, and
disease.
One study by Lutz (1983) attempted to relate levels of ambient pollution to pulmonary
diseases seen in a family practice clinic in Salt Lake City, UT, during the winter of 1980-
5 1981, when heavy smog conditions prevailed. Data on patient diagnoses; local climatological
conditions; and levels of CO, O3, and paniculate matter were obtained over a 13-week period.
Pollutant levels were measured daily and then averaged for each week of the study; absolute
values were not reported. For each week, weighted simple linear regression and correlation
analyses were performed. Significant correlations (p = 0.01) between pollution-related
10 diseases and the environmental variables were found for paniculate (r = 0.79), O3
(r = -0.67), percent of smoke and fog (r = 0.79), but not for CO (r = 0.43) or percent of
cloud cover (r = 0.33). The lack of a significant correlation with CO was explained by a
small fraction (2%) of diagnoses for ischemic heart disease compared to a predominance of
respiratory tract diseases such as asthma, bronchitis, bronchiolitis, and emphysema.
15 Daily lung function in a large community population exposed to indoor and outdoor air
pollution was measured in Tucson, AZ, by Lebowitz et al. (1983a,b, 1984, 1985, 1987),
Lebowitz (1984), and Robertson and Lebowitz (1984). Subsets of both healthy subjects and
subjects with asthma, allergies, and airway obstructive disease were drawn from a symptom-
stratified, geographic sample of 117 middle-class households. Symptoms, medication use,
20 and peak flow measurements were recorded daily over a two-year period. Indoor and outdoor
monitoring was conducted in a random sample of 41 representative houses. Maximum one-
hour concentrations of O3, CO, and NO2 and daily levels of TSP, allergens, and meteoro-
logical variables were monitored at central stations within one-half mile of each population
subset. Indoor pollutant measurements were made for particles and CO, indicating that gas
25 stoves and tobacco smoking were the predominant indoor sources. Levels of CO were low,
averaging less than 2.4 ppm indoors and 3.8 to 4.9 ppm outdoors. Spectral time series
analysis was used to evaluate relationships between environmental exposure and pulmonary
effects over time (Lebowitz et al., 1987; Robertson and Lebowitz, 1984). Asthmatics were
the most responsive while healthy subjects showed no significant responses. Outdoor O3,
30 NO2, allergens, meteorology, and indoor gas stoves were significantly related to symptoms
and peak flow.
March 12, 1990 10-12 DRAFT-DO NOT QUOTE OR CITE
-------
10.2.4 Summary
Currently available studies on the effects of CO exposures producing COHb
concentrations of up to 39% fail to find any consistent effects on lung parenchyma and
vasculature (Hugod, 1980; Fisher et al., 1969) or on alveolar macrophages (Chen et al.,
5 1982; Weissbecker et al., 1969). The lack of significant changes in lung tissue is consistent
with the lack of histologic changes in the pulmonary and coronary arteries (see
Section 10.3.4). Alveolar epithelial permeability to 51Cr-EDTA increased in rabbits (Fein
et al., 1980) exposed to high concentrations of CO (63% COHb), and increased capillary
endothelial permeability to 131I-labeled human serum albumin was reported in early human
10 studies (Parving, 1972) following acute, high-level CO exposure (23% COHb); however, no
accumulation of lung water was found in dogs (Halebian et al., 1984a,b) with COHb levels of
59% and no edema was found in the lungs of rats chronically exposed to CO concentrations
as high as 1300 ppm (Penney et al., 1988a). In addition, no changes in diffusing capacity of
the lung were found in dogs with COHb levels up to 18% (Fisher et al., 1969). It is
15 unlikely, therefore, that CO has any direct effect on lung tissue except at extremely high
concentrations. The capillary endothelial and alveolar epithelial edema found with high levels
of CO exposure in victims of CO poisoning may be secondary to cardiac failure produced by
myocardial hypoxia (Fisher et al., 1969) or may be due to acute cerebral anoxia (Naeije
etal., 1980).
20 Ventilatory responses to CO are related to the CO concentration as well as to the
experimental conditions and the animal species being studied. In conscious goats (Chapman
et al., 1980; Doblar et al., 1977; Santiago and Edelman, 1976) and cats (Gautier and Bonora,
1983), after an initial depression, ventilation suddenly increases, particularly at high CO
concentrations (>2000 ppm). This response may result from the direct effects of hypoxia
25 (and possibly central acidosis) and/or a specific CNS effect of CO (see Section 10.3). No
effects on ventilation and perfusion distribution were found, however, in dogs exposed to 1 %
CO for 10 min, resulting in COHb levels of 59% (Robinson et al., 1985). At very high
concentrations of CO (COHb >60%) total pulmonary resistance, measured indirectly by
trachea! pressure, was reported to increase (Mordelet-Dambrine et al., 1978; Mordelet-
30 Dambrine and Stupfel, 1979).
March 12, 1990 10-13 DRAFT-DO NOT QUOTE OR CITE
-------
Human studies on the pulmonary function effects of CO are complicated by the lack of
adequate exposure information, the small number of subjects studied, and the short exposures
explored. Occupational or accidental exposure to the products of combustion and pyrolysis,
particularly indoors, may lead to acute decrements in lung function if the COHb levels are
5 > 17% (Sheppard et al., 1986) but not at concentrations <2% (Evans et al., 1988; Hagberg
et al., 1985; Cooper and Alberti, 1984). It is difficult, however, to separate the potential
effects of CO from those due to other respiratory irritants in the smoke and exhaust.
Community population studies on CO in ambient air have not found any relationships with
pulmonary function, symptomatology, and disease (Lebowitz et al., 1987; Robertson and
10 Lebowitz, 1984; Lutz, 1983).
10.3 CARDIOVASCULAR EFFECTS OF CARBON MONOXIDE
10.3.1 Introduction
15 Carbon monoxide exposure exerts deleterious effects in humans by several known
mechanisms. Carbon monoxide combines with Hb to form COHb, which directly decreases
the O2 content of blood. In addition, CO shifts the oxyhemoglobin dissociation curve to the
left, providing less O2 to the tissues at a given tissue PO2. Although no clinical studies have
been done, in vitro studies suggest that CO also may exert a deleterious effect in man with
20 coronary artery disease (injured vascular endothelium) by inhibiting the effects of oxyhemo-
globin on the action of acetylcholine (Ignarro et al., 1987). Acetylcholine causes a release of
endothelium-derived relaxing factor (EDRF). Carbon monoxide exposure in patients with
diseased endothelium could accentuate acetylcholine-induced vasospasm and aggravate silent
ischemia.
25 This section will discuss studies in man dealing with the effects in healthy individuals, in
patients with heart disease, and in other susceptible populations.
March 12, 1990 10-14 DRAFT-DO NOT QUOTE OR CITE
-------
10.3.2 Experimental Studies in Humans
10.3.2.1 Cardiorespiratory Response to Exercise
Effects in Healthy Individuals
The most extensive human studies on the Cardiorespiratory effects of CO are those
5 involving the measurement of O2 uptake during exercise. Tnese studies were discussed in the
previous CO criteria document (U.S. Environmental Protection Agency, 1979), an addendum
to that document (U.S. Environmental Protection Agency, 1984), and in other published
reviews (Horvath 1981; Shephard, 1983,1984).
Healthy young individuals were used in most of the studies evaluating the effects of CO
10 on exercise performance (see Table 10-1); healthy older individuals were studied in only two
(Raven et al., 1974a; Aronow and Cassidy, 1975). In these studies, O2 uptake during
submaximal exercise for short durations (5 to 60 min) was not affected by COHb levels as
high as 15 to 20% (Table 10-1). Under conditions of short-term maximal exercise, however,
statistically significant decreases (3 to 23%) in maximal O2 uptake (VO2 max) were found at
15 COHb levels ranging from 5 to 20% (Klein et al., 1980; Stewart et al., 1978; Weiser et al.,
1978; Ekblom and Huot, 1972; Vogel and Gleser, 1972; Pirnay et al., 1971). In another
study by Horvath et al. (1975), the critical level at which COHb marginally influenced VO2
max (p<0.10) was approximately 4.3%. The data obtained by Horvath's group and others
are summarized in Figure 10-1. There is a linear relationship between decline in VO2 max
20 and increase in COHb that can be expressed as percent decrease in VO2 max = 0.91 (%
COHb) + 2.2 (U.S. Environmental Protection Agency, 1979; Horvath (1981). Short-term
maximal exercise duration also has been shown to be reduced (3 to 38%) at COHb levels
ranging from 2.3 to 7% (Horvath et al., 1975; Drinkwater et al., 1974; Raven et al., 1974
a,b; Weiser et al., 1978; Ekblom and Huot, 1972). (See Table 10-1.)
25 Acute effects of cigarette smoke on maximal exercise performance are apparently similar
to those described above in subjects exposed to CO. Hirsch et al. (1985) studied the acute
effect of smoking on the Cardiorespiratory function during exercise in nine healthy male
subjects who were current smokers. They were tested twice-once after smoking three
cigarettes per hour for five hours and once after not having smoked. The exercise tests were
30 done on a bicycle ergometer with analysis of gas exchange and intra-arterial blood gases and
March 12, 1990 10-15 DRAFT-DO NOT QUOTE OR CITE
-------
I
TABLE 10-1. SUMMARY OF EFFECTS OF CARBON MONOXIDE ON MAXIMAL AND
SUBMAXIMAL EXERCISE PERFORMANCE
Exposure*'
50 and 100 ppm CO
4-h
treadmill exercise
at 85 % maximal
heart rate (HR)
50 ppm CO, 25 and
35 °C
5-min
treadmill exercise
to exhaustion
50 ppm CO, 35 °C
20-min
treadmill exercise
to exhaustion
50 ppm CO, 25 °C
5-min
treadmill exercise
to exhaustion
75 and 100 ppm CO
1 5-min
treadmill exercise
to exhaustion
100 ppm CO
Ih
treadmill exercise
to exhaustion
0.5% CO
2.5-3.5 min
5-min submaximal
exercise at
1.84L/minVO2
COHbc
2.17% (50 ppm)
4.15% (100 ppm)
2.3%
(nonsmokers)
5.1% (smokers)
2.5%
(nonsmokers)
4.1% (smokers)
2.7
(nonsmokers)
4.5% (smokers)
3.3-4.3%
3.95%
3.95%
Subjects(s)
23 males
20-38 years
(8 smokers)
16 males
40-57 years
(7 smokers)
20 young males
equally divided
by smoking
history
20 males
21-30 years
equally divided
by smoking
history
4 males
24-33 years
(1 smoker)
9 male
1 female
nonsmokers
44-55 years
10 nonsmokers
x = 30 years
Observed Effects'1
Mean exercise duration was
19 s shorter on CO days;
coagulation variables,
cholesterol, and triglycerides
were not significantly changed
•
No change in VO2 max; total
work time decreased at 25 °C in
older nonsmokers
•
No change in VO2 max; exercise
duration decreased in non
smokers; change in respiratory
pattern in both smokers and
nonsmokers
•
No change in VO2 max or work
time; no smoking effect
VO2 max decreased (p< 0.10) at
4.3% COHb; lower work times
and ventilatory volumes at all
COHb levels (p< 0.05)
Mean exercise time until
exhaustion decreased 5%
(p<0.001)
•
No change in mean VO2; O2 debt
per VO2 increased 14%
Conclusions
Submaximal exercise duration
decreased significantly at
4% COHb
No significant decrease in
VO2 max in older men exposed
to CO but work time to
exhaustion decreased in
nonsmokers at 2.3% COHb
Work time decreased in
nonsmokers at 2.5% COHb
No significant decrease in
maximal exercise performance
Maximal exercise performance
decreased at COHb >4%
Exercise time decreased
in older nonsmokers at
3.95% COHb
Work at 4% COHb was per-
formed with greater metabolic
cost
Reference*
Brinkhouse (1977)
Raven et al. (1974a)
Drinkwater et al. (1974)
Raven et al. (1974b)
Horvath et al. (1975)
Aronow and Cassidy
(1975)
Chevalier et al. (1966)
-------
0
1
e6
Conclusions
Observed Effects'1
Subjects(s
y
ffi
0
u
"a
g
1
.•s1
g
•a
«s
1
5
k o
"2 °
S 2 -S
No major change in i
respiratory response
submaximal work wi
levels <7%
Stroke volume decreased with
higher ambient temperature; HR
increased with CO exposure but
no change in cardiac output or
stroke volume
|l
G up
Ov 00
^
OO
vo
vq
Tf
C
CO
S *
CJ « x
S 8 |
&P 'g r.
|
1
X
|
g
o
3
3
O
II
Maximal exercise pei
and VO2 max decrea
increasing COHb
During maximal exercise, work
time and VO2 max significantly
decreased at 7-20% COHb; no
change in VO2 with submaximal
exercise
S $
8 !
?*
CO CO
*S
*"*?
£!
oo
Tf
X
£
e? « <5*
'•s E §•>
? a 'S *
.a 2 o S
H g S2
i?
•3
S
•§
3
£
Maximal O2 uptake
decreased at 5% COI
• *
VO2 max decreased while VE
and HR both increased
6 male non
nonsmokers
25-39 years
^
o
V)
o
30 ppm CO
5-h
treadmill exercis
until exhaustion
So1
2
•3
i,
'3
i
8
•gls
Maximal exercise pe:
in Denver, CO, (161
decreased at 5% CO!
Total exercise time decreased
3.8%; total work performed
decreased 10%, and VO2 max
decreased 2.8%
2
S
>>
|2 °8
1 1 «": | i
<* § S e (S
ss
»n
9
20-min rebreathi
to achieve target
COHb; treadmill
exercise until
exhaustion
Sj
•5
•s
«
4
i
li
Maximal exercise pe
decreased at 5.5% C
•
Maximal exercise time and VO2
decreased while HR and VE
increased
6 male
nonsmokers
25-39 years
&
"5
VI
4)
yi
O a U, 'g .-3
a » s B 1
i?
c
•a
.s
J
*
tf
Its
CO did not affect pu
function, subjective i
or exercise metabolU
k
W H
o „
1
ill
?
M
§ c
c S5 S3
S% -S *£
P) c V
r~ S oi
O 'S «
O u O
C O* ^
a -s *
>»°
*-t -1« j5 '«
B
*«3
o
.a
6
,
Maximal O2 uptake
decreased at 8% CO
•
VE and breathing
frequency (f^ jn-
creased while VO2 max
and (A-a) O2 difference
decreased with exercise
9 male
nonsmokers
g*
00
r~
_3
"o
?*-
if!
2" ll*
March 12, 1990
10-17 DRAFT-DO NOT QUOTE OR CITE
-------
cr
o
9
i— »
oo
TABLE 10-1 (cont'd). SUMMARY OF EFFECTS OF CARBON MONOXIDE ON MAXIMAL AND
SUBMAXIMAL EXERCISE PERFORMANCE
Exposure'-1" COHb'
15-min rebreathing 12.8-15.8%
to achieve target
COHb; exercise at
30, 70, and 100%
VO2 max
0.05% CO 15.4%
5-min
moderate exercise
for 15 min at
4 km/hr
0.15-0.35% CO 16-52%
>70 min
225 ppm CO 18-20%
1-h
bicycle exercise
aj 50, 75, and 100%
VO2max
225 ppm CO 20.3%
1-h
bicycle exercise
at 45, 75, and 100%
VO2max
Subjects(s)
9 males
23-34 years
5 males
24-35 years
4 males
21-33 years
8 males
20-23 years
(3 smokers)
16 males
(6 smokers)
Observed Effects'1
VO2 max decreased 14.2% with
maximal exercjse; no change in
ventilation or VO2 with sub-
maximal exercise
Increased HR but no change in
VO2 or ventilation with sub-
maximal exercise; VO2 max
decreased 15.1%
No hyperpnea at rest; arterial
PCO2 increased and pH
decreased; cardiac output
increased with increasing COHb
VO2 max decreased 23%
(p<0.001); with submaximal
exercise HR increased (p<0.05)
while VO2 was unchanged
VO2 max decreased 24%
(p<0.001); no change in work
efficiency or VO2 with
submaximal exercise
Conclusions
Maximal exercise performance
decreased after CO exposure
Maximal O2 uptake
decreased at 15% COHb
CO has a depressive action on
the respiratory center
Maximal O2 uptake
decreased at >18% COHb
Maximal O2 uptake
decreased at >20% COHb
Reference"
Ekblom et al. (1975)
Pirnay et al. (1971)
Chiodi et al. (1941)
Vogel and Gleser (1972)
Vogel et al. (1972)
'Exposure concentration, duration, and activity level.
bl ppm = 1.145 mg/m3; 1 mg/m3 = 0.873 ppm at 25°C, 760 mm Hg; \%
'Estimated or measured blood carboxyhemoglobin (COHb) levels.
dSee glossary of terms and symbols for abbreviations and acronyms.
'Cited in U.S. Environmental Protection Agency (1979; 1984).
10,000 ppm.
-------
X
o
IN
40
35-
30-
Z 25-1
LJ
§20-1
o:
O
is-
le-
s'
10 15 20 25
PERCENT COHb
30
35
40
Figure 10-1. Relationship^ between carboxyhemoglobin level (COHb) and decrement in
maximal oxygen uptake (VO2 max) for healthy nonsmokers.
Source: Adapted from U.S. Environmental Protection Agency (1979) and Horvath (1981).
March 12, 1990
10-19 DRAFT-DO NOT QUOTE OR CITE
-------
pressures. On the smoking day, the maximal O2 uptake was significantly decreased by 4%
and the anaerobic threshold was decreased by 14%. The rate-pressure product was a
significant 12% higher at comparable work loads of 100 watts on the smoking day compared
to the nonsmoking day. There were no changes due to smoking, however, on the duration of
5 exercise or on the mean work rate during maximal exercise testing. The blood COHb level
before exercise was 1.8% on the nonsmoking and 6.6% on the smoking day. At peak
exercise the COHb was 0.9% and 4.8%, respectively, on the nonsmoking and smoking day.
The authors concluded that the main adverse effect of smoking was due to CO, although the
increase in rate-pressure product also might be the result of the simultaneous inhalation of
10 nicotine. They felt that the magnitude of change in performance indicators corresponded well
with earlier reports.
It would be interesting, therefore, to determine if smokers and nonsmokers had different
responses to CO exposure. Unfortunately, smokers and nonsmokers were not always
identified in many of the studies on exercise performance, making it difficult to interpret the
15 available data. Information derived from studies on cigarette smoke is also sparse. As a
result, attempts to sort out the acute effects of CO from those due to other components of
cigarette smoke have been equivocal. Seppanen (1977) reported that the physical work
capacities of cigarette smokers decreased at 9.1% COHb levels after breathing either boluses
of 1100 ppm CO or after smoking cigarettes. The greatest decrease in maximal work,
20 however, was observed after CO inhalation.
Klausen et al. (1983) compared the acute effects of cigarette smoking and inhalation of
CO on maximal exercise performance. They studied 16 male smokers under three different
conditions: after eight hours without smoking (control), after inhalation of the smoke of three
cigarettes, and after CO inhalation. Just before maximal exercise testing the arterial CO
25 saturation reached 4.51 and 5.26% after cigarette smoke and CO inhalation, respectively,
compared to 1.51% for controls. Average maximal O2 uptake decreased by about 7% with
both smoke and CO. Exercise time, however, decreased 20% with smoke but only 10% with
CO, suggesting that nicotine, smoke particles, or other components of tobacco smoke may
contribute to the observed effects. The authors, therefore, concluded that a specified COHb
30 level induced by either smoke or CO decreased maximal work performance to the same
March 12, 1990 10-20 DRAFT-DO NOT QUOTE OR CITE
-------
degree. Of note is the more marked decrease in work time compared to maximal O2 uptake
induced by CO, a finding that agrees with the Ekblom and Huot (1972) results (see
Table 10-1).
If the magnitude of the effect of CO exposure is due only to a decrease in O2-carrying
5 capacity proportional to the COHb concentration, the magnitude should be roughly the same
as if the O2 capacity is decreased by anemia. Celsing et al. (1987) found in a series of very
carefully performed studies in normal subjects that maximal O2 uptake decreased by
19 mL/min/kg per gram per liter change in Hb over a wide range of Hb concentrations from
137 to 170 g/L. This decrease corresponds to a 2% decrease in maximal O2 uptake for every
10 3% decrease in Hb concentration in a well-trained subject. The decrease also corresponds to
the decrease in VO2 max reported by Ekblom and Huot (1972) and Horvath et al. (1975).
However, Ekblom and Huot found a much more marked effect on maximal work time (i.e.,
work on a constant load until exhaustion with a duration of about 6 min). An explanation for
the marked decrease in maximal work time could be that CO has a negative effect on the
15 oxidative enzymatic system wheareas the decrease in work time is due to a combination of a
decrease in O2 capacity and a less efficient oxidative enzymatic system. If the data are
extrapolated to lower COHb values, a 3% level of COHb should decrease the maximal work
time by about 20%. However, this decrease is more than the 10% average decrease reported
by Klausen et al. (1983) where they also found more marked effects in less well-trained
20 subjects compared to well-trained subjects.
Effects in Individuals with Heart Disease
The previous criteria document (U.S. Environmental Protection Agency, 1979)
concluded that patients with heart disease are especially at risk to CO exposure sufficient to
25 produce 2.5 to 3% COHb. This statement was based primarily on studies initiated by
Aronow et al. (1972) and Aronow and Isbell (1973) demonstrating that patients with angina
pectoris, when exposed to low levels of CO, experienced reduced time to onset of exercise-
induced chest pain as a result of insufficient O2 supply to the heart muscle. A study by
Anderson et al. (1973) reported similar results at mean COHb levels of 2.9 and 4.5% (see
30 Table 10-2).
March 12, 1990 10-21 DRAFT-DO NOT QUOTE OR CITE
-------
S
to
100 ppm CO for
60 min; postexposure
incremental exercise
at 48.6 L/min on
a cycle ergometer
117 or 253 ppm CO
for 50-70 min;
pre- and post-
exposure incremental
exercise at -6 METS
on a treadmill
(modified Naughton
protocol)
TABLE 10-2. SUMMARY OF EFFECTS OF CARBON MONOXIDE EXPOSURE IN
PATIENTS WITH ANGINA
Exposure"'1" COHbc COHbd
50 or 100 ppm CO 2.9% (SP) ND
for 50 min of 4.5% (SP)
each hour x 4 h;
postexposure
exercise on a
treadmill
ACOHb" Subject(s)
1.6% 10 males, 5 smokers
3.2% and 5 nonsmokers,
with reproducible
exercise-induced
angina; 49.9 years
Observed Effects
Duration of exercise before onset
of angina was significantly
shortened at 2.9 and 4.5% COHb
(p<0.005); duration of angina was
significantly prolonged at 4.5%
(p <0.01) but not at 2.9% COHb.
The response of smokers was not
significantly different from that
of nonsmokers.
Reference
Anderson et at.
(1973)f
3.0% (CO-Ox) 2.8% (CO-Ox) 1.5% 24 male nonsmokers
with reproducible
exercise-induced
angina; 59 ± 1 years
(49-66 years)
3.2% (CO-Ox) 2.7% (CO-Ox) 2.0% 63 male nonsmokers
5.6% (CO-Ox) 4.7% (CO-Ox) 4.4% with reproducible
exercise-induced
2.4% (GC) 2.0% (GC) 1.8% angina; 62 ± 8 years
4.7% (GC) 3.9% (GC) 4.0% (41-75 years)
Time to onset of angina decreased
5.9% (p = 0.046); no significant
effect on the duration of angina.
O2 uptake at angina was
reduced about 3% (p = 0.04).
There were no significant changes
in heart rate or systolic blood
pressure at angina.
Earlier onset of myocardial
ischemia was found with CO exposure:
time to ST endpoint decreased 5.1 and
12.1 % (p <0.05) and time to angina
onset decreased 4.2 and 7.1%
-------
I
TABLE 10-2 (cont'd). SUMMARY OF EFFECTS OF CARBON MONOXIDE EXPOSURE IN
PATIENTS WITH ANGINA
Exposure"'11
COHbc
COHb"
ACOHb"
Subjects)
Observed Effects
Reference
100-200 ppm CO
for 60 min; postex-
posure incremental
exercise at 317 KPM
on a cycle ergometer
3.8% (CO-Ox) 3.6% (CO-Ox) 2.2% 25 male and 5 female
nonsmokers with
evidence of exercise-
induced angina on at
least one day; 58 ±
11 years (36-75 years)
No significant difference in time
to onset or duration of angina. No
significant difference in maximal
exercise time, maximal ST segment
depression, or time to significant
ST segment depression during
exercise. No significant difference
in maximal ejection fraction; small
decreases in blood pressure
(p = 0.031) and change in ejection
fraction (p = 0.049) during CO
exposure requires further evaluation.
Actuarial analysis (Bissette et al.,
1986) including 3/30 subjects
experiencing angina only on the CO
exposure day showed a significant
decrease in time to onset of angina
after CO exposure.
Sheps et al.
(1987)
O
o
z
3
I
s
o
100-200 ppm CO
for 60 min; postex-
posure incremental
exercise @300 KPM
on a cycle ergometer
5.9% (CO-Ox) 5.2% (CO-Ox) 4.2% 22 male and 8 female
nonsmokers with evidence
of exercise-induced
angina on at least one
day; 58 ± 11 years
(36-75 years)
Earlier onset of ventricular
dysfunction, angina, and poorer
exercise performance was found with
CO exposure; mean duration of
exercise was shorter (p<0.05);
subjects were likely to experience
angina earlier during exercise with
CO (p<0.05) using actuarial analysis.
Both the level (p = 0.05) and change
in left ventricular ejection fraction
at submaximal exercise (p = 0.05)
were less on CO-exposure compared
to air-exposure day. There was no
significant difference in the peak
exercise left ventricular ejection
fraction.
Adams etal. (1988)
"Exposure concentration, duration, and peak activity level. NOTE: Because oxygen consumption was not measured, it is not possible to determine the actual level of exercise at which angina occurred.
1 = 1.145 mg/m3; 1 mg/m3 = 0.873 ppm at 25° C, 760 mm Hg; 1% = 10,000 ppm.
'Measured blood carboxyhemoglobin (COHb) level after CO exposure; GC = gas chromatograph; CO-Ox = CO-Oximeter; SP = spectrophotometric method of Buchwald (1969).
dMeasured blood carboxyhemoglobin (COHb) level after exercise stress test; GC = gas chromatograph; CO-Ox = CO-Oximeter; ND = not determined.
"Postexposure increase in COHb over baseline.
'Cited in U.S. Environmental Protection Agency (1979; 1984).
-------
In 1981, Aronow reported an effect of 2% COHb on time to onset of angina levels in
15 patients. The protocol was similar to previously reported studies, with patients exercising
until onset of angina. Only 8 of the 15 subjects developed more than 1 mm ischemic ST
segment depression at the onset of angina during the control periods. This was not
5 significantly affected by CO. It is questionable, therefore, as to whether the remaining
patients truly met adequate criteria for ischemia despite angiographically documented cardiac
disease. After breathing 50 ppm of CO for one hour, the patients' times to onset of angina
significantly decreased from a mean of 321.7 ± 96 s to 289.2 ± 88 s.
In 1983, the studies by Aronow and his colleagues were reevaluated by an ad hoc
10 committee to the EPA (especially the 1981 study). The committee concluded that the results
of Aronow's studies did not meet a reasonable standard of scientific quality and, therefore,
should not be used by the Agency in defining the critical COHb level at which adverse health
effects of CO are occurring. A summary of the committee findings and a reevaluation of the
key health effects information reported to be associated with relatively low level CO exposure
15 can be found in an addendum to the 1979 criteria document for CO (U.S. Environmental
Protection Agency, 1984).
The experimental design used in the Aronow studies on CO exposure effects in patients
with angina set the stage for subsequent, more precisely designed studies. Aronow and his
colleagues used the subjective measure of time to onset of angina as their only significant
20 variable of CO effect. In an attempt to improve upon these earlier preliminary studies, the
more recent studies employed electrocardiogram (EKG) changes as objective measures of
ischemia. Another consideration in the conduct of the newer studies on angina was to
establish better the dose response relationships for low levels of CO exposure. While the
COHb level is accepted as the best measure of the effective dose of CO, the reporting of low
25 level effects is problematic because of inconsistencies in the rigor with which the devices for
measuring COHb have been validated. The most frequently used technique for measuring
COHb has been the optical method found in the IL series of CO-Oximeters (CO-Ox). Not
only is there a lot of individual variability in these machines, but recent comparisons with the
gas chromatographic (GC) technique of measuring COHb have suggested that the optical
30 method may not be a suitable reference technique for measuring low levels of COHb. (See
Chapter 8, Section 8.5 for more details.) Several additional studies have appeared in the
March 12, 1990 10-24 DRAFT-DO NOT QUOTE OR CITE
-------
literature to help define better the precise COHb levels at which cardiovascular effects occur
in angina patients (see Table 10-2).
Sheps et al. (1987) studied 30 patients with ischemic heart disease age 38 to 75 years
and assessed not only symptoms during exercise, but radionuclide evidence of ischemia (left
5 ventricular ejection fraction changes). This study was designed to be representative of a
broad group of patients with myocardial ischemia. Patients were nonsmokers with ischemia,
defined either by exercise-induced ST segment depression, angina, or abnormal ejection
fraction response (i.e., all patients had documented evidence of ischemia).
Patients were exposed to CO (100 ppm) or air during a three-day, randomized double-
10 blind protocol to achieve a postexposure level of 4% COHb (CO-Ox measurement). Resting
preexposure levels were 1.7%, and postexposure levels averaged 3.8% on the CO exposure
day, thus the study examined acute elevation of COHb levels from 1.7% to an average of
3.8%, or an average increase of 2.2% from resting values. Comparing exposure to CO to
exposure to air, there was no significant difference in time to onset of angina, maximal
15 exercise time, maximal ST segment depression (1.5 mm for both), or time to significant ST
segment depression. The conclusion of this study was that 3.8% COHb produces no
clinically significant effects on this patient population.
Interestingly, further analysis of the time to onset of angina data in this paper
demonstrated that 3 of the 30 patients experienced angina on the CO exposure day but not on
20 the air control day. These patients had to be deleted from the classical analysis of differences
between time to onset of angina that was reported in the publication. However, actuarial
analysis of time to onset of angina including these patients revealed a statistically significant
difference in time to onset of angina favoring an earlier time under the CO-exposure
conditions (Bissette et al., 1986). None of the patients had angina only on the air exposure
25 day.
Subsequent work from these same investigators (Adams et al., 1988) focused on
repeating the study at 6% COHb (CO-Ox measurement) using an identical protocol and a
similar patient population. Postexposure COHb levels averaged 5.9 + 0.1% compared to
1.6 + 0.1% after air exposure, representing an increase of 4.3%. The mean duration of
30 exercise was significantly longer after air compared to CO exposure (626 + 50 s for air vs.
585 + 49 s for CO, p<0.05). Actuarial methods suggested that subjects experienced angina
March 12, 1990 10-25 DRAFT-DO NOT QUOTE OR CITE
-------
earlier during exercise on the day of CO exposure (p<0.05). In addition this study showed
that, at a slightly higher level of CO exposure, both the level and change in ejection fraction
at submaximal exercise were greater on the air day than on the CO day. The peak exercise
left ventricular ejection fraction, however, was not different for the two exposures.
5 These results demonstrated earlier onset of ventricular dysfunction and angina and
poorer exercise performance in patients with ischemic heart disease after acute CO exposure
sufficient to increase COHb to 6%. It is of interest that in both the 4% study and the 6%
study reported by this group, some of the patients experienced angina on the CO day, but not
on the air exposure day. There were no patients who experienced angina in the reverse
10 sequence.
Kleinman and Whittenberger (1985) and Kleinman et al. (1989) studied nonsmoking
male subjects with a history of stable angina pectoris and positive exercise tests. All but two
of the 26 subjects had additional confirmation of ischemic heart disease, such as previous
myocardial infarction, positive angiogram, positive thallium scan, prior angioplasty, or prior
15 bypass surgery. Subjects were exposed for one hour in a randomized double-blind crossover
fashion to either 100 ppm CO or to clean air on two separate days. Subjects performed an
incremental exercise test on a cycle ergometer to the point at which they noticed the onset of
angina. For the study group, the one-hour exposure to 100 ppm CO resulted in an increase
of COHb from 1.4% after clean air to 3% (CO-Ox measurement) after CO. For the entire
20 study group (n = 26), the one-hour exposure to 100 ppm resulted in a decrease of the time to
onset of angina by 6.9% from 6.5 to 6.05 min (Kleinman and Whittenberger, 1985). This
difference was significant in a one-tailed paired T-test (p = 0.03). When using a two-tailed
test the difference loses statistical significance at the p = 0.05 level.
In a subsequent publication of results from this study (Kleinman et al., 1989), the two
25 subjects with inconsistencies in their medical records and histories were dropped from the
analysis. For this study group (n = 24), the one-hour exposure to 100 ppm CO (3% COHb
by CO-Ox measurement) resulted in a decrease of time to onset of angina by 5.9%
(p = 0.046). There was no significant effect on the duration of angina but O2 uptake at
angina point was reduced 2.7% (p = 0.04). Only eight of the subjects exhibited depression
30 in the ST segment of their EKG traces during exercise. For this subgroup, there was a 10%
March 12, 1990 10-26 DRAFT-DO NOT QUOTE OR CITE
-------
reduction (p< 0.036) in time to onset of angina and a 19% reduction (p< 0.044) in the time
to onset of 0.1 mV ST segment depression.
A multicenter study of effects of low levels of COHb has been conducted on a relatively
large sample (n = 63) of individuals with coronary artery disease from three different cities
5 (Allred et al., 1989a,b). The purpose of this study was to determine the effects of 2.0% and
3.9% COHb (GC measurement) on time of onset of significant ischemia during a standard
treadmill exercise test. Significant ischemia was measured subjectively by the time of
exercise required for the development of angina (time of onset of angina) and objectively by
the time required to demonstrate a 1-mm change in the ST segment of the ECG (time to ST).
10 Male subjects, ages 41 to 75 (mean = 62.1 years) with stable exertional angina pectoris and
positive stress test, as measured by > 1-mm ST segment change, were studied. Further
evidence that these subjects had coronary artery disease was provided by the presence of at
least one of the following criteria: angiographic evidence of >70% narrowing of at least one
coronary artery, documented prior myocardial infarction, or a positive stress thallium
15 demonstrating an unequivocal perfusion defect. Thus, as opposed to some previous studies
reported, this study critically identified patients with documented coronary artery disease.
The protocol for this study was similar to that used in the Aronow studies because two
exercise tests were performed on the same day. The two tests were separated by a recovery
period and double-blind exposure period. On each of the three exposure days the subject
20 performed a symptom-limited exercise test on a treadmill, then was exposed for 50 to 70 min
to the test environment (clean air, 117 ppm CO, or 253 ppm CO), and then performed a
second symptom-limited exercise test. The time to the onset of angina and the time to 1-mm
ST change was determined for each test. The percent change following exposure at both 2.0
and 3.9% COHb (GC measurement) then were compared to the same subject's response to the
25 randomized exposure to room air (less than 2 ppm CO.)
When potential exacerbation of the exercise-induced ischemia by exposure to CO was
tested using the objective measure of time to 1-mm ST segment change, exposure to 2.0%
COHb resulted in a 5.1 % decrease (p = 0.01) in the time to attain this level of ischemia. At
3.9% COHb the decrease in time to the ST criterion was 12.1% (p^O.OOOl) relative to the
30 air-day results. At the 3.9% COHb level this reduction in time to ST depression was
accompanied by a significant (p = 0.03) reduction in the heart rate-blood pressure product
March 12, 1990 10-27 DRAFT-DO NOT QUOTE OR CITE
-------
(double product), an index of myocardial work. The maximal amplitude of the ST segment
change also was significantly affected by the CO exposures: at 2% COHb the increase was
11% (p = 0.002) and at 3.9% COHb the increase was 17% relative to the air day
(p
-------
drug interaction with the effects of CO). The major medications being used in this group
were betablockers (38/63), nitrates (36/63), and calcium-channel blockers (40/63). The other
major concern was the influence of the severity of the disease. The simplest approach to this
was to evaluate the influence of the duration of the exercise because the subjects with more
5 severe disease were limited in their exercise performance. No significant correlation was
found between duration of exercise and the change in the time to angina or ST criterion.
There also was no relationship between the average time of exercise until the onset of angina
and either of the endpoints. There also was no relationship between the presence of a
previous myocardial infarction and the study endpoints.
10 The duration of exercise was significantly shortened by the 3.9% COHb but not by the
2.0% level. This finding must be used cautiously because these subjects were not exercised
to their maximum capacity in the usual sense. The major reason for termination of the
exercise was the progression of the angina (306/376 exercise tests.) The subjects were to
grade their angina on a four-point scale, and when the exercise progressed beyond level two
15 they were stopped. Therefore this significant decrease in exercise time of 40 s at the 3.9%
COHb level is undoubtedly due to the earlier onset of angina followed by the normal rate of
progression of the severity of the angina.
The individual center data provides insight into the interpretation of other studies that
have been conducted in this area. Each of the centers enrolled the numbers of subjects that
20 have been reported by other investigators. The findings reported above were not substantiated
in all instances at each center. When one considers the responses of the group to even the
3.9% COHb, it is clear as to why one might not find significance in one parameter or
another. For the decrease in ST segment at 3.9%, only 49 of 62 subjects demonstrated this
effect on the day tested. The potential for finding significance at this effect rate with a
25 smaller sample size is reduced. Random sampling of this population with a smaller sample
easily could provide subjects that would not show significant effects of these low levels of CO
on the test day.
The recent report (Allred et al., 1989b) of the multicenter study conducted by the Health
Effects Institute (HEI) discusses some reasons for differences between the results of the
30 studies cited above (also see Table 10-2 and Table 10-3). The studies have different designs,
types of exercise tests, inclusion criteria (and, therefore, patient populations), exposure
March 12, 1990 10-29 DRAFT-DO NOT QUOTE OR CITE
-------
s
er
9
s
TABLE 10-3. COMPARISON OF SUBJECTS IN STUDIES OF THE EFFECT OF CARBON MONOXIDE
EXPOSURE ON OCCURRENCE OF ANGINA DURING EXERCISE
Subiect Characteristics
Number of
Study Subjects Gender
Anderson et al. (1973) 10 male
Kleinman et al. (1989) 24 male
Allied et al. 63 male
(1989a,b)
Medication
1 subject took
digitalis; drug
therapy basis
for exclusion
14 on betablockers;
19 on nitrates;
8 on Ca-channel
blockers
38 on betablockers;
36 on nitrates; 40
Smoking
History
5 smokers
(refrained for
12 h prior
to exposure)
No current
smokers
No current
smokers
Description of
Disease
Stable angina pectoris,
positive exercise test
(ST changes); reproducible
angina on treadmill
Ischemic heart disease,
stable exertional angina
pectoris
Stable exertional angina
and positive exercise test
Age
(years)
(mean = 49.9)
49-66
(mean = 59)
41-75
(mean - 62.1)
on calcium antagonists
Sheps et al. (1987)
30
(23 with
angina)
25 male
5 female
Adams et al. (1988)
30
22 male
8 female
26 subjects on
medication; 19 on beta
blockers; 11 on
Ca-channel blockers; 1
on long-acting nitrates
25 subjects on
medication; 13 on beta
blockers + Ca-channel
blockers; 6 on beta
blockers; 5 on
Ca-channel blockers;
1 on long-acting nitrates
No current
smokers
No current
smokers
(ST changes) plus 1 or more
of the following: (1) >70%
lesion by angiography in 1
or more major vessels,
(2) prior Ml, (3) positive
exercise thallium test
Ischemia during exercise
(ST changes or abnormal
ejection fraction response)
and 1 or more of the
following: (1) angic-
graphically proven CAD,
2) prior MI, (3) typical
angina
Ischemia during exercise
(ST changes or abnormal
ejection fraction response)
and 1 or more of the
following: (1) angio-
graphically proven CAD,
(2) prior MI, (3) typical
angina
36-75
(mem = 58.2)
36-75
(mean = 58)
Source: Adapted from Allred et al. (1989b)
-------
conditions, and means of measuring COHb. All of the studies have shown an effect of COHb
elevation on the time to onset of angina (see Figure 10-2). Results form the Kleinman et al.
(1989) study showed a 6% decrease in exercise time to angina at 3.0% COHb (CO-Ox
measurement) measured at the end of exposure. Allred et al. (1989a,b) reported a 5 and 7%
5 decrease in time to onset of angina after increasing COHb levels to 3.2 and 5.6% (CO-Ox
measurement), respectively, at the end of exposure. Both Sheps et al. (1987) and Adams
et al. (1988) reported a significant decrease in the time to onset of angina on days when
COHb levels at the end of exposure were 3.8 and 5.9% (CO-Ox measurement), respectively,
if the data analysis by actuarial method included subjects who experienced angina on the CO
10 but not the air day.
The multicenter study (Allred et al., 1989a,b) demonstrated a dose-response effect of
COHb on time to onset of angina. The only other single study that investigated more than a
single target level of COHb was Anderson et al. (1973) and their results did not show a dose
response for angina.
15 The time to onset of significant ECG ST-segment changes, which are indicative of
myocardial ischemia in patients with documented cornary artery disease (CAD), is a more
objective indicator of ischemia than angina. Allred et al. (1989a,b) reported a 5.1 and
12.1 % decrease in time to ST depression at COHb levels of 2.0 and 3.9% (GC
measurement), respectively, measured at the end of exercise. An additional measurement of
20 the ST change was made by Allred et al. (1989b) to confirm this response - all the leads
showing ST segment changes were summed. This summed ST score also was significantly
affected by both levels of COHb. The significant finding for the summed ST score indicates
that the effect reported for time to ST was not dependent upon changes observed in a single
EKG lead.
25 The differences between the results of these five studies on exercise-induced angina can
largely be explained by differences in experimental methodology and analysis of data and, to
some extent, by differences in subject populations and sample size. For example, the
Kleinman study and the Allred study used one-tailed p values. The Sheps and Adams studies
used two-tailed p values. If a two-sided p value was utilized on the time to onset of angina
30 variable observed at the lowest COHb level in the Allred study, it would become 0.054 rather
than 0.027, a result that would be considered borderline significant. If a two-sided p value
March 12, 1990 10-31 DRAFT-DO NOT QUOTE OR CITE
-------
o
UJ
30
25 -
20 -
15 -
10 -
O 5 -
UJ ^ ^
a
ui 0 -
o
a:
LJ
a.
-5 -
-10
ANDERSON
KLEINMAN
111
ALLRED
ALLRED
SHEPS
ADAMS
PERCENT COHb BY OPTICAL METHODS
Figure 10-2. The effect of CO exposure on time to onset of angina. For comparison across
studies, data are presented as the mean percent differences between air and CO exposure days
for individual subjects that were calculated from each of the studies. Bars indicate calculated
standard errors of the mean. (See text and Table 10-2 and Table 10-3 for more detailed
information).
Source: Adapted from Allred et al. (1989b).
March 12, 1990
10-32 DRAFT-DO NOT QUOTE OR CITE
-------
were used in the Kleinman study, the difference in time to onset of angina would lose
significance at the p = 0.05 level.
The entry criteria in the Allred study were more rigorous than in the other studies. All
subjects were required to have stable exertional angina and reproducible exercise-induced ST
5 depression and angina. Besides these criteria, all subjects were required to have either a
previous myocardial infarction (MI), angiographic disease, or a positive thallium stress test.
In addition, only men were studied. These strict entry criteria were helpful in allowing the
investigators to measure more precisely an adverse effect of CO exposure. However, because
of the difficulty the investigators had in recruiting subjects, some questions remain about how
10 representative the study population is of all patients in the United States with exercise-
induced ischemia. In addition, the protocol for the multicenter study was slightly different
from some of the protocols previously reported. On each test day, the subject performed a
symptom-limited exercise test on a treadmill, then was exposed for approximately one hour to
air or one of two levels of CO in air, and then underwent a second exercise test. Time to the
15 onset of ischemic EKG changes and time to the onset of angina were determined for each
exercise test. The percent difference for these endpoints from the pre- and postexposure test
then was determined. The results on the 2% target day and then the 4% target day were
compared to those on the control day.
The statistical significance reported at the low-level CO exposure is only present when
20 the differences between pre- and postexposure exercise tests are analyzed. Analysis of only
the postexposure test results in a loss of statistical significance for the 2% COHb level. Some
of the differences between the results of this multicenter study and previous studies may be
related to the fact that the exposure was conducted shortly after patients exercised to angina.
Because the effects of ischemia may have a variable duration (radionuclide studies have shown
25 metabolic effects of ischemia to last for over an hour), differences between pre- and
postexposure tests may have been due to effects of CO exposure on recovery from a previous
episode of exercise-induced ischemia rather than detrimental effects only during exercise.
In conclusion, five key studies have investigated the potential for CO exposure to
enhance the development of myocardial ischemia at <6% COHb during progressive exercise
30 tests. Despite differences between these studies, it is impressive that Figure 10-2 shows a
consistent relationship in percent decrease in time to onset of angina across multiple studies.
March 12, 1990 10-33 DRAFT-DO NOT QUOTE OR CITE
-------
Therefore, there are clearly demonstrable effects of low-level CO exposure in patients with
ischemic heart disease. The decrements in performance that have been described at the lowest
levels (<3% COHb) are probably in the range of reproducibility of the test and would not be
alarming to most physicians. The adverse health consequences of these types of effects,
5 however, are very difficult to project.
Effects in Individuals with Chronic Obstructive lung Disease
Aronow et al. (1977) studied the effects of a one-hour exposure to 100 ppm CO on
exercise performance in 10 men, aged 53 to 67 years, with chronic obstructive lung disease.
10 The resting mean COHb levels increased from baseline levels of 1.4% to 4.1% after
breathing CO. The mean exercise time until marked dyspnea significantly decreased (33%)
from 218 s in the air-control period to 147 s after breathing CO. The authors speculated that
the reduction in exercise performance was due to a cardiovascular limitation rather than
respiratory impairment.
15 Only one other study in the literature by Calverley et al. (1981) looked at the effects of
CO on exercise performance in older subjects with chronic lung disease. They evaluated
15 patients with severe reversible airway obstruction due to chronic bronchitis and
emphysema. Six of the patients were current smokers but they were asked to stop smoking
for 12 h before each study. The distance walked within 12 min was measured before and
20 after each subject breathed 0.02% CO in air from a mouthpiece for 20 to 30 min until COHb
levels were 8 to 12% above their initial levels. A significant decrease in walking distance
was reported when the mean COHb concentration reached 12.3%, a level much higher than
most of those reported in the studies on healthy subjects.
Thus, while it is possible that individuals with hypoxia due to chronic lung diseases such
25 as bronchitis and emphysema may be susceptible to CO during submaximal exercise typically
found during normal daily exercise, these effects have not been studied adequately at relevant
COHb concentrations of <5%.
Effects in Individuals with Chronic Anemia
30 An additional study by Aronow et al. (1984) on the effect of CO on exercise
performance in anemic subjects found a highly significant decrease in work time (16%)
March 12, 1990 10-34 DRAFT-DO NOT QUOTE OR CITE
-------
induced by a 1.24% increase in COHb. The magnitude of change seems to be very unlikely,
however, even considering the report by Ekblom and Huot (1972). The study was double-
blind and randomized, but with only 10 subjects. The exercise tests were done on a bicycle
in the upright position with an increase in workload of 25 watt every 3 min. However, no
5 measure of maximal performance such as blood lactate was used. The mean maximal heart
rate was only 139 to 146 beats/min compared to a predicted maximal heart rate of
170 beats/min for the mean age of the subjects. A subject repeating a test within the same
day, which was the case in the Aronow et al. (1984) study, often will remember the time and
work load and try to do the same in the second test. Normally, however, some subjects will
10 increase while others will decrease the time. This situation was apparent on the air-control
day, with an increase demonstrated in 6 out of 10 subjects, despite the high reproducibility
for such a soft, subjective endpoint. Also, comparing the control tests on the air day with the
CO day, 7 out of 10 subjects increased their work time. After CO exposure, however, every
subject decreased their time between 29 and 65 s. These data appear to be implausible given
15 the soft endpoint used, when two to three of the subjects would be expected to increase their
time even if there was a true effect of CO.
10.3.2.2 Arrhythmogenic Effects
The literature until recent years has been confusing with regard to potential
20 arrhythmogenic effects of CO.
Davies and Smith (1980) studied the effects of moderate CO exposure on healthy
individuals. Six matched groups of human subjects lived in a closed, environmental-exposure
chamber for 18 days and were exposed to varying levels of CO. Standard 12-lead
electrocardiograms were recorded during five control, eight exposure, and five recovery days,
25 respectively. P-wave changes of at least 0.1 mV were seen in the electrocardiograms during
the CO exposure period in 3 of 15 subjects at 2.4% COHb and in 6 of 15 at 7.1 % COHB
compared to none of 14 at 0.5% COHb. The authors felt that CO had a specific toxic effect
on the myocardium in addition to producing a generalized decrease in O2 transport to tissue.
Several methodological problems create difficulties of interpretation for this study. The
30 study design did not use each subject as his own control. Thus, only one exposure was
conducted out for each subject. Half of the subjects were tobacco smokers who were required
March 12, 1990 10-35 DRAFT-DO NOT QUOTE OR CITE
-------
to stop and certainly some of the ECG changes could have been due to the effects of nicotine
withdrawal. Although the subjects were deemed to be normal, no screening stress tests were
performed to uncover latent ischemic heart disease or propensity to arrhythmia. Most
importantly, no sustained arrhythmias or measurable effects on the conduction system were
5 noted by the authors. If p-wave changes of clinical significance are representative of a toxic
effect of CO on the atrium, then an effect on conduction of arrhythmias should be
demonstrated.
Knelson (1972) reported that 7 of 26 individuals, aged 41 to 60 years, had abnormal
electrocardiograms after exposure to 100 ppm CO for four hours (COHb levels of 5 to 9%).
10 Two of them developed arrhythmias. No further details were given regarding specifics of
these abnormalities. Among 12 younger subjects aged 25 to 36 years, all electrocardiograms
were normal.
Hinderliter et al. (1989) reported on effects of low-level CO exposure on resting and
exercise-induced ventricular arrhythmias in patients with CAD and no baseline ectopy. They
15 studied 10 patients with ischemic heart disease and no ectopy according to baseline
monitoring. After an initial training session, patients were exposed to air, 100 ppm CO, or
200 ppm CO on successive days in a randomized, double-blinded crossover fashion. Venous
COHb levels after exposure to 100 and 200 ppm CO averaged 4 and 6%, respectively.
Symptom-limited supine exercise was performed after exposure. Eight of the ten patients had
20 evidence of exercise-induced ischemia - either angina, ST segment depression, or abnormal
left ventricular ejection fraction response - during one or more exposure days. Ambulatory
electrocardiograms were obtained for each day and analyzed for arrhythmia frequency and
severity. On air- and CO-exposure days, each patient had only zero to one ventricular
premature beat per hour in the two hours prior to exposure, during the exposure period,
25 during the subsequent exercise test, and in the five hours following exercise. The authors
concluded that low-level CO exposure is not arrhythmogenic in patients with CAD and no
ventricular ectopy at baseline.
When patients with other levels of ventricular ectopy were studied (Sheps et al., 1989),
there was an increase in exercise-related arrhythmia (both simple and complex). Although no
30 definite evidence exists to date relating effects of CO exposure and lethal arrhythmias, the
recent epidemiologic study of Stern and colleagues (Stern et al., 1988) indicates that an excess
March 12, 1990 10-36 DRAFT-DO NOT QUOTE OR CITE
-------
of cardiovascular mortality in tunnel workers could be due to exposure to high levels of CO
(see Section 10.3.3.1). Their findings that risk decreased after job cessation and that risk was
not related to length of exposure suggest an acute effect of CO exposure maybe the causative
factor (perhaps due to arrhythmia production). These findings are consistent with the general
5 lack of effect of CO exposure on the development or progression of atherosclerosis.
10.3.2.3 Effects on Coronary Blood Flow
The effects of breathing CO on myocardial function in patients with and without
coronary heart disease have been examined by Ayres et al. (1969; 1970). Acute elevation of
10 COHb from 0.98 to 8.96% by a bolus exposure using either 1000 ppm CO for 8 to 15 min or
5000 ppm for 30 to 45 s caused a 20% average decrease in coronary sinus O2 tension without
a concomitant increase in coronary blood flow in the patients with coronary artery disease.
Observations in patients with coronary disease revealed that acute elevation of COHb to
approximately 9% decreased the extraction of O2 by the myocardium. However, overall
15 myocardial O2 consumption did not change significantly because an increase in coronary blood
flow served as a mechanism to compensate for a lower overall myocardial O2 extraction. In
contrast, patients with noncoronary disease increased their coronary blood flow with an
insignificant decrease in coronary sinus O2 tension as a response to increased COHb. The
coronary patients also switched from lactate extraction to lactate production. Thus, a
20 potential threat exists for patients with coronary heart disease who inhale CO because of their
inability to increase coronary blood flow to compensate for the effects of increased COHb.
Although in this study the coronary sinus PO2 dropped only slightly (reflecting average
coronary venous O2 tension), it is certainly possible that, in areas beyond a significant
coronary arterial stenosis, tissue hypoxia might be precipitated by very low tissue PO2 values.
25 Tissue hypoxia might be further exacerbated by a coronary-steal phenomena whereby
increased overall coronary flow diverts flow from areas beyond a stenosis to other normal
areas. Therefore the substrate for the worsening of ischemia and consequent precipitation of
arrhythmias is present with CO exposure.
30
March 12, 1990 10-37 DRAFT-DO NOT QUOTE OR CITE
-------
10.3.3 Relationship between CO Exposure and Risk of Cardiovascular
Disease in Man
10.3.3.1 Risk of bchemic Heart Disease
Epidemiologic studies on the relation between CO exposure and ischemic heart disease
5 are scarce. Earlier epidemiological data were summarized by Kuller and Radford (1983).
They concluded that mortality and morbidity studies have been negative or equivocal in
relating CO levels to health effects, but studies in human subjects with compromised coronary
circulation support an effect of acute exposure to CO at blood levels equivalent to about
20 ppm over several hours. They calculate that based on health surveys, probably over
10 10 million subjects in the United States are exposed to potentially deleterious levels of CO
and that perhaps 1250 excess deaths related to low-dose environmental CO exposure occur
each year.
Stern et al. (1981) reported a study performed by the NIOSH. They investigated the
health effects of chronic exposure to low concentrations of CO by conducting a historical
15 prospective cohort study of mortality patterns among 1558 white, male motor vehicle
examiners in New Jersey. The examiners were exposed to 10 to 24 ppm. The COHb levels
were determined in 27 volunteers. The average COHb level before a work shift was 3.3%
and the post-shift level was 4.7% in the whole group and 2.1 and 3.7%, respectively, in
nonsmokers only. The death rates were compared to the rates in the U.S. population based
20 on vital statistics. The cohort demonstrated a slight overall increase in cardiovascular deaths
but a more pronounced excess was observed within the first 10 years following employment.
The study has several important limitations, however, including the use of historical controls.
A second limitation is the lack of knowledge about smoking habits. A third is that the
individuals' values of COHb were not known.
25 Stern et al. (1988) published coronary heart disease mortality data among bridge and
tunnel officers exposed to CO. They investigated the effect of occupational exposure to CO
on mortality from arteriosclerotic heart disease in a retrospective study of 5529 New York
City bridge and tunnel officers. There were 4317 bridge officers and 1212 tunnel officers.
Among former tunnel officers, the standardized mortality ratio was 1.35 (90% confidence
30 interval 1.09 to 1.68) compared to the New York City population. Using the proportional
hazards model, the authors compared the risk of mortality from arteriosclerotic heart disease
March 12, 1990 10-38 DRAFT-DO NOT QUOTE OR CITE
-------
among tunnel workers with that of the less-exposed bridge officers. They found an elevated
risk in the tunnel workers that declined within as little as five years after cessation of
exposure. The 24-h average CO level in the tunnel was around 50 ppm in 1961 and around
40 ppm in 1968. However, higher values were recorded during rush hours. In 1971 the
5 ventilation was further improved and the officers were allowed clean-air breaks. Although
the authors concluded that CO exposure may play an important role in the pathophysiology of
cardiovascular mortality, other factors must be taken into consideration. Mortality from
arteriosclerotic heart disease has a complex multifactor etiology. The presence of other risk
factors, such as cigarette smoke, hypertension, hyperlipidemia, family history of heart
10 disease, obesity, socioeconomic status, and sedentary living all can increase the risk of
developing coronary heart disease. In addition, detailed exposure monitoring was not done in
this study. The bridge and tunnel workers were not only exposed to CO but also were
exposed to other compounds emitted from motor vehicle exhaust and to the noise and stress of
their environment. These other factors could have contributed to the findings.
15 Intoxication with CO that induces COHb levels around 50 to 60% is often lethal.
However, even levels around 20% COHb have been associated with death, mainly coronary
events, in patients with severe coronary artery disease. Balraj (1984) reported on 38 cases of
individuals dying immediately or within a few days following exposure to 10 to 50% COHb,
usually nonlethal levels of CO. All of the subjects had coronary artery disease, and 29 of
20 them had severe cases. The author concluded that the CO exposure, between 10 to 30%
COHb in 24 cases, triggered the lethal event in subjects with a restricted coronary flow
reserve. Similar associations between CO exposure and death or myocardial infarction have
been reported by several other authors. Atkins and Baker (1985) reported two cases with 23
and 30% COHb; McMeekin and Finegan (1987) reported one case with 45% COHb; Minor
25 and Seidler (1986) reported one case with 19% COHb; and Ebisuno et al. (1986) reported
one case with 21% COHb.
Forycki et al. (1980) described electrocardiographic changes in 880 patients treated for
acute poisoning. Effects were observed in 279 cases, with the most marked changes in cases
with CO poisoning. In those, the most common change was a T-wave abnormality and in
30 six cases a pattern of acute myocardial infarction was present. Conduction disturbances also
were common in CO poisoning but arrhythmias were less common.
March 12, 1990 10-39 DRAFT-DO NOT QUOTE OR CITE
-------
The association between smoking and cardiovascular disease (CVD) is fully established.
Although little is known about the relative importance of CO compared to nicotine, most
researchers consider them to be equally important. The nicotine component clearly aggravates
the decrease in O2 capacity induced by CO through an increase in the O2 demand of the heart.
5 This is exemplified in the study by Deanfield et al. (1986) using positron emission
tomography. They found that smoking one cigarette induced perfusion abnormalities in six
out of eight patients with CAD and exercise-induced angina. However, the smoke-induced
ischemia was without angina or silent ischemia in all of the patients and without ST
depression in seven of the patients. This raises an important question regarding analyses of
10 the effect of CO. Ischemia is not always associated with angina and/or ST depression.
However, most of the reports used to develop the guidelines for CO exposure have used
angina and/or ST depression as a sign of ischemia. Only the two studies by Sheps et al.
(1987) and Adams et al. (1988) used additional techniques to diagnose ischemia (see
Section 10.3.2). Both used gated radionuclide angiography to measure changes in ejection
15 fraction and wall motion induced by exercise, allowing the detection for signs of ischemia in
the absence of angina and/or ST depression. Positron emission tomography is even more
sensitive, however. Future studies on the effects of CO in patients with CVD, therefore, will
need to include more sensitive measures of ischemia than angina and/or ST depression.
Passive smoking exposes an individual to all components in the cigarette smoke, but the
20 CO component dominates heavily because only 1 % or less of the nicotine is absorbed from
sidestream smoke compared to 100% in an active smoker (Wall et al., 1988; Jarvis, 1987).
Therefore, exposure to sidestream smoke will be the closest to pure CO exposure even if the
resultant levels of COHb are low (about 2 to 3%). Two recent studies report on the
relationship between passive smoking and risk of coronary heart disease (CHD). Svendsen
25 et al. (1987) report from the multiple risk factor intervention trial (MRFIT) study on
1245 married, never-smoking men and 286 men married to women who smoke. The relative
risk for exposure to sidestream smoke was 1.48 (p = 0.13, 95% confidence interval [CI] =
0.89 to 2.47) for nonfatal and fatal coronary events and 1.96 (p = 0.08, 95% CI = 0.93 to
4.11) for all causes of mortality. Even more significant results were reported by Helsing
30 et al. (1988) who studied 4162 men and 14,873 women that were nonsmokers, some of which
had been exposed to sidestream smoke. The relative risk for exposure to passive smoke was
March 12, 1990 10-40 DRAFT-DO NOT QUOTE OR CITE
-------
1.31 (95% CI = 1.1 to 1.6) in men and 1.24 (95% CI = 1.1 to 1.4) in women for
arteriosclerotic heart disease death. Even though it is impossible to rule out an effect of the
other component in sidestream smoke, the data suggest an increase in risk of CHD associated
with a prolonged exposure to low levels of CO. In the United States, a population study by
5 Cohen et al. (1969) suggested an association between atmospheric levels of CO and increased
mortality from myocardial infarction in Los Angeles, but potential confounders were not
effectively controlled. In contrast, a similar study in Baltimore (Kuller et al., 1975) showed
no association between ambient CO levels and myocardial infarction or sudden death. In a
Finnish study (Hernberg et al. 1976), the prevalence of angina among foundry workers
10 showed an exposure-response relationship with regard to CO exposure, but no such result was
found for ischemic EKG findings during exercise.
In a cross-sectional study of 625 smokers, age 30 to 69, Wald et al. (1984) reported that
the incidence of CVD was higher in subjects with COHb >5% compared to subjects below
3%, a relative risk of 21.2 (95% CI = 3.3 to 34.3). Even if all of the subjects were
15 smokers, the association between COHb and CVD might be due to the fact that percent
COHb is a measure of smoke exposure.
Low to intermediate levels of COHb might interfere with the early course of an acute
myocardial infarction. The increase in COHb can be due to recent smoking or environmental
exposure. Mall et al. (1985) reported on a prospective study in smoking and nonsmoking
20 patients with acute myocardial infarction who were separated by their baseline COHb levels.
A total of 66 patients were studied. Thirty-one patients presented with a COHb level of 1.5%
and 35 with a level of 4.5%. In the group with elevated COHb, more patients developed
transmural infarction, but the difference was not significant. Patients with transmural
infarction had higher maximum creatine phosphokinase values when COHb was over 2%.
25 During the first six hour after admission to the hospital, these patients needed an
antiarrhythmic treatment significantly more frequently. Differences in rhythm disorders were
still present at a time when nicotine, due to its short half-life, was already eliminated. The
authors concluded that moderately elevated levels of COHb may aggravate the course of an
acute myocardial infarction.
30
March 12, 1990 10-41 DRAFT-DO NOT QUOTE OR CITE
-------
10.3.3.2 Risk of Hypertension
Because short-term exposure to CO does not change the arterial blood pressure, it is
unlikely that a prolonged CO exposure will induce hypertension. However, in a study by
Ahmad and Ahmad (1980), a higher prevalence of hypertension in the population was
5 observed in a large city compared to a small village where COHb levels were 16.4 and 5.6%,
respectively. The main difference between the two locations was that greater exposures to
noise and CO in the large city are known to induce hypertension, which might explain the
findings. On the other hand, laboratory animal studies (see Section 10.3.4) indicate that
prolonged CO exposure induces cardiac hypertrophy. If a similar effect occurs in humans, it
10 may be significant because hypertension increases the risk of developing left ventricular
hypertrophy (Frohlich, 1987). Therefore, it is important to study further the association
between CO exposure and risk of hypertension, which should be one of the research tasks
identified for future investigation.
15 10.3.4 Studies in Laboratory Animals
10.3.4.1 Introduction
The mechanisms by which CO exerts its toxic effects are detailed in Chapter 9. In
brief, CO combines with blood Hb to form COHb; this decreases the O2-carrying capacity of
the blood and shifts the O2Hb dissociation curve to the left. The cardiovascular system is
20 sensitive to alterations in O2 supply, and because inhaled CO limits O2 supply, it might be
expected to adversely affect the cardiovascular system; the degree of hypoxia and the extent
of tissue injury will be determined by the dose of CO. The effect of CO on the
cardiovascular system has been the subject of several recent reviews (Turino, 1981; McGrath,
1982; Penney, 1988). This section will discuss studies in animals that have evaluated the
25 effects of CO on ventricular fibrillation, hemodynamics, cardiomegaly, hematology, and
atherosclerosis. In this review, CO concentrations, times of exposure, and COHb levels are
provided whenever they were mentioned in the original manuscript. An attempt has been
made to focus, where possible, on those studies that have used the most relevant concentra-
tions of CO. For a more detailed treatment of the effects of higher concentrations of CO the
30 reader is referred to the review by Penney (1988).
March 12, 1990 10-42 DRAFT-DO NOT QUOTE OR CITE
-------
10.3.4.2 Ventricular Fibrillation Studies
Data obtained from animal studies suggest that CO can disturb cardiac conduction and
cause cardiac arrhythmias (see Table 10-4). In dogs exposed for six weeks in environmental
chambers to CO (50 and 100 ppm; COHb = 2.6 to 12.0%) intermittently and continuously,
5 Preziosi et al. (1970) reported abnormal electrocardiograms; the changes appeared during the
second week and continued throughout the exposure. The blood cytology, hemoglobin, and
hematocrit values were unchanged from control values. DeBias et al. (1973) studied the
effects of breathing CO (96 to 102 ppm; COHb = 12.4%) continuously (23 h/day; for
24 weeks) on the electrocardiograms of healthy monkeys and monkeys with myocardial
10 infarcts induced by injecting microspheres into the coronary circulation. The authors
observed higher P-wave amplitudes in both the infarcted and non-infarcted monkeys and a
higher incidence of T-wave inversion in the infarcted monkeys. The authors concluded there
was a greater degree of ischemia in the infarcted animals breathing CO. Although there was
a greater incidence of T-wave inversion in the infarcted monkeys the effects were transient
15 and of such low magnitude that accurate measurements of amplitude were not possible.
In other long-term studies, however, several groups have reported no effects of CO
either on the EKG or on cardiac arrhythmias. Musselman et al. (1959) observed no changes
in the EKG of dogs exposed continuously to CO (50 ppm COHb = 7.3%) for three months.
These observations were confirmed by Malinow et al. (1976) who reported no effects on the
20 EKG in cynomolgus monkeys exposed to CO (500 ppm-pulsed; COHb = 21.6%) for 14 mo.
Several research groups have investigated the effects of CO on the vulnerability of the
heart to induced ventricular fibrillation. DeBias et al. (1976) reported that CO (100 ppm
inhaled for 16 h; COHb = 9.3%) reduced the threshold for ventricular fibrillation induced by
an electrical stimulus applied to the myocardium of monkeys during the final stage of
25 ventricular repolarization. The voltage required to induce fibrillation was highest in normal
animals breathing air and lowest in infarcted animals breathing CO. Infarction alone and CO
alone each required significantly less voltage for fibrillation; when the two were combined the
effects on the myocardium were additive. These observations were confirmed in both
anesthetized, open-chested dogs with acute myocardial injury (Aronow et al., 1978) and in
30 normal dogs (Aronow et al., 1979) breathing CO (100 ppm; COHb = 6.3 to 6.5%) for
two hours. However, Kaul et al. (1974) reported that anesthetized dogs inhaling 500 ppm
March 12, 1990 10-43 DRAFT-DO NOT QUOTE OR CITE
-------
S TABLE 1O4. VENTRICULAR FIBRILLATION AND HEMODYNAMIC
1 STUDIES IN LABORATORY ANIMALS
sy
Exposurea>b COHb0 Animal Dependent Variable4 Results
VO CO = 50 ppm - Dog (n = 4) EKG, heart rate No effects
® continuously for 3 mo Rabbit (n = 4)
Rat (n = 100)
Comments Reference
Musselman et al. (1959)
CO = 50-100 ppm for
6 weeks intermittently or
continuously
2.6-12%
CO = 100 ppm for 24 wk, 12.4%;
23 h/day; CO = 9.3%
100 ppm for 6 h
CO = 500 ppm, pulsed 21.6%
12 h/day for 14 mo
C0= 100 ppm, 2 h 6.3%
Coronary artery
ligated; normal
CO = 5,000 ppm 4.9-17.0%
5 sequential exposures
to produce desired
COHb; Coronary artery
ligated
CO = 100 ppm 6.8-14.6%
Coronary artery
occluded briefly
Dog (n = 28)
EKG and pathology
Cynomolgus monkey EKG and susceptibility
(n = 52; 20) to induced fibrillation
Cynomolgus monkey EKG, arterial pressure,
(n = 26) left ventricular
pressure, dP/dt, Vamli
Dog (n = 21, 20) Ventricular fibrilla-
tion threshold (VFT)
Dog(n= 11)
Dog (n = 14)
EKG, coronary blood
flow
Arrhythmia; conduction
slowing in ischemic
myocardium
Abnormal EKG, heart
dilation, myocardial
thinning, some subjects
showed scarring and
degeneration in heart
muscle
Abnormal EKG and
increased sensitivity
to fibrillation
voltage
No effects
Reduced VFT in normal
and ligated dogs
Elevated ST segment;
increased flow to
non-ischemic myocar-
dium
No changes
Preziosi et al. (1970)
Infarcted animals
showed greatest effect
of COHb on both
dependent variables
DeBias et al. (1973)
Subjects on normal and Malinow et al. (1976)
high cholesterol diets
Studies were conducted Aronow et al. (1978,
blind 1979)
CO can augment
ischemia in acute
myocardial infarction
Concluded CO is not
arrhythmogenic during
early minutes of
infarction
Becker and Haak (1979)
Foster (1981)
CO = 200 ppm for 60
and 90 min; Paced
hearts and introduced
premature stimulus
1.64-6.30%
Dog (n = ?)
Threshold for ventri-
cular arrhythmias and
refractory period
No effects
Hutcheon et al. (1983)
-------
w
EJ
2
<
s
HEMOD
§3
^ ^
IBRILLATION
ATORYANIM
£ o
«? CQ
i"
£• 2 "o
f- § a
O C
?
03
O
3
1
a.
T3 2
1 i
o a
1 °
C. (S
3 c
Arrhythmia woi
at 20% COHb i
9 animals
ta
CO
'S
>^
Ji
<
CK
II
c
I
tR
o
""
CO = ?; previous
myocardial infarction
and 2 min coronary
artery occlusion
C
0
1
No changes
.2
f*»
S
g" tt
ill
.« o 5
O > o
«-s §
pgl
s2|
l'|S
e e -S.J
^Ift
1-l-f
S-k'l:
'Exposure concentrati
bl ppm = 1.145 mg/:
"Measured blood carb
dSee glossary of term
March 12, 1990
10-45 DRAFT-DO NOT QUOTE OR CITE
-------
CO (COHb = 20 to 35%) for 90 min were resistant to direct electrocardiographic changes.
At 20% COHb there was evidence of enhanced sensitivity to digitalis-induced ventricular
tachycardia but there was no increase in vulnerability of the ventricles to
hydrocarbon/epinephrine or to digitalis-induced fibrillation following exposure to 35% COHb.
5 Several workers have investigated the effect of breathing CO shortly after cardiac injury
on the electrical activity of the heart. Becker and Haak (1979) evaluated the effects of CO
(five sequential exposures to 5,000 ppm, producing COHb = 4.9 to 17.0%) on the
electrocardiograms of anesthetized dogs one hour after coronary artery ligation. Myocardial
ischemia, as judged by the amount of ST-segment elevation in epicardial electrocardiograms,
10 increased significantly at the lowest COHb levels (4.9%) and increased further with increasing
CO exposure; there were no changes in heart rate, blood pressure, left atrial pressure, cardiac
output, or blood flow to the ischemic myocardium. Similar results were noted by Sekiya
et al. (1983) who investigated the influence of CO (3,000 ppm for 15 min followed by
130 ppm CO for one hour, COHb = 13 to 15%) on the extent and severity of myocardial
15 ischemia in dogs. This dose of CO inhaled prior to coronary artery ligation increased the
severity and extent of ischemic injury, and the magnitude of ST segment elevation, more than
did ligation alone. There were no changes in heart rate or arterial pressure. Vanoli et al.
(1986) reported that CO (COHb = 10 and 20%) worsened the arrhythmia, following a
two minute coronary artery occlusion, in two of nine dogs with one-month-old anterior
20 myocardial infarcts. They were unable, however, to reproduce their results. The authors
conclude that while acute exposure to CO is unlikely to produce detrimental effects in post-
Mi dogs, the CO-dependent tachycardia may enhance the risks and consequences of any
ischemic episode.
On the other hand, several groups have reported no effects of CO on the EKG or on
25 cardiac arrhythmias. Musselman et al. (1959) observed no changes in the EKG of dogs
exposed continuously to CO (500 ppm) for three months. Their observations were confirmed
by Malinow et al. (1976) who reported no effects on the EKG in cynomolgus monkeys
exposed to CO for 14 mo (500 ppm-pulsed; COHb = 21.6%). Foster (1981) concludes that
CO (100 ppm CO for six to nine minutes; COHb = 10.4%) is not arrhythmogenic in dogs
30 during the early minutes of acute myocardial infarction following occlusion of the left anterior
descending coronary artery. This level of CO did not effect either slowing of conduction
March 12, 1990 10-46 DRAFT-DO NOT QUOTE OR CITE
-------
through the ischemic myocardium or the incidence of spontaneous ventricular tachycardia.
These results were confirmed by Hutcheon et al. (1983) in their investigation of the effects of
CO on the electrical threshold for ventricular arrhythmias and the effective refractory period
of the heart. They conclude that CO (200 ppm CO for 60 and 90 min; COHb = 5.1 to
5 6.3%) does not alter the effective refractory period or the electrical threshold for ventricular
arrhythmias in dogs. These results are consistent with those of Mills et al. (1987), who
studied the effects of 0 to 20% COHb on the electrical stability of the heart in chloralose-
anesthetized dogs during coronary occulsion. There were no major effects on heart rate,
mean arterial blood pressure, effective refractory period, vulnerable period, or ventricular
10 fibrillation threshold.
The effects of CO (1500 ppm, COHb = 15%) during acute myocardial ischemia on
arrhythmias in dogs with a healed anterior myocardial infarction and at low or high risk for
ventricular fibrillation were investigated by Vanoli et al. (1989). Following a two minute
coronary artery occlusion, malignant arrhythmias occurred in two dogs at low risk but in none
15 of the dogs at high risk for ventricular fibrillation. The authors conclude that, in dogs at high
risk for ventricular fibrillation, arrhythmogenic effects seldom can be expected from acute
exposure to CO.
On balance the results from animal studies suggest that mhaled CO can cause
disturbances in cardiac rhythm in both healthy and compromised hearts. Depending on the
20 exposure regime and species tested, the threshold for this response varies between 50 and
100 ppm CO (COHb = 2.6 to 12%) in dogs and 100 ppm (COHb = 12.4%) in monkeys
inhaling CO for 6 to 24 weeks; and 500 ppm CO (COHb = 4.9 to 17.0%) in dogs and
100 ppm (COHb = 9.3%) in monkeys inhaling CO for 0.6 to 16 h.
25 10.3.4.3 Hemodynamic Studies
The effects of CO on coronary flow, heart rate, blood pressure, cardiac output,
myocardial O2 consumption, and blood flow to various organs have been investigated in
laboratory animals. The results are somewhat contradictory (partly because exposure regimes
differed); however, most workers agree that CO in sufficiently high doses can affect many
30 hemodynamic variables.
March 12, 1990 10-47 DRAFT-DO NOT QUOTE OR CITE
-------
Adams et al. (1973) described increased coronary flow and heart rate and decreased
myocardial O2 consumption in anesthetized dogs breathing 1500 ppm CO for 30 min
(COHb = 23.1%). The decreased O2 consumption indicates that the coronary flow response
was not great enough to compensate for the decreased O2 availability. The authors noted that
5 although there was a positive chronotropic response, there was no positive inotropic response.
The authors speculated that (1) the CO may have caused an increase in the endogenous
rhythm or blocked the positive inotropic response or (2) the response to CO was mediated
reflexly through the cardiac afferent receptors to give a chronotropic response without the
concomitant inotropic response. When they used /3-adrenergic blocking agents, the heart-rate
10 response to CO disappeared, suggesting possible reflex mediation by the sympathetic nervous
system.
In a later study in chronically-instrumented, awake dogs exposed to 1000 ppm CO
producing COHb levels of 30%, Young and Stone (1976) reported an increase in coronary
flow with no change in myocardial O2 consumption. The increased coronary flow occurred in
15 animals with hearts paced at 150 beats/min, as well as in nonpaced animals, and in animals
with propranolol and atropine blockade. Because the changes in coronary flow with arterial
O2 saturation were similar whether the animals were paced or not, these workers conclude that
the increase in coronary flow is independent of changes in heart rate. Furthermore, the
authors reasoned that if the coronary vasodilation was caused entirely by the release of a
20 metabolic vasodilator, associated with decreased arterial O2 saturation, the change in coronary
flow in animals with both /J-adrenergic and parasympathetic blockade should be the same as in
control dogs. Young and Stone conclude that coronary vasodilation observed with an arterial
O2 saturation reduced by CO is mediated partially through an active neurogenic process.
Increased myocardial blood flow after CO inhalation in dogs was confirmed by Einzig
25 et al. (1980), who also demonstrated the regional nature of the blood flow response. Using
labeled microspheres, these workers demonstrated that whereas both right and left ventricular
beds were dilated maximally at COHb levels of 41% (CO = 15,000 to 20,000 ppm for
10 min), subendocardial/subepicardial blood flow ratios were reduced. The authors conclude
that in addition to the global hypoxia associated with CO poisoning, there is also an
30 underperfusion of the subendocardial layer, which is most pronounced in the left ventricle.
March 12, 1990 10-48 DRAFT-DO NOT QUOTE OR CITE
-------
These results were confirmed by Kleinert et al. (1980) who investigated the effects of
lowering O2 content by about 30% with low O2 or CO gas mixtures (CO = 10,000 ppm for
three minutes, COHb = 21 to 28%) on regional myocardial relative tissue PO2, perfusion and
small vessel blood content. In anesthetized, thoracotomized rabbits, both hypoxic conditions
5 increased regional blood flow to the myocardium, but to a lesser extent in the endocardium.
Relative endocardial PO2 fell more markedly than epicardial PO2 in both conditions. Small
vessel blood content increased more with CO than with low PO2, whereas regional 02
consumption increased under both conditions. The authors conclude that the response to
lowered O2 content (whether by inhaling low O2 or CO gas mixtures) is increased flow,
10 metabolic rate, and the number of open capillaries, and the effects of both types of hypoxia
appear more severe in the endocardium.
A decrease in tissue PO2 with CO exposure also has been reported by Weiss and Cohen
(1974). These workers exposed anesthetized rats to 80 and 160 ppm CO for 20-min periods
and measured tissue O2 tension as well as heart rate. A statistically significant decrease in
15 brain PO2 occurred with inhalation of 160 ppm CO, but there was no change in heart rate.
Horvath (1975) investigated the coronary flow response in dogs with COHb levels of
6.2 to 35.6% produced by continuous administration of precisely measured volumes of CO.
Coronary flow increased progressively as blood COHb increased and was maintained for the
duration of the experiment. However, when animals with complete atrioventricular block
20 were maintained by cardiac pacemakers and exposed to COHb levels of 6 to 7%, there was
no longer an increase in coronary blood flow. These results are provocative, because they
suggest an increased danger from low COHb levels in cardiac-disabled individuals.
Patajan et al. (1976) exposed unanesthetized rats to 1500 ppm CO for 80 min to achieve
COHb levels of 60 to 70%. After a slight transient increase, heart rate as well as blood
25 pressure decreased throughout the exposure. The authors interpret their data as indicating that
the lowering of blood pressure was more important than the degree of hypoxia to the
neurological impairment seen in their studies.
The effects of CO hypoxia and hypoxic hypoxia on arterial blood pressure and other
vascular parameters also were studied in carotid baroreceptor and chemoreceptor-denervated
30 dogs (Traystman and Fitzgerald, 1977). Arterial blood pressure was unchanged by CO
hypoxia but increased with hypoxic hypoxia. Similar results were seen in carotid
March 12, 1990 10-49 DRAFT-DO NOT QUOTE OR CITE
-------
baroreceptor-denervated animals with intact chemoreceptors. Following carotid
chemodenervation, arterial blood pressure decreased equally with both types of hypoxia.
In a subsequent report from the same laboratory (Sylvester et al., 1979), the effects of
CO hypoxia (CO = 10,000 ppm followed by 1000 ppm for 15 to 20 min; COHb = 61 to
5 67%) and hypoxic hypoxia, were compared in anesthetized, paralyzed dogs. Cardiac output
and stroke volume increased during both CO and hypoxic hypoxia whereas heart rate was
variable. Mean arterial pressure decreased during CO hypoxia, but increased during hypoxic
hypoxia. Total peripheral resistance fell during both hypoxias, but the decrease was greater
during the CO hypoxia. After resection of the carotid body, the circulatory effects of hypoxic
10 and CO hypoxia were the same and were characterized by decreases in mean arterial pressure
and total peripheral resistance. In a second series of closed-chest dogs, hypoxic and CO
hypoxia caused equal catecholamine secretion before carotid body resection. After carotid
body resection, the magnitude of the catecholamine response was doubled with both hypoxias.
These workers conclude that the responses to hypoxic and CO hypoxia are different and that
15 the difference is dependent on intact chemo- and baroreflexes and on differences in arterial O2
tension, but not on differences in catecholamine secretion or ventilatory response.
In cynomolgus monkeys exposed to 500 ppm intermittently for 12 h a day for 14 mo
(COHb = 21.6%), Malinow et al. (1976) reported no changes in arterial pressure, left
ventricular pressure, dp/dt, and Vmax. On the other hand, Kanten et al. (1983) studied the
20 effects of CO (150 ppm, COHb up to 16%) for 0.5-2 h on hemodynamic parameters in
open-chest, anesthetized rats, and reported that heart rate, cardiac output, cardiac index,
dF/dtmax (aortic), and stroke volume increased significantly, whereas mean arterial pressure,
total peripheral resistance, and left ventricular systolic pressure decreased. These effects were
evident at COHb levels as low as 7.5% (0.05 h). There were no changes in stroke work, left
25 ventricular dp/dtmax, and stroke power.
The effects of CO on blood flow to various vascular beds has been investigated in
several animal models, and most of the studies have been conducted at rather high CO or
COHb levels. In general, CO increases cerebral blood flow. However, the effects of CO on
the cerebral circulation are discussed in detail in Section 10.4.1.
30 In recent studies Oremus et al. (1988) reported that in the anesthetized rat breathing CO
(500 ppm; COHb = 23%) for one hour that CO reduces mean arterial pressure through
March 12, 1990 10-50 DRAFT-DO NOT QUOTE OR CITE
-------
peripheral vasodilation predominantly in the skeletal muscle vasculature. There were no
differences in heart rate or mesenteric or renal resistances between the CO-exposed and
control groups. This was confirmed by Gannon et al. (1988) who reported that in the
anesthetized rat breathing CO (500 ppm; COHb = 24%) for one hour that CO increased
5 inside vessel diameter (36 to 40%), increased flow rate (38 to 54%), and decreased mean
arterial pressure to 79% of control in the cremaster muscle. There was no change in the
response of 3A vessels to topical applications of phenylephrine as a result of CO exposure.
King et al. (1984, 1985) compared whole-body and hindlimb blood flow responses in
anesthetized dogs exposed to CO or anemic hypoxia. Arterial O2 content was reduced by
10 moderate (50%) or severe (65%) CO-hypoxia (produced by dialysis with 100% CO) or
anemic hypoxia (produced by hemodilution). These workers noted that cardiac output was
elevated in all groups at 30 min and in the severe CO group at 60 min. Hindlimb blood flow
remained unchanged during CO hypoxia in the animals with intact hindlimb innervation but
was greater in animals with denervated hindlimbs. There was a decrease in mean arterial
15 pressure in all groups associated with a fall in total peripheral resistance. Hindlimb resistance
remained unchanged during moderate CO hypoxia in the intact groups but was increased in
the denervated group. The authors concluded that the increase in cardiac output during CO
was directed to nonmuscle areas of the body and that intact sympathetic innervation was
required to achieve this redistribution. However, aortic chemoreceptor input was not
20 necessary for the increase in cardiac output during severe CO hypoxia, nor for the diversion
of the increased flow to nonmuscle tissues.
King et al. (1987) investigated the effects of severe CO (1000 to 10,000 ppm to lower
arterial O2 content to 5 to 6 vol) and hypoxic hypoxia on the contracting gastrocnemius
muscle of anesthetized dogs. Oxygen uptake decreased from the normoxic level in the CO
25 group but not in the hypoxic hypoxia group. Blood flow increased in both groups during
hypoxia but more so in the CO group. Oxygen extraction increased further during
contractions in the hypoxic group but fell in the CO group. The authors observed that the O2
uptake limitation occurring during CO hypoxia and isometric contractions was associated with
a reduced O2 extraction and concluded that the leftward shift in the O2Hb dissociation curve
30 during CO hypoxia may have impeded O2 extraction.
March 12, 1990 10-51 DRAFT-DO NOT QUOTE OR CITE
-------
Melinyshyn et al. (1988) investigated the role of /J-adrenoreceptors in the circulatory
responses to severe CO (about a 63% decrease in arterial O2 content obtained by dialyzing
with 100% CO) of anesthetized dogs. One group was 0 blocked with propanolol (0, and ft
blockade), a second with ICI 118,551 (ft blockade), and a third was a time control. Cardiac
5 output increased in all groups during CO hypoxia with the increase being greatest in the
unblockaded group. Hindlimb blood flow rose during CO hypoxia only in the unblockaded
group. The authors conclude that 35% of the rise in cardiac output occurring during CO
hypoxia (COH) depended on peripheral vasodilation mediated through ft-adrenoreceptors.
Thus, the results from animal studies indicate that inhaled CO can adversely affect
10 several hemodynamic parameters. The threshold for these effects may be near 150 ppm CO
(COHb = 7.5%).
10.3.4.4 Cardiomegaly
The early investigations of cardiac enlargement following prolonged exposure to CO
15 have been confirmed in different animal models and extended to characterize the development
and regression of the cardiomegaly (see Table 10-5). Theodore et al. (1971) reported cardiac
hypertrophy in rats breathing 500 ppm CO (COHb = 32 to 38%) for 168 days, but not in
dogs, baboons, or monkeys. Penney et al. (1974a) also noted cardiomegaly in rats breathing
500 ppm CO; heart weights were one-third greater than predicted within 14 days of exposure,
20 and 140 to 153% of controls after 42 days of exposure. The cardiomegaly was accompanied
by changes in cardiac lactate dehydrogenase (LDH) isoenzyme composition that were similar
to those reported in other conditions that cause cardiac hypertrophy (e.g., aortic and
pulmonary artery constriction, coronary artery disease, altitude acclimation, severe anemia).
To further characterize the hypertrophy and determine its threshold, Penney et al.
25 (1974b) measured heart weights in rats exposed continuously to 100, 200, and 500 ppm CO
(COHb = 9.26, 15.82, and 41.14%), for various times (20 to 46 days); they noted
significant increases in heart weights at 200 and 500 ppm CO, with changes occurring in the
left ventricle and septum, right ventricle, and atria especially. The authors concluded that
whereas the threshold for the Hb response is 100 ppm CO (COHb = 9.26%), the threshold
March 12, 1990 10-52 DRAFT-DO NOT QUOTE OR CITE
-------
I
TABLE 10-5. CARDIAC HYPERTROPHY STUDIES IN LABORATORY ANIMALS
Exposure"'1"
400-500 ppm for
168 days
100 ppm, 46 days
200 ppm, 30 days
500 ppm, 20-42 days
CO = 60, 125, 250,
500 ppm for 21 days
COHb°
32-38%
(dogs and
monkeys
only)
9.2%
15.8%
41.12%
-
Animal
Monkey (n = 9)
Baboon (n = 3)
Dog (n = 16)
Rat (n = 136)
Mouse (n = 80)
Rat (n = 32)
Fetal rats
(n = 75)
Dependent Variable*
Cardiovascular damage
in rat heart
Heart size; LDH
Hb, Hot, HW
Results
No changes except slight
hypertrophy
Hypertrophy of both left
and right ventricles;
LDH increases
Hb and Hot depressed with
60 ppm and elevated by
Comments
Threshold for cardiac
enlargement near 200 ppm
HW increase probably not
due to increased viscosity
Reference
Theodore et al. (1971)
Penney et al. (1974a,b)
Prigge and Hochraincr (1977)
gestation
250 and 500 ppm; HW
increased at all concen-
trations
or pulmonary hypertension
o
o
c
I
8
n
CO = 400 ppm, or 35-58%
500 ppm increased
to 1,100 ppm
CO = 500 ppm 38-42%
until 50 days of age
CO = 500 ppm for 38-42%
1-42 days. Open-
chest, anesthetized
preparation.
CO = 150 ppm (15% in
throughout gestation adult rats)
Rat (n = 30)
Rat (n > 200)
5 and 25 days
old
Rat
(n = 25)
Rat
(n = 88)
Cardiomegaly (HW/BW)
and LDH
HW
Right ventricle (RV)
Left ventricle (LV)
Stoke index (SI),
Mean stroke power (SP),
Mean cardiac output (CO),
Systemic resistance (SR),
Pulmonary resistance (PR).
BW
Wet-Heart Weight (WHW)
HW/BW and %M LDH
subunits increased with
low and high CO; after
removal of CO HW/BW and
%M LDH remained high
HW, LV, and RV increased
response greater in
younger group
SI, SP, CO increased;
SR and PR decreased
BW depressed;
WHW increased
Potential for cardiac DNA
synthesis and hyperplasia
ends between 5-25 days
postnatal life
Concluded that increased
CO via increased stroke
volume is compensation for
CO intoxication; increased
work may cause cardiomegaly
Increased HW due
to increased water content
Styka and Penney (1978)
Penney and Weeks (1979)
Penney et al. (1979)
Fechter et al. (1980)
-------
vo
^o
O
TABLE 10-5 (cont'd). CARDIAC HYPERTROPHY STUDIES IN LABORATORY ANIMALS
Exposure"'11
CO = 500 ppm
for 32 days
(1982)
COHbc
38-42%
Animal
Rat
(n = 140)
Dependent Variable"*
Cardiomegaly
Results
HW/BW higher after
70 days of exposure
and after 30 days
Comments
Cannot be explained by
changes in DNA or
hydroxyproline
Reference
Penney et al. (1982)
of recovery; both RV
and LV were affected
CO = 157-200 ppm 21.8-33.5% Rat
last 17 days (n = 96)
gestation
RBC count, HW,
placenta! weight
(PW), cardiac
LDH M subunit,
Mb
Depressed RBC; HW and
PW increased; LDH(M)
increased; Mb increased
Cardiomegaly not due to
elevated water content
(Disagrees with Fechter
et al., 1980)
Penney et al. (1983)
CO = 500 ppm
for 38-47 days
38-40%
Rat (n = 25)
Cardiac compliance and
dimensions
No change in compliance;
LV length and outside
diameter increased
Chronic COHb produces
eccentric Cardiomegaly
with no change in wall
stiffness
Penney et aJ. (1984a)
CO = 200 ppm from
Day 7 of pregnancy
until parturition,
and for 28 days fol-
lowing parturition
Rat (n > 180) HW, RV, and LV weight
RV increased with CO
during fetal period,
HW and LV increased
with CO during
postnatal period
Hemodynamic load caused
by CO during fetal period
results in Cardiomegaly
due to myocyte hyperplasia
Clubb et al. (1986)
'Exposure concentration and duration.
bl ppm = 1.145 mg/m3; 1 mg/m3 = 0.873 ppm at 25 °C, 760 mm Hg; 1%
"Measured blood carboxyhemoglobin (COHb) levels.
dSee glossary of terms and symbols for abbreviations and acronyms.
10,000 ppm.
-------
for cardiac enlargement is near 200 ppm CO (COHb = 12.03%), and unlike cardiac
hypertrophy caused by altitude, which primarily involves the right ventricle, cardiac
hypertrophy caused by CO involves the whole heart.
The regression of cardiac hypertrophy in rats exposed continuously to moderate
5 (400 ppm; COHb = 35%) or severe (500-1100 ppm; COHb = 58%) CO for six weeks was
followed by Styka and Penney (1978). Heart weight to body weight ratio (HW/BW)
increased from 2.65 in controls, to 3.52 and 4.01 with moderate and severe CO exposure,
respectively. Myocardial LDH M subunits (M LDH) were elevated 5 to 6% by moderate and
12 to 14% by severe CO exposure. Forty-one to 48 days after terminating the CO exposure,
10 Hb concentrations among groups did not differ significantly; HW/BW ratios were similar in
the control and moderately exposed animals, but remained significantly elevated in the
severely exposed animals.
In addition to cardiomegaly, Kjeldsen et al. (1972) has reported ultrastructual changes in
the myocardium of rabbits breathing 180 ppm CO (COHb = 16.7%) for two weeks. The
15 changes included focal areas of necrosis of myofibrils and degenerative changes of the
mitochondria. In addition, varying degrees of injury were noted in the blood vessels. These
included edema in the capillaries; stasis and perivascular hemorrhages on the venous side; and
endothelial swelling, subendothelial edema, and degenerative changes in myocytes on the
arterial side.
20 The hemodynamic consequences of prolonged CO exposure have been examined in rats
breathing 500 ppm CO (COHb = 38 to 42%) for 1 to 42 days, (Penney et al., 1979) and in
goats breathing 160 to 220 ppm CO (COHb = 20%) for two weeks (James et al., 1979). In
rats, cardiomegaly developed and stroke index, stroke power, and cardiac index increased;
total systemic and pulmonary resistances decreased. Left and right ventricular systolic
25 pressures, mean aortic pressure, maximum left ventricular dp/dt, and heart rate did not
change significantly. Penney et al. concluded that enhanced cardiac output, via an increased
stroke volume, is a compensatory mechanism to provide tissue oxygenation during CO
intoxication and that increased cardiac work is the major factor responsible for the
development of cardiomegaly. In chronically-instrumented goats, James et al. noted that
30 cardiac index, stroke volume, left ventricular contractility, and heart rate were all unchanged
during exposure to CO, but were depressed significantly during the first week following
March 12, 1990 10-55 DRAFT-DO NOT QUOTE OR CITE
-------
termination of the exposure. Discrepancies between the Penney and James studies may be
caused by differences in the CO concentrations or in the species used.
Penney et al. (1984a) studied the compliance and measured the dimensions of
hypertrophied hearts from rats breathing 500 ppm CO (COHb = 38 to 40%) for 38 to
5 47 days. Heart weight to body weight ratios increased from 2.69 to 3.34. Although
compliance of the right and left ventricles was higher in the CO group, the differences
disappeared when the heart weight was normalized by body weight. Left ventricular apex-to-
base length and left ventricular outside diameter increased 6.4% and 7.3%, respectively; there
were no changes in left ventricle, right ventricle, or septum thickness. The authors conclude
10 that chronic CO exposure produces eccentric cardiomegaly with no intrinsic change in wall
stiffness.
The consequences of breathing CO also have been investigated in perinatal animals.
Prigge and Hochrainer (1977) reported elevated heart weights in fetuses from pregnant rats
exposed for 21 days to CO concentrations as low as 60 ppm. Because these animals
15 developed anemia rather than polycythemia, these workers discounted increased blood
viscosity as a cause of the cardiomegaly. Penney and Weeks (1979) examined the effects of
inhaling 500 ppm CO (COHb = 38 to 42%) until 50 days of age on cardiac growth in young
(5 days) and old (25 days) rats. They observed that the younger rats experienced the greatest
change in heart weight and DNA synthesis and concluded that the potential for cardiac DNA
20 synthesis and muscle cell hyperplasia ends in rats during the 5th through 25th days of
postnatal development.
Fechter et al. (1980) reported elevated wet-heart weights at birth in neonatal rats from
dams exposed throughout gestation to 150 ppm CO (COHb = 15%). There were no
differences in dry-heart weight, total protein, or RNA or DNA levels; the differences between
25 groups in wet-heart weight disappeared after four days. These workers concluded that the
increased heart weight seen at birth in the CO-exposed rats is caused by cardiac edema.
These results were not verified by Penney et al. (1983) in offspring of pregnant rats
exposed to 157, 166, and 200 ppm CO (COHb = 21.8 to 33.5%) for the last 17 of 22 days
gestation. These workers observed that wet- and dry-heart weights increase proportionately
30 and conclude that cardiomegaly, present at birth, is not due to elevated myocardial water
content. They also determined that cardiac LDH M subunit composition and Mb
March 12, 1990 10-56 DRAFT-DO NOT QUOTE OR CITE
-------
concentration were elevated at 200 ppm CO. They conclude that maternal CO inhalation
exerts significant effects on fetal body and placenta! weights, heart weight, enzyme
constituents, and composition. Moreover, in newborn rats inhaling 500 ppm CO (COHb =
38 to 42%) for 32 days and then developing in air, Penney et al. (1982) observed that
5 HW/BW ratio increased sharply after birth, peaked at 14 days of age, and then fell
progressively; it remained higher in rats exposed prenatally to CO than in control rats for up
to 107 days of age. The persistent cardiomegaly could not be explained by changes in DNA
or hydroxyproline.
Ventricular weights (wet and dry) and myocyte size and volume were measured in
10 perinatal rats exposed to 200 ppm CO by Clubb et al. (1986). Pregnant rats were exposed to
air or CO, and, at birth, pups from these two groups were subdivided into four groups:
(1) control group (air/air), maintained in air in utero and postpartum; (2) air/CO group,
received CO only postpartum; (3) CO/CO group; received CO in utero and postpartum; and
(4) CO/air group, received CO in utero, but in air postpartum. Right ventricle weights were
15 increased in animals exposed to CO during the fetal period, but left ventricular weights were
increased by CO during the neonatal period. Although HW/BW ratios increased to that of the
CO/CO group by 12 days of age in animals exposed to CO postnatally only (air/CO),
HW/BW ratios decreased to that of controls (air/air) by 28 days of age in animals exposed to
air postnatally following fetal CO exposure (CO/air). There was no difference in myocyte
20 volume between groups at birth. Left ventricle plus septum and right ventricle cell volumes
of the CO/CO group were smaller than the controls at 28 days of age despite the heavier wet
and dry weights of the CO/CO neonates. At birth, the CO-exposed animals had more
myocytes in the right ventricle than the air-exposed controls; CO exposure after birth resulted
in left ventricular hyperplasia.
25 Clubb et al. (1986) concluded that the increased hemodynamic load caused by CO
during the fetal period results in cardiomegaly, characterized by myocyte hyperplasia, and this
cellular response is sustained throughout the early neonatal period in animals exposed to CO
postpartum.
Thus, results from animal studies indicate that inhaled CO can cause cardiomegaly, and
30 that the threshold for this response is near 200 ppm (COHb = 12%) in adult rats, and
60 ppm in fetal rats.
March 12, 1990 10-57 DRAFT-DO NOT QUOTE OR CITE
-------
10.3.4.5 Hematology Studies
Increase in Hb concentration, as well as hematocrit ratio, is a well-documented response
to hypoxia, which serves to increase the O2 carrying-capacity of the blood. Guyton and
Richardson (1961) and Smith and Crowall (1967) however, suggest that changes in hematocrit
5 ratio not only affect the O2-carrying capacity of the blood, but blood flow as well. Therefore,
when hematocrit ratios increase much above normal, O2 delivery to the tissues may be
reduced because the resultant decrease in blood flow can more than offset the increased O2
carrying capacity of the blood. Smith and Crowall conclude that there is an optimum
hematocrit ratio at sea level that shifts to a higher value with altitude acclimation.
10 Presumably a similar compensation would occur when O2 transport is reduced by CO.
Changes in Hb concentration and hematocrit ratio have been reported in numerous
animal studies (see Table 10-6). In dogs exposed to 50 ppm CO (COHb = 7.3%) for
three months, Musselman et al. (1959) reported a slight increase in Hb concentration (12%),
hematocrit ratio (10%), and in red blood cells (10%). These observations were extended by
15 Jones et al. (1971) to include several species of animals exposed to 51 ppm or more CO
(COHb = 3.2 to 20.2%), intermittently or continuously, for up to 90 days. There were no
significant increases in the Hb and hematocrit values observed in any of the species at
51 ppm CO (COHb = 3.2 to 6.2%). At 96 ppm CO (COHb = 4.9 to 12.7%), significant
increases were noted in the hematocrit value for monkeys (from 43 to 47%) and in the Hb
20 (from 14.0 to 16.49%) and hematocrit values (from 46 to 52%) for rats. Hemoglobin and
hematocrit values were elevated in rats (14 and 10%, respectively) guinea pigs (8 and 10%,
respectively), and monkeys (34 and 26%, respectively) exposed to 200 ppm CO
(COHb = 9.4 to 12.0%); they also were elevated in dogs, but there were too few animals to
determine statistical significance. However, in dogs exposed to CO (195 ppm;
25 COHb = 30%) for 72 h, Syvertsen and Harris (1973) reported that hematocrit and Hb
increased from 50.3 to 57.8% and 15.0 to 16.2 g%, respectively. The differences in
hematocrit and Hb occurred after 72 h exposure and were attributed to increased
erythropoiesis. Penney et al. (1974b) observed significant increases in Hb (from 15.6 to
16.7 g%) in rats exposed to 100 ppm CO over several weeks and conclude that the threshold
30 for the Hb response is close to 100 ppm (COHb = 9.26%).
March 12, 1990 10-58 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 10-6. HEMATOLOGY STUDIES IN LABORATORY ANIMALS
CJ
1— »
ro
\f
VO
s
£
VO
O
»
Exposure*'1"
CO = 50 ppra
continuously for 3 mo
CO = 51,96, or 200 ppm
for 90 days
CO = 67.5 ppm
22 h/day, 7 day/
for 2 years
CO = 195 for 72 h
CO = 100 ppm, 46 days;
200 ppm, 30 days;
500 ppm, 20-42 days
CO = 200 ppm last
18 days gestation
COHbc
7.3%
3.296
1.8%
3.2-6.2%
4.9-12.7%
9.4-20.2%
depending
upon species
1.9-5.5%
and
2.8-10.2%
-30%
9.20%
15.82%
41.12%
27%
Animal Dependent Variable1*
Dog (n = 4) Hb, Hct, and RBC,
Rabbit (n = 40) EKG
Rat (n = 100)
Rat (n = 35) Hb
Guinea pig (n = 35)
Monkey (n = 9)
Dog (n = 6)
Cynomolgus Hct, Hb RBC counts
monkey (n = 27)
Dog (n = 12) Hct and Hb
Rat (n = 32) Hb
Rat Hb, Hct, and RBC
Results
Hb, Hct, RBC increased
in dogs and rabbit; no
change in EKG in dog
Increases in rats at 96,
106, and 200 ppm;
increases in all animals
at 200 ppm
No effects
Both increased
Increased at all levels
Hb, Hct, and RBC ail
lower
Comments
No toxic signs in dogs,
rabbits or rats
Unusual variation in
COHb
Increase due to
erythropoiesis
About 30 days until Hb
approached asymptotic
values
Reference
Musselman et at. (1959)
Jones et al. (1971)
Eckhardt et al. (1972)
Syvertsen and Harris (1973)
Penney et al. (1974b)
Penney et al. (1980)
50 and 100 ppm
for 6 weeks on various
2.6-12%
Dog (n = 46)
Hb
No effects
Preziosi et al. (1970)
o
o
§
3
o
c
o
a
2
so
n
intermittent daily
schedules
CO = 50 ppm, 95 h/week
whole natural life
expectancy up to
2 years (also short-term)
Rat (n = 336) EKG, organ weights, No effects Also showed no effects Stupfel and Bouley (1970)
Mouse (n = 767) Hb, Hct, and RBC on other variables
"Exposure concentration and duration.
bl ppm = 1.145 mg/m3; 1 mg/m3 =
°Measured blood carboxy hemoglobin
0.873 ppm at 25°C, 760 mm Hg; 1 % = 10,000 ppm.
(COHb) levels.
dSee glossary of terms and symbols for abbreviations and acronyms.
-------
Several groups have reported no change in Hb or hematocrit following CO exposure.
Thus, Preziosi et al. (1970) observed no significant change in Hb concentration in dogs
exposed to 50 and 100 ppm CO (COHb = 2.6 to 12.0%) for six weeks. In monkeys,
exposed to 20 and 65 ppm CO (COHb = 1.9 to 10.2%) for two years, Eckardt et al. (1972)
5 noted no compensatory increases in Hb concentration or hematocrit ratio. In mice exposed
5 days a week to 50 ppm CO for one to three months, Stupfel and Bouley (1970) observed no
signfiicant increase in Hb.
Interestingly, in fetuses removed from pregnant rats after 21 days exposure to CO,
Prigge and Hochrainer (1977) reported a significant increase in fetal hematocrit (from 33.3 to
10 34.5%) at 60 ppm and a significant decrease in Hb and hematocrit at 250 ppm (from 9.1 to
8.0 g% and from 33.3 to 28.4%, respectively) and 500 ppm (from 9.1 to 6.5 g% and from
33.3 to 21.9%, respectively). These results were confirmed by Penney et al. (1980) who
reported significantly lower Hb (12.6 vs. 15.8 g%), hematocrit (46.2 vs. 54.4%), and RBC
counts (27.2 vs. 29.1%) in newborns from pregnant rats exposed to 200 ppm CO (COHb =
15 27.8%) for the final 18 days of development than in controls. However, in a later study,
Penney et al. (1983) reported that although RBC count was depressed in neonates from
pregnant rats exposed to 157, 166, and 200 ppm CO (COHb = 21.8 to 33.5%) for the last
17 out of 22 days gestation, mean corpuscular Hb and volume were elevated.
The results from animal studies indicate inhaled CO can increase Hb concentration and
20 hematocrit ratio and that the threshold for this response, at least in rats, appears to be near
100 ppm (COHb = 9.26%). Small increases in Hb and hematocrit probably represent a
compensation for the reduction in O2-transport caused by CO. At higher CO concentrations,
excessive increases in Hb and hematocrit may impose an additional workload on the heart and
compromise blood flow to the tissue. The O2 transport system of the fetus is especially
25 sensitive to CO inhaled by the mother, and it may be affected by CO concentrations as low as
60 ppm.
10.3.4.6 Atherosclerosis and Thrombosis
The section dealing with cholesterol and atherosclerosis in the previous air quality
30 criteria document for CO (U.S. Environmental Protection Agency, 1979) described about
March 12, 1990 10-60 DRAFT-DO NOT QUOTE OR CITE
-------
12 publications. These studies generally utilized animal models of atherosclerosis or animal
models describing arterial wall cholesterol uptake in response to COHb concentrations ranging
from 4.5 to 41.1% (see Table 10-7). The conclusion was that the evidence failed to support
conclusively a relationship between CO exposure and atherosclerosis in animal models. Since
5 completion of the 1979 air quality criteria document, a number of additional studies have
been published (Table 10-7). However, taken in aggregate, the studies still fail to
conclusively prove an atherogenic effect of exposure to low doses of CO.
Astrup et al. (1967) described atheromatosis as well as increased cholesterol
accumulation in aortas of rabbits fed cholesterol and exposed to CO (170 to 350 ppm;
10 COHb = 17 to 33%) for 10 weeks. These observations were not verified, however, by
Webster et al. (1970) who observed no changes in the aorta or carotid arteries or in serum
cholesterol levels in squirrel monkeys fed cholesterol and exposed intermittently to CO (100
to 300 ppm; COHb = 9 to 26%) for seven months; they did note enhanced atherosclerosis in
the coronary arteries. Davies et al. (1976) confirmed that coronary artery atherosclerosis was
15 significantly higher in rabbits fed cholesterol and exposed intermittently to CO for 10 weeks
(250 ppm; COHb = 20%); but they also reported no significant differences between groups
in aortic concentrations of triglycerides, cholesterol, phospholipids, or plasma cholesterol.
In cynomolgus monkeys fed cholesterol and exposed intermittently to CO for 14 mo (50
to 500 ppm; COHb = 21.6%), Malinow et al. (1976) observed no differences in plasma
20 cholesterol levels or in coronary or aortic atherosclerosis. Armitage et al. (1976) confirmed
that intermittent CO (150 ppm; COHb = 10% for 52 and 84 weeks) did not enhance the
extent or severity of atherosclerosis in the normal White Carneau pigeon. While CO
exposure did increase the severity of coronary artery atherosclerosis in birds fed cholesterol;
the difference between groups, noted at 52 weeks, was not present after 84 weeks.
25 Stender et al. (1977) exposed rabbits that were fed high levels of cholestrol to CO for
six weeks continuously and intermittently (200 ppm; COHb = 17%). In the cholesterol-fed
group, CO had no effect on free- and esterified-cholesterol concentrations in the inner layer of
the aortic wall. In the normal group, CO increased the concentration of cholesterol in the
aortic arch, but there was no difference in the cholesterol content of the total aorta.
30 Hugod et al. (1978), using a blind technique and the same criteria to assess intimal
damage as was used in earlier studies (Kjeldsen et al., 1972; Thomsen, 1974; Thomsen and
March 12, 1990 10-61 DRAFT-DO NOT QUOTE OR CITE
-------
1
cr
to
TABLE 10-7. ATHEROSCLEROTIC STUDIES IN LABORATORY ANIMALS
Exposure"'1"
COHbc
Animal
Dependent Variable"1
Results
Comments
Reference
CO = 170 ppm for
8 week, then 350 ppm
for last 2 week, fed
cholesterol
17-33%
Rabbit (a = 24)
Atherosclerotic
changes
Increased aortic ather-
omatosis and cholesterol;
Local degenerative
signs and hemorrhages in
hearts
Not verified in subsequent
studies
Astrup et al. (1967)
CO = 100-300 ppm
4 h/day, 5 days/week
for 7 mo,
fed cholesterol
9-26%
Squirrel monkey
Atherosclerosis
in various blood
vessels plus serum
cholesterol
Increased coronary athero-
sclerosis but no effects
on aorta or carotid
arteries or serum
cholesterol
Webster et al. (1970)
CO = 250 ppm 20.6%
continuously for 2 week
CO = 50, 100, and 4.5%
180 ppm for periods 9.0%
ranging from 30 min -
24 h, and from
2-11 days
Cynomolgus monkey Coronary artery
(n = 20) pathology
Rabbit (n = 61) Aortic damage
Subcndothelial edema,
gaps between endothelial
cells, infiltration cells
containing lipid droplets
Increased aortic intimal
lesions at 180 ppm CO
for 4 h or more
Lipid-laden cell findings
suggest greater sensitivity
of monkeys than of rabbits
Postulates 180 ppm CO for
4 h is threshold for
injury
Thomsen (1974)
Thomsen and Kjeldsen (1975)
CO = 150 ppm 6 h/ 10%
day, 5 day/week for
52 and 84 weeks, fed
cholesterol
CO = 250 ppm 4 h/ 20%
day, 7 day/week,
10 week
White carncau Severity of
pigeon (n = 180) alherosclerosis
Rabbit (n = 24) Blood cholesterol;
coronary artery
atherosclerosis;
aortic cholesterol
content
No effect in normocholes-
terolemic birds; coronary
artery atherosclerosis
significantly enhanced in
hypercholesterolemic
birds at 52 weeks
Increased atherosclerosis
in coronary arteries but
no differences in aortic
or plasma cholesterol
No significant changes in
coronary arteries after
84 weeks
Study disagrees with Astrup
et al. (1967)
Armitage et al. (1976)
Davies et al. (1976)
-------
TABLE 10-7 (cont'd). ATHEROSCLEROTIC STUDIES IN LABORATORY ANIMALS
o\
Exposure"-1" COHb°
CO = 50-500 ppm, 21.6%
12 h/day for 14 mo
CO = 200 ppm, 17%
continuously or
12 h/day for 6 week
CO = 200 ppm for -
5-12 weeks, 2,000 ppm
for 320 min; 4,000 ppm
for 205 min
CO = 400 ppm for 23%
10 alternate half-
hours of each day
for 12 mo
Animal Dependent Variable'1
Cynomolgus monkey Aortic and coronary
(n = 26) atherosclerosis
Rabbit (n = 30) Cardiovascular
pathology
Rabbit (n = 150) Coronary artery
and aortic damage
Cynomolgus monkey Cholesterol content
(n = 11) of vessels and
plasma
Results
No effects
No differences in
atherosclerosis but CO
produced higher serum
cholesterol levels
No effect
No effect on plasma-free
cholesterol, cholesterol
ester, tri- and diglycerides,
and phospholipids; no
Comments
Subjects on high- and low-
cholesterol diets;
disagrees with Astrup
et al. (1967).
Serum cholesterol was con-
trolled by adjusting
individual diets; apparently
coronary atherosclerosis
in Astrup et al. (1967) was
caused by increased serum
cholesterol
Inability to reproduce
earlier results may be due
to lack of blind technique
and smaller number of
animals in earlier studies
Agrees with Malinow et al.
(1976)
Reference
Malinow et al. (1976)
Slender et al. (1977)
Hugod et al. (1978)
Bing et al. (1980)
Smoked 43 cigarettes
per day for 14-19
mo; fed choles-
terol
CO = 200-300 ppm
continuously for
0.6-1.9% Baboons (n = 36)
Serum cholesterol
Rabbits (n = 140)
Myocardial mor-
phology using
electron microscopy
significant increase in
cholesterol content of
aorta; no histologic
damage and no fat
deposition
No significant differences
in serum total cholesterol,
VLDL + LDL cholesterol,
HDL cholesterol, or
triglyceride concentrations
No histotoxic effects
Rogers et al. (1980)
Hugod (1981)
-------
i
H-*
to
TABLE 10-7 (cont'd). ATHEROSCLEROTIC STUDIES IN LABORATORY ANIMALS
Exposure*'11
Smoked 43 cigarettes
per day for up to
33 mo; fed
cholesterol
COHb0 Animal
0.64-2.0% Male baboons
(n = 36)
0.35-1.13% Female baboons
(n = 25)
Dependent Variable*1 Results Comments
Serum cholesterol No significant differences
in serum total cholesterol,
VLDL -f LDL cholesterol,
HDL cholesterol, or
triglyceride concentrations;
slightly enhanced plague
formation in carotid artery;
no difference in lesions or
vascular content of lipid
or prostaglandin in aorta
or coronary arteries
Reference
Rogers el al. (1988)
CO = 100 ppm
8 h/day, 5 day/week
for 4 mo,
fed cholesterol
6.8-7.6%
Pigs (n = 38)
(normal or
homozygous and
heterozygous
for von
Willebrand's
disease) with
balloon-catheter
injury of coro-
nary arteries
Coronary artery
and aortic lesions
No significant changes
Sultzer et al. (1982)
O
o
"Exposure concentration and duration.
bl ppm = 1.145 mg/m3; 1 mg/m3 = 0.873 ppm at 25°C, 760 mm Hg; 1% = 10,000 ppm.
"Measured blood carboxyhemoglobin (COHb) levels.
dSee glossary of terms and symbols for abbreviations and acronyms.
-------
Kjeldsen, 1975), noted no histologic changes in the coronary arteries or aorta in rabbits
exposed to CO (200, 2000, or 4000 ppm) for 0.5 to 12 weeks. These workers suggested that
the positive results obtained earlier were due to the non-blind evaluation techniques and the
small number of animals used in the earlier studies. Later, Hugod (1981) confirmed these
5 negative results using electron microscopy.
Only a few of the studies published since completion of the 1979 criteria document have
demonstrated a significant atherogenic effect of low-level CO exposure. Turner et al. (1979)
showed that CO enhanced the development of coronary artery lesions in White Carneau
pigeons that were fed a diet of 0.5 and 1 %, but not 2%, cholesterol. The exposure was to
10 150 ppm for six hours, five days each week for 52 weeks (COHb = 10%-20%). Plasma
cholesterol levels may have been increased slightly by the CO, but this was significant
(p<0.5) only at Week 11. Marshall and Hess (1981) exposed minipigs to 160, 185, and
420 ppm CO for four hours per day for 1 to 16 days (COHb = 5 to 30%). The higher
concentrations were associated with adhesion of platelets to arterial endothelium and to fossae
15 of degenerated endothelial cells. Additional changes at the higher concentration included an
increased hematocrit, an increase in blood viscosity, and an increase in platelet aggregation.
Alcindor et al. (1984) studied rabbits with induced hypercholesterolemia. Three sets of
rabbits were studied. The first was a control group receiving a normal diet and breathing air.
The second group was given a 2% cholesterol diet. The third group was given the same diet
20 and exposed to 150 ppm CO. COHb levels were not reported. Low-density lipoprotein
particles in the CO-exposed rabbits were richer in cholesterol and had a higher cholesterol-to-
phospholipid molar ratio than did the particles from the nonexposed rabbits after 45 days
Other animal studies have given generally negative results. Bing et al. (1980) studied
25 cynomolgus monkeys (Macacafascicularis). Four animals were used as controls. Seven
were exposed to CO at a level of 400 ppm for 10 alternate half-hours of each day during
12 mo. Carboxyhemoglobin levels showed a gradual increase to a peak at five hours of 20%.
The monkeys had no histologic evidence of atherosclerosis, vessel wall damage, or fat deposi-
tion in the arterial wall. There was no significant change in cholesterol or in lipoprotein
30 levels. High density to total cholesterol ratios did not differ between the CO-exposed and air-
exposed animals. These animals were on a normal diet with no augmentation of cholesterol
March 12, 1990 10-65 DRAFT-DO NOT QUOTE OR CITE
-------
or fat content. The study demonstrated that even high levels of CO exposure are not
invariably followed by arterial injury or abnormal lipid accumulation.
Similar negative results were reported by Sultzer et al. (1982) who studied swine. Pigs
with and without von Willebrand's disease were divided into groups that were exposed to
5 intermittent, low-level CO or to air. CO was delivered at 100 ppm for eight hours each week
day for four months. COHb levels averaged 7% after five hours of exposure. The degree of
coronary and aortic atherosclerotic lesion development in response to a 2% cholesterol diet
was similar in the two exposure groups. There was no effect of the ambient CO on the
degree of hypercholesterolemia induced by the diet. The findings showed no obvious effect
10 of CO on atherogenesis in hypercholesterolemic pigs.
A number of studies have examined the contribution of CO in cigarette smoke to the
purported effects of smoking on atherogenesis and thrombosis. Rogers et al. (1980) fed a
high-cholesterol diet to 36 baboons for up to 81 weeks. The animals were taught to puff
either cigarette smoke or air by operant conditioning using a water reward. Half of the
15 baboons smoked 43 cigarettes each day. The baboons were given a cigarette or sham every
15 min during a 12-h day except during times of blood drawing. Average COHb levels in
smokers were about 1.9%. Only slight differences in the very low-density lipoprotein
(VLDL), LDL and high-density lipoprotein (HDL) levels were noted between the smokers
and nonsmokers. Additionally, platelet aggregation with adenosine 5'-phosphate (ADP) and
20 collagen was similar in the two groups.
Rogers et al. (1988) extended their previous study of male baboons for an additional
1.2 years of diet and smoking (total diet, 3.2 years; total smoking, 2.8 years). They also
studied a separate group of 25 female baboons that received the diet for 2.6 years and were
exposed to cigarette smoke for 1.6 years. Blood levels of COHb were determined by GC and
25 reported both as total concentration in milligrams per deciliter and as percent saturation of
Hb, as calculated by a validated linear regression equation. Levels of COHb in the male
baboons averaged 0.64% at baseline, whereas COHb was on an average of 0.35% in female
baboons at baseline. The weekly averages of COHb levels determined after 57 weeks were
2.01 and 1.13% in male and female baboons, respectively. The baseline cholesterol levels
30 were 105 mg/dL and 88 mg/dL in the two groups of baboons. Levels at 16 weeks were
226 mg/dL in males and 291 mg/dL in females. There were no significant differences in total
March 12, 1990 10-66 DRAFT-DO NOT QUOTE OR CITE
-------
cholesterol, HDL cholesterol, or LDL cholesterol between smokers and controls. There were
slightly more fatty streaks and fibrous lesions in the carotid arteries of smokers than in
controls. No differences in lesion prevalence, vascular content of lipids, or prostaglandins
were seen in aorta or coronary arteries.
5 The results reported by Rogers et al. (1980; 1988) suggest little if any effect of cigarette
smoking on atherosclerotic lesion development in baboons. How these findings can be
extrapolated to effects of smoking in humans is difficult to know. The COHb levels attained
in the experimental animals were barely 2%. Levels in human smokers are probably 4% or
more during the waking hours of the day. On the other hand, the findings are consistent with
10 most studies of the effects of low levels of CO on atherogenesis.
A study of cockerels by Penn et al. (1983) has shown negative results as well. Three
groups of cockerels, each including seven animals, were used to determine if the atherogenic
effect of cigarette smoke could be separated from an effect due solely to CO. Cockerels
develop aortic fibromuscular atherosclerotic lesions spontaneously. Various agents, including
15 some carcinogens, have been shown to accelerate the growth in thickness and extent of these
lesions. The authors used this model by exposing one group of animals to the smoke from
40 cigarettes each day for five days each week. The cockerels were exposed from about six
weeks of age until about 22 weeks of age. A similar group of animals was exposed to CO
calibrated to give a similar COHb level to that achieved in the animals exposed to cigarette
20 smoke. The third group of animals was exposed to filtered air. Carboxyhemoglobin levels
following an exposure session were measured at 6, 9, 12, and 15 weeks. The average level
in the air-exposed animals was 1.6%. Levels in the cigarette smoke and CO groups were 6.7
and 7.2%, respectively. Atherosclerosis was quantified both by the extent of the aorta
involved and by the cross-sectional area of the intimal thickening. The cigarette smoke-
25 exposed group had more aortic lesions and lesions with greater cross-sectional area than did
either the CO-exposed group or the air-exposed group. This difference was significant at
p<0.05 in a one-tailed chi-square test. The data suggest that atherogenic effects of cigarette
smoke are not solely attributable to CO.
It has been postulated that a possible atherogenic effect of CO may be mediated through
30 an ability of CO to enhance platelet aggregation or some other component of thrombosis.
This possibility was raised in the study by Marshall and Hess (1981) noted above. Other
March 12, 1990 10-67 DRAFT-DO NOT QUOTE OR CITE
-------
studies, however, have demonstrated that the effect of CO is to depress platelet aggregation.
In one study (Mansouri and Perry, 1982), platelet aggregation to epinephrine and arachidonic
acid was reduced in in vitro experiments in which CO was bubbled through platelet-rich
plasma. Similarly, platelets from smokers aggregated less well than platelets from
5 nonsmokers, although this inhibition of aggregation was not correlated with the level of
COHb.
Madsen and Dyerberg (1984) extended these observations by studying effects of CO and
nicotine on bleeding time in humans. Smoke from high-nicotine cigarettes caused a
significant shortening of the bleeding time. Smoke from low-nicotine cigarettes caused no
10 significant change in bleeding time. CO inhalation sufficient to raise the COHb to 15% was
followed by a shortening of the bleeding time (6.0 minutes to 4.8 min), but for a short period
of time (< 1.5 h). After administration of aspirin, neither nicotine nor CO affected bleeding
times or platelet aggregation. The findings suggest that the proaggregating effects of cigarette
smoke are mediated through an inhibitory effect of nicotine on prostacyclin (PGIj) production.
15 Effects of CO in the smoke seem to be minor and short lived.
These findings were corroborated by Renaud et al. (1984). The effects of smoking
cigarettes of varying nicotine content on plasma clotting times and on aggregation of platelets
with thrombin, ADP, collagen, and epinephrine were studied in 10 human subjects. Both the
clotting functions and platelet aggregation were increased with increasing nicotine content in
20 cigarettes. There was no correlation of these parameters, however, with COHb levels.
COHb levels, reported as percent increase from baseline, achieved about a 60% increase.
Effeney (1987) has provided convincing evidence that these effects of nicotine and CO
on platelet function are mediated through opposing effects on PGI2 production. Four rabbits
were exposed to CO in an exposure chamber at 400 ppm for 7 to 10 days. Carboxy-
25 hemoglobin levels averaged about 20%. Ten rabbits received an infusion of nicotine for 7 to
10 days. Full-thickness samples of atrial and ventricular myocardium were incubated with
arachidonic acid for determination of PGI2 production by radioimmunoassay of 6-keto-PGF,a
and by inhibition of platelet aggregation. Carbon monoxide exposure increased PGI2
production which was significant in ventricular myocardium. Nicotine exposure reduced
30 PGI2 production in all tissues examined. The combination of nicotine and CO caused a net
March 12, 1990 10-68 DRAFT-DO NOT QUOTE OR CITE
-------
increase in PGI2 production. The effect of CO may be to induce hypoxemia, a known
stimulant of PGI2 production. This effect of CO would serve to reduce aggregation.
Another explanation for the antiaggregatory effect of CO exposure recently has been
provided by Brune and Ullrich (1987). These investigators bubbled CO through platelet-rich
5 plasma and then challenged the platelets with various agonists. The CO exposure was much
greater than that encountered in physiologic or even toxic states. The results, however,
indicated that inhibition of aggregation was related to enhancement of guanylate cyclase action
and associated increased cyclic guanosine monophosphate (cGMP) levels.
10 10.3.5 Summary and Conclusions
The 1984 Addendum to the 1979 Air Quality Criteria Document for Carbon Monoxide
(U.S. Environmental Protection Agency, 1984) reported what appears to be a linear
relationship between level of COHb and decrements in human exercise performance,
measured as maximal O2 uptake. Exercise performance consistently decreases at a blood level
15 of about 5.0% COHb in young, healthy, nonsmoking individuals (Klein et al., 1980; Stewart
et al., 1978; Weiser et al., 1978). Some studies have even observed a decrease in
performance at levels as low as 2.3 to 4.3% COHb (Horvath et al., 1975; Drinkwater et al.,
1974; Raven et al., 1974a); however, this decrease is so small as to be of concern mainly for
competing athletes rather than for ordinary people conducting the activities of daily life.
20 Cigarette smoking has a similar effect on cardiorespiratory response to exercise in nonathletic
human subjects indicating a reduced ability for sustained work (Hirsch et al., 1985; Klausen
etal., 1983).
Since the 1979 Air Quality Criteria Document, several important studies appearing in
the literature have expanded the cardiovascular data base. Adverse effects in patients with
25 reproducible exercise-induced angina (Allred et al., 1989a,b) have been noted with
postexposure COHb levels (CO-Ox measurement) as low as 3.2% (corresponding to an
increase of 2.0% from baseline). Sheps et al. (1987) also found a similar effect in a group of
patients with angina at COHb levels of 3.8% (representing an increase of 2.2% from
baseline). Kleinman et al. (1989) studied subjects with angina and found an effect at 3%
30 COHb representing an increase of 1.5% from baseline. Thus, the lowest observed adverse
effect level in patients with stable angina is somewhere between 3 and 4% COHb (CO-Ox
March 12, 1990 10-69 DRAFT-DO NOT QUOTE OR CITE
-------
measurement), representing an increase from baseline of from 1.5 to 2.2%. Effects on silent
ischemia episodes, which represent the majority of episodes in these patients, have not been
studied (see Chapter 12).
Exposure sufficient to achieve 6% COHb recently has been shown to adversely affect
5 exercise-related arrhythmia in patients with CAD (Sheps et al., 1989). This finding combined
with the epidemiologic work of Stern et al. (1988) in tunnel workers is suggestive but not
conclusive that CO exposure may provide an increased risk of sudden death from arrhythmia
in patients with CAD.
There is also strong evidence from both theoretical considerations and experimental
10 studies in animals that CO can adversely affect the cardiovascular system. Tables 10-4
through 10-7 are summaries of the data pertinent to the effects of CO on the cardiovascular
systems of experimental animals. Accordingly, disturbances in cardiac rhythm and con-
duction have been noted in healthy and cardiac-impaired animals at CO concentrations of 50
to 100 ppm (COHb = 2.6 to 12%); alterations in various hemodynamic parameters have been
15 observed at CO concentrations of 150 ppm (COHb = 7.5%); cardiomegaly has been reported
at CO concentrations of 200 ppm (COHb = 12%) and 60 ppm in adult and fetal animals,
respectively; changes in Hb concentrations have been reported at CO concentrations of
100 ppm (COHb = 9.26%) and 60 ppm in adult and fetal animals, respectively.
There is conflicting evidence that CO exposure will enhance development of
20 atherosclerosis in laboratory animals; and most studies show no measurable effect. Similarly,
the possibility that CO will promote significant changes in lipid metabolism that might
accelerate atherosclerosis is suggested in only a few studies. Any such effect must be subtle
at most. Finally, CO probably inhibits rather than promotes platelet aggregation. Except for
the studies by Rogers et al. (1980, 1988) on baboons, the CO exposures used in the studies
25 on atherosclerosis created COHb levels of 7% or higher; sometimes much higher. While
occupational exposures in some workplace situations might regularly lead to COHb levels of
10% or more, such high-exposure levels are almost never encountered in the
nonoccupationally exposed general public. In this general population, exposures are rarely as
much as 25 to 50 ppm, and COHb levels typically are below 3% in nonsmokers (see
30 Chapter 8). When examined in this context, this review, therefore, provides little data to
March 12, 1990 10-70 DRAFT-DO NOT QUOTE OR CITE
-------
indicate that an atherogenic effect of exposure would be likely to occur in human populations
at commonly encountered levels of ambient CO.
5 10.4 CEREBROVASCULAR AND BEHAVIORAL EFFECTS OF
CARBON MONOXIDE
10.4.1 Control of Cerebral Blood Flow and Tissue PO2 with Carbon
Monoxide and Hypoxic Hypoxia
10.4.1.1 Introduction
10 The effect of CO on cerebral blood flow (CBF) and cerebral O2 consumption (CMROj)
is complicated by the relationship between CBF and cerebral O2 delivery or availability.
Alterations in cerebral neurological function, as evaluated by neurological symptoms or
changes in evoked potential responses, are particularly difficult to correlate with changes in
CBF or cerebral O2 delivery. One of the most fundamental challenges to the organism is to
15 obtain O2 from its environment and deliver it to the tissues. However, each tissue or organ
may have regulatory mechanisms to obtain O2 which differ from other tissues or organs.
Literature concerning the cerebrovascular effects of CO is incomplete and in many cases
conflicting, and, despite the enormous literature concerning hypoxia and the cerebrovascula-
ture, the mechanisms that regulate the cerebral vessels during hypoxia are unclear.
20 Kety and Schmidt (1948) demonstrated that CMRO2 is about 3.5 mL of O2 per 100 g of
brain (cerebral hemispheres) per minute in normal adult man. This consumption of O2 is
virtually unchanged under a variety of conditions. About one-half of CMRO2 is dedicated to
synaptic transmission (Astrup, 1982; Donegan et al., 1985) and this remains relatively
constant under all conditions. Half of the remaining, vegetative O2 consumption, a quarter of
25 the overall value, maintains resting neuronal membrane potentials. The remaining quarter is
consumed by a variety of unidentified, but presumably irreducible, processes (Astrup, 1982).
In order for the brain to maintain its CMRO2 it has but two adaptations: (1) the brain could
extract more O2 from the blood or (2) CBF could increase. In fact, the brain generally relies
largely on increasing CBF for its major adaptability mechanism to provide more O2 to the
30 tissue. Thus, the following discussion concerns the regulation of CBF with hypoxia, with
little discussion of the regulation of O2 extraction.
March 12, 1990 10-71 DRAFT-DO NOT QUOTE OR CITE
-------
The cerebral vasculature responds to decreases in O2 availability by increasing CBF in
order to maintain cerebral O2 delivery and/or by increasing O2 extraction in order to maintain
cerebral O2 utilization when cerebral O2 delivery is limited. Compared with other forms of
cerebral O2 deprivation, such as hypoxic hypoxia (lowered inspired O2 concentration) and
5 anemia, CO hypoxia may interfere with O2 delivery and cerebral O2 utilization through effects
on the Hb dissociation curve and on the cytochrome oxidase system. During O2 deprivation
(hypoxia) CBF and cerebral O2 delivery may be altered by hemodynamic responses,
specifically changes in cerebral perfusion pressure, as well as the absolute amount of O2
limitation (arterial O2 content). Because hypoxia adversely effects cerebral autoregulation,
10 hypertension during hypoxia may result in an increased CBF and, hence, cerebral O2
availability. Conversely, hypotension during hypoxia may decrease CBF and cerebral O2
delivery. In the following sections, the effects of hypoxia (hypoxic and CO) on the cerebro-
vasculature will be demonstrated and the potential mechanisms of action of hypoxia on
cerebral blood vessels will be examined. The effects of CO on global and regional CBF, and
15 the effects of both high and low levels of CO on CBF and CMRO2 also will be examined.
An attempt will be made to demonstrate the potential mechanisms of action of hypoxia on the
cerebrovasculature, and the synergistic effects of CO and cyanide hypoxia on the cerebral
circulation will be examined.
20 10.4.1.2 Effects on Global Cerebral Blood Flow
At the present time, there is conflicting information concerning whether the
cerebrovascular response to CO is similar to other forms of cerebral hypoxia, such as hypoxic
hypoxia and anemic hypoxia. Few studies are available in which other types of hypoxia have
been compared to CO hypoxia, especially at similar levels of O2 deprivation. In addition,
25 comparison of cerebrovascular effects of CO and other types of hypoxia from laboratories of
different investigators is difficult because of differences in anesthetic techniques, use of
different animal species, and use of different CBF techniques. An important point to
emphasize when comparing CO hypoxia to hypoxic hypoxia is that although arterial O2
content is reduced with both hypoxic and CO hypoxia, there is no reduction in arterial O2
30 tension with CO hypoxia. Comparisons of the equivalent effects of both CO and hypoxic
hypoxia on CBF and CMRO2 in the same animal preparations have been made by Traystman's
March 12, 1990 10-72 DRAFT-DO NOT QUOTE OR CITE
-------
laboratory in several studies (Traystman and Fitzgerald, 1977; Traystman et al., 1978;
Fitzgerald and Traystman, 1980; Traystman and Fitzgerald, 1981; Koehler et al., 1982;
Koehler et al., 1984; Koehler et al., 1983; Koehler et al., 1985). The concept of equivalent
effects of both types of hypoxia (hypoxic and CO) has been described by Permutt and Farhi
5 (1969) and involves the comparison of physiologic effects of elevated COHb and low O2 at
equal reductions in Hb, arterial O2 content, arterial or venous O2 tension, or blood flow. The
focus of investigation of several laboratories, including Traystman's, concerning the effects of
hypoxia on the cerebral vasculature is not merely to describe the well-known vasodilation that
occurs, but to examine the mechanism that produces this vasodilation. This is where much of
10 the controversy lies. Here issues focus on the importance of local mechanisms in controlling
CBF such as tissue acidosis (Kety and Schmidt, 1948; Molnar and Szanto, 1964) and the
direct effect of hypoxia on vascular smooth muscle (Detar and Bohr, 1968). Some years ago
it also had been postulated that O2-sensitive carotid arterial chemoreceptors might play a role
in the CBF response during hypoxia (Ponte and Purves, 1974). Other groups of investigators
15 additionally suggested that carotid baroreceptor stimulation produces cerebral vasoconstriction
(James and MacDonnell, 1975; Ponte and Purves, 1974). This vasoconstrictor response to an
elevation in blood pressure could contribute to the autoregulatory responses of the cerebral
vessels. Because the carotid and aortic chemoreceptors are stimulated by certain forms of
hypoxia, as is systemic arterial blood pressure, the underlying mechanism of the cerebral
20 vasodilator response to hypoxia is complicated.
Traystman et al. (1978) previously reported that the increase in CBF in dogs during a
reduction in arterial O2 content, produced by breathing the animal with a low O2 gas mixture
(hypoxic hypoxia), was not different from the increase in CBF when O2 content was
decreased by adding CO to the breathing gas mixture (CO hypoxia) (Figure 10-3). This was
25 true both before and after carotid sinus chemodenervation. This study also demonstrated that
with hypoxic hypoxia mean arterial blood pressure increased, whereas it decreased with CO
hypoxia. Because the CBF increase with CO and hypoxic hypoxia was similar,
cerebrovascular resistance actually decreased more with CO hypoxia. This represents the
effect of the carotid chemoreceptors on systemic blood pressure during each type of hypoxia.
30 When the carotid chemoreceptors were denervated, cerebrovascular resistance decreased to the
March 12, 1990 10-73 DRAFT-DO NOT QUOTE OR CITE
-------
O
O
o
-J
CD
-------
same level as with CO hypoxia. They concluded that the carotid chemoreceptors do not play
an important role in the global cerebral vasodilator response to either CO or hypoxic hypoxia.
There were two important limitations of that study, however. One involved the
possibility that the aortic chemoreceptors (aortic bodies) might have a role in controlling
5 CBF, and this was not considered at that time. The aortic bodies have been reported to have
a role in the control of pulmonary blood flow (Stern et al., 1964). The other involved the
fact that sectioning of the carotid sinus nerves denervated not only the carotid
chemoreceptors, but also the carotid sinus baroreceptors. Because hypoxic hypoxia and CO
hypoxia affect blood pressure, they therefore could modulate chemoreceptor input by
10 baroreceptor input. Traystman and Fitzgerald (1981) demonstrated that the carotid and aortic
chemoreceptors are not necessary for the increase in CBF with hypoxia and that the increase
in CBF is not modified by the carotid and aortic baroreceptors (Figure 10-4). They also
showed that the cerebral vasodilation to hypoxia in carotid chemoreceptor-denervated animals
and in carotid sinus nerve sectioned and vagotomized animals resembles that occurring in
15 animals exposed to CO hypoxia, with intact chemoreceptors in which both the arterial O2
tension is high and the chemoreceptors may not be activated (Figure 10-5). CMRO2 remained
at control values under both hypoxic hypoxia and CO hypoxia conditions and was unchanged
by any denervation condition. These data support the notion that the brain increases its blood
flow in response to its O2 needs with both hypoxic and CO hypoxia in control or baroreceptor
20 and chemoreceptor denervated dogs in order to maintain CMRO2 constant. The cerebral
blood vessels appear to be relatively unresponsive to reflex stimuli (Heymans and Bouckaert,
1932; Heistad and Marcus, 1978; Heistad et al., 1976), and the CBF responses to low-
inspired O2 or elevated CO are not dependent on either the carotid or aortic chemoreceptors.
These responses also are not modified by either the carotid or aortic baroreceptors. These
25 findings would be most compatible with the idea that control of the cerebral vasculature
during hypoxia is mediated locally, however it remains possible that central (brainstem)
mechanisms are involved. These brainstem mechanisms concerning the CBF responses to
hypoxic hypoxia have been studied little, and their possible involvement in CBF responses to
CO have not been examined at all.
30
March 12, 1990 10-75 DRAFT-DO NOT QUOTE OR CITE
-------
•Jo _
i*i
UIQ X
KO e
"o 1
< o
o
e
<
60
40
20
0
140
120
100
BO
60
0
80
60
40
20
S"l
(E —
UJ
U
CONTROL
ntzi
» « HTPO'IC HTPQXIA
O •- • O CO HYPOKIA
( ) % el CONTROL
"5 To Ft 5b
BARO
DENERVATED
.
-------
CONTROL
VAGOTOMY
£
-! o _
ecu. -f
co E
UJQ V
CCO "c
o° -
CO
UJ
cc
3
_l V)
< CO —
cc <£ x
UJQ. £
cc o 3
< 0
o
CO
cc
<
3 uj !s
<=M
j'S i
< io x
CC yj c
2 * E
f^ **•*
cc
UJ
u
60
40
20
160
1 40
120
100
4
4
O1
•
•"••'.i,
. .d.u'K^
^s**v».
•IP*
•
• (112) i
^^\^
(98) J ?^
» • HYPOXIC HYPOXIA
• 0 0 CO HYPOXIA
5 ( ) •/. of CONTROL
4
•
t(l78)»
•(I"»'*S.
^«s«^
^T
•
.,M»I
^^^.^
^^V^^
^^i
X-T
x»''
• (83) H(,''
J
i . « •
6.0
4.0
2.0
°C
j£
• (63) t*f^''
• (50) •¥
• 1 « ^^^^J
/S
*<59) JS .•'
T ^x
. .(46)*'
I i i J
) 5 10 15 20 0 5 10 15 20
ARTERIAL 02 CONTENT (vol %)
Figure 10-5. Effects of hypoxic and carbon monoxide hypoxia on cerebral blood flow, mean
arterial blood pressure, and cerebral vascular resistance in control and vagotomized animals.
Data points and bars represent means ± SE of five animals. Numbers in parentheses are
percent of control. *p < 0.05.
Source: Traystman and Fitzgerald (1981).
March 12, 1990
10-71 DRAFT-DO NOT QUOTE OR CITE
-------
The idea, however, that CBF responses to CO are always similar to those of hypoxic
hypoxia is not universal. Studies on fetal (Jones et al., 1978) and newborn lambs (Jones
et al., 1981) have demonstrated that CBF increases during hypoxic hypoxia may correlate
better with decreased arterial O2 content than with decreased arterial O2 partial pressure. The
5 description of hypoxia in terms of arterial O2 content rather than arterial O2 partial pressure is
more than simply an arbitrary choice between two similar variables. When one considers
hypoxic hypoxia as a fall in arterial O2 content, this emphasizes the importance of cerebral O2
delivery to the microvascular exchange site, whereas O2 partial pressure emphasizes diffusion
from the exchange site to the parenchyma. The studies previously mentioned (Jones et al.,
10 1978; Jones et al., 1981) demonstrated that the relationship between CBF and arterial O2
content is such that the product of CBF and arterial O2 content, which equals cerebral O2
delivery, is essentially constant as arterial O2 partial pressure falls. The study in newborn
lambs (Jones et al., 1978) demonstrated that arterial fractional O2 extraction (CMRO2 per
amount of O2 delivered) was well maintained in both anemic and hypoxic hypoxia conditions.
15 The maintenance of cerebral O2 delivery and fractional O2 extraction during anemic and
hypoxic hypoxia is not unique to the lamb and does apply to adults of other species (Jones
etal., 1981).
Koehler et al. (1982), working with a newborn-lamb model, tested the hypothesis that
CBF and CMRO2 bear relationships to arterial O2 content during CO hypoxia that are not
20 different from those occurring during hypoxic hypoxia. They reasoned that if these
relationships differ between hypoxic hypoxia and CO hypoxia, then other effects of CO
exposure, such as the shift in the O2Hb dissociation curve or histotoxic effects, need to be
considered. Koehler et al. (1982) found that CO hypoxia causes a 47% greater increase in
CBF compared with hypoxic hypoxia for a similar reduction in arterial O2 content
25 (Figure 10-6). A greater CBF response to CO hypoxia than to anemic hypoxia also has been
reported in humans (Paulson et al., 1973). In the study of Koehler et al. (1982), CMRO2 and
O2 delivery were constant during hypoxic hypoxia. Thus, fractional O2 extraction, which
equals O2 consumption divided by O2 delivery, remained constant with hypoxic hypoxia.
During CO hypoxia, although CMRO2 remained constant, O2 delivery increased and fractional
30 O2 extraction decreased. This decline in fractional O2 extraction was correlated with the
leftward shift of the O2Hb dissociation curve that accompanied CO hypoxia. In this situation,
March 12, 1990 10-78 DRAFT-DO NOT QUOTE OR CITE
-------
250
200
Cerebrol
Blood ,50
Flow
(ml/min/lOOg)
100
50
•— Control ond HH
0— COM
0.4 0.6 0.8
Fractional Arterial O Saturation
1.0
Figure 10-6. Cerebral blood flow as a function of fractional arterial O2 saturation. Circles
represent control or hypoxic hypoxia; squares represent CO hypoxia.
Source: Koehler et al. (1982).
March 12, 1990
10-79 DRAFT-DO NOT QUOTE OR CITE
-------
the additional increase in CBF with CO hypoxia could be explained because the shift in the
curve lessens the O2 diffusion gradient into the tissue and further lowers the arterial O2 partial
pressure. In other words, while increases in CBF maintain O2 availability at the micro-
vascular exchange site, overall O2 transport to the cells becomes relatively more diffusion-
5 dependent with CO hypoxia. Although these investigators believe that the best explanation
for the difference in the CBF response to hypoxic hypoxia and CO hypoxia is the leftward
shift of the O2Hb dissociation curve, they cannot completely rule out a potential histotoxic
effect of CO resulting from the competition of CO and O2 for cytochrome aa3 oxidase. This
effect generally is considered to be insignificant at low CO levels because in vitro cytochrome
10 oxidase remains completely oxidized until very low tissue O2 partial pressure levels are
reached. However, Hempel et al. (1977) has shown that cerebral cytochrome aa3 in vivo is in
a substantially reduced state, raising the possibility that CO may readily compete with O2 at
relatively low CO levels. This binding relationship between CO and the oxidase also has
been demonstrated by Piantadosi et al. (1985; 1987). On the other hand, if CO were exerting
15 a histotoxic effect, CMRO2 would be expected to fall, and this was not observed in Koehler's
experiments. It also is possible, however, that CO could have a direct histotoxic effect on
cerebral vascular smooth muscle, independently of brain tissue metabolism.
In a subsequent study, Koehler et al. (1984) compared the effect of hypoxic hypoxia and
CO hypoxia on CBF in adult and newborn sheep in which arterial O2 content was reduced to
20 50 to 60% of control with both types of hypoxia. During hypoxic hypoxia, CBF increased to
maintain cerebral O2 delivery in both adults and newborns; however CMRO2 did not change.
Although CMRO2 was higher in newborns, the responses of CBF/CMRO2 to hypoxic hypoxia
was not different in newborns and adults. In newborns and adults, CBF increased to a greater
extent with CO hypoxia than with hypoxic hypoxia for similar reductions in arterial O2
25 content (Figure 10-7). This resulted in an increase in cerebral O2 delivery with CO hypoxia.
As discussed previously, the degree to which CO hypoxic differed from hypoxic hypoxia
correlated with the magnitude of the leftward shift of the O2Hb dissociation curve that
accompanies CO hypoxia. In the adult animals with CO hypoxia, CMRO2 was reduced by
16%, however, CMRO2 was maintained in the newborns. These data allowed for the
30 conclusion that maintenance of cerebral O2 delivery during hypoxic hypoxia is a property of
CBF regulation common to both newborn and adult sheep. During CO hypoxia, the position
March 12, 1990 10-80 DRAFT-DO NOT QUOTE OR CITE
-------
.50
I
** .40
•
"o
>
~ .30
^.
N
O
> 20
I
CSJ
O JO
o
o
8
(vol. %)
12
16
Figure 10-7. Comparison of newborn and adult responses of the reciprocal of the cerebral
arteriovenous O2 content difference (C.O2 - QO^-l to a reduction in arterial O2 content
during hypoxic hypoxia (HH). Open circles, room-air control in newborns; solid circles, HH
in newborns; open triangles, room-air control in adults; solid triangles, HH in adults.
Regression lines were fitted to the reciprocal of C.O2. For newborns (solid line), (C.O2 -
CAH = 1.74 C.O2-1 + 0.01 (r = 0.91). For adults (dashed line), (C.O2 - CvO2)-l =
1.64 C.O2 + 0.02 (r = 0.67). Responses of blood flow per unit O2 consumption are not
significantly different between newborns and adults.
Source: Koehler et al. (1985).
March 12, 1990
10-81 DRAFT-DO NOT QUOTE OR CITE
-------
of the OzHb dissociation curve is an additional factor that sets the level of O2 delivery. The
fetal conditions of low arterial-O2 content and a left-shifted OaHb dissociation curve may have
provided the newborn with a microcirculation better suited for maintaining CMRO2 during CO
hypoxia.
5 Some points need to be considered when comparing the cerebrovascular effects of
hypoxic hypoxia and CO hypoxia as described in the dog studies (Traystman and Fitzgerald,
1981; Traystman et al., 1978) with those described in the newborn and adult sheep. In the
studies concerning the anesthetized dogs, the CBF response to hypoxic hypoxia and CO
hypoxia was not statistically different although the mean blood flows tended to be higher with
10 moderate levels of CO. This response was statistically significant in the newborn and adult
sheep experiments. One likely explanation for the different result is that arterial blood
pressure declined during CO hypoxia in the anesthetized dog, whereas it was well maintained
in the unanesthetized sheep. Because cerebral autoregulation is impaired during severe
hypoxia (Haggendal and Johannsson, 1965), a drop in perfusion pressure during CO hypoxia
15 may have limited the increase in CBF in the dog. A reexamination of data from the dog
study indicated that cerebral O2 delivery increased and fractional O2 extraction decreased
during CO hypoxia. Thus, the dog study is consistent with the data obtained in the sheep
study. Similar results have been reported in humans (Paulson et al., 1973) and in goats
(Doblar et al., 1977). Another possible explanation for the differences in CBF responses to
20 hypoxia in the dogs versus the sheep is that the dogs were anesthetized with sodium
pentobarbital, whereas the sheep were studied in the unanesthetized state. Pentobarbital
anesthesia reduces CBF and CMRO2 so that differences in flow would be minimized in the
dog studies. Finally, the Px of sheep Hb is considerably higher than in the dog (44 mmHg
for sheep vs. 27 mmHg for dogs) so that the leftward shift of the O2Hb dissociation curve
25 would be larger in sheep and therefore result in a greater increase in CBF with CO hypoxia
versus hypoxic hypoxia. Evidence supporting a role for Px in the CBF response to CO was
obtained by Koehler et al. (1983) in experiments in which lambs were first exchange
transfused with high-Pjo donor blood, which resulted in an increase in cerebral fractional O2
extraction. With the induction of CO hypoxia to return Px to the pretransfusion level,
30 cerebral O2 delivery and O2 extraction also returned to pretransfusion levels. These
investigators suggested that since PK can affect capillary and tissue O2 partial pressure
March 12, 1990 10-82 DRAFT-DO NOT QUOTE OR CITE
-------
independent of arterial O2 content, the position of the O2Hb dissociation curve appears to set
the level of cerebral O2 delivery about which CBF is regulated when arterial O2 content is
reduced. These data, taken together with the previously mentioned work from this group, are
consistent with the existence of a tissue O2 tension-dependent mechanism controlling the
5 cerebral vasculature in which tissue O2 tension is a function of CMRO2, cerebral O2 delivery
to the microcirculation, the position of the O2Hb dissociation curve, and microcirculatory
morphology.
10.4.1.3 Effects on Regional Cerebral Blood Plow
10 Both human and animal histopathology studies have suggested that there are regional
differences in tissue injury following severe CO exposure. One potential source of these
differences is regional differences in the CBF response to CO exposure. Two logical
comparisons of the regional CBF response to CO hypoxia are (1) anatomical, (i.e., rostral to
caudal [cortex to brainstem] comparison) and (2) physiological (i.e. brain areas with a func-
15 tional blood brain barrier versus brain areas without an intact blood brain barrier). Koehler
et al. (1984) observed interesting regional CBF responses to hypoxic and CO hypoxia in
newborn lambs and adult sheep (Table 10-8; Figure 10-8). In adults, regions with high
normoxic blood flows such as the caudate nucleus and midbrain showed a large response to
hypoxia, whereas lower blood flow regions with large white matter tracts, such as the cervical
20 spinal cord, pons, diencephalon, and piriform lobe, showed a relatively lower response.
Other brain regions were essentially homogeneous in their responses. CO hypoxia increased
CBF to a greater extent than hypoxic hypoxia in all brain regions, but the overall pattern of
regional CBF was similar for the two types of hypoxia in the adults. In Table 10-8 the
regions are listed in order from highest to lowest responsivity, and the particular groups of
25 regions that are significantly different are separated by pairs of vertical brackets. In the
newborns, regional responses differed for each type of hypoxia. With hypoxic hypoxia in the
newborns, the brainstem regions had a significantly greater response than all other regions
except the caudate nucleus, whereas all cerebral lobes responded significantly less than all
other regions. With CO hypoxia, the difference between brainstem responses and those of
30 other regions was less marked. In the adults, in contrast, there was no significant interactive
effect between the type of hypoxia and the pattern of regional response. With both types of
March 12, 1990 10-83 DRAFT-DO NOT QUOTE OR CITE
-------
jEMiar JBEI •'fipa® %jeq^n^a^g HE
nn
E
r
r
4-
I
r r
,j ,i
I I
fit
-------
I
i
i
I
,t
n
I
P
,
it
41
I
lit
"i.1
flri
i 11 r
IRINAU
G^MIHILLUM
MIOUUUA
P3NS
MiDlRAIN
CAUDATE N.
I,
L-:
F NONTAX I
1
> E
-------
reported. Okeda et al. (1987) also demonstrated CO-induced regional CBF differences in
cats. This group attempted to demonstrate that there is a selective vulnerability of the
pallidum and cerebral white matter and showed low CBF values for these brain areas. In the
newborn lambs (Koehler et al., 1984), unlike the adult sheep, the patterns of regional CBF
5 responses were not similar with the two types of hypoxia. Brainstem regions, especially the
medulla, had marked responses relative to the cerebrum during hypoxic hypoxia. Peelers
et al. (1979) also have made this observation in unanesthetized fetal lambs. Rosenberg et al.
(1982) found in the fetal and neonatal lamb, but not in the adult sheep, that the brainstem also
had a greater CBF response to arterial CO2 than other regions, and Cavazzuti and Duffy
10 (1982), in newborn dogs, observed results consistent with those in the lamb in that brainstem
regions displayed a greater blood flow response to hypoxic hypoxia and to hypercapnia. The
most likely explanation in the puppy for the higher brainstem response to hypoxia and
hypercapnia is that this region has a higher CMRO2. Normoxic glucose consumption in vivo
(Cavazzuti and Duffy, 1982) and O2 consumption in vitro (Himwich and Fazekas, 1941) are
15 relatively high in the brainstem of the neonatal puppy. This also may be true in the neonatal
lamb. However, if this were the only explanation, there should be a much greater CBF
response in brainstem regions relative to the rest of the brain during CO hypoxia, as well as
hypoxic hypoxia, but this has not been observed. The larger CBF response of caudate
nucleus with hypoxia compared with cortical lobes may be explained by the high fraction of
20 grey matter in the caudate nucleus. The large response of the brainstem region also may be
partly explained by a relatively high proportion of grey matter. It also is likely that increased
activation of cardiovascular and respiratory centers in the brainstem during hypoxia produces
local increases in metabolism and O2 demand, which in turn would produce an additional
increase in blood flow in this area. The capability of the central nervous system to increase
25 CMRO2 during hypoxic hypoxia has been demonstrated in certain strains of rats (Berntman
et al., 1979). An alternative explanation for the regional differences in sensitivity to hypoxia
is that stimulation of the peripheral chemoreceptors by hypoxia produced sympathetic
vasoconstriction preferentially in the cerebral hemispheres. This explanation is considered
unlikely because previous studies have shown that neither carotid nor aortic chemoreceptor
30 denervation alters CBF from cortical regions during hypoxic or CO hypoxia (Traystman and
Fitzgerald, 1981; Traystman et al., 1978).
March 12, 1990 10-86 DRAFT-DO NOT QUOTE OR CITE
-------
There is at least one area of the brain that does not respond to alterations in arterial O2
content and partial pressure as do other brain areas, the neurohypophysis. The
neurohypophysis is an anatomically unique region of the brain, and the regulation of blood
flow to this area appears to be different from other areas of the brain. Hanley et al. (1986)
5 demonstrated that when arterial O2 content was reduced equivalently with hypoxic hypoxia
and CO hypoxia, global CBF increased by 239 and 300%, respectively. Regional CBF also
showed similar responses for all brain areas except the neurohypophysis. With hypoxic
hypoxia, neurohypophysis blood flow increased markedly (320%), but it was unchanged with
CO (Figure 10-9). These blood flow responses of the neurohypophysis occur independently
10 of alterations in blood pressure.
Wilson et al. (1987) determined the role of the chemoreceptors in the neurohypophyseal
response to hypoxia and found that chemoreceptor denervation completely inhibited the
increase in neurohypophyseal blood flow associated with hypoxia. The response to CO was
unaltered (Figure 10-10). These data (Hanley et al., 1986; Wilson et al., 1987) demonstrated
15 that the mechanism responsible for the increase in neurohypophyseal blood flow with hypoxia
is unique when compared to other brain regions. The only animals in which
neurohypophyseal blood flow did not respond were the denervated animals and those given
CO. Both of these conditions are ones in which the chemoreceptors have been shown to be
inactive (Traystman and Fitzgerald, 1981; Comroe, 1974). Although for most brain regions
20 the blood flow response to hypoxia does not involve the peripheral chemoreceptors, this is not
true for the neurohypophysis. Thus, here is an example of one regional brain area that does
not respond to CO hypoxia (i.e., a change in arterial O2 content) but does respond to a change
in arterial O2 tension. This suggests that the chemoreceptor represents the mechanism
involved in the neurohypophyseal response to hypoxic hypoxia and that local changes in
25 arterial O2 content are not involved in this response, because the neurohypophysis does not
respond to CO. It is unclear whether other regional brain areas have similar responses and
mechanisms to hypoxic and CO hypoxia.
10.4.1.4 Effect of Low Levels of Carbon Monoxide on Cerebral Blood Flow
30 Little information is available concerning the effects of low levels of CO on the cerebral
vasculature. This is particularly unfortunate because many investigators have shown
March 12, 1990 10-87 DRAFT-DO NOT QUOTE OR CITE
-------
CONTROL I CD CONTROL 2
HYPOXIC HYPOXIA
CO-HYPOXIA
O
O
z
2
13
u.
CD
U
200
160
160
140
120
100
60
60
40
20
0
CEftCBMAl
HEMISPMCJU
CAUOATC
WMITC
MATTER
KTPOTHALAMUS COVKUJLM
MCOIAN
CMINtMX
Figure 10-9. Effect of hypoxic hypoxia and carbon monoxide (CO) hypoxia on
neurohypophyseal and regional cerebral blood flow (rCBF). Each bar represents mean ± SE
of five dogs. Both types of hypoxia (diagonal and cross-hatched bars) produced significant
increases from control (open and dark bars) in blood flow to all regions except
neurohypophysis. Both parts of neurohypophysis, median eminence and neural lobe, showed
no change from control with CO hypoxia but did not have significant flow responses to
hypoxic hypoxia. Note changes in vertical axis at right for median eminence and neural lobe
blood flow.
Source: Hanley et al. (1986).
March 12, 1990
10-88 DRAFT-DO NOT QUOTE OR CITE
-------
8
CD
125
CO
75
50
25
KNCRVATO
INTACT
2000
o
CD
1600 3?
1200
600
400
5
i
*>
§
NORMOXIA
HYPOXIC
HYPOXIA
NORMOXIA
HYPOXIC
HYPOXIA
CEREBRAL HEMISPHERES
NEUROHYPOPHYS1S
Figure 10-10. Effect of complete chemoreceptor denervation on total cerebral and
neurohypophyseal blood flow. Each line represents mean ± SEM of six dogs. (Note change
in y axis for neurohypophyseal blood flow.)
Source: Wilson et al. (1987).
March 12, 1990
10-89 DRAFT-DO NOT QUOTE OR CITE
-------
behavioral and electrophysiological abnormalities with various levels of CO exposure
(Xintaras et al., 1966; Beard and Wertheim, 1967; Fodor and Winneke, 1972; Horvath et al.,
1971), and it is conceivable that these effects could result from abnormalities of CBF.
Carbon monoxide hypoxia results in an increase in CBF, and this has been demonstrated by a
5 number of investigators (Traystman and Fitzgerald, 1981; Traystman et al., 1978; Sjostrand,
1948; Haggendal and Norbeck, 1966; Paulson et al., 1973). However, many difficulties have
been encountered in these experiments such as extracranial contamination of the measured
CBF, surgical trauma to the cerebral vasculature, inadequate control of blood gases, and
failure to measure COHb concentrations.
10 Traystman (1978) examined the CBF responses to CO hypoxia in anesthetized dogs,
particularly in the range of COHb less than 20% (Figure 10-11). A COHb level as low as
2.5% resulted in a small, but significant, increase in CBF to 102% of control. With
reductions in O2-carrying capacity of 5, 10, 20, and 30% (COHb 5, 10, 20, and 30%) CBF
increased to approximately 105, 110, 120, and 130% of control, respectively. At each of
15 these levels, CMRO2 remained unchanged. At COHb levels above 30%, CBF increased out
of proportion to the decrease in O2-carrying capacity, but the brain could no longer maintain
CMRO2 constant. At a COHb level of 50%, CBF increased to about 200% of control. These
findings are in general agreement with those of MacMillan (1975) who demonstrated that as
COHb increased to 20, 50, and 65%, CBF increased to 200, 300, and then 400%,
20 respectively, in cats. These CBF increases at 20% COHb are higher than those reported in
Traystman's (1978) study but the reason for this is not known. Haggendal and Norbeck
(1966) demonstrated a 50 to 150% of control increase in CBF with COHbs of 30 to 70%, and
Paulson et al. (1973) showed a 26% increase in CBF with a COHb of 20%. These findings
also indicate that CBF increases progressively with increasing COHb concentrations, and that
25 CMRO2 is maintained constant even at a COHb level of 30%. This has important
implications regarding the behavioral and electrophysiological consequences of CO exposure.
These findings also would be consistent with those of Dyer and Annau (1978) who found that
superior colliculus-evoked potential latencies are not affected by COHb levels up to 40%. At
levels above this, the brain cannot increase blood flow enough to compensate for decreased
30 tissue O2 delivery. At these high COHb levels, then, behavioral and neurophysiological
abnormalities should be quite evident. At lower COHb levels, these abnormalities should not
March 12, 1990 10-90 DRAFT-DO NOT QUOTE OR CITE
-------
c
O
u
Q
O
O
m
m
LU
o:
Ld
O
220
210 -
200 -
190 -
180 -
170 -
160 -
150 -
140 -
130 -
120 -
110 -
100
10 20 30
CARBOXYHEMOGLOBIN, %
40
50
Figure 10-11. Effect of increasing carboxyhemoglobin levels on cerebral blood flow, with
special reference to low-level administration (below 20% COHb). Each point represents the
mean ± SE of 10 dog preparations.
Source: Traystman (1978).
March 12, 1990
10-91 DRAFT-DO NOT QUOTE OR CITE
-------
OdiPMOHiiSy., m x oanpnoMBBd timriliiji IHH irialanr fLc.,, nations wife
|miijyit cmiaB ant •«.««•«?. ifc Hnnd Ham or rmttaeA^
CadtGKtiOB tD I
clfal* of hjyuua at these Icwds CLci.,, vpto
r «'Jm MiiijAjMii J
20
iiKjibp^
addfeakiH. the idea of a Ifltao&nld teiidl bdov vlodi donees M CGHb varid ant
T 197g|
I978|. A JftErzataA! fcicB
15 a^dtoiJiq|iiyairinEpk^dBaoi^^
BeaBnd and Weriham, 1967). Beczne Tn^tmn (1978|
CfflF and ititaMe CMRO, lavtoCXHblevdbof 30%,tiUssflgge^aliiii»t
M^ Sevoal
20 alimillin^iaiCTF^Alreiiomiiypa^ McDowan (1966) reported a tinesfaoM xnciial OL
rf SP mmUjg m samto&far^ Ajfp. CBF bpgan to increase as aiterial Qz tension
H|u ^ KognrcetaL
oaofinnBd MdXwnalS''s ^'"'^1^c
-------
tcfQK. Aho, becaanecf Inssrf fp^ftaniliiiflirii artcrinies(DBiiagctaL,
1979) and fesjgn«rt slope of
•jpednfic finKtaan. flat applies owr a. wide aange of Oj,
10
absorption is 2
in fine deadis (Radfanl et >L, 19H6). Cyanide also has been
2©
«*r - -
1959; Meyer, 1963; RriedyetaL, 1976). Gonatose stales and deep
i of dectncal activity have been obsetvod in a vaiiety of y*"»f following, cyanide
(WanlandWheafley, 1947; Brierty, 1975). Cyanide appeared to sdectmahj
te matter (South et aL, 1963; Levine, 1967), but it is unclear wteHief
25 dfonennipalholo^iRas due to diiect effects of cyanide on n^
i or ischemia. Uris neuropathology aho may have been due to n^analty
Ldradati
The action of cyanide on ffle cendbnl vasodatnie and on CMRO^ has not been studied in
any peat Asaffl nnder cimiiiJlrf cnMBiiMif Hin^iffiH-it in meihuds, anhnal species.,,
dosages, finhne to nxasnie blood or tissue cyanides, lack of control in regard toother
as respiration and consequently CO^ and diffiailty in securing pore cerebral
March 12, 1990 1O^3 DRAFT-DO NOT QUOTE OR CHE
-------
venous samples, have led to some inconsistencies in observations and interpretations of
cyanide and the cerebral circulation. This has been so for CO as well. McGinty (1929)
observed an increase in the outflow from the sagittal sinus of anesthetized dogs following a
small dose of sodium cyanide. Paulet (1958) observed an increase in cerebrospinal fluid
5 pressure following a bolus injection of cyanide, and concluded that CBF must have increased.
Studies in cats and rabbits showed large increases in blood flow through the sagittal sinus
without any alteration in the arterio-venous O2 content difference across the blood brain
barrier (Russek et al., 1963). Russek observed an increase in CMRO2 to 300% of control in
these experiments. He concluded that stimulation of the carotid chemoreceptors caused the
10 increase in CMRO2 and that the increase in CBF was secondary to the metabolic change.
Brierly et al. (1976) speculated that CBF decreased during cyanide administration due to
cyanide's depressive effect on the myocardium. However, he never actually measured CBF,
but rather he made speculations on CBF that were based upon vascular pressure changes in
the sagittal sinus of monkeys. Aliukhin et al. (1974) also observed increases in CBF
15 following acute cyanide poisoning in rats.
The cerebral metabolic response to cyanide has been studied in vitro and in vivo.
McGinty (1929) observed an outpouring of lactic acid into the cerebral venous drainage
during moderate cyanide intoxication, and this was the first indication of an altered cerebral
aerobic metabolism in cyanide hypoxia. Fazekas et al. (1939) reported a decreased CMRO2
20 in dogs following administration of potassium cyanide, however, these workers did not
measure CBF. Doses of cyanide sufficient to decrease cerebral cytochrome oxidase activity
by 50% in rats led to increases in lactate, inorganic phosphate, and triphosphate (Albaum
et al., 1946). Olsen and Klein (1947) reported similar metabolic findings in rats and
calculated that glucose consumption must have increased to account for the rise in lactic acid.
25 They also discussed how the addition of cyanide to brain slices, in vitro, failed to decrease O2
consumption until high cyanide levels were achieved. Gasteva and Raize (1975) demon-
strated a reduction in CMRO2 with cyanide.
Early studies of the combined effects of CO and cyanide were inconclusive in describing
any interaction of these two agents (Hofer, 1926; Moss et al., 1951). Two inhalation
30 toxicological studies have reported no interaction between CO and cyanide with respect to
lethal-dose levels (Higgins et al., 1972; Yamamoto, 1976). Smith et al. (1976b) using
March 12, 1990 10-94 DRAFT-DO NOT QUOTE OR CITE
-------
behavioral assessments reported at least an additive effect of cyanide and CO on the time to
cause incapacitation and death in exercising rats. However, few studies have ever examined
the combined effects of these agents on the cerebrovasculature. Pitt et al. (1979) examined
individual and combined effects of cyanide and CO on CBF and CMRO2 because many of the
5 deleterious effects of the fire environment may be due to altered cerebral nervous system
function. They found that CBF increases during CO hypoxia and it also increases with
cyanide hypoxia. CMRO2 remained unchanged until the higher levels of either cyanide or CO
were reached. These data are consistent with those previously presented for CO. When CO
and cyanide were administered simultaneously, CBF increased in an additive manner
10 (Figure 10-12), but significant decreases in CMRO2 occurred at the combination of the lower
concentrations (Figure 10-13). These data suggest that CO and cyanide are physiologically
additive in producing changes in CBF, but may act synergistically on CMRO2.
Figure 10-14 (from Pitt et al., 1979) demonstrates the relationship between CBF and
CMRO2 (VEOz) in CO and cyanide hypoxia. Three aspects of this figure suggest that cyanide
15 and CO may act through similar mechanisms with respect to changes in CBF and CMRO2.
First, low doses of either agent alone produce increases in CBF that maintain CMRO2
constant. Second, higher doses of CO or cyanide increase CBF to 200% of control while
CMRO2 decreased to around 80% of control. Finally, combinations of cyanide and CO
hypoxia result in an increased CBF with a decreased CMRO2 that would be predicted on the
20 basis of an additive effect. Although CO binds to nonhemoglobin proteins, including
cytochrome oxidase in vitro, it is unlikely that in vivo this binding contributes to the hypoxic
effect of CO (Root, 1965). Because we demonstrated previously that the effect of CO
hypoxia on the brain was similar to an equivalent reduction in arterial O2 content with hypoxic
hypoxia, it may be that the mechanism that mediates the increase in CBF to maintain CMRO2
25 relatively constant with CO and hypoxic hypoxia, may be similarly affected by blocking
cellular respiration with cyanide hypoxia. The nonspecificity of the hypoxic response
suggests a common mechanism of cerebral vasodilation with hypoxia, and it lends support to
a metabolic control of cerebral vessels although the precise mediator is unknown.
As Pitt et al. (1979) explains, there are several explanations for the loss of maintaining
30 CMR02 constant when CBF increased to 200% of control, or more, for combinations of CO
and cyanide hypoxia. First, it is possible that the cortical cells themselves were damaged by
March 12, 1990 10-95 DRAFT-DO NOT QUOTE OR CITE
-------
-I 400
O
o:
t; 300
o
o
„. 200-
o
^ 100-
u.
CQ o
o
• Q 1.5 ^.g/ml CN
,5 1.0 pg/ml CN
No CN
10 20 30 40 50
% COHb
Figure 10-12. Effect of cyanide (CN) and CO hypoxia, alone and in combination, on
cerebral blood flow. Each point represents mean + SE. Closed circles = CO alone
(19 animals); open circles =1.0 /tg/mL blood CN (12 animals); and open squares =
1.5 Mg/inL blood CN (seven animals).
Source: Pitt et al. (1979).
March 12, 1990
10-96 DRAFT-DO NOT QUOTE OR CITE
-------
o
cc
H
z
o
o
CM
o
125
100
75
50
25
f^T
No CN
1.0 pg/ml CN
1.5 /ig/ml CN
10 20 30 40 50
% COHb
Figure 10-13. Effect of CN and CO hypoxia, alone and in combination, on cerebral oxygen
consumption. Each point represents mean ± SE. Closed circles = CO alone (19 animals);
open circles =1.0 /ig/mL blood CN (12 animals); and open squares = 1.5 /tg/mL blood CN
(seven animals).
Source: Pitt et al. (1979).
March 12, 1990
10-97 DRAFT-DO NOT QUOTE OR CITE
-------
300
o 250
§200
150
100
CD
O
50
• CO ALONE
o CN ALONE
o CN/CO
4 CONTROL
25 50 75 100 125
V 02 (% of Control)
Figure 10-14. Relationship of CBF to cerebral O2 consumption (VOj) during CN and CO
hypoxia. The mean ± SEM of CBF (percent of control) and VO2 (percent of control) is
plotted for the groups previously described. A straight line was fitted by regression for the
effects of CO or CN alone, and extended up to a CBF of 300% of control.
Source: Pitt et al. (1979).
March 12, 1990
10-98 DRAFT-DO NOT QUOTE OR CITE
-------
combination of cyanide and CO and were rendered incapable of utilizing the O2 and substrate
delivered. Because the observed physiological parameters returned to control when blood
cyanide and COHb levels were returned to control, the cortical cell damage was temporary
and reversible. Second, it is possible that within the brain, regional inequalities existed
5 between the increase in CBF and the metabolic demands of the tissue. Third, it is possible
that in severe hypoxia, although the energy state of the brain may be normal and CBF
increases in an apparent adequate manner, a compensatory decrease in neuronal activity is
produced (Duffy et al., 1972). Finally, a fourth mechanism of how cyanide and CO hypoxia
could act to synergistically reduce CMRO2 is based on the characteristics of the multi-enzyme
10 cellular respiratory chain (Chance et al., 1970). Under normal metabolic conditions,
cerebral mitochondria! respiration is zero order with respect to O2, until mitochondria! arterial
O2 tension is less than 1 mmHg (Chance et al., 1962). During low levels of cyanide hypoxia,
the abundance of cytochrome oxidase (Lubbers, 1968), the ability of unblocked respiratory
chains to branch out and oxidize cyanide blocked chains, and the ability of a multi-enzyme
15 system to maintain a constant electron flow for a wide range of steady-state changes in the
terminal oxidase (Chance et al., 1970) enable the brain to maintain CMRO2 constant by
increasing CBF. However, further reduction in O2 supply with the addition of CO result in a
further reduction in the enzyme system.
20 10.4.1.6 Mechanism of Regulation of Cerebral Blood Flow in Hypoxia
Although it is clear that hypoxia produces cerebral vasodilation and an increase in CBF,
the precise mechanism by which this occurs is not clear. Hypotheses to explain this
mechanism include direct effects of O2, neurogenic (which was referred to earlier in this
chapter), and chemical or metabolic theories. Little evidence exists concerning the direct
25 effects of O2 on cerebral vessels, however, there is some evidence that O2 may act directly on
the smooth muscle of cerebral vessels, with a high O2 tension leading to vasodilation. Garry
(1928) showed that spirals of carotid artery of sheep contract with high O2 tension. This
response was confirmed later by Smith and Vane (1966) and Detar and Bohr (1968), and in
addition, Detar and Bohr (1968) showed that isolated rabbit aortas dilated when perfused with
30 blood or saline with low O2 tension. The dependence of the contractile response to O2 tension
is explained if one assumes that O2 plays a metabolic role within the mitochondria of smooth
March 12, 1990 10-99 DRAFT-DO NOT QUOTE OR CITE
-------
ft fa
OE fc «•• (pi ^mij|» tamaxr, 'Kittam aad Mag (lyo| fantr
1Q,
da
- --.- »_ ^» - »• - -
qpaon atauaiifc •> 1% »••• OGMB enat
mmrMarteriBJB
rfi
to lit Ih ••iliMiim nf ii inlail i imiliHiiai imift fcjjnrii QTrtr, I"T" Tiiiimjj ITrfi)
1971).
i
-------
15
1L^»^ ^u^^—^- "•• J. __
-------
increased CBF with hypercapnia has been demonstrated (Shalit et al., 1967). Other
investigators also have demonstrated the potential involvement of higher brain centers in the
regulation of CBF (Langfitt and Kassell, 1968; Molnar and Szanto, 1964; Stavraky, 1936),
and it remains possible that these central neurogenic centers may be involved in the CBF
5 response to hypoxia.
Carbon monoxide can compromise tissue oxygenation in three ways: a fall in arterial O2
content, an increase in OjHb affinity, and theoretically, a direct cellular effect (Coburn,
1979). Arterial O2 content falls as CO occupies O2-binding sites, but when CO occupies
binding sites, the O2 affinity of the remaining sites increases (Paulson et al., 1973; Coburn,
10 1979; Roughton and Darling, 1944). As a result, cerebral venous O2 tension, and presumably
tissue O2 tension, decrease. Both types of hypoxia, hypoxic and CO, produce essentially
identical decreases in cerebral venous O2 tension. This reduction in cerebral venous O2
tension could be translated into an increased CBF via some of the mechanisms previously
described (i.e., adenosine, O2 receptors, neurogenic mechanisms, etc). Because arterial O2
15 tension falls only with hypoxic hypoxia, one has difficulty in ascribing changes in CBF to
alterations in arterial O2 tension. In CO hypoxia, arterial O2 content is reduced, not O2
tension, thus one must consider the possibility of O2 content-type receptors, which is unlikely.
Another controversial potential effect of CO is an identifiable direct effect of CO on cellular
metabolism. If the CO/O2 tension ratio in the mitochondrion is sufficiently high, CO can
20 combine with cytochrome a3 (Coburn, 1979). This would prevent oxidation at the terminal
electron transport chain and would be equivalent to a lack of molecular O2. Whether this
potential mechanism operates in vivo is unclear at the present time.
10.4.1.7 Summary
25 The data reviewed indicate that CO hypoxia increases cerebral blood flow, even at very
low exposure levels. Cerebral O2 consumption is well maintained until levels of COHb reach
upwards of 30%. The overall responses of the cerebrovasculature are similar in the fetus,
newborn, and adult animal; however, the mechanism of the increase in cerebral blood flow is
still unclear. In fact, several mechanisms working simultaneously to increase cerebral blood
30 flow appear likely and these may involve metabolic and neural aspects as well as the O2Hb
dissociation curve, tissue O2 levels, and even a histotoxic effect of CO. These potential
March 12, 1990 10-102 DRAFT-DO NOT QUOTE OR CITE
-------
mechanisms of CO-induced alterations in the cerebral circulation need to be investigated
further. The interaction of CO with cyanide (additive and synergistic) on the cerebral
vasculature is clear, however the interaction of CO with other agents and their combined
effects on brain blood vessels is unknown. This also is true for the long-term (chronic)
5 effects of CO alone, or in combination with other agents in low- or high-dose levels on the
cerebral vasculature. Finally, under normal circumstances the brain can increase its blood
flow or its O2 extraction in order to compensate for a reduced O2 environment. Whether these
compensatory mechanisms continue to operate successfully in a variety of conditions where
the brain, or its vasculature are compromised (i.e., stroke, head injury, atherosclerosis,
10 hypertension) is unknown and requires further investigation.
10.4.2 Behavioral Effects of Carbon Monoxide
10.4.2.1 Introduction
The following is an evaluative review of the literature concerning the behavioral and
15 nervous system effects of elevated COHb. An effort was made to organize the findings by
subject matter, devoting a section of the review to each of several subtopics. Such an
organizational scheme is always arbitrary and therefore occasionally strained. The
organization of the material is, however, a benefit which outweighs the disadvantages.
Extensive use is made of tables in each subtopic to help summarize the findings and give
20 a critique of each study. For each published report the following information is given:
duration of CO exposure, range of COHb achieved, number of subjects studied (n), the
dependent variable studied, the authors' conclusions about effects, a comment about the
features of the study, and technical critique notes.
The technical critique notes refer to technique problems that frequently exist in studies
25 in the CO literature concerned with behavior. A critique code has been devised to facilitate
reference to two of the most common problems - blinding and multiple statistical tests
conducted on the same data base. The following paragraphs describe the problems and a table
is given to define the codes used.
One of the features of a study is the so-called blinding of subjects and experimenters to
30 the exposure conditions. To avoid the effects of suggestion and expectation on the part of the
subject, the subject should not be informed about his own exposure condition until after the
March 12, 1990 10-103 DRAFT-DO NOT QUOTE OR CITE
-------
completion of the study (i.e., the subject should be kept 'blind1 regarding exposure. To avoid
unintentional bias in the handling of subjects or making unintentional suggestions to the
subjects, experimenters who deal directly with subjects also should be blind. All subjective
scoring of data should be performed blindly. An experiment in which both subjects and
5 experimenters were blind is called 'double blind.' When only subjects are blind, a single-
blind condition is said to exist. When no blinding was used, the study will be called
nonblind.
The other technical aspect of an experiment that was included in the summary tables
involves the statistical significance test methodology that was employed. When many
10 individual tests are conducted on the same data set the probability is increased that at least one
of them will be significant by chance alone. Thus, the 'experiment-wise' Type I error rate is
increased (Muller et al., 1984). Studies in which data were analyzed in the above manner
will have an increased probability of reporting a significant effect even when no effect exists
in the population. This bias toward significant effect worsens with the number of tests
15 conducted. Statistical methods exist that can be used to test multiple hypotheses for each data
set without increased probability of false significant effects.
The two technical problems, blinding and statistical methods, are noted in the summary
tables by use of the following code letters in the column labeled 'Technical Critique. 1' If
nothing appears in the Technical Critique column, the experiment was conducted double-
20 blind and multiple significance testing was not done. The following list defines the code.
A - No or unspecified statistical test
B - Multiple-significance tests on the same data set
C - Single-blind study
25 D - Nonblind study
Examination of the summary tables reveals that many of the same references appear
repeatedly. These were instances in which the experimenters measured several variables in
the same study. Thus, the reader should recognize that the number of experiments reported in
30 the literature is considerably smaller than the number of summary table entries.
March 12, 1990 10-104 DRAFT-DO NOT QUOTE OR CITE
-------
10.4.2.2 Sensory Effects
Vision
Absolute Threshold. Studies of the effects of COHb on absolute visual threshold are
summarized in Table 10-9. In an experiment using four well-trained young subjects it was
5 demonstrated that visual sensitivity was decreased in a dose-related manner by COHb levels
of 4.5, 9.4, 15.8, and 19.7% (McFarland et al., 1944). Various aspects of these data were
subsequently reported by Halperin et al. (1959) and McFarland (1970). COHb elevations
were accomplished by inhalation of boluses of high-concentration CO. Visual thresholds were
measured repeatedly over a five-minute period at each COHb level. Experimenters were not
10 blind to the exposure conditions and the subjects could have easily deduced the conditions
from the experimental design, because no air-only condition was included to control for the
effects of the testing scheme itself. Data from only one typical subject were presented.
Thresholds were measured at only one level of dark adaptation (0.002 foot candles).
The McFarland et al. (1944) study (above) stands in disagreement with several other
15 studies of absolute threshold effects of COHb. An early study of dark adaptation was
reported by Abramson and Heyman (1944) in which effects were inconsistent and not
statistically significant. Nine subjects were tested and the COHb level ranged up to 30%.
Documentation of the study was, however, very sparse so that it was difficult to consider the
study critically but the power was apparently quite low. McFarland (1973), in a scantly
20 documented article, reported that similar threshold shifts occurred at the end of a CO-
exposure period (17% COHb) and an air-only session. Thus it is possible that the effects
reported by McFarland et al. (1944) were due to fatigue or some other time-on-task related
variable. Von Restorff and Hebisch (1988) found no dark adaptation effects on subjects with
COHb levels ranging from 9 to 17%. Luria and McKay (1979) found no effect of 9% COHb
25 on scotopic visual threshold.
The effect of 17% COHb (bolus administration, followed by maintenance CO level for
135 min) on the entire dark adaptation curve was studied by Hudnell and Benignus (1989)
using 21 young men in a double-blind study. No difference between CO and air groups was
observed. A power of 0.7 was calculated for the test employed so that the conclusions are
30 reasonably defensible. From the above evidence, it appears that if COHb elevation affects
visual sensitivity, it remains to be demonstrated.
March 12, 1990 10-105 DRAFT-DO NOT QUOTE OR CITE
-------
o
tr
TABLE 10-9. EFFECTS OF COHb ON ABSOLUTE VISUAL THRESHOLD
o
o
o
o
cj
O
»
n
Exposure
Duration,
min
7
Bolus
18
Elevated
COHb Range
% n
10.0 - 30.0 9
17.0 21
+ 135
9.0 18
CO
Effect Comment
No Stimuli, methods not well specified.
No statistics given. Only one subject
at 10% COHb.
No Power for McFarland-size effect was ca.
0.7. Entire dark adaptation curve was
measured. Stimulus was 10.0 deg visual
angle p31 phosphor CRT with neutral
density filters.
No Measured only scotopic sensitivity.
Technical
Critique* Reference
A,D Abramson
and Heyman (1944)
Hudnell and
Benignus (1989)
B,C Luria and
McKay (1979)
0.5
4.5 - 19.7
Various
6.0 - 17.0
Bolus
9.0 17.0
+60
4 Yes No control group for air only. Only
typical data for one subject. Tested only
at adaptation to 0.002 ft candle. Effect
was COHb-ordinal, beginning at ca. 5%.
27 No Smokers and nonsmokers tested, n not
given. Few methods, specifications, or
statistics. Minutes of exposure (not
given) adjusted to target COHb.
5 No Dependent variables were time to
adaptation and sensitivity after
adaptation.
A,C
A.C
McFarland et al.
(1944); Halperin
et al. (1959)
McFarland (1970)
McFarland (1973)
Von Restorff
and Hebisch (1988)
"Technical problems: A=No or unspecified statistical tests; B = Multiple-significance tests on the same data; C=Single-blind study; D=Nonblind study. If
no technical problems are noted, the experiment was conducted under double-blind conditions and multiple-significance testing was not done.
-------
Temporal Resolution. The temporal resolution of the visual system has been studied in
the form of critical flicker fusion (CFF). In the CFF paradigm, subjects report the frequency
at which light flashes begin to appear as a continuous light. Studies of the effects of COHb
on CFF are summarized in Table 10-10.
5 Seppanen et al. (1977) reported dose-ordinal decreases of CFF for COHb values of ca.
4.0, 6.1, 8.4, 10.7, and 12.7%. The experiment was conducted with 22 healthy subjects
whose age ranged from 20 to 62 yrs. COHb was induced by breathing high concentrations of
CO from a Douglas bag. Subjects were blind as to the condition but apparently
experimenters were informed. Appropriate controls for fatigue were included and the
10 exposure levels were randomized.
A study was reported by Von Post-Lingen (1964) in which COHb levels ranged up to
23%. COHb was induced in 100 subjects by breathing CO-contaminated air from a
spirometer for about seven minutes in a single-blind procedure. One group of subjects was
given an injection of Evipan (sodium hexobarbitone, see Reynolds, 1982), a drug previously
15 shown to have produced decreases in CFF only if patients had demonstrable brain damage.
In the nondrug group, CFF was unaffected until ca. 14% COHb. In the drug group,
however, effects began at COHb levels as low as 6% and were dose proportional up to the
highest COHb value. When the drug + CO study was repeated in a small (n=15)
double-blind replication, no effects were seen. The latter replication study was given only
20 one paragraph in the report and thus it is not clear exactly what was done.
Beard and Grandstaff (1970) reported significant effects on CFF in an earlier study in
which four subjects had been exposed to CO level of 50, 150, or 250 ppm for one hour.
COHb was estimated to have reached 3.0, 5.0, and 7.5%, respectively, by the end of the
exposure. Documentation was extremely sparse and with only four subjects, power was
25 probably low. Even though the elevated COHb groups had decreased CFF, the results were
not dose ordinal. There is a comparatively large literature published before the Seppanen
et al. (1977) article, in which the effect of elevated COHb on CFF was tested. In none of the
earlier studies was CFF found to be affected, even though much higher levels of COHb were
tried. The studies and their maximum COHb levels were Fodor and Winneke (1972) -
30 7.5%, Guest et al. (1970) - 8.9%, Lilienthal and Fugitt (1946) - 15.4%, O'Donnell et al.
March 12, 1990 10-107 DRAFT-DO NOT QUOTE OR CITE
-------
TA!til040s OP eOHb ON
Ikpiure
Buration, 6§HI Range
Bin %
Comment
Te§hai§al
Refereaee
lelus
i
41
1
11,§«17,§
1,9 --11,7
7,1-11,1
f,0 - 11,7
7J*17J
4 Yes
II No
No
i Ne
m
17
No
Test speiifieatioa, methods, aad statisties
net given, Iffetts disgfdinal in
estimated from ereath sample,
Few test spe@ifleatiens, m\y grand name sf
test devige, gQHk estimated frem exposure by
eriginal authers,
Tested with red ae§8 limp, 0.7 deg visual
viewed binaeularly, Luntinanee net
Tested with nesn lamp, 1 deg visual angle!
viewed mgneeularty; Luminanee ml given
Minutes 3f exposure adjusted t§ target
Tested with red light, 1J deg visual angle,
viewed 6ino@ularly; Lumiaanee not given,
Tested in neisy eavir§ament alter 69 exposure
No speeiflgations for 6FF test,
Tested with white light, IJ deg visual angle,
apparently viewed toeularlv; l
-------
•us
o
S
IS
oe
(XQ
as
£
55
tC
fjP W
as 1Z.-S «=•£ ffi
3K S. S 'f a *
in
os
ar
us:
S'.JJ
*3
-
« gj« SB «l
S _
S "*
55 £ ffi
2 ~
..an.
£
^^ fR •'IB"*
3i;S
Ha|8
M Jab ueu
-------
(1971a) - 12.7%, Ramsey (1973) -11.2%, Vollmer et al. (1946) - 17.5%, and Winneke
(1974) - 10.0%. To be sure, there was much variation in size of the subject group, method,
and experimental design among the above studies, but no pattern emerges as to why the
Seppanen et al. (1977), Beard and Grandstaff (1970), and Von Post-Lingen (1964) studies
5 found significant effects when the others did not. It is noteworthy that the studies reporting
significant effects were all conducted in a single- or nonblind manner.
Miscellaneous Visual Functions. A number of researchers reported the results of
experiments in which visual parameters other than absolute threshold or CFF were measured
10 as part of a battery of tests. Many of these experiments studied a large group of subjects.
Table 10-11 summarizes these studies.
Beard and Grandstaff (1970) reported a study in which four subjects were exposed to
CO sufficient to produce estimated COHb levels of 3.0, 5.0, and 7.5%. The measurements
made were CFF (see above), brightness-difference thresholds, visual acuity, and absolute
15 threshold. Data for the latter variable were unreliable and not reported. Dose-related
impairments in acuity and brightness-difference sensitivity were reported. The scant
documentation of methods, plus the few subjects, make the results difficult to evaluate.
Five other reports of significant visual function effects by COHb elevation are extant.
Two of the studies (Bender et al., 1972; and Fodor and Winneke, 1972) reported that
20 tachistoscopic pattern detection was impaired by COHb levels of 7.3 and 5.3%, respectively.
Weir et al. (1973), Ramsey (1972), and Salvatore (1974) reported that brightness
discrimination was adversely affected by COHb levels of 6 to 20%.
Tests of visual function after COHb elevation conducted by other authors have been
uniformly nonsignificant. Table 10-11 summarizes these data. Especially noteworthy are
25 studies by Hudnell and Benignus (1989) and Stewart et al. (1972) both of which found no
acuity effects as reported by Beard and Grandstaff (1970). Brightness discrimination was
similarly not found to be affected (Ramsey, 1973), in contradiction with the reports of others.
The latter study is especially interesting in that it represents a failure to replicate an earlier
study by the same author (Ramsey, 1972). The first of the pair of studies by Ramsey was
30 conducted in a single-blind manner, the second was double blind.
March 12, 1990 10-110 DRAFT-DO NOT QUOTE OR CITE
-------
2 TABLE 10-11. EFFECTS OF COHb ON MISCELLANEOUS VISUAL FUNCTIONS
K
s-
i_> Exposure Elevated
1° Duration, COHb Range
i_* min %
0 60 1.8-6.7b
150 7.3°
780 5.3C
60 27.0-41.0
9
£ Bolus 17.0
Various 6.0 - 17.0
O
i£
H 45 5.0
6
^ 45 7.6-11.2
2!
s
*Q 390 5.4
O
W 30 4.0 - 12.7
Q
Dependent CO
n Variable Effect
4 Brightness Yes
42 Pattern Yes
12 Pattern Yes
5 See No
comment
21 Acuity and No
motion
27 Peripheral No
vision %
20 Brightness Yes
and depth
60 Brightness No
and depth
6 Brightness Yes
discrimination
22 Perceptual No
speed
Technical
Comment Critique* Reference
Test specification, methods, and statistics A,D Beard and Grandstaff (1970)
not given. Effects and acuity disordinal
in COHb (both variables). COHb estimated
from breath sample.
Pattern displayed for unspecified short B,C Bender et al. (1972)
time. COHb estimated from breath sample.
Pattern displayed for 0.1 s. COHb B.C Fodor and Winneke (1972)
estimated by original authors from
exposure.
Tested detection of dim objects in glare A,D Forbes et al. (1937)
and approach/recession comments of
objects. No specifications or statistics
given.
Stimulus was x x deg p 31 phosphor CRT. Hudnell and Benignus (1989)
Tested both photopic + 135 and scotopic.
Few methods, specifications, or statistics. A,C McFariand (1973)
Minutes of exposure (not given), adjusted
to target COHb.
No specification for tests. Only B,C Ramsey (1972)
brightness discrimination affected.
No specifications for tests. Results did B Ramsey (1973)
not support previous study by Ramsey
(1972).
None. D Salvatore (1974)
Test not well specified. B,C Seppanen et al. (1977)
n
-------
I
h— *
O
8
TABLE 10-11 (cont'd). EFFECTS OF COHb ON MISCELLANEOUS VISUAL FUNCTIONS
H1 '
O
5
o
K>
?
6
0
25
o
H
0
Exposure
Duration,
min
Var up
to 1440
150-
300
5
90-120
Bolus
Elevated
COHb Range
%
Continuous
distribution
Continuous
distribution
up to 20.0
7.5 - 17.5
2.0
5.6°
n
11
27
17
15
50
Dependent CO
Variable Effect
See No
comment
Defect No
detection
Red field No
size
Brightness Yes
discrimination
See No
comment
Technical
Comment Critique. Reference
Little documentation. Tested acuity, B Stewart et al. (1970)
depth, color, and phoria using clinical
instruments.
Subject inspected small parts. Not Stewart et al. (1972)
well specified.
Tested with perimeter bar and red B,C Vollmer et al. (1946)
sample patch.
Tested intensity matching with red, green, B Weir et al. (1973)
and white.
Poor documentation. Tested target detection B Wright et al. (1973)
in "dim" light, during glare, recovery after
glare, and depth.
Technical problems: A=No or unspecified statistical tests; B= Multiple-significance tests on the same data; C= Single-blind study; D=Nonblind study. If no technical problems are noted, Ihe
experiment was conducted under double-blind conditions and multiple-significance testing was not done.
''Values of COHb were not reported
The original
by the original
authors estimated COHb from expired
authors. Values given in the table
air.
were estimated by the present author using exposure parameters and the method of Cobum et al. (1965).
o
-------
The most thorough modern tests of visual function was performed by Hudnell and
Benignus (1989) who tested absolute threshold (see above), acuity, and motion detection with
COHb levels of 17% and found no effects due to COHb. The acuity and motion detection
were tested at both scotopic and photopic levels.
5 It would appear that the results of studies of the effects of COHb elevation on
miscellaneous visual function are not supportive of significant effects. Results that were
significant in two studies (Beard and Grandstaff, 1970; Weir et al., 1973) were contradicted
by other reports using relatively large groups of subjects. One author (Ramsey, 1972, 1973)
failed to confirm his own findings.
10
Audition
Surprisingly little work has been done concerning the effects of COHb on auditory
processes. Table 10-12 is a summary of the studies. Stewart et al. (1970) reported that the
audiogram of subjects exposed to as high as 12.0% COHb was not affected. Haider et al.
15 (1976) exposed subjects to a 105 dB, one-octave bandwidth random noise (center frequency of
2 kHz) for 15 min while COHb level was elevated to 13%. Under continued COHb
elevation, the temporary threshold shifts (ITS) were measured after noise cessation. No
effects of COHb on TTS was observed. Guest et al. (1970) tested the effects of elevated
COHb (8.9%) on auditory flutter fusion and found no significant effect. The flutter fusion
20 test is analogous to CFF in vision and was tested by having the subject judge the rate at which
an interrupted white noise became apparently continuous. From these data it would appear
that the functioning of the auditory system is not comparatively sensitive to COHb elevation,
but little research has been done.
25 10.4.2.3 Motor and Sensorimotor Performance
Fine Motor Skills
Bender et al. (1972) found that manual dexterity and precision (Purdure pegboard) were
impaired by 7% COHb. Winneke (1974) reported that hand steadiness was affected by 10%
COHb, but no supportive statistical test was presented.
30 Similar motor functions were evaluated by a number of other investigators and found
not to be affected, even at higher COHb levels. Table 10-13 summarizes the literature.
March 12, 1990 10-113 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 10-12. EFFECTS OF COHb ON MISCELLANEOUS AUDITORY FUNCTIONS
Exposure Elevated
Duration, COHb Range
min %
65 8.9
H-»
-------
TABLE 10-13. EFFECTS OF COHb ON FINE MOTOR SKILLS
K
o
to
v^5
10
o
>— »
O
i— »
>— k
L/l
O
>
H
i
0
0
o
H
*8
g
a
I.AJ
o
Exposure
Duration,
min
150
780
150
180
30
Varup
to 480
Varup
to 1440
up to 20.0
5
300
90 - 120
Bolus
Technical
exnerimer
Elevated
COHb Range
%
7.3"
5.3b
5.5
3.0- 12.4
4.0 - 12.7
Continuous
distribution
up to 12.0
Continuous
distribution
7.5 - 17.5
10.0b
20.0
5.6b
problems: A=No or unspecified
it was conducted under double-blii
Dependent
n Variable
42 Tapping
and
pegboard
12 Tapping,
pegboard,
and
steadiness
16 Tapping
9 Ataxia
22 Tapping
11 See
comments
27 See
comment
17 Postural
stability
18 See
comments
15 See
comment
50 Steadiness
CO
Effect
Yes
No
No
No
No
No
No
No
Yes
No
No
statistical tests; B=Multiple-significance
id conditions and multiDle-siffnificance te
Technical
Comment Critique*
Some aspect of each task declared affected. B,C
COHb estimated from breath sample.
COHb estimated by original authors from B,C
exposure.
Tested tapping with and without simultaneous C
arithmetic task.
No data above 6.6% COHb shown, only results B
of significance tests. Noisy environment.
None. B,C
Little documentation of procedure or results. B
Tested collar/pin, screw, Flanagan
coordination, tapping, and hand steadiness.
Tested collar/pin, spiral drawing, and hand
steadiness.
Tested both eyes open and closed. B,C
Tested tapping, steadiness, and Purdue hand A,C
precision. Only steadiness declared affected.
Tested tapping, star tracing, and rail walking. B
None. B
tests on the same data; C=Single-bIind study; D=Nonblind study. If no
stine was not done.
Reference
Bender et al. (1972)
Fodor and Winneke (1972)
Mihevic et al. (1983)
O'Donnell (1971b)
Seppanen et al. (1977)
Stewart et al. (1970)
Stewart et al. (1972)
Vollmer et al. (1946)
Winneke (1974)
Weir et al. (1973)
Wright et al. (1973)
technical problems are noted, the
o
Originial authors estimated COHb from expired air.
-------
Vollmer et al. (1946) reported that 20% COHb did not affect postural stability. O'Donnell
et al. (1971b) used the Pensacola Ataxia Battery to measure various aspects of locomotion and
postural stability. Subjects with 6.6% COHb were not affected. Stewart et al. (1970, 1972)
tested the ability of subjects to manipulate small parts using the Crawford collar and pin test
5 and screw test, the AAA hand steadiness test and the Flanagan coordination test. COHb
levels up to 15% had no effect on any of the measures. Two subjects were taken to 33% and
40% COHb, however, and in these subjects, the collar and pin performance was impaired and
the subjects reported hand fatigue. Manual dexterity (Purdue Pegboard), rapid precision
movement (Purdue hand precision), and static hand steadiness (pen in hole) and tapping tests
10 were not affected by COHb levels of ca. 5.3% (Fodor and Winneke, 1972). Wright et al.
(1973) reported that hand steadiness was not affected by COHb levels of 5.6%. Weir et al.
(1973) found no effects of 14% COHb on tapping, star tracing, and rail walking. Mihevic
et al. (1983) discovered no effect on tapping when the task was performed alone or
simultaneously with an arithmetic task. Finally, Seppannen et al. (1977) demonstrated that
15 tapping speed was unaffected by 12.7% COHb. Most of the above nonsignificant studies
used a moderate-to-large number of subjects. The overwhelming evidence in the area of fine
motor control indicates that COHb levels below ca. 20% (the highest level tested) do not
produce effects.
20 Reaction Time
Table 10-14 summarizes the literature with respect to the effects of elevated COHb on
reaction time. Of the 11 experiments that studied reaction time, only one reported a
significant result (Weir et al., 1973), and that effect occurred only at 20% COHb. A number
of the nonsignificant effects were from studies using a large number of subjects. The
25 pervasive finding that COHb elevation does not affect reaction time is especially impressive
because of the wide range of COHb levels employed (5.3 to 27.8%).
Tracking
Tracking is a special form of fine motor behavior and hand-eye coordination that
30 requires a subject to either follow a moving target or compensate for a moving target's
motion by manipulation of a lever, for example. The literature on tracking is summarized in
March 12, 1990 10-116 DRAFT-DO NOT QUOTE OR CITE
-------
1
fc
O
u
^•j
w
§
a
i
u
s
1
I— 1
6
T"H
H
Reference
11
H O
g
o
CJ
8l
w
£
c
U)
c
15
-> MM *JQ
Jj ^M Q^
«0
2 B"
| -2 B
|| B
|
JS
Fodor and Wi
0
«
1
•'§• •
° §
COHb estimated
authors from exf
o
JS o
9- •-
111
~^
CO
1
f^1 So
C*l OO
vv CTv
^* *— i
•a -a
1$ IS
1 1
£ =
q
<
—
u
2 H v. % o
o o
Z Z
u u
1 1
C/5 O
NO to
'q
o
ts >o
§ 1
ON
C-
& p
s s;
" C
^ T>
•a s
S -G
•s %
3 2
u o
« <
ca *^ ^>
u O u
•^ C W)
'I "g S
« 5 2
b o -0
a X*l
•g15 t
4> w
1 ill
< 1
JH u 8
!,-§ 1
1 S •§ u
2 R
q
i
q q
Ot tO
CO
S
2 £
R
ON
•a
.b
O
cd
CO
O
«"
a
CQ
1
1
o —
S 8
<
CB J) >,
2 5.3
S 2 '3
.S S Sf d
•o * " 3
2 j< T3 -3
ss-s s
H H a a
sr
o E
0 O 0
2 2 Z
O U bo
i
o
o
'
-------
I
to
d
O
O
TABLE 10-14 (cont'd). EFFECTS OF COHb ON REACTION TIME
Exposure Elevated
Duration, COHb Range
min %
300 10.0C
O
,L
°° Bolus 5.6C
CO
n Type Effect Comment
18 Simple No COHb estimated from exposure
and by original authors.
choice
50 Simple No None.
Technical
Critique* Reference
A,C Winneke (1974)
B Wright et al. (1973)
Technical problems: A = No or unspecified statistical tests; B = Multiple-significance tests on the same data; C = Single-blind study; D = Nonblind study. If no technical problems are noted, the
experiment was conducted under double-blind conditions and multiple-significance testing was not done.
bOriginal authors estimated COHb from expired air.
°COHb estimated from exposure by original authors.
8
n
-------
Table 10-15. Of the 11 studies on the topic, four reported significant effects and one of those
found effects only at 20% COHb. The matter is more complicated, however, and the
literature in the area offers some clues to the reasons for the diversity among the reports.
O'Donnell et al. (1971a,b) used critical instability compensatory tracking in which the
5 task was to keep a meter needle centered. Simultaneous performance of detection tasks also
was required in one of the studies. No effects were demonstrated for COHb levels of
12 to 13%. The critical instability tracking task also was used by Gliner et al. (1983) in
conjunction with peripheral light detection. COHb levels up to 5.8% had no effect on
performance. Pursuit rotor tracking also was reported to be unaffected at 5.3% COHB
10 (Fodor and Winneke, 1972). Weir et al. (1973) reported that pursuit rotor performance was
slightly affected beginning at 20% COHb. In a 1988 study, Bunnell and Horvath used a two-
dimensional tracking task in which the stimulus was presented on a CRT and controlled with a
joystick. No effect of COHb or exercise or combination of the two was seen for COHb
levels up to 10.2%. Schaad et al. (1983) reported that pursuit and compensatory tracking
15 were not affected by COHb of 20% even during simultaneous performance of monitoring
tasks.
In a pair of careful studies of different design, Putz et al. (1976, 1979) studied
compensatory tracking by having the subject try to keep a vertically moving spot in the center
of an oscilloscope screen. The tracking was performed while simultaneously performing a
20 light-brightness detection task. In both studies, tracking was significantly affected by COHb
levels of 5%. The fact that both studies demonstrated significant results despite differences in
experimental design lends credibility to the finding. Additional credibility was gained when
the Putz et al. (1976) study was replicated with similar results by Benignus et al. (1987).
The consistency of the compensatory tracking results in the Putz et al. (1976) protocol was
25 not continued when Benignus et al. (1989a) attempted to demonstrate a dose-effect
relationship using the same experimental design. In the latter study, independent groups were
exposed to CO sufficient to produce COHb levels of control, 5, 12, and 17% COHb. CO
was administered via Douglas bag breathing and then COHb was maintained by low-level CO
in room air. A fifth group was exposed to CO in the chamber only and this group served as a
30 positive control because it was treated in exactly the same ways as the subjects in Putz et al.
(1976) and in Beningus et al. (1987). No significant effects were demonstrated on tracking in
March 12, 1990 10-119 DRAFT-DO NOT QUOTE OR CITE
-------
I
^ J
1 — »
JsJ
£
0
9
H-ft
to
O
•art
£
?
O'
O
g
s
O
d
tn
^^
s
O
TABLE 10-15. EFFECTS OF COHb ON TRACKING
Exposure Elevated
Duration, COHb Range
min %
240 8.2
Bolus + 5.6 - 17.0
240
Bolus + 7.0 - 10.0
55
780 5.3b
150 5.8
540 5.9 - 12.7
180 3.0 - 12.4
240 3.0-5.1
240 3.5 - 4.6
CO
n Type Effect
22 Compensatory Yes
74 Compensatory No
15 Compensatory No
12 Pursuit No
rotor
15 Compensatory No
4 Compensatory No
9 Compensatory No
30 Compensatory Yes
30 Compensatory Yes
Comment
CRT display with simultaneous
monitoring, same task as Putz
et al. (1976, 1979).
CRT display with simultaneous
monitoring, same task as Putz
et al. (1976, 1979).
Two-dimensional tracking test
performed for 9 min using
CRT and joystick.
COHb estimated from exposure
by original authors.
Critical instability method
using a line on a CRT screen.
Task was to keep meter needle
centered while monitoring
other meters and lights.
Tested in noisy environment.
CO exposure during sleep.
Critical instability method
using a meter needle
centering. Noisy environment.
CRT display with simultaneous
monitoring. Significant at
5.1% COHb only.
CRT display with simultaneous
monitoring. Significant at
4.6% COHb only.
Technical
Critique* Reference
Benignus et al. (1987)
Benignu. et al. (1989a)
C Bunnell and Horvath (1988)
B,C Fodor and Winneke (1972)
C Gliner et al. (1983)
B O'Donnell et al. (1971a)
B O'Donnell et al. (1971b)
Putz et al. (1976)
Putz et al. (1979)
-------
***
5
6
o
TABLE 10-15 (cont'd). EFFECTS OF COHb ON TRACKING
Exposure Elevated
Duration, COHb Range
min %
270 20.0
90 - 120 7.0 - 20.0
n
10
15-
25
CO
Type Effect
Pursuit and No
compensatory
Pursuit Yes
Comment
Light-monitoring and arithmetic
tasks performed simultaneously.
No consistent effects until
20% COHb.
Technical
Critique* Reference
B,D Schaad et at. (1983)
B Weir et al. (1973)
Technical problems: A=No or unspecified statistical tests; B=Multiple-significance tests on the same data; C=Single-blind study; D = Nonblind study. If no technical problems are noted, the
experiment was conducted under double-blind conditions and multiple-significance testing was not done.
Original authors estimated COHb from expired air.
O
c!
I
O
W
r>
-------
any group. The means for the tracking error were elevated in a nearly dose-ordinal manner
but not to a statistically significant extent.
At present there is no apparent reason for the lack of consistency among the reports of
tracking performance. The largest study, with the widest dose range (Benignus et al., 1989a),
5 appears to be the strongest indicator of no significant effects of COHb elevation. However, it
is difficult to ignore the several other studies which were controlled well and did demonstrate
significant effects. At this point the best summary seems to be that COHb elevation produces
small decrements in tracking which are sometimes significant. The possible reasons for such
high variability are unclear. Benignus et al. (1989b) discussed the issues in a speculative
10 manner. The latter article will be reviewed later in the present document.
10.4.2.4 Vigilance
A dependent variable, which is possibly affected by elevated COHb, is the performance
of extended, low-demand tasks characterized as vigilance tasks. Because of the low-demand
15 characteristic of vigilance tasks, they are always of a single-task type. Table 10-16 is a
summary of the literature on the subject. Of the eight reports, four reported significant
effects. Despite the seemingly greater unanimity in this area, it is noteworthy that for each
report of significant effects, there exists a failed attempt at direct replication.
Horvath et al. (1971) reported a significant vigilance effect at 6.6% COHb. A second
20 study, conducted in the same laboratory (Christensen et al., 1977), failed to find significant
effects of 4.8% COHb on the same task. To be sure, the second study used slightly lower
COHb levels, but the means left no suggestion of an effect. Roche et al. (1981) reported that
performance of the same task using bolus exposure to produce 5% COHb was not affected.
Fodor and Winneke (1972) reported a study in which 5.3% COHb significantly impaired
25 performance of a vigilance task. The same task and protocol were tried again in the same
laboratory (Winneke, 1974). In the replication attempt, no significant effects were found,
even for COHb levels up to 10%.
Groll-Knapp et al. (1972) reported dose-related significant effects of COHb ranging
from estimated values of 3 to 7.6%. Effects were large but apparently the study was not
30 blind. Haider et al. (1976) reported similar effects at low COHb levels but not at higher
levels. The authors have mentioned twice failures to replicate the results (Haider et al., 1976;
March 12, 1990 10-122 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 10-16. EFFECTS OF COHb ON VIGILANCE
O
1— *
1°
1— «
O
9
NJ
O
§
H
6
O
Exposure
Duration,
min
120
780
120
210
120 - 240
135
Bolus
+60
300
Technical
Elevated
COHb Range
% n
4.8 10
5.3b 12
3.0 - 7.6b 20
6.0 - 12.0 20
3.0-13.0 20
2.3 - 6.6 15
5.0 18
10.0° 18
CO Technical
Type Effect Comment Critique* Reference
Light No None. Christensen et al. (1977)
White Yes Effect disordinal in COHb. B,C Fodor and Winneke (1972)
noise COHb estimated from exposure
by original authors.
Tone Yes Effects were significant at A,D Groll-Knapp et al. (1972)
3 % COHb and increased with
dose.
Click No None. B Groll-Knapp et al. (1978)
Tone Yes Two experiments were reported, A Haider et al. (1976)
one gave effect at ca. 7.6%
COHb and the other gave no
effect at 13%. No data
presented, only conclusions.
Light Yes No effect at 2.356 COHb. C Horvath et al. (1971)
Light No None. Roche et al. (1981)
White No COHb estimated from exposure A,C Winneke (1974)
noise by original authors.
problems: A=No or unspecified statistical tests; B=MuItiple-significance tests on the same data; C= Single-blind study; D= Nonblind study. If no technical problems are noted, the
*y experiment was conducted under double-blind conditions
A '"Original authors estimated COHb from expired air.
and multiple-significance testing was not done.
t-3 cCOHb estimated from exposure by original authors.
O
n
-------
Groll-Knapp et al., 1978). A similar experiment using a different stimulus failed to produce
significant effects at 12% COHb (Groll-Knapp et al., 1978).
The fact that all replication attempts for each of the reported significant effects of COHb
on vigilance have failed to verify the original reports is evidence for some unreliability or the
5 operation of unknown and uncontrolled variables. That the non-verifications were conducted
by the original researchers, as well as by others, makes the case for unreliability even more
binding. If vigilance is affected by COHb elevation, a convincing demonstration remains to
be made. Perhaps a case may be made for small effects similar to the argument advanced in
Benignus et al. (1989b).
10
10.4.2.5 Miscellaneous Measures of Performance
Continuous Performance
Continuous performance is a category of behavior which is related to vigilance. The
difference is that many tasks that are performed over a long period of time are more
15 demanding and involve more than simple vigilance. Sometimes the continuous performance
tasks are not performed for a sufficiently long period of time to involve decrements in
vigilance or are interrupted too frequently. Table 10-17 is a summary of the literature
regarding the effects of COHb on continuous performance.
Putz et al. (1976, 1979) reported that monitoring performed simultaneously with
20 tracking was impaired at COHb values as low as 5%. In a replication attempt of the Putz
et al. studies, Benignus et al. (1987) failed to find any effects of ca. 8% COHb. O'Donnell
et al. (1971a) also failed to find effects of COHb on a monitoring task performed
simultaneously with tracking. Schaad et al. (1983) found no effects on monitoring
simultaneously with tracking even when COHb was 20%. Gliner et al. (1983) reported that
25 signal detection was affected by 5.8% COHb when performed singly, but not when performed
simultaneously with tracking. The latter results are in conflict with Putz et al. (1976, 1979).
Isogna and Warren (1984) reported that the total game score on the performance of a
multi-task video game was reduced by COHb levels of 4.2%. Separate task scores were not
collected. Schulte (1963) reported that letter, word, and color detection tasks were dose-
30 ordinally impaired by COHb levels of as low as 5% and ranging up to 20%. Reported COHb
March 12, 1990 10-124 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 10-17. EFFECTS OF COHb ON CONTINUOUS PERFORMANCE
r>
sr
ts)
s
9
to
o
o
1
g
n
Exposure Elevated
Duration, COHb Range
min %
240 8.2
200 4.6 - 12.6
150 5.8
120 2.1 - 4.2
540 5.9 - 12.7
240 3.0 - 5.1
240 3.5 - 4.6
270 20.0
CO
n Type Effect
22 Light No
52 Numeric No
display
15 Light Yes
9 Complex Yes
target
detection
4 Meters No
and
lights
30 Light Yes
and
tone
tasks
30 Light Yes
and
tone
tasks
10 Light No
moni-
toring
Comment
Light monitoring simulta-
neously with tracking.
Task was to detect numerals
of unmatched parity in a
series of three.
Same task as Putz et al. (1976,
1979) but under one condition
performed without tracking.
Only the latter condition was
affected.
Video game. Targets tracked
and "shot down."
Monitoring meters and lights
while tracking.
Light monitoring simulta-
neously with tracking. Tone
monitoring as separate task.
Only light tasks affected. No
effect at 3.0% COHb.
Light monitoring simulta-
neously with tracking. Tone
monitoring as separate task.
Both light and tone task
affected. No effect at 3.5%
COHb.
Pefonned simultaneously with
tracking.
Technical
Critique* Reference
Benignus et al. (1987)
Benignus et al. (1977)
C Gliner et al. (1983)
C Insogna and Warren (1984)
B O'Donnell et al. (1971a)
Putz et al. (1976)
Putz et al. (1979)
B,D Schaad et al. (1983)
-------
o
9
to
i
O
2
O
c
TABLE 10-17 (cont'd). EFFECTS OF COHb ON CONTINUOUS PERFORMANCE
Exposure Elevated
Duration, COHb Range CO
min % n Type Effect
? 0 - 20.0 49 Letter, Yes
word,
and
color
detection
Comment
COHb values larger than expected
asymptotic value. All values
affected even for low COHb.
Technical
Critique" Reference
B,C Schulte (1963)
Technical problems: A=No or unspecified statistical tests; B=Multiple-significance tests on the same data; C=Single-blind study; D=Nonblind study. If no technical problems are noted, the
experiment was conducted under double-blind conditions and multiple-significance testing was not done.
s
n
-------
levels were at considerable variance with values expected from the exposure parameters
(Laties and Merigan, 1979). Benignus et al. (1977) reported that 8.2% COHb did not effect
a numeric monitoring task.
Again, there is disturbing lack of replicability in the literature. The two most credible
5 studies showing effects of COHb on continuous performance (Putz et al., 1976, 1979) were
not verified by Benignus et al. (1987). In the latter study, the tracking effects of the Putz
et al. work were verified. Similar studies of monitoring during tracking (O'Donnell et al.,
1971a; Schaad et al., 1983; Gliner et al., 1983) also reported no effects of COHb, even with
levels of up to 20%. It seems necessary to suspend judgment regarding the continuous
10 performance results until further data and understanding are available. Perhaps the best
judgment is to hypothesize small effects.
Time Estimation
In 1967 Beard and Wertheim reported that COHb produced a dose-related decrement in
15 single-task time estimation accuracy beginning at 2.7%. Various versions of the same task
were tested by others (see Table 10-18) with COHb levels ranging up to 20% without effects
being demonstrated (Stewart et al., 1970, 1972, 1973b; O'Donnell et al., 1971b; Weir and
Rockwell, 1973, Wright and Shephard, 1978b). An exact replication was conducted by Otto
et al. (1979) which also did not find significant results. It seems safe to assume that time
20 estimation is not affected by COHb elevation.
Cognitive Effects. Table 10-19 is a summary of the literature concerning the effects of
COHb elevation on the performance of cognitive tasks. Five of the 11 experiments that have
been reported found cognitive effects of COHb. Bender et al. (1972) reported effects of
7.3% COHb on a variety of tasks. Groll-Knapp et al. (1978) reported memory to be affected
25 after exposure to CO during sleep (11 % COHb), but a very similar study performed by the
same group later found no effects of 10% COHb (Groll-Knapp et al., 1982). Arithmetic
performance was affected slightly in a nondose-ordinal manner when a simultaneous tapping
task was performed (Mihevic et al., 1983). Schulte (1963) reported a dose-ordinal effect on
arithmetic performance beginning at 5% and ranging to 20% COHb. COHb levels in the
30 latter study were at considerable variance with expected values from the exposure parameters
(Laties and Merigan, 1979). Similar variables were tested by others, sometimes at higher
March 12, 1990 10-127 DRAFT-DO NOT QUOTE OR CITE
-------
g
1
-p
t— »
VO
o
o
t— k
to
oo
O
^
^
1
0
0
2J
9
O
a
9
w
0
TABLE 10-18. EFFECTS OF COHb ON TIME ESTIMATION
Exposure Elevated
Duration, COHb Range
min %
120 2.7 - 12.5b
180 3.0 - 12.4
120 3.7 - 7.8
Var up Continuous
to 1440 distribu-
tion up
to 12.0
150-300 Continuous
distribu-
tion up
to 20.0
90 20.0
2 2.0 - 8.0°
n
18
9
13
11
27
15
13
CO
Type Effect
Duration Yes
discrimination
Duration No
and lime
Duration No
discrimination
Duration No
discrimination
See No
comment
Duration No
estimation
Duration No
discrimination
"Technical problems: A=No or unspecified statistical tests; B= Multiple-significance tests on
Technical
Comment Critique*
Effects were COHb ordinal B,C
beginning at ca. 2.7% COHb.
Tone duration discrimination B
and time interval estimation.
Noisy environment.
Replication of Beard and
Wertheim (1967).
Tone duration compared to B
light duration.
Used duration discrimination,
time estimation, and Marquette
test.
Various tone duration judg- B
ments were used.
Results of three experiments. D
the same data; C = Single-blind study; D=Nonblind study.
Reference
Beard and Wertheim (1967)
O'Donnell et al. (1971b)
Otto et al. (1979)
Stewart et al. (1970)
Stewart et al. (1972, 1973b)
Weir et al. (1973)
Wright and Shephard (1978b)
If no technical problems are noted, the
experiment was conducted under double-blind conditions and multiple-significance testing was not done.
bCOHb values were not reported by the
°Original authors estimated COHb from
originial authors.
expired air.
Values given in the table were estimated by the present authors using exposure parameters and
the method of Coburn et al. (1965).
O
-------
TABLE 10-19. EFFECTS OF COHb ON MISCELLANEOUS COGNITIVE TASKS
i— » Exposure Elevated
J° Duration, COHb Range
h— > min %
NT5
VO v.
0 150 - 300 7.3b
>55 7.0 - 10.0
210 6.0 - 12.0
O 410 11. Oc
t— »
to
480 10.0
ffl 120-240 3.0-13.0
h-^
i
i
o 15° 5-5
.,
Q 540 5.9 - 12.7
O
H
tn 270 20.0
0
CO
n Task Effect
42 Digit span, Yes
nonsense syllables,
intelligence test
15 Memory, stroop Yes
test, visual
search, and
arithmetic
20 Arithmetic, No
nonsense syllables,
mood scale
10 Memory, mood Yes
20 Verbal learning No
and memory
20 Attention, memory, No
arithmetic.
16 Arithmetic Yes
4 Arithmetic No
10 Arithmetic No
Comment
Some aspect of each
declared affected. COHb
estimated from breath
sample.
Stroop test affected in
nondose-ordinal manner.
Visual search affected
interactively by CO and
exercise.
None.
Only memory declared
affected. Exposure
during sleep.
Tested before and after
exposure. Was exposed
during sleep.
None.
Effect was present only
during multiple-task
performance and was not
dose ordinal.
Tested in noisy
environment.
Performed simultaneously
with tracking.
Technical
Critique" Reference
B,C Bender et al. (1972)
C Bunnell and Horvath (1988)
B Groll-Knapp et al. (1978)
B Groll-Knapp et al. (1978)
B Groll-Knapp et al. (1982)
A Haider et al. (1976)
C Mihevic et al. (1983)
B O'Donnell et al. (1971a)
B,D Shaad et al. (1983)
n
-------
TABLE 10-19 (cont'd). EFFECTS OF COHb ON MISCELLANEOUS COGNITIVE TASKS
o
o
O
O
2,
Exposure Elevated
Duration, COHb Range
min %
? 0.0 - 20.0
150 - 300 Continuous
n Task
49 Arithmetic
27 Arithmetic
distribution
up to 20.0
CO
Effect Comment
Yes COHb values larger than
expected asymptotic value.
Significant effects even
at low COHb.
No None.
Technical
Critique" Reference
B,C Schulte (1963)
Stewart et al. (1972)
"A=No or unspecified statistical tests; B=Multiple-significance tests on the same data; C=Single-blind study; D=Nonblind study.
bOriginal authors estimated COHb from expired air.
"Original authors estimated COHb via Coburn, Forster, and Kane equation.
1
g
n
-------
levels of COHb and with relatively large groups of subjects, without finding effects
(O'Donnell et al., 1971a; Stewart et al., 1972; Haider et al., 1976; GroU-Knapp et al., 1978;
Schaad et al., 1983). The conclusions are, at best, equivocal.
A recent study by Bunnell and Horvath (1988) utilized a wide range of cognitive effects
5 involving short-term memory, manikin rotation, stroop word-color tests, visual search, and
arithmetic problems (the latter as part of a divided attention task performed simultaneously
with tracking). COHb was formed by bag breathing followed by a CO level in room air
designed to maintain a constant COHb level. Subjects were exercised at either 0, 35, or 60%
of VO2 max before cognitive tests were performed. The stroop test performance was slightly
10 but significantly affected by either 7 or 10% COHb by the same amount but exercise had no
effect. The authors suggested that negative transfer effects (difficultly in reversing
instructional sets) were responsible for the decrement. Visual searching improved for both
COHb levels at rest and at medium exercise. At the high exercise level, however, COHb
produced dose-ordinal impairments in performance. The authors conjectured that hypoxic
15 depression of cortical function interacted with hypoxic stress and exercise stress to produce
the effects.
Most of the data on cognitive effects of COHb elevation are not sufficiently consistent to
consider. The study by Bunnell and Horvath (1988), however, is suggestive of potentially
important effects of interactions of COHb and exercise. Before any conclusions may be
20 drawn about the results, the study must be expanded to unravel the mechanisms by which the
interactions were produced.
10.4.2.6 Automobile Driving
Complex behavior, in the form of automobile driving, has been tested a number of times
25 for effects of COHb elevation. Not only is automobile driving potentially more sensitive to
disruption because of its complexity, but it is an inherently interesting variable because of its
direct applicability to nonlaboratory situations. The well-practiced nature of the behavior, on
the other hand, may make performances more resistant to disruption. The complexity of the
behavior also leads to methodological difficulties. To exhaustively measure the complex
30 behaviors usually leads investigators to measure many dependent variables. Statistically
March 12, 1990 10-131 DRAFT-DO NOT QUOTE OR CITE
-------
analyzing a large number of variables in a defensible way requires many subjects and leads to
greater expense.
Table 10-20 is a summary of the studies of automobile driving as affected by COHb. In
an early study by Forbes et al. (1937), using only five subjects, steering accuracy in a
5 simulator was investigated with COHb levels of up to 27.8%. No effects were demonstrated.
A sparsely documented experiment by Wright et al. (1973), using 50 subjects with 5.6%
COHb, tested a number of functions of simulator performance but found no effects. Weir
et al. (1973) performed an experiment with actual automobile driving on a highway in which
a great number of variables were measured and tested. None of the variables were reliably
10 affected until COHb exceeded ca. 20%. Wright and Shephard (1978a) failed to find effects
of 7% COHb on driving. In the latter study, the authors reported effects but only after
misapplication of the chi-square test. The only effect of COHb on driving at a lower level
(7.6%) was reported by Rummo and Sarlanis (1974), who found that the ability to follow
another car at a fixed distance was impaired.
15 The difference between the experiments of Rummo and Sarlanis (1974) and Weir et al.
(1973) is troubling. Both measured following distance but only the experiment employing the
lower-level COHb found effects. If automobile driving is affected by COHb elevation, it
remains to be demonstrated in a conclusive manner.
20 10.4.2.7 Brain Electrical Activity
Electrical activity of the brain (see review by Benignus, 1984) offers the possibility of
testing the effects of COHb without the problem of selecting the most sensitive behavioral
dependent variable. It is less dependent upon subject cooperation and effort and may be a
more general screening method. The major disadvantage of the measures is the lack of
25 functional interpretability. The area has been plagued with poor quantification and,
frequently, a lack of statistical significance testing.
The electroencephalogram (EEG) is a recording of the continuous voltage fluctuations
emitted by the intact brain. The slow-evoked potential originally called the contingent
negative variation (CNV) is computed by averaging over trials and was linked to (among
30 other things) cognitive processes or expectancy (Donchin et al., 1977). The evoked potential
(EP) is the electrical activity in the brain resulting from sensory stimulation, either auditory
March 12, 1990 10-132 DRAFT-DO NOT QUOTE OR CITE
-------
tr
to
TABLE 10-20. EFFECTS OF COHb ON AUTOMOBILE DRIVING TASKS
8
Exposure
Duration,
min
Elevated
COHb Range
Task
CO
Effect
Comment
Technical
Critique*
Reference
60
20
27.0-41.0
7.6
Steering accuracy No
Steering wheel
reversals and
following
distance
Yes
In auto simulator
(unspecified).
Only following distance
was affected. Tested
in auto simulator.
A,D
B,C
Forbes et al. (1937)
Rummo and Sarlanis (1974)
9
UJ
O
O
90 - 120
Bolus
Bolus
7.0 - 20.0
5.6b
7.0
12
50
10
See comments
See comment
Brake-reaction
time
Yes
No
No
Instrumented automobile
driven on highway. Many
measures of control
functions plus measures
of driving stability and
information processing
load were utilized.
The latter was affected
at 20% but not reliably
below that. Authors
debated functional
importance of findings.
Poor documentation.
Tested automobile
simulator performance-
brake, accelerator,
steering, and signals.
None.
Weir et al. (1973)
Wright etal. (1973)
Wright and Shephard (1978a)
s
n
"Technical problems: A=No or unspecified statistical tests; B=MultipIe-significance tests on the same data; C=Single-blind study; D=Nonblind study. If no technical problems are noted, the
experiment was conducted under double-blind conditions and multiple-significance testing was not done.
''Original authors estimated COHb from expired air.
-------
(AEP) or visual (VEP). EEC, CNV, and EPs have been studied with COHb elevation.
Table 10-21 is a summary of results from these studies.
Groll-Knapp et al. (1972) reported that the CNV was decreased in amplitude in a dose-
related manner for COHb values ranging from 3 to 7.6%. In a second study, Groll-Knapp
5 et al. (1978) again reported that CNV amplitude was reduced by 12% COHb when subjects
missed a signal in a vigilance task. More evidence is required before the functional
significance of such an effect can be deduced but it is a potentially important finding.
Clinical EEGs were analyzed by visual inspection by Stewart et al. (1970, 1973a) and
Hosko (1970) after exposure to sufficient CO to produce COHb levels ranging up to 33%.
10 No effects were noticed. Groll-Knapp et al. (1978) reported similar results using spectrum
analysis on EEGs from subjects with 12% COHb. Haider et al. (1976) reported slight
changes in the EEG spectrum for COHb levels of 13%, but no tests of significance were
conducted. In view of the above studies, it seems reasonable to assume that no EEG effects
of COHb levels below at least 10% should be expected.
15 O'Donnell et al. (1971b) reported that sleep stages (as determined from the EEG) were
not distributed by COHb levels up to 12.4%. Groll-Knapp et al. (1978) and Haider et al.
(1976), however, both reported distributed sleep stages at similar COHb levels using EEG
spectra. Groll-Knapp et al. (1982) repeated their earlier study and found essentially the same
effects.
20 The VEP was consistently not affected by COHb elevation below ca. 22% and usually
the lowest level for effects was higher (see Table 10-21). At higher levels the effects were
dose-related (Dyer and Annau, 1977; Stewart et al., 1970; Hosko, 1970). The above is true
for both rats and man.
A single study of visual electrophysiology has reported low-level effects of COHb
25 (Ingenito and Durlacher, 1979). The electroretinogram (ERG) of anesthetized cats was
reported to have exhibited a reduced /3-wave amplitude beginning at 7.5%. Effects were
dose-ordinal up to 42% COHb. The contribution of the anesthesia (chloralose) to the effect
as a possible potentiator of COHb effects was not tested in the study. The authors reported
that the effect outlasted the COHb elevation and possibly was due to direct cellular CO
30 toxicity.
March 12, 1990 10-134 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 10-21. EFFECTS OF COHb ON BRAIN ELECTRICAL ACTIVITY
t— > Exposure
* Duration,
•— » min
vn
O 120
Injection
120
210
o
8
410
O
^
H 480
6
o
^ 420
g
H
O
§ 21°
H
W
0
W 120
n
M
Elevated
COHb Range Dependent CO
% n Variable Species Effect
6.0-55.0 15 VEP Rat Yes
10.0-75.0 10 Tone AEP Rat Yes
3.0 - 17.6b 20 CNV Human Yes
6.0 - 12.0 20 VEP, Human Yes
click
AEP,
CNV, and
EEG
spectra
11.0 10 Click, Human Yes
AEP, and
EEG sleep
stages
10.0 20 EEG sleep Human Yes
stages
12.0 20 Sleep Human Yes
stages
and EEG
spectra
5.3 55 VEP Human No
7.5 - 42.0 6 ERG Cat Yes
Comment
First significant effect increased amplitude
at 22% COHb in cortex (at 38% in superior
colliculus). Changes were dose related.
COHb-ordinal effects beginning at ca. 45%
in rat. Small effect possible near 25% when
CO was injected ip.
Significant differences at all COHb levels
above endogenous.
CNV only declared affected. No data
given, only conclusions stated.
Both affected.
Same conclusions as Groll-Knapp et al.
(1978)
Both changed slightly (no significance
test).
Both young (n = 33) and elderly (n = 22)
were tested. Mean age for young: 22.8, and
for old: 68.7.
Decreased /?-wave amplitude. Dose-related
effect beginning at 7.5% COHb.
Technical
Critique* Reference
Dyer and Annau (1977)
Fechter et al. (1987)
A,D Groll-Knapp et al. (1972)
B Groll-Knapp et al. (1978)
B Groll-Knapp et al. (1978)
B Groll-Knapp et al. (1982)
A Haider et al. (1976)
Harbin et al. (1988)
Ingenito and Durlacher (1979)
-------
TABLE 10-21 (cont'd). EFFECTS OF COHb ON BRAIN ELECTRICAL ACTIVITY
o
1— >
1°
I— I
o
0
0
6
0
1
o
1
o
n
tn
Exposure Elevated
Duration, COHb Range
min % n
18 9.0 18
180 3.0-12.4 9
Var up Continuous 1 1
to 1440 distribution
up to 33.0
Var 3.2 - 15.2 6
240 3.0-5.1 30
120 7.0 - 62.0C 6
Dependent CO
Variable Species Effect
VEP Human No
Sleep Human No
stages
VEP and Human Yes
EEC
VEP and Human No
EEC
Tone AEP Human Yes
VEP Rat Yes
Technical
Comment Critique* Reference
None. B,C Luna and McKay (1974)
No data above 6.6% COHb shown only results B O'Donnell et al. (1971b)
of signficance tests. Noisy environment.
No effects until COHb ca.2\%. Only 2 subjects B Stewart et al. (1970); Hosko (1970)
were tested in the high range. Effect was an
increased amplitude of peaks Nl, P2, and N2.
No statistical tests.
None. B,D Stewart et al. (1973a)
p-p Amplitude of N1-P1 peak increased in COHb- Putz et al. (1976)
ordinal manner beginning at 3%.
Increased amplitude at both cortex and superior Xintaras et al. (1966)
colliculus at 62% COHb. Late component amplitude
decreased at 7% COHb in superior colliculus. No
statistics, only typical data given.
Technical problems: A=No or unspecified statistical tests; B=Multiple-significance tests on the same data; C= Single-blind study; D=Nonblind study. If no technical problems are noted, the
experiment was conducted under double-blind conditions and multiple-significance testing was not done.
bCOHb was estimated by comparison to a separate series of animals.
°COHb was estimated from exposure by published data (Montgomery and Rubin, 1971) by present author.
i
-------
Groll-Knapp et al. (1978) found no effect of COHb (8.6%) on auditory (click)-evoked
potentials during waking, but reported increased positive-peak amplitudes when subjects were
tested during sleep at ca. 11 % COHb. The finding was verified by Groll-Knapp et al.
(1982). The fact that the data were collected during sleep is potentially important.
5 Putz et al. (1976) conducted a double-blind study in which 30 persons were exposed to
70 ppm CO for 240 min (5% COHb at the end of the session). Among other variables, the
auditory-evoked potential was measured. The peak-to-peak amplitude of the N1-P1
components was increased in a dose-ordinal manner beginning at ca. 3% COHb.
Ten millisecond tone bursts were used by Fechter et al. (1987) to produce AEPs in rats
10 exposed to graded doses of injected CO. COHb levels ranged up to 75%. Slight evaluations
of the mean threshold began at about 25% COHb. Effects were first seen at higher
frequencies. All effects were reversible. During the exposure, normal body temperature was
maintained to avoid hypothermia (Annau and Dyer, 1977).
Many of the brain electrical activity measures seem to be altered by COHb elevation.
15 The functional significance of these changes is not clear. Sometimes an alteration is not an
indication of a deleterious effect but merely implies some change in processing. When
induced by low levels of COHb, however, any change should be viewed as potentially
serious.
20 10.4.2.8 Schedule-Controlled Behavior
Because of the high levels of COHb that can be employed in studies of laboratory
animals, using schedule-controlled behavior, effects of COHb are reported in all articles on
the subject. Table 10-22 is a summary of the literature. There are a number of problems
with the published literature, however, as seen in Table 10-22. Only a few investigators
25 measured COHb; instead, they simply specified the exposure parameters. Another problem is
that of hypothermia, which occurs in rats when COHb levels rise (Annau and Dyer, 1977;
Mullin and Krivanek, 1982). If hypothermia develops as a consequence of COHb elevation,
behavioral effects may be secondary to the hypothermia, not the COHb directly. None of the
experimenters attempted to control for hypothermia effects. Thus behavioral effects of COHb
30 may be overestimated in the rat.
March 12, 1990 10-137 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 10-22. EFFECTS OF COHb ON SCHEDULE-CONTROLLED BEHAVIOR
\^ Exposure
M" Duration,
O min
P 12°
\o 1>44°
£
75
90
48
Injection
9 240
Co
oo
a
» 30
p>
H
a 90
o
S90
0
O Variable
3
o
^ «/-T>UK ..,<.<,
Elevated
COHb Range
%
9.0 - 58.0"
9.0 - 58.8'
35.0-55.0'
8.0 - 54.0«
15.0-55.0"
9.0 - 58.0
12.2 - 54.9
Continuous
distribution
up to 32.0
34.0 - 53.0
40.0 - 66.0
15.0 - 40.01
CO
n Scheds. Species Effect
5 CRF Rat Yes
8 Body Rat Yes
weight
15 MULTcombi- Rat Yes
nations of
FI3 and FR30
4 DRL21 Rat Yes
? FB, FR25 Rat Yes
VE5, VR15
VR25, DRL
22 CRF brain Rat Yes
stimulation
6 Behavioral Rat Yes
screen
3 Appetitive Monkey Yes
shuttling
3 MULT FR 30, Rat Yes
DRL 18
3 Multiple Rat Yes
sequential
responses
4 FCN Rat Yes
Comment
Rates fell inversely at COHb beginning at ca.20%
Weight fell inversely at COHb beginning at 22%. Food and
water consumption also fell.
Rates fell inversely at COHb beginning between 32 and
48% for schedules.
Rates fell inversely at COHb beginning at ca. 37%.
Temporal discrimination was undisturbed.
Effects were COHb ordinal beginning at ca. 14%. DRL was
affected at COHb < 1 %. Methods were poorly described
CO was injected ip. Effects were first noted near 45%
COHb.
Behavioral screen included reflexes, grasping, and
conditioned avoidance. Lowest level effect was on
conditioned avoidance at 12.2%. COHb levels measured
by authors were much lower than predicted from data of
Montgomery and Rubin (1971).
Shuttling velocity decreased as COHb beginning at 16 - 22%.
Rates fell beginning ca. 45% COHb.
Rats repeated releamed response chain after extinction.
More time to relearn was required beginning ca. 50% COHb.
Rates fell inversely at COHb beginning between 20 and 28%
Reference
Annau (1975)
Annau (1975)
Ator (1982)
Ator et al.
(1976)
Beard and
Wcrtheim (1967)
Fountain et al.
(1986)
Mullin and
Krivanek (1982)
Purser and
Berrill (1983)
Schrot and
Thomas (1986)
Schrot et al.
(1984)
Smith et al.
(1976a)
O
-------
The level of COHb may be estimated for studies in which it was not given by use of the
data from Montgomery and Rubin (1971). The latter-published normative curves can be used
for such estimates. Schrot and Thomas (1986) and Schrot et al. (1984) have published
corroborating curves. For all rat studies in which COHb was not measured, the present
5 reviewer estimated the levels from exposure parameters.
With one exception (Mullin and Krivanek, 1982), effects of COHb did not occur on
schedule-controlled behavior until COHb exceeded ca. 20%. In some studies no effect was
observed until even higher levels. It is possible, however, that a number of the studies were
insensitive because of the small numbers of subjects employed. In the study by Mullin and
10 Krivanek (1982), it was reported that conditioned-avoidance behavior was affected at COHb
levels as low as 12.2%. The COHb level reported in the latter study is, however, about half
of the value that would be estimated from the Montgomery and Rubin (1971) data. It seems
likely that either exposure or COHb values were erroneous in the report. If the exposure data
were correct, the effects threshold would fit the other data in the literature. It thus appears
15 that COHb does not affect schedule-controlled data in laboratory animals until levels exceed
20%.
When there were frank effects on schedule-controlled behavior, they seemed all to be in
the direction of a slowing of rate or speed of response. Schedule-produced patterns of
behavior were not disrupted, in general. Thus, it appears that the effect of elevated COHb is
20 on some general aspect of behavioral control having to do with the rate of processing.
10.4.2.9 Summary and Discussion of Behavioral Literature
The literature regarding the effects of COHb on behavior, as seen in the above review,
does not allow clear-cut conclusions. Results of studies frequently were not replicable or
25 were not supported by related studies. In this section an attempt will be made to discover
what, if any, general conclusions can be made.
Analysis of Technical Problems
It is possible that the many technical problems that were noted in the summary tables in
30 the 'technical critique' column may account for some of the lack of agreement among experi-
mental results. In the following analysis, tabulations were made of all of the studies in which
March 12, 1990 10-139 DRAFT-DO NOT QUOTE OR CITE
-------
either blinding or statistical analysis problems occurred. Studies were cast into 2x2 tables
according to the presence or absence of a particular condition and according to the occurrence
(or not) of a COHb effect. If multiple-dependent variables were measured in a particular
experiment, that experiment was tallied as having reported a significant effect if any one or
5 more of the variables was reported as affected. A study was tallied as having reported a
significant effect only if the effect occurred below 10% COHb to avoid the inclusion of frank
effects that will occur if COHb is made sufficiently high. All studies were included,
regardless of what dependent variables were studied. Only human studies were included. To
decide whether the technical problem in question can be inferred to have influenced the
10 results, a Fisher's exact test was conducted on each table. Two such tests were conducted;
therefore, the alpha level selected for each test was 0.05/2 or 0.025. Given signficant results,
exploratory tests also were conducted.
Table 10-23 is a tabulation of studies according to their blinding practices. Non-blind
and single-blind studies were pooled because of the few non-blind studies. The Fisher test
15 yielded significant results, p = 0.0154. It is impressive that the rate of reported COHb
effects was about 2.24 times as high for studies not using a double-blind design. When the
five non-blind studies were dropped and the data reanalyzed in an exploratory manner, the
rate of finding significant effects for single-blind studies was 2.275 times as high as for
double-blind studies (p = 0.0172).
20
TABLE 10-23. EFFECT OF BLIND CONDITIONS
25 Non-Double Blind Double Blind
Effects 16 6
No Effects 9 15
30
Fisher's exact test, p = 0.0154.
35
March 12, 1990 10-140 DRAFT-DO NOT QUOTE OR CITE
-------
Table 10-24 is a tabulation of studies according to the employment of multiple-
significance tests on the same data set. The Fisher test was not significant (p = 0.221).
5 TABLE 10-24. EFFECT OF STATISTICAL METHODOLOGY
Multiple- Conservative-
Significance Significance
10 Test Methods Test Methods
Effects 13 9
No Effects 13 11
15
Fisher's exact test, p = 0.221.
20 From the above analyses it may be concluded that studies that were not conducted in a
double-blind manner tend to demonstrate more apparent COHb effects. It may be argued that
the bias thus introduced into the study is added to whatever COHb effect may be present. No
evidence was demonstrated, however, for the hypothesis that multiple statistical tests tended to
produce an inflated Type I experiment-wise error rate.
25
Evaluation of the Literature
In the evaluation of the literature on the behavioral effects of COHb it is not clear how
to treat the results from studies not conducted in a double-blind manner. Although they are
biased toward reporting COHb effects, they clearly contain useful information. In the
30 following summary, results from double-blind studies will be tabulated separately from results
of studies not conducted in a double-blind manner.
Table 10-25 summarizes literature on the behavioral effects of COHb. Both double-
blind and non-double-blind studies are tabulated. For each family of dependent variables, the
table gives the number of studies in the double-blind and non-double-blind categories.
35 Finally, the proportion of studies reporting a COHb effect, p(E), is given for both double-
blind and non-double-blind studies. Several conclusions may be drawn from Table 10-25.
March 12, 1990 10-141 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 10-25. PROBABILITY OF EFFECTS OF COHbm
10
15
20
25
Dependent Variable
Absolute visual threshold
Critical flicker fusion
Misc. visual functions
Misc. auditory functions
Fine motor skills
Reaction time
Tracking
Vigilance
Continuous performance
Time estimation
Misc. cognitive function
Automobile driving
Brain electrical activity
Non-
Double
Blind
(n)
4
7
9
0
6
7
4
4
4
2
5
3
3
Double
Blind
(n)
1
3
5
3
4
5
7
4
5
4
5
2
6
Non-
Double
Blind,
P(E)
.20
.33
.55
N/A
.33
.00
.00
.75
.75
.50
.80
.33
.33
Double
Blind
P(E)
.00
.00
.00
.00
.00
.00
.43
.25
.40
.00
.00
.00
.50
'Based on numbers of studies in each category.
Those conclusions are as follows:
30
35
40
A. Non-double-blind studies produce a greater proportion of reported effects, even
for the individual dependent variables. This is true with the exception of tracking
and brain electrical activity. This observation supports the previous inference that
non-double-blind studies are biased toward finding more effects than justified.
B. Studies using double-blind procedures found effects on four of the 14 dependent
variable families: tracking, vigilance, continuous performance, and brain
electrical activity.
C. In the five dependent variable categories where COHb was found to affect
behavior, usually less than half of the reported studies found effects in double-
blind studies.
D. Sensory and cognitive effects were not found to be affected by COHb in double-
blind studies.
March 12, 1990
10-142 DRAFT-DO NOT QUOTE OR CITE
-------
E. In most instances, the rate of finding COHb effects in non-double-blind studies
was high in the same dependent variables where the double-blind studies reported
effects. This observation lends further support to the findings in the double-
blind studies.
Continuous Performance. It may be argued that the dependent variables of tracking,
vigilance, and continuous performance are related functionally. Each of the dependent
variables in the three categories require the continuous performance of some sort of behavior.
10 The response rate and/or attention demand in vigilance behavior is low compared to the other
two groups, otherwise, the behaviors are similar. Tracking is clearly a particular form of
continuous performance which was categorized separately simply because of the homogeneity
of a group of studies that existed in the literature. It also may be argued that, despite the
high response rates in the tracking and continuous performance studies, both of these
15 behaviors require a strong component of sustained attention. It seems fair to conjecture,
therefore, that behaviors that require sustained attention and/or sustained performance are
most sensitive to disruption by COHb.
The group of studies of tracking, vigilance, and continuous performance offer the most
consistent and defensible evidence of COHb effects on behavior. The results across studies
20 is, however, far from consistent. Further examination of the three areas seems appropriate.
Compensatory tracking was studied by two groups of investigators using virtually
identical task parameters and equipment (Putz et al., 1976, 1979; Benignus et al., 1987,
1989a). Both of the studies by Putz et al. (1976, 1979) found significant and moderately
large effects of 5% COHb. Benignus et al. (1987) reported similar but smaller significant
25 effects in a nearly identical experiment to Putz et al. (1976). However, in a dose-effects
study including another direct replication group, Benignus et al. (1989a) found no significant
effects, even for COHb levels of 17%. In the latter study, the means were nearly dose
ordinal but too small to be statistically significant. It is particularly puzzling why the latter
study, using a large number of subjects on an identical task, should find no significant effects
30 for even 17% COHb when three other studies found effects at lower levels. Three other
March 12, 1990 10-143 DRAFT-DO NOT QUOTE OR CITE
-------
double-blind tracking studies of various methods found no effects of COHb levels of 12% or
greater.
As discussed in the above literature review, there is a similar disunity among studies on
the effects of COHb on vigilance. Because of the many failed attempts at direct replication,
5 the conclusions seem weaker than for tracking.
Of the five double-blind experiments in which continuous performance was measured,
three were mentioned earlier in the discussion of tracking. In these studies (direct
replications), continuous performance was measured simultaneously with tracking (Putz et al.,
1976, 1979; Benignus et al., 1987). The latter of the three found no effects. A small study
10 reported continuous performance effects that were disordinal in COHb (O'Donnel et al.,
1971a). The remaining study (Benignus et al., 1977) used a different task and obtained no
COHb effects.
Multiple Performance. It is possible that COHb impairs task performance more when
15 multiple tasks are performed simultaneously, thus decreasing the amount of behavioral reserve
capacity. To test this exploratory hypothesis, behavioral studies (sensory and
electrophysiological excluded) were tabulated according to single/multiple task performance
and significant/nonsignificant effects of COHb. If any given study used both single-task and
multiple-task behaviors, the study was tabulated according to the multiple-task results only.
20 For studies in which only single tasks were required, the study was classed as having shown
significant results if one or more variables were significant, otherwise as nonsignificant. The
above rules prevented any study from being tabulated more than once.
Table 10-26 is the result of the above tabulation. The exploratory Fisher's exact test
yielded p = 0.081. While the result would have been nonsignificant by a priori rules, the
25 table shows a slight tendency toward more multitask studies showing a significant effect,
whereas more of the single-task studies found no effect.
The question of multitask vs. single-task performance sensitivity to COHb disruption
would be answered best by experiments in which the tasks were performed singly and
together within the same study. In the literature there were only two cases in which this was
30 done. In both of these, one of the single tasks alone was affected more than multitask
performance (Bender et al., 1972; Gliner et al., 1983). Experiments designed to specifically
March 12, 1990 10-144 DRAFT-DO NOT QUOTE OR CITE
-------
10
TABLE 10-26. EFFECT OF SINGLE VS. MULTIPLE
TASK PERFORMANCE
Single Task Multiple Task
Effects 7
No Effects 14
Fisher's exact test, p = 0.081.
15
answer the question of single-task vs. multitask performance sensitivity need to be done.
COHb Formation Rate. It is possible that the rate at which COHb is formed is an important
variable in the effects of COHb on the CNS. To explore this possibility, studies were cast
20 into a table according to their rate of COHb formation (fast or slow) and whether effects were
found or not. When COHb was formed to its target value in 10 min or less, the
study was tabulated as fast. Studies in which effects were found only above 10% COHb were
tabulated as no effects to avoid consideration of frank effects.
Table 10-27 is the result of the above classification. The exploratory two-sided Fisher's
25 exact test yielded p = 0.002. It appears that studies using slow COHb formation are more
likely to find significant effects. Experiments explicitly designed to test the hypothesis would
be needed before firm conclusions can be drawn.
30 TABLE 10-27. EFFECT OF RATE OF COHb FORMATION
Slow Fast
35 Effects
No Effects
20
12
2
9
40 Fisher's exact test, p = 0.002.
March 12, 1990 10-145 DRAFT-DO NOT QUOTE OR CITE
-------
10.4.2.10 Hypotheses
Dose-Effect Function
An effort has been made to unify the dose-effects literature concerning CO and behavior
(Benignus et al., 1989b). The analyses and hypotheses of the latter article will be reviewed
5 below. Both laboratory animal and human data were considered. Only dose-effects studies
were considered so that comparisons and extrapolations could be made. The literature
concerning dose-effects functions in humans, as above, was found to be inconsistent. As has
been pointed out in the present review, effects of COHb elevation do not become significant
until ca. 20% in laboratory animals. The literature for such higher levels of COHb is quite
10 consistent. Nonlinear, positively-accelerated functions were fitted to the laboratory animal
data to describe dose-effects relationships.
To compare the human data to the laboratory animal data, it was necessary to select
from the divergent group of human dose-effects studies. The argument was made that the
most accurate estimate of the dose-effects function was that the effects below ca. 20% COHb
15 were either zero or very small in size. This argument was based on the observations about
nonverification of findings in the literature as well as upon the findings of Benignus et al.
(1989a), which demonstrated small but nonsignificant elevations in mean tracking error in a
large study (n=74) with COHb levels up to 17%.
The data from Benignus et al. (1989a) were fitted with the same form of function as
20 fitted to the laboratory animal data. Extrapolation of the curve projected the human data as
passing through laboratory animals curves. The latter observation was used to imply that the
human and laboratory animal findings were similar and that frank effects of COHb elevation
in humans should not be expected below ca. 20% COHb.
An argument was made for the possibility and importance of small effects for low
25 COHb levels but no unimpeachable evidence could be marshalled. The possibility of small
effects below ca. 20% COHb was supported by the observation that in many studies, using
both laboratory animals and humans, means were usually shifted in the direction of
deleterious effects at low levels, but not in a statistically significantly manner. Means were
much more rarely shifted in directions implying improvement of behavioral abilities with
30 small levels of COHb. Thus, there may exist small effects below 20% COHb (or some
March 12, 1990 10-146 DRAFT-DO NOT QUOTE OR CITE
-------
individuals are affected while most are not). The latter possibility cannot be ignored but it
cannot be confirmed.
Compensatory Mechanisms
5 As discussed above in Section 10.4.1, a proportional vasodilation occurs in the brain in
response to COHb elevation. This vasodilation is sufficient, on the average, to keep the
cerebral O2 consumption from being reduced even though the COHb has reduced the blood's
O2-carrying capacity 20 to 30% and the presence of COHb has shifted the OjHb dissociation
curve to the left. The cerebral vasodilation may be viewed, Ideologically, as a closed-loop
10 compensatory mechanism to assure adequate oxygenation of the brain in the presence of
elevated COHb.
If the cerebral vasodilation is adequate in any individual and if the vasodilation is
homogeneous for all cerebral tissue, then that individual should not be behaviorally impaired
by COHb elevation. This statement assumes that the sole mechanism for CO toxicity is the
15 hypoxic effect of COHb.
The agreement between the behavioral literature and the compensatory mechanism
hypothesis is noteworthy. According to the compensatory mechanism data, O2 consumption
in the brain does not begin to decrease until COHb exceeds 20 to 30%. Data from behavioral
studies in laboratory animals demonstrate that significant effects in schedule-controlled
20 behavior do not occur below 20 to 30% COHb. Behavioral effects in humans have not been
unambiguously demonstrated below 20 to 30% COHb.
10.4.2.11 Conclusions
At the present stage of evidence, it seems unfounded to conclude that COHb elevation
25 below ca. 20% is deleterious to the behavioral abilities in humans. It seems unwise,
however, to ignore the frequent evidence in favor of effects. Even if effects are small or
occasional, they might be important to the performance of critical tasks.
Some of the differences among studies of the effect of COHb on the behavior of humans
is due apparently to technical problems in the execution of experiments, because single-blind
30 or nonblind experiments tend to yield a much higher rate of significant effects than do double-
blind studies. Even when non-double-blind experiments are eliminated from consideration,
March 12, 1990 10-147 DRAFT-DO NOT QUOTE OR CITE
-------
however, a substantial amount of disparity remains among results of studies. It is possible
that such residual disagreement is due to the action of an unsuspected variable that is not
being controlled across experiments.
If the compensatory CNS blood flow hypothesis has validity, it is possible that there
5 exist groups that are at higher risk to COHb elevation than the usual subjects who were
studied in the behavioral experiments. Disease or injury might either impair the
compensatory mechanism or reduce the non-exposed O2 delivery. Aging increases the
probability of such injury and disease. It also is possible that there exist individual
differences with regard to COHb sensitivity and/or compensatory mechanisms. Too little is
10 known about the compensatory process to make conjectures, but the matters seem important
to investigate.
The literature on the behavioral effects of COHb elevation has grown considerably since
the last Criteria Document was written (U.S. Environmental Protection Agency, 1979). It
seems safe to state that the effect of the new information did not increase the certainty about
15 COHb effects. Unless some key piece of information is uncovered by new research, there
does not seem to be much hope of gaining clarification in the conflicting findings. The
solution to the puzzle would seem to lie in the conduct of more research into mechanisms of
action of CO rather than in further attempts to show reliable behavioral effects. The latter
approach, which has not been successful in the past, should be resumed only when
20 mechanisms of toxicity are understood better. More findings of behavioral effects of COHb
would not appreciably alter the conclusions of the present section unless future studies were to
show an unusual unanimity.
25 10.5 DEVELOPMENTAL TOXICITY OF CARBON MONOXIDE
10.5,1 Introduction
Developmental toxicity has been described in the U.S. Environmental Protection
Agency's Guidelines for the Health Assessment of Suspect Developmental Toxicants (Federal
Register, 1986) as including death of the developing organism, structural abnormalities,
30 altered growth, and functional deficits resulting from toxic exposures that occur prior to the
subject's attaining sexual maturity. The appearance of toxic effects may occur at any time
March 12, 1990 10-148 DRAFT-DO NOT QUOTE OR CITE
-------
throughout life. Concern for special vulnerability of immature organisms to toxic compounds
focuses on the possibilities that (1) a toxic exposure that is not sufficient to produce maternal
toxicity or toxicity in the adult organism will adversely affect the fetus or neonate or (2) at a
level of exposure that does produce a toxic consequence to the adult or mother that the fetus
5 or neonate suffers a qualitatively different toxic response. Toxic responses that occur early in
life, but which are not permanent, may or may not be a cause of concern. In some cases they
truly may be transient events with no persisting consequences. However, in other cases such
results may have their own consequences for development of the organism or may reappear
under conditions of ill health produced by other toxic exposures, environmental stresses, or
10 exposure to pathogenic agents. Therefore, even seemingly transitory toxic effects must be
viewed as serious consequences of exposure when they occur in humans. In this section, data
are presented that describe the toxic consequences of CO exposure early in development.
These data describe the types of toxic outcomes that the immature subject may show and also
help to identify the dosage at which such toxicity is seen.
15 There are theoretical reasons and supporting experimental data that suggest that the fetus
and developing organism are especially vulnerable to CO. One reason for approaching the
fetus as a separate entity for purposes of regulation is that the fetus is likely to experience a
different CO exposure than the adult given identical concentrations of the gas in air. This is
due to differences in uptake and elimination of CO from fetal hemoglobin which are
20 documented below. Less studied is the possibility that tissue hypoxia may differ between the
fetus and adult even at equivalent carboxyhemoglobin concentrations as a result of differences
in perfusion of critical organs, in maturation of adaptive cardiovascular responses to hypoxia
and as a result of tissue requirements for oxygen. Inferences concerning these factors are
obtained principally from experimentation performed in laboratory animals in which the
25 immature organism does show enhanced toxicity relative to the adult. Concern must be
expressed, too, for the development of sensitive and appropriate animal models. It is
necessary to bear in mind the relative state of development of the human and laboratory
animal in question at the time of birth in developing useful animal models. For example, the
neonatal rat is very immature relative to the neonatal human at birth, with respect to
30 development of the central nervous system, (Fechter et al., 1986) and so a combined prenatal
and neonatal exposure model may be more accurate in predicting consequences of prenatal
March 12, 1990 10-149 DRAFT-DO NOT QUOTE OR CITE
-------
exposure in the human. Further, differences appear to exist among species in the relative
affinity of fetal and adult hemoglobin for CO. These data are reviewed by Longo (1970).
There exist a variety of relevant data bases that will be reviewed. These include
experimental investigations conducted using laboratory animals (and these are most
5 numerous), case report data collected in offspring of women exposed to generally high-level,
acute CO poisoning during pregnancy, and epidemiological data. One large, but
problematical, literature from the standpoint of this document concerns maternal cigarette
smoking.
Cigarette smoking constitutes a major source of exposure of the individual to CO and
10 this is particularly relevant for the fetus because of the high affinity of fetal hemoglobin for
CO (Longo, 1977). Maternal smoking has been associated with a variety of untoward
consequences ranging from increased incidence of placenta previa, abruptio placentae,
spontaneous abortion and subsequent fetal deaths to depressed birthweight, and increased
numbers of hospital admissions for a broad range of complaints during at least the first
15 5 years of life, to poorer than predicted school performance during the first 11 years of life.
This literature has been thoroughly reviewed as a report to the U.S. Surgeon General
(National Institute of Child Health and Human Development, 1979). These outcomes should
be cause for significant concern. However, it is not easy to determine the extent to which CO
is a causative factor. Cigarette smoke contains a large number of toxic chemicals other than
20 CO and these other agents either alone or in combination may be responsible /or the untoward
outcomes that are readily associated with maternal smoking. A few epidemiological reports
are reviewed below in which it is concluded that CO, either from ambient sources or cigarette
smoke, is responsible for developmental disruption. However, these reports generally are
deficient in characterizing the level of CO exposure or in ruling out potential contributions by
25 other toxic agents contained in smoke. Investigations with laboratory animals exposed to CO
rather than cigarette smoke early in development have demonstrated developmental anomalies
and persisting neurobehavioral disorders that are most relevant to this document and that are
reviewed below. Because such effects are seen at CO levels approaching values observed in
the offspring of cigarette smokers, this must be cause for serious concern. However, it is not
30 possible to use the cigarette smoker literature in establishing criteria for permissible CO
exposure.
March 12, 1990 10-150 DRAFT-DO NOT QUOTE OR CITE
-------
10.5.2 Theoretical Basis for Fetal Exposure to Excessive Carbon Monoxide
and for Excess Fetal Toxicity
Hill et al. (1977) aptly described mathematical models for predicting fetal exposure to
CO based upon placental transport and the differences between maternal and fetal hemoglobin
5 affinity for CO and O2. They predicted that for any maternal CO exposure of moderate
duration that fetal COHb levels would lag behind maternal COHb levels for several hours, but
would, given sufficient time, surpass maternal COHb levels by as much as 10% (in humans)
due to the higher affinity of fetal hemoglobin for CO than adult hemoglobin. Moreover, they
predicted a far longer wash-out period for the fetal circulation to eliminate CO following
10 termination of exposure than that found in the adult. Data, accumulated in both laboratory
animal and human studies, support these conclusions.
10.5.2.1 Evidence for Elevated Fetal Carboxyhemoglobin Relative to Maternal
Hemoglobin
15 A fairly wide range of neonatal and maternal COHb levels has been published for
humans, probably due to wide differences in cigarette smoking patterns prior to and during
labor. In one recent study, measurement of fetal cord blood in the offspring of cigarette
smokers who smoked during labor showed fetal COHb levels 2.55 times higher than in
maternal blood. Cord blood averaged 10.1 % COHb at delivery while maternal blood
20 averaged 5.6% on the mother's arrival at the hospital and 4.1 % at delivery (Bureau et al.,
1982). These values for fetal COHb are fairly high relative to other published sources
(Longo, 1977 - Table IV). However, greater fetal COHb levels have been found in
laboratory studies across a broad range of animal species if sufficient time was allowed for
COHb to equilibrate in the fetal compartment. Christensen et al. (1986) ultimately observed
25 higher CO levels in fetal guinea pigs than in their dams following CO exposure given near
term. Immediately following maternal exposure, at gestational age 62 to 65 days, to a bolus
of CO (5 mL given over 65 seconds), they reported a faster elevation in maternal COHb
levels than in fetal levels, a finding consistent with the models of Hill et al. (1977).
Anders and Sunram (1982) exposed gravid rats to 22 ppm CO for one hour on Day 21
30 of gestation and reported that fetal COHb levels averaged 12% higher than levels taken at the
same time period in the dam. These results are consistent with those of Garvey and Longo
(1978) whose study involved chronic CO exposures in rats. Dominick and Carson (1983)
March 12, 1990 10-151 DRAFT-DO NOT QUOTE OR CITE
-------
exposed pregnant sows to CO concentrations of 150 to 400 ppm for 48 to 96 h between
gestation days (GDs) 108 to 110. They reported fetal COHb levels that exceeded maternal
levels by 3 to 22% using a CO-oximeter.
Longo and Hill (1977) similarly reported that COHb levels in fetal lambs do exceed
5 maternal levels once equilibrium is reached in the fetal compartment and that this washout
from the fetal blood exceeds that observed for maternal blood.
Fetal COHb kinetics may not be static throughout pregnancy. Bissonnette and Wickham
(1977) studied transplacental CO uptake in guinea pigs at approximate gestational ages 45 to
68 days. They report that placental diffusing capacity increases significantly with increased
10 gestational age and appears to be correlated with fetal weight rather than placental weight.
Longo and Ching (1977) also showed increases in CO diffusion rates across the placenta of
the ewe during the last trimester of pregnancy. However, they did not find a consistent
increase when diffusion rate was corrected for fetal weight (i.e., when diffusing capacity was
expressed on a per kilogram fetal weight basis).
15
10.5.2.2 Effect of Maternal Carboxyhemoglobin on Placental O2 Transport
Gurtner et al. (1982) studied the transport of O2 across the placenta in the presence of
CO by cannulating both the maternal and fetal vessels of anesthetized sheep preparation.
They measured the transport of O2 across the placenta compared to transport of argon, urea,
20 and tritiated water when CO was introduced. They showed a reduction in O2 diffusing
capacity relative to Ar that appeared to be related to the level of maternal COHb. Reduction
of O2 transport was observed below 10% COHb and O2 transport approached zero at COHb
values of 40 to 50%. They interpreted these data as supporting the role of carrier-mediated
transport for O2 and suggest that CO competitively binds to this carrier. An alternative
25 explanation is that the introduction of CO simply reduces the amount of fetal hemoglobin
available to bind O2. Moreover, Longo and Ching (1977), for example, were unable to alter
CO diffusing capacity across the placenta by administration of a series of drugs that bind to
cytochrome P-450. Gilbert et al. (1979) have underscored the low concentration of
cytochrome P-450 in human placenta, as compared to liver and to the very low levels found
30 in many other species.
March 12, 1990 10-152 DRAFT-DO NOT QUOTE OR CITE
-------
Christensen et al. (1986) suggest that maternal CO exposure may independently impair
O2 diffusion across the placenta due to the enhanced affinity of maternal hemoglobin for O2 in
the presence of COHb (the Haldane effect). Using the guinea pig, these authors demonstrated
an initial almost instantaneous fall in fetal paO2 levels and an increase in fetal pCOj, which
5 subsequently was followed by an increase in fetal COHb between approximately 5 to 10 min
(the last time point studied but a time when fetal COHb values were still far below maternal
values). They calculated that the decrease in fetal O2 transfer was due mostly to a decrease in
maternal O2-carrying capacity, but also, perhaps up to one-third, by the increased affinity of
hemoglobin for O2 in the presence of CO. This model assumes that uterine perfusion remains
10 constant under the experimental conditions used. Longo (1976) also showed a significant
dose related drop in fetal O2 levels in blood taken from the fetal descending artery and fetal
inferior vena cava after pregnant ewes were exposed to variable levels of CO for durations
sufficient to yield COHb equilibration in both the fetal and maternal compartments. To
summarize, it has been demonstrated that the presence of maternal COHb over a range of
15 values results in depressed O2 levels in fetal blood. The simplest explanations for the inverse
relationship between maternal COHb and fetal O2 levels are reduced maternal O2-carrying
capacity, impaired dissociation of O2 from maternal hemoglobin (the Haldane effect), and
reduced availability of free fetal hemoglobin able to bind O2.
20 10.5.3 Measurement of Carboxyhemoglobin Content in Fetal Blood
Zwart et al. (1981) and Huch et al. (1983) have called into question the accuracy of
spectrophotometric measurements of COHb in fetal blood using the IL 182 and 282 CO-
oximeters. The CO-oximeter is effectively a spectrophotometer preset to read samples at
specific wavelengths that correspond to absorbance maxima for oxy- carboxy- and met-
25 hemoglobin determined using adult blood samples. Different plug-in modules (IL 182) or
programmed absorbance values (IL 282) can be used to correct for species differences in the
absorbance spectrum of rat, human, dog, and cow. Some investigators have used these
instruments for estimating COHb levels in species for which the instrument has not been
calibrated such as the pig and guinea pig. Typically, individual investigators have calibrated
30 the CO-oximeter using blood standards fully saturated with CO and with O2. The adequacy
of such a procedure is not certain. (See Chapter 8, Section 8.5 for more details on the
March 12, 1990 10-153 DRAFT-DO NOT QUOTE OR CITE
-------
measurement of COHb.) Further, the correspondence of absorbance maxima between adult
and fetal hemoglobin for species upon which the CO-oximeter is calibrated at the factory is
an empirical question for which little data are published. Noting the finding of higher
apparent COHb levels in the venous cord blood of humans than in the uterine artery, Huch
5 et al. (1983) examined the possibility that oxyhemoglobin in fetal blood might interfere with
accurate measurement of COHb levels in the fetus due, presumably, to different absorbance
maxima for fetal than adult blood. Working in vitro, Huch et al. (1983) deoxygenated fetal
and maternal blood by flushing a tonometer with nitrogen and 5% CO2. They introduced a
"small volume" of CO gas, measured the blood gases using the IL 282 CO-oximeter, and then
10 studied the effect of stepwise addition of O2 to the apparent COHb levels. They showed little
influence of O2 saturation upon maternal COHb, but indicate that O2 saturation did affect
readings of fetal COHb so as to overestimate COHb. This confound is particularly likely at
high oxyhemoglobin concentrations. Zwart et al. (1981) suggest an apparent elevation of
COHb levels of approximately 2% with 40% oxyhemoglobin saturation and of approximately
15 6% with oxyhemoglobin levels of 90 to 95%. Such errors do not invalidate the finding that
fetal COHb exceeds maternal values, but do bring into question the magnitude of this
difference. Whether similar errors also occur in measuring fetal COHb levels in animal blood
is uncertain and should be subjected to experimental test. The calibration of
spectrophotometers based upon fetal hemoglobin absorbance spectra rather than automated
20 analysis based upon adult absorbance spectra is recommended to achieve greater accuracy in
determining absolute levels of CO in fetal blood. Vreman et al. (1984) have described a gas
chromotographic method for measuring COHb which has been applied to human neonates.
Because of the very small volume of blood needed to make these measurements and because
they eliminate the problem of absorbance spectra of fetal hemoglobin, this may be considered
25 a useful means of accurately assessing COHb in developing organisms. There also is a new
model of the Co-oximeter (#482) which apparently allows for use of absorbance spectra based
on calibration of fetal blood.
10.5.4 Consequences of Carbon Monoxide in Development
30 This section presents the evidence that CO exposure during early development has the
potential of producing untoward effects. The four types of toxic outcomes - fetotoxicity,
March 12, 1990 10-154 DRAFT-DO NOT QUOTE OR CITE
-------
gross teratogenicity, altered growth, and functional deficiencies in sensitive organ systems -
are considered in order. As is the case in adult organisms, the nervous and cardiovascular
systems appear to be most sensitive to CO exposure.
5 10.5.4.1 Fetotoxic and Teratogenic Consequence of Prenatal Carbon Monoxide
Exposure
There is clear evidence from human and animal studies that very high levels of CO
exposure may be fetotoxic. However, there exists some question concerning the level of
exposure that causes fetal death as both the duration and concentration of maternal exposure
10 are critical values in determining fetal exposure. More important for the setting of ambient
air standards is the suggestion of a causal relationship between sudden infant death syndrome
(SIDS) and ambient CO levels. These data are extremely limited at present and no conclusive
correlation can be drawn between SIDS and CO.
The lowest-observed-effect level (LOEL) for fetotoxicity in laboratory animals appears
15 to be in the range of 500 pprn for rodents, but one experiment conducted in pigs suggests that
this species may be especially sensitive to this effect showing significant fetotoxicity at
250 ppm for 2 to 4 days late in gestation. Evidence of fetotoxicity in animals also has come
from acute, high-dose experiments that are not included here because they are not directly
relevant to standard setting.
20 The data that suggest that prenatal CO exposure produces terata is extremely limited
and, again, comes largely from quite high exposure levels. Table 10-28 presents the reported
effects of prenatal CO exposure on fetotoxicity, teratogenicity, and growth abnormalities.
Perinatal Carbon Monoxide Exposure and Sudden Infant Death Syndrome (SIDS)
It has been suggested that CO may be a causative factor in SIDS. Hoppenbrouwers
25 et al. (1981) reported a statistical association between the frequency of SIDS and levels of
several airborne pollutants including CO, sulfur dioxide (SO2), NO2, and hydrocarbons.
SIDS was reported more commonly in the winter, at a time when the burning of fossil fuels
for heating would be greatest. It is interesting to note that there is a phase lag of
approximately 7 weeks between the increase in pollutant levels and the increase of SIDS
30 cases. Whether this phase lag for SIDS represents a failure to identify the true cause of SIDS
or a lag with some meaning in terms of the physiology is uncertain. Further correlations
March 12, 1990 10-155 DRAFT-DO NOT QUOTE OR CITE
-------
cr
to
TABLE 10-28. TERATOGENIC CONSEQUENCES OF PRENATAL CARBON MONOXIDE EXPOSURE
IN LABORATORY ANIMALS
Species (Strain)
Maternal Treatment
Maternal Toxicity
Development Abnormality
References
o
ON
Mouse (strain NR)
Rat (Ames-Wilson)
Rat (Sprague-Dawley)
Rabbit (strain NR)
Mouse (CF-1) and
Rabbit (New Zealand)
5,900 or 15,000 ppm CO
for 5-8 min every
other day of
gestation
3,400 ppm CO for 1 h/day
for 3, 6, or 8.3 mo
750 ppm CO for 3 h/day
on GDs 7, 8, or 9
90 or 180 ppm CO from
mating to the day before
parturition
7 or 24 h/day of 250 ppm
CO on GDs 6-15 for mice
and on GDs 6-18 for rabbits
Acute effects: unconsciousness
(no COHb levels)
Decreased body weight, appetite,
and muscle tone; lack of grooming
(COHb levels of 60-70%)
NR
(no COHb levels)
NR
(COHb levels of 8-9% and
and 16-18%)
Transient increase in body
weight for mice in 7-h/day
group; COHb levels of 10-11 %
(mice) and 13-15% (rabbits)
for 7-h/day exposures
Abortions, resorptions, and
abnormal growth of survivors
Decrease of litter size, decrease
of preweaning survival (50%
reduction of pregnancy at 3 mo,
no pregnancies induced with
longer exposures; 19% increase
of estrous cycle)
Absorptions, stillbirths, and
skeletal anomalies, decreased
fetal body weight and crown-
rump length
180 ppm: 35% mortality of
neonates, 11 % decrease in
birth weights, and increase
in malformations 90 ppm:
9.9% mortality of neonates,
13% decrease in birth weights
Mice: increase in resorptions
and body weight w/7-h/day
exposure, decrease in body
weight and crown-rump length
w/24-h day exposure; both
exposures increased skeletal
anomalies (GD 18)
Rabbits: increase in body
weight and crown-rump length
w/7-h/day exposure
Wells (1933)
William and Smith (1935)
Choi and Oh (1975)
Astrup et al. (1972)
Scwetz et al. (1979)
-------
»—*
to
>«
VO
TABLE 10-28 (cont'd). TERATOGENIC CONSEQUENCES OF PRENATAL CARBON MONOXIDE EXPOSURE
IN LABORATORY ANIMALS
Species (Strain)
Maternal Treatment
Maternal Toxicity
Development Abnormality
References
Rat (Long-Evans)
Mice (CD-I)
Mice (CD-I)
Pig
Rabbit
0, 30, or 90 ppm CO or
13% oxygen in nitrogen
on CDs 3-20
125
250
500 ppm CO CDs 8-18
0, 65, 125 ppm CO
GDs 7-18
150-450 ppm for 48-
96 h between GDs
108-110
12 "puffs" of 2700-
5400 ppm CO daily from
GDs 6-18
Decrease in successful
pregnancies; COHb levels
of 4.8 and 8.8%
None
Decreased maternal respiration rate
Significant maternal death rate
13% oxygen: 12% decrease in
body weight
90 ppm CO: 14% increase in brain
weight, 24% decrease in lung
weight, serotonin concentration
decrease in brain
Increased fetal mortality
significant w/500 ppm, no
effect on number of
implantation sites
Impaired righting reflex on
PD 1 for 125-ppm group,
impaired negative geotaxis
on PD 10 for 125-ppm group,
impaired serial righting on
PD 14 for both 65- and 125-ppm
CO groups
Linear increase in number of
stillbirths significant when
maternal COHb exceeded 23%
(approximately 2500 ppm)
Larger number of fetal deaths
No terata
Garvey and Longo (1978)
Singh and Scott (1984)
Singh (1986)
Dominick and Canon (1983)
Rosenkrantz et al. (1986)
Rat (Wistar)
1,000-6,000 ppm CO
2x/day for2h
40 min total from
GDs
0-6
7-13
14-20
0-20
Decreased fetal weight
at GD 21
Tachi and Aoyama (1983)
Tachi and Aoyama (1986)
-------
were obtained between SIDS and the predicted level of CO and lead for the child's birth
month and between SIDS and the level of pollution at the reporting station closest to the
infant's home. These correlations are not compelling without more information on the
methods by which other possible risk factors were controlled in making the geographical
5 correlations. While it is technically difficult, it would be very useful to obtain COHb levels
close to the time of death in SIDS victims as this would greatly assist in determining the
incidence of elevated CO exposure in such cases.
There have been several studies linking maternal cigarette smoking with SIDS (these are
reviewed in the National Institute of Child Health and Human Development, 1979 report on
10 "Smoking and Health"), but it is uncertain what the role of CO might be in such a
relationship. Thus, it is clear that severe, acute CO intoxication can be fetotoxic although
specification of maternal and fetal COHb levels is difficult because such exposures rarely
involve the achievement of steady-state COHb levels or permit careful and rapid
determination of COHb levels. More relevant to the issue of standards for ambient exposure
15 is the possible link between CO and SIDS, but this literature currently is insufficient to
determine whether such a relationship exists.
Fetotoxicity in Laboratory Animals
Working with CD-I strain mice, Singh and Scott (1984) found significant increases in
20 the number of dead or resorbed fetuses per litter and an increase in fetal mortality with
continuous CO exposure of 500 ppm from Gestation Day (GD) 8 until subjects were
sacrificed at GD 18. Although not statistically significant, they found a dose-related trend in
these measures beginning at approximately 125 ppm. The no-observed-effect level (NOEL)
for these measures was 250 ppm. There was no effect of CO on the number of implantation
25 sites suggesting a fetotoxic rather than an embryopathic event. Schwetz et al. (1979) also
exposed mice to 250 ppm CO for 7 and 24 h per day on GDs 6 to 15. They found no effect
on number of implantation sites or number of live fetuses per litter but did show a significant
elevation in resorptions with the 7-h exposure (10 to 11% COHb) but not with 24-h per day
exposures.
30 Dominick and Carson (1983) exposed pregnant sows to CO concentrations of 150 to
400 ppm for 48 to 96 h between GDs 108 to 110 (average gestation was 114 days). They
March 12, 1990 10-158 DRAFT-DO NOT QUOTE OR CITE
-------
showed a significant linear increase in the number of stillbirths as a function of increasing CO
exposure. Stillbirths were significantly elevated above control levels when the maternal
COHb levels exceeded 23% saturation. These saturation levels were obtained at
approximately 250 ppm. COHb levels were measured using an IL 182 CO-oximeter equipped
5 with a human blood board; pig blood fully saturated with CO and with O2 were run each day
to calibrate the instrument. There was a very large variability among Utters at a given
concentration level in the percentage of stillbirths that occurred. Penney et al. (1980) found
evidence of reduced litter size in rats exposed for the last 18 days of gestation to 200 ppm CO
(maternal COHb levels averaged 28%). However, Fechter et al. (1987) did not observe
10 similar effects on litter size in rats exposed to levels of CO as high as 300 ppm (maternal
COHb levels of 24%).
Teratogenicity
There are very limited data (Astrup et al., 1972) suggesting increased fetal mortality and
15 malformations among rabbits exposed to 180 ppm CO throughout gestation (COHb levels =
16 to 18%). The frequency of malformations reported was very small, the historical rate of
such anomalies in the laboratory undocumented, and so these results require replication by
other workers before they can be considered as the basis for regulation. Rosenkrantz et al.
(1986) exposed rabbits to high doses (12 puffs of 2,700 to 5,400 ppm CO) for short time
20 periods daily from CDs 6 to 18. COHb levels were estimated at 16% although animals had
not equilibrated with the inhaled mixture. Despite a large number of fetal deaths, there was
no evidence of terata in the CO-exposed animals.
Choi and Oh (1975) reported skeletal anomalies in rats exposed to 750 ppm CO for 3 h
per day on GDs 7, 8, or 9. They also reported an excess in fetal absorptions and stillbirths
25 and a decrease in body length. Schwetz et al. (1979) reported no teratogenic effects but an
increase in minor skeletal variants in CF-1 mice exposed to 250 ppm CO for 24 h per day
from GDs 6 to 15.
10.5.4.2 Carbon Monoxide and Body Weight
30 One of the best studied and possibly one of the most sensitive measures of early CO
exposure is a depression in birthweight. The effect seen in animals following fetal CO
March 12, 1990 10-159 DRAFT-DO NOT QUOTE OR CITE
-------
exposure is generally transitory and occurs despite the fact that maternal body weight growth
through pregnancy does not appear to be adversely affected. The LOEL is in the range of
150 to 200 ppm in laboratory animals. In as much as the depressed birthweight observed is a
transient event, its significance is not clear. However, in humans, low birthweight babies
5 may be at particular risk for many other developmental disorders so that the effect cannot be
disregarded casually. Moreover, in humans there is a strong correlation between maternal
cigarette smoking and reduced birthweight. Whether the causative agent here is CO, nicotine,
or a combination of these or other agents is uncertain.
Studies relating human CO exposure from ambient sources or cigarette smoking to
10 reduced birthweight frequently have failed to take into account all sources of CO exposure.
Alderman et al. (1987), for example, studied the relationship between birthweight and
maternal CO exposure based upon neighborhood ambient CO data obtained from stationary
air monitoring sites in Denver. They failed to show a relationship between these factors, but
failed to control for maternal cigarette smoking or possible occupational exposures to CO.
15 COHb measurements were not made either among the mothers or their offspring to estimate
net exposure levels. A similar design problem is found in the study of Wouters et al. (1987)
where cord blood COHb and birthweight were correlated. The authors report a significant
correlation between cigarette smoking and reduced birthweight, but no correlation between
cord blood COHb and birthweight. Such data might be interpreted to mean that CO is not the
20 component in cigarette smoke responsible for reduced birthweight. Such a conclusion appears
to be unjustified based upon Wouters et al. (1987), because COHb is a good estimate of
recent CO exposure only. Thus it may indicate only how recently women in this study
smoked their last cigarette before delivery of the child rather than estimating smoking rates or
history throughout pregnancy.
25 Other studies have related indirect exposure to smoke in pregnancy with reduced
birthweights. Martin and Bracken (1986) showed an association between passive smoking
(exposure to cigarette smoke for at least 2 h per day) and reduced birthweight.
Unfortunately, side-stream smoke contains significant nicotine as well as CO and so it is not
possible to relate this effect to CO exposure.
30 Mochizuki et al. (1984) attempted to evaluate the role of maternal nicotine intake in
reduced birthweight and did present evidence of possibly impaired utero-placental circulation
March 12, 1990 10-160 DRAFT-DO NOT QUOTE OR CITE
-------
among smokers. These changes were not related specifically to the nicotine content of the
cigarettes and failed, moreover, to take into account the possible synergistic effects between
reduced perfusion that might have resulted from the vasoconstrictive effects of nicotine and
the reduced O2 availability that might have resulted from CO exposure. As noted in the
5 section of this report that deals with the effects of high altitude, many of the outcomes of
maternal CO exposure also are observed in offspring of women living at high altitude. These
include reduced birthweight, increased risk of perinatal mortality, and increased risk of
placenta! abnormalities. Limited data exist on the possibility of increased risk of CO
exposure to the fetus being carried at high altitude. Such findings are considered in the
10 section on high altitude.
Fechter and Annau (1980) replicated earlier data from their laboratory showing
significantly depressed birthweights in pigmented rats exposed throughout gestation to
150 ppm CO. Penney et al. (1980) also found a significant depression in birthweight among
rats exposed for the last 18 days of gestation to 200 ppm CO. Penney et al. (1983) showed a
15 trend toward divergence in bodyweight among fetuses exposed to 200 ppm CO, which
developed progressively during the last 17 days of parturition suggesting that late gestational
exposure to CO may be essential to observe the effect. Storm et al. (1986) reported that in
the following CO exposure from the beginning of gestation through postnatal day (PD) 10
that body weight was depressed in a dose-dependent fashion at 75, 150, and 300 ppm CO.
20 Moreover, these values were all significantly lower than air-control subjects. By age 21 days
no significant body weight differences were seen among the test groups. At no time have
Fechter and colleagues observed evidence of maternal toxicity as identified by death, reduced
maternal weight gain, or gross physical appearance. Morris et al. (1985b) exposed neonatal
piglets chronically to CO for 21 days starting at approximately 28 days of age (200- and 300-
25 ppm COHb levels averaged 16 and 21%, respectively). They reported a significant
impairment in weight gain in pigs exposed to 300 ppm, but no effect in pigs exposed to
200 ppm CO.
10.5.4.3 Alteration in Cardiovascular Development following Early Carbon Monoxide
30 Exposure
It is known that a variety of cardiovascular and hematopoietic changes can accompany
hypoxia in neonates and adult subjects including elevation in hemoglobin, hematocrit, and
March 12, 1990 10-161 DRAFT-DO NOT QUOTE OR CITE
-------
heart weight. Data gathered in adult laboratory animals suggest that these changes may be
related. Cardiomegaly resulting from hypoxia reflects the amount of work performed to
extract an adequate supply of O2. Whether the same processes occur in prenatal and neonatal
CO-induced hypoxia has been the subject of several reports (these reports are summarized in
5 Table 10-29). For prenatal exposure, the accumulated laboratory animal data suggest that
CO-induced cardiomegaly may be proportionately greater than in adult animals at a given
maternal CO-exposure level. Whether this is due to higher fetal COHb levels, as a
consequence of fetal hemoglobin's affinity for CO is not clear. While the cardiomegaly may
resolve when the neonatal subject is placed in a normal air environment, there is evidence for
10 a persisting increase in the number of muscle fibers. The functional significance of these
changes, if any, is uncertain. The LOEL for fetal cardiomegaly has not been well
determined. One experiment has shown significant elevation of heart weight following CO
exposures as low as 60 ppm (Prigge and Hochrainer, 1977) and there are no published dose-
response experiments that provide a NOEL. Chronic prenatal CO exposure at approximately
15 200 ppm results in a significant increase in the number of muscle fibers in the heart. The
NOEL for this change has not been determined.
Fechter et al. (1980) measured wet-and-dry heart weight and protein and nucleic acid
levels at several time points between birth and weaning in rats prenatally exposed to 150 ppm
CO or to air. They reported that neonates had significantly elevated wet-heart weights despite
20 a slightly reduced body weight at birth. Groups did not differ at birth in dry-heart weight,
total protein, or RNA or DNA levels in whole heart. No significant differences between
groups on any measure were present at PD 4 or subsequent ages studied. The data were
interpreted as evidence for cardiac edema rather than a change in heart muscle mass itself.
The finding of a heavier heart at birth replicated the finding of Prigge and Hochrainer (1977),
25 who exposed rats prenatally to CO at levels as low as 60 ppm and observed a similar increase
in heart weight. Clubb et al. (1986) have conducted a comprehensive experiment in which
rats were exposed to 200 ppm CO either prenatally from GD 7 (CO/air), prenatally and
postnatally until age 28 days (CO/CO), only postnatally for various durations (air/CO), or to
air (air/air). Subjects were sacrificed at different ages and ventricular wet and dry weights
30 and myocyte volume and number were measured histologically so that estimates of cell size
and cell number could be made. They observed increases in right ventricular weight due to
March 12, 1990 10-162 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 10-29. CONSEQUENCES OF PRENATAL CARBON MONOXIDE EXPOSURE ON CARDIOVASCULAR
&
jo
I— '
^£>
\O
O
^
<*
0
1
H
a
o
^
2
H
O
cj
O
w
._
M
r*»
UH V ULAJl'JVU:
Wet- Heart/
COHb Body Heart Body
Exposure (%) Weight Weight Weight
150 ppm 15 — Increased Increased
CO, CDs
1-21
230 ppm 24 Decreased ND Increased
CO, GD 24 PD 5
2-PD 21
60, 125, ND Decreased Increased Increased
250,
500 ppm
CO
157, 166, 24.9 Decreased Increased Increased
200 ppm ventricles
CO, CDs
5-22
200 ppm 27.8 Decreased Increased Increased
CO
30, 90 ppm 4.8-8.8 -
CO
200 ppm
CO from
GD 7-PD 28 ND - Increased Increased
at birth
GD 7-PD 1 Decreased Increased
at PD 28
1IN1 1IN J-.AJ5UKAHJKT KAld
Dry-
Heart Total
Weight Hematocrit Hemoglobin Other
Decreased ND ND Nucleic acid
protein
unchanged, no
significant
differences at
PDs 4-21
ND Increased Increased
PD 5 PD 5
ND Decreased, Decreased,
250-500 ppm 250-500 ppm
Increased — — Increased LDH M
subunit, increased
DNA content
ND - - No lasting effects
of prenatal
exposure
ND
ND ND Prenatal CO
increased myocytes
ND in right ventricle,
postnatal CO
References
Fechter et al.
(1980)
Hoffman and
Campbell
(1977)
Prigge and
Hochainer
(1977)
Penney et al.
(1983)
Penney et al.
(1980)
Garvey and
Longo
(1978)
Clubb et al.
(1986)
-------
TABLE 10-29 (cont'd). CONSEQUENCES OF PRENATAL CARBON MONOXIDE EXPOSURE ON CARDIOVASCULAR
DEVELOPMENT IN LABORATORY RATS
i
,_!
O
t— *
•^
O
"i>
3
j
i
0
0
^x
Q
H
O
COHb
Exposure (%)
PD 1-PD 28
500 ppm ND
CO PDs
1-32 (CO
gradually
increased
from
PDs 1-7)
300 and Approxi-
700 ppm mately
CO PDs 30 and
1-32 (CO 50%
gradually
increased
from
PDs 1-7)
500 ppm ND
CO PDs
1-32 (CO
gradually
increased
from
PDs 1-7)
Body
Weight
Decreased
at PD 21
and PD 28
—
Decreased
in 700-ppm
CO group
only in
adulthood
Decreased
only during
CO exposure
Wet-
Heart
Weight
Increased
at birth
Elevated in
adulthood
Increased
in 700-ppm
CO group
only in
adulthood
ND
Heart/
Body
Weight
Increased
Elevated in
adulthood
Increase
in 700-ppm
CO group
only in
adulthood
Increased
during
exposure
Dry-
Heart Total
Weight Hematocrit Hemoglobin Other References
ND increased myocytes
in left ventricle
ND Increase at ND Heart rate Penney et al.
some ages in elevated 10- (1984)
adulthood 15% in
adulthood.
No effect on
blood pressure
Decrease Increased ND No consistent Penney et al.
in dry/wet in females effect on heart (1988)
heart in adulthood rate at either
weight in exposure level.
females Elevated ventri-
cular DNA levels
in adulthood
ND ND ND Clubbetal.
(1989)
-------
!
cr
TABLE 10-29 (cont'd). CONSEQUENCES OF PRENATAL CARBON MONOXIDE EXPOSURE ON CARDIOVASCULAR
DEVELOPMENT DSf LABORATORY RATS
com
Exposure (%)
500 ppm ND
CO PDs
1-32 (CO
gradually
increased
from
PDs 1-7)
Body
Weight
Decreased
during CO
exposure
Wet-
Heart
Weight
Increased
during CO
exposure
Heart/ Dry-
Body Heart
Weight Weight
Increased ND
during CO
exposure
Total
Hematocrit Hemoglobin Other
Increased ND Exercise in
during CO adulthood
exposure increased
atrium to body
weight ratio.
CO in neonatal
period
produced small
additional effect
References
Penney et al.
(1989)
-------
fetal CO exposure, and increases in left ventricular weight following neonatal CO exposure.
As in the case of Fechter et al. (1980), they showed a gradual return to normal heart weight
when prenatally exposed neonates were placed in an air environment neonatally. They
attributed the reversal of cardiomegaly in the CO/air group to a loss in cell volume rather
5 than a loss in cell number (which remains elevated). Myocyte volume did not differ between
CO and air subjects at birth. Left ventricle plus septum and right ventricle cell volumes of
the CO/CO group were smaller than controls at 28 days of age despite the heavier wet heart
weight shown by the CO/CO subjects. Clubb et al. (1986) report that prenatal CO increased
right ventricular myocyte number and that neonatal CO exposure increased left ventricular
10 myocytes suggesting that cardiomegaly in early development is related to increased
hemodynamic load. This possibility is supported by reports from Penney et al. (1983)
showing that prenatal exposures to CO levels between 157 and 200 ppm during the last
17 days of gestation did lead to a significant elevation in DNA content among treated
subjects. Moreover, hydroxyproline content, an indicator of collagen, also was increased
15 following the CO exposure as was cardiac lactate dehydrogenase M (LDH M) subunit among
the 200-ppm CO subjects.
Penney et al. (1980) compared the effects of prenatal CO exposure at a dose of 200 ppm
with exposure both prenatally and neonatally until age 29 days. Neonatal CO concentrations
were elevated to 500 ppm. Cardiomegaly and depressed hemoglobin, hematocrit, and red
20 blood cell counts were found following CO exposure. In subjects allowed to survive until
young adulthood, the heart weight to body weight ratio of subjects receiving CO both
prenatally and neonatally still was elevated, while those in the prenatal CO condition did not
differ from control subjects in this measure. Penney and colleagues have published a series of
papers that propose that the neonatal period is a time when CO exposure might produce
25 persisting cardiovascular consequences. Typically their experimental protocol involves
exposure of rats to CO from soon after birth until PD 33. CO levels are increased in step-
wise fashion during the first week of neonatal life reaching the nominal CO exposure level by
PD 7. Spectrophotometric determination of COHb levels were reported in one manuscript
published by these authors to be approximately 30% for subjects exposed to peak CO values
30 of 350 ppm, 40% for subjects receiving 500 ppm, and 50% for those exposed to 700 ppm
CO. (Penney et al., 1988). The treatment produces significant reductions in body weight
March 12, 1990 10-166 DRAFT-DO NOT QUOTE OR CITE
-------
(Penney et al., 1988), elevated hematocrit (Penney et al., 1988; Penney et al., 1989), and
significant increases in heart weight (left ventricle plus septum and right ventricle) above
control subjects at the end of CO exposure to 350 ppm (Penney et al., 1989). The elevation
in heart weight partially recovers as subjects mature although in some studies (e.g., Penney
5 et al., 1984) persistent effects were observed into adulthood when neonatal CO levels were
500 ppm. In other studies, the elevation in heart weight or the heart weight:body weight ratio
resulting from 500-ppm CO exposure neonatally was no longer present in adulthood (e.g.,
Clubb et al., 1989). Penney et al. (1984) also suggested a 10-15% increase in adult heart
rates associated with neonatal exposure to 500 ppm CO, but this effect was not replicated
10 using 350- and 750-ppm CO exposure (Penney et al., 1988). Further, there was no evidence
for an increase in blood pressure or other functional changes that might explain the
tachycardia associated with 500-ppm CO exposure. Finally, a recent paper (Penney et al.,
1989) suggested possible additive effects of neonatal exposure to 500 ppm CO and exercise-
induced changes on adult heart size. Analysis of these effects are complicated by particularly
15 large effects of exercise on atrial weight rather than ventricular weight.
To summarize, there is good evidence for the development of severe cardiomegaly
following early life CO exposure at doses between 60 to 200 ppm. These effects are
transitory if exposure is prenatal and it is not clear whether they alter cardiac function or
produce latent cardiovascular effects that may become overt later in life. Persisting elevation
20 in heart weight results from combined prenatal CO exposure at 200 ppm and neonatal
exposure at 500 ppm. The LOEL for this effect has not been determined.
There are many published reports suggesting some residual increase in heart weight
associated with neonatal CO exposures of 500 ppm and greater, maintained over the first
33 days of life. Even granting a small (10-15%) increase in heart rate found in one study
25 among subjects exposed to 500 ppm CO neonatally, there is no evidence that neonatal CO
exposure has functional consequences for experimental subjects.
10.5.4.4 Neurobehavioral Consequences of Perinatal Carbon Monoxide Exposure
The Developmental Toxicology Guidelines published by the U.S. Environmental
30 Protection Agency (Federal Register 1986) recognized the importance of neurobehavioral
investigations as a means of assessing nervous system function. Behavior is an essential
March 12, 1990 10-167 DRAFT-DO NOT QUOTE OR CITE
-------
function of the nervous system and abnormalities in this outcome can be diagnostic for
particular neurological disorders or for nervous system dysfunctions. The LOEL for such
effects appears to be in the range of 125 to 150 ppm using a variety of behavioral tasks in
experimental animals, though isolated studies suggest possible anomalies in the range of 60 to
5 65 ppm. These studies are summarized in Table 10-30. There are a limited number of
human case reports that also are described here (see Table 10-31 for summary). However,
those reports generally are not adequate for evaluating a threshold for persisting
neurobehavioral impairments in children.
Crocker and Walker (1985) reported on the consequences of acute CO exposure in 28
10 children of which 16 had COHb levels over 15% and were considered to have had
"potentially toxic" COHb levels. These children were between the ages of 8 mo and 14
years. The authors report nausea, vomiting, headache, lethargy, and syncope to be the most
common signs and symptoms. A very limited follow-up investigation was performed with
these children and so no conclusions can be drawn from this work concerning persisting
15 effects. In addition to very large differences in the nature (dose and duration) of exposure,
the extreme variability in patient age limits the potential value of the data presented in this
work. The absence of any reports from children having COHb levels of 15% or less (a very
significant COHb level) is regrettable because these are the children one must study to
develop an understanding of the relationship between dose and effect for the purpose of
20 setting standards for ambient air.
Klees et al. (1985) conducted a more comprehensive study of the consequences of
childhood CO exposure on subsequent behavioral development. They report that the age
atwhich exposure occurred, its severity, and also the child's intellectual level at the time of
exposure also play a role in the outcome. Younger children tended to show somewhat milder
25 symptoms if they did recover fully than did children who were older at the time of CO
exposure. Subjects who had higher intellectual function prior to accidental exposure also
appeared to fare better after CO exposure. However, the authors stress that long-term
perceptual and intellectual consequences of CO exposure may occur that are not identified
well in short-term cursory examinations. Of 14 children followed up for 2 to 11 years after
30 intoxication only 1 showed no sequelae (despite COHb levels of 42% on admission to hospital
at the age of 9 years 10 mo). Seven children had impairment of visual memory and
March 12, 1990 10-168 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 10-30. NEUROBEHAVIORAL CONSEQUENCES OF PRENATAL CARBON MONOXIDE
EXPOSURE IN LABORATORY ANIMALS
Species (strain)
Maternal/Neonatal Treatment
Maternal/Embryonic Toxicity
Developmental Abnormality
References
Rat
(Sprague-Dawley)
Rat
(Long-Evans)
Rat
(Long-Evans)
10,000 ppm CO for 2 or 3 h on
GD 15; no cross-fostering
150 ppm CO throughout
gestation; no cross-fostering
150 ppm CO throughout
gestations; no cross-
fostering
Acute effects: loss of righting
reflex followed by coma. Litter
size normal; COHb levels of
approximately 50%
No difference in litter size or
fetal mortality; COHb levels
of 12.2-14%
Litter size normal; no
differences in neonatal
mortality; COHb levels of 15%
26% increase in exploratory
activity in figure-8 maze at PD 30
(3-h exposure)
3.3 % decreased birth weights and
decrease in preweaning weights;
decreased locomotor response to
L-dopa in open field (PD 4 and PD 14);
increased rate of habituation
(PD 14)
4.9% decreased birth weights and
decrease in preweaning weights;
decreased response to L-dopa (in
open field) at PD 1, PD 4 (also
decreased dopamine levels);
increase in rate of habituation
of activity (PD 14)
Daughtrey and Norton
(1983)
Fechter and Annau
(1976)
Fechter and Annau
(1977)
Mouse
(Swiss-Webster)
Rat
(Long-Evans)
CO exposure throughout
gestation; no cross-
fostering
CO exposure throughout
gestation; no cross-
fostering
ND
Maternal COHb levels of
6-11%
Maternal weight gain,
gestation length, and
litter size normal;
COHb levels of 15.6%
Increased errors in heat-
motivated Y-nuze at PD 40
Decreased acquisition and (24-h)
retention of two-way active
avoidance (PD 30); neither multiple
measures nor use of pseudo-
conditioning controls similarly
affected
AbbatieDoandMohmann
(1979)
Mactutui and Fechter
(1984)
-------
I
TABLE 10-30 (cont'd). NEUROBEHAVIORAL CONSEQUENCES OR PRENATAL CARBON MONOXIDE
EXPOSURE IN LABORATORY ANIMALS
Species (strain)
Maternal/Neonatal Treatment
Maternal/Embryonic Toxic ity
Developmental Abnormality
Rat
(Long-Evans)
150 ppm CO throughout
gestation; cross-
fostering for weight
measures
ND
(no COHb levels)
(PD 120), but minimal (24-h)
and pronounced (28-day)
decreased retention; decreased
acquisition and retention of
two-way avoidance (PDs 300-360)
7.6% decrease in birth weights
and decreased preweaning weights;
decreased negative geotaxis (PD 3);
decreased homing behavior (PDs 3-5)
References
Rat
(Long-Evans)
CO exposure throughout
gestation; no cross-
fostering
ND
COHb levels of 15.6%
Normal two-way avoidance
acquisition with moderate or
difficult task requirements
Mactutus and Fechter
(1985)
Fechter and Annau
(1980)
-------
I
TABLE 10-31. CONSIJQUHNCliS OF HUMAN CARBON MONOXIDE INTOXICATION DURING
HARLY DEVELOPMENT
Characterization of
Exposure1
Approximate
COHb Level
Immediate
Symptoms and Their Frequency
Persisting
Symptoms and Their
Frequency After Hyperbaric
O2 and Normobaric O2 Therapy
References
%-^
£
o
o
2;
3
§
g
n
a
Light
Medium
Severe
Accidental at 13 weeks
of age
Accidental at 21 days'
old
28 pediatric exposures
4-27%
6-36%
37%
60%
6.7% 4 h after exposure
(> 15 % at time of removal
from CO)
Threshold value at which
symptom was first observed in
any subject
15%
16.7%
19.8%
18.6%
24.5%
36.9%
Hyperreflexia (1/3) auditive
memory impairment and spatial
orientation problems (1/3)
ND (6/12)
Coma (1/12)
Unconscious (1/12)
Normal (2/12)
Coma-developmental level
Regression (language and
motor)
Violent anger/nervousness
(1/1)
Convulsion, hypotonic,
unconscious (1/1)
Lethargy, vomiting (1/1)
Asymptomatic
Nausea/headache
Vomiting
Lethargy
Visual symptoms/syncope
Seizures
Auditive and visual memory impairment
(1/3)
Anxiety or emotional instability 0/12)
Memory impairment (2/12)
Spatial/temporal disorganization
and perceptual problems (3/12)
None or questionable effect (4/12)
Persistent emotional instability
(1/1)
Recovery of minor neurologic
deficits by 6 weeks (1/1)
None
Headaches, memory deficit, decline
in school performance (3/28)
Kleei et al. (1985)
Kleet et «1. (1985)
Kleei et al. (1985)
Venningetal.(1982)
O'Sullivan (1983)
Crocker and Walker
(1985)
-------
o
TABLE 10-31 (cont'd). CONSEQUENCES OF HUMAN CARBON MONOXIDE INTOXICATION DURING
EARLY DEVELOPMENT
) .
0
5
o
jg
6
o
z
3
0
G
3
tn
O
K
n
H
W
Persisting
Symptoms and Their
Characterisation of Approximate Immediate Frequency After Hyperbaric
Exposure1 COHb Level Symptoms and Their Frequency O2 and Normobaric O2 Therapy
Light (6/14) 19-42% Somnolence (2/6) Perceptual (2/6)
Headache/nausea (3/6) Memory (3/6)
Emotional (1/16)
Psychomotor (1/16)
Cognitive (3/6)
Medium (5/14) 16-42% Incontinent (1/5) Perceptual (1/5)
Unconscious (4/5) Memory (1/5)
Emotional (2/5)
Cognitive (2/5)
Severe (3/14) ?-13% Vigil coma (3/3) Perceptual (2/3)
Cognitive (3/3)
Notes: Exposure duration and level of CO exposure are poorly defined in all studies. Exposures are grouped according to authors' descriptive characterization when
level. The latter varies widely with group.
See glossary of terms and symbols for abbreviations and acronyms.
Reference*
Klee* et al. (1985)
Klees et al. (1985)
Klees et al. (1985)
available rather than COHb
-------
concentration, but normal IQ scores. These children had "slight or medium" exposure
(COHb levels in the low to mid-twenties) and no coma. The six children with serious
learning disorders did not have more severe CO exposures as judged from thek COHb levels.
They include several cases where exposures did occur at a young age and in children who had
5 psychological difficulties prior to CO exposure. This study leaves some question concerning
the relative vulnerability of children to CO as a function of their age as several of the
youngest children did make full recovery while others did not. It seems likely that the child's
age may have an influence on the duration of CO exposure which is survivable and perhaps
also on the promptness with which either hospitalization or measurement of COHb levels is
10 made. Further study of the outcomes of childhood CO exposures will be useful in
determining whether there are differences with respect to vulnerability to CO level. Yenning
et al. (1982) report on a case of acute CO poisoning in a 13-week-old baby who had
profoundly elevated COHb levels (60% 2 h after removal from the automobile in which she
accidentally was exposed to CO). Her parents had much lower COHb values though this may
15 reflect differences in concentration of CO inhaled. The child was reported to be unconscious
for 48* h, to go through convulsions over the next 18 days, but, again, to show recovery from
"minor neurological abnormalities" by 6 weeks later. There have been a series of
experiments reported in rodents that identify both persistent neurobehavioral effects of
prenatal CO exposure and also transient effects that may be symptomatic of functional delays
20 in development. Fechter and Annau (1980) reported delays in the development of negative
geotaxis and homing in rats exposed prenatally to 150 ppm CO (maternal COHb levels were
not reported in this paper, but levels previously reported in this laboratory under that
exposure regimen are 15 to 17% [Fechter and Annau, 1977]). These data were replicated by
Singh (1986) using CD-I mice exposed from GD 7 to 18 to 0, 65, and 125 ppm CO. He
25 found that exposure at 125 ppm significantly impaired the righting reflex on PD 1 and
negative geotaxis on PD 10. He also reported impaired aerial righting among subjects
exposed prenatally to 65 or 125 ppm CO. Morris et al. (1985a) studied the consequences of
moderate CO exposure given very late in gestation. They exposed pigs to 200 and 250 ppm
CO (COHb levels of 20 and 22%) from GD 109 until birth. They found impairment of
30 negative-geotaxis behavior and open field activity 24 h after birth in pigs exposed to
250 ppm. Activity in the open field was significantly reduced in subjects exposed to both 200
March 12, 1990 10-173 DRAFT-DO NOT QUOTE OR CITE
-------
and 250 ppm 48 h after birth. The significance of these behavioral dysfunctions is that they
point to delays in behavioral development that may themselves contribute to impairments in
the way in which the individual interacts with its environment.
There also are reports of impaired cognitive function produced by prenatal CO which
5 may be related to permanent neurological damage. Mactutus and Fechter (1984) showed
poorer acquisition and retention of a learned active avoidance task in rats of 30 to 31 days of
age that had received continuous prenatal exposures of 150 ppm CO. This study is
noteworthy because very careful efforts were made to distinguish cognitive deficits and
performance deficits such as motivational, emotional, or motoric factors. These findings
10 were replicated and extended by Mactutus and Fechter (1985). They studied the effects of
prenatal exposures to 150 ppm CO (16% maternal COHb) on learning and retention in
weanling, young adult, and aging (1-year-old) rats. They found that both the weanling and
young adult rats showed significant retention deficits, while in aging adults impairments were
found in both learning and retention relative to control subjects. They interpreted these
15 results to mean that there are permanent neurological sequelae of prenatal CO exposure.
They raise the important issue that sensitivity of tests for consequences of early toxic exposure
may reflect the developmental status of the test subject and complexity of the task. In this
case, a learning impairment not observed in the early adult period was detected by working
with aged subjects. No systematic attempts have been made to replicate these effects using
20 lower levels of CO. One earlier study (Abbatiello and Mohrmann, 1979) suggested an
increase in the number of errors made by mice prenatally exposed to CO throughout gestation
(maternal COHb levels were 6 to 11%) and required to learn a maze discrimination task at
6 weeks of age. The absence of many details concerning the manner in which the control
subjects were handled during pregnancy and the absence of details in the exposure protocol
25 make it difficult to draw firm conclusions from this paper.
10.5.4.5 Neurochemical Consequences of Prenatal and Perinatal Carbon Monoxide
Exposure
A significant number of studies have appeared concerning the consequence of acute and
30 chronic prenatal and perinatal CO exposure upon a variety of neurochemical parameters.
These experiments are important because the transmission of information between nerve cells
is based upon neurochemical processes. Neurotransmitters can sometimes serve as markers
March 12, 1990 10-174 DRAFT-DO NOT QUOTE OR CITE
-------
for the development of specific neurons in the brain, thereby serving as a sensitive alternative
to histopathologic investigation particularly when the toxic agent selectively lesions neurons
based upon a biochemical target. The absence of a specific cell group identified by
neurochemical methods may have important consequences for subsequent brain development
5 because the absence of targets for synapse formation can have additional consequences on
brain development. Altered neurochemical development has been observed at CO-exposure
levels lower than those necessary to produce signs of maternal toxicity or gross teratogenicity
in the neonates. Chronic prenatal and perinatal exposures to 150-300 ppm CO have been
shown to yield persisting alterations in norepinephrine, serotonin, and GABA levels and in
10 GABA uptake in rats. There also are a substantial number of acute exposure studies that have
demonstrated neurochemical effects of CO. However, these generally have been conducted at
life-threatening levels and are not particularly relevant to setting ambient air standards for
CO.
Storm and Fechter (1985a) and Storm et al. (1986) have carefully studied the developing
15 cerebellum because this structure shows a rather slow developmental pattern and has been
shown to be sensitive to hypoxic injury. The cerebellum plays prominent roles in many
diverse functions. It is a part of the extra-pyramidal motor system, and it plays an important
role in maintaining balance. The cerebellum also receives diverse sensory inputs and plays a
role in sensory-motor integration. The cerebellar cortex contains a diverse group of neurons
20 whose organization has been studied very well. The intrinsic neurons of the cerebellum -
those having their cell bodies and axonal processes within this structure - consist of the
granule, pyramidal, stellate, basket, and Golgi cells. The granule cells use the excitatory
amino acid, glutamate, as their neurotransmitter and synapse on the pyramidal cells. The
other intrinsic cells of the cerebellum appear to use the inhibitory amino acid, GABA, as their
25 neurotransmitter. The Purkinje cells, being extremely large in size, probably contribute
considerably to the total GABA levels found in the cerebellum.
The cerebellum receives several different neurotransmitter inputs from other brain
regions. Most important of these is a noradrenergic input from the brainstem, a cholinergic
link via mossy fibers and possibly aspartate or glutamate climbing fibers.
30 Storm and Fechter (1985a) reported that chronic prenatal CO exposures of 150 ppm CO
(approximately 16 to 18% COHb based upon other research in this laboratory) decreased
March 12, 1990 10-175 DRAFT-DO NOT QUOTE OR CITE
-------
cerebellar wet weight, but increased norepinephrine levels in this structure when expressed
either in terms of concentration (nanograms per milligram wet weight) or total cerebellar
content above control values between the ages of 14 to 42 days. This period represented the
duration of the experiment. While this persisting elevation in norepinephrine cannot be
5 considered permanent, it is the case that rats do obtain normal adult values of monoamine
neurotransmitters at about the age of 40 to 45 days. There was no effect of CO treatment on
norepinephrine levels in the cerebral cortex. Since noradrenergic neurons have their cell
bodies outside of the cerebellum and project axons that terminate on cell bodies in this
structure, Storm and Fechter's data may reflect an effect of increased noradrenergic
10 innervation secondary to toxic injury to target neurons in the cerebellum. Consistent with this
hypothesis, Storm et al. (1986) reported deficits in cerebellar weight, but more importantly
deficits in markers of GABA-ergic activity in the cerebellum following prenatal and perinatal
CO exposures. GABA is thought to be an inhibitory neurotransmitter present in several
neurons that are endogenous to the cerebellum. Subjects in this experiment received 0, 75,
15 150, and 300 ppm CO (corresponding maternal COHb levels were 2.5, 11.5, 18.5, and
26.8%) from the beginning of gestation until PD 10. Neurochemical measurements were
made either on PD 10 or PD 21. They showed reduced total GABA levels in the cerebellum
following either 150- or 300-ppm CO exposure at both measurement times. They also
reported a significant reduction in total GABA uptake, but not glutamate uptake in
20 synaptosomes prepared from cerebella of 21-day-old neonates. Glutamate is an excitatory
neurotransmitter found within the cerebellum. Histological markers of cerebellar toxicity also
were obtained that were compatible with the neurochemical data and these are described
below under histopathology.
In a subsequent paper, Storm and Fechter (1985b) evaluated norepinephrine and
25 serotonin levels at PDs 21 and 42 in four different brain regions (pons/medulla, neocortex,
hippocampus, and cerebellum) following chronic prenatal exposures to 75, 150, and 300 ppm
CO. They reported that norepinephrine and serotonin concentrations decreased linearly with
dose in the pons/medulla at 21 but not 42 days of age (i.e., evidence of a transient effect) the;
LOEL was 150 ppm. Norepinephrine increased linearly with CO dose in neocortex at 42, but
30 not at 21 days of age. They also showed that cerebellar weight was significantly reduced at
March 12, 1990 10-176 DRAFT-DO NOT QUOTE OR CITE
-------
150 and 300 ppm CO when measured on PD 21 and for the 300-ppm-exposed rats at 42 days
of age.
10.5.4.6 Morphological Consequences of Acute Prenatal Carbon Monoxide
5 Profound acute CO exposures do result in obvious neurological pathology that can be
predicted to be inconsistent with life or with normal neurological development. These data
are not as relevant to setting standards for ambient air quality as they are in demonstrating
both the danger of accidental high-level CO exposures and in providing possible insight into
the susceptibility of the developing brain to toxic exposure. The one possible exception is a
10 study conducted in fetal pigs exposed via the mother to 300 ppm CO for 96 h (Dominick and
Carson, 1983). The authors reported quite marked sensitivity to the CO as reflected in
fetotoxicity, but also identified multifocal hemorrhages and vacuolation of the neuropile
throughout the cortical white matter and brain stem. They also observed cerebellar edema
with swollen oligodendrocytes and astxocytes, two non-neuronal cell types that have important
15 roles in supporting neural function.
Okeda et al. (1986) studied the effects of 2,000- to 3,000-ppm CO exposure for 76 to
150 min in cats of different gestational ages. They suggest a different pattern of neurological
damage in cats exposed late versus those exposed early to the CO. In late gestation, the
primary changes were seen in cerebral white matter and brain stem. Basal ganglia and
20 thalamus were affected less and cerebral cortex even less affected. Kittens exposed to CO at
an early gestational age show less histopathology than those exposed later. Cerebral white
matter and the basal ganglia tended to be most affected by early CO exposure.
Daughtrey and Norton (1982) studied the effect of exposing pregnant rats (GD 15) to
1,000 ppm CO exposure for 3 h upon central nervous system development of the fetuses on
25 GD 16. Estimated maternal COHb levels reached about 50%. They reported 13 to 28%
fetotoxicity (lethality) and described hemorrhagic infarcts and the most consistent damage to
the ventral germinal matrix overlying the caudate nucleus. Further study (Daughtrey and
Norton, 1983) indicated damage to the dendritic branches of Golgi type II neurons in the CO
exposed fetuses.
30 Storm et al. (1986) showed that exposure of rats to 300 ppm CO (maternal COHb levels
of 26.8%) throughout gestation and until day 10 after gestation resulted in a noticeably
March 12, 1990 10-177 DRAFT-DO NOT QUOTE OR CITE
-------
smaller cerebellum at PD 21. The cerebellum of exposed neonates had fewer fissures than
normal controls.
10.5.5 Summary
5 The data reviewed provide strong evidence that CO exposures of 150 to 200 ppm
produce reductions in birthweight, cardiomegaly, delays in behavioral development, and
disruption in cognitive function in laboratory animals of several species. Isolated experiments
suggest that some of these effects may be present at doses as low as 60 to 65 ppm maintained
throughout gestation. The current data from human children suggesting a link between
10 environmental CO exposures and SIDS are weak, but further study should be encouraged.
Human data from cases of accidental high dose CO exposures are difficult to use in
identifying a LOEL or NOEL for this agent because of the small numbers of cases reviewed
and problems in documenting levels of exposure. However, such data if systematically
gathered and reported could be useful in identifying possible ages of special sensitivity to CO
15 and cofactors or other risk factors that might identify sensitive subpopulations.
10.6 OTHER SYSTEMIC EFFECTS OF CARBON MONOXIDE
Studies (see Table 10-32) reviewed in the previous criteria document (U.S.
20 Environmental Protection Agency, 1979) and again in Chapter 9 of this document suggest that
enzyme metabolism and the P-450-mediated metabolism of xenobiotic compounds may be
affected by CO exposure (Montgomery and Rubin, 1971; Kustov et al., 1972; Pankow and
Ponsold, 1972, 1974; Martynjuk and Dacenko, 1973; Swiecicki, 1973; Pankow et al., 1974;
Roth and Rubin, 1976a,b). Most of the authors have concluded, however, that effects on
25 metabolism at low COHb levels (<15%) are attributable entirely to tissue hypoxia produced
by increased levels of COHb because they are no greater than the effects produced by
comparable levels of hypoxic hypoxia. At higher levels of exposure, where COHb
concentrations exceed 15 to 20%, there may be direct inhibitory effects of CO on the activity
of mixed-function oxidases but more basic research is needed (see Chapter 9, Section 9.4).
30 The decreases in xenobiotic metabolism shown with CO exposure might be important to
March 12, 1990 10-178 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 10-32. OTHER SYSTEMIC EFFECTS OF CARBON MONOXIDE
o
sr
J°
»—*
^o
v^5
0
i
>— »
-j
VO
O
£•
^Tl
>-3
I
O
o
z
o
H
d
O
0
n
H
w
Exposure*'1"
Increasing exposure
to 3000 ppm at
100 days
0.8 or 3.0%
until death
group
250, 500, and
1000 ppm for 24 h
Accidental
exposure
50 ppm
17 ppm
250-3000 ppm
for 90 min
50 ppm for
3 mo
Subcutaneous CO
at 7.2 and 9.6
mol/fcg; 40
injections in
S3 days
COIIb0 Subject(s)
ND Rat
n = 36-45
71.3% or Rat
79.2% n = 5 per
group
ND Rat
n = 36
3.4% to Man
32% n = 6
ND Rat
n = 92
ND Rat
n = 95
20-60% Rat
n = 10-20
per group
ND Rat
n = 100;
Rabbit
n = 40;
Dog
n = 4
50% Rat
n = 20-30
Observed Effects'1
Slightly few weight gains during first
100 days; increased weight gain in last
200 days
Increased plasma levels of leucine aminopepti-
dase. No change in state-4 respiration of
mitochondria; decreased state-3 rate in CO-
exposed rats
Decreased food and water intake; decreased
weight gain
Increased serum phosphocreatine-kinase
Decreased liver cytochrome oxidase; increased
liver succinate dehydrogenase
Increased aspartate and alanine amino
transferase activity
Prolonged response to hexobarbital at
1000 ppm and to zoxazolamine at 250 ppm
No effect on body weight
Increased leucine aminopeptidase
activity in the liver with single
and repeated injections; increased
liver weight with repeated injections
Conclusions
No significant body weight
effect
Acute CO poisoning caused
damage to liver mitochondria
Significant body weight
effect
Diffuse myolysis indicative
of acute renal failure
Tissue hypoxia
Tissue hypoxia
Decreased xenobiotic
metabolism
No significant body weight
effect
Reference
Campbell (1934)°
Katsumata et al.
(1980)
Koob et al.e
(1974)
Kuska et al.
(1980)
Kustov et al.e
(1972)
Martynjuck and0
Dacenko (1973)
Montgomery and0
Rubin (1971)
Musselman0
et al. (1959)
Pankow and
Ponsold (1972,
1974)e
-------
eren
Obse
8
Pank
Ponso
Pank
(1974
ope
leucine aminp
the liver; enlarge
ethanol
CO exposure
combined with
1% ethanol
c
I
g
-s.s
.2
1
CO
o
a|
•B-s
li
X J3
£.a
og.
a£
^
"S
H
Dec
'
« " S3
A c a.
1
e
I
1.
II
"SCO
Sa!?
18"
11
o£
ii
X *
Jjp-a
r; C
rate of
live
e h
11
•s
Q
J3
1
g
I§
e S
P5 It
ill Is
3*
"•5.
fe -5
v.
£§£
'
s*
* — c
E M'"
3 C US
li*
o .S3 c
on o
Q.S5 £
•Oc
li!
•3 II h
o C S.
Q
IN S
-
o *o «->
**..>,
•o c .•=
1
I
1
e
I
'i
•3 |
Z u
,
"8
JO
s
8
•a
a;
rgic syste
I
5
i
n,
1
o
•o
1
.
-s ts
S
*• n
00
u II
2 e
§ B I a s
« 8
E
e
1
-
£
£
e
0 0, T
-------
individuals receiving treatment with drugs. The implications of this effect are discussed in
Chapter 12, Section 12.3.
The effects of CO on tissue metabolism noted above may partially explain the body
weight changes associated with CO. Short-term exposure to 250-1000 ppm for 24 h was
5 reported previously to cause weight loss in laboratory rats (Koob et al., 1974) but no
significant body weight effects were reported in long-term exposure studies in laboratory
animals at CO concentrations ranging from 50 ppm for 3 mo to 3000 ppm for 300 days
(Theodore et al., 1971; Musselman et al., 1959; Campbell, 1934; Stupfel and Bouley, 1970).
It is quite probable that the initial hypoxic stress resulted in decreased weight gain followed
10 by compensation for the hypoxia with continued exposure by adaptive changes in the blood
and circulatory system (see Section 10.3). It is known, however, that CO-induced hypoxia
during gestation will cause a reduction in the birthweight of laboratory animals. While a
similar effect has been difficult to demonstrate in humans exposed to CO alone, there is a
strong correlation between maternal cigarette smoking and reduced birthweight. (See
15 Section 10.5 for a more complete discussion of fetal effects of CO exposure.)
Inhalation of high levels of CO, leading to COHb concentrations greater than 10 to
15%, have been reported to cause a number of systemic effects in laboratory animals as well
as effects in humans suffering from acute CO poisoning. Tissues of highly active oxygen
metabolism, such as heart, brain, liver, kidney, and muscle, may be particularly sensitive to
20 CO poisoning. The impairment of function in the heart and brain caused by CO exposure is
well known and has been described in other sections of this chapter. Other systemic effects
of CO poisoning are not as well known and are, therefore, less certain. There are reports in
the literature (see Table 10.6-1) of effects on liver (Katsumata et al., 1980), kidney (Kuska
et al., 1980), and bone (Zebro et al., 1983). Results from one additional study in adult
25 guinea pigs suggest that immune capacity in the lung and spleen was affected by intermittent
exposure to high levels of CO for 3 to 4 weeks (Snella and Rylander, 1979). It generally is
agreed that these systemic effects are caused by the severe tissue damage occurring during
acute CO poisoning due to (1) ischemia resulting from the formation of COHb, (2) inhibition
of O2 release from HbO2, (3) inhibition of cellular cytochrome function (e.g., cytochrome
30 oxidases), and (4) metabolic acidosis.
March 12, 1990 10-181 DRAFT-DO NOT QUOTE OR CITE
-------
The effects of CO on visual acuity and dark adaptation, caused primarily by CNS
alterations, were described previously (see Section 10.4). Besides central effects, however,
acute CO exposure can cause damage at the sensory end-organ level. Observed ocular effects
from acute CO poisoning range from retinal hemorrhages (Dempsey et al., 1973; Kelley and
5 Sophocleus, 1978) to blindness (Duncan and Gumpert, 1983; Katafuchi et al., 1985). In
addition, peripheral neuropathy and tortuous retinal vessels have been described after chronic,
intermittent exposure to low levels of CO over a 16-mo period (Trese et al., 1980). The
authors of the latter report speculated that increased blood flow from low-level, chronic
exposure to CO may lead to the development of a compensatory retinal vascular tortuosity.
10 With high-level, acute exposures to CO, the compensation will not take place and localized
vascular hemorrhages result.
Finally, exposure to CO has been associated with direct and indirect mutagenic activity
(van Houdt et al., 1987), as measured by the Salmonella/microsome test (Ames et al., 1975).
When tested in human populations living in the San Francisco Bay area, however, no
15 significant association was found between ambient levels of CO and site-specific cancer
incidence (Selvin et al., 1980). A case-control study of persons with newly diagnosed
multiple myeloma living in four geographical locations of the United States found an
increased risk for self-respondents reporting an exposure to combustion products including CO
(Morris et al., 1986). The difference in number of cases was small when compared to
20 geographically matched controls and the exposure was not defined well. Giver; the limited
information available at this time, it is unlikely that exposure to CO would contribute
significantly to the development of cancer in nonoccupationally exposed individuals.
25 10.7 ADAPTATION, HABITUATION, AND COMPENSATORY
RESPONSES TO CARBON MONOXIDE EXPOSURE
This section considers whether exposure to CO eventually will lead to the development
of physiological responses that tend to offset some of the deleterious effects. While there is
possibly a temporal continuum in such processes, in this review the term "adaptation" will be
30 used to refer to long-term phenomena, and the term "habituation" will refer to short-term
processes. Allusions will be made, where possible, to the physiological chain of events by
March 12, 1990 10-182 DRAFT-DO NOT QUOTE OR CITE
-------
which adaptation and habituation come about, but extensive reductive explanations will be
avoided. The term "compensatory mechanism" will be used to refer to those physiological
responses that tend to ameliorate deleterious effects, whether in the long-term or short-term
case.
5
10.7.1 Short-Term Habituation
Arguments have been made for the possibility that there exist short-term compensatory
mechanisms for CO exposure. These hypothetical mechanisms have been (1) based upon
physiological evidence, and (2) used to account for certain behavioral findings reported in the
10 literature.
There is physiological evidence for responses that would compensate for the deleterious
effects of CO in a very short time span. As discussed in Section 10.4, CO exposure has been
demonstrated to produce an increased cerebral blood flow which is apparently produced by
cerebrovascular vasodilation. It also has been shown (Doblar et al., 1977; Miller and Wood,
15 1974; Traystman, 1978; Zorn, 1972), however, that the tissue PO2 values for various CNS
sites fall in proportion to COHb, despite the increased blood flow. Apparently, the PO2
values would fall considerably more without the increased blood flow. Although the
published graphs of these data do not show very short time intervals, it appears that tissue PO2
falls immediately and continuously as COHb rises. Although there is no evidence for time
20 delays or for threshold effects in these data, it is noteworthy that only very high CO levels
were employed. Thus, the saturation rates were high, and time lags or thresholds would be
difficult to detect.
As discussed in Section 10.3, both coronary blood flow and O2 extraction in the
peripheral musculature increase as COHb rises. These, too, are compensatory mechanisms,
25 but mechanisms that have been shown to be only partly effective. None of the studies present
evidence of time lags or threshold effects, because only terminal or near-asymptotic values
were reported.
The behavioral work upon which short-term habituation hypotheses have been predicated
are mostly human studies, where CO exposure at very low levels or at very early exposure
30 times (well before asymptotic saturation) have shown performance decrements that were not
apparent with higher or longer exposures (Section 10.4). Depending upon the particular
March 12, 1990 10-183 DRAFT-DO NOT QUOTE OR CITE
-------
version, the habituation hypothesis holds that there might exist some threshold value of CO
below which no compensation would be initiated, or that there might be some time lag in the
compensatory mechanism so that the early effects of CO exposures would later subside. The
behavioral data to support this contention have been controversial and need to be examined
5 closely.
Most of the hypotheses about compensatory mechanisms were based, however, upon
post hoc reasoning to explain empirical findings, not upon results form experiments to test the
existence or nature of such mechanisms. Disregarding hypothesized time lags and thresholds,
without the compensatory mechanisms CO would apparently have even more deleterious
10 effects and the threshold for such effects would be lower.
10.7.2 Long-Term Adaptation
Adaptation is an all-inclusive term that incorporates all of the acute or chronic
adjustments of an organism to a stressor. It does not indicate (or predict) whether the
15 adjustments are initially or eventually beneficial or detrimental. Acclimatization is an
adaptive process that results in reduction of the physiological strain produced by exposure to a
stressor. Generally, the main effect of repeated, constant exposure to the stressor is
considered to result in an improvement of performance or a reduced physiological cost. Both
of these phenomenon tend to exploit the reserve potential of the organism.
20 It generally is agreed that adaptation to lowered levels of oxygen tension and oxygen-
carrying capacity can occur with continued hypoxic exposure. This is evident especially in
healthy individuals living for lengthy periods of time at high terrestrial altitudes. It should be
noted, however, that there is no assurance that individuals moving from low altitudes to
higher ones will attain the physiological status to the higher altitudes that is observed in
25 natives (viz. natives of the Andes and Himalayas). Prominent features of prolonged altitude
exposures are increases in hemoglobin concentration and hematocrit. Additional alterations
are right ventricular hypertrophy, pulmonary artery vasoconstriction, possible changes in
cardiac output, and increased blood volume due to increases in the red cell mass.
Whether or not adaptation can occur in individuals chronically exposed to various
30 ambient concentrations of CO remains unresolved. Concern for CO intoxication in England
and Scandinavia led to the speculation that adaptational adjustments could occur in man (Grut,
March 12, 1990 10-184 DRAFT-DO NOT QUOTE OR CITE
-------
1949; Killick, 1940). These concerns were directed to situations where high ambient CO
concentrations were present. There are only a few available studies conducted in humans.
Killick (1940), using herself as a subject, reported that she developed acclimatization as
evidenced by diminished symptoms, slower heart rate, and the attainment of a lower COHb
5 equilibrium level following exposure to a given inspired CO concentration. Interestingly,
Haldane and Priestly (1935) already had reported a similar finding as to the attainment of a
different COHb equilibrium following exposure, to a fixed level of CO in the ambient air.
Killick (1948) repeated her CO-exposure studies in an attempt to obtain more precise
estimations of the acclimatization effects she had noted previously. The degree of
10 acclimatization was indicated by (1) a diminution in severity of symptoms during successive
exposure to the same concentrations of CO, and (2) a lower COHb level after acclimatization
than that obtained prior to acclimatization during exposure to the same concentrations of
inhaled CO.
Before using herself as a subject, Killick (1937) studied the effects of CO on laboratory
15 animals. Mice were exposed to successively higher concentrations of CO, which in a period
of 6 to 15 weeks reached levels of 2300 to 3275 mg/m3 (2000 to 2850 ppm) CO and produced
60 to 70% COHb. The nonadapted mice exhibited much more extreme symptoms when
exposed to such levels. A control group was used to partially rule out effects of selection of
CO-resistant animals.
20 Clark and Otis (1952) exposed mice to gradually increasing CO levels over a period of
14 days until a level of 1380 mg/m3 (1200 ppm) was reached. When exposed to a simulated
altitude of 34,000 ft, survival of the CO-adapted groups was much greater than controls.
Similarly, Clark and Otis (1952) acclimatized mice to a simulated altitude of 18,000 ft and
showed that these altitude-adapted mice survived 2875 mg/m3 (2500 ppm) CO better than
25 controls. Wilks et al. (1959) reported similar effects in dogs.
Gorbatow and Noro (1948) showed that rats given successive daily short-term exposures
could tolerate, without loss of consciousness, longer and longer exposures. Their CO-
exposure levels were 2875 to 11,500 mg/m3 (2000 to 10,000 ppm). Increases in tolerance to
CO began to be evident as early as the fourth or fifth day of exposure and still were occurring
30 as late as the 47th day. Nonexposure for several days eliminated some of the adaptation.
Similar results were reported by Zebro et al. (1976).
*
March 12, 1990 10-185 DRAFT-DO NOT QUOTE OR CITE
-------
Chronic CO exposure of rats increases hemoglobin concentration, hematocrit, and
erythrocyte counts via. erythropoietin production (see Section 10.3.4). Penney et al. (1974)
concluded that the threshold for the erythropoietin response was 100 ppm (9.26% COHb).
Cardiac enlargement, involving the entire heart during CO exposure (compared to right
5 ventricular hypertrophy with high-altitude exposure), is induced when ambient CO is near
200 ppm, producing COHb levels of 15.8% (Penney et al., 1974). Blood volume of the rat
exposed for 7.5 weeks to CO exposures peaking at 1300 ppm nearly doubled while
erythrocyte mass more than tripled (Penney et al., 1988). After 42 days of continuous
exposure to 500 ppm, rat blood volume almost doubled, primarily as a consequence of
10 increases in erythrocytes (Davidson and Penney, 1988). It should be noted that all the
demonstrated effects on tissues and fluids are induced by long-term exposures to high CO
concentrations. McGrath (1989) exposed rats for 6 weeks to altitudes ranging from 3,300 ft
(ambient) to 18,000 ft and to concentrations of CO ranging from 0 to 500 ppm. At 9 and
35 ppm CO, where COHb levels ranged from 0.9 to 3.3%, there were no significant changes
15 in body weight, right ventricular weight, hematocrit, or hemoglobin. Small but
nonsignificant changes in these variables were measured when the CO concentration was
100 ppm and COHb levels ranged from 9.4 to 10.2%. This is consistent with the
observations noted above (Penney et al., 1974) - that the threshold for erythropoietin effects
was 100 ppm.
20 Besides the level of exposure, the time course of exposure to CO also is important. As
discussed in Section 10.3.4, Hb increases in laboratory animals exposed to CO after about
48 h, and continues to increase in the course of continued exposure until about 30 days,
depending perhaps upon exposure level. This hemopoietic response to long-term CO
exposure is similar to that shown for long-term hypoxic hypoxia, except that it is slower to
25 start and tends to offset CO hypoxic effects.
Most investigators have at least implied that increased Hb level is the mechanism by
which adaptation occurs. Certainly this explanation is reasonable for the studies showing
increased survival in groups adapted for several days. Little has been done, however, to
elucidate the extent to which such increases offset the deleterious effects of CO. The
30 probability that some adaptation occurs is supported theoretically due to Hb increases, and
March 12, 1990 10-186 DRAFT-DO NOT QUOTE OR CITE
-------
empirically in the findings of laboratory animal studies measuring survival time. But
adaptation has not been demonstrated for specific health effects other than survival time.
Compensatory increases in Hb are not without deleterious consequences of their own,
such as cardiac hypertrophy (see Section 10.3.4). The Hb increases also are not entirely
5 compensatory at all CO levels in view of the fact that deleterious effects still occur at some
CO levels for many physiological systems. It is possible, however, that without such
mechanisms as Hb increases, CO effects would be worse or would occur at lower exposure
concentrations.
10 10.7.3 Summary
The only evidence for short- or long-term COHb compensation in man is indirect.
Experimental animal data indicate that COHb levels produce physiological responses that tend
to offset other deleterious effects of CO exposure. Such responses are (1) increased coronary
blood flow, (2) increased cerebral blood flow, (3) increased hemoglobin through increased
15 hemopoiesis, and (4) increased O2 consumption in muscle.
Short-term compensatory responses in blood flow or O2 consumption may not be
complete or might even be lacking in certain persons. For example, from laboratory animal
studies it is known that coronary blood flow is increased with COHb, and from human
clinical studies it is known that subjects with ischemic heart disease respond to the lowest
20 levels of COHb (5%, or less). The implication is that in some cases of cardiac impairment,
the short-term compensatory mechanism is impaired.
From neurobehavioral studies, it is apparent that decrements due to CO have not
occurred consistently in all subjects, or even in the same studies, and have not demonstrated a
dose-response relationship with increasing COHb levels. The implication from this data
25 suggests that there might be some threshold or time lag in a compensatory mechanism such as
increased cerebral blood flow. Without direct physiological evidence in either laboratory
animals or, preferably humans, this concept only can be hypothesized. The observed results
from the neurobehavioral studies could be explained by differences or problems in
experimental protocols or due to possible nonrandom sampling.
30 The idea of a threshold or a time lag in compensatory mechanisms should not be
rejected entirely, however. There simply is no direct evidence. Studies need to be performed
March 12, 1990 10-187 DRAFT-DO NOT QUOTE OR CITE
-------
to (1) measure cerebral blood flow and tissue PO2 with low COHb levels at various ambient
concentrations of CO to determine early and low level effects accurately, and (2) design
behavioral studies where threshold effects or time lags are factors in the experimental
protocols that can be explicitly studied.
5 The mechanism by which long-term adaptation would occur, if it could be demonstrated
in humans, is assumed to be an increased Hb concentration via a several-day increase in
hemopoiesis. This alteration in Hb production has been demonstrated repeatedly in animal
studies but no recent studies have been conducted indicating or suggesting that some
adaptational benefit has or would occur. Furthermore, even if the Hb increase is a signature
10 of adaptation, it has not been demonstrated to occur at low ambient concentrations of CO.
The human studies of the 1940s have not been replicated, so the question of adaptation
remains unresolved.
15 10.8 REFERENCES
Abbatiello, E. R.; Mohrmann, K. (1979) Effects on the offspring of chronic low exposure carbon monoxide
during mice pregnancy. Clin. Toxicol. 14: 401-406.
20 Abramson, E.; Heyman, T. (1944) Dark adaptation and inhalation of carbon monoxide. Acta Physiol. Scand.
7: 303-305.
Adams, J. D.; Erickson, H. H.; Stone, H. L. (1973) Myocardial metabolism during exposure to carbon
monoxide in the conscious dog. J. Appl. Physiol. 34: 238-242.
25
Adams, K. F.; Koch, G.; Chatterjee, B.; Goldstein, G. M.; O'Neil, J. J.; Bromberg, P. A.; Sheps, D. S.;
McAllister, S.; Price, C. J.; Bissette, J. (1988) Acute elevation of blood carboxyhemoglobin to 6%
impairs exercise performance and aggravates symptoms in patients with ischemic heart disease. J. Am.
Coll. Cardiol. 12: 900-909.
30
Ahmad, I.; Ahmad, J. (1980) Environmental carbon monoxide and hypertension. J. Tenn. Med. Assoc.
73: 563-566.
Albaum, H. G.; Tepperman, J.; Bodansky, O. (1946) The in vivo inactivation by cyanide of brain cytochrome
35 oxidase and its effect on glycolysis and on the high energy phosphorus compounds in brain. J. Biol.
Chem. 164: 45-51.
Alcindor, L. G.; Belegaud, J.; Aalam, H.; Piot, M. C.; Heraud, C.; Boudene, C. (1984) Teneur en cholesterol
des lipoproteines legeres chez le lapin hypercholesterolemique intoxique ou non par 1'oxyde de carbone
40 [Low-density lipoprotein levels in hypercholesterolemic rabbits intoxicated or not by carbon monoxide].
Ann. Nutr. Metab. 28: 117-122.
March 12, 1990 10-188 DRAFT-DO NOT QUOTE OR CITE
-------
Alderman, B. W.; Baron, A. E.; Savitz, D. A. (1987) Maternal exposure to neighborhood carbon monoxide and
risk of low infant birth weight. Public Health Rep. 102: 410-414.
Aliukhin, Y.; Antonov, L. M.; Gasteva, S. V. (1974) Oxygen consumption in the brain in histotoxic hypoxia.
5 Sechonov Physiol. J. USSR 60: 1376-1381.
Allied, E. N.; Bleecker, E. R.; Chaitman, B. R.; Dahms, T. E.; Gottlieb, S. O.; Hackney, J. D.; Pagano, M.;
Selvester, R. H.; Walden, S. M.; Warren, J. (1989a) Short-term effects of carbon monoxide exposure on
the exercise performance of subjects with coronary artery disease. N. Engl. J. Med. 321: 1426-1432.
10
Allred, E. N.; Bleecker, E. R.; Chairman, B. R.; Dahms, T. E.; Gottlieb, S. O.; Hackney, J. D.; Hayes, D.;
Pagano, M. Selvester, R. H.; Walden, S. M.; Warren, J. (1989b) Acute effects of carbon monoxide
exposure on individuals with coronary artery disease. Cambridge, MA: Health Effects Institute; research
report no. 25.
15
Ames, B. N.; McCann, J.; Yamasaki, E. (1975) Methods for detecting carcinogens and mutagens with the
Salmonella/mammalian-microsome mutagenicity test. Mutat. Res. 31: 347-363.
Anders, M. W.; Sunram, J. M. (1982) Transplacental passage of dichloromethane and carbon monoxide. Toxicol.
20 Lett. 12: 231-234.
Anderson, E. W.; Andelman, R. J.; Strauch, J. M.; Fortuin, N. J.; Knelson, J. H. (1973) Effect of low-level
carbon monoxide exposure on onset and duration of angina pectoris: a study in ten patients with ischemic
heart disease. Ann, Intern. Med. 79: 46-50.
25
Annau, Z. (1974) The comparative effects of hypoxic and carbon monoxide hypoxia on behavior. In: Weiss, B.;
Laties, V. G., eds. Behavioral toxicology. New York, NY: Plenum Publishing; pp. 105-127.
Annau, Z. (1975) The comparative effects of hypoxic and carbon monoxide hypoxia on behavior. In: Weiss, B.;
30 Laties, V. G., eds. Behavioral toxicology. New York, NY: Plenum Press; pp. 151-155.
Annau, Z.; Dyer, R. S. (1977) Effects of environmental temperature upon body temperature in the hypoxic rat.
Fed. Proc. Fed. Am. Soc. Exp. Biol. 36: 599.
35 Armitage, A. K.; Davies, R. F.; Turner, D. M. (1976) The effects of carbon monoxide on the development of
atherosclerosis in the White Carneau pigeon. Atherosclerosis 23: 333-344.
Aronow, W. S. (1981) Aggravation of angina pectoris by two percent carboxyhemoglobin. Am. Heart J. 101:
154-157.
40
Aronow, W. S.; Cassidy, J. (1975) Effect of carbon monoxide on maximal treadmill exercise: a study in normal
persons. Ann. Intern. Med. 83: 496-499.
Aronow, W. S.; Isbell, M. W. (1973) Carbon monoxide effect on exercise-induced angina pectoris. Ann. Intern.
45 Med. 79: 392-395.
Aronow, W. S.; Harris, C. N.; Isbell, M. W.; Rokaw, S. N.; Imparato, B. (1972) Effect of freeway travel on
angina pectoris. Ann. Intern. Med. 77: 669-676.
50 Aronow, W. S.; Ferlinz, J.; Glauser, F. (1977) Effect of carbon monoxide on exercise performance in chronic
obstructive pulmonary disease. Am. J. Med. 63: 904-908.
Aronow, W. S.; Stemmer, E. A.; Wood, B.; Zweig, S.; Tsao, K.-P.; Raggio, L. (1978) Carbon monoxide and
ventricular fibrillation threshold in dogs with acute myocardial injury. Am. Heart J. 95: 754-756.
March 12, 1990 10-189 DRAFT-DO NOT QUOTE OR CITE
-------
Aronow, W. S.; Stemmer, E. A.; Zweig, S. (1979) Carbon monoxide and ventricular fibrillation threshold in
normal dogs. Arch. Environ. Health 34: 184-186.
Aronow, W. S.; Schlueter, W. J.; Williams, M. A.; Petratis, M.; Sketch, M. H. (1984) Aggravation of exercise
5 performance in patients with anemia by 3% carboxyhemoglobin. Environ. Res. 35: 394-398.
Astrup, J. (1982) Energy-requiring cell functions in the ischemic brain. J. Neurosurg. 56: 482-497.
Astrup, P.; Kjeldsen, K.; Wanstrup, J. (1967) Enhancing influence of carbon monoxide on the development of
10 atheromatosis in cholesterol-fed rabbits. J. Atheroscler. Res. 7: 343-354.
Astrup, P.; Olsen, H. M.; Trolle, D.; Kjeldsen, K. (1972) Effect of moderate carbon-monoxide exposure on fetal
development. Lancet (0000): 1220-1222.
15 Atkins, E. H.; Baker, E. L. (1985) Exacerbation of coronary artery disease by occupational carbon monoxide
exposure: a report of two fatalities and a review of the literature. Am. J. Ind. Med. 7: 73-79.
Ator, N. A. (1982) Modulation of the behavioral effects of carbon monoxide by reinforcement contingencies.
Neurobehav. Toxicol. Teratol. 4: 51-61.
20
Ator, N. A.; Merigan, W. H., Jr.; Mclntire, R. W. (1976) The effects of brief exposures to carbon monoxide on
temporally differentiated responding. Environ. Res. 12: 81-91.
Ayres, S. M.; Mueller, H. S.; Gregory, J. J.; Giannelli, S., Jr.; Penny, J. L. (1969) Systemic and myocardial
25 hemodynamic responses to relatively small concentrations of carboxyhemoglobin (COHB). Arch.
Environ. Health 18: 699-709.
Ayres, S. M.; Giannelli, S., Jr.; Mueller, H. (1970) Myocardial and systemic responses to carboxyhemoglobin.
In: Coburn, R. F., ed. Biological effects of carbon monoxide. Ann. N. Y. Acad. Sci. 174: 268-293.
30
Ayres, S. M.; Evans, R. G.; Buehler, M. E. (1979) The prevalence of carboxyhemoglobinemia in New Yorkers
and its effects on the coronary and systemic circulation. Prev. Med. 8: 323-332.
Balraj, E. K. (1984) Atherosclerotic coronary artery disease and "low" levels of carboxyhemoglobin; report of
35 fatalities and discussion of pathophysiologic mechanisms of death. J. Forensic Sci. 29: 1150-1159.
Beard, R. R.; Grandstaff, N. (1970) Carbon monoxide exposure and cerebral function. Ann. N. Y. Acad. Sci.
174: 385-394.
40 Beard, R. R.; Wertheim, G. A. (1967) Behavioral impairment associated with small doses of carbon monoxide.
Am. J. Public Health 57: 2012-2022.
Becker, L. C.; Haak, E. D., Jr. (1979) Augmentation of myocardial ischemia by low level carbon monoxide
exposure in dogs. Arch. Environ. Health 34: 274-279.
45
Bender, W.; Goethert, M.; Malorny, G. (1972) Effect of low carbon monoxide concentrations on psychological
functions. Staub Reinhalt. Luft 32: 54-60.
Benignus, V. A. (1984) EEG as a cross species indicator of neurotoxicology. Neurobehav, Toxicol. Teratol.
50 6: 473-483.
Benignus, V. A.; Otto, D. A.; Prah, J. D.; Benignus, G. (1977) Lack of effects of carbon monoxide on human
vigilance. Percept. Mot. Skills 45: 1007-1014.
March 12, 1990 10-190 DRAFT-DO NOT QUOTE OR CITE
-------
Benignus, V. A.; Muller, K. E.; Barton, C. N.; Prah, J. D. (1987) Effect of low level carbon monoxide on
compensatory tracking and event monitoring. Neurotoxicol. Teratol. 9: 227-234.
Benignus, V. A.; Muller, K. E.; Pieper, K. S.; Prah, J. D. (1989a) Compensatory tracking in humans with
5 elevated carboxyhemoglobin. Neurotoxicol. Teratol.: in press.
Benignus, V. A.; Muller, K. E.; Malott, C. M. (1989b) Dose-effects functions for carboxyhemoglobin and
behavior. Neurotoxicol. Teratol.: in press.
10 Beme, R. M.; Rubio, R.; Curnish, R. R. (1974) Release of adenosine from ischemic brain, effect on cerebral
vascular resistance and incorporation into cerebral adenine nucleotides. Circ. Res. 35: 262-271.
Berntman, L.; Carlsson, C.; Siesjo, B. K. (1979) Cerebral oxygen consumption and blood flow in hypoxia:
influence of sympathoadrenal activation. Stroke (Dallas) 10: 20-25.
15
Betz, E. (1972) Cerebral blood flow: its measurement and regulation. Physiol. Rev. 52: 595-630.
Bicher, H. I.; Bruley, D. F.; Reneau, D. D.; Kinsley, M. H. (1973) Autoregulation of oxygen supply to micro
areas of brain tissue under hypoxic and hyperbaric conditions. Bibl. Anat. 11: 526-531.
20
Bing, R. J.; Sarma, J. S.; Weishaar, R.; Rackl, A.; Pawlik, G. (1980) Biochemical and histological effects of
intermittent carbon monoxide exposure in cynomolgus monkeys (Macaco fascicularis) in relation to
atherosclerosis. J. Clin. Pharmacol. 20: 487-499.
25 Bissette, J.; Carr, G.; Koch, G. G.; Adams, K. F.; Sheps, D. S. (1986) Analysis of (events/time at risk) ratios
from two-period crossover studies. In: American Statistical Association 1986 proceedings of the
Biopharmaceutical Section; August; Chicago, IL. Washington, DC: American Statistical Association;
pp. 104-108.
30 Bissonnette, J. M.; Wickham, W. K. (1977) Placental diffusing capacity for carbon monoxide in unanesthetized
guinea pigs. Respir. Physiol. 31: 161-168.
Borgstrom, L.; Johannsson, H.; Siesjo, B. K. (1975) The relationship between arterial PQ and cerebral blood
flow in hypoxic hypoxia. Acta Physiol. Scand. 93: 423-432.
35
Brierly, J. B. (1975) Comparison between effects of profound arterial hypotension and cyanide on the brain of
macaca mulatta. Adv. Neurol. 10: 213-221.
Brierly, J. B.; Brown, A. W.; Calverly, J. (1976) Cyanide intoxication in the rat: physiological and
40 neuropathological aspects. J. Neurol. Neurosurg. Psych. 39: 129-140.
Brinkhous, K. M. (1977) Effects of low level carbon monoxide exposure: blood lipids and coagulation
parameters. Research Triangle Park, NC: U. S. Environmental Protection Agency, Health Effects
Research Laboratory; EPA report no. EPA-600/1-77-032. Available from: NTIS, Springfield, VA;
45 PB-269340.
Brune, B.; Ullrich, V. (1987) Inhibition of platelet aggregation by carbon monoxide is mediated by activation of
guanylate cyclase. Mol. Pharmacol. 32: 497-504.
50 Buchwald, H. (1969) A rapid and sensitive method for estimating carbon monoxide in blood and its application in
problem areas. Am. Ind. Hyg. Assoc. J. 30: 564-569.
Bunnell, D. E.; Horvath, S. M. (1988) Interactive effects of physical work and carbon monoxide on cognitive
task performance. Aviat. Space Environ. Med. 59: 1133-1138.
March 12, 1990 10-191 DRAFT-DO NOT QUOTE OR CITE
-------
Bureau, M. A.; Monette, J.; Shapcott, D.; Pare, C.; Mathieu, J.-L.; Lippe, J.; Blovin, D.; Berthiaume, Y.;
Begin, R. (1982) Carboxyhemoglobin concentration in fetal cord blood and in blood of mothers who
smoked during labor. Pediatrics 69: 371-373.
5 Burgess, D. W.; Bean, J. W. (1970) The effect of neuronal activity on local cerebral PO2 and blood flow in the
mammal. In: Russell, R. W. R., ed. Brain and blood flow. London, United Kingdom: Pittman Medical
& Scientific Publishing Co., Ltd.; pp. 120-124.
Bums, T. R.; Greenberg, S. D.; Cartwright, J.; Jachimczyk, J. A. (1986) Smoke inhalation: an ultrastructural
10 study of reaction to injury in the human alveolar wall. Environ. Res. 41: 447-457.
Calverley, P. M. A.; Leggett, R. J. E.; Flenley, D. C. (1981) Carbon monoxide and exercise tolerance in
chronic bronchitis and emphysema. Br. Med. J. 283: 878-880.
15 Campbell, J. A. (1934) Growth, fertility, etc. in animals during attempted acclimatization to carbon monoxide.
Exp. Physiol. 24: 271-281.
Cavazzuti, M.; Duffy, T. E. (1982) Regulation of local cerebral blood flow in normal and hypoxic newborn
dogs. Ann. Neurol. 11: 247-257.
20
Celsing, F.; Sverdenhag, J.; Pihlstedt, P.; Ekblom, B. (1987) Effects of anemia and stepwise-induced
polycythemia on maximal aerobic power in individuals with high and low haemoglobin concentration.
Acta Physiol. Scand. 129: 47-54.
25 Chance, B.; Cohen, P.; Jobsis, F.; Schoener, B. (1962) Intracellular oxidation-reduction states in vivo. Science
(Washington, DC) 137: 499-508.
Chance, B.; Erecinska, M.; Wagner, M. (1970) Mitochondrial responses to carbon monoxide toxicity. In:
Coburn, R. F., ed. Biological effects of carbon monoxide. Ann. N. Y. Acad. Sci. 174: 193-204.
30
Chapman, R. W.; Santiago, T. V.; Edelman, N. H. (1980) Brain hypoxia and control of breathing:
neuromechanical control. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 49: 497-505.
Chen, S.; Weller, M. A.; Penney, D. G. (1982) A study of free lung cells from young rats chronically exposed
35 to carbon monoxide from birth. Scanning Electron Microsc. 2: 859-867.
Chevalier, R. B.; Krumholz, R. A.; Ross, J. C. (1966) Reaction of nonsmokers to carbon monoxide inhalation:
cardiopulmonary responses at rest and during exercise. JAMA J. Am. Med. Assoc. 198: 1061-1064.
40 Chiodi, H.; Dill, D. B.; Consolazio, F.; Horvath, S. M. (1941) Respiratory and circulatory responses to acute
carbon monoxide poisoning. Am. J. Physiol. 134: 683-693.
Choi, K. D.; Oh, Y. K. (1975) A teratological study on the effects of carbon monoxide exposure upon the fetal
development of albino rats. Chungang Uihak 29: 209-212.
45
Christensen, C. L.; Gliner, J. A.; Horvath, S. M.; Wagner, J. A. (1977) Effects of three kinds of hypoxias on
vigilance performance, Aviat. Space Environ. Med. 48: 491-496.
Christensen, P.; Gronlund, J.; Carter, A. M. (1986) Placental gas exchange in the guinea-pig: fetal blood gas
50 tensions following the reduction of maternal oxygen capacity with carbon monoxide. J. Dev. Physiol.
8: 1-9.
Clark, R. T., Jr.; Otis, A. B. (1952) Comparative studies on acclimatization of mice to carbon monoxide and to
low oxygen. Am. J. Physiol. 169: 285-294.
March 12, 1990 10-192 DRAFT-DO NOT QUOTE OR CITE
-------
Clubb, J. J., Jr.; Penney, D. G.; Baylerian, M. S.; Bishop, S. P. (1986) Cardiomegaly due to myocyte
hyperplasia in perinatal rats exposed to 200 ppm carbon monoxide. J. Mol. Cell. Cardiol. 18: 477-486.
Clubb, F. J., Jr.; Penney, D. G.; Bishop, S. P. (1989) Cardiomegaly in neonatal rats exposed to 500 ppm carbon
5 monoxide. J. Mol. Cell. Cardiol. 21: 945-955.
Cobum, R. F. (1977) Oxygen tension sensors in vascular smooth muscle. In: Reivich, M.; Coburn, R. F.;
Lahiri, S.; Chance, B., eds. Tissue hypoxia and ischemia. Adv. Exp. Med. Biol. 78: 101-115.
10 Coburn, R. F. (1979) Mechanisms of carbon monoxide toxicity. Prev. Med. 8: 310-322.
Coburn, R. F.; Forsters R. E.; Kane, P. B. (1965) Considerations of the physiological variables that determine
the blood carboxyhemoglobin concentration in man, J. Clin. Invest. 44: 1899-1910.
15 Cohen, S. I.; Deane, M.; Goldsmith, J. R. (1969) Carbon monoxide and survival from myocardial infarction.
Arch. Environ. Health 19: 510-517.
Collier, C. R.; Workman, J. M.; Mohler, J. G.; Aaronson, J.; Cabula, O. (1972) Physiological adaptations to
carbon monoxide levels and exercise in normal men. Research Triangle Park, NC: U. S. Environmental
20 Protection Agency, Division of Health Effects Research; report no. EPA-RI-72-002. Available from:
NTIS, Springfield, VA; PB-213834.
Comroe, J. H., Jr. (1974) Physiology of respiration. 2nd ed. Chicago, IL: Yearbook Medical Publishers.
25 Cooper, K. R.; Alberti, R. R. (1984) Effect 01 kerosene heater emissions on indoor air quality and pulmonary
function. Am. Rev. Respir. Dis. 129: 629-631.
Crocker, P. J.; Walker, J. S. (1985) Pediatric carbon monoxide toxicity. J. Emerg. Med. 3: 443-448.
30 Daughtrey, W. C.; Norton, S. (1982) Morphological damage to the premature fetal rat brain after acute carbon
monoxide exposure. Exp. Neurol. 78: 26-37.
Daughtrey, W. C.; Norton, S. (1983) Caudate morphology and behavior of rats exposed to carbon monoxide in
utero. Exp. Neurol. 80: 265-278.
35
Davidson, S. B.; Penney, D. G. (1988) Time course of blood volume change with carbon monoxide inhalation
and its contribution to the overall cardiovascular response. Arch. Toxicol. 61: 306-313.
Davies, D. M.; Smith, D. J. (1980) Electrocardiographic changes in healthy men during continuous low-level
40 carbon monoxide exposure. Environ. Res. 21: 197-206.
Davies, R. F.; Topping, D. L.; Turner, D. M. (1976) The effect of intermittent carbon monoxide exposure on
experimental atherosclerosis in the rabbit. Atherosclerosis 24: 527-536.
45 Deanfield, J. B.; Shea, M. J.; Wilson, R. A.; Horlock, P.; de Landsheere, C. M.; Selwyn, A. P. (1986) Direct
effects of smoking on the heart: silent ischemic disturbances of coronary flow. Am. J. Cardiol.
57: 1005-1009.
DeBias, D. A.; Banerjee, C. M.; Birkhead, N. C.; Harrer, W. V.; Kazal, L. A. (1973) Carbon monoxide
50 inhalation effects following myocardial infarction in monkeys. Arch. Environ. Health 27: 161-167.
DeBias, D. A.; Banerjee, C. M.; Birkhead, N. C.; Greene, C. H.; Scott, S. D.; Harrer, W. V. (1976) Effects of
carbon monoxide inhalation on ventricular fibrillation. Arch. Environ. Health 31: 42-46.
March 12, 1990 10-193 DRAFT-DO NOT QUOTE OR CITE
-------
DeLucia, A. J.; Whitaker, J. H.; Bryant, L. R. (1983) Effects of combined exposure to ozone and carbon
monoxide (CO) in humans. In: Lee, S. D.; Mustafa, M. G.; Mehlrnan, M. A., eds. International
symposium on the biomedical effects of ozone and related photochemical oxidants; March; Pinehurst,
NC. Princeton, NJ: Princeton Scientific Publishers, Inc.; pp. 145-159. (Advances in modern
5 environmental toxicology: v. 5).
Dempsey, L. C.; O'Donnell, J. J.; Hoff, J. T. (1973) Carbon monoxide retinopathy. Am. J. Ophthalmol. 82:
691.
10 Detar, R.; Bohr, D. F. (1968) Oxygen and vascular smooth muscle contraction. Am. J. Physiol. 214: 241-244.
Doblar, D. D.; Santiago, T. V.; Edelman, N. H. (1977) Correlation between ventilatory and cerebrovascular
responses to inhalation of CO. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 43: 455-462.
15 Dominick, M. A.; Carson, T. L. (1983) Effects of carbon monoxide exposure on pregnant sows and their
fetuses. Am. J. Vet. Res. 44: 35-40.
Donegan, J. H.; Traystman, R. J.; Koehler, R. C.; Jones, M. D., Jr.; Rogers, M. C. (1985) Cerebrovascular
hypoxic and autoregulatory responses during reduced brain metabolism. Am. J. Physiol.
20 249: H421-H429.
Drinkwater, B. L.; Raven, P. B.; Horvath, S. M.; Gliner, J. A.; Ruhling, R. O.; Bolduan, N. W.; Taguchi, S.
(1974) Air pollution, exercise, and heat stress. Arch. Environ. Health 28: 177-181.
25 Dubeck, J. K.; Penney, D. G.; Brown, T. R.; Sharma, P. (1989) Thyroxine treatment of neonatal rats suppresses
normal and stress-stimulated heart cell hyperplasia and abolishes persistent cardiomegaly. J. Appl.
Cardiol. 4: 195-205.
Duffy, T. E.; Nelson, S. R.; Lowry, O. H. (1972) Cerebral carbohydrate metabolism during acute hypoxia and
30 recovery. J. Neurochem. 19: 959-977.
Duling, B. R.; Kuschinsky, W.; Wahl, M. (1979) Measurements of the perivascular PO2 in the vicinity of the
pial vessels of the cat. Pfluegers Arch. 383: 29-34.
35 Duncan, J. S.; Gumpert, J. (1983) A case of blindness following carbon monoxide poisoning, treated with
dopamine [letter]. J. Neurol. Neurosurg. Psychiatry 46: 459.
Dyer, R. S.; Annau, Z. (1977) Carbon monoxide and flash evoked potentials from rat cortex and superior
collkiulus. Pharmacol. Biochem. Behav. 6: 461-465.
40
Dyer, R.; Annau, Z. (1978) Carbon monoxide and superior colliculus evoked potentials. In: Otto, D. A., ed.
Multidisciplinary perspectives in event-related brain potential research: proceedings of the fourth
international congress on event-related slow potentials of the brain (EPIC IV); April 1976;
Hendersonville, NC. Washington, DC: U. S. Environmental Protection Agency, Office of Research and
45 Development; pp. 417-419; EPA report no. EPA-600/9-77-043. Available from: NTIS, Springfield, VA;
PB-297137.
Ebisuno, S.; Yasuno, M.; Yamada, Y.; Nishino, Y.; Hori, M.; Inoue, M.; Kamada T. (1986) Myocardial
infarction after acute carbon monoxide poisoning: case report. Angiology 37: 621-624.
50
Eckardt, R. E.; MacFarland, H. N.; Alarie, Y. C. E.; Busey, W. M. (1972) The biologic effect from long-term
exposure of primates to carbon monoxide. Arch. Environ. Health 25: 381-387.
March 12, 1990 10-194 DRAFT-DO NOT QUOTE OR CITE
-------
Effeney, D. J. (1987) Prostacyclin production by the heart: effect of nicotine and carbon monoxide. J. Vase.
Surg. 5: 237-247.
Einzig, S.; Nicoloff, D. M.; Lucus, R. V., Jr. (1980) Myocardial perfusion abnormalities in carbon monoxide
5 poisoned dogs. Can. J. Physiol. Pharmacol. 58: 396-405.
Ekblom, B.; Huot, R. (1972) Response to submaximal and maximal exercise at different levels of
carboxyhemoglobin. Acta Physiol. Scand. 86: 474-482.
10 Ekblom, B.; Huot, R.; Stein, E. M.; Thorstensson, A. T. (1975) Effect of changes in arterial oxygen content on
circulation and physical performance. J. Appl. Physiol. 39: 71-75.
Evans, R. G.; Webb, K.; Homan, S.; Ayres, S. M. (1988) Cross-sectional and longitudinal changes in
pulmonary function associated with automobile pollution among bridge and tunnel officers. Am. J. Lid.
15 Med. 14: 25-36.
Fazekas, J. F.; Colyer, H.; Himwich, H. E. (1939) Effect of cyanide on cerebral metabolism. Proc. Soc. Exp.
Biol. Med. 42: 496-498.
20 Fechter, L. D.; Annau, Z. (1976) Effects of prenatal carbon monoxide exposure on neonatal rats. Adverse Eff.
Environ. Chem. Psychotropic Drugs 2: 219-227.
Fechter, L. D.; Annau, Z. (1977) Toxicity of mild prenatal carbon monoxide exposure. Science (Washington,
DC) 197: 680-682.
25
Fechter, L. D.; Annau, Z. (1980) Persistent neurotoxic consequences of mild prenatal carbon monoxide
exposure. In: Di Benedetta, C.; et al, eds. Multidisciplinary approach to brain development. Dev.
Neurosci. 9: 111-112.
30 Fechter, L. D.; Annau, Z. (1980) Prenatal carbon monoxide exposure alters behavioral development.
Neurobehav. Toxicol. 2: 7-11.
Fechter, L. D.; Thakur, M.; Miller, B.; Annau, Z.; Srivastava, U. (1980) Effects of prenatal carbon monoxide
exposure on cardiac development. Toxicol. Appl. Pharmacol. 56: 370-375.
35
Fechter, L. D.; Mactutus, C. F.; Storm, J. E. (1986) Carbon monoxide and brain development. Neurotoxicology
7: 463-473.
Fechter, L. D.; Karpa, M. D.; Proctor, B.; Lee, A. G.; Storm, J. E. (1987) Disruption of neostriatal
40 development in rats following perinatal exposure to mild, but chronic carbon monoxide. Neurotoxicol.
Teratol. 9: 277-281.
Fechter, L. D.; Thome, P. R.; Nuttall, A. L. (1987) Effects of carbon monoxide on cochlear electrophysiology
and blood flow. Hear. Res. 27: 37-45.
45
Federal Register. (1986) Guidelines for the health assessment of suspect developmental toxicants. F. R.
(September 24) 51: 34028-34040.
Fein, A.; Grossman, R. F.; Jones, J. G.; Hoeffel, J.; McKay, D. (1980) Carbon monoxide effect on alveolar
50 epithelial permeability. Chest 78: 726-731.
Fisher, A. B.; Hyde, R. W.; Baue, A. E.; Reif, J. S.; Kelly, D. F. (1969) Effect of carbon monoxide on
function and structure of the lung. J. Appl. Physiol. 26: 4-12.
March 12, 1990 10-195 DRAFT-DO NOT QUOTE OR CITE
-------
Fitzgerald, R. S.; Traystman, R. J. (1980) Peripheral chemoreceptors and the cerebral vascular response to
hypoxemia. Fed. Proc. Fed. Am. Soc. Exp. Biol. 39: 2674-2677.
Fodor, G. G.; Winneke, G. (1972) Effect of low CO concentrations on resistance to monotony and on
5 psychomotor capacity. Staub Reinhalt. Luft 32: 46-54.
Forbes, W. H.; Dill, D. B.; De Silva, H.; Van deVenter, F. M. (1937) The influence of moderate carbon
monoxide poisoning upon the ability to drive automobiles. J. Ind. Hyg. Toxicol. 19: 598-603.
10 Forycki, Z.; Swica, P.; Krasnowiecki, A.; Dubicki, J.; Panow, A. (1980) Zmiany w obrazie
elektrokardiograficznym w przebiegu ostrych zatruc [Electrocardiographic changes during acute
poisonings]. Pol. Tyg. Lek. 35: 1941-1944.
Foster, J. R. (1981) Arrhythmogenic effects of carbon monoxide in experimental acute myocardial ischemia: lack
15 of slowed conduction and ventricular tachycardia. Am. Heart J. 102: 876-882.
Fountain, S. B.; Raffaele, K. C.; Annau, Z. (1986) Behavioral consequences of intraperitoneal carbon monoxide
administration in rats. Toxicol. Appl. Pharmacol. 83: 546-555.
20 Frohlich, E. D. (1987) Potential mechanisms explaining the risk of left ventricular hypertrophy. Am. J. Cardiol.
59: 91A-97A.
Gannon, B. J., et al. (1988) Effects of acute carbon monoxide exposure on precapillary vessels in the rat
cremaster muscle. Fed. Proc. Fed. Am. Soc. Exp. Biol. 2: A743.
25
Garry, R. C. (1928) The effect of oxygen lack upon surviving smooth muscle. J. Physiol. (London) 66: 235-248.
Garvey, D. J.; Longo, L. D. (1978) Chronic low level maternal carbon monoxide exposure and fetal growth and
development. Biol. Reprod. 19: 8-14.
30
Gaskell, T. W. H. (1880) On the toxicity of the heart and blood vessel. J. Physiol. 3: 48-75.
Gasteva, S. V.; Raize, T. E. (1975) Intensity of respiration and phospholipid metabolism in isolated rat brain
tissue at different temperatures in the presence of KCN. Byull. Eksp. Biol. Med. 79: 53-55.
35
Gautier, H.; Bonora, M. (1983) Ventilatory response of intact cats to carbon monoxide hypoxia. J. Appl.
Physiol.: Respir. Environ. Exercise Physiol. 55: 1064-1071.
Gilbert, R. D.; Cummings, L. A.; Juchau, M. R.; Longo, L. D. (1979) Placental diffusing capacity and fetal
40 development in exercising or hypoxic guinea pigs. J. Appl. Physiol.: Respir. Environ. Exercise Physiol.
46: 828-834.
Gliner, J. A.; Raven, P. B.; Horvath, S. M.; Drinkwater, B. L.; Sutton, J. C. (1975) Man's physiologic
response to long-term work during thermal and pollutant stress. J. Appl. Physiol. 39: 628-632.
45
Gliner, J. A.; Horvath, S. M.; Mihevic, P. M. (1983) Carbon monoxide and human performance in a single and
dual task methodology. Aviat. Space Environ. Med. 54: 714-717.
Gorbatow, O.; Noro, L. (1948) On acclimatization in connection with acute carbon monoxide poisonings. Acta
50 Physiol. Scand. 15: 77-87.
Groll-Knapp, E.; Wagner, H.; Hauck, H.; Haider, M. (1972) Effects of low carbon monoxide concentrations on
vigilance and computer-analyzed brain potentials. Staub Reinhalt. Luft 32: 64-68.
March 12, 1990 10-196 DRAFT-DO NOT QUOTE OR CITE
-------
Groll-Knapp, E.; Haider, M.; Hoeller, H.; Jenkner, H.; Stidl, H. G. (1978) Neuro- and psychophysiological
effects of moderate carbon monoxide exposure. In: Otto, D. A., ed. Multidisciplinary perspectives in
event-related brain potential research: proceedings of the fourth international congress on event-related
slow potentials of the brain (EPIC IV); April 1976; Hendersonville, NC. Washington, DC: U. S.
5 Environmental Protection Agency, Office of Research and Development; pp. 424-430; EPA report no.
EPA-600/9-77-043. Available from: NTIS, Springfield, VA; PB-297137.
Groll-Knapp, E.; Harder, M.; Jenkner, H.; Liebich, H.; Neuberger, M.; Trimmel, M. (1982) Moderate carbon
monoxide exposure during sleep: neuro- and psychological effects in young and elderly people.
10 Neurobehav. Toxicol. Teratol. 4: 709-716.
Grut, A. (1949) Chronic carbon monoxide poisoning: a study in occupational medicine. Copenhagen, Denmark:
Ejnar Munksgaard.
15 Guest, A. D. L.; Duncan, C.; Lawther, P. J. (1970) Carbon monoxide and phenobarbitone: a comparison of
effects on auditory flutter fusion threshold and critical flicker fusion threshold. Ergonomics 13: 587-594.
Gurtner, G. H.; Traystman, R. J.; Burns, B. (1982) Interactions between placental 02 and CO transfer. J. Appl.
Physiol.: Respir. Environ. Exercise Physiol. 52: 479-487.
20
Guyton, A. C.; Richardson, T. Q. (1961) Effect of hematocrit on venous return. Circ. Res. 9: 157-164.
Hagberg, M.; Kolmodin-Hedman, B.; Lindahl, R.; Nilsson, C.-A.; Norstrom, A. (1985) Irritative complaints,
carboxyhemoglobin increase and minor ventilatory function changes due to exposure to chain-saw
25 exhaust. Eur. J. Respir. Dis. 66: 240-247.
Haggendal, E.; Johannsson, E. (1965) Effect of arterial carbon dioxide tension and oxygen saturation on cerebral
blood flow autoregulation in dogs. Acta Physiol. Scand. 66: 27-53.
30 Haggendal, E.; Norbeck, B. (1966) Effect of viscosity on cerebral blood flow. Acta Cbir. Scand. Suppl.
364: 13-22.
Haider, M.; Groll-Knapp, E.; Hoeller, H.; Neuberger, M.; Stidl, H. (1976) Effects of moderate carbon
monoxide dose on the central nervous system-electrophysiological and behaviour data and clinical
35 relevance. In: Finkel, A. J.; Duel, W. C., eds. Clinical implications of air pollution research: air
polution medical research conference; December 1974; San Francisco, CA. Acton, MA: Publishing
Sciences Group, Inc.; pp. 217-232.
Haldane, J. S.; Priestley, J. G. (1935) Respiration. New Haven, CT: Yale University Press.
40
Halebian, P. H.; Barie, P. S.; Robinson, N.; Shires, G. T. (1984a) Effects of carbon monoxide on pulmonary
fluid accumulation. Curr. Surg. 41: 369-371.
Halebian, P.; Sicilia, C.; Hariri, R.; Inamdar, R.; Shires, G. T. (1984b) A safe and reproducible model of
45 carbon monoxide poisoning. Ann. N. Y. Acad. Sci. 435: 425-428.
Halperin, M. H.; McFarland, R. A.; Niven, J. L; Roughton, F. J. W. (1959) The time course of the effects of
carbon monoxide on visual thresholds. J. Physiol. (London) 146: 583-593.
50 Hanley, D. F.; Wilson, D. A.; Traystman, R. J. (1986) Effect of hypoxia and hypercapnia on neurohypophyseal
blood flow. Am. J. Physiol. 250: H7-H15.
Harbin, T. J.; Benignus, V. A.; Muller, K. E.; Barton, C. N. (1988) The effects of low-level carbon monoxide
exposure upon evoked cortical potentials in young and elderly men. Neurotoxicol. Teratol. 10: 93-100.
March 12, 1990 10-197 DRAFT-DO NOT QUOTE OR CITE
-------
Harik, S. I.; LaManna, J. C.; Light, A. I.; Rosenthal, M. (1979) Cerebral norepinephrine: influence on cortical
oxidative metabolism in situ. Science (Washington, DC) 206: 69-71.
Heistad, D. D.; Marcus, M. L. (1978) Controversies in cardiovascular research: evidence that neural mechanisms
5 do not have important effects on cerebral blood flow. Circ. Res. 42: 295-302.
Heistad, D. D.; Wheller, R. C. (1972) Effect of carbon monoxide on reflex vasoconstriction in man. J. Appl.
Physiol. 32: 7-11.
10 Heistad, D. D.; Marcus, M. L.; Ehrhardt, J. C.; Abboud, F. M. (1976) Effect of stimulation of carotid
chemoreceptors on total and regional cerebral blood flow. Circ. Res. 38: 20-25.
Helsing, K. J.; Sandier, D. P.; Comstock, G. W.; Chee, E. (1988) Heart disease mortality in nonsmokers living
with smokers. Am. J. Epidemiol. 127: 915-922.
15
Hempel, F. G.; Jobsis, F. F.; LaManna, J. L.; Rosenthal, M. R.; Saltzman, H. A. (1977) Oxidation of cerebral
cytochrome aa3 by oxygen plus carbon dioxide at hyperbaric pressures. J. Appl. Physiol.: Respir.
Environ. Exercise Physiol. 43: 873-879.
20 Hernberg, S.; Karava, R.; Koskela, R.-S.; Luoma, K. (1976) Angina pectoris, ECG findings and blood pressure
of foundry workers in relation to carbon monoxide exposure. Scand. J. Work Environ. Health 2(suppl.
1): 54-63.
Heymans, C.; Bouckaert, J. J. (1932) Sinus carotidien et regulation relfexe de la circulation arterielle
25 encephalo-bulbaire. C. B. Seances Soc. Biol. Paris 110: 996-999.
Higgins, E. A.; Fiorca, V.; Thomas, A. A.; Davis, H. V. (1972) Acute toxicity of brief exposures to HF, HCL,
NO2, and HCN with and without CO. J. Fire Tech. 8: 12-30.
30 Hill, E. P.; Hill, J. R.; Power, G. G.; Longo, L. D. (1977) Carbon monoxide exchanges between the human
fetus and mother: a mathematical model. Am. J. Physiol. 232: H311-H323.
Himwich, H. E.; Fazekas, J. F. (1941) Comparative studies of the metabolism of the brain of infant and adult
dogs. Am. J. Physiol. 132: 454-459.
35 Hinderliter, A. L.; Adams, K. F., Jr.; Price, C. J.; Herbst, M. C.; Koch, G.; Sheps, D. S. (1989) Effects of
low-level carbon monoxide exposure on resting and exercise-induced ventricular arrhythmias in patients
with coronary artery disease and no baseline ectopy. Arch. Environ. Health 44: 89-93.
Hirsch, G. L.; Sue, D. Y.; Wasserman, K.; Robinson, T. E.; Hansen, J. E. (1985) Immediate effects of cigarette
40 smoking on cardiorespiratory responses to exercise. J. Appl. Physiol. 58: 1975-1981.
Hofer, R. (1926) Uber die Wirkung von Gasgemischen. Arch. J. Exp. Path. Pharmakol. Ill: 183-205.
Hoffman, D. J.; Campbell, K. I. (1977) Postnatal toxicity of carbon monoxide after pre- and postnatal exposure.
45 Toxicol. Lett. 1: 147-150.
Hoppenbrouwers, T.; Calub, M.; Arakawa, K.; Hodgman, J. (1981) Seasonal relationship of sudden infant death
syndrome and environmental pollutants. Am. J. Epidemiol. 113: 623-635.
50 Horvath, S. M. (1975) Influence of carbon monoxide on cardiac dynamics in normal and cardiovascular stressed
animals. Sacramento, CA: Air Resources Board; final report on grant ARB-2096.
Horvath, S. M. (1981) Impact of air quality in exercise performance. Exercise Sport Sci. Rev. 9: 265-296.
March 12, 1990 10-198 DRAFT-DO NOT QUOTE OR CITE
-------
Horvath, S. M.; Dahms, T. E.; O'Hanlon, J. F. (1971) Carbon monoxide and human vigilance: a deleterious
effect of present urban concentrations. Arch. Environ. Health 23: 343-347.
Horvath, S. M.; Raven, P. B.; Dahms, T. E.; Gray, D. J. (1975) Maximal aerobic capacity at different levels of
5 carboxyhemoglobin. J. Appl. Physiol. 38: 300-303.
Hosko, M. J. (1970) The effect of carbon monoxide on the visual evoked response in man and the spontaneous
electroencephalogram. Arch. Environ. Health 21: 174-180.
10 Huch, R.; Huch, A.; Tuchschmid, P.; Zijlstra, W. G.; Zwart, A. (1983) Carboxyhemoglobin concentration in
fetal cord blood. Pediatrics 71: 461-462.
Hudnell, H. K.; Benignus, V. A. (1989) Carbon monoxide exposure and human visual function thresholds.
Neurotoxicol. Teratol. 11: 363-371.
15
Hugod, C. (1980) The effect of carbon monoxide exposure on morphology of lungs and pulmonary arteries in
rabbits: a light- and electron-microscopic study. Arch. Toxicol. 43: 273-281.
Hugod, C. (1981) Myocardial morphology in rabbits exposed to various gas-phase constituents of tobacco smoke:
20 an ultrastructural study. Atherosclerosis 40: 181-190.
Hugod, C.; Astrup, P. (1980) Exposure of rabbits to carbon monoxide and other gas phase constituents of
tobacco smoke. MMW Muench. Med. Wochenschr. 122(suppl. 1): S18-S24.
25 Hugod, C.; Hawkins, L. H.; Kjeldsen, K.; Thomsen, H. K.; Astrup, P. (1978) Effect of carbon monoxide
exposure on aortic and coronary intimal morphology in the rabbit. Atherosclerosis 30: 333-342.
Hutcheon, D. E.; Doorley, B. M.; Oldewurtel, H. A. (1983) Carbon monoxide inhalation and the vulnerability
of the ventricles to electrically induced arrhythmias. Fed. Proc. Fed. Am. Soc. Exp. Biol. 42: 1114.
30
Ignarro, L. J.; Byrns, R. E.; Buga, G. M.; Wood, K. S. (1987) Endothelium-derived relaxing factor from
pulmonary artery and vein possesses pharmacologic and chemical properties identical to those of nitric
oxide radical. Circ. Res. 61: 866-879.
35 Ingenito, A. J.; Durlacher, L. (1979) Effects of carbon monoxide on the b-wave of the cat electroretinogram:
comparisons with nitrogen hypoxia, epinephrine, vasodilator drugs and changes in respiratory tidal
volume. J. Pharmacol. Exp. Ther. 211: 638-646.
Insogna, S.; Warren, C. A. (1984) The effect of carbon monoxide on psychomotor function. In: Mital, A., ed.
40 Trends in ergonomics/human factors I. Amsterdam, The Netherlands: Elsevier/North-Holland;
pp. 331-337.
James, I. M.; MacDonnell, L. (1975) The role of baroreceptors and chemoreceptors in the regulation of the
cerebral circulation. Clin. Sci. 49: 465-471.
45
James, W. E.; Tucker, C. E.; Grover, R. F. (1979) Cardiac function in goats exposed to carbon monoxide. J.
Appl. Physiol.: Respir. Environ. Exercise Physiol. 47: 429-434.
Jarvis, M. J. (1987) Uptake of environmental tobacco smoke. In: O'Neill, I. K.; Brunnemann, K. D.; Dodet, B.;
50 Hoffmann, D., eds. Environmental carcinogens: methods of analysis and exposure measurements: vol. 9 -
passive smoking. Lyon, France: International Agency for Research on Cancer; pp. 43-58. (IARC
scientific publications: v. 81).
March 12, 1990 10-199 DRAFT-DO NOT QUOTE OR CITE
-------
Jobsis, F. F.; Rosenthal, M. (1978) Behaviour of the mitochondrial respiratory chain in vivo. In: Cerebral
vascular smooth muscle and its control. Amsterdam, The Netherlands: Elsevier; pp. 149-169. (Ciba
Foundation symposium series no. 56).
5 Jones, M. D., Jr.; Traystman, R. J. (1984) Cerebral oxygenation of the fetus, newborn, and adult. Semin.
Perinatal. 8: 205-216.
Jones, R. A.; Strickland, J. A.; Stunkard, J. A.; Siegel, J. (1971) Effects on experimental animals of long-term
inhalation exposure to carbon monoxide. Toxicol. Appl. Pharmacol. 19: 46-53.
10
Jones, M. D., Jr.; Sheldon, R. E.; Peelers, L. L.; Makowski, E. L.; Meschia, G. (1978) Regulation of cerebral
blood flow in the ovine fetus. Am. J. Physiol. 235: H162-H166.
Jones, M. D., Jr.; Traystman, R. J.; Simmons, M. A.; Molteni, R. A. (1981) Effects of changes in arterial O2
15 content on cerebral blood flow in the lamb. Am. J. Physiol. 240: H209-H215.
Kanten, W. E.; Penney, D. G.; Francisco, K.; Hill, J. E. (1983) Hemodynamic responses to acute
carboxyhemoglobinemia in the rat. Am. J. Physiol. 244: H320-H327.
20 Katafuchi, Y.; Nishimi, T.; Yamaguchi, Y.; Matsuishi, T.; Kimura, Y.; Otaki, E.; Yamashita, Y. (1985)
Cortical blindness in acute carbon monoxide poisoning. Brain Dev. 7: 516-519.
Katsumata, Y.; Aoki, M.; Oya, M.; Yada, S.; Suzuki, O. (1980) Liver damage in rats during acute carbon
monoxide poisoning. Forensic Sci. Int. 16: 119-123.
25
Kaul, B.; Calabro, J.; Hutcheon, D. E. (1974) Effects of carbon monoxide on the vulnerability of the ventricles
to drug-induced arrhythmias. J. Clin. Pharmacol. 14: 25-31.
Kelley, J. S.; Sophocleus, G. J. (1978) Retinal hemorrhages in subacute carbon monoxide poisoning: exposures
30 in homes with blocked furnace flues. JAMA J. Am. Med. Assoc. 239: 1515-1517.
Kety, S. S.; Schmidt, C. F. (1948) The effects of altered arterial tensions of carbon dioxide and oxygen on
cerebral blood flow and cerebral oxygen consumption in normal young men. J. Clin. Invest. 27: 484-492.
35 Killick, E. M. (1937) The acclimatization of mice to atmospheres containing low concentrations of carbon
monoxide. J. Physiol. (London) 91: 279-292.
Killick, E. M. (1940) Carbon monoxide anoxemia. Physiol. Rev. 20: 313-344.
40 Killick, E. M. (1948) The nature of the acclimatization occurring during repeated exposure of the human subject
to atmospheres containing low concentrations of carbon monoxide. J. Physiol. 107: 27-44.
King, C. E.; Cain, S. M.; Chapler, C. K. (1984) Whole body and hindlimb cardiovascular responses of the
anesthetized dog during CO hypoxia. Can. J. Physiol. Pharmacol. 62: 769-774.
45
King, C. E.; Cain, S. M.; Chapler, C. K. (1985) The role of aortic chemoreceptors during severe CO hypoxia.
Can. J. Physiol. Pharmacol 63: 509-514.
King, C. E.; Dodd, S. L.; Cain, S. M. (1987) O2 delivery to contracting muscle during hypoxic or CO hypoxia.
50 J. Appl. Physiol. 63: 726-732.
Kjeldsen, K.; Astrup, P.; Wanstrup, J. (1972) Ultrastructural intimal changes in the rabbit aorta after a moderate
carbon monoxide exposure. Atherosclerosis 16: 67-82.
March 12, 1990 10-200 DRAFT-DO NOT QUOTE OR CITE
-------
Klausen, K.; Andersen, C.; Nandrup, S. (1983) Acute effects of cigarette smoking and inhalation of carbon
monoxide during maximal exercise. Eur. J. Appl. Physiol. Occup. Physiol. 51: 371-379.
Klees, M.; Heremans, M.; Dougan, S. (1985) Psychological sequelae to carbon monoxide intoxication in the
5 child. Sci. Total Environ. 44: 165-176.
Klein, J. P.; Forster, H. V.; Stewart, R. D.; Wu, A. (1980) Hemoglobin affinity for oxygen during short-term
exhaustive exercise. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 48: 236-242.
10 Kleinert, H. D.; Scales, J. L.; Weiss, H. R. (1980) Effects of carbon monoxide or low oxygen gas mixture
inhalation on regional oxygenation, blood flow, and small vessel blood content of the rabbit heart.
Pfluegers Arch. 383: 105-111.
Kleinman, M.; Whittenberger, J. (1985) Effects of short-term exposure to carbon monoxide in subjects with
15 coronary artery disease. Sacramento, CA: California State Air Resources Board; report no.
ARB-R-86/276. Available from: NTIS, Springfield, VA; PB86-217494.
Kleinman, M. T.; Davidson, D. M.; Vandagriff, R. B.; Caiozzo, V. J.; Whittenberger, J. L. (1989) Effects of
short-term exposure to carbon monoxide in subjects with coronary artery disease. Arch. Environ. Health
20 44: 361-369.
Knelson, J, H. (1972) U.S. air quality criteria and ambient standards for carbon monoxide. Staub-Reinhalt, Luft
32: 60-64.
25 Knelson, J. H. (1972) The carbon monoxide air quality standard: implications for inner city children [memo to
Dr. J. F. Finklea]. Research Triangle Park, NC: U. S. Environmental Protection Agency; March 23.
Koehler, R. C.; Jones, M. D., Jr.; Traystman, R. J. (1982) Cerebral circulatory response to carbon monoxide
and hypoxic hypoxia in the lamb. Am. J. Physiol. 243: H27-H32.
30
Koehler, R. C.; Traystman, R. J.; Rosenberg, A. A.; Hudak, M. L.; Jones, M. D., Jr. (1983) Role of
O2-hemoglobin affinity on cerebrovascular response to carbon monoxide hypoxia. Am. J. Physiol. 245:
H1019-H1023.
35 Koehler, R. C.; Traystman, R. J.; Zeger, S.; Rogers, M. C.; Jones, M. D., Jr. (1984) Comparison of
cerebrovascular responses to hypoxic and carbon monoxide hypoxia in newborn and adult sheep. J.
Cereb. Blood Flow Metab. 4: 115-122.
Koehler, R. C.; Traystman, R. J.; Jones, M. D., Jr. (1985) Regional blood flow and O2 transport during
40 hypoxic and CO hypoxia in neonatal and adult sheep. Am. J. Physiol. 248: H118-H124.
Kogure, K.; Scheinberg, P.; Reinmuth, O. M.; Fujishima, M.; Busto, R. (1970) Mechanisms of cerebral
vasodilatation in hypoxia. J. Appl. Physiol. 29: 223-229.
45 Kontos, H. A.; Wei, E. P.; Raper, A. J.; Rosenblum, W. I.; Navari, R. M.; Patterson, J. L., Jr. (1978) Role of
tissue hypoxia in local regulation of cerebral microcirculation. Am. J. Physiol. 234: H582-H591.
Koob, G. F.; Annau, Z.; Rubin, R. J.; Montgomery, M. R. (1974) Effect of hypoxic hypoxia and carbon
monoxide on food intake, water intake, and body weight in two strains of rats. Life Sci. 14: 1511-1520.
50
Korner, P. I. (1965) The role of the arterial chemoreceptors and baroreceptors in the circulatory response to
hypoxia of the rabbit. J. Physiol. 180: 279-303.
March 12, 1990 10-201 DRAFT-DO NOT QUOTE OR CITE
-------
Kuller, L. H.; Radford, E. P. (1983) Epidemiological bases for the current ambient carbon monoxide standards.
EHP Environ. Health Perspect. 52: 131-139.
Kuller, L. H.; Radford, E. P.; Swift, D.; Perper, J. A.; Fisher, R. (1975) Carbon monoxide and heart attacks.
5 Arch. Environ. Health 30: 477-482.
Kuska, J.; Kokot, F.; Wnuk, R. (1980) Acute renal failure after exposure to carbon monoxide. Mater. Med. Pol.
(Engl. Ed.) 12: 236-238.
10 Kustov, V. V.; Belkin, V. L; Abidin, B. L; Ostapenko, O. F.; Malkuta, A. N.; Poddubnaja, L. T. (1972)
Aspects of chronic carbon monoxide poisoning in young animals. Gig. Tr. Prof. Zabol. 5: 50-52.
Lahiri, S.; Delaney, R. G. (1976) Effect of carbon monoxide on carotid chemoreceptor activity and ventilation.
In: Paintal, A. S., ed. Morphology and mechanisms of chemoreceptors: proceedings of an international
15 satellite symposium; October 1974; Srinagar, India. New Delhi, India: Navchetan Press; pp. 340-344.
Langfitt, T. W.; Kassell, N. F. (1968) Cerebral vasodialatation produced by brainstem stimulation: neurogenic
control vs. autoregulation. Am. J. Physiol. 215: 90-97.
20 Laties, V. G.; Merigan, W. H. (1979) Behavioral effects of carbon monoxide on animals and man. Annu. Rev.
Pharmacol. Toxicol. 19: 357-392.
Lebowitz, M. D. (1984) The effects of environmental tobacco smoke exposure and gas stoves on daily peak flow
rates in asthmatic and non-asthmatic families. In: Rylander, R.; Peterson, Y.; Snella, M.-C., eds. ETS -
25 environmental tobacco smoke: report from a workshop on effects and exposure levels; March 1983;
Geneva, Switzerland. Eur. J. Respir. Dis. 65(suppl. 133): 90-97.
*
Lebowitz, M. D.; Holberg, C. J.; O'Rourke, M. K.; Corman, G.; Dodge, R. (1983a) Gas stove usage, CO and
TSP, and respiratory effects. Research Triangle Park, NC: U. S. Environmental Protection Agency,
30 Health Effects Research Laboratory; EPA report no. EPA-600/D-83-107. Available from: NTIS,
Springfield, VA; PB83-250357.
Lebowitz, M. D.; Holberg, C. J.; Dodge, R. R. (1983b) Respiratory effects on populations from low-level
exposures to ozone. Presented at: 76th annual meeting of the Air Pollution Control Association; June;
35 Atlanta, GA. Pittsburgh, PA: Air Pollution Control Association; paper no. 83-12.5.
Lebowitz, M. D.; Corman, G.; O'Rourke, M. K.; Holberg, C. J. (1984) Indoor-outdoor air pollution, allergen
and meteorological monitoring hi an arid southwest area. J. Air Pollut. Control Assoc. 34: 1035-1038.
40 Lebowitz, M. D.; Holberg, C. J.; Boyer, B.; Hayes, C. (1985) Respiratory symptoms and peak flow associated
with indoor and outdoor air pollutants in the southwest. J. Air Pollut. Control Assoc. 35: 1154-1158.
Lebowitz, M. D.; Collins, L.; Holberg, C. J. (1987) Time series analyses of respiratory responses to indoor and
outdoor environmental phenomena. Environ. Res. 43: 332-341.
45
Levine, S. (1967) Experimental cyanide encephalopathy: gradients of susceptibility in the corpus callosum. J.
Neuropathol. Exp. Neurol. 26: 214-222.
Levine, S.; Stypulkowski, W. (1959) Experimental cyanide encephalopathy. Arch. Pathol. 67: 306-323.
50
Lilienthal, J. L., Jr.; Fugitt, C. H. (1946) The effect of low concentrations of carboxyhemoglobin on the
"altitude tolerance" of man. Am. J. Physiol. 145: 359-364.
March 12, 1990 10-202 DRAFT-DO NOT QUOTE OR CITE
-------
Longo, L. D. (1970) Carbon monoxide in the pregnant mother and fetus and its exchange across the placenta. In:
Coburn, R. F., ed. Biological effects of carbon monoxide. Ann. N. Y. Acad. Sci. 174: 313-341.
Longo, L. D. (1976) Carbon monoxide: effects on oxygenation of the fetus in utero. Science (Washington, DC)
5 194: 523-525.
Longo, L. D. (1977) The biological effects of carbon monoxide on the pregnant woman, fetus, and newborn
infant. Am. J. Obstet. Gynecol. 129: 69-103.
10 Longo, L. D.; Ching, K. S. (1977) Placental diffusing capacity for carbon monoxide and oxygen in
unanesthetized sheep. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 43: 885-893.
Longo, L. D.; Hill, E. P. (1977) Carbon monoxide uptake and elimination in fetal and maternal sheep. Am. J.
Physiol. 232: H324-H330.
15
Lubbers, D. W. (1968) The oxygen pressure field of the brain and its significance for the normal and critical
oxygen supply of the brain. In: Lubbers, D. W.; Luft, V. C.; Thews, G., Witzleb, E., eds. Oxygen
transport in blood and tissue. Stuttgart, Federal Republic of Germany: Verlag; pp. 124-139.
20 Luria, S. M.; McKay, C. L. (1979) Effects of low levels of carbon monoxide on visions of smokers and
nonsmokers. Arch. Environ. Health 34: 38-44.
Lutz, L. J. (1983) Health effects of air pollution measured by outpatient visits. J. Fam. Pract. 16: 307-313.
25 MacMillan, V. (1975) Regional cerebral blood flow of the rat in acute carbon monoxide intoxication. Can. J.
Physiol. Pharmacol. 53: 644-650.
Mactutus, C. F.; Fechter, L. D. (1984) Prenatal exposure to carbon monoxide: learning and memory deficits.
Science (Washington, DC) 223: 409-411.
30
Mactutus, C. F.; Fechter, L. D. (1985) Moderate prenatal carbon monoxide exposure produces persistent, and
apparently permanent, memory deficits in rats. Teratology 31: 1-12.
Madsen, H.; Dyerberg, J. (1984) Cigarette smoking and its effects on the platelet-vessel wall interaction. Scand.
35 J. Clin. Lab. Invest. 44: 203-206.
Malinow, M. R.; McLaughlin, P.; Dhindsa, D. S.; Metcalfe, J.; Ochsner, A. J., Ill; Hill, J.; McNulty, W. P.
(1976) Failure of carbon monoxide to induce myocardial infarction in cholesterol-fed cynomolgus
monkeys (Macacafascicularis). Cardiovasc. Res. 10: 101-108.
40
Mall, Th.; Grossenbacher, M.; Perruchoud, A. P.; Ritz, R. (1985) Influence of moderately elevated levels of
carboxyhemoglobin on the course of acute ischemic heart disease. Respiration 48: 237-244.
Mansouri, A.; Perry, C. A. (1982) Alteration of platelet aggregation by cigarette smoke and carbon monoxide.
45 Thromb. Haemostasis 48: 286-288.
Marshall, M.; Hess, H. (1981) Akute Wirkungen niedriger Kohlenmonoxidkonzentrationen auf Blutrheologie,
Thrombozytenfunktion und Arterienwand beim Miniaturschwein [Acute effects of low carbon monoxide
concentrations on blood rheology, platelet function, and the arterial wall in the minipig]. Res. Exp. Med.
50 178: 201-210.
Martin, T. R.; Bracken, M. B. (1986) Association of low birth weight with passive smoke exposure in
pregnancy. Am. J. Epidemiol. 124: 633-642.
March 12, 1990 10-203 DRAFT-DO NOT QUOTE OR CITE
-------
Martynjuk, V. C.; Dacenko, I. I. (1973) Aktivnost' transaminaz v usloviyakh khronicheskoi intoksikatsii okis'yu
ugleroda [Aminotransferase activity in chronic carbon monoxide poisoning]. Gig. Naselennykh. Mest.
12: 53-56.
5 Mason, G. R.; Usder, J. M.; Effiros, R. M.; Reid, E. (1983) Rapid reversible alterations of pulmonary epithelial
permeability induced by smoking. Chest 83: 6-11.
McDowall, D. G. (1966) Interrelationships between blood oxygen tensions and cerebral blood flow. In: Payne, J.
P.; Hill, D. W., eds. Symposium on oxygen measurements in blood and tissue and their significance.
10 London, United Kingdom: Churchill; pp. 205-214.
McFarland, R. A. (1970) The effects of exposure to small quantities of carbon monoxide on vision. In: Coburn,
R. F., ed. Biological effects of carbon monoxide. Ann. N. Y. Acad. Sci. 174: 301-312.
15 McFarland, R. A. (1973) Low level exposure to carbon monoxide and driving performance. Arch. Environ.
Health 27: 355-359.
McFarland, R. A.; Roughton, F. J. W.; Halperin, M. H.; Niven, J. I. (1944) The effects of carbon monoxide
and altitude on visual thresholds. J. Aviat. Med. 15: 381-394.
20
McFarland, R. A.; Forbes, W. H.; Stoudt, H. W.; Dougherty, J. D.; Crowley, T. J.; Moore, R. C.; Nalwalk,
T. J. (1973) A study of the effects of low levels of carbon monoxide upon humans performing driving
tasks. Boston, MA: Harvard School of Public Health. Available from: NTIS, Springfield, VA;
PB-233894.
25
McGinty, D. (1929) The regulation of respiration: XXV. Variations in the lactic acid metabolism in the intact
brain. Am. J. Physiol. 88: 312-329.
McGrath, J. J. (1982) Physiological effects of carbon monoxide. In: McGrath, J. J.; Barnes, C. D., eds. Air
30 pollution - physiological effects. New York, NY: Academic Press; pp. 147-181.
McGrath, J. J. (1989) Cardiovascular effects of chronic carbon monoxide and high-altitude exposure. Cambridge,
MA: Health Effects Institute; research report number 27.
35 Mchedlishvili, G.; Nikolaischvili, L.; Antia, R. (1976) Are the pial arterial responses dependent on the direct
effect of intravascular pressure and extravascular PO2, PCO2 and pH? Microvasc. Res. 10: 298-311.
McMeekin, J. D.; Finegan, B. A. (1987) Reversible myocardial dysfunction following carbon monoxide
poisoning. Can. J. Cardiol. 3: 118-121.
40
Melinyshyn, M. J.; Cain, S. M.; Villeneuve, S. M.; Chapler, C. K. (1988) Circulatory and metabolic responses
to carbon monoxide hypoxia during /3-adrenergic blockade. Am. J. Physiol. 255: H77-H84.
Meyer, A. (1963) Intoxications. In: Blackwood, W.; McMenemey, W. H.; Meyer, A.; Norman, R. M.; Russell,
45 D. S., eds. Greenfield's neuropathology. 2nd ed. London, United Kingdom: Edward Arnold (Publishers)
Ltd.; pp. 235-287.
Mihevic, P. M.; Gliner, J. A.; Horvath, S. M. (1983) Carbon monoxide exposure and information processing
during perceptual-motor performance. Int. Arch. Occup. Environ. Health 51: 355-363.
50
Miller, A. T.; Wood, J. J. (1974) Effects of acute carbon monoxide exposure on the energy metabolism of rat
brain and liver. Environ. Res. 8: 107-111.
March 12, 1990 10-204 DRAFT-DO NOT QUOTE OR CITE
-------
Mills, A. K.; Skornik, W. A.; Valles, L. M.; O'Rourke, J. J.; Hennessey, R. M.; Verrier, R. L. (1987) Effects
of carbon monoxide on cardiac electrical properties during acute coronary occlusion. Fed. Proc. Fed.
Am. Soc. Exp. Biol. 46: 336.
5 Minor, M.; Seidler, D. (1986) Myocardial infarction following carbon monoxide poisoning. W. Va. Med. J.
82: 25-28.
Minty, B. D.; Royston, D. (1985) Cigarette smoke induced changes in rat pulmonary clearance of "TcDTPA: a
comparison of participate and gas phases. Am. Rev. Respir. Dis. 132: 1170-1173.
10
Minty, B. D.; Jordan, C.; Jones, J. G. (1981) Rapid improvement in abnormal pulmonary epithelial permeability
after stopping cigarettes. Br. Med. J. 282: 1183-1186.
Mochizuki, M.; Maruo, T.; Masuko, K.; Ohtsu, T. (1984) Effects of smoking on fetoplacental-matemal system
15 during pregnancy. Am. J. Obstet. Gynecol. 149: 413-420.
Molnar, L.; Szanto, J. (1964) Effect of electrical stimulation of the bulbar vasomotor center on the cerebral blood
flow. Q. J. Exp. Physiol. 49: 184-193.
20 Montgomery, M. R.; Rubin, R. J. (1971) The effect of carbon monoxide inhalation on in vivo drug metabolism
in the rat. J. Pharmacol. Exp. Ther. 179: 465-473.
Mordelet-Dambrine, M.; Stupfel, M. (1979) Comparison in guinea-pigs and in rats of the effects of vagotomy
and of atropine on respiratory resistance modifications induced by an acute carbon monoxide or nitrogen
25 hypoxia. Comp. Biochem. Physiol. A 63: 555-559.
Mordelet-Dambrine, M.; Stupfel, M.; Duriez, M. (1978) Comparison of tracheal pressure and circulatory
modifications induced in guinea pigs and in rats by carbon monoxide inhalation. Comp. Biochem.
Physiol. A 59: 65-68.
30
Morris, G. L.; Curtis, S. E.; Simon, J. (1985a) Perinatal piglets under sublethal concentrations of atmospheric
carbon monoxide. J. Anim. Sci. 61: 1070-1079.
Morris, G. L.; Curtis, S. E.; Widowski, T. M. (1985b) Weanling pigs under sublethal concentrations of
35 atmospheric carbon monoxide. J. Anim. Sci. 61: 1080-1087.
Morris, P. D.; Koepsell, T. D.; Baling, J. R.; Taylor, J. W.; Lyon, J. L.; Swanson, G. M.; Child, M.; Weiss,
N. S. (1986) Toxic substance exposure and multiple myeloma: a case-control study. JNCI J. Natl.
Cancer Inst. 76: 987-994.
40
Moss, R. H.; Jackson, C. F.; Seiberlich, J. (1951) Toxicity of carbon monoxide and hydrogen cyanide gas
mixtures: a preliminary report. AMA Arch. Ind. Hyg. Occup. Med. 4: 53-64.
Muller, K. E.; Barton, C. N.; Benignus, V. A. (1984) Recommendations for appropriate statistical practice in
45 toxicologic experiments. Neurotoxicology 5: 113-126.
Mullin, L. S.; Krivanek, N. D. (1982) Comparison of unconditioned reflex and conditioned avoidance tests in
rats exposed by inhalation to carbon monoxide, 1,1,1-trichloroethane, toluene or ethanol.
Neurotoxicology 3: 126-137.
50
Musselman, N. P.; Groff, W. A.; Yevich, P. P.; Wilinski, F. T.; Weeks, M. H.; Oberst, F. W. (1959)
Continuous exposure of laboratory animals to low concentration of carbon monoxide. Aerosp. Med. 30:
524-529.
March 12, 1990 10-205 DRAFT-DO NOT QUOTE OR CITE
-------
Naeije, R.; Peretz, A.; Cornil, A. (1980) Acute pulmonary edema following carbon monoxide poisoning.
Intensive Care Med. 6: 189-191.
National Institute of Child Health and Human Development. (1979) Pregnancy and infant health. In: Smoking
5 and health: a report of the Surgeon General. Washington, DC: U. S. Department of Health, Education,
and Welfare; DHEW publication no. PHS 79-50066.
Neubauer, J. A.; Santiago, T. V.; Edelman, N. H. (1981) Hypoxic arousal in intact and carotid chemodenervated
sleeping cats. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 51: 1294-1299.
10
Niden, A. H. (1971) The effects of low levels of carbon monoxide on the fine structure of the terminal airways.
Am. Rev. Respir. Dis. 103: 898.
Niden, A. H.; Schulz, H. (1965) The ultrastructural effects of carbon monoxide inhalation on the rat lung.
15 Virchows Arch. A: Pathol. Anat. 339: 283-292.
Nilsson, B.; Norberg, K.; Nordstrom, C.; Siesjo, B. K. (1976) Influence of hypoxia and hypercapnia on cerebral
blood flow in rats. In: Harper, A. M.; Jennett, W. B.; Miller, J. D.; Rowan, J. O., eds. Blood flow and
metabolism of the brain. Edinburgh, Scotland: Livingstone; pp. 9.19-9.23.
20
Norberg, K.; Siesjo, B. K. (1975) Cerebral metabolism in hypoxic hypoxia. I. Pattern of activation of glycolysis:
a re-evaluation. Brain Res. 86: 31-44.
O'Donnell, R. D.; Mikulka, P.; Heinig, P.; Theodore, J. (1971a) Low level carbon monoxide exposure and
25 human psychomotor performance. Toxicol. Appl. Pharmacol. 18: 593-602.
O'Donnell, R. D.; Chikos, P.; Theodore, J. (1971b) Effect of carbon monoxide exposure on human sleep and
psychomotor performance. J. Appl. Physiol. 31: 513-518.
30 O'Sullivan, B. P. (1983) Carbon monoxide poisoning in an infant exposed to a kerosene heater. J. Pediatr. 103:
249-251.
Okeda, R.; Matsuo, T.; Kuroiwa, T.; Tajima, T.; Takahashi, H. (1986) Experimental study on pathogenesis of
the fetal brain damage by acute carbon monoxide intoxication of the pregnant mother. Acta Neuropathol.
35 69: 244-252.
Okeda, R.; Matsuo, T.; Kuroiwa, T.; Nakai, M.; Tajima, T.; Tkahashi, H. (1987) Regional cerebral blood flow
of acute carbon monoxide poisoning in cats. Acta Neuropathol. 72: 389-393.
40 Olsen, N. S.; Klein, J. R. (1947) Effect of cyanide on the concentration of lactate and phosphates in brain. J.
Biol. Chem. 167: 739-746.
Opitz, E.; Schneider, M. (1950) Uber die SauerstoffVersorgung des Gehirns und den Mechanisms von
Mangelwirken. Curr. Top. Pathol. 46: 126-260.
45
Oremus, R. A., et al. (1988) Effects of acute carbon monoxide (CO) exposure on cardiovascular (CV)
hemodynamics in the anesthetized rat. Fed. Proc. Fed. Am. Soc. Exp. Biol. 2: A1312.
Otto, D.; Reiter, L. (1978) Neurobehavioral assessment of environmental insult. In: Otto, D. A., ed.
50 Multidisciplinary perspectives in event-related brain potential research: proceedings of the fourth
international congress on event-related slow potentials of the brain (EPIC IV); April 1976;
Hendersonville, NC. Washington, DC: U. S. Environmental Protection Agency, Office of Research and
Development; pp. 409-416; EPA report no. EPA-600/9-77-043. Available from: NTIS, Springfield, VA;
PB-297137.
March 12, 1990 10-206 DRAFT-DO NOT QUOTE OR CITE
-------
Otto, D. A.; Benigus, V. A.; Prah, J. D. (1979) Carbon monoxide and human time discrimination: failure to
replicate Beard-Wertheim experiments. Aviat. Space Environ. Med. 50: 40-43.
Pankow, D.; Ponsold, W. (1972) Leucine aminopeptidase activity in plasma of normal and carbon monoxide
5 poisoned rats. Arch. Toxicol. 29: 279-285.
Pankow, D.; Ponsold, W. (1974) Kombinationswirkungen von Kohlenmonoxid mit anderen biologischaktiven
Schadfaktoren auf den Organismus [The combined effects of carbon monoxide and other biologically
active detrimental factors on the organism]. Z. Gesamte Hyg. Ihre Grenzgeb. 20: 561-571.
10
Pankow, D.; Ponsold, W.; Fritz, H. (1974) Combined effects of carbon monoxide and ethanol on the activities of
leucine aminopeptidase and glutamic-pyruvic transaminase in the plasma of rats. Arch. Toxicol.
32: 331-340.
15 Parving, H.-H. (1972) The effect of hypoxia and carbon monoxide exposure on plasma volume and capillary
permeability to albumin. Scand. J. Clin. Lab. Invest. 30: 49-56.
Paulet, G. (1958) Sur le mechanisme de la perte de 1'action hypertensive de 1'adrenaline au cours de 1'anoxie
cyanhydrique. J. Physiol. (Paris) 50: 31-52.
20
Paulson, O. B.; Parving, H.-H.; Olesen, J.; Skinhoj, E. (1973) Influence of carbon monoxide and of
hemodilution on cerebral blood flow and blood gases in man. J. Appl. Physiol. 35: 111-116.
Peelers, L. L. H.; Sheldon, R. E.; Jones, M. D., Jr.; Makowski, E. L.; Meschia, G. (1979) Blood flow to fetal
25 organs as a function of arterial oxygen content. Am. J. Obstet. Gynecol. 135: 637-646.
Penn, A.; Butler, J.; Snyder, C.; Albert, R. E. (1983) Cigarette smoke and carbon monoxide do not have
equivalent effects upon development of arteriosclerotic lesions. Artery 12: 117-131.
30 Penney, D. G.; Caldwell, T. M. (1984) Growth of the young heart: influence of nutrition and stress. Growth
48: 483-498.
Penney, D. G.; Weeks, T. A. (1979) Age dependence of cardiac growth in the normal and carbon
monoxide-exposed rat. Dev. Biol. 71: 153-162.
35
Penney, D.; Dunham, E.; Benjamin, M. (1974a) Chronic carbon monoxide exposure: time course of hemoglobin,
heart weight and lactate dehydrogenase isozyme changes. Toxicol. Appl. Pharmacol. 28: 493-497.
Penney, D.; Benjamin, M.; Dunham, E. (1974b) Effect of carbon monoxide on cardiac weight as compared with
40 altitude effects. J. Appl. Physiol. 37: 80-84.
Penney, D. G.; Sodt, P. C.; Cutilletta, A. (1979) Cardiodynamic changes during prolonged carbon monoxide
exposure in the rat. Toxicol. Appl. Pharmacol. 50: 213-218.
45 Penney, D. G.; Baylerian, M. S.; Fanning, K. E. (1980) Temporary and lasting cardiac effects of pre- and
postnatal exposure to carbon monoxide. Toxicol. Appl. Pharmacol. 53: 271-278.
Penney, D. G.; Baylerian, M. S.; Thill, J. E.; Fanning, C. M.; Yedavally, S. (1982) Postnatal carbon monoxide
exposure: immediate and lasting effects in the rat. Am. J. Physiol. 243: H328-H339.
50
Penney, D. G.; Baylerian, M. S.; Thill, J. E.; Yedavally, S.; Fanning, C. M. (1983) Cardiac response of the
fetal rat to carbon monoxide exposure. Am. J. Physiol. 244: H289-H297.
March 12, 1990 10-207 DRAFT-DO NOT QUOTE OR CITE
-------
Penney, D. G.; Barthel, B. G.; Skoney, J. A. (1984a) Cardiac compliance and dimensions in carbon monoxide-
induced cardiomegaly. Cardiovasc. Res. 18: 270-276.
Penney, D. G.; Stryker, A. E.; Baylerian, M. S. (1984b) Persistent cardiomegaly induced by carbon monoxide
5 and associated tachycardia. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 56: 1045-1052.
Penney, D. G.; Davidson, S. B.; Gargulinski, R. B.; Caldwell-Ayre, T. M. (1988a) Heart and lung hypertrophy,
changes in blood volume, hematocrit and plasma renin activity in rats chronically exposed in increasing
carbon monoxide concentrations. JAT J. Appl. Toxicol. 8: 171-178.
10
Penney, D. G.; Gargulinski, R. B.; Hawkins, B. J.; Santini, R.; Caldwell-Ayre, T. M.; Davidson, S. B. (1988b)
The effects of carbon monoxide on persistent changes in young rat heart: cardiomegaly, tachycardia and
altered DNA content. JAT J. Appl. Toxicol. 8: 275-283.
15 Penney, D. G.; Clubb, F. J., Jr.; Allen, R. C.; Jen, C.; Benerjee, S.; Hull, J. A. (1989) Persistent early heart
changes do alter the cardiac response to exercise training in adulthood. J. Appl. Cardiol. 4: 223-237.
Permutt, S.; Farhi, L. (1969) Tissue hypoxia and carbon monoxide. In: Effects of chronic exposure to low levels
of carbon monoxide on human health, behavior, and performance. Washington, DC: National Academy
20 of Sciences; pp. 18-24.
Petajan, J. H.; Packham, S. C.; Frens, D. B.; Dinger, B. G. (1976) Sequelae of carbon monoxide-induced
hypoxia in the rat. Arch. Neurol. (Chicago) 33: 152-157.
25 Pfluger, E. (1875) Beitrage zue lehre von der Respiration. I. Ueber die physiologische Verbrennung in der
Lebendigeen Organismen. Pfluegers Arch. Gesamte Physiol. Menschen Tiere 10: 251-367.
Piantadosi, C. A.; Sylvia, A. L.; Saltzman, H. A.; Jobsis-Vandervliet, F. F. (1985) Carbon
monoxide-cytochrome interactions in the brain of the fluorocarbon-perfused rat. J. Appl. Physiol.
30 58: 665-672.
Piantadosi, C. A.; Sylvia, A. L.; Jobsis-Vandervliet, F. F. (1987) Differences in brain cytochrome responses to
carbon monoxide and cyanide in vivo. J. Appl. Physiol. 62: 1277-1284.
35 Pirnay, F.; DuJardin J.; DeRoanne, R.; Petit, J. M. (1971) Muscular exercise during intoxication by carbon
monoxide. J. Appl. Physiol. 31: 573-575.
Pitt, B. R.; Radford, E. P.; Gurtner, G. H.; Traystman, R. J. (1979) Interaction of carbon monoxide and
cyanide on cerebral circulation and metabolism. Arch. Environ. Health 34: 354-359.
40
Pittman, R. N.; Duling, B. R. (1973) Oxygen sensitivity of vascular smooth muscle. Microvasc. Res.
6: 202-122.
Ponte, J.; Purves, M. J. (1974) The role of the carotid body chemoreceptors and carotid sinus baroreceptors in
45 the control of cerebral blood vessels. J. Physiol. (London) 237: 315-340.
Preziosi, T. J.; Lindenberg, R.; Levy, D.; Christenson, M. (1970) An experimental investigation in animals of
the functional and morphologic effects of single and repeated exposures to high and low concentrations of
carbon monoxide. Ann. N. Y. Acad. Sci. 174: 369-384.
50
Prigge, E.; Hochrainer, D. (1977) Effects of carbon monoxide inhalation on erythropoiesis and cardiac
hypertrophy in fetal rats. Toxicol. Appl. Pharmacol. 42: 225-228.
March 12, 1990 10-208 DRAFT-DO NOT QUOTE OR CITE
-------
Purser, D. A.; Berrill, K. R. (1983) Effects of carbon monoxide on behavior in monkeys in relation to human
fire hazard. Arch. Environ. Health 38: 308-315.
Putz, V. R.; Johnson, B. L.; Setzer, J. V. (1976) Effects of CO on vigilance performance: effects of low level
5 carbon monoxide on divided attention, pitch discrimination, and the auditory evoked potential.
Cincinnati, OH: U. S. Department of Health, Education, and Welfare, National Institute for Occupational
Safety and Health. Available from: NTIS, Springfield, VA; PB-274219.
Putz, V. R.; Johnson, B. L.; Setzer, J. V. (1979) A comparative study of the effects of carbon monoxide and
10 methylene chloride on human performance. J. Environ. Pathol. Toxicol. 2: 97-112.
Radford, E. P.; Pitt, B.; Halpin, B.; Caplan, Y.; Fisher, R.; Schweda, P. (1976) Study of fire deaths in
Maryland: Sept. 1971-Jan. 1974. In: Physiological and toxicological aspects of combustion products:
international symposium; March 1974; Salt Lake City, UT. Washington, DC: National Academy of
15 Sciences; pp. 26-35.
Ramsey, J. M. (1972) Carbon monoxide, tissue hypoxia, and sensory psychomotor response in hypoxaemic
subjects. Clin. Sci. 42: 619-625.
20 Ramsey, J. M. (1973) Effects of single exposures of carbon monoxide on sensory and psychomotor response.
Am. Ind. Hyg. Assoc. J.: 212-216.
Raven, P. B.; Drinkwater, B. L.; Horvath, S. M.; Ruhling, R. O.; Gliner, J. A.; Sutton, J. C.; Bolduan, N. W.
(1974a) Age, smoking habits, heat stress, and their interactive effects with carbon monoxide and
25 peroxyacetylnitrate on man's aerobic power. Int. J. Biometeorol. 18: 222-232.
Raven, P. B.; Drinkwater, B. L.; Ruhling, R. O.; Bolduan, N.; Taguchi, S.; Gliner, J.; Horvath, S. M. (1974b)
Effect of carbon monoxide and peroxyacetyl nitrate on man's maximal aerobic capacity. J. Appl.
Physiol. 36: 288-293.
30
Renaud, S.; Blache, D.; Dumont, E.; Thevenon, C.; Wissendanger, T. (1984) Platelet function after cigarette
smoking in relation to nicotine and carbon monoxide. Clin. Pharmacol. Ther. 36: 389-395.
Reynolds, J. E. F., ed. (1982) Martindale's: the extra pharmacopoeia. 28th ed. London, United Kingdom: The
35 Pharmaceutical Press; p. 804.
Robertson, G.; Lebowitz, M. D. (1984) Analysis of relationships between symptoms and environmental factors
over time. Environ. Res. 33: 130-143.
40 Robinson, N. B.; Barie, P. S.; Halebian, P. H.; Shires, G. T. (1985) Distribution of ventilation and perfusion
following acute carbon monoxide poisoning. In: 41st annual forum on fundamental surgical problems held
at the 71st annual clinical congress of the American College of Surgeons; October; Chicago, IL. Surg.
Forum 36: 115-118.
45 Roche, S.; Horvath, S.; Gliner, J.; Wagner, J.; Borgia, J. (1981) Sustained visual attention and carbon
monoxide: elimination of adaptation effects. Hum. Factors 23: 175-184.
Rogers, W. R.; Bass, R. L., Ill; Johnson, D. E.; Kruski, A. W.; McMahan, C. A.; Montiel, M. M.; Mott, G.
E.; Wilbur, R. L.; McGill, H. C., Jr. (1980) Atherosclerosis-related responses to cigarette smoking in
50 the baboon. Circulation 61: 1188-1193.
Rogers, W. R.; Carey, K. D.; McMahan, C. A.; Montiel, M. M.; Mott, G. E.; Wigodsky, H. S.; McGill, H.
C., Jr. (1988) Cigarette smoking, dietary hyperlipidemia, and experimental atherosclerosis in the baboon.
Exp. Mol. Pathol. 48: 135-151.
March 12, 1990 10-209 DRAFT-DO NOT QUOTE OR CITE
-------
Root, W. S. (1965) Carbon monoxide. In: Fenn, W. O.; Rahn, H., eds. Handbook of physiology: a critical,
comprehensive presentation of physiological knowledge and concepts. Section 3: Respiration, volume II.
Washington, DC: American Physiological Society; pp. 1087-1098.
5 Rosenberg, A. A.; Jones, M. D., Jr.; Traystman, R. J.; Simmons, M. A.; Molteni, R. A. (1982) Response of
cerebral blood flow to changes in P^Q in fetal, newborn, and adult sheep. Am. J. Physiol. 242:
H862-H866. 2
Rosenkrantz, H.; Grant, R. J.; Fleischman, R. W.; Baker, J. R. (1986) Marihuana-induced embryotoxicity in the
10 rabbit. Fundam. Appl. Toxicol. 7: 236-243.
Rosenthal, M.; LaManna, J. C.; Jobsis, F. F.; Levasseur, J. E.; Kontos, H. A.; Patterson, J. L. (1976) Effects
of respiratory gases on cytochrome a in intact cerebral cortex: is there a critical PQ ? Brain Res. 108:
143-154. 2
15
Roth, R. A., Jr.; Rubin, R. J. (1976a) Role of blood flow in carbon monoxide- and hypoxic hypoxia-induced
alterations in hexobarbital metabolism in rats. Drug Metab. Dispos. 4: 460-467.
Roth, R. A., Jr.; Rubin, R. J. (1976b) Comparison of the effect of carbon monoxide and of hypoxic hypoxia. II.
20 Hexobarbital metabolism in the isolated, perfused rat liver. J. Pharmacol. Exp. Ther. 199: 61-66.
Roughton, F. J. W.; Darling, R. C. (1944) The effect of carbon monoxide on the oxyhemoglobin dissociation
curve. Am. J. Physiol. 141: 17-31.
25 Roy, C. S.; Brown, J. G. (1879) The blood pressure and its variations in the arterioles, capillaries, and smaller
veins. J. Physiol. 2: 323-359.
Rubio, R.; Berne, R. M. (1969) Release of adenosine by the normal myocardium in dogs and its relationship to
the regulation of coronary arteries. Circ. Res. 25: 407-415.
30
Rubio, R.; Beme, R. M.; Bockman, E. L.; Curnish, R. R. (1975) Relationship between adenosine concentration
and oxygen supply in rat brain. Am. J. Physiol. 228: 1896-1902.
Rummo, N.; Sarlanis, K. (1974) The effect of carbon monoxide on several measures of vigilance in a simulated
35 driving task. J. Saf. Res. 6: 126-130.
Russek, M.; Fernandez, F.; Vega, C. (1963) Increase of cerebral blood flow produced by low dosages of
cyanide. Am. J. Physiol. 204: 309-312.
40 Salvatore, S. (1974) Performance decrement caused by mild carbon monoxide levels on two visual functions. J.
Saf. Res. 6: 131-134.
Santiago, T. V.; Edelmain, N. H. (1976) Mechanism of the ventilatory response to carbon monoxide. J. Clin.
Invest. 57: 977-986.
45
Schaad, G.; Kleinhanz, G.; Piekarski, C. (1983) Zum einfluss von kohlenmonoxid in der atemluft auf die
psychophysische leistungsfahigkeit. Wehrmed. Monatsschr. 10: 423-430.
Schrot, J.; Thomas, J. R. (1986) Multiple schedule performance changes during carbon monoxide exposure.
50 Neurobehav. Toxicol. Teratol. 8: 225-230.
Schrot, J.; Thomas, J. R.; Robertson, R. F. (1984) Temporal changes in repeated acquisition behavior after
carbon monoxide exposure. Neurobehav. Toxicol. Teratol. 6: 23-28.
March 12, 1990 10-210 DRAFT-DO NOT QUOTE OR CITE
-------
Schulte, J. H. (1963) Effects of mild carbon monoxide intoxication. Arch. Environ. Health 7: 30-36.
Schwetz, B. A.; Smith, F. A.; Leong, B. K. J.; Staples, R. E. (1979) Teratogenic potential of inhaled carbon
monoxide in mice and rabbits. Teratology 19: 385-392.
5
Sekiya, S.; Sato, S.; Yamaguchi, H.; Harumi, K. (1983) Effects of carbon monoxide inhalation on myocardial
infarct size following experimental coronary artery ligation. Jpn. Heart J. 24: 407-416.
Selvin, S.; Sacks, S. T.; Merrill, D. W.; Winkelstein, W. (1980) Relationship between cancer incidence and two
10 pollutants (total suspended particulate and carbon monoxide) for the San Francisco bay area. Berkeley,
CA: Lawrence Berkeley Laboratory. Available from: NTIS, Springfield, VA; LBL-10847.
Seppanen, A. (1977) Physical work capacity in relation to carbon monoxide inhalation and tobacco smoking.
Ann. Clin. Res. 9: 269-274.
15
Seppanen, A.; Hakkinen, J. V.; Tenkku, M. (1977) Effect of gradually increasing carboxyhemoglobin saturation
on visual perception and psychomotor performance of smoking and nonsmoking subjects. Ann. Clin. Res.
9: 314-319.
20 Shalit, M. N.; Reinmuth, O. M.; Shimojyo, S.; Scheinberg, P. (1967) Carbon dioxide and cerebral circulatory
control. III. The effects of brainstem lesions. Arch. Neurol. 17: 342-353.
Shephard, R. J. (1983) Carbon monoxide the silent killer. Springfield, IL: Charles C. Thomas. (American lecture
series no. 1059).
25
Shephard, R. J. (1984) Athletic performance and urban air pollution. Can. Med. Assoc. J. 131: 105-109.
Sheppard, D.; Distefano, S.; Morse, L.; Becker, C. (1986) Acute effects of routine firefighting on lung function.
Am. J. Ind. Med. 9: 333-340.
30
Sheps, D. S.; Adams, K. F., Jr.; Bromberg, P. A.; Goldstein, G. M.; O'Neil, J. J.; Horseman, D.; Koch, G.
(1987) Lack of effect of low levels of carboxyhemoglobin on cardiovascular function in patients with
ischemic heart disease. Arch. Environ. Health 42: 108-116.
35 Sheps, D. S.; Herbst, M. C.; Hinderliter, A. L.; Adams, K. F.; Ekelund, L. G.; O'Neil, J. J.; Goldstein, G.
M.; Bromberg, P. A.; Herdt, J.; Ballenger, M.; Davis, S. M.; Koch, G. (1989) Effects of 4% and 6%
carboxyhemoglobin on arrhythmia production in patients with coronary artery disease. Submitted for
publication.
40 Siesjo, B. K.; Nilsson, L. (1971) The influence of arterial hypoxemia upon labile phosphates and upon
extracellular and intracellular lactate and pyruvate concentrations in the rat brain. Scand. J. Clin. Lab.
Invest. 27: 83-96.
Siggaard-Andersen, J.; Bonde-Peterson, F.; Hansen, T. J.; Mellemgaard, K. (1968) Plasma volume and vascular
45 permeability during hypoxia and carbon monoxide exposure. Scand. J. Clin. Lab. Invest. Suppl. 103:
39-48.
Singh, J. (1986) Early behavioral alterations in mice following prenatal carbon monoxide exposure.
Neurotoxicology 7: 475-482.
50
Singh, J.; Scott, L. H. (1984) Threshold for carbon monoxide induced fetotoxicity. Teratology 30: 253-257.
Sjostrand, T. L. (1948) Brain volume, diameter of the blood vessels in the piamater, and intracranial pressure in
acute carbon monoxide poisoning. Acta Physiol. Scand. 15: 351-361.
March 12, 1990 10-211 DRAFT-DO NOT QUOTE OR CITE
-------
Skinhoj, E. (1966) Regulation of cerebral blood flow as a single function of interstitial pH. Acta Neurol. Scand.
42: 604-607.
Smith, E. E.; Crowell, J. W. (1967) Role of an increased hematocrit in altitude acclimatization. Aerospace Med.
5 38: 39-43.
Smith, D. J.; Vane, J. B. (1966) Effects of oxygen on vascular and other smooth muscles. J. Physiol. (London)
186: 284-294.
10 Smith, A. D. M.; Duckett, S.; Waters, A. H. (1963) Neuropathological changes in chronic cyanide intoxication.
Nature (London) 200: 179-181.
Smith, M. D.; Merigan, W. H.; Mclntire, R. W. (1976a) Effects of carbon monoxide on
fixed-consecutive-number performance in rats. Pharmacol. Biochem. Behav. 5: 257-262.
15
Smith, P. W.; Crane, C. R.; Sanders, D. C.; Abbott, J. K.; Endecott, B. S. (1976b) Effects of exposure to
carbon monoxide and hydrogen cyanide. In: Physiological and toxicological aspects of combustion
products: international symposium; March 1974; Salt Lake City, UT. Washington, DC: National
Academy of Sciences; pp. 75-88.
20
Snella, M.-C.; Rylander, R. (1979) Alteration in local and systemic immune capacity after exposure to bursts of
CO. Environ. Res. 20: 74-79.
Sokoloff, L. (1978) Local cerebral energy metabolism: its relationships to local functional activity and blood
25 flow. In: Cerebral vascular smooth muscle and its control. Amsterdam, The Netherlands:
Elsevier/North-Holland, Inc.; pp. 171-197. (Ciba Foundation symposium no. 56).
Sparrow, D.; Bosse, R.; Rosner, B.; Weiss, S. T. (1982) The effect of occupational exposure on pulmonary
function: a longitudinal evaluation of fire fighters and nonfire fighters. Am. Rev. Respir. Dis. 125:
30 319-322.
Stavraky, G. W. (1936) Response of cerebral blood vessels to electrical stimulation of the thalamus and
hypothalamus regions. AMA Arch. Neurol. Psychiatry 35: 1002-1028.
35 Stender, S.; Astrup, P.; Kjeldsen, K. (1977) The effect of carbon monoxide on cholesterol in the aortic wall of
rabbits. Atherosclerosis 28: 357-367.
Stern, S.; Ferguson, R. E.; Rapaport, E. (1964) Reflex pulmonary vasoconstriction due to stimulation of the
aortic body by nicotine. Am. J. Physiol. 206: 1189-1195.
40
Stern, F. B.; Lemen, R. A.; Curtis, R. A. (1981) Exposure of motor vehicle examiners to carbon monoxide: a
historical prospective mortality study. Arch. Environ. Health 36: 59-66.
Stern, F. B.; Halperin, W. E.; Hornung, R. W.; Ringenburg, V. L.; McCammon, C. S. (1988) Heart disease
45 mortality among bridge and tunnel officers exposed to carbon monoxide. Am. J. Epidemiol. 128:
1276-1288.
Stewart, R. D.; Peterson, J. E.; Baretta, E. D.; Bachand, R. T.; Hosko, M. J.; Herrmann, A. A. (1970)
Experimental human exposure to carbon monoxide. Arch. Environ. Health 21: 154-164.
50
Stewart, R. D.; Newton, P. E.; Hosko, M. J.; Peterson, J. E.; Mellender, J. W. (1972) The effect of carbon
monoxide on time perception, manual coordination, inspection, and arithmetic. In: Weiss, B.; Laties, V.
G., eds. Behavioral toxicology. New York, NY: Plenum Press; pp. 29-60.
March 12, 1990 10-212 DRAFT-DO NOT QUOTE OR CITE
-------
Stewart, R. D.; Peterson, J. E.; Fisher, T. N.; Hosko, M. J.; Baretta, E. D.; Dodd, H. C.; Herrmann, A. A.
(1973a) Experimental human exposure to high concentrations of carbon monoxide. Arch. Environ. Health
26: 1-7.
5 Stewart, R. D.; Newton, P. E.; Hosko, M. J.; Peterson, J. E. (1973b) Effect of carbon monoxide on time
perception. Arch. Environ. Health 27: 155-160.
Stewart, R. D.; Newton, P. E.; Kaufman, J.; Forster, H. V.; Klein, J. P.; Keelen, M. H., Jr.; Stewart, D. J.;
Wu, A.; Hake, C. L. (1978) The effect of a rapid 4% carboxyhemoglobin saturation increase on maximal
10 treadmill exercise. New York, NY: Coordinating Research Council, Inc.; report no.
CRC-APRAC-CAPM-22-75. Available from: OTIS, Springfield, VA; PB-296627.
Storm, J. E.; Fechter, L. D. (1985a) Alteration in the postnatal ontogeny of cerebellar norepinephrine content
following chronic prenatal carbon monoxide. J. Neurochem. 45: 965-969.
15
Storm, J. E.; Fechter, L. D. (1985b) Prenatal carbon monoxide exposure differentially affects postnatal weight
and monoamine concentration of rat brain regions. Toxicol. Appl. Pharmacol. 81: 139-146.
Storm, J. E.; Valdes, J. J.; Fechter, L. D. (1986) Postnatal alterations in cerebellar GABA content, GABA
20 uptake and morphology following exposure to carbon monoxide early in development. Dev. Neurosci. 8:
251-261.
Stupfel, M.; Bouley, G. (1970) Physiological and biochemical effects on rats and mice exposed to small
concentrations of carbon monoxide for long periods. Ann. N. Y. Acad. Sci. 174: 342-368.
25
Stupfel, M.; Mordelet-Dambrine, M.; Vauzelle, A. (1981) COHb formation and acute carbon monoxide
intoxication in adult male rats and guinea-pigs. Bull. Eur. Physiopathol. Respir. 17: 43-51.
Styka, P. E.; Penney, D. G. (1978) Regression of carbon monoxide-induced cardiomegaly. Am. J. Physiol. 235:
30 H516-H522.
Sultzer, D. L.; Brinkhous, K. M.; Reddick, R. L.; Griggs, T. R. (1982) Effect of carbon monoxide on
atherogenesis in normal pigs and pigs with von Willebrand's disease. Atherosclerosis 43: 303-319.
35 Svendsen, K. H.; Kuller, L. H.; Martin, M. J.; Ockene, J. K. (1987) Effects of passive smoking in the multiple
risk factor intervention trial. Am. J. Epidemiol. 126: 783-795.
Swiecicki, W. (1973) Wplyw wibracji i treningu fizycznego na przemiane weglowodanowa u szczurow zatrutych
tlenkiem wegla [The effect of vibration and physical training on carbohydrate metabolism in rats
40 intoxicated with carbon monoxide]. Med. Pr. 34: 399-405.
Sylvester, J. T.; Scharf, S. M.; Gilbert, R. D.; Fitzgerald, R. S.; Traystman, R. J. (1979) Hypoxic and CO
hypoxia in dogs: hemodynamics, carotid reflexes, and catecholamines. Am. J. Physiol. 236: H22-H28.
45 Syvertsen, G. R.; Harris, J. A. (1973) Erythropoietin production in dogs exposed to high altitude and carbon
monoxide. Am. J. Physiol. 225: 293-299.
Tachi, N.; Aoyama, M. (1983) Effect of cigarette smoke and carbon monoxide inhalation by gravid rats on the
conceptus weight. Bull. Environ. Contam. Toxicol. 31: 85-92.
50
Tachi, N.; Aoyama, M. (1986) Effect of restricted food supply to pregnant rats inhaling carbon monoxide on
fetal weight, compared with cigarette smoke exposure. Bull. Environ. Contam. Toxicol. 37: 877-882.
March 12, 1990 10-213 DRAFT-DO NOT QUOTE OR CITE
-------
Theodore, J.; O'Donnell, R. D.; Back, K. C. (1971) Toxicological evaluation of carbon monoxide in humans and
other mammalian species. JOM J. Occup. Med. 13: 242-255.
Thomsen, H. K. (1974) Carbon monoxide-induced atherosclerosis in primates: an electron-microscopic study on
5 the coronary arteries of Macaco irus monkeys. Atherosclerosis 20: 233-240.
Thomsen, H. K.; Kjeldsen, K. (1975) Aortic intimal injury in rabbits: an evaluation of a threshold limit. Arch.
Environ. Health 30: 604-607.
10 Traystman, R. J. (1978) Effect of carbon monoxide hypoxia and hypoxic hypoxia on cerebral circulation. In:
Otto, D. A., ed. Multidisciplinary perspectives in event-related brain potential research: proceedings of
the fourth international congress on event-related slow potentials of the brain (EPIC IV); April 1976;
Hendersonville, NC. Washington, DC: U. S. Environmental Protection Agency, Office of Research and
Development; pp. 453-458; EPA report no. EPA-600/9-77-043. Available from: NTIS, Springfield, VA;
15 PB-297137.
Traystman, R. J.; Fitzgerald, R. S. (1977) Cerebral circulatory responses to hypoxic hypoxia and carbon
monoxide hypoxia in carotid baroreceptor and chemoreceptor denervated dogs. Acta Neurol. Scand.
56(suppl. 64): 294-295.
20
Traystman, R. J.; Fitzgerald, R. S. (1981) Cerebrovascular response to hypoxia in baroreceptor- and
chemoreceptor-denervated dogs. Am. J. Physiol. 241: H724-H731.
Traystman, R. J.; Fitzgerald, R. S.; Loscutoff, S. C. (1978) Cerebral circulatory responses to arterial hypoxia in
25 normal and chemodenervated dogs. Circ. Res. 42: 649-657.
Traystman, R. J.; Gurtner, G. H.; Rogers, M. C.; Jones, M. D., Jr.; Koehler, R. C. (1981) A possible role of
oxygenases in the regulation of cerebral blood flow. In: Kovach, A. G. B.; Sandor, P.; KoMai, M., eds.
Cardiovascular physiology: natural mechanisms. Budapest, Hungary: Pergamon Press; pp. 167-177.
30 (Adv. Physiol. Sci. v. 9).
Trese, M. T.; Krohel, G. B.; Hepler, R. S. (1980) Ocular effects of chronic carbon monoxide exposure. Ann.
Ophthalmol. 12: 536-538.
35 Turino, G. M. (1981) Effect of carbon monoxide on the cardiorespiratory system; carbon monoxide toxicity:
physiology and biochemistry. Circulation 63: 253A-259A.
Turner, D. M.; Lee, P. N.; Roe, F. J. C.; Gough, K. J. (1979) Atherogenesis in the White Carneau pigeon:
further studies of the role of carbon monoxide and dietary cholesterol. Atherosclerosis 34: 407-417.
40
U. S. Environmental Protection Agency. (1979) Air quality criteria for carbon monoxide. Research Triangle
Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and Assessment
Office; EPA report no. EPA-600/8-79-022. Available from: NTIS, Springfield, VA; PB81-244840.
45 U. S. Environmental Protection Agency. (1984) Revised evaluation of health effects associated with carbon
monoxide exposure: an addendum to the 1979 EPA air quality criteria document for carbon monoxide.
Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria
and Assessment Office; EPA report no. EPA-600/9-83-033F. Available from: NTIS, Springfield, VA;
PB85-103471.
50
Van Houdt, J. J.; Alink, G. M.; Boleij, J. S. M. (1987) Mutagenicity of airborne particles related to
meteorological and air pollution parameters. Sci. Total Environ. 61: 23-36.
March 12, 1990 10-214 DRAFT-DO NOT QUOTE OR CITE
-------
Vanoli, E.; De Ferrari, G. M.; Stramba-Badiale, M.; Bickestaff, L. H.; Dickey, T.; Farber, J. P.; Schwartz, P.
J. (1986) Carbon monoxide and transient ischemia in conscious dogs with a healed myocardial infarction.
Presented at: 70th annual meeting of the Federation of American Societies for Experimental Biology;
April; St. Louis, MO. Fed. Proc. Fed. Am. Soc. Exp. Biol. 45: 778.
5
Vanoli, E.; De Ferrari, G. M.; Stramba-Badiale, M.; Farber, J. P.; Schwartz, P. J. (1989) Carbon monoxide and
lethal arrhythmias in conscious dogs with a healed myocardial infarction. Am. Heart J. 117: 348-357.
Venning, H.; Roberton, D.; Milner, A. D. (1982) Carbon monoxide poisoning in an infant. Br. Med. J. 284:
10 651.
Vogel, J. A.; Gleser, M. A. (1972) Effect of carbon monoxide on oxygen transport during exercise. J. Appl.
Physiol. 32: 234-239.
15 Vogel, J. A.; Gleser, M. A.; Wheeler, R. C.; Whitten, B. K. (1972) Carbon monoxide and physical work
capacity. Arch. Environ. Health 24: 198-203.
Vollmer, E. P.; King, B. G.; Birren, J. E.; Fisher, M. B. (1946) The effects of carbon monoxide on three types
of performance, at simulated altitudes of 10,000 and 15,000 feet. J. Exp. Psychol. 36: 244-251.
20
Von Post-Lingen, M.-L. (1964) The significance of exposure to small concentrations of carbon monoxide. Proc.
R. Soc. Med. 57: 1021-1029.
Von Restorff, W.; Hebisch, S. (1988) Dark adaptation of the eye during carbon monoxide exposure in smokers
25 and nonsmokers. Aviat. Space Environ. Med. 59: 928-931.
Vreman, H. J.; Kwong, L. K.; Stevenson, D. K. (1984) Carbon monoxide in blood: an improved microliter
blood-sample collection system, with rapid analysis by gas chromatography. Clin. Chem.
(Winston-Salem, NC) 30: 1382-1386.
30
Wahl, M.; Kuschinsky, W. (1976) The dilatory action of adenosine on pial arteries of cats and its inhibition by
theophylline. Pfluegers Arch. 362: 55-59.
Wahl, M.; Kuschinsky, W. (1979) The dilating effect of bistamine on pial arteries of cats and its mediation by
35 H2 receptors. Circ. Res. 44: 161-165.
Wahl, M.; Deetjin, P.; Thuran, D.; Ingvar, D. H.; Lassen, N. A. (1970) Micropuncture evaluation of the
importance of perivascular pH for the arteriolar diameter on the brain surface. Pfluegers Arch. 316:
152-163.
40
Wald, N. J.; Boreham, J.; Bailey, A.; Ritchie, C.; Haddow, J. E.; Knight, G. (1984) Urinary cotinine as marker
of breathing other people's tobacco smoke. Lancet (8370): 230-231.
Wall, M. A.; Johnson, J.; Jacob, P.; Benowitz, N. L. (1988) Cotine in the serum, saliva, and urine of
45 nonsmokers, passive smokers, and active smokers. Am. J. Public Health 78: 699-701.
Ward, A.; Wheatley, M. (1947) Sodium cyanide: sequence of changes of activity induced at various levels of the
central nervous system. J. Neuropathol. Exp. Neurol. 6: 292-294.
50 Webster, W. S.; Clarkson, T. B.; Lofland, H. B. (1970) Carbon monoxide-aggravated atherosclerosis in the
squirrel monkey. Exp. Mol. Pathol. 13: 36-50.
March 12, 1990 10-215 DRAFT-DO NOT QUOTE OR CITE
-------
Weir, F. W.; Rockwell, T. H.; Mehta, M. M.; Attwood, D. A.; Johnson, D. F.; Herrin, G. D.; Anglen, D.
M.; Safford, R. R. (1973) An investigation of the effects of carbon monoxide on humans in the driving
task: final report. Columbus, OH: The Ohio State University Research Foundation; contract no.
68-02-0329 and CRC-APRAC project CAPM-9-69. Available from: NTIS, Springfield, VA; PB-224646.
5
Weiser, P. C.; Morrill, C. G.; Dickey, D. W.; Kurt, T. L.; Cropp, G. J. A. (1978) Effects of low-level carbon
monoxide exposure on the adaptation of healthy young men to aerobic work at an altitude of 1,610
meters. In: Folinsbee, L. J.; Wagner, J. A.; Borgia, J. F.; Drinkwater, B. L.; Gliner, J. A.; Bedi, J. F.,
eds. Environmental stress: individual human adaptations. New York, NY: Academic Press, Inc.; pp.
10 101-110.
Weiss, H. R.; Cohen, J. A. (1974) Effects of low levels of carbon monoxide on rat brain and muscle tissue PQ .
Environ. Physiol. Biochem. 4: 31-39.
15 Weissbecker, L,; Carpenter, R. D.; Luchsinger, P. C.; Osdene, T. S. (1969) In vitro alveolar macrophage
viability: effect of gases. Arch. Environ. Health 18: 756-759.
Wells, L. L. (1933) The prenatal effect of carbon monoxide on albino rats and the resulting neuropathology.
Biologist 15: 80-81.
20
Wilks, S. S.; Tomashefski, J. F.; Clark, R. T., Jr. (1959) Physiological effects of chronic exposure to carbon
monoxide. J. Appl. Physiol. 14: 305-310.
Williams, I. R.; Smith, E. (1935) Blood picture, reproduction, and general condition during daily exposure to
25 illuminating gas. Am. J. Physiol. 110: 611-615.
Wilson, D. F.; Erecinska, M.; Drown, C.; Silver, I. A. (1979) The oxygen dependence of cellular energy
metabolism. Arch. Biochem. Biophys. 195: 485-493.
30 Wilson, D. A.; Manley, D. F.; Feldman, M. A.; Traystman, R. J. (1987) Influence of chemoreceptors on
neurohypophyseal blood flow during hypoxic hypoxia. Circ. Res. 61: II94-II101.
Winn, H. R.; Rubio, R.; Berne, R. M. (1981) Brain adenosine concentration during hypoxia in rats. Am. J.
Physiol. 241: H235-H242.
35
Winneke, G. (1974) Behavioral effects of methylene chloride and carbon monoxide as assessed by sensory and
psychomotor performance. In: Xintaras, C.; Johnson, B. L.; de Groot, I., eds. Behavioral toxicology:
early detection of occupational hazards [proceedings of a workshop; June 1973; Cincinnati, OH].
Cincinnati, OH: U. S. Department of Health, Education, and Welfare, National Institute for Occupational
40 Safety and Health; pp. 130-144; DHEW publication no. (NIOSH) 74-126. Available from: NTIS,
Springfield, VA; PB-259322.
Wouters, E. J. M.; de Jong, P. A.; Cornelissen, P. J. H.; Kurver, P. H. J.; van Oel, W. C.; van Woensel, C.
L. M. (1987) Smoking and low birth weight: absence of influence by carbon monoxide? Eur. J. Obstet.
45 Gynecol. Reprod. Biol. 25: 35-41.
Wright, G. R.; Shephard, R. J. (1978a) Brake reaction time - effects of age, sex, and carbon monoxide. Arch.
Environ. Health 33: 141-149.
50 Wright, G. R.; Shephard, R. J. (1978b) Carbon monoxide exposure and auditory duration discrimination. Arch.
Environ. Health 33: 226-235.
Wright, G.; Randell, P.; Shephard, R. J. (1973) Carbon monoxide and driving skills. Arch. Environ. Health 27:
349-354.
March 12, 1990 10-216 DRAFT-DO NOT QUOTE OR CITE
-------
Xintaras, C.; Johnson, B. L.; Ulrich, C. E.; Terrill, R. E.; Sobecki, M. F. (1966) Application of the evoked
response technique in air pollution toxicology. Toxicol. Appl. Pharmacol. 8: 77-87.
Yamamoto, K. (1976) Acute combined effects of HCN and CO with the use of the combustion products from
5 PAN (polyacrylonitrile) - gauze mixtures. Z. Rechtsmed. 78: 303-311.
Young, S. H.; Stone, H. L. (1976) Effect of a reduction in arterial oxygen content (carbon monoxide) on
coronary flow. Aviat. Space Environ. Med. 47: 142-146.
10 Zebro, T.; Littleton, R. J.; Wright, E. A. (1976) Adaptation of mice to carbon monoxide and the effect of
splenectomy. Virchows Arch. A: Pathol. Anat. Histopathol. 371: 35-51.
Zebro, T.; Wright, E. A.; Littleton, R. J.; Prentice, A. I. D. (1983) Bone changes in mice after prolonged
continuous exposure to a high concentration of carbon monoxide. Exp. Pathol. 24: 51-67.
15
Zorn, H. (1972) The partial oxygen pressure in the brain and liver at subtoxic concentrations of carbon
monoxide. Staub Reinhalt. Luft 32: 24-29.
Zwart, A.; Buursma, A.; Oeseburg, B.; Zijlstra, W. G. (1981) Determination of hemoglobin derivatives with the
20 IL 282 CO-oximeter as compared with a manual spectrophotometric five-wavelength method. Clin. Chem.
(Winston-Salem, NC) 27: 1903-1907.
Zwart, A.; Buursma, A.; Van Kempen, E. J.; Oeseburg, B.; van der Ploeg, P. H. W.; Zijlstra, W. G. (1981) A
multi-wavelength spectrophotometric method for the simultaneous determination of five haemoglobin
25 derivatives. J. Clin. Chem. Clin. Biochem. 19: 457-463.
March 12, 1990 10-217 DRAFT-DO NOT QUOTE OR CITE
-------
11. COMBINED EXPOSURE OF CARBON MONOXIDE
WITH OTHER POLLUTANTS, DRUGS,
AND ENVIRONMENTAL FACTORS
11.1 HIGH ALTITUDE EFFECTS OF CARBON MONOXIDE
11.1.1 Introduction
Precise estimates of the number of people exposed to CO at high altitude are not readily
available. As of 1980, however, more than 4.2 (Lindsey, 1989) million people were living at
10 altitudes in excess of 1524 m (5000 ft). Moreover, estimates obtained from several states
with mountainous regions (i.e. California, Nevada, Hawaii, and Utah) indicate that more than
35 million tourists may sojourn in high altitude areas during the summer and winter months.
The potential effects on human health of inhaling CO at high altitudes are complex.
Whenever CO binds to Hb it reduces the amount of Hb available to carry O2. People at high
15 altitudes already live in a state of hypoxemia, however, because of the reduced partial
pressure of oxygen (POj) in the air. Carbon monoxide, by binding to Hb, intensifies the
hypoxemia existing at high altitudes by further reducing transport of O2 to the tissues. Hence,
the effects of CO and high altitude usually are considered to be additive.
This consideration does not take into account the fact that within hours (perhaps sooner)
20 of arrival at high altitude, however, hemoconcentration occurs which increases the Hb
concentration. The increased Hb concentration offsets the decreased O2 saturation and
restores O2 concentration to pre-ascent levels. Consequently, the simple additive model of
COHb and altitude hypoxemia may be valid only during early altitude exposure.
The visitor newly arrived to higher altitudes may be at greater risk from CO than the
25 adapted resident, however, because of a noncompensated respiratory alkalosis from
hyperventilation, lower arterial Hb saturation without a compensatory absolute polycythemia
(therefore greater hypoxemia) and hypoxia-induced tachycardia. (See Chapter 12,
Section 12.5 for further discussion of this topic.)
Several factors tend to exacerbate ambient CO levels at high altitude (Kirkpatrick and
30 Reeser, 1976). For example, in mountain communities, automobile emissions are higher.
Automobiles tuned for driving at 1610 m (5280 ft) emit almost 1.8 times more CO when
March 12, 1990 11-1 DRAFT - DO NOT QUOTE OR CITE
-------
driven at 2438 m (8000 ft). Automobiles tuned for driving at sea level emit almost four times
more CO when driven at 2438 m. Moreover, automobile emissions are increased by driving
at reduced speeds, along steep grades, and under poor-driving conditions. Therefore, large
influxes of tourists driving automobiles tuned for sea-level conditions into high-altitude resort
5 areas may drastically increase pollutant levels in general, and CO levels in particular
(National Research Council, 1977). Newer automobile engine technologies, however, should
significantly reduce CO emissions in general, as well as CO emissions at high altitude.
Heating devices (space heaters and fireplaces) used for social effect, as well as warmth, are a
second factor contributing to CO emissions in mountain resort areas. Finally, population
10 growth in mountain areas is concentrated along valley floors; this factor combined with the
reduced volume of air available for pollutant dispersal in valleys causes pollutants, including
CO, to accumulate in mountain valleys. As a result of these factors, the NAAQS for CO of
9 ppm is exceeded frequently in Denver, CO, (altitude 1610 m) during the winter months
(Haagenson, 1979).
15 Because of concern for CO exposure at high altitudes, it has been suggested that the
NAAQS for CO set at sea level is probably too high for altitudes of 1500 m and above. An
example of supporting data for this opinion were studies conducted before 1950 on the
psychophysiological effects of high altitude and CO-induced hypoxia. These studies provided
evidence for a concept that there are physiologically equivalent altitudes dependent on the
20 ambient concentration of CO. In 1976, the states of California and Nevada adopted ambient
standards for the Lake Tahoe air basin (1900 m; 6231 ft), which were more stringent than the
NAAQS (i.e., 6 ppm rather than 9 ppm).
Mitchell et al. (1979) justified this concept by stating that "equivalent
carboxyhemoglobin levels observed at sea level would occur during exposure to lower
25 ambient CO concentrations at 1500 m." The high-altitude standard was calculated from the
model developed by Coburn et al. (1965). This model was developed for quasi-steady-state
responses to low CO concentrations, such as those produced endogenously, however, and was
not intended to apply to other, exogenous sources of CO (see Chapter 9 for a description of
the model). Collier and Goldsmith (1983) acknowledged that an error was made in the
30 original calculations for the California-Nevada high-altitude standard. They expanded the
March 12, 1990 11-2 DRAFT - DO NOT QUOTE OR CITE
-------
10
15
20
25
30
35
computations to include factors relating to the ambient CO concentrations and altitude and
concluded, based on their model calculations, that the expected altitude effect would be small.
11.1.2 Carboxyhemoglobin Formation
The effects of high altitude on COHb formation have been considered in a theoretical
paper by Collier and Goldsmith (1983). Transforming and rearranging the Coburn-Foster-
Kane equation (Coburn et al., 1965), these workers derived an equation expressing COHb in
terms of endogenous and exogenous sources of CO. Thus,
FICO (PB-47) + VCO Z
where:
SCO =
SCO
FICO
?B
VCO
K
Z
106K
(11-1)
K
and
and
M
SO, =
DLCO
(YO
COHb (%)
Fraction inspired CO (ppm)
Barometric Pressure (torr)
Rate of CO production (mL«min-' STPD)
P7O2/(M x SO,)
1/DLCO + (PB-47/CVg
mean partial pressure of pulmonary capillary O2 (torr)
Haldane constant
O2 Hb
CO diffusing capacity (mL-min-'-torr-1)
alveolar ventilation (mL-min-1 STPD)
According to this relationship, a given PCO will result in a higher percent COHb at
high altitudes (where PO2 is reduced). Thus, Collier and Goldsmith calculate that humans
breathing 8 ppm CO will have equilibrium COHb levels of 1.4% at sea level and 1.6, 1.8,
and 1.8%, respectively, at 1530, 3050, and 3660 m (Table 11-1). Moreover, these workers
calculate an increase in COHb at altitude even in the absence of inhaled CO (due to
endogenous production of CO).
40
March 12, 1990
11-3 DRAFT - DO NOT QUOTE OR CITE
-------
10
15
TABLE 11-1. CALCULATED EQUILIBRIUM VALUES OF PERCENT COHb
AND PERCENT 0^ IN HUMANS EXPOSED
TO AMBIENT CO AT VARIOUS ALTITUDES
Ambient CO
(ppm)
0
4
8
12
16
Sea
%
COHb
0.20
0.8
1.4
2.1
2.7
level
%
02Hb
97.3
96.8
96.2
95.6
95.1
530
%
COHb
0.26
0.9
1.6
2.3
2.9
m
%
02Hb
93.6
93.0
92.5
91.9
91.4
3050
%
COHb
0.35
1.1
1.8
2.5
3.2
m
%
02Hb
82.4
82.1
81.7
81.3
80.9
3660
%
COHb
0.37
1.1
1.8
2.5
3.2
m
%
O2Hb
73.3
73.1
72.9
72.7
72.5
20
Notes: The table is for unacclimatized, sedentary individuals at one level of activity (v*O2 = 500 ml^min-1).
Source: Adapted from Collier and Goldsmith (1983).
25
11.1.3 Cardiovascular Effects
30 There are studies comparing the cardiovascular responses to CO with those to high
altitude, but there are relatively few studies of the cardiovascular responses to CO at high
altitude (see Table 11-2). Forbes et al. (1945) reported that CO uptake increased during six
minutes of exercise of varying intensity on a bicycle ergometer at 4877 m (16,000 ft). The
increased CO uptake was caused by altitude hyperventilation stimulated by decreased arterial
35 O2 tension and not by diminished barometric pressure.
Pitts and Pace (1947) reported that pulse rate increased in response to the combined
stress of high altitude and CO. The subjects were 10 healthy men who were exposed to
simulated altitudes of 2134, 3048 and 4572 m (7000, 10,000, and 15,000 ft) and inhaled
3000 or 6000 ppm CO to obtain COHb levels of 6 or 13%, respectively. The mean pulse
40 rate during exercise and the mean pulse rate during the first five minutes after exercise were
correlated with and increased with the COHb concentration and simulated altitude. The
authors concluded that the response to a 1 % increase in COHb level was equivalent to that
March 12, 1990 11-4 DRAFT - DO NOT QUOTE OR CITE
-------
2
§•
TABLE 11-2. SUMMARY OF EFFECTS OF CARBON MONOXIDE AT ALTITUDE
Exposure COHb
Alt = 4,877 m
3,000-4,000 ppm CO
6 min exercise
Alt = 1,524- 1,848m 5-10%
CO = 1,500-2,000 ppm
Alt - 3.070-4.555 m 5-22%
CO = 2,800-5.600 ppm
Simulated alt - 2,134, 6 and 13%
3,048, 4,572 m (16, 14.
11% 0;+ Nj) 3,000 or
6,000 ppm
CO treadmill exercise
Alt = 2,134-4,877 m 1.1-20.5%
CO = 100-300 ppm
MCO in rebrealhing -
system P,O2 varied from
650 to 40 mm Hg
CO administered at 2-75 %
constant rate
All = 305-3.109 m 4.77-6.66%
smokers
Subject
Human
(n=3
Human
(n=5)
Human
(n=20)
Human
(n=10)
Human
(n=4)
Dog
(n=31)
Dog
(n=4)
Human
(n=62)
Dependent Variable
Blood CO
Flicker fusion
frequency (FFF)
Critical flicker
frequency (CFF),
body sway (BS),
red visual field
(RVF)
Pulse rate
Visual sensitivity
MCO in blood
Rale of increase in
COHb
COHb levels
Results
CO uptake increased
with altitude
FFF decreased with
CO at altitude
CFF, BS, and RVF
impaired by altitude;
no added effect of
CO
Pulse rate during and
5 min after exercise
increased with altitude and
COHb
CO decreased visual sensitivity
No change in I4CO activity
in blood when P,O2 varied
from 40 to 650 mmHg; MCO
decreased lo 50% control
when P.O2 decreased below
40 mmHg
COHb increased at
constant up to 50%; at
50%, rate of COHb forma-
tion decreased
COHb in smokers higher
at altitude than at sea
level
Comments
Caused by altitude hyperventilation.
FFF not affected by altitude or COHb
alone; 8-10% COHb reduced altitude
tolerance by 1,215 m.
No correlation of any response with
COHb. Effects of CO may be masked by
compensatory effect.
Response to 1 % COHb equivalent to
increase in altitude of 335 ft.
The effects of CO and altitude
are additive.
Recovery from the detrimental
effects of CO lagged behind elimina-
tion of CO from blood.
With severe arterial hypoxemia
(P,O2 <40 mmHg) MCO shifts into
extravascular tissue.
Suggests that at high COHb levels
CO shifts into extravascular space.
Reference
Forbes et al. (1945)
Lillianthal & Fugitt
(1946)
Vollmer et al. (1946)
Pitta & Pace (1947)
Halperin et al. (1959)
Luomanmaki & Cobum
(1969)
Luomanmaki & Coburn
(1969)
Brewer et al. (1970)
-------
I
»— *
JO
\o
8
,_,
>— *
ON
O
l>
!-H
H
I
23
0
H
O
cj
O
i—3
M
0
O
TABLE 11-2 (cont'd). SUMMARY OF EFFECTS OF CARBON MONOXIDE AT ALTITUDE
Exposure COHb
Alt = 2,438 m 5%
Alt = 4,500 m 20%
CO = 4,300 ppm every
second hour for
3-5 hours
Alt = 3,109 m 0.4-7.14%
Smokers
Alt = 3,048 m 4.2%
CO bolus followed by
40 ppm.
Bicycle exercise
Alt = 1,610m 5%
CO = 100% bolus
Alt = 1,524 m 20%
CO = 160-200 ppm
6 weeks
Alt = 3, 100m 1.8-6.2%
Smokers
Subject
Human
Human
(n=16)
Human
(n=49)
Human
(n=12)
Human
(n=9)
Goat
(n=6)
Human
(n=44)
Dependent Variable
Visual sensitivity
Capillary perme-
ability to protein
(CP)
COHb and O2 affinities
Cardiac output, stroke
volume (SV), arterial-
mixed venous O2
difference (A-V)
Work performance
Cardiac index (CO,
Stoke volume (SV),
Heart rate (HR),
Ventricular con-
tractility (V^j)
Infant birth weights
Results
Visual sensitivity
decreased by CO at
altitude
CP increased with CO
but not with altitude;
plasma volume decreased
with altitude
COHb levels and O2
affinities decreased in poly-
cythemic smokers on
cessation of smoking
At altitude, cardiac
output increased and
SV and A-V decreased
in nonsmokers; no
effect on smokers.
Increased working HR,
and shortened post-
exercise LV ejection
lime; towered anaerobic
threshold
No effect on CI, SV,
HR, and V^ during
exposure
Maternal smoking
associated with 2-3
times greater reduc-
tion in infant birth
weight than at sea
level
Comments
5 % COHb depresses visual sensitivity
as much as 2,438-3,048 m. The effects
of altitude and CO are additive.
Increase in CP appears unique to CO
(nonhypoxic effect).
O2 dropped to lower than normal sea
level values in polycythemic
smokers on cessation of
smoking.
Smokers may be partially adapted to
hypoxic environments and CO.
CO impaired exercise performance to
same degree as at KB level.
After removal from CO, both
HR and V,,^, were depressed.
COHb levels measured in mothers
were inversely related to infant
birth weight.
Reference
McFarland (1970);
McFarland et al. (1944)
Parving (1972)
Brewer et al. (1974)
Wagner et al. (1978)
Weiser et al. (1978)
lames et al. (1979)
Moore et al. (1982)
-------
1
jo
\o
,_,
^
-J
O
Tl
H
1
O
z
o
H
c
o
H
M
O
O
H
w
Exposure
Alt = 4,572 m
CO = 500 ppm
6 weeks
Alt = 5,486 m
CO = 50, 100,
500 ppm
6 weeks
Alt = 4,572 m
CO = 100 ppm
6 weeks
Alt = 55, 1,524,
2,134, and 3, 048 m;
CO = 0, 50, 100,
and 150 ppm
Alt = 55 and 2,134 m;
CO = 0 and 9 ppm for
8h
TABLE 1 1-2 (cont'd)
COHb Subject
36.2% and Rat
34.1% (n=24)
5.8, 11.1, Rat
and 4.26% (n=22)
8.4% Rats
(n=24)
2.56 - 4.42% Human
(n=23)
(11 men;
12 women)
0.2 - 0.7% Human
(n=IT)
. SUMMARY OF
Dependent Variable
Hematocrit (Hct),
mean electrical
axis (MEA),
HW/BW ratios
Cardiac hypertrophy,
coronary capillarity
Hct ratio and weights:
BW, HW, RV, LV + S,
Pituitary (PiT)
Maximum aerobic
capacity (VO2 max)
Maximum^aero'bic
capacity (VO2 max)
EFFECTS OF CARBON
Results
Hct increased by altitude
and CO; MEA shifted
left with CO, right
with altitude
RV hypertrophy and
coronary capillarity
increased with altitude
Alt 1BW, tHct, TRY,
tHT, TPiT; CO tHct,
TLV+S
VO2 max decreased at 2,134m
(4%) and 3,048m (8%); VO2 max
decreased slightly with
increasing ambient CO.
VO2 max decreased 7-10% with
increasing altitude at 0 ppm CO;
similar effects were found after
8-h exposure to 9 ppm CO, regard-
less of exercise level
MONOXIDE AT ALTITUDE
Comments Reference
Effects of altitude and CO on Hct, MEA, Cooper et el. (1985)
and HW/BW were additive.
Increase in coronary McDonagh et al. (1986)
capillarity was blocked by CO.
Effects produced by altitude were McGrath (1988)
not intensified by 100 ppm CO.
Altitude- and CO-hypoxia independ- Horvath et al. (1988 a,b)
ently affect VO2 max; decreased COHb
with increasing altitude was due, in
part, to decreased driving CO pressure
VO2 max was reduced in all subjects Horvath and Bed! (1989)
at altitude regardless of the ambient
CO level
-------
obtained by raising a normal group of men 102 m (335 ft) in altitude. This relationship was
stated for a range of altitudes from 2134 to 3048 m (7000 to 10,000 ft) and for increases in
COHbupto 13%.
Weiser et al. (1978) studied the effects of CO on aerobic work at 1610 m (5280 ft) in
5 young subjects inhaling 100% CO until COHb levels reached 5%. They reported that this
level of COHb impaired exercise performance at high altitude to the same extent as that
reported at sea level (Horvath et al., 1975). Because these subjects were Denver residents
and fully adapted to this altitude, however, they would have had an arterial O2 concentration
the same as at sea level (about 20 mL O2/dL). Hence, 5% COHb would lower arterial O2
10 concentration about the same amount at both altitudes and impair work performance at
altitude to the same extent as at sea level. In the Weiser study, breathing CO during
submaximal exercise caused small but significant changes in cardiorespiratory function; the
working heart rate increased and the postexercise left ventricular ejection time shortened, but
not to the same extent as when filtered air was breathed. CO lowered the anaerobic threshold
15 and, at work rates heavier than the anaerobic threshold, increased minute ventilation.
Wagner et al. (1978) studied young smokers and nonsmokers who exercised at 53% of
their VO2 max at 760 and 523 torr. Carboxyhemoglobin levels were raised to 4.2%. While
at altitude with these elevated COHb levels, nonsmokers increased their cardiac output and
decreased their arterial-mixed venous O2 differences. Smokers did not respond in a similar
20 manner. Smokers, with their initial higher Hb concentrations, may have developed some
degree of adaptation to CO and/or high altitude.
Horvath et al. (1988) in a complex study involving four altitudes (up to 3050 m) and
four ambient CO concentrations (up to 150 ppm) evaluated COHb levels during a maximal
aerobic capacity test. They concluded that VO2 max values determined in men were only
25 slightly diminished due to increased ambient CO. Carboxyhemoglobin concentrations attained
at maximum were highest at 55 m (4.42%) and lowest at 3035 m (2.56%) while breathing
150 ppm CO (Figure 11-1). This was attributed to the reduced partial pressure of CO at high
altitude. No additional effects that could be attributed to the combined exposure to high
altitude and CO were found. Independence of the altitude and CO hypoxia was demonstrated
30 under the condition of performing a maximum aerobic capacity test. The reductions in VO2
March 12, 1990 11-8 DRAFT - DO NOT QUOTE OR CITE
-------
MAX
5 MIN POST MAX
.a
E
O
CJ
w
o
-2
Control COHb % = 0.67 ±0.15
0 50 100150 0 50 100150 0 50 100150 0 50 100150
CARBON MONOXIDE LEVEL (PPM)
B
O
O
5
w
a
-2
MAX
ES3 5 MIN POST MAX
Control COHb 0.81 ± 0.23
0 50 100150 0 50 100150 0 50 100150 0 50 100150
CARBON MONOXIDE LEVEL (PPM)
Figure 11-1. Increment in percent carboxyhemoglobin (A% HbCO) over basal (control) levels
at the end of a maximum aerobic capacity test and at the 5th min of recovery from a test in a
typical (A) male and (B) female subject. Altitudes are 55, 1524, 2134, and 3048 m, whereas
exercise was conducted with ambient concentrations of 0, 50, 100, and 150 ppm CO.
Source: Horvath et al. (1988).
March 12, 1990
11-9 DRAFT - DO NOT QUOTE OR CITE
-------
max due to high altitude and to the combined exposure of ambient CO and high altitude were
similar.
Horvath and Bedi (1989) studied 17 nonsmoking young men to determine the alterations
in COHb during exposure to 0 or 9 ppm ambient CO for eight hours at sea level or
5 an altitude of 2134 m (7000 ft). Nine subjects rested during the exposures and eight
exercised for the last 10 minutes of each hour at a mean ventilation of 25 L (BTPS). All
subjects performed a maximal aerobic capacity test at the completion of their respective
exposures. At the low CO concentrations studied, the Coburn, Forster, and Kane (CFK)
equation estimated COHb levels to be 1.4% (Petersen and Stewart, 1975). Carboxy-
10 hemoglobin concentrations fell in all subjects during their exposures to 0 ppm CO at sea level
or 2134 m. During the eight hour exposures to 9 ppm CO, COHb levels rose linearly from
approximately 0.2 to 0.7% (Figures 11-2). No significant differences in uptake were found
whether the subjects were resting or intermittently exercising. Levels of COHb were similar
at both altitudes. A portion of the larger estimate of COHb determined by the CFK equation
15 could be accounted for by the use of an assumed blood volume. Maximal aerobic capacity
was reduced approximately 7 to 10% consequent to altitude exposure during 0 ppm CO.
These values were not altered following eight-hour exposure to 9 ppm CO in either resting or
exercising individuals.
20 11.1.4 Chronic Studies
There have been few studies of the long-term effects of CO at altitude (see Table 11-2).
James et al. (1979) studied cardiac function in six unsedated goats that were chronically
instrumented and exposed to 160 to 200 ppm CO (COHb = 20%) for six weeks at 1524 m.
Cardiac index and stroke volume were unchanged during and after the exposure. Heart rate
25 and contractility (V^J of the left ventricular myocardium were unchanged during exposure to
CO, but both were depressed during the first week after removal of the CO. The authors
concluded that if there was a decrease in intrinsic myocardial function during the CO
exposure, it may have been masked by increased sympathetic activity.
McGrath (1988; 1989) studied cardiovascular, body, and organ weight changes in rats
30 exposed continuously for six weeks to (1) ambient altitude, (2) ambient altitude + CO,
March 12, 1990 11-10 DRAFT - DO NOT QUOTE OR CITE
-------
1.0
08
.a
x
o 0.6
o
* 0.4
0.2
O.O
D-D 2134m
*-A. 55m
O-O 0 ppm
Resting Subjects
\
\\
CO 9ppm
-ex..
12345678 M
Hours
B
1.0
0.8
.o
x
00.6
o
0.4
0.2
0.0
D-D 2134m
Ar-A 55m
o-oOppm
Active Subjects
PM
Figure 11-2. Change in carboxyhemoglobin concentration (% COHb) during eight-hour
exposures to 0 to 9 ppm CO for (A) resting and (B) exercising subjects. Altitudes are sea
level (55 m) and 2134 m (7000 ft).
Source: Horvath and Bedi (1989).
March 12, 1990
11-11 DRAFT - DO NOT QUOTE OR CITE
-------
(3) simulated high altitude, and (4) CO at high altitude. Altitudes ranged from 3300 ft
(1000 m) to 18,000 ft (5486 m ) and CO concentrations from 0 to 500 ppm.
Carbon monoxide had no effect on body weight at any altitude. There was a tendency
for hematocrit to increase even at the lowest concentration of CO (9 ppm), but the increase
5 did not become significant until 100 ppm. At 10,000 ft, there was a tendency for the total
heart weight to increase in rats inhaling 100 ppm CO. Although its effects on the heart at
high altitude are complex, CO, in concentrations of 500 ppm or less, had little effect on the
right ventricle; it did not exacerbate any effects due to altitude. There was a tendency for the
left ventricle weight to increase with exposure to 35 ppm carbon monoxide at high altitude,
10 but the increase was not significant until 100 ppm CO. Heart rate, blood pressure, cardiac
output, and peripheral resistance were unaffected by exposure to 35 ppm CO or 10,000-ft
altitude, singly or in combination. The author concluded that six weeks of exposure to 35
ppm CO does not produce measurable effects in the healthy laboratory rat, nor does it
exacerbate the effects produced by exposure to 10,000-ft altitude.
15 The data reported by McGrath (1988; 1989) are generally in agreement with findings
reported by other investigators. Carboxyhemoglobin obtained at the end of the six weeks of
exposure to CO are presented in Figure 11-3. The COHb concentrations were (at 3300 ft)
0.6, 0.9, 2.4, 3.7, and 8.5% for ambient CO levels of 0, 9, 35, 50, and 100 ppm,
respectively. This relationship can be expressed as:
20
%COHb = 0.115 + O.OSx (11-2)
where x is the CO exposure, in ppm. The correlation coefficient (r) for this relationship was
0.99. The changes at other altitudes were not sufficient to calculate the rate of increase.
25 Exposure of rats to 500 ppm and altitudes up to 18,000 ft resulted in COHb levels of 40 to
42%. An interesting, but not unexpected, finding in this study was that high altitude
residence in the absence of exogenous CO resulted in increased basal COHb concentrations.
These values were 0.6, 1.3, 1.7, and 1.9% for altitudes from 3300 to 18,000 ft. These
increases can be expressed as:
30
%COHb = 0.0000914 + 0.26687x (11-3)
March 12, 1990 11-12 DRAFT - DO NOT QUOTE OR CITE
-------
HUD HO HXOnb JLON Od - JLdVHd CMI
066T 'Zl
Percent COHb
o -»
a
ore
EL
cr.
I
I
I-F
8
I
S*
a
I
-------
where x is altitude, in feet. The correlation coefficient (r) for this relationship was 0.99.
Whether similar increases in basal COHb concentrations would be observed in humans
adapted to altitude needs to be determined. Presumably, because there are marked elevations
in Hb and Hct in residents of high altitudes, a greater endogenous CO production might be
5 present. In this study, 100 ppm CO had no effect on body, right ventricle, total heart,
adrenal, spleen, or kidney weights, but it did increase Hct ratios and left ventricle weights.
There was no significant interaction between altitude and CO on any parameter except kidney
weight. The author concluded that although there was a tendency for hematocrit ratios,
spleen weights, and total heart weights to be elevated by combined CO-altitude exposure, the
10 results were not significant and, in general, the effects produced by 4572-m altitude were not
intensified by exposure to 100 ppm CO.
McDonagh et al. (1986) studied cardiac hypertrophy and ventricular capillarity in rats
exposed to 5486 m (18,000 ft) and 50, 100, and 500 ppm CO. Coronary capillarity
increased after exposure to 5486 m for six weeks, but this response was blocked by CO.
15 Right ventricular thickness was increased by altitude, but was not increased further by CO.
At 500 ppm, CO the right ventricular hypertrophy was attenuated, but the results are
uncertain due to the high mortality in this group. Left ventricular thickness also was
increased at 5486 m (18,000 ft) and increased further by CO. The authors concluded that
because the ventricular thickness is increased while capillarity is reduced, it is possible that
20 the myocardium can be underperfused in the altitude plus CO group.
Cooper et al. (1985) evaluated the effects of CO at altitude on EKGs and cardiac
weights in rats exposed for six weeks to (1) ambient (amb), (2) ambient + 500 ppm CO
(amb+CO), (3) 4572 m (15,000 ft) (alt), and (4) 4572 m + 500 ppm CO (alt+CO). COHb
values were 36.2 and 34.1 % in the amb+CO and alt+CO groups, respectively. Hematocrits
25 were 54 ± 1, 77± 1, 68 ± 1, and 82± 1%, in the amb, amb+CO, alt, and alt+CO
groups, respectively. In the amb+CO, alt, and alt+CO groups, respectively, the mean
electrical axis shifted 33.2° left, 30° right, and 116.4° right. Heart weight to body weight
ratios were 2.6, 3.2, 3.2, and 4.0 x 10"3 in the amb, amb+CO, alt, and alt+CO groups,
respectively. Whereas CO increased left ventricular weight, and alt increased right
30 ventricular weight, alt+CO increased both. EKG changes were consistent with changes in
cardiac weight.
March 12, 1990 11-14 DRAFT - DO NOT QUOTE OR CITE
-------
These results indicate that whereas CO inhaled at ambient altitude causes a left
electrical axis deviation, CO inhaled at 4572 m exacerbates the well-known phenomenon of
right electrical axis deviation. Thus, the results from chronic animal studies indicate that
there is little effect of CO on the cardiovascular system of rats exposed to CO concentrations
5 of 100 ppm or less and altitudes up to 3030 m (10,000 ft).
Exposure to CO from smoking may pose a special risk to the fetus at altitude. Moore
et al. (1982) reported that maternal smoking at 3100 m is associated with a two- to threefold
greater reduction in infant birth weight than has been reported at sea level. Moreover, COHb
levels of 1.8 to 6.2% measured in all pregnant subjects, were inversely related to infant birth
10 weight. Earlier, Brewer et al. (1970, 1974) reported that the mean COHb level in smokers at
altitude is higher than smokers at sea level, and that subjects who smoked had greater O2
affinities than nonsmokers. Moreover, cessation of smoking by polycythemic individuals at
altitude results in a marked reduction in COHb and a decrease in hemoglobin-O2 affinity to
values less than those reported for normal individuals at sea level. The chronic effects of
15 altitude and CO exposure are summarized in Table 11-3.
TABLE 11-3. CHRONIC EFFECTS OF ALTITUDE AND CARBON
MONOXIDE EXPOSURE
20 Effect Altitude Carbon Monoxide
Hemoglobin t t
Hematocrit t T
Pulmonary arterial t
25 pressure
Cardiac hypertrophy
Right ventricle t
Both ventricles - I
Cardiac output" tt ?
30 Blood volume t t
Body weight 4
"Initial increase that later returns to baseline value.
March 12, 1990 11-15 DRAFT - DO NOT QUOTE OR CITE
-------
11.1.5 Neurobehavioral Effects
The neurobehavioral effects following CO exposure are controversial and are described
in detail in Chapter 10, (Section 10.4). Those neurobehavioral studies specifically concerned
with CO exposure at altitude are reviewed briefly in this section.
5 McFarland et al. (1944) reported changes in visual sensitivity occurring at a COHb
concentration of 5% or at a simulated altitude of approximately 2432 m (8000 ft). Later,
McFarland (1970) expanded on the original study and noted that a pilot flying at 1829 m
(6000 ft) breathing 0.005% CO in air is at an altitude physiologically equivalent to
approximately 3658 m (12,000 ft). McFarland stated that sensitivity of the visual acuity test
10 was such that even the effects of small quantities of CO absorbed from cigarette smoke were
clearly demonstrable. In subjects inhaling smoke from three cigarettes at 2286 m (7500 ft),
there was a combined loss of visual sensitivity equal to that occurring at 3048 to 3353 m
(10,000 to 11,000 ft). This report was confirmed by Halperin et al. (1959), who also
observed that recovery from the detrimental effects of CO on visual sensitivity lagged behind
15 elimination of CO from the blood.
Lilienthal and Fugitt (1946) reported that combined exposure to altitude and CO
decreased flicker-fusion frequency (FFF) (i.e., the critical frequency in cycles per second at
which a flickering light appears to be steady). Whereas mild hypoxia (that occurring at 2743
to 3658 m (9000 to 12,000 ft)) alone impaired FFF, COHb levels of 5 to 10% decreased the
20 altitude threshold for onset of impairment to 1524 to 1829 m (5000 to 6000 ft).
The psychophysiological effects of CO at altitude are a particular hazard in high-
performance aircraft (Denniston et al., 1978). Acute ascent to altitude increases ventilation
via the stimulating effects of a reduced PO2 on the chemoreceptors. The increased ventilation
causes a slight increase in blood pH and a slight leftward shift in the O2Hb dissociation curve.
25 Although such a small shift would probably have no physiological significance under normal
conditions, it may take on physiological importance for aviators required to fly under a
variety of operational conditions and to perform tedious tasks involving a multitude of
cognitive processes. The leftward shift of the O2Hb dissociation curve may be further
aggravated by the persisting alkalosis caused by hyperventilation resulting from anxiety. The
30 potential for this effect has been reported by Pettyjohn et al. (1977), who reported that
respiratory minute volume may be increased by 110% during final landing approaches
March 12, 1990 11-16 DRAFT - DO NOT QUOTE OR CITE
-------
requiring night-vision devices. Thus, the hypoxia-inducing effects of CO inhalation would
accentuate the cellular hypoxia caused by stress and altitude-induced hyperventilation.
11.1.6 Compartmental Shifts
5 Studies by Luomanmaki and Coburn (1969) suggest that CO in very high
concentrations may pose a special threat at higher altitudes. These workers report that during
hypoxia, in anesthetized dogs, CO shifts out of the blood and into the tissues. In
experiments using 14CO, they observed that radioactivity in blood did not change when arterial
O2 tension increased from 50 to 500 mmHg. However, 14CO activity in blood decreased to
10 50% of control levels when arterial PO2 decreased below 40 mmHg; 14CO shifted back into
the blood when arterial PO2 returned to normal. Because there was no significant difference
between splenic and central venous WCO radioactivity either before or after the MCO shift,
these workers excluded the possibility that the 14CO had been sequestered in the spleen.
Luomanmaki and Coburn (1969) also studied the shift of CO out of the blood during
15 hypoxia by administering CO into a rebreathing system and measuring the rate at which blood
COHb increased. They reasoned that if the partition of CO between vascular and
extravascular stores remained constant, the increase in blood COHb should be proportional to
the amount of CO administered. They found that COHb increased at a constant rate up to a
saturation of 50%. With additional CO, there was a decrease in the rate at which COHb
20 increased; this suggests that proportionally greater amounts of CO were entering the
extravascular stores. At 50% COHb (corresponding to an arterial PO2 of 90 mmHg), the
rate of COHb formation became nonlinear. Agostoni et al. (1980) presented a theoretical
model supporting these observations; they developed equations predicting that decreasing
venous PO2 causes CO to move out of the vascular compartment and into skeletal and heart
25 muscle. This increases the rate at which COMb is formed in the tissues.
The shift of CO out of the blood has been further demonstrated in studies (Horvath
et al., 1988) conducted on both men and women undergoing maximal aerobic capacity tests
at altitudes of 55, 1524, 2134, and 3058 m and CO concentrations of 0, 50, 100, and 150
ppm. Carbon monoxide at maximum work shifted into extravascular spaces and returned to
March 12, 1990 11-17 DRAFT - DO NOT QUOTE OR CITE
-------
the vascular space within five minutes after exercise stopped (Figure 11-4). This liberation of
CO was related to the concentration of COHb achieved as noted by the regression equation:
y = 0.0017 + 0.3047x (11-4)
5
where x is the COHb concentration at exhaustion.
11.1.7 Conclusions
While there are many studies comparing and contrasting inhaling CO with exposure to
10 altitude, there are relatively few reports on the effects of inhaling CO at altitude. There are
data to support the possibility that the effects of these two hypoxia episodes are at least
additive. These data were obtained at CO concentrations that are too high to have much
meaning for regulatory concerns. There also are data that indicate decrements in visual
sensitivity and flicker-fusion frequency in subjects exposed to CO (COHb = 5 to 10%) at
15 higher altitudes. These data, however, are somewhat controversial.
There are even fewer studies of the long-term effects of CO at high altitude. These
studies generally indicate few changes at CO concentrations below 100 ppm and altitudes
below 4572 m (15,000 ft). A provocative study by McDonagh et al. (1986) suggests that the
increase in ventricular capillarity seen with altitude exposure may be blocked by CO. The
20 fetus may be particularly sensitive to the effects of CO at altitude; this is especially true with
the high levels of CO associated with maternal smoking.
11.2 CARBON MONOXIDE INTERACTIONS WITH DRUGS
25 11.2.1 Introduction
There is little direct information on the possible enhancement of CO toxicity by
concomitant drug use or abuse; however, there are some data suggesting cause for concern.
There is evidence that interactions of drug effects with CO exposure can occur in both
directions, that is, CO toxicity may be enhanced by drug use and the toxic or other effects of
30 drugs may be altered by CO exposure. Nearly all the published data that are available on
CO combinations with drugs concern psychoactive drugs. Possible interactions of CO with
March 12, 1990 11-18 DRAFT - DO NOT QUOTE OR CITE
-------
Y = 0.0017 + 0.3047 X
Sy = 0.1337
x
1 1.4 1.8 2.2 2.6 3 3.4 3.8 4.2 4.6
MAX COHb
Figure 11-4. Higher concentrations of COHb observed at the end of a five-minute recovery
period after attainment of the subject's maximum aerobic capacity indicate that liberation of
CO from tissue stores is linearly related to COHb concentration present at exhaustion.
Source: Horvath et al. (1988).
March 12, 1990
11-19 DRAFT - DO NOT QUOTE OR CITE
-------
other classes of drugs (e.g., those likely to be used in patients with cardiovascular disease
who also are at risk for CO exposure) will be discussed elsewhere in this document (see
Chapter 12, Section 12.4). Another related area of concern that will be reviewed elsewhere is
interactions of CO with other toxicants (see Section 11.3).
5 The use and abuse of psychoactive drugs and alcohol is ubiquitous in society. Because
of CO's well-established effects on brain functioning, interactions between CO and
psychoactive drugs could be anticipated. Unfortunately, very little systematic research has
addressed this question. In addition, very little of the research that has been done has utilized
models for expected effects for treatment combinations. Thus, often it is not possible to
10 assess whether the combined effects of drugs and CO exposure are additive or differ from
additivity. It is important to recognize that even additive effects of combinations can be of
clinical significance, especially when the individual is unaware of the combined hazard.
11.2.2 Alcohol
15 The effects of combined CO exposure and alcohol (ethanol) administration have been the
most extensively studied interaction. The previous criteria document (U.S. Environmental
Protection Agency, 1979) reviewed two human studies which examined combinations of CO
and alcohol. A study from the Medical College of Wisconsin (1974) found no effects of
alcohol doses resulting in blood alcohol levels of about 0.05% and COHb levels in the general
20 range of 8 to 9%, either alone or in combination, on a number of psychomotor behavioral
tasks. The lack of sensitivity of these measures to alcohol doses known to affect performance
under many other conditions, as well as other problems in the study design, raises the
question of the adequacy of this study to detect interactive effects. Rockwell and Weir (1975)
studied the interaction of CO exposures resulting in nominal 0, 2, 8, and 12% COHb levels
25 with alcohol doses resulting in nominal 0.05% blood alcohol levels for effects on actual
driving performance in young, nonsmoking college students. Dose-related effects of CO for
perceptual narrowing and decreased eye movement were observed. In addition, effects were
observed on some measures by this dose of alcohol alone. An effect-addition model was used
to evaluate the alcohol-CO interaction. In combination, the effects of CO and alcohol were
30 often additive, and there was a supra-additive alcohol-CO interaction at 12% COHb levels.
Although the 1979 review highlighted the lack of an interaction effect except at high COHb
March 12, 1990 11-20 DRAFT - DO NOT QUOTE OR CITE
-------
concentrations, it should be noted that interaction effects for this study were defined as effects
greater than the sum of the effects of the treatments alone. Thus, this extensive study in
human subjects provides some evidence that driving-related performances already disrupted by
alcohol could be further compromised by CO exposure.
5 Because of a concern that persons exposed to CO may not be able to detect odors that
would indicate a fire or other hazardous condition, especially when consuming alcohol, Engen
(1986) conducted a carefully controlled study of combined CO-alcohol exposure in human
subject volunteers. The detection of a threshold concentration of the smoky odor of quaiacol
was evaluated using signal-detection analysis. The dose of alcohol given resulted in blood
10 alcohol levels of about 0.04 to 0.07% and CO exposure resulted in COHb values of 7.0 to
7.7%. In signal detection studies, d1 is a measure of detection threshold, with higher values
reflecting greater detection. The average d' for the four treatment conditions was as follows:
air only (1.95), CO only (2.34), alcohol + air (2.20), and alcohol + CO (1.64). Although
not statistically significant, there was a tendency for both alcohol and CO to improve odor
15 detection compared to air only. When alcohol and CO were combined, the odor detection
was significantly poorer than after either treatment alone, but it was not significantly poorer
than the air control. One of the features of signal detection analysis is that it allows the
separation of treatment effects on sensory sensitivity from effects on performance that would
influence the reporting of the signal. Thus, in this study it was found that these changes in
20 odor detection produced by alcohol and CO occurred in the absence of an effect of any of the
experimental treatments on reporting bias. Thus, one could conclude that the results of
combined alcohol and CO exposure was to eliminate the small improvement in odor
sensitivity produced by exposure to either treatment alone. The relevance and importance of
these small changes in odor detection are not readily apparent, especially because none of the
25 treatments were significantly different from air control; however, they do suggest that a CO-
alcohol interaction on odor thresholds may exist. An incidental finding of this study was that
alcohol did not alter COHb concentrations after exposure to CO; nor did CO exposure affect
blood alcohol levels produced by a fixed dose of oral alcohol.
There also have been a number of animal studies of combinations of alcohol and CO.
30 Although there is some evidence that alcohol metabolism can be reduced in rat liver in situ by
a COHb level of 20% (Toppling et al., 1981), an in vivo study in mice found no effects of
March 12, 1990 11-21 DRAFT - DO NOT QUOTE OR CITE
-------
CO exposure on alcohol metabolism (Kim and Carlson, 1983). Compared to levels in control
mice, either 1, 3, or 5 days of eight hour per day exposure to 500 ppm CO (COHb levels
averaged 28%) had no effect on blood alcohol levels when 2.2 g/kg of alcohol was
administered ip after 5.5 h of exposure on each of these days. On the other hand, Pankow
5 et al. (1974) provide some evidence that high doses of CO associated with COHb levels of
greater than 50% decreased blood alcohol levels in rats 30 min after a very large dose of
alcohol (4.8 g/kg). They also reported that this dose of alcohol significantly lowered COHb
levels associated with a very large subcutaneous dose of CO. These high-dose combinations
were also associated with additive effects on enzyme markers of hepatotoxicity, but no
10 interactions were observed when lower doses of CO were given.
In contrast to the inconsistent metabolic effects seen with combinations of CO and
alcohol, results of two behavioral studies in animals have both shown substantial effects.
Mitchell et al. (1978) studied the interaction of inhaled CO with two doses of alcohol
(0.6 and 1.2 g/kg) in rats using two behavioral measures. Sensorimotor incapacitation was
15 assessed by failure to remain on a rotating rod. An additional measure of motor effects was
the inability to withdraw the leg from a source of electric shock. The length of exposure to
about 2000 ppm CO before the animals failed in these performances was decreased in a dose-
dependent manner by alcohol. Carboxyhemoglobin determinations made at the time of
behavioral incapacitation was inversely related to alcohol dose. For example, nearly 50%
20 COHb levels were required to impair rotorod performance in the absence of alcohol, whereas
after 1.2 g/kg ethanol, less than 45% COHb levels produced the same effect. Unfortunately,
data was not provided on the effects of these doses of alcohol in the absence of CO exposure
to help determine the nature and magnitude of the interaction effects.
Knisely et al. (1989) recently reported a large interaction of CO exposure and alcohol
25 administration on operant behavior in animals. Mice that had been trained to lever press for
water reinforcement were tested with 1.1 g/kg alcohol and various doses of CO, alone and in
combination. An unusual feature of this study was that the both the alcohol and CO were
administered by ip injection. The authors provide evidence that this route of CO exposure
results in COHb formation and behavioral effects comparable to those seen after inhalation
30 exposure. The results of the study were evaluated by comparing the effects of the
combinations to those expected by summing the effects of each treatment alone. A dose of
March 12, 1990 11-22 DRAFT - DO NOT QUOTE OR CITE
-------
alcohol that had little effect on rates of lever pressing when given alone resulted in large rate-
decreasing effects when given in combination with doses of CO that also had no effects when
given alone. Supra-additive effects with alcohol were obtained by a dose of CO as low as
7.5 mL/kg, which when given alone was associated with COHb levels of about 20%.
5 Significant supra-additive effects also were obtained with higher doses of CO. Typically,
behavioral effects of CO alone were not seen under these test conditions until COHb
saturations greater than 40 to 50% were obtained (Knisely et al., 1989). Thus, alcohol about
doubled the acute toxicity of CO in this study.
10 11.2.3 Barbiturates
There has been some interest in the interaction of CO with barbiturates because
prolongation of barbiturate effects can reflect effects of toxicants on drug metabolism. In an
early evaluation of the functional significance of the binding of CO to cytochrome P-450,
Montgomery and Rubin (1971) examined the effects of CO exposure on the duration of action
15 of hexobarbital and the skeletal muscle relaxant zoxazolamine in rats. Both drugs are largely
deactivated by the hepatic mixed-function oxidase (MFO) system. Although CO was found to
dose-dependently enhance both hexobarbital sleeping time and zoxazolamine paralysis,
subsequent research indicated that this was probably not due to a specific inhibition of the
MFO system by CO, but rather a nonspecific effect of hypoxia, because even greater effects
20 could be produced at a similar level of arterial O2 produced by hypoxic hypoxia (Montgomery
and Rubin, 1973; Roth and Rubin, 1976). In support of the lack of effects of CO on drug
metabolism, Kim and Carlson (1983) found no effect of CO exposure on the plasma half-life
for either hexobarbital or zoxazolamine in mice. This would suggest that something other
than a metabolic interaction may be responsible for the enhancement of in vivo effects of these
25 drugs by CO.
There have been two studies of the interaction of CO and pentobarbital using operant
behavior in laboratory animals. McMillan and Miller (1974) found that exposure of pigeons
to 380 ppm CO, a concentration that had little effect on behavior when given alone, reduced
the response rate of schedule-controlled behavior, thereby increasing the effect of an
30 intermediate dose of pentobarbital. On the other hand, the disruptive effects of all doses of
pentobarbital on the temporal patterning of fixed-interval responding was enhanced markedly
March 12, 1990 11-23 DRAFT - DO NOT QUOTE OR CITE
-------
by 1030 ppm CO. This concentration of CO by itself did not alter response patterning, but
did lower overall rates of responding. In the study described more fully above in the section
on alcohol interactions (11.2.2), Knisely et al. (1989) found generally additive effects of ip
CO administration with the effects of pentobarbital in mice responding under a fixed-ratio
5 schedule. In that study the interaction of CO with pentobarbital was not as evident as the
interaction with alcohol, suggesting that general conclusions about CO interactions with
central nervous system depressant drugs may not be possible.
11.2.4 Other Psychoactive Drugs
10 Even more limited data are available on interactions of CO exposure with other
psychoactive drugs. In the study by Knisely et al. (1989), described above (Section 11.2.2),
of interactions of ip CO administration with psychoactive drugs on operant behavior of mice,
d-amphetamine, chlorpromazine, nicotine, diazepam, and morphine were studied in addition
to alcohol and pentobarbital. As with alcohol, a suggestion of greater than additive effects
15 were obtained from combinations of CO with both d-amphetamine and chlorpromazine;
however, in these cases the differences from additivity did not reach statistical significance.
Effects of CO in combination with nicotine, caffeine, and morphine were additive. McMillan
and Miller (1974) also found evidence for an interaction of CO and d-amphetamine on
operant behavior in pigeons. In this study, CO concentrations as low as 490 and 930 ppm
20 were able to modify the behavioral effects of d-amphetamine.
11.3 COMBINED EXPOSURE TO CARBON MONOXIDE AND OTHER
AIR POLLUTANTS AND ENVIRONMENTAL FACTORS
25 Exposure to a single air pollutant at ambient concentrations may have no harmful
biological effects. In real life, however, exposure occurs not only to a single agent but also
k> multiple agents, resulting in potential interactions between them. The result of the
interactions may be of an additive, synergistic, or antagonistic nature. Another possible
interaction is potentiation, a condition in which a pollutant that is noneffective at a given
30 exposure level may enhance the toxicity of another pollutant given simultaneously. Exposure
March 12, 1990 11-24 DRAFT - DO NOT QUOTE OR CITE
-------
to CO frequently occurs in the natural environment in combination with other combustion
products and air pollutants.
In this section both human and animal effects associated with combined exposure to CO
and other air pollutants and environmental factors are reviewed. Although a number of
5 studies in the literature have tested exposure to combined pollutants, fewer studies actually
have been designed to test specifically for interactions between CO and the other exposure
components. Therefore, this section emphasizes only those studies providing a combined
treatment group where pollutant exposure levels are reported. The COHb levels resulting
from CO exposure also are given if they were reported in the original manuscripts. The
10 toxicity data discussed stress the newer literature published since 1979 in order to update the
information reviewed in the previous Air Quality Criteria Document (U.S. Environmental
Protection Agency, 1979).
11.3.1 Exposure in Ambient Air
15 Photochemical air pollution usually is associated with two or more pollutants, consisting
mainly of CO, sulfur oxides, ozone, nitric oxides, peroxyacetyl nitrates, and organic
peroxides. The gaseous compounds that constitute tobacco smoke are CO, hydrogen cyanide,
and nitric oxide. As urban living, industrial employment, and cigarette smoking bring man
into direct contact with CO and other pollutants, it seems appropriate to determine if
20 combined exposure to these pollutants has detrimental health effects.
Several studies have been conducted to determine the effects resulting from combined
exposure to CO and other pollutants. The experimental details (e.g., concentrations and
duration of treatment) and the associated effects for each study are summarized in
Table 11-4. A brief discussion of the major findings follows.
25 Murphy (1964) observed an increase in blood COHb levels in mice and rats exposed to
CO+O3 for six hours as compared with mice exposed to CO alone. However, another study
(DeLucia et al., 1983) in adults exposed to CO+O3 during exercise, showed no synergistic
effects on blood COHb levels or pulmonary or cardiorespiratory thresholds. Similarly,
simultaneous exposure to CO-f O3+NO2 for two hours produced no consistent changes
30 (synergistic or additive) in pulmonary function indices and physiological parameters in young,
male subjects (Hackney et al., 1975a,b).
March 12, 1990 11-25 DRAFT - DO NOT QUOTE OR CITE
-------
TABLE 11-4. COMBINED EXPOSURE TO CARBON MONOXIDE AND OTHER POLLUTANTS
£3- Pollutant
Concentration
No./sex/species
Treatment
Observed Effects
Reference
K)
ON
to co
CO
CO
o,
CO
NO2
CO
Peroxy-
acetyl-
nitrate
(PAN)
300 ppm
0.75 ppm
280 ppm
3 ppm
100 ppm
0.3 ppm
30 ppm
0.25 ppm
0.30 ppm
50 ppm
0.27 ppm
9-10/-"/mouse and
rat
9/-/mice
24/M/human
24/F/human
(smokers and
nonsmokers)
8/M/human
10/M/human
(smokers)
10/F/human
(nonsmokers)
Exposed to 300 ppm CO alone or
0.75 ppm O3 + 300 ppm CO for
6 h; blood COHb levels were
determined
Exposed to 280 ppm CO alone or
3 ppm O3 + 280 ppm CO for 6 h
Exposed during exercise (four 1-h
rides on a bicycle to filtered
air only; 0.3 ppm O3 alone; 100 ppm
CO alone; or 0.3 ppm O3 + 100 ppm
CO); blood COHb, pulmonary
function, cardiorespiratory per-
formance, blood laclate levels,
and subjective symptoms were
examined
Exposed to O3 alone and in
combination with NO2 and CO for
2 h with secondary stress of
heat and intermittent light
exercise; subjective symptoms
were recorded; pulmonary
function, and physiological
studies were conducted
Subjects exposed to filtered air
only; 50 ppm CO alone; 0.27 ppm
PAN alone; or 50 ppm CO +
0.27 ppm PAN for 5 min at
two different temperatures,
25° and 35 °C (relative
humidity 30%); were tested for
maximal aerobic power, metabolic
temperature, and cardio-
respiratory responses
Simultaneous exposure produced
higher COHb levels (30.4% in
rats and 18.9% in mice) than
exposure to CO alone (25.8% in
rats and 14.8% in mice).
COHb levels were 24.3% in mice
exposed CO + O3 compared with
19.2% in mice exposed to CO alone.
Exposure to O3 + CO did not
elicit a synergistic effect.
Combined exposure did not
alter the threshold(s) of any
subject for appearance of adverse
effects due to O3 alone. Exposure
to CO alone caused a mean increase
in COHb (5.8%) levels compared
with exposure not involving CO.
No consistent synergistic or
additive effects were observed
in subjects exposed to 03 +
NO2 + CO in any parameter
measured, except for increases
in blood COHb (levels not
reported).
Maximal aerobic power was not
affected by any pollutant
conditions. The heart rate
was significantly (p<0.05)
greater in the CO group
compared with the filtered-air
group. Metabolic and thermo-
regulatory responses were not
different in the various
pollutant environments. Increases
in COHb levels of smokers during
the CO or CO + PAN exposures
were observed.
Murphy (1964)
Murphy (1964)
DeLucia et al. (1983)
Hackney et al. (1975 a,b)
Drinkwater et al. (1974);
Gliner et al. (1975);
Raven et al. (1974a,b)
-------
TABLE 11-4 (cont'd). COMBINED EXPOSURE TO CARBON MONOXIDE AND OTHER POLLUTANTS
Pollutant
Concentration
No./sex/species
Treatment
Observed Effects
Reference
tb
-J
I
O
O
O
d
I
§
O
h- CO
\o
\o
O NO
CO
NO
HCN
CO
NO,
CO
SO2
CO
so,
100 ppm
500 ppm
10 ppm
50 ppm
200 ppm
5 ppm
0.5 ppm
20 ppm
67.5 ppm
0.5 ppm
7.5 ppm
20 ppm
67.5 ppm
0.5 ppm
10 ppm
250 ppm
25 ppm
15/M/rat
(Long-Evans)
12/M/rabbit
(New Zealand white)
24/M/rat
24/F/rat
(Sprague-Dawley)
24/M/rat
24/F/rat
(Sprague-Dawley)
32-40/F/mouse
(CF-1)
Exposed to clean air only; 100 or
500 ppm CO alone; 10 or 50 ppm
NO alone; 100 ppm CO + 10 ppm
NO; or 500 ppm CO + 50 ppm NO
for 3 h; changes in
discrimination learning and
brain activity were measured
Exposed to 0.5 ppm HCN + 5 ppm
NO + 200 ppm CO for 2 weeks
Exposed to clean air only; 0.5 or
7.5 ppm NO2 alone; 20 or
67.5 ppm CO alone; 0.5 ppm
NO2 + 67.5 ppm CO; or 7.5 ppm
NO2 + 20 ppm CO continuously,
24 h/day, 7 days/week for
52 weeks; chronic toxicity was
assessed
Exposed to clean air only, 0.5 ppm
or 10 ppm SO2 alone; 20 or 67.5 ppm
CO alone; 0.5 ppm SO2 + 67.5 ppm
CO; or 10 ppm SO2 + 20 ppm CO
continuously, 24 h/day,
7 days/week for 52 weeks; chronic
toxicity was assessed
Exposed to filtered air only; 25 ppm
SO2 alone; or 25 ppm SO + 250 ppm
CO for 7 h/day during Days 6 through
15 gestation; teratogenic potential
was evaluated
No significant changes were observed in
COHb levels between any treatment groups.
Exposure to 100 ppm CO + 10 ppm NO
signinificantly (p<0.01) increased mean
metHb levels when compared to NO (10 ppm)
alone. Combined exposure caused significant
behavioral effects at both levels. Com-
bined exposure also affected early auditory-
evoked potential components (P10 and N^; the
effect was more pronounced at higher dose
level (500 ppm CO + 50 ppm NO) than at the
lower levels, indicating a dominant role
for NO.
Combined exposure to the three noxious
gases caused no morphological changes in
the lung, pulmonary, and coronary arteries,
or aorta.
No consistent changes in pulmonary function
indices were observed in any of the groups
exposed to the pollutants alone or in
combination with CO. Hematological and
biochemical changes were within the
normal range. Combined exposure to CO
+ NO2 did not increase the severity of
the histopalhological changes observed
in the lungs of rats exposed to NO2 alone.
Combined exposures caused no consistent
changes in pulmonary function
indices, hematology, or biochemical
or histological parameters.
Exposure to SO2 alone or SO2 +
CO caused no teratogenic
effects.
Groll-Knapp et al. (1988)
Hugod (1979)
Busey (1972)
Busey (1972)
Murray et al. (1978)
-------
TABLE 11-4 (cont'd). COMBINED EXPOSURE TO CARBON MONOXIDE AND OTHER POLLUTANTS
i
Pollutant
CO
S02
Concentration
250 ppm
70 ppm
No ./sex/species
20/F/rabbit
(New Zealand white)
Treatment Observed Effects
Exposed to filtered air only, 70 ppm No teratogenicity was
SO2 alone, or 70 ppm SO2 + 250 ppm observed.
CO for 7 h/day during Days
6 to 18 of gestation; teratogenic
potential was evaluated
Reference
Murray et al. (1978)
CO
SO2
CO
PbClBr
N>
00
3 mg/m3
6 mg/m3
0.5 mg/m3
67.5 ppm
0.6 ppm
100 ppm
1,000 ppm
3/-/human
24/M/rat
24/F/rat
(Sprague-Dawley)
5/M/rat
(Wistar)
Exposed to pure air for 5 min;
6 mg/m3 CO for 20 min; 6 mg/m3
CO + 0.5 mg/m3 SO2 for 5 min;
0.5 mg/m3 SO for 25 min; 6 mg/m3
CO + 0.5 mg/m3 for 25 min; or
3 mg/m3 CO +0.5 mg/m3 SO2 for
25 min; variations in ocular
sensitivity to light and color vision
were tested
Exposed to clean air alone, 67.5 ppm
CO alone; 0.6 ppm PbClBr alone; or
to 0.6 ppm + 67.5 ppm CO continuously
24 h/day, 7 days/week for
52 weeks; chronic toxicity was
assessed
Exposed to clean air only; 100 ppm
CO alone; 1,000 ppm CH2C12
only; or 100 ppm CO + 1,000 ppm
CH2CI2 for 3 h; mixed-
function oxidase activity and
blood COHb levels were examined
Inhalation of 6 mg/m3 CO +
0.5 mg/m3 SO2 for 5 min
or CO alone for 20 min
caused significant differences
in light and color sensitivity
when compared to controls.
No consistent changes in
pulmonary function, hematology,
or biochemistry were observed
between the treatment groups.
Combined exposure to CO +
PbClBr did not increase the
incidence of histological changes
observed in the kidneys of rats
exposed to PbClBr alone.
Combined exposure had an additive
effect. Blood COHb was significantly
(p<0.001) increased in rats exposed
to CO + CH2CL2 (14.6%) compared with
with CO alone (8.8%). Combined
exposure significantly (p< 0.005)
increased the ethoxycoumarin-
0-deethylase activity in the
kidneys. Treatment had no effect
on liver microsomal oxidation.
Mamatsashvili (1967)
Busey (1972)
Kuippa et al. (1981)
1,500 ppm
2,000 ppm
-/-/dog
(Cowenose Mongrel)
Exposed to 1,500 ppm CO for
25 min followed by a 2-h
exposure to 2,000 ppm CH2C12 in
air also containing 150 ppm CO;
effects on the cardiovascular
system were evaluated
Combined exposure of CO + CH2C12
had no effect on the phsiologic
response due to CO, instead CO
antagonized the responses due to
Adams (1975)
Information was not reported in the original manuscript.
-------
Combined exposure to CO and peroxyacetylnitrate (PAN) exerted no greater effect on
the work capacity of healthy men (young and middle-aged smokers and nonsmokers) than did
exposure to CO alone. Increases in blood COHb levels of smokers during the CO or
CO+PAN exposures were observed (Drinkwater et al., 1974; Raven et al., 1974a,b; Gliner
5 etal., 1975).
Groll-Knapp et al. (1988) reported that combined exposure of rats to CO+NO for
three hours caused a significant (p<0.01) increase in mean methemoglobin (metHb) levels
when compared with metHb levels in rats exposed to NO alone. No significant changes were
observed in blood COHb levels as compared with exposure to CO alone or to CO+NO.
10 Combined exposure also caused significant behavioral changes. Hugod (1979) reported that
combined exposure to CO+NO+HCN for two weeks produced no morphological changes in
the lungs, pulmonary arteries, coronary arteries, or aortas of rabbits.
In a one-year inhalation toxicity study, no adverse toxic effects were seen in groups of
rats exposed to relatively low levels of CO+NO2 or CO+SO2 as compared with rats exposed
15 to one of these pollutants alone (Busey, 1972). Murray et al. (1978) observed no teratogenic
effects in offspring of mice or rabbits exposed to CO+SO2 for 7 hours/day during gestation
Days 6 to 15 or 18, respectively.
Halogenated hydrocarbons, such as polybrominated and polychlorinated biphenyls, are
widely used as organic solvents. These chemicals are metabolized in the body to produce CO
20 which is readily bound to Hb. Therefore, any additional exposure to CO, producing higher
COHb levels, could possibly cause greater health effects. For example, up to 80% of inhaled
methylene chloride (CHjClz) will be metabolized to CO. Inhalation of 500 to 1000 ppm,
therefore, would result in COHb levels of over 14%. This elevation in COHb can not only
have a significant effect when combined with CO exposure, but the CO resulting from
25 metabolism generally requires a longer time to dissipate (Kurppa, 1984).
In one study, combined exposure to CO+CH2C12 for three hours had an additive effect
on blood COHb levels in rats (Kurppa et al., 1981). On the other hand, Adams (1975)
reported that combined exposure to CO+CH2C12 did not have an additive effect on the
physiologic response in the cardiovascular systems of dogs due to CO, instead, CO
30 antagonized the responses due to CH2C12.
March 12, 1990 11-29 DRAFT - DO NOT QUOTE OR CITE
-------
11.3.2 Exposure to Combustion Products
A common condition in an atmosphere produced by a fire is the presence of a rapidly
changing combination of potentially toxic gases (primarily, CO, CO2, and HCN), reduced O2
levels (hypoxic hypoxia), and high temperatures. Combined exposure to these gases occurs
5 during smoke inhalation under conditions of hypoxic hypoxia. In addition, both CO and CO2
are common products of carbon-containing materials; consequently, accidental exposure to
high levels of CO will rarely occur without simultaneous exposure to CO2. Exposure to CO
and HCN is of concern because both CO and HCN produce effects by influencing tissue O2
delivery. Increased COHb reduces O2-carrying capacity and may interfere with tissue O2
10 release, while HCN inhibits tissue respiration. Studies were conducted to determine the
lexicological interactions of the combustion products with and without reduced O2. (Also see
Chapter 10, Section 10.4.1.5 for more discussion on CO and HCN.)
Several studies have investigated the effects resulting from combined exposure to CO
and combustion products from fires. The experimental details and the associated effects for
15 each study are summarized in Table 11-5. The following is a brief discussion of the major
findings.
Rodkey and Collison (1979) reported a significant (p<0.02) decrease in mean survival
time in mice jointly exposed until death to CO+CO2 compared with mice exposed to CO
alone. In contrast, Crane (1985) observed no differences in the times-to-incapacitation or
20 times-to-death in rats exposed until death to various concentrations of CO+CO2. In a recent
study, Levin et al. (1987a) demonstrated a synergistic effect between CO and CO2 in rats
exposed to various concentrations of CO+CO2. Simultaneous exposure to nonlethal levels of
CO2 (1.7 to 17.3%) and to sublethal levels of CO (2500 to 4000 ppm) caused deaths in rats
both during and following (up to 24 h) a 30-min exposure. The rate of COHb formation was
25 1.5 times greater in rats exposed to CO+C02 than in rats exposed to CO alone.
Combined exposure to CO+HCN had an additive effect in rats as evidenced by
increases in mortality rate and changes in COHb levels. COHb levels decreased as the CO
level decreased and HCN level increased (Levin et al., 1987b). HCN had a depressive effect
on CO uptake and COHb formation, an effect that may explain the reason for the low COHb
30 levels (<50%) seen in some people who died in a fire (Levin et al., 1987b). In contrast,
March 12, 1990 11-30 DRAFT - DO NOT QUOTE OR CITE
-------
1 —
ST Combustion
H-> Product
TABLE 11-5. COMBINED EXPOSURE TO CARBON MONOXIDE AND COMBUSTION PRODUCTS
Concentration
No./Sex/Species
Treatment
Observed Effects
Reference
o
53
n
CO
6,000 ppm
2.1,2.3,
4.5, 5.4%
6/F/rat
(NMRI)
CO
CO,
5,000-14,000
ppm
4-13%
V-/rat
d
?> CO
H co2
I
d
O
1,470-6,000
ppm
1.7-17.3%
6/M/rat
(Fischer 344)
Exposed to 6,000 ppm CO alone;
6,000 ppm CO + 2.1 or 4.5% CO2;
or 6,000 ppm CO + 2.3 or 5.4%
CO2 until death; O2 concentrations
were either 14 or 21 %; mean
survival time (MST) and fatal blood
COHb level were measured
Exposed to concentrations ranging
from 5,000 to 14,000 ppm CO
alone or with CO2 concentrations
ranging from 4 to 13%
continuously until death;
synergistic effects (times-to-
incapacitation ([ti]) or the times-
to-death ([td]) were evaluated.
Exposed to 1,470 to 6,000 ppm
CO or 2,500 to 4,000 ppm CO +
1.7 to 17.3% CO2for30min;
toxicological interactions
(mortality and COHb formation)
were evaluated
The MST was significantly
(p<0.02) decreased in rats
exposed to 6,000 ppm CO + 2.1 %
CO2 (18.4 min) or 4.5% CO2
(16.8 min) compared with CO
alone (22.4 min) in the presence
of 21% O2. Exposure to 14%
O2 + 6,000 ppm CO decreased
the MST to 9.6 min; addition
of 2.3% or 5.4% CO2had no
further effect on MST. Combined
exposure (CO + CO^ had no
effect on fatal blood COHB.
No synergistic effects were
observed; no significant CO2
changes were observed in the
endpoints (t; or tj) for added
CO2 compared to endpoints for
CO alone.
Exposure to CO alone caused
deaths at levels of 4,600 to
6,000 ppm and at COHb levels of
> 83 %. Deaths were primarily
due to the high COHb, low
O2Hb, and hypoxia. Combined
exposure to >2500 ppm CO +
1.7 to 17.3% CO2 caused deaths
during exposure and the follow-
up period (24 h). No
mortality occurred in rats at
< 2,500 ppm CO alone regardless
of CO2 concentrations. The rate
of COHb formation was 1.5 times
greater in rats exposed to
2,500 ppm CO + 5.25% CO2 than
in rats exposed to 2,500 ppm CO
alone. The COHb equilibrium level
was the same
Rodkey and Collison (1979)
Crane (1985)
Levin et al. (1987a)
-------
TABLE 11-5 (cont'd). COMBINED EXPOSURE TO CARBON MONOXIDE AND COMBUSTION PRODUCTS
ET Combustion
i—> Product
rO
Concentration
No./Sex/Species
Treatment
Observed Effects
Reference
CO
HCN
1,242-4,600
ppm
43.2-126.4
ppm
6/M/rat
(Fischer 344)
HCN
200 ppm
0.5 ppm
12-24/M/rabbit
(albino)
0.63-0.66%
0.325-0.375%
4-9 mg/kg
1-6.35 mg/kg
10/M/mouse
(ICR)
Exposed to 4,600 ppm CO alone or
1,242 to 3,450 ppm CO + 43.2 to
126.4 ppm HCN for 30 min;
lethality and COHb formation were
measured as toxicological
endpoints
Exposed to 0.5 ppm HCN alone or
0.5 ppm HCN + 200 ppm CO for 1 or
4 weeks; morphological changes
in the lung, pulmonary arteries,
coronary arteries, or aorta were
evaluated
Mice were exposed to clean air or
to atmospheric concentrations of
0.63-0.66% CO for 3 min (pre-
treatment) and then injected
ip with 4-9 mg/kg KCN
(78%) for the combined exposure, but
was reached in 10 min in the
presence of CO2 and 20 min in the
absence of CO2. Combined exposure
increased the concentration of
COHb, caused severe acidosis, and
prolonged the recovery of acidosis
following cessation of exposure.
Exposure to CO2 alone produced no
mortality or incapacitation.
Combined exposure to CO + HCN
had an additive effect as
evidenced by increased mortality.
As the concentrations of HCN
increased, the animals died at
lower CO concentrations and
presented lower levels of COHb
at death. When rats were
exposed to 1,470 ppm CO alone
or 1,450 ppm CO + 100 ppm HCN,
the initial rate of COHb
formation was the same in the
presence or absence of HCN;
however, the final COHb level
was lower in the presence of
HCN, indicating a depressive
effect of HCN on CO uptake and
the low COHb formation.
Exposure to HCN alone or in
combination with CO produced
no morphological changes in
the lung, pulmonary arteries,
coronary arteries, or aorta.
The LDjo value was significantly
(p< 0.05) lower for KCN
(6.51 mg/kg) in CO-pretreated
mice than in air-pretreated
mice (7.90 mg/kg).
Levin et al. (1987b)
Hugod (1979)
Norris et al. (1986)
-------
£
8*
TABLE 11-5 (cont'd). COMBINED EXPOSURE TO CARBON MONOXIDE AND COMBUSTION PRODUCTS
Combustion
Product
Concentration
No./Sex/Species
Treatment
Observed Effects
Reference
Ui
LO
CO
KCN
1,000 ppm
2,500 ppm 7.5
rag/kg
20/M/mouse
(Swiss Webster)
In another experiment, mice were
pretreated with either saline
(0.1 mL/10, ip) or KCN (1 to 6.35
mg/kg, ip) and then were exposed
via inhalation to CO in the range
of 0.325 to 0.375% CO for 4
min; lethality and blood CO
and cyanide concentrations were
measured
Preexposed to 1 ,000 ppm CO for
4 h followed by a single ip
injection of 7.5 mg/kg KCN,
24 h later; effects on KCN-
induced letahlity were studied
Preexposed to 7.5 mg/kg KCN (ip)
24 h prior to exposure to
2,500 ppm CO for 2 h; effects
on KCN-induced lethality were
studied
Sublethal doses of KCN (3.5 to
6.35 mg/kg) produced a synergistic
effect in mortality: 40-100%
mortality in KCN pretreated mice
compared to 10-20% in saline
pretreated mice. There were
no differences in CO or cyanide
blood levels between these
treatment groups.
No alterations in lethality in
CO + KCN group as compared
with control + KCN group.
Pretreatment with KCN had no
significant effect on lethality
associated with subsequent
exposure to CO.
Winston and Roberts (1975)
I
o
o
§
CO
(under
con-
ditions
of
hypoxic
hypoxia)
O co
a (under
con-
O ditions of
^ hypoxic
O hypoxia)
CO: 6,000
ppm
O2: 14 or
21%
6/F/rat
(NMRI)
CO: 500,
1,000 or
2,500 ppm
O2: 7 or
10%
20/M/mouse
(Swiss Webster)
Exposed to 6,000 ppm CO until
death in the presence of either
14 or 21 % O2; MST and fatal blood
COHb levels were measured
Mice were preexposed to 500 or
1,000 ppm CO for 4 h and then
exposed to 2,500 ppm for 2 h,
24 h laterexposure to CO.
Preexposed to 500 or 1,000 ppm CO
for 4 h and then exposed to
7% 02for2h, 24 h
later
MST was significantly
(p<0.01) decreased in the
presence of low (14%) O2
(9.6 min) compared to that
of high (21%) O2 (22.4 min)
levels. A significantly
(p <0.01) higher level of COHb
was observed in rats treated with
14% O2 (89.4%) compared with those
treated with 21% O2 (83.4%).
Preexposure to CO caused a
significant (p<0.05) decrease
in lethality during subsequent
Preexposure to CO followed by
exposure to O2 had no effect
on lethality. Preexposure to
CO had no protective effect
against hypoxic hypoxia.
Rodkey and Collison (1979)
Winston and Roberts (1975)
-------
I
TABLE 11-5 (cont'd). COMBINED EXPOSURE TO CARBON MONOXIDE AND COMBUSTION PRODUCTS
Combustion
Product
Concentration
No./Sex/Species
Treatment
Observed Effects
Reference
Preexposed to 10% O2 for 4 h
and then exposed to 2,500 ppm CO
for 2 h, 24 h later
Preexposed to 1,000 ppm CO or 10%
O2 for 4 h and then exposed
to 2,500 ppm CO for up to 2 h,
24 h later
Preexposure to O2 followed by
exposure to CO significantly
(p<0.05) decreased lethality
compared to controls.
Preexposure to either CO or O2
had no significant effect on
O2-consumption level.
Alterations in CO lethality
were not associated with
alterations in COHb levels.
H
I
O
O
25
O
CO
(under
condi-
tions of
hypoxic
hypoxia)
CO: 500 or
1,000 ppm
O2: 6-21 % or
11.8-20.5%
*-/M/mouse
(Swiss Webster)
Exposed to reduced O2 (6-21%)
alone or reduced O2 (11.8 to
20.5%) + 500 ppm CO, or 20.2%
O2 + 1,000 ppm CO for 20 min;
animals were subjected to
behavioral tests that determined
reaction time and performance of
the animals in a mouse pole-jump
apparatus
Reaction time gradually
increased with a decrease in
O2to 10%. At <10% O2,
reaction time increased
dramatically and animal
performance decreased almost
immediately. At reduced O2
levels + CO, the decreases in
performance were even greater
than those seen in mice
exposed to reduced O2 levels
only. At 20% O2 (close to
ambient level) + 1,000 ppm CO,
performance was nearly
completely degraded.
Cagliostro and Islas (1982)
'Data not provided in the published manuscript.
n
-------
Hugod (1979) reported that exposure to HCN alone or to CO+HCN for one to four weeks
produced no morphological changes in the lung, pulmonary and coronary arteries, or aorta of
rabbits.
Combined exposures to CO+KCN have produced conflicting results. Norris et al.
5 (1986) reported that the LDjo values were significantly lower in mice pretreated with CO
prior to intraperitoneal injection of KCN. Sublethal doses of KCN produced a synergistic
effect on mortality. On the other hand, Winston and Roberts (1975) observed no alterations
in lethality in mice pretreated with CO and then treated with intraperitoneal injections of
KCN.
10 A number of studies examined the effects of CO administered under conditions of
hypoxic hypoxia. Rodkey and Collison (1979) observed a lower mean survival time and a
higher level of COHb in mice exposed to CO in the presence of low O2 (14%) than in those
exposed to an ambient O2 (21%) level. Winston and Roberts (1975) showed that preexposure
of mice to CO, followed by exposure to 7% O2 24 h later, had no effect on lethality as
15 compared with controls exposed to 7% O2 only. Thus, preexposure to CO had no protective
effect against hypoxic hypoxia. However, preexposure to 10% O2 caused a significant
decrease in lethality in mice exposed 24 h later to CO. Alterations in colethality were not
associated with alterations in COHb levels. In a behavioral study in mice, Cagliostro and
Islas (1982) showed that reaction times gradually increased with a decrease in O2 levels to
20 10%. At < 10% O2, reaction time increased dramatically. At reduced O2 levels and in the
presence of CO, the decreases in performance were even greater than those observed in mice
exposed to reduced O2 levels alone.
11.3.3 Exposure to Other Environmental Factors
25 11.3.3.1 Environmental Heat
Several of the studies (Drinkwater et al., 1974; Raven et al., 1974a,b; Gliner et al.,
1975) describing the effects of CO exposure alone and CO combined with peroxyacetyl nitrate
(PAN) on exercise performance in healthy adult men, reviewed in Sections 10.3.2 and
11.3.1, also evaluated the effects of heat stress. Subjects were exposed to 50 ppm CO and/or
30 0.27 ppm PAN in environmental exposure chamber conditions of 30% RH at 25 and 30°C.
In these studies, O2 uptake and exercise duration were assessed during both maximal and
March 12, 1990 11-35 DRAFT - DO NOT QUOTE OR CITE
-------
submaximal exercise. Heat stress was more effective in reducing maximal exercise
performance than exposure to the polluted environments. The combination of heat stress with
CO exposure was found to be important, however, in producing symptom complaints during
submaximal exercise at 35 °C that were not found at 25 °C. Further work in the same
5 laboratory (Bunnell and Horvath, 1989) also demonstrated that subjects experienced
significant levels of symptoms, particularly exertion symptoms, associated with elevated
COHb when exercising in the heat. These studies suggest, therefore, that heat stress may be
an important determinant of changes in exercise performance when combined with exposure
to CO.
10 Yang et al. (1988) studied the combined effects of high temperature and CO exposure in
laboratory mice and rats. They were exposed one hour per day for 23 consecutive days to
environmental chamber temperatures of 25 and 35 °C at CO concentrations ranging from 580
to 607 ppm. Carboxyhemoglobin levels after one hour of exposure ranged from 31.5 to
46.5%. The toxicity of CO to mice, based on the LCjo and survival time, was found to be
15 three times higher at 35 °C. High temperature also was found to enhance the effects of CO
on the function of oxidative phosphorylation of liver mitochondria in rats. Body temperature
regulation and heat tolerance also was affected by CO exposure. The authors speculate that
these effects of combined exposure to CO and high temperature are due to the production of
higher COHb, possibly due to hyperventilation.
20
11.3.3.2 Environmental Noise
Fechter et al. (1987; 1988) and Fechter (1988) speculated that the cochlea would be
particularly susceptible to injury when exposed to both CO and environmental noise. The
rationale for this potential effect was that CO exposure could impair cochlear oxygenation at a
25 time when auditory metabolism was likely to be enhanced by noise exposure. Using
laboratory rats exposed to high levels of CO (250 to 1200 ppm for 3.5 h) with and without
broad-band noise (105 dBA for 120 min), the authors were able to show that CO acts in a
dose-dependent manner to potentiate noise-induced auditory dysfunction. While CO or noise
alone did not have an effect, CO combined with noise produced a more severe loss of hair
30 cells at the basal end of the cochlea. Auditory threshold loss for the combined exposure was
evident at all frequencies tested but was greatest for high-frequency tones. A previous pilot
March 12, 1990 11-36 DRAFT - DO NOT QUOTE OR CITE
-------
study by Young et al. (1987) conducted at 1200 ppm CO also showed that combined exposure
to noise and CO produced high-frequency shifts of greater magnitude than those produced by
exposure to noise alone.
Results from the toxicologic studies in rats suggest that combined exposure to noise and
5 CO may be important in evaluating potential risk to exposed individuals. The CO levels used
in these studies, however, are much greater than those encountered in the typical ambient, or
even in the typical occupational environment. Thus, it is difficult to predict how relevant
these studies are to actual conditions of human exposure. An early epidemiologic study by
Lumio (1948) in operators of CO-fueled vehicles found significantly greater permanent
10 hearing loss than expected after controlling for possible confounding factors. More recently,
Sulkowski and Bojarski (1988) studied age-matched workers with similar length of duty
employed in foundry, cast iron, and cast steel positions of a mining devices factory where CO
and noise exposure varied. Careful otological and audiometric examinations were performed
on these workers. The group exposed to the combined effects of 95 dBA noise and a mean
15 concentration of 41 ppm CO did not experience any greater hearing loss than the groups
exposed only to noise (96 dBA) or CO (45 ppm). In fact, a permanent threshold shift was
significantly larger in workers exposed to noise alone than those exposed to the combined
influence of CO and noise. This study needs to be verified, however, at similar, relevant
exposure levels before any definitive conclusions can be made regarding the potential of
20 lower-level CO to potentiate noise-induced auditory loss in humans.
11.3.4 Summary
Much of the data concerning the combined effects of CO and other pollutants found in
the ambient air are based on animal experiments. Only a few human studies are available.
25 Early studies in healthy human subjects by Hackney et al. (1975a,b), Raven et al. (1974a,b),
Gliner et al. (1975), and Drinkwater et al. (1974) on common air pollutants such as NO2, O3,
or PAN and more recent work on CO + O3 by DeLucia et al. (1983) failed to show any
interaction from combined exposure.
In animal studies, no interaction was observed following combined exposure of CO and
30 pollutants such as HCN, NO2, SO2, or PbClBr (Hugod, 1979; Busey, 1972; Murray et al.,
1978). However, an additive effect was observed following combined exposure of high levels
March 12, 1990 11-37 DRAFT - DO NOT QUOTE OR CITE
-------
of CO + NO (Groll-Knapp et al., 1988), and a synergistic effect was observed after
combined exposure to CO and O3 (Murphy, 1964).
Toxicological interactions of combustion products, primarily CO, CO2) and HCN, from
indoor and outdoor fires, have shown a synergistic effect following CO + CO2 exposure
5 (Rodkey and Collison, 1979; Levin et al., 1987a) and an additive effect with CO + HCN
(Levin et al., 1987b). Additional studies are needed, however, to evaluate the effects of CO
under conditions of hypoxic hypoxia.
Finally, laboratory animal studies (Yang et al., 1988; Fechter et al., 1988; Young
et al., 1987) suggest that the combination of environmental factors such as heat stress and
10 noise may be important determinants of health effects occurring in combination with exposure
to CO. Of the effects described, the one potentially most relevant to typical human exposures
is a greater decrement in exercise performance seen when heat stress is combined with
50 ppm CO (Drinkwater et al., 1974; Raven et al., 1974a,b; Gliner et al., 1975).
15
11.4 ENVIRONMENTAL TOBACCO SMOKE
A common source of CO for the general population comes from tobacco smoke, along
with other primary sources arising from the environment. Exposure to tobacco smoke not
only affects the COHb level of the smoker himself, but under some circumstances, such as in
20 poorly ventilated spaces, tobacco smoke exposure also can affect nonsmokers. For example,
acute exposure (1 to 2 h) to smoke-polluted environments has been reported to cause an
incremental increase in nonsmokers' COHb of about 1% (Jarvis, 1987). In addition to CO,
other products inhaled by the affected individuals, such as NO2, HCN, nicotine, and potential
carcinogens contained in tobacco smoke, may produce subtle physiological and biochemical
25 effects in both the smoker and nonsmoker. Possible pathological changes due to the
interaction of CO and these other constituents of tobacco smoke that may occur in the lungs
and other tissues remain to be elucidated.
A detailed discussion of the possible health effects due to CO emitted from tobacco
smoke is beyond the scope of this document. Those interested in the problems related to
30 smoking tobacco (i.e., carcinogenesis and cardiovascular and pulmonary disease) should refer
to review documents specifically concerned with these matters (U.S. Department of Health
March 12, 1990 11-38 DRAFT - DO NOT QUOTE OR CITE
-------
and Human Services, 1983; U.S. Department of Health, Education and Welfare, 1979,
1972). In addition, a number of sources have reviewed the potential health effects of tobacco
smoke on nonsmokers (Fielding and Phenow, 1988; Hulka, 1988; Mohler, 1987; Surgeon
General of the United States, 1986; National Research Council, 1986).
5 Tobacco smoking has been found to result in higher COHb levels than exposure to
ambient concentrations of CO. The actual quantity of CO entering the lung depends on the
form in which tobacco is smoked, the pattern of smoking, and the depth of inhalation
(Robinson and Forbes, 1975). Very little CO (approximately 5%) is absorbed in the mouth
and larynx, therefore most of the CO available for binding to Hb must reach the alveoli in
10 order to raise the level of COHb present in the blood. The CO concentration in tobacco
smoke is approximately 4.5% (45,000 ppm). It has been estimated that a smoker may be
exposed to 400 to 500 ppm CO for the approximately 6 min that it takes to smoke a typical
cigarette, producing an average baseline COHb of 4%, with a typical range of 3 to 8%.
Heavy smokers can have COHb levels as high as 15%. In comparison, nonsmokers average
15 about 1% COHb in their blood. (See Chapter 8 for more information on CO exposure in the
population.) As a result of the higher baseline COHb levels, smokers are actually excreting
CO into the air rather than inhaling it from the ambient environment. Smokers may even
show an adaptive response to the elevated COHb levels, as evidenced by increased red cell
volumes or reduced plasma volumes (Smith and Landaw, 1978a,b). For these reasons, EPA
20 previously has not considered active smokers in determining the need for a margin of safety
for the CO NAAQS (Federal Register, 1980). This position was affirmed by the Clean Air
Scientific Advisory Committee of EPA's Science Advisory Board during review of the
previous CO criteria document (U.S. Environmental Protection Agency, 1979).
The effects of CO from tobacco smoke have been discussed in other sections of the
25 document. Human experimental studies suggested that acute effects of tobacco smoke on
maximal exercise performance are similar to those described for healthy subjects exposed to
CO. Prospective and retrospective epidemiological studies identified tobacco smoke as one of
the major factors in the development of cardiovascular disease. Tobacco smoke may
contribute to the development and/or aggravation of effects in exposed individuals through the
30 action of several independent or complementary mechanisms, one of which is the formation
of significant levels of COHb. Unfortunately, attempts to separate the CO effects of tobacco
March 12, 1990 11-39 DRAFT - DO NOT QUOTE OR CITE
-------
smoke from the potential effects of other substances present in the smoke have been
unsuccessful. (For a discussion of these studies, see Section 10.3).
In summary, although tobacco smoke is another source of CO for smokers as well as
nonsmokers, it is also a source of other chemicals with which environmental CO levels could
5 interact. Available data strongly suggest that acute and chronic CO exposure attributed to
tobacco smoke can affect the cardiopulmonary system, but the potential interaction of CO
with other products of tobacco smoke confounds the results. In addition, it is not clear if
incremental increases in COHb caused by environmental exposure would actually be additive
to chronically elevated COHb levels due to tobacco smoke, because some physiological
10 adaptation may take place. There is, therefore, a need for further research to describe these
relationships better.
11.5 REFERENCES
15
Adams, J. D. (1975) The effects of carbon monoxide and methylene chloride on the canine heart [Ph.D. thesis].
College Station, TX: Texas A&M University. Available from: University Microfilms, Ann Arbor, MI;
20 publication no. 75-25,077.
Agostoni, A.; Stabilini, R.; Viggiano, G.; Luzzana, M.; Samaja, M. (1980) Influence of capillary and tissue PQ
on carbon monoxide binding to myoglobin: a theoretical evaluation. Microvasc. Res. 20: 81-87.
25 Brewer, G. J.; Eaton, J. W.; Weil, J. V.; Grover, R. F. (1970) Studies of red cell glycolysis and interactions
with carbon monoxide, smoking, and altitude. In: Brewer, G. J., ed. Red cell metabolism and function:
proceedings of the first international conference on red cell metabolism and function; October 1969; Ann
Arbor, MI. New York, NY: Plenum Press; pp. 95-114. (Advances in experimental medicine and biology:
v. 6).
30
Brewer, G. J.; Sing, C. F.; Eaton, J. W.; Weil, J. V.; Brewer, L. F.; Grover, R. F. (1974) Effects on
hemoglobin oxygen affinity of smoking in residents of intermediate altitude. J. Lab. Clin. Med.
84: 191-205.
35 Bunnell, D. E.; Horvath, S. M. (1989) Interactive effects of heat, physical work, and CO exposure on
metabolism and cognitive task performance. Aviat. Space Environ. Med. 60: 428-432.
Busey, W. M. (1972) Chronic exposure of albino rats to certain airborne pollutants [unpublished material].
Vienna, VA: Hazleton Laboratories, Inc.
40
Cagliostro, D. E.; Islas, A. (1982) The effects of reduced oxygen and of carbon monoxide on performance in a
mouse pole-jump apparatus. J. Combust. Toxicol. 9: 187-193.
March 12, 1990 11-40 DRAFT - DO NOT QUOTE OR CITE
-------
Cobura, R. F. (1979) Mechanisms of carbon monoxide toxicity. Prev. Med. 8: 310-322. Coburn, R. F.; Forster,
R. E.; Kane, P. B. (1965) Considerations of the physiological variables that determine the blood
carboxyhemoglobin concentration in man. J. Clin. Invest. 44: 1899-1910.
5 Collier, C. R.; Goldsmith, J. R. (1983) Interactions of carbon monoxide and hemoglobin at high altitude. Atmos.
Environ. 17: 723-728.
Cooper, R. L.; Dooley, B. S.; McGrath, J. J.; McFaul, S. J.; Kopetzky, M. T. (1985) Heart weights and
electrocardiograms in rats breathing carbon monoxide at altitude. Fed. Proc. Fed. Am. Soc. Exp. Biol.
10 44: 1048.
Crane, C. R. (1985) Are the combined toxicities of CO and CO2 synergistic? J. Fire Sci. 3: 143-144.
DeLucia, A. J.; Whitaker, J. H.; Bryant, L. R. (1983) Effects of combined exposure to ozone and carbon
15 monoxide (CO) in humans. In: Lee, S. D.; Mustafa, M. G.; Mehlman, M. A., eds. International
symposium on the biomedical effects of ozone and related photochemical oxidants; March; Pinehurst,
NC. Princeton, NJ: Princeton Scientific Publishers, Inc.; pp. 145-159. (Advances in modern
environmental toxicology: v. 5).
20 Denniston, J. C.; Pettyjohn, F. S.; Boyter, J. K.; Kelliher, J. C.; Hiott, B. F.; Piper, C. F. (1978) The
interaction of carbon monoxide and altitude on aviator performance: pathophysiology of exposure to
carbon monoxide. Fort Rucker, AL: U. S. Army Aeromedical Research Laboratory; report no. 78-7.
Available from: NTIS, Springfield, VA; AD-A055 212.
25 Drinkwater, B. L.; Raven, P. B.; Horvath, S. M.; Gliner, J. A.; Ruhling, R. O.; Bolduan, N. W.; Taguchi, S.
(1974) Air pollution, exercise, and heat stress. Arch. Environ. Health 28: 177-181.
Engen, T. (1986) The combined effect of carbon monoxide and alcohol on odor sensitivity. Environ. Int. 12:
207-210.
30
Fechter, L. D. (1988) Interactions between noise exposure and chemical asphyxiants: evidence for potentiation of
noise induced hearing loss. In: Berglund, B.; Berglund, U.; Karlsson, J.; Lindvall, T., eds. Noise as a
public health problem - volume 3, performance behaviour, animal, combined agents, and community
responses; proceedings of the 5th international congress on noise as a public health problem; August;
35 Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building Research; pp. 117-122.
Fechter, L. D.; Thorne, P. R.; Nuttall, A. L. (1987) Effects of carbon monoxide on cochlear electrophysiology
and blood flow. Hear. Res. 27: 37-45.
40 Fechter, L. D.; Young, J. S.; Carlisle, L. (1988) Potentiation of noise induced threshold shifts and hair cell loss
by carbon monoxide. Hear. Res. 34: 39-47.
Federal Register. (1980) Carbon monoxide; proposed revisions to the national ambient air quality standards:
proposed rule. F. R. (August 18) 45: 55066-55084.
45
Fielding, J. E.; Phenow, K. J. (1988) Health effects of involuntary smoking. N. Engl. J. Med. 319: 1452-1460.
Forbes, W. H.; Sargent, F.; Roughton, F. J. W. (1945) The rate of carbon monoxide uptake by normal men.
Am. J. Physiol. 143: 594-608.
50
Gliner, J. A.; Raven, P. B.; Horvath, S. M.; Drinkwater, B. L.; Sutton, J. C. (1975) Man's physiologic
response to long-term work during thermal and pollutant stress. J. Appl. Physiol. 39: 628-632.
March 12, 1990 11-41 DRAFT - DO NOT QUOTE OR CITE
-------
Groll-Knapp, E.; Haider, M.; Kienzl, K.; Handler, A.; Trimmel, M. (1988) Changes in discrimination learning
and brain activity (ERP's) due to combined exposure to NO and CO in rats. Toxicology 49: 441-447.
Haagenson, P. L. (1979) Meteorological and climatological factors affecting Denver air quality. Atmos. Environ.
5 13: 79-85.
Hackney, J. D.; Linn, W. S.; Mohler, J. G.; Pedersen, E. E.; Breisacher, P.; Russo, A. (1975a) Experimental
studies on human health effects of air pollutants: II. four-hour exposure to ozone alone and in
combination with other pollutant gases. Arch. Environ. Health 30: 379-384.
10
Hackney, J. D.; Linn, W. S.; Law, D. C.; Karuza, S. K.; Greenberg, H.; Buckley, R. D.; Pedersen, E. E.
(1975b) Experimental studies on human health effects of air pollutants: III. two-hour exposure to ozone
alone and in combination with other pollutant gases. Arch. Environ. Health 30: 385-390.
15 Halperin, M. H.; McFarland, R. A.; Niven, J. L; Roughton, F. J. W. (1959) The time course of the effects of
carbon monoxide on visual thresholds. J. Physiol. (London) 146: 583-593.
Horvath, S. M.; Bedi, J. F. (1989) Alteration in carboxyhemoglobin concentrations during exposure to 9 ppm
carbon monoxide for 8 hours at sea level and 2134 m altitude in a hypobaric chamber. JAPCA
20 39: 1323-1327.
Horvath, S. M.; Raven, P. B.; Dahms, T. E.; Gray, D. J. (1975) Maximal aerobic capacity at different levels of
carboxyhemoglobin. J. Appl. Physiol. 38: 300-303.
25 Horvath, S. M.; Bedi, J. F.; Wagner, J. A.; Agnew, J. (1988) Maximal aerobic capacity at several ambient
concentrations of CO at several altitudes. J. Appl. Physiol. 65: 2696-2708.
Hugod, C. (1979) Effect of exposure to 0.5 ppm hydrogen cyanide singly or combined with 200 ppm carbon
monoxide and/or 5 ppm nitric oxide on coronary arteries, aorta, pulmonary artery, and lungs in the
30 rabbit. Int. Arch. Occup. Environ. Health 44: 13-23.
Hulka, B. S. (1988) The health consequences of environmental tobacco smoke. Environ. Technol. Lett.
9: 531-538.
35 James, W. E.; Tucker, C. E.; Grover, R. F. (1979) Cardiac function in goats exposed to carbon monoxide. J.
Appl. Physiol.: Respir. Environ. Exercise Physiol. 47: 429-434.
Jarvis, M. J. (1987) Uptake of environmental tobacco smoke. In: O'Neill, I. K.; Brunnemann, K. D.; Dodet, B.;
Hoffmann, D., eds. Environmental carcinogens: methods of analysis and exposure measurement: volume
40 9 - passive smoking. Lyon, France: International Agency for Research on Cancer; pp. 43-58. (IARC
scientific publications no. 81).
Kim, Y. C.; Carlson, G. P. (1983) Effect of carbon monoxide inhalation exposure in mice on drug metabolism
in vivo. Toxicol. Lett. 19: 7-13.
45
Kirkpatrick, L. W.; Reeser, W. K., Jr. (1976) The air pollution carrying capacities of selected Colorado
mountain valley ski communities. J. Air Pollut. Control Assoc. 26: 992-994.
Knisely, J. S.; Rees, D. C.; Balster, R. L. (1989) Effects of carbon monoxide in combination with behaviorally
50 active drugs on fixed-ratio performance in the mouse. Neurotoxicol. Teratol. 11: 447-452.
Kurppa, K. (1984) Carbon monoxide. In: Aitio, A.; Riihimaki, V.; Vainio, H., eds. Biological monitoring and
surveillance of workers exposed to chemicals. New York, NY: Hemisphere Publishing Corp.;
pp. 159-164.
March 12, 1990 11-42 DRAFT - DO NOT QUOTE OR CITE
-------
Kurppa, K.; Kivisto, H.; Vainio, H. (1981) Dichloromethane and carbon monoxide inhalation:
carboxyhemoglobin addition, and drug metabolizing enzymes in rat. Int. Arch. Occup. Environ. Health
49: 83-87.
5 Levin, B. C.; Paabo, M.; Gurman, J. L.; Harris, S. E.; Braun, E. (1987a) Toxicological interactions between
carbon monoxide and carbon dioxide. Toxicology 47: 135-164.
Levin, B. C.; Paabo, M.; Gurman, J. L.; Harris, S. E. (1987b) Effects of exposure to single or multiple
combinations of the predominant toxic gases and low oxygen atmospheres produced in fires. Fundam.
10 Appl. Toxicol. 9: 236-250.
Lilienthal, J. L., Jr.; Fugitt, C. H. (1946) The effect of low concentrations of carboxyhemoglobin on the
"altitude tolerance" of man. Am. J. Physiol. 145: 359-364.
15 Lindsey, T. K. (1989) [Letter to Dr. James McGrath concerning U.S. population living at 5,000 feet or more].
Forestville, MD: July 4.
Lumio, J. S. (1948) Hearing deficiencies caused by carbon monoxide (generator gas). Helsinki, Finland:
Oto-Laryngological Clinic of the University.
20
Luomanmaki, K.; Cobum, R. F. (1969) Effects of metabolism and distribution of carbon monoxide on blood and
body stores. Am. J. Physiol. 217: 354-363.
Mamatsashvili, M. I. (1967) Determination of the reflex effect of sulfur dioxide and carbon monoxide by
25 adaptometry and by tests of color vision. Gig. Sanit. 32: 226-230.
McDonagh, P. F.; Reynolds, J. M.; McGrath, J. J. (1986) Chronic altitude plus carbon monoxide exposure
causes left ventricular hypertrophy but an attenuation of coronary capillarity. Presented at: 70th annual
meeting of the Federation of American Societies for Experimental Biology; April; St. Louis, MO. Fed.
30 Proc. Fed. Am. Soc. Exp. Biol. 45: 883.
McFarland, R. A. (1970) The effects of exposure to small quantities of carbon monoxide on vision. In: Coburn,
R. F., ed. Biological effects of carbon monoxide. Ann. N. Y. Acad. Sci. 174: 301-312.
35 McFarland, R. A.; Roughton, F. J. W.; Halperin, M. H.; Niven, J. I. (1944) The effects of carbon monoxide
and altitude on visual thresholds. J. Aviat. Med. 15: 381-394.
McGrath, J. J. (1988) Body and organ weights of rats exposed to carbon monoxide at high altitude. J. Toxicol.
Environ. Health 23: 303-310.
40
McGrath, J. J. (1989) Cardiovascular effects of chronic carbon monoxide and high-altitude exposure. Cambridge,
MA: Health Effects Institute; research report number 27.
McMillan, D. E.; Miller, A. T., Jr. (1974) Interactions between carbon monoxide and d-amphetamine or
45 pentobarbital on schedule-controlled behavior. Environ. Res. 8: 53-63.
Medical College of Wisconsin. (1974) Exposure of humans to carbon monoxide combined with ingestion of ethyl
alcohol and the comparison of human performance when exposed for varying periods of time to carbon
monoxide. Milwaukee, WI: Medical College of Wisconsin; report no. MCOW-ENVM-CO-74-2.
50 Available from: NTIS, Springfield, VA; PB-242099.
Miranda, J. M.; Konopinski, V. J.; Larsen, R. I. (1967) Carbon monoxide control in a high highway tunnel.
Arch. Environ. Health 15: 16-25.
March 12, 1990 11-43 DRAFT - DO NOT QUOTE OR CITE
-------
Mitchell, D. S.; Packham, S. C.; Fitzgerald, W. E. (1978) Effects of ethanol and carbon monoxide on two
measures of behavioral incapacitation of rats. Proc. West. Pharmacol. Soc. 21: 427-431.
Mitchell, R. S.; Judson, F. N.; Moulding, T. S.; Weiser, P.; Brock, L. L.; Kelble, D. L.; Pollard, J. (1979)
5 Health effects of urban air pollution: special consideration of areas at 1,500 m and above. JAMA J. Am.
Med. Assoc. 242: 1163-1168.
Mohler, S. E. (1987) Passive smoking: a danger to children's health. J. Pediatr. Health Care 1: 298-304.
10 Montgomery, M. R.; Rubin, R. J. (1971) The effect of carbon monoxide inhalation on in vivo drug metabolism
in the rat. J. Pharmacol. Exp. Ther. 179: 465-473.
Montgomery, M. R.; Rubin, R". J. (1973) Oxygenation during inhibition of drug metabolism by carbon monoxide
or hypoxic hypoxia. J. Appl. Physiol. 35: 505-509.
15
Moore, L. G.; Rounds, S. S.; Jahnigen, D.; Grover, R. F.; Reeves, J. T. (1982) Infant birth weight is related to
maternal arterial oxygenation at high altitude. J. Appl. Physiol.: Respir. Environ. Exercise Physiol.
52: 695-699.
20 Murphy, S. D. (1964) A review of effects on animals of exposure to auto exhaust and some of its components. J.
Air Pollut. Control Assoc. 14: 303-308.
Murray, F. J.; Schwetz, B. A.; Crawford, A. A.; Henck, J. W.; Staples, R. E. (1978) Teratogenic potential of
sulfur dioxide and carbon monoxide in mice and rabbits. In: Mahlum, D. D.; Sikov, M. R.; Hackett, P.
25 L.; Andrew, F. D., eds. Developmental toxicology of energy-related pollutants: proceedings of the
seventeenth annual Hanford biology symposium; October 1977; Richland, WA. Oak Ridge, TN: U. S.
Department of Energy, Technical Information Center; pp. 469-478. Available from: NTIS, Springfield,
VA; CONF-771017.
30 National Research Council. (1977) Carbon monoxide. Washington, DC: National Academy of Sciences. (Medical
and biologic effects of environmental pollutants).
National Research Council. (1986) Environmental tobacco smoke: measuring exposures and assessing health
effects. Washington, DC: National Academy Press.
35
Norris, J. C.; Moore, S. J.; Hume, A. S. (1986) Synergistic lethality induced by the combination of carbon
monoxide and cyanide. Toxicology 40: 121-129.
Pankow, D.; Ponsold, W.; Fritz, H. (1974) Combined effects of carbon monoxide and ethanol on the activities of
40 leucine aminopeptidase and glutamic-pyruvic transaminase in the plasma of rats. Arch. Toxicol.
32: 331-340.
Parving, H.-H. (1972) The effect of hypoxia and carbon monoxide exposure on plasma volume and capillary
permeability to albumin. Scand. J. Clin. Lab. Invest. 30: 49-56.
45
Peterson, J. E.; Stewart, R. D. (1975) Predicting the carboxyhemoglobin levels resulting from carbon monoxide
exposures. J. Appl. Physiol. 39: 633-4538.
Pettyjohn, F. S.; McNeil, R. J.; Akers, L. A.; Faber, J. M. (1977) Use of inspiratory minute volumes in
50 evaluation of rotary and fixed wing pilot workload. Fort Rucker, AL: U. S. Army Aeromedical Research
Laboratory; report no. 77-9.
Pitts, G. C.; Pace, N. (1947) The effect of blood carboxyhemoglobin concentration on hypoxia tolerance. Am. J.
Physiol. 148: 139-150.
March 12, 1990 11-44 DRAFT - DO NOT QUOTE OR CITE
-------
Raven, P. B.; Drinkwater, B. L.; Horvath, S. M.; Ruhling, R. O.; Gliner, J. A.; Sutton, J. C.; Bolduan, N. W.
(1974a) Age, smoking habits, heat stress, and their interactive effects with carbon monoxide and
peroxyacetylnitrate on man's aerobic power. Int. J. Biometeorol. 18: 222-232.
5 Raven, P. B.; Drinkwater, B. L.; Ruhling, R. O.; Bolduan, N.; Taguchi, S.; Gliner, J.; Horvath, S. M. (1974b)
Effect of carbon monoxide and peroxyacetyl nitrate on man's maximal aerobic capacity. J. Appl. Physiol.
36: 288-293.
Robinson, J. C.; Forbes, W. F. (1975) The role of carbon monoxide in cigarette smoking: I. Carbon monoxide
10 yield from cigarettes. Arch. Environ. Health 30: 425-434.
Rockwell, T. J.; Weir, F. W. (1975) The interactive effects of carbon monoxide and alcohol on driving skills.
Columbus, OH: The Ohio State University Research Foundation; CRC-APRAC project CAPM-9-69.
Available from: NTIS, Springfield, VA; PB-242266.
15
Rodkey, F. L.; Collison, H. A. (1979) Effects of oxygen and carbon dioxide on carbon monoxide toxicity. J.
Combust. Toxicol. 6: 208-212.
Roth, R. A., Jr.; Rubin, R. J. (1976) Role of blood flow in carbon monoxide- and hypoxic hypoxia-induced
20 alterations in hexobarbital metabolism in rats. Drug Metab. Dispos. 4: 460-467.
Smith, J. R.; Landaw, S. A. (1978a) Smokers' polycythemia. N. Engl. J. Med. 298: 6-10.
Smith, J. R.; Landaw, S. A. (1978b) Smokers' polycythemia [letter to the editor]. N. Engl. J. Med. 298: 973.
25
Sulkowski, W. J.; Bojarski, K. (1988) Hearing loss due to combined exposure to noise and carbon monoxide - a
field study. In: Berglund, B.; Berglund, U.; Karlsson, J.; Lindvall, T., eds. Noise as a public health
problem - volume 1, abstract guide: proceedings of the 5th international congress on noise as a public
health problem; August; Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building
30 Research; p. 179.
Surgeon General of the United States. (1986) The health consequences of involuntary smoking: a report of the
Surgeon General. Rockville, MD: U. S. Department of Health and Human Services, Office on Smoking
and Health.
35
Toppling, D. L.; Fishlok, R. C.; Trimble, R. P.; Storer, G. B.; Snoswell, A. M. (1981) Carboxyhemoglobin
inhibits the metabolism of ethanol by perfused rat liver. Biochem. Int. 3: 157-163.
U. S. Department of Health, Education, and Welfare. (1972) The health consequences of smoking: a report of the
40 Surgeon General. Washington, DC: Public Health Service; report no. DHEW(HMS) 72-7516.
U. S. Department of Health, Education, and Welfare. (1979) Smoking and health: a report of the Surgeon
General. Washington, DC: Public Health Service; publication no. DHEW (PHS) 79-50066.
45 U. S. Department of Health and Human Services. (1983) The health consequences of smoking: cardiovascular
disease - a report of the Surgeon General. Washington, DC: Public Health Service; publication no.
DHHS (PHS) 84-50204.
U. S. Environmental Protection Agency. (1979) Air quality criteria for carbon monoxide. Research Triangle
50 Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and Assessment
Office; EPA report no. EPA-600/8-79-022. Available from: NTIS, Springfield, VA; PB81-244840.
Vollmer, E. P.; King, B. G.; Birren, J. E.; Fisher, M. B. (1946) The effects of carbon monoxide on three types
of performance, at simulated altitudes of 10,000 and 15,000 feet. J. Exp. Psychol. 36: 244-251.
March 12, 1990 11-45 DRAFT - DO NOT QUOTE OR CITE
-------
Wagner, J. A.; Horvath, S. M.; Andrew, G. M.; Cottle, W. H.; Bedi, J. F. (1978) Hypoxia, smoking history,
and exercise. Aviat. Space Environ. Med. 49: 785-791.
Weiser, P. C.; Morrill, C. G.; Dickey, D. W.; Kurt, T. L.; Cropp, G. J. A. (1978) Effects of low-level carbon
5 monoxide exposure on the adaptation of healthy young men to aerobic work at an altitude of 1,610
meters. In: Folinsbee, L. J.; Wagner, J. A.; Borgia, J. F.; Drinkwater, B. L.; Gliner, J. A.; Bedi, J. F.,
eds. Environmental stress: individual human adaptations. New York, NY: Academic Press, Inc.;
pp. 101-110.
10 Winston, J. M.; Roberts, R. I. (1975) Influence of carbon monoxide, hypoxic hypoxia or potassium cyanide
pretreatment on acute carbon monoxide and hypoxic hypoxia lethality. J. Phannacol. Exp. Ther. 193:
713-719.
Yang, L.; Zhang, W.; He, H.; Zhang, G. (1988) Experimental studies on combined effects of high temperature
15 and carbon monoxide. J. Tongji Med. Univ. 8: 60-65.
Young, J. S.; Upchurch, M. B.; Kaufman, M. J.; Fechter, L. D. (1987) Carbon monoxide exposure potentiates
high-frequency auditory threshold shifts induced by noise. Hear. Res. 26: 37-43.
March 12, 1990 11-46 DRAFT - DO NOT QUOTE OR CITE
-------
12. EVALUATION OF SUBPOPULATIONS POTENTIALLY
AT RISK TO CARBON MONOXIDE EXPOSURE
12.1 INTRODUCTION
Most of the information on the human health effects of carbon monoxide discussed in
Chapter 10 of this document has concentrated on two carefully defined population groups -
young healthy, predominantly male adults and patients with diagnosed coronary artery
10 disease. On the basis of the known effects described, patients with reproducible exercise-
induced angina appear to be best established as a sensitive group within the general population
that is at increased risk for experiencing health effects (i.e., decreased exercise duration due
to exacerbation of cardiovascular symptoms) of concern at ambient or near-ambient CO-
exposure concentrations that result in COHb levels of <5%. A smaller sensitive group of
15 healthy individuals experience decreased exercise duration at similar levels of CO exposure,
but only during short-term maximal exercise. Decrements in exercise duration in the healthy
population, therefore, would be mainly of concern to competing athletes rather than for
nonathletic people carrying out the common activities of daily life.
It is known, however, from both theoretical work and from experimental research in
20 laboratory animals that certain other groups in the population are at potential risk to exposure
from CO. The purpose of this chapter is to explore the potential effects of CO in population
groups that have not been studied adequately, but which could be expected to be susceptible
to CO because of underlying physiological status either due to gender differences, aging,
preexisting disease, or because of the use of medications or alterations in their environment.
25 These probable risk groups include (1) fetuses and young infants; (2) pregnant women; (3) the
elderly, especially those with compromised cardiopulmonary or cerebrovascular functions;
(4) individuals with obstructed coronary arteries, but not yet manifesting overt
symptomatology of coronary artery disease; (5) individuals with congestive heart failure;
(6) individuals with peripheral vascular or cerebrovascular disease; (7) individuals with
30 hematological diseases (e.g., anemia) that affect oxygen-carrying capacity or transport in the
blood; (8) individuals with genetically unusual forms of hemoglobin associated with reduced
March 5, 1990 12-1 DRAFT - DO NOT QUOTE OR CITE
-------
oxygen-carrying capacity; (9) individuals with chronic obstructive lung diseases;
(10) individuals using medicinal or recreational drugs having CNS depressant properties;
(11) individuals exposed to other pollutants (e.g., methylene chloride) that increase
endogenous formation of CO; and (12) individuals who have not been adapted to high altitude
5 and are exposed to a combination of high altitude and CO.
Little empirical evidence currently is available by which to specify health effects
associated with ambient or near-ambient CO exposures in these probable risk groups. Where
the previous chapters dealt with documented evidence of CO exposure through controlled or
natural laboratory investigations, this chapter will be more speculative. An effort will be
10 made to determine the anticipated effects of CO in special subpopulations that form a
significant proportion of the population at large.
12.2 AGE AND GENDER AS RISK FACTORS
15 The fetus and newborn infant are theoretically susceptible to CO exposure for several
reasons. Fetal circulation is likely to have a higher COHb level than the maternal circulation
due to differences in uptake and elimination of CO from fetal hemoglobin. Because the fetus
also has a lower oxygen tension in the blood than adults, any further drop in fetal oxygen
tension due to the presence of COHb could have a potentially serious effect. The newborn
20 infant with a comparatively high rate of oxygen consumption and lower hemoglobin blood
oxygen-transport capacity than most adults also would be potentially susceptible to the
hypoxic effects of increased COHb. Newer data from laboratory animal studies on the
developmental toxicity of CO suggest that prolonged exposure to high levels (> 100 ppm) of
CO during gestation may produce a reduction in birthweight, cardiomegaly, and delayed
25 behavioral development (see Chapter 10, Section 10.5). Human data are scant and more
difficult to evaluate, but further research is warranted. Additional studies, therefore, are
needed in order to determine if chronic exposure to CO, particularly at low, near-ambient
levels, can compromise the already marginal conditions existing in the fetus and newborn
infant.
30 The effects of CO on maternal-fetal relationships are not understood well. In addition to
fetuses and newborn infants, pregnant women also represent a susceptible group because
March 5, 1990 12-2 DRAFT - DO NOT QUOTE OR CITE
-------
pregnancy is associated with increased alveolar ventilation and an increased rate of oxygen
consumption that serves to increase the rate of CO uptake from inspired air. A perhaps more
important factor is that pregnant women experience hemodilution due to the disproportionate
increase in plasma volume as compared to erythrocyte volume. This group, therefore, should
5 be studied to evaluate the effects of CO exposure and elevated COHb levels.
Seventy percent of the population of the United States survive to 65 years of age and
30% reach 80 years of age or more (Brody et al., 1987). In 1982 about 40% were 75 years
of age or older, corresponding to about 10.7 million subjects. The percentage of the
population reaching 75 years of age or older is expected to increase to 49% by the year 2000
10 and to 56% by the year 2080, making it one of the fastest growing segments of the
population. Thus, the aging population represents a potentially large subgroup that may be at
risk to CO exposure.
Changes in metabolic capability with age may make the aging population particularly
susceptible to CO. Maximal oxygen uptake declines steadily with age at a rate of about
15 0.9 mL/kg/min/year on the average. However, the rate is only 0.65 in an active person and
about 1.3 in an inactive person. Since inactivity is the most prevalent condition, especially in
the elderly, the maximal oxygen uptake at age 65 will be about 16 to 21 mL/kg/min or 1.2-
1.6 L/min. At age 75 the maximal oxygen uptake will be about 10 to 15 mL/kg/min or 0.75-
1.1 L/min. Note that these values refer to a healthy but inactive male person.
20 A person needs about 1 L/min or 10 mL/kg/min in maximal oxygen uptake to meet
daily metabolic requirements. Thus, at age 75 many healthy subjects are on the border line
with respect to performing ordinary activities, implying that even a low level of COHb might
be enough to critically impair oxygen delivery to the tissues and severely limit daily metabolic
requirements.
25 The above given data refer to inactive males. However, the decline by age in maximal
oxygen uptake seems to be the same in inactive females. Because females have about 25%
lower maximal oxygen uptake expressed in milliliters per kilogram per minute, the critical
age for a female will be about 70 years. If a person is physically active on a regular basis,
the critical age with respect to maximal oxygen uptake would be expected to increase by 15 to
30 20 years. Because females have a longer life expectancy than males, the aging female
March 5, 1990 12-3 DRAFT - DO NOT QUOTE OR CITE
-------
population potentially at risk to CO exposure would be expected to be larger than the aging
male population.
5 12.3 RISK OF CARBON MONOXIDE EXPOSURE IN INDIVIDUALS
WITH PREEXISTING DISEASE
12.3.1 Subjects with Coronary Artery Disease
Coronary heart disease remains the major cause of death and disability in the United
States both in males and females. According to the most recent data compiled by the
10 American Heart Association (1988), persons with diagnosed coronary artery disease numbered
about 5.4 million in 1985 and current estimates are as high as seven million (U.S.
Department of Health and Human Services, 1987; Collins, 1988). These individuals have
myocardial ischemia, which occurs when the heart muscle receives insufficient oxygen
delivered by the blood. For some, chest discomfort called angina pectoris can occur. The
15 predominant type of ischemia, as identified by ST segment depression, in all patients with
coronary artery disease, however, is asymptomatic (i.e., silent). Especially patients who
experience angina usually have more ischemic episodes that are asymptomatic. About
10% of middle-aged men develop a positive exercise test, one of the signs of ischemia.
Nationally, more than one million heart disease deaths occur each year, half of them being
20 fatal. About 20% of all myocardial infarctions are silent. Of the 500,000 survivors of
hospitalized myocardial infarction, about 10% are asymptomatic but have signs of ischemia.
Thus, many more persons than currently known are not aware that they have coronary heart
disease and may constitute a high-risk group.
Persons with both asymptomatic and symptomatic coronary artery disease have a limited
25 coronary flow reserve and, therefore, will be sensitive to a decrease in oxygen-carrying
capacity induced by CO exposure (see Section 10.3.2). In addition, CO might interfere with
different vasomotor-active substances and thereby induce a vasoconstriction or prevent a
vasodilation in the coronary arteries which will produce ischemia that is likely to be silent.
Unfortunately, studies have not been performed specifically addressing the effect of CO
30 exposure on asymptomatic ischemia.
March 5, 1990 12-4 DRAFT - DO NOT QUOTE OR CITE
-------
12.3.2 Subjects with Congestive Heart Failure
Congestive heart failure is a major and growing public health problem in the United
States. It has been estimated that approximately three million Americans suffer from heart
failure, and moreover, because the prevalence of heart failure is known to increase with age,
5 improvements in the average life expectancy of the general population would be expected to
increase the magnitude of the problem over the next few decades.
Today about 75% of the patients with heart failure are above the age of 60 years (Brody
et al., 1987). About 400,000 new cases of heart failure are diagnosed every year in the
United States, resulting in about 1.6 million hospitalizations. The mortality rate is high,
10 between 15 to 60% per year. The cause of death is often sudden death and because about
65% of heart failure patients have serious arrhythmias, sudden death is thought to be due to
arrhythmia. Each year 200,000 patients die. The mortality is highest in Class 4 patients or
in patients with a low maximal oxygen uptake (below 10 mL/kg/min).
Patients with congestive heart failure have a markedly reduced circulatory capacity and,
15 therefore, may be very sensitive to any limitations in oxygen-carrying capacity. Thus,
exposure to CO certainly will reduce their exercise capacity and even be dangerous, especially
if CO is determined to be proarrhythmogenic (see Section 10.3.2). The etiology of heart
failure is diverse but the dominating disease is coronary artery disease. The large portion of
heart failure patients with coronary artery disease, therefore, might be even more sensitive to
20 CO exposure.
12.3.3 Subjects with Other Vascular Diseases
Peripheral vascular disease is present in about 7% of both the male and female
population and is more prevalent above age 65 years. Cerebrovascular disease also is present
25 in about 6.6% of both the male and female population of the same ages. Both of these
conditions often are found in subjects with coronary artery disease. Both conditions also are
associated with a limited blood flow capacity and, therefore, should be sensitive to CO
exposure. It is not clear, however, how low levels of exposure to CO will affect these
individuals. Only one study (Aronow et al., 1974), reviewed in the previous criteria
30 document (U.S. Environmental Protection Agency, 1979), has been reported on patients with
peripheral vascular disease. Ten men with diagnosed intermittent claudication experienced a
March 5, 1990 12-5 DRAFT - DO NOT QUOTE OR CITE
-------
significant decrease in time to onset of leg pain when exercising on a bicycle ergometer after
breathing 50 ppm CO for 2 h (2.8% COHb). Further research is needed, therefore, to better
determine the sensitivity of subjects with vascular disease to CO.
5 12.3.4 Subjects with Anemia and Other Hematologic Disorders
Clinically diagnosed low values of hemoglobin, characterized as anemia, is a relatively
prevalent condition in the United States. If the anemia is mild to moderate, an inactive
person is often asymptomatic. However, due to the limitation in the oxygen-carrying capacity
resulting from the low hemoglobin values, an anemic person should be more sensitive to low-
10 level CO exposure than a person with normal hemoglobin levels (see Section 10.3.2). If
anemia is combined with other prevalent diseases, such as coronary artery disease, the
individual also will be at an increased risk to CO exposure. Anemia is more prevalent in
women and in the elderly, already two potentially "high" risk groups. An anemic person also
will be more sensitive to the combination of CO exposure and high altitude. Additional
15 studies are needed, therefore, in order to determine the susceptibility of this group to CO
exposure.
Individuals with hemolytic anemia often have higher baseline levels of COHb because
the rate of endogenous CO production from heme catabolism is increased. One of the many
causes of anemia is the presence of abnormal hemoglobin in the blood. For example, in
20 sickle-cell disease the average lifespan of red blood cells with abnormal hemoglobin S
(Hb S) is 12 days compared to an average of 88 days in healthy individuals with normal
hemoglobin (Hb A). As a result, baseline COHb levels can be as high as 4% (Solanki et al.,
1988). In subjects with hemoglobin Zurich, where affinity for CO is 65 times that of normal
hemoglobin, COHb levels range from 4 to 7% (Zinkham et al., 1980).
25 There are over 350 variants to normal human hemoglobin (Zinkham et al., 1980). In
the Hb S variant, sickling takes place when deoxy Hb S in the red blood cell reaches a critical
level and causes intracellular polymerization. Oxygenation of the Hb S molecules in the
polymer, therefore, should lead to a change in molecular shape, breakup of the polymer, and
unsickling of the cell. Carbon monoxide was considered at one time to be potentially
30 beneficial because it ultimately would reduce the concentration of deoxy Hb S by converting
March 5, 1990 12-6 DRAFT - DO NOT QUOTE OR CITE
-------
part of the hemoglobin to COHb. Exposure to CO, however, was not considered to be an
effective clinical treatment because high COHb levels (>20%) were required.
Other hematologic disorders can cause elevated concentrations of COHb in the blood.
Ko and Eisenberg (1987) studied a patient with Waldenstrom's Macroglobulinemia. Not only
5 was the COHb saturation elevated, but the half-life of COHb was about three times longer
than in a normal individual. Presumably, exogenous exposure to CO, in conjunction with
higher endogenous CO levels, could result in critical levels of COHb. However, because CO
also can modify the characteristics of unstable hemoglobin, as demonstrated in patients with
Hb S it is not known how ambient or near-ambient levels of CO would affect individuals with
10 these disorders.
12.3.5 Subjects with Obstructive Lung Disease
Chronic obstructive pulmonary disease (COPD) is a prevalent disease especially among
smokers. It is estimated (U.S. Department of Health and Human Services, 1987; Collins,
15 1988) that 14 million persons (-6%) suffer from COPD in the United States and that a large
number (>50%) of these individuals have limitations in their exercise performance
demonstrated by a decrease in oxygen saturation during mild to moderate exercise. In spite
of their symptoms, many of them (-30%) continue to smoke and already may have COHb
levels of 4 to 8%. Subjects with hypoxia are also more likely to have a progression of the
20 disease resulting in severe pulmonary insufficiency, pulmonary hypertension, and right heart
failure. Studies by Aronow et al. (1977) and Calverley et al. (1981), reviewed in
Chapter 10, suggest that individuals with hypoxia due to chronic lung disease such as
bronchitis and emphysema may be susceptible to CO during submaximal exercise typically
found during normal daily activity.
25 The prevalence of chronic asthma in the United States is estimated to be as high as
9 million persons or about 4% of the total population (U.S. Department of Health and Human
Services, 1987; Collins, 1986, 1988). There has been evidence that hospital admissions for
asthma have increased considerably in the past few years, particularly among individuals less
than 18 years of age. Because asthmatics also can experience exercise-induced airflow
30 limitation, it is likely that they also would be susceptible to hypoxia. It is not known,
however, how exposure to CO would affect these individuals.
March 5, 1990 12-7 DRAFT - DO NOT QUOTE OR CITE
-------
12.4 SUBPOPULATIONS AT RISK FROM COMBINED EXPOSURE TO
CARBON MONOXIDE AND OTHER CHEMICAL SUBSTANCES
12.4.1 Interactions with Psychoactive Drugs
There is almost a complete lack of data on the possible toxic consequences of combined
5 CO exposure and drug use. The most extensively studied interaction has been combined
exposure to CO and alcohol. The previous criteria document (U.S. Environmental Protection
Agency, 1979) reviewed an extensive human study of alcohol-CO interactions on driving
performance (Rockwell and Weir, 1975). In this study of actual driving behavior, alcohol
and CO effects were often additive, and at 12% COHb concentrations, combined effects were
10 observed that were greater than the sum of the effects of CO and alcohol alone. Two animal
studies of alcohol-CO combinations (Mitchell et al., 1978; Knisely et al., 1989) also provide
evidence that the effects of alcohol on behavior can be enhanced by high concentrations of
CO exposure.
Thus is seems prudent to tentatively conclude that the behavioral effects of alcohol may
15 be exacerbated under some conditions of CO exposure. What is not known is the range of
behavioral effects for which this occurs, the quantitative nature of the interaction, the
mechanism of the combined effects, or the minimal COHb concentrations needed to see an
interaction. Further research on this clearly is needed. This is particularly the case when one
considers the role of alcohol in our society and the likelihood of frequent opportunities for
20 combined alcohol use and CO exposure. Some statistics from the recent report to Congress
on Alcohol and Health illustrate the potential problem (National Institute on Alcohol Abuse
and Alcoholism, 1987), In 1984, the estimated per capita alcohol consumption per year in the
United States was 2.65 gal of pure alcohol per person over the age of 14. The National
Institute on Alcohol Abuse and Alcoholism estimates that two-thirds of the U.S. population
25 over the age of 18 drink alcohol, and one-half of these are moderate to heavy drinkers.
Nearly 50% of all accidental deaths are alcohol related. Even a small interaction of CO
exposure with alcohol would be magnified by the high incidence of these combinations.
Other studies of interactions of CO and drugs have been conducted; however not nearly
enough data exist upon which one could draw conclusions concerning populations at risk.
30 Some evidence from animal research indicates that CO exposure may alter the effects of
pentobarbital, d-amphetamine, and chlorpromazine (McMillan and Miller, 1974; Knisely
March 5, 1990 12-8 DRAFT - DO NOT QUOTE OR CITE
-------
et al., 1989). Because these drugs represent diverse classes of psychoactive drugs, and many
other classes have not been examined at all, it must be concluded that this is an area of
concern for which it is difficult at the present time to make recommendations that will have
an effect on air quality standards. The lack of data on possible interactions of CO exposure
5 and drug use was identified in both the 1979 Criteria Document and an addendum to that
document (U.S. Environmental Protection Agency, 1979, 1984). Little has changed since
then.
12.4.2 Interactions with Cardiovascular Drugs
10 There are limited data currently available to determine if there is a possible interaction
between CO exposure and different cardiovascular drugs. Drugs used to treat patients with
coronary artery disease, such as betablockers, calcium channel blockers, and nitrates, should
be tested for potential interaction with CO because those patients already are high-risk
subjects. Patients with angina that were used as subjects in studies on the effects of CO
15 exposure (see Chapter 10, Section 10.3) also were treated with these classes of drugs.
Unfortunately, drug interactions were not investigated in most of the studies. Only Allred
et al. (1989a,b) analyzed their data for potential medication effect and no interactions with
CO were found. The only other available data dealt with the interaction of CO with
betablockers and calcium blockers in smokers. Deanfield et al. (1984) studied 10 smoking
20 patients with stable angina in a double-blind placebo controlled study. He studied two
betablockers, atenolol and propranolol, and one calcium blocker, nifedipine. The patients
underwent exercise tests and Holter monitoring both when they still were smoking and after
they had stopped smoking for one month. The performance and results from Holter
monitoring showed improvement after the patients refrained from smoking. The difference
25 was largest for nifedipine. Blood levels of propranolol were increased when the patients
stopped smoking; levels of nifedipine and atenolol were unchanged. Part of the decreased
efficacy of the drugs while smoking might be due to lower plasma levels and part of it might
be due to some interaction on a cellular level. However, it currently is not known if the
interaction was due to nicotine and/or CO.
30 Another of the high risk groups using multiple medications are heart failure patients.
They often use digitalis, diuretics, vasodilators, and recently, inhibitors of angiotension
March 5, 1990 12-9 DRAFT - DO NOT QUOTE OR CITE
-------
converting enzyme. If CO exposure modifies the responses to those drugs, the patients' status
may deteriorate when the plasma levels of a drug are lower or the patients may develop side
effects when the plasma levels of a drug are higher. Due to the large number of high-risk
patients with coronary artery disease and heart failure that use often very potent and multiple
5 mediations, this area needs to be addressed carefully through further research.
12.4.3 Mechanisms of Carbon Monoxide Interactions with Drugs: Need
for Further Research
Because data are generally lacking on CO-drug interactions, it should be useful to
10 speculate on some of the mechanisms by which CO might be expected to alter drug effects, or
vice versa, and discuss possible populations at risk due to these potential interaction effects.
12.4.3.1 Metabolic Effects
A mechanism by which CO might be expected to interact with many drugs is through
15 the modification of drug metabolism. CO is known to bind to cytochrome P-450 in vitro
(Gray, 1982), but the significance of this under physiological conditions is not known.
Another section of this document (see Section 9.4) reviews the interactions of CO with
oxidative metabolism and concludes that clinically relevant inhibition of these systems
probably does not occur under most conditions of exposure. If further research provides
20 evidence that these important drug-metabolizing systems are significantly compromised as a
result of ambient CO exposure, then drugs dependent upon these systems for activation or
deactivation would interact with CO exposures. If changes in drug metabolism occur as a
result of CO exposure, this would be of considerable practical importance. It might be
necessary to alter prescribing practices in heavily exposed populations. More research on this
25 is needed,
12.4.3.2 Central Nervous System (CNS) Depression
In the absence of systematic data on the interactions of CO with psychoactive drugs, it is
necessary to hypothesize mechanisms by which such interactions might occur. In the 1984
30 Addendum (U.S. Environmental Protection Agency, 1984), it was speculated that "drugs with
primary or secondary CNS depressant effects should be expected to exacerbate the
March 5, 1990 12-10 DRAFT - DO NOT QUOTE OR CITE
-------
neurobehavioral effects of CO," presumably because of the generally depressant effects on the
nervous system of CO itself. On the other hand, one might also argue the converse (i.e., that
CNS depressant drugs, because they might reduce cerebral metabolism and hence oxygen
utilization, could lessen the neurobehavioral effects of CO). It should be obvious that
5 speculation on these matters, in the absence of data, cannot be expected to yield answers upon
which regulatory decisions could be made. On the other hand, because of the overall
sensitivity of the CNS to perturbations, it is possible that interactions of these types could
occur and may even be quite pronounced. Clearly, more research on this is needed because
we cannot rely on scientific speculation.
10
12.4.3.3 Alteration in Cerebral Blood Flow
Another mechanism by which CO could be speculated to interact with certain drugs is
through modification of cerebral blood flow. Brain hypoxia resulting from CO exposure may
result in compensatory increases in cerebral blood flow (Dobler et al., 1977). Drugs that
15 have vasoconstrictive effects on cerebral circulation could be hypothesized to interfere with
this compensatory mechanism and thus exacerbate the neurobehavioral toxicity of CO. The
methylxanthines, such as caffeine and theophylline, have well-established central
vasoconstrictive effects (Rail, 1980) and thus could be hypothesized to enhance CO-induced
brain hypoxia. On the other hand, their vasodilatory effects in the periphery (Rail, 1980)
20 might enhance the vasodilatory effects of CO. Because of the widespread use of
methylxanthines, these possible interactions may be of particular significance.
As the oxygen-carrying capacity of the blood decreases with CO poisoning, many
organs, including the brain, will compensate their blood flow to try and maintain proper
tissue oxygenation. Several studies using radiolabeled microspheres to measure cerebral
25 blood flow have demonstrated autoregulation and increased blood flow in response to CO
(Koehler et al., 1982). However, if the brain oxygen supply is inadequate despite increased
blood flow, further metabolic changes will undoubtedly occur. The consequences of these
metabolic changes with respect to their effect on the regulation of brain blood flow is
uncertain. More simply, damage resulting from lack of proper brain oxygenation may alter
30 the brain vasculature's ability to regulate brain blood flow. Damage to the vasculature
previously has been shown to alter the vasculature's response to vasoactive agents. For
March 5, 1990 12-11 DRAFT - DO NOT QUOTE OR CITE
-------
example, it is known that following ischemia-induced injury and other types of brain injury,
the brain releases polyunsaturated fatty acids from its phospholipids (Gardiner et al., 1981).
These free fatty acids may be metabolized enzymatically to compounds such as prostaglandins
and leukotrienes with effects on the cerebral vasculature. For example, the metabolism of
5 arachidonic acid generates prostaglandins and oxygen-free radicals that can cause cerebral
vasodilation in normal animals. However, when these radicals are produced in excess, as in
acute, extreme hypertensive episodes, the free radicals initiate peroxidation of other
unsaturated fatty acids (Kukreja et al., 1986). These peroxides and oxygen radicals cause
damage to the vascular endothelium and decrease the brain's capacity to regulate blood flow
10 in response to changes in arterial CO2 (Wei et al., 1981). Additionally, the vascular damage
caused by these oxygen radicals alters the normal cerebral arterial response to vasoactive
agents, including neurotransmitters. For example, acetylcholine, which is normally a dilator
of cerebral arterioles, produces vasoconstriction after free radical-induced damage (Wei et al.,
1985). Potentially, therefore, in an injured brain, acetylcholine may decrease an already
15 inadequate blood flow.
Conceptually, tissue hypoxia produced by CO may stimulate arachidonic acid
metabolism and production of prostaglandins and free radicals in a manner similar to hypoxia
caused by ischemia or trauma. Production of vasodilator free radicals or vasodilator
prostaglandins may be a mechanism by which the brain increases its blood flow in response to
20 CO. Assuming this is the case, therapeutic agents or drugs of abuse that modify the
arachidonic acid cascade may alter the brain's capacity to increase blood flow in response to
CO. For example, if aspirin, indomethacin, or other cyclooxygenase inhibitors are present
during exposure to CO, the brain's capacity to increase its flow may be diminished due to
decreased capacity to form dilator prostanoids and free radicals. An instance where this sort
25 of possibility is known to occur is in neonatal animals, and possibly in neonatal humans.
Investigators recently have shown in neonatal animals that indomethacin markedly diminished
the brain's capacity to increase its blood flow in response to hypoxia (Leffler and Busija,
1987).
A possible interaction between CO and nitrite exposure also might be predicted. Nitrites
30 can be expected to oxidize hemoglobin to methemoglobin leaving less hemoglobin to bind
either to O2 or CO. However, since CO has a greater affinity than O2 for hemoglobin, it is
March 5, 1990 12-12 DRAFT - DO NOT QUOTE OR CITE
-------
most likely to expect additive effects in reduction of oxyhemoglobin. In addition, some
organic nitrites such as amyl nitrite, used to relieve angina pain, and butyl nitrite, an abused
substance, produce significant peripheral vasodilation and, consequently, an abrupt drop in
blood pressure and subsequent tachycardia. These potent cardiovascular effect's which result
5 from carbon monoxide exposure, might interfere with the enhanced cardiac output,
particularly the output to sensitive organs such as the brain. To date there are no published
data on the combined effect of CO and nitrite exposure. However, there are limited data
showing reasonably parallel consequences on auditory function (Fechter et al., 1987, 1989)
when CO and butyl nitrite are given individually to rats.
10 Additionally, recent evidence shows that acetylcholine stimulates arachidonic acid
metabolism (Busija et al., 1987). Whether hypoxia increases arachidonic acid metabolism via
stimulation of acetylcholine release is uncertain, however, exogenous acetylcholine is known
to stimulate brain prostaglandin production. Assuming that endogenous acetylcholine release
in response to CO-induced hypoxia is important, other agents such as atropine or
15 scopolamine, which block muscarinic receptors, could reduce the vasodilator response to CO.
In an opposite manner, cholinesterase inhibitors (e.g., organophosphate or carbamate
insecticides) that penetrate the blood brain barriers may magnify the dilator response to CO.
Other agents that may modify the capacity of the brain's blood flow to regulate in response to
CO are agents that cause an acute, large increase in blood pressure, thus inducing excess free
20 radical production, lipid peroxidation, and abnormal vascular reactivity. Such an agent might
include for example, cocaine, which when administered in large doses causes acute transient
hypertension.
While the above points are speculative, the possibility that therapeutic agents and drugs
of abuse may alter the brain vasculature's capacity to respond to CO is a subject that bears
25 further consideration and investigation.
12.4.4 Interactions with Other Chemical Substances in the Environment
Besides direct ambient exposure to CO, there are other chemical substances in the
environment that can lead to increased COHb saturation when inhaled. Halogenated
30 hydrocarbons used as organic solvents undergo metabolic breakdown by cytochrome P-450 to
form CO and inorganic halide. Possibly the greatest concern regarding potential risk in the
March 5, 1990 12-13 DRAFT - DO NOT QUOTE OR CITE
-------
population comes from exposure to one of these halogenated hydrocarbons, methylene
chloride (CHjCy, and some of its derivatives. Almost a million kilograms are produced each
year, making it the second highest source of CO in the environment. Although it is present in
ambient air emissions, the highest concentrations of CH2C12 occur from various sources such
5 as paint removers, cleaners, propellants, and from industrial manufacturing.
From available experimental studies (see Section 11-3), it is not clear if combined
exposure to CO and CH2C12 would produce an additive effect in humans. Theoretically, acute
CH2C12 exposure can result in a steady production of endogenous CO in tissues such as the
lung, liver, kidney, heart, and brain that contain cytochrome P-450. Any histotoxic hypoxia
10 produced at the tissue level combined with hypoxic hypoxia due to the formation of COHb
from endogenous as well as exogenous CO exposure could place exposed individuals at risk.
12.5 SUBPOPULATIONS EXPOSED TO CARBON MONOXIDE AT
15 HIGH ALTITUDES
For patients with coronary artery disease, restricted coronary blood flow limits oxygen
delivery to the myocardium. Carbon monoxide also has the potential for compromising
oxygen transport to the heart. For this reason, such patients have been identified as the
subpopulation most sensitive to the effects of CO. A reduction in the partial pressure of
20 oxygen (POj) in the atmosphere, as at high altitude, also has the potential for compromising
oxygen transport. Therefore, patients with coronary artery disease who visit higher elevations
might be unusually sensitive to the added effects of atmospheric CO.
Before considering the combined effects of CO and atmospheric hypoxia, it is important
to distinguish between the long term resident of high altitude, as compared with the newly
25 arrived visitor from low altitude. Specifically, the visitor will be more hypoxemic than the
fully adapted resident for the following reasons. Initially, the visitor will exhibit relative
hypoventilation, particularly during sleep, since ventilatory adaptation requires several days.
The result will be a lowering of arterial P02, a fall in arterial O2 saturation, and a reduction in
arterial O2 content. This hypoxemia will stimulate the sympathetic nervous system to increase
30 heart rate, myocardial contractility, and systemic arterial blood pressure (Grover et al., 1986).
These factors combine to increase cardiac work, calling for an increase in coronary blood
March 5, 1990 12-14 DRAFT - DO NOT QUOTE OR CITE
-------
flow. In addition, the initial increase in ventilation will produce a respiratory alkalosis which,
in turn, will increase the affinity of hemoglobin for oxygen and thereby interfere with oxygen
release to the tissues.
Over several days following arrival at high altitude, a number of mechanism will
5 operate to lessen the initial impact of atmospheric hypoxia. Ventilation will increase
progressively, and this will evaluate arterial O2 tension, saturation, and content. A decrease
in plasma volume increases hematocrit (hemoconcentration), with an associated increase in the
O2-carrying capacity (hemoglobin concentration) of the blood; at this point, the polycythemia
is only relative not absolute. Nevertheless, this will increase further the arterial O2 content.
10 Although increased sympathetic activity persists, cardiac beta receptor responsiveness
decreases, mitigating the initial tachycardia. This combined with a decreased in cardiac
stroke volume leads to a return of cardiac output to normal (or even subnormal) levels
(Grover et al., 1986). Compensation for the initial respiratory alkalosis returns blood pH
towards normal. Concurrently, there is an increase in the concentration of 2,3-
15 disphosphoglycerate (2,3-DPG) within the red cells, the net effect being not only a return of
hemoglobin-oxygen affinity to normal but actually to levels lower than prior to ascent. This
facilitates the release of oxygen to the tissues, an effect that more than offsets the slight
decrease in arterial O2 saturation. For the heart, this is particularly important, for it removes
the demands for increased coronary blood flow at moderate altitude (Grover et al., 1976).
20 For the long-term resident at high altitude, systemic blood pressure returns to (or below)
values normal for sea level (Marticorena et al., 1969). Cardiac output remains at (or below)
levels normal for sea level (Hartley et al., 1967). Tissue capillary density increases, thereby
enhancing oxygen delivery. Consequently, demands on the coronary circulation are not
increased. An absolute polycythemia develops, i.e., total red cell mass plateaus at levels
25 greater than at sea level. As a consequence, the normal turn-over of this greater mass of red
cells results in an increase in the endogenous production of CO (Johnson, 1968).
Based on these considerations, the population subgroup at greatest risk from CO
exposure would be the newly-arrived transient visitors to high altitudes. By binding
hemoglobin, CO would reduce further arterial O2 content, i.e., increase hypoxemia. In
30 addition, CO would augment the effect of alkalosis by increasing further the affinity of
hemoglobin for oxygen, thereby impairing O2 delivery even more. Both factors would
March 5, 1990 12-15 DRAFT - DO NOT QUOTE OR CITE
-------
increase demands for greater coronary blood flow. These initial risks would decline
progressively if the visitor remains long enough to complete physiological adaptation. The
period of increased risk is probably prolonged in the elderly, since adaptation to high altitude
proceeds more slowly with increasing age (Dill et al., 1963, 1985; Robinson et al., 1973).
5 Not surprisingly, the total number of transient visitors to high altitude far exceeds the
resident population. Ironically, it is these same transient visitors who contribute most to
atmospheric pollution with CO in the mountains (automobile engines not tuned to high
altitude, inefficient wood burning fireplaces used for social effect in vacation cabins, etc.). In
addition, newly arrived visitors are often unaware of the physiological effects of high altitude
10 (plus CO), and hence are prone to over-exertion which would increase the potential hazard.
For a variety of reasons, COHb concentrations tend to be higher in high altitude residents
than seen at low altitude (Johnson, 1968).
One would postulate that the combination of high altitude with CO would pose the
greatest risk to persons newly arrived at high altitude who have underlying cardiovascular
15 disease, particularly since they are usually older individuals. Surprisingly, this hypothesis has
never been tested adequately. In fact, there are virtually no data on how patients with known
or suspected coronary artery disease responde to a sojourn at moderately high altitude (8,000
to 12,000 ft or higher) with or without added CO exposure. In two pilot studies, the risk
from altitude alone at least appears to be minimal (Okin, 1970; Khanna et al., 1976). Among
20 148,000 persons (10% over 50 years of age) trekking in Nepal to altitudes up to 18,000 ft,
there were no cardiac deaths and only three helicopter evacuations for cardiac problems
(Shlim and Houston, 1989). Nevertheless, the need remains for a rigorous test of the
hypothesis.
If the cardiovascular effects of atmospheric hypoxia at high altitude are augmented by
25 added exposure to CO, then patients already hypoxic from chronic obstructive lung disease
should also be at increased risk from CO at altitude. Paradoxically, this does not appear to
be true, again at least for brief exposure to altitude alone, even though hypoxemia is
exaggerated (Graham and Houston, 1978; Schvvart et al., 1984). This may reflect the
decrease in air density at high altitude which reduces both the work of breathing (Thoden
30 et al., 1969) as well as the effective degree of airway obstruction in such patients (Kryger
etal., 1978).
March 5, 1990 12-16 DRAFT - DO NOT QUOTE OR CITE
-------
Although limited observations do not indicate an increased risk from exposure to
moderate altitude (without added CO) for patients with either cardiovascular or obstructive
airway disease, this does not imply that prolonged residence at altitude is well tolerated.
Individuals with these disorders, while successfully living at higher altitudes initially, tend to
5 leave these altitudes as they reach older age. Outward migration of older individuals with
these disorders has been described for the higher elevations in the state of Colorado
(Regensteiner and Moore, 1985). These elderly residents living at altitudes above 8,000 ft
left primarily due to poor health. Heart disease and lung disease (each 41%) accounted for
the majority of reasons for leaving their high altitude homes.
10 It is known that low birth weights occur in both infants born at altitudes above 6,000 ft
as well as infants born near sea level whose mothers had elevated COHb levels due to
cigarette smoking (see Chapter 11, Section 11.1). It has also been shown that COHb levels in
smokers at high altitude are higher than in smokers at sea level (Brewer et al., 1970). While
it is probable that the combination of hypoxic hypoxia and hypoxia resulting from ambient
15 exposure to CO could further reduce birth weight at high altitude and possibly modify future
development, no data are presently available to support this hypothesis. A study conducted in
Colorado (Alderman et al., 1987) failed to find a strong relationship between risk of low birth
weight and maternal exposure to neighborhood CO estimated from stationary monitors. The
combination, however, of maternal smoking and 6,000 ft altitude did result in lower birth
20 weights than those due to altitude alone.
12.6 REFERENCES
25
Alderman, B. W.; Baron, A. E.; Savitz, D. A. (1987) Maternal exposure to neighborhood carbon monoxide and
risk of low infant birth weight. Public Health Rep. 102: 410-414.
Allred, E. N.; Bleecker, E. R.; Chaitman, B. R.; Dahms, T. E.; Gottlieb, S. O. Hackney, J. D.; Pagano, M.;
30 Selvester, R. H.; Walden, S. M.; Warren, J. (1989a) Short-term effects of carbon monoxide exposure on
the exercise performance of subjects with coronary artery disease. N. Engl. J. Med. 321: 1426-1432.
Allred, E. N.; Bleecker, E. R.; Chairman, B. R.; Dahms, T. E.; Gottlieb, S. O.; Hackney, J. D.; Hayes, D.;
Pagano, M. Selvester, R. H.; Walden, S. M.; Warren, J. (1989b) Acute effects of carbon monoxide
35 exposure on individuals with coronary artery disease. Cambridge, MA: Health Effects Institute; research
report no. 25.
American Heart Association. (1988) 1988 heart facts. Dallas, TX: American Heart Association.
March 5, 1990 12-17 DRAFT - DO NOT QUOTE OR CITE
-------
Aronow, W. S.; Stemmer, E. A.; Isbell, M. W. (1974) Effect of carbon monoxide exposure on intermittent
claudication. Circulation 49: 415-417.
Aronow, W. S.; Ferlinz, J.; Glauser, F. (1977) Effect of carbon monoxide on exercise performance in chronic
5 obstructive pulmonary disease. Am. J. Med. 63: 904-908.
Brewer, G. J.; Eaton, J. W.; Weil, J. V.; Grover, R. F. (1970) Studies of red cell glycolysis and interactions
with carbon monoxide, smoking, and altitude. In: Brewer, G. J., ed. Red cell metabolism and function:
proceedings of the first international conference on red cell metabolism and function; October 1969; Ann
10 Arbor, MI. New York, NY: Plenum Press; pp. 95-114. (Advances in experimental medicine and
biology: v. 6).
Brody, J. A.; Brock, D. B.; Williams, T. F. (1987) Trends in the health of the elderly population. Annu. Rev.
Public Health 8: 211-234.
15
Busija, D. W.; Wagerle, L. C.; Pourcyrous, M.; Leffler, C. W. (1987) Acetylcholine dramatically increases
prostanoid synthesis in piglet parietal cortex. Brain Res. 439: 122-126.
Calverley, P. M. A.; Leggett, R. J. E.; Flenley, D. C. (1981) Carbon monoxide and exercise tolerance in
20 chronic bronchitis and emphysema. Br. Med. J. 283: 878-880.
Collins, J. G. (1986) Prevalence of selected chronic conditions: United States, 1979-81 Hyattsville, MD: U. S.
Department of Health and Human Services, National Center for Health Statistics; DHHS publication no.
(PHS) 86-1583. (Data from the national health survey, series 10, no. 155).
25
Collins, J. G. (1988) Prevalence of selected chronic conditions, United States, 1983-85. Hyattsville, MD: U. S.
Department of Health and Human Services, National Center for Health Statistics; DHHS publication no.
(PHS) 88-1250. (Advance data from vital and health statistics no. 155).
30 Deanfield, J.; Wright, C.; Rrikler, S.; Ribeiro, P.; Fox, K. (1984) Cigarette smoking and the treatment of
angina with propranolol, atenolol, and nifedipine. N. Engl. J. Med. 310: 951-954.
Dill, D. B.; Robinson, S.; Balke, B. (1963) Respiratory responses to exercise as related to age. In: Cunningham,
D. J. C.; Lloyd, B. B., eds. The regulation of human respiration. Oxford, United Kingdom: Blackwell;
35 pp. 453-460.
Dill, D. B.; Alexander, W. C.; Myhre, L. G.; Whinnery, J. E.; Tucker, D. M. (1985) Aerobic capacity of
D. B. Dill, 1928-1984. Fed. Proc. Fed. Am. Soc. Exp. Biol. 44: 1013.
40 Doblar, D. D.; Santiago, T. V.; Edelman, N. H. (1977) Correlation between ventilatory and cerebrovascular
responses to inhalation of CO. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 43: 455-462.
Fechter, L. D.; Thorne, P. R.; Nuttall, A. L. (1987) Effects of carbon monoxide on cochlear electrophysiology
and blood flow. Hear. Res. 27: 37-45.
45
Fechter, L. D.; Richard, C. L.; Mungekar, M.; Gomez, J.; Strathern, D. (1989) Disruption of auditory function
by acute administration of a "room odorizer" containing butyl nitrite in rats. Fundam. Appl. Toxicol.
12: 56-61.
50 Gardiner, M.; Nilsson, B.; Rehncrona, S.; Siesjo, B. K. (1981) Free fatty acids in the rat brain in moderate and
severe hypoxia. J. Neurochem. 36: 1500-1505.
Golanki, D. L.; McCurdy, P. R.; Cuttitta, F. F.; Schechter, G. P. (1988) Hemolysis in sickle cell disease
measured by endogenous CO production. Am. J. Clin. Pathol. 89: 221-225.
March 5, 1990 12-18 DRAFT - DO NOT QUOTE OR CITE
-------
Graham, W. G. B.; Houston, C. S. (1978) Short-term adaptation to moderate altitude: patients with chronic
obstructive pulmonary disease. JAMA J. Am. Med. Assoc. 240: 1491-1494.
Gray, R. D, (1982) Kinetics and mechanism of carbon monoxide binding to purified liver microsomal cytochrome
5 P-450 isozymes. J. Biol. Chem. 257: 1086-1094.
Grover, R. F.; Lufschanowski, R.; Alexander, J. K. (1976) Alterations in the coronary circulation of man
following ascent to 3,100 m altitude. J. Appl. Physiol. 41: 832-838.
10 Grover, R. F.; Weil, J. V.; Reeves, J. T. (1986) Cardiovascular adaptation to exercise at high altitude. In:
Pandolf, K. B., ed. Exercise and sport sciences reviews: v. 14. New York, NY: Macmillan; pp. 269-302.
Hartley, L. H.; Alexander, J. K.; Modelski, M.; Grover, R. F. (1967) Subnormal cardiac output at rest and
during exercise at 3,100 m altitude. J. Appl. Physiol. 23: 839-848.
15
Johnson, R. L., Jr. (1968) Rate of red cell and hemoglobin destruction after descent from high altitude. Brooks
Air Force Base, TX: USAF School of Aerospace Medicine; contract no. AF 41609-68-C-0032.
Khanna, P. K.; Dham, S. K.; Hoon, R. S. (1976) Exercise in an hypoxic environment as a screening test for
20 ischaemic heart disease. Aviat. Space Environ. Med. 47: 1114-1117.
Knisely, J. S.; Rees, D. C.; Balster, R. L. (1989) Effects of carbon monoxide in combination with behaviorally
active drugs on fixed-ratio performance in the mouse. Neurotoxicol. Teratol. 11: 447-452.
25 Ko, B. H.; Eisenberg, R. S. (1987) Prolonged carboxyhemoglobin clearance in a patient with Waldenstrom's
macroglobulinemia. Am. J. Emerg. Med. 5: 503-508.
Koehler, R. C.; Jones, M. D., Jr.; Traystman, R. J. (1982) Cerebral circulatory response to carbon monoxide
and hypoxic hypoxia in the lamb. Am. J. Physiol. 243: H27-H32.
30
Kryger, M.; Aldrich, F.; Reeves, J. T.; Grover, R. F. (1978) Diagnosis of airflow obstruction at high altitude.
Am. Rev. Respir. Dis. 117: 1055-1058.
Kukreja, R. C.; Kontos, H. A.; Hess, M. L.; Ellis, E. F. (1986) PGH synthase and lipoxygenase generate
35 superoxide in the presence of NADH or NADPH. Circ. Res. 59: 612-619.
Leffler, C. W.; Busija, D. W. (1987) Arachidonic acid metabolites and perinatal hemodynamics. Semin.
Perinatal. 11: 31-42.
40 Marticorena, E.; Ruiz, L.; Severino, J.; Galvez, J.; Penaloza, D. (1969) Systemic blood pressure in white men
born at sea level: changes after long residence at high altitudes. Am. J. Cardiol. 23: 364-368.
McMillan, D. E.; Miller, A. T., Jr. (1974) Interactions between carbon monoxide and a-amphetamine or
pentobarbital on schedule-controlled behavior. Environ. Res. 8: 53-63.
45
Mitchell, D. S.; Packham, S. C.; Fitzgerald, W. E. (1978) Effects of ethanol and carbon monoxide on two
measures of behavioral incapacitation of rats. Proc. West. Pharmacol. Soc. 21: 427-431.
National Institute on Alcohol Abuse and Alcoholism. (1987) Sixth special report to the U.S. Congress on alcohol
50 and health from the Secretary of Health and Human Services. Rockville, MD: U. S. Department of
Health and Human Services, Alcohol, Drug Abuse, and Mental Health Administration; DHHS publication
no. (ADM)87-1519.
March 5, 1990 12-19 DRAFT - DO NOT QUOTE OR CITE
-------
Okin, J. T. (1970) Response of patients with coronary heart disease to exercise at varying altitude. Adv. Cardiol.
5: 92-96.
Rail, T. W. (1980) Central nervous system stimulants: the xanthines. In: Oilman, A. G.; Goodman, L. S.;
5 Oilman, A., eds. The pharmacological basis of therapeutics. 6th ed. New York, NY: Macmillan
Publiching Co., Inc.; pp. 592-607.
Regensteiner, J. G.; Moore, L. G. (1985) Migration of the elderly from high altitudes in Colorado. JAMA J.
Am. Med. Assoc. 253: 3124-3128.
10
Robinson, S.; Dill, D. B.; Ross, J. C.; Robinson, R. D. (1973) Training and physiological aging in man. Fed.
Proc. Fed. Am. Soc. Exp. Biol. 32: 1628-1634.
Rockwell, T. J.; Weir, F. W. (1975) The interactive effects of carbon monoxide and alcohol on driving skills.
15 Columbus, OH: The Ohio State University Research Foundation; CRC-APRAC project CAPM-9-69.
Available from: NTIS, Springfield, VA; PB-242266.
Schwartz, J. S.; Bencowitz, H. Z.; Moser, K. M. (1984) Air travel hypoxemia with chronic obstructive
pulmonary disease. Ann. Int. Med. 100: 473-477.
20
Shlim, D. R.; Houston, R. (1989) Helicopter rescues and deaths among trekkers in Nepal. JAMA J. Am. Med.
Assoc. 261: 1017-1019.
Solanki, D. L.; McCurdy, P. R.; Cuttitta, F. F. (1988) Hemolysis in sickle cell disease as measured by
25 endogenous carbon monoxide production: a preliminary report. Am. J. Clin. Pathol. 89: 221-225.
Thoden, J. S.; Dempsey, J. A.; Reddan, W. G.; Bimbaum, M. L.; Forster, H. V.; Grover, R. F.; Rankin, J.
(1969) Ventilatory work during steady-state exercise. Fed. Proc. Fed. Am. Soc. Exp. Biol.
28: 1316-1321.
30
U. S. Department of Health and Human Services. (1987) Current estimates from the national health interview
survey. Series 10: no. 166. Data from the National Health Interview Survey. Hyattsville, MD: Public
Health Service, National Center for Health Statistics. Available from: GPO, Washington, DC; GPO
stock number 017-022-01052-6.
35
U. S. Environmental Protection Agency. (1979) Air quality criteria for carbon monoxide. Research Triangle
Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and Assessment
Office; EPA report no. EPA-600/8-79-022. Available from: NTIS, Springfield, VA; PBS 1-244840.
40 U. S. Environmental Protection Agency. (1984) Revised evaluation of health effects associated with carbon
monoxide exposure: an addendum to the 1979 EPA air quality criteria document for carbon monoxide.
Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria
and Assessment Office; EPA report no. EPA-600/9-83-033F. Available from: NTIS, Springfield, VA;
PB85-103471.
45
Wei, E. P.; Kontos, H. A.; Dietrich, W. D.; Povlishock, J. T.; Ellis, E. T. (1981) Inhibition by free radical
scavengers and by cyclooxygenase inhibitors of pial arteriolar abnormalities from concussive brain injury
in cats. Circ. Res. 48: 95-103.
50 Wei, E. P.; Kontos, H. A.; Christman, C. W.; DeWitt, D. S.; Povlishock, J. T. (1985) Superoxide generation
and reversal of acetylcholine-induced cerebral arteriolar dilation after acute hypertension. Circ. Res.
57: 781-787.
March 5, 1990 12-20 DRAFT - DO NOT QUOTE OR CITE
-------
Zinkham, W. H.; Houtchens, R. A.; Caughey, W. S. (1980) Carboxyhemoglobin levels in an unstable
hemoglobin disorder (Hb Zuerich): effect on phenotypic expression. Science (Washington, DC)
209: 406-408.
March 5, 1990 12-21 DRAFT - DO NOT QUOTE OR CITE
-------
APPENDIX A: GLOSSARY OF TERMS AND SYMBOLS
Abbreviations and Acronyms
X
12C16O
12C18O
.31J
2,3-DPG
51Cr
7-mode
85Kr
9%TcDTPA
A/F
a.k.a.
A-aDO2
ACD
ACGIH
ADD (m)
AEP
AIRS
Alt
amb
ANOVA
ANSI
Ar
atm
BEI
BF
BUS
BS
BTPS
Btu
C
Ca
CAA
CAD
CALINE3 Model
CASAC
Q
CBF
cc
CL
Atmospheric lifetime
Chi
Carbon monoxide containing oxygen isotope 16
Carbon monoxide containing oxygen isotope 18
Iodine-131
2,3-diphosphoglycerate
Chromium-51
137 second driving cycle test
Krypton-85
Radiolabeled diethylene triamine pentacetic acid
Air-to-fuel ratio
Also known as
Alveolar-arterial O2 gradient
Acid citrate dextrose
American Conference of Governmental Industrial Hygienists
Additive constant specification
Auditory evoked potential
Aerometric Information Retrieval System (U.S. EPA)
Altitude above sea level
Ambient
Analysis of variance
American National Standards Institute
Argon
Atmosphere
Biological exposure index
Blue flame (heater)
Bibliographic literature information system
Body sway
Body temperature, barometric pressure, and saturated
British thermal unit
Celsius
Calcium
Clean Air Act
Coronary artery disease
A form of dispersion modeling
Clean Air Scientific Advisory Committee
Concentration of carbon monoxide for a bulk mixture
Cerebral blood flow
Cubic centimeter *(see also cm3)
Concentration of carbon monoxide in the sample
March 12, 1990
A-1 DRAFT - DO NOT QUOTE OR CITE
-------
CEQ
CFF
Cff
CFK
CFKE
cGMP
CH2C12
GH3CC13
CH4
CHD
CI
Q(t)
CID
cm
cm3
CMRO2
CN
CNS
CNV
CO
CO
CO-Ox
CO2
COM
COHb
COMb
COPD
CP
CVD
CVS-72
CVS-75
Cyt
d1
dBA
dF/dtmax
DLCO
DA
dP/dt
DpCO
DPG
EDRF
EDTA
EEC
EKG
EP
EPA
Presidents Council on Environmental Quality
Critical flicker frequency
Critical flicker fusion
Coburn-Forster-Kane
Coburn-Forster-Kane equation
Cyclic guanosine monophosphate
Methylene chloride
Methylchloroform
Methane
Coronary heart disease
Confidence interval
The air pollutant concentration to which an individual is exposed at any
point in time t.
Cubic inch displacement
Centimeter(s)
Cubic centimeter
Cerebral O2 consumption
Cyanide
Central nervous system
Contingent negative variation (slow-evoked potential)
Carbon monoxide
Cardiac output
CO-oximeter
Carbon dioxide
Carbon monoxide hypoxia
Carboxyhemoglobin
Carboxymyoglobin
Chronic obstructive pulmonary disease
Capillary permeability to protein
Cardiovascular disease
Constant volume sample cold start test
Constant volume sample test including cold and hot starts
Cytochrome
Measure of detection threshold
Decibels (A-scale)
Derivative of maximal force
Diffusing capacity for CO, mL min'1 ton"1 (STPD)
Diffusing capacity for O2, mL min'1 (STPD)
Derivative of pressure with time
Carbon monoxide diffusion coefficient across the placenta
Diphosphoglycerides
Endothelium-derived relaxing factor
Ethylenediaminetetraacetic acid
Electroencephalogram
Electrocardiogram (also ECG)
Evoked potential
Environmental Protection Agency
March 12, 1990
A-2 DRAFT - DO NOT QUOTE OR CITE
-------
ERG
ETS
f
fB
Fco
FDA
FEF
FEV,
FFF
F.CO
FID
pio2
FMVCP
FRC
FVC
g
GC
GD
GFC
GMP
h
H20
HANES
Hb
Hb5
HbA
HBO
HbO2
HCN
HCs
Hct
HDL
He
HEI
Hg
HgO
HH
HO2*
HR
HW/BW
H*
I205
K
K2HPO4
Electroretinogram
Environmental tobacco smoke
Fetal
Flow rate of carbon monoxide for a bulk mixture
Air flow rate
Breathing frequency
Carbon monoxide flow rate
Food and Drug Administration
Forced expiratory flow
Forced expiratory volume (at one minute)
Flicker-fusion frequency
Volumetric fractional concentration of CO in dry inspired air, ppm
Flame ionization detector
Fraction of inspired O2
Federal Motor Vehicle Control Program
Functional residual capacity
Forced vital capacity
Gram(s)
Gas chromatograph
Gestation day
Gas filter correlation
Guanosine monophosphate
Hour
Water
Health and Nutrition Examination Survey
Hemoglobin concentration in blood, g dL"1
Abnormal hemoglobin found in individuals with sickle-cell disease
Normal hemoglobin
Hyperbaric oxygen
Oxyhemoglobin
Hydrogen cyanide
Hydrocarbons
Hematocrit
High-density lipoprotein
Helium
Health Effects Institute
Mercury
Mercuric oxide
Hypoxic hypoxia
Hydroperoxyl radical
Heart rate
Heart weight to body weight ratio
Hydrogen atom
Iodine pentoxide
Intraperitoneal
Warburg partition coefficient
Mono-hydrogen potassium arthrophosphate
March 12, 1990
A-3 DRAFT - DO NOT QUOTE OR CITE
-------
K4Fe(CN)6
kg
kJ
km
K,
KPM
L
LC*
LDH
LDL
LDV
LOEL
LPG
LV
MLDH
M
m
m3
Mb
MEA
MEMs
metHb
METS
MFO
mg
mi
MI
mL
MMFR
mo
mol
MRFIT
MSA
MSHA
MST
MULT (m)
n
N2
N2O
NAAQS
NADPH
NaHCO3
NAMS
NASA
Potassium ferrpcyanide
The effective reaction rate constant
Kilogram(s)
Kilo Joule, IxlO10 ergs, 0.948 Btu
Kilometer
Michaelis-Menten constant
Kilopondmeters per minute
Liter(s)
Concentration that is lethal to 50% of test subjects (used in inhalation
studies)
Dose that is lethal to 50% of test subjects
Lactate dehydrogenase
Low-density lipoprotein
Light-duty vehicle
Lowest-observed-effect level
Liquefied petroleum gas
Left ventricle
Myocardial lactate dehydrogenase
Haldane constant
Maternal
Cubic meter
Myoglobin
Mean electrical axis
Microenvironmental monitors
Methemoglobin
Basal metabolic equivalent
Mixed-function oxidase
Milligram(s)
Mile
Myocardial infarction
Milliliter(s)
Maximum mid-expiratory flow rate
Month(s)
Mole
Multiple risk factor intervention trial
Metropolitan Statistical Area
Mine Safety and Health Administration
Mean survival time
Multiplicative constant specification
Number
Nitrogen
Nitrous oxide
National Ambient Air Quality Standards
Reduced nicotinamide adenine dinucleotide phosphate
Acid sodium carbonate
National Air Monitoring Stations
National Aeronautics and Space Administration
March 12, 1990
A-4 DRAFT - DO NOT QUOTE OR CITE
-------
NBS
ND
NDIR
NEDS
NEM
NG
NIOSH
NIST
nm
NO
NO2
NO3
NOEL
NOX
NR
02
O2Mb
03
OH*
P
P
P*,
P.C02
PAC02
PAH
PAN
PA
PB
PCN
"co
PA
PCO2
PD
PEMs
PfCO
PGI2
P,CO
PiT
PmCO
PMN
P02
ppbv
ppm
ppmm
PR
PW
Q
National Bureau of Standards, now NIST
Not determined
Nondispersive infrared
National Emissions Data System
NAAQS Exposure Model
Natural gas
National Institute for Occupational Safety and Health
National Institute of Standards and Technology
Nanometer
Nitric oxide
Nitrogen dioxide
Nitrate
No-observed-effect level
Nitrogen oxides
No response
Oxygen
Oxymyoglobin
Ozone
Hydroxyl radical
Pressure in atmospheres
Propane
Partial pressure of O2 at 50% saturation of hemoglobin
Partial pressure of CO2 in arterial blood, torr
Partial pressure of CO2 in alveolar gas, torr
Polyaromatic hydrocarbons
Peroxyacetyl nitrate
Partial pressure of O2 in arterial blood
Barometric pressure, torr
Potassium cyanide
Partial pressure of CO
Mean partial pressure of pulmonary capillary 02 (torr)
Partial pressure of CO2
Postnatal day
Personal exposure monitors
Partial pressures of carbon monoxide in the fetal placental capillaries
Prostacyclin
Partial pressure of CO in humidified inspired air, ton-
Pituitary
Partial pressures of CO in the maternal placental
Polymorphonuclear neutrophil leukocytes
Partial pressure of O2
Parts per billion by volume
Parts per million by volume (milligrams per liter)
Parts per million by mass (milligrams per kilogram)
Pulmonary resistance
Placental weight
Overall perfusion
March 12, 1990
A-5 DRAFT - DO NOT QUOTE OR CITE
-------
r
rCBF
R
R2
RBC
RV
RV
RVF
s
SAROAD
scf
SCO
SD
SE
SF,
SHAPE
SHED
SI
SIDS
SIPs
SLAMS
S02
SO2
SP
SR
SRMs
ST
STPD
SV
t
t1
td
TEAM
TEM
Tg
TH
THC
t,
TLC
torr
TSP
TTS
TV
TWA
UV
VD
VEP
Correlation coefficient
Regional cerebral blood flow
CO/O2 at 50% inhibition
Coefficient of determination
Red blood cell
Right ventricle
Residual volume
Red visual field
Second
U.S. EPA centralized data base; superceded by AIRS (q.v.)
Standard cubic foot
Percent COHb of total Hb
Standard deviation
Standard error
Sulfur hexafluoride
Simulation of Human Activity and Pollutant Exposure
Sealed housing for evaporative determination
Stroke index
Sudden infant death syndrome
State Implementation Plans
State and Local Air Monitoring Stations
%O2Hb of total Hb
Sulfur dioxide
Mean stroke power
Systemic resistance
Standard Reference Materials
Segment of the EKG (see definition of electrocardiogram)
Standard temperature and pressure, dry
Stroke volume
Time, minute
Postexposure time in minutes
Times-to-death
Total exposure assessment methodology
Transmission electron microscopy
Teragram(s); 1012 grams; 106 metric tons
Total hydrocarbon
Total hydrocarbon content
Times-to-incapacitation
Total lung capacity
A unit of pressure equal to 1/760 of an atmosphere
Total suspended particulates
Temporary threshold shifts
Tidal volume (also VT)
Time-weighted average
Ultraviolet
Physiological deadspace
Visual evoked potential
March 12, 1990
A-6 DRAFT - DO NOT QUOTE OR CITE
-------
VFT
VLDL
" max
VMT
VPD
YT
y
VA
vco
yo2
V02max
w/
WBC
WF
WHW
x
[COHb]
[C0]ave
[OH']ave
2
A
Ventricular fibrillation threshold
Very low-density lipoprotein
Ventricular contractility
Vehicle-miles traveled
Vehicles per day
Tidal volume
Ventilation
Alveolar ventilation, mL min'1 (STPD)
Rate of endogenous production of CO, mL min'1 (STPD)
Minute ventilation; expired volume per minute
Oxygen uptake by the body
Oxygen consumption of tissues or cells (also,
Maximal oxygen uptake
With
White blood cell
White flame (heater)
Wet heart weight
Mean
Concentration of COHb in blood, as milliliters of CO per milliliter of
blood (STPD)
Average concentration of CO
Average concentration of the hydroxyl radical
Sigma (sum of terms)
Angstrom
Micrometer(s)
Definitions
Acclimatization: The physiological and behavioral adjustments of an organism to changes in its
environment.
Adaptation: Changes in an organism's structure or habit that help it adjust to its surroundings.
Additivity: A pharmacologic or toxicologic interaction in which the combined effect of two or
more chemicals is approximately equal to the sum of the effect of each chemical alone.
(Compare with: antagonism, synergism.)
Adiabatic warming: The temperature increase produced in a descending air mass as pressure
increases with decreasing altitude.
March 12, 1990
A-7 DRAFT - DO NOT QUOTE OR CITE
-------
Air Pollutant: Any substance in air that could, if in high enough concentration, harm humans,
other animals, vegetation, or material. Pollutants may include almost any natural or
artificial composition of matter capable of being airborne. They may be in the form of
solid particles, liquid droplets, gases, or in combinations of these forms. Generally, they
fall into two main groups: (1) those emitted directly from identifiable sources and (2) those
produced in the air by interaction between two or more primary pollutants, or by reaction
with normal atmospheric constituents, with or without photoactivation. Exclusive of pollen,
fog, and dust, which are of natural origin, about 100 contaminants have been identified and
fall into the following categories: solids, sulfur compounds, volatile organic chemicals,
nitrogen compounds, oxygen compounds, halogen compounds, radioactive compounds, and
odors.
Air Pollution: The presence of contaminant or pollutant substances in the air that do not disperse
properly and interfere with human health or welfare, or produce other harmful
environmental effects.
Air Pollution Episode: A period of abnormally high concentration of air pollutants, often due
to low winds and temperature inversion, that can cause illness and death.
Air Quality Criteria: The levels of pollution and lengths of exposure above which adverse health
and welfare effects may occur.
Air Quality Standards: The level of pollutants prescribed by regulations that may not be exceeded
during a specified time in a defined area.
Alveolar-arterial oxygen pressure difference [P(A-a)OJ: The difference in partial pressure of
oxygen in the alveolar gas spaces and that in the systemic arterial blood, measured in torr.
Alveolar carbon dioxide pressure (PACO2): Partial pressure of carbon dioxide in the air contained
in the lung alveoli.
Alveolar oxygen partial pressure (Ptf)J: Partial pressure of oxygen in the air contained in the
alveoli of the lungs.
Alveolus: Hexagonal or spherical air cells of the lungs. The majority of alveoli arise from the
alveolar ducts which are lined with the alveoli. An alveolus is an ultimate respiratory unit
where the gas exchange takes place.
Ambient Air: Any unconfined portion of the atmosphere: open air, surrounding air.
Ambient Air Quality Standards: (See: Criteria Pollutants and National Ambient Air Quality
Standards).
Anatomical dead space (VDHWl): Volume of the conducting airways down to the level where,
during air breathing, gas exchange with blood can occur, a region probably situated at the
entrance of the alveolar ducts.
March 12, 1990 A-8 DRAFT - DO NOT QUOTE OR CITE
-------
Antagonism: A pharmacologic or toxicologic interaction in which the combined effect of two
chemicals is less than the sum of the effect of each chemical alone; the chemicals either
interfere with each other's actions, or one interferes with the action of the other. (Compare
with: additivity, synergism).
Arrhythmia: Any variation from the normal rhythm of the heartbeat.
Arterial oxygen saturation (SaOj): Percent saturation of dissolved oxygen in arterial blood.
Arterial partial pressure of carbon dioxide (PaCOj): Partial pressure of dissolved carbon dioxide
in arterial blood.
Arterial partial pressure of oxygen (PaOz): Partial pressure of dissolved oxygen in arterial blood.
Atmosphere [atm\: A standard unit of pressure representing the pressure exerted by a 29.92-in
column of mercury at sea level at 45° latitude and equal to 1000 g/cm2. The whole mass
of air surrounding the Earth, composed largely of oxygen and nitrogen.
ATPS condition (ATPS): Ambient temperature and pressure, saturated with water vapor. These
are the conditions existing in a water spirometer.
BTPS conditions (BTPS): Body temperature, barometric pressure, and saturated with water
vapor. These are the conditions existing in the gas phase of the lungs. For humans the
normal temperature is taken as 37°C, the pressure as the barometric pressure, and the
partial pressure of water vapor as 47 torr.
Carbon dioxide (COJ: A colorless, odorless, non-poisonous gas, which results from fossil fuel
combustion and is normally a part of the ambient air.
Carbon dioxide production (VCOj): Rate of carbon dioxide production by organisms, tissues, or
cells. Common units: mL CO2 (STPD)/kg.min.
Carbon monoxide (CO): An odorless, colorless, toxic gas formed by incomplete combustion,
with a strong affinity for hemoglobin and cytochrome; it reduces oxygen absorption
capacity, transport, and utilization.
Carboxyhemoglobin (COHb): Hemoglobin in which the iron is associated with carbon monoxide.
The affinity of hemoglobin for carbon monoxide is about 240 to 250 times greater than for
oxygen.
Central nervous system (CNS): The portion of the nervous system that includes the brain and
spinal cord, and their connecting nerves.
March 12, 1990 A-9 DRAFT - DO NOT QUOTE OR CITE
-------
Chronic obstructive lung disease (COLD): This term refers to diseases of uncertain etiology
characterized by persistent slowing of airflow during forced expiration. It is recommended
that a more specific term, such as chronic obstructive bronchitis or chronic obstructive
emphysema, be used whenever possible. Synonymous with chronic obstructive pulmonary
disease (COPD).
Combustion: Burning, or rapid oxidation, accompanied by release of energy in the form of heat
and light. A basic cause of air pollution.
Combustion product: Substance produced during the burning or oxidation of a material.
Criteria: Descriptive factors taken into account by EPA in setting standards for various pollutants.
These factors are used to determine limits on allowable concentration levels, and to limit
the number of violations per year. When issued by EPA, the criteria provide guidance to
the states on how to establish their standards.
Criteria pollutants: The 1970 amendments to the Clean Air Act required EPA to set National
Ambient Air Quality Standards for certain pollutants known to be hazardous to human
health. EPA has identified and set standards to protect human health and welfare for six
pollutants: ozone, carbon monoxide, total suspended particulates, sulfur dioxide, lead,
and nitrogen oxide. The term "criteria pollutants" derives from the requirement that EPA
must describe the characteristics and potential health and welfare effects of these pollutants.
It is on the basis of these criteria that standards are set or revised.
Diffusing capacity of the lung (DL, DLO2, DLCO2, DLCO): Amount of gas (O2, CO,
commonly expressed as mL gas (STPD) diffusing between alveolar gas and pulmonary
capillary blood per torr mean gas pressure difference per minute, that is, mL 02/(min-
torr). Synonymous with transfer factor and diffusion factor.
Dose-response relationship: A relationship between (1) the dose, often actually based on
"administered dose" (i.e., exposure) rather than absorbed dose, and (2) the extent of toxic
injury produced by that chemical. Response can be expressed either as the severity of
injury or proportion of exposed subjects affected.
Electrocardiogram (ECG, EKG): A graphic tracing of the variations in electrical potential caused
by the excitation of the heart muscle and detected at the body surface. The normal
electrocardiogram shows deflections resulting from arterial and ventricular activity. The
first deflection, P, is due to excitation of the atria. The QRS deflections are due to
excitation (depolarization) of the ventricles. The T wave is due to recovery of the ventricles
(repolarization). The U wave is a potential undulation of unknown origin immediately
following the T wave, seen in normal electrocardiograms and accentuated in hypokalemia.
Emission: Pollution discharged into the atmosphere from smokestacks, other vents, and surface
areas of commercial or industrial facilities; from residential chimneys; and from motor
vehicle, locomotive, or aircraft exhausts.
March 12, 1990 A-10 DRAFT - DO NOT QUOTE OR CITE
-------
Emission factor: The relationship between the amount of pollution produced and the amount of
raw material processed. For example, an emission factor for a blast furnace making iron
would be the number of pounds of particulates per ton of raw materials.
Emission inventory: A list, by source, of the amount of air pollutants discharged into the
atmosphere of a community. It is used to establish emission standards.
Emission standard: The maximum amount of air polluting discharge legally allowed from a
single source, mobile or stationary.
Environment: The sum of all external conditions affecting the life, development, and survival of
an organism.
EPA: The U.S. Environmental Protection Agency; established in 1970 by Presidential Executive
Order, bringing together parts of various government agencies involved with the control of
pollution.
Episode (pollution): An air pollution incident in a given area caused by a concentration of
atmospheric pollution reacting with meteorological conditions that may result in a significant
increase in illnesses or deaths. Although most commonly used in relation to air pollution,
the term also may be used in connection with other kinds of environmental events such as
a massive water pollution situation.
Exceedance: Violation of environmental protection standards by exceeding allowable limits or
concentration levels.
Exposure: The amount of radiation or pollutant present in an environment that represents a
potential health threat to the living organisms in the environment.
Fetus: The post-embryonic stage of the developing young. In humans, from the end of the
second month of pregnancy up to birth.
Forced expiratory flow (FEFx): Related to some portion of the FVC curve. Modifiers refer to
the amount of the FVC already exhaled when the measurement is made.
Forced expiratory volume (FEV): Denotes the volume of gas that is exhaled in a given time
interval during the execution of a forced vital capacity. Conventionally, the times used
are 0.5, 0.75, or 1 s, symbolized FEV0.3, FEV0.7j, FEV,.0. These values often are expressed
as a percent of the forced vital capacity, e.g. (FEVLO/VC) X 100.
Forced vital capacity (FVC): Vital capacity performed with a maximally forced expiratory effort.
Hematocrit (Hct): The percentage of the volume of red blood cells in whole blood.
March 12, 1990 A-ll DRAFT - DO NOT QUOTE OR CITE
-------
Hemoglobin (Hb): A hemoprotein naturally occurring in most vertebrate blood, consisting of
four polypeptide chains (the globulin) to each of which there is attached a heme group.
The heme is made of four pyrrole rings and a divalent iron (Fe2+-protoporphyrin) which
combines reversibly with molecular oxygen.
Hydrocarbons (HC): Chemical compounds that consist entirely of carbon and hydrogen.
Hypoxemia: A state in which the oxygen pressure and/or concentration in arterial and/or venous
blood is lower than its normal value at sea level. Normal oxygen pressures at sea level are
85 to 100 torr in arterial blood and 37 to 44 torr in mixed venous blood. In adult humans
the normal oxygen concentration is 17 to 23 mL (yiOO mL arterial blood; in mixed venous
blood at rest it is 13 to 18 mL O2/100 mL blood.
Hypoxia: Any state in which the oxygen in the lung, blood, and/or tissues is abnormally low
compared with that of a normal resting human breathing air at sea level. If the Pm is low
in the environment, whether because of decreased barometric pressure or decreased
fractional concentration of oxygen, the condition is termed environmental hypoxia.
Hypoxia when referring to the blood is termed hypoxemia. Tissues are said to be hypoxic
when their fm is low, even if there is no arterial hypoxemia, as in "stagnant hypoxia" which
occurs when the local circulation is low compared to the local metabolism.
In vitro: 1. "In glass"; a test-tube culture. 2. Any laboratory test using living cells taken from
an organism.
In vivo: In the living body of a plant or animal. In vivo tests are those laboratory experiments
carried out on whole animals or human volunteers.
Indoor air: The breathing air inside a habitable structure or conveyance.
Indoor air pollution: Chemical, physical, or biological contaminants in indoor air.
Inversion: An atmospheric condition caused by a layer of warm air preventing the rise of cooling
air trapped beneath it. This prevents the rise of pollutants that might otherwise be dispersed
and can cause an air pollution episode.
Isotope: A variation of an element that has the same atomic number but a different weight because
of its neutrons. Various isotopes of the same element may have different radioactive
behaviors.
Lapse rate: Vertical temperature gradient in the atmosphere; usually negative (i.e., decreasing
with altitude) (see "inversion").
Lowest-observed-adverse-effect level (LOAEL): The lowest dose or exposure level of a chemical
in a study at which there is a statistically or biologically significant increase in the frequency
or severity of an adverse effect in the exposed population as compared with an appropriate,
unexposed control group.
March 12, 1990 A-12 DRAFT - DO NOT QUOTE OR CITE
-------
Lowest-observed effect level (LOEL): In a study, the lowest dose or exposure level at which a
statistically or biologically significant effect is observed in the exposed population compared
with an appropriate unexposed control group.
Lung volume (VJ: Actual volume of the lung, including the volume of the conducting airways.
Maximal aerobic capacity (max VOJ: The rate of oxygen uptake by the body during repetitive
maximal respiratory effort. Synonymous with maximal oxygen consumption.
Methemoglobin (MetHb): Hemoglobin in which iron is in the ferric state. Because the iron is
oxidized, methemoglobin is incapable of oxygen transport. Methemoglobins are formed
by various drugs and occur under pathological conditions. Many methods for hemoglobin
measurements utilize methemoglobin (chlorhemiglobin, cyanhemiglobin).
Minute ventilation (VE): Volume of air breathed in one minute. It is a product of tidal volume
(VT) and breathing frequency (fB). (See ventilation)
Minute volume: Synonymous with minute ventilation.
Modeling: An investigative technique using a mathematical or physical representation of a system
or theory that accounts for all or some its known properties. Models are often used to test
the effect of changes of system components on the overall performance of the system.
Monitoring: Periodic or continuous surveillance or testing to determine the level of compliance
with statutory requirements and/or pollutant levels in various media or in humans, animals,
and other living things.
National Ambient Air Quality Standards (NAAQS): Air quality standards established by EPA
that apply to outside air throughout the country. (See criteria pollutants)
Nitric oxide (NO): A gas formed by combustion under high temperature and high pressure in
an internal combustion engine. It changes into nitrogen dioxide in the ambient air and
contributes to photochemical smog.
Nitrogen dioxide (NOJ: The result of nitric oxide combining with oxygen in the atmosphere.
A major component of photochemical smog.
Nitrogen oxides (NOJ: Compounds of nitrogen and oxygen in ambient air; that is, nitric oxide
(NO) and others with a higher oxidation state of nitrogen, of which nitrogen dioxide is the
most important lexicologically.
No-observed-adverse-effect level (NOAEL): The highest experimental dose at which there is no
statistically or biologically significant increases in frequency or severity of adverse health
effects, as seen in the exposed population compared with an appropriate, unexposed
population. Effects may be produced at this level, but they are not considered to be
adverse.
March 12, 1990 A-13 DRAFT - DO NOT QUOTE OR CITE
-------
No-observed-effect level (NOEL): The highest experimental dose at which there is no statistically
or biologically significant increases in frequency or severity of toxic effects seen in the
exposed compared with an appropriate, unexposed population.
Oxygen consumption (VO2, QOz): Rate of oxygen uptake of organisms, tissues, or cells.
Common units: mL O2 (STPD)/(kg»min) or mL O2 (STPD)/(kg«hr). For whole organisms
the oxygen consumption commonly is expressed per unit surface area or some power of the
body weight. For tissue samples or isolated cells Q02 = fj.L O2/h/mg dry weight.
Oxygen saturation (SOz): The amount of oxygen combined with hemoglobin, expressed as a
percentage of the oxygen capacity of that hemoglobin. In arterial blood, SaO2.
Oxygen uptake (VOj): Amount of oxygen taken up by the body from the environment, by the
blood from the alveolar gas, or by an organ or tissue from the blood. When this amount
of oxygen is expressed per unit of time one deals with an "oxygen uptake rate. " "Oxygen
consumption" refers more specifically to the oxygen uptake rate by all tissues of the body
and is equal to the oxygen uptake rate of the organism only when the oxygen stores are
constant.
Ozone (O3): Found in two layers of the atmosphere, the stratosphere and the troposphere. In
the stratosphere (the atmospheric layer beginning 7 to 10 miles above the Earth's surface)
ozone is a form of oxygen found naturally which provides a protective layer shielding the
earth from ultraviolet radiation's harmful health effects on humans and the environment.
In the troposphere (the layer extending up 7 to 10 miles from the Earth's surface), ozone
is a chemical oxidant and major component of photochemical smog. Ozone can seriously
affect the human respiratory system and is one of the most prevalent and widespread of all
the criteria pollutants for which the Clean Air Act required EPA to set standards. Ozone
in troposphere is produced through complex chemical reactions of nitrogen oxides, which
are among the primary pollutants emitted by combustion sources; hydrocarbons, released
into the atmosphere through the combustion, handling and processing of petroleum
products; and sunlight.
Peroxyacetyl nitrate (PAN): Pollutant created by action of UV component of sunlight on
hydrocarbons and nitrogen oxides in the air; an ingredient of photochemical smog.
pH: A measure of the acidity or alkalinity of a liquid or solid material.
Photochemical smog: Air pollution caused by chemical reactions of various pollutants emitted
from different sources.
Physiological dead space (VD): Calculated volume that accounts for the difference between the
pressures of carbon dioxide in expired and alveolar gas (or arterial blood). Physiological
dead space reflects the combination of anatomical dead space and alveolar dead space, the
volume of the latter increasing with the importance of the nonuniformity of the
ventilation/perfusion ratio in the lung.
March 12, 1990 A-14 DRAFT - DO NOT QUOTE OR CITE
-------
Pollutant: Generally, any substance introduced into the environment that adversely affects the
usefulness of a resource.
Pollution: Generally, the presence of matter or energy whose nature, location or quantity
produces undesired environmental effects. Under the Clean Water Act, for example, the
term is defined as the man-made or man-induced alteration of the physical, biological, and
radiological integrity of water.
Population: A group of interbreeding organisms of the same kind occupying a particular space.
Generically, the number of humans or other living creatures in a designated area.
Respiratory frequency (f,0: The number of breathing cycles per unit of time. Synonymous with
breathing frequency (fB).
Smog: Air pollution associated with oxidants. (See photochemical smog)
Smoke: Particles suspended in air after incomplete combustion of materials.
Sulfur dioxide (SOJ: Colorless gas with pungent odor, primarily released from burning of fossil
fuels, such as coal, containing sulfur.
Synergism: A pharmacologic or toxicologic interaction in which the combined effect of two or
more chemicals is greater than the sum of the effect of each chemical alone. (Compare
with: additivity, antagonism.)
STPD conditions (STPD): Standard temperature and pressure, dry. These are the conditions of
a volume of gas at 0°C, at 760 torr, without water vapor. A STPD volume of a given gas
contains a known number of moles of that gas.
Tidal volume (TV): That volume of air inhaled or exhaled with each breath during quiet
breathing, only used to indicate a subdivision of lung volume. When tidal volume is used
in gas exchange formulations, the symbol VT should be used.
Time-weighted average (TWA): The average concentration to which a worker may be exposed
continuously for 8 h without damage to health.
Torr: A unit of pressure equal to 1,333.22 dynes/cm2 or 1.33322 millibars. The torr is equal
to the pressure required to support a column of mercury 1 mm high when the mercury is
of standard density and subjected to standard acceleration. These standard conditions are
met at 0°C and 45° latitude, where the acceleration of gravity is 980.6 cm/s2. In reading
a mercury barometer at other temperatures and latitudes, corrections, which commonly
exceed 2 torr, must be introduced for these terms and for the thermal expansion of the
measuring scale used. The torr is synonymous with pressure unit mm Hg.
Total lung capacity (TLC): The sum of all volume compartments or the volume of air in the
lungs after maximal inspiration. The method of measurement should be indicated, as with
RV.
March 12, 1990 A-15 DRAFT - DO NOT QUOTE OR CITE
-------
Ventilation: Physiological process by which gas is renewed in the lungs. The word ventilation
sometimes designates ventilatory flow rate (or ventilatory minute volume) which is the
product of the tidal volume by the ventilatory frequency. Conditions usually are indicated
as modifiers; that is,
VE = Expired volume per minute (BTPS),
and
Vj = Inspired volume per minute (BTPS).
Ventilation often is referred to as "total ventilation" to distinguish it from "alveolar
ventilation". (See ventilation, alveolar)
Ventilation, alveolar (VJ: Physiological process by which alveolar gas is removed completely and
replaced with fresh gas. Alveolar ventilation is less than total ventilation because when a
tidal volume of gas leaves the alveolar spaces, the last part does not get expelled from the
body but occupies the dead space, to be reinspired with the next inspiration. Thus the
volume of alveolar gas actually expelled completely is equal to the tidal volume minus the
volume of the dead space. This truly complete expiration volume times the ventilatory
frequency constitutes the alveolar ventilation.
Ventilation, dead-space (VD): Ventilation per minute of the physiologic dead space (wasted
ventilation), BTPS, defined by the following equation:
VD = VE(PaC02 - PEC02)/(PaC02 - P.CO,)
Ventilation/perfusion ratio (VA/Q): Ratio of the alveolar ventilation to the blood perfusion volume
flow through the pulmonary parenchyma. This ratio is a fundamental determinant of the
oxygen and carbon dioxide pressure of the alveolar gas and of the end-capillary blood.
Throughout the lungs the local ventilation/perfusion ratios vary, and consequently the local
alveolar gas and end-capillary blood compositions also vary.
Vital capacity (VC): The maximum volume of air exhaled from the point of maximum
inspiration.
Warbug partition coefficient (K): The carbon monoxide/oxygen ratio that produces 50% inhibition
of the oxygen uptake of the enzyme or, in the case of myoglobin, a 50% decrease in the
number of available oxygen-binding sites.
Wood-burning stove pollution: Air pollution caused by emissions of particulate matter, carbon
monoxide, total suspended particulates, and polycyclic organic matter from wood-burning
stoves.
March 12, 1990 A-16 DRAFT - DO NOT QUOTE OR CITE
-------
References
American College of Chest Physicians - American Thoracic Society (1975). Pulmonary terms
and symbols: a report of the ACCP-ATS Joint Committee on pulmonary nomenclature.
Chest 67: 583-593.
Bartels, H.; Dejours, P.; Kellogg, R. H.; Mead, J. (1973) Glossary on respiration and gas
exchange. Journal Applied Physiology 34: 549-558.
Collier, C. R.; Goldsmith, J. R. (1983) Interactions of carbon monoxide and hemoglobin at high
altitude. Atmospheric Environment 17: 723-728.
U. S. Environmental Protection Agency (1989) Glossary of environmental terms and acronym
list. Washington, DC: Office of Communications and Public Affairs; report no. 19K-
1002.
U. S. Environmental Protection Agency (1989) Glossary of terms related to health, exposure,
and risk management. Research Triangle Park, NC: Air Risk Information Support Center;
report no. EPA/450/3-88/016. Available from: NTIS, Springfield, VA; PB89-
184584/XAB.
ftU.S.GOVERNMENTPRINTINGOFFICE:1990-7'*1'- 15? ° ° " 3 5
March 12, 1990 A-17 DRAFT - DO NOT QUOTE OR CITE
------- |