&EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-452/R-92-004
AUGUST 1992
Air
REVIEW OF THE NATIONAL AMBIENT AIR QUALITY STANDARDS
FOR CARBON MONOXIDE-'"
ASSESSMENT OF SCIENTIFIC AND TECHNICAL INFORMATION
OAQPS STAFF PAPER
-------
The cover illustration is an air quality map of the U.-S. which
displays the highest second maximum nonoverlapping 8-hour average
carbon monoxide concentration by metropolitan statistical area
(MSA) for 1990. This illustration can be found on page 4-16 of
the National Air Quality and Emission Trends Report, 1990 (EPA-
450/4-91-023).
•"- a- -'.•• .s -' •' •• ••.'. • -1 j •
This report has been reviewed by the OfSee of Air Quality
Planning and Standards, tJSEPA, and approved for publication.
Mention of trade names or commercial products is not intended to
constitute endorsement or recommendation for use.
' ••
-------
Preface
This document was finalized by the Office of Air Quality
Planning and Standards (OAQPS) in August 1992 and reviews
information from relevant studies of carbon monoxide (CO) health
effects and exposure analysis. The assessment contained in this
staff paper reflects information in the document "Air Quality
Criteria for Carbon Monoxide" (EPA/600/8-90/045F).
-------
11
TABLE OF CONTENTS
' Page
I. Purpose 1
II. Background 1
A. Legislative Requirements 1
B. Establishment of Carbon Monoxide Standards 2
C. First Review of Carbon Monoxide Standards 3
D. Current Review of Carbon Monoxide Standards 5
III. Approach 6
IV. Air Quality Information 7
A. Introduction 7
B. Nonattainment Status . 7
C. Air Quality Trends 8
V. Health Effects of Carbon Monoxide 9
A. Introduction 9
B. Mechanisms of Toxicity. 9
C. Measurement and Estimation of Carboxyhemoglobin
Levels . , 10
1. Measurement of Carboxyhemoglobin Levels 10
2. Estimation of Carboxyhemoglobin Levels Using
the Coburn, Forster, Kane Equation., 11
D. Health-Effects and Effect Levels of Carbon Monoxide..14
1. Effects on Persons with Cardiovascular Disease...15
2. Effects on Exercise Capacity and Oxygen Uptake...22
3 . Central Nervous System Effects 23
4. Developmental Toxicity Effects ; 25
5. Environmental Factors, Drugs, Other Pollutants...26
E. Populations Potentially at Risk To Carbon Monoxide...26
F. Carboxyhemoglobin Levels of Concern 27
-------
Ill
VI. Averaging Times .;... 33
VII. Exposure Analysis .... 34
* • '
A. Overview . .34
B. Estimating COHb Levels in Exposed Populations..'... 1. .35
C, Results ........; 36
VIII. Summary of Staff Conclusions and Recommendations. 43
References. 45
Appendix A. CASAC Closure Letter on the Criteria Document .A-l
Appendix B. Overview of Exposure Analysis Model B-l
Appendix C. Estimating COHb Levels In Exposed Populations .C-l
Appendix D. CASAC Closure Letter on the Staff Paper and
Response from Administrator to CASAC Chairman D-l
-------
IV
TABLES
Number
Page
V-l Predicted COHb Response to Exposure to Constant
CO Concentrations « 13
V-2 Key Health Studies for Establishing NAAQS for
Carbon Monoxide 16
V-3 Summary of Subpopulations at Risk 28
VII-1 Percent of Heart Disease Population That Expei-iences
One or More 1-Hr and 8-Hr Daily Maximum Exposures
> Concentration Level Shown for Four Alternative
Scenarios - • •»
VII-2 Heart Disease Person-Days With at Least One Hourly
COHb Estimate > Value Shown for Four Alternate
Scenarios • • « «• •
VII-3 Heart Disease Person-Hours With a 1-Hr COHb Estimate
> Value Shown for Four Alternative Scenarios
C-l Summary of Important Variables Used in COHb Module
of pNEM/CO C-2
FIGURE
Number • Page
V-l The Effect of CO Exposure on Time To Onset of
Angina • 21
-------
Acknowledgments
This Staff Paper is the product of the OAQPS. The principal
authors include Dr. David J. McKee (Carbon Monoxide NAAQS Review
Program Manager) , Thomas R. McCurdy, and Harvey M. Richmond.
This document reflects many helpful comments within the
Environmental Protection Agency (EPA) from individuals in OAQPS
(Michael H. Jones, John H. Haines, Dr. Thomas C. Curran, Barbara
A. Beard), the Environmental Criteria and Assessment Office
(ECAO) (James A. Raub), the Atmospheric Research and Exposure
Assessment Laboratory (Gerald G. Akland), the Office of General
Counsel (Gerald K. Gleason), the Office of Policy, Planning, and
Evaluation (John C. Chamber1in), and Region VIII (Dr. Suzanne M.
Wuerthele). Comments of the Clean Air Scientific Advisory
Committee (CASAC) made at meetings held on April 3.0, 1991 to
review the external review draft of the Air Quality Criteria for
Carbon Monoxide and on March 5, 1992 to review the external
review draft of this OAQPS Staff Paper also have been carefully
considered in revising this draft. A copy of the CASAC closure
letter for the criteria document is included as Appendix A of
this Staff Paper. Finally, the clerical and support services
provided by Patricia R. Crabtree and Barbara Miles are
acknowledged-and greatly appreciated.
-------
-------
I.
REVIEW OF THE NATIONAL AMBIENT AIR QUALITY STANDARDS FOR
CARBON MONOXIDE
1992 ASSESSMENT OF SCIENTIFIC AND TECHNICAL INFORMATION
PURPOSE
The purpose of this Office of Air Quality Planning and
Standards (OAQPS) Staff Paper is to evaluate the key studies and
scientific information contained in the revised EPA document, Air
Quality Criteria for Carbon Monoxide (USEPA, 1991; henceforth
referred to as CD) and to identify the critical elements that the
EPA staff believes should be considered in the review of the
national ambient air quality standards (NAAQS) for carbon
monoxide (CO). Factors relevant to the evaluation of the current
primary NAAQS, as well as staff conclusions and recommendations,
are provided in this Staff Paper.
II. BACKGROUND
A. Legislative Requirements
Since 1970 the Clean Air Act (Act) has provided authority
and guidance for the listing of certain ambient air pollutants
which may endanger public health and/or welfare and for revising,
as necessary, the NAAQS for those pollutants. Primary standards
must be based on health effects criteria and must provide an
adequate margin of safety to ensure protection of public health.
Judicial decisions have provided clear guidance that economic and
technological feasibility are not to be considered in setting the
primary NAAQS, although these factors may be considered in the
development of State implementation plans to implement NAAQS
(Lead Industries Association v. EPA, 1980; American Petroleum
Institute v. EPA, 1981) . Guidance was also provided in the
legislative history of the Act that the standards should be set
-------
at "the maximum permissible ambient air level . . . which will
protect the health of any (sensitive) group of the population,"
and further that margins of safety are to be included such that
NAAQS provide "a reasonable degree of^protection . . . against
.I'vi.f,,,1-;. • - , • i ^
hazards which research^has not yet identified" (U.S. Senate,
1974). In weighing risks for margin of safety purposes, EPA
takes into account factors such as size of sensitive
population(s), nature and severity of all health effects
involved, and kind and degree of other uncertainties. It is,
however, a policy choice left specifically to the EPA
Administrator to select a particular approach for setting a
primary NAAQS which provides an adequate margin of safety (Lead
Industries Association v. EPA, 1980).
The Act, as amended in 1977 and 1991, requires periodic
review of criteria and, as appropriate, revision of existing
NAAQS. Thus, if the Administrator determines that review of
criteria makes proposal of new or revised NAAQS necessary, such
standards are to be revised and promulgated in accordance with
section 109 (b) of the Act. However, the Administrator may find
that revision of NAAQS is not necessary and may conclude review
by reaffirmation.
B. Establishment of Carbon Monoxide Standards
On April 30, 1971, the EPA promulgated NAAQS for CO under
section 109 of the Act (36 FR 8186). Identical primary and
secondary NAAQS were set at 9 ppm as an 8-hr average and 35 ppm
as a 1-hr average, neither to be exceeded more than once per
year. Scientific and technical bases for these NAAQS are
provided in the document, Air Quality Criteria Document for
Carbon Monoxide (U.S. Dept. of Health, Education and Welfare,
1970). The NAAQS promulgated in 1971 were based largely upon
research by Beard and Wertheim (1967) who reported that CO
exposures which produced carboxyhemoglobin (COHb) levels of 2 to
3% were associated with central nervous system (CNS) effects such
as impaired ability to discriminate time intervals.
-------
C. First Review of Carbon Monoxide Standards
A revised Air Quality Criteria Document for Carbon Monoxide
(USEPA, 19793), prepared by the ECAO, and a Staff Paper (USEPA,
1979b), prepared by OAQPS, identified several major factors
pertinent to subsequent action taken on the NAAQS for CO. The
CASAC met on June 14-15, 1979, to review drafts of these
documents and provide advice on the CO standards. As discussed
in a notice of proposed rulemaking (45 FR 55066) published on
August 18, 1980, although the Beard and Wertheim (1967) study no
longer could serve as a basis for the CO NAAQS, other studies
available in 1980 provided alternative evidence of decreased time
to onset of angina attack at COHb levels as low as 2.7 to 3.0%.
This and other scientific evidence served as a basis for EPA to
propose: (1) retaining the 8-hour primary standard level of 9
ppm, (2) revising the 1-hr primary standard level from 35 ppm to
25 ppm, (3) revoking the existing secondary CO NAAQS due to a
lack of evidence of adverse welfare effects at or near ambient CO
levels, (4) changing the form of the standard from deterministic
to statistical by stating allowable exceedances as .expected
values rather than as explicit values, (5) adopting a daily
interpretation for exceedances of the CO NAAQS so exceedances
would be determined on the basis of days on which the 8- or 1-hr
average concentrations were above the standard levels.
On June 18, 1982, EPA announced (47 FR 26407) that a second
public comment period was necessary to open discussion on several
important issues and additional analyses. These issues included:
(1) the role of the Aronow (1981) study in assessing CO effects,
(2) consideration of a multiple exceedance 8-hr standard for CO,
(3) technical adequacy of the revised draft sensitivity analysis
(Biller and Richmond, 1982) on the Coburn, Forster, and Kane
model predictions of COHb levels, and (4) technical adequacy of
the revised exposure analysis (Johnson and Paul, 1982). The
CASAC met on July 6, 1982 to discuss these issues and provide
advice, a summary of which was sent to the Administrator on
August 31, 1982 (Friedlander, 1982).
-------
The 1980 proposal (45 FR 55066) was based in large part on
studies by Dr. Wilbert Aronow (Aronow, et al., 1972, 1973, 1974,
1975, 1977, 1978), which provided CASAC and EPA staff with, a
basis for concluding that COHb levels of 2.7-3.0% posed ashealth
risk of concern in individuals.with, angina and other types of
cardiovascular disease. A subsequent disclosure in March 1983 by
the Food and Drug Administration (FDA) concerning work conducted
for FDA by Dr. Aronow caused the EPA to question the scientific
credibility of Dr. Aronow's research on CO. -As a result, EPA
decided it would be prudent to conduct an independent review of
his CO research prior to making a decision on the CO standards.
A committee of experts was convened and chaired by Dr. Steven
Horvath (University of California, Santa Barbara). Following
meetings with Dr. Aronow and examination of limited data and
records available from his CO studies, the committee concluded in
its report (Horvath et al., 1983) that the EPA should not rely on
Dr. Aronow's studies for a decision on the level of the CO
standards due to problems regarding data collection and analysis.
As a result of this finding, the ECAO prepared a draft
Addendum to the 1979 Air Quality Criteria for Carbon Monoxide
(U.S. EPA, 1984a), hereafter referred to as Addendum. OAQPS
prepared a draft Review of the NAAQS for Carbon Monoxide:
Reassessment of Scientific and Technical Information (U.S. EPA,
1984b), hereafter referred to as Staff Reassessment. These
documents were prepared to reevaluate the scientific and
technical evidence on health effects of CO at or near ambient
levels in consideration of the reduced usefulness of the Aronow
studies. Both documents were reviewed by the CASAC at a public
meeting on September 25, 1983. CASAC sent a closure letter to
the Administrator on May 17, 1984, which concluded that the
Addendum and the Staff Reassessment represented a scientifically
balanced and defensible summary of health effects literature for
CO. On August 9, 1984, the, EPA announced (49 FR 31923)
availability of the final Addendum (I984a) and final Staff
Reassessment (1984b), both of which were revised to reflect CASAC
-------
and public comment. In the same notice, the EPA reviewed the
basis for proposal to revise the CO standards and solicited
additional public comment. In a subsequent notice (50 FR 37484)
on September 13, 1985, the EPA announced its final decision not
to revise the existing primary (health based) standard and to
revoke the secondary (welfare based) standard for CO. In doing
so the Administrator determined that the existing 1-hr and 8-hr
primary NAAQS provided adequate protection from exposure to
ambient CO.
D. Current Review of Carbon Monoxide Standards
In 1987, the ECAO initiated action on a revised CD, which
was released for public review in April 1990. The revised CD
included discussion of several new studies of effects of CO on
angina patients which had been initiated in light of the
controversy discussed above. The CASAC reviewed the CD at a
meeting on April 30, 1991 and concluded in a letter to the.
Administrator that the document ". . . provides a scientifically
balanced and defensible summary of current knowledge of the
effects of this pollutant and provides an adequate basis for the
EPA to make a decision as to the appropriate primary NAAQS for
CO." (See Appendix A.)
-------
-------
III. APPROACH
The approach taken in this Staff Paper is to evaluate arid
integrate information provided in the 1991 revised CD and to
incorporate consideration of an exposure analysis used to
evaluate the adequacy of the current CO primary NAAQS. Due to
the EPA decision (50 FR 37484) to revoke the secondary NAAQS and
the lack of any substantial new information suggesting the need
for a secondary NAAQS for CO to protect public welfare from any
known or anticipated adverse welfare effects, only health
information and analysis related to evaluation of the primary CO
NAAQS are reviewed here.
Critical elements have been identified which the staff
believes should be considered in review of the primary CO NAAQS.
Particular attention is drawn to those judgments that must be
based on careful interpretation of incomplete or uncertain
evidence. In such instances, the Staff Paper provides the
staff's evaluation of evidence as it relates to a specific
judgment, sets forth alternatives that the staff believes should
be considered, and recommends a course of action. After a short
discussion of air quality information in Section IV, Section V of
this Staff Paper provides:
(a) description of probable mechanisms of CO toxicity,
(b) discussion of measurement and estimation of COHb,
(c) evaluation of effects of concern and effect levels,
(d) identification of populations at risk to ambient CO,
and,
(e) discussion of COHb levels of concern.
Section VI is a rationale for selection of CO NAAQS
averaging times; and Section VII is a discussion of exposure
analysis and resulting estimates of exposure occurrences and
number of people exposed to the current 1-hr and alternative 8-hr
CO levels and various COHb levels upon attainment of the current
CO NAAQS. Finally, drawing on information in Sections V, VI, and
VII, Section VIII presents staff conclusions and recommendations.
-------
-------
IV. AIR QUALITY INFORMATION
A. Introduction
Carbon monoxide is a colorless, odorless gas which can be
emitted into ambient air as a result of both natural processes
and human activity. Although CO exists as a trace constituent of
the troposphere, much of human CO exposure which results in
elevated 'levels of COHb in the blood is due to incomplete fossil
fuel combustion. In 1990, transportation sources of CO accounted
for 63% of total ambient CO emissions in the U.S., while other
combustion processes (e.g., steam boilers, industrial processes,
solid waste disposal) were responsible for most of the rest (CD,
p. 1-4). AS discussed in Section VII, other significant sources
of CO exposure include combustion sources and gas stoves in
homes, and passive smoking in many microenvironments. See
Chapter 6 of the CD for an extended discussion of CO emission
sources, air quality patterns, and CO diffusion models.
B. Nonattainment Status
On November 6, 1991, the EPA published a list of areas that
legally do not attain the current 8-hr CO NAAQS (56 FR 56694).
There are 42 areas on the list, most of which are metropolitan
statistical areas. For the most part, nonattainment designations
are based upon 1988-1989 air quality data.
Almost 55 million people live in areas that violate the
current 8-hr CO NAAQS (AQMD, 1991). It should not be inferred,
however, that everyone in these areas is exposed to the
concentration values recorded by ambient monitors used to
determine whether or not an area violates a CO NAAQS. Actual
exposures experienced by residents of a non-attainment area may
be lower or higher than the ambient levels recorded at a fixed-
site monitor (CD, Chapter 8). For this reason, EPA undertakes an
exposure analysis to better estimate total human exposures to CO
as people go about their daily lives. One such analysis for
Denver, Colorado is described in Chapter VII.
-------
There are very few violations of the 1-hir CO NAAQS of 35
ppm, which is determined by evaluating the second-highest 1-hr
value in a year. In fact, in the 1988-1989 time period, only two
metropolitan statistical areas (MSA's) recorded a violation of
the 1-hr NAAQS: Denver, CO and Steubenville, OH. There were
fewer than 20 such violations recorded in all site-years of data
for all MSA's contained in EPA's Aerometric Information Retrieval
Service (AIRS) data base for the 1980 through 1990 time period
(AIRS, 1991). In all areas that exceed both.the 1-hr and the 8-
hr NAAQS, the 8-hr exceedance(s) is(are) proportionally further
away from the CO standard than are the 1-hr exceedance(s). Thus,
exceedances of the 8-hr CO standard are more significant in
of control program development than exceedances of the 1-hr CO
standard.
C. Air Quality Trends
The EPA analyzes long-term trends of NAAQS pollutants in the
Technical Services Division's (TSD) annual Trends Report; the
most recent one includes 1990 data (TSD, 1991). During the 1980-
1990 time period, the composite national average of the second-
highest non-overlapping 8-hr CO average concentration (often
known as the "design value") decreased by approximately 30%. The
median rate of improvement was about 3% per year. The downward
trend in the annual number of exceedances of the 8-hr NAAQS is
even more impressive during the same time period. The composite
average of the estimated number of annual exceedances (often
known as "expected exceedances11) .decreased about 86% between 1980
and 1990. The Trends Report (TSD, 1991) also provides
information regarding trends in CO emissions and probable reasons
for the decreasing ambient CO levels monitored around the
country; the interested reader should review that report for
additional information.
-------
V. HEALTH EFFECTS OF CARBON MONOXIDE
A. Introduction
This section draws on elements which are critical in
reviewing primary NAAQS for CO. Identification of the principle
mechanism of toxicity is important to establish a linkage between
CO exposure and resultant health effects. Methodologies used to
measure biomarkers of CO exposure are evaluated based on known
limitations discussed in the CD (Chapter 8). Health effects of
concern are presented in a manner which discusses the adverse
nature of effects and estimates a lowest effects level for
purposes of standard setting. Populations at risk to CO exposure
also are identified. Finally, staff recommendations are made
regarding the range of COHb levels that should be considered in
setting CO standards that protect public health with an adequate
margin of safety.
B. Mechanisms ofToxicity
The mechanism of toxicity principally associated with health
effects of greatest concern from CO exposure is hypoxia induced
by elevated COHb levels. The primary exchange route for CO to
human tissues is through the lungs. Although CO is a naturally
occurring chemical in blood being produced endogenously by normal
catabolic processes, blood COHb levels do not often exceed 0.5 to
0.7% in normal individuals unless exogenous CO is breathed. Some
individuals with high endogenous CO production can have COHb
levels of 1.0 to 1.5% (e.g. anemics). Exogenous CO diffuses
through the respiratory zone (alveoli) to the blood where it
binds to hemoglobin (Hb) to form COHb. The chemical affinity of
CO for Hb is 218 to 250 times greater than that of oxygen (O2)
(Roughton, 1970; Wyman et al., 1982; Rodkey et al., 1969). This
preferential binding of CO to Hb limits the availability of Hb
for O2 transport to tissues throughout the body. As COHb levels
increase, the dissociation curve for normal human blood is
shifted to the left resulting in more reduced delivery of O2 to
-------
10
tissues and a greater degree of CO-induced hypoxia. It is this
reduced O2 delivery to heart muscle tissue which is-of great
concern for individuals with ischemic heart disease because their
already compromised condition puts them at increased ^cisk.
Although several other mechanisms of toxicity are discussed
in the CD, these are not considered to be as well understood as
COHb hypoxia. Intracellular effects of CO (CD, pp.. 9-22 to 9-31)
have been associated with CO toxicity. Preferential bindina of
;"". s-_e '^\./s~-^ *\-s~ • • , • - ' ^
CO £6 myoglobin, cytochrome P-450, and cytochrome c oxidase has
been-studied and could lead to impairment of intracellular oxygen
transport to mitochondria. However, mechanisms of toxicity
associated with CO-induced inhibition of these hemoproteins at
relevant CO levels are not well understood at this time and will
require additional research.
Based on the review and conclusions drawn in the CD, COHb
levels provide the most useful estimate of exogenous CO exposures
and serve as the best biomarker of CO toxicity for ambient-level
exposures to CO. Thus, COHb levels are used in this Staff Paper
as the measure of health effects and to identify the lowest
effects level for CO.
C. Measurement and Estimation of Carboxvhemoalobin Levels
1. Measurement of Carboxvhemoqlobin Levels
As noted in the section above, the best indicator of CO
exposure to relate to health effects of concern is blood COHb
levels. Since the last CO NAAQS review, significant concerns
have been raised about the accuracy of CO-Oximeters (CO-Ox),
which are the instruments most commonly used to measure COHb in
health effects studies, at levels in the range of o to 5% COHb.
While CO-Ox measurements have been shown to be very precise
(i.e. replicable), research has shown that the accuracy (i.e.,
ability to detect the true level) of these optical instruments is
not always sufficient to use alone at levels below 5% COHb
(Allred et al., 1989a,b, 1991). As indicated in the CD (pp. 8-72
to 8-73), the results from linear regression analyses of
-------
11
comparisons between Ik 282 CO-Ox instruments and various
reference instruments (involving gas chromatography, GC) show a
fairly linear slope (range 0.895 to 1.122) and a wide range of
intercept values (range -1.279 to +1.17). In the only health
effect study that used both CO-Ox and GC methods to measure COHb
levels in subjects with cardiovascular disease, researchers also
found the spread of COHb values to be much greater for the CO-Ox
values than for the GC values (Allred et al., 1989a,b, 1991).
In order for optical instruments such as CO-Ox to be used to
measure COHb levels accurately at low levels they must be
calibrated routinely with an alternative method (CD, p. 8-64).
When properly calibrated, CO-Ox instruments can provide useful
information on mean values. However, variation in individuals'
oxyhemoglobin (O2Hb) levels appears to influence COHb readings
(Allred et al., 1989a,b) and, as noted above, CO-Ox instruments
also give too broad a range of COHb values when compared to GC
measurements on the same samples (Allred et al., 1989a,b, 1991).
In the section assessing health effects and effect levels
associated with CO exposure, the measurement method used to
obtain COHb values will be indicated in parentheses (GC or CO-
Ox) . The fact that most of the health effects literature for CO
has relied on CO-Ox measurements of COHb levels introduces
additional uncertainty as one attempts to assess lowest effects
levels and dose-response relationships.
2. Estimation of Carboxvhemoglobin Levels Using the
Coburn. Forster. Kane Equation
In order to set ambient CO standards based on an assessment
of health effects at various COHb levels, it is necessary to
estimate the ambient CO concentrations that are likely to result
in COHb levels of concern. The CD (p. 9-21) concludes that the
best all around model for predicting COHb levels is the Coburn,
Forster, Kane (CFK) differential equation (Coburn et al., 1965).
The CFK model estimates COHb levels resulting from exposure to CO
concentrations as a function of time and various physiological
-------
12
and environmental factors (e.g., blood volume, endogenous CO
production rate, ventilation rate, altitude).
Over the last 20 years, modelers have developed and
evaluated both linear and nonlinear solutions to the CFK model.
The linear CFK model assumes that O2Hb is constant and does not
vary with COHb level. The nonlinear CFK model incorporates the
interdependence between O2Hb and COHb. At COHb levels below 6%
both approaches give estimates that are within 0.5% COHb (Smith,
1990). While the linear CFK model is easier to solve and gives
approximately the same COHb estimate, in the range of interest
(i.e., 1 to 5% COHb), we have chosen to use the nonlinear
solution in the remainder of this Staff Paper because it is more
accurate physiologically. With the assumption of a linear
relationship between O2Hb and COHb, there is an analytical
solution to the nonlinear CFK equation (Muller and Barton, 1987),
Table V-l presents baseline estimates (i.e., typical
physiological parameters are used) of COHb levels expected to.be
reached by nonsrookers exposed to various constant concentrations;
of CO for either 1 or 8 hrs based on the CFK model. (Smokers are
not included because they have voluntarily exposed themselves to
high CO levels.)* There are, however, two major uncertainties
involved in estimating COHb levels resulting from exposure to CO
concentrations. First, among the population with cardiovascular-
disease, or any other group of interest, there is a distribution
for each of the physiological parameters used in the CFK model.
Past work (Biller and Richmond, 1982) has shown that these
variations are sufficient to produce noticeable deviations from
the COHb levels in Table V-l. Second, predictions based on
exposure to coristant CO concentrations can underestimate the
* 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 cigarrette, producing an average COHb of 4%, with a
typical range of 3 to 8%. Heavy smokers can have COHb levels as
high as 15 %. (CD, p. 11-39)
-------
13
Table V-l. Predicted COHb Response to Exposure to
Constant CO Concentrations
Percent COHb Based on Nonlinear CFK Model1
1-hour Exposure 8-hour Exposure
CO
(ppm)
7.0
9.0
12.0
15.0
20.0
25.0
35.0
50. 0
Intermittent
• Rest/Light
Activity
0.7
0.7
0.8
0.9
1.1
1.3
1.6
2.1
Moderate
Activity
0.7
0.8
1.0
1.1
1.4
1.7
2.2
3.0
Intermittent
Rest/Light
Activity
1.1
a. 4
1.7
2.1
2.7
3.4
4.6
6.4
Moderate
Activity
1.1
1.4
1.8
2.2
2.9
3.6
5.0
7.0
ssumed parameters: alveolar ventilation rates = 10 STPD liters/min
ntermittent rest/light activity) and 20 STPD liters/min (moderate
itivity); hemoglobin = 15 g/100 ml (normal male); altitude = sea level;
titial COHb =0.5 percent; endogenous CO production rate •= 0.00 STPD ml/min;
.ood volume = 5500 ml; Haldane coefficient (measure of affinity.of
imoglobin for CO) =218; lung diffusivity for CO = 35.9 and 44.3 STPD
./min/loss at 10 and 20 STPD liters/min alveolar ventilation rates
>spectively. '
-------
14
response of individuals exposed to widely fluctuating CO levels
that typically occur in the ambient environment (Biller and
Richmond, 1982).
• Section VII and Appendix C discuss the use of the nonlinear
CFK model to estimate the distribution of COHb levels in persons
with ischemic heart disease in Denver, Colorado under "as is"
conditions and where the current 9 ppm, 8-hour CO NAAQS is just
attained. A more detailed discussion of the EPA's application of
the CFK model can be found in Biller and Richmond (1992).
D- Health Effects and Effect Levels of CO
Health effects information which is pertinent to review of
the NAAQS for CO has been thoroughly reviewed in Chapters 10 and
11 of the CD. The CD has reviewed hundreds of health effects
studies, some of which were conducted at extremely high levels of
CO (i.e. much higher than typically found in ambient air).
However, the focus of this Staff Paper is only on those key
health studies, which generally were conducted with human
subjects at COHb levels that are most relevant to regulatory
decision making.
Health effects associated with exposure to CO include
cardiovascular system, central nervous system (CNS), and
developmental toxicity effects, as well as effects of combined
exposure to CO and other pollutants, drugs, and environmental
factors. Cardiovascular effects of CO are directly related to
reduced O2 content of blood caused by combination of CO with Hb
to form COHb and resulting in tissue hypoxia. Most healthy
individuals have mechanisms (e.g. increased blood flow, blood
vessel dilation) which compensate for this reduction in tissue O2
levels, although the effect of reduced maximal exercise capacity
has been reported in healthy persons even at low COHb levels.
Several medical conditions such as occlusive vascular disease
chronic obstructive lung disease, and anemia can increase
susceptibility to potential adverse effects of CO during
exercise.
-------
15
The best documented cardiovascular effects in patients with
angina pectoris are decreased time to onset of chest pain and ST
segment depression during exercise stress. The commonly accepted
criterion of exercise-induced myocardialr ischemia is 1 mm or
greater ST segment depression. The ST segment is a portion of
the electrocardiogram (ECG), depression of which is an indication
of insufficient O0 supply to heart muscle tissue. Other
£t - - .'; ' '
cardiovascular effects of CO such as increased blood flow,
arrhythmogenic effects, or effects on individuals with chronic
anemia or chronic obstructive lung disease either have not been
studied adequately or do not appear to pose a significant health
threat.
Effects of CO on the CNS involve both behavioral and
physiological changes. These include modification of visual
perception, hearing, motor and sensorimotor performance,
vigilance, and cognitive ability. Developmental toxicity effects
of CO, though not well studied in humans, do pose a potentially
serious threat to the fetus. Finally, environmental factors
(e.g. altitude), drug interaction, and pollutant interaction also
can play a role in the public health impact of ambient CO
exposure.
Table V-2 is a summary of key health studies which have been
identified by staff as being most pertinent to a regulatory
decision on the NAAQS for CO and is not intended to be
comprehensive. Each of the key studies is considered in light of
limitations discussed in the CD and this Staff Paper. Due to the
fact that most of the studies used optical methods, COHb levels
are presented as optical measurements.
1. Effects on Persons with.Cardiovascular Disease
Five key studies on cardiovascular effects of CO (Allred et
al., 1989a,b, 1991; Kleinman et al., 1989; Adams et al., 1988;
Sheps et al., 1987; Anderson et al., 1973) have provided evidence
of the potential for CO to enhance development of exercise-
induced myocardial ischemia in patients who suffer from angina
pectoris. (Angina pectoris is a disease marked by brief,
-------
16
Table V-2
KEY HEALTH STUDIES FOR ESTABLISHING NAAQS
FOR CARBON MONOXIDE
COHb
Concentration
Percent3
Health Effects
Reference*
2.3-7.0
2.9-5.9
5.0-20.0
5.0-20.0
Decreased short-term
maximal exercise duration
in young healthy men
Decreased exercise dura-
tion due to increased
chest pain (angina) in
patients with ischemic
heart disease
Decreased maximal oxygen
consumption with short-
term strenuous exercise
in young healthy men
Equivocal effects on
visual perception, audi-
tion, motor and sensori-
motor performance, vigi-
lance, and other measures
of neurobehavioral per-
formance
Drinkwater et .al. . (19 7 4)
Ekblom and Huot (1972)
Horyath et al. (1975)
Raven et al. (1974a,b)
Weiser et al. (1978)
Adams et al. (1988)
Allred et al. (1989a,b,
1991)
Anderson et al. (1973)
Kleinman et al. (1989)
Sheps et al. (1987)
Ekblom and Huot (1972)
Klein et al. (1980)
Pirnay et al. (1971)
Stewart et al. (1978)
Vogel and Gleser (1972)
Weiser et al. (1978)
Benignus et al. (1977,
1987, 1990a,b)
Bunnell and Horvath (1988)
Christensen et al. (1977)
Gliner et al. (1983)
Harbin et al. (1988)
Hudnell and Benignus (1989)
McFarland (1970, 1973)
McFarland et al. (1944)
Mihevic et al. (1983)
O'Donnell et al. (1971)
Putz et al. (1976, 1979)
Roche et al. (1981)
Rummo and Sarlanis (1974)
Seppannen et al. (1977)
Von Post-Lingen (1964)
Winneke (1974)
•aBlood COHb levels determined by optical methods.
•"References also found in U.S. Environmental Protection Agency (1991).
-------
,17
recurrent attacks often precipitated by deficient oxygenation of
heart muscle tissue.) As was noted in Section II.C (pp. 3-5) of
this Staff Paper, earlier angina' studies conducted by Dr. Wilbert
Aronow reporting decreased time to onset of exercise- induced
angina for COHb levels in the range of 2-3% (CO-Ox) have been
called into question. On.ly one other early study by Anderson et
al. (1973) reported decreased time to onset of angina pain for
COHb levels as low as 2.9 (CO-Ox), representing a 1.6% increase
in average COHb levels over baseline. Details of this study were
reported at length in the Addendum (U.S. EPA, 1984a).
Recent controlled exposure studies of angina patients have
provided substantial new evidence of decreased exercise duration
due to early onset of chest pain. (See discussion in CD, pp. 10-
21 to 10-35). A study which provides strong evidence of the
health effects of CO is the multicenter study of Allred et al.
(1989a,b, 1991). There are several reasons why this study is
important to the CO NAAQS review: 1) dose-response relationships
are shown; 2) information on ST-segment depression of subjects is
available; 3) both GC and CO-Ox measurements were taken; 4) a
large number of subjects was used; and 5) it was conducted at
multiple laboratories around the U.S. This study involved 63
males (ages 41-75) with coronary artery disease living in three
different U.Si cities. The objective was to assess the impact of
exposure to CO on time to onset of significant ischemia during a
standard treadmill test. Unusual care was taken to establish
presence of coronary artery disease in all subjects prior to
testing. The protocol for the study was quite similar to that
used in the Aronow studies; i.e. two exercise tests were
performed on the same day separated by a recovery period and a
double-blind exposure period. Subjects were exposed to either
clean air, 117 ppm CO, or 253 ppm CO for 50 to 70 minutes while
performing symptom-limited exercise on a treadmill. Time to
onset of angina and time to ST segment depression were determined
for each test following exposure to both CO levels and compared
to clean air (< 2 ppm CO) exposure. After exposure to 117 ppm
-------
18
and 253 ppm CO, COHb levels measured before the exercise stress
test were 2.4 and 4.7% COHb (GC) and 3.2 and 5.6% COHb (CO-Ox),
respectively* After the stress test COHb levels were 2.q;and
3.9% (GC) and 2.7 and 4.7% ^(CO-Ox). Using the objective^ measure
of time to ST-segment depression, CO exposure which produced 3.2%
COHb (CO-Ox, pre-test) resulted in a 5.1% (p = o.Ol) decrease in
time to the ST criterion, and 5.6% COHb (CO-Ox, pre-test)
decreased time to the ST criterion by 12.1% (p<0.000l) relative
to clean air exposure. Combining slopes for the 62 individuals
yielded a significant (p<0.005) regression which indicates that
there was a 3.9% decrease in time to ST criterion for every 1%
increase in COHb. Time to onset of angina, signalled by chest
pain, also was reduced in the same subjects, and regression
analysis yielded a significant relationship (p<0.025). Both
endpoints (time to angina and time to ST change) were highly
correlated.
In another study (Sheps et al., 1987), 30 non-smokers with
ischemic heart disease (ages 38-75) were exercised during
exposure to 100 ppm CO or air using a 3-day, randomized double
blind protocol. Following CO exposure average COHb levels were
4.1% (CO-Ox), representing a 2.2% COHb increase from the initial
COHb level. (In this and other studies using the CO-Ox, the
initial high COHb readings may have been due in part to the
inaccuracies of the CO-Ox at very low COHb levels.) In comparing
results of air-exposed subjects to CO-exposed subjects as a
group, no statistically significant differences were reported in
time to onset of angina, maximal exercise time, maximal ST-
segment depression, or time to significant ST-segment depression.
Although the authors concluded that 4.1% COHb did not produce
clinically significant effects in the subject group, 3 of 30
patients did experience angina on CO-exposure days but not on
air-exposure days. Further analysis of time to onset of angina
that includes these three patients indicated a statistically
significant decrease for CO exposure compared to air exposure
(Bissette et al., 1986). The same group of researchers (Adams et
-------
19
al., 1988) exposed 30 subjects with obstructive coronary artery
disease to either air or sufficient CO to reach COHb levels of
5.9% (CO-Ox), representing an average increase of 4.2% COHb above
initial COHb levels. Results"of this study provide evidence that
exposure to CO induces earlier onset of angina and ventricular
dysfunction as well as poorer exercise performance in patients
with ischemic heart disease. As in the earlier study, several
patients.experienced, angina on the CO-exposure .day and not on the
air-exposure day but never the reverse. "Although the Sheps et
al. (1987) and Adams et al. (1988) studies did not observe
statistically significant changes in time to onset of angina
using conventional statistical procedures, the results of these
studies are not incompatible with the rest of the studies
reporting an effect of CO." (CD, p. 10-32)
A separate study of the effects of CO exposure was conducted
with 26 non-smoking male, angina patients (Kleinman and
Whittenberger, 1985; Kleinman et al., 1989). One hour of
exposure to 100 ppm CO was sufficient to raise COHb levels to
3.0% (CO-Ox), representing an average increase of 1.5% COHb over
initial COHb level. For the group, CO exposure resulted in a
decrease in time to onset of angina by 6.9% compared to clean air
exposure (Kleinman and Whittenberger, 1985). This was a
"statistically significant difference (p=0.03). Reanalysis,
necessitated by dropping two subjects due to inconsistent medical
records, resulted in an average decrease of 5.9% (p=0.046) in
time to onset of angina for CO exposure compared to air exposure
(Kleinman et al., 1989). For the eight patients who exhibited
depression in the ST segment of EGG traces during exercise there
was a decrease of 10% (p<0.036) in time to onset of angina and a
decrease of 19% (p<0.044) in time to onset of ST segment
depression.
Allred et al. (1989b, 1991) discuss possible reasons for
some differences in results of the above-cited studies. These
studies have different designs, types of exercise tests,
inclusion criteria (e.g., patient populations), exposure
-------
20
conditions, and measurement methods for COHb. Of the studies
only two (Allred et al., 1989a,b, 1991; Anderson et al.., 1973)
investigate more than a single target level of COHb, and of those
two only Allred et al. (1989a,b, 1991) demonstrate a dose-
response effect of COHb on time to onset of angina. Different
measurement methodologies for COHb may account for some of the
discrepancies between studies. As discussed in Section V.C.I, of
this Staff Paper and in the CD (pp. 8-70 to 8-74), only Allred et
al. (1989a,b, 1991) used both the GC and CO-pximeter to measure
COHb and found the spread of COHb values to be much greater for
the CO-Ox than for the GC. Another difference in the studies was
that Allred et al. (I989a,b, 1991) used more rigorous subject
entry criteria. All were male subjects, all were required to
have stable exertional angina and reproducible exercise-induced
ST-segment depression and angina, and all were required to have
either a previous myocardial infarction, angiographic .disease or
a positive, thallium test. ;
The major conclusion which is drawn in the CD regarding all
of the studies discussed above is that all ". . . show a decrease
in time to onset of angina at postexposure COHb levels ranging
from 2.9 to 5.9%. This represents incremental increases of 1.5
to 4.4% COHb from preexposure baseline levels. Therefore, there
are clearly demonstrable effects of low-level CO exposure in
patients with ischemic heart disease" (CD, pp. 10-34 to 10-35).
Across-study comparison is depicted in Figure V-l, which the CD
presents (p. 10-33) as an adaptation from Allred et al. (1989b,
1991). For purposes of comparison, optical methods (CO-Ox) are
used to avoid confusion. The .adverse nature of the effects
described in the five key studies is uncertain due to the range
of.professional judgments on the clinical significance of small
performance decrements produced by exercise and CO exposure.
Although some physicians may not be greatly concerned about
decrements in performance occurring around 3.0% COHb (CO-Ox),
consistency across studies of response for both decrease in time
to onset of angina and ST-segment depression suggest that-the
-------
21
to
ffl
-------
22
effect does occur and may limit activity of persons with ischemic
heart disease. Furthermore, Bassan (1990) indicates- that 58% of
cardiologists believe that recurrent exercise-induced angina
attacks are associated with substantial risk of precipitating
myocardial infarction, fatal arrhythmia, or slight but cumulative
myocardial damage (CD, p. 10-35). Based on discussions in the CD
and at the April 30, 1991 CASAC meeting, staff recommends that
2.9 - 3.0% COHb (CO-Ox), representing an increase above initial
COHb of 1.5 to 2.2% COHb, be considered a level of potential
'adversity for individuals at risk.
2. Effects on Exercise Capacity and Oxygen Uptake
Maximal oxygen uptake and maximal exercise capacity are
indirect measures of cardiovascular capacity and can provide
insight into the impact of CO on the cardiovascular systems of
even healthy individuals. Although decreases in these attributes
may not be very serious in healthy persons for CO exposures
typically found in the ambient air, they can be indicative of the
extent to which an individual's ability to function normally may
be affected while engaging in activities which require high.
levels of sustained exercise.
Numerous researchers have studied the effects of CO on
oxygen uptake and exercise performance in healthy individuals.
Several investigators (Klein et al., 1980; Stewart et al., 1978;
Weiser et al., 1978; Ekblom and Huot, 1972; Vogel and Gleser,
1972; Pirnay et al., 1971) found statistically significant
decreases (3 to 23%) in maximal oxygen uptake under conditions of
short-term maximal exercise at COHb levels ranging from 5 to 20%
(CO-Ox). Horvath et al. (1975) found that the lowest level at
which COHb marginally influenced maximal oxygen uptake (p<0.10)
was about 4.3% (CO-Ox); COHb levels of 3.3% and 4.3% (CO-Ox)
reduced work time to exhaustion by 4.9% and 7% respectively.
Similar results were found following exhaustive treadmill
exercise at 5% COHb (CO-Ox) (Stewart et al., 1978; Klein et al.,
1980). Short-term maximal exercise duration has been shown to be
reduced by 3 to 38% at COHb levels ranging from 2.3 to 7% (CO-Ox)
-------
23
(Horvath et al./ 1975; Drinkwater e't al. , 1974; Raven et al.,
1974a,b; Weiser et al., 1978; Ekblom and Huot, 1972). While
these effects may be of only limited concern for healthy
individuals, possible impairment of work capacity or of ability
to engage in normal activities for persons with respiratory
^disease should be considered in evaluating whether CO NAAQS
provide an adequate margin of safety.
3. Central Nervous System Effects
A variety of central nervous system (CNS) effects have been
found to be associated with CO exposures which result in COHb
levels of 5 to 20% (CO-Ox). These effects include changes in
visual perception, hearing, motor and sensorimotor performance,
vigilance, and other measures of neurobehavioral performance.
In a study conducted by McFarland et al. (1944) visual
sensitivity was reported to decrease in a dose-related manner at
COHb levels ranging from 4.5 to 19.7% (CO-Ox). Subsequent
analyses (McFarland, 1970, 1973) were published which suggested
that threshold shifts occurred at the end of a CO-exposure
period, Which resulted in a 17% COHb (CO-Ox) level. However,
Hudnell and Benignus (1989) studied dark adaptation and found no
difference between CO and air groups, thus leading to the CD (p.
10-108) conclusion that if COHb elevation does affect visual
sensitivity, it has not yet been demonstrated.
In either studies, Christensen e't al.. (1977) failed to find
significant vigilance effects at 4.8% COHb (CO-Ox), while Bunnell
and Horvath (1988) reported significant impairments in
performance at either 7 or 10% COHb (CO-Ox). Critical flicker
fusion (CFF) was used as a measure of temporal resolution by
estimating a subject's ability to discriminate flashes of light.
Even though several studies (Winneke, 1974; O'Donnell et al.,
1971a) found CFF not to be affected, Seppanen et al. (1977) and
Von Post-Lingen (1964) reported a relationship between decreases
of CFF at COHb levels ranging up to 20% (CO-Ox) and higher.
Motor and sensorimotor performance has been investigated
using a variety of measures (e.g. fine motor skills, reaction
-------
24
time, and tracking). Although Winneke (1974) found some effects
on steadiness and precision at 10% COHb (CO-Ox), several other
investigators (Mihevic et al., 1983; O'Donnell, 1971b; Seppanen
' et al., 1977) reported no COL effect at COHb levels ranging from
5.5 to 12.7% .(CO-Ox). Reaction time was unaffected by COHb
levels of 7 and 10% (CO-Ox) (Rummo and Sarlanis, 1974; Winneke,
1974), and the pervasive finding is that COHb elevation does not
affect reaction time for COHb levels as high as 20% (CO-Ox) (CD,
p. 10-118). Compensatory tracking was not significantly affected
by COHb levels of 5.8% (CO-Ox) (Gliner et al., 1983) or by levels
of 12 to 13 % (CO-Ox) (O'Donnell et al., 1971); however, tracking
tasks were significantly affected by COHb levels of 5% (CO-Ox)
(Putz et al., 1976,1979). Results of the Putz et al. (1976)
study were confirmed by Benignus et al. (1987) but not by
Benignus et al. (1990a) when attempting to demonstrate a dose-
effect relationship using the .same experimental design. Benignus
et al. (1990b) discusses possible reasons for high variability
between studies, and the CD (p. 10-121) concludes that COHb
elevation produces small decrements in tracking that ar,e
sometimes significant.
It can be seen from the discussion above that a wide range
of neurobehavioral effects of CO exposure have been investigated
but only in healthy young subjects. Even though new information
regarding neurobehavioral effects of COHb levels in the range of
5-20% (CO-Ox) has been published during the past decade,
conditions under which these effects occur are poorly understood
(CD, 10-143). Because neurobehavioral effects have not yet been
seen for COHb levels below 5%, (CO-Ox) staff recommends focussing
on the cardiovascular effects which have been demonstrated at
lower COHb levels. Resulting standards, therefore, should
provide an adequate margin of safety against neurobehavioral
effects of CO occurring in the exposed population.
4. Developmental Toxicity Effects
Developmental toxicity covers a variety of effects in the
developing organism including fetal death, structural
-------
25
abnormalities, altered growth and functional deficits. The fetus
may be particularly vulnerable to the toxic effects of CO
exposure because fetal development often occurs at or near
critical tislue bxygenation levels (Longo, 1977). d6lft> levels
tend to be naturally elevated in the fetus due to differences in
uptake and elimination of CO from fetal hemoglobin.
Human data on developmental toxicity of CO is very limited
for obvious ethical reasons. Maternal smoking, however, has been
associated with a number of adverse health effects, many of which
can be attributed to very high CO exposures (500-1000 ppm) from
cigarette smoke. These effects include spontaneous abortion and
subsequent fetal death due to depressed birth weight, increased
number of hospital admissions during the first 5 years of life,
and poorer than predicted school performance during the first 11
years of life. These and other effects of smoking are reviewed
in a report to the U.S. Surgeon General (National Institute of
Child Health and Human Development, 1979). Current data
(Hoppenbrouwers et al., 1981) suggesting a link between
environmental CO and sudden infant death syndrome (SIDS) are
suggestive, but further study is needed before any causal
relationship can be inferred.
Finally, animal studies have provided evidence of fetal
mortality, teratogenicity, reduced body weight, morphological
changes, altered cardiovascular development, and neurochemical
changes. However, animal studies investigating these effects are
typically conducted at CO exposures much greater than found in
the ambient air, and extrapolation to human health effects
remains very difficult. Due to the large uncertainties
associated with developmental toxicity effects in humans for
ambient CO exposures, staff recommends that these effects be
treated as margin of safety considerations.
5. 'Environmental Factors. Drugs, Other Pollutants
Several additional factors have been investigated for
potential interactions with CO that may alter health effects.
Among the more important are altitude, drugs, coexposure to other
-------
26
pollutants and heat stress. Altitude is a matter of concern
because of the large populations exposed to CO while living in
cities above 1500 m. While there are some data to support the
;*.L- •• -• • - - . ^^
possibility that effects of inhaling CO and effects of high
altitude may be additive (Cooper et al.., 1985; Mcbonagh et al.,
1986), several studies even at 2,000 m to 4500 m show little or
no additivity (McGrath, 1988; Horvath, 1988; Horvath and Bedi,
1989). Most other studies have been conducted at CO levels which
are too high to be of regulatory use.
There is evidence that interactions of drug effects with CO
toxicity can occur in both directions, i.e., CO toxicity may be
enhanced by drug use and toxic or other effects of drugs may be
altered by CO exposure. A recent study (Knisely et al., 1989)
reported a large interaction of CO exposure and alcohol in mice
demonstrating that alcohol doubled the acute toxicity of CO. In
the same study CO exposure in combination with administration of
barbiturates and other psychoactive drugs produced additive but
not synergistic effects. Combined exposures of CO and other
pollutants have been investigated primarily using animal subjects
with only a few human studies being published. No interaction
was observed in humans for CO in combination with common ambient
air pollutants such as NO2, O3, and PAN (Raven et al., I974a,b;
Drinkwater et al., 1974; Gliner et al., 1975), although a greater
decrement in exercise performance was reported in these studies
when heat stress was combined with 50 ppm CO. In conclusion,
staff recommends that information on CO in combination with other
pollutant exposures and environmental stress be treated as a
margin of safety consideration^
Et Populations Potentially at Risk to Carbon Monoxide
This section identifies subpopulations most likely to be
susceptible to low-level CO exposures based on the health effects
evidence reviewed in the revised CD and Section V-D of this Staff
Paper. Table V-3 summarizes the population groups at risk to low
level CO exposures (i.e., resulting in COHb levels below 5%)
-------
27 ' . - .
based on current evidence and mechanistic considerations. The
table includes the coronary artery disease group, which is most
clearly defined as an "at risk" population based on the
collection of studies discussed^ in Section V-D, and other groups
which may be more susceptible to CO based on more limited and
'uncertain evidence and plausible biological mechanisms. Except
for persons with ischemic heart disease and peripheral vascular
disease, there is little specific experimental evidence to
clearly demonstrate increased risk for CO-induced health^effects
at levels below 5% COHb. However, it is reasonable to expect
that individuals with preexisting illness or physiological
conditions which limit oxygen absorption or oxygen transport to
body tissues would be somewhat more susceptible to the hypoxic
(i.e., oxygen starvation) effects of CO. Table V-3 provides
population estimates for each subpopulation and a brief summary
of why each group is suspected of being more susceptible than
average to CO exposures.
The current health effects evidence suggests that the
population group at greatest risk from exposure to ambient levels
of CO is individuals with stable exercise-induced angina. Given
the likely mechanisms of CO effects on the cardiovascular system,
individuals with other indications of ischemic heart disease and
those with silent ischemia are considered to be similarly at risk
for low-level ambient CO exposures.
F. COHb Levels of Concern
In selecting the appropriate level for the primary NAAQS for
CO, the Administrator must first determine the COHb levels of
concern taking into account a large and diverse health effects
data base.. The scientific quality and strength of health data
are assessed in the CD and in Section V of this Staff Paper.
Based on these assessments, judgments are made here to identify
those studies that are most useful in establishing a range of
COHb levels to be considered in standard setting. In addition,
the more uncertain or less quantifiable evidence discussed in
-------
28
Table V-3. Summary of Subpopulations Potentially At Risk*
Croup at Risk to
Ixaw Level CO
Exposures
Coronary Artery
Disease
Peripheral Vascular
Disease
Cerebrovascular
Disease
Anemia
Chronic Obstructive
Lung Disease
Fetuses and Young
Infants
Rationale
Strongest evidence
is for group with
symptomatic angina
pectoris, although
asymptomatic
individuals have
limited coronary
flow reserve and
are likely -to be
sensitive to CO—
induced decrease in
Oj carrying
capacity.
This condition ie
associated with
limited blood flow
capacity and should
be sensitive to CO
exposures.
This condition is
associated with
limited blood flow
to the brain and
may be sensitive to
CO exix?8ures.
02 carrying
capacity of blood
is already reduced
increasing
likelihood of OO-
induced hypoxia
effects at lower CO
exposure levels
than for non-anemic
individuals
These subgroups
have reduced
reserve capacities
for dealing with
cardiovascular
stresses and have
reduced Oo supply
in blood which may
hasten onset of co-
induced hypoxic
effects .
Several animal
studies report
deleterious effects
in offspring (e.g..
reduced
birthweight, .
increased newborn
mortality, and
lower behavioral
activity levels ) .
Population
Estimates
Prevalence of
diagnosed ischemic
heart disease - 7.0
million (in 1989).
Prevalence of
undiagnosed (silent
ischemia) estimated
to be 3 to 4
million (in 1989).
l.S million heart
Attacks /yr ( in
1987)
S 13,700 heart
attack
fatalities/year (in
1987)
O.7S million (in
1979)
2.5 million persons
(in 1983-1985)
••
3.2 million (in
1987)
Bronchitis - 11.1
million
Emphysema — 2.O
million
Asthma — 8.6
million
(above for 1983-
1985)
3.6 million live
births per year
(1983)
Percent of
U.S.
Population^
About 2.9%
About 1.4%
About O.3«
About 1%
About 1.3%
About 4.6%
About O.8%
*
About 3.5%
About 1.5%
References
DOHS, 199O
American Heart
Association, 1989
American Heart
Association, 1989
American Heart
Association, 1989
DKEW, 1974
i
DHHS, 1988a
DHHS, 1988b
DHHS, 1988
DHHS, 1990
'All subgroups listed are not necessarily sensitve to CO exposure at normal ambient levels.
^Percentages were calculated based on 1989 V.S. population base of 243.5 million and assumed the absolute
numbers in the previous column were the same for 1989. Neither the absolute numbers nor the percentages can
be added because of significant overlap among these groups.
-------
29 '
Section V are, reviewed to determine the lower end of the range
that would provide an adequate margin of safety from effects of
clear concern. Those judgments relevant to the establishment of
an appropriate range of COHb levels are summarized below.
1. Cardiovascular effects, as measured by decreased time to
onset of anaina pain and bv decreased time to onset of
significant EGG ST-seoment depression; are nudged to be the
health effects of greatest concern which clearly have been
associated with CO exposures at levels observed in the ambient
air. Decrease in time to onset of exercise-induced angina pain
is well documented in studies of angina patients whose
postexposure COHb levels have been raised to 2.9-5.9% (CO-Ox),
which represents incremental increases of 1.5 to 4.2% COHb from
baseline levels (Allred et al., 1989a,b, 1991; Kleinman et al.,
1989; Adams et al., 1988; Sheps et al., 1987; Anderson et al.,
1973). Figure V-l, depicted earlier in Section V, shows
consistency between percent of time to onset of angina and COHb
levels for each of the five studies. Time to onset of
significant ECG ST-segment change, which is indicative of
myocardial ischemia in patients with documented coronary artery
disease and a more objective indicator of ischemia than angina
pain, provides supportive evidence of health effects occurring as
low as 2.9-3.0% COHb (CO-Ox). In light of the above data and
discussions of adverse health consequences in the CD (p. 10-35)
and at the April 30, 1991 CASAC meeting, staff concludes that CO
exposures resulting in COHb levels of 2.9-3.0% (CO-Ox) or higher
in persons with heart disease has the potential to increase the
risk of myocardial ischemia, an adverse health effect. It is
important that standards be set which appropriately reduce the
risk of ambient exposures which may produce such COHb levels.
2. Clinical importance of cardiovascular effects associated
with exposures to CO resulting in COHb levels of 2 to 3% remains
uncertain. Although one recent study (Allred et al., 1989a,b)
provides evidence of a 5.1% decrease in time to ST-segment
depression at 2.0% COHb when using the GC, health significance of
-------
30
this effect is more uncertain than at higher COHb levels.
Several earlier studies (Aronow, 1981; Aronow and Isbell, 1973;
Aronow et al., 1974) reported that aggravation of angina and
peripheral vascular disease may occur at COHb levels as low as
2%; however, only limited use should be made of them because
"...results of Aronow's studies did not meet a reasonable
standard of scientific quality and, therefore, should hot be used
by the Agency in defining the.critical COHb level at which
adverse health effects of CO are occurring." (CD, p. 10-24). It
is, therefore, recommended by staff that results suggesting
cardiovascular effects in angina patients when COHb levels are
between 2.0 and 2.9% be considered by the Administrator in
evaluating whether the current CO standards provide an adequate
margin of safety.
3. Elevated COHb levels in exercising, healthy, young
persons affect exercise performance and reduce oxygen uptake.
Numerous studies (Drinkwater et al., 1974; Ekblom and Huot, 1973;
Horvath et al., 1975; Raven et al., 1974; Weiser et al., 1978)
provide evidence of decreased short-term maximal exercise
duration (3-38%) in healthy young males when COHb levels are in
the range of 2.3-7.0% (CO-Ox). Submaximal exercise for short
periods of oxygen uptake was not affected by COHb levels of 15-
20% (CO-Ox) (Raven et al., 1974); however, under conditions of
short-term maximal exercise, maximal oxygen uptake was decreased
significantly (3-23%) when COHb levels ranged from 5-20% (CO-Ox)
(Klein et al., 1980; Stewart et al., 1978; Weiser et al., 1978;
Ekblom and Huot, 1972). Although these effects may not be of
clear health significance in the healthy population, possible
impairment of work capacity of individuals with respiratory
illness should be considered in evaluating whether CO standards
provide an adequate margin of safety.
-------
31
4. There are several other factors which staff believes
should be considered in evaluating the adequacy of the current CO
, *%•; •, .
NAAOS.
(a) The wide range of human susceptibility to CO exposures
and ethical considerations in selecting subjects for experimental
purposes together suggest that the most sensitive individuals
have not been studied.
(b) * Animal studies of developmental toxicity and human
studies of the effects of maternal smoking provide evidence that
exposure to high concentrations of CO can be detrimental to fetal
development, although very little is known about the effects of
ambient CO exposures on the developing fetus.
(c) Although little is known about effects of CO on
potentially sensitive populations other than those with
cardiovascular disease/ staff believes there is reason for
concern about visitors to high altitudes, individuals with anemia
or respiratory disease, and the elderly.
(d) Impairment of visual perception> sensorimotor
performance, vigilance or other CNS effect's has not been
demonstrated to be caused by CO concentrations commonly found in
the ambient air; however, short-term peak CO exposures may be
responsible for impairments which could be a matter of concern
for complex activities such as driving a car.
(e) Limited evidence suggests concern for individuals
exposed to CO concurrently with drug use (e.g. alcohol), during
heat stress, or coexposure to other pollutants.
(f) Large uncertainties remain regarding modelling COHb
formation and estimating human exposure to CO which could lead to
over or underestimation of COHb levels in the population
associated with attainment of a given CO NAAQS.
(g) COHb measurements made using the CO-Ox may not reflect
COHb levels in angina patients studied, thereby creating
uncertainty in establishing a lowest effects level for CO.
In summary, staff concludes that the lowest COHb level at
which adverse effects have been demonstrated in persons with
-------
32
angina is around 2.9-3.0%, representing an increase of 1.5% above
baseline when using the CO-Ox to measure COHb. These data serve
to establish the upper end of the range^of COHb levels of
concern. Taking into account uncertainties in the data, the less
significant health end points and less quantifiable data on other
potentially sensitive groups, staff recommends that the lower end
of the range be established at 2.0% COHb. Below this level, the
potential for public health risk appears to be small.
-------
33
VI. AVERAGING TIMES
When the EPA promulgated CO primary NAAQS on April 30, 1971.
(36 FR 8186), two averaging times— 1-hr and 8-hr — were
selected. The 8-hr standard was chosen because most individuals,
even at rest, appear to approach equilibrium levels of COHb after
8 hr of exposure. In addition the 8-hr period approximates
blocks of time for which people are often exposed in a particular
location or activity (e.g. sleeping, working) and provides a good
indicator for tracking continuous exposures that occur during any
24-hr period. The 1-hr standard was chosen because a 1-hr
averaging period provides a better indicator of short-term health
effects of CO. The 1-hr standard provides reasonable protection
from effects which might be encountered from very short duration
peak (bolus) exposures in the urban environment. Review of
current scientific information in the CD indicates that these
reasons for choosing averaging times for the CO standards remain
valid and there are no compelling arguments for selecting new or
different averaging times. It is, therefore, recommended by
staff that both averaging times be retained for primary CO
standards.
-------
-------
34
VII. EXPOSURE ANALYSIS
A. Overview
This section discusses an exposure analysis intended to
provide estimates of CO exposure and their resultant COHb levels
for people living in one city for different exposure scenarios.
The analysis provides a basis for assessing protection afforded
by the current CO standards and preliminary insight into the
relative impact of certain indoor sources to total CO exposure,
at least for Denver, Colorado, the city chosen for analysis.
Denver was chosen because (1) in 1988 it violated both the
current 1-hr and 8-hr CO NAAQS (one of only two areas that
exceeds both standards; see Section IV.B), (2) it has a
relatively high 8-hr design value, 16.2 ppm—the second-highest
design value in the U.S., and (3) CO personal monitoring data are
available for a rough validity check of the modeling effort. The
four scenarios relate to (1) current air quality versus future
air quality associated with just-attaining the 8-hr CO NAAQS, and
(2) common indoor sources present versus ambient air without
internal sources. Only the 8-hr NAAQS is evaluated since
previous analyses indicate that it is the controlling standard
from a control strategy development viewpoint (EPA, 1979b). In
the current analysis indoor sources that are removed include only
residential gas stoves and passive smoking. Other indoor
sources, such as running automobiles in private or public garages
and CO intrusion into a motor vehicle from the vehicle itself,
are not removed from the analysis; their inclusion affects the
exposure results. .
The model used for exposure analysis is a new version of the
CO NAAQS Exposure Model (NEM); see Johnson et-al. (1992).
Because the new version incorporates Monte Carlo sampling and
multiple runs, or realizations, of the model, it is known as
pNEM/CO (probabilistic NEM applied to CO).
The major model outputs of interest are estimates of the
number of person-days of exposure to various CO levels for the
four scenarios mentioned above for adults with cardiovascular
-------
35
disease in Denver. In addition, estimates also are made of the
percentage of the cardiovascular heart disease population in
Denver that would exceed selected COHb levels one or more times
per year under the four scenarios. These latter estimates are
derived by applying a modified version of the CFK equation (see
the next Section and Appendix C for a discussion of the new CFK
model used in pNEM/CO). It is estimated that there are about
36,800 non-smoking people in Denver with diagnosed or undiagnosed
(silent) ischemia. Some of these are children- <18 years of age.
Excluding them results in a "sensitive population" estimate in
Denver of about 36,645 non-smoking adult persons. See Appendix B
for a description of model mechanics, cohorts, and
microenvironments. Staff does not know to what extent findings
of this analyses can be generalized to other U.S. urban areas.
It always is speculative to make inferences to a population from
a sample size of one, even if that sample is almost the "worst
case" as Denver is with respect to ambient CO concentrations. In
addition, the functional relationships developed to relate CO
monitoriong values to outside-the-microenvironment concentrations
rely heavily on the 1982-83 personal monitoring data available in
Denver. (The only other place these data exist is in Washington,
DC or the same time period.) The "transferability" of these
functional relationships is unknown at the present time.
B. Estimating COHb Levels in Exposed Populations
As discussed earlier in this staff paper, the most relevant
indicator for evaluating the potential for being at risk to CO-
induced health effects is the distribution of blood COHb levels
in sensitive populations. The original CO-NEM (Johnson and Paul,
1983) used the CFK model (see Section V.C.2 of this Staff .Paper)
to estimate population COHb levels for the exposed population.
The new exposure analysis also contains a module for computing
COHb blood level distributions in exposed populations. The COHb
module in pNEM/CO uses the nonlinear solution to the CFK
differential equation and assumes a linear relationship between
-------
36
O2Hb and COHb. With the assumption of a linear relationship
between O2Hb and COHb there is an analytical solution to ..the CFK
equation (Muller and Barton, 1987). A brief summary of the COHb
^module used in pNEM/CO is given in Appendix C. A more detailed
discussion of how Muller's solution is used to estimate COHb
levels within pNEM/CO can be found in Biller and Richmond (1992)
and Johnson et al. (1992).
C. Results
Summary results of pNEM/CQ exposure modeling appear as
Tables VII-1 through VTI-3. All three tables focus on the number
and relative amount of person-days or person-hours of exposure
above specified CO levels or above specified COHb percentages. A
person-hour of exposure to 9 ppm, for instance, means that one
person is exposed to 9 ppm for one hour. One hundred person-
hours of exposure to 9 ppm could mean that 100 people are exposed
for 1 hour; 1 person is exposed for 100 hours; or any combination
of people exposed and hours of exposure that when multiplied,
equals 100. (Ten people each exposed 10 times to 9 ppm is one
example.)
Our estimates must be expressed in terms of person-hours (or
person-days) because the human activity data base used to model
activities is in the form of single days of activity that are
"strung together" to represent population cohorts. Thus, the
"sampling frame" used in pNEM/CO is a person-day comprised of 24
person-hours, and inferences can thus be made only to person-days
or person-hours of exposure. See Appendix B for additional
discussion of this point.
As shown in the tables, between 21 and 25 runs of pNEM/CO
were made for each of four scenarios. Unfortunately, these are
not adequate to obtain tight distributions for the relatively
extreme values of interest (i.e., % of person-days >£.!% COHb).
Many of the distributions are generally log-normal in shape and
are quite "lumpy." If more runs were made, distributions would
become more normalized and "smoother." The practical impact of
-------
37 .
this is that nonparametric tests have to be used to compare
scenario results, rather than the "usual" tests only suited for
random samples from normal distributions.
% . •&
Table VII-1 presents model results relating to the relative
percentage of person-days of exposure to the adult population
with heart disease in Denver that would experience air quality
concentrations associated with the four scenarios. Remember that
the two "just attain" scenarios relate only to attainment of the
current 8hr NAAQS of 9 ppm. Four air quality indicators are
evaluated:
1-hour daily maximum (ihDM) >35 ppm
8-hour DM (8hDM) >9ppm
8hDM >;12 ppm
8hDM >15 ppm
In all cases, the mean and its 95th percentile confidence
interval (C.I.) are presented. If the mean is statistically
different than 0 (at a p<0.05), it is noted with an 6. If the
statistic is less than 0.05% (in the case of Table VII-1) but is
not 0, the * symbol is used.
The data presented in Table VII-1 indicate that attaining
the current 8-hr CO NAAQS greatly reduces the estimated median
number and percentage of person-days of exposure to the various
1-hour and 8-hour CO indicators used. The relative reduction
from the "as is/all-sources" scenario is approximately 90% for
the 1-hr indicators and nearly 100% for the three 8-hr
indicators. These percentage reductions hold for both "attain"
scenarios: with and without the two internal sources that were
explicitly analyzed—gas stoves and passive smoking. Thus, the
data presented in Table VII-1 indicate that attaining the current
8-hr NAAQS, and consequently the current 1-hr NAAQS, apparently
will result in virtually all persons with heart disease in the
study area receiving CO exposures at or below ambient air.quality
concentrations of concern.
A Kolmogorov-Smirnov (K-S) test of the two "all sources"
scenarios was undertaken to determine if attaining the 8-hr CO
-------
38
Table VII-1
HEART DISEASE PERSON-DAYS WITH
ONE OR MORE 1H AND 8H DAILY MAXIMUM (DM) EXPOSURES
^CONCENTRATION LEVEL SHOWN FOR FOUR ALTERNATIVE SCENARIOS
(Any Exercise Level)
"As Is" Air Quality
Ambient Air Ambient Air
"Just Attain" Air Quality
Ambient Air
Ambient Air
posure and Internal w/o Certain
licators CO Sources Internal Sources
)M>35ppm
adian P-D 67,000
san Percent 0.50®
5% C.I. ±0.02
65,900
0.49@
±0.03
)M:>9ppm
fledian P-D 634,300 565,900
4ean Percent 4.75®
95% C.I. ±0.07
DM>12ppm
Median P-D 113,600
flean Percent 0.89®
95% C.I. ±0.05
DM>15ppm
«edian P-D 23,200
4ean Percent 0.17®
35% C.I. ± *
liber of Runs 21
bes: *: Less than 0.0005%
NA: Not applicable
@: Reject HQ that |i
P-D: Person-days of ex
4.19®
±0.06
91,900
0.69®
±0.02
14,800
0.11®
±0.01
25
(but not 0) .
== o at p <. 0.
posure
and Internal
CO Sources
7,300
0.058®
±0,007
1,300
0.011®
±0.003
130
0.002®
±0.001
10
*
±*
23
05
w/o Certain .
Internal Sources
6,900
0.052®
±0.004
500
0.004®
±0.001
30
0.001®
± *
0
*
± *
,22
-------
: 39
air quality standard significantly reduced exposures. It does so
for all of the exposure indicators shown in Table VIT-1. When
the K-S test is applied to the sources/no sources scenarios for
the "as is" air quality situation, however, not all exposure
estimates are statistically different. The three 8hDM exposure
estimates are different, but the IhDM estimates are not (DN =
0.19, ~ p>0.99).
These estimates—and those that follow—have to be used with
caution. The validation effort of pNEM/CO described in Johnson
et al. (1992) indicates that the model replicates the IhDM
distribution of person-days of exposure fairly well. It also
replicates the 8hDM distribution fairly well between the 30th and
the 95th percentile levels, but underestimates higher percentile
CO concentrations and overestimates lower percentile
concentrations. The main reason for the tails of the
distribution being different than the monitored data is that the
cohort-sampling method of obtaining activity-days (see Appendix
B) conflates together people having systematically low CO
exposures due to their lifestyle with people having
systematically high CO exposures within a particular cohort.
Thus, we "lose" the person who systematically cooks for long
periods of time using a gas stove. We "lose" the person who
consistently spends a lot of time in smoky indoor areas, such as
a tavern, or in high CO microenvironments, such as a service
station or garage. These people have some probability of being
included in our modeling effort, but not often enough to affect
the high tail of the person-days of exposure distribution. This
must be recognized as a shortcoming of the modeling effort, but
there is little alternative to proceeding in the way that we do
since year-long human activity data are not available. In fact,
such data are likely never to be available, since people would
have to account for their activities for an entire year rather
than the day or two that they now report on. The reporting
burden is too great to ensure cooperation for that length of
-------
40
time. All exposure modeling efforts face this shortcoming
whether or not they acknowledge it explicitly.
The next.pNEM/CO series of outputs are depicted in Table
VII-2. The indicator of interest is a surrogate measure of CO
dose received by exposed individuals—the percentage of COHb in
TxLood. Five COHb levels are analyzed, based on the range
discussed in Section V and talcing into account those used in past
CO NAAQS regulatory actions: 2.1, 2.3, 2.5, 2.7V and 3.0 percent
(30 FR 37484). The body of Table VII-2 depicts (1) the median
estimate of the number of adult heart disease person-days with at
least one hourly COHb > the COHb value depicted, and (2) the
relative percentage of total heart disease person-days that these
estimates represent. There are 13,412,070 total person-days that
were modeled (36,645 adults estimated to have heart disease times
366 days).
The absolute estimates of the number of person-days above
specified COHb levels are quite large for the lower COHb
percentages under the "as is" scenario. Even these estimates are
r-elativelv small, however, due to the large base of 13.4 million
total person-days. For instance, Table. VII-2 shows an estimate
of 63,300 person-days > 2.1% COHb, but this amounts to only 0.5%
of total possible person-days in the Denver heart disease cohort.
When the 8-hour DM NAAQS is attained, we only expect 100-200 or
so person-days of exposure at a COHb >2.1%; this is fewer than
0.002% of total person-days in the Denver group.
Attainment of the 8-hr CO NAAQS has a statistically
significant impact on COHb estimates. All of the "attain"
estimates are significantly lower than their counterpart "as is"
estimates using a K-S test of the distribution at ~p<0.05. In
fact, removing selected indoor sources (gas stoves and passive
smoking) also reduces COHb estimates significantly for the "as
is" scenario except for the >3% COHb indicator. The >3%
estimates—small to begin with—are-not statistically different
(DN = 0.22, ~p>0.99). The relative reductions in COHb person-
days are between 24-36% for the remaining COHb .levels when the
-------
41
Table VII-2
HEART DISEASE PERSON-DAYS WITH AT LEAST ONE HOURLY COHb
ESTIMATE >VALUE SHOWN FOR FOUR ALTERNATE SCENARIOS
(Any Exercise Level)
Exposure
Indicators
"As Is" Air Quality
Ambient Air Ambient Air
and Internal w/o Certain
CO Sources Internal Sources
"Just Attain" Air Quality
Ambient Air
and Internal
CO Sources
Ambient Air
w/o Certain
Internal Sources
COHb >2.1%
Median P-D
Mean Percent
95% C.I.
COHb >2.3%
Median P-D
Mean Percent
95% C.I.
COHb >2.5%
Median P-D
Mean Percent
95% C.I.
COHb >2.7%
Median P-D
Mean Percent
95% C.I.
COHb >3.0%
Median P-D
Mean Percent
95% C.I.
Number of Runs
Notes : *1
NA:
6:
P-D:
63,300
0.46@
±0.03
26,800
0.20e
±0.02
11,900
0.09@
±0.01
4,600
0.046
± *
1,600
o.oie
± *
42,600
0.32@
±0.01
17,700
0.13@
±0.01
7,600
0.06®
±0.01
3,500
Ov03@
± *
1,700
0.018
± *
200
0.0026
±0.001
80
o.ooie
±0.001
30
*@
±*
0
*
±*
0
*
±*
100
0.0016
±0.001
20
*@
±*
10
*e
±*
0
*@
±*
0
*
±*
Less than 0.0005% (but not 0).
Not applicable
Reject H0 that
Person-days of
U = 0 at p < 0.
exposure
05
-------
,42
"w/o certain internal sources" case is compared with the
"internal sources included" scenario. Thus, these two types of
internal sources play an important, but not dominant, role in
total CO exposure.
The final pNEM/CO result table is Table VII-3. It depicts
the absolute number and relative percentage of adult heart
disease person-hours that exceed the depicted COHb indicators for
the four scenarios. There are 321,889,680 adult heart disease
person-hours in Denver (36,645 x 366 x 24 hours/day).
Compared with the large base of 322 million person-hours,
the absolute and relative estimates shown in Table VII-3 are
quite small for all scenarios. Most of the estimates are
significantly different than O except for those associated with
the "just attain" scenarios for the higher COHb percentages.
Excluding gas stove and passive smoking indoor sources reduces
person-hours of COHb >2.1% and COHb >2.3% by 41% for the "as is"
air quality case. The comparable figure for the "attain" case is
59% and 73% for the respective COHb values. However, the
absolute estimates for the "attain" air quality situation are
very small (e.g., <30b person-hours). Again, the relatively
large reductions in COHb levels due to removal of certain indoor
sources indicates that these sources play an important role in
total CO exposure. The limited ability of an ambient standard to
affect CO exposures from these sources should be recognized.
As previously noted, estimates of CO emitted from indoor
sources are based on limited data. Therefore, CO exposure and
COHb estimates associated with these sources should be viewed
cautiously. While we used the best data available from the Gas
Research Institue on gas stove usage, we have not fully captured
the upper end of the usage pattern in pNEM/CO. We also have not
captured the distribution of CO concentrations associated with
passive smoking in indoor microenvironments.
While uncertainties associated with indoor sources are
important for estimating total CO exposure, they are less
important in reaching judgments regarding a CO NAAQS because
-------
Table VII-3
HEART DISEASE PERSON-HOURS WITH AN 1H COHfo ESTIMATE
RVALUE SHOWN FOR FOUR ALTERNATIVE SCENARIOS
(Any Exercise Level)
Dose
Indicators
"As Is" Air Quality
Ambient Air Ambient Air
and Internal w/o Certain
CO Sources Internal Sources
"Just Attain" Air Quality
Ambient Air
and Internal
CO Sources
Ambient Air
w/o Certain
Internal Sources!
COHb >2.1%
Median P-H
Mean Percent
95% C.I.
COHb >2.3%
Median P-H
Mean Percent
95% C.I.
COHb >2.5%
Median P-H
Mean Percent
95% C.I.
COHb >2.7%
Median P-H
Mean Percent
95% C.I.
COHb >3.0%
Median P-H
Mean Percent
95% C.I.
..Number of Runs
Notes : * :
NA:
176,800
0.050®
±0.004
63,100
0.019®
±0.002
22,600
0.007®
±0.001
8,200
0.003®
± *
2,900
0.001®
± *
21
Less than
Not applic
104,300
0.033®
±0.002
37,200
0.012®
±0.001
15,000
O.005®
±0.001
6,800
0.002®
± *
2,400
0.001®
± *
25
0,00005% (but not 0).
sable
270
±*
110
*@
±*
40
*
±*
0
*
±*
0
*
±*
23
no II
*@ 1
H
*@ N
±A II
10 II
*@ II
— * H
° H
±: I
o I]
+** II
22 U
P-H:
Reject HQ that n = 0 at p < 0.05
Person-hours of exposure
-------
44 ;
passive smoke and gas stove emissions are not affected by
alternative NAAQS levels. The focus of this Staff Paper is not .
to establish the need for controlling indoor sources but for
assessing the adequacy or need for revision of the CO NAAQS.
In conclusion, results from the Denver exposure analysis
indicate that the current 8-hr CO standard, if attained, provides
adequate protection against ambient exposures to CO. This is
true for attainment of the current 1-hr CO NAAQS also, since it
is attained when the 8-hr NAAQS is attained. The analysis also
reveals that indoor sources of CO exposures that cannot be
mitigated by a NAAQS play a role in total CO exposure. Given
uncertainties in the database used to estimate indoor source
contributions, however, this finding must be viewed with caution.
-------
-------
• • "45 ' ' '..,-, -:
VIII f SUMMARY OF STAFF CONCLUSIONS AND RECOMMENDATIONS
This summary of staff conclusions and recommendations draws
-- ...••• • ' • ' . • Vsftfr;-.
upon the discussions in Section V, VI, and VII of this paper.
The key findings are:
1) The staff concludes that the lowest COHb level at which
adverse effects have been demonstrated in small populations of
persons with angina is around 3.0% as measured by co-Ox. In
these populations, post-exposure COHb levels represent an
increase of about 1.5% above baseline. These data serve to
establish the upper-end of the range of COHb levels of concern
for people with coronary artery disease. Taking into account
uncertainties in the data, the less significant health end
points, and the less quantifiable data on other potentially
sensitive groups, staff recommends that the lower-end of the
range be established at 2.0% COHb. Below this level, the
potential for public health risk appears to be small, if any.
Based on this assessment, and considering the 1985 review of
similar CO effects and effects levels, staff recommends that the
evaluation of adequacy of the current CO standards should focus
on reducing the number of individuals with cardiovascular disease
from being exposed to CO levels in the ambient air that would
result in COHb levels of 2.1% or greater. Standards that protect
against COHb levels at the lower end of the range should provide
an adequate margin of safety against effects of uncertain
occurrence as well as those of clear concern that have been
associated with COHb levels in the upper-end of the range.
2) The staff again recommends that both the 1-hr and 8-hr
averaging times be retained for the primary CO standards.
3) Based on the exposure analysis results described in
Section VII, staff concludes that relatively few people of the
cardiovascular sensitive population group analyzed will
experience COHb levels >2.1% when exposed to CO levels in the
absence of indoor sources when the current ambient standards are
attained. The analysis also suggests, however, that certain
-------
.46
indoor sources (e.g. passive smoking; gas stove usage) contribute
to total CO exposure. In addition, other indoor CO sources such
as wood stoves and fireplaces also contribute to total CO
exposure, but they were not explicitly modeled. While the,
contribution of indoor sources cannot be effectively mitigated by
ambient air quality standards, these sources of exposure may be
of concern for such high risk groups as individuals with
cardiovascular disease, pregnant women, and their unborn
children.
The potential contribution of indoor sources to total CO
exposure should be examined in greater detail outside the context
of the NAAQS review. Such an examination should address all
potential indoor sources of CO exposure in greater detail than
could be accomplished within this NAAQS review. Until such a
detailed review is complete, results of the exposure analysis
concerning indoor exposures should be viewed as being
preliminary.
In conclusion, based on its assessment of the available
scientific and technical information, staff recommends that the
current 1-hr (35 ppm) -and 8-hir (9 ppm) primary standards be
reaffirmed.
-------
47
REFERENCES
8
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.
AIRS. Aerometric Information Retrieval System (April 16,
1991) Office of Air Quality Planning and Standards, U.S.
EPA. Quick Look Report.
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.
Allen, T.H.; Peng, M.T.; Chen, K.P.; Huang, T.F.; Chang, C.;
Fang, H.S. (1956) Prediction of Blood Volume and Adiposity
in Man from Body Weight and Cube of Height. Metabolism 5:
328-345.
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, «T. (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.; Chaitman, 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.
Allred, E.N.; Bleecker, E.R.; Chaitman, B.R.; Dahms, T.E.;
Gottlieb, S.M.; Hackney, J.D.; Pagano, M.; Selvester, R.H.;
Walden, S.M.; Warren, J. (1991) Effects of carbon monoxide
on myocardial ischemia. Environ. Health Perspect. 91: -89-
132.
American Petroleum Institute v. Costle (1982), 665 F.2d 1176
(D.C. Cir. 1981), cert, denied 102 S. Ct* 1737.
American Heart Association. (1989) 1990 heart and stroke
facts. Dallas, TX: American Heart Association.
-------
48
10.
11.
12.
13.
14.
15.
16.
17,
18.
19.
20.
21.
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 on 10 patients with ischemic heart disease. Ann,
Intern. Med. 79: 46-50.
- --•• * ' *•'•''•**•;:£'••£• '. ' ^'^^Ifjjs-**' - ,-«.'*- •$=.- • " '.
Air Quality Management Division (AQMD) October 26, 1991.
Office of Air Quality Planning and Standards, U.S. EPA.
Ozone and Carbon Monoxide Areas Designated Nonattainment.
Aronow, W.S. (1981) Aggravation of angina pectoris by two
percent carboxyhemoglobin. American Heart Journal 101: 154-
156. .
Aronow, W.S. (1978) Effect of passive smoking on angina
pectoris. N. Eng. J. Med. 229: 21-24.
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.; Cassidy, J. ; Vangrow, J.S.; March, H.; Kern,
J.C.; Goldsmith, J.R.; Khemka, M.; Pagano, J.; Vawter, M.;
(1974) Effect of cigarette smoking and breathing carbon
monoxide on cardiovascular hemodynamics on anginal patients.
Circulation 50: 340-347.
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.; Harris, D.N.; Isbell, M.W.; Rokaw, S.N.;
Imparato, B. (1972) Effect of freeway travel on angina
pectoris. Ann. Intern. Med. 77: 669676.
Aronow, W.S.; and Isbell, M.W. (1973) Carbon monoxide
effect on exercise-induced angina pectoris. Ann. Intern.
Med. 79: 392-395.
Aronow, W.S.; Stemmer, E.A.; Isbell, M.W. (1974) Effect of
carbon monoxide exposure on intermittent claudication.
Circulation 49: 415-417. . -
Bassan, M.M.
332: 272.
(1990) Letter to the Editor. N. Engl. J. Med.
Beard, R.R.; Wertheim, G.A. (1967) Behavioral impairment
associated with small doses of carbon monoxide. Am. J.
Public Health 57: 2012-2022.
-------
I
. 49 .V, . .
22. Benignus, y.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.
23. ^Benignus, V.A.; Muller, K.E.; Smith, M.J.; Pieper, K.S.;
^Prah, J.D. (1990a) Compensatory tracking in humans with
elevated carboxyhemoglobin. Neurotoxicol. Teratol. 12:
105-110.
24. Benignus, V.A.; Muller, K.E.; Mallot, C.M. (1990b) Dose-
effects functions for carboxyhemoglobin and behavior.
Neurotoxicol. Teratol. 12: 111-118.
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.
26. Biller, W.F.; Richmond, H.M. (1982) Sensitivity analysis
on Coburn model predictions of COHb levels 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.
27. Biller, W.F.; Richmond, H.M. (1992) COHb Module for a
Probabilistic CO NEM. Research Triangle Park, NC: U.S.
Environmental Protection Agency.
28. 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.
29. 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.
30. Christensen, C.L.; Gliner, J.A.; Horvath, S.M.; Wagner, J.A.
(1977) Effects of three kinds of hypoxias on vigilance
performance. Av. Sp. Env. Med. 48: 491-496.
31. 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.
32. 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. 44: 1040.
-------
50
33. DHEW (1974) U.S. Department of Health, Education, and
Welfare. Prevalence of Chronic circulatory Conditions:
United States, 1972. DHEW Publication No. (HRA) 75-1521.
Rockville, Maryland. 1974. *^
34. DHHS (1990) U.S. Department of Health and Human Services.
Vital and health statistics: current estimates from the
National Health Interview Survey, 1989, Hyattsviile,, MD:
Public Health Service, National Center'^for Health
Statistics; DHHS publication no. (PHS)90-1504. (Series 10:
data from the National Health Survey no. 176).
35. DHHS (1982) Publication NO. (PHS)83-1682. Hematological and
nutritional biochemistry reference data for persons 6
months-74 years of age: United States, 1976-80. U.S., Dept.
of Health and Human Services, Hyattsville, MD. •
36. Drinkwater, B.L. ; Raven, P.B.; Hdrvath, 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 carboxyhemoglobin.
Acta Physiol. Scand. 86: 474-482.
Friedlander, S.K, Chairman, Clean Air Scientific Advisory
Committee (1982) Letter to Ann M. Gorsuch, EPA
Administrator.
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.
40. 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: 7±4-717.
41. 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.
42. 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 arrythmias in patients with coronary artery
disease and no baseline ectopy. Arch. Environ. Health 44:
89-93.
37,
38.
39.
-------
51
43,
-44.
45.
46.
47.
48.
49.
50.
51.
52.
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.
Horvath, S.M.; Raven, P.B.; Dahms, T.E.; Gray, D.j.
Maximal aerobic capacity at different levels of -,)„ •
carboxyhemoglobin. J. Appl. Physiol. 38: 300-303.
(1975)
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.
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 39: 1323-1327.
Horvath, S.M., Ayres, A.M.; Sheps, D.S.; Ware, J. (1983)
[Letter to Dr. Lester D. Grant, including the peer-review
committee report on Dr. Aronow's studies.] May 25, 1983.
Available from: U.S. Environmental Protection Agency,
Central Docket Section, Washington, D.C.; docket no. OAQPS
79-7, IV-H-58.
Hudnell, H.K.; Benignus, Y.A. (1989) Carbon monoxide
exposure and human visual function thresholds.
Neurotoxicol. Teratol. 11: 363-371.
Johnson, T.; Capel, J. ; Paul, R; Wijnberg, L. (1992)
Estimation of Carbon Monoxide Exposure and associated
Carboxyhemoglobin Levels in Denver Residents Using a
Probablistic version of NEM. Prepared by International
Technology for U.S. EPA Office of Air Quality Planning and
Standards. Durham, N.C.
Johnson, T.; Paul, R.A. (1982) The NAAQS Exposure Model
(NEM) Applied to Carbon Monoxide (Draft). Prepared by PEDCo
Environmental, Inc., for U.S. EPA, Office of Air Quality
Planning and Standards. Durham, N.C. -
Johnson, T.R. (1984) A Study of Personal Exposure to
Carbon Monoxide in Denver, Colorado. Research Triangle
Park, NC: U.S. Environmental Protection Agency (EPA-600/84-
014). .
Kane, D.M. (1985) Investigation of the method to determine
carboxyhaemoglobin in blood. Ontario, Canada: Department of
National Defense - Canada, Defense and Civil institute of
Environmental Medicine; DCIEM no. 85-R-32.
-------
52
53,
54.
55.
56.
57.
58.
59,
60.
61.
62.
63.
Klein, J.P. ; Forster, H.V.; Stewart, R.D.; Wu, A. (1980)
Hemoglobin affinity for oxygen during short-terra exhaustive
exercise. J. Appl. Physioli-48: 236-242.
,.£..,. , '. ;. ..5-sv*.'; •:*;>;-..»•-. -:- . • ; •- " . ••
Kleinman, M.T.; Davidson, D.M.; Vandagriff, R.B.'; Caiozzo,
V.J.; Whittenberger, J.L. (1989) Effects 6f short-term
exposure to carbon monoxide in subjects with coronary artery
disease. Arch. Environ. Health 44: 361-369.
Kleinman, M.; Whittenberger, J. (1985) Effects of short-
term exposure to carbon monoxide in subjects with coronary
artery disease. Sacramento, CA: California State Air
Resources Board; report no. ARB-R-86/276. Available from:
NTIS, Springfield, VA; PB86-217494.
Lead Industries Association. Inc. v. EPA (1980) 647 F.2d
1130 (D.C. Cir. 1980), cert, denied 101 S. Ct. 621.
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.
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. Proc. Fed. Am. Soc. Exp. Biol.
45: 883.
McGrath, J.J. (1989) Cardiovascular effects of chronic
carbon monoxide and high-altitude exposure. Cambridge, MA:
Health Effects Institute; research report number 27.
McFarland, R.A. (1973) Low level exposure to carbon
monoxide and driving performance. Arch. Environ. Health 27:
355-359.
McFarland, R.A. (1970) The effects of exposure to small
quantities of carbon monoxide on vision. In: Coburh, R.F.,
ed. Biological effects of carbon monoxide. Ann. N.Y. Acad.
Sci. 174:.301-312.
j
McFarland, R.A.; Roughton, F. J. W.; Halperiii, M.H.; Niven,
J.R. (1944) The effects of carbon monoxide and altitude on
visual thresholds. J. Aviat. Med. 15: 381-394.
, 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.
-------
53
64.
65.
66.
67.
68.
69.
70.
71,
72,
73
74.
Muller, K.E.; Barton, C.N. (1987) A nonlinear version of
the Coburn, Fprster and Kane model of blood
carboxyhemoglobin. Atmos. Environ. 21: 1963-1968.
•• . , . , - .. ..-. ,.
NagdaY N.L. ; Rector, H.E.; Koontzy M.D. (1987) Guidelines
for Monitoring Air Quality. Hemisphere Publishing Corp.
National Institute of Child Health and Human Development.
(1979) Pregnancy and infant health. In: Smoking and health-
a report of the Surgeon General. Washington, DC: U. S.
Department of Health, Education, and Welfare; DHEW
publication no. PHS 79-50066.
O'Donneil, R.D.; Mikulka, P; Heinig, P.; Theodore, J. (1971)
Low level carbon monoxide exposure and human psychomotor
performance. Toxicbl. Appl. Pharma-col. 18: 593-602.
Pirnay, F.; Dujardin, J. ; Deroanne, R.; Petit, J.M. (1971)
Muscular exercise during intoxication by carbon monoxide J
Appl. Physiol. 31: 573-575.
Putz, V.R. (1979) The effects of carbon monoxide on dual-
task performance. Human Factors 21: 13-24.
Putz, V.R.; Johnson, B.L.; Setzer, J.V. (1979) A
comparative study of the effects of carbon monoxide and
human performance. In: Proc. 1st. Ann.
••» » —- — — __ • .... m*~m. ^r •.,•• i. ^ ^*JU a**^** *^^^jj* « ^
NIOSH Sci. Svmp. Chicago: Pathotox Publishing Co.
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. '
Raven, P.B.; Drinkwater, B.L.; Horvath, S.M.; Ruhling, R.o.;
Gliner, J.A.; Sutton, J.C. Bolduan, N.W. (1974) 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.
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.
Roche, S.; Horvath, S.; Gliner, J.; Wagner, J.; Borgia, J.
.(1981) Sustained visual attention and carbon monoxide;
elimination of adaption effects Hum. Factors 23:175-184.
-------
54
75. Rodkey, F.L., J.D. O'Neal, and H.A. Collison (1969) Oxygen
and carbon monoxide equilibria of human adult hemoglobin at
atmospheric and elevated pressure. Blood 33: 57-65.
76, Roughton, F.J.W. (1970) .The equilibrium ^of carbon monoxide
with human hemoglobin in whole bipod. In: Biological
effects of carbon monoxide, proceedings of a conference;
New York, N.Y. Ann. N.Y. Acad. Sci. 174 (art.l) 177-188.
77. Salorinne, Y. (1976) Single-breath pulmonary diffusing
capacity. In: Scandanavian J. Resp. diseases,
Supplementation 96.
78. 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.
79. Seppanen, A. (1977) Physical work capacity in relation to
carbon monoxide inhalation and tobacco smoking. Ann.. Clin.
Res. 9: 269-274.
80. 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.
81. 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.; Dalton, J.L.; Ballenger, M.N.; Davis, S.M.; Koch,
G.G. (1990) Production of arrythmias by elevated
carboxyhemoglobin in patients with coronary artery disease.
Ann. Intern. Med. 113: 343-351.
82. 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. (1991)
Effects of 4% and 6% carboxyhemoglobin on arrhythmia
production in patients with coronary artery disease.
Cambridge, MA: Health Effects Institute; research report
no. 41.
83. Smith, M.V. (1990) Comparing solutions to the linear and
nonlinear CFK equations for predicting COHb formation.
Math. Biosci. 99: 251-263. .
-------
55
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
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 treadmill exercise. New
YorkfiNY: Coordinating Research Council, Inc.; report no;
CRC-APRAC-CAPM-2 2-75. Available from NTIS, Springfield, VA-
PB-296627.
Technical Services Division (TSD) (1991) Office of Air
Quality Planning and Standards; U.S. EPA. National Air
Quality and Emissions Trends Report, 1989. U. S. EPA No
450/4-91-023.
Tikuisis, p., Kane, D.M., McLellan, T.M., Buick, F.,
Fairburn, S.M. (1992) Rate of formation of carboxyhemoglobin
in exercising humans exposed to carbon monoxide. J. Apnl
Physiol. 72: 4. rr
U.S. Department of Health, Education, and Welfare (1970) Air
Quality Criteria for Carbon Monoxide. AP-62.
U.S. Environmental Protection Agency. Environmental
Criteria and Assessment Office (1978) Altitude as a Factor
in Air Pollution. U.S. EPA No. 600/9-78-015. Research
Triangle Park, NiC.
U.S. Environmental Protection Agency. Environmental
Criteria and Assessment Office (1979a) Air Quality Criteria
for Carbon Monoxide. U.S. EPA No. 600/8-79-022. Research
Triangle Park, N.C.
U.S. Environmental Protection Agency. Office of Air Quality
Planning and Standards (1979b) Assessment of Adverse Health
Effects from Carbon Monoxide and Implications for Possible
Modifications of the Standard. Research Triangle Park, N.C.
U.S. Environmental Protection Agency. Office of Air Quality
Planning and Standards (1982) Sensitivity Analysis on
Coburn.Model Predictions of Carboxyhemoglobin Levels
Associated with Alternative Carbon Monoxide Standards.
Research Triangle Park, N.C.
U;S. Environmental Protection Agency. Office of Air Quality
Planning and Standards (1984aj Review of the NAAQS for
Carbon Monoxide:. Reassessment of Scientific and Technical
Information. Research Triangle Park, N.C.
U.S. Environmental Protection Agency. Environmental Criteria
Assessment Office (1984b) 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, N.C.
-------
56
94.
95.
96,
97.
98.
99.
U.S. Environmental Protection Agency (1992)
Quality Criteria for Carbon Monoxide. U.S.
90/045F. Research Triangle Park, N.C.
Draft Air
EPA No. 600/8-
Vogel, J.A.; Gleser, M.A. (1972) Effects of carbon monoxide
on oxygen transport during exercise. J. Appl.Physa.bl. 32:
234-239.
Von Post-Lingen, M.L. (1964) The significance of exposure
to small concentrations of carbon monoxide. Prpc. R. Soc.
Med. 57: 1021-1029.
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: Environmental
Stress. Individual Human Adaptations. L.J. Folinsbee, J.A.
Wagner, J.F. Borgia, B'.L. Drinkwater, J.A. Gliner, and J.F.
Bedi, eds., Academic Press,.New York. pp. 101-110.
Winneke, G. (1974) Behavioral effects of methylene chloride
and carbon monoxide as assessed by sensory and psychomotor
performance. In; Behavioral Toxicology. Early Detection of
Occupational Hazards, Proceedings of a Workshop, National
Institute for Occupational Safety and Health and University
of Cincinnati, Cincinnati, Ohio, June 24-29, 1973. C.
Xintaras, B.L. Johnson, and I. de Groot, eds., HEW
Publication No. (NIOSH) 74-126, U. S. Department of Health,
Education and Welfare.
Wyman, J.; Bishop, G.; Richey B.; Spokane, R.; Gill, S.
(1982) Examination of Haldane's first law for the partition
of carbon monoxide and oxygen to hemoglobin" AQ. Biopolymers
21: 1735-1747.
-------
APPENDIX A
-------
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, O.C. 20460
EPA-SAB-CASAC-91-015
July 17, 1991
,
THEAOMINiSnUTOR
The Honorable William K. Reilly
Administrator
U.S. Environmental Protection Agency .
401 M Street. SW
Washington. DC 20460
* • '" . .
Dear Mr. Reilly: .
At a public meeting held on April 30. 1991. the Clean Air Scientific Advisory
Committee (CASAC) completed its review of the draft EPA Air Quality Criteria for
Carbon Monoxide dated March 1990. The Committee unanimously concluded that
this document, with minor revisions (currently being incorporated by ECAO Staff).
provides a scientifically balanced and defensible summary of the current knowledge
of the effects of this pollutant and provides an adequate basis for the EPA to
make a decision as to the appropriate primary, J^AAQS for carbon monoxide.
The first external review draft of this document was released for public
comment on April 30. 1990 with the comment period ending on July 31. 1990.
CASAC is pleased with the responsiveness of ECAO in producing a comprehensive]
well-written document to support Agency decision-making. For the record. I have
attached brief responses to the major issues which were addressed in the
Committee charge.
The CASAC is ready to review the Staff Paper on Carbon Monoxide as soon
as it is available. The Committee urges the Agency to move forward as rapidly as
possible with completion of the Staff Paper and. ultimately, the issuance of a
reaffirmed or revised NAAQS for carbon monoxide based on the current scientific
data.
We appreciate the opportunity to present our views on this important
environmental health issue.
Sincerely,
Roger O. McClellan
Chairman. Clean Air
Scientific Advisory Committee
Attachment
PrirXed on Recycled Paper
-------
Clean Air Scientific Advisory Committee
Review of
Draft Air Quality Criteria for Carbon Monoxide
On April 30.1991. the Clean Air Scientific Advisory Committee convened
to review the draft document Air Quality Criteria for Carbon Monoxide, dated
March 1990. Development of this document stems from requirements of section
108 of the Clean Air Act. This section requires that the Administrator identify
pollutants that may reasonably be anticipated to endanger public health or
welfare and to issue air quality for them. These criteria must incorporate the
latest scientific information available to indicate the type and extent of
identifiable effects that may occur from exposure to the pollutant in ambient air.
Section 109 of the Act requires periodic review/revision of existing
criteria and standards. If the Administrator concludes that the revised criteria
make appropriate the proposal of new standards, such standards are to be
promulgated in accordance with section 109{b). Conversely, if the Administrator
concludes that the revisions to the standards are unnecessary, they remain
unchanged.
In accordance with the Clean Air Act EPA's Environmental Criteria and
Assessment Office is revising the criteria for carbon monoxide, incorporating
new data which have become available since the completion of the last criteria
document (1979) and the addendum to that document (1984).
The draft carbon monoxide document review consisted of a chapter by
chapter review and focused on addressing the following issues:
1) What method of analysis of blood carboxyhemoglobin levels, optical or
gas chromatography, should be used to determine lowest observed adverse
effect levels for CO? Should end-exposure or end-exercise COHb levels be
used as an input to the exposure models of COHb formation developed by
Coburn, Foster and Kane?
Due to the large amount of variability in spectroscopic measurement of
carboxyhemoglobin by CO oxymeters, gas chromatography should be the
method of choice,
Coburn-Foster-Kane-based models yield the expectant net increase in
COHb for a given exposure to CO (concentration and duration), and the level of
activity/exercise (alveolar ventilation and diffusing lung capacity for CO). Input
to the model requires the preexposure COHb level, with the post exposure level
-------
predicted by the model. The model does not accurately predict the rate of
appearance of COHb at the blood sampling point Because of the lag in the
delivery of CO due to lung washing and blood circulation factors.
2) How important is tissue action of CXD. given the likelihood of typical
ambient exposures of the population to low levels of CO for 1 to 8 hours in
duration?" -*>?••>••*•••
Although it is difficult to expand on the information contained within the
document it should be noted that elevated levels of CO. particularly from bolus
exposures, may potentially affect the electron transport chain. Also, some
studies conclude that CO dissolved in plasma is more dangerous than elevated
COHb levels. Low levels of dissolved CO may be significant in cellular .
respiration.
3) What fraction of the total population with ischemic heart disease (IHD) is
represented by the study populations used in the recent key clinical
investigations of Sheps, et al. (1987), Adams et al. (1988), Kleinman et al.
(1989) and Allred et al. (1989)?
The study by Allred et al. and the Coronary Artery Surgery Study
(CASS) provide a wide representation of patients with ischemic heart disease:,
and the CASS study is a good source of information on the variability of
characteristics of IHD (almost 25.000 patients enrolled). AH subjects studied for'
the effects of CO fall within this variability.However, since no Coronary Artery
Disease (CAD) registry was developed for the CO studies, coupled with the
change in characterization of CAD in recent years, it is difficult to assess the
representativeness of the study populations.
4) Were appropriate statistical analyses used in the key studies on subjects
with IHD? Should there be a more rigorous comparison of statistical
approaches, including discussion of primary versus secondary analyses, use of
trimmed or non-trimmed means, and choice of one- or two-tailed tests of
significance? Could other formal techniques (meta-analysis) be used to
provide a better assessment of data?
The analyses provided in the document were adequate and appropriate.
In general statistical analyses need not be uniform, but should be tailored to the
data being collected, and the distinction between one- and two-tailed tests is
insignificant. Meta-analysis is useful, but graphic presentations such as those
provided in figure 10-2 are satisfactory. However, error bars should be
highlighted and made a common basis for data presentation.
5) Could differences in the study designs utilized in the key studies on the
subjects with Ischemic Heart Disease account for some of the differences in the
results?
-------
It is unlikely that"variations in^uBy design resulted in vanations in
results The protocols for each study are described in sufficient detail and the
authors have done an excellent job of presenting and interpreting the results.
6) Are the small changes reported in jhe key studies on subjectsAivfih
Ischemic Heart Disease of dinted significance? What is the definition of ah
adverse health effect in this population?
There Is a wide distribution of opinioiY^hclinlng this issue. The panel
agrees that the effect observed at these levels are small performance
decrements and that they are consistent across the populations studied. It is
important to note that the ST segment changes and decrements in the time to
onset of angina appear to be a consistent response to tow levels of CO
exposure. Among health professionals there is a range of views as to the
clinical significance of these changes with the dominant view being that the
changes should be considered as adverse or a harbinger of adverse effects.
-------
APPENDIX B
-------
-------
Appendix B
Overview of Exposure Analysis Model
The exposure analysis simulates CO pollutant concentrations
in "microenvironments" (defined below) which people enter as they
go through their daily lives. In pNEM/CO, people are represented
by a set of eleven demographic groups that are based upon age,
sex, occupation, and commuting breakdowns. These groups are
further subdivided into "cohorts" that have a distinct
combination of: home district, work district (if applicable), and
type of residential cooking fuel (natural gas or not). These
subdivisions are used because the personal exposure monitoring
(PEM) study conducted in 1983 in Denver (Johnson et al., 1984)
indicated that these factors contribute significantly to personal
CO exposure.
Home and work districts are defined by U.S. Census tract
data on where people live and work in Denver. Ambient air
quality for each district is defined by -"transformed" ambient air
quality data monitored at fixed-site stations located in each
district. .
Cohort daily activities are simulated by sampling from human
activity data obtained in Denver, Washington (DC), and
Cincinnati. These human activity data bases were combined for
pNEM/CO modeling purposes to increase the number of diary-days of
data available for analyses. In addition, using all three
studies provides more seasonal .variation in activities than use
of the Denver study alone provides. The human activity data are
in the form of a "person-day" of activities specific to the
demographic group and cohort being modeled. Since the modeling
"sampling frame," so to speak, is a person-day, inferences to the
total Denver population CO exposures associated with the air
quality scenarios are most appropriately made only to "person-
days" or "person-hours" of exposure.
as inputs to the CFK model in the COHb module of pNEM/CO. Many
of the parameters are physiological variables which vary from
individual to individual within the populations of interest. The
COHb module assigns a height and weight to the person
representing each cohort and keeps this constant for a 24-hr
period. The procedure relies on height and weight data collected
as part 'of the National Health and Nutrition Examination Survey
1 Q*7 C. \
-------
. B-4
-------
APPENDIX D
-------
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON. D.C. 2Q460
EPA-SAB-CASAC-LTR-92-016
Honorable William K. Reilly
Administrator
U.S. Environmental Protection Agency
401 M St., S.W.
Washington, D.C, 20460
.August 11, 1992
OFFICE OF
THE ADMINISTRATOR
Subject: Clean Air Scientific Advisory Committee Closure on the OAQPS Staff
Paper for Carbon Monoxide
Dear Mr. Reilly:
Clean Ajr Scientific Advisory Committee (CASAC) of EPA's Science
ndard? fnr
-QAOPg
M
document entitled Review of the National Amhipnt Air Qualit
Monoxide: Assessment of Scientific and Technical informati
Inri SZT'f66 "^^ sStisfaction the ^Provements made in the sciemffic ot
and completeness of the staff paper. It has been modified in accordance with the
recommendations made by the CASAC in March and April, 1992.
in to* ?*«£°SUment ''f 5°nsi5ent wlth a" asp60** of the scientific evidence presented
in the cntena document for carbon monoxide. It has organized the relevant
information in a logical fashion and the Committee believes that it provides a
scientifically adequate basis for regulatory decisions on carbon monoxide. The staff
paper concludes, and the CASAC concurs, that a standard of the present form and
with a numencal value similar to that of the present standard would be supported by
the present scientific data on health effects of exposure to carbon monoxidT
The Committee looks forward to receiving notice of the revised or reaffirmed
carbon monoxide standard when ft is proposed. reanirmea
Sincerely,
Roger O. McClellan, D.V.M.
Chairman /
Clean Air Scientific Advisory Committee
f..- Pfiritea en Recycled Paper
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
DEC 2 81982
•THEAOMINISFRATOR
Roger O. McClellan, D.V.M.
Chairman
Clean Air Scientific Advisory Committee
U.S. Environmental Protection Agency
401 M Street, S.W.
Washington, D.C. 20460
Dear Dr. McClellan:
ria - °f ^9^ 11* "92 concerning the
Clean Air Scientific Advisory Committee (CASAC) review of the
staff document entitled Review of the National Ambient pir
Quality Standards for Carbon Monoxide : Assessment o"f~ScTe'nt J f -5 c*.
and Technical Information - OAOPS Staff PaperT --
v I am pleased that the CASAC found that the document is
consistent with the scientific evidence presented in the draft
criteria document for carbon monoxide and that it provides a
scientifically adequate basis for completing the review of the
air quality standards for carbon monoxide. The guidance and
recommendations provided by the CASAC greatly facilitated
preparation of the staff paper and will greatly assist us in
reaching a decision as to whether standard revisions are
appropriate at this time.
I want to thank you, the other Committee members and
consultants for your contribution to the review of the carbon
monoxide standards.
Sincerely yours,
/e/ William X. Eeflly
William K. Reilly
Printed on Recycled Paper
-------
TECHNICAL REPORT DATA
(Please read tnstruclioris on the reverse before completing}
I REPORT NO.
EPA-452/R-92-004
2,
| TITLE AND SUBTITLE
Review of the National Ambient Air Quality Standards
for Carbon Monoxide: 1992 Reassessment of Scientific
and Technical Information — OAQPS Staff Paper
6. PERFORMING ORGANIZATION CODE
| AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
D.J.; McCurdy, T. R. ; Richmond, H. M.
| PERFORMING ORGANIZATION NAME AND ADDRESS
Office of Air and Radiation
Office of Air Quality Planning and Standards
U. S» Environmental Protection Agency
Research Triangle Park, NC 27711 ' ''.
. SPONSORING AGENCY NAME AND ADDRESS
3. RECIPIENT'S ACCESSION NO.
REPORT DATE" V
1O. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
13. TYPE OF REPORT AND PERIOD COVERED
Final . „,-,..,
14. SPONSORING AGENCY CODE
Is. SUPPLEMENTARY NOTES
16. ABSTRACT . . . _ . ,,.,...
This paper evaluates and interprets the updated scientific and technical information
that EPA staff believes is most relevant to the review of primary (health) national
ambient air quality standards for carbon monoxide. This assessment is intended to
bridge the gap between the scientific review in the EPA criteria document for carbon".
monoxide and the judgments required of the Administrator in setting ambient air
quality standards for carbon monoxide. The major recommendations of the staff paper
include the following: (1) There continues to be a need to control ambient levels of
carbon monoxide to protect public health; (2) Both 1-hour and 8-hour averaging times
should be retained for primary carbon monoxide standards; (3) Exposure analysis
results indicate relatively few individuals with angina pectoris would experience
carboxyhemoglobin (COHb) levels of 2.1 % or greater when exposed to carbon monoxide
levels in ambient air only if current standards are attained; (4) Public health risk
for COHb levels of 2.0 % or lower appears to be small, if any; (5) Current 1-hour
(35 ppm) and 8-hour (9 ppm) standards for carbon monoxide should be reaffirmed.
KEYWORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
Carbon Monoxide
Air Pollution
Ambient Air Quality
Air Standards
Health 'Effects
Air .Quality Standards
COSATI Field/Group
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (T.
Unclassified
92
2O. SECURITY CLASS (Tllispagej
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
-------
INSTRUCTIONS
1> Insert the UPA «port numbcr,;«"it:app«"s on the cover of the publication.
2. LEAVE BLANK ,.._.« .
3. RECIPIENTS ACCESSION NUMBER
Reserved for use by each report recipient. „ • , •
^^
number and include subtitle for the specific title.
AaU carry, date indicating at least month and year. Indicate the l-asis o« which it was selected (c.f..Ja,f «'»«<-.
approval, date of preparation. etc.J. .-•• ,,
6. PERFORMING ORGANIZATION CODE
Leave blank. '
7- Civ™mRe§ in conventional order fSobi *./**.
zation. ' .
S. PERFORMING ORGANIZATiPN REPORT NUMBER
Insect if performing organoaiion wishes to assign this number.
. rfr./. List author^ aff.liatio,, if it differs from the performing organ!-
•
report was prepared. Subordinate numbers nuy bo incU.dcd in parcu.hcs,-,
11. CONTRACT/GRANT NUMBER
Insert contract or grant number under which, report was prepared.
12. SPONSORING AGENCY NAME AND ADDRESS
Include ZIP code. •
13. TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc,, and if applicable, dates covered.
14. SPONSORING AGfc-NCY CODE '
Insert appropriate code. V
To be published in, Supersedes, Supplements, etc.
significant bibliography or literature survey, mention it here.
' **« *'"" * "
"*'
the primary posting
------- |