FINAL: 07/2008
ACUTE EXPOSURE GUIDELINE LEVELS
              (AEGLs)
        CARBON MONOXIDE
       (CAS Reg. No. 630-08-0)
              July 2008

-------
CARBON MONOXIDE                                                  FINAL: 07/2008
                                       PREFACE

       Underthe authority of the Federal Advisory Committee Act (FACA) P. L. 92-463 of 1972, the
National Advisory Committee for Acute Exposure Guideline Levels for Hazardous Substances
(NAC/AEGL Committee) has been established to identify, review and interpret relevant toxicologic
and other scientific data and develop AEGLs for high priority, acutely toxic chemicals.

       AEGLs represent threshold  exposure limits for the general public and are applicable to
emergency exposure periods ranging from 10 minutes to 8 hours. AEGL-2 and AEGL-3 levels, and
AEGL-1 levels as appropriate, will  be developed for each of five exposure periods (10  and  30
minutes,  1 hour, 4 hours,  and 8 hours) and will be distinguished by varying degrees of severity of
toxic effects. It is believed that the  recommended exposure levels are applicable to the general
population including infants and children, and other individuals who may be sensitive or susceptible.
The three AEGLs have been defined as follows:

       AEGL-1 is the airborne concentration (expressed as ppm or mg/m3) of a substance above
which it is predicted that the general  population, including susceptible individuals, could experience
notable discomfort,  irritation,  or certain asymptomatic, non-sensory effects. However, the effects are
not disabling and are transient and  reversible upon cessation of exposure.

       AEGL-2 is the airborne concentration (expressed as ppm or mg/m3) of a substance above
which it is predicted that the general  population, including susceptible individuals, could experience
irreversible or other serious, long-lasting adverse health effects, or an  impaired ability to escape.

       AEGL-3 is the airborne concentration (expressed as ppm or mg/m3) of a substance above
which it is predicted that the general  population, including susceptible individuals, could experience
life-threatening health effects or death.

       Airborne concentrations below the AEGL-1 represent exposure levels that could produce
mild  and progressively increasing odor,  taste,  and  sensory  irritation, or certain asymptomatic,
non-sensory effects. With increasing airborne concentrations above each AEGL level, there is a
progressive increase in the likelihood of occurrence and the severity of effects described for each
corresponding AEGL level. Although the AEGL values represent threshold levels for the general
public, including sensitive subpopulations, it is recognized that certain individuals, subject to unique
or idiosyncratic responses,  could experience the effects described at concentrations below the
corresponding AEGL level.

-------
CARBON MONOXIDE                                                 FINAL: 07/2008


                               TABLE OF CONTENTS


PREFACE	ii

EXECUTIVE SUMMARY	vii

1.      INTRODUCTION	1

2.      HUMAN TOXICITY DATA	2
       2.1.   Acute Lethality	3
             2.1.1  Case Studies	4
       2.2.   Nonlethal Toxicity	7
             2.2.1  Experimental Studies	7
             2.2.2  Case Studies	14
       2.3.   Developmental/Reproductive Toxicity	20
       2.4.   Genotoxicity	23
       2.5.   Carcinogen icity	23
       2.6.   Summary	23

3.      ANIMAL TOXICITY DATA	24
       3.1.   Acute Lethality	24
             3.1.1. Rats	24
             3.1.2. Mice	25
       3.2.   Nonlethal Toxicity	28
             3.2.1  Monkeys	28
             3.2.2  Dogs	29
       3.3.   Developmental/Reproductive Toxicity	29
             3.3.1  Pigs	29
             3.3.2  Rabbits	30
             3.3.3  Rats	30
             3.3.4  Mice	30
       3.4.   Genotoxicity	31
       3.5.   Carcinogen icity	31
       3.6.   Summary	31

4.      SPECIAL CONSIDERATIONS	32
       4.1.   Stability, Metabolism and Disposition	32
       4.2.   Mechanism of Toxicity	32
       4.3    Issues related to post-mortem CO determination in humans
             4.3.1. Potential factors influencing COHb levels
             4.3.2. Influence on collection site on measured COHb levels

       4.4.   Other Relevant Information	33
             4.4.1. Species Variability	33
             4.4.2. Intraspecies Variability	35

-------
CARBON MONOXIDE                                            FINAL: 07/2008
            4.4.3.  Time Scaling	36
            4.4.4.  Mathematical models of COHb formation	37

5.     DATA ANALYSIS FOR AEGL-1 	40
      5.1.   Human Data Relevant to AEGL-1	40
      5.2.   Animal Data Relevant to AEGL-1	40
      5.3.   Derivation of AEGL-1	40

6.     DATA ANALYSIS FOR AEGL-2 	41
      6.1.   Human Data Relevant to AEGL-2	41
      6.2.   Animal Data Relevant to AEGL-2	42
      6.3.   Derivation of AEGL-2	42

7.     DATA ANALYSIS FOR AEGL-3 	46
      7.1.   Human Data Relevant to AEGL-3	46
      7.2.   Animal Data Relevant to AEGL-3 	46
      7.3.   Derivation of AEGL-3	47

8.     SUMMARYOFAEGLs	48
      8.1.   AEGL Values and Toxicity Endpoints	48
      8.2.   Comparison with Other Standards and Criteria	52
      8.3.   Data Adequacy and Research Needs	54

9.     REFERENCES	55

APPENDIX A Time Scaling Calculations for AEGLs	63
      AEGL-2	64
      AEGL-3	65

APPENDIX B Mathematical Model for Calculating COHb and Exposure Concentrations	66

APPENDIX C Derivation Summary for Carbon Monoxide AEGLs	75
      AEGL-1 	76
      AEGL-2	77
      AEGL-3	80
                                      IV

-------
CARBON MONOXIDE                                       FINAL: 07/2008

                           LIST OF TABLES

TABLE 1: CHEMICAL AND PHYSICAL DATA	2
TABLE 2: SYMPTOMS ASSOCIATED WITH COHb IN HEALTHY ADULT HUMANS	3
TABLE 3: INCIDENCE OF ATHEROSCLEROTIC CORONARY ARTERY DISEASE AND COHb IN
     FATALITIES THAT INVOLVED CO EXPOSURE	6
TABLE 4: EFFECTS OF ACUTE CO EXPOSURE IN A HUMAN SUBJECT	11
TABLE 5: SYMPTOM THRESHOLD VALUES FOR PEDIATRIC CO TOXICITY	16
TABLE 6: SEVERITY OF CO POISONING	19
TABLE 7: COHb, EXPOSURE DURATION AND LACTATE CONCENTRATIONS IN RELATION TO
     SEVERITY OF CO POISONING	19
TABLES: SEVERITY OF CO POISONING	20
TABLE 9: OVERVIEW OF CLINICAL SCORING, COHb AND FETAL OUTCOME	20
TABLE 10: OVERVIEW OF MATERNAL CLINICAL EFFECTS, COHb AND FETAL OUTCOME22
TABLE 11: SUMMARY OF LC50 DATA IN LABORATORY ANIMALS	26
TABLE 12: COHb AFTER 48 HOURS CONTINUOUS EXPOSURE TO CO	34
TABLE 13: AEGL-1 VALUES FOR CARBON MONOXIDE	41
TABLE 14: AEGL-2 VALUES FOR CARBON MONOXIDE	46
TABLE 15: AEGL-3 VALUES FOR CARBON MONOXIDE	48
TABLE 16: SUMMARY/RELATIONSHIP OF AEGL VALUES FOR CARBON MONOXIDE	49
TABLE 17: EXTANT STANDARDS AND GUIDELINES FOR CARBON MONOXIDE	52
TABLE 18: CONCENTRATION-TIME COMBINATIONS RESULTING IN 4 % COHb	71
TABLE_19: COHb VALUES FOR AEGL-2 CONCENTRATION-TIME COMBINATIONS IN
     DIFFERENT SUBPOPULATIONS 	71
TABLE 20: CONCENTRATION-TIME COMBINATIONS RESULTING IN 40 % COHb	72
TABLE 21: COHb VALUES FOR AEGL-3 CONCENTRATION-TIME COMBINATIONS IN
     DIFFERENT SUBPOPULATIONS 	72
TABLE 22: COMPARISON OF REPORTED AND CALCULATED COHb VALUES FOR THE DATA
     BYHALDANE(1895)	74

-------
CARBON MONOXIDE                                        FINAL: 07/2008

                           LIST OF FIGURES

FIGURE 1: LC50 VALUES FOR CO IN DIFFERENT SPECIES	27
FIGURE 2: COHb FOR DIFFERENT EXPOSURE CONCENTRATION-TIME COMBINATIONS38
FIGURE 3: CATEGORICAL REPRESENTATION OF ALL CO INHALATION DATA	51
FIGURE 4: COHb VS. EXPOSURE TIME PLOTS 	69
FIGURE 5: CALCULATION OF 60-MINUTE AEGL-2 FOR HEALTHY ADULT	71
FIGURE 6: CALCULATION OF 60-MINUTE EXPOSURE CONCENTRATION THAT WOULD
     RESULT IN 40% COHb IN A HEALTHY ADULT 	74

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

                                EXECUTIVE SUM MARY

       Carbon monoxide (CO) is  a tasteless, non-irritating, odorless and  colorless  gaseous
substance.  The main  source  of CO production  is the  combustion of fuels. Exposure at the
workplace occurs in blast furnace operations in the steel industry and when gasoline- or propane-
powered forklifts, chain-saws or other machines are used in confined spaces, such as companies,
tunnels and mines. Environmental exposure to CO can occur while traveling in motorvehicles(9-25
and up to 35 ppm), visiting urban locations with heavily traveled roads (up  to 50 ppm), or cooking
and heating with domestic gas, kerosene, coal or  wood (up to 30 ppm) as well as in fires and by
environmental tobacco smoke. Endogenous CO formation during normal  metabolism leads to a
background carboxyhemoglobin concentration (COHb) of about 0.5-0.8 %. Smokers are exposed to
considerable CO concentrations leading to a COHb of about 3-8 %.

       CO binds to hemoglobin forming COHb and thereby renders the hemoglobin molecule less
able to bind oxygen. Due to this mechanism, the oxygen transport by the blood and the release of
bound oxygen in the tissues are decreased. Tissue  damage results from local hypoxia. Organs with
a high oxygen requirement, such as the heart and the brain, are especially sensitive forthis effect.

       AEGL-1 values were not recommended because susceptible persons may experience more
serious effects (equivalent to AEGL-2 level) at concentrations, which  do  not yet cause AEGL-1
effects in the general population.

       Patients with coronary artery disease show  health  effects at lower COHb levels than
children,  pregnant women or  healthy adults  and, thus, constitute  the most  susceptible
subpopulation. For the derivation of AEGL-2 values  a level  of 4  % COHb was chosen. At this
exposure level, patients with coronary artery disease may experience a reduced time until onset of
angina (chest pain) during physical exertion (Allred etal., 1989; 1991). In the available studies, the
CO exposure alone (i.e. with subjects at rest) did  not cause  angina, while exercise alone did so.
However, since all studies used  patients with stable  exertional angina, who  did not experience
angina while at rest, it cannot be ruled out that in more susceptible individuals (a part of the patients
with unstable angina pectoris  might belong to this group) CO exposure alone could cause or
increase angina symptoms. The changes in the electrocardiogram (ST-segment depression of 1
mm (corresponding to  0.1  mV) or greater) associated with  angina symptoms were considered
reversible, but is indicative of clinically relevant myocardial ischemia requiring medical treatment.
An exposure level of 4 % COHb is unlikely to cause a significant increase in the frequency of
exercise-induced arrhythmias. Ventricular arrhythmias have been observed at COHb of 5.3%, but
not at 3.7 % (Sheps et al., 1990;  1991), while in another study no effect of CO exposure on
ventricular arrhythmia was found at 3 or 5 % COHb (Dahms et al., 1993). This exposure level,
which corresponds to COHb values of 5.0-5.6% in newborn and children was considered protective
of acute  neurotoxic effects in  children, such  as  syncopes,  headache, nausea, dizziness and
dyspnea (Klasner  et al.,  1998; Crocker and Walker,  1985), and long-lasting neurotoxic effects
(defects in the cognitive development and behavioral alterations) in children (Klees et al., 1985). A
mathematical model (Coburn et al., 1965; Peterson and Stewart, 1975)  was used  to calculate
exposure concentrations in air resulting in a COHb of 4 % in adults at the end of exposure periods
of 10 and 30 minutes and 1, 4 and 8 hours. A total  uncertainty factor of 1 was used. A level of 4 %
COHb was the NOEL for AEGL-2 effects in patients with coronary artery disease, while the LOEL
was estimated at 6-9 %. In comparison, the LOEL  was about 10-15 % in children and 22-25 % in
                                          VII

-------
CARBON MONOXIDE
                                                                  FINAL: 07/2008
pregnant women. Since AEGL-2 values were based on experimental data on the most susceptible
subpopulation, they were considered protective also for other subpopulations and a total uncertainty
factor of 1 was used.

       It is acknowledged that apart from emergency situations, certain scenarios could lead to CO
concentrations which may cause serious effects in persons with cardiovascular diseases. These
scenarios include e.g. extended exposure to traffic fume emissions (e.g., in tunnels or inside cars
with  defect car exhaust systems), charcoal or wood fire furnaces, and  indoor air pollution by
tobacco smoking.

       The derivation of AEGL-3 values was based on a weight of evidence analysis of human
lethal and non-lethal observations. Analysis of lethal cases reported by Nelson (2005a) indicated
that most lethal poisoning cases occurred atCOHb levels higherthan 40% and that survival of CO-
exposed humans were likely to be seen at levels below 40%.  Thus, 40%COHb level seems a
reasonable threshold for lethality.

       This level is supported by experimental studies performed in healthy human subjects.
Studies by Chiodi etal. (1941), Henderson etal. (1921), and Haldane (1895) suggest that a COHb
of about 34-56 % does not cause lethal effects in healthy individuals. Further support come from the
studies  by  Kizakevich et  al.  (1994), Stewart et  al.  (1970),  and Nielsen (1971) that  reported
headache as the only symptom when subjects were exposed to 20-33 % COHb. A level  of 40 %
COHb was used as the basis for AEGL-3 derivation. This point of departure is supported by studies
in animals reporting minimum lethal COHb levels in rats and mice of about 50-70 % (E.I. du Pont de
Nemours and Co., 1981; Rose et al., 1970). A mathematical model (Coburn et al., 1965; Peterson
and Stewart, 1975) was used to calculate exposure concentrations in air resulting in a COHb of 40
% at the end of exposure periods of 10 and 30 minutes and 1, 4 and 8 hours. A total uncertainty
factor of 3 was used. A total  uncertainty factor of 3 for intraspecies variability was considered
adequate based on supporting evidence for susceptible subpopulations: 1) Exposure to the derived
AEGL-3 concentrations will result in COHb values of about 14-17 % in adults, which, based on case
reports, was considered to protect heart patients against CO-induced myocardial  infarction. It
should be  noted, however, that a clear threshold for this endpoint cannot be defined  because
myocardial  infarction might be triggered  at lower COHb in hypersusceptible individuals. 2) This
COHb level was considered protective of lethal effects in the unborn, because in the case studies
available, stillbirths were found only after measured maternal COHb of about 22-25 % or higher
(Caravati et al., 1988; Koren et al., 1991) and the level was supported by animal studies..

       The AEGL values are listed in the table below.
SUMMARY TABLE OF AEGL VALUES FOR CARBON MONOXIDE
Classificatio
n
AEGL-1
(Nondisablin
g)

10-Minute
N.R.a

30-
Minute
N.R.

1-Hour
N.R.

4-Hour
N.R.

8-Hour
N.R.

Endpoint
(Reference)
-

                                          VIM

-------
CARBON MONOXIDE
                                                                  FINAL: 07/2008
AEGL-2 b
(Disabling)
AEGL-3 c
(Lethal)













420 ppm
(480
mg/m3)
1700 ppm
(1900
mg/m3)












150 ppm
(170
mg/m3)
600 ppm
(690
mg/m3)












83 ppm
(95
mg/m3)
330 ppm
(380
mg/m3)












33 ppm
(38
mg/m3)
150 ppm
(170
mg/m3)












27 ppm
(31
mg/m3)
130 ppm
(150
mg/m3)












Cardiac effects in
humans with
coronary artery
disease (Allred et
al., 1989; 1991)
Lethal poisoning
was associated with
a COHb >= 40 % in
most lethal
poisoning cases
reported by Nelson
(2005a); no severe
or life-threatening
effects in healthy
humans at COHb of
34-56 % (Chiodi et
al., 1941;
Henderson et al.,
1921; Haldane,
1895)
a N.R., not recommended because susceptible  persons  may experience more serious effects
(equivalent to AEGL-2 level) at concentrations,  which do not yet cause AEGL-1 effects in the
general population.

b It was estimated that exposure to the AEGL-2 concentration-time combinations result in COHb
levels of 5.3-5.6 % in newborns, 4.9-5.2 % in 5-year-old children, 4.0 % in adults and 6.2-11.5 % in
adult smokers.

c It was estimated that exposure to the AEGL-3 concentration-time combinations result in COHb
levels of 19.5-20.1 % in newborns, 18.1-18-7 % in 5-year-old children,  13.8-17.2 % in adults and
16.1-23.0 % in adult smokers.
References

Allred, E.N., E.R. Bleecker, B.R. Chaitman, I.E. Dahms, S.O. Gottlieb, J.D. Hackney, M. Pagano,
R.H. Selvester, S.M. Walden and J. Warren, 1989. Short-term effects of carbon monoxide exposure
on the exercise performance of subjects with coronary artery disease. New England Journal of
Medicine 321, 1426-1432.

Allred, E.N., E.R. Bleecker, B.R. Chaitman, I.E. Dahms, S.O. Gottlieb, J.D. Hackney, M. Pagano,
R.H. Selvester,  S.M. Walden and J. Warren, 1991. Effects of carbon monoxide on  myocardial
ischemia. Environmental Health Perspectives 91, 89-132.
Atkins, E.H. and E.L. Baker, 1985. Exacerbation of coronary artery disease by occupational carbon

-------
CARBON MONOXIDE                                                 FINAL: 07/2008

monoxide exposure: A report of two fatalities and a review of the literature. American Journal of
Industrial Medicine 7, 73-79.

Caravati, E.M., C.J.  Adams, S.M. Joyce and N.C. Schafer, 1988. Fetal toxicity associated with
maternal carbon monoxide poisoning. Annals of Emergency Medicine 17, 714-717.

Chiodi, H., D.B. Dill, F. Consolazio and S.M. Horvath, 1941. Respiratory and circulatory responses
to acute carbon monoxide poisoning. American Journal of Physiology 134, 683-693.

Coburn, R.F., R.E. Forsterand P.B. Kane, 1965. Considerations of the physiological variables that
determine the blood carboxyhemoglobin concentration in man. Journal of Clinical Investigation 44,
1899-1910.
Crocker, P.J. and J.S.Walker, 1985. Pediatric carbon monoxide toxicity. The Journal of Emergency
Medicine 3, 443-448.

Crocker,  P.J. and J.S. Walker, 1985.  Pediatric carbon monoxide toxicity. Journal of Emergency
Medicine 3, 443-448.

Dahms, I.E., L.T. Younis, R.D. Wiens, S. Zarnegar, S.L. Byers and B.R. Chaitman, 1993. Effects of
carbon monoxide exposure in  patients with documented cardiac arrhythmias. Journal of the
American College of Cardiology 21, 442-450.

Ebisuno, S., M. Yasuno, Y.  Yamada, Y.  Nishino, M. Hori, M. Inoue and T. Kamada, 1972.
Myocardial infarction after acute carbon monoxide poisoning: case report. Angiology37, 621-624.

Grace, T.W. and F.W. Platt, 1981. Subacute carbon monoxide poisoning. Journal of the American
Medical Association  246, 1698-1700.

Haldane, J.,  1895. The action of carbonic acid on man. Journal of Physiology 18, 430-462.
Klasner,  A.E., S.R.  Smith,  M.W. Thompson and A.J. Scalzo, 1998. Carbon  monoxide mass
exposure in a pediatric population. Academic Emergency Medicine 5, 992-996.

Henderson, Y., H.W.  Haggard, M.C. Teague, A.L. Prince and R.M. Wunderlich, 1921. Physiological
effects of automobile exhaust gas and standards of ventilation for brief exposures. Journal of
Industrial Hygiene 3, 79-92.

Kizakevich, P.M., M.L. McCartney, M.J. Hazucha, L.H.  Sleet, W.J. Jochem, A.C. Hackney and K.
Bolick, 2000. Noninvasive ambulatory assessment of cardiac function in healthy men exposed to
carbon monoxide during upper and lower body exercise. Eurpean Journal of Applied Physiology83,
7-16.

Klasner,  A.E., S.R.  Smith,  M.W. Thompson and A.J. Scalzo, 1998. Carbon  monoxide mass
exposure in a pediatric population. Academic Emergency Medicine 5, 992-996.

Klees,  M., M. Heremans and S.  Dougan, 1985. Psychological sequelae to carbon monoxide
intoxication in the child. The Science of the Total Environment 44, 165-176.

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

Koren, G., R. Sharav, A. Pastuszak, L.K. Garrettson, K. Hill, I. Samson, M. Rorem, A. King and J.E.
Dolgin,  1991. A multicenter, prospective  study  of fetal outcome following  accidental carbon
monoxide poisoning in pregnancy. Reproductive Toxicology 5, 397-403.

Marius-Nunez, A.L., 1990. Myocardial infarction with normal coronary arteries after acute exposure
to carbon monoxide. Chest 97, 491-494.

Nelson.G. 2005a. Effects of Carbon Monoxide in Man. In: Carbon Monoxide and  Human
Lethality: Fire and Non-fire studies, M.M. Hirschler (Ed.). Taylor and Francis, New York, 2005,
pp 3-60.

Nielsen, B., 1971. Thermoregulation during work in carbon monoxide poisoning. Acta Physiologica
Scandinavica 82, 98-106.

Peterson,  J.E. and R.D. Stewart, 1975. Predicting the carboxyhemoglobin levels resulting from
carbon monoxide exposures.  Journal of Applied Physiology 39, 633-638.

Sheps, D.S., M.C. Herbst, A.L. Hinderliter, K.F. Adams, L.G. Ekelund, J.J. O'Neill, G.M. Goldstein,
P.A. Bromberg, J.L. Dalton,  M.N.  Ballenger,  S.M.  Davis  and G.G. Koch, 1990.  Production of
arrhythmias by elevated carboxyhemoglobin in patients with coronary artery  disease. Annals of
Internal Medicine 113, 343-351.

Sheps, D.S., M.C. Herbst, A.L. Hinderliter, K.F. Adams, L.G. Ekelund, J.J. O'Neill, G.M. Goldstein,
P.A. Bromberg,  M.  Ballenger, S.M. Davis and G. Koch, 1991. Effects of 4 Percent and 6 Percent
Carboxyhemoglobin on Arrhythmia Production in Patients with Coronary Artery  Disease. Research
Report No. 41, Health  Effects Institute, Cambridge, Massachusetts.

Stewart, R.D., J.E.  Peterson,  E.D. Baretta, R.T. Bachand, M.J. Hosko and A.A. Herrmann, 1970.
Experimental human exposure to carbon monoxide. Archives of Environmental /-/ea/f/?21,154-164.

-------

-------
CARBON MONOXIDE                                                 FINAL: 07/2008

1.      INTRODUCTION

       Carbon monoxide (CO) is a tasteless, odorless and colorless gaseous substance (WHO,
1999a). CO is produced by both natural and anthropogenic processes. The main source of CO
production is the combustion of fuels. The burning of any carbonaceous fuel produces CO and
carbon dioxide as the primary products. The production of carbon dioxide predominates when the
air or oxygen supply is in  excess of the stoichiometric needs for complete combustion. If burning
occurs under fuel-rich conditions, with less air or oxygen than  is needed, CO will be produced in
abundance (WHO, 1999a). Emission sources include gasoline-and diesel-powered motor vehicles,
stationary  combustion equipment, such as heating and power generating  plants, industrial
processes, such as blast furnace operation  in steel industry, indoor sources, such as gas ovens,
unvented kerosene and gas space heaters and coal and wood stoves, as well as wildfires and
tobacco smoking. Exposure at the workplace occurs in blast furnace operations in the steel industry
and when  gasoline- or propane-powered forklifts, chain-saws or other  machines are used  in
confined spaces, such as companies, tunnels and mines. Low concentrations are produced in the
atmosphere by reactions of hydroxyl radicals with methane and other hydrocarbons as well as by
the reactions of alkenes with ozone.

       In addition to exogenous sources, humans are  also exposed to small amounts  of CO
produced endogenously.  In the process of  natural degradation of hemoglobin to bile pigments,
oxidation of the tetrapyrrol ring of heme leads to opening of the  ring and formation of biliverdin and
CO (WHO, 1999a). The endogenous CO formation  leads to a background  carboxyhemoglobin
concentration  in blood (COHb) of about 0.5 to 0.8 % (NIOSH, 1972).

       Increased destruction of red blood cells, e.g. caused by hematomas, blood transfusion or
intravascular hemolysis, and accelerated breakdown of other heme proteins will lead to increased
production of CO. In patients with hemolytic anemia, the CO production rate was 2-8 times higher
and blood COHb was 2-3 times higher than  in  healthy individuals  (Coburn et al., 1966).

       Smokers are exposed to considerable CO concentrations leading to an average COHb of 4
%, with a usual range of 3-8 % (Radford and Drizd, 1982).

       Exposure to CO can also be caused indirectly by  exposure to  certain halomethanes,
particularly dichloromethane (synonym: methylene chloride), because these solvents are at least
partly metabolized oxidativelyto CO by cytochrome P450 (Gargas et al., 1986; see ATSDR, 1998
for review).

       Environmental exposure to CO can occur while traveling in motorvehicles, working, visiting
urban locations associated with combustion sources, or cooking and heating with domestic gas,
charcoal or wood fires, as well as by environmental tobacco smoke. WHO (1999a) summarized
environmental concentrations as follows: CO concentrations in ambient air monitored from fixed-site
stations are generally below 9 ppm (8-hour average). However, short-term peak concentrations up
to 50 ppm are  reported on heavily traveled roads. The CO levels in homes are usually lowerthan 9
ppm; however, the peak value in homes could be up to 18 ppm with gas stoves, 30 ppm with wood
combustion and 7 ppm with kerosene heaters. The CO concentrations inside motorvehicles are
generally around 9-25 ppm and occasionally over 35 ppm. Similar exposure levels were reported by
EPA (2000).

                                          1

-------
  CARBON MONOXIDE
                                                                    FINAL: 07/2008
TABLE 1: CHEMICAL AND PHYSICAL DATA
Parameter
Molecular formula
Molecular weight
CAS Registry Number
Physical state
Color
Synonyms
Density
Melting point
Boiling point
Solubility
Odor
Explosive limits in air
Conversion factors
Value
CO
28.01
630-08-0
gaseous
colorless
none
1 .250 g/l at 0 °C
1.145g/lat25°C
-199°C
-191.5°C
35.4 ml/l at 0 °C
21 .4 ml/l at 25 °C
odorless
12.5 % (LEL) to 74.2 % (UEL)
1 ppm = 1 . 1 45 mg/m3
1 mg/m3 = 0.873 ppm
Reference
WHO, 1999a
WHO, 1999a
WHO, 1999a
WHO, 1999a
WHO, 1999a

WHO, 1999a
WHO, 1999a
WHO, 1999a
WHO, 1999a
WHO, 1999a
WHO, 1999a
WHO, 1999a
  2.
HUMAN TOXICITY DATA
|        Based on older literature^ the COHb in the blood has been correlated with symptoms in
  healthy adults, shown in the left half of Table 2 (WHO, 1999a). Very similar tables or descriptions
  are found  in different publications (e.g. AIHA, 1999; Winter and Miller,  1976;  Holmes, 1985;
  Stewart, 1975; Roos 1994). However, with respect to both lethal and nonlethal effects of CO,
  susceptible subpopulations have been identified and effects on these are depicted in the right half
  of Table 2 for comparison (see subsequent sections for references). The unborn fetus and adults
  with coronary artery disease are considerably more susceptible for lethal effects of CO than healthy
  adults. For nonlethal effects of CO, subjects with coronary artery disease (increased frequency of
  arrhythmias and reduced time to onset of angina and to changes in the electrocardiogram and
  children  (syncopes, long-lasting neurotoxic effects) constitute susceptible subpopulations.

-------
CARBON MONOXIDE
                                                                 FINAL: 07/2008
TABLE 2: SYMPTOMS ASSOCIATED WITH COHb IN HEALTHY ADULT HUMANS AND
SUSCEPTIBLE SUBPOPULATIONS
Healthy Adults;
adopted from WHO, 1999a
COHb
(%)
••-'I
3-8
10
20
30
40-50
60-70
80
Symptoms
physiologic background concentration
background concentration in smokers
no appreciable effect, except shortness of
breath on vigorous exertion, possible
tightness across the forehead, dilation of
cutaneous blood vessels
shortness of breath on moderate exertion,
occasional headache with throbbing in
temples
decided headache, irritable, easily fatigued,
judgment disturbed, possible dizziness,
dimness of vision
headache, confusion, collapse, fainting on
exertion
unconsciousness, intermittent convulsion,
respiratory failure, death if exposure is long
continued
rapidly fatal
Susceptible Subpopulations
COHb
(%)
2
5-6
7
13
15
25
25

Symptoms
during physical exertion
reduced time to onset of angina
and electrocardiogram signs of
myocardial ischemia in subjects
with coronary artery disease
increase in cardiac arrhythmias
in subjects with coronary artery
disease
headache, nausea in children
cognitive development deficits
in children
myocardial infarction in subjects
with coronary artery disease
syncopes in children
Stillbirths

2.1.   Acute Lethality

      Mortality from CO  poisoning is high: for England and Wales,  1365 deaths due to CO
exposure were reported in 1985. In the USA, more than 3800 people annually die from accidental or
intentional CO exposure (WHO, 1999a).

-------
CARBON MONOXIDE                                                 FINAL: 07/2008

       Immediate death  from CO is  most likely caused by effects on the heart,  because the
myocardial tissue is most sensitive to  hypoxic effects of CO. Severe poisoning results in marked
hypotension and lethal arrhythmias, which have been considered responsible for a large number of
pre-hospital deaths. Rhythm disturbances include sinus tachycardia, atrial flutter and fibrillation,
premature ventricular contractions, ventricular tachycardia and fibrillation (WHO, 1999a).

       The susceptible subpopulations for lethal effects are subjects with coronary artery disease
and the unborn fetus (see Section 2.3). The review on death causes by Balraj (1984) shows an
association between coronary artery disease and relatively lowCOHb. A number of case studies is
presented in which CO exposure contributed to myocardial infarction (all cases of infarction are
presented in this section irrespective of whetherthe patients were rescued from death by intensive
medical care or not).

       The British Standards Institution (BSI, 1989) has published the following concentration-time
combinations as  lethal exposures to  CO (used for hazard estimation  in fires): 40000 ppm x 2
minutes, 16000 ppmx 5 minutes, 8000 ppmx 10 minutes, 3000 ppm x 30 minutes and 1500 ppmx
60 minutes. The International Standard Organization has published lethal exposure concentrations
of 12000-16000 ppm for 5 minutes and 2500-4000 ppm for 30 minutes (for an adult engaged in light
activity) (ISO, 1989).  From the documents it was concluded that the  published values are  for
normal, healthy adults and that the values were based on animal data (especially monkeys; Purser
and Berrill, 1983); the documents did not discuss the issue of subpopulations at higher risk for lethal
effects.

2.1.1   Case Studies

       Nelson (2005a) reported data from unvented space heaters related to human lethality cases
related to CO poisoning. Sixteen out of 22 lethal cases had  COHb levels more than 40%.  Six out
of 22 victims had COHb < 40% and 2/6 cases had  pre-existing conditions such as arteriosclerotic
disease and cardiorespiratory failure. A1942 fatality study reported by Nelson (2005a) summarized
COHb  data for 68 victims that were found dead in a gas-filled room or in a garage containing
exhaust gases at high concentrations.  CO concentrations were not provided. Sixty-seven percent
of the 68 cases died with 40-88% COHb levels.   Three-percent of the  cases died with 30-40%
COHb  levels. Summary  of another fatality study  from Poland showed  a similar trend of COHb
levels (Nelson, 2005a). Individual data were not provided and the CO source was not discussed.
However, the Polish study considered 321 lethal  CO poisonings from  1975-1976 and provided
COHb levels for survivors (n=220) and fatal cases (n=101). The survivors had a mean COHb level
of 28.1% (SD=14.1), whereas the lethal cases showed an average COHb level of 62.3% (SD=10.1).
 Over 80% of the survivors had COHb levels below 40%.  In contrast, around 90% of the deceased
had COHb levels above 50%. Similar percentages of survivors and deceased were  observed at
COHb levels between 40-50% with a slight increase in the number of survivors when compared to
that of the lethal cases.  These three  studies showed a trend that most lethal cases occurred at
COHb levels higher than 40% and that survivorship was likely to be seen at levels below 40%.

       Another study from the Center of Forensic Sciences in Canada evaluated 304 fatal cases
from 1965-1968 (Nelson, 2005a).  The mean lethal COHb level was 51 ±  12% with a majority range
between 40 and 59% and the highest single frequency range at 45-59%.  A report on CO exposure
from exhaust fumes in the state of Maryland during 1966-1971 showed COHb levels in the 40-79%

-------
CARBON MONOXIDE                                                 FINAL: 07/2008

range for 98% of lethal cases (Nelson, 2005a). The Institute of Forensic Medicine in Oslo reported
a study of COHb levels in automobile exhaust victims (n=54). The mean fatal COHb level was 70
percent and 40% was the minimum COHb level exhibited by less than 2% of the cases (Nelson,
2005a). Another forensic study (Nelson et al. 2005c) examining 2241 fatalities between the years
of 1976-1985 found that the mean COHb  level of all the cases was 64.20% with  a standard
deviation of 17.47.  The data showed that 34% of victims had COHb levels of less than 60%. Of
those who died in fires, 41 % had COHb levels of less than 60% compared to 22% of the non-fire
deaths.

       Pach et al. (1978; 1979) reviewed a cases of carbon  monoxide in the Toxicological Clinic,
Cracow, Poland in the years 1975-1976. Excluded from this study were mixed intoxications, e.g., by
CO and medicaments. Group A were 101 persons (60 men  and 41  women, mean age 48 +/-15
years) that had died from CO poisoning before arrival at the clinic.  Measurement of COHb and
autopsy was done on these subjects. Group B comprised 220 subjects (95 men and 125 women,
mean age  38 +/-18  years) that were treated for CO  poisoning. COHb was determined upon arrival
at the clinic.  Patients for which the time between the end of exposure and blood drawing at the
clinic was longer than 120 minutes (N = 62) were excluded from further analysis. For the patients,
the COHb  at the end of exposure was recalculated. Mean COHb values for Groups A and B were
62 +/-10 % and 28  +/- 14 %, respectively. In Group A, the  percentages of subjects with COHb
between 30-40, 40-50, 50-60, 60-70, 70-80 and 80-90 % were 2, 6, 26, 44, 21 and 2, respectively,
while 3, 25, 32, 24, 12, 3, 0.6 and 0.6 % of the patients in the corrected Group B had COHb values
between 0-10,10-20, 20-30, 30-40,  40-50, 50-60, 60-70 and 70-80  %, respectively. Within each
group no correlation  between  COHb and either sex,  blood alcohol  above 0.1% or poisoning
circumstances (accidental or suicidal) were found. Group A showed a higher percentage (34 %) of
subjects that were 60 years or older than Group B (13 %), while Group  B had a higher percentage
of subjects younger than 30.

       Grace and Platt(1981) reported two cases of myocardial infarction due to CO poisoning. In
the first case, a 67-year-old man  was exposed to  increased CO concentrations for about a few
weeks in his home due to a rusted-outflue of a gas-furnace. The man presented to the emergency
room after three days of persistent light-headedness with vertigo, brief stabbing anterior chest pain
that worsened with deep inspiration, a dry cough, chills and a mild headache. His wife experienced
similar malaise and dizziness that had been resolving over the past week. At the hospital, his
symptoms were explained with a  diagnosis of viral  syndrome, hypokalemia of unclear origin and
diabetes mellitus with diabetic peripheral and autonomic neuropathy. Ten days after discharge he
was seen in the emergency room with true vertigo, palpitations and nausea, but was sent home to
be  followed  up as  an outpatient. Four days  later he  returned to the  emergency  room after
development of rectal urgency and an explosive incontinent diarrheal stool, followed by a severe
crushing anterior chest pain. With the pain he collapsed on the floor. The electrocardiogram showed
an acute myocardial infarction. His COHb (measured on arterial blood gases) was 15.6 %, the level
of the patient's wife  was 18.1 %. The patient survived and recovered completely.

       In the second case, a 69-year-old  man came to the emergency room after awakening two
days earlier with confusion,  nausea and vomiting. He then passed out and awoke the next day in
the bathroom. He crawled to the living room, where he again  passed out for an undetermined
amount of time, awoke to open his door for fresh air, and then went  to bed. He later experienced
auditory and visual  hallucinations and phoned his neighbor for help.  An acute inferior myocardial

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

infarction with secondary mild congestive heart failure and chronic obstructive pulmonary disease
was diagnosed. During his hospitalization, his sister and daughter-in-law spent a night in his mobile
home. They arrived at the emergency room early the next morning with throbbing headaches,
vomiting and vertigo. Their COHb values were 28 and 32 %. A faulty gas water heater had caused
CO exposure. The patient survived and recovered completely.

      Atkins and Baker (1985) described two fatal cases of workers with severe atherosclerotic
coronary artery disease. The first worker (age not stated) was a shipping employee in a plant that
reconditioned steel dyes. A gas-fired furnace was used fortempering the dyes, but also for heating
the plant. One day the worker was found unconscious and resuscitation efforts at a nearby hospital
were unsuccessful. Autopsy showed a severe two-vessel coronary artery disease and old scarring,
and a COHb of 30 %. Four other workers of the plant complaining of nausea were seen in the
emergency room, but COHb was not obtained.  The second worker (age not stated) was operating a
bale press in a used-clothing company. As well as gas- and oil-fired  heaters, there were a number
of propane-fueled  forklifts  used to transport  bales of  clothing  and ventilation  was  poor.
Resuscitation was unsuccessful after his collapse. Autopsy revealed three-vessel coronary artery
disease and global subendocardial ischemia. Two blood samples showed COHb of 24.1 and 21.5
%. Five other workers from the  same company were also seen,  complaining of light nausea,
lightheadedness and headache. One was hospitalized with a COHb of 35 %; the others had levels
from 4.1 to 12.8 %.  CO measurement was performed in the company the next day and revealed
concentrations of 135-310 ppm. Concentrations were highest near forklifts (250-310 ppm) and near
the bale press (120-230 ppm), which was where the patient had been working at the time of his
death.

       Ebisuno et al. (1972) reported a case of myocardial infarction after acute CO poisoning in a
healthy  young man.  A  28-year-old male  ironworker was admitted  to  the  emergency  room
complaining of precordial pain.  Two  hours  before admission the patient had been exposed
accidentally to CO for about one hour while  working at  a  blast furnace. After the exposure he
experienced  a sense  of fullness  of  the   head  and precordial  pain  following  transient
unconsciousness. Blood samples two  hours  after the exposure contained COHb of 21 %. The
electrocardiogram was interpreted as an  acute anterior  myocardial infarction.  The coronary
arteriogram one month after onset of infarction showed no significant narrowing on both left and
right coronary arteries. The left ventriculogram showed a giant aneurysm in the apical  portion. At
operation from the ventricular aneurysmectomy, a massive transmural myocardial necrosis was
observed. After surgical treatment, the patient is free of symptoms.

       Marius-Nunez (1990) reported the  case of  a 46-year-old man, who suffered an acute
myocardial infarction  after CO exposure. He was found  unconscious in a doorway of a burning
apartment. Artificial  respiration  was initiated until arrival at the emergency  room.  The
electrocardiogram showed sings of myocardial infarction, which was confirmed by high levels of
cardiac enzymes in the patient's serum. Blood gas analysis revealed a COHb of 52.2  %. After 3
hours treatment with 100 % oxygen, the patient became alert and oriented, COHb was 23 %. After
7 hours, he was extubated and a COHb of 13.4 % was measured. The patient's medical profile was
negative for coronary  heart disease risk factors, such as smoking, hypertension, diabetes mellitus
or coronary artery disease.  A coronary  angiogram performed one week later failed to  reveal
evidence of coronary  obstructive lesions.

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

       Balraj (1984) reviewed all deaths that were certified by the Cuyahoga County Coroner's
Office from the years 1958-1980, wherein asphyxia by CO was the primary caused of death and a
natural disease was the ,,other" diagnosis or vice versa. During the 23-year period, 38 deaths were
certified. These were divided into two groups: Group 1  consisted of 28 cases where all diagnosis
including the abnormal COHb was documented by complete postmortem examination. Group 2
consisted of 10 cases where the diagnosis of the "other" condition was based on review of medical
records, including results of coronary angiogram, serum enzymes, and clinical history; autopsy was
not performed on these 10 cases. The Group 3 served for comparison and comprised all deaths
that occurred in individuals 35 to 86 years of age in whom the COHb was 60 % and more (n = 100).
A complete autopsy had been performed in each of these cases.

       Of the 28 cases in Group 1, the primary cause of death was asphyxia  by CO in 21 cases.
The "other" condition in 19 of these cases was atherosclerotic coronary artery disease. Of these, 8
persons had hypertensive cardiovascular disease and 2 had pulmonary emphysema in addition. In
the remaining 7 cases of this group, the primary cause of death was atherosclerotic coronary artery
disease and the "other" condition was asphyxia by CO. In Group 2 atherosclerotic coronary artery
disease was the primary cause of death and asphyxia by CO was the "other" condition in 3 cases.
In the remaining 7 cases asphyxia by CO was the primary cause of death and in all but one of these
cases, the "other" condition was atherosclerotic coronary artery disease, two of the individuals had
hypertensive cardiovascular disease in addition. The results are presented in  Table 3.

-------
CARBON MONOXIDE
                                                                   FINAL: 07/2008
TABLE 3: INCIDENCE OF ATHEROSCLEROTIC CORONARY ARTERY DISEASE AND COHb IN
FATALITIES THAT INVOLVED CO EXPOSURE;
adopted from Balraj, 1984

Total
Age (years)





COHb (%)


Delayed deaths
Coronary
atherosclerosis


Myocardial infarct

Heart weight (g)

30-40
41-50
51-60
61-70
71-80
81-90
10-30
40-50
60 and more

mild
moderate
severe
recent
old
415 and more
Number of cases
Group 1
28
1
1
7
10
5
4
14
4
0
10
2
2
24
1
4
20
Group 2
10
0
0
2
4
2
2
5
3
0
2
unknown
unknown
5
0
1
unknown
Group 3
100
22
31
28
10
6
3
0
0
100
0
89
5
6
0
2
13
2.2.    Nonlethal Toxicity
       Nonlethal effects of CO on humans have been reported in experimental studies in both
healthy individuals and in patients with coronary artery disease (see Section 2.2.1). Case studies
(see Section 2.2.2) are presented  for children and adults and identify  children  as another
susceptible subgroup for nonlethal CO effects.
2.2.1   Experimental Studies
2.2.1.1 Subjects with Coronary Disease

-------
CARBON MONOXIDE                                                  FINAL: 07/2008
       A large number of studies investigated the effects of low CO exposure (COHb <10 %) on
healthy individuals and high-risk groups. These experiments have been reviewed extensively by
WHO (1999a) and  EPA (2000). In healthy individuals, symptoms, such as decreases in work
capacity and decrements of neurobehavioral function, start at COHb of 5 % (WHO, 1999a; EPA,
2000; Hazucha, 2000). With respectto high-risk groups, studies evaluating ST-segment changes in
the electrocardiogram and cardiac arrhythmogenic effects in patients with coronary artery disease
will be presented here, because these gave the most consistent results and also were considered
most relevant for AEGL  derivation (for review see WHO 1999a; EPA, 2000).

       Caracteristic  points of  an  electrocardiogramm  are  the  P wave,  reflecting  atrial
depolarization, the QRS-complex, representing the ventricular muscle depolarization, and the T-
wave, reflecting ventricular muscle  repolarization. In the  normal electrocardiogramm,  the ST-
segment is isoelectric, resting at the same potential as the interval between the T-wave and the
next P wave. Horizontal depression or a downsloping ST-segment merging into the T-wave occurs
as a result of ischemia,  ventricular strain,  changes in the pattern of ventricular depolarization or
drug effects. In chronic ischemic heart disease, there may be moderate degrees of horizontal ST-
segment depression or a downward sloping ST-segment, flattening or inversion of T-waves and
prominent U-waves.  It  is difficult to define an abnormal  ST-segment depression in precise
quantitative terms. However, a myocardia ischemia has to be considered if the beginning of the ST-
segment is more than 0.5 mm (corresponding to 0.05 mV) below the isoelectric line and there is an
associated T-wave abnormality  (Wilson  et al., 1991).

       Allred et al. (1989a; b; 1991) conducted a multicenter study of effects of low COHb on 63
individuals with coronary artery disease.  Male subjects aged 41-75 (mean = 62.1 years) with stable
exertional angina pectoris  (diagnosis established  for >3 months;  no at rest symptoms) and a
positive stress test (measured  by  a greater than 1-mm change  in the ST-segment of the
electrocardiogram and occurrence of angina symptoms), were studied in three different test centers
using standardized test protocols. Only patients showing reproducible effects before and after a test
stay in the exposure chamber on the qualifying visit were included. On the subsequent exposure
days, the stress test was repeated before the exposure and if the result was not reproducible
compared to the qualifying visit, the visit was repeated on another date and at the second failure in
the pretest the subject was dropped from the study. Further evidence that these subjects had
coronary artery disease was provided by the presence of at least  one of the following criteria:
angiographic evidence of narrowing (-70  %) of at least one coronary artery, documented  prior
myocardial infarction or a positive stress  thallium test demonstrating an unequivocal perfusion
defect.
       All patients were tested three times on separate days in a double-blind fashion. On each of
the 3 exposure days, the subject performed a symptom-limited exercise test on a treadmill (pretest),
he was then exposed for 50-70  minutes randomly to air and to CO (subjects were exposed to CO
concentrations that were experimentally determined to produce end-exposure COHb of 2.2% or4.4
%; these COHb values were 10 % higher than the targeted  concentrations to compensate for the
CO loss during exercise) and afterwards he performed a second symptom-limited exercise test. The
mean exposure levels and ranges for the test environment were clean air(0 ppm), 117 ppm (range
42-202 ppm) for COHb of 2 % and 253 ppm (range 143-357 ppm) for COHb  of 4  %.  Gas
chromatographic measurements of COHb  were performed 1 minute afterthe pretest, after 30 and
40 minutes into exposure, at the end of exposure and 1 minute after the second stress test and

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

revealed postexercise COHb of 2.0+/-0.1 and 3.9+/-0.1 %, respectively. The time to onset of angina
and the time to 1-mm ST-segment change were determined for each test. The percent changes
following exposure at both 2 % and 4 % COHb were then compared with the same subject's
response to the randomized exposure to room air.
       When potential exacerbation  of the exercise-induced ischemia by exposure to CO was
tested using the objective measure of time to 1-mm ST-segment change, exposure to CO levels
producing COHb of 2 % resulted in a overall statistically significant 5.1 % decrease in the time to
attain this level of ischemia. For  individual centers,  results were significant  in one,  borderline
significant in one and nonsignificant in one center. At 4 % COHb, the decrease in time to the ST
criterion was 12.1% (statistically significant for all patients, the effect was found in 49/62 subjects)
relative to the air-day results. Significant effects were found in all three test centers. The maximal
amplitude of the ST-segment change was also significantly affected by the  carbon monoxide
exposures: at 2 % COHb the maximal increase was 11 % and at 4 % COHb the increase was 17 %
relative to the air day.
       At 2  % COHb, the time to angina was reduced  by 4.2 % in all patients (effects were
significant in two test centers and nonsignificant in one center). At 4 % COHb, the time was reduced
by 7.1 % in all  patients (effects were significant in one,  borderline significant in one and
nonsignificant in one center). The two end-points (time to angina and time to ST change) were also
significantly correlated.
       Only at 4  % COHb a significant reduction in the total exercise time and in  the  heart rate-
blood pressure product was found (this double product provides a clinical index of the work of the
heart and myocardial oxygen consumption).

       A number of other studies also evaluated the same endpoints. A reduced time to onset of
exercise-induced chest pain was reported at COHb  of 2.5-3.0 % (Aronow et al., 1972),  3 %
(Kleinman et al.,  1989), 2.9 % and 4.5 % (Anderson et al., 1973) and at 3.9 % (Kleinman et al.,
1998). No significant depression of the ST-segment was found at COHb of 3.8 % (Sheps et al.,
1987) and 3.9 % (Kleinman et al., 1998). WHO (1999a) has tried to explain the differences between
these studies by differences in experimental methodology and analysis of data and by differences in
subject populations and sample size.

       Sheps etal. (1990; 1991) assessed the effect of CO exposure on ventricular arrhythmias. 41
subjects with established coronary artery disease (36 men and 5 women) with a mean age of 62.8
+/- 1.1 years were analyzed. Patients were categorized based on arrhythmia frequency on the
training  day before,  during and 6 hours after exercise: 10 had no arrhythmias (0-2 ventricular
premature  depolarizations  (VPD)/h), 11  had low-level arrhythmias   (3-50  VPD/h),  11  had
intermediate-level arrhythmias (51-200 VPD/h) and 9 had high-level arrhythmias (>200 VPD/h). The
protocol was performed over 4 consecutive days. Day 1 was the familiarization session and
instructions were  given how to use the 24-hour ambulatory electrocardiogram recorder; a symptom-
limited maximal bicycle exercise test was done. Days 2 to 4 were exposure days with either pure
room air or CO (100 or 200 ppm) administered in a randomized double-blind fashion. COHb
measurements were performed before exposure, after 30 and 60 minutes into the exposure, at the
end of the exposure and before and after exercise using an IL-282 CO-oximeter. Exposures were
stopped when the target levels of 4 or 6 % COHb was reached. Exposure durations were 94.2 +/-
4.2 (SE) minutes  (range 40 to 170 min) for the 4 % level and 82.3 +/- 2.9 (SE) minutes (range 39 to
135 min) for the 6 % level. On all three test days, the mean pre-exposure COHb was 1.8 %.  The
post-exposure and post-exercise COHb measured were 1.46 and 1.36 % for air exposure, 4.01 and

                                          10

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

3.93 % forthe 4-% group and 5.91 and 5.02 % forthe 6-% group. Comparisons of arrhythmia data
were done at COHb of 1.41, 3.71 and 5.33 %, respectively.
       During the exposure period, the mean number of single VPD/h on the room air day was
significantly higher than on the 4 % COHb day, while no significant difference in the mean number
of VPD/h was noted between  room air and 6 % COHb exposure. When the baseline level of VPD
frequency was controlled for by calculating the difference between the  VPD frequency during
exposure and the VPD frequency before exposure, there was no significant difference between the
room air and 4 % COHb exposure.
       During exercise period, the frequency of single VPD/h was greater in the 6-% day than on
room-air day (167 +/- 38 vs. 127+/-28 VPD/h; p=0.03). This effect was still significant, when the
baseline VPD level was controlled for (117+/-S4 vs. 74+/-26, p=0.04). For this analysis, data from
subjects in the low, medium and high VPD frequency groups were pooled. The difference remained
significant when all subjects, including those categorized in the "no arrhythmia" group were included
in the analysis. The VPD frequency was  not significantly increased at 4 % COHb.

       The initial findings (essentially negative) of this  study in  10  patients with  ischemic heart
disease and no ectopy during baseline monitoring were also published separately (Hinderliteretal.,
1989).

       Dahms et al. (1993) studied 28 men and 5 women with documented  coronary artery disease
and a minimum of 30 ventricular ectopic beats per hour over a 20-h period studied. On three testing
days, the subjects were exposed in  a randomized double-blind fashion to either room  air or
sufficient CO to elevate their COHb to 3 or 5% in  1 hour. The mean exposure concentrations during
this hour were 159+/-25 ppmand 292+/-31 ppm, respectively.  This was followed by a maintenance
exposure to mean concentrations of 19.3 and 31 ppm, respectively,  for an additional 90 minutes,
which included the exercise test (after 60 minutes of equilibrium exposure) and immediate
postexercise phase. The subjects then left the laboratory and resumed their normal daily activity to
determine changes in ventricular ectopic beats  after CO exposure. To this end, continuous 20-h
ambulatory electrocardiograms were  obtained with the  recorder placed on the patients  2  hours
before CO exposure. There  was no significant change in the frequency of single ventricular ectopic
beats at rest from 115+/-28 (in room air) to 121+/-31 at 3 % and 94+/-2S at 5 % COHb. Exercise
itself increased the frequency of ventricular ectopic beats (from a baseline of 116 to 206 during
exercise and 375 during exercise recovery forthe room air exposure), but there was no additional
effect of CO exposure. Analysis of the data based on grouping of the subjects by the severity of
disease (ventricular  ectopic  beat frequency,  ejection  fraction,  presence  of exercise-induced
ischemia) indicated no proarrhythmic  effect of CO.

2.2.1.2 Healthy Adults

       Chiodi et al. (1941) exposed each of 4 male subjects (aged 21-33 years) repeatedly to CO
concentrations of 0.15-0.35 % (1500-3500 ppm) for 70 minutes  or longer. During 1 hour before
exposure, basal oxygen consumption, ventilation, pulse rate and blood pressure were recorded and
arterial blood for pH determination was obtained. The subject, remaining in rest during exposure,
then breathed  CO-containing air from a 600-liter gasometer. The  measurement of the above
mentioned parameters was  continued during exposure. In one set of experiments the test subjects
reached COHb between 3.4 and 10.4 % (8 experiments in total with the following COHb at the end
of exposure: 4.6, 6.3,  7.2, 9.2 and 9.8% in subject H.C. and 3.4, 9.5 and 10.4 % in subject  F.C.). In

                                          11

-------
CARBON MONOXIDE                                                 FINAL: 07/2008

another set of experiments, three subjects reached COHb of 27-52% at the end of exposure (in 11
of a total of 22 experiments COHb between 40 and 52 % were measured). The following COHb
values were measured at the end of exposure: 0, 31, 32,  32, 33, 39, 41, 42, 43, 45 and 52 % in
subject H.C., 0, 27, 35, 41, 43 and 48 % in subject F.C. and 0, 0, 41, 42 and 44 % in subject S.H.
No statement was made on whether any symptoms were observed. The cardiac output increased
20-50 % at COHb >40 %, while the changes were negligible at COHb of <30 %. No effects on the
other parameters measured were found.

       Henderson et al. (1921) exposed volunteers in a 6.4-m3 gas-tight, steel-walled exposure
chamber. CO was generated by dripping formic acid into strong sulfuric acid. A defined volume of
CO was led into the chamber and mixed with an electric fan. Analysis of the exposure concentration
in the chamber was done using the iodine  pentoxide method. Subjects (9 men and  1 woman;
number of subjects at each concentration  given in brackets) were exposed for 1 hour at 200 ppm
(2), 300 ppm (3), 400 ppm (11), 500 ppm (1), 600 ppm (9),  800 ppm (4), 900 ppm (1) or 1000 ppm
(1) CO. Blood samples were taken before exposure, at 30 minutes into the exposure, at the end of
the exposure (60 minutes) and once ortwice  during the next three hours after end ofthe exposure.
The COHb was determined using the carmine method. Directly after leaving the exposure chamber,
subjects breathed several times into a bladder bag and CO was determined in the exhaled air using
the iodine pentoxide method. CO concentrations in alveolar air after 60 minutes was  130-136 ppm
at an exposure concentration of 400 ppm, 120-230 ppm at 600 ppm and 140-230 ppm at 800 ppm.
The COHb percentage ranged from 11-12 % at 200 ppm, 10-14 % at 300 ppm, 14-22 % at 400
ppm, 16-26 % at 600 ppm, 26-34 % at 800 ppm, 34 % at 900 ppm and 38 % at 1000 ppm. After
exposure to up to 500 ppm for 60 minutes, no symptoms were observed. At 600 ppm, 2/9 subjects
reported slight frontal headache. At 800 ppm all subjects reported decided frontal headache during
4-8 hours. At 900 ppm insomnia and  irritability occurred in addition to headache. At  1000 ppm,
irritability, throbbing frontal headache and  at times Cheyne-Stokes breathing were observed. The
Romberg test (ability to stand erect with eyes closed) showed a marked loss of equilibrium after a
60-minute  exposure to 800 ppm or higher.

       Haldane (1895) reported  on a series of 11 studies  in which the author exposed himself to
different CO concentrations for different exposure times. The exposure conditions and effects are
summarized in the following Table 4. The subject breathed the CO atmosphere from a mouthpiece.
No mentioning of an analytical measurement of the exposure concentrations used was made. At the
end or one or more times during the  exposure, the exposure  was interrupted and the  subject
walked in the room or ran up a flight  of stairs (once or a few times) to investigate the effect of
physical exertion at different COHb levels. The COHb was determined colorimetrically by measuring
the amounts of carmine solution  that had to  be added to the diluted blood sample or to an equal
dilution of normal, oxygenated blood to adopt the color of a  CO-saturated blood dilution.  For COHb
<70 %, the author found his COHb determinations accurate within a 5 % error.  Although the
exposure measurement of this study does not meet today's standards, the reported COHb values
are in fairly well agreement with the values calculated from the given exposure concentration and
exposure time using the mathematical model  of Coburn, Forsterand Kane (see Section 4.4.4) when
assuming a resting ventilation rate (see Table 21 in Appendix B).
                                         12

-------
CARBON MONOXIDE
                                                                      FINAL: 07/2008
TABLE 4: EFFECTS OF ACUTE CO EXPOSURE IN A HUMAN SUBJECT;
adopted from Haldane, 1895
No
1
2
3
4
5
6
Exp. cone.
(vol. %) [ppm]
0.50 [5000]
0.39 [3900]
0.40 [4000]
0.36 [3600]
0.41 [4100]
0.12 [1200]
Total
exposure
time
(min)
11.5
30.5
24
29
29
120
Observations
no symptoms; hyperpnea after running upstairs
no symptoms
slight feeling of palpitation, pulse 102
palpitation, respiration 18, pulse 120, feeling abnormal
after running upstairs became giddy, much out of
breath, palpitations, slightly impaired vision
no symptoms except unusual hyperpnea and giddiness
after running upstairs
-
on walking throbbing in the head and palpitations, on
running giddy, short of breath
-
very slight hyperpnea and palpitations
after running marked giddiness and impairment of
vision and hearing (for 1-2 min)
-
slight tendency to palpitations, pulse 96
no symptoms
slight palpitations, sleepy
after running (no exposure) distinct dimness of vision
and hearing, slight tendency to stagger, abnormal
hyperpnea
slight hyperpnea while sitting
distinct hyperpnea, feeling uneasy, dull and abnormal;
after running: weak in the legs, markedly impaired
vision and hearing, confusion
at time (min)/
COHb (%)

15 min 723%
22 min
29 min
30.5 min 739
%
24 min 7 27 %
18 min 7 26%
29 min 7 37 %
15 min 7 13%
28 min
29 min 7 35 %
1 5 min 7 8 %
33 min
46 min 7 18%
67 min
90 min 7 27 %
104 min
120 min 7 37%
                                            13

-------
CARBON MONOXIDE
                                                                      FINAL: 07/2008
7
8
9
10
11
0.21 [2100]
irregular due
to
disconnected
tubing, 0.43 %
for last 10 min
0.027 [270]
0.021 [210]
0.046 [460]
71.5
35
210
240
240
-
very slight feeling of fullness, throbbing in the head
-
feeling decidedly abnormal, slight hyperpnea, marked
throbbing
breathing decidedly deeper, pulse 104
feeling decidedly abnormal, impaired vision, slight
feeling of giddiness
hyperpnea more distinct, beginning to look
pale/yellowish
-
feeling worse shortly after any movement in the chair
hyperpnea marked, slight confusion of mind
vision dim, limbs weak, difficulty in getting up and
walking without assistance; at 6 min after exposure stop
very unsteady walking, nearly falling, very indistinct
vision
hardly able to stand, no walking alone without falling
down
-
-
-
no symptoms; after running: very slight unusual
shortness of breath and palpitations
-
-
-
no symptoms
-
20 min 717%
34 min
40 min / 39 %
43 min
45 min
54 min
59 min
61 min 744.5
%
63 min
65 min
71 min / 49 %
35 min, 56 %
60 min / 7 %
120 min 711 %
180 min 715%
210 min 7 14%
60 min 7 8 %
120 min 7 13%
180 min 7 13%
240 min 7 13%
60 min 7 17%
                                            14

-------
CARBON MONOXIDE
                                                                  FINAL: 07/2008



-
-
no symptoms; after running: unusual hyperpnea, slight
palpitations
120min/28%
180min/28%
240 min / 23 %
       Stewart et al. (1970) performed 25 inhalation exposure experiments on a total of 18 healthy
men (age 24-42). These were exposed sedentary in an exposure chamber at <1, 25, 50, 100, 200,
500 or 1000 ppm for periods of 30 minutes to 24 hours. The chamber atmosphere was monitored
continuously by  infrared spectroscopy  and periodically by gas chromatography. The subjects
performed the following psychoneurological tests: hand and foot reaction time in a driving simulator,
Crawford collar and pin test, Crawford screwtest, hand steadiness test, Flanagan coordination test,
othorator  visual  test,  complete  audiogram,  resting   12-lead  electrocardiogram,  standard
electroencephalogram, visual evoked response and time estimation-hand reaction time test. No
subjective symptoms or objective  signs of illness were  noted during or in the 24-hour period
following exposure to 25 ppm for 8 hours, 50 ppm for 1, 3 or 8 hours, or 100 ppm for 1, 3 or 8 hours.
There was no detectable change from control values in the clinical tests. A significant relationship
between the Crawford collar and pin test  and CO concentration was considered a chance finding by
the authors. Of  11  subjects exposed to 200  ppm for 4 hours, 3 subjects  reported they had
developed a ,,mild sinus" headache in the final  hour. In the clinical tests, no detectable statistical
change from control values was observed. In the first exposure to 500 ppm for 1.8 hours, one of the
two subjects reported light-headedness after20 minutes of exposure, which was believed to be due
to his  hyperventilation. After 1 hour of exposure, both subjects were aware of a 10 % increase in
heart rate with the minimal exertion of walking to the blood port. After 90 minutes of exposure the
second subject noted the onset of mild frontal headache. During the second exposure to 500 ppm
for 2.3 hours, the same subjects both developed mild frontal headaches after 1 hour of exposure.
Minimal exertion caused a transient intensification of the pain.  Both  headaches remained mild
during the first postexposure hour, then they intensified into excruciatingly severe  occipitofrontal
headaches, reaching a pain  peak 3.5 hours after exposure, and persisted for 7 hours. During the
third exposure to 500 ppm, the occurrence of mild frontal headaches was noted after 1  hour of
exposure.  Immediately after exposure,  both subjects were placed in a hyperbaric chamber and
administered oxygen and the mild headaches were gone within minutes. The mean COHb reached
after 2.3-h exposure to 500 ppm was about 25.5%, after 4-h exposure to 200 ppm about 16.0 %
and after 8-h exposure to 100 ppm about 12.5 %.

       In another experiment (Kizakevich et al., 1994) evaluating cardiovascular responses of
exercising individuals, 16 health young men performed a sequence of brief (5 minutes) multi-level
treadmill and hand-crank exercises at <2 % COHb  and after attaining 5, 10,  15 or 20  % COHb on
different days . Non-invasive impedance cardiography was used to estimate cardiac output, stroke
volume, heart rate, cardiac contractility and time-to-peak ejection time. The electrocardiogram was
used to assess myocardial irritability and ischemia and  changes in cardiac rhythm. The results
showed that compensatory cardiovascular responses to submaximal upper-  and lower-body
exercise (e.g., increased heart rate, cardiac contractility, cardiac output) occur after CO exposures.
These changes were highly significant for exposures attaining 20% COHb. The authors concluded
that  healthy  young  men can  perform submaximal exercise without overt  impairment  of
cardiovascular function after CO exposures attaining  20 % COHb.
                                          15

-------
CARBON MONOXIDE                                                 FINAL: 07/2008
       Nielsen (1971) investigated the effect of CO exposure on thermoregulation. Experiments
were performed repeatedly on two subjects. Subject JHB reached COHb levels of 25% (mean of 8
experiments) and 33 % (4 experiments) and subject PJC reached 30 % (4 experiments). After
reaching the desired COHb  level, the subjects exercised on a chair-ergometer for 1 hour at a
medium to high workload (mean heart rate 120-170 beats per minute). The subjects were not
exposed continuously to CO  during exercise, but the COHb level was maintained by breathing a
calculated volume of CO from an anesthesia bag for 1-1.5 minutes every 15 minutes during
exercise. CO exposure  led to an increase in the plateau level of the deep body temperature during
exercise of 0.3-0.5 °C. The  lactic acid concentration  was not  increased after exercise  at air
exposure (120 mg/l in JHB and 79 mg/l in PJC),  but increased during CO exposures (309-660 mg/l
in both subjects). The authors stated neitherthe absence northe presence of any symptoms of CO
exposure.

2.2.2   Case Studies
2.2.2.1 Children

       Klasner et al. (1998)  published a retrospective  chart review on a mass poisoning at an
elementary school. The CO leak was discovered at noon, about 4 hours after school started. Of the
564 people at  school,  504 were children. Any child who showed evidence or  complained  of
symptoms was sent to  a hospital by ambulance or school bus. 177 children (mean age 8.7+/-1.8
years, range 4-12 years) were taken to one of three hospitals. All children were given 100 % oxygen
by face mask in the hospital (the authors stated that only few of these received simple face mask
oxygen en route to the  hospital). The level of poisoning  was assessed according to standardized
poison center data sheets (TESS,  toxic exposure surveillance sheets) and was recorded as
unknown (n=6), no effect (n=16), minor effect (n=124) or moderate effect (n=30). One child, for
whom the data sheet classification was that of a  major effect, was considered miscoded by the
authors because the medical  record showed that this child was sent home from the hospital without
further treatment. Symptoms  were present in  155 children and a mean COHb of 7.0 % (95 %-C.I.
6.6-7.5 %) was measured in a total of 147 children (blood was drawn at the same time oxygen
therapy began). The authors estimated that the children were exposed at least 60 minutes (in some
cases 90 to 120 minutes) to fresh air prior to obtaining their initial COHb. In the 177 children the
following symptoms (number of mentionings) were observed (some children reported more than
one symptom):  headache (139),  nausea (69),  dizziness (30),  dyspnea (19),  vomiting (13),
abdominal pain (11), drowsiness (9), other symptoms (0). The authors found a correlation between
the total number of symptoms reported and the COHb, such that children with higher COHb were
slightly more likely to report more symptoms. The authors did not mention how many of the 60
adults  experienced symptoms, but stated that symptomatic adults were taken adult  hospital
facilities.
                                         16

-------
CARBON MONOXIDE
                                                                  FINAL: 07/2008
       Crocker and Walker (1985) analyzed 28 patients with CO poisoning that were 14 years old
or younger. 25/28 CO exposures were secondary to faulty venting or faulty combustion of gas
furnaces, 2/28 were secondary to faulty combustion of a gas stove and 1/28 to motor vehicle
exhaust.  12 patients had COHb of less than 15 % and were completely asymptomatic. These
patients were considered to have nontoxic exposures, and they were not studied further. Of the 16
patients (mean age 7.0+/-3.8 years, 3 children were younger than 5 years) with  COHb of 15 % or
higher, 16/16 experienced nausea, 12/16 experienced associated vomiting, 13/14 (no information
on 2) complained of headache and 11/16 patients were reported to be lethargic.3/14 patients
reported visual problems, such as blurred or double vision. 9/16 reported at least one syncopal
episode with an average COHb of 31.6 % and a threshold level of 24.3 %. Every patient with a
COHb of 24.5% or higher experienced syncope. Lethargy was reported in  11/16 patients at a mean
COHb of 25.9 % and a threshold of 18.6 %. Symptoms and COHb are presented in Table 5. All
patients were  successfully treated with hyperbaric oxygen. The authors  provided the COHb
measured after hospital admission, but did not give any information on the delay between the end
of exposure and measurement and on (probable) oxygen administration before hospital admission,
e.g. oxygen by face mask during ambulance transport.
       Patient follow-up utilizing parental telephone interview and medical record review 3-12
months after the poisoning was used  to screen for neurologic  sequelae. Three patients had
developed problems: a 12-year-old boy with 36.1 %COHb had developed chronic headaches, a 6-
year-old girl with 36.9 % COHb had developed memory difficulties after suffering a major motor
seizure during  the poisoning episode and an 8-year-old girl with 24.5 % COHb developed poor
school performance, which were attributed to her long-standing poor reading ability; psychological
evaluation revealed no cognitive deficits. The former two children reported complete resolution of
their symptoms by 9 months post exposure.
TABLE 5: SYMPTOM THRESHOLD VALUES FOR PEDIATRIC CO TOXICITY; adopted from
Crocker and Walker, 1985
Symptom
none
Nausea
Vomiting
Headache
Lethargy
Visual symptoms
Syncope
Seizures
Threshold COHb (%)
<15
16.7
19.8
16.7
18.6
24.5
24.5
36.9
Average COHb (%)
<15
27.1
29.4
28.3
25.9
32.5
31.6
36.9
Percentage of patients * (%)
100
100
78.6
91.6
78.6
25.0
64.3
6.3
* The percentage of patients showing the respective symptom refers to the 16 patients with COHb >15 '
except for asymptomatic patients ("none"), which refers to the 12 patients with COHb <15 %.
                                          17

-------
CARBON MONOXIDE                                                  FINAL: 07/2008
       Kleesetal. (1985) investigated the neurotoxic sequelae of CO poisoning in children that had
been brought to the emergency department of St. Pierre Hospital, Brussels following CO poisoning
(irrespective of whetherthey were subsequently hospitalized or not). Cases were only studied when
follow-up was possible: in a short-term follow-up of 20 children that were submitted to psychological
tests at the time of the intoxication and who were re-examined again about 3 months later, and in a
long-term follow-up of 14 children that were re-examined between 2-11 years after the intoxication.
The authors listed the COHb measured after hospital admission, but did not give any information on
the delay between the end of exposure and measurement neither did they indicate a (probable)
oxygen administration before hospital admission,  e.g.  oxygen  by face mask during ambulance
transport.
       In the long-term follow-up, 6 of the 14 children (age 2.8-12.1 years at the time of intoxication;
mean age  7.8 years)  exhibited serious disorders (spatial  organization  problems, constructive
apraxia, deterioration of lexical activity,  as well  as spelling and arithmetic); two of them had a
previous history of psychological difficulties, but displayed additional difficulties afterthe poisoning.
COHb between 13 and 32 % (mean 21 %) have been reported for 4/6 children (no data on the other
2 children were available). Seven of the 14 children (age  >6 years, except for one 3.5-year old child;
mean age 9.8 years) exhibited slight impairment of visual memory and concentration; these children
had COHb between 16 and  26 % (mean 22 %). One child of this group  did not display any
sequelae.
       In the short-term follow-up, the authors grouped the 20 children according to age. In children
below 3 years of age (n=6, 2.0-2.9 years), medium intoxications (n=5, symptoms included  loss of
consciousness, but no coma; reported COHb 16-27 %)did not produce manifest sequelae except
fora momentary standstill in the child's progress of about 2 months, but their negative behavior was
found to be amplified (children were more nervous, more irritable, more anxious); however it was
not possible to determine if these behavioral disturbances were a direct effect of the CO intoxication
ofwhetherthey were due to neurophysiological causes ortothe stressful psychological conditions
surrounding the intoxication. In one case of severe intoxication (symptoms included  coma;  COHb
37 %) developmental level regression (motricity and language), violent anger and nervosity were
observed.
       In children  from 4 to 9 years old (n=8),  the  intoxication did  not alter  the intellectual
capacities, but in 6 cases (reported COHb of 4, 6, 25 and 27 %;  missing data  for two children) the
mnesticand instrumental aspects of the cognitive development were modified (the othertwo were
difficult to  evaluate due  to  intellectual  retardation and language retardation).  Visuo-spatial
perceptions and topographical memory were particularly perturbed, as was auditory memory.
       In children over 10 years of age (n=10), difficulty in perceiving and organizing the material to
be memorized either auditory or visually was found in the three children less than 12 years (COHb
of 26, 27 and 36 %). With the three children over 14 years, one case (30 % COHb) of serious
balance impairment was observed and only some slowness and instability with  the othertwo (COHb
of 26 and 30%).

       Meert et al. (1998) evaluated clinical characteristics and neurologic outcome  of all children
with CO poisoning admitted to the  Children's Hospital of Michigan,  Detroit between January 1987
and December 1996. Exposures were categorized as 1) severely toxic when COHb was >25 %, 2)
toxic when COHb was between 10.1 and 25 %, 3) suspected toxic when COHb was <=10 % with
acute neurologic manifestations, or 4) nontoxic when COHb was <=10 % without acute neurologic
manifestations. Of 106 cases (median age 3.5 years, range 0.1 to 14.9 years) were investigated, 37

                                          18

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

with severe toxic, 37 with toxic, 13 with suspected toxic and 19 with nontoxic exposures. The most
common  presenting   symptoms  included   altered   level   of   consciousness  (lethargy,
unresponsiveness),  metabolic  acidosis,  tachycardia and hypertension. All  exposures were
accidental, occurring as a result of smoke inhalation during house fires in 95, motor vehicle exhaust
in 6 and defective heating system in 5 cases. Forty-three children had an associated cutaneous
burn injury. All patients received normobaric oxygen fora median period of 5.5 hours (range 0.6 to
44 hours). Fifteen patients died, 8 from hypoxic-ischemic encephalopathy after cardiopulmonary
arrest at presentation, 3 from massive burn injury and 4 from late complications of burn injury. Nine
survivors suffered neurologic sequelae: 1) 6 had persistent deficits,  such as cognitive and motor
deficits or developmental delay (of these  4  had presented with respiratory or cardiorespiratory
arrest with COHb between 31.5 and 45 % and the other two; COHb 14.8 and 5.9 %; had severe
burns with  40  and 75  %, respectively, of the body surface  area  affected) and 2) 3  patients
developed delayed neurologic syndromes (2  children; COHb  33.3  and 34.8 %; with  transient
tremors,  cognitive  deficits  and hallucinations  starting  after  4 and  14 days,  that  resolved
spontaneously after about 2 months, and 1 child; COHb 3.1 %; that developed deficits in  cognitive
and interpersonal skills after 51 days and in whom brain imaging revealed bilateral occipital lobe
infarcts).

       Further information on pediatric CO poisoning can be found in the review of White (2000).

2.2.2.2 Adults

       Burney et al. (1982)  reported an epidemiologic and clinical investigation of 184 persons
exposed to  CO in a public school. CO release was from a furnace and was caused because of a
door to the exhaust chamber had been inadvertently left ajar. The CO was distributed throughout
the school building by a forced air heating system. Exposure began at 7.30 a.m. and  ended at
10.00 a.m. Of the 184 exposed persons (146 students and 38 teachers, mean age for all exposed
was 20 years) 160 became ill and 96 were transported to four hospitals fortreatment. COHb levels
were  measured on 66  persons and showed a mean of 18.2+7-6.4 %, with almost half falling
between 21 and 25 %. Persons in whom COHb levels were drawn had a mean exposure time of
107+/-33 minutes. Of the 160 persons who became ill, the following  symptoms were reported for
159 persons: headache (90 %), dizziness (82 %),  weakness (53  %), nausea (46  %), trouble
thinking (46 %), shortness of  breath (40%), trouble with vision (26%),  and loss of consciousness (6
%). For headache, dizziness, muscle weakness,  trouble with vision  and trouble with thinking, a
strong correlation between symptom and duration of exposure was found, while nausea, shortness
of breath and loss of consciousness did  not show this correlation. The authors corrected  the
measured COHb level for the delay between exposure  and the drawing of blood samples and
reported a corrected mean COHb of 20.7+/-7.0 %.

       Ely etal. (1995) reported a poisoning incident in a warehouse of a small sewing company. A
propane-fueled forklift was in use in the warehouse, in which a total of 30 people worked. The
forklift was  parked in a position where its exhaust focused directly into an air intake duct, that
communicated with a vent opening  above a table in the inspection  and packing area, where 5
people worked. On the day of the  incident, one man reported pronounced nausea,  vomiting,
dizziness and had a tonic-clonic seizure. Simultaneous, other coworkers developed chest pain and
dyspnea. The warehouse was evacuated immediately. Air CO measurements were 386 ppm in the
sewing area and 370 ppm in an unrelated work area. Thirty persons with complaints of severe

                                          19

-------
CARBON MONOXIDE
                                                                  FINAL: 07/2008
headaches (93 %), dizziness (63 %), weakness (63 %), nausea (60 %), chest pain ortightness (57
%), shortness of breath (50 %), vomiting (37 %), abdominal pains (33 %), muscle cramping (30 %),
difficulty concentrating (23 %), visual changes (20 %) and confusion (17 %) were treated for CO
exposure. Twenty-six patients had expiratory CO analyses after being treated with 100 % oxygen
for over 2 hours. Expiratory CO was higher in those from the inspection and packing area (21.1+/-
0.7 % versus 8.4+7-4.8 %).  These  persons were among the most severely ill.  The authors
extrapolated the mean expiratory CO concentration of 21.1 % back to a COHbof about 35 % at the
end of exposure. Two years after the incident, follow-up was obtained for25 (83%) of the patients:
11 (44 % of those reached) reported seeing physicians for persisting symptoms (numbness in arms
or legs, 36 %; restlessness, 36 %; persistent headaches, 32 %; irritable or violent behavior, 16 %;
confusion, 16%; incontinence, 16%; difficulty walking or moving arms/legs, 16%; memory loss, 16
%; difficulty speaking, 4 %).

       Sokal and Kralkowska  (1985)  analyzed  39 patients (18 men, 21 women) that were
hospitalized foracute CO poisoning. 25 patients were intoxicated by household gas and 14 patients
by coal-stove gas. The patient's ages ranged from 13 to 78 years. The duration of the poisoning
varied between 1 and 14 hours  and was established on the basis of an epidemiological review of
the circumstances of poisoning. The severity of poisoning evaluated  on admission to hospital
according to the clinical criteria presented in Table 6. On basis of the clinical criteria, 16 cases were
classified as degree I, 12 as degree  II, 8 as degree III and 4 as degree IV. For statistical analysis
the mild and moderate cases (I and II) were pooled into one group and the severe and very severe
cases (III and IV) into another. Results presented in Table 7 show that mean COHb in severe and
very severe poisonings were only slightly higher (not statistically significant) than those in the mild
and moderate group. On the other hand, the average duration of exposure which induced severe or
very severe poisonings was  about  twice  as long as that  associated with mild and moderate
poisonings. In the severe and very severe poisonings, the lactic acid concentration in blood, as an
indicator of metabolic acidosis, was significantly higher. For pyruvate and glucose concentrations
no significant differences were found (not shown).
TABLE 6: SEVERITY OF CO POISONING; from Sokal and Kralkowska, 1985
Grade I (mild)
Grade II (moderate)
Grade III (severe)
Grade IV (very
severe)
headache, vomiting, tachycardia, no disturbances of consciousness
disturbances or loss of consciousness without other neurological symptoms,
tachycardia, pain-induced reflexes still intact
loss of consciousness, intense muscular tonus, neurological symptoms,
tachycardia and tachypnea, circulatory and respiratory disturbances not
observed
loss of consciousness, clinical signs of central nervous system damage,
circulatory and respiratory disturbances
                                          20

-------
CARBON MONOXIDE
                                                                  FINAL: 07/2008
TABLE 7: COHb, EXPOSURE DURATION AND LACTATE CONCENTRATIONS IN RELATION TO
SEVERITY OF CO POISONING; from Sokal and Kralkowska, 1985

COHb (%)
Exposure duration
(h)
Blood lactate
concentration
(nmol/ml)a
Mild and moderate
poisonings (I and II)
(n)
27+7- 12
(27)
4.6 +/- 3.3
(27)
4.1 +1-3.6
(27)
Severe and very severe
poisonings (III and IV) (n)
34+7- 13
(11)
9.1 +1-3.5
(12)
8.8+7- 3.1
(11)
Very severe
poisonings (IV) (n)
31 +7-14
(3)
10.3+7-1.3 (4)
1 1 .0 +7- 2.2 (3)
a Blood lactate concentrations in 12 control individuals was 1.4 +7- 0.3 ^mol/ml.

       Deschamps et al. (2003), in a prospective study, measured effects on memory one month
after an acute CO intoxication. Of all patients examined in the hospital for suspicion of acute CO
intoxication over 4 years (N=944), 230 patients fulfilled the inclusion criterion of a COHb level of 11
%  or higher in the first blood sample measured at the hospital. After applying further inclusion
criteria,  i.e., age between 18 and 60, fluent in French language, no disease or risk factor which
might impair memory, e.g., excessive alcohol consumption, treatment with psychotropic drugs, drug
abuse, neurological or psychiatric diseases and exposure to solvents or heavy metals, 38 patients
were suitable for inclusion, of which 32 were examined. The median COHb in the first blood sample
was 23 %. Median blood  CO at the end of exposure was calculated as 30 %. The median number
of  days  between intoxication and psychometric testing was 31. Each patient was paired with a
control with respect to gender,  age and educational level. Tests were selected to study several
types of memory,  i.e.,  long term and working memory (verbal Buschke's  test) and  short term
memory (digit  span  (verbal) and Corsi's test (visual)). Other tests addressed disturbances of
attention (simple reaction time test, verbal fluence test) and divided attention (reaction time test with
double task and color/word decoding test). The only tests indicating a lower performance of patients
were for number recall and fatagability (mean reaction time was higher for the second part of the
trial than for the first part. The results did  not correlate with the end-of-exposure COHb. In several
other tests, patients  showed a better performance than controls, some  of these tests showed a
positive  correlation between result and the end-of-exposure COHb. The authors concluded that one
month afterthe incident, the memory of the  patients was not lowerthan in paired controls, and was
even higher for learning and word recall.

2.3.    Developmental/Reproductive Toxicity

       Koren et al. (1991) described a prospective, multicenter study of acute CO poisoning during
pregnancy. Between December 1985 and March 1989, a total of 40 cases of CO poisoning during
pregnancy were collected. All pregnant women were in good health  prior to the CO poisoning and
had not suffered from a known chronic illness. The 40  pregnancies included 3 twin births, 1
termination of  pregnancy at 16 weeks of gestation,  and 4 births that were pending. The CO
                                          21

-------
CARBON MONOXIDE
                                                                  FINAL: 07/2008
poisoning was caused by malfunctioning furnaces (n = 23), malfunctioning water heaters (n = 7),
car fumes (n  = 6),  methylene chloride exposure (n = 3)  and yacht engine fumes (n = 1). The
exposure occurred during the first trimester (n = 12), second trimester (n = 14) or third trimester (n =
14). The clinical grade of poisoning was based on clinical signs and symptoms as shown in Table 8.
Cases in which COHb values were available or could be estimated from the known ambient CO
concentrations are presented  in Table 9. Adverse fetal outcome occurred only after Grade 4 or 5
poisoning.
TABLE 8: SEVERITY OF CO POISONING; adopted from Keren et al., 1991
Grade 1
Grade 1 +
Grade 2
Grade 3
Grade 4
Grade 5
Alert, oriented, headache, dizziness, nausea
As Grade 1 , but another person exposed in the same incidence was unconscious
Alert, alterations of mental state, more pronounced headache, dizziness, nausea
Not alert, disorientation, loss of recent memory, muscle weakness, incoordination
Disoriented, depressed sensorium, limited and inappropriate response to simple
commands
Comatose, responding only to pain or not responding to any stimulus
TABLE 9: OVERVIEW OF CLINICAL SCORING, COHb AND FETAL OUTCOME;
adopted from Koren etal., 1991
Grade
5
5
4
4
4
2
1
1
1
COHb (%)
40-50
26
39
25
21
13.8
18
14
6.2
Time of exam
after exposure
(h)
2
1
2
2
2
1
unknown
unknown
1.5
Treatment a
HfO, 2 h
HfO, 3 h
HybO, 2 h
HfO, 2 h
HybO, 2 h
HfO, 7 h and
HybO, 2 h
HfO, 12 h
none
none
Outcome
Elective termination (in the text the
authors state: fetal death at term
followed by maternal demise)
Stillborn
Normal
Cerebral palsy compatible with
postanoxic encephalopathy
Normal
Normal
Normal
Normal
Normal
                                          22

-------
CARBON MONOXIDE
                                                                 FINAL: 07/2008
1
1
1
2.4
0.8
2
unknown
1
unknown
none
none
none
Normal
Normal
Normal
Cases with indirect measures of exposure
1 +
1 +
1
1
1
1
32,
measured in
affected son
32
32
14
14
5
2
-
-
-
-
-
HfO, 12 h
none
none
none
none
none
Normal
Fetal bradycardia
Normal
Normal
36-week gestation
Normal
' HfO = high-flow oxygen; HybO = hyperbaric oxygen
       Caravati et al. (1988) reported on six cases of acute CO poisoning during pregnancy (all
cases of patients with CO poisoning during pregnancy admitted to two teaching hospitals in Salt
Lake City during a two-year period). Results of COHb measurements and outcomes are given in
Table 10.  Cases 5 and 6 were  treated with  100 % oxygen for 5 hours before the COHb
measurement, which is between 3  and 4 half-lifetimes of CO under this condition, using a half-life
time of 80 minutes for treatment with  100 % oxygen  (Peterson and Stewart, 1970).  It can be
concluded that the end-of-exposure COHb values were about 8-16 fold higher and thus were about
40-80 % in Case 5 and  22-44  % in Case 6. In conclusion,  the three cases of stillbirths  were
associated with maternal COHb concentrations of 22 % or higher.
                                         23

-------
CARBON MONOXIDE
                                                                      FINAL: 07/2008
TABLE 10: OVERVIEW OF MATERNAL CLINICAL EFFECTS, COHb AND FETAL OUTCOME;
adopted from Caravati et al., 1988
Case




1
28-year-
old,
pregnancy
week 20






2
32-year-
old,
pregnancy
week 16


3
1 9-year-
old,
pregnancy
week 30


4
1 8-year-
old,
pregnancy
week 41




5
20-year-
old,
pregnancy
COHb
(%)



9.6










23






39






32








5



Time
between end
of exposure
and blood
sampling (h)
8










not stated






not stated






not stated








5 hours with
oxygen
treatment

Treatment




1 00 % oxygen by
face mask for 10 h;
then COHb had
reduced to 1 .7 %







1 00 % oxygen by
face mask for 10 h;
after 2.5 and 9.5 h
COHb was 8.9 and
1 .8 %, respectively.


1 00 % oxygen by
face mask for 8 h;
after 5 h COHb had
reduced to 4 %



oxygen treatment
using iron lung







1 00 % oxygen by
face mask during
ambulance and
helicopter transport
Maternal Effects and Fetal
Outcome



Poisoning was caused by a gas-leak
in the restaurant where the woman
worked; during a 6-hour working
period, she developed severe
headache, nausea and dizziness; she
visited hospital 6 hours later with
persisting headache, lethargy and
dizziness; she was discharged in
good health and delivered a normal
female infant weighing 2900 g four
months later.
Poisoning was caused by clogged
furnace; she complained of
headache, nausea and dizziness of
48 hours duration; she was
discharged 36 hours later in good
health and delivered a term healthy
male infant weighing 2920 g.
Poisoning was caused by a
malfunctioning heater; after 18 hours
exposure she complained of severe
headache and nausea; she was
discharged after 8 hours of oxygen
therapy and delivered a healthy 3940-
g male infant.
The woman was found unconscious
and was combative on arrival in the
emergency department; her mental
status rapidly improved an she
recalled having nausea, vomiting and
headache earlier that day; fetal heart
tones were absent and the woman
delivered a stillborn female infant the
next day.
The woman was found awake outside
her home together with case 6; they
had occluded the furnace the evening
before to improve heating; she
                                             24

-------
CARBON MONOXIDE
                                                                FINAL: 07/2008
week 38
6
1 8-year-
old,
pregnancy
week 13














2.8


















5 hours with
oxygen
treatment















to the hospital
1 00 % oxygen by
face mask during
ambulance and
helicopter transport
to the hospital













delivered a stillborn 3380-g male fetus
36 hours later.
The woman was found unconscious
together with case 5; fetal heart rate
was 1 36 per min at the scene and
1 90-200 per min 5 hours after the
exposure; after 5 hours, she was
somnolent but oriented and regained
full mental alertness during the next 2
hours; fetal heart rate decreased to
1 50-1 60 per min the next day and the
woman was discharged; she
delivered a nonviable 1210-g fetus at
33 weeks of gestation; autopsy
revealed brachycephaly,
craniosynostosis, multiple organ
cavity anomalies, multiple
contractures of extremities,
hypoplastic lungs and a small brain
with hydrocephalus.
       Farrow et al. (1990) reported a case of fetal death in a 20-year-old woman, who was
exposed to CO due to use of a portable propane heater in her unventilated mobile home. She
arrived by ambulance at the hospital approximately 60 minutes after being found unconscious at
her mobile home.  En  route to hospital she had  been  intubated and  had  received 100  %
supplemental oxygen. Her measured COHb at the time of admission was 7 %. On the second day
in  hospital,  the patient delivered a 1050-g stillborn  female fetus. On gross autopsy, bright red
discoloration of the skin and visceral organs was noted. A fetal COHb of 61 % was measured. The
authors assumed that the  mother had reached a minimal COHb of 40 to 50 % since she was found
unconscious.

2.4.    Genotoxicity

       No studies documenting genotoxic effects of CO in humans were located in the available
literature.

2.5.    Carcinogenicity

       No studies documenting carcinogenic effects of CO in humans were located in the available
literature.

2.6.    Summary

       In healthy adults, death from CO poisoning occurs at COHb larger than 50 % (AIHA, 1999;
WHO, 1999a; Steward, 1971; Steward etal, 1970; Pachetal., 1978; 1979). At COHb of about 16%
headaches can develop (Steward etal., 1970). Subtle (non-adverse) effects, such as decrements in
neurobehavioral function start at about 5 % COHb (WHO, 1999a; EPA, 2000).
                                         25

-------
CARBON MONOXIDE                                                 FINAL: 07/2008
       Analysis of lethal cases reported by Nelson (2005a) indicated that most lethal poisoning
cases occurred at COHb levels higher than 40% and that survival of CO-exposed humans were
likely to be seen at  levels  below 40%.  Persons with coronary artery disease constitute  a
subpopulation that is much more susceptible to the effects of CO. Case reports indicate that death
through myocardial infarction can occur at COHb around 20-30 % and as low as about 15 % in this
group (Balraj, 1984; Atkins and Baker,  1985; Ebisuno et al., 1986; Grace and Platt,  1981). In
individuals with coronary artery disease, a COHb of 2.0 or 4.0 % can significantly reduce the time to
onset of angina and the time to 1 -mm ST-segment change in the electrocardiogram during physical
exercise (Allred etal.,  1989a; b; 1991). At 5.3 %, but not at 3.7 % COHb an increased arrhythmia
frequency was observed in subjects with coronary artery disease (Sheps et al., 1990; 1991).

       Children and the unborn also constitute susceptible subpopulations: Measured COHb of
higher than
22-25% in the mothers' blood may lead to stillbirths (Koren etal., 1991; Caravatietal., 1988). After
CO poisonings associated with mean COHb of 21 % (range 13-32%) irreversible neurotoxic effects
resulting in defects in the cognitive development and in behavioral alterations were observed in a
long-term follow-up study, especially in young children (mean COHb 21 %) (Klees et al., 1985).
Acute symptoms of CO poisoning in children include effects, such as nausea, vomiting, headache
and lethargy. These symptoms were  reported to occur already at a COHb of 7 % in one study
(Klasner et al., 1998), while in another study a threshold of 16.7-19.8 % COHb was found (Crocker
and Walker, 1985). Visual symptoms and syncopes occurred at a threshold of 24.5 % COHb, at
higher COHb every child experienced at least one syncope (Crocker and Walker, 1985).
3.      AN IM AL TOXICITY DATA
3.1.    Acute Lethality

       Lethality data for acute inhalation exposure have been reported for rats, mice and guinea
pigs. The lethality data are summarized in Table 12 and graphically presented in Figure 1.

3.1.1.  Rats

       E.I. du Pont de Nemours and Co. (1981) determined LC50 values for male Crl:CD rats
(weight 250+/-25 g) at exposure times of 5,15, 30, and 60 minutes. The experiment was performed
in duplicate with one set of animals exposed head only to the test gas while the other set was
unrestrained inside a 175-liter rectangular exposure chamber. In restrained rats, respiration rate
was  monitored by recording pressure fluctuations due to breathing in a body plethysmograph.
During CO exposures the chamber atmosphere was monitored continuously for oxygen (BioMarine
Industries model 225 oxygen meter), carbon dioxide and CO (InfraRed Industries model 702-D non-
dispersive analyzer) using infrared analyzers. Blood from CO exposed rats that died during or within
30 minutes post-exposure was  collected by cardiac  puncture. The blood was  measured for
hemoglobin, COHb and  oxyhemoglobin by  an  Instrumentation Laboratories model  282 CO-
Oximeter. The post-exposure observation period was 14 days during which time body weights were
monitored.
                                         26

-------
CARBON MONOXIDE                                                 FINAL: 07/2008

       Nearly all of the deaths occurred during the exposure period; of all animals that died only 2
of 216  restrained and 3 of 148 unrestrained rats died after the end of the exposure period. The
authors reported LC50 values for the 5-,  15-, 30-, and 60-minute exposure periods  for the
unrestrained rats of 10151  ppm (95% C.I., 9580-10953 ppm), 5664 ppm (95% C.I., 5218-6078
ppm), 4710 ppm  (95% C.I., 4278-5254  ppm), and 3954  ppm  (95% C.I., 3736-4233 ppm),
respectively. The LC50 values were lower (higher toxicity) for restrained rats. For the respective
exposure durations values of 10754, 4318, 2890 and 1888 ppm were obtained. The RD50 for rats
exposed to CO was 15000 ppm. The COHb values were 60 % or higher in rats that had died after
unrestrained exposure and  50 % or higher in rats that had died after restrained exposure.

       Darmer et al.  (1972) reported a  LC50  of 14200 ppm  for  5 minutes exposure. Haskell
Laboratory (1978) obtained a LC50 of 4070 ppm for 30  minutes exposure. Hartzell et al. (1985)
reported a LC50 of 8636 ppm for 15  minutes exposure and 5207 ppm for 30 minutes exposure.
Kimmerle  (1974) reported a LC50 of 5500 ppm for 30  minutes and 4670 ppm for 60  minutes
exposure.

       Rose et al. (1970) reported  a LC50 of 2070 mg/m3 (95 % C.I. 1831-2241 mg/m3) (1807,
1598-1956 ppm) for 4 hours exposure in male Sprague-Dawley rats. The COHb in animals that had
died was between 50 and 80 %.

3.1.2.  Mice

       Pesce et al. (1987) exposed groups of about 100 OFrstrain  mice/age group/sexto S.STorr
(about 7200 ppm; final analytical concentration) for 76 minutes or to 4.4 Torr (about 5800 ppm) for
146 minutes. For the 76-minute exposure, survival rates for males were 36 % for 31-day-old males
and 22 % for 184-day-old males. Of the exposed females, 57 % of 31-day-old females and 63 % of
184-day-old females survived. After exposure for 146 minutes, survival rates were 40 %for34-day-
old males,  27 % for 85-day-old males, 24 % for 230-day-old males and 27 % for 387-day-old males
and 48 % for 34-day-old females, 67  % for 85-day-old females and  56 % for 387-day-old females.
Except for the about  1-month-old mice,  male  mice showed  a significantly  lower survival  than
females. Survival was not significantly influenced by age.

       Winston and Roberts (1978)  investigated the influence of age on  lethal effects of CO on
mice (strain not stated;  male mice were used in all groups, except for the two youngest groups that
comprised both males and females). Animals of different age were exposed to 2000 ppm CO for up
to 6 hours  in stainless steel  exposure chambers. The analytical concentration was determined by
an automated gas chromatograph. Mortality occurred in  3/37 two-day old  mice, 21/32 17-day-old
mice, 16/20 30-day-old mice, 11/1754-day-old mice, 10/20 108-day-old mice and 6/18 150-day-old
mice. The animals of the youngest and that of the oldest age group were found to be more resistant
to CO. These two groups were also found  less  susceptible to lethal effects from hypoxic hypoxia
when mice were exposed to a reduced oxygen  concentration of 7.5 %.

       Hilado et al. (1978) reported 30-minutes LC50 values of 3570 ppm for Swiss-Webster mice
and 8000 ppm for ICR mice. Respiratory distress was the only sign observed during the exposures.

       Rose et al. (1970) reported  a LC50 of 2800 mg/m3 (95 % C.I. 2679-2926 mg/m3) (2444,
2339-2554 ppm) for 4 hours exposure in male Swiss albino mice. COHb was not determined.

                                         27

-------
CARBON MONOXIDE
                                                               FINAL: 07/2008
3.1.3. Guinea Pigs

      Rose et al. (1970) reported a LC50 of 6550 mg/m3 (95 % C.I. 5509-7788 mg/m3) (5718,
4809-6799 ppm) for 4 hours exposure in Hartley guinea pigs. The COHb in animals that had died
was between 57 and 90 %.
TABLE 11: SUMMARY OF LC50 DATA IN LABORATORY ANIMALS
Specie
s
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Mouse
Mouse
Mouse
Mouse
Guinea
pig
Concentration
(ppm)
14200
10151
8636
5664
5607
5500
5207
4710
4070
4670
3954
1807
10127
3570
8000
2444
5718
Exposure
time (min)
5
5
15
15
30
30
30
30
30
60
60
240
15
30
30
240
240
Remark

Crl:Cd strain, male

Crl:Cd strain, male



Crl:Cd strain, male


Crl:Cd strain, male
Sprague-Dawley strain,
male

Swiss-Webster strain
ICR strain
Swiss albino strain, male
Hartley strain, male
Reference
Darmeretal., 1972
E.I. du Pontde Nemours
and Co., 1981
Hartzelletal., 1985
E.I. du Pontde Nemours
and Co., 1981
Herpoletal., 1976
Kimerle, 1974
Hartzelletal., 1985
E.I. du Pontde Nemours
and Co., 1981
Haskell Laboratories, 1978
Kimerle, 1974
E.I. du Pontde Nemours
and Co., 1981
Rose etal., 1970
Kishitani etal., 1979
Hiladoetal., 1978
Hiladoetal., 1978
Rose etal., 1970
Rose etal., 1970
                                        28

-------
CARBON MONOXIDE
                                                                 FINAL: 07/2008
                          LC50 values in  different species
  log concentration (ppm)
4 0

3

3

3.4
3 9
°
^-\,^








CJ
""""--—-_,
™"^\li
•
c:







K~-\_ '— '
^----^






A


--.^
""^•\^_

0,5              1.0               1.5               2,0
                             log tim^min)
       •  rat (DuPont) l~l  rat (other)  O  mouse     A  guinea pig
                                                                              2,5
FIGURE 1: LC50 VALUES FOR CO IN DIFFERENT SPECIES
The solid line was calculated by Probit analysis from the data in E.I. du Pont de Nemours and Co.
(1981). The slope of this line indicates a time scaling exponent of n=2.6. Analysis of all data yielded
a value of n=2.8. The LC50 values are taken from Table 11.
3.2.    Nonlethal Toxicity

       A large number of  studies investigated  nonlethal  effects of single and  repeated CO
exposures in animals (see WHO, 1999a for review). Reported here are only studies that support or
add information to the effects seen  in humans,  because these studies were  considered most
relevant. These include syncope-like observations and behavioral effects in monkeys, effects on
heart function in dogs as well as developmental/reproductive toxic effects in different species.

3.2.1   Monkeys

       Purser and Berrill (1983) studied the behavioral  effects of CO exposure on cynomolgus
monkeys (3 male animals, 4-5 years old). The basic behavioral model consisted of an individual
monkey placed in a chamber with a lever press at one end a reward (chocolate candy) dispenser at
the other. At 5-minute intervals throughout the test session a buzzer was sounded and a light
flashed over the lever. If the monkey pressed the lever within a 1-minute period, a candy was
                                         29

-------
  CARBON MONOXIDE                                                 FINAL: 07/2008

  presented in the dispenser. The monkey then moved the length of the chamber to pick up the
  candy. The  major performance parameter measured was the time from the animal releasing the
  leverto its first touch of the dispenser, i.e., the time taken to traverse the chamber. Each session
  consisted of the following stages: 1) a 25-minute preexposure period during which baseline carbon
  dioxide production and behavioral task performance times were established, 2) 2.5 % CO was
  introduced into the chamber at a sufficiently high flow rate to increase the concentration to 900 ppm
  within 1 minute, 3) 900 ppm CO were maintained for 30 minutes, during which the effects on clinical
  condition, carbon dioxide  production and behavioral task performance were  examined, 4) the
  chamber was flushed of CO, decreasing the concentration to less than 100 ppm within 4 minutes,
  5) animals were maintained for anther 45 minutes in the chamber while their clinical  condition,
  carbon  dioxide production  and behavioral task performance were monitored. Carbon dioxide and
  CO  concentrations were  monitored continuously using infrared  analyzers. Five preliminary
  experiments were conducted at a 1000 ppm CO, followed by the main experimental series that
  consisted of 10 exposures at 900 ppm,  3 for each animal, and 1 preliminary run. For 3 exposures
  (one for each animal) the animals were removed from the chambers minutes after the  end of the
|  exposure period so that venous blood samples could be taken for COHb analysis.

         During the 4 preliminary exposures to 1000 ppm CO, there was generally no visible effect of
  the animals  until 18-20 minutes of exposure had elapsed, at which time they generally became less
  active, occasionally sitting down for short periods. At approximately 25 minutes a dramatic change
  occurred over a period of 1-2 minutes and the animals went from an apparently normal state to one
  of severe intoxication. This change was preceded by one or more warning signs at approximately
  23 minutes, which consisted of momentary closure of eyes,  yawning and shaking of the head.
  Immediately priorto collapse the animals sometimes paced around in a mechanical fashion, often
  swaying as they walked. As few as 20 seconds later the animals were lying or rolling on the floor,
  sometimes attempting to rise before sitting on the floor or lying down again. During recovery, the
  animals remained in a state of severe intoxication for approximately 30 minutes,  lying  down with
  their eyes closed. On three occasions animals vomited during this period. After 25-30 minutes the
  animals were usually sufficiently recovered to get up and move around the chamber, in response to
  the buzzer they would sometimes move towards or even press the  lever although they made no
  attempt to fetch the candy. The performance of the behavioral task was unaffected during the first
  15 minutes  of exposure, but before the first minor clinical signs there was generally a  slowing of
  response.

         During exposures to 900 ppm, the first signs generally occurred after 20-25 minutes when
  the animals  became less active, followed by the minor warning signs at approximately 26 minutes.
  Although in most cases the animals were lying down at the end of the exposure period, they did not
  appear to be severely intoxicated and in 6 of 9 exposures the signs were mild and the animals did
  not  reach a  state of collapse. During  the recovery  period the animals remained in  a  state of
  intoxication for approximately 16 minutes. Recovery was more rapid than that following exposure to
  1000 ppm, as all animals performed the behavioral task within 25 minutes of the end of exposure.
  The first effects upon the chamber traverse time occurred at 15 minutes into the exposure as a
  slight, statistically significant decrement in performance. The decrement at 20 minutes was not
  statistically significant while at 25 minutes it was highly significant, as the mean response time was
  twice the preexposure  response time (1.10 sec vs. 0.62 sec). The first time that the test was
  conducted successfully on all occasions was after 25 minutes of recovery when the mean chamber
  traverse time was 3 times as long as the mean preexposure time. From 30 to  45 minutes the

                                            30

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

animals were  more  active and  response times gradually  improved but did not reach the
preexposure level.
       The mean COHb measured at the end of the exposure was 32.9 % (range 31.7-34.8 %).
Carbon dioxide  production,  indicating the metabolism  in the  animals,  decreased  gradually
throughout the exposure (statistically significant at 25 and 30  minutes  of exposure) and then
increased gradually towards preexposure levels during the recovery period (significantly lower until
15 minutes into the recovery period).

       From earlier experiments, the  authors estimated COHb of 16-21 % for the period of 15-20
minutes when deficits in behavioral task performance were started during the exposure period. In
the state of severe intoxication,  the animals  were capable of performing some  coordinated
behavioral actions when they were sufficiently stimulated, e.g. by loud noise or removing them from
the chamber. The authors report that in unpublished experiments using higher CO concentrations
the animals passed rapidly from this stage to one of deep coma.

       DeBias et al. (1979) reported that CO exposure (100 ppm for 16 h; resulting in a COHb of
9.3 %) reduced the threshold for ventricular fibrillation induced by an electrical shock applied to the
myocardium of monkeys during the final stage of ventricular repolarization. The voltage required to
induce fibrillation was highest in normal animals breathing air and lowest in  infarcted animals
breathing CO. Additivity was found for the effects of infarction alone and CO exposure alone each
of which required significantly less voltage for fibrillation.

3.2.2   Dogs

       Aronow et al. (1979) reported  that CO exposure increased the  vulnerability of the heart to
induced ventricular fibrillation in normal dogs breathing 100 ppm CO for 2 hours (resulting COHb
was  6.3-6.5 %). The ventricular fibrillation was induced by an electrical stimulus applied to the
myocardium.

       Sekiya  et al. (1983) reported that exposure to CO concentrations of  3000 ppm for 15
minutes followed by 130 ppm for 1 hour (resulting COHb was 13-15 %) increased the  severity and
extent of ischemic injury and the magnitude of ST-segment elevation  which was induced in
anaesthetized dogs by coronary artery ligation more than  did ligation alone.

3.3.    Developmental/Reproductive Toxicity
3.3.1   Pigs

       Dominick and Carson  (1983) exposed pregnant sows to CO concentrations between 150
and 400 ppm for 48-96 hours between gestational days 108-110 (average gestation was 114 days).
They showed a significant linear increase in the number of stillbirths as  a function of increasing CO
concentration. Stillbirths were significantly elevated above control levels when the maternal COHb
exceeded 23 % saturation.  These saturation levels were obtained at approximately 250 ppm.

       Morris et al. (1985)  exposed 16 pigs to 0, 200 or 250 ppm from gestational day 109 until
birth (maternal COHb at 24  hours into  the exposure was 0,13.6 and 17.1 %, respectively). Stillbirth
rate for the 3 groups (total of 123 piglets) were2.3, 2.4 and 4.8  %,  respectively. The study authors
stated that the stillbirth rate was not affected because the observed  rates were lower than the

                                          31

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

industrial norm  of 5-10 %. The COHb in  neonatal piglets at birth were 0, 19.8 and 22.4 %,
respectively. The authors found impairment of negative geotaxis behavior and open field activity 24
hours after birth in the 250-ppm  group. Activity in open field was significantly reduced at 48 hours
after birth in piglets from both exposure groups.

3.3.2   Rabbits

       Astrup et al. (1972) reported an increase  in fetal mortality and malformations in rabbits
exposed to 180  ppm CO continuously throughout gestation. Maternal COHb was 16-18 %.

       Rosenkrantz et al. (1986) exposed rabbits to high concentrations of CO-containing cigarette
smoke (12 puffs of 2700-5400 ppm CO; exposure to puffs of cigarette smoke by face mask; each
puff sequence consisted of 30 seconds cigarette smoke and 30 seconds fresh air) for 12 minutes
daily from gestational days 6-18.  The COHb level reached at the end of each exposure was 16 %. A
large number of fetal deaths,  but no malformations were observed in exposed animals.

3.3.3   Rats

       Choi and Oh (1975) exposed rats to 750 ppm CO for 3 h/d on gestational days 7, 8 or 9. An
excess of fetal absorptions and stillbirths as well as a decrease in body length and an increase in
skeletal anomalies were observed. COHb was not determined.

       Penney et al. (1980) exposed pregnant COBS rats for the last 18 days of gestation to 200
ppm CO. The mean maternal COHb was about 27.8 % and the mean fetal level was 27.0 %. The
body weight of the pups  was significantly lower than that  of controls. The heart weight of both
exposed females and pups was significantly increased.

       Mactutus and Fechter (1985) exposed Long-Evans rats continuously throughout gestation to
0 or 150 ppm CO. Mean COHb  was 15.6 % vs.  1 % in control subjects. At 120 days of age,  CO-
exposed rats acquired a conditioned avoidance response equally well as control animals. However,
following a 24-h interval  the  CO-exposed  rats  failed to demonstrate significant retention.  In a
second  experiment, in which animals received  50 training trials per day until a criterion of ten
consecutive avoidance responses was met,  the prenatal CO-exposed rats again acquired the task
as well as control animals. When tested for retention 28 days later, a significant memory impairment
was again observed in terms  of trials required to retain the avoidance criterion as well as in total
percent avoidance. At one year of age, the CO-exposed rats showed  impairment relative to  air-
exposed controls in both the original learning and retention of the two-way avoidance response.

3.3.4   Mice

       Singh and Scott (1984) exposed groups of 17 pregnant CD-1 mice to CO concentrations of
0, 65, 125,  250 or 500 ppm  for 24 h/d on gestational days 6 to 17.  Mice  were sacrificed and
examined on day 18. No signs of maternal toxicity were observed at any dose. The mean  percent
fetal  mortality per litter was 4.52, 5.89, 12.50, 15.50 and 55.30 %, respectively. Besides a dose-
dependent increase in embryolethality, fetus weights were significantly reduced at exposure levels
of 125 ppm or higher. No fetal malformations were detected. COHb was not determined.
                                          32

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

       Singh (1986) exposed CD-1 mice to 0, 65 or 125 ppm CO continuously during gestational
days 7 to 18(COHb not determined). No signs of maternal toxicity were observed. Exposure did not
affect the number of live pups born per litter or their birth weight.  Prenatal exposure to 125 ppm
significantly increased the time required by pups for righting reflex on day 1 of birth and negative
geotaxis on day 10. Prenatal exposure at both concentrations significantly decreased the mean
aerial righting score of pups on day 14.

3.4.    Genotoxicity

       No information regarding the carcinogenicity of CO in animals was located in the available
       literature.

3.5.    Carcinogenicity

       No information regarding the carcinogenicity of CO in animals was located in the available
literature.

3.6.    Summary

       Several studies reported LC50 values in rats, mice and guinea pigs. In the study of E.I. du
Pont de Nemours and Co. (1981), the following LC50 values were calculated by Probit analysis:
10151 ppm for 5 minutes, 5664 ppm for 15 minutes, 4710 ppm for 30 minutes and 3954 ppm for 60
minutes.

       In a study in  cynomolgus monkeys, at exposure to 900 ppm no signs of intoxication occurred
during the first 20-25 minutes (corresponding to COHb of about 16-21  %), at 25 minutes the
animals' performance in a behavioral test significantly decreased  and at the end of the exposure
period (30 min) animals became less active and were lying down. After about 25 minutes at a 1000
ppm, within 1- 2 minutes the animals went into a  state of severe  intoxication, virtually unable to
perform coordinated movements (Purser and Berrill, 1983).

       In developmental toxicity tests, CO caused an  increase in the rate of stillbirths or fetal
mortality in pigs after a 2-3 day-exposure to COHb above 23 % (Dominick and Carson, 1983), in
rabbits after continuous  exposure to 16-18 % COHb throughout gestation (Astrup et al., 1972) as
well as  after daily  exposure to  high  CO concentrations in cigarette smoke (exposure for 12
minutes/day on gestational days 6-18, resulting in COHb of 16%)(Rosenkrantzetal., 1986), in rats
after 3 exposures to 750 ppm for 3 h/d (Choi and Oh, 1975) and in mice after exposure to 125 ppm
for 11 d  (Singh and  Scott,  1984). Significant memory impairment in behavioral tests were found in
young rats after continuous CO exposure  throughout gestation (mean maternal COHb was 15.6%)
(Mactutus and Fechter,  1985).

       In monkeys, a COHb of 9.3 % resulted in reduced threshold for electric shock-induced
ventricular fibrillation (DeBias et al., 1979). A similar effect was found in dogs at 6.3-6.5 % COHb
(Aronowet al., 1979). A COHb of 13-15 % increased the severity and extent of ischemic injury and
the magnitude of ST-segment elevation  in a myocardial infarction model in dogs (Sekiya et al.,
1983).
                                          33

-------
  CARBON MONOXIDE                                                 FINAL: 07/2008

  4.     SPECIAL CONSIDERATIONS

  4.1.   Stability, Metabolism and Disposition

        CO is produced endogenously in normal metabolism: when an a-methylene bridge in the
  heme group of hemoglobin is broken during the catabolic process, one molecule of CO is released.
  It has been estimated that this production amounts to approximately 0.3 to 1.0 ml/h with an
  additional 0.1  ml/h resulting from a  similar catabolic process involving other heme-containing
  compounds (e.g., myoglobin as well as cytochrome and catalase enzymes). This endogenous
  production of CO gives rise to an baseline or back ground level approximately of 0.5-0.8 % COHb
  (NIOSH, 1972).

        Almost all of the CO that has been  inhaled is eliminated through the lungs when the
  previously exposed person enters an atmosphere free of CO. Carbon monoxide not only binds to
  hemoglobin forming COHb, but 10-50 % of the total  body store of CO is also distributed to
  extravascular sites such as skeletal muscle, where it can bind to myoglobin. Extravascular CO can
  be slowly metabolized to carbon dioxide (Fenn, 1970). Inside the cells, CO can bind to all heme
  proteins capable of binding oxygen, such as myoglobin, cytochrome c oxidase, cytochrome P450
  enzymes and tryptophan oxygenase  (WHO, 1999a). However, the exact extent of this binding in
  vivo as well as the physiological consequences in terms of inhibition of protein and enzyme function
  and the existence and  relevance of possible toxic effects has not been clearly shown up until now
  (cf. extensive discussion in WHO, 1999a).

         The time required to eliminate half of the gas is 3 to 5 hours (Landaw, 1973), depending on
  the amount of respiration, which acts to wash  it out of the body. Peterson and Stewart  (1970)
  reported a range for the elimination half-time from 128 to 409 minutes from 39 experiments, with an
  average of 320 minutes in human subjects that breathed normal air after CO exposure. Increased
  oxygen pressure helps to dislodge it from the hemoglobin. One hundred percent oxygen given at
  atmospheric pressure reduces the half-elimination rate to about 80 minutes (Peterson and Stewart,
  1970). Weaver et al. (2000) reported a half-life of 74+7-25 minutes for COHb in CO-poisoned
  patients receiving 100 % oxygen. Klasner et al.  (1998) reported a half-time of 44 minutes for
  children (n=26, 4-12 years old) when given 100 % oxygen via face mask. Hyperbaric oxygen at 3
|  bar pressure reduces the half time to about 20-25 minutes (Beard, 1982; Landaw, 1973).

  4.2.   Mechanism of Toxicity

        If not stated otherwise, the information on the mechanism of toxicity is taken from the
  extensive recent reviews of WHO (1999a) and EPA (2000). The best understood biologic effect of
  CO is its  combination with hemoglobin (Hb) to form COHb, thereby rendering the hemoglobin
  molecule less able to bind with oxygen. Although the rate of CO binding with hemoglobin is about
  one-fifth slower and the rate of dissociation from hemoglobin is an order of magnitude slower than
  the  respective rates for oxygen, the CO chemical affinity for hemoglobin (represented  by the
  Haldane coefficient M) is about 245 times greater than that for oxygen. One part of CO and 245
  parts of oxygen would form equal parts of oxyhemoglobin and COHb (50 % each), which would be
  achieved  by breathing  air containing 21 % oxygen and  570 ppm CO. The steady-state ratio of
  COHb/oxyhemoglobin is proportional to the ratio of their respective partial pressures:
               COHb/O2Hb = M (Pco / P02)

                                           34

-------
CARBON MONOXIDE                                                  FINAL: 07/2008
       Under dynamic conditions, competitive binding of oxygen andCOto hemoglobin is complex:
the greater the number of heme groups bound to CO, the greater the affinity of free heme groups
for oxygen. CO not only occupies oxygen binding sites, molecule for molecule, thus reducing the
amount of available oxygen, but also alters the characteristic relationship between oxyhemoglobin
and the partial pressure of oxygen, which in normal blood is S-shaped. The difference in the partial
pressure of oxygen between  freshly oxygenated arterial blood (P(O2) = 100 mm Hg) and mixed
venous blood (P(O2) = 40 mm Hg) represents a release to the tissues of approximately 5 ml O2/100
ml blood (NIOSH, 1972). With increasing COHb in blood, the dissociation curve is shifted gradually
to the left, and its shape is transformed into that  of a  rectangular hyperbola.  This changes the
release of oxygen to the tissues appreciably: the oxygen content of the blood is not only lowered
during exposure to CO, but the shift of the oxyhemoglobin dissociation curve to the left decreases
the amount of remaining oxygen that is made available to the tissues. Both  mechanisms serve to
effectively lower the tissue  partial pressure of oxygen and hence can create a generalized tissue
hypoxia. Because the shift occurs over a critical saturation range for release of oxygen to tissues, a
reduction  in oxyhemoglobin by CO poisoning  will have more severe effects  on the  release of
oxygen than the equivalent reduction of hemoglobin due to anemia.

       While the brain has a higher requirement for oxygen than the heart,  in contrast to the
cerebral circulation the coronary  circulation must supply an even increased amount  of oxygen
during periods of generalized tissue hypoxia; since underthese circumstances the heart is forced to
increase both its rate and its output in orderto meetthe normal oxygen demands of the  body. This
increase  in myocardial activity demands an increased  oxygen supply to the myocardium, which
must be met by the coronary circulation. Under hypoxic conditions increased oxygen supply to the
peripheral tissues can be accommodated by increased blood flow (via vascular dilatation) and/or
increased oxygen extraction by the tissues. The myocardium underthese circumstances appears
only to increase the flow of blood rather than to extract an additional amount of oxygen from the
coronary circulation. While the peripheral tissues normally extract only 25 percent of the oxygen
content of the perfusing arterial  blood  during resting  conditions, the myocardium extracts 75
percent, thus leaving the mixed venous blood only 25 percent saturated. This mechanism has the
overall effect of maintaining the myocardial oxygen tension at a higher level than would be present
in other muscle tissue and thus insures a continual aerobic metabolism, even  under hypoxic duress.
In terms of oxygen  partial  pressure,  the mixed venous blood  of the  peripheral  tissues is
approximately 40 mm Hg while the mixed venous blood of the coronary circulation is only 20 mm
Hg. In the presence of COHb (and the shift to the left of the oxyhemoglobin dissociation curve),
however, the arterio-venous difference can only be maintained by an increased flow in the coronary
circulation. In an individual with diminished coronary circulation because of coronary heart disease,
however, this situation may result in a decrease  in  the venous oxygen partial  pressure of the
myocardium precipitated by an inability to maintain the normal arterio-venous gradient. Studies in
dogs suggest that exercise plus an increased COHb, in addition to the global myocardial hypoxia,
leads especiallyto areas of relative hypoxia in the leftventricle secondaryto redistributive changes
in subendocardial blood flow(Einzig, 1980). This hypoxic effect is further enhanced, as mentioned
above, by an increase in  cardiac rate and output as  a general response to peripheral tissue
hypoxemia. A person with diminished  coronary circulation caused by coronary heart disease,
consequently, may be constantly near the point of myocardial tissue hypoxia, which can ultimately
lead to myocardial infarction.
                                          35

-------
CARBON MONOXIDE                                                 FINAL: 07/2008

4.3    Issues related to post-mortem CO determination in humans

4.3.1.  Potential factors influencing COHb levels

       Data on the post-mortem decay of carboxyhemoglobin is limited. Rodat et al. (1987)
reported about the stability of carbon monoxide after death.  A CO poisoining due to a running
truck engine in a non-ventilated area was discussed. The autopsy was performed 10 months
after the death due to insurance claims.  The body was decomposed but some muscle tissue
was recovered and tested for carboxyhemoglobin levels, which were 26%. Muscle tissue as
well as other human tissues, such as brain, lung, and kidney may be used for diagnosing death
due to lethal exposure to CO (Vreman et al., 2006).  The report did not discuss measurements
from blood samples,  presumably due to the decomposition of blood.  Following death, the report
indicated that carboxyhemoglobin levels disintegrate overtime releasing reduced hemoglobin.
In addition, the report indicated that the formation of sulfur compounds in a putrefied cadaver
makes it difficult to interpret the absorption spectra of the  carboxyhemoglobin measurements, a
phenomenon that has been acknowledged by others (Kojima et al., 1986; Winek and Prex
1981).  Rodat et al. (1987) and Kojima et al. (1986) also suggested that the endogenous
formation of CO after death is very low,

       People who die of CO poisoning often show sub-lethal COHb levels in their blood
(Ronald Coburn, personal communication).  The lungs rapidly absorb carbon monoxide, which
avidly combines with hemoglobin at 230 to 270 times greater than oxygen (Ellenhorn, 1997;
Larsen 2005). Oxygen therapy increases oxygen delivery and pulmonary excretion of carbon
monoxide by displacing carbon monoxide from the hemoglobin and decreasing the half-life time
of COHb (Ellenhorn,  1997; Roos 1994), in turn explaining the sub-lethal COHb levels in blood
samples from deceased people exposed to CO.

       Carbon monoxide shifts from the blood into the muscle tissue have been reported in the
literature (Bruce and  Bruce 2006; Luomanmaki and Coburn, 1969).  In order for shifts to occur,
blood must  be flowing in capillaries. Presumably during the moving of corpses after death, blood
could be pushed through capillaries to a small extent (Ronald Coburn, personal communication)
leading to carbon monoxide shifts, but no studies were found reporting this phenomenon in
dead people.

       Oxidation of CO to CO2 has been reported in living animals and humans. The rate of
oxidation in  skeletal and cardiac muscle was found to be small but still measurable (Fenn and
Cobb, 1932; Clark 1950; Luomanmaki and Coburn, 1969). It is unknown whether this oxidation
occurs in cadavers and its effects on the CO decay rates  after death.

       Once blood is collected from a cadaver, the postmortem samples may be measured
more than once and results would depend on storage and treatment of samples. Levin et al.
(1990) found a 19% decrease in measured COHb levels when blood was refrigerated for three
months and then frozen for three months. The refrigerated blood samples were first tested by
microdiffusion techniques (sensitivity of -5%) within one month of being obtained from victims of
a hotel fire on December 31, 1986. The samples remained refrigerated until being  mailed
frozen to a second laboratory for testing (received March  26, 1987). The samples were frozen
upon arrival and were thawed and refrozen several times  during the next three months for

                                         36

-------
CARBON MONOXIDE                                                 FINAL: 07/2008

experimental purposes. After sonication and filtration, the samples were analyzed on a CO-
oximeter IL-282 (sensitivity not given).  The authors concluded that aging of blood samples and
methods of storage could affect accuracy of analytical results. This result was supported by
another study which determined that the contact of the sample with air could decrease the
%COHb saturation  (Chace et al. 1986). This is in contrast to reports that CO would be stable
for months to years in stored samples (vacutainer tubes, especially heparin-anticoagulated
tubes) (Hampson 2008, Kunsman et al. 2000). Proper storage of samples would prevent loss of
CO (Nelson 2005b).

4.3.2   Influence on collection site on measured COHb levels

       Reductions  in the %COHb saturation are also associated with differences between
carboxyhemoglobin measurements derived from heart blood and peripheral blood specimens.
Levine et al. (2002) studied data from 42 CO poisoning cases.  The Office of the Chief Medical
Examiner in the state of Maryland provided the data. Blood samples from the heart and the
subclavian veins were analyzed in a CO-oximeter. The specific heart site for blood collection
was not reported. Also, the report did not indicate whether the deceased individuals with
decreased carboxyhemoglobin were given oxygen therapy.  Blood samples with COHb
saturation levels greater than 12% were confirmed and quantitated by gas chromatography.
The latter analysis measured both CO  content and CO capacity, and not hemoglobin
concentration, which tends to vary in postmortem specimens (Levine et al. 2002).  Samples
were normalized for hemoglobin ensuring that differences between the heart blood and
peripheral blood were not caused by significant differences in hemoglobin between the two
blood samples.  The average heart blood COHb level was 42 (range= 11-79; standard
deviation=19.95, median=38) and the average peripheral blood COHb level was 39 (range= 4.2-
71;  standard deviation =17.07, median=37). The average heart blood to peripheral blood (H/P)
ratio was 1.09.  Sixty-two percent (26 of 42) of the cases had an H/P ratio between 0.9 and 1.1,
whereas 74% (31 of 42) of the cases had an H/P ratio between 0.8 and 1.2.  Statistical analysis
showed no statistically significant differences in COHb levels between heart and peripheral
blood samples (Levine et al. 2002). The report acknowledged that there might be instances
(e.g. cardiopulmonary resuscitation) where differences between heart blood and peripheral
blood COHb levels  might occur in isolated cases, but in general, there were no significant
differences  between the two blood sources.

       Dalpe-Scott et al. (1995) calculated the H/P ratio of drug concentrations in postmortem
blood samples for 113 drugs representing  320 cases. Thirty-five carbon monoxide poisoning
cases were examined.  The average H/P ratio was 1.0, range 0.9-1.5.  The specific COHb
levels were not provided.  Data from Dalpe-Scott et al.  (1995) confirmed those findings
presented by Levine et al.  (2002).

       Differences  between  COHb levels in the heart blood when compared to those found in
the  periphery (e.g. femoral vein) have been reported in cases that received cardiopulmonary
resuscitation.  Rice (1976) found wide variation in carboxyhemoglobin levels in 300 consecutive
fatal cases of CO poisoning. Source of CO poisoning (fire, gas heaters, etc.) was not identified
in most of the  300 cases with the exception of 4 case studies discussed in the paper.  The
author hypothesized that levels below 50% COHb were probably low due to the dissociation of
carboxyhemoglobin after death when oxygen therapy was given as an attempt to resuscitate the

                                          37

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

person. A summary of the case findings is given below.

•  Casel:  a child (14 months) was found apparently dead in a smoldering room fire. Artificial
   respiration was given on the way to the hospital and continued for about an hour before
   death was pronounced.  Subclavian blood showed carboxyhemoglobin levels of 15%.  The
   report did not indicate if it was collected from the subclavian artery or vein.  Blood from the
   femoral vein reported a 31% carboxyhemoglobin level, or a 2-fold difference between sites.
   Subclavian blood to femoral blood  ratio: 0.48

•  Case 2:  a man of 57 yrs died of CO poisoning. The CO source was a disconnected coal
   gas supply pipe. The emergency personnel found him cold but gave him artificial respiration
   on the way to the hospital where he was pronounced dead. Subclavian blood showed
   carboxyhemoglobin levels of 32%.  The report did not indicate if it was collected from the
   subclavian  artery or vein. Blood from the femoral vein reported a 52% carboxyhemoglobin
   level, or a 1.6-fold difference between sites. Subclavian blood to femoral blood  ratio: 0.62

•  Case 3:  a woman of 43 yrs was exposed to CO during a fire.  Fire personnel recovered her
   and attempted resuscitation using artificial respiration and pure oxygen. Subclavian blood
   showed carboxyhemoglobin levels of 42%. The report did not indicate if it was collected
   from the subclavian artery or vein.  The common iliac vein showed 45% carboxyhemoglobin
   level. Blood from the femoral vein  reported a 59% carboxyhemoglobin level, or a 1.2-fold
   difference when compared to the subclavian vein.  Subclavian blood to femoral  blood ratio:
   0.71
   Subclavian blood to iliac blood ratio:  0.93

•  Case 4:  an infant of 5 months old died in a room fire.  Artificial respiration was performed
   with the infant.  Femoral samples were not provided. Blood draining the blood cavity was
   taken and reported a 48% level of carboxyhemoglobin, whereas the subclavian  blood
   reported a 34% level of carboxyhemoglobin.  The report did not indicate if it was collected
   from the subclavian artery or vein.  Subclavian  blood to peripheral blood ratio:  0.71

       Rice (1976) explained the results by pointing out that blood with high concentration of
carboxyhemoglobin does not coagulate and artificial respiration would have pushed blood to
move into and out of the lungs.  Thus,  oxygen therapy would have increased the dissociation of
carboxyhemoglobin in the blood, and the amount of the dissociation would have depended on
the vigor and the duration of the artificial respiration. The disassociation would be higher in the
blood from the  lungs, the heart, and the blood vessels in close proximity to the lungs and heart.
       Currently, the standard forensic practice is to collect blood from suitably isolated
peripheral sites (e.g. femoral vein) which are less likely to be subject to postmortem chemical
redistribution (Flanagan et al. 2005;  Drummer 2007). The common practice of procuring blood
samples from live persons has been venipuncture of the antecubital area of the arm (Ernst
2005).

       Gas chromatography is considered the most precise and accurate technique to measure
COHb levels, but other techniques such as spectrophotometric analyses worked well (Lee et al.

                                          38

-------
CARBON MONOXIDE
                                                                 FINAL: 07/2008
1975; Mahonoey et al. 1993; Ronald Coburn personal communication).

4.4.    Other Relevant Information
4.4.1.  Species Variability

       With regard to lethal effects, COHb of 50-80 % have been reported as lethal level in rats and
guinea pigs (Rose et al., 1970; E.I. du Pont de Nemours and Co., 1981). In apparently healthy
people that died from CO poisoning, usually COHb of 60 % or higher are found (Balraj, 1984; AIHA,
1999; Winter and Miller,  1976, Holmes, 1985, Stewart,  1975).

       Syncopes have been reported to occur in children at a threshold of 24.5 % COHb (Crocker
and Walker, 1985).  In monkeys, at COHb little higher than 16-21 % syncope-like effects occurred
(Purser and Berrill, 1983). The lowest COHb that resulted in cognitive development defects in
children in a long-term follow-up study was 13 % (Klees et al., 1985). In mice, memory impairment
was found in the offspring of rats exposed continuously at 15.6 % COHb during gestation (Mactutus
and Fechter, 1985).

       Taken together, these studies imply a limited species variability for different effects with
regard to the COHb at which these effects occur. However, the exposure conditions necessary to
reach a certain  COHb differ between species due to different affinities of their hemoglobin for CO.

The equilibrium COHb of different species is determined by the species-specific Haldane (affinity)
constant M. Reported values are  228  for dogs (Sendroy and  O'Neal, 1955),  195 for monkeys
(Sendroy and O'Neal, 1955), 170 for rats (Rodkey and O'Neal, 1970) and 117 for guinea pigs
(Rodkey and O'Neal, 1970). Jones etal. (1971) reported equilibrium COHb in different species after
48-hour continuous exposure as shown in Table 12. Using the  mathematical model described in
Appendix B, corresponding COHb values for a 70-kg man can be calculated as 7.9, 13.8 and 25.0
% for 51, 96 and 200 ppm, respectively^
TABLE 12: COHb AFTER 48 HOURS CONTINUOUS
CO; adopted from Jones et al., 1971
CO concentration
(ppm)
51
51
51
51
96
96
96
species
dog
monkey
rat
guinea pig
dog
monkey
rat
EXPOSURE TO
COHb in blood (%) (n)
5.7
5.3
5.1
3.2
12.5
10.3
7.5
(2)
(3)
(15)
(15)
(2)
(3)
(15)
                                         39

-------
CARBON MONOXIDE
                                                                  FINAL: 07/2008
96
200
200
200
200
guinea pig
dog
monkey
rat
guinea pig
4.9
20.8
20.0
16.4
9.4
(15)
(2)
(3)
(15)
(15)
4.2.2.  Intraspecies Variability

       Experiments in mice did not indicate that very young or very old animals were more
susceptible to lethal effects of CO exposure (Pesce et al., 1987; Winston and Roberts, 1978).
However, there is considerable variability within human subpopulations: aCOHb of about 15% only
leads to very slight symptoms, such as headache, in healthy adults (Stewart et al., 1970; WHO,
1999a). In contrast, the same COHb was reported to cause long-lasting defects in the cognitive
development and  behavioral alterations in children (Klees et al., 1985) or even to contribute to
death from myocardial infarction in individuals with coronary artery disease (Grace and Platt, 1981;
Balraj,  1984). In case reports of myocardial infarction, other subjects that were exposed underthe
same conditions (and sometimes had higher COHb) did not experience effects above the AEGL-2
level (Atkins and Baker, 1985; Grace and Platt, 1981).

       Subpopulations at higher risk for toxic effects of CO include the following groups:

a) fetuses because of higher CO affinity and slower CO elimination (see Sections 2.3 and 4.4.4).
The severity of exposure and maternal clinical signs appear to be associated with fetal mortality
(Koren et al. 1991).  A review by Greingor et al. (2001) noted that carbon  monoxide crosses the
placenta through passive diffusion or facilitated by a carrier. As the fetus increases in age and
weight, placental carbon monoxide diffusion increases.  When the mother is exposed to carbon
monoxide, the  amount of oxygen in her blood decreases, and oxygen transported across the
placenta decreases and puts the fetus  in a hypoxic state. Carbon monoxide crosses the placenta
as it dissociates from maternal hemoglobin and binds to fetal hemoglobin. The only source of fetal
oxygen is the mother, and maternal treatment for carbon monoxide poisoning reduces fetal COHb
levels (Greingor etal. 2001). Clearance of carbon monoxide in the fetus would be dependent on the
mother's oxygen intake.  Children have the  same type of hemoglobin as adults, but are more
susceptible than adults because they breathe a greater amount of air per body weight than adults.

The binding affinities of embryonic hemoglobins suggest fetuses are more susceptible to  CO
intoxication when  compared to that of the adult hemoglobin.  Embryonic hemoglobins (Gower I,
Gower II, Portland) are present until about 8-12 weeks of gestation.  Fetal hemoglobin is expressed
from about 5 weeks of gestation until 9 months after birth. Adult hemoglobin starts being produced
between 3-6 months after birth (Orville, 2008).   Under physiological  conditions, the binding
constants of fetal and adult hemoglobin to carbon monoxide are 0.09 uM and 0.13 uM, respectively,
meaning that the  binding affinity for carbon monoxide is higher for fetal  hemoglobin than adult
                                          40

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

hemoglobin (Di Cera et al. 1989).  The rate constants for the binding of carbon monoxide to the
three embryonic hemoglobins are 3.Ox 10~6 M/s (Gower I), 2.0 x10~6 M/s (Gower II), and 3.5x 10~6
M/s (Portland) compared to 4.0 x 10~6 M/s for adult hemoglobin at pH 6.5  (Hofmann and Brittain,
1996).   No data were located reporting if embryos  are more susceptible than  fetuses.   Data
regarding susceptibility of embryos to carbon monoxide is mostly qualitative.

b) children because they develop acute neurotoxic effects (e.g. headaches, nausea), long-lasting
neurotoxic effects (e.g. memory deficits) and impaired ability to escape (i.e. syncopes)  at lower
COHbthan adults (see Section 2.2.2.1). Children also have developing organs (brain, lungs)which
may be affected differently than the developed organs  of adults (ATSDR 2002).  Children tend to
be more susceptible than adults because they breathe  a greater amount of air per body weight than
adults.

c) people with pre-existing diseases, either known or unknown, that  already  decrease the
availability of oxygen to critical tissues, including subjects with coronary artery disease (see
Sections  2.2.1  and 2.2.1.1),  chronic obstructive  lung  disease,   chronic   anemia  and
hemoglobinopathies, such as sickle cell anemia._For example, in sickle-cell disease, the  average
lifespan of red blood  cells with abnormal hemoglobin  is 12 days compared to an average of 120
days in healthy individuals with normal hemoglobin. "As a result, baseline  COHb levels can be as
high as 4%. Presumably, exogenous exposure to CO, in conjunction with higher endogenous CO
levels, could result in critical levels of COHb. However,  it is not known how ambient or near-ambient
levels of CO would affect individuals with these disorders" (EPA, 2000; see also WHO, 1999a). Due
to the physiologic adaptation in these subpopulations, they are not considered more susceptible
than patients with coronary artery disease.

d) people at high altitude, especially those not living there long enough for physiological adaptation.
"It is important to distinguish between the long-term resident of high altitude and the newly arrived
visitor from low altitude. Specifically,  the visitor will be more hypoxemic than the fully  adapted
resident. One would postulate that the combination of high altitude with carbon monoxide would
pose the  greatest  risk  to  persons newly arrived at high altitude  who have underlying
cardiopulmonary disease, particularly because they are usually older individuals. Surprisingly, this
hypothesis  has never been tested adequately" (WHO,  1999a). Due to physiologic adaptation,
people living at high altitude are not  considered generally more  susceptible than patients with
coronary artery disease. Since it is generally not advisable for patients with severe coronary artery
disease to travel to places at high altitude, it is not considered necessary to especially take that
part of the identified  susceptible subpopulation (i.e. patients with coronary  artery disease; see
below)  into account when deriving AEGL values.

       An estimated 62 million people in the United States (about 20% of the population) have one
or more types  of cardiovascular disease (American Heart Association, 2003).  For the major
diseases within the category of total cardiovascular disease, about 50 million Americans have high
blood pressure, 13 million have coronary heart disease, 4.9 million have heart failure, 4.7 million
have cerebrovascular disease (stroke), and 1 million have congenital cardiovascular defects.

       The prevalence of cardiovascular diseases increases with age. It is  10% for males and 4%
for females at age 25-34, 51 % for males and 48 % for females at age  55-64 and 71 % for males
and 79 % for females at age 75 or  older  (American Heart Association,  2003).

                                          41

-------
  CARBON MONOXIDE                                                  FINAL: 07/2008
        Coronary heart disease caused more than 1 of every 5 deaths in the USA in 2000. Coronary
  heart disease was mentioned  as cause of death in 681,000 cases and  myocardial infarction in
  239,000 deaths. Fifty percent of men and 63 % of women who died suddenly of coronary heart
  disease had no previous symptoms of this disease (American Heart Association, 2003).

        Within the group of people with coronary heart disease, 7.6 million had myocardial infarction
  (heart attack) and 6.6 million of angina pectoris (chest pain) (American Heart Association, 2003).
  The prevalence of angina  pectoris in the  British adult population is about 4 % (Williams and
  Stevens, 2002).

        Angina pectoris is a symptom of coronary heart disease. Common  features of an attack are
  central chest pain, pain radiating to the lower jaw,  or arms, and shortness of breath. The pain
  occurs when there is insufficient oxygen delivery to the heart, leading to ischemia. This is usually,
  although not exclusively, a  result of an atheromatous narrowing (stenosis) in one or more of the
  coronary arteries. Angina can classified broadly as stable or unstable, depending on its severity and
  pattern of occurrence. Stable angina is typically provoked by exercise (e.g., hurrying across a street
  or climbing a long flight of stairs), stress or extremes of temperature and is relieved by either rest or
  sublingual nitrates or both. Unstable angina is understood as anginal pain that occurs with lesser
  degrees of exertion, with increasing frequency, or at rest (i.e., without exertion). The pain may be
  more severe,  last longer, and requires  more intensive intervention (usually hospitalization for
  initiation of medication under cardiac monitoring). If left untreated, unstable angina may result in a
  heart attack and irreversible damage to the heart. The diagnosis of angina is generally based on
  clinical history, electrocardiograph stress testing (where patients are exercised on a treadmill to look
  at the effect on their electrocardiogram), and coronary angiography (looking for narrowings in the
|  coronary arteries) (Williams and Stevens, 2002).

  4.4.3.  Time Scaling

        The  L.C50 values for different exposure periods are shown in Figure 1. Overall the distribution
  does not seem to argue against a linear relationship between log(concentration) and log(time) and
  from the data from E.I. du Pontde Nemours and Co. (1981) a value of 2.6 can be calculated for the
  exponent n  from the slope. Regression analysis of all data yielded a value of n=2.8. However,
  taking a closer look at the data from this study suggests that the data might  be  distributed  non-
  linearly and that the slope decreases with increasing exposure time.

        The  AEGL-2 and AEGL-3 exposure concentrations were derived from a mathematical model
  based on the same COHb at the end of the respective exposure periods. These values are also
  distributed non-linearly in a log-log  plot: the slope between the two shortest exposure periods (10
  and 30 min) is equivalent to n=1.0-1.1 and the slope  between the two longest exposure periods (4
  and 8 h) is equivalent to n=2.9-3.4. This non-linearity is probably caused by the fact that the COHb
  depends strongly on the ventilation  rate and lung blood flow for short exposure rates, while for long
  exposure rates the COHb becomes independent of these parameters and exclusively depends on
  the affinity of hemoglobin for CO (represented by the Haldane constant M).  Since rats have a higher
  ventilation rate per kg body weight than humans, their COHb reaches the steady state faster and
  therefore forthe same exposure time the slope for rats is smallerthan the  corresponding slope for
  humans, i.e., COHb depends stronger on the ventilation rate in humans compared to rats.

                                            42

-------
CARBON MONOXIDE                                              FINAL: 07/2008
4.4.4. Mathematical models of COHb formation

      In 1965, Coburn,  Forster and Kane developed a differential equation (CFK model) to
describe the major physiological variables that determine the COHb in blood using data from
patients with increased endogenous production of CO due to anemia (Coburn etal., 1965). The
CFK model is represented by the following equation:

                 d(COHb\     Vco        COHb,*P02           Pco
                        dt            Vb      M*B*Vb*OHb      B*Vb
where:       B = 1 / DL + PL/ VA
            M = Ratio of affinity of blood for CO to that for O2; M = 21 8
            OHb = ml of O2 per ml blood; OHb = 0,2
            COHbt = ml of CO per ml blood at time
            P02 = average partial pressure of oxygen in the lung capillaries; P02 = 100 mm Hg
            Vco = rate of endogenous CO production; Vco = 0.007 ml/min
            DL = diffusivity of the lung for CO; DL == 30 ml / min mm Hg
            PL = barometric pressure minus the vapor pressure of water at body temperature,
            PL = 713 mm Hg
            Vb = blood volume; Vb = 5500 ml
            PCO = partial pressure of CO in the air inhaled (mm Hg)
            VA = alveolar ventilation rate; VA = 6000 ml/min (awake), 4000 ml (sleeping)
            t = exposure duration (min)
      Peterson and Stewart (1970) reported that the CFK model well predicted COHb measured in
18 healthy male students, aged between 24 and 42 years, that were exposed to the following
combinations of CO concentrations and exposure times: about 50 ppmforSO minutes to 24 hours,
about 1 00 ppm for 1 5-480 minutes, about 200 ppm for 1 5-1 20 minutes and about 500 ppm for 1 5-
114 minutes. They used the following integrated form of the CFK equation and  parameters:
               A*COHb,-B*l''co-P,0
where       A=P02/MOHb
            B = 1 / DL + PL / VA
            M = Ratio of affinity of blood for CO to that for O2; M = 21 8
            OHb = ml of O2 per ml blood; OHb = 0,2
            COHbt = ml of CO per ml blood at time
            COHbo = ml of CO per ml blood at beginning of the exposure
            P02 = average partial pressure of oxygen in the lung capillaries; P02 = 100 mm Hg
            Vco = rate of endogenous CO production; Vco = 0.007 ml/min
            DL = diffusivity of the lung for CO; DL == 30 ml / min mm Hg
            PL = the vapor pressure  of water at body temperature, PL = 71 3 mm Hg
            Vb = blood volume; Vb = 5500 ml
            PCO = partial pressure of CO in the air inhaled (mm Hg)


                                       43

-------
 CARBON MONOXIDE
                                                                  FINAL: 07/2008
              VA = alveolar ventilation rate; VA = 6000 ml/min (awake), 4000 ml (sleeping)
              t = exposure duration (min)
                      PB = 750mmHg
                      p
                  'QO-,
                                                                   IOOO pp/r.
                              3456 789100
                                                  345 67891000   2   34 5000
                                 Exposure duration, minutes

FIGURE 2: COHb FOR DIFFERENT EXPOSURE CONCENTRATION-TIME COMBINATIONS
(from Peterson and Stewart, 1975)
       In another study by Peterson and Stewart (1975), data from a series of human exposures to
CO were analyzed to determine the fit to the theoretical CFK equation.  19 men and 3 women were
exposed to concentrations of 50, 100 or 200 ppm for 0.33-5.25 hours.  Three exercise levels from
sedentary to 0,  150 or 300  kpm/min on  an ergometer  were used  (15  subjects  in  total). These
resulted in mean ventilation rates of 10.1 (9.1  for women), 14.0, 24.0  (19.7 forwomen) and 29.7
l/min, respectively. The CFK model predicted COHb for both men and women as well as for resting
and exercising subjects within a standard error of about 2 %. In contrast to the original model, which
assumes  all variables to  be constant except t, PL,  COHbt and Pco,  the following parameter
alterations were introduced:

       P02: When the partial pressure of oxygen in inspired air (Pi02) is less than the 149 mm Hg
             found under normal conditions,  the partial pressure of oxygen in the lung capillaries
             will be less than the value of 100 mm Hg assumed by  Coburn and coworkers.  From
             measurements of oxygen partial pressure in arterial blood, which is assumed to be
             the same as the oxygen partial pressure in lung capillaries, the following equation
             was derived: P02 = 1 / (0.072-0.00079 Pi02 + 0.000002515 (Pi02)2) and
             Pio2 =  Fi02  (PB - 47 - Pico)  with Fi02 =  fraction of  oxygen  in inspired  air,  PB =
             barometric pressure (mm Hg), Pico = partial pressure of CO in  inspired air
       DL: Body size effects on diffusivity at rest were was calculated  from published data as-
             DL = 1 / (-0.0287 + 0.1188/A) with A = body surface in m2
       Vb: the published blood volume relationship of 74 mg/kg of body weight for men and 73
                                         44

-------
CARBON MONOXIDE                                                 FINAL: 07/2008

             ml/kg for women was used.
       VA: The alveolar ventilation rate was expressed as: VA = VE - f VD; with VE = total rate of
             ventilation (ml/min), f = respiration rate (min~1) and VD = dead space (ml)
       OHbt: At standard concentrations, 1 g of hemoglobin will hold  1.38 ml of oxygen and thus
             OHbmax = 1.38 [Hb]/100,  with [Hb] being the hemoglobin  concentration in blood
             (g/100 ml). During and after CO exposure, the value of OHbt that must be used is
             actually OHbt = OHbmax - COHbt. In this case, the CFK equation can only be solved
             by iterative procedures.
       COHb: This value can be converted to the more conventional  ..percentage saturation" by:
             % carboxyhemoglobin = COHb 100/OHbmax

       Tikuisis et al. (1992) studied the rate of formation of COHb in healthy young males at a low
(45 W) and moderate (90 W) exercise load. Ten nonsmoking subjects were exposed to CO on two
separate occasions distinguished by the activity level. Each experiment began with an exposure to
3000 ppm for 3 minutes during a rest period followed by 3 intermittent exposures ranging from 3000
ppm for 1  minute at low exercise to 667 ppm for 3 minutes at moderate exercise. The net increase
in COHb after all  exposures (about 10 %) deviated  by <1 % between the measured and values
predicted from the CFK model. Within this deviation, there was a general tendency of the CFK
equation to underpredict the increase in COHb for  the exposures at rest and the  first exercise
exposure and to overpredict levels for the latter two  exposures at exercise.

       Benignusetal. (1994) exposed 15 men to 7652 mg/m3(6683 ppm) CO for 3.1-6.7 minutes
at rest. Except for the Haldane constant M, which was assumed to be 245, all other physiological
parameters of the CFK equation were measured for each individual  from the very beginning of
exposure. Arterial COHb was considerably higher than the venous COHb. The rate of increase in
blood  COHb and the arterial-venous COHb differences varied widely among individuals. The peak
arterial COHb at the end of exposure ranged from 13.9 to 20.9 %. The peak venous levels reached
during the recovery period ranged from 12.4 to 18.1 %. The arterial-venous difference ranged from
2.3 to  12.1 % COHb. The CFK equation overestimated venous blood COHb, whereas arterial blood
levels were significantly and consistently underestimated.

       Hilletal. (1977) developed a mathematical model to predict values of blood COHb in mother
and fetus for prolonged exposures to 30-300 ppm CO. During CO exposure, fetal COHb lag behind
maternal  COHb by several  hours. During  prolonged  uptake, fetal  levels eventually overtake
maternal levels and approach equilibrium values as much as  10% higher than the mother's, due to
the higher affinity of CO for fetal hemoglobin compared to adult hemoglobin. During CO washout
the fetal levels again lag behind the mothers.

5.      DATA ANALYSIS FOR AEGL-1
5.1.    Human Data Relevant to AEGL-1

       CO has no odor and does not cause irritative effects. A large number of studies investigated
the effects of low CO exposure (COHb <10 %)on healthy individuals and high-risk groups. In these,
effects on healthy persons, such as decreases in work capacity and decrements of neurobehavioral
function, start at COHb of 5% (WHO, 1999a; EPA, 2000).

       In  patients  with  coronary artery  disease, which constitute the  most  susceptible

                                         45

-------
CARBON MONOXIDE
                                                                 FINAL: 07/2008
subpopulation, the time to onset of angina and the time to 1-mm ST-segment change in the
electrocardiogram during physical exercise were significantly reduced at COHb of 2.0 or 4.0 %
(Allredetal. 1989a; b; 1991).

5.2.    Animal Data Relevant to AEGL-1

       No studies in experimental  animals were located that were considered relevant for the
derivation of AEGL-1 values. The studies describing effects of CO on cardiac function, such as
Sekiya et al. (1983), DeBias et al. (1979) and Aronow et al. (1979), normally employ models in
which the heart was damaged additionally by an electric stimulus or by coronary artery ligation.
Effects of CO exposure  found in these systems can hardly be extrapolated quantitatively to
humans.

5.3.    Derivation of AEGL-1

       CO is an imperceptible toxic gas. Until very severe symptoms occur (inability to walk) none
or only nonspecific symptoms were noted in healthy humans and monkeys (Haldane, 1895; Purser
and Berrill, 1983).

       In  patients  with   coronary  artery disease,  which  constitute  the  most  susceptible
subpopulation, effects, such as significant electrocardiogram changes, reduced time to the onset of
angina and  increased cardiac arrhythmia, start occurring at exposure concentrations little higher
than current ambient air quality guidelines, e.g. the U.S. National Air Quality Guideline of 9 ppmfor
8 hours (National Air Pollution Control Administration,  1970; FR , 2000; EPA, 2000; Raub, 2000),
the WHO Air Quality Guideline of 10 mg/m3 (9 ppm) for 8 hours (based on 2.5 % COHb) (WHO,
1999a) and the designated European Union Limit Value of 10 mg/m3 (9 ppm) for 8 hours (EC,
1999). These  cardiac  effects were considered  above the AEGL-1 level  and thus would not
constitute a suitable basis for the derivation of AEGL-1 values.

       AEGL-1 values are not recommended because susceptible persons may experience more
serious effects (equivalent to AEGL-2 level) at concentrations, which do not yet cause AEGL-1
effects in the general population.

       In addition, CO exposures encountered frequently  in everyday life are at or above the
concentration range, in which AEGL-1 level would have to be set: smokers have COHb in the range
of 3-8 % (Radford and Drizd,  1982) and CO concentrations of about 10-50 ppm, which  can be
found on heavily traveled roads, inside  motor vehicles and in homes with gas-, coal-, wood- or
kerosene-fired heaters and stoves, correspond to an equilibrium COHb of 1.8-7.5 % (see Figures 2
and 4).
TABLE 13: AEGL-1 VALUES FOR CARBON MONOXIDE
AEGL Level
AEGL-1
10 minutes
N.R.a
30 minutes
N.R.
1 hour
N.R.
4 hours
N.R.
8 hours
N.R.
a N.R., not recommended because susceptible persons may experience more serious effects
(equivalent to AEGL-2 level) at concentrations, which do not yet cause AEGL-1 effects in the
                                         46

-------
CARBON MONOXIDE                                                 FINAL: 07/2008

general population

6.     DATA ANALYSIS FOR AEGL-2
6.1.    Human Data Relevant to AEGL-2

       In patients with coronary artery disease, COHb of 2 or4 % significantly reduced the time to
angina and the time to 1-mm change in the ST-segment of the electrocardiogram upon physical
exercise; at 4 % the total exercise  time and the heart  rate-blood pressure product were also
significantly reduced (Allred et  al., 1989a; b; 1991). A reduced time to onset of exercise-induced
chest pain at COHb between 2.5 and 4.5 % was also reported by several other studies (Aronow et
al., 1972; Anderson et al., 1973; Sheps et al., 1987; Kleinman et al., 1989; 1998).

       Sheps et al. (1990;  1991) reported that  in patients with  coronary artery disease the
frequency of ventricular premature depolarizations was significantly increased at an COHb of 5.3 %,
but not at 3.7  %, compared to room air exposure.  Dahms et al. (1993) found no  increased
frequency of ventricular ectopic beats at COHb of 3 or 5 %.

       Klasner et al. (1998) analyzed a mass poisoning of 504 school children. In 147 of 155
children that showed symptoms, the mean COHb measured about 1 hour (up to 2 hours) after
removal from the CO atmosphere was 7.0 % COHb. Of all children that were examined in hospital
(177) (mean age 8.7 years) the  following symptoms were observed: headache (139), nausea (69),
dizziness (30), dyspnea (19), vomiting  (13), abdominal pain (11) and drowsiness (9).

       In an analysis of CO poisonings in 16 children (up to 14 years of age) with COHb of 15 % or
higher, Crocker and Walker (1985)  reported thresholds for effects, such as nausea, vomiting,
headache and  lethargy between  16.7 and 19.8 % COHb (average concentrations  in children
displaying these  symptoms were  25.9-29.4 %). Visual  symptoms  and  syncopes occurred at a
threshold of 24.5 % COHb (average 31.6-32.5 %). All  9 children  with a COHb of 24.5 % or higher
experienced at least one syncope.

       In an investigation on long-term effects of CO poisoning  in children, evaluated 2-11 years
after the poisoning, Kleesetal.  (1985) reported that 6 of the 14 children exhibited serious disorders
(spatial  organization problems, constructive apraxia,  deterioration  of lexical activity, as well as
spelling and arithmetic). Compared to the other 7 children, that exhibited only slight impairment of
visual memory and concentration, the first group of more severely affected children were younger
(mean age 7.8 years; range 2.8-12.1 years) than the latter group (mean age 9.8 years; range 3.5-
14.5); there was no difference in measured COHb (mean 21 (range 13-32)% in the first vs. 22(16-
26) % in the latter group). A short-term follow-up (3 months after the poisoning) suggested that
medium intoxications (reported COHb  16-27 %) did not produce manifest sequelae except for a
momentary standstill in the child's progress of about 2 months.

       Kizakevich etal. (1994)  reported that healthy young men can  perform submaximal exercise
without overt impairment of cardiovascular function after CO exposures attaining  20  % COHb.
Stewart et al. (1970)  found  that a CO exposure of healthy subjects resulting in 12.5 to 25.5 %
COHb did not affect the results of several neurophysiological tests. Nielsen (1971) did not report on
severe effects in three subjects that were repeatedly exposed to  CO resulting in concentrations of
25-33% COHb. In a poisoning incident at the workplace, severe headaches, dizziness, weakness,

                                          47

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

nausea, chest pain, shortness of breath and other symptoms were reported for a COHb of about 35
%(Elyetal., 1995).

6.2.    Animal Data Relevant to AEGL-2

       In a study in cynomolgus monkeys, Purser and Berrill (1983) reported that during exposure
to 900 ppm CO for a total of 30 minutes, no signs of intoxication occurred until 20-25  minutes
(corresponding to COHb of about  16-21  %). At  25  minutes  into the exposure, the animals'
performance in a behavioral test significantly decreased. At the end of the  exposure period, the
animals became less active, most of them were lying down,  but animals did not collapse. At a 1000
ppm, no effects were observed during the first 16-20 minutes. At this time the animals became less
active and sat down for short periods. At about 25 minutes, the animals went into a state of severe
intoxication within 1-2 minutes, in which animals were lying down with eyes closed, they sometimes
vomited and were virtually unable to perform coordinated movements.

       Significant memory impairment in behavioral tests were found in young rats after continuous
CO exposure throughout gestation (mean maternal COHb was 15.6 %) (Mactutus and Fechter,
1985).

       In monkeys, a COHb of 9.3 % resulted in  reduced threshold for electric shock-induced
ventricular fibrillation  (DeBias et al., 1979). Aronow et al. (1979) reported that CO exposure
increased the vulnerability of the heart to induced ventricular fibrillation in normal  dogs breathing
100 ppm CO for 2 hours (resulting COHb was 6.3-6.5 %). The ventricular fibrillation was induced
by an electrical stimulus applied to the myocardium.  A COHb of 13-15 % increased the severity and
extent of ischemic injury and the magnitude of ST-segment  elevation in a  myocardial infarction
model in dogs (Sekiya et al., 1983).

6.3.    Derivation of AEGL-2

       The derivation of AEGL-2 values was based on effects in patients with coronary artery
disease. An estimated 62 million people in the United States (about 20 % of the population) have
one or more types of cardiovascular disease (American Heart Association,  2003). For the major
diseases within the category of total cardiovascular disease, about 50 million Americans have high
blood  pressure, 13 million have ischemic (coronary) heart disease, 5 million have heart failure,  4
million have cerebrovascular disease (stroke), and 2 million  have rheumatic fever or heart disease.

       For the derivation of AEGL-2 values a level of 4 % COHb was  chosen. At this exposure
level,  patients with coronary artery disease may experience a reduced time until onset of angina
(chest pain) during physical exertion (Allred  et al., 1989a; b; 1991).

       Caracteristic  points of  an electrocardiogramm  are  the  P  wave,  reflecting atrial
depolarization, the QRS complex, representing the ventricular muscle depolarization, and the  T
wave, reflecting  ventricular muscle repolarization. In the  normal electrocardiogramm, the ST
segment is isoelectric, resting  at the same potential as the interval between the T wave and the
next P wave. Horizontal depression or a downsloping ST segment  merging into the T wave occurs
as a result of ischemia, ventricular strain,  changes in the pattern  of ventricular depolarization or
drug effects. In chronic ischemic heart disease, there may be moderate degrees of horizontal ST

                                          48

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

segment depression or a downward sloping ST segment, flattening or inversion of T waves and
prominent U  waves. It is  difficult to  define an abnormal ST segment  depression  in precise
quantitative terms. However, a myocardia ischemia has to be considered if the beginning of the ST
segment is more than 0.5 mm (corresponding to 0.05 mV) below the isoelectric line and there is an
associated T wave abnormality (Wilson et al., 1991).

       According  to the practice guidelines for chronic stable angina (Gibbons et  al., 1999),  a
ST-segment depression at rest is a marker for adverse cardiac events in patients with and without
known coronary artery disease. Additional exercise-induced ST-segment depression  in the patient
with >=1 mm  rest ST-segment depression is a reasonably sensitive indicator of coronary artery
disease. The ST-segment depression is indicative of clinically  relevant myocardial ischemia
requiring medical  treatment. From the ST-segment depression, the Duke treadmill score can be
calculated. It equals the exercise time in minutes minus (5x the ST-segment deviation, during or
after exercise, in millimeters) minus (4x the angina index, which has a value of "0"  if there is no
angina, "1" if angina occurs, and "2" if angina is the  reason for stopping the test). Among
outpatients with suspected coronary artery disease, the two thirds of patients with scores indicating
low risk (score >=5) had a four-year survival rate of 99% (average annual mortality rate 0.25%), and
the 4% who had  scores indicating high risk (score <-10) had a four-year survival  rate of 79%
(average annual mortality rate 5%) (Gibbons et al.,  1999).

       In the available experimental studies, the CO exposure alone (i.e. with subjects at rest) did
not cause angina, while exercise alone did so. Moreover, the changes in the electrocardiogram (ST-
segment depression of  1 mm or greater) as well as the angina symptoms can be considered fully
reversible after a  single incident. This  effect level was considered to be below that defined for
AEGL-2. It should  be noted that all experimental studies used patients with stable exertional angina,
who did not experience angina while at rest. Thus, it is considered  likely that in more susceptible
individuals (a part of the patients with unstable angina pectoris might belong to this group) CO
exposure alone could increase angina symptoms. In  hypersusceptible patients more severe effects,
even including myocardial infarction cannot be ruled out.

       It should be noted that in contrast to the anecdotal case reports on myocardial infarction
discussed in the derivation of AEGL-3, the studies  investigating electrocardiogram changes and
angina symptoms in patients with coronary artery disease, used here for the derivation  of AEGL-2
values,  are high-quality, well-conducted experimental studies with well-characterized exposure
conditions and information on interindividual variability.

       An exposure level of 4 % COHb is unlikely to cause a significant increase in the frequency
of exercise-induced arrhythmias. This effect has been observed at COHb of 5.3 %, but not at 3.7 %
(Sheps et al., 1990; 1991),  while in another  study  no effect of CO  exposure on  ventricular
arrhythmia was found at 3 or 5 % COHb  (Dahms et al., 1993). No experimental studies in  heart
patients are available that used significantly higher  levels of COHb.

       Use of a level of 4 % COHb as a point of departure for the derivation of AEGL-2 values  is
supported by the studies in animals: a COHb of 9.3 % resulted in a reduced threshold  for electric
shock-induced ventricular fibrillation in  monkeys (DeBias et al., 1979) and a COHb  of 6.3-6.5 %
increased the vulnerability of the heart to electrically induced ventricular fibrillation in  healthy dogs
(Aronow et al., 1979). These animal studies suggest  that a level below 6-9 % COHb should be

                                          49

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

selected for AEGL-2 derivation in order to protect individuals with compromised cardiac function.

       A total uncertainty factor of 1 forintraspecies variability was considered adequate based on
supporting evidence in other susceptible subpopulations (children, pregnant women, elderly people
and smokers):

1) The derived AEGL-2 values would result in a COHb of 4.9-5.2 % in 5-year-old children (see
Table 19 in Appendix B). This level is considered protective of neurotoxic effects in children: 1) in
the study by Klasneret al. (1998) acute neurotoxic effects, such as headache, nausea, dizziness,
dyspnea and vomiting were found at a mean COHb of 7.0% (measured after a mean time of 1 hour
(up to 2 hours) after removal of the children from the CO atmosphere). This suggests that the end of
exposure COHb had been  between  10 and 14 % (these values  were estimated  using the
mathematical model of Coburn et al. (1965) and Peterson and Stewart (1975). 2) In the study by
Crocker and Walker (1985) a threshold of 24.5 % COHb for syncopes in children, an effect that was
considered to impair the ability to escape, was reported. 3) In the study by Klees et al. (1985), that
investigated long-lasting neurotoxic effects (defects in the cognitive development and behavioral
alterations) in children, the lowest concentration resulting in cognitive development defects was 13
% COHb in the long-term follow-up study. The COHb reported in the Crocker and Walker (1985) as
well as in the Klees et al. (1985) studies were measured after hospital admission and may have
been considerably lower than levels  at the time of the end of the CO exposure, as has also been
described in the Klasneret al. (1998) study. Also the percentage of children that received oxygen
before hospital admission was probably considerably higher in these two studies since after acute
exposure to high CO concentrations (e.g. by fires in homes) severe poisoning symptoms occurred.
Oxygen administration reduces the elimination half time in children to about 44 minutes (Klasner et
al., 1998).

       The observations in children are supported by observations in experimental animals. In the
study by Purser and Berrill (1983) at COHb little higherthan 16-21  % syncope-like effects occurred
in monkeys and in mice memory impairment was found in the offspring of rats exposed continuously
at COHb of 15.6 % during gestation  (Mactutus and Fechter, 1985).

2) Caravati etal. (1988) and  Koren et al. (1991) described cases of stillbirth after CO exposure of
pregnant women. In these cases, the COHb measured in the maternal blood were higherthan 22-
25 %. There are no studies reporting effects on the unborn after a single acute exposure resulting in
lower COHb levels (EPA, 2000). Cigarette smoking of pregnant women is associated with a lower
birth weight, however,  these effects cannot be clearly attributed to CO only because cigarette
smoke is a complex mixture of chemicals (EPA, 2000). There is no evidence that a single elevation
of COHb has  any negative effects on pregnancy.

3) There is no evidence that elderly people without cardiovascular disease are more susceptible to
an acute CO exposure than younger adults (EPA, 2000; WHO, 1999a). Therefore, AEGL-2 values
derived on effects in coronary artery disease patients are likely to protect other elderly people.

4) In  smokers with  a  background  COHb  of 3-8  % from smoking, exposure to the AEGL-3
concentration-time combinations will result in 6.2. and 11.5 % COHb (see Table 19 in Appendix B).
Smokers may show an adaptive response to their chronically elevated COHb levels, as evidenced
by increased  red blood cell  volumes or reduced plasma volumes (EPA,  2000). This adaptive

                                          50

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

response is likely to reduce the effect level in smokers compared to non-smokers exposed to the
same total COHb level. The estimated COHb exposure level in smokers who are healthy adults is
unlikely to lead to significant health effects (Kizakevich et al., 1994; Stewart et al., 1970; Nielsen,
1971).  For pregnant women, cigarette smoking alone may cause effects on the unborn (EPA,
2000). A single additional exposure to COHb levels of 6.2-11.5% overthe "smoking background" of
3-8 %  COHb is considered unlikely to significantly contribute to the effects of smoking during
pregnancy. No study is available that compared the effects on the cardiovascular system of a 4 %
elevation of the background COHb level in non-smoking and smoking patients with coronary artery
disease. However, a single exposure to COHb levels of 6.2-11.5% overthe "smoking background"
of 3-8 % COHb is considered unlikely to  significantly  contribute to the effects of smoking on the
cardiovascular system.

       In conclusion, patients with coronary artery disease must be considered more susceptible to
the effects of CO than other subpopulations that may be more susceptible than healthy adults, i.e.,
children, elderly people and pregnant women. A level of 4 % COHb was  the NOEL for AEGL-2
effects in  patients with coronary artery  disease,  while the LOEL was estimated at 6-9 %.  In
comparison,  the LOEL was about 10-15 %  in children and 22-25 % in pregnant women. Since
AEGL-2 values were based on experimental data on the most susceptible subpopulation, they were
considered protective also for other subpopulations and a total uncertainty factor of 1  was used.

       Using the  CFK model (Coburn  et al.,  1965;  Peterson  and Stewart, 1975), exposure
concentrations were calculated for 10 minutes, 30 minutes, 1 hour, 4 hours and 8 hours, that would
result in a end-of-exposure COHb of 4  % in adults (see Appendix B). It should be noted that
calculations were performed for a 70-kg man with a starting COHb of 0.75 % due to endogenous
CO production and using a ventilation rate of 23 m3/day. Somewhat higher end-of-exposure COHb
would result for children. For a 5-kg child with an alveolar ventilation rate of 3580 mg/min, COHb
values between 4.9 to 5.2  % were calculated for the different AEGL time  points. For a 3.5-kg
newborn with an alveolarventilation rate of 1250 ml/min, COHb values between 5.3 and 5.6% were
calculated. Higher COHb value will also be obtained in people having a higher starting  COHb as a
result from other exposures. For smokers having typical starting COHb levels between 3 and 8 %,
COHb  values between 6.2  and 11.5 % will  result from exposure to AEGL-2 concentration-time
combinations.

       A total uncertainty factor of 1 was  used.  An intraspecies uncertainty factor of 1  was
considered adequate because the values are based  on observations in the most susceptible human
subpopulation (patients with coronary artery disease).

       It is acknowledged that apart from emergency situations, certain scenarios could lead to CO
concentrations which may cause  serious effects in  persons with cardiovascular diseases. These
scenarios  include e.g. extended exposure to traffic fume emissions (e.g., in tunnels or inside cars
with  defect car exhaust systems), charcoal or wood fire furnaces, and  indoor air pollution by
tobacco smoking.

       The values are  listed in Table  14  below.
TABLE 14: AEGL-2 VALUES FOR CARBON MONOXIDE
i i i i
i
                                          51

-------
  CARBON MONOXIDE
                                                                  FINAL: 07/2008
AEGL Level
AEGL-2
10 minutes
420 ppm
(480 mg/m3)
30 minutes
150 ppm
(170 mg/m3)
1 hour
83 ppm
(95 mg/m3)
4 hours
33 ppm
(38 mg/m3)
8 hours
27 ppm
(31 mg/m3)
  7.     DATA ANALYSIS FOR AEGL-3
  7.1.   Human Data Relevant to AEGL-3

        A large number of deaths occur annually due to acute poisonings in fires and in closed
  locations (e.g.  in private homes and workplaces). In the latter instance, poisoning usually occurs
  because gas-, oil- or coal-fired furnaces or stoves are operated without sufficient ventilation.  In
  apparently healthy people that died  from CO poisoning, usually COHb of 60 % or higher are found
  (Balraj, 1984; AIHA, 1999; AIHA, 1999; Winter and Miller, 1976, Holmes, 1985, Stewart, 1975).  In
  early experimental studies,  healthy subjects were  exposed to sufficient concentration-time
  combinations to  reach levels of about 40 to 55 % COHb  (Haldane, 1895; Chiodi et al., 1941).
  Effects described at this level of CO exposure included hyperpnea, confusion of mind, dim vision
  and unsteady/inability to walk (Haldane, 1895). Henderson etal. (1921) exposed subjects for 1 hour
  to 34-38 % COHb. Subjects showed a marked loss of equilibrium in the Romberg test, irritability,
  throbbing frontal headache and at times Cheyne-Stokes breathing was observed.

        Nelson (2005a) reported data from unvented space heaters related to human lethality cases
  related to CO poisoning. Sixteen out of 22 lethal cases had COHb levels more than 40%. Six out
  of 22 victims had COHb < 40% and  2/6 cases had pre-existing conditions such as arteriosclerotic
|  disease and cardiorespiratory failure._A 1942 fatality study reported by Nelson (2005a) summarized
  COHb data for 68 victims that were found dead in a gas-filled room or in a  garage containing
  exhaust gases at high concentrations. CO concentrations were not provided. Sixty-seven percent
  of the  68 cases died with 40-88%  COHb levels. Three-percent of the cases died with  30-40%
  COHb levels.  Summary of another fatality study from Poland showed a similar trend of COHb
  levels (Nelson, 2005a).  Individual data were not provided and the CO source was not discussed.
  However,  the  Polish study considered 321 lethal CO poisonings from 1975-1976 and provided
  COHb levels for survivors (n=220) and fatal cases (n=101).  The survivors had a mean COHb level
  of 28.1% (SD=14.1), whereas the lethal cases showed an average COHb level of 62.3% (SD=10.1).
  Over 80% of the survivors had COHb levels below 40%. In contrast, around 90%  of the deceased
  had COHb levels above 50%. Similar percentages of survivors and deceased were observed  at
  COHb levels between 40-50% with  a slight increase in the number of survivors when compared  to
  that of the lethal cases. These three studies showed a trend that most lethal cases occurred  at
  COHb levels higher than 40% and that survivorship was likely to be seen at levels below 40%.

        Another study from the Center of Forensic Sciences in Canada evaluated 304 fatal cases
  from 1965-1968 (Nelson, 2005a). The mean lethal COHb level was 51 ± 12% with  a majority range
  between 40 and 59% and the highest single frequency range at 45-59%. A report on CO exposure
  from exhaust fumes in the state of Maryland during 1966-1971 showed COHb levels in the  40-79%
  range for 98%  of lethal cases (Nelson, 2005a). The Institute of Forensic Medicine  in Oslo reported
  a study of COHb levels in automobile exhaust victims (n=54). The mean fatal COHb level was 70
  percent and 40% was the minimum COHb level exhibited by less than 2% of the cases (Nelson,
  2005a). Another forensic study (Nelson et al. 2005c) examining 2241 fatalities  between the years
  of  1976-1985  found that the mean COHb level of all the cases was 64.20% with a standard
                                          52

-------
CARBON MONOXIDE                                                 FINAL: 07/2008

deviation of 17.47. The data showed that 34% of victims had COHb levels of less than 60%.  Of
those who died in fires, 41 % had COHb levels of less than 60% compared to 22% of the non-fire
deaths.

       Kizakevich etal. (1994) reported that healthy young men can perform submaximal exercise
without overt impairment of cardiovascular function after CO exposures attaining 20  % COHb.
Stewart et al. (1970) found that a CO exposure of healthy subjects resulting in 12.5 to 25.5 % COHb
did not affect the results of several neurophysiological tests. Nielsen (1971) did not report on severe
effects in three subjects that were repeatedly exposed to CO resulting in concentrations of 25-33%
COHb.

       In susceptible groups of the population, deaths  may be caused by considerable lower
exposure to CO: Caravati etal. (1988) and Koren etal. (1991) described cases of stillbirth after CO
exposure of pregnant women. In these cases, the COHb measured in the maternal blood were
higher than 22-25%.

       Persons with coronary artery disease constitute another susceptible subpopulation (Balraj,
1984).  Several case reports  indicate that death through myocardial infarction can occur after
repeated or prolonged exposure, the corresponding COHb levels measured after transport to the
hospital (and thus not representing the end-of-exposure concentrations) were around 20-30 % and
as low as about  15 % (Atkins  and Baker, 1985; Ebisuno etal., 1986; Grace and Platt, 1981).
7.2.    Animal Data Relevant to AEGL-3

       Several studies reported LC50 values for rats, mice and guinea pigs for exposure durations
between 5 minutes and 4 hours. The values are given in Table 11 and are shown in Figure 1.
Similarto humans, the minimum lethal COHb in rats and mice were about 50-70 % (E.I. du Pont de
Nemours and Co., 1981; Rose et al., 1970).

       An increase in the rate of stillbirths was  reported in pigs after a 2-3 day-exposure to CO
resulting in maternal COHb above 23 % (Dominick and Carson, 1983).  Increased rates in fetal
mortality were also observed in rabbits after continuous exposure maternal  COHb of 16-18 %
throughout gestation (Astrup et al., 1972) as well as after daily exposure to high CO concentrations
in cigarette smoke (exposure for 12 minutes/day on gestationaldays6-18, resulting in COHb of 16
%) (Rosenkrantz et al., 1986).

7.3.    Derivation of AEGL-3

       Most of the human reports did not document for how long the victims were acutely exposed
to CO.  Despite this uncertainty in the exposure duration, it was possible to set AEGL-3 values by
using the CFK model,  which calculated the  exposure  concentrations at the various AEGL time
durations (i.e. 10 min, 30 min, 1 hr,  4 hr, 8 hr) that would produce a certain COHb% in the blood
associated with  a lethality threshold.  Although  victims of CO poisoning exhibit a wide range of
COHb levels, a weight-of-evidence analysis of numerous lethal human cases and their COHb levels
at their time of death helped setting the lethality threshold to a 40% COHb level. Note that the

                                         53

-------
CARBON MONOXIDE                                                FINAL: 07/2008

database included reports of COHb levels in individual cases or summaries where COHb data were
averaged or reported by COHb ranges. The approach of using all available data was preferred over
the selection of an individual  key study for AEGL-3 derivations because it was the only way the
evaluation could have a  broad  picture of COHb levels reported  in humans with different
demographics (e.g.  sex, age, disease status), type of CO exposure source, possible variation in
sample collection, and absence or presence of oxygen therapy to humans priorto death. Also, the
weight-of-evidence approach would average out the studies' uncertainties.

       Nelson (2005a) reported data from unvented space heaters related to deaths due to CO
poisoning. Twenty-two lethal cases were reported and 16/22 (73%) had COHb levels  more than
40%. Six out of 22 victims (27%) had COHb < 40% and 2/6 cases had pre-existing conditions such
as arteriosclerotic disease and cardiorespiratory failure.  A 1942 fatality study reported  by Nelson
(2005a) summarized COHb data for 68 victims that were found dead in a gas-filled  room or in a
garage containing exhaust gases at high concentrations. CO concentrations were not provided.
Sixty-seven percent of the 68 cases  died with 40-88% COHb levels. Three-percent  of the cases
died with  30-40% COHb levels.  Summary of another fatality study from Poland  showed a similar
trend of COHb levels (Nelson, 2005a). Individual data were not provided and the CO source was
not discussed. However, the Polish  study considered 321  lethal CO poisonings from 1975-1976
and provided COHb levels for survivors  (n=220) and fatal cases (n=101). The survivors had a
mean COHb level of 28.1%(SD=14.1), whereas the lethal cases showed an average COHb level of
62.3% (SD=10.1).  Over 80% of the  survivors had COHb levels below 40%.  In contrast, around
90% of the deceased had COHb levels above 50%. Similar percentages of survivors and deceased
were observed at COHb levels between 40-50% with a slight increase in the number of survivors
when compared  to that of the lethal cases. These three studies showed a trend that most lethal
cases occurred at COHb levels higher than 40% and that survivorship was  likely to be seen  at
levels below 40%.  Thus, 40%COHb level seems a reasonable threshold for lethality.

      Additional support comes from a  study conducted by the Center of Forensic Sciences in
Canada evaluated 304 fatal cases from 1965-1968 (Nelson, 2005a).  The mean lethal COHb level
was 51 ± 12% with a majority range between 40 and 59% and the highest single frequency range at
45-59%. A report on CO exposure from exhaust fumes in the state of Maryland during 1966-1971
showed COHb levels in the  40-79% range for 98% of lethal cases (Nelson, 2005a).  The  Institute of
Forensic Medicine in Oslo reported a study of COHb levels in automobile exhaust  victims (n=54).
The mean fatal COHb level was 70 percent and 40% was the minimum COHb level  exhibited by
less than 2% of the cases (Nelson, 2005a). Another forensic study (Nelson etal. 2005c)  examining
2241 fatalities between the years of  1976-1985 found that the mean COHb level of all the cases
was 64.20% with a standard deviation of 17.47. The data showed that 34% of victims had COHb
levels of  less than 60%.  Of those who died in  fires, 41% had COHb levels of  less  than 60%
compared to 22% of the non-fire deaths.
      The 40% COHb level is also supported by experimental studies performed in healthy human
subjects. Studies by Chiodi  etal. (1941), Henderson etal. (1921), and Haldane (1895) suggest that
a COHb of about 34-56 % does not cause lethal effects in healthy individuals. Further support
comes from the studies by Kizakevich etal. (1994), Stewart et al. (1970), and Nielsen (1971) that
reported headache as the only symptom when subjects were exposed to 20-33 % COHb. Several
case reports indicate that in patients  with coronary artery disease, CO exposure can  contribute to
myocardial infarction.  In the published cases of myocardial infarction, the following COHb values
were measured after transport to the  hospital: 52.2 % (Marius-Nunez, 1990), 30 %, 22.8 % (Atkins

                                         54

-------
CARBON MONOXIDE                                                 FINAL: 07/2008

and Baker, 1985), 21 % (Ebisuno et al., 1986),  15.6 % (Grace and Platt, 1981). A level of 40 %
COHb was used as the basis for AEGL-3 derivation. This point of departure is further supported by
studies in animals reporting minimum lethal COHb levels in rats and mice of about 50-70% (E.I. du
Pont de Nemours and Co., 1981; Rose et al., 1970).

       Another uncertainty of the human reports used to support a lethality threshold level of
40% COHb was that they did not address whether the carboxyhemoglobin measurement was
derived from a peripheral site (e.g. femoral vein) or from central blood. This type of information
is missing in many of the carbon monoxide poisoning reports. Although it remains uncertain
where the blood samples were taken from, data from Levine et al. (2002) and Dalpe-Scott et al.
(1995) ruled out significant postmortem changes in COHb levels demonstrated by similar heart
blood to peripheral blood (H/P) ratios between central and peripheral  blood.

       Using the CFK model (Coburn et  al.,  1965; Peterson  and  Stewart, 1975), exposure
concentrations were  calculated that would result in a COHb of 40% at the end of exposure periods
for 10 and 30 minutes as well as for 1, 4 and 8 hours (see Appendix B).

       AEGL-3 values calculated with an intraspecies uncertainty factor of 10 would lead to an
approximately 4% COHb level in exposed healthy adults. The values would be conservative and
more protective of susceptible populations including the developing fetus, children, and those with
compromised circulatory systems, especially at longer exposure durations. However, this is the
approximate background level in smokers (WHO 1999a). At that level,  healthy individuals have
decreases in work capacity and decrements of neurobehavioral functions (WHO 1999a; EPA 2000;
Hazucha2000). Furthermore, workers complained of light nausea, lightheadedness, and headache
at  COHb levels  of 4.1-12.8% (Atkins  and  Platt 1985).  These effects  are  below  the lethality
threshold. At slightly higher COHb level (5-6%), there may be  an increase in cardiac activity in
subjects with coronary artery disease (WHO 1999a). Therefore, a total uncertainty factor of 3 for
intraspecies variability was considered adequate based on the  following supporting evidence in
susceptible subpopulations:

1) Exposure to the derived AEGL-3 concentrations will result in COHb values of about 14-17 % in
adults (see Table 21  in Appendix B). In the reported cases of myocardial infarction, the measured
COHb was normally  above 20 %, except in one case in which the measured COHb was about 15
%.  In this  case (Grace and  Platt, 1981), the man was exposed during  several weeks to
(presumably) the same high CO concentration in  his  home  and  presented two times to the
emergency room with signs of CO intoxication (which  were misdiagnosed) until the infarction
occurred. Therefore, the derived AEGL-3 values are considered to protect heart patients against
CO-induced myocardial infarction. It should  be noted,  however, that a  clear threshold for this
endpoint cannot  be defined  because myocardial infarction might be triggered at lower COHb in
hypersusceptible individuals  and myocardial infarction can also occur spontaneously or by trigger
effects (e.g. psychological stress, physical exertion) which have no relevant effects on  the health of
normal subjects.

2) With regard to stillbirths, a COHb of 14-17 % was considered  protective of lethal effects on the
unborn, because in the case  studies available, stillbirths were found only after measured maternal
COHb of about  22-25 % or higher (Caravati et al., 1988; Koren  et al., 1991). In  the clinic, a
measured COHb of about 15-20 % in pregnantwomen (implicating a higher end-of-exposure level)

                                         55

-------
CARBON MONOXIDE
                                                                 FINAL: 07/2008
is considered a severe CO intoxication that could require hyperbaric oxygen treatment (Ellenhorn,
1997; Tomaszewski, 1998). Available animal studies reported increased rates of stillbirths after a 2-
3 day exposure at maternal COHb above 23 % (Dominick and Carson, 1983), after continuous
exposure at maternal COHb of 16-18 %(Astrupetal., 1972), and repeated short-term exposures at
16 % maternal COHb (Rosenkrantz et al., 1986). Taken together, the animal data support the
conclusion that pregnant women should not be exposed to COHb levels higherthan about 14-17%
in order to prevent lethal effects on the unborn.

3)  In  smokers with a  background COHb of 3-8 % from smoking,  exposure to the AEGL-3
concentration-time combinations will result in COHb levels between 16.1 and 23.0% (see Table 21
in Appendix B). Smokers may show an adaptive response to their chronically elevated COHb levels,
as evidenced by increased red  blood cell volumes or reduced plasma volumes (EPA, 2000). This
adaptive response is likely to reduce the effect level in smokers compared to non-smokers exposed
to the same total COHb level. The estimated  COHb exposure level in smokers is  considered
protective of lethal effects if they are healthy adults.  Also, from the discussion above, it is
considered unlikely that smoking pregnant women will have an increase risk of stillbirths at the
AEGL-3 exposure level. As discussed above, a threshold forthe induction of myocardial infarction
by CO exposure cannot be defined. Therefore, heavy smokers with coronary artery disease, which
have a higher risk for myocardial infarction already from smoking (American Heart Association,
2003), may be at somewhat higher risk compared to non-smoking patients.

       The values are listed in Table 15  below.
TABLE 15: AEGL-3 VALUES FOR CARBON MONOXIDE
AEGL Level
AEGL-3
10 minutes
1700 ppm
(1900
mg/m3)
30 minutes
600 ppm
(690 mg/m3)
1 hour
330 ppm
(380 mg/m3)
4 hours
150 ppm
(170 mg/m3)
8 hours
130 ppm
(150 mg/m3)
8.     SUMMARYOFAEGLs
8.1.   AEGL Values and Toxicity Endpoints

      The AEGL values for various levels of effects and various time periods are summarized in
Table 16. They were derived using the following key studies and methods.

      AEGL-1 values are not recommended because susceptible persons may experience more
serious effects (equivalent to AEGL-2 level) at concentrations, which do not yet cause AEGL-1
effects in the general population.

      The AEGL-2 was based on cardiovascular effects in patients with coronary artery disease,
which constitute the most susceptible subpopulation. Forthe derivation of AEGL-2 values a level of
4 % COHb was chosen. At  this  exposure level,  patients with coronary artery disease  may
experience a reduced time until onset of angina (chest pain) during physical exertion. The changes
in the electrocardiogram (ST-segment depression of 1 mm or greater) associated with angina
symptoms were fully reversible. An exposure level of 4 % COHb is unlikely to cause a significant
                                         56

-------
CARBON MONOXIDE
                                                                  FINAL: 07/2008
increase in the frequency of exercise-induced arrhythmias. A mathematical model (Coburn et al.,
1965; Peterson and Stewart, 1975) was used to calculate exposure concentrations resulting in a
COHb of 4 % at the end of exposure periods of 10 and 30 minutes and 1,  4 and 8 hours. An
intraspecies uncertainty factor  of 1 was used.  A total uncertainty factor of 1 was  used. An
intraspecies uncertainty factor of 1  was considered adequate because the values are based on
observations in the most susceptible human subpopulation (patients with coronary artery disease).

      The AEGL-3 values were based on the study on COHb levels of 40%  in human blood
derived from a weight-of-evidence analysis of lethal and non-lethal poisoning  cases (Nelson,
2005a).  A threshold for lethality of 40 % is also supported by experimental studies  by Chiodi et
al. (1941), Henderson et al. (1921), and Haldane (1895), in which  exposures resulting in COHb
of 34-56 % did not cause lethal effects in healthy individuals. Further support  comes from the
studies of Kizakevich et al. (1994), Stewart et al.  (1970), and Nielsen (1971) that reported
headache as the only symptom when health adults were exposed  to 20-33 % COHb. A level of
40 % COHb was used as the basis for AEGL-3 derivation. A mathematical model (Coburn et al.,
1965; Peterson and Stewart, 1975) was used to calculate exposure concentrations resulting in a
COHb of 40 % at the end of exposure periods of 10 and 30 minutes and 1, 4 and 8 hours. An
intraspecies uncertainty factor of 3 was used. The derived values (corresponding to a COHb
value of about 15%)  are supported by information on effects, such as myocardial infarction and
stillbirths, reported in more susceptible subpopulations.
TABLE 16: SUMMARY/RELATIONSHIP OF AEGL VALUES FOR CARBON MONOXIDE
Classification
AEGL-1
(Non^disabling)
AEGL-2
(Disabling)
AEGL-3
(Lethal)
10-Minute
N.R. a
420 ppm
(480 mg/m3)
1700 ppm
(1900 mg/m3)
30-Minute
N.R.
150 ppm
(170 mg/m3)
600 ppm
(690 mg/m3)
1-Hour
N.R.
83 ppm
(95 mg/m3)
330 ppm
(380 mg/m3)
4-Hour
N.R.
33 ppm
(38 mg/m3)
150 ppm
(170 mg/m3)
8-Hour
N.R.
27 ppm
(31 mg/m3)
130 ppm
(150 mg/m3)
a N.R., not recommended because susceptible persons may experience more serious effects
(equivalent to AEGL-2 level) at concentrations, which do not yet cause  AEGL-1 effects in the
general population

       All inhalation data are summarized in Figure 3 below. The data were classified into severity
categories chosen to fit into definitions of the AEGL level health effects.  The category severity
definitions are "No effect"; "Discomfort"; "Disabling"; "Some Lethality"; "Lethal" and "AEGL". In the
figure depicting the COHb levels, the AEGL lines are drawn at the COHb levels for adults. The grey
boxes above the lines indicate the range of COHb levels in neonates,  children and smokers (with 8
% COHb from smoking).

       The single exposure animal data point in the AEGL-2 COHb  box represents the study by
Aronowetal. (1979) using dogs with electrically damaged hearts. The two single exposure human
data points in  the  bos represent the study  by Sheps et al. (1990;  1991) reporting  increase
arrhythmiajn heart patients and the study by Klasner et al. (1998) reporting moderate neurotoxic  |
                                          57

-------
CARBON MONOXIDE



effects in children.
                                                                           FINAL: 07/2008
                                  Consistency of Data for Carbon Monoxide

                                      with Derived AEGL Values
                                        o
                                           0
                                    180     240     300

                                        Time (minutes)
                                Consistency of Data for Carbon Monoxide

                                     with Derived AEGL Values
                   °
                       c
                       o
                      Single exposure
                                                                I
                                                Repeated or continuous exposure
FIGURE 3: CATEGORICAL REPRESENTATION OF ALL CO INHALATION DATA
                                               58

-------
CARBON MONOXIDE                                                  FINAL: 07/2008

8.2.    Comparison with Other Standards and Criteria

       Other standards and guidance levels for workplace and community exposures are listed in
Table 17. The German BAT(BiologischerArbeitsstoff-Toleranz-Wert; biological exposure index) is
5 % COHb, equivalent to a concentration of 30 ppm CO (Greim und Lehnert, 1994). The ACGIH
BEI (biological exposure index) is 3.5 % COHb at the end of shift, equivalent to a CO concentration
in end exhaled air of 20 ppm (ACGIH,  1999).
TABLE 17: EXTANT STANDARDS AND GUIDELINES FOR CARBON MONOXIDE
Guideline
AEGL-1
AEGL-2
AEGL-3
ERPG-1 (AIHA)a
ERPG-2 (AIHA)
ERPG-3 (AIHA)
EEGL(NRC)b
PEL-TWA (OSHA) c
IDLH(NIOSH)d
REL-TWA(NIOSH)6
TLV-TWA (ACGIH)'
MAK (Germany) 9
MAK Spitzenbegrenzung
(Germany) h
MAC (The Netherlands) '
Einsatztoleranzwert J
WHO Air Quality Guideline k
U.S. National Ambient Air
Quality Standard '
EU Ambient Air Limit Value m
Exposure Duration
10
minutes
N.R.
420 ppm
1700 ppm



1500 ppm








87 ppm
for 15 min


30
minutes
N.R.
150 ppm
600 ppm



800 ppm

1200 ppm



60 ppm


52 ppm


1 hour
N.R.
83 ppm
330 ppm
200 ppm
350 ppm
500 ppm
400 ppm








26 ppm
35 ppm

4 hours
N.R.
33 ppm
150 ppm











100 ppm



8 hours
N.R.
27 ppm
130 ppm



50 ppm
[24 hours]
50 ppm

35 ppm
[200 ppm
ceiling]
25 ppm
30 ppm

25 ppm

9 ppm
9 ppm
9 ppm
 N.R., not recommended because susceptible persons may experience more serious effects (equivalent to
AEGL-2 level) at concentrations, which do not yet cause AEGL-1 effects in the general population.
                                          59

-------
CARBON MONOXIDE                                                      FINAL: 07/2008

a ERPG (Emergency Response Planning Guidelines, American Industrial Hygiene Association) (AIHA,
       1999)
       The ERPG-1 is the maximum airborne concentration below which it is believed nearly all individuals
       could be exposed for up to one hour without experiencing other than mild, transient adverse health
       effects or without perceiving a clearly defined objectionable odor. The ERPG-1 value is based on a
       COHbof 5-6%, which, based on the original CFK model using a ventilation rate at rest, is considered
       to be produced by 1-hour CO exposure to 200 ppm.. This exposure level is not expected to produce
       any effects during a 1 -hour exposure period. While delayed transient effects, such as headache, are
       possible, no permanent effects in more susceptible individuals are expected.
       The ERPG-2 is the maximum airborne concentration below which it is believed nearly all individuals
       could be exposed for up to one hour without experiencing or developing irreversible or other serious
       health  effects or symptoms that could impair an individual's ability to take protective  action. The
       ERPG-2 value is based on a COHb of 10-12 %, which, based on the original CFK model using a
       ventilation rate at rest, is considered to be produced by 1 -hour CO exposure to 350 to 500 ppm.. This
       exposure level is expected to cause slight neurological symptoms (increased threshold of visual light)
       in healthy individuals and chest pain  at less exertion  in heart patients. (Comment:  The  ERPG
       derivation does not discuss the CO effects on children. Moreover, model calculation  for deriving
       ERPG  values assumed a resting ventilation rate, while for derivation of AEGL values a ventilation rate
       corresponding to light to moderate activity was assumed).
       The ERPG-3 is the maximum airborne concentration below which it is believed nearly all individuals
       could be exposed for up to one hour without experiencing or developing life-threatening health effects.
       The ERPG-3 is based on the believe that humans can generally tolerate COHb of 20 % for brief
       periods without substantial toxicity. Based on the original CFK model using a ventilation rate at rest, it
       was considered that exposure to 500 ppm for 1 hour will lead to a COHb of about 15 %. (Comment:
       The ERPG derivation does not discuss the CO effects on children. Moreover, model calculation for
       deriving ERPG values assumed a resting ventilation rate, while for derivation of AEGL values a
       ventilation rate corresponding to light to moderate activity was assumed).
b EEGL (Emergency Exposure Guidance Levels, National Research Council) (NRC, 1987)
       is the concentration of contaminants that can cause discomfort or other evidence of irritation or
       intoxication in or around the workplace, but avoids death, other severe acute effects and long-term or
       chronic injury. The  NRC document states that 400 ppm (460 mg/m3) was  determined  as the
       concentration of CO to which a 1 -hour exposure would result in a carboxyhemoglobin (COHb) level of
       less than  10% in resting individuals. The committee cautions that sensitive individuals, such as
       persons with angina or heart disease,  should not be exposed to concentrations approaching the
       EEGL as they may incur serious adverse health effects (Comment: The EEGL derivation excludes
       patients with coronary artery disease. Moreover, model calculation for deriving EEGL values assumed
       a resting ventilation rate, while for derivation of AEGL values a ventilation rate corresponding to light
       to moderate activity was assumed).
c OSHA PEL-TWA (Occupational Health and Safety Administration, Permissible Exposure Limits -Time
Weighted Average) (OSHA, website)
       is defined analogous to the ACGIH-TLV-TWA, but is for exposures of no more than 10 hours/day, 40
       hours/week.
                                              60

-------
CARBON MONOXIDE                                                     FINAL: 07/2008
d IDLH (Immediately Dangerous to Life and Health, National Institute of Occupational Safety and
Health) (NIOSH, 1996)
       represents the maximum concentration from which one could escape within 30 minutes without any
       escape-impairing symptoms, or any irreversible health effects. The IDLH value is based on the
       observation by Henderson etal., 1921, that exposure of a healthy man at 1000 ppmfor 1 hour caused
       unpleasant but no dangerous symptoms,  and that more severe symptoms develop at 40 % COHb
       (Steward, 1975). According to the CFK model, a 30-minute exposure at 1200 ppm will produce a
       COHb of 10-13 %.  (Comment: The IDLH  derivation  does not discuss patients with coronary artery
       disease. In the Henderson et al. (1921)  study, the subject was sitting still during exposure and
       developed Cheyne-Stokes breathing at the end of exposure, which is considered a serious effect.
       Moreover,  model calculation in  the IDLH derivation assumed a  resting ventilation rate, while for
       derivation of AEGL values a ventilation rate corresponding to light to moderate activity was assumed).
e NIOSH REL-TWA (National Institute of Occupational Safety and  Health,  Recommended Exposure
Limits - Time Weighted Average) (NIOSH, 1996)
       is defined analogous to the ACGIH-TLV-TWA.
f ACGIH TLV-TWA (American Conference of Governmental Industrial Hygienists, Threshold LimitValue
- Time Weighted Average) (ACGIH, 2001)
       is the time-weighted average concentration fora normal 8-hourworkday and a 40-hour workweek, to
       which nearly  all workers may be repeatedly exposed, day  after day, without adverse effect. "This
       value is intended to maintain blood COHb levels below 3.5 %, to minimize the potential for adverse
       neurobehavioral changes, and to maintain cardiovascular work and exercise capacities".
9 MAK  (Maximale  Arbeitsplatzkonzentration  [Maximum  Workplace  Concentration], Deutsche
Forschungsgemeinschaft [German Research Association], Germany) (Henschler, 1981;  DFG, 1999)
       is defined analogous to the ACGIH-TLV-TWA.
h MAK Spitzenbegrenzung (Kategorie II,2) [Peak Limit Category 11,1] (DFG, 1999)
       constitutes the maximum average concentration to which workers can be exposed for a period up to
       30 minutes, with no more than 4 exposure periods per workshift; total exposure may not exceed 8-
       hour TWA MAK.
1 MAC ([Maximum Workplace Concentration], Dutch Expert Committee for Occupational Standards,
The Netherlands) (MSZW, 1999)
       is defined analogous to the ACGIH-TLV-TWA.
1  Einsatztoleranzwert  [Action Tolerance  Levels]  (Vereinigung  zur Forderung des deutschen
Brandschutzes e.V.  [Federation for the Advancement of German Fire Prevention]) (Greim, 1996)
       constitutes a concentration to which unprotected firemen and the general population can be exposed
       to for up to 4  hours without any health risks.
"Air Quality Guideline (WHO, 1999a)
       is based on a COHb of 2.5 %, which should not be exceeded even when  a normal subject engages in
       light or  moderate exercise.
'U.S. National Ambient Air Quality Standard (National Air Pollution Control Administration, 1970; FR.2000;
EPA, 2000)
m EU Limit Value for Ambient Air (EC, 1999)
                                            61

-------
CARBON MONOXIDE                                                  FINAL: 07/2008
8.3.    Data Adequacy and Research Needs

       A sufficient number of experimental and case studies  in humans is available for the
derivation of AEGL values.
       CO is the classical example of an imperceptible toxic gas. Until very severe symptoms occur
(inability to walk) none or only nonspecific symptoms were noted in monkeys and healthy humans.
For this reason no AEGL-1 values for CO are recommended.

       AEGL-2 values were based on cardiac effects in subjects with coronary artery disease.
Several high quality  studies  are available  addressing  endpoints such  as time to the onset of
exercise-induced angina, time to the onset of exercise-induced 1-mm ST-segment changes in the
electrocardiogram and frequency of exercise-induced  arrhythmias.  However, no  experimental
studies in heart patients are available that used significantly higher levels of COHbthan about 5 %
COHb.

       AEGL-3 values were based analysis of clinical  cases of lethal and non-lethal  poisoning
supporting 40% COHb as the lethality threshold. The AEGL-3 values derived using an intraspecies
uncertainty factor of 3 (corresponding to an COHb of about 15 %) are supported by the available
case reports of lethal effects (myocardial infarction, stillbirths) in more susceptible subpopulations.
Lethal effects from myocardial infarction in hypersusceptible patients with coronary artery disease
at even  lower  CO  concentrations, which could  be at  the upper end  of  the  range of CO
concentrations found inside buildings and in ambient air outside,  cannot be excluded.

       Most studies relating COHb on health effects do not investigate whether the frequency or
severity of the  effects increase with  exposure time (at  a  constant COHb). There is thus  an
uncertainty concerning the increase of effects with time at a constant COHb. This is true for all
AEGL levels. Studies elucidating this exposure-effect-time relationship could support the derived
AEGL-2 and AEGL-3 values.

9.  REFERENCES

ACGIH, American Conference of Governmental Industrial Hygienists, 2001. Carbon monoxide. Documentation
of the Threshold Limit Values and Biological Exposure Indices. ACGIH, Cincinnati, OH, USA, Suppl. pp.1-4.

ACGIH, American Conference of Governmental Industrial Hygienists, 1999. 1999 TLVs and BEIs. ACGIH,
Cincinnati, OH, USA.

AIHA, American Industrial Hygiene Association, 1999. Carbon Monoxide. Emergency Response Planning
Guidelines. AIHA Press, Fairfax, VA, USA.

Allred, E.N., E.R. Bleecker,  B.R. Chaitman, T.E. Dahms, S.O. Gottlieb, J.D. Hackney, D. Hayes, M. Pagano,
R.H. Selvester, S.M. Walden and J. Warren,  1989a. Acute Effects of Carbon  Monoxide Exposure on
Individuals with Coronary Artery  Disease. Research Report  No. 25, Health  Effects Institute, Cambridge,
Massachusetts, USA, 1989, cited in WHO, 1999a.

Allred, E.N., E.R. Bleecker, B.R. Chaitman, T.E. Dahms,  S.O. Gottlieb, J.D. Hackney, M. Pagano, R.H.


                                          62

-------
CARBON MONOXIDE                                                      FINAL: 07/2008

Selvester,  S.M. Walden and J. Warren,  1989b.  Short-term effects of carbon monoxide exposure on the
exercise performance of subjects with coronary artery disease. New Engl. J. Med. 321, 1426-1432.

Allred,  E.N., E.R.  Bleecker, B.R. Chaitman, I.E. Dahms, S.O. Gottlieb, J.D. Hackney, M. Pagano, R.H.
Selvester,  S.M. Walden and J. Warren, 1991. Effects of carbon monoxide on myocardial ischemia. Environ.
Health  Perspect. 91, 89-132.

American  Heart Association, 2003. Heart Disease and Stroke Statistics — 2003 Update. American Heart
Association, Dallas, Texas; available on the internet at http://www.americanheart.org.

Anderson, E.W., R.J. Andelman, J.M. Strauch, N.J. Fortuinand J.H. Knelson, 1973. Effect of low-level carbon
monoxide  exposure on onset and duration of angina pectoris: a study in ten patients with ischemic heart
disease. Ann. Intern. Med. 79, 46-50, cited in EPA, 2000.

Aronow, W.S., R. Charter and G. Seacat, 1979. Effect of 4% carboxyhemoglobin on human performance in
cardiac patients. Preventive Med. 8, 562-566.

Aronow, W.S., C.N. Harris, M.W. Isbell, S.N. Rokawand B. Imparato, 1972. Effect of freeway travel on angina
pectoris. Ann. Intern. Med. 77, 669-676, cited in WHO, 1999a.

Astrup, P., H.M. Olsen, D. Trolle and K. Kjeldsen, 1972. Effect of moderate carbon monoxide exposure on
fetal development. Lancet 7789, 1220-1222.

ATSDR, Agency  for  Toxic Substances and Disease Registry,  1998.  Methylene  Chloride.  Update.
U.S.Department of Health and Human Services,  Public Health Service, Agency for Toxic Substances and
Disease Registry, Atlanta, Georgia, USA.

Atkins,  E.H. and E.L. Baker, 1985. Exacerbation of coronary artery disease by occupational carbon monoxide
exposure:  A report of two fatalities and a review of the literature. Am. J. Ind. Med. 7, 73-79.

Balraj,  EX., 1984. Atherosclerotic coronary artery disease and "low" levels of carboxyhemoglobin: Report of
fatalities and discussion of pathophysiologic mechanisms of death. J. Foren. Sci. 29, 1150-1159.

Beard,  R.R., 1982. Inorganic compounds of oxygen, nitrogen, and carbon. In: Patty's Industrial Hygiene and
Toxicology. Vol. 2C,  G.D.  Clayton and F.E. Clayton  (Eds.), John Wley & Sons,  New York, 1982, pp.
4114-4124.

Benignus,  V.A., M.J.  Hazucha, M.V.  Smith and  P.A.  Bromberg, 1994.  Prediction of carboxyhemoglobin
formation due to transient exposure to carbon monoxide. J. Appl. Physiol. 76, 1739-1745, cited in  WHO,
1999a.

Bruce,  M. C., Bruce, E. N., 2006. Analysis of factors that influence rates of carbon monoxide uptake,
distribution, and washout from blood and extravascular tissues using a multicompartment model. J Appl
Physiol. 100, 1171-80.

BSI, British Standards Institution, 1989. Guide for the assessment of toxic hazards in fire in buildings and
transport. Draft for Development. British Standards Institution, Milton Keynes, UK, 1989.

Caravati, E.M., C.J. Adams, S.M. Joyce  and N.C. Schafer, 1988. Fetal toxicity associated with maternal
carbon monoxide poisoning. Ann. Emerg.  Med. 17, 714-717.

Chase, D.H., L.R. Goldbaum and NT. Lappas. 1986. Factors affecting the loss of carbon monoxide from

                                              63

-------
  CARBON MONOXIDE                                                      FINAL: 07/2008

  stored blood samples. J. Anal. Toxicol. 10: 181-189.

  Chiodi, H., D.B. Dill, F. Consolazio and S.M. Horvath, 1941. Respiratory and circulatory responses to acute
  carbon monoxide poisoning. Am. J. Physiol. 134, 683-693.

  Choi, K.D. and Y.K. Oh, 1975. [A teratological study on the effects of carbon monoxide exposure upon the
|  fetal development of albino rats] (in Korean). Chungang Uihak 29, 209-213, cited in WHO, 1999a.

  Clark, R. T., Jr., 1950. Evidence for conversion of carbon monoxide to carbon dioxide by the intact animal.
  Am J Physiol.  162,560-4.

  Coburn, R.F. (personal communication dated April 08, 2008)

  Coburn, R.F., R.E. Forsterand P.B. Kane, 1965. Considerations of the physiological variables that determine
  the blood carboxyhemoglobin concentration in man. J. Clin. Invest. 44, 1899-1910.

  Coburn, R.F., W.J. Williams, and S.B. Kahn, 1966. Endogenous carbon monoxide production in patients with
  hemolytic anemia. J. Clin. Invest. 45, 460-468, cited in WHO, 1999a.
  Crocker, P.J. and J.S. Walker, 1985. Pediatric carbon monoxide toxicity. J. Emerg. Med. 3, 443-448.

  Dahms, I.E., L.T. Younis, R.D. Wens, S. Zarnegar, S.L. Byersand B.R. Chaitman, 1993. Effects of carbon
  monoxide exposure in patients with documented cardiac arrhythmias. J. Am.  Coll. Cardiol. 21, 442-450.

  Dalpe-Scott, M., M. Degouffe,  D.  Garbutt and M. Drost, 1995. A comparison of drug concentrations in
  postmortem cardiac and peripheral blood in 320 cases. Can. Soc.  Forens. Sci. J. 28, 113-121.

  Darmer, MacEwen and Smith, 1972. AMRL-TR-72-130. Paper No. 22, cited in E.I. du Pontde Nemoursand
  Co., 1981.

  DeBias, D.A.,  C.M. Banerjee, N.C. Birkhead, C.H. Greene, S.D. Scott, and  W.V. Harrer, 1976. Effects of
  carbon monoxide inhalation on ventricular fibrillation. Arch.  Environ. Health 31, 42-46, cited in WHO, 1999a.

  Deschamps, D., C. Geraud, H. Julien, F.J. Baud and S. Dally, 2003. Memory one  month after acute carbon
  monoxide intoxication: a prospective study. Occup. Environ. Med. 60, 212-216.

  DFG, Deutsche Forschungsgemeinschaft, 1999. MAK- und BAT-Werte-Liste 1999. Senatskommission zur
  Prufung gesundheitsschadlicher Arbeitsstoffe, Deutsche Forschungsgemeinschaft, Wley-VCH Verlag GmbH,
  Weinheim, Germany, 1999.

  Dominick,  M.A. and T.L. Carson,  1983. Effects of carbon monoxide exposure on pregnant sows and their
  fetuses. Am. J. Vet. Res. 44, 35-40, cited in WHO, 1999a.

  Drummer,  O.H. 2007. Requirements for bioanalytical procedures in postmortem toxicology. Anal. Bioanal.
  Chem.  388: 1495-1503.

  Ebisuno, S., M. Yasuno, Y. Yamada, Y. Nishino, M. Hori, M. InoueandT. Kamada, 1972. Myocardial infarction
  after acute carbon monoxide poisoning: case report. Angiology 37,  621-624.

  E.I. du Pont de Nemours and Co.,  1981. Inhalation Toxicity of  Common Combustion Gases. Haskell
  Laboratory Report No. 238-81. Haskell Laboratory, Newark, DE, USA, 1981.

  Einzig,  S., D.M. Nicoloff and R.V. Lucas Jr., 1980. Myocardial perfusion abnormalities in carbon monoxide

                                               64

-------
CARBON MONOXIDE                                                      FINAL: 07/2008

poisoned dogs. Can. J. Physiol. Pharmacol. 58, 396-405, cited in Penney etal., 1980.

EC, European Commission, 1999. Vorschlag fur eine Richtlinie des Rates uber Grenzwerte fur Benzol und
Kohlenmonoxid in der Luft (1999/C 53/07), issued by the European Commission on 20.01.1999.

Ellenhorn,  M.J., 1997. "Carbon Monoxide" in: Ellenhorn's Medical Toxicology. 2nd ed. Lippincott, Williams &
Wilkins, Baltimore, pp.1465-1476.

Ely, E.W., B. Moorehead and E.F. Haponik, 1995. Warehouse workers' headache: emergency evaluation and
management of 30 patients with carbon monoxide poisoning. Am. J. Med. 98, 145-155.

EPA, U. S. Environmental Protection Agency, 2000. Air quality criteria for carbon monoxide. EPA 600/P-
99/001F, U. S. Environmental Protection Agency, Office of Research and Development, Washington, D.C.

Ernst, D.J. 2005. Performing the Venipuncture.  Chapter 3 in: Applied Phlebotomy. Lippincott Wlliams &
Wilkins, Baltimore, Maryland, pp. 59-97.

Farrow, J.R., G.J. Davis, T.M.  Roy, L.C. McCloud and G.R. Nichols II, 1990. Fetal death due to nonlethal
maternal carbon monoxide poisoning. J. Foren. Sci. 35,  1448-1452.

Fenn,W.O., 1970. The burning of CO in tissues. In: Biological effects of carbon monoxide, R.F. Coburn (Ed.).
Ann. N. Y.  Acad.  Sci. 174, 64-71, cited in WHO, 1999a.

Fenn, W.O. and D. M. Cobb, 1970. The burning of carbon monoxide by heart and skeletal muscle.  Am. J.
Physiol. 102:393-401.

Flanagan,  R. J., G. Connally, and J. M. Evans. 2005. Analytical toxicology: guidelines for sample collection
postmortem. Toxicol. Rev. 24, 63-71.

Fowles, J.R., G.V. Alexeeff and D.  Dodge, 1999. The use of  benchmark dose methodology with acute
inhalation lethality data. Reg. Toxicol. Pharmacol. 29, 262-278.

FR, Federal Register, 2000. Air Quality Criteria for Carbon Monoxide (Final). Federal Register Volume 65,
Number 160, August 17, 2000.

Gargas, M.L., H.J.  Clewell and M.E. Andersen, 1986. Metabolism of inhaled dihalomethanes in vivo:
differentiation of kinetic constants for two independent pathways. Toxicol. Appl. Pharmacol. 82, 211 -223.

Gibbons R.J.,, K. Chatterjee, J. Daley, J.S. Douglas, S.D. Fihn, J.M. Gardin, M.A. Grunwald, D. Levy, B.W.
Lytle, R.A.  O'Rourke, W.P. Schafer, S.V. Wlliams, 1999. ACC/AHA/ACP-ASIM guidelinesforthe management
of patients with chronic stable angina: a  report  of the American  College  of Cardiology/American Heart
Association Task Force on Practice_Guidelines (Committee on the Management of Patients Wth Chronic  |
Stable Angina). J. Am. Coll. Cardiol. 33, 2092-2197.

Grace,  T.W. and F.W. Platt, 1981. Subacute carbon monoxide poisoning. J. Am. Med. Assoc. 246,1698-1700.

Greim,  H.,  1996. Entwicklung von Verfahren zurAbschatzung dergesundheitlichen Folgen von GroUbranden.
Bericht zum Forschungsvorhaben 4b/92 des Bundesamtes fur Zivilschutz, Germany.

Greim,  H. and G.  Lehnert, 1994. Kohlenmonoxid. Biologische Arbeitsstoff-Toleranz-Werte (BAT-Werte) und
Expositionsaquivalente  fur  krebserzeugende  Arbeitsstoffe (EKA), Arbeitsmedizinisch-toxikologische
Begrundungen, Band 1. 7. Lieferung. DFG, Deutsche Forschungsgemeinschaft, VCH Verlag Weinheim,

                                             65

-------
CARBON MONOXIDE                                                      FINAL: 07/2008

Germany, 1994, pp.1-2.

Haldane, J., 1895. The action of carbonic acid on man. J. Physiol. 18, 430-462.

Hampson, N.B. 2008. Stability of carboxyhemoglobin in stored and mailed blood samples. Am. J. Emerg.
Med.26: 191-195.

Hartzell, G.E., D.N.  Priest and W.G.  Switzer,  1985.  Modeling  of toxicological effects of fire gases. II.
Mathematical modeling of intoxication of rats by combined carbon monoxide and hydrogen cyanide. J.  Fire
Sci. 3, 115-128, cited in NIOSH, 1996.

Haskell Laboratories, 1978. No further information, cited in E.I. du Pontde Nemours and Co.,  1981.

Hazucha, M.J., 2000. Effect of Carbon Monoxide on Work and Exercise Capacity in Humans.  Chapters in:
Carbon Monoxide Toxicity. D.G. Penney (Ed.). CRC Press, Boca Raton, London, New York, Washington D.C.,
2000, pp. 101-134.

Henderson, Y., H.W. Haggard, M.C. league, A.L. Prince and R.M. Wunderlich, 1921. Physiological effects of
automobile exhaust gas and standards of ventilation for brief exposures. J. Ind. Hyg. 3, 79-92.

Henschler,  D.,  1981. Monochloressigsaure.   In:  Gesundheitsschadliche  Arbeitsstoffe,  Toxikologisch-
arbeitsmedizinische  Begrundungen von  MAK-Werten,  Loseblattsammlung,  8.  Lfg.,  DFG, Deutsche
Forschungsgemeinschaft, VCH Verlag Weinheim, Germany.

Herpol, C., R. Minne and E. VanOutryve, 1976. Biological evaluation of the toxicity of gases produced under
fire conditions by synthetic materials. Parti. Methods and preliminary experiments concerning the reaction of
animals. Combust. Sci. Technol. 12, 217-228, cited in Fowlesetal., 1999.

Hilado, C.J.,  H.J.  Cumming, A.M. Machado, C.J. Casey and A.  Furst, 1978. Effect of individual gaseous
toxicants on mice. Proc. West. Pharmacol. Soc.  21, 159-160, cited in AIHA, 1999.

Hill, E.P., J.R. Hill,  G.G. Power and L.D. Longo, 1977. Carbon monoxide exchanges between the human fetus
and mother: a mathematical  model. Am. J. Physiol. 232, H311-H323.
Hinderliter, A.L., K.F. Adams, C.J. Price, M.C. Herbst, G.  Koch and D.S. Sheps.  Effects of low-level carbon
monoxide exposure on resting and exercise-induced ventricular arrhythmias in patients with coronary artery
disease and no baseline ectopy. Arch. Environ. Health  44, 89-93.

Holmes, R.S., 1985.  Genetic variants of enzymes of alcohol and aldehyde metabolism. Alcoholism: Clinical
and Experimental  Research 9, 535-538.

ISO, International  Standard Organization, 1989.  Toxicity testing of fire effluents- Parti. ISO/TR 9122-1.

Jones, R.A., J.A. Strickland, J.A. Stunkard and J. Siegel, 1971. Effects on experimental animals of long-term
inhalation exposure to carbon monoxide. Toxicol. Appl. Pharmacol. 19, 46-53.

Kimmerle, 1974. Combustion Toxicol. 1, 4-51, cited in E.I. du Pontde Nemours and Co., 1981.

Kishitani, K., K. Nakamura, 1979. Research on evaluation of toxicities of combustion gases generated during
fires. In Proceedings of the Third Joint Panel Conderence of the U.S.-Japan Cooperative Program in Natural
Resources held March 13-17, 1978, at the National Bureau of Standards, Gaithersburg, MD, M.D. Sherald
(Ed.). U.S. Department of Commerce, Washington,  D.C., cited in Fowlesetal., 1999.
                                              66

-------
CARBON MONOXIDE                                                      FINAL: 07/2008

Kizakevich, P.M., M.L. McCartney, M.J. Hazucha, L.H. Sleet, W.J. Jochem, A.C. Hackney and K. Bolick, 2000.
 Noninvasive ambulatory assessment of cardiac function in healthy men exposed to carbon monoxide during
upper and lower body exercise. Eur. J. Appl. Physiol. 83, 7-16.

Klasner, A.E., S.R. Smith, M.W. Thompson and A.J. Scalzo,  1998. Carbon monoxide mass exposure in a
pediatric population. Academic Emerg. Med. 5, 992-996.

Klees, M.,  M. Heremansand S. Dougan, 1985. Psychological sequelae to carbon monoxide intoxication in the
child. Sci. Total Environm. 44, 165-176.

Kleinman,  M.T., D.M. Davidson, R.B. Vandagriff, V.J. Caiozzo and J.L. Whittenberger, 1989.  Effects of short-
term exposure to carbon monoxide in subjects with coronary artery disease. Arch. Env. Health 44, 361 -369.

Kleinman,  M.T., D.A. Leaf, E. Kelly, V.  Caiozzo,  K. Osann and T. O'Niell, 1998.  Urban angina in the
mountains: effects of carbon monoxide and mild hypoxemia on subjects with chronic stable angina. Arch.
Environ. Health 53, 388-397.                                                                       |

Kojima, T., I. Okamoto, M. Yashiki, T. Miyazaki, F. Chikasue,  K. Degawa, S. Oshida, and K. Sagisaka.
1986.  Production of carbon monoxide in cadavers. Forensic Sci. Int.  32, 67-77.

Koren, G., R. Sharav, A. Pastuszak,  L.K. Garrettson, K.  Hill, I. Samson, M. Rorem, A.  King and J.E. Dolgin,
1991.  A multicenter, prospective study of fetal outcome  following accidental carbon monoxide poisoning in
pregnancy. Reprod. Toxicol. 5, 397-403.

Kunsman, G.W., C.L. Presses and P. Rodriguez. 2000.  Carbon monoxide stability in stored postmortem
blood  samples. J. Anal. Toxicol. 24: 572-578.

Landaw, S.A., 1973. The effects of cigarette smoking on total body burden and excretion  rates of carbon
monoxide. J. Occup. Med. 15, 231-235, cited in WHO, 1999a.

Larsen, J.B. 2005. Physiological Effects of Carbon Monoxide. In: Carbon Monoxide and Human
Lethality: Fire and  Non-fire studies, M.M. Hirschler (Ed.). Taylor and  Francis, New York, 2005, pp 111-172.

Lee, K.T.,  O.A. Kit and E. Jacob. 1975. Determination of carboxyhaemoglobin in blood.
Mikrochimica Acta. II: 657-663.

Levine, B.C., P.R.  Rechani, J.L. Gurman, F. Landron, H.M. Clark, M.F. Yoklavich, J.R.
Rodriguez, L.  Droz, F.M., de Cabrera and S. Kaye.  1990. Analysis of carboxyhemoglobin and cyanide in
blood  from victims of the Dupont Plaza Hotel fire in  Puerto Rico. J. Forensic Sci. 35: 151-168.

Levine, B., K.A. Moore, J.M. Titus and D.  Fowler. 2002.  A comparison of carboxyhemoglobin
saturation  values in postmortem heart blood and peripheral blood specimens. J. Forensic Sci. 47: 1388-
1390.

Luomanmaki, K., Coburn, R. F., 1969. Effects of metabolism and distribution of carbon monoxide on blood
and body stores. Am. J.  Physiol. 217, 354-63.

Mactutus,  C.F. and L.D. Fechter, 1985. Moderate prenatal carbon monoxide exposure produces persistent
and apparently permanent memory deficits in rats. Teratol. 31, 9-12.

Mohoney,  J. H.J. Vreman, O.K. Stevenson and A.L. Van Kessel. 1993. Measurement of
carboxyhemoglobin and total hemoglobin by five specialized spectrophotometers (CO-oximeters) in

                                              67

-------
CARBON MONOXIDE                                                      FINAL: 07/2008

comparison with reference methods. Clin. Chem. 39: 1693-1700.

Marius-Nunez, A.L., 1990. Myocardial infarction with normal coronary arteries after acute exposure to carbon
monoxide. Chest 97, 491-494.

Meert, K.L., S.M. Heidemann and A. P. Sarnaik, 1998. Outcome of children with carbon monoxide poisoning
treated with normobaric oxygen. J. Trauma Injury Infect. Crit. Care 44, 149-154.

Morris, G.L., S.E. Curtis and J. Simon, 1985. Perinatal piglets under sublethal concentrations of atmospheric
carbon monoxide. J. Anim. Sci, 61, 1080-1087, cited in WHO, 1999a.

MSZW, Ministerie van Sociale Zaken en Werkgelegenheid, 1999. Nationale MAC-lijst2000. Sdu Vitgeers, Den
Haag, 1999.

National Air Pollution Control Administration, 1970. Air quality criteria for carbon monoxide. US Department of
Health, Education, and Welfare, Public Health  Service, Environmental Health Service, Washington, DC,
publication no. AP-62.

Nelson.G. 2005a. Effects of Carbon Monoxide in Man. In: Carbon Monoxide and  Human Lethality: Fire
and Non-fire studies, M.M. Hirschler (Ed.). Taylor and Francis,  New York, 2005, pp 3-60.

Nelson.G. 2005b. Carbon monoxide determination in human blood. In: Carbon Monoxide and Human
Lethality:  Fire and Non-fire studies, M.M. Hirschler (Ed.). Taylor and Francis, New York, 2005, pp 173-177.

Nelson.G., D.V. Canfield  and J.B. Larsen.  2005. Carbon monoxide and fatalities. In: Carbon Monoxide
and Human Lethality: Fire and Non-fire studies,  M.M. Hirschler (Ed.). Taylor and Francis, New York, 2005,
pp 179-196.

Nielsen, B., 1971. Thermoregulation during work in carbon monoxide poisoning. Acta Physiol. Scand. 82, 98-
106.

NIOSH, National Institute for Occupational Safety and Health, 1972 . Occupational Exposure to Carbon
Monoxide. U.S. Department of Health, Education and Welfare.

NIOSH, National Institute of Occupational Safety and Health, 1996. Carbon Monoxide. Documentation for
Immediately Dangerous  to Life and Health Concentrations.  Http://www.cdc.gov/niosh/idlh/630080.html.
Download date 15.03.2000.

NRC, National Research  Council, 1987. Emergency and continuous exposure guidance levels for selected
airborne contaminants. Vol. 7. National Research Council, Commission of Life Sciences, Board on Toxicology
and Environmental Health Hazards, Committee on Toxicology, National Academy Press, Washington, DC,
USA, pp.  1738, cited in NIOSH,  1996.

Numa, A.H. and C.J. Newth, 1996. Anatomic dead space in infants and children. J. Appl. Physiol. 80, 1485-
1489.

OSHA, Occupational Health and Safety Administration. Code of Federal  Regulations 29,  Part 1910,
1910.1000. http://www.osha-slc.gov/OshStd_data/1910_1000_TABLE_Z-1 .html. Download date 04.05.2000.

Pach, J.,  L. Cholewa, Z.  Marek, M. Bogusz and B. Groszek,  1978. Analysis of predictive factors in acute
carbon monoxide poisoning. Vet. Hum. Toxicol. 21 Suppl, 158-159.
                                              68

-------
CARBON MONOXIDE                                                       FINAL: 07/2008

Pach, J., L. Cholewa, Z. Marek, M. Bogusz and B. Groszek, 1978. Various factors influencing the clinical
picture and mortality in acute carbon monoxide poisoning [in Polish]. Folia Medica Cracoviensia, 20,159-168.

Penney, D.G., M.S. Baylerian and K.E. Fanning, 1980. Temporary and lasting cardiac effects of pre- and
postnatal exposure to carbon monoxide. Toxicol. Appl. Pharmacol. 53, 271-278.

Pesce, V.H.D., M. Stupfel, V. Gourletand C. Lemercerre, 1987. Age and survival of an acute carbon monoxide
intoxication: an animal model. Sci. Total Environ. 65, 41-51.

Peterson, J.E. and R.D. Stewart, 1970. Absorption and elimination of carbon monoxide by inactive young men.
Arch. Environ. Health 21, 165-171.

Peterson,  J.E. and R.D. Stewart, 1975. Predicting the carboxyhemoglobin levels resulting  from carbon
monoxide  exposures. J. Appl. Physiol. 39, 633-638.

Purser, D.A. and K.R. Berrill, 1983. Effects of carbon monoxide on behavior in monkeys in relation to human
fire hazard. Arch. Environ. Health 38, 308-315.
                                                                                                I
Radford, E.P. and T.A.  Drizd, 1982. Blood carbon monoxide levels in  persons 3-74 years of age: United
States, 1976-80. Hyattsville, Maryland, US Department of Health and Humans Services, National Center for
Health  Statistics (DHHS Publication No. (PHS) 82-1250), cited in WHO, 1999a.

Raub, J.A., 2000. The Setting of Health-Based Standards for Ambient Carbon Monoxide and Their Impact on
Atmospheric Levels. Chapter 4 in: Carbon Monoxide Toxicity. D.G. Penney (Ed.). CRC Press, Boca Raton,
London, New York, Washington D.C., 2000, pp. 83-99.                                                |

Rice, H. M., 1976. Carboxyhaemoglobin dissociation in the cadaver following attempted resuscitation. J.
Clin. Pathol. 29, 27-9.

Rodat,  O., G.  Nicolas, A. Lugnierand P. Mangin, 1987.  Stabilite tissulaire du monoxyde de carbone apres
la mort. La Presse Medicale 16, 826-827.

Roos, R.A.C., 1994.  Neurological complications of carbon monoxide intoxication. Chapters in: Handbookof
Clinical  Neurology, Vol. 20 (64): Intoxications of the nervous system, Part I. P.J. Vinken and G.W. Bruyn
(Editors). Elsevier Science, Amsterdam, pp. 31-38.

Rose, C.S., R.A. Jones, L.J. Jenkins and J. Siegel, 1970. The acute hyperbarictoxicity of carbon monoxide.
Toxicol. Appl. Pharmacol. 17, 752-760.

Rosenkrantz,  H., R.J. Grant, R.W. Fleischman and R.J. Baker, 1986. Marihuana-induced embryotoxicity inthe
rabbit. Fundam. Appl. Toxicol. 7, 236-243.

Sekiya, S., S.  Sato, H. Yamaguchi and K. Harumi, 1983. Effects of carbon monoxide inhalation on myocardial
infarct size following experimental coronary artery ligation. Jpn.  Heart J. 24, 407-416, cited in WHO, 1999a.

Sheps, D.S., K.F. Adams Jr., P.A. Bromberg, G.M. Goldstein, J.J. O'Neil, D. Horstman and G. Koch, 1987.
Lack of effect of low levels of carboxyhemoglobin on cardiovascular function in patients with ischemic heart
disease. Arch. Environ. Health 42, 108-116.

Sheps, D.S.,  M.C. Herbst,  A.L. Hinderliter, K.F. Adams, L.G. Ekelund, J.J.  O'Neill, G.M. Goldstein, P.A.
Bromberg, J.L.  Dalton,  M.N. Ballenger, S.M. Davis and G.G.  Koch, 1990. Production  of arrhythmias by


                                              69

-------
  CARBON MONOXIDE                                                      FINAL: 07/2008

  elevated carboxyhemoglobin in patients with coronary artery disease. Ann. Intern. Med. 113, 343-351.

  Sheps, D.S., M.C. Herbst, A.L.  Hinderliter, K.F. Adams, L.G.  Ekelund, J.J. O'Neill, G.M. Goldstein, P.A.
  Bromberg,  M.  Ballenger,  S.M. Davis  and G.  Koch,  1991.  Effects of  4 Percent and  6  Percent
  Carboxyhemoglobin on Arrhythmia Production in Patients with Coronary Artery Disease. Research Report No.
  41, Health Effects Institute, Cambridge, Massachusetts.

  Singh,  J., 1986.  Early  behavioral alterations in mice following prenatal  carbon monoxide exposure.
  Neurotoxicol. 7, 475-482.

  Singh, J. and L.H. Scott, 1984. Threshold for carbon monoxide  induced fetotoxicity. Teratol. 30, 253-257.

  Sokal, J.A. and E. Kralkowska, 1985. The relationship between exposure duration, carboxyhemoglobin, blood
  glucose, pyruvate and lactate and the severity of intoxication in 39 cases of acute carbon monoxide poisoning
  in man. Arch. Toxicol. 57, 196-199.

  Stewart, R.D., 1975. The effect of carbon monoxide on humans. Annual Review of Pharmacology 15, 409-423.

  Stewart, R.D., J.E. Peterson, E.D. Baretta, R.T. Bachand, M.J. HoskoandA.A.  Herrmann, 1970. Experimental
  human exposure to carbon monoxide. Arch. Environ. Health 21, 154-164.

  Tikuisis,  P., D.M.  Kane,  T.M.  McLellan, F. Buick  and S.M. Fairburn,  1992.  Rate of formation of
  cyrboxyhemoglobin in exercising humans exposed to carbon monoxide. J. Appl. Physiol. 72, 1311-1319.

  Tomaszewski, C., 1998.  Carbon  Monoxide. Chapter 96 in: GoldfranksToxicologic Emergencies, 6th Ed., L.R.
  Goldfrank, N.E. Flomenbaum, N.A. Lewin, R.S. Weisman, M.A.  Howland and R.S. Hoffman (Eds.). Appleton
|  and Lange, Stamford, CT, 1998, p. 1551-1563.

  Vreman, H. J., R. J. Wong, D. K. Stevenson, J. E. Smialek, D. R. Fowler, L. Li, R. D. Vigorito, and H. R.
  Zielke. 2006. Concentration of carbon monoxide (CO) in postmortem human  tissues: effect of
  environmental CO exposure. J. Forensic Sci. 51, 1182-90.

  Weaver, L.K., S. Howe, R. Hopkins and K.J. Chan, 2000. Carboxyhemoglobin half-life in carbon monoxide-
  poisoned patients treated with 100% oxygen at atmospheric pressure. Chest 107, 801 -808.

  White, S.R., 2000. Pediatric Carbon Monoxide Poisoning.  Chapter 21  in: Carbon Monoxide Toxicity. D.G.
|  Penney (Ed.). CRC Press, Boca Raton, London, New York, Washington D.C., 2000, pp. 463-491.

  WHO, World Health Organization, 1999a. Environmental  Health Criteria 213, Carbon Monoxide (Second
  Edition),  IPCS,  International Programme on Chemical  Safety; World  Health  Organization,  Geneva,
  Switzerland, 1999.

  WHO (World Health Organization), 1999b. Principles for the Assessment of Risks to Human  Health from
  Exposure to Chemicals. Environmental Health Criteria 210. IPCS, International Programme on Chemical
  Safety; World Health Organization, Geneva, 110 pp.

  Wlson, J.D., E.Braunwald, K.J.  Isselbacher, R.G. Petersdorf, J.B. Martin, A.S. Fauci and R.K.  Root (Eds.)
  1991. Harrison's Principles of Internal Medicine. 12th Ed. McGraw-Hill, Inc., New York.

  Wnek, C. L., Prex, D. M., 1981.  A comparative study of analytical methods to determine postmortem
  changes in carbon monoxide concentration. Forensic Sci. Int. 18, 181-7.
                                                70

-------
CARBON MONOXIDE                                                      FINAL: 07/2008

Winston, J.M. and R.J. Roberts, 1978. Influence of increasing age on lethality induced by carbon monoxide or
hypoxic hypoxia. Biol. Neonate 34, 199-202.

Winter, P.M. and J.N. Miller, 1976. Carbon monoxide poisoning. J. Am. Med. Assoc. 236, 1503.
                                              71

-------
CARBON MONOXIDE                                             FINAL: 07/2008
                                  APPENDIX A




                       Time Scaling Calculations for AEGLs
                                      72

-------
CARBON MONOXIDE




Key study:

Toxicity endpoint:
                                                                FINAL: 07/2008
                        Time Scaling Calculations for AEGL-2

                   Allred et al. (1989a; b; 1991); Sheps et al. (1990; 1991)

                   In an experimental study in 63 subjects with coronary artery disease, a
                   significantly   reduced  time   to  ST-segment  depression  in  the
                   electrocardiogram and a significantly reduced time to  onset of angina
                   pectoris during physical exercise were found at 2 or 4 % COHb (Allred et al.,
                   1989a; b; 1991).  At higher COHb of 5.3,  but not at 3.7 %, a significantly
                   increased frequency of exercise-induced arrhythmias was found (Sheps et
                   al., 1990; 1991). AEGL-2 values were derived on a COHb of 4 %.

Mathematical model: The CFK model (Coburn et al., 1965; Peterson and Stewart, 1975) was used
                   to calculate exposure concentrations resulting in a COHb of 4 % at the end
                   of the exposure periods.  Concentrations were calculated  for 10 and 30
                   minutes, 1, 4 and  8 hours (see Appendix B).
Scaling:



Uncertainty factors:


Calculations:

10-minute AEGL-2

30-minute AEGL-2

1-hour AEGL-2

4-hour AEGL-2

8-hour AEGL-2
                   Instead of a time scaling according to C"xT = const., a mathematical model
                   was used to calculated exposure concentrations for the relevant time periods
                   (see Appendix B).

                   Uncertainty factor of 1
                   1 for intraspecies variability
                   10-min AEGL-2 = 424 ppm/1 = 420 ppm (480 mg/m3)

                   30-min AEGL-2 = 150 ppm/1 = 150 ppm (170 mg/m3)

                          1-hour AEGL-2 = 83 ppm/1 = 83 ppm (95 mg/m3)

                          4-hour AEGL-2 = 33 ppm/1 = 33 ppm (38 mg/m3)

                          8-hour AEGL-2 = 27 ppm/1 = 27 ppm (31 mg/m3)
                                         73

-------
CARBON MONOXIDE
                                                                FINAL: 07/2008
Key study:
Toxicity endpoint:
                        Time Scaling Calculations for AEGL-3

                   Nelson (2005a) ;Chiodi et al. (1941); Henderson et al. (1921); Haldane
                   1895)

                   Exposure of healthy subjects to sufficient concentration-time combinations to
                   reach levels of about 34 to 56 % COHb did not result in severe or life-
                   threatening  effects. At this level of CO exposure, Haldane  described
                   symptoms  including  hyperpnea,  confusion of  mind,  dim vision  and
                   unsteady/inability to walk. Also, analysis of lethal cases reported by Nelson
                   (2005a) indicated that most lethal poisoning cases occurred at COHb levels
                   higher than 40% and that survival  of CO-exposed humans were likely to be
                   seen at levels below 40%.  Thus, 40%COHb level seems a reasonable
                   threshold for lethality.

Mathematical model: The CFK model (Coburn et al., 1965; Peterson and Stewart, 1975) was used
                   to calculate exposure concentrations resulting in a COHb of 40 % at the end
                   of the exposure periods. Concentrations were calculated for 10 and 30
                   minutes and 1, 4 and 8 hours (see Appendix B).
Scaling:
                   Instead of a time scaling according to C"xT = const., a mathematical model
                   was used to calculated exposure concentrations for the relevant time periods
                   (see Appendix B).
Uncertainty factors:  Total uncertainty factor of 3
                   3 for intraspecies variability
Calculations:

10-minute AEGL-3   10-min AEGL-3 = 5120 ppm/3 = 1700 ppm (1900 mg/m3)

30-minute AEGL-3   30-min AEGL-3 = 1810 ppm/3 = 600 ppm (690 mg/m3)

1-hour AEGL-3             1-hour AEGL-3 = 998 ppm/3 = 330 ppm (380 mg/m3)

4-hour AEGL-3             4-hour AEGL-3 = 439 ppm/3 = 150 ppm (170 mg/m3)

8-hour AEGL-3             8-hour AEGL-3 = 403 ppm/3 = 130 ppm (150 mg/m3)


The COHb levels corresponding to the AEGL-3 values are given in Table 21 in Appendix B.
                                         74

-------
CARBON MONOXIDE                                             FINAL: 07/2008
                                 APPENDIX B




        Mathematical Model for Calculating COHb and Exposure Concentrations
                                      75

-------
CARBON MONOXIDE                                                 FINAL: 07/2008

        Mathematical Model for Calculating COHb and Exposure Concentrations

Study describing
model:       Coburn et al. (1965); Peterson and Stewart (1975)

Model:       Forthe calculation of concentration-time combinations that result in a certain COHb,
             the model of Coburn, Forsterand Kane (CFK model) (see Section 4.4.4) was used.

             Since this model in the formulation of Peterson and Stewart (1975) calculates COHb
             larger than 100 % at high exposure concentrations, the following  correction
             proposed by Peterson and Stewart (1975) was used: the amount of bound oxygen is
             actually not constant,  but is dependent on the COHb, therefore:
                   OHbt=OHbmax-COHbt.

             Since in this case, the CFK equation can only be solved iteratively, calculations
             were done using time steps (At) of 1 minute for the period of 0-10 minutes, steps of
             5 minutes between  10 and  60 minutes, steps of 15 minutes between 60 and 240
             minutes, and steps of 20 minutes between 240 and 480 minutes. In each step, the
             COHb of the step before  was  used to  calculate  OHbt. For  the  first step, a
             background COHb of  0.75 % was assumed.

             The alveolar ventilation rate was calculated as:
                   VA = VE - f VD (Peterson and Stewart, 1975);
             with   VE = total rate  of ventilation (ml/min),
                   f = respiration  rate (min~1) and
                   VD = dead space (ml).

             Derivations were done for  a 70-kg  man, assuming a blood volume of 5500  ml
             (Coburn et al., 1965) and a daily inhalation  volume (VE) of 23 m3 (8 hours resting
             and 16 hours light/non-occupational activity; WHO, 1999b), a respiration rate (f) of
             18 min"1 and a dead space (VD) of 2.2 ml/kg (Numa and Newth, 1996).
                                         76

-------
CARBON MONOXIDE
                                                              FINAL: 07/2008
            Calculations  using the following equation  were carried  out in a spreadsheet
            computer program:
              A (COHb)t =
                                  V,
                                   CO
                                         COHbt_, * P02
P,
                                  Vb    M*B*Vb(OHbm^-COHbt_l}    B*
where:
temperature,
COHbt = ml of CO per ml blood at time t (min)
      Conversion: % carboxyhemoglobin = COHb 100 /OHbmax
Vco = rate of endogenous CO production; Vco = 0.007 ml/min
Vb = blood volume; Vb (70-kg man) = 5500 ml; Vb (5-yr child, 20 kg) = 1500 ml;
            Vb (newborn, 3.5 kg) = 400 ml
M = Ratio of affinity of blood for CO to that for O2; M = 218 (newborn: M = 240)
B = 1 / DL + PL / VA
with:  DL = diffusivity of the lung for CO; DL = 30 ml / min mm Hg
      PL = barometric  pressure  minus the vapor pressure of water at body

      PL = 713 mm Hg
      VA = alveolar ventilation rate;
VA (70-kg man) = 23 m3/d * 1 • 106 ml/m3 * 1/1440 min/d -18 /min * 2,2 ml/kg * 70 kg
VA (70-kg man) = 13200 ml/min
VA (5-yr child) = 3580 ml/min
VA (newborn) = 1250 ml/min
OHbmax = ml of O2 per ml blood under normal conditions; OHb = 0,2
P02 = average partial pressure of oxygen in the lung capillaries; P02 = 100 mm Hg
PCO = partial pressure of CO in the air inhaled (mm Hg);
      Conversion: PCo (mm Hg) = PCo (ppm) /1316
t = exposure duration (min)
                                       77

-------
CARBON MONOXIDE
                                                       FINAL: 07/2008
       100
     .0
     I
     o
     o
10
                                      10                        100

                                           Exposure time (min)
                                                                                 1.000
FIGURE 4: COHb VS. EXPOSURE TIME PLOTS

Data are shown for CO exposure concentrations indicated (70-kg man).
                                   78

-------
CARBON MONOXIDE
                                                                     FINAL: 07/2008
  CFK Modell for Calculation of COHb
  Dr. Peter Gr em                                 Physiologic parameters.
  Mtxielby Cotium, Former anc Kane (1965) w!t?i ccfrecGore Introduced By Petersor ana 70-kg    20-kg crrld 3.5-kg
  sie*ar: n&75)                                    ad:ult           newborn
  Model parameters (see TSD):
PL
M
OHb
PO2
Vb
Voo
D
Va
DL
COHbt
COHbo
ExpT.ro
Dtp. Cone.
CO
713 mm Hg
218
0.2 ml/nil blood
100 mmHg
5500 ml
0.007 ml/irtn
0.0015 ml CO/rrJ blood
13200 ml/nln
30 ml.'min mm Hg
0.02 ml CO/if. blood
00015 nil CC.'rr MOOT
60 min


Auxiliary expressions:
A
B
COHbt
COHbo
a

calculates COHb
a'ter exps&jre to
COHb: ^J^^^^_^^^^_
2.293578
0.0373485
0.02
0.0015
0.7509257

>?:c:rc ic to orlg ns CFK node :



in %: 0.75


in %: 10
in %: 0.75



Results for exposure to
time(rrin) COHb (%)
10 1.3631498
30 2.5132273
60 4.0330397
240 9.4791254
4&0 11.619731

83 ppm for 60 min:
4 07OQ77
200
0.27
SE:O 1500 400


13200 3580 1250






83 ppm:








Cacu:31ec COHb according la n*sdel by CBbum.
Footer and Kane
F=:=-sor and ste
time (min)

1
2
3
4
5
6
7
8
9
10
15
20
25
30
35
40
45
50
55
60
15?5| '.vtn ccr-ectoria hfodLt=d by
t3rt :19T5;
dt

1
1








5
5
5
5
5
5
5
5
5
5


dHbCO HbCO
0.0015
0.0001253 0.0016253
0.0001247 0.0017501
0.0001241 0.0018742
0.0001235 0.0019977
0.0001229 0.0021206
0.0001223 0.002243
0.0001217 0.0023647
0.0001211 0.0024358
0.0001235 0.0026064
0.0001199 0.0027263
0.0005968 0.0033231
0.0005821 0.0039052
0.0005677 0.0044729
0.0005536 0.0050265
0.0005397 0.0055661
0.0005261 0.0060922
0.0005128 0.006505
0.0004997 0.0071047
0.0004369 0.0075917
0.0004744 0.0080661


%
0.75
0.81267
C. 875:35
0.937098
0.998659
i.:6-:3i9
1.12148
1.182343
1.2425:6
1.303176
1.36315
1.661547
1 .9526
2.236448
2.513227
2.753274
3.2^E124
3.302513
3.552373
3.795838

FIGURE 5: CALCULATION OF 60-MINUTE AEGL-2 FOR HEALTHY ADULT
Calculations:  For the derivation of AEGL-2 values, exposure concentrations were calculated that
              would result in a COHb of 4 %. A representation of the spreadsheet for the 60-
              minute AEGL-2 is shown in Figure 5. Results are shown in the following Table 18.

-------
CARBON MONOXIDE
                                                                 FINAL: 07/2008
TABLE 18: CONCENTRATION-TIME COMBINATIONS RESULTING IN
Exposure time (min)
10
30
60
240
480
4%COHb
for a 70-kg adult man
exposure concentration
(ppm)
424
150
83
33
27
exposure concentration
(ppm), rounded
420
150
83
33
27
             For children, newborns and adult smokers, the end-of-exposure COHb values for
             exposure to the concentrations calculated in Table 18 were computed using the
             CFK model:
TABLE 19: COHb VALUES FOR AEGL-2 CONCENTRATION-TIME COMBINATIONS IN
DIFFERENT SUBPOPULATIONS
Exposure Exposure 5-yr Newborn
time concentration Child
(min) (ppm)
10 420 5.2 5.5
30 150 5.2 5.6
60 83 5.2 5.6
240 33 5.0 5.4
480 27 4.9 5.3
Healthy adult
4.0
4.0
4.0
4.0
4.0
Smoker
(3%
COHb)
6.2
6.3
6.4
6.6
6.7
Smoker
(8%
COHb)
11.2
11.3
11.4
11.5
11.5
             For the derivation of AEGL-3 values, exposure concentrations were calculated that
             would result in a COHb of 40 %. A representation of the spreadsheet for the  60-
             minute value is shown in Figure 6. Results are shown in the following Table 20.
                                         80

-------
CARBON MONOXIDE
                                                                FINAL: 07/2008
TABLE 20: CONCENTRATION-TIME COMBINATIONS RESULTING IN
Exposure time (min)
10
30
60
240
480
40%COHb
for a 70-kg adult man
exposure concentration
(ppm)
5120
1810
998
439
403
exposure concentration
(ppm), rounded
5100
1800
1000
440
400
       For children, newborns, healthy non-smoking adults and smokers, the end-of-exposure
       COHb values for exposure to the AEGL-3 exposure concentration-time combinations were
       computed using the CFK model. For all subpopulations, the endogenous CO production
       rate was adjusted so that the starting level of 0.75 % for children and newborn and 3 and 8
       % for smokers were constant without additional CO exposure.
TABLE 21 : COHb VALUES FOR AEGL-3 CONCENTRATION-TIME COMBINATIONS IN
DIFFERENT SUBPOPULATIONS
Exposure Exposure 5-yr Newbor
time concentration Child n
(min) (ppm)
10 1700 18.7 19.9
30 600 18.5 19.8
60 330 18.3 19.6
240 150 18.6 20.1
480 130 18.1 19.5
Healthy
adult
13.8
14.0
14.1
16.4
17.2
Smoker
(3 % COHb)
16.1
16.2
16.4
18.6
19.2
Smoker
(8% COHb)
21.1
21.1
21.2
22.7
23.0
                                         81

-------
CARBON MONOXIDE
                                                                   FINAL: 07/2008
         CFK Model! for Calculation of COHb
         Dr. Peter Griem                                    Physiologic parameters:
         MotSil by Cabjrr ForsteranQ Kans I19C-5I witncorrac-:tons IntroDucea by Ps:efsor ard JQ^Q    20-kg child 3 5-kg
                                                       adult            newborn
5le*ai:i1J75)
         Model oa'amelers (see TSD',:
PL 713mmHg
M 218
OHb 0.2 ml/ml blood
PO2 1DD mm Hg
Vb 55DD ml
Vco 0.007 rnlj'min
D 0.0015 mICO/ml biood
Va 13200 ml/min
DL 3D ml/min mm Hg
COKbt C.D2 ml CO.'ml blood
COhco C.CC15 ml CO'm b coo
Exp. Time £3 mm
Exp. Cone-
CO 868 ppm
Aux.liary expressions:
A 2-29357B
B 0.0373485
COHbt 0.02
COHbo D.B015
a 0.7508257

caiDi atea CCHS :aco:rc 13 :o o-gns CFK nocsn
arer exposure to
COHb: 0-083547B ml;'ml blood
CalsialeJ CC-lc aeCD-dhg to -ncOeiBj CoCun.
For&ter ans Kaie i;J9€Ei WEI correcusns irfocjc^c ty
Peterson anfl Ste*art (1975}:
time (min) dt

t 1
2 1
3 1
4 1
5 1
6 1
7 1
e 1
8 1
10 1
15 5
20 5
25 5
3D 5
35 5
4C E
4£ 5
SO 5
55 5
6D 5






in %:


in %:
in%:









0.76


10
D.7S



Results for e>:posjre tc
lime (mm)
10
3D
SB
240
48D

3S3 po-71




dHbCO

0.0015726
0.0015648
0.0015572
0.0015493
0.0015414
0.0015333
0.0015252
D. DO" 5'7
0.0015087
0.0015003
0,0074583
0.0072378
0.0070043
0.0067554
O.OQ64998
0.0062291
3. D 359462
0.0056522
0.0053477
0.0050342
COHb (%)
8.434959S
22.884894
40.019622
62.314719
62.3289

for 80 r-in:




HbCO
0.0015
0.0030726
0.0046375
O.OQ61S47
D. 007744
0.0092553
0.01081S7
0.012343S
0.0138608
0.0153696
0.0168695
0.0243292
0.0315671
0.0385714
0.0453298
0.0518297
Q.058058S
0.0640051
0.0696572
D.D75005
O.OB00392




5500 1500


1 3200 3580






398 ppm:








^H



%
n.75
1.536301
2.31878
3.D87335
3.871833
4.642682
5.^09326
6.171832
6.&3D437
7.684784
8.43496
12.1646
15.79355
18.28572
22.884.B9
25.81485
2S.D2839
32.D0254
34.S28S2
37.50243
4Q.D18S2

200
0.27

400


1250









































FIGURE 6: CALCULATION OF 60-MINUTE EXPOSURE CONCENTRATION THAT WOULD.
RESULT IN 40 % COHb IN A HEALTHY ADULT

-------
CARBON MONOXIDE
                                                                   FINAL: 07/2008
             The following end-of-exposure COHb values were calculated for the series of
             experiments reported by Haldane (1895). Since exposure occurred while the subject
             was sitting on a chair,  a ventilation rate of 7.5 l/min was used for the calculation
             (WHO, 1999b). The alveolar ventilation rate was calculated as:
             VA (70-kg man) = 3600  I/8 h * 1 • 103 ml/l* 1/480 min/8 h -18 /min * 2,2 ml/kg * 70 kg
             VA (70-kg man) = 4700 ml/min
TABLE 22:
Experiment
No.
1
2
3
4
5
6
7
8
9
10
11
COMPARISON O
VALUES FORTH
Concentration
(ppm)
5000
3900
4000
3600
4100
1200
2100
irregular
270
210
460
F REPOR1
E DATA B
Time
(min)
11.5
30.5
24
29
29
120
71
35
210
240
240
FEDANDCALCUL
Y HALDANE (189S
COHb
measured (%)
not done
39
27
37
35
37
49
56
14
13
23
ATEDCOHb
)
COHb
calculated (%)
22
43
35
38
43
46
50
-
17
15
30
                                           83

-------
CARBON MONOXIDE                                             FINAL: 07/2008
                                 APPENDIX C




                 Derivation Summary for Carbon Monoxide AEGLs
                                      84

-------
   CARBON MONOXIDE                                                FINAL: 07/2008

   ACUTE EXPOSURE GUIDELINES FOR CARBON MONOXIDE
  	(CAS NO. 630-08-0)	
10 minutes
N.R.
30 minutes
N.R.
1 hour
N.R.
4 hours
N.R.
8 hours
N.R.
AEGL-1 VALUES
a N.R., not recommended because susceptible persons may experience more serious effects
(equivalent to AEGL-2 level) at concentrations, which do not yet cause AEGL-1 effects in the general
population	
Reference: Not applicable
Test Species/Strain/Number: Not applicable / not applicable / not applicable
Exposure Route/Concentrations/Durations: Not applicable / not applicable / not applicable
Effects: Not applicable
Endpoint/Concentration/Rationale:
CO is the classical example of a tasteless, non-irritating, odorless and colorless toxic gas. Until very
severe symptoms occur (inability to walk) none or only nonspecific symptoms were noted in monkeys
and healthy humans (Haldane,  1895; Purser and Berrill, 1983). In patients with coronary artery disease,
which constitutes the most susceptible subpopulation, effects, such as significant electrocardiogram
changes, reduced time to the onset and increased cardiac arrhythmia, start occurring at exposure
concentrations little higher than current ambient air quality guidelines, e.g. the U.S. National Air Quality
Guideline of 9 ppm for 8 hours (National Air Pollution Control Administration, 1970; FR , 2000; EPA,
2000), the WHO Air Quality Guideline of 10 mg/m3 (9 ppm) for 8 hours (based on 2.5 % COHb) (WHO,
1999a) and the designated European Union Limit Value of 10 mg/m3 (9 ppm) for 8 hours (EC, 1999).
These effects were considered  above the AEGL-1 effect level and thus would not constitute a suitable
basis for the derivation of AEGL-1 values. AEGL-1 values were not recommended because susceptible
persons may experience more serious effects (equivalent to AEGL-2 level) at concentrations, which do
not yet cause AEGL-1 effects in the general population._ln addition, CO exposures encountered    |
frequently in everyday life are at or above the concentration range,  in which AEGL-1 level would have
to be set: smokers have COHb  in  the range  of 3-8 % (Radford and Drizd, 1982) and CO concentrations
between about 10 and 50 ppm, which can be found on heavily traveled roads, inside motor vehicles
and in homes with gas-, coal-, wood- or kerosene-fired heaters and stoves, correspond to an
equilibrium COHb of 1.8-7.5 %  (see Figures 2 and 4).	
Uncertainty Factors/Rationale: Not applicable
Modifying Factor: Not applicable
Animal to Human Dosimetric Adjustment: Not applicable
Time Scaling: Not applicable
Data Adequacy: Not applicable
                                            85

-------
CARBON MONOXIDE                                                 FINAL: 07/2008

       ACUTE EXPOSURE GUIDELINES FOR CARBON MONOXIDE
	(CAS NO. 630-08-0)	
10 minutes
420 ppm
30 minutes
150 ppm
1 hour
83 ppm
4 hours
33 ppm
8 hours
27 ppm
  AEGL-2 VALUES
  Reference: Allred, E.N., E.R. Bleecker, B.R. Chaitman, I.E. Dahms, S.O. Gottlieb, J.D. Hackney, D.
  Hayes, M. Pagano, R.H. Selvester, S.M. Walden and J. Warren,  1989a. Acute Effects of Carbon
  Monoxide Exposure on Individuals with Coronary Artery Disease. Research Report No. 25, Health
  Effects Institute, Cambridge, Massachusetts, USA, 1989; Allred,  E.N., E.R. Bleecker, B.R. Chaitman,
  I.E. Dahms, S.O. Gottlieb, J.D. Hackney,  M. Pagano, R.H. Selvester, S.M. Walden and J. Warren,
  1989b. Short-term effects of carbon monoxide  exposure on the exercise performance of subjects with
  coronary artery disease. New England Journal of Medicine 321, 1426-1432; Allred, E.N., E.R. Bleecker,
  B.R. Chaitman, I.E. Dahms, S.O. Gottlieb, J.D. Hackney,  M. Pagano, R.H. Selvester, S.M. Walden and
  J. Warren, 1991. Effects of carbon monoxide on  myocardial ischemia. Environmental Health
  Perspectives 91, 89-132.	
  Test Species/Strain/Sex/Number: Humans with coronary artery disease / not applicable / male / 63
  Exposure Route/Concentrations/Durations: Inhalation / mean concentrations of 0, 117 or 253 ppm for
  50-70 minutes were used, adjusted individually to reach carboxyhemoglobin concentrations of 2.2 % or
  4.4 % at the end of exposure (about 2 or 4 % COHb in the subsequent exercise tests)	
  Effects:
  When potential exacerbation of the exercise-induced ischemia by exposure to CO was tested using the
  objective measure of time to 1-mm ST-segment change in the electrocardiogram, exposure to CO
  levels producing  COHb of 2 % resulted in a overall statistically significant 5.1  % decrease in the time to
  attain this level of ischemia. For individual centers (patients were tested in one of three centers), results
  were significant in one, borderline significant in one and nonsignificant in one center. At 4 % COHb, the
  decrease in time to the ST criterion was 12.1% (statistically significant for all patients, the effect was
  found in 49/62 subjects) relative to the air-day results. Significant effects were found in all three test
  centers. The maximal amplitude of the ST-segment change was also significantly affected by the
  carbon monoxide exposures: at 2 % COHb the maximal increase was 11 % and at 4 % COHb the
  increase was 17  % relative to the air day.
  At 2 % COHb, the time to exercise-induced angina was reduced by 4.2  % in all patients (effects were
  significant in two test centers and nonsignificant in  one center). At 4 % COHb, the time was reduced by
  7.1 % in all patients (effects were significant in one, borderline significant in one and nonsignificant in
  one center). The two end-points (time to  angina and time to ST change) were also significantly
  correlated. Only at 4 % COHb a significant reduction in the total exercise time and in the heart rate-
  blood pressure product was found (this double product provides a clinical index of the work of the heart
  and myocardial oxygen consumption).	
  Endpoint/Concentration/Rationale:
  Patients with coronary artery disease show health effects at lower COHb levels than children, pregnant
  women or healthy adults and, thus, constitute the most susceptible subpopulation. For the derivation of

                                         86

-------
CARBON MONOXIDE                                                  FINAL: 07/2008
 AEGL-2 values a level of 4 % COHb was chosen. At this exposure level, patients with coronary artery
 disease may experience a reduced time until onset of angina (chest pain) during physical exertion
 (Allred et al., 1989; 1991). In the available studies, the CO exposure alone (i.e. with subjects at rest) did
 not cause angina, while exercise alone did so. However, it should be noted that all studies used
 patients with stable exertional angina, who did not experience angina while at rest. Thus, it cannot be
 ruled out that in more susceptible individuals (a part of the patients with unstable angina pectoris might
 belong to this group) CO exposure alone could increase angina symptoms. The changes in the
 electrocardiogram (ST-segment depression of 1 mm or greater) associated with angina symptoms were
 considered reversible, but is indicative of clinically relevant myocardial  ischemia requiring medical
 treatment. An exposure level of 4 % COHb is unlikely  to cause a significant increase in the frequency of
 exercise-induced arrhythmias. Ventricular arrhythmias have been observed at COHb of 5.3 %, but not
 at 3.7 % (Sheps et al., 1990; 1991), while in another study no effect of  CO exposure on ventricular
 arrhythmia was found at 3 or 5 % COHb (Dahms et al., 1993). An exposure level of 4 % COHb was
 considered protective of acute neurotoxic effects in children, such as syncopes, headache, nausea,
 dizziness and dyspnea (Klasner et al., 1998; Crocker  and Walker, 1985), and long-lasting neurotoxic
 effects (defects in the cognitive development and behavioral alterations) in children (Klees et al., 1985).
 It is acknowledged that apart from emergency situations, certain scenarios could lead to CO
 concentrations which may cause serious effects in persons with cardiovascular diseases. These
 scenarios include e.g.  extended exposure to traffic fume emissions (e.g., in tunnels or inside cars with
 defect car exhaust systems), charcoal or wood fire furnaces,  and indoor air pollution by tobacco
 smoking.

 Uncertainty  Factors/Rationale:
 Total uncertainty factor: 1
 Interspecies:  Not applicable
 Intraspecies:  1 - A  level of 4 % COHb was the NOEL for AEGL-2 effects in patients with coronary
               artery disease, while the LOEL was estimated at 6-9 %. In comparison, the LOEL was
               about 10-15 % in children and 22-25 % in pregnant women. Since AEGL-2 values were
               based on experimental data on the most susceptible subpopulation, they were
               considered  protective also for other subpopulations and a total uncertainty factor of 1
               was used.
 Modifying Factor: Not applicable
 Animal to Human Dosimetric Adjustment: Not applicable
 Time Scaling:
 A mathematical model (Coburn et al., 1965; Peterson and Stewart, 1975) was used to calculate
 exposure concentrations in air resulting in a COHb of 4 % at the end of exposure periods of 10 and 30
 minutes and 1, 4 and 8 hours.	
 Data Adequacy:
 AEGL-2 values were based on cardiac effects in subjects with coronary artery disease, which constitute
 the most susceptible subpopulation. Several high quality studies are available addressing endpoints
 such as time to the onset of exercise-induced angina, time to the onset of exercise-induced 1-mm ST-
 segment changes in the electrocardiogram and frequency of exercise-induced arrhythmias. However,
 no experimental studies in  heart patients are available that used significantly higher levels of COHb
 than about 5 % COHb.


                                          87

-------
CARBON MONOXIDE
                                                                FINAL: 07/2008
      ACUTE EXPOSURE GUIDELINES FOR CARBON MONOXIDE
     	(CAS NO. 630-08-0)	
10 minutes
1700 ppm
30 minutes
600 ppm
1 hour
330 ppm
4 hours
150 ppm
8 hours
130 ppm
 AEGL-3 VALUES
 Reference:  Nelson.G. 2005a. Effects of Carbon Monoxide in Man. In: Carbon Monoxide and Human
 Lethality: Fire and Non-fire studies, M.M. Hirschler (Ed.). Taylor and Francis, New York, 2005, pp 3-60;
 Chiodi, H., D.B. Dill, F. Consolazio and S.M. Horvath, 1941. Respiratory and circulatory responses to
 acute carbon monoxide poisoning. American Journal of Physiology 134, 683-693; Haldane, J., 1895.
 The action of carbonic acid on man. Journal of Physiology 18, 430-462; Henderson, Y., H.W. Haggard,
 M.C. Teague, A.L.  Prince and R.M. Wunderlich, 1921. Physiological effects of automobile exhaust gas
 and standards of ventilation for brief exposures. Journal of Industrial Hygiene 3, 79-92.	
 Test Species/Strain/Sex/Number: Nelson (2005a): Humans/not applicable/both sexes/~3010 subjects
 Chiodi et al. (1941), Haldane (1895), Henderson et al. (1921): Humans (healthy young males) / not
 applicable / males / 4 (total)	
 Exposure Route/Concentrations/Durations: Inhalation /Nelson (2005a) reported COHb levels iff	[ Formatted Table
 deceased subjects poisoned by inhalaling CO; Chiodi et al. (1941): repeated test on three subjects that
 reached COHb of 27-52% at the end of exposure; individual COHb values were 31, 32, 32, 33, 39, 41,
 42, 43, 45 and 52 % in subject H.C., 27, 35, 41, 43 and 48 % in subject F.C. and 41, 42 and 44 % in
 subject S.H.; Haldane (1895): repeated exposure of one subject reaching the following COHb at the
 end of exposure (time in min): 13 % (240 min), 14 % (210 min), 23 % (240 min), 27 % (24 min), 35 %
 (29 min), 37 % (29 min), 37 % (120 min), 39 % (30.5 min), 49 % (71 min), 56 % (35 min).
 Effects:  At a COHb of about 40-56 %, Haldane (1895) described symptoms included hyperpnea,
 confusion of mind, dim vision and unsteady/inability to walk. Chiodi et al. (1941) found no effect on
 oxygen consumption, ventilation, pulse rate, blood pressure and blood pH; the cardiac output increased
 20-50 % at COHb >40 %, while the changes were negligible at COHb of <30 %.  Nelson (2005a)
 reported COHb measurements in lethal poisoning human cases and the data indicated that most lethal
 poisoning cases occurred at COHb levels higher than 40% and that survival of CO-exposed humans
 were likely to be seen at levels below 40%.	
 Endpoint/Concentration/Rationale:
        The derivation of AEGL-3 values was based on observations in humans. Analysis of lethal
 cases reported by Nelson (2005a) indicated that most lethal poisoning cases occurred at COHb levels
 higher than 40% and that survival of CO-exposed humans were likely to be seen at levels below 40%.
 Thus, 40%COHb level seems a reasonable threshold for lethality.  This level is supported by
 experimental studies performed in healthy human subjects. Studies by Chiodi et al. (1941), Henderson
 et al. (1921), and Haldane (1895) suggest that a COHb of about 34-56 % does not cause lethal effects
 in healthy individuals. Further support come from the studies by Kizakevich et al. (1994), Stewart et al.
 (1970), and Nielsen (1971) that reported headache as the only symptom when subjects were exposed
 to 20-33 % COHb. The point of departure of 40% COHb is also supported by studies in animals
                                        88

-------
CARBON MONOXIDE                                                  FINAL: 07/2008
 reporting minimum lethal COHb levels in rats and mice of about 50-70 % (E.I. du Pont de Nemours and
 Co., 1981; Rose et al., 1970).  Further support comes from published cases of myocardial infarction
 that measured COHb levels after transport to the hospital:  52.2 % (Marius-Nunez, 1990), 30 %, 22.8 %
 (Atkins and Baker, 1985), 21 % (Ebisuno et al., 1986), 15.6 % (Grace and Platt, 1981).	
 Uncertainty Factors/Rationale:
 Total uncertainty factor: 3
 Interspecies:   Not applicable
 Intraspecies:   3 - an intraspecies uncertainty factor of 3 was supported by information on effects, such
 	as myocardial infarction and stillbirths, reported in more susceptible subpopulations.
 Modifying Factor: Not applicable
 Animal to Human Dosimetric Adjustment: Not applicable
 Time Scaling:
 A mathematical model (Coburn et al., 1965; Peterson and Stewart, 1975) was used to calculate
 exposure concentrations in air resulting in a COHb of 40 % at the end of exposure  periods of 10 and 30
 minutes and 1, 4 and 8 hours.	
 Data Adequacy:
 AEGL-3 values were based on 40% COHb levels derived from the analysis of clinical cases of lethal
 and non-lethal poisoning. The AEGL-3 values derived using an intraspecies uncertainty factor of 3
 (corresponding to an COHb of about 15 %) are supported by the available case reports of lethal effects
 (myocardial infarction, stillbirths) in more susceptible subpopulations. Lethal effects from myocardial
 infarction in hypersusceptible patients with coronary artery disease at even lower CO concentrations,
 which could be at the upper end of the range of CO concentrations found inside buildings and in
 ambient air outside, cannot be excluded.
                                          89

-------