INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY
ENVIRONMENTAL HEALTH CRITERIA 13
Carbon Monoxide
This report contains the collective views of an international group of
experts and does not necessarily represent the decisions or the stated
policy of either the World Health Organization or the United Nations
Environment Programme.
Published under the joint sponsorship of
the United Nations Environment Programme
and the World Health Organization
World Health Organization
Geneva, 1979
ISBN 92 4 154073 7
(c) World Health Organization 1979
Publications of the World Health Organization enjoy copyright
protection in accordance with the provisions of Protocol 2 of the
Universal Copyright Convention. For rights of reproduction or
translation of WHO publications, in part or in toto, application
should be made to the Office of Publications, World Health
Organization, Geneva, Switzerland. The World Health Organization
welcomes such applications.
The designations employed and the presentation of the material in
this publication do not imply the expression of any opinion whatsoever
on the part of the Secretariat of the World Health Organization
concerning the legal status of any country, territory, city or area or
of its authorities, or concerning the delimitation of its frontiers or
boundaries.
The mention of specific companies or of certain manufacturers'
products does not imply that they are endorsed or recommended by the
World Health Organization in preference to others of a similar nature
that are not mentioned. Errors and omissions excepted, the names of
proprietary products are distinguished by initial capital letters.
CONTENTS
ENVIRONMENTAL HEALTH CRITERIA FOR CARBON MONOXIDE
1. SUMMARY AND RECOMMENDATIONS FOR FURTHER STUDIES
1.1. Summary
1.1.1. Properties and analytical methods
1.1.2. Sources of environmental pollution
1.1.3. Environmental levels
1.1.4. Effects on experimental animals
1.1.5. Effects on man
1.1.6. Evaluation of health risk
1.2. Recommendations for further studies
2. CHEMISTRY AND ANALYTICAL METHODS
2.1. Physical and chemical properties
2.2. Methods of measuring carbon monoxide in ambient air
2.3. Biological monitoring
3. SOURCES OF CARBON MONOXIDE IN THE ENVIRONMENT
3.1. Natural occurrence
3.2. Man-made sources
4. ENVIRONMENTAL DISTRIBUTION AND TRANSFORMATION
4.1. Atmospheric transport and diffusion
4.2. Environmental absorption and transformation
5. ENVIRONMENTAL LEVELS AND EXPOSURES
5.1. Ambient air concentrations and exposures
5.2. Indoor concentrations and exposure
5.3. Occupational exposure
5.4. Carboxyhaemoglobin levels in the general population
6. METABOLISM
6.1. Endogenous carbon monoxide production
6.2. Absorption
6.3. Reactions with body tissues and fluids
6.4. Excretion
7. EFFECTS ON EXPERIMENTAL ANIMALS
7.1. Species differences
7.2. Cardiovascular system and blood
7.3. Central nervous system
7.4. Behavioural changes and work performance
7.5. Adaptation
7.6. Embryonal, fetal, neonatal, and teratogenic effects
7.7. Carcinogenicity and mutagenicity
7.8. Miscellaneous changes
7.9. Interactions
8. EFFECTS ON MAN
8.1. Healthy subjects
8.1.1. Behavioural changes
8.1.2. Work performance and exercise
8.1.3. Adaptation
8.1.4. Effects on the cardiovascular system and other
effects
8.1.5. Carboxyhaemoglobin levels resulting from exposure
to methane-derived halogenated hydrocarbons
8.1.6. Levels and effects of carboxyhaemoglobin resulting
from smoking
8.1.7. Interactions
8.2. High-risk groups
8.2.1. Individuals with cardiovascular and chronic
obstructive lung disease
8.2.2. Anaemic individuals
8.2.3. Embryo, fetus, neonate, and infants
8.2.4. Individuals living at high altitudes
8.3. Summary table
9. EVALUATION OF HEALTH RISKS
9.1. Introduction
9.2. Exposure
9.2.1. Assessment of exposure
9.2.2. Endogenous production
9.2.3. Outdoor environmental exposure
9.2.4. Indoor exposure
9.2.5. Exposures related to traffic
9.2.6. Occupational exposure
9.2.7. Tobacco smoking
9.2.8. Multiple exposures
9.3. Effects
9.3.1. Cardiovascular system
9.3.1.1 Development of atherosclerotic
cardiovascular disease
9.3.1.2 Acute effects on existing heart illness
9.3.1.3 Acute effects on existing vascular disease
9.3.2. Nervous system
9.3.3. Work capacity
9.4. Recommended exposure limits
9.4.1. General population exposure
9.4.2. Working population exposure
9.4.3. Derived limits for carbon monoxide concentrations
in air
REFERENCES
ANNEX 1
ANNEX 2
NOTE TO READERS OF THE CRITERIA DOCUMENTS
While every effort has been made to present information in the
criteria documents as accurately as possible without unduly delaying
their publication, mistakes might have occurred and are likely to
occur in the future. In the interest of all users of the environmental
health criteria documents, readers are kindly requested to communicate
any errors found to the Division of Environmental Health, World Health
Organization, Geneva, Switzerland, in order that they may be included
in corrigenda which will appear in subsequent volumes.
In addition, experts in any particular field dealt with in the
criteria documents are kindly requested to make available to the WHO
Secretariat any important published information that may have
inadvertently been omitted and which may change the evaluation of
health risks from exposure to the environmental agent under
examination, so that the information may be considered in the event of
updating and re-evaluation of the conclusions contained in the
criteria documents.
WHO TASK GROUP ON ENVIRONMENTAL HEALTH CRITERIA FOR CARBON MONOXIDE
Members
Dr H. Buchwald, Assistant Deputy Minister, Alberta Department of
Labour, Edmonton, Alberta, Canada (Chairman)
Dr V.A. Cizikov, Central Institute for Advanced Medical Training,
Moscow, USSR
Dr E. Haak, Physiology Branch, Clinical Studies Division, Health
Effects Research Laboratory, US Environmental Protection Agency,
Research Triangle Park, NC, USA
Dr P. Iordanidis, National Technical University of Athens, Athens,
Greece
Dr K. Ishikawa, Department of Public Health, School of Medicine, Chiba
University, Chiba, Japan
Dr V. Kodat, Hygiene Department, Ministry of Health of the Czech
Socialist Republic, Prague, Czechoslovakia (Vice-Chairman)
Dr K. Kurppa, Department of Occupational Medicine, Institute of
Occupational Health, Helsinki, Finland
Professor P.J. Lawther, Clinical Section, Medical Research Council
Toxicology Unit, St Bartholomew's Hospital Medical College,
London, England
Mr I.R.C. McDonald, Chemistry Division, Department of Scientific &
Industrial Research, Petone, New Zealand (Rapporteur)
Dr G. Winneke, Institute for Air Hygiene & Silicosis Research,
Düsseldorf, Federal Republic of Germany
Representatives of other organizations
Mr J. Janczak, Economic Commission for Europe, Geneva, Switzerland
Dr D. Djordjevic, International Labour Office, Geneva, Switzerland
Mr C. Satkunanthan, International Register of Potentially Toxic
Chemicals of the United Nations Environment Programme, Geneva,
Switzerland
Observers
Dr J.J. Vostal, Biomedical Science Department, General Motors Research
Laboratories, Warren, MI, USA
Secretariat
Mrs B. Goelzer, Scientist, Office of Occupational Health, Division of
Noncommunicable Disease, WHO, Geneva, Switzerland
Dr Y. Hasegawa, Medical Officer, Control of Environmental Pollution &
Hazards, Division of Environmental Health, WHO, Geneva,
Switzerland
Dr R. Horton, Senior Research Adviser, Health Effects Research
Laboratory, US Environmental Protection Agency, Research Triangle
Park, NC, USA (Temporary Adviser)
Dr J. Korneev, Scientist, Control of Environmental Pollution &
Hazards, Division of Environmental Health, WHO, Geneva,
Switzerland (Secretary)
Dr H. de Koning, Scientist, Control of Environmental Pollution &
Hazards, Division of Environmental Health, WHO, Geneva,
Switzerland
Dr M. Vandekar, Medical Officer, Pesticides Development & Safe Use,
Division of Vector Biology & Control, WHO, Geneva, Switzerland
Dr V.B. Vouk, Chief, Control of Environmental Pollution & Hazards,
Division of Environmental Health, WHO, Geneva, Switzerland
ENVIRONMENTAL HEALTH CRITERIA FOR CARBON MONOXIDE
A WHO Task Group on Environmental Health Criteria for Carbon
Monoxide met in Geneva from 11 to 17 October 1977. Dr V.B. Vouk,
Chief, Control of Environmental Pollution and Hazards, opened the
meeting on behalf of the Director-General. The Task Group reviewed and
revised the second draft of the criteria document and made an
evaluation of the health risks from exposure to carbon monoxide.
The first and second drafts were prepared by Dr S.M. Horvath of
the Institute of Environmental Studies, University of California,
Santa Barbara, USA. The comments on which the second draft was based
were received from the national focal points for the WHO Environmental
Health Criteria Programme in Bulgaria, Canada, Czechoslovakia, France,
Netherlands, Poland, USSR, and USA and from the International Labour
Organisation (ILO), Geneva, the Food and Agriculture Organization of
the United Nations (FAO), Rome, the United Nations Educational,
Scientific and Cultural Organization (UNESCO), Paris, the United
Nations Industrial Development Organization (UNIDO), Vienna, the
Permanent Commission and International Association on Occupational
Health, the Commission on Atmospheric Environment, International Union
of Pure and Applied Chemistry (IUPAC), and from the Pan American
Sanitary Engineering Center (CEPIS).
The collaboration of these national institutions, international
organizations and WHO collaborating centres is gratefully
acknowledged. Without their assistance this document would not have
been completed. The Secretariat wishes to thank, in particular,
Professor P.J. Lawther and Mr R.E. Waller of the Medical Research
Council Toxicology Unit, St Bartholomew's Hospital Medical College,
London, and Dr G. Winneke of the Institute for Air Hygiene and
Silicosis Research, Düsseldorf, for their help in the scientific
editing of the document.
This document is based primarily on original publications listed
in the reference section. However, several recent publications broadly
reviewing health aspects of carbon monoxide have also been used
including those of the Commission of the European Communities (1974),
NAS/NRC (1977), US Department of Health, Education and Welfare (1970,
1972), and Committee on the Challenges of Modern Society (1972).
Details of the WHO Environmental Health Criteria Programme,
including some of the terms frequently used in the documents, may be
found in the introduction to the publication "Environmental Health
Criteria 1 -- Mercury" published by the World Health Organization,
Geneva, 1976, and now available as a reprint.
The following conversion factorsa have been used in this document:
carbon monoxide 1 ppm = 1145 µg/m3 1 µg/m3 = 0.873 ppm
methylene chloride 1 ppm = 3480 µg/m3 1 µg/m3 = 0.288 ppm
nitrogen dioxide 1 ppm = 1880 µg/m3 1 µg/m3 = 0.532 ppm
ozone 1 ppm = 2000 µg/m3 1 µg/m3 = 0.500 ppm
peroxyacetyl nitrate 1 ppm = 5000 µg/m3 1 µg/m3 = 0.200 ppm
1 Torr = 1.333 × 102 pascals = 1 mmHg
a All conversion factors for atmospheric pollutants refer to 25°C
and 101 kPa (1 atm) pressure.
1. SUMMARY AND RECOMMENDATIONS FOR FURTHER STUDIES
1.1 Summary
1.1.1 Properties and analytical methods
Carbon monoxide (CO) is a colourless, odourless, tasteless gas
that is slightly less dense than air. It is a product of incomplete
combustion of carbon-containing fuels and is also produced by some
industrial and biological processes. Its health significance as a
contaminant of air is largely due to the fact that it forms a strong
coordination bond with the iron atom of the protohaem complex in
haemoglobin forming carboxyhaemoglobin (HbCO) and thus impairs the
oxygen-carrying capacity of the blood. The dissociation of
oxyhaemoglobin is also altered by the presence in blood of
carboxyhaemoglobin so that the supply of oxygen to tissues is further
impaired. The affinity of haemoglobin for carbon monoxide is roughly
240 times that of its affinity for oxygen; the proportions of
carboxyhaemoglobin and oxyhaemoglobin in blood are largely dependent
on the partial pressures of carbon monoxide and oxygen. Carbon
monoxide is absorbed through the lungs and the concentrationa of
carboxyhaemoglobin in the blood at any time will depend on several
factors. When in equilibrium with ambient air, the carboxyhaemoglobin
content of the blood will depend mainly on the concentrations of
inspired carbon monoxide and oxygen. However, if equilibrium has not
been achieved, the carboxyhaemoglobin concentration will also depend
on the time of exposure, pulmonary ventilation, and the
carboxyhaemoglobin originally present before inhalation of the
contaminated air. Formulae exist by which these estimates can be made.
In addition to its reaction with haemoglobin, carbon monoxide combines
with myoglobin, cytochromes, and some enzymes; the health significance
of these reactions is not clearly understood but is likely to be of
less importance than that of the reaction of the gas with haemoglobin.
Methods available for the measurement of carbon monoxide in
ambient airb range from fully automated methods using the non-
dispersive infrared technique and gas chromatography to very simple
semiquantitative manual methods using detector tubes. Since the
formation of carboxyhaemoglobin in man is dependent on many factors
a Throughout the document, the word concentration refers to mass
concentration, unless otherwise stated.
b Selected Methods of Measuring Air Pollutants, WHO Offset
Publication No. 24 (1976) published under the joint sponsorship of
the United Nations Environment Programme and the World Health
Organization, Geneva.
including the variability of ambient air concentrations of carbon
monoxide, carboxyhaemoglobin concentrations should be measured rather
than calculated. Several relatively simple methods are available for
determining carbon monoxide either by analysis of the blood or of
alveolar air that is in equilibrium with the blood. Some of these
methods have been validated by careful comparative studies.
1.1.2 Sources of environmental pollution
At present, the significance of natural sources of carbon monoxide
for man is uncertain. Estimates of man-made carbon monoxide emissions
vary from 350 to 600 million tonnes per annum. By far the most
important source of carbon monoxide at breathing level is the exhaust
of petrol-powered motor vehicles. The emission rate depends on the
type of vehicle, its speed, and its mode of operation. Other sources
include heat and power generators, some industrial processes such as
the carbonization of fuel, and the incineration of refuse. Faulty
domestic cooking and heating appliances may be important sources that
are often overlooked.
1.1.3 Environmental levels
Natural background levels of carbon monoxide are low
(0.01-0.9 mg/m3 or 0.01-0.8 ppm). Carbon monoxide concentrations in
urban areas are closely related to motor traffic density and to
weather and vary greatly with time and distance from the sources. The
configuration of buildings is important and concentrations fall
sharply with increasing distance from the street.
There are usually well-marked diurnal patterns with peaks
corresponding to the morning and evening "rush hours". Data from Japan
and the USA show that 8-h mean concentrations of carbon monoxide are
generally less than 20 mg/m3 (17 ppm). However, maximum 8-h mean
concentrations of up to 60 mg/m3 (53 ppm) have occasionally been
recorded. Much higher relatively transient peaks may be observed in
still weather where there is traffic congestion, and high
concentrations can be found in confined spaces such as tunnels,
garages, and loading bays in which vehicles operate and in vehicles
with faulty exhaust systems. There may be relatively high pollution by
carbon monoxide in workplaces and in some homes with cooking and
heating appliances that are faulty or do not have flues.
By far the commonest cause of high carboxyhaemoglobin
concentrations in man is the smoking of tobacco and the inhalation of
the products by the smoker.
1.1.4 Effects on experimental animals
Many experiments on animals have yielded valuable information
about the effects of carbon monoxide. There is general agreement that
most animals die when carboxyhaemoglobin levels exceed about 70% and
that the rate of administration of the gas is important in determining
the outcome. It is also agreed that carboxyhaemoglobin levels
exceeding 50% are often associated with damage to organs including the
brain and the heart. When animals are exposed to lower concentrations,
the effects are more difficult to discern and may be manifested as
changes in metabolism and biochemistry, alterations in the blood, or
changes in behaviour. There is evidence that some animals adapt to
exposure to comparatively low concentrations of carbon monoxide. As
might be expected, the variability of reported results of experiments
increases as the effects become less marked and the need for
scrupulous experimental design and technique becomes more important.
Of particular importance is the interpretation of the claims of some
workers that continuous intermittent exposure of animals to
concentrations resulting in carboxyhaemoglobin levels of 10-20% can
produce demonstrable histological changes in the myocardium, blood
vessels, and central nervous system. There are claims that such
exposures affect cholesterol uptake in the aorta and the coronary
arteries. The relevance of these findings, if accepted as real, are
obvious for the aetiology of cardiovascular diseases in man and,
therefore, must be assessed with great caution. It is also important
to study carefully the reports of research workers who have failed to
find evidence of damage.
1.1.5 Effects on man
The effects on man of exposure to high concentrations of carbon
monoxide are well documented and the diagnosis, treatment, and
sequelae of acute carbon monoxide poisoning are adequately dealt with
in standard texts. Recently much attention has been paid to the
possible effects on function and structure of exposure to carbon
monoxide concentrations resulting in carboxyhaemoglobin levels of 10%
or less. Carbon monoxide acts primarily by interfering with oxygen
transport and as the central nervous system is more sensitive to
hypoxia than the other systems of the body, much work has been done on
the impairment of vigilance, perception, and the performance of fine
tasks following exposure to concentrations of carbon monoxide too low
to produce clinical signs or symptoms. Many common drugs, beverages,
food, and fatigue can alter alertness, efficiency, and dexterity and
reported observed effects of low concentrations of carbon monoxide are
difficult, if not impossible, to interpret when no account is given of
precautions taken in the experimental design to eliminate or assess
the separate effects of other stresses. Again, great attention must be
paid to reports of impeccable experiments that have failed to
reproduce effects already reported. There would seem to be some
justification for accepting the possibility that concentrations of
carboxyhaemoglobin exceeding 2.5% might be associated with some
impairment of vigilance and other modes of perception. However, it
cannot be emphasized too strongly that when assessing the significance
(health and social) of this, the effects that many other commonly
acceptable factors might have on the tests should be taken into
account. It is possible, even likely, that the damaged heart and
respiratory system are more prone to impairment by carbon monoxide
than the intact brain. Skeletal muscle is sensitive to hypoxia and
obviously its sensitivity is enhanced by arterial disease. However, of
much greater importance is the effect of carbon monoxide on the
ischaemic myocardium which is especially vulnerable to additional
hypoxia. Evidence has been reported of changes in cardiac function and
the time of onset of angina pectoris on exercise when
carboxyhaemoglobin levels exceed 2.5%. Changes in oxygen uptake and
transfer are theoretically possible at or below these levels, and thus
there will be some patients whose cardiac function is so impaired that
any further hypoxic stress from carbon monoxide or from other factors,
will be intolerable. Similarly, the gross hypoxia of all tissues seen
in cases of severe respiratory disease renders the body even more
susceptible to the effects of low concentrations of carbon monoxide.
It follows that there is reason to regard carbon monoxide in this
sense as a pollutant for which the "threshold" is that concentration
which would be in equilibrium with the carboxyhaemoglobin produced
endogenously by the breakdown of blood pigments. However, it must be
realized that at these extremes of illness other usually trivial
stresses, such as ambient and body temperatures, infection, noise, and
anxiety, may be of much greater importance.
In addition to patients with diseases of the heart and the lungs,
it is likely that other groups, such as the anaemic, elderly,
postoperative patients, or those with cerebrovascular arteriosclerosis
may be at special risk. The effects on the fetus in utero of carbon
monoxide, especially that derived from maternal smoking, are of
special interest. The effects of carbon monoxide on people living at
high altitudes are greater than on those living at sea level and this
added risk must be assessed. There is a distinct possibility that
healthy man may adapt to the mild hypoxia caused by carboxyhaemo-
globin levels of about 3-5% (or to even higher values) as he does to
high altitude. Many workers in industry, and even more smokers,
repeatedly have carboxyhaemoglobin values such as these and there have
been few attempts to correlate symptoms or pathological findings
specifically with these levels of carboxyhaemoglobin. There is little
evidence that exposure to comparatively low concentrations of carbon
monoxide causes disease though it is suspected of being an etiological
factor in the association of heart disease with smoking.
1.1.6 Evaluation of health risk
There can be no doubt that the assessment of risk of exposure to
the lower concentrations of carbon monoxide in inspired air in
populations containing the sick and the fit, the smoker and nonsmoker,
the very young and the very old, would be a complex, if not
impossible, task, even if the ambient concentration of carbon monoxide
remained constant in time and place. Rough guidance, therefore, is all
that is possible in the light of available evidence derived from sound
scientific work. Those responsible for the welfare of specially
susceptible groups must refer critically to the evidence from
published works and to that reviewed in this document and make their
special decisions. There is general agreement that any individual
should be protected from exposure to carbon monoxide that would result
in carboxyhaemoglobin levels of 5% for any but transient periods, and
that especially susceptible persons ought not to be subjected to
concentrations giving carboxyhaemoglobin levels exceeding 2.5%. Advice
concerning such subjects must depend on individual assessment of their
clinical status and of other environmental factors including the
demands of the tasks they have to perform (persons engaged in driving,
monotonous tasks, keeping watch, etc., though healthy, might require
special consideration). The real possibility that adaptation occurs
makes consideration of the smoker and industrially exposed worker
difficult; ethical as well as health factors might affect action but
it would seem reasonable to have the same limit of 5% carboxy-
haemoglobin for industrial workers as for the rest of the healthy
population. The smoker inflicts high carboxyhaemoglobin values on
himself, by choice; he ought to be told of the evidence that this
habit might be harmful and then be subject to the levels of protection
recommended above.
1.2 Recommendations for Further Studies
(a) Though some would maintain that there are enough data on
levels of carbon monoxide in urban air, there is a need for further
surveys of air and blood levels so that some more precise correlations
may be established. Various populations in various places should be
studied properly to assess the magnitude of the problem posed by
carbon monoxide in the air of towns, houses, and workplaces.
(b) Opinions concerning levels of carbon monoxide below which no
adverse effects are seen and above which impairment of mental or
bodily functions seems to occur are based on comparatively few data.
These have been obtained from experiments concerning vigilance tests
and other tests of perception and performance, or the effects of
carbon monoxide on exercising cardiac and skeletal muscle and on
symptoms in patients with cardiovascular disease. There is an obvious
need to provide further data concerning larger numbers of subjects in
soundly designed experiments, the results of which should be properly
analysed. By such means it is hoped that dose-response and
concentration-response relationships may be established. The continued
refinement and application of epidemiological techniques must
complement experimental work.
(c) There are few data for assessing the possible consequences of
exposure of man to long-term, low concentrations of carbon monoxide.
There is a need to evaluate the possible effects of such exposures and
to determine the role of adaptation.
(d) Evidence that carbon monoxide plays a role in the observed
deleterious effects of smoking has been produced by experiments on
animals but more work is needed to confirm and extend these findings.
The hazards to the fetus of maternal smoking need to be evaluated and
the susceptibility of the fetus to carbon monoxide from whatever
source needs to be studied.
(e) The effects of comparatively low levels of carboxyhaemoglobin
on such skills as driving, and on perception in other tasks, need
further careful investigation. Not only must the possibility of the
enhancement of effects of carbon monoxide by other commonly occurring
factors be assessed, but there is a need to compare the effects of
carbon monoxide on vigilance and performance with the effects of such
common agents as therapeutic drugs, alcohol, fatigue, and food.
Attention is drawn to the difficulties inherent in the design of such
tests as well as to the problems involved in the assessment of the
health and social significance of the results.
(f) The possible effects of carbon monoxide on people living and
working at high altitudes (including aircraft pilots) have received
too little attention. There is a need for further work on this
subject.
2. CHEMISTRY AND ANALYTICAL METHODS
2.1 Physical and Chemical Properties
Carbon monoxide (CO) is a colourless, odourless, and tasteless gas
which is commonly formed during the incomplete combustion of
carbonaceous material. It is slightly lighter than air and only
slightly soluble in water. Carbon monoxide absorbs electromagnetic
radiation in the infrared region with the main absorption band centred
at 4.67 µm; this property is used for the measurement of carbon
monoxide concentrations in air. Some other physical properties of
carbon monoxide are listed in Table 1.
Table 1. Physical properties of carbon monoxide
Relative molecular mass 28.01
Critical point -140.2°C at 34.5 atm (3.5 MPa)
Melting point -205.1°C
Boiling point -191.5°C
Density, at 0°C, 1 atm 1.250 g/litre
at 25°C, 1 arm 1.145 g/litre
Specific gravity relative to air 0.967
Solubility in water at 0°C, 1 atm 3.54 ml/100 ml
at 25°C, 1 atm 2.14 ml/100 ml
at 37°C, 1 atm 1.83 ml/100 mla
Conversion factors:
at 0°C, 1 atm 1 mg/m3 = 0.800 ppmb
1 ppm = 1.250 mg/m3
at 25°C, 1 atm 1 mg/m3 = 0.873 ppm
1 ppm = 1.145 mg/m3
a Value obtained by graphic or calculated interpolation (Altman et al., 1971).
b Parts per million by volume.
While carbon monoxide is chemically inert under normal conditions
of temperature and pressure (25°C; 1 atm (101 kPa)), it becomes
reactive at higher temperatures and can act as a strong reducing
agent. At 90°C, it reacts with iodine pentoxide to produce iodine
vapour. At 150°C, it also releases mercury vapour from mercury(II)
oxide. Both reactions are used in the analytical chemistry of carbon
monoxide. The oxidation of carbon monoxide to carbon dioxide (CO2)
is accelerated by metallic catalysts such as palladium on silica gel,
or by a mixture of manganese and copper oxides (Hopcalite).
In forming carboxyhaemoglobin (HbCO), carbon monoxide reacts with
the iron in protohaem -- a constituent of haemoglobin -- and forms
strong coordination bonds. Thus carboxyhaemoglobin is toxic because it
is about 200 times more stable than oxyhaemoglobin (HbO2). Carbon
monoxide also combines reversibly with myoglobin and cytochromes,
including P-450.
The environmental chemistry of carbon monoxide is discussed in
section 4.2.
2.2 Methods of Measuring Carbon Monoxide in Ambient Air
Three methods are most commonly used for the routine estimation of
carbon monoxide in air. These are the continuous analysis method based
upon nondispersive infrared absorption spectroscopy (NDIR); the semi-
continuous analysis method using gas chromatographic techniques and a
semiquantitative method employing detector-tubes. Other methods
include catalytic oxidation, electrochemical analysis, mercury
displacement, and the dual isotope technique (WHO, 1976).
In the NDIR method, infrared radiation is divided into two beams
that are directed through a reference and a sample cell, respectively.
Any carbon monoxide introduced into the sample cell will absorb
radiation at the characteristic band centred at 4.67 µg, causing the
detector to produce an output signal proportional to the concentration
of carbon monoxide in the sample cell. NDIR analysers are produced by
several manufacturers in the form of continuous, automated
instruments. Good commercial instruments have a detection limit of
about 1 mg/m3 (0.87 ppm). Carbon dioxide and water vapour interfere
but there are several techniques to minimize this interference.
In chromatographic methods, carbon monoxide is first separated
from water vapour, carbon dioxide, and hydrocarbons. It is then
catalytically reduced to methane and passed through a flame ionization
detector, the output signal of which is proportional to the carbon
monoxide concentration in the air sample. The most common
concentration range in commercial instruments is from about 1 to
350 mg/m3 (1-300 ppm) but others are available with a range of about
0.02 to 1.00 mg/m3 (0.017-0.87 ppm). Gas chromatography is
particularly suitable, when low concentrations of carbon monoxide have
to be measured with a high degree of specificity.
The detector tube method is very simple and can be used for
estimating concentrations above 5 mg/m3. Air is drawn through
specially manufactured tubes containing a chemical agent that changes
colour if carbon monoxide is present and can be used to estimate
concentrations. The advantages and limitations of detector tubes are
further discussed in a WHO manual (WHO, 1976).
A well known method is based on the measurement of the temperature
rise caused by the catalytic oxidation of carbon monoxide. The limit
of detection is about 1 mg/m3. Most hydrocarbons will interfere
unless removed (NAS/NRC, 1977). For measurements in ambient air, the
sensitivity may not always be sufficient.
Electrochemical analysers (Hersch, 1964, 1966) are based on the
liberation by carbon monoxide of iodine from iodine pentoxide (at
150°C), which is then reduced at the cathode of a galvanic cell. The
current developed is a measure of the carbon monoxide concentration
present in the air sample.
A further highly sensitive method is based on the reduction of
mercury(II) oxide by carbon monoxide at a temperature between 170 and
210°C. Mercury vapour generated during this reaction is determined by
absorption spectrophotometry at 253.7 nm. This method, as modified by
Seller & Junge (1970), has a reported detection limit of about
3 µg/m3.
The slight difference in the fluorescence spectra of 16CO and
18CO is used for carbon monoxide determination by the so-called dual
isotope fluorescence method. Instruments using this principle are
available with ranges of 0-20 mg/m3 (0-17.5 ppm) and 0-200 mg/m3
(0-175 ppm) with a reported detection limit of about 0.2 mg/m3
(0.17 ppm). Other pollutants present cause very little interference
(McClatchie et al., 1972).
An important part of any carbon-monoxide measurement procedure is
the calibration technique. Many publications deal with this topic and
the Deutsche Industrienormen Ausschuss (DIN) and the International
Standard Organization (ISO) have special groups for establishing
suitable calibration standards.
2.3 Biological Monitoring
Blood carboxyhaemoglobin can be satisfactorily determined in a
venous blood sample, which should be collected in a closed container
containing an anticoagulant (dry sodium heparin or di-sodium ethylene-
diaminotetracetic acid, EDTA). Blood samples may be preserved for
several days prior to analysis if kept cold (4°C) and in the dark.
Complete mixing of blood must be attained if carbon monoxide and
haemoglobin are to be measured separately. Total haemoglobin is
conveniently determined by conversion to cyanmethaemoglobin (Van
Kampen & Zijlstra, 1961), which is then determined spectrophoto-
metrically (Drabkin & Austin, 1935).
Various methods are available for the determination of carboxy-
haemoglobin by spectrophotometry or by the liberation of carbon
monoxide (WHO, 1976). One method consists of measuring the absorbance
at 4 wavelengths in the Soret region (390-440 nm) of a blood sample
diluted to about 1:70 with an aqueous solution of ammonia (Small et
al., 1971). Carboxyhaemoglobin and methaemoglobin are estimated from
absorbance values, and oxyhaemoglobin is obtained from the difference.
The method is precise at low carboxyhaemoglobin concentrations (up to
25% saturation). A very convenient method is the automated
differential spectrophotometer (Malenfant et al., 1968), which is
available commercially as CO-oximeter. Simultaneous absorbance
measurements are made to determine the three component system (reduced
haemoglobin, oxyhaemoglobin and carboxyhaemoglobin) contained in a
haemolysed blood sample. The signals are processed and displayed in a
digital form as haemoglobin (g/100 ml) and the percentage of
oxyhaemoglobin and carboxyhaemoglobin. A method has recently been
described (Rossi-Bernardi et al., 1977) for the simultaneous
determination of four haemoglobin derivatives (deoxyhaemoglobin,
oxyhaemoglobin, methaemoglobin, carboxyhaemoglobin) and total oxygen
contents of 10 µl of whole blood. The spectrophotometric method of
Commins & Lawther (1965), which has been validated by Lily et al.
(1972), has the advantage of requiring only 0.01 ml blood obtained
from a finger prick sample.
Gas chromatography on molecular sieves with a suitable detection
system is probably the most satisfactory procedure for the measurement
of carbon monoxide liberated from carboxyhaemoglobin. Carbon monoxide
liberation is achieved through acidification. The method described by
Sotnikov (1971) requires only 0.1 ml of blood and has a limit of
detection of 0.01 ml of carbon monoxide per 100 ml of blood. Dahms &
Horvath (1974) have proposed a very accurate method for estimating low
concentrations of carboxyhaemoglobin. Table 2 compares some techniques
for the analysis of carboxyhaemoglobin in blood.
Another approach to the estimation of exposure to carbon monoxide
is by the analysis of expired air. The subject takes a deep breath and
holds it for 20 sec. The first 350-500 ml of expired air (dead air
space) is discarded and the remaining gas (alveolar air) is collected
in an aluminized mylar bag for analysis, using an NDIR instrument. The
value of the alveolar technique, pioneered by Sjöstrand (1948), is
based on the assumption that, during breath-holding, the lung is a
closed vessel in which blood carboxyhaemoglobin equilibrates with lung
gas and that Haldane's relationship (see section 6) applies.
Theoretically, the slope of the straight line relating %
carboxyhaemoglobin to alveolar pCO in ppm should be approximately
0.155 at sea level for % carboxyhaemoglobin values equivalent to a
carbon monoxide concentration between 0 and 50 ppm, and progressively
lower for higher concentrations (Coburn et al., 1965). Values
approximating to this theoretical ratio have been found experimentally
(Forbes et al., 1945; Malenfant et al., 1968). Although the alveolar
air method is less precise than the direct measurement of
carboxyhaemoglobin in blood, it can be used in epidemiological studies
and for general monitoring (McFarland, 1973) but cannot be used on
persons with chronic pulmonary disease.
Table 2. Comparison of techniques for the analysis of carboxyhaemoglobin in blooda
Detection method Sample Resolutionb Sample CVc Reference
volume (ml/dl) analysis (%)
(ml) time
(min)
Gasometric
Van Slyke 1.0 0.3 15 6 Horvath & Roughton (1942)
syringe-capillary 0.5 0.02 30 2-4 Roughton & Root (1945)
Optical
spectrophotometric 2.0 0.006 30 1.8 Coburn et al. (1964)
spectrophotometric 0.1 0.08 10 Small et al. ( 1971 )
spectrophotometric 0.4 0.10 3 Maas et al. (1970)
spectrophotometric 0.01 0.10 20 Commins & Lawther (1965)
Chromatographic
thermal conductivity 1.0 0.005 20d 1.8 McCredie & Jose (1967)
flame ionization 0.1 0.002 20 1.8 Collison et al. (1968)
thermal conductivity 1.0 0.001 30 2.0 Ayres et al. (1966)
thermal conductivity 0.25 0.006 3 1.7 Dahms & Horvath (1974)
a Adapted from: Dahms & Horvath (1974).
b Smallest detectable difference between duplicate determinations.
c Coefficient of variation based on samples containing less than 2.0 ml of carbon monoxide
per decilitre.
d Best estimate.
3. SOURCES OF CARBON MONOXIDE IN THE ENVIRONMENT
3.1 Natural Occurrence
The amount of carbon monoxide produced globally by natural sources
is at present uncertain. Several investigators have estimated that
natural sources (primarily oxidation of methane in the atmosphere and
emissions from the oceans) produce about ten times as much carbon
monoxide as man-made sources (Spedding, 1974). On the other hand, a
recent study concluded that the natural production of carbon monoxide
was much smaller and might be somewhat less than the emissions from
man-made sources (Seiler, 1975). If this is the case, man-made
emissions of carbon monoxide may play an important role in the global
carbon monoxide cycle.
Several estimates of the production of carbon monoxide by
atmospheric reactions have been made. Stevens et al. (1972) estimated
that in the northern hemisphere alone, more than 3 × 109 metric
tonnes of carbon monoxide are produced annually by the oxidation of
methane and other organic constituents. McConnell et al. (1971)
considered that biologically produced methane could be the source of
approximately 2.5 × 109 metric tonnes of carbon monoxide annually.
Subsequently, it was calculated, by Weinstock & Nicki (1972), that the
oxidation of methane alone could produce twenty-five times the
quantity of carbon monoxide generated by man's activities. Estimates
by Levy (1972) indicated that the oxidation of methane was a much
larger source of carbon monoxide that man-made sources.
The surface layers of the ocean are a second major natural source
of carbon monoxide. Linnenbom et al. (1973) calculated an upper limit
of 220 × 106 metric tonnes of carbon monoxide emitted from the oceans.
Using a model of the flux of gases across the air-sea interface, Liss
& Slater (1974) estimated a total ocean flux of carbon monoxide of
43 × 106 metric tonnes per year.
Among other natural sources of carbon monoxide are forest and
grass fires, volcanoes, marsh gases, and electric storms. Some carbon
monoxide is also formed in the upper atmosphere (above 75 km) by the
photo-dissociation of carbon dioxide (Altshuller & Bufallini, 1965;
Bates & Witherspoon, 1952). Another natural source of carbon monoxide
is rainwater, where production of carbon monoxide in the clouds is
tentatively attributed to the photochemical oxidation of organic
matter or the slight dissociation of carbon dioxide induced by
electrical discharges or both (Swinnerton et al., 1971). Some carbon
monoxide is also formed during germination of seeds and seedling
growth (Siegel et al., 1962; Wilks, 1959), by the action of
microorganisms on plant flavonoids (Westlake et al., 1961), and in
higher plants (Delwiche, 1970). Kelp may be a significant source of
carbon monoxide. Chapman & Tocher (1966) reported that some float
cells of kelp contained carbon monoxide concentrations as high as
916 mg/m3 (800 ppm).
Carbon monoxide is produced in measurable quantities in man and
animals as a by-product of haem catabolism.
3.2 Man-Made Sources
According to Jaffe (1973), global emissions of man-made carbon
monoxide in 1970 amounted to 360 million tonnes. Seller (1975)
calculated the carbon monoxide emissions for 1973 as 600 million
tonnes. A breakdown of Jaffe's estimate according to the type of
source is shown in Table 3. The motor vehicle was by far the largest
contributor accounting for 55% of total emissions. Other
transportation sources, certain industrial processes, waste disposal
and miscellaneous burning activities were responsible for the
remaining carbon monoxide emissions.
Table 3. Estimated global man-made emissions of carbon monoxide,
1970a
Source Emissions (106 metric tonnes)
Mobile
Motor vehicles: gasoline 197
diesel 2
Aircraft 5
Watercraft 18
Railroads 2
Other (nonhighway) motor vehicles 26
Stationary
Coal combustion 4
Oil combustion 1
industrial processes 41
Refuse disposal 23
Miscellaneous 41
Total 360
a From: Jaffe (1973).
The tremendous increase in the number and use of motor vehicles
during the past 30 years has been accompanied by a rapid increase in
carbon monoxide emissions. In the USA for example, the emission of
carbon monoxide rose from approximately 73 million tonnes in 1940 to
about 100 million tonnes in 1970 (US Environmental Protection Agency,
1973a). In 1968, the upward trend was reversed because of the initial
impact of motor vehicle emission controls. The rate at which carbon
monoxide is emitted from motor vehicles varies not only with vehicle
but also with the mode of operation of the vehicle. The emissions of
carbon monoxide by other mobile sources are comparatively small;
however, the emissions from locomotives, boats, and aircraft may
create local problems.
Among the stationary sources, the burning of waste material and
certain industrial processes generate substantial amounts of carbon
monoxide. Petroleum refineries, iron foundries, kraft-pulp mills,
carbon-black plants, and sintering processes are the major sources.
Emission rates for some of these processes are given in Table 4. The
burning of refuse, either in incinerators or openly, is an important
source of carbon monoxide. If uncontrolled, the emission rate of
carbon monoxide from incinerators is about 17.5 kg per tonne of refuse
burned. If burned openly, the emission rates can vary from about 25 to
60 kg per tonne, depending upon the type of refuse (US Environmental
Protection Agency, 1973b). The combustion of fossil fuels in electric
generating plants, industries, and the home, while resulting in the
emission of smaller quantities of carbon monoxide individually, may
constitute a major source when combined.
Any industrial process or operation, where incomplete combustion
of carbonaceous material occurs, may easily be of importance as far as
occupational exposure to carbon monoxide is concerned. Smelting of
iron ore, gas production works, gasworks and coke ovens, distribution
and use of both natural gas and coal gas, automobile manufacturing,
garages, and service stations are among the most important sources for
occupational exposure to carbon monoxide (Ministry of Labour, 1965).
It should also be emphasized that tobacco smoke is a most
significant source of man-made carbon monoxide in a closed environment
and that the carboxyhaemoglobin levels found in smokers are
consistently higher than those in nonsmokers.
Table 4. Emission rates for carbon monoxide in selected industrial
processesa
Source Emissions (uncontrolled)
Petroleum refineries
Fluid catalytic cracking units 39.2 kg/103 litre
Moving bed catalytic cracking units 10.8 kg/103 litre
Steel mills
Blast furnaces--ore charge 875 kg/tonne
Sintering 22 kg/tonne
Basic oxygen furnace 69.5 kg/tonne
Gray iron foundries
Cupola 72.5 kg/tonne
Carbon black
Channel process 16 750 kg/tonne
Thermal process negligible
Furnace process 2650 kg/tonne
a From: US Environmental Protection Agency (1973b).
4. ENVIRONMENTAL DISTRIBUTION AND TRANSFORMATION
4.1 Atmospheric Transport and Diffusion
The ambient air concentrations of carbon monoxide at locations
removed from man-made sources are low and variable. Junge et al.
(1971) reported that background levels of carbon monoxide in the lower
atmosphere may range from 0.01 to 0.23 mg/m3 (0.009-0.2 ppm).
Concentrations have been reported of 0.025 to 0.9 mg/m3
(0.022-0.8 ppm) in North Pacific marine air; 0.04 to 0.9 mg/m3
(0.036-0.8 ppm) in rural areas of California; 0.07 to 0.30 mg/m3
(0.06-0.26 ppm) at Point Barrow, Alaska; 0.06 to 0.8 mg/m3
(0.05-0.7 ppm) in Greenland; and about 0.07 mg/m3 (0.06 ppm) in the
South Pacific (Cavanagh et al., 1969; Goldman et al., 1973; Robbins et
al., 1968; Robinson & Robbins, 1970). Seiler & Junge (1969) observed
similar average concentrations over the North and South Atlantic Ocean
(0.20 mg/m3 and 0.06 mg/m3 (0.18 and 0.05 ppm) respectively).
Background levels of carbon monoxide are influenced by the origin of
the air masses and vertical distributions of carbon monoxide have been
reported by Seiler & Junge (1969, 1970). They found consistent carbon
monoxide concentrations of 0.15 mg/m3 (0.13 ppm) at an altitude of
10 km in both northern and southern hemispheres. In the troposphere,
average concentrations of 0.11 mg/m3 (0.10 ppm) were observed,
while, in the stratosphere, the concentrations ranged from 0.03 to
0.06 mg/m3 (0.027-0.05 ppm). A recent study (Goldman et al., 1973)
showed that a gradual decrease in carbon monoxide concentrations
occurred with increasing altitude ranging from approximately
0.09 mg/m3 (0.08 ppm) at 4 km to 0.05 mg/m3 (0.04 ppm) at 15 km.
4.2 Environmental Absorption and Transformation
The residence time of carbon monoxide in the atmosphere is
believed to be approximately 0.2 years. Background levels do not
appear to be increasing, indicating the presence of various scavenging
and removal mechanisms (sinks). Oxidation in the atmosphere and take-
up by the soil, vegetation, and inland fresh waters have been
identified as the major removal mechanisms.
The oceans act as reservoirs for carbon monoxide, since
considerable quantities are dissolved in the water. Because of the
equilibrium that exists, carbon monoxide is dissolved or released
according to conditions depending on the partial pressure of carbon
monoxide in the atmosphere and on the water temperature.
The carbon monoxide produced at the earth's surface migrates by
diffusion and eddy currents to the troposphere and stratosphere where
it is oxidized to carbon dioxide (CO2 by the hydroxyl (OH) radical.
This process, however, can account for only a portion of the carbon
monoxide oxidation. Calvert (1973) suggested several possible
reactions of carbon monoxide with other pollutants also involving the
hydroxyl radical and Westberg et al. (1971) demonstrated that carbon
monoxide can accelerate both the oxidation of nitric acid (NO) to
nitrogen dioxide (NO2) and the rate of ozone formation.
Microorganisms can metabolize carbon monoxide and large quantities
of these organisms are present in the soil. Ingersoll et al. (1974)
showed that desert soils took up carbon monoxide at the lowest rates
and tropical soils at the highest. Cultivated soils had a lower carbon
monoxide uptake rate than uncultivated soils, presumably because there
is less organic matter in the surface layer.
It has been reported that some plant species can remove carbon
monoxide from the atmosphere by oxidation to carbon dioxide or by
conversion to methane (Bidwell & Fraser, 1972). However, Ingersoll et
al. (1974) could not measure carbon monoxide removal by any plants
tested in an artificial atmosphere containing a carbon monoxide
concentration of 115 mg/m3 (100 ppm). The process of plant
respiration as a carbon monoxide sink still requires considerable
further study.
It is believed that inland fresh waters may remove some carbon
monoxide from the atmosphere. Rainwater contains appreciable
quantities of carbon monoxide and the runoff into the lakes and rivers
may account for additional removal.
5. ENVIRONMENTAL LEVELS AND EXPOSURES
5.1 Ambient Air Concentrations and Exposures
Carbon monoxide concentrations in the ambient air have been
measured in many large urban areas for a number of years. Although a
substantial body of data is now available, it is still not possible to
assess the overall human exposure to carbon monoxide adequately. Both
the extreme temporal and spatial variability of carbon monoxide
concentrations and the small number of monitoring stations in each
urban area make the assessment of human exposure difficult. The
available data, however, provide some indication of the patterns and
trends of urban carbon monoxide concentrations.
Concentrations in urban areas usually follow a very pronounced
diurnal pattern, and, although influenced by factors such as location
and meteorological conditions, their values are closely correlated
with the amount of motor vehicle traffic. Thus, although the exact
shape of the curve representing the temporal variation in carbon
monoxide concentrations over a day, varies with the situation, it
usually shows two peaks, corresponding to the morning and evening
traffic rush hours. Such curves have been established by many authors
(Colucci & Begeman, 1969; Göthe et al., 1969; Waller et al., 1965), by
means of continuous carbon monoxide monitoring. In general, an initial
peak is detected between 07h00 and 09h00 coinciding with heavy morning
traffic, and another, in the late afternoon, as illustrated in Fig. 1.
However, there may be exceptions to this typical pattern, as shown in
Fig. 2, which presents data from a monitoring station in New York City
(Martin & Stern, 1974). In this case, the traffic conditions were such
that carbon monoxide concentrations remained high during most of the
day. Because of changes in traffic patterns, the carbon monoxide
concentrations are usually lower at weekends than on weekdays and
follow a different diurnal pattern.
At any location, the concentration of carbon monoxide due to motor
vehicle traffic depends on the following specific variables:
(a) number of vehicles operating;
(b) engine characteristics of the operating vehicles (capacity,
gasoline or diesel, use of emission control devices);
(c) speed of traffic and gradient;
(d) temperature (as it affects the operating efficiency of the
engine).
More general variables include the meteorological conditions
(including wind speed and direction, and temperature gradients) and
the geometry of locality (shape and height of buildings, width of
street, etc.).
Carbon monoxide concentrations are normally reported in terms of
8-h average concentrations. This averaging time has been used because
it takes from 4 to 12 h for the carboxyhaemoglobin levels in the human
body to reach equilibrium with the ambient carbon monoxide
concentrations. Two types of approaches have been used for calculating
the 8-h average (McMullen, 1975). One approach is to examine all
possible 8-h intervals and calculate a moving 8-h average (24, 8-h
averages each day). The other approach is to examine three
consecutive, nonoverlapping, 8-h intervals per day. It would appear
that the moving average approach offers some advantages in that it
approximates the human body's integrating response to cumulative
carbon monoxide exposure. In addition to the 8-h averages, carbon
monoxide concentrations are also reported in terms of other averaging
times and frequency distributions.
Carbon monoxide concentrations in the ambient air vary
considerably, not only among urban areas but within cities as well.
The maximum 8-h mean concentrations measured at more than 200
monitoring stations in the USA in 1973 ranged from less than
10 mg/m3 to 58 mg/m3 (8.7-51 ppm) with most of the values being
less than 30 mg/m3 (26 ppm) (Martin & Stern, 1974). The highest 8-h
mean concentration of 59 mg/m3 was observed at a monitoring station
in New York City. Data from a 38 station network in Japan showed that
the 8-h standard of 23 mg/m3 (20 ppm) was not exceeded (Environment
Agency, 1973). That carbon monoxide concentrations are extremely
variable within an urban area is illustrated by the maximum 8-h means
observed at monitoring stations in the metropolitan Los Angeles area
in 1973. They ranged from 7 mg/m3 (6 ppm) in the outlying areas to
49 mg/m3 (42.6 ppm) near the centre of the city. The maximum 1-h
mean concentrations exhibit a similar variability ranging from less
than 10 mg/m3 (8.7 ppm) in some areas to over 90 mg/m3 (78.3 ppm)
in others. The highest 1-h mean concentration observed in 1973 was
92 mg/m3 (80 ppm) at a monitoring station in Philadelphia; at more
than 80% of the stations the maximum 1-h concentration was below
50 mg/m3 43.5 ppm). In Japan the maximum 1-h concentrations ranged
from 5 mg/m3 to 48 mg/m3 (4.3-41.8 ppm).
Although the data presented above refer to measurements carried
out in the USA and Japan, it is indicated that very similar conditions
are prevalent in many parts of the world.
Annual average concentrations of carbon monoxide are not of much
value for assessing human exposure, although they do provide an
indication of the long-term trends. Annual average concentrations of
carbon monoxide in most locations fall well below 10 mg/m3 (8.7 ppm)
(Stewart et al., 1976).
Usually, the monitoring stations are located so that they can
provide representative information of air pollution within the
community. However, in certain areas, such as loading platforms and
underpasses, the concentrations found are much higher than those in
city streets. At Chicago post office loading platforms (Conlee et al.,
1967), ambient carbon monoxide concentrations ranged from 10 to
88 mg/m3 at various locations. Wright et al. (1975) determined the
levels of carbon monoxide encountered by pedestrians and street
workers in Toronto. They reported carbon monoxide levels ranging from
11 to 57 mg/m3, with much higher concentrations in poorly ventilated
underpasses and underground garages.
Industry also contributes to the pollution of the ambient air. Gas
generator plants (Nowara, 1975), smelters (Morel & Szemberg, 1971),
steelworks (Butt et al., 1974; Maziarka et al., 1974); plastics works
(Argirova, 1974, unpublished data)a, electric generating plants
(Grigorov et al., 1968), and mines (Notov, 1959) have all been
suggested, among other industrial activities, as sources of
environmental carbon monoxide pollution. Rural work places, especially
those involving intensive livestock production facilities, can provide
high ambient levels of carbon monoxide. Concentrations up to
136 mg/m3 (119 ppm) have been found in these conditions, with high
levels of carboxyhaemoglobin in both animals and workers (Long &
Donham, 1973).
5.2 Indoor Concentrations and Exposure
Carbon monoxide is widely generated indoors by heating, cooking,
and tobacco smoking. According to Yocom et al. (1971), gas heating
systems did not appear to affect indoor carbon monoxide
concentrations, but gas stoves, water heaters, and automobile activity
in attached garages could be major sources. These authors also
reported that carbon monoxide concentrations were generally unrelated
to outdoor levels. Obviously, peak concentrations in the kitchen were
observed during meal times, Sofoluwe (1968) reported extremely high
concentrations of carbon monoxide in Nigerian dwellings, where
firewood was used for cooking. During the preparation of meals,
average carbon monoxide concentrations were reported to be over
1000 mg/m3 (870 ppm) with peak levels as high as 3400 mg/m3
(2960 ppm). Sidorenko et al. (1970) found an indoor level of carbon
monoxide of 32.3 mg/m3 (28 ppm) some two and a half hours after
domestic gas combustion began, when there was no ventilation, but only
13.4 mg/m3 (11.6 ppm) when ventilation was provided. The use of
improved stoves resulted in better conditions (Sidorenko et al.,
1972).
a "Argirova, M. [Atmospheric air pollution by a plant processing
plastics.] In: Proceedings of the First National Conference on
Sanitary Chemistry of the Air, Sofia, Bulgaria, 26-27 November,
1974 (in Bulgarian).
Rench & Savage (1976) have also investigated winter levels of
carbon monoxide in the home. Outdoor concentrations were lower than
those found indoors. Age of building, type of appliance, heating
sources, and socio-economic status were statistically significantly
related to indoor levels of carbon monoxide, higher levels occurring
in the kitchens of old houses and in homes belonging to families in
lower income groups. Levels were also higher in kitchens with space
wall heaters compared with those with forced air, gravity feed, or hot
water heating systems. Places of recreation may be problem areas.
Excessive levels of carbon monoxide were found in ice-skating areas
where ice-resurfacing machines were used. Levels as high as
348 mg/m3 (304 ppm) were found in such an arena after complaints of
illness among children skating there were reported to the local health
department (Johnson et al., 1975). Improperly regulated space heaters
in such premises could also produce high concentrations of carbon
monoxide. This was recently reported to have occurred in an Alaskan
ice-skating arena.
That outdoor concentrations of carbon monoxide can sometimes
strongly influence indoor levels was illustrated in a study conducted
in New York City (Lee, 1972). As shown in Fig. 3, the concentrations
inside and outside an apartment building followed the same general
pattern and were closely related to nearby traffic flows.
Relatively high concentrations of carbon monoxide have also been
observed inside the passenger compartment of motor vehicles (Borst,
1970). Carbon monoxide may enter the compartment from faulty or
damaged exhaust systems or from the surrounding air, in road traffic.
The carbon monoxide level in the compartment is often higher than that
found outside the vehicle. Haagen-Smit (1966) found an average
concentration in a vehicle passenger compartment of 42 mg/m3
(36.5 ppm) on a Los Angeles freeway during the rush hour traffic
(higher concentrations have been reported by Aronow et al. (1972).
It should be emphasized that cigarette smoking is the most common
source of carbon monoxide for the general population (Table 5).
Exposure to smoking primarily affects the carboxyhaemoglobin level
of the smoker himself (Kahn et al., 1974; Landaw, 1973). In some
circumstances, such as poorly ventilated enclosed spaces, the tobacco
smoke may affect all of the occupants (section 8.1.6).
Table 5. Percentage carboxyhaemoglobin levels in smokers and
nonsmokersa
Description Nonsmokers Smokers
mean range mean range
UK pregnant women 1.1 3.6
Meat porters 1.6 5.1
Office workers 1.3 6.2
London office workers 1.12 0.1-2.7 5.5 2.2-13.0
29 000 USA blood donors 1.39 0.4-6.9 5.57 0.8-11.9
3311 California longshoremen 1.3 5.9
Munich population 2.36 7.38
Rural Bavarians 1.03 6.06
a From: Wakeham (1976).
5.3 Occupational Exposure
Many occupational groups are subject to high carbon monoxide
exposure. These include, traffic policemen, garage personnel, workers
in the metallurgical, petroleum, gas, and chemical industries, and
firefighters.
An average carbon monoxide level of 172 mg/m3 (149.6 ppm) was
recorded in the air of a Paris police garage between 07h30 and 08h00
and 205 mg/m3 (179 ppm) between 19h30 and 20h00 (Chovin, 1967).
Similarly, Trompeo et al. (1964) found an average level of 112 mg/m3
(97.4 ppm) in 12 underground garages in Rome; a maximum concentration
of 570 mg/m3 (498.5 ppm) was recorded. In similar studies in
enclosed garages with capacities of 300-500 vehicles, where the only
ventilation was via the entrance, the average concentration of carbon
monoxide for a period from 08h00 to 17h00 was 72 mg/m3 (62.6 ppm) in
the summer and 61 mg/m3 (53 ppm) in the winter (Ramsey, 1967a).
The same author (Ramsey, 1967b) reported that the carboxyhaemo-
globin levels of 14 nonsmoking parking garage employees, exposed to an
average concentration of carbon monoxide of 68 mg/m3 (59 ppm),
increased from 1.5 to 7.3%. He also noted that, in smokers exposed to
the same environment, initial carboxyhaemoglobin levels of 2.9% rose
to 9.3% at the end of the day. Smokers not exposed to this environment
had final levels of only 3.9%. Ramsey stated that occupational
exposure played a greater role than smoking in increasing
carboxyhaemoglobin levels. However, in a study of some 350 Canadian
garage and service station personnel, Buchwald (1969) found that
cigarette smoking was the more significant contributor to the high
levels of carboxyhaemoglobin measured. Of the smokers, 70% had levels
in excess of 5%, while a similar level was found in only 30% of the
nonsmokers. Breysse & Bovee (1969) used expired air samples to
determine exposure to carbon monoxide in stevedores, gasoline-powered
lift truck drivers, and winch operators. They made some 700 estimates
of carboxyhaemoglobin, almost 6% of which exceeded 10%. Seven percent
of the stevedores and 18% of the lift truck operators had
carboxyhaemoglobin levels over 10%. Smoking contributed substantially
to the attainment of these high levels. Carboxyhaemoglobin levels as
high as 10% were also found in dock workers in studies by Petrov
(1968). Inspectors at USA-Mexico border crossing stations were found
to be exposed to ambient levels of carbon monoxide that fluctuated
between 6 and 195 mg/m3 (5 and 170.5 ppm). During one hour of an
evening shift, ambient carbon monoxide averaged 131 mg/m3 (114 ppm).
Carboxyhaemoglobin levels of smokers and nonsmokers which were 4.0 and
1.4% respectively, prior to duty rose to 7.6 and 3.8% (Cohen et al.,
1971).
Data obtained by Balabaeva & Kalpazanov (1974) in studies on
traffic policemen in 4 large towns in Bulgaria were similar to those
Chovin (1967) obtained in his study on the exposure of Paris policemen
to carbon monoxide. However, Göthe et al. (1969) found relatively low
levels of carboxyhaemoglobin in Swedish traffic policemen. When blood
lead levels were studied in 50 London taxi drivers using their
carboxyhaemoglobin levels as an index of exposure to exhaust products,
carboxyhaemoglobin levels were higher in day drivers than in night
drivers and in smokers than in nonsmokers. Concentrations in all
groups ranged from 0.4 to 9.7%, and those in nonsmokers (day and
night) from 0.4 to 3.0%.
Direct measurement of carboxyhaemoglobin levels in firefighters
engaged in prolonged firefighting indicated that 10% had values
exceeding 10% (Gordon & Rogers, 1969). However, the high levels of
carboxyhaemoglobin found in the control (non-firefighting) groups make
these conclusions somewhat uncertain. Goldsmith's study (1970) of
longshoremen suggested that the expired alveolar concentrations of
carbon monoxide in smokers were age-related. For example, subjects
smoking one packet of 20 cigarettes per day in the 45-54 year age
group had an average alveolar value of 31 mg/m3 (27 ppm), while the
75-84 year age group had an average of only 16 mg/m3 (14.4 ppm).
Nonsmokers did not exhibit this age-related pattern, since even up to
84 years of age, alveolar concentrations remained at the same level
4 mg/m3 (3.6 ppm).
Exposure to low levels of carbon monoxide may have a significant
influence on the health and efficiency of these workers but this
awaits further study. Many other instances of occupational exposure
are available but most studies are complicated by the unknown or
unreported smoking habits of the workers under study (Grut, 1949).
5.4 Carboxyhaemoglobin Levels in the General Population
The most extensive study of blood carboxyhaemoglobin levels in the
general population was carried out by Stewart and his associates
(1973c and 1974), who took blood samples in 18 urban areas and in some
small towns in the States of New Hampshire and Vermont, USA. Similar
blood samples were evaluated for carboxyhaemoglobin levels by Kahn et
al. (1974), Davis & Gantner (1974), and Wallace et al. (1974) in
metropolitan St Louis. A total of 45 649 blood donors provided blood
for analysis. Stewart's subjects (29 000 including 1016 from St Louis)
were studied in March 1971 and Kahn's (16 649 subjects all from St
Louis) from October 1971 to October 1972. Stewart et al. concluded
that 45% of all the non-smoking donors exposed to ambient carbon
monoxide had a carboxyhaemoglobin saturation greater than 1.5%, while
Kahn's group reported a level of less than 1%.
6. METABOLISM
The primary factors that determine the final level of
carboxyhaemoglobin are: the amount of inspired carbon monoxide; minute
alveolar ventilation at rest and during exercise; endogenous carbon
monoxide production; blood volume; barometric pressure; and the
relative diffusion capability of the lungs. The rate of diffusion from
the alveoli and the binding of carbon monoxide with the blood
haemoglobin are the steps limiting the rate of uptake into the blood.
6.1 Endogenous Carbon Monoxide Production
Carbon monoxide can be produced endogenously from the catabolism
of pyrrole rings, originating from haemoglobin, myoglobin,
cytochromes, and other haem-containing pigments. Haem catabolism is
the main source of endogenous carbon monoxide production, but recent
in vitro investigations suggest additional endogenous sources, e.g.,
lipid peroxidation (Wolff & Bidlack, 1976).
The endogenous carboxyhaemoglobin level in man is estimated to be
about 0.1-1.0%. Increased production of endogenous carbon monoxide has
been found in haemolytic anaemias (Coburn et al., 1966), and could be
expected in haematomas and after exposure to certain toxic chemicals
capable of causing haemolysis. The liver is probably the major source
of endogenous carbon monoxide as a consequence of an increase in liver
cytochromes induced by certain drugs (Coburn, 1970a), or in porphyria
cutanea tarda (acquired or symptomatic porphyria) (White, 1970).
Similarly, bone marrow can become a major source of carbon monoxide in
haematological diseases, such as sideroblastic anaemia, being
characterized by ineffective erythropoiesis (White, 1970). In neonates
endogenous carbon monoxide can be markedly elevated, as well as in
females during the progesterone phase of the menstrual cycle
(Delivoria-Papadopoulos et al., 1970; Longo, 1970), and even more so
during pregnancy (see section 8.2.3). Another important cause of
carboxyhaemoglobin elevation is exposure to several methane-derived
halogenated hydrocarbons (see section 8.1.5).
6.2 Absorption
The classical absorption curves of Forbes et al. (1945) have been
re-evaluated for man at rest by Peterson & Stewart (1970) who exposed
volunteers to a variety of different concentrations of carbon monoxide
for periods ranging from 0.5 to 24 h. Using a regression approach,
they derived the following empirical relationship for blood carboxy-
haemoglobin as a function of ambient carbon monoxide concentration
and exposure time:
log10 y = 0.85753 log10 x + 0.62995 log10 t - 2.29519
where y = % carboxyhaemoglobin
x = carbon monoxide concentration in ppm
t = time in minutes
Although these new data come closer to presenting potential uptake in
individuals exposed to present-day ambient concentrations of carbon
monoxide, they do not apply to the uptake that would occur in active
man. Furthermore, they fit only the linear, non-steady-state portion
of the absorption curve. Ott & Mage (1974) have objected to the
Peterson & Stewart equation as being basically a static model. They
indicate that the use of averaging periods as long as 1 h compared
with 10-15 min introduces an error into recorded urban concentrations.
This error may be serious if many sharp, ambient peaks are present.
The data presented by Forbes et al. (1945) are still the only
experimental information available that takes ventilation into account
and, even so, this information is inadequate since the full range of
inspired, ventilatory volumes possible in exercising man was not
considered. Coburn et al. (1965) have developed an equation from which
it is possible to calculate blood carboxyhaemoglobin (Peterson &
Stewart, 1970) as a function of time, considering appropriate
physiological and physical factors. Their basic differential equation
is as follows:
d(CO) [HbCO] pCo2 1 pIco
= Vco - × × +
dt [HbO2] M 1 pB - 47 1 pB - 47
+ +
DL VA DL VA
d(CO)
where is the rate of change of CO in the body (ml/min)
dt
[HbCO] is the concentration of CO in the blood (ml CO/ml blood)
[HbO2] is the concentration of O2 in the blood (ml O2/ml blood)
DL is the diffusion capacity of the lung (ml/min/mmHg)
VA is the alveolar ventilation rate (ml/min)
pB is barometric pressure (mmHg)
pIco is inspired carbon monoxide pressure (mmHg)
pCo2 is the mean pulmonary capillary oxygen pressure (mmHg)
M is the Haldane constant (220-240 at pH 7.4)
Vco is the rate of endogenous CO production (ml/min)
One solution of the equation depends on the assumption that
pCo2, and HbO2 are constant and independent of HbCO. However,
HbO2 concentration depends upon HbCO in a complex way and, in
general, solution of the equation requires some special computer
techniques using a second approximation; these have been attempted and
general solutions are available (Coburn et al., 1965). Further
development of Coburn's concepts will undoubtedly improve the basis on
which theoretical uptakes can be calculated. For immediate, practical
purposes, calculations based on the Haldane formula (see section 6.3)
can be used.
Table 6 indicates the equilibrium percentage saturation of the
haemoglobin with carbon monoxide at various alveolar pressures of
carbon monoxide, calculated from the Haldane formula:a
% HbCO 230 pAco
=
% HbO2 pAo2
where pAco and pAo2, are the alveolar pressures of carbon
monoxide and oxygen respectively.
Table 6. Percent carboxyhaemoglobin versus carbon monoxide alveolar pressure
% HbCO mg/m3 ppm % in air Pa Torr
0.87 5.7 5 0.0005 0.506 0.0038
1.73 11.5 10 0.001 1.013 0.0076
3.45 23.0 20 0.002 2.026 0.0152
5.05 34.5 30 0.003 3.305 0.0248
6.63 46.0 40 0.004 4.052 0.0304
8.16 57.5 50 0.005 5.065 0.0380
9.63 69.0 60 0.006 6.078 0.0456
11.08 80 70 0.007 7.091 0.0532
12.46 92.0 80 0.008 8.104 0.0608
13.80 103.0 90 0.009 9.117 0.0684
15.11 114.5 100 0.010 10.130 0.0760
16.37 126.0 110 0.011 11.143 0.0836
17.60 130.0 120 0.012 12.156 0.0912
18.78 149.0 130 0.013 13.170 0.0988
19.95 160.0 140 0.014 14.183 0.1064
21.05 172.0 150 0.015 15.196 0.1140
22.15 183.0 160 0.016 16.209 0.1216
23.23 195.0 170 0.017 17.209 0.1291
24.26 206.0 180 0.018 18.235 0.1368
25.25 218.0 190 0.019 19.221 0.1442
26.22 229.0 200 0.020 20.261 0.1520
a The alveolar oxygen pressure is assumed to be 13 kPa (98 Torr)
6.3 Reactions with Body Tissues and Fluids
An adequate oxygen supply to maintain tissue metabolism is
provided by the integrated functioning of the respiratory and
cardiovascular systems to transport oxygen from the ambient air to the
various tissues of the body. Nearly all of the oxygen, except that
dissolved in plasma, is bound reversibly to the haemoglobin contained
within the erythrocytes. The most significant chemical characteristic
of carbon monoxide is that it also is reversibly bound by haemoglobin.
Therefore, it is a competitor with oxygen for the four binding sites
on the haemoglobin molecule.
The equilibrium constant M expresses the relative affinity of
haemoglobin for carbon monoxide and oxygen when the concentration of
reduced haemoglobin is minimal. This Haldane constant (Douglas et al.,
1912) is defined by the following equation:
HbCO pCO
= M x
HbO2 pO2
where pCO and pO2 represent the equilibrium partial gas
pressures: each pressure being the same in the erythrocytes or
haemoglobin solution as in the equilibrated gas phase. [HbCO] and
[HbO2] are the concentrations of carboxyhaemoglobin and
oxyhaemoglobin, respectively. The value of M is about 200 in most
species, in spite of the fact that carbon monoxide combines with
haemoglobin more slowly than oxygen. Carboxyhaemoglobin dissociates
very slowly due to the tight binding of carbon monoxide to
haemoglobin. Technically, it is not possible to measure the rate of
dissociation of carbon monoxide from partly saturated haemoglobin. The
dissociation velocity constant has been measured by only a few
investigators on sheep and human haemoglobin fully saturated with
carbon monoxide. Roughton (1970), however, has presented the most
comprehensive analysis of the interaction of carbon monoxide with
erythrocyte haemoglobin, showing clearly that, for man, M is higher
than commonly thought, i.e., it is more likely to be between 240 and
250; however, the value for M depends on the point of reference on
the dissociation curve.
Solution of Haldane's equation would give an approximate level of
carboxyhaemoglobin, e.g., exposure to ambient air containing carbon
monoxide levels of 28.7, 57.5, or 115 mg/m3 (25, 50, or 100 ppm)
would lead to carboxyhaemoglobin saturations of approximately 4.8, 9.2
and 16.3%, if the arterial oxygen pressure were 10.7 kPa (80 Torr).
The carbon monoxide enters the lungs with each breath and diffuses
across the alveolar-capillary membrane in a manner similar to oxygen.
If air with a constant concentration of carbon monoxide is breathed
for several hours, the rate of uptake of carbon monoxide decreases
approximately exponentially until an equilibrium state is attained in
which the partial pressure of carbon monoxide in the pulmonary
capillary blood is the same as that in the alveolar.
Oxygen transport in the blood is best described by the
oxyhaemoglobin dissociation curve (Fig. 4). In the presence of
carboxyhaemoglobin, this curve is no longer typically sigmoid but is
shifted to the left so that a lower oxygen pressure is present for the
same oxyhaemoglobin saturation compared with blood without
carboxyhaemoglobin (Roughton & Darling, 1944). Fig. 5 illustrates the
extent of the Haldane shift to the left more clearly than the typical
curves of Fig. 4. Mulhausen et al. (1968) illustrated this shift by
showing that the half saturation oxygen tension shifted from 3.6 to
3.1 kPa (26.7 to 23.2 Torr) in subjects who were intermittently
exposed to a high concentration of ambient carbon monoxide. Carbon
monoxide not only diminishes the total amount of oxygen available by
direct replacement of oxygen (Fig. 4) but also alters the dissociation
of the remaining oxygen so that it is held more tenaciously by
haemoglobin and released at lower oxygen tensions. The oxyhaemoglobin
curve in the presence of carboxyhaemoglobin progressively resembles
the simple oxygen dissociation curve of myoglobin. Myoglobin is a haem
compound with only one haem unit per molecule and does not exhibit
haem-haem interactions. It is possible that the combination of one or
more of the four haem groups in haemoglobin with carbon monoxide
decreases the haem-haem interactions of the remaining haem units and
results in a molecule approaching the behaviour of myoglobin.
Any consideration of the toxicity of carbon monoxide must include
not only the decrease in the oxygen carrying capacity of haemoglobin
but also the interference with oxygen release at the tissue level.
The venous pO2 values expected to result from various
carboxyhaemoglobin levels can be calculated (Forster, 1970; Permutt &
Fahri, 1969). If blood flow and metabolic rate remain constant,
equilibration with an ambient carbon monoxide concentration of
230 mg/m3 or 200 ppm (25% carboxyhaemoglobin) will lower venous
pO2 from 5.3 to less than 4 kPa (40 to less than 30 Torr). A
similar degree of venous hypoxaemia results from an ascent to an
altitude of 3658 metres or a 35% reduction in oxygen capacity in an
anaemic patient. It can also be calculated that, at 5%
carboxyhaemoglobin, there will be only a slight drop in the mixed
venous pO2. Even more valuable relationships can be obtained by
plotting oxygen content against the partial pressure of oxygen
(Roughton & Darling, 1944). The difference in oxygen content at
various percentages of carboxyhaemoglobin from 0 to 20 reveal the
minimal magnitude of the relative unavailability of oxygen due to the
Haldane effect (Fig. 4). It was this evaluation that led Roughton &
Darling to conclude that carboxyhaemoglobin concentrations of less
than 40% produce relatively easily compensated restrictions in the
amount of oxygen available for tissue delivery. This can only be
applied to subjects with normal respiratory and circulatory systems.
The small reductions in oxygen content at 5-10% carboxyhaemoglobin may
be quite critical for patients suffering from cardiovascular diseases
or chronic obstructive lung disease. Coburn et al. (1965) published a
detailed theoretical analysis of the physiology and variables that
determine blood carboxyhaemoglobin levels in man. The details of the
formulae used in these calculations are presented in section 6.2.
With regard to the intracellular effects of carbon monoxide,
consideration must be given to the interactions of all substances
within the tissue cells that are involved with oxygen delivery. Since
haemoglobin and myoglobin are structurally related, they react with
carbon monoxide in a similar manner. The function of myoglobin, in
vivo, may be to act as a reservoir for oxygen within the muscle
fibre. The carbon monoxide and oxygen equilibria of human myoglobin
has been studied in vitro and a hyperbolic oxygen dissociation curve
established (Rossi-Fanelli & Antonini, 1958). This curve, unlike that
for haemoglobin, is not affected by the hydrogen ion concentration,
the ionic strength, or the concentration of myoglobin. The relative
affinity constant, M, is approximately 40 but is still sufficient to
induce appreciable formation of carboxymyoglobin. Both Coburn et al.
(1970b, 1973) and Luomanmaki (1966) have studied the interrelation-
ships between carboxyhaemoglobin and carboxymyoglobin. Coburn, using
14CO, showed that identical carbon monoxide exposures can produce
different degrees of saturation of haemoglobin, depending upon the
partial pressures of oxygen in blood and tissue. Coburn (1970b)
determined the ratio of the carbon monoxide content in muscle to the
content in blood as a function of arterial pO2. This ratio, for
skeletal muscle, is approximately 1, but in myocardial tissue it was
found to be 3. When arterial pO2 fell below 5.3-4 kPa (40-30
Torr), carbon monoxide disappeared from the blood, presumably entering
the muscle. Considerable amounts of extravascular carbon monoxide are
stored in muscle. The higher ratio for cardiac tissue may be of
considerable significance. In an individual with a blood
carboxyhaemoglobin level of 10%, some 30% of cardiac myoglobin may be
saturated with carbon monoxide. Coburn and associates (1973) estimated
the mean pO2 of skeletal muscle and myocardium and found that they
were 0.8-1.1 and 0.5-0.8 kPa (6-8 and 4-6 Torr), respectively.
Although no final judgement can be made regarding the next lower
step involved in oxygen transport, i.e., the role of cytochromes a3
and P-450, the fact that, experimentally, they react with carbon
monoxide in the same way as other haem-containing substances suggests
that they may play a role in carbon monoxide poisoning. Available
evidence suggests that interactions between carbon monoxide and
cytochrome oxidases are of minor significance at the concentrations of
carbon monoxide found in community air pollution. All of the data on
the cytochromes have been obtained from in vitro experiments.
Whether similar events occur in vivo remains uncertain. The most
likely oxidase for in vivo inhibition is P-450. Cooper et al. (1965)
reported that the ratio of carbon monoxide to oxygen required for 50%
inhibition is close to 1, in contrast to a similar ratio of between
2.2 and 28 for cytochrome a3. Root (1965) believes that at a pCO
compatible with life, only nonsignificant blocking of the oxygen
consumption system occurs. In terms of the total distribution
throughout the body of an inhaled dose of carbon monoxide, the amounts
bound to these haemoproteins are small compared with haemoglobin and
myoglobin. A diagrammatic representation of the factors influencing
body carbon monoxide stores has been presented by Coburn (1970b). The
possible significance of the role of these haemoproteins lies in the
concept that, under conditions where tissue pO2 is decreased, the
affinity of intracellular haemoproteins for carbon monoxide may
increase.
6.4 Excretion
Adequate data are available on the rate of absorption of carbon
monoxide but there is considerably less information concerning the
rates of carbon monoxide egress from the lungs. The same factors that
determine how much carbon monoxide is taken up by the blood should
apply in reverse when clearance of carbon monoxide from blood is
considered. The primary factors involved are the amounts of carbon
monoxide and oxygen present, the magnitude of ventilation, and the
quality of the diffusion barrier. Age influences the quality of the
barrier and it appears that with advancing age the barrier becomes
"thicker" and there are fewer gas exchange membranes. Sedov et al.
(1971) presented data on the elimination of carbon monoxide at various
atmospheric pressures and ambient temperatures. Neither lower
barometric pressures nor high temperatures appreciably altered the
rates of elimination. It has been implied by Pace et al. (1950) that a
sex difference in elimination may also exist, a faster rate occurring
in females than in males, at least under the experimental conditions
of their study. Some sex differences in excretion have been reported
by Goldsmith et al. (1963).
Available evidence suggests that there is a biphasic decline in
the percentage of carboxyhaemoglobin in the arterial blood (Godin &
Shephard, 1972; Wagner et al., 1975). There is a rapid, initial,
exponential decline (distribution phase), probably related to the
distribution of carbon monoxide from the circulating blood to splenic
blood, myoglobin, and cytochrome enzymes. Elimination of carbon
monoxide through the lungs also occurs during this phase. The
distribution phase, which persists for the first 20-30 min, is
followed by a slower linear decline (elimination phase). This phase
probably reflects the rates of release of carbon monoxide from
haemoglobin and myoglobin, pulmonary diffusion, and ventilation, as
well as the fact that pCO decreases with time. Myhre (1974) found
that a similar biphasic excretion pattern occurred at an altitude of
1630 metres. However, he noted that the half-time of carboxyhaemo-
globin was much longer (5.5 h). After continuous exposure to carbon
monoxide for 49 h, 50% was eliminated in 30-180 min and 90% within
180-420 min (Tiunov & Kustov, 1964). Other investigators (Godin &
Shephard, 1972; Peterson & Stewart, 1970) reported exponential
carboxyhaemoglobin elimintation curves over many hours. However,
because of inadequate sampling in the early phase of elimination, they
were unable to observe the more rapid initial decline. The absolute
level of carboxyhaemoglobin, when the elimination studies began,
apparently modified the rate of disappearance of carbon monoxide from
the blood. Considerable individual variability has been observed and
the duration of exposure may also be an important factor. It has been
shown that prolonged exposure (more than 3 years) to concentrations of
carbon monoxide of 11.5-115 mg/m3 (10-100 ppm) results in a markedly
retarded elimination of carbon monoxide (Kodat, 1971).
In summary, discharge of carbon monoxide occurs rapidly at first
becoming slower with time and the lower the initial level of
carboxyhaemoglobin, the slower the rate of elimination. Apparently, no
studies have been made to determine elimination rates at the low
levels of carboxyhaemoglobin (2-4%) that might be present following
exposure to ambient concentrations of carbon monoxide.
Several procedures have been tested that could accelerate the
excretion of carbon monoxide from the blood of individuals with high
levels of carboxyhaemoglobin (40-70%). Pace et al. (1950) reported
that treatment of such individuals in a recompression chamber with an
oxygen level equivalent to 2.5 atmospheres (partial pressure of oxygen
equal to 253 kPa (1900 Torr) or an alveolar oxygen pressure of 239 kPa
(1801 Tort)) would facilitate removal of carbon monoxide. They
indicated that 1 h in such a chamber would result in the reduction of
the carboxyhaemoglobin level to 10 to 15% of the initial level. In a
revival of the old Henderson & Haggard treatment concept, Malorny et
al. (1962) evaluated the influence of breathing different gas mixtures
on the excretion of carbon monoxide in animals having high
carboxyhaemoglobin levels (60-74%). It was determined that 50% of
carbon monoxide could be excreted in 19 min if 5% carbon dioxide and
95% oxygen were breathed, compared with a time of 28 min for 100%
oxygen, and 41 min for ambient air. Another approach was suggested by
Agostini et al. (1974), who employed a total body asanguineous, hypo-
thermic procedure. These approaches to the removal of the body burden
of excessive amounts of carbon monoxide remain to be evaluated fully
in clinical trials.
7. EFFECTS ON EXPERIMENTAL ANIMALS
Great caution must be used in applying the resultsa obtained
from animal experiments to man. Nonetheless, animal studies have
provided valuable insight into both the potentially adverse effects of
carbon monoxide and the basic mechanisms by which this substance
influences physiological processes. However, many studies have used
extraordinarily high levels of carbon monoxide, rarely found in air.
These studies are not referred to in this document, since the toxic
effects of very high levels of carbon monoxide have been well
documented for both animals and man.
7.1 Species Differences
The oxygen dissociation curves for different animal species used
in carbon monoxide studies are not the same. There are also questions
concerning the relative affinities for carbon monoxide of the
haemoglobin in various animal species. Fodor & Winneke (1971.)
reported in vitro studies that showed different affinities for
different species, the highest being in man followed by rat, mouse,
and rabbit, in descending order. Klimisch et al. (1975) found the
following sequence of affinities: hamster, rat, pig, rabbit. Apart
from different affinities there may well be species differences with
respect to susceptibility to carbon monoxide. According to Alexandrov
(1973) certain mammals can be classified in order of decreasing carbon
monoxide susceptibility as follows: mouse, rat, cat, dog, guineapig,
and rabbit. This difference might, in part, be related to different
ventilation/body weight-ratios.
Species differences in reaction are ideally illustrated in the
studies by De Bias et al. (1972a,b, 1973) on dogs and cynomolgus
monkeys (Macaca fascicularis = M. irus). Chronic exposure (23 h per
day) over several months to 115 mg/m3 (100 ppm) resulted in
carboxyhaemoglobin levels of 14 and 12.4% in dogs, and monkeys,
respectively. The dogs (De Bias et al, 1972a) remained clinically in
good health with no untoward signs that could be interpreted as
induced by carbon monoxide. Serum enzymes, haematological variables,
and electrocardiograms did not change significantly. Carbon monoxide
exposure of normal monkeys resulted in myocardial effects.
Experimentally infarcted monkeys had greater P-wave amplitudes and
increased incidence of T-wave inversions than normal monkeys similarly
exposed. One important fact that may partly explain this difference
between dogs and monkeys is, that monkeys, like man, have end-
arteries, whereas dogs have a well developed collateral circulation.
a Animal studies have been reported in other sections of this
document wherever necessary. Consequently, some data on animals
(such as sheep) are not repeated in this section.
7.2 Cardiovascular System and Blood
Increase of coronary blood flow is the normal response of the
myocardium to carbon monoxide exposure. Permutt & Farhi (1969)
calculated theoretically, that, when carboxyhaemoglobin levels are
approximately 5%, an increase in coronary blood flow of about 20% more
than the resting level would be necessary to prevent coronary sinus
pO2 from falling below minimum levels. This calculation has been
confirmed experimentally by Adams et al. (1973) and Horvath (1973),
both of whom demonstrated approximately similar increases in coronary
blood flow in spite of using very different methods to increase
carboxyhaemoglobin levels.
There is considerable controversy concerning the cardiovascular
effects of carbon monoxide in dogs. Long-term exposures of animals to
carbon monoxide concentrations sufficiently high to produce
carboxyhaemoglobin levels in excess of 20% can induce pathological
changes in the heart and brain. As in acute high level intoxication in
man, serious sequelae develop. Lindenberg et al. (1961) exposed 8 dogs
to carbon monoxide concentrations of 115 mg/m3 (100 ppm). Four were
exposed continuously for 24 h a day, 7 days per week, for 6 weeks and
another 4 were exposed intermittently. All dogs had abnormal
electrocardiograms and some of the hearts showed histological
degeneration of muscle. In another study, Preziosi et al. (1970)
exposed dogs continuously to 115 mg/m3 for 6 weeks and reported
abnormal electrocardiograms, right and left heart dilatation, and
myocardial thinning. Histological examination showed older scarring in
some cases and fatty degeneration of heart muscle in others.
Carboxyhaemoglobin levels of 7.7 to 12% were lower than would be
predicted from the exposure. When 4 dogs were exposed intermittently
to a carbon monoxide concentration of 115 mg/m3 for 11 weeks,
central nervous system and cardiac effects were found (Lewey &
Drabkin, 1944).
Ehrich et al. (1944) also exposed 4 dogs to 115 mg/m3 on an
intermittent schedule. They observed that electrocardiographic changes
occurred at variable times during the 11 weeks of exposure. The gross
appearance of the hearts after exposure were normal but microscopic
examination revealed marked degenerative changes in individual fibres.
Lindenberg et al. (1961) exposed dogs to a carbon monoxide
concentration of 57 mg/m3 (50 ppm) on the same schedule as for his
115 mg/m3 exposure studies. Carboxyhaemoglobin levels of 2.6-5.5%
were observed. No changes in haemoglobin levels or haematocrit were
noted. Electrocardiographic changes were observed in the third week of
exposure similar to, but less severe than, those observed in animals
exposed to the higher ambient level of carbon monoxide. Dogs were
exposed by Musselman et al. (1959) to a carbon monoxide concentration
of 57 mg/m3 for 24 h per day, 7 days per week, for 3 months. No
changes in the electrocardiogram or heart rates were observed.
Pathological examination of organs and tissues did not reveal any
changes after exposure. Experimental data have been presented by
Orellano et al. (1976) who claim that carbon monoxide injected
intraperitoneally into dogs is nontoxic. This study, as well as other
reports from these investigators, raise some intriguing questions as
to the mechanisms involved in carbon monoxide toxicity.
Continuous exposure of cynomolgus monkeys to a carbon monoxide
concentration of 115 mg/m3 resulted in demonstrable electro-
cardiographic effects in the myocardium of both normal monkeys
and monkeys with myocardial infarction (De Bias et al., 1972b, 1973).
The same investigators (De Bias et al., 1976) also studied the
susceptibility of the ventricles to induced fibrillation. Normal and
infarcted monkeys were exposed to a carbon monoxide concentration of
115 mg/m3 for 6 h. Application of high voltages were required to
produce fibrillation in normal monkeys in air but when infarcted
animals were exposed to carbon monoxide, the voltage level required
was very low. Animals with either infarction alone or carbon monoxide
exposure alone also required significantly less voltage to produce
fibrillation; when the two were combined, the effects were additive.
Intermittent exposure to carbon monoxide (30 min per h, 12 h per day,
over a period of 14 months) of cynomolgus monkeys fed either a normal
or a semipurified cholesterol diet did not result in myocardial
infarctions or electrocardiographic abnormalities (Malinow et al.,
1976). In these animals, the blood carboxyhaemoglobin concentration
reached 21.6% at the end of the daily period of breathing carbon
monoxide at 529 mg/m3 (460 ppm).
Carbon monoxide exposure did not increase the aortic and coronary
atherosclerosis induced in cynomolgus monkeys by cholesterol feeding.
Eckardt et al. (1972) exposed cynomolgus monkeys (22 h per day, 7 days
per week, for 2 years) to carbon monoxide concentrations of 23 or
75 mg/m3 (20 or 65 ppm). Carboxyhaemoglobin levels, which showed
considerable variation during the experimental period, ranged from 2.0
to 5.5% and 4.8 to 10.2% for low and high carbon monoxide ambient
concentrations, respectively. These levels of carboxyhaemoglobin did
not lead to compensatory increases in haematocrit, haemoglobin, or
erythrocyte counts nor to cardiac fibrosis or pathological effects in
the brain. Cholesterol-fed squirrel monkeys were exposed to carbon
monoxide at 229-344 mg/m3 (200-300 ppm) for several hours per day
for 7 months (Webster et al., (1968). No differences were found in
plasma cholesterol or in aortic and carotid atherosclerosis which
could be ascribed to carbon monoxide. However, the authors did observe
an increase in coronary atherosclerosis. Somewhat similar results were
later reported in studies on cynomolgus monkeys (Malinow et al.,
1976). Jones et al. (1971) exposed rats, guineapigs, dogs, and monkeys
to 58, 110, or 229 mg/m3 (51, 96 or 200 ppm) continuously for 90
days. Haematocrit and haemoglobin levels remained constant at the
lowest level of carbon monoxide exposure but were significantly
elevated at the two higher levels in all species except the dog.
Cynomolgus monkeys (Macaca irus) were exposed to a concentration of
286 mg/m3 (250 ppm) for 2 weeks by Thomsen (1974). In all the
exposed animals, the coronary arteries showed widening of the
subendothelial spaces in which cells with or without lipid droplets
were accumulating. He suggested that monkeys were more sensitive to
carbon monoxide than the rabbits studied by Astrup et al. (1967).
Experimental studies on rabbits exposed to relatively high
concentrations of carbon monoxide at 195-206 mg/m3 (170-180 ppm) for
extended periods indicated the presence of high levels of cholesterol
in the arteries or enhanced vascular disease (Astrup et al., 1970).
The lesions observed, which included subendothelial oedema, a gap
between endothelial cells, and increased infiltration of cells with
lipid droplets, might be early precursors of atherosclerotic disease.
However, such lesions occurred only in animals concurrently on a high
cholesterol or fat diet or on both. Exposure to carbon monoxide alone
induced some changes such as endothelial hypertrophy and
proliferation.
One experimental study on the effects of carbon monoxide on the
natural history of heart disease in the cynomolgus monkey has been
reported. De Bias et al. (1973) exposed animals to a carbon monoxide
concentration of 137 mg/m3 (120 ppm) for 24 weeks. The average
carboxyhaemoglobin level of 12.4% resulted in a polycythaemia with an
increase in haematocrit from 35 to 50%. All animals developed
increased P-wave amplitude and T-inversion which suggested nonspecific
myocardial stress rather than ischaemia. Animals in which an
experimental myocardial infarction was produced prior to exposure to
carbon monoxide had more marked electro-cardiographic changes than
animals breathing room air. In 1976, Ramsey & Casper reported that
erythrocytic 2,3-diphosphoglycerate (2,3-DPG) played neither a
compensatory nor an aggravating role in the hypoxia induced by the
presence of 20 or 30% carboxyhaemoglobin. Stupfel & Bouley (1970)
exposed mice and rats for 95 h per week to a carbon monoxide
concentration of 57 mg/m3 (50 ppm) for either 1 to 3 months or for
their natural life expectancy of up to 2 years. A large number of
measurements were made during exposure followed by pathological
examination after death. The authors did not observe any important
effects of carbon monoxide exposure on the animals. In a study by
Penney et al. (1974a,b), the influence of hypoxic hypoxia on the
development of cardiac hypertrophy in the rat was compared with that
of carbon monoxide hypoxia. Exposure to various levels of carbon
monoxide resulted in hypertrophy of both the fight and left ventricles
in contrast with the right ventricle hypertrophy observed in response
to the hypoxic hypoxia stress.
Cardiac hypertrophy and a reduction in cytochrome oxidase levels
were demonstrated in chick embryos exposed to carbon monoxide for 144
and 168 h (Tumasonis & Baker, 1972). The resistance of young chickens
to carbon monoxide decreased with age. Body temperatures decreased
during exposure to carbon monoxide with the greatest fall in
temperature and the longest survival time occurring in the youngest
chickens. Total body asanguineous hypothermic perfusion (total body
exsanguination exchange transfusion) has been suggested as a
therapeutic measure for carbon monoxide poisoning (Agostini et al.,
1974). In an attempt to explain this beneficial effect, Ramirez et al.
(1974) compared the survival of normal dogs exposed to high levels of
carbon monoxide with that of acutely anaemic dogs transfused with
carboxyhaemoglobin blood to normal blood volumes. All normal dogs with
carboxyhaemoglobin levels of 54-100% died within 0.25-10 h but the
transfused animals, having a final mean carboxyhaemoglobin level of
80% after transfusion, survived. The authors suggested that hypoxic
anaemia was not the principal mechanism of carbon monoxide toxicity
but rather a blocking out of the energy supply on the cellular level,
governed by the cytochrome system.
Most of the investigations using rabbits for carbon monoxide
related research originated in Astrup's laboratories, where they first
demonstrated that low carboxyhaemoglobin levels enhanced the
development of atheromatosis (Astrup et al., 1967). Additional studies
(Hellung-Larsen et al., 1968; Thomsen & Kjeldsen, 1975) have shown
that lactate dehydrogenase isoenzymes (M subunits) increase and that a
higher incidence of focal intimal changes occur in rabbits exposed to
carbon monoxide. The myocardial ultrastructure of rabbits exposed to a
concentration of 206 mg/m3 (180 ppm) for at least 4 h showed
degenerative changes such as contraction bands, myofibrillar
disintegration, myelin body formation, and dehiscence of the
intercalated discs (Thomsen & Kjeldsen, 1974). Exposure of rabbits to
a similar concentration of carbon monoxide for 2 weeks resulted in
more extensive myocardial damage (Kjeldsen et al., 1974).
Astrup et al. (1970) furnished evidence that carbon monoxide
increases endothelial membrane permeability. They found that rabbits
exposed to carbon monoxide developed arterial lesions resulting in a
considerable accumulation of lipids. It has also been shown that human
coronary arteries exposed to carbon monoxide in vitro have a higher
uptake of cholesterol, although no significant changes in lipid
synthesis were observed (Sarma et al., 1975). Myocardial damage and
impaired myocardial performance have also been reported in animals and
man exposed to carbon monoxide (Ayres et al., 1970). Although there is
some evidence which suggests that exposure to carbon monoxide can
induce changes in blood vessels and the myocardium, there is also
evidence to the contrary (section 8.1).
7.3 Central Nervous System
Because of the brain's high oxygen demand, cerebral function
should be influenced at low carboxyhaemoglobin levels. However, data
concerning this are contradictory. Sensitivity to carbon monoxide may
follow a circadian rhythm (Stupfel, 1975; Stupfel et al., 1973).
Maximum sensitivity in rats occurred during the dark period of a
12-12 h light-dark cycle. Dyer & Annau (1976) could find no effect on
superior colliculus evoked potentials of rats until the level of
ambient carbon monoxide had reached 573 mg/m3 (500 ppm) in marked
contrast to Xintaras et al. (1966), who observed a 20% increase in the
amplitude of superior colliculus evoked potentials after only 1 h
exposure to 57 mg/m3 (50 ppm), and a 50% increase after a 2-h period
of exposure at this level. This study has, however, been criticized
for not taking into account the effects of dark adaptation on the
amplitude of visual evoked potentials in rats (Dyer & Annau, 1976).
In view of the conflicting reports it is of some interest to
examine the available data on cerebral pO2 tensions, cerebral
blood flow (CBF) and cerebral metabolism. Zorn (1972) studied the
effects of carbon monoxide inhalation on brain and liver pO2 using
platinum electrodes. Tissue pO2 fell in both organs, even at a
carboxyhaemoglobin concentration of 2%, and the fall was approximately
linear to increases in carboxyhaemoglobin. There was a decrease in
pO2 of 0.027-0.24 kPa (0.2-1.8 Torr) for each 1% fall in
oxyhaemoglobin percentage saturation. These data suggest that the
carbon monoxide influenced levels other than the intracellular level,
since if its effects were limited to this area then tissue pO2
would have been expected to increase. Similar studies were performed
by Weiss & Cohen (1974) on rat brain and muscle. They found a decrease
in cerebral cortical pO2 following inhalation of low levels of
carbon monoxide. Unfortunately, they did not measure
carboxyhaemoglobin levels in these rats but, in a group of sham-
operated animals exposed to similar levels of inhaled carbon monoxide,
carboxyhaemoglobin had increased to 3.3%. During progressive
administration of carbon monoxide to dogs, CBF did not increase until
carboxyhaemoglobin levels reached 20%. Thereafter, CBF increased
progressively and was double that in the controls when the
carboxyhaemoglobin level reached 50% (Häggendal et al., 1966). On the
other hand, Traystman (1976) did observe a progressive increase in CBF
in dogs even at very low carboxyhaemoglobin values. The lowest level
studied was 2.5%, which produced a slight but significant CBF
increase. Thus, Traystman did not believe in a threshold effect. The
effects were produced with both hypoxic hypoxia and carbon monoxide
hypoxia. It remains to be seen, how these data relate to cerebral
circulation in man.
The respiratory centre or arterial chemoreceptors were not
stimulated to increase respiratory minute volume even when
carboxyhaemoglobin levels were as high as 40% (Chiodi et al., 1941).
Mills & Edwards (1968) measured the frequency of electrical impulses
in the afferent nerves from the aortic and carotid chemoreceptors and
showed that administration of carbon monoxide did result in
chemoreceptor stimulation. The response appeared to have an
approximately linear relationship with the carboxyhaemoglobin
concentration (at least above 8%). These findings suggest that carbon
monoxide might stimulate breathing. Failure to observe an increased
minute volume may be explained by the fact that, in the presence of
carbon monoxide, the chemoreceptor stimulation was offset by hypoxic
depression of brain structures involved in breathing. There is some
evidence that this balancing between chemoreceptors and central
nervous depression is operative in anaemia (Santiago & Edelman, 1972).
Additional investigations are needed to clarify this effect of carbon
monoxide inhalation.
7.4 Behavioural Changes and Work Performance
In extensive studies on rabbits, guineapigs, rats, and mice
(Gadaskina, 1960; Ljublina, 1960; Rylova, 1960) exposure to a carbon
monoxide level of 30 mg/m3 (26 ppm), although not inducing any
morphological blood changes, did result in a number of unfavourable
physiological changes. Among these were decreased work capacity, poor
adjustment to postural shifts (orthostatic tests), and increased
thyroid activity. These changes were more evident during the initial
period of the exposure but reverted towards normal later, suggesting
some adaptation to carbon monoxide exposure.
7.5