INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY
ENVIRONMENTAL HEALTH CRITERIA 4
OXIDES OF NITROGEN
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 1977
ISBN 92 4 154064 8
(C) World Health Organization 1977
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 OXIDES OF NITROGEN
1. SUMMARY AND RECOMMENDATIONS FOR FURTHER RESEARCH
1.1 Summary
1.1.1 Chemistry and analytical methods
1.1.2 Sources of oxides of nitrogen
1.1.3 Environmental levels and exposures
1.1.4 Effects on experimental animals
1.1.5 Effects on man
1.1.5.1 Controlled exposures
1.1.5.2 Accidental and industrial exposures
1.1.5.3 Community exposures
1.1.6 Evaluation of health risks
1.2 Recommendations for further research
2. CHEMISTRY AND ANALYTICAL METHODS
2.1 Chemical and physical properties
2.2 Atmospheric chemistry
2.3 Analytical methods
2.3.1 Sampling
2.3.2 Evaluation of analytical methods
2.3.2.1 Manual methods
2.3.2.2 Automatic methods
2.3.3 Calibration procedures
3. SOURCES OF OXIDES OF NITROGEN
3.1 Natural sources
3.2 Man-made sources
3.2.1 Stationary sources
3.2.2 Mobile sources
3.2.3 Non-combustion sources
3.2.4 Other sources
4. ENVIRONMENTAL LEVELS AND EXPOSURES
4.1 Background concentrations
4.2 Urban concentrations
4.3 Indoor exposure
4.4 Smoking
5. EFFECTS ON EXPERIMENTAL ANIMALS
5.1 Local effects on the respiratory system
5.1.1 Morphological changes
5.1.2 Functional changes
5.1.3 Biochemical effects
5.2 Other effects
5.2.1 Effects on growth and body weight
5.2.2 Immunological effects
5.2.3 Haematological effects
5.2.4 Miscellaneous biochemical effects
5.2.5 Effects on reproduction
5.2.6 Effects on the central nervous system
5.2.7 Behavioural changes
5.2.8 Carcinogenicity, mutagenicity and teratogenicity
5.3 Interaction of nitrogen dioxide and infectious agents
5.4 Summary table
6. EFFECTS ON MAN
6.1 Controlled exposures
6.2 Accidental and industrial exposures
6.3 Community exposures
6.3.1 Effects on pulmonary function
6.3.2 Effects on the incidence of acute respiratory
disease
6.3.3 Effects on the prevalence of chronic respiratory
disease
6.4 Summary tables
7. EVALUATION OF HEALTH RISKS FROM EXPOSURE TO OXIDES OF NITROGEN
7.1 Exposure levels
7.2 Experimental animal studies
7.3 Controlled studies in man
7.4 Effects of accidental and industrial exposures
7.5 Effects of community exposures
7.6 Evaluation of health risks
REFERENCES
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 information may be considered in the event of
updating and re-evaluating the conclusions contained in the criteria
documents.
WHO TASK GROUP ON ENVIRONMENTAL
HEALTH CRITERIA FOR OXIDES OF NITROGEN
Tokyo, 23-27 August 1976
Participants
Members
Dr K. Biersteker, Medical Research Division, Municipal Health
Department, Rotterdam, Netherlands
Professor K. A. Bustueva, Department of Community Hygiene, Central
Institute for Advanced Medical Training, Moscow, USSR
Dr R. G. Derwent, Environmental and Medical Sciences Division,
Atomic Energy Research Establishment, Harwell, England
Professor L. Friberg, Department of Environmental Hygiene, The
Karolinska Institute, Stockholm, Sweden (Chairman)
Dr D. E. Gardner, Biomedical Research Branch, Clinical Studies
Division, Health Effects Research Laboratory, Environmental
Protection Agency, Research Triangle Park, NC, USA
(Rapporteur)
Dr J. Jager, Centre of General and Environmental Hygiene, Institute
of Hygiene and Epidemiology, Prague, Czechoslovakia
Dr T. Nakajima, Division of Environmental Health Research, Osaka
Prefectural Institute of Public Health, Osaka, Japan
(Vice-Chairman)
Dr G. von Nieding, Laboratorium fur Atmung und Kreislauf,
Krankenhaus Bethanien, Moers, Federal Republic of Germany
Mr E. A. Schuck, US Environmental Protection Agency, Environmental
Monitoring and Support Laboratory, Las Vegas, NV, USA
Observers
Dr J. Kagawa, Department of Medicine, Tokai University,
Kanagawa, Japan
Professor K. Maeda, Department of Medicine, Tokyo University, Tokyo,
Japan
Dr T. Okita, Department of Community Environmental Sciences, The
Institute of Public Health, Tokyo, Japan
Dr H. Watanabe, Hyogo Prefectural Institute of Public Health, Kobe,
Japan
Professor N. Yamaki, Faculty of Engineering, Saitama University,
Saitama, Japan
Dr N. Yamate, First Section of Environmental Chemistry, National
Institute of Hygienic Sciences, Tokyo, Japan
Dr G. Freeman, Department of Medical Sciences, Stanford Research
Institute, Menlo Park, CA, USA (Temporary Adviser)
Dr Y. Hasegawa, Medical Officer, Control of Environmental Pollution
and Hazards, World Health Organization, Geneva, Switzerland
Mr G. Ozolins, Scientist, Control of Environmental Pollution and
Hazards, World Health Organization, Geneva, Switzerland
(Secretary)
Professor C. M. Shy, Institute for Environmental Studies and
Department of Epidemiology, School of Public Health,
University of North Carolina, Chapel Hill, NC, USA
(Temporary Adviser)
Dr T. Suzuki, The Institute of Public Health, Tokyo, Japan
(Temporary Adviser)
Professor T. Toyama, Department of Preventive Medicine, Keio
University, Tokyo, Japan (Temporary Adviser)
ENVIRONMENTAL HEALTH CRITERIA FOR OXIDES OF NITROGEN
A WHO Task Group on Environmental Health Criteria for Oxides of
Nitrogen met in Tokyo from 23 to 27 August 1976. Dr Y. Hasegawa,
Medical Officer, Control of Environmental Pollution and Hazards,
Division of Environmental Health, WHO, opened the meeting on behalf of
the Director-General and expressed the appreciation of the
Organization to the Government of Japan for kindly acting as host to
the meeting. In reply the group was welcomed by Dr M. Hashimoto,
Director-General of the Air Quality Bureau, Environment Agency, Japan.
The Task Group reviewed and revised the second draft criteria document
and made an evaluation of the health risks from exposure to oxides of
nitrogen.
The first and second drafts of the criteria document were prepared
by Dr G. Freeman, Director, Department of Medical Sciences, Stanford
Research Institute, Menlo Park, CA, 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, Federal Republic of Germany, India, Japan, New
Zealand, Poland, Sweden, the USA and the USSR; and from the Food and
Agriculture Organization of the United Nations (FAO), Rome, and the
World Meteorological Organization (WMO), Geneva. The collaboration of
these national institutions and international organizations is
gratefully acknowledged.
The Secretariat also wishes to acknowledge the most valuable
collaboration in the final phase of the preparation of this document,
of Professor C. M. Shy, School of Public Health, University of North
Carolina, NC, USA, Dr D. E. Gardner, Chief, Biomedical Research
Branch, Health Effects Research Laboratory, Environmental Protection
Agency, Research Triangle Park, NC, USA, and Dr R. G. Derwent,
Environmental and Medical Sciences Division, Atomic Energy Research
Establishment, Harwell, England.
This document is based primarily on original publications listed
in the reference section. Much valuable information may also be found
in other published criteria documents (North Atlantic Treaty
Organization, 1973; US Department of Health, Education, and Welfare,
1976; US Environmental Protection Agency, 1971a) and in the reviews on
oxides of nitrogen by Cooper & Tabershaw (1966), Morrow (1975), and
Stern, ed. (1968). Details of the WHO Environmental Health Criteria
Programme including some terms frequently used in the documents may be
found in the general introduction to the Environmental Health Criteria
Programme published together with the environmental health criteria
document on mercury (Environmental Health Criteria 1, Geneva, World
Health Organization, 1976).
The following conversion factors have been used in this document.a
nitric 1 ppm = 1230 µg/m3 carbon 1 ppm = 1150 µg/m3
oxide monoxide
nitrogen 1 ppm = 1880 µg/m3 ozone 1 ppm = 2000 µg/m3
dioxide
nitrous 1 ppm = 1800 µg/m3 sulfur 1 ppm = 2600 µg/m3
oxide dioxide
a When converting values expressed in ppm to µg/m3, the numbers
have been rounded up to 2 or, exceptionally 3 significant figures and,
in most cases, concentrations higher than 10,000 µg/m3 have been
expressed in mg/m3.
1. SUMMARY AND RECOMMENDATIONS FOR FURTHER RESEARCH
1.1 Summary
1.1.1 Chemistry and analytical methods
In the context of this criteria document, the term oxides of
nitrogen is understood to include nitric oxide (NO) and nitrogen
dioxide (NO2). Other oxides of nitrogen which exist in the
atmosphere are not known to have any biological significance and have
not been referred to in this document. At the point of discharge from
man-made sources, the predominant oxide of nitrogen is nitric oxide
which is readily converted to nitrogen dioxide by chemical reactions
in the atmosphere.
Nitric oxide and nitrogen dioxide can be measured separately or
collectively by manual or automated techniques. However, whereas a
certain analytical method can be quite reliable for one compound
("chemiluminescence" for nitric oxide: "Saltzman method" for nitrogen
dioxide), difficulties may arise in the simultaneous monitoring of
both oxides. Gas-phase titration, permeation tubes, and gravimetric
standards have been used for the accurate calibration of these
analytical procedures.
1.1.2 Sources of oxides of nitrogen
On a global scale, quantities of nitric oxide and nitrogen dioxide
produced naturally by bacterial and volcanic action and by lightning
by far outweigh those generated by man's activities. However, as they
are distributed over the entire earth's surface, the resulting
background atmospheric concentrations are very small.
The major source of man-made emissions of oxides of nitrogen into
the atmosphere is the combustion of fossil fuels in stationary sources
(heating, power generation) and in motor vehicles (internal combustion
engines). Other contributions to the atmosphere come from specific
non-combustion industrial processes, such as the manufacture of nitric
acid and explosives. Indoor sources include smoking, gas-fired
appliances, and oil stoves. Differences in the nitrogen dioxide
emission of various countries are mainly due to differences in fossil
fuel consumption.
Worldwide emissions of oxides of nitrogen in 1970 were estimated
at approximately 53 million tonnes.
1.1.3 Environmental levels and exposures
The natural background concentration of nitrogen dioxide over land
areas is usually in the range of 0.4-9.4 µg/m3 (0.0002-0.005 ppm).
This concentration is 1-2 orders of magnitude lower than the
concentrations normally found in urban areas. Annual mean nitrogen
dioxide concentrations in urban areas throughout the world are
typically in the range of 20 90 µg/m3 (0.01-0.05 ppm), although it
is exceedingly difficult to generalize.
Data for shorter averaging periods show considerable variations
depending on meteorological and seasonal conditions and on the
proximity and nature of local sources of pollution. Generally, the
highest monthly means of nitrogen dioxide levels in large urban areas
are about 60-110 µg/m3 (0.03 0.06 ppm), the highest daily means
136-400 µg/m3 (0.07-0.22 ppm), and the highest hourly values
240-850 µg/m3 (0.13-0.45 ppm).
In contrast with typical primary air pollutants, nitrogen dioxide
concentrations do not show consistent seasonal behaviour throughout
all urban areas of the world and are not necessarily highest during
the months of maximum photochemical activity.
Exposure from indoor sources such as home appliances and smoking
should not be underestimated. In the immediate proximity of domestic
gas-fired appliances, nitrogen dioxide concentrations of up to
2000 µg/m3 (1.1 ppm) have been measured. Tobacco smoke has been
reported to contain nitric oxide levels of about 98-135 mg/m3
(80-110 ppm) and nitrogen dioxide levels of about 150-226 mg/m3
(80-120 ppm), but these levels may fluctuate considerably with the
conditions of combustion.
1.1.4 Effects on experimental animals
Reversible and irreversible adverse effects may be caused by
exposure to nitrogen dioxide, depending upon the concentration,
length, and mode of exposure, the species of animal tested, and the
presence of infectious agents.
Morphological changes reported in a number of animal species
including the mouse, rat, rabbit, guineapig, and monkey, appeared to
be most prominent in the terminal bronchiolar and alveolar duct
epithelia. Exposure to about 470-1900 µg/m3 (0.25-1.0 ppm) resulted
in numerous pathophysiological changes including bronchitis,
bronchopneumonia, atelectasis, protein leakage into the alveolar
space, changes in collagen, elastin, and mast cells of the lungs,
reduction or loss of cilia and adenomatous changes.
At concentrations of 3800-47 000 µg/m3 (2.0-25 ppm) these
effects became more pronounced. The more sensitive ciliated
bronchiolar and type 1 alveolar lining cells were injured first and
were replaced by the proliferation of more resistant nonciliated
cells, and type 2 cells, respectively. Prolonged exposure resulted in
a reduction in diameter of small airways by exudate, hypertrophy of
the respiratory epithelium, and swelling of the basement membrane.
In studies on the effect of nitrogen dioxide on lung function,
increased respiratory rates were reported in rats exposed to
concentrations as low as 1500 µg/m3 (0.8 ppm). Reductions in both
diffusion capacity and peak expiratory flow rates were demonstrated in
beagles exposed to a combination of nitrogen dioxide at
1210 µg/m3 (0.64 ppm) and nitric oxide at 310 µg/m3 (0.25 ppm).
Biochemical changes included alterations in the action of several
pulmonary enzymes, in the lipid content of the lungs, in the stability
of pulmonary surfactant, and a decrease in the lung glutathione
levels. As the nitrogen dioxide concentration increased to
11-75 mg/m3 (6-40 ppm), the effects became more pronounced.
A number of extrapulmonary effects have been reported at nitrogen
dioxide concentrations of 560-3700 µg/m3 (0.3 to 2.0 ppm).
Examination of blood from exposed animals showed changes in the number
of circulating erythrocytes, in enzyme activity, and in antibody
titres. Within the range of these concentrations, effects were also
noted on the conditioned reflexes of the central nervous system
(600 µg/m3, 0.32 ppm) and on the endocrine and reproductive systems
(2400 µg/m3, 1.3 ppm) of rats.
With increasing levels of exposure, a variety of other effects
were demonstrated. These included decrease in growth rate
(5000 µg/m3, 2.7 ppm) and loss in physical performance
(9400 µg/m3, 5.0 ppm).
Exposure to nitrogen dioxide increased the susceptibility of
experimental animals to both bacterial and viral respiratory
infections; this response was clearly dose-related. Results indicated
that concentration had a much greater influence on the toxicity of
nitrogen dioxide than length of exposure, i.e. equal
concentration-time products at different exposure times were not
equally hazardous.
When mice were exposed to 940 µg/m3 (0.5 ppm) for 90 days and
then artificially infected, a significant increase in the mortality
rate was observed. A similar response was noted in squirrel
monkeys-exposed to much higher concentrations of 9400-19 000 µg/m3
(5-10 ppm) for 1 or 2 months. When mortality rates due to respiratory
infection were compared after continuous and intermittent exposure to
nitrogen dioxide, there was a significant increase in both treatments
with increasing length of exposure. However, for each given length of
exposure, there was no statistical difference between the continuous
and the intermittent exposure groups.
Nitrogen dioxide interferes with the lung's ability to remove
inhaled deposited particles efficiently by altering the phagocytic,
enzymatic, and functional processes of the alveolar macrophages and of
the ciliated epithelial cells.
1.1.5 Effects on man
1.1.5.1 Controlled exposures
Studies on exposure to nitrogen dioxide in man have been conducted
to determine the lowest levels at which odour can be detected and at
which dark adaptation is altered. For odour perception the lowest
nitrogen dioxide level was approximately 200 µg/m3 (0.11 ppm). The
lowest level for impairment of dark adaptation was reported to be
140 µg/m3 (0.074 ppm).
Exposure to nitrogen dioxide levels of 1300-3800 µg/m3
(0.7-2.0 ppm) for 10 min gave rise to an increase in inspiratory and
expiratory flow resistance. In another study, inhalation of nitrogen
dioxide concentrations of 3000-3800 µg/m3 (1.6-2.0 ppm) for 15 min
caused a significant increase in total airway resistance, which became
more pronounced at concentrations above 3800 µg/m3 (2.0 ppm). A
number of authors have reported that exposure to 7500-9400 µg/m3
(4-5 ppm) produced an increase in airway resistance and a decrease in
the arterial partial pressure of oxygen and carbon monoxide diffusion
capacity. However, prolongation of the exposure time to 60 min did not
enhance the effect further. A recent report showed that in 13 out of
20 asthmatic subjects, the reaction to the inhalation challenge with a
bronchoconstrictor (carbachol) increased significantly after exposure
to a nitrogen dioxide level of 190 µg/m3 (0.1 ppm) for 1h. Similar
results were reported in a study in which healthy subjects were
exposed to a combination of nitrogen dioxide at 100 µg/m3
(0.05 ppm), ozone at 50 µg/m3 (0.025 ppm), and sulfur dioxide at
260 µg/m3 (0.10 ppm) for a period of 2h.
1.1.5.2 Accidental and industrial exposures
Exposure to high concentrations of oxides of nitrogen has been
reported in various occupations.
Farmers who were exposed to silo gases from the fermentation of
harvested crops were acutely affected by oxides of nitrogen, some of
them fatally. It has been estimated that exposure to nitrogen dioxide
levels of 560-940 mg/m3 (300-500 ppm) may result in fatal pulmonary
oedema or asphyxia and that levels of 47-140 mg/m3 (25-75 ppm) can
cause bronchitis or pneumonia.
Miners who used explosives repeatedly in their work were reported
to develop chronic respiratory diseases. Analysis of the products of
explosion showed the presence of oxides of nitrogen at concentrations
of 88-167 ppm.
A study on surviving victims who had been exposed to the fumes of
burning nitrocellulose did not reveal any differences in survival
between the exposed groups and the unexposed controls over the
following 30 years. Unfortunately quantitative exposure data for the
various groups were not available in this study.
There are very few studies on the acute or chronic effects of
low-level industrial exposures.
1.1.5.3 Community exposures
Several studies have been reported in which an attempt has been
made to relate pulmonary function to nitrogen dioxide exposure.
However, the results of all these studies have either failed to
demonstrate a significant difference in lung function between the
groups exposed to different levels of nitrogen dioxide, or have been
confounded by the fact that relatively high concentrations of other
pollutants were present.
This also applies to studies conducted to correlate the frequency
of acute respiratory disease and chronic respiratory illness with
concentrations of nitrogen dioxide.
For example, a study to evaluate the effects of nitrogen dioxide
on the incidence of acute respiratory disease in children and their
parents living near a large point source of this pollutant
demonstrated an excess rate of respiratory illness in comparison with
a control group. However, the probable contribution of other
pollutants such as sulfuric acid aerosols, nitric acid fumes, and
suspended nitrates, made it difficult to attribute this excess to the
presence of nitrogen dioxide.
Similarly, because of relatively high exposure to other air
pollutants, it has not been possible to associate observed increases
in the frequency of chronic respiratory illness with a measured level
of nitrogen dioxide. It has been noted, however, that these
epidemiological studies seem to confirm the results of controlled
studies on man and experimental animal studies.
1.1.6 Evaluation of health risks
As it has not yet been shown that the concentrations in ambient
air of oxides of nitrogen, other than nitrogen dioxide, have any
significant biological activity, a guideline for the protection of
public health has been developed only for nitrogen dioxide.
Compared with experimental toxicological studies, there are very
few epidemiological studies on the effects of either occupational or
community exposures which can provide sufficient information for the
assessment of health risks due to exposure to nitrogen dioxide. Thus,
a health protection guideline has been developed based on data from
controlled human studies and animal experiments. As previously stated,
available epidemiological data tend to support these results.
A nitrogen dioxide concentration of 940 µg/m3 (0.5 ppm) has been
selected as an estimate of the lowest level at which adverse health
effects due to short-term exposure to nitrogen dioxide can be expected
to occur. Although the Task Group was aware that one study in man
showed effects at a lower concentration, it was of the opinion that
this required confirmation.
By adopting a minimum safety factor of 3-5, the Task Group agreed
that a maximum one hour exposure of 190-320 µg/m3 (0.10-0.17 ppm)
should be consistent with the protection of public health and that
this exposure should not be exceeded more than once per month.
A caution has been added that it might be prudent to lower this
exposure limit in view of biological evidence of the interaction of
nitrogen dioxide with other air pollutants present and also in view of
the fact that some populations are highly sensitive to this substance.
Owing to lack of information on the effects of long-term exposure to
nitrogen dioxide in man, only a short-term exposure limit has been
suggested.
1.2 Recommendations for Further Research
In discussing the health risks of nitrogen dioxide exposure, the
Task Group concentrated on the biological activity of nitrogen dioxide
alone, rather than in conjunction with other compounds with which it
is commonly associated in the ambient atmosphere. However, the Task
Group was particularly concerned with the potential for enhanced
biological effects in ambient situations in which peak concentrations
of nitrogen dioxide and photochemical oxidants occur together. The
Task Group also expressed concern over atmospheric oxidation products
of nitrogen dioxide, such as nitrous and nitric acid and various
nitrate compounds. Taking into consideration the biological data on
the combined effects of nitrogen dioxide and oxidants, and of the
nitrates in the ambient air, the Task Group made the following
recommendations for research on the health effects of these
substances:
a) Controlled studies on man and experimental studies on animals
should be conducted to compare the reaction of sensitive
biological systems at typical peak concentrations of nitrogen
dioxide and ozone alone, and in combination. In animals, the
infectious disease model appears to be particularly appropriate
for these studies. In man, the effects should be studied of
nitrogen dioxide and oxidants, alone or in combination, on airway
resistance before and after administration of bronchoconstrictors.
b) Similar experimental animal and controlled human studies should be
conducted to evaluate the biological effects of nitric acid and
nitrates at concentrations found in ambient air.
c) The possibility of delayed effects from exposure to nitrogen
dioxide and its oxidation products should be considered. These
possibilities may be pursued by means of epidemiological studies
and recently developed experimental techniques to assess
carcinogenicity and mutagenicity.
d) While the Task Group is aware that epidemiological studies alone
cannot provide a quantitative basis for evaluating the health
risks of exposure to nitrogen dioxide, the importance of
epidemiological studies of occupational and community groups
should not be minimized. There is a particular need for long-term
follow-up studies which may identify chronic or delayed and often
subtle effects in cohorts of exposed populations.
e) Studies of highly sensitive subjects should be given careful
consideration. Asthmatic subjects and persons with
cardio-pulmonary disease should be studied with respect to
functional and symptomatic changes associated with variations in
the average hourly concentrations of nitrogen dioxide, nitrates,
and related compounds. Controlled exposure of asthmatic subjects
and other highly sensitive persons to nitrogen dioxide and
nitrates at concentrations found in the ambient air may be
undertaken, with their consent and with due consideration for the
protection of subjects so exposed. It would be highly desirable to
study animal models of human asthma and hypersensitivity.
2. CHEMISTRY AND ANALYTICAL METHODS
2.1 Chemical and Physical Properties
Oxides of nitrogen are usually classified in terms of the
oxidation state of nitrogen (Table 1).
Table 1. Oxides of nitrogen
Name Chemical formula
nitrous oxide N2O
nitric oxide NO
dinitrogen trioxide N2O3
nitrogen dioxide NO2
dinitrogen tetroxide N2O4
dinitrogen pentoxide N2O5
Nitrous oxide (dinitrogen oxide) is the most prevalent oxide of
nitrogen in the atmosphere. This compound is generated by anaerobic
processes in the soil and in the surface layers of the oceans and is
present in the atmosphere in concentrations of about 450 µg/m3
(0.25 ppm) (Robinson & Robbins, 1972). Although this species may play
an important role in stratospheric chemistry, it is of little
importance in the lower atmosphere and has no direct significance for
human health.
Nitric oxide (nitrogen oxide) and nitrogen dioxide, the most
abundant man-made oxides of nitrogen in urban areas, are derived from
air used in high temperature combustion processes. Both nitric oxide
and nitrogen dioxide are found in combustion gases but nitric oxide
predominates because its formation is favoured by high temperatures.
The formation of nitric oxide can be described by the following
reactions (Spedding, 1974):
O + N2 <=> NO + N(1)
O2 + N <=> NO + O(2)
Atomic oxygen needed in reaction (1) is produced in the flame by 2
parallel reactions:
CO + OH <=> CO2 + H(3)
H + O2 <=> OH + O(4)
The amount of nitric oxide formed depends on the temperature of
the flame, the concentrations of nitrogen and oxygen, and the
residence time of gases in different zones of temperature, pressure,
and concentration. Temperature is the most significant variable in the
production of nitric oxide under normal combustion conditions
(McKinnon, 1974). The production of nitric oxide per unit mass of fuel
burned decreases with decreasing mean combustion temperature of
different fuels, i.e.: coal, oil, natural gas. Because the internal
combustion engine operates at a high temperature, motor vehicles are
an important source of nitric oxide.
Some physical properties of nitric oxide and nitrogen dioxide are
given in Table 2. Nitric oxide is a colourless, odourless gas that is
slightly soluble in water. Although the boiling point of nitrogen
dioxide is 21.2°C, it only exists in the gaseous form at normal air
temperatures because of its low partial pressure.
Table 2. Physical properties of nitric oxide and nitrogen
dioxide a
Oxides of Molecular Melting Boiling Solubility
nitrogen weight point point in water
°C °C ml/litre
nitric
oxide 30.01 -- 163.6 -- 151.8 73.40
nitrogen
dioxide 46.01 -- 11.20 21.2 --
a From: West (1976)
Nitrogen dioxide is in equilibrium with its dimer, dinitrogen
tetroxide (2NO2 <=> N2O4) but at atmospheric concentrations
the fraction of nitrogen dioxide present in dimer form is negligible.
Dinitrogen trioxide can be formed from nitric oxide and/or
nitrogen dioxide. However, at the low concentrations of nitric oxide
and nitrogen dioxide found even in very heavily polluted air, the
chemical equilibrium data predict negligible concentrations of
dinitrogen trioxide which thus has no significance as an air pollutant
(Leighton, 1961).
Dinitrogen pentoxide is thought to be an important reactive
intermediate in photochemical air polution, formed mainly by the
oxidation of nitrogen dioxide with ozone (Demerjian et al., 1974).
However, there are no specific analytical methods for the measurement
of this species in ambient air and there is no evidence that it has
any significance for human health.
2.2 Atmospheric Chemistry
The atmospheric chemistry of the oxides of nitrogen is very
complex, particularly when other air pollutants such as hydrocarbons
are present. For this reason, only a simplified account can be
presented here.
Nitric oxide is fairly reactive and is readily oxidized in the
atmosphere to nitrogen dioxide. The conversion takes place by means of
several reactions depending on the concentrations of nitric oxide. At
high concentrations, as much as 10% of nitric oxide can be oxidized by
the reaction:
2NO + O2 -> 2NO2(5)
The rate of this reaction decreases with dilution and rapidly
becomes insignificant. At low concentrations, an important reaction
leading to nitrogen dioxide formation is:
NO + O3 -> NO2 + O2(6)
Nitrogen dioxide absorbs strongly in the ultraviolet region
between 300 and 400 µm and is decomposed by sunlight yielding nitric
oxide and ozone (Leighton, 1961). Thus, in daylight, reaction (6)
proceeds in the opposite direction and eventually an equilibrium
NO + O3 <=> NO2 + O2(7)
is established.
The position of equilibrium (7) is a function of the rate of
reaction (6) and the rate of light absorption by nitrogen dioxide
which varies with the time of the day, latitude, and other atmospheric
variables (Calvert, 1976). Generally, however, in unpolluted and rural
areas, the daytime concentrations of nitric oxide are only a small
fraction of the concentration of nitrogen dioxide.
The air is more polluted in urban areas than in rural areas and
the concentration of oxides of nitrogen is markedly higher. Thus, it
is possible, particularly at night time, that reaction (6) proceeds to
completion and that all ozone is removed leaving substantial
concentrations of both nitric oxide and nitrogen dioxide in the
atmosphere. During the day, the equilibrium (7) shifts in favour of
ozone formation.
Thus, in polluted air, the position of equilibrium (7) and the
resulting nitric oxide, nitrogen dioxide, and ozone concentrations
depend on a large number of meteorological and other factors, and
particularly on the simultaneous presence of hydrocarbon pollutants.
The pattern of nitrogen dioxide concentrations in urban air is
therefore quite different from that of primary pollutants such as
nitric oxide, carbon monoxide, and sulfur dioxide. It is much more
similar to that of typical secondary pollutants such as ozone and
photochemical aerosol species. The differences in behaviour between
nitrogen dioxide and other primary pollutants show in the relationship
between peak values and long-term mean values, and in diurnal and
seasonal variability. These aspects are illustrated in section 4.
The main atmospheric sink for oxides of nitrogen appears to
involve its oxidation to nitric acid. This is an example of a general
reaction in atmospheric chemistry where pollutants are oxidized to
species which are more readily removed from the atmospheric
circulation. This is particularly true for the oxides of nitrogen
since nitric acid is much more soluble in water and much more readily
adsorbed on the surface of suspended particulate matter.
In view of the possible effects on human health of nitrate
particles, this atmospheric conversion may have an added significance.
The mechanism of the conversion most probably involves hydroxyl
radicals as shown by the equation (8)
OH + NO2 <=> HNO3(8)
2.3 Analytical Methods
2.3.1 Sampling
Although nitric oxide and nitrogen dioxide are chemically reactive
they behave quite predictably in glass and teflon sampling moulds.
Residence time in the sampling manifold requires specific
consideration, when sampling air containing nitric oxide, nitrogen
dioxide, and ozone during daylight. Since the equilibrium (7) is
disturbed inside the dark sampling manifold, nitrogen dioxide
concentrations may be overestimated when sampling takes much over
10 s (Butcher & Ruff, 1971).
Collectors based on solid adsorbents that have been developed for
the selective sampling of nitric oxide and nitrogen dioxide have great
potential because of their stability and simplicity and because a wide
selection of methods can be used for subsequent analysis. Nitrogen
dioxide can be quantitatively absorbed on columns packed with an inert
material coated with triethanolamine without affecting the nitric
oxide concentrations (Levaggi et al., 1972). The nitric oxide is
subsequently adsorbed on a second column treated with cobalt (II)
oxide.
2.3.2 Evaluation of analytical methods
Detailed description of the various methods has been omitted since
they are discussed elsewhere (US Department of Health, Education and
Welfare, 1965; US Environmental Protection Agency, 1971a; World Health
Organization, 1976). Instead, a critical evaluation is given of the
important methods used for measurements in the ambient air and in the
health-effects studies discussed in subsequent sections.
Nitric oxide and nitrogen dioxide may be measured separately or
collectively by manual or automated techniques.
The colorimetric manual methods are based on a specific reaction
in which nitrite ions and diazotizing reagents produce a deeply
coloured azo-dye (Mulik et al., 1974; Saltzman, 1954; US Environmental
Protection Agency, 1971b). These methods can be automated to give mean
concentrations over averaging periods of 15-60 min (Japanese Standards
Association, 1974; Lyshkow, 1965: Saltzman, 1960).
The chemiluminescence techniques are automatic and specific and
have revolutionized the measurement of oxides of nitrogen. The method
is based on the measurement of red light produced by the reaction
O3 + NO -> O2 + NO2 + light. The major features are selective
response to nitric oxide, sensitivity into the 1 µg/m3 range,
linearity over a factor of 105 in concentration, and rapid (<1 s)
time response (Fontijn et al., 1970; Stedman et al., 1972). High cost,
complexity, and the requirement of some form of data logging system if
long-term mean values are required for averaging periods from 1 day to
1 year are some drawbacks of these techniques. They are ideally suited
for the measurement of peak concentrations over averaging periods of
from 15 s to 1 h which are largely inaccessible with manual methods.
Besides the two basic methods there are a large number of other
methods for the measurement of nitric oxide and nitrogen dioxide. Gas
chromatography, long-path infrared spectroscopy, and electrochemistry
have been used but either they are generally cumbersome or they have
comparatively low detection limits for use in atmospheric measurements
(World Health Organization, 1976).
2.3.2.1 Manual methods
The most widely used manual methods for nitrogen dioxide are the
Saltzman method and the Jacobs-Hochheiser method; detailed analytical
procedures are described elsewhere (World Health Organization, 1976).
Nitric oxide can also be measured by these methods if oxidized to
nitrogen dioxide prior to analysis. Although various solid and liquid
oxidizing agents have been proposed for the conversion of nitric oxide
to nitrogen dioxide, they are not very reliable and tend to
underestimate ambient nitric oxide concentrations (World Health
Organization, 1976).
Colorimetric procedures are complicated because of the time
required for the colour to develop. This colour is not permanent and
the Saitzman procedure, in particular, is not therefore recommended
where the samples cannot be analysed after a short time delay. The
Jacobs-Hochheiser method is a useful modification of the diazotization
method, applicable to 24 h samples, which can be analysed up to 1
month after collection (Jacobs & Hochheiser, 1958). However, this
method has several deficiencies including a variable collection
efficiency for nitrogen dioxide (Hauser & Shy, 1972).
Various modifications of the Jacobs-Hochheiser method are being
evaluated. One of these, the arsenite method (Christie et al., 1970)
suffers from serious interference by nitric oxide (Merryman et al.,
1973). The TGS-ANSAa method (Mulik et al., 1974) appears to
eliminate all the deficiencies of the Jacobs-Hochheiser method.
The Saltzman procedure has been used extensively in Europe, Japan,
and the USA and has been tested by many workers. The method requires
simple and inexpensive apparatus and its detection limit is adequate
for most pollution studies. However, in view of the problems
associated with the oxidation of nitric oxide to nitrogen dioxide, it
is only suitable for measuring nitrogen dioxide. If nitrogen dioxide
has to be monitored over periods of more than 2 h, or if the presence
of relatively high concentrations of other oxidizing or reducing
agents is suspected, a series of short-term samples (15-30 min each)
should be collected and analysed as soon as possible. This may require
some form of automatic sampling scheme.
2.3.2.2 Automatic methods
Continuous analysers based on the Saltzman procedure have been
used extensively, but this should not be encouraged, since they are
quite complicated, requiring excessive operator attention (World
Health Organization, 1976). In addition, certain commercial versions
suffer from interference by ozone (Baumgardner et al., 1975).
Furthermore, since none of the proposed oxidizing agents for the
nitric oxide-nitrogen dioxide conversion is wholly satisfactory under
field conditions, continuous analysers based on the Saitzman procedure
are suitable only for the measurement of nitrogen dioxide
concentrations.
a TGS -- absorbing solution consisting of 20 g triethanolamine +
0.5 g guaicol + 0.25 g sodium metabisulfite/litre of distilled water.
ANSA -- reagent consisting of 0.1% 8-anilino-1-naphthalenesulfonic
acid in absolute methanol.
Chemiluminescence methods are ideally suited to the measurement of
nitric oxide concentrations and are accurate and reproducible over a
wide range of concentrations. There are no important sources of
interference.
Most commercial chemiluminescence analysers for nitric oxide are
also equipped with some form of converter which reduces the nitrogen
dioxide to nitric oxide before reaction with ozone to yield a combined
measurement of nitric oxide and nitrogen dioxide. Considerable
problems may occur in the mechanics of the subtraction of the nitric
oxide signal from the combined nitric oxide-nitrogen dioxide signal
when the nitric oxide concentrations are much higher than the nitrogen
dioxide concentrations.
Although the chemiluminescence determination of nitric oxide is
interference-free, this is not always the case with the measurement of
nitrogen dioxide. The conversion of atmospheric ammonia to nitric
oxide can be eliminated by the appropriate choice of converter
material and operating temperature. Ammonia derived from animal waste
products may interfere with the determination of nitrogen dioxide
exposures in the animal experiments described later. Certain nitrogen
species such as nitric acid and peroxyacetylnitrate (PAN) decompose in
most commercial thermal converters (Winer et al., 1974). This
contributes minor interference during photochemical air pollution
episodes.
2.3.3 Calibration procedures
There are 3 independent procedures for calibrating methods for
measuring oxides of nitrogen. One technique involves the use of
permeation tubes for nitrogen dioxide (Lindqvist & Lanting, 1972).
Another technique is based on the gas-phase titration of nitric oxide
with ozone which provides simultaneous calibration for nitric oxide,
nitrogen dioxide, and ozone using the reaction:
NO + O3 -> NO2 + O2(6)
Finally, dynamic dilution can be used to prepare flowing mixtures
of nitric oxide and nitrogen dioxide in air for calibration purposes
(Japanese Standards Association, 1976).
There has been much discussion in the literature concerning the
"Saltzman factor" i.e. the conversion factor for sodium nitrite to
nitrogen dioxide. This problem arises when the method is calibrated
against standard solutions of nitrite ions, and can be obviated by the
calibration methods discussed above. However, it must be borne in mind
when interpreting literature data where this form of calibration has
been used and a value assumed for the "Saltzman factor". This factor
is usually about 0.72 (Forweg, 1975), but may vary with experimental
conditions and concentration.
3. SOURCES OF OXIDES OF NITROGEN
3.1 Natural Sources
Nitric oxide and nitrogen dioxide present in the air are produced
by natural processes including lightning, volcanic eruptions, and
bacterial action in the soil, as well as by man-made activities. It
has been estimated that the annual, natural global emissions of these
oxides of nitrogen are of the order of 1100 million tonnes (Robinson &
Robbins, 1972). This by far surpasses emissions of oxides of nitrogen
generated by man-made activities which were estimated in 1970 to be
approximately 53 million tonnes. However, since natural emissions are
distributed over the entire globe, the resulting air concentrations
are practically negligible.
3.2 Man-made Sources
The major source of man-made emissions of oxides of nitrogen is
the combustion of fossil fuels. The predominant oxide of nitrogen
emitted by combustion processes is nitric oxide; nitrogen dioxide is
produced in much smaller amounts. The observed percentage of nitric
oxide in the total emission of oxides of nitrogen is 90-95% by volume
although it depends on a number of factors and varies substantially
from one source to another.
The distribution of emissions from different sources, in selected
countries is shown in Table 3. Because emissions of oxides of nitrogen
are extremely variable, these estimates provide only a general guide
on the nature and magnitude of the more important sources. Generally,
the differences between the various countries illustrated in Table 3
can be readily accounted for by differences in fuel use. For example,
in the Netherlands a significant fraction of the electricity
generating plants use natural gas whereas this use of natural gas is
negligible in the UK. This greater reliance on natural gas in the
Netherlands may account for the much smaller relative contribution
from stationary sources. Table 3 also indicates that transportation
sources are relatively more significant in Japan and the USA than in
the Netherlands or the UK.
Projection of emission data into the future must be treated with
some caution. It is evident, however, that in the absence of any
abatement strategies, oxides of nitrogen emissions in most urban areas
will increase steadily within the next decade (Organization for
Economic Cooperation and Development, 1973). Prior to the energy
crisis in the 1970s, oxides of nitrogen emissions had been expected to
about double between 1968 and 1980. Such projections may require
reevaluation in the light of changing patterns of fuel use.
Table 3. Emissions of oxides of nitrogen from various sources in
selected countries expressed as 106 tonnes per year
Japana Netherlandsb UKc USAd
Source
1972 1972 1970 1970
Transportation 0.96 0.13 0.46 11.7
Fuel combustion in )
stationary sources ) 0.19 1.98 10.0
Non-combustion ) 1.44
industrial processes ) -- -- 0.2
Miscellaneous -- -- -- 0.9
Total 2.40 0.32 2.43 22.8
From: a Central Council for Environmental Pollution Control
(1977).
b National Air Pollution Council, the Netherlands (1976).
c Derwent & Stewart (1973).
d US Environmental Protection Agency (1973).
3.2.1 Stationary sources
As shown in Table 3, stationary combustion sources in Japan, the
Netherlands and the UK account for 60, 59, and 82% of total emissions,
respectively. In the USA this figure is about 44%. These emissions
include a substantial contribution from power generating plants. In
the UK and the USA, these large power plants contribute 52% and 21%,
respectively, of the total emissions (Derwent & Stewart, 1973; Mason
et al., unpublished dataa).
The combustion of fuel in the home makes only a minor contribution
to total emissions of oxides of nitrogen. In the UK and the USA
domestic fuel use accounts for only 5% and 6% of total emissions
respectively.
Fuel combustion by the commercial and industrial sectors provides
a substantial source of oxides of nitrogen emissions in certain urban
areas, particularly through space heating during the winter season.
a Paper presented at the Sixty-second Annual Meeting, Air Pollution
Control Association, June 1969, Paper No. 96-101, p. 19.
3.2.2 Mobile sources
Transportation sources include personal motor vehicles, buses,
trucks, railroad vehicles, aircraft, and ships on inland waterways. Of
these many categories, petrol-powered motor vehicles provide by far
the largest contribution to total emissions. As a whole,
transportation sources make a substantial contribution to the total
emissions of the countries listed in Table 3. For Japan, the
Netherlands, UK, and USA, transportation sources account for 40, 41,
18, and 51% respectively, of total emissions.
3.2.3 Non-combustion sources
Although the total emissions from industrial processes (other than
from fuel combustion) are relatively small, certain processes are
significant local sources of oxides of nitrogen. Examples of these
non-combustion sources include the manufacture of nitric acid,
electroplating, and processes involving concentrated nitric acid such
as the manufacture of explosives and the manufacture of sulfuric acid
by the chamber process. The manufacture of nitric acid is usually the
most significant of these non-combustion sources (Bagg, 1971).
Bacterial degradation of silage material can be a significant
source of oxides of nitrogen and has led to certain occupational
hazards which are mentioned in section 6.
3.2.4 Other sources
Exposure to oxides of nitrogen from home appliances such as gas
stoves and from tobacco smoking should not be underestimated. Exposure
levels due to these sources are discussed in section 4.
4. ENVIRONMENTAL LEVELS AND EXPOSURES
4.1 Background Concentrations
Available data indicate that natural background levels of nitrogen
dioxide over land areas range from 0.4 to 9.4 µg/m3
(0.0002-0.005 ppm) and those of nitric oxide from 0 to 7.4 µg/m3
(0-0.006 ppm) (Robinson & Robbins, 1972). In Panama, for example,
Lodge & Pate (1966) found average nitrogen dioxide values ranging from
1.7 µg/m3 (0.0009 ppm) during the dry season to 6.8 µg/m3
(0.0036 ppm) in the rainy season. The natural background level of
nitrogen dioxide in remote areas of Western Europe ranged from below
2.0 to 4.2 µg/m3 (below 0.0011 to 0.0022 ppm) (Georgii & Weber,
1962). These concentrations are 1-2 orders of magnitude lower than
those typically found in urban areas.
4.2 Urban Concentrations
Urban concentrations of nitric oxide, nitrogen dioxide, and oxides
of nitrogen have been measured in a number of countries in recent
years. The results have usually been reported as 1 h, 24 h, or annual
averages. Annual average nitric oxide levels in large cities have been
reported to range from 49 to 95 µg/m3 (0.040-0.077 ppm) (Environment
Agency, 1974: US Environmental Protection Agency, 1971a). An
indication of long term average concentrations of nitrogen dioxide can
be obtained from Table 4 where the annual mean concentrations are
shown for selected urban areas. It is important to recognize that
these concentrations did not necessarily represent the maximum
exposure levels in these cities, since nitrogen dioxide concentrations
vary greatly within a given urban area. It should also be remembered
that different measurement methods were used in different countries
and that these might have changed during the years tabulated.
Thus, additional information and interpretation are required before
comparisons between cities or determination of trends can be made.
Table 4. Annual mean concentrations of nitrogen dioxide in selected
cities (µg/m3)a
Washington, Frankfurt/
Year Rotterdam DC Main Tokyo
1962 56 19
1963 56 23
1964 75 28
1965 56 30
1966 35 47
1967 43 75 34
1968 43 41
1969 43 63 77
1970 45 94 82 73
1971 75 80 58
aFrom: Commissie Bodem, Water en Lucht, 1970; Environment Agency,
1976; Jost & Rudolph, 1975; US Environmental Protection Agency
1962-1971.
Tables 5 and 6 illustrate the observed short-term mean
concentrations of nitrogen dioxide. In addition to annual means, Table
5 presents the maximum 1-h, 24-h, and monthly mean concentrations of
nitrogen dioxide recorded at selected sites in 5 cities in Japan
(Environment Agency, 1974). Data on maximum 24-h concentrations and
annual means for 5 US cities are given in Table 6 (US Environmental
Protection Agency, 1976a). The maximum 24-h mean concentrations of
nitrogen dioxide were generally within the range of 100-400 µg/m3
(0.054).22 ppm) and the maximum 1-h concentrations over 800 µg/m3
(0.43 ppm). The maximum 24-h mean value refers to the day of the year
with the highest mean concentration. The maximum 1-h value refers to
the highest 1-h value in the year and the maximum monthly to the
highest monthly mean in the year.
Table 5. Nitrogen dioxide concentrations (µg/m3) recorded in
selected cities in Japan during 1973 using the Saltzman methoda
Cityb Annual Maximum Maximum Maximum
mean 1 -h value 24-h mean monthly mean
Sendai 46 240 134 60
Tokyo 86 840 426 105
Kawasaki 90 440 200 113
Osaka 86 640 228 115
Matsue 10 60 22 14
a From: Environment Agency, 1974.
b Densely populated, except Matsue.
Table 6. Annual mean and maximum 24-h nitrogen dioxide
concentrations recorded in selected cities in the USA during
1974 using the chemiluminescence methoda
City Annual mean Maximum 24-h mean
San Jose 67 285
Philadelphia 73 166
Washington, DC 68 130
New York 80 243
Chicago 47 114
a From: US Environmental Protection Agency (1976a)
Since most air pollutant concentrations are approximately
log-normally distributed, a fairly consistent relationship can be
established between annual averages and the averages calculated for
shorter averaging periods (Larsen, 1969). For nitrogen dioxide, the
maximum 24-h mean is about 2 5 times higher than the annual mean at a
given site. The relationship of the maximum 1-h value to the annual
mean is not as consistent. Available data show that the hourly maximum
value is approximately 5-10 times the annual mean. This relationship
does not hold for averaging periods of less than 1 h or for unusual
situations. The model also appears to overpredict maximum monthly mean
nitrogen dioxide concentrations. Examples of the seasonal variation in
nitric oxide and nitrogen dioxide concentrations for selected sites in
the USA and Japan are shown in Fig. 1 and 2, respectively. These
variations are caused by meteorological factors and to a lesser degree
by seasonal changes in emission rates. Ambient temperature, wind
speed, and inversion height are important factors affecting the
dilution of air pollutants. In addition, variations in nitrogen
dioxide production by the atmospheric chemical reactions discussed in
section 2 play a substantial role in the seasonal changes observed. In
the cities cited, mean winter nitrogen dioxide values were 2-3 times
higher than summer concentrations. Considering the complexity of the
factors involved, this observation is probably not universal and other
sites may show the opposite trends.
The frequency of occurrence of nitrogen dioxide hourly maximum
concentrations in an urban area in the USA is shown in Table 7
(California Air Resources Board, 1975). During the period cited, the
hourly maximum concentration of nitrogen dioxide exceeded
200 µg/m3 (0.11 ppm) on more than 50% of the days. On one day a 1-h
value of 839 µg/m3 (0.46 ppm) was observed which is similar to the
high 1-h value noted in Table 5 for Tokyo.
Table 7. Distribution of hourly maximum concentrations of nitrogen
dioxide during July-September 1975a
Number of days with hourly maximum
in nitrogen dioxide concentration range (µg/m3)
Site 200-400 400-600 600-800 > 800
Los Angeles 43 5 2 1
Azusa 51 2 0 0
Burbank 52 9 1 0
a From: California Air Resources Board, 1975.
An example of urban, diurnal, seasonal variations in nitric oxide
and nitrogen dioxide concentrations is given in Fig. 3 with reference
to Delft in the Netherlands (Guicherit, 1975).
Features of interest include 2 peak concentrations of both nitric
oxide and nitrogen dioxide found in the morning and evening which can
be ascribed to the influence of automotive sources and occur typically
on clear days. A time shift in the nitrogen dioxide peak is shown
during spring and summer indicating increased photochemical conversion
of nitric oxide into nitrogen dioxide.
Stationary sources involving fuel combustion for space heating can
also produce early morning peaks.
4.3 Indoor Exposure
Exposure to oxides of nitrogen in the home due to the use of
gas-fired appliances is usually underestimated. The recent expansion
in the use of natural gas may have increased this exposure.
Measurements conducted by Schwarzbach (1975) concerning nitrogen
dioxide formation by gas-fired domestic appliances such as space
heaters, boilers, and cookers showed concentrations of up to
2000 µg/m3 (1.1 ppm) at breathing height in the immediate vicinity
of cookers.
The concentrations of nitrogen dioxide measured in a normally
ventilated room using an oil-fired stove ranged from 380 to
1700 µg/m3 (0.2-0.9 ppm) depending on the type of stove and from 750
to 940 µg/m3 (0.4-0.5 ppm) when a gas-fired stove was used (Watanabe
et al., 1966). Occupational exposure is discussed in section 6.2.
4.4 Smoking
Special mention must be made of the intense, deliberate exposure
of man to oxides of nitrogen in tobacco smoke. Bokhoven & Niessen
(1961) reported that tobacco smoke contained nitric oxide and nitrogen
dioxide levels of 98-135 mg/m3 (80-110 ppm) and 150-226 mg/m3
(80-120 ppm)b, respectively. This is equivalent to a nitric oxide
intake of 160-500 µg per cigarette (Horton et al., 1974).
Haagen-Smit et al. (1959) made no distinction between nitric oxide
and nitrogen dioxide in reporting oxides of nitrogen levels of
145-655 ppm in tobacco smoke.
5. EFFECTS ON EXPERIMENTAL ANIMALS
A considerable amount of toxicological data is available relating
exposure to nitrogen dioxide with a variety of respiratory effects.
The purpose of this section is to review and summarize selected animal
studies which are most relevant for the evaluation of the health
hazards resulting from exposure to nitrogen dioxide.
Very few studies have been reported on the effects of nitric oxide
on experimental animals and, even in the most recent studies, the
concentrations used have been much higher than ambient air levels
(Greenbaum et al., 1967; Oda et al., 1975; Wagner, 1977, unpublished
dataa). Thus, the following discussion has been almost entirely
limited to studies on the effects of exposure to nitrogen dioxide.
5.1 Local Effects on the Respiratory System
5.1.1 Morphological changes
There are several reports that describe alterations in the
morphological integrity of the lung after exposure to nitrogen dioxide
concentrations of 1900 µg/m3 (1.0 ppm) and below. Salamberidze
(1969) did not find any pathological or histological changes in rats
exposed for 90 days to a nitrogen dioxide level of 100µg/m3
(0.05 ppm). However, electron microscopic studies by Buell (1970)
revealed damage to insoluble collagen fibres isolated from the lungs
of rabbits exposed to a nitrogen dioxide level of 470 µg/m3
(0.25 ppm) for 4 h/day, 5 days/week, for 24-36 days.
Jakimcuk & Celikanov (1968) reported that continuous exposure of
rats to a nitrogen dioxide concentration of 600 µg/m3 (0.32 ppm) for
90 days resulted in morphological changes such as peribronchitis,
bronchitis, and light pneumosclerosis. Similar studies with a nitrogen
dioxide concentration of 150 µg/m3 (0.08 ppm) did not produce
significant changes.
a Report on research work performed under the USA/Federal Republic of
Germany Cooperative Program in Natural Resources, Environmental
Pollution and Urban Development. Institut for Wasser-, Boden-, und
Lufthygiene des Bundesgesund-heitsamtes, 1977.
b According to G. Freeman, the values given for nitrogen dioxide are
too high. He considers levels of 19-95 mg/m3 (10-50 ppm) to be more
probable (Personal communication, 1977).
Inhalation of nitrogen dioxide concentrations of 1900 µg/m3
(1.0 ppm) for 1 h or 940 µg/m3 (0.5 ppm) for 4 h led to significant
morphological changes in the mast cells of the lung in rats (Thomas et
al., 1967). In exposed animals, the cells were ruptured and there was
evidence of loss of cytoplasmic granules. These changes, which were
observed in the pleura, bronchi, and surrounding tissue with more
marked effects around the mediastinum, were reversible in 24 h.
Sherwin & Carlson (1973) found a relative increase in protein
content in the lung lavage fluid of guineapigs continuously exposed to
a nitrogen dioxide level of 750 µg/m3 (0.4 ppm) for 1 week in
comparison with that of control animals. While the meaning of the
elevated protein levels is not yet clear, the authors believe that
both protein leakage from the capillary bed and the increased rate of
cell turnover within the exposed lung were responsible.
Blair et al. (1969) exposed mice to a nitrogen dioxide
concentration of 940 µg/m3 (0.5 ppm) for 6, 18 and 24 h daily and
studied the sequential alterations in lung morphology. After 3-12
months, the alveoli were expanded in all exposed mice. The authors
stated that the overall lesions appeared to be consistent with the
development of early focal emphysema. Inflammation of the bronchioles,
surface erosion of the epithelium, and blockage of the
bronchiolar-alveolar junction were also observed.
Continuous exposure of mice to nitrogen dioxide concentrations of
940-1500 µg/m3 (0.5-0.8 ppm) for 1 month produced numerous
structural changes. These effects included proliferation of the
epithelial cells in the mucous membrane; degeneration and ablation of
mucous membranes: oedematous changes in alveolar epithelial cells:
shortening of cilia; and influx of monocytes (Hattori et al., 1972;
Nakajima et al., 1969). Chen et al. (1972) studied recovery processes
after nitrogen dioxide exposure. Immediately following exposure to
nitrogen dioxide levels of 1900-2800 µg/m3 (1.0-1.5 ppm) for 1
month, the histological changes in exposed mice were identical to
those reported above. However, when the animals were allowed to
recover in clean air for 1-3 months there was a pronounced
infiltration of lymphocytes around the brochioles which was not found
in mice killed either during or immediately following exposure to
nitrogen dioxide. The authors suggested that this response resembled
those of an autoimmune disease.
Freeman et al., (1966) found slight bronchiolar, epithelial
hypertrophy and the development of a moderate degree of tachypnea in
rats continuously exposed for 33 months (approximately natural
life-time) to a nitrogen dioxide level of 1500 µg/m3 (0.8 ppm). The
authors repeated these long-term exposure studies at a concentration
of 3800 µg/m3 (2.0 ppm). (Freeman et al., 1968a, 1968b, 1969;
Freeman, 1970; Stephens et al., 1971a, 1971b, 1972). Exposure of rats
to this concentration of nitrogen dioxide resulted in a number of
microscopic and ultrastructural changes in the terminal bronchioles,
alveolar ducts, and alveoli. The lungs were about 10% heavier than
normal and the animals continued to exhibit tachypnea. There was
homogeneous and uniform hypertrophy of the bronchiolar epithelium,
loss of bronchiolar cilia, depression of natural cellular exfoliation,
and blebbing of bronchiolar cells. Intra-cytoplasmic, crystalloid
inclusion bodies appeared later. Electron microscopy revealed
thickening of lung collagen fibrils and of the alveolar basement
membranes.
Cell renewal rates were also studied in rats exposed to nitrogen
dioxide (Evans et al., 1972, 1973a, 1973b, 1975), by measuring the
uptake of tritiated thymidine by actively dividing alveolar cells.
Continuous exposure to 3800 µg/m3 (2.0 ppm) caused a marked increase
in number of type 2 alveolar cells. The labelling index reached a
maximum at 48 h and by the seventh day had returned to its normal
baseline level.
Monkeys (Macaca speciosa) exposed continuously for 14 months to
a nitrogen dioxide concentration of 3800 µg/m3 (2.0 ppm) developed
hypertrophy of the bronchiolar epithelium. Mixing an aerosol of sodium
chloride at 330 µg/m3 with the nitrogen dioxide did not appear to
alter the response (Furiosi et al., 1973). Several species of
laboratory animals were also exposed to nitrogen dioxide levels of
19 mg/m3 (10.0 ppm) or more in order to evaluate effects which could
possibly lead to chronic obstructive pulmonary disease. At nitrogen
dioxide levels of 19-47 mg/m3 (10-25 ppm) for 26 and 13 weeks
respectively, rats developed large, air-filled lungs that did not
collapse under atmospheric pressure. The lungs became grossly
emphysematous and the thoracic cage enlarged with dorsal kyphosis
(Freeman & Haydon 1964; Freeman et al., 1968a, 1968b, 1969).
Connective tissue changes involving both collagen and elastic tissue
were observed. Animals began to die of respiratory failure after 16
months.
For further information and for detailed descriptions of
morphological effects at these high concentrations the following
publications are suggested: Freeman & Haydon (1964); Kleinerman &
Cowdrey (1968); Kleinerman & Wright (1961); Parkinson & Stephens
(1973).
5.1.2 Functional changes
Both short-term and long-term exposure to concentrations of
nitrogen dioxide exceeding 1500 µg/m3 (0.8 ppm) have been reported
to cause changes in pulmonary function.
Rats exposed for 990 days to a nitrogen dioxide concentration of
1500 µg/m3 (0.8 ppm) maintained elevated respiratory rates
throughout their life (Freeman et al., 1966; Haydon et al., 1965).
Beagles exposed daily to a mixture of nitrogen dioxide and nitric
oxide at approximate levels of 1210 µg/m3 (0.64 ppm) and 310 µg/m3
(0.25 ppm) respectively, for 61 months demonstrated reductions in both
diffusion capacity and peak expiratory flow rates (Lewis et al.,
1974). However, exposure to mixtures of nitrogen dioxide at
940-1900 µg/m3 (0.5-1.0 ppm) and nitric oxide at 250 µg/m3
(0.2 ppm) for 16 h per day, for 72 weeks, did not result in any
changes in carbon monoxide diffusion capacity, compliance, or total
expiratory resistance to airflow (Vaughan et al., 1969).
Neither transpulmonary resistance nor compliance was affected in
rats exposed to 3800 µg/m3 (2.0 ppm) throughout their lifetimes
(approx. 2 years) although tachypnoea was consistently present
(Freeman et al., 1968c). Nonhuman primates also breathed more rapidly
when exposed for over 7 years to nitrogen dioxide levels of
3800 µg/m3 (2.0 ppm) (Freeman & Juhos, 1976, Freeman et al., 1969)
or to 9400 µg/m3 (5.0 ppm) for 2 months (Henry et al., 1970). In the
latter study, tidal volumes were also significantly reduced.
Guineapigs exposed to 9400 µg/m3 (5 ppm) for 7´ h per day, 5
days per week for 5´ months showed no changes in expiratory flow
resistance (Balchum et al., 1965). Murphy et al., (1964) exposed
guineapigs to 9800 µg/m3 (5.2 ppm) for 4 h and recorded increased
respiratory rate and decreased tidal volumes. Pulmonary function
returned to normal when the animals were returned to clean air.
Rats exposed to a nitrogen dioxide level of 5500 µg/m3
(2.9 ppm) for 24 h each day, 5 days per week for 9 months showed a
significant decrease (13%) in lung compliance compared with controls
(Arner & Rhoades, 1973). However, Wagner and co-workers (1965) were
unable to detect any significant effects in rabbits exposed to a
nitrogen dioxide concentration of 9400 µg/m3 (5 ppm) for 6 h daily
over a period of 18 months.
Davidson et al. (1967) exposed rabbits for 24 h/day for 3 months
to nitrogen dioxide levels of 15-22.6 mg/m3 (8-12 ppm) and observed
reversible increases in non-elastic resistance and in functional
residual capacity as well as a diminution in compliance.
The dose-effect relationship was studied in the lungs of rats and
cats exposed to nitrogen dioxide concentrations of 940-38 000 µg/m3
(0.5-20 ppm) (Zorn, 1975). A tendency towards an increase in
respiratory rates and a decrease in arterial oxygen pressure was shown
at concentrations as low as 1900 µg/m3 (1.0 ppm). A single 2-h
exposure to nitrogen dioxide levels of 19, 28, 66, and 94 mg/m3
(10, 15, 35, and 50 ppm) affected the pulmonary function in squirrel
monkeys. At concentrations of 19-28 mg/m3 (10-15 ppm) the tidal
volume decreased with little change in respiratory rate (Henry et
al., 1969).
5.1.3 Biochemical effects
Nakajima & Kusumoto (1968) reported an initial reduction in the
quantity of reduced glutathione in the lung and liver of mice exposed
to a nitrogen dioxide concentration of 1500 µg/m3 (0.8 ppm)
continuously for 5 days. On the fifth day, the level of glutathione
approached normal and was no longer significantly different from that
of the controls. However, with continuous exposure over 6 months the
animals tended to lose weight and the glutathione level fell once more
(Nakajima et al., 1969, Nakajima, 1973).
Chow et al. (1974) observed a rise in glutathione peroxidase
(1.11.1.9) activity in the lungs of rats exposed to nitrogen dioxide
concentrations of 1900 and 4300 µg/m3 (1.0 and 2.3 ppm) for 4 days.
At about 12 mg/m3 (6.2 ppm) there was a significant increase in the
activities of glutathione reductase (NAD(P)H) (1.6.4.2) and
glucose-6-phosphate dehydrogenase (1.1.1.49) in comparison with the
controls. The authors believe that alterations in such enzyme systems
are sensitive and specific bioindicators of tissue damage.
Fukase et al. (1976) reported that exposure to nitrogen dioxide at
about 11 mg/m3 (6 ppm) for 4 h every day for 30 days caused an
increase in glutathione reductase (NAD(P)H) (1.6.4.2) and
glucose-6-phosphate dehydrogenase (1.1.1.49) activities. Exposure to a
nitrogen dioxide level of 28 mg/m3 (15 ppm) for 7 days resulted in a
significant increase in glutathione levels. Exposure to 53 mg/m3
(28 ppm) for 7 days resulted in significant increases in glutathione
levels and in the activities of glutathione reductase (NAD(P)H)
(i.6.4.2), glucose-6-phosphate dehydrogenase (1.1.1.49) and
glutathione peroxidase (1.11.1.9).
Biochemical evidence indicating that on exposure to nitrogen
dioxide there is a proliferation of type 2 alveolar cells to replace
the injured type I cells in lung tissue has been submitted by Sherwin
et al. (1972). In these investigations, guineapigs were exposed to a
concentration of 3800 µg/m3 (2 ppm) continuously for 1-3 weeks; a
significant increase occurred in the lactate dehydrogenase
(cytochrome) (1.1.2.3) index of the lower lobes.
Oxygen consumption and lactate dehydrogenase (cytochrome)
(1.1.2.3), and aldolase (4.1.2.13) activity levels were all elevated
in lung, liver, kidney, and spleen tissue following short-term and
long-term exposure of guineapigs to nitrogen dioxide (Buckley &
Balchum 1965, 1967a, 1967b). In the short-term treatment the
guineapigs exposed to 75 mg/m3 (40 ppm) for a total of 4´ h were
killed 2 h after treatment. The long-term treatment included exposure
to 28 mg/m3 (15 ppm) continuously for 10 weeks. The mechanisms
involved in these changes have not yet been indentified but they may
reflect an acute response to stress.
By applying the disc electrophoresis method, Sherwin & Carlson
(1973) demonstrated higher protein levels in the lavage fluid of
guineapigs exposed for 1 week to a nitrogen dioxide concentration of
750 µg/m3 (0.4 ppm).
Thomas et al. (1968) found that short-term exposure (4 h) to
1900 µg/m3 (1.0 ppm) resulted in the lipoperoxidation of lung lipids
in rats. Rats fed on a vitamin E-deficient diet and then exposed to
nitrogen dioxide had more peroxidation of surfactant and tissue lipids
than did rats on a vitamin E-supplemented diet (Roehm et al., 1971).
Anti-oxidants appeared to serve as protection against peroxidation and
free radical formation (Menzel et al., 1972).
Arner & Rhoades (1973) reported a significant decrease (8.7%) in
the lung lipid content of rats exposed to a nitrogen dioxide
concentration of 5500 µg/m3 (2.9 ppm) for 9 months and also a marked
decrease in the percentage of total saturated phospholipid fatty
acids. This reduction in saturation was primarily due to a decrease in
the percentage of hexadecanoic (palmitic) acid. There were also
significant changes in the surface properties of the lung washings
from animals exposed to nitrogen dioxide indicating an increase in
surface tension and a decrease in stability of the pulmonary
surfactant. Continuous exposure for 14 days to a nitrogen dioxide
concentration of 9400 µg/m3 (5.0 ppm) markedly decreased the
lecithin turnover rate in rat lungs (Thomas & Rhoades, 1970). The
authors suggested that the pulmonary phospholipid synthesis might be
altered by nitrogen dioxide exposure.
5.2 Other Effects
Although the primary target for nitrogen dioxide exposure is the
lung, data are accumulating that indicate that it may effectively
alter a wide range of other systems.
5.2.1 Effects on growth and body weight
Numerous investigators have measured the growth rate of animals
during exposure to nitrogen dioxide and have produced conflicting
results. Body weights of rats and hamsters were reported to be
significantly lower than those of the controls in studies by: Kaut el
al. (1966) with a nitrogen dioxide exposure of 5000 µg/m3 (2.7 ppm)
for 8 weeks; Freeman & Haydon (1964) with a concentration of
24 mg/m3 (12.5 ppm) for 213 days; and Kleinerman & Rynbrandt (1976)
with a concentration of 38 mg/m3 (20 ppm) for 24 h. Other
investigators using guineapigs, hamsters, mice, rabbits, dogs, and
rats failed to find any such effects (Nakajima et al. (1969) with
nitrogen dioxide exposures of 1300 1500 µg/m3 (0.7 0.8 ppm) for 1
month; Salamberidze (1969) with a concentration of 100 µg/m3
(0.05 ppm) for 90 days; and Wagner et al. (1965) with a concentration
of 1900-47 000 µg/m3 (1.0-25 ppm) for 18 months).
5.2.2 Immunological effects
Nitrogen dioxide exposure seems to alter the immunological
reaction of experimental animals. Continuous exposure to
9400 µg/m3 (5.0 ppm) for 3-5 months appeared to depress the squirrel
monkey's ability to form protective serum neutralizing antibodies.
However, when monkeys were exposed for 16 months to a nitrogen dioxide
concentration of 1900 µg/m3 (1.0 ppm), the exposed animals
consistently showed higher serum neutralizing antibody titres than the
control (Ehrlich & Fenters, 1973; Fenters et al., 1971). The same
workers (Ehrlich et al., 1975) exposed mice to 940 µg/m3 (0.5 ppm)
continuously while superimposing a 1-h peak of 3800 µg/m3 (2.0 ppm)
each day over a period of 3 months in order to determine if this
exposure produced changes in the circulating immunoglobulins.
Non-vaccinated mice showed a marked decrease in levels of IgA and an
increase in serum IgM, IgG1, and IgG2. These investigators also
measured haemagglutination-inhibition (HI) and serum neutralization
(SN) antibody formation and found that nitrogen dioxide depressed the
SN antibody formation but did not alter the HI titres.
Contrary to these findings, Antweiler et al. (1975) did not find
any alteration in the ability of the guineapig to produce antibodies,
even after 33 days of exposure to 10 mg/m3 (5.3 ppm).
A month of continuous exposure to a nitrogen dioxide level of
1700 µg/m3 (0.9 ppm) reduced the ability of the mouse spleen to
produce primary antibodies (Nakamura et al., 1971).
Balchum et al. (1965) reported a circulating substance in the
serum of guineapigs exposed to nitrogen dioxide which had properties
similar to a lung antibody. This substance reacted in vitro with
proteins extracted from the lung tissue of control animals. The titres
of this reactive substance increased with the intensity and duration
of exposure to nitrogen dioxide. Exposure concentrations were 9400 and
28 000 µg/m3 (5 and 15 ppm) for periods of up to 1 year.
5.2.3 Haematological effects
Salamberidze (1969) studied the effects of continuous exposure
(24 h/day) of rats to a nitrogen dioxide concentration of 100 µg/m3
(0.05 ppm) for periods of up to 90 days. The author did not find any
effects on haemoglobin or erythrocytes.
Polycythemia, with reduced mean corpuscular volume but a normal
mean corpuscular haemoglobin concentration was found in rats and
monkeys (Macaca speciosa) following continuous exposure for 3 months
to 3800 µg/m3 (2.0 ppm) (Furiosi et al., 1973). Mitina (1962)
reported leucocytosis in rabbits exposed to nitrogen dioxide
concentrations of 2400-5700 µg/m3 (1.3-3.0 ppm) for 2 h per day for
15 and 17 weeks. This effect was followed by a phagocytic depression
of circulating leucocytes. The leucocytic response was accelerated by
the presence of sulfur dioxide.
Carson et al. (1962) exposed dogs to nitrogen dioxide
concentrations of 73-310 mg/m3 (39-164 ppm) for periods ranging from
5 min to 1 h. They did not find any changes in haematocrit or blood
platelet counts 4, 24, 48, or 72 h after exposure. Wagner et al.
(1965) did not find any haematological effect when dogs were exposed
for 18 months to nitrogen dioxide concentrations of 1900 or
9400 µg/m3 (1 or 5 ppm).
The addition of nitrogen dioxide at concentrations of
940-1500 µg/m3 (0.5-0.8 ppm) to carbon monoxide at 58 mg/m3
(50 ppm) failed to affect the carboxyhaemoglobin concentrations in the
blood of mice in studies reported by Nakajima & Kusumoto (1970).
The chemical action of nitrogen dioxide on the circulating
erythrocyte in vivo is poorly understood. Although
methaemoglobinaemia could possibly result from exposure to low levels
of nitrogen dioxide, evidence confirming this is not available. Wagner
(1977, unpublished data a) could not find a detectable increase in
the concentration of methaemoglobin in rats exposed to 9400 µg/m3
(5 ppm) for as long as 10 months. However, exposure to 19 mg/m3
(10 ppm) for 1 h induced a significant increase in the concentration
of methaemoglobin in arterial blood. Nakajima & Kusumoto (1968) did
not find any increase in the concentration of methaemoglobin in the
blood of mice continuously exposed to a nitrogen dioxide concentration
of 1500 µg/m3 (0.8 ppm) for 5 days.
5.2.4 Miscellaneous biochemical effects
Continuous exposure of rats to a nitrogen dioxide concentration of
100 µg/m3 (0.05 ppm) for 90 days did not produce any effects on the
activities of cholinesterase (3.1.1.8), catalase (1.11.1.6), and
SH-groups in blood (Salamberidze, 1969).
Veninga & Lemstra (1975) reported that a single 2-h exposure to
560-7500 µg/m3 (0.3-4.0 ppm) produced elevated levels of ascorbic
acid in the liver of mice.
Kosmider & Misiewicz (1973) exposed guineapigs to a nitrogen
dioxide concentration of 1900 µg/m3 (1.0 ppm), continuously, for a
total of 180 days. The authors observed increased aminotransferase
activity in the blood serum and heart homogenates but decreased levels
in the brain and in the liver. No significant alterations were evident
in the basic alkaline phosphatase (3.1.3.1) and magnesium-activated
phosphatase activities in the blood serum of dogs exposed to nitrogen
dioxide levels of either 1900 or 9400 µg/m3 (1 or 5 ppm) for 18
months (Wagner et al., 1965).
However, Wagner (1972) found an elevation in serum cholesterol in
rats after continuous exposure to 9400 µg/m3 (5 ppm) for
1 year.
Drozdz and co-workers (1973, 1974, 1975) measured the effects of 6
months continuous exposure to a nitrogen dioxide concentration of
2000 µg/m3 (1.1 ppm) on the guineapig. They studied alterations in
the activity of various enzymes in both the blood and liver as well as
in the central nervous system. The experiments demonstrated that
prolonged exposure to nitrogen dioxide led to disturbances in the
levels of glucose, lactic acid, total lipids, seromucoids, hexoses,
hexamines, and sialic acid. Kaut et al. (1966) reported a decrease in
the albumin/globulin (A/G) quotient as well as in the vitamin C
content of the suprarenal glands, in rats exposed to a nitrogen
dioxide concentration of 5000 µg/m3 (2.7 ppm) for 2-8 weeks, 6 h per
day, 5 days per week.
Continuous exposure to high nitrogen dioxide concentrations of
28-216 mg/m3 (15 115 ppm) produced numerous systemic biochemical
changes. Nitrogen dioxide has been reported to produce increases in:
serum protease inhibitor activity; plasma corticosterone levels;
oxygen consumption in the spleen and kidney; lactate dehydrogenase
(cytochrome) (1.1.2.3) activity in liver and kidney; and aldolase
(4.1.2.13) activity in the liver, kidney, spleen, and serum (Buckley &
a See footnote, p. 32.
Balchum, 1965; Kleinerman & Rynbrandt, 1976; Tusl, 1975). Svorcova &
Kaut (1971) reported an elevation in urinary nitrites and nitrates in
rabbits immediately after exposure for 15 min to a concentration of
45 mg/m3 (23.9 ppm). It is possible that this indicates that
nitrogen dioxide is rapidly converted to nitrite and nitrate ions and
that these ions are excreted in the urine shortly afterwards.
5.2.5 Effects on reproduction
Salamberidze & Cereteli (1971) studied changes in the female rat's
reproductive and endocrine systems resulting from exposure to
2360 µg/m3 (1.3 ppm), 12 h per day for 3 months. There was a
prolongation of the estrus cycle associated with an increased
interestrual period and a decrease in the number of monthly cycles.
The litter size and the fetal weights decreased but the capacity for
pregnancy was not affected. These effects may reflect the effect of
nitrogen dioxide exposure on endocrine or reproductive function.
5.2.6 Effects on the central nervous system
Salamberidze (1969) did not find any effects on the central
nervous system in rats exposed to a nitrogen dioxide concentration of
100 µg/m3 (0.05 ppm) for 90 days.
At a slightly higher concentration, Jakimcuk & Celikanov (1968)
reported a significant delay in the conditioned reflexes of the
central nervous system in rats after 90 days of exposure to a nitrogen
dioxide concentration of 600 µg/m3 (0.32 ppm).
5.2.7 Behavioural changes
Murphy et al. (1964) reported that the voluntary running activity
of male mice was depressed when the concentration of nitrogen dioxide
reached 14 mg/m3 (7.7 ppm) and the animals were exposed for 6 h. A
similar loss in activity was reported by Tusl et al. (1973). They
measured the influence of nitrogen dioxide on the performance of rats
during physical exertion as measured by swimming. In rats exposed to
9400 µg/m3 (5.0 ppm), a decrease of 25% in performance occurred in
the fifth to sixth week of the experiment. Animals exposed to
1900 µg/m3 (1.0 ppm) also showed a tendency towards decreased
performance, although this was not statistically significant.
5.2.8 Carcinogenicity, mutagenicity, and teratogenicity
In order to study the possible formation of nitrosamines by the
reaction of nitrogen dioxide with tissue amines, mice were exposed to
a nitrogen dioxide concentration of 75 mg/m3 (40 ppm) for periods up
to 1´ years. Although proliferative alterations at the terminal
bronchioles were always present, no carcinomas were found in the lungs
of these animals (Henschler & Ross, 1966).
Kaut (1970) analysed lung tissue to detect nitro- and
nitroso-compounds, especially nitrosamines in white rats exposed to
mixtures of oxides of nitrogen ranging from 5 to 250 ppm for 3 h. The
compounds were found in vitro in tissues exposed to high
concentrations of oxides of nitrogen, but not in vivo.
Rats were exposed for periods ranging from 2´ months to a lifetime
(2 years) to automotive exhaust gas containing carbon monoxide, oxides
of nitrogen, carbon dioxide, and aldehydes at concentrations of
58 mg/m3 (50 ppm), 23 ppm, 6700 mg/m3 (3700 ppm), and 2.0 ppm,
respectively. According to the author, spontaneous tumours and
abscesses were more frequent in the group exposed to the gas than in
the control group but none occurred in the lung tissue (Stupfel et al.
1973).
These studies cannot be considered to provide any evidence of the
carcinogenic effect of oxides of nitrogen.
No evidence is available on the mutagenicity and teratogenicity of
oxides of nitrogen per se, but nitrous acid has been reported to be
mutagenic in some laboratory tests.
5.3 Interaction of Nitrogen Dioxide and Infectious Agents
The influence of nitrogen dioxide on susceptibility to respiratory
infection and its adverse effect on the pulmonary defence system of
the host has been clearly demonstrated in several species of animals.
Ehrlich (1966) and Henry et al. (1969) showed that exposure of
mice, hamsters, and squirrel monkeys to nitrogen dioxide made them
more vulnerable to respiratory infection with Klebsiella
pneumoniae, the mouse being the most sensitive. With a 2-h exposure,
the minimum concentration of nitrogen dioxide required to produce a
significant rise in the mortality rate was 6600 µg/m3 (3.5 ppm). No
effect was observed at 4700 µg/m3 (2.5 ppm). However, when mice were
exposed continuously for 1 year to 940 µg/m3 (0.5 ppm), a
statistically significant increase in mortality rate occurred after 90
days (Blair et al., 1969; Ehrlich & Henry, 1968).
Ito et al. (1971) exposed female mice to nitrogen dioxide
concentrations of 940-1900 µg/m3 (0.5-1.0 ppm) continuously for 39
days and studied the influence of the nitrogen dioxide on infection
with influenza virus histopathologicaily. Advanced interstitial
pneumonia and adenomatous proliferation in the epithelium of the
peripheral bronchi were noted. Intermittent exposure to 19 mg/m3
(10.0 ppm) for 2 h daily, for 5 days, also significantly increased the
susceptibility of mice to influenza virus infection as demonstrated by
increased mortality.
Motomiya et al. (1972) studied the interaction of nitrogen dioxide
and the influenza virus. They reported a high incidence of adenomatous
proliferation of peripheral, bronchial, epithelial cells in mice
exposed for 3 months to nitrogen dioxide levels of 560-940 µg/m3
(0.3-0.5 ppm) followed by infection with the influenza virus. The
effect was more serious than that seen in infected mice kept in clean
air. Continuous exposure for an additional 3 months did not enhance
the effect.
Continuous exposure of squirrel monkeys to nitrogen dioxide levels
of 9400 µg/m3 and 19 mg/m3 (5 and 10 ppm) for 1 or 2 months
increased their susceptibility to both bacterial and viral infections.
All the animals exposed to 19 mg/m3 (10 ppm) died within 2-3 days of
infection with the influenza virus. At 9400 µg/m3 (5 ppm) 1 of the 3
experimental monkeys died. All control monkeys had symptoms of vital
infection but no deaths occurred. When the nitrogen dioxide exposed
monkeys were challenged with Klebsiella pneumoniae, 2 out of 7 monkeys
exposed to 9400 µg/m3 (5 ppm) for 2 months died and the remainder
had the infectious agent in the lungs at autopsy. At 19 mg/m3
(10 ppm) for 1 month 1 of 4 monkeys died and 2 had the infectious
agent in the lungs at autopsy (Henry et al., 1969, 1970). Both
squirrel monkeys and hamsters showed a reduction in resistance to
Klebsiella pneumoniae after a single 2-h exposure to nitrogen
dioxide concentrations of 66-75 mg/m3 (35-40 ppm) (Ehrlich 1966).
Gardner et al. (1977) and Coffin et al. (1976) studied the
time-dose-response for nitrogen dioxide -- Streptococcus pyogenes
interaction. Mice were exposed to nitrogen dioxide concentrations
ranging from 940 µg/m3-53 mg/m3 (0.5 28 ppm) for various periods
of time ranging from 10 min to 12 months before treatment with the
bacterial aerosol. When comparisons were made, it was evident that the
mortality rate increased with increasing concentrations of nitrogen
dioxide. Different relationships between concentration and time
produced significantly different mortality responses. The data suggest
that concentration has a greater influence on the effect of nitrogen
dioxide than length of exposure.
Studies were also conducted to compare continuous with
intermittent exposure to nitrogen dioxide at concentrations of 2800
and 6600 µg/m3 (1.5 and 3.5 ppm). There was a significant increase
in mortality rate for each of the experimental groups with increasing
duration of exposure. When the data were adjusted for the total
difference in concentration x time, the mortality rate was essentially
the same for both groups.
There are numerous studies that show that nitrogen dioxide is
injurious to specific pulmonary defence systems in the lung. It
interferes with the efficiency of clearing inhaled particles including
bacteria and viruses from the airway and with the phagocytosis and
digestion of such particles by the alveolar macrophage. As this is the
major defence against infection by inhalation, any alteration in this
system would increase the risk of infection (Green & Kass, 1964; Kass
et al. 1966).
Aranyi & Ehrlich (1973, unpublished dataa) isolated alveolar
macrophages from mice continuously exposed to a nitrogen dioxide
concentration of 4700 µg/m3 (2.5 ppm) for 3 h daily, 5 days per
week, for one month. Scanning electron microscopy revealed changes in
the surface of these cells and a reduction in the ability of the cells
to phagocytize Escherichia coli, in vitro.
Goldstein et al. (1973) measured the effect of nitrogen dioxide on
antibacterial activity in the mouse. Pulmonary bactericidal activity
decreased progressively with exposure to increasing concentrations of
nitrogen dioxide. This defect was present in animals exposed for 4 h
to concentrations of nitrogen dioxide of 13 mg/m3 (7 ppm) or more.
With exposure for 17 h, this bactericidal dysfunction occurred at
levels as low as 4300 µg/m3 (2.3 ppm). These reports are consistent
with the earlier report of Gardner et al. (1969) who obtained
macrophages from the lungs of rabbits after in vivo phagocytosis and
found a pronounced inhibition of phagocytic activity (50%) after a 3-h
exposure to nitrogen dioxide at 19 mg/m3 (10 ppm). A 2-h exposure to
15 mg/m3 (8.0 ppm) increased the proportion of polymorphonuclear
leucocytes in the lavage fluid. This condition persisted for more than
72 h after the cessation of exposure.
Acton & Myrvik (1972) found that rabbits exposed to nitrogen
dioxide concentrations of 28 mg/m3 (15.0 ppm) for a short period of
time (3 h) had alveolar cells with a lower capacity to develop
virus-induced resistance and to phagocytize BCG vaccine. Valand et al.
(1970) demonstrated that macrophages washed from the lungs of rabbits
exposed for 3 h to a nitrogen dioxide concentration of 47 mg/m3
(25 ppm) and injected with parainfluenza-3 virus failed to develop
resistance to rabbit pox virus. The alveolar macrophages obtained from
these exposed animals failed to produce interferon.
Buckley & Loosli (1969) exposed mice for 6 weeks to a nitrogen
dioxide concentration of 71 mg/m3 (38.0 ppm) and were unable to
detect any alteration in the rate of clearance of an aerosol of
staphylococci.
a Illinois Institute of Technology Research Institute Report No.
L6070-2, EPA Contract No. 68-02-0761.
The second mechanism of host defence is the mechanical or physical
removal of inhaled and deposited particles by means of the mucociliary
escalator.
Giordano & Morrow (1972) determined that exposure to a nitrogen
dioxide concentration of 11 mg/m3 (6 ppm) continuously for 6 weeks
depressed this mucociliary transport.
5.4 Summary Table
Experimental animal studies which provide useful quantitative
information for the establishment of guidelines for the protection of
public health with respect to nitrogen dioxide exposure are summarized
in Table 8.
Table 8 Experimental animal studies
I. Local effects on the respiratory system
Nitrogen dioxide
concentration Length of
exposure
Effects Responsea Species Number Reference
(µg/m3) (ppm) (number (h/day) of
of animals
days)
1900 1.0 4 24 glutathione peroxidase rat 5 (11) Chow et al. (1974)
(1.11.1.9) activity significantly
increased in lung
1900 1.0 1 4 peroxidation of lung lipids rat 6 (10) Thomas et al. (1968)
1500 0.8 990 24 elevated respiratory rates 9/9 (0/12) rat 9 (12) Freeman et el. (1966)
throughout their lives
1500 0.8 990 24 minimal bronchiolar epithelial 9/9 (0/12) rat 9 (12) Freeman et al. (1966)
hypertrophy
1500 0.8 5 24 significant decrease in lung mouse 10 (10) Nakajima & Kusumoto
reduced glutathione (1968)
940-1500 0.5-0.8 30 24 degeneration and desquamation 10/10 (0/5) mouse 10 (5) Nakajima et al (1969)
of mucous membrane;
I. Local effects on the respiratory system cont'd
Nitrogen dioxide
concentration Length of
exposure
Effects Responsea Species Number Reference
(µg/m3) (ppm) (number (h/day) of
of animals
days)
940-1500 0.5-0.8 30 24 shortening and reduction of the 6/10 (0/5) mouse 10 (5) Hattori et al. (1972)
cilia of ciliated epithelial cells;
oedematous change of alveolar
epithelial cells
proliferation of epithelial cells
of the peripheral bronchus
(adenomatous changes)
1200 plus 0.64 plus 61 16 reduction in pulmonary 6/11 (3/8) dog 11 (18) Lewis et al. (1974)
310 nitric 0.25 months diffusion capacity
oxide nitric
oxide
1200 plus 0.64 plus 61 16 decreased peak expiratory flow 5/11 (1/8) dog 11 (18) Lewis et al. (1974)
310 nitric 0.25 months rate
oxide nitric
oxide
940 plus 0.5 plus 18 16 no changes in carbon 0/12 (0/20) dog 12 (20) Vaughan et al.(1969)
250 nitric 0.2 months monoxide diffusion capacity,
oxide nitric compliance, or total expiratory
oxide resistance
940 0.5 90-360 6,18,24 evidence of focal emphysema 12/12 (0/4) mouse 12 (4) Blair et al. (1969)
I. Local effects on the respiratory system cont'd
Nitrogen dioxide
concentration Length of
exposure
Effects Responsea Species Number Reference
(µg/m3) (ppm) (number (h/day) of
of animals
days)
940 0.5 1 4 reduction in mitochondria of 6/6 (0/6) rat 6 (6) Thomas et al. (1967)
alveolar cells and degradation
of mast cells
750 0.4 7 24 significant increase in protein guineapig 9 (9) Sherwin & Carlson (1973)
content of lung lavage fluid
600 0.32 90 24 morphological changes such as rat 16 (15) Jakimcuk & Celikanov
bronchitis, peribronchitis, and (1969)
light pneumosclerosis; no effect
observed at 150 µg/m3
(0.08 ppm)
470 0.25 24-36 4 structural changes in lung 2/3 (0/1) rabbit 3 (1) Buell (1970)
collagen fibres (electron
microscope)
100 0.05 90 24 no pathological or histological 0/10 (0/10) rat 10 (10) Salamberidze (1969)
effects
II. Other effects
Nitrogen dioxide
concentration Length of
exposure
Effects Responsea Species Number Reference
(µg/m3) (ppm) (number (h/day) of
of animals
days)
1900 1.0 16 months 24 higher serum neutralizing antibody titres than squirrel 5 Ehrlich & Fenters (1973)
control monkey
1900 1.0 180 24 decreased aminotransferase activity in brain guineapig 30 (50) Kosmider & Misiewicz
and liver, increased activity in blood serum (1973)
and heart
1700 0.9 30 24 reduction in antibody production in spleen mouse 9 (9) Nakamura et al. (1971)
1300-1500 0.7-0.8 1 month 24 no change in growth rate mouse 20 (20) Nakajima et al. (1969)
940 with 0.5 with 3 months 24 changes in circulating immunoglobulins; mouse 112-160 Ehrlich et al. (1975)
1 h peak 1 h peak depression in serum neutralizing antibody titres (112-160)
(3800) (2) daily
daily
940-1500 + 0.5-0.8 + 1-1.5 24 no change in blood carboxyhaemoglobin mouse 94 (49) Nakajima & Kusumoto
58 000 50 months (1970)
carbon carbon
monoxide monoxide
600 0.32 90 24 significant changes in conditioned reflexes of rat 15 (15) Jakimcuk & Celikanov
the central nervous system (1968)
II. Other effects cont'd
Nitrogen dioxide
concentration Length of
exposure
Effects Responsea Species Number Reference
(µg/m3) (ppm) (number (h/day) of
of animals
days)
560-7500 0.3-4.0 1 2 increase in ascorbic acid levels in liver mouse 20 (114) Veninga & Lemstra
(1975)
100 0.05 90 24 no effects on weight gain, central nervous rat 10 (10) Salamberidze (1969)
system, activities of cholinesterase (3.1.1.8),
catalase (1.11.1.6) and SH-groups in blood,
haemoglobin, or erythrocytes
a Number of animals showing effects/total number of animals; numbers in brackets refer to control groups.
Table 8. Experimental animal studies--continued
III. Interaction with infectious agents
Nitrogen dioxide
concentration Length of
exposure
Effects Responsea Species Number Reference
(µg/m3) (ppm) (number (h/day) of
of animals
days)
940-1900 0.5-1.0 39 24 higher incidence of 7/12 (1/12) mouse 12 (12) Ito et al. (1971)
adenomatous proliferation of
bronchial and bronchiolar
epithelium than unexposed
challenged group
940 0.5 12 24 increased susceptibility to mouse 4 (4) Blair et al. (1969)
months infection; first statistical Ehrlich & Henry (1968)
significance evident at 90 days;
reduced pulmonary clearance of
inhaled microbes
560-940 0.3-0.5 3 24 more severe adenomatous mouse 12 (8) Motomiya et al. (1972)
months proliferation of the peripheral
bronchial cells than unexposed
challenged group
560-940 0.3-0.5 6 24 no further enhancement by an mouse 12 (8) Motomiya et al. (1972)
months additional 3 months exposure
a Number of animals showing effects/total number of animals; numbers in brackets refer to control groups.
6. EFFECTS ON MAN
6.1 Controlled Exposures
The effects of nitrogen dioxide on both healthy subjects and
patients have been studied (with their consent) under controlled
conditions. Although the studies are few in number and only concern
short-term exposure, much useful information has been obtained for
assessing health effects in man.
Henschler et al. (1960) studied normal, healthy males, aged 20-35
years to obtain data concerning the threshold of odour perception for
nitrogen dioxide. When the concentrations reached 230 µg/m3
(0.12 ppm), 3 out of 9 subjects perceived the odour immediately and 8
out of 13 could detect concentrations of 410 µg/m3 (0.22 ppm). At a
higher concentration of 790 µg/m3 (0.42 ppm), 8 out of 8 subjects
immediately recognized the odour. When the nitrogen dioxide
concentration was increased very gradually from 0 to 51 mg/m3
(27 ppm), the volunteers failed to detect the odour. Awareness of the
odour increased when the humidity was increased from 60% to 80%.
Similar studies on odour perception were carried out by Feldman (1974)
and galamberidze (1967) and in these experiments the nitrogen dioxide
olfactory thresholds were found to be 200 and 230 µg/m3 (0.11 and
0.12 ppm), respectively.
Threshold values for the impairment of dark adaptation by nitrogen
dioxide have also been reported. Salamberidze (1967) determined that
the threshold after 5 and 25 min of nasal inhalation of nitrogen
dioxide was 140 µg/m3 (0.074 ppm).
Studies with human volunteers were also made to determine subtle
changes in respiratory function during nitrogen dioxide exposure.
Volunteer patients with moderate degrees of chronic respiratory
diseases were studied as well as healthy individuals.
Airway resistance increased significantly compared with
pre-exposure resistance following 5-min exposure of 15 healthy
individuals to nitrogen dioxide levels of 5600-75000 µg/m3
(3-40 ppm) (Nakamura, 1964). Suzuki & Ishikawa (1965) exposed 10
healthy subjects to nitrogen dioxide concentrations ranging from
1300-3800 µg/m3 (0.7-2.0 ppm) for 10 min. The inspiratory and
expiratory flow resistance rose to about 150 and 110% of control
values, respectively, 10 min after the exposure.
Abe (1967) reported studies in which 5 healthy males were
exposed to nitrogen dioxide levels of 7500-9400 µg/m3 (4-5 ppm)
for 10 min. Inhalation of nitrogen dioxide caused an increase in
both expiratory and inspiratory flow resistance reaching a maximum
30 min after the end of the exposure. Values for effective
compliance obtained 30 min after cessation of exposure showed a 40%
decrease compared with controls.
Orehek et al. (1976) measured the bronchomotor sensitivity of
asthmatic patients to a bronchoconstrictor agent (carbachol) before
and after exposure to nitrogen dioxide. These authors attempted to
establish dose-response curves for the specific airway resistance
(SRaw) of 20 asthmatics who were exposed for 1 h to 190 and
380 µg/m3 (0.1 and 0.2 ppm). The degree of enhancement of
bronchial sensitivity by the nitrogen dioxide was variable among the
individuals tested. Nitrogen dioxide at 190 µg/m3 (0.1 ppm)
induced a significant increase in initial SRaw and enhanced the
bronchoconstrictor effect in 13 subjects. In 7 subjects this level
of nitrogen dioxide did not modify either of these effects. Several
possible reasons were advanced by the authors to explain why some
asthmatic subjects responded and others did not. It seems clear that
sensitivity to nitrogen dioxide can vary among individuals. Yokoyama
(1968, 1970, 1972) also reported considerable individual variation
in response among volunteers exposed from 10 to 120 min to several
concentrations of nitrogen dioxide.
Nieding and his associates conducted a series of studies on the
effects of nitrogen dioxide on pulmonary function in man.
In 1970, they reported pulmonary function studies on 13 healthy
subjects and 88 patients with chronic bronchitis who were exposed
for 15 min to nitrogen dioxide levels of 940 9400 µg/m3
(0.5-5.0 ppm). Inhalation of concentrations below 2800 µg/m3
(1.5 ppm) had no significant effect. Concentration