The Effects of Nitrogen Dioxide on Human Health

15. In considering the health effects of nitrogen dioxide, the Panel have been guided by the conclusions of the Department of Health's Advisory Group on the Medical Aspects of Air Pollution Episodes, who considered the evidence with respect to this pollutant in their third report. In very high concentrations, such as have occurred in certain industrial accidents, nitrogen dioxide can cause very severe and sometimes fatal lung damage. In this report the Panel are concerned with the health effects that may be observed at the much lower concentrations that occur during pollution episodes in our towns and cities. It has been suggested that the gas may have both acute, short-term, and chronic, longer-term, effects on health, particularly in people with asthma.
16. The mechanism by which nitrogen dioxide acts is most probably related to its properties as an oxidising agent which can damage cell membranes and proteins. At relatively high concentrations nitrogen dioxide causes acute inflammation of the airways. In addition, short-term exposure can affect the immune cells of the airways in a manner that might predispose people to an increased risk of respiratory infections.
17. The health effects of nitrogen dioxide have been assessed in four main ways: (i) by experimental exposure of volunteers with and without asthma to the gas; (ii) by assessment of the effects on groups of people of variations in ambient concentrations of nitrogen dioxide, using daily symptoms or lung function measurements; (iii) by assessing changes in hospital admissions or mortality in a population in relation to changes in ambient concentrations; and (iv) by comparing the health of groups of people who have had different long-term exposures.
18. People with healthy lungs, whether at rest or exercising, show little response to experimental inhalation of nitrogen dioxide at concentrations well above those occurring in the ambient air, even during extreme pollution episodes. Very small changes in sensitive tests of lung function have been recorded at exposures between 2500 and 7500 ppb. However, in people with asthma, some studies have shown changes in these tests of lung function to have occurred at exposures of around 300 ppb when the subjects have been exercising, though other studies have shown no changes at higher concentrations. Measurements of the responsiveness of the lung to inhalation of irritant chemicals have shown that the airways of some people with asthma may become more sensitive to such stimuli after exposure to nitrogen dioxide at concentrations down to about 200 ppb. It should be noted that this concentration is only reached in occasional episodes of outdoor air pollution in the United Kingdom. In 1976, an early study reported similar effects in people with asthma at concentrations as low as 100 ppb. However, these findings have never been repeated in any other study, and the Panel considered that there was sufficient doubt about the validity of the measurements of nitrogen dioxide for this isolated result to be disregarded.
19. Other studies have investigated the possibility that inhalation of nitrogen dioxide at moderate concentrations may cause an inflammatory reaction in the lungs or may increase the susceptibility of individuals to subsequent inhalation of allergens, such as those from house dust mite or grass pollen. The inflammatory reaction, if repeated frequently, might act to decrease the resistance of individuals to infection, and is more relevant to repeated exposures to elevated indoor levels than to exposure to the outdoor concentrations typically occurring in the United Kingdom. At present, the evidence from both human and animal studies is equivocal. However, there is some evidence that exposure to nitrogen dioxide can enhance the response of someone with asthma to inhalation of allergen. After exposure of volunteers with asthma to 400 ppb for one hour, inhalation of house dust mite extract has been shown to cause a significantly greater fall in lung function than occurred in a control experiment, suggesting that the gas may have primed the airways to react more readily to the allergen. A similar finding has been reported following ozone exposure and grass pollen inhalation and it may be that this subtle effect is a more important consequence of irritant gas exposure than the more direct effects on lung function. Nevertheless, to date all these effects have only been demonstrated at concentrations of nitrogen dioxide that occur in the ambient air of the United Kingdom only in the most exceptional circumstances.
20. Although the exposure chamber studies discussed above have been unable to demonstrate effects at low concentrations of nitrogen dioxide, a number of epidemiological studies have shown associations between such health effects and exposure to concentrations below the 1987 WHO guidelines. Not all of these studies have taken appropriate account of the many confounding factors, such as concurrent exposure to other pollutants and exposure to indoor sources. In two studies, one of attacks of croup in German children and the other of childhood asthma admissions to hospital in Birmingham, an increase in concentrations of nitrogen dioxide was associated with an increase in illness.
21. Studies of episodes of pollution, in which much higher concentrations have occurred, may perhaps provide a better method of assessing the magnitude of health effects. The most important of these was the December 1991 episode in London, in which the maximum recorded hourly average concentration of nitrogen dioxide was 423 ppb. Particles, measured as black smoke, and sulphur dioxide were also raised during this episode to daily average concentrations of 228 µg/m3 of black smoke in Ilford and 54 ppb of sulphur dioxide at both Lambeth and the City of London, as measured in the national smoke and sulphur dioxide network. During this episode the mortality rate in London from all causes was raised by 10% and there was an increase in admission rates to hospital among older people with chronic lung disease. No effects on younger people with asthma were detected. It was not clear whether these effects were primarily associated with exposures to nitrogen dioxide or to particles. An episode in 1992 in Birmingham showed an effect on lung function in a group of people with severe asthma, but no effects in people with milder asthma, and in this episode the association seemed to be stronger with particles, measured as black smoke, than with the gases. However, it is likely that in all such episodes it is the mixture of pollutants, perhaps combined with other factors associated with meteorological conditions, that is the cause, rather than any individual constituent.
22. The concentrations of nitrogen dioxide to which people have been exposed in these episodes are generally lower than those levels producing effects in exposure chamber studies. This raises the question as to whether the health effects recorded are really due to nitrogen dioxide or, rather, whether they may be a consequence of parallel exposure to other pollutants, perhaps in combination with nitrogen dioxide. If they are due to nitrogen dioxide alone, it may be that they are a consequence of different patterns of exposures to those studied in volunteers, perhaps with earlier lower concentrations acting to potentiate the effect of later higher levels. The Panel are of the view that this question remains unresolved.
23. The question as to whether exposure to nitrogen dioxide causes long-term damage to the lungs is less well studied and the information is more difficult to interpret. There is less information, what there is is contradictory, and the evidence tends to be indirect, relying on surrogate measures of exposure rather than direct measurements. Studies from Switzerland and the United States have suggested that those living in areas with higher exposures to nitrogen dioxide have poorer lung function, but other similar studies have failed to find such an association. The contrast between the former East and West Germany, with higher levels of nitrogen dioxide from traffic pollution and higher levels of hay fever and asthma in the West, led to some speculation that the two might be causally connected. However, although there is, as mentioned above, some experimental evidence that nitrogen dioxide may enhance susceptibility to allergens and some studies have suggested that those living near to busy roads or in areas with heavy traffic pollution may be at greater risk of asthma, a recent study in East London failed to find such an association.
24. Since exposure to nitrogen dioxide indoors is often an important contributor to the overall exposure of individuals, some studies have specifically investigated relationships between such exposure and health. Indoor concentrations of nitrogen dioxide in the kitchens of homes with gas cookers average, over a year, around 15 ppb and peak concentrations may be as high as almost 600 ppb over an hour. The outdoor concentration of nitrogen dioxide is the main determinant of indoor concentrations in homes without gas cookers, the concentrations generally being somewhat lower indoors. In homes with such cookers, indoor levels are usually at or above outdoor levels, being higher in the winter months when homes are less well ventilated and more use is made of gas appliances. Thus people living in homes with gas cookers are exposed to more nitrogen dioxide indoors than those living in other homes. Studies of children living in homes with gas cookers have led to suggestions that the increased symptoms reported in these children may be due to their increased exposure to nitrogen dioxide from this source. There are fewer studies of the association between lung health and gas cooking in adults and the associations reported have not been consistent, but a recent study in East Anglia showed that young women who had gas cookers had substantially more respiratory symptoms and poorer lung function on average than women from homes with electric cookers. Although these effects could be due to nitrogen dioxide, it should be borne in mind that gas cooking has much wider effects on the indoor environment and that these may also affect respiratory health. Taking all this evidence into account, the Panel have taken the view that a long-term effect of chronic exposure to nitrogen dioxide, whilst not yet demonstrated, cannot be ruled out.
25. The possibility that nitrogen dioxide may cause damage to genes in cells has been examined in a few studies. Mutations have been described in bacterial and mammalian cells exposed to high concentrations of the gas, but the evidence is limited. Standard tests for chromosomal damage in mammalian cells and in animals have been negative. There is no evidence that nitrogen dioxide is carcinogenic but it could, under appropriate conditions, react with amines in the body to form carcinogenic N-nitroso compounds. The generation of such substances has been reported in animals but the subsequent development of tumours has not been described. Nitrogen dioxide may combine with other atmospheric chemicals, notably polycyclic aromatic hydrocarbons to produce nitroarenes. Many nitroarenes are in vitro mutagens and some are carcinogenic in animals exposed to high doses by routes other than inhalation. The International Agency for Research on Cancer has categorised six such compounds as 'possibly carcinogenic to humans'. These various indirect effects have not been included in the Panel's considerations for recommending an air quality standard for nitrogen dioxide.