Levels of air pollution considered safe by the Environmental Protection Agency (EPA) increase stroke risk, and do so within hours of exposure, according to a new study. The results suggest that there may be no “safe” lower limit for airborne particulate matter, raising the question of how to weigh the cost of reducing pollution against the benefits of reducing stroke.
AN INCREASE in 10 micrograms per cubic meter of fine pollutants increases the relative risk of stroke by 18 percent, investigators reported, far above estimates from previous studies, which have ranged from no effect to 6 percent.
The health effects of low levels of particulate matter have been long studied and have clearly been tied to cardiovascular events, but the effect of fine particulates on stroke has been controversial, according to Gregory A. Wellenius, ScD, assistant professor of epidemiology at Brown University in Providence, RI, and lead author of the study published in the Feb. 13 Archives of Internal Medicine.
Levels of particulate matter are regulated by the EPA, and particles smaller than 2.5 micrometers in diameter (PM2.5) are of special concern. That size is important, Dr. Wellenius said, because, unlike larger particles, the smaller ones can deposit deep in the airways. While the exact mechanism of increased stroke risk is unknown, it may involve a combination of autonomic dysfunction, direct vessel damage, and inflammatory changes.
According to the EPA Air Quality Index, a PM2.5 of 15 micrograms per cubic meter indicates that “air quality is satisfactory and poses little or no risk,” while levels from 15 to 40 micrograms indicate that “air quality is acceptable; however, there may be a moderate health concern for a very small number of people.”
Past studies of stroke risk from pollution have almost all drawn their data from hospital billing records, rather than medical charts. The weakness of that approach, Dr. Wellenius said, is that billing records don't provide accurate information on timing of symptom onset.
“We decided these limitations were important and we thought we could do better by reviewing the medical records of individual patients,” he said. Because all patients in the study were seen by neurologists, and because “neurologists are particularly good at taking a careful history,” Dr. Wellenius said he obtained a very accurate estimate of when each patient had his or her stroke.
The investigators also had access to hourly records of airborne particulate matter in the Boston area from a recording station operated by Harvard University. The study was performed while Dr. Wellenius was a Fellow in Cardiovascular Medicine at Beth Israel Deaconess Medical Center in Boston, which is affiliated with Harvard.
DR. GREGORY A. WELLENIUS reported that the odds ratio for stroke rose approximately linearly with increasing air pollution, even when the level was within the EPA level considered “satisfactory.”
By comparing variables, including weather, between the day of the stroke and a day one week before, each patient served as his or her own control. Dr. Wellenius explained. “We asked, ‘Why did they have a stroke today, rather than the same day last week?’”
Over a nine-year period, Dr. Wellenius enrolled 1705 patients with a mean age of 73 years. About one quarter of the strokes were small-vessel strokes, about one quarter were due to cardioembolisms, and one fifth were due to large-artery atherosclerosis.
The PM2.5 level met the EPA standard for “satisfactory” on 83 percent of the days of the study, and for “acceptable” on 16 percent of the days. The odds ratio for stroke rose approximately linearly with increasing air pollution, even when the level was within the “satisfactory” zone. Compared with the stroke risk at PM2.5 of less than 5 micrograms per cubic meter (very clean air), the risk at 15-20 micrograms per cubic meter (the upper end of “satisfactory”) was approximately 1.4 times as great, and at 20 micrograms and higher, was 1.5 times as great.
Strokes were more common in the 24 hours following a transient elevation in air pollution, Dr. Wellenius found. Looking at the relationship between an elevation of PM2.5 and stroke onset, he found that the risk of stroke rose steadily for the 12 hours after a rise in air pollution, then fell again. Twenty-four hours after the transient elevation, the effect of the increase had disappeared.
The effect of air pollution was seen in patients with small vessel or atherosclerotic large vessel stroke, but not cardioembolism, suggesting an effect at the level of the vessel, Dr. Wellenius said.
Extrapolating from the data, Dr. Wellenius calculated that an increase in 10 micrograms per cubic meter of fine pollutants increases the relative risk of stroke by 18 percent, far above estimates from previous studies, which have ranged from no effect to 6 percent. The greater increase, he said, is likely due to the greater accuracy with which symptom onset was recorded in his study. In other studies, he suggested, the effect of pollution would be diluted because patients may have had their strokes days before admission.
He also estimated that a 2 microgram reduction in mean PM2.5 across the board might have averted approximately 6100 stroke hospitalizations in 2007, or 3 percent of the 184,000 stroke hospitalizations in the Northeast for that year.
Leslie McClure, PhD, associate professor of biostatistics in the School of Public Health at the University of Alabama at Birmingham, told Neurology Today that among the strengths of the study was its attention to the effect of short-term exposures, and its focus on a relatively low-pollution environment. “One of the key messages in this paper is that they were looking at an area of the country where they're meeting the EPA standards, and yet they very clearly found associations between short-term exposure to small particulate matter and stroke. So even within the allowable limits of the law we are finding there is still room for improvement.”
Dr. McClure was the lead author on a study examining regional and racial differences in exposure to cigarette smoke in the REGARDS (Reasons for Geographic And Racial Differences in Stroke) study of stroke risk published in 2011 in the CDC journal, Preventing Chronic Disease.
Future studies should factor in socioeconomic status, she suggested. “We know there are large health disparities among people with large differences in socioeconomic status, and that people who live closer to factories and highways tend to be of lower socioeconomic status.” Whether those factors interact to affect risk should be addressed in future studies, she said.
If the results are replicated elsewhere, the question becomes whether a change in regulations should be considered. “That's a political decision,” Dr. Wellenius said. “If these results are replicated, there would be no doubt that further reductions in particular matter would reduce the risk of strokes,” as well as heart attacks, based on other studies. “But it is up to the politicians to put a price tag on that.”
DR. LESLIE MCCLURE said future studies should factor in whether socioeconomic status affects the associative risk, adding: “We know there are large health disparities among people with large differences in socioeconomic status, and that people who live closer to factories and highways tend to be of lower socioeconomic status.”
For the individual patient at risk, avoiding moderate levels of air pollution may be warranted. “But the most important thing is to reduce the traditional risk factors,” Dr. Wellenius said, including smoking. “Doing that will go a long way toward helping them not have the stroke in any case triggered by any mechanism.”
• McClure LA, Murphy HL, Malarcher A, et al. Regional and racial differences in smoking and exposure to secondhand smoke: the Reasons for Geographic and Racial Differences in Stroke (REGARDS) Study. 2011;8(5):A108