2003 Progress Report: Relationship of Ambient Particulate Matter to Heart Rate Variability and Cardiac Arrhythmias in Elderly Adults with Coronary Artery DiseaseEPA Grant Number: R826783
Title: Relationship of Ambient Particulate Matter to Heart Rate Variability and Cardiac Arrhythmias in Elderly Adults with Coronary Artery Disease
Investigators: Ostro, Bart , Tager, Ira , Woo, Mary , Lipsett, Michael , Bolton, Merle
Institution: Public Health Institute , California Office of Environmental Health Hazard Assessment , University of California - San Francisco , Eisenhower Medical Center , University of California - Berkeley , University of California - Los Angeles
Current Institution: Public Health Institute , California Office of Environmental Health Hazard Assessment , Eisenhower Medical Center , University of California - Berkeley , University of California - Los Angeles , University of California - San Francisco
EPA Project Officer: Chung, Serena
Project Period: October 1, 1998 through September 30, 2000
Project Period Covered by this Report: October 1, 2002 through September 30, 2003
Project Amount: $436,964
RFA: Health Effects of Particulate Matter and Associated Air Pollutants (1998) RFA Text | Recipients Lists
Research Category: Air Quality and Air Toxics , Human Health , Particulate Matter , Air
Epidemiological studies have demonstrated consistent associations between exposure to ambient particulate matter (PM) and increased daily morbidity and mortality. Mortality effects appear to be greatest among elderly people with preexisting cardiorespiratory diseases, especially ischemic heart disease. We reported previously that exposures to PM are associated with an increased risk of daily mortality, including deaths from cardiovascular diseases, in the Coachella Valley, California, a desert resort area where the PM mass historically has been dominated by coarse particles. Assuming a causal relationship, the mechanisms underlying reported associations between exposures to ambient PM and daily mortality are unknown but may involve disturbances of the autonomic nervous system and the electrical activity of the heart. Decreased heart rate variability (HRV) measured over 24-hour intervals has been linked to an increased risk of cardiovascular mortality in several subsets of patients with cardiovascular disease. The objective of this research project was to examine whether exposures to ambient PM10, PM2.5, and the coarse fraction (PM10-PM2.5) were associated with altered cardiac autonomic nervous system regulation, measured as decreased HRV, in a population of older adults with documented coronary artery disease. The specific objectives of this research project were the following:
- In older adults with established coronary artery disease, are exposures to elevated levels of PM10, PM2.5, the coarse fraction, or ultrafine particles associated with altered cardiac autonomic nervous system regulation, measured as decreased HRV?
- In older adults with established coronary artery disease, are exposures to elevated levels of PM10, PM2.5, the coarse fraction, or ultrafine particles associated with increases in the number and frequency of specific cardiac arrhythmias?
To assess HRV in this study population, we conducted serial 24-hour electrocardiographic (Holter) monitoring at weekly intervals for up to 12 consecutive weeks. After first assessing the feasibility of such monitoring in a pilot study in spring 1999, we conducted the full study among 19 adults, ages 61-81, from February through May 2000. Exposure assessment for these analyses consisted of real-time monitoring of ambient PM10 and PM2.5, ozone, nitrogen dioxide, and carbon monoxide at two fixed-site monitors located in Palm Springs and Indio, at either end of the population corridor in the Coachella Valley. (Technical problems with the ultrafine particle monitoring data made the latter unsuitable for use in this analysis.) In addition, we undertook limited concurrent indoor and outdoor particle measurements at a subset of the subjects’ homes.
We used random-effects models (generalized least squares estimators) to examine relationships between the various pollutant metrics and changes in HRV, after examining for possible confounding by meteorological factors. Although the Holter monitoring took place over 24-hour intervals, the time-domain and geometric HRV estimates (standard deviation of normal-to-normal beats [SDNN], standard deviation of the means of all normal-to-normal intervals within successive 5-minute blocks [SDANN], root mean square successive differences [rMSSD], and triangular index [TRII]) were calculated over two 2-hour periods chosen to maximize contrast in particle levels (6:00-8:00 a.m. and p.m., separately). Frequency-domain HRV variables were estimated over two 5-minute intervals chosen to minimize interindividual variability in activity (the second 5 minutes after the monitor hook-up, when the subjects were resting comfortably, and a 5-minute interval at 3:00 a.m., when the subjects would likely be asleep). We also examined relationships between pollutant variables and ventricular and supraventricular tachyarrhythmias.
The results suggested associations between decrements in several time-domain and geometric HRV variables (SDNN, SDANN, and TRII) over the two-evening monitoring period in relation to both PM10 and the coarse fraction but not PM2.5. For both PM10 and coarse particles, the magnitude of these associations increased as the averaging time increased for up to 4 hours but diminished to nonsignificance when the pollutant averaging time was extended to the prior 24 hours. There was no association of the particle metrics with decrements in rMSSD and no evidence of an association between the pollutant gases and these time-domain and geometric variables during the evening interval.
There were fewer associations during the morning monitoring period between pollutant metrics and time-domain variables. The coarse fraction was associated with both SDNN and SDANN at lags up to 4 hours but not at 24 hours. PM10, PM2.5, and ozone were not associated with any HRV metrics during the 2-hour morning period.
Analysis of the frequency-domain variables during sleep (3:00 a.m.) also indicated an association between HRV and particulate metrics measured during the preceding 1 to 4 hours. Total power was associated with all three particulate metrics. The strongest associations for PM10 and the coarse fraction were averaged over the prior 4 hours. Ozone measured over 8 hours also was associated with decreased total power, though the coefficient was of marginal statistical significance. There were several borderline associations with changes in the high and low frequency components, with no obvious pattern.
The analysis of frequency-domain variables calculated during the second 5 minutes of the monitoring period, after the initial Holter hook-up, indicated associations of 1-hour, but not 8-hour, ozone concentrations with decreases in total power, as well as both the high and low frequency components. In contrast to all of the other analyses, however, there was no discernible pattern of association between any of the particulate metrics and any measure of HRV.
Refinement of exposure through the use of indicator variables and interaction terms for air conditioning appeared to increase the association of decreased time-domain HRV variables with PM10 and more modestly for coarse particles. Other individual behaviors that potentially could affect exposures, including opening windows or going outside for more than an hour, did not significantly alter the PM-HRV relationships identified without adjustment for these factors.
One ancillary objective of this research project was to examine which subject characteristics might be associated with an increased susceptibility to PM-related decreases in HRV. The principal factors that were so associated were left ventricular ejection fraction, an important clinical measure of heart function, and pulmonary function. Gender and use of beta-blockers, a class of cardiovascular drugs commonly administered to cardiac patients, were both less strongly associated with pollutant-related HRV decrements. Nonetheless, the small number of subjects involved in this analysis warrants a caveat that these findings should be viewed as exploratory only.
The analyses of potential relationships of various pollutant metrics with ventricular and supraventricular arrhythmias during the evening 2-hour interval were constrained by relatively few outcome events. Therefore, although there were no statistically significant associations detected in these analyses, this may have been a function of limited statistical power.
Analysis of indoor versus outdoor concentrations of fine and coarse particles indicated lower levels of coarse versus fine particles indoors. Indoor concentrations of coarse particles were lower than those outdoors in this data set, with the average indoor-to-outdoor mass concentration ratio equal to about 0.66 ( ± 0.27). There was a modest correlation between the outdoor and indoor coarse particle data (R2 = 0.35), suggesting that a substantial fraction of this PM subset likely was to have been generated by indoor sources and occupant activities, such as vacuuming, dusting, cleaning, walking, and other behaviors that could result in the resuspension of settled particles. Outdoor PM2.5 concentrations, however, could explain only about 37 percent of the variability of the indoor concentrations, suggesting that (as with the coarse fraction) there may have been significant contributions by indoor sources to the overall PM2.5 concentrations in the subjects’ residences as well. These results indicate the possibility of similar degrees of exposure misclassification for these two PM subfractions for individuals spending most of their time indoors, suggesting in turn that differential exposure misclassification would not be a likely explanation of our results.
Several previous reports have linked exposures to ambient and indoor fine particles with decreases in HRV; to our knowledge this is the first report linking exposure to ambient coarse particles to decrements in HRV in individuals with established heart disease. The clinical significance of small PM-associated changes in HRV remains to be established.
- Both PM10 and the coarse fraction consistently were associated with decrements in time-domain (SDNN and SDANN) and geometric (TRII) measures of HRV. These particle metrics also were associated with decrements of frequency-domain variables (total power as well as high and low frequency components).
- PM2.5 was either not associated or less strongly associated with decreases in geometric, time-, and frequency-domain HRV variables.
- Ozone was associated with decrements in frequency-domain variables only.
- The associations between the pollutant variables and decrements in HRV tended to be strongest when the pollutant data were obtained either concurrent with or within the few hours preceding the HRV measurement.
- The HRV changes observed in this investigation are slightly lower than, but of the same order of magnitude as, those observed in other studies of air pollution and HRV.
- Left ventricular ejection fraction and pulmonary function were the two individual-level characteristics most strongly associated with increased susceptibility to particle-associated HRV decrements. Because of the relatively small number of subjects and observations, these observations need to be evaluated in other populations as well.
- Refinement of exposure through the use of indicator variables and interaction terms for air conditioning appeared to increase the association of decreased time-domain HRV variables with PM10 and more modestly for coarse particles. Other individual behaviors that potentially could affect exposures, including opening windows or going outside for more than an hour, did not significantly alter the PM-HRV relationships identified without adjustment for these factors. In addition, the null effect of PM2.5 was generally not affected by these adjustments.
- There were no significant associations between pollutant metrics and either ventricular or supraventricular tachyarrhythmias. This analysis was constrained, however, by the few outcome events, likely resulting in low statistical power.
- Whereas the particle concentrations observed in this study were relatively high overall, the PM2.5/PM10 ratio, both at the Palm Springs monitor and inside and outside a subset of the subjects’ residences, was greater than would have been expected from prior monitoring efforts in the Valley. Indoor/outdoor correlations, however, were comparable for fine and coarse particles, suggesting that for individuals spending a significant portion of their time indoors, differential measurement error would be an unlikely explanation of our findings.
The clinical significance of these findings of PM-associated HRV decrements remains to be established. Most studies linking decreases in HRV with increased cardiac mortality have examined 24-hour baseline HRV as a predictor of adverse events over the course of months to years. Although it is possible that our results may offer a partial mechanistic explanation for the repeated observations in time-series studies of PM-related cardiovascular mortality, it also is possible that the acute decrements in HRV represent an epiphenomenon of some other unmeasured, underlying pathophysiologic process.
Journal Articles on this Report : 2 Displayed | Download in RIS Format
|Other project views:||All 6 publications||2 publications in selected types||All 2 journal articles|
||Geller MD, Chang M, Sioutas C, Ostro BD, Lipsett MJ. Indoor/outdoor relationship and chemical composition of fine and coarse particles in the southern California deserts. Atmospheric Environment 2002;36(6):1099-1110.||
||Lipsett MJ, Tsai FC, Roger L, Woo M, Ostro BD. Coarse particles and heart rate variability among older adults with coronary artery disease in the Coachella Valley, California. Environmental Health Perspectives 2006;114(8):1215-1220||