T-wave alternans, air pollution and traffic in high-risk subjects

Abstract

Particulate pollution has been linked to risk for cardiac death; possible mechanisms include pollution-related increases in cardiac electrical instability. T-wave alternans (TWA) is a marker of cardiac electrical instability measured as differences in the magnitude between adjacent T waves. In a repeated-measures study of 48 patients aged 43 to 75 years, associations of ambient and home indoor particulate pollution, including black carbon (BC) and reports of traffic exposure, with changes in 0.5-hourly maximum TWA (TWA-MAX), measured by 24-hour Holter electrocardiographic monitoring, were investigated. Each patient was observed up to 4 times within 1 year after percutaneous intervention for myocardial infarction, acute coronary syndromes without infarction, or stable coronary artery disease, for a total of 5,830 0.5-hour observations. Diary data for each 0.5-hour period defined whether a patient was home or not home, or in traffic. Increases in TWA-MAX were independently associated with the previous 2-hour mean ambient BC (2.1%, 95% confidence interval 0.9% to 3.3%) and with being in traffic in the previous 2 hours (6.1%, 95% confidence interval 3.4% to 8.8%). When subjects were home, indoor home BC effects were largest and most precise; when subjects were away from home, ambient central site BC effects were strongest. Increases in pollution increased the odds of TWA-MAX > or =75th percentile (odds ratio 1.4, 95% confidence interval 1.2 to 1.6 for a 1 microg/m(3) increase in 6-hour mean BC). In conclusion, after hospitalization for coronary artery disease, being in traffic and short-term ambient or indoor BC exposure increased TWA, a marker of cardiac electrical instability.

Figure 1

Twenty-four hour estimated circadian pattern of TWA-MAX, plotted as the mean of TWA-MAX values for each half-hour period of the day, controlling for subject, day of the week, being in traffic, average heart rate, hour of the day, date, mean temperature, and BC. The curve and pointwise 95% confidence interval (dotted line) are estimated using a penalized spline.

Figure 2

The percent change in maximum TWA for increasing averaging times for ambient PM_2.5 and BC. PM_2.5 effects are scaled to 10 μg/m³; BC effects are scaled to 1 μg/m³.