OBJECTIVE: Fine particulate matter (PM2.5) air pollution and variations in ambient temperature have been linked to increased cardiovascular morbidity and mortality. However, no large-scale study has assessed their effects on directly measured aerobic functional capacity among high-risk patients. METHODS: Using a cross-sectional observational design, we evaluated the effects of ambient PM2.5 and temperature levels over 7 days on cardiopulmonary exercise test results performed among 2078 patients enrolling into a cardiac rehabilitation programme at the University of Michigan (from January 2003 to August 2011) using multiple linear regression analyses (controlling for age, sex, body mass index). RESULTS: Peak exercise oxygen consumption was significantly decreased by approximately 14.9% per 10 mug/m(3) increase in ambient PM2.5 levels (median 10.7 mug/m(3), IQR 10.1 mug/m(3)) (lag days 6-7). Elevations in PM2.5 were also related to decreases in ventilatory threshold (lag days 5-7) and peak heart rate (lag days 2-3) and increases in peak systolic blood pressure (lag days 4-5). A 10 degrees C increase in temperature (median 10.5 degrees C, IQR 17.5 degrees C) was associated with reductions in peak exercise oxygen consumption (20.6-27.3%) and ventilatory threshold (22.9-29.2%) during all 7 lag days. In models including both factors, the outcome associations with PM2.5 were attenuated whereas the effects of temperature remained significant. CONCLUSIONS: Short-term elevations in ambient PM2.5, even at low concentrations within current air quality standards, and/or higher temperatures were associated with detrimental changes in aerobic exercise capacity, which can be linked to a worse quality of life and cardiovascular prognosis among cardiac rehabilitation patients.