BACKGROUND: Panel studies have shown a consistent association between changes in the cardiac autonomic nervous system with particulate matters (PM) but less with gaseous pollutants. This study examined the linkage between nitrogen dioxide (NO2) and heart rate variability (HRV) in a susceptible population. METHODS: We recruited a panel of 83 patients from the National Taiwan University Hospital Cardiology Clinic to measure their 24-h HRV by ambulatory electrocardiography. Thirty-nine patients had coronary heart disease (CHD) and another 44 patients had more than one major CHD risk factor. Ambient concentrations of NO2, sulphur dioxide (SO2), carbon monoxide (CO), ozone, and PM less than 10 microm in diameter (PM10) at each participant\'s close-by monitoring station were used to represent study participants\' exposures. We used linear mixed-effects models to analyse the association between individual air pollutants and log10-transformed HRV, with key personal and environmental attributes and co-pollutants being adjusted. RESULTS: We found that an increase in 10 ppb NO2 at 4-h to 8-h moving averages was associated with 1.5-2.4% decreases in the standard deviation of all normal-to-normal intervals (SDNN) in our participants. For every 10 ppb NO2 at 5 and 7-h moving averages, our participants\' low frequency was decreased by 2.2 and 2.5%, respectively. In contrast, HRV was not associated with PM10, CO, SO2, or O3. CONCLUSION: Increasing NO2 exposure was found to be associated with decreasing SDNN and low frequency in susceptible populations.