OBJECTIVES: Mass poisoning by carbon disulfide (CS2) occurred in a viscose rayon factory in Korea. Up until 1998, 830 employees, including 38 who had died, were diagnosed with CS2 poisoning. Among the CS2 poisoned subjects, heart rate variability (HRV) was evaluated to investigate whether the toxic effect of CS2 persists after the exposure has ceased. The dose-response relationship between carbon disulfide exposure and HRV was also evaluated. METHODS: The case group was comprised of 71 retired male workers diagnosed as being CS2 poisoned. The control group was comprised of 127 males of same age-range who had no history of CS2 exposure and cardiovascular diseases. Information on individual age, height, weight, cigarette smoking, alcohol drinking, regular exercise, medical and occupational history, chest x-ray, and ECG recording of the two groups were collected through a self-administered questionnaire and with a medical examination. Time (maximum, average, minimum RR interval) and frequency domain measures (low frequency - LF, high frequency - HF, total power spectrum TPS, and LF/HF ratio) of the two groups were analyzed. CS2 exposure indices of the case group (duration of employment, exposure level per work department, cumulative exposure index and duration of retirement) were investigated. RESULTS: Using a univariate analysis, the frequency domain measures for the case group were significantly lower than those in the control group, except for HF. In the multivariate analysis, previous history of CS2 poisoning was inversely related to all frequency domain parameters and it significantly affected the LF (p<0.05) and the LF/HF ratio (p<0.05). There was no significant dose-response relationship between CS2 exposure indices and HRV parameters in the case group. CONCLUSION: This study suggests that further studies are necessary to evaluate the residual effects of CS2 poisoning even after the CS2 exposure has ceased.