The purpose of this study was to examine the correlation between previous arsenic exposure and peripheral vascular disease after stopping consumption of high-arsenic artesian well water for more than two decades in blackfoot disease endemic villages in Taiwan. A total of 582 adults (263 men and 319 women, aged 52.6 +/- 10.6 years) living in these villages underwent Doppler ultrasound measurement of systolic pressures on bilateral ankle (posterior tibial and dorsal pedal) and brachial arteries and estimation for long-term arsenic exposure. The diagnosis of peripheral vascular disease was based on an ankle-brachial index (the ratio between ankle and brachial systolic pressures) <0.90 on either side. Three indices of arsenic exposure were estimated: (1) duration of living in blackfoot disease endemic villages; (2) duration of artesian well water consumption; and (3) cumulative arsenic exposure in mg/l-years based on the detailed history of residential addresses and artesian well water consumption and the arsenic concentration in artesian well water. Multiple logistic regression analysis was used to assess the association between peripheral vascular disease and arsenic exposure. A dose-response relation was observed between the prevalence of peripheral vascular disease and the long-term arsenic exposure. The odds ratios (95% confidence intervals) after adjustment for age, sex, body mass index, cigarette smoking, serum cholesterol and triglyceride levels, diabetes mellitus and hypertension were 2.77 (0.84-9.14), and 4.28 (1.26-14.54) for those who had cumulative arsenic exposure of 0.1-19.9 and > or = 20.0 mg/l-years, respectively, compared with those who were not exposed. This study suggests a close relation between long-term arsenic exposure and peripheral vascular disease in blackfoot disease endemic villages in Taiwan after stopping consumption of artesian well water.