Abstract
BACKGROUND: Arsenic has been associated with diabetes and impaired glucose tolerance in many studies, although some reports have shown null findings. METHODS: We conducted a cross-sectional study among 300 adults in Bangladesh. Participants were randomly selected from a roster of 1800 people who previously participated in studies of arsenic and skin lesions. We measured fasting glucose and insulin levels. We assessed drinking water arsenic concentration using graphite furnace atomic absorption spectrophotometry (GF-AAS) and toenail arsenic concentration using inductively coupled mass spectrometry (ICP-MS). We ran covariant-adjusted, linear regression and spline models to examine associations of arsenic concentrations with the homeostatic model assessment of insulin resistance (HOMA-IR), a marker of insulin resistance, and HOMA of beta-cell function (HOMA-β), a marker of beta-cell function. RESULTS: Among 285 participants with complete data, the median (IQR) arsenic concentration was 4.0 (6.9) μg/g in toenails and 39.0 (188.5) μg/L in drinking water. Arsenic concentrations were not associated with insulin resistance or beta-cell function. HOMA-IR was 0.67% lower and HOMA-β was 0.28% lower per µg/g increment in toenail arsenic, but these effect estimates were small, and confidence intervals crossed the null value. CONCLUSIONS: Although arsenic exposure has been associated with diabetes, we found no evidence of an adverse effect on insulin resistance or beta-cell function.