Arsenic Exposure, Arsenic Metabolism, and Glycemia: Results from a Clinical Population in New York City.

Little information is available regarding the glycemic effects of inorganic arsenic (iAs) exposure in urban populations. We evaluated the association of total arsenic and the relative proportions of arsenic metabolites in urine with glycemia as measured by glycated blood hemoglobin (HbA1c) among 45 participants with prediabetes (HbA1c ≥ 5.7-6.4%), 65 with diabetes (HbA1c ≥ 6.5%), and 36 controls (HbA1c < 5.7%) recruited from an academic medical center in New York City. Each 10% increase in the proportion of urinary dimethylarsinic acid (DMA%) was associated with an odds ratio (OR) of 0.59 (95% confidence interval (CI): 0.28-1.26) for prediabetes, 0.46 (0.22-0.94) for diabetes, and 0.51 (0.26-0.99) for prediabetes and diabetes combined. Each 10% increase in the proportion of urinary monomethylarsonic acid (MMA%) was associated with a 1.13% (0.39, 1.88) increase in HbA1c. In contrast, each 10% increase in DMA% was associated with a 0.76% (0.24, 1.29) decrease in HbA1c. There was no evidence of an association of total urinary arsenic with prediabetes, diabetes, or HbA1c. These data suggest that a lower arsenic methylation capacity indicated by higher MMA% and lower DMA% in urine is associated with worse glycemic control and diabetes. Prospective, longitudinal studies are needed to evaluate the glycemic effects of low-level iAs exposure in urban populations.