Urine arsenic and risk of acute myocardial infarction, stroke, and heart failure – A prospective case-cohort study in Danish never smokers

Objective: To examine the association of low-level urine As with incident stroke, acute myocardial infarction (AMI), and heart failure (HF) among never smokers. Methods: Between 1993 and 1997, the Danish Diet, Cancer and Health (DCH) cohort enrolled 19,394 participants, ages 50–64 yrs, who reported never smoking. Among these never smokers with baseline urine samples available for arsenic speciation, we identified incident cases of stroke (N = 457), AMI (N = 696), and HF (N = 803) using the Danish National Patient Registry. We also randomly selected a sub-cohort of 541 men and 548 women. We used modified Cox proportional hazards models to estimate adjusted hazard ratios (HR) for each incident CVD outcome in relation to urine As adjusted for creatinine. Results: We did not find an association with stroke, AMI, or HF across unadjusted and adjusted models using our primary exposure metric, sum of As species (iAs + MMA + DMA), AsB-corrected. We also did not find evidence of an association of secondary metrics (%DMA, %MMA) with AMI or HF, although HRs were consistently elevated for upper quartile %MMA. We observed associations for stroke with upper quartile %DMA (HR = 0.57; 95 % CI: 0.40–0.82), and with third quartile %MMA (HR = 1.52; 95 % CI: 1.07–2.20) but not the upper quartile %MMA (HR = 1.20; 95 % CI: 0.85–1.70). Conclusions: We found little evidence of an association between low-level urine As species and AMI, HF, or stroke among Danish never smokers, although further research is needed to disentangle a potential cardiovascular role of %MMA.