Comment on Menke et al. Metals in Urine and Diabetes in U.S. Adults. Diabetes 2016;65:164-171.

It is with great interest that we noted the important work by Menke et al. (1) on the relationship between metal excretion and the incidence of type 2 diabetes. We would like to draw attention to the potential confounders when interpreting data by relying on urinary metal levels as a proxy for environmental exposure to examine potential associations with overt diabetes. Specifically, it has previously been shown that subtle changes in renal physiology as well as changes in insulin and glucose levels result in changes in renal excretion rates of cadmium and other metals (2–10). Therefore, changes in renal physiology that are known to develop early in the course of frank diabetes (11) may present a confounder when examining the association between overt diabetes and urinary metal excretion. For this reason, we believe that including prediabetes as an outcome variable in parallel with type 2 diabetes when examining potential associations between exposure to nonessential transition metals and dysglycemia is a sensible approach to overcome this limitation given the high rate of progression from prediabetes to overt diabetes. Using this approach, we previously reported a nonlinear association between prediabetes and urinary cadmium excretion in the National Health and Nutrition Examination Survey (NHANES) 2005–2010 data set (12).