Weight and Mortality in Adults With Diabetes

To the Editor: Dr Carnethon and colleagues reported higher total and noncardiovascular mortality in adults with diabetes who were normal weight vs overweight or obese. Close examination of the data suggests that there is a potential limitation in the analysis that might have affected the conclusions. The absence of association between weight status and cardiovascular mortality was attributed to the small number of observed cardiovascular events. However, participants who had comorbid hypertension at the time of incident diabetes were not excluded from the analysis. The prevalence of hypertension was higher in overweight and obese participants than in the cohort with normal weight, leading to higher risk of cardiovascular events in the former. Also, systolic blood pressure rather than hypertension was adjusted for in the multivariable models, which may have attenuated the association of weight status with cardiovascular mortality. In addition, obesity is an independent risk factor for the development of cardiovascular diseases. Although this study excluded patients who had preexisting cardiovascular disease at baseline, the morbidity of cardiovascular disease might be higher in overweight and obese patients than in patients with normal weight during follow-up of the study. This may have attenuated the obesity paradox that patients with normal weight had higher cardiovascular mortality than overweight and obese patients. Previous studies found the obesity paradox between body mass index and cardiovascular mortality because patients who had preexisting cardiovascular disease were not excluded.