Abstract

BACKGROUND: Previous epidemiological evidence on the sex-specific association of obesity, particularly abdominal obesity, as reflected by larger waist circumference (WC), with incident diabetic neuropathy (DN) remains limited. METHODS: We used data from a patient cohort with a median 10-year history of type 2 diabetes mellitus at the time of recruitment. A composite outcome of four types of predefined DN (Neuro 1-4) was used as the outcome of interest. Because of sex differences in abdominal obesity, analyses were conducted separately for men and women. RESULTS: Among the 7442 participants (4551 men and 2891 women) recruited in the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study, 3999 cases of incident DN were documented (2529 in men and 1470 in women). Larger WCs were associated with a higher risk of DN among both men and women. Compared with the lowest quintile, the hazard ratio (HR) for the highest quintile was 1.30 (95% CI 1.13-1.49) among men (P_trend <.001). For women, the HR for the highest vs lowest quintile was 1.25 (95% CI 1.04-1.51) (P_trend <.001). A linear relationship between WC and DN was observed in men, and in women, the risk of DN increased as the WC quintile increased before it appeared to plateau. The relationship between body mass index and incident DN was similar to the results observed for WC. CONCLUSION: General and abdominal obesity were both associated with an increased risk of incident DN among individuals with type 2 diabetes mellitus, regardless of sex.