Kramer et al. (1) studied whether serum uric acid (UA) predicts incident type 2 diabetes by glucose tolerance status in older community-dwelling adults. The regression models adjusted for age, sex, BMI, diuretic use, and estimated glomerular filtration rate showed that UA levels independently predicted incident type 2 diabetes among participants who had impaired fasting glucose (odds ratio 1.75 [95% CI 1.1–2.9], P = 0.02) (1). However, relationships between serum UA and type 2 diabetes are controversial.
Taniguchi et al. (2) investigated the association between serum UA levels and incident diabetes in a prospective cohort study of 6,365 Japanese male employees aged 35–61 years and reported that UA was not significantly associated with incident diabetes. Chien et al. (3) conducted a community-based prospective cohort study of a median 9-year follow-up among 2,690 Chinese subjects aged 35–97 years and reported that the adjusted relative risk (RR) for diabetes in the fifth quintile of serum UA was 1.63 (95% CI 1.20–2.23) compared with the first quintile and, after additional adjustment for metabolic syndrome, was 1.40 (1.02–1.92). However, after further adjustment for homeostasis model assessment of insulin resistance index, the adjusted RR for diabetes in the 5th quintile of serum UA became 1.22 (0.85–1.74) (3). Chien et al. did not study men and women separately. Dehghan et al. (4) suggested that serum UA is an independent risk factor for diabetes; however, they also did not study men and women separately, though serum levels of UA are substantially different between men and women. In our cross-sectional study in 2,449 Japanese men and 1,448 Japanese women (5), serum UA was positively associated with metabolic syndrome but negatively associated with diabetes in men, not in women. We would deeply appreciate it if the authors would re-study the data separately in men and women to find whether UA levels independently predict incident diabetes in men.
