POS-343 BODY MASS INDEX AND DIABETIC NEPHROPATHY: A MENDELIAN RANDOMIZATION STUDY

Abstract

The association between obesity and diabetic nephropathy (DN) has been studied but the results of these epidemiological studies are not entirely consistent. The causal effect of body mass index (BMI) on DN and kidney traits in type 2 diabetes was evaluated by a two-sample Mendelian randomization (MR) analysis. A total of 59 independent genetic variants were selected as instrumental variables for BMI in 158,284 participants from Biobank Japan, and their effects on DN risk, eGFR and proteinuria were estimated in 3972 Chinese individuals with type 2 diabetes. Then, subgroup analysis was performed with different BMI levels or sex. In addition, the sensitivity analyses were performed including MR-Egger regression for detecting genetic directional pleiotropic effects, and power calculation for evaluating the reliability of the study. The generalized summary-data-based MR (GSMR) method showed that a one standard deviation increase in BMI was causally associated with a 3-fold increase in the risk of DN (odds ratio [OR], 4.05; 95% confidence interval [95% CI], 2.10 to 7.82; P<0.001), and a 35% decrease of eGFR level (OR, 0.65; 95% CI, 0.54 to 0.78; P<0.001). However, BMI was not associated with the level of proteinuria (OR, 1.26; 95% CI, 0.89 to 1.79; P=0.19). In addition, subgroup analysis showed a positive association between BMI and the risk of DN in individuals with different BMI levels (BMI≥25 vs BMI<25 kg/m2) or sex. Analyses stratified by sex indicated the casual effect of BMI on DN was stronger in women (OR, 9.82; 95% CI, 1.83 to 52.67; P=0.008) than in men (OR, 3.39; 95% CI, 1.23 to 9.34; P=0.02). In sensitivity analyses, the MR-Egger regression intercepts indicated that the results were less likely to be affected by pleiotropy. Based on 1,314 DN cases and 2,658 controls, our study had ≥80% power to detect the causal effect of BMI on DN risk if the genuine OR≥1.69 per standard deviation increase of BMI. Genetic evidence showed that higher BMI levels was causally associated with increased risk of DN and decreased eGFR level, but not with proteinuria. The DN risk for women was more susceptible to BMI levels than men.