Causal relationship between obesity and diabetic neuropathy: A 2-sample Mendelian randomization study

A growing body of evidence suggests a potential correlation between obesity and diabetic neuropathy (DN), but the causal nature of this relationship remains uncertain. We conducted a Mendelian randomization (MR) analysis to investigate the causal relationship between obesity and DN. Independent single nucleotide polymorphisms strongly associated with DN were identified from genome-wide association studies. Summary statistics for body mass index (BMI), waist circumference (WC), hip circumference (HC), and waist-to-hip ratio (WHR) were also obtained from genome-wide association studies datasets. The causal effects were estimated using inverse-variance weighted analysis, MR-Egger regression, and weighted median estimation. Sensitivity analyzes were performed to assess the robustness of the findings. Inverse-variance weighted analysis revealed significant negative associations of BMI (P = .0006, OR = 0.809, 95% CI: 0.716-0.913), WC (P = .008, OR = 0.792, 95% CI: 0.665-0.943), and HC (P = .032, OR = 0.871, 95% CI: 0.768-0.988) with DN risk. No significant association was observed for WHR (P = .425, OR = 0.954, 95% CI: 0.849-1.071). After Bonferroni correction (α = 0.0125), there was a statistically significant relationship between BMI and DN (P ≤ .0125), indicating a causal relationship. However, the causal relationships between WC, HC, WHR, and DN were not statistically significant, with P-values all > .0125. However, the OR value of MR analysis is close to 1, indicating that BMI may have a slight protective effect on the risk of DN.