Relationship between BMI, indicators of lipid metabolism and diabetic neuropathy: a Mendelian randomization study.

Abstract

Background: To identify the relationship between BMI or lipid metabolism and diabetic neuropathy using a Mendelian randomization (MR) study.

Methods: Body constitution-related phenotypes, namely BMI (kg/m²), total cholesterol (TC), and triglyceride (TG), were investigated in this study. Despite the disparate origins of these data, all were accessible through the IEU OPEN GWAS database ( https://gwas.mrcieu.ac.uk/ ). Instrumental variables and F-statistics for each exposure-outcome pair were determined in weighted mode, weighted median, MR-Egger and Inverse-Variance Weighted (IVW) MR analyses. The p-value threshold was consistently set at 5.00E-08, following established methodology. The preliminary analysis utilized the IVW method to explore potential causal relationships between body constitution-related phenotypes and diabetic neuropathy. Inverse variance weighting, a technique amalgamating random variables, assigns weights inversely proportional to each variable's variance, commonly used for merging findings from independent studies. The weighted median method provides a causal estimate even when up to 50% of the instruments are invalid, enhancing robustness. The weighted mode method identifies the most common causal effect, reducing bias when some instruments exhibit horizontal pleiotropy. The Wald ratio method was utilized to calculate exposure-outcome effects, employing a range of methodologies to ensure result accuracy across different scenarios. This study addresses the critical gap in understanding the causal relationship between BMI, lipid metabolism, and diabetic neuropathy (DN). Employing a MR approach, it highlights BMI as a predictive factor for DN progression, providing insights into potential risk management strategies.

Results: IVW analysis showed that BMI (P = 0.033, OR = 2.53, 95% CI 1.08-5.96) and triglycerides level (P = 0.593, OR = 1.11, 95% CI 0.77-1.60) were positively associated with the initiation of DN, indicating that the values of BMI and triglycerides are potentially the risk factors in DN development. Additionally, TC was negatively associated with the DN (P = 0.069, OR = 0.72, 95% CI = 0.50-1.03).The forest plot of advanced analysis between BMI and DN relationship indicated a positive correlation between increasing BMI and the risk of DN. In addition, it is evident that with the increase in BMI, the risk of diabetic polyneuropathy also rises. This research demonstrates a positive association between BMI and DN risk (P = 0.033, OR = 2.53, 95% CI = 1.08-5.96). However, no significant correlation was observed between triglycerides (P = 0.593) or total cholesterol (P = 0.069) and DN development, underscoring the complex interplay between lipid metabolism and DN.

Conclusion: This research demonstrates a positive association between the risk of DN and BMI, while no significant correlation exists between TG or TC and the development of DN. These results imply that BMI may serve as a predictive factor for the progression of DN.