Type 1 diabetes mellitus increases the risk of sudden sensorineural hearing loss: A two-sample Mendelian randomization study.

OBJECTIVES

Diabetes mellitus is closely associated with sudden sensorineural hearing loss (SSNHL), but no definitive evidence has established a causal relationship between type 1 diabetes mellitus (T1DM) and SSNHL. This study aims to investigate the impact of T1DM on SSNHL from a genetic perspective, providing insights for risk prediction and treatment strategies.

METHODS

Genetic data related to exposure (T1DM) and outcome (SSNHL) were obtained from publicly available genome-wide association studies (GWAS). Instrumental variables were selected, and Mendelian randomization (MR) analysis was conducted to explore the causal association between T1DM and SSNHL. Inverse variance weighted (IVW) analysis was used as the primary method, with random-effects IVW serving as the main analytical approach. MR-Egger, weighted median, simple mode, and weighted mode analyses were utilized as supplementary methods. Cochran's test was applied to evaluate the heterogeneity of the selected instrumental variables, MR-PRESSO was applied to detect outliers, MR-Egger regression was used to assess horizontal pleiotropy and leave-one-out analysis was conducted to examine the robustness of individual single nucleotide polymorphisms (SNPs) on the overall results.

RESULTS

A total of 127 SNPs were selected as instrumental variables for the MR analysis. IVW analysis demonstrated a genetically determined association between T1DM and SSNHL (=1.036, 95% 1.002 to 1.071, =0.038). Forest plots and scatter plots indicated a causal relationship, suggesting that T1DM increases the risk of SSNHL. Cochran's test demonstrated no significant heterogeneity among SNPs (MR-Egger: =126.030, =0.356; IVW: =126.450, =0.373). The funnel plot appeared symmetrical, indicating that the selected instrumental variables were primarily related to exposure rather than potential confounding factors. The MR-Egger intercept was not significantly different from zero (=0.527), indicating no evidence of horizontal pleiotropy among the SNPs. MR-PRESSO analysis did not identify any outlier SNPs (=0.356). Leave-one-out analysis confirmed the robustness of the findings, as the results remained stable after removing individual SNPs.

CONCLUSIONS

Two-sample MR analysis supports the conclusion that T1DM patients have an increased risk of developing SSNHL.