The causal relationship between type 2 diabetes (T2D) and frozen shoulder is unclear. This study aims to explore the genetic causal association between T2D and glycemic traits (fasting glucose [FG], fasting insulin [FI], glycated hemoglobin [HbA1c], and 2-hour postprandial glucose [2hGlu]) on frozen shoulder.
Methods
Using 2-sample Mendelian randomization (MR), we analyzed nonconfounded estimates of the effects of T2D and glycemic traits on frozen shoulder. Single-nucleotide polymorphisms (SNPs) strongly associated (P < 5 × 10–8) with exposures from genome-wide association studies (GWAS) were identified. We employed fixed effect mode inverse variance weighting (IVW-FE), random effect mode IVW (IVW-MRE), MR-Egger, and weighted median to assess the association of exposures and outcome. Sensitivity analysis was conducted to test for heterogeneity and multidirectionality bias in MR.
Results
We found a significant genetic causal correlation between T2D (IVW-MRE P = .007, odds ratio [OR] 1.093, 95% confidence interval [CI] 1.03-1.16), FG (IVW-FE P < .001, OR 1.455, 95% CI 1.173-1.806), and frozen shoulder, but no evidence for causal correlation between FI, HbA1c, and 2hGlu and frozen shoulder. Although there was certain heterogeneity, sensitivity analysis reveals no deviation from the MR assumptions.
Conclusion
This study supports a genetic causal relationship between T2D and FG and frozen shoulder.