Rheumatol Ther. 2022 Dec 1. doi: 10.1007/s40744-022-00510-4. Online ahead of print.
ABSTRACT
INTRODUCTION: The association between diabetes mellitus (DM) and risk of osteoarthritis (OA) is inconsistent based on published observational studies. This study aimed to conduct a two-sample Mendelian randomization (MR) analysis to explore the causal link between glycated hemoglobin (HbA1c) level and OA risk.
METHODS: Genome-wide association studies (GWAS) summary statistics were obtained from the publicly available Integrative Epidemiology Unit (IEU) OpenGWAS database. A series of screening processes were performed to select qualified instrumental single-nucleotide polymorphisms (SNPs) strongly related to exposure. The inverse-variance-weighted method, weighted-median method, and MR-Egger method were performed to ensure robust and reliable results. The MR-Egger intercept test, Cochran’s Q test, and the leave-one-out sensitivity analysis were utilized to assess the horizontal pleiotropy, heterogeneities, and stability of these genetic variants for OA. Odds ratio (OR) and 95% confidence interval (CI) were calculated.
RESULTS: MR analyses found a robust causal association of genetically determined HbA1c with knee OA (OR = 1.561; 95% CI 1.110-2.197; P = 0.011), but not with hip OA (OR = 1.073; 95% CI 0.674-1.710; P = 0.766) or overall OA (OR = 1.141; 95% CI 0.904-1.441; P = 0.804). Sensitivity analyses showed that there was a strong association between SNPs and HbA1c (F = 21.138), no evidence of heterogeneity (Q = 150.625, P = 0.402), and no potential SNPs affecting the causal link.
CONCLUSION: Our MR study supported a causal effect of genetically increased HbA1c on knee OA risk.
PMID:36451030 | DOI:10.1007/s40744-022-00510-4
