Causal association between type 1 diabetes mellitus and inflammatory bowel disease: A two-sample bidirectional Mendelian randomization study

Abstract

Abstract Background: Type 1 diabetes mellitus (T1DM) and inflammatory bowel disease (IBD) are common autoimmune diseases. Previous studies reported an association between these two diseases. However, due to the limitations of observational research, it cannot provide sufficient evidence to support the existence of causality or reverse causality. Objective: This study aimed to evaluate the existence of the causal relationship between T1DM and IBD. Methods: A two-sample bidirectional Mendelian randomization (MR) was employed to investigate a possibility of the causal relationship between T1DM and IBD, including Crohn's disease (CD) and ulcerative colitis (UC). Independent genetic variants of T1DM and IBD were retrieved from the IEU OpenGWAS project，and the individuals were of European ancestry. After filtering the single nucleotide polymorphisms (SNPs), several MR methods, including MR–Egger regression, weighted median, inverse variance-weighted (IVW), and weighted mode with IVW as the primary analysis method, were used to test the causal association between the two diseases. Furthermore, a sensitivity analysis was conducted to detect the heterogeneity and pleiotropy of the instrumental variables. Results: Bidirectional MR analysis indicated a negative correlation between T1DM and IBD (IVW OR: 0.95, 95% Confidence Interval [CI]: 0.94 to 0.97, P = 1.19E-12). However, a separate analysis of CD and UC showed that this negative association appeared to exist only for T1DM and CD (IVW OR: 0.95, 95% CI: 0.91 to 0.99, P = 0.0063) and not for UC (IVW OR: 1.02, 95% CI: 0.98 to 1.05, P = 0.3315). In contrast, we found no difference in the prevalence of T1DM among patients with IBD, CD, and UC compared to the general population. Conclusions: This study using MR has confirmed that individuals diagnosed with T1DM exhibit a decreased susceptibility to IBD in comparison to the general population，indicating that new therapeutic targets for IBD based on the pathogenesis of T1DM should be explored.