Abstract
BACKGROUND: To assess the causal role of lipid traits and lipid-lowering agents in inflammatory bowel disease (IBD). METHODS: Univariable mendelian randomization (MR) and multivariable MR (MVMR) analyses were conducted to evaluate the causal association between low-density lipoprotein cholesterol (LDL-C), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C) and IBD. Drug-targeted MR analyzed the effects of lipid-lowering drugs on IBD, and network MR was used to analyze potential mediation effects. RESULTS: The levels of HDL-C had an inverse relationship with the risk of Crohn's disease (CD, OR: 0.85, 95% CI: 0.73-0.98, P = 0.024). In MVMR, the inverse relationships were found in all three outcomes. Drug-targeted MR analyses showed that with one-SD LDL-C decrease predicted by variants at or near proprotein convertase subtilisin/kexin type 9 (PCSK9), the OR values of people diagnosed with IBD, ulcerative colitis (UC) and CD were 1.75 (95%CI: 1.13-2.69, P = 0.011), 2.1 (95%CI: 1.28-3.42, P = 0.003) and 2.24 (95%CI: 1.11-4.5, P = 0.024), respectively. With one-SD LDL-C decrease predicted by variants at or near cholesteryl ester transfer protein (CETP), the OR value of people diagnosed with CD was 0.12 (95%CI: 0.03-0.51, P = 0.004). Network-MR showed that HDL-C mediated the causal pathway from variants at or near CETP to CD. CONCLUSION: Our study suggested a causal association between HDL-C and IBD, UC and CD. Genetically proxied inhibition of PCSK9 increased the risk of IBD, UC and CD, while inhibition of CETP decreased the risk of CD. Further studies are needed to clarify the long-term effect of lipid-lowering drugs on the gastrointestinal disorders.