Causal relationships between plasma lipids and sepsis: A Mendelian randomization study

Abstract

Although observational studies have indicated that plasma lipids are associated with an increased risk of sepsis, due to confounders and reverse causality, the causal relationship remains unclear. This study was designed to assess the causal effects of plasma lipid levels on sepsis. We used a 2-sample Mendelian randomization (MR) method to evaluate the causal effect of plasma lipids on sepsis. MR analysis employs methods such as inverse variance weighted, MR-Egger regression, weighted median regression (WME), simple mode and weighted mode. The inverse variance weighted (IVW) method was predominantly utilized to assess causality. Heterogeneity was affirmed by Cochran Q test, while pleiotropy was corroborated by MR-Egger regression analysis. The robustness and reliability of the results were demonstrated through "leave-one-out" sensitivity analysis. Instrumental variables included 226 single-nucleotide polymorphisms (SNPs), comprising of 7 for triglyceride (TG), 169 for high-density lipoprotein cholesterol (HDL-C), and 50 for low-density lipoprotein cholesterol (LDL-C). The risk of sepsis appeared to increase with rising LDL-C levels, as indicated by the inverse variance weighted analysis (OR 1.11, 95% CI from0.99 to1.24, P = 0.068). However, no causality existed between LDL-C, HDL-C, TG and sepsis. Two-sample MR analysis indicated that increased LDL-C level is a risk factor for sepsis, while TG and HDL-C levels have protective effects against sepsis. However, no significant causal relationship was found between TG, HDL-C, and LDL-C levels and sepsis.