Abstract
Background: Whether glucocorticoids are beneficial in sepsis and related deaths has been controversial, and the exact causal relationship is unclear. This study used Mendelian randomization to investigate the causal relationships between glucocorticoids and sepsis and 28-day mortality of sepsis.
Methods: We obtained SNPs (p-value < 5 × 10-8) strongly associated with the exposure factor glucocorticoids (n=5440) from a genome-wide association study (GWAS) with glucocorticoids as the exposure factor and sepsis and 28-day mortality from sepsis as the outcome factor. Causality studies were conducted using variance weighted inverse methods supplemented with MR Egger, weighted median and weighted mode and MR simple mode analyses. The results were analyzed for heterogeneity and tested for horizontal multiplicity, leave-one-out sensitivity tests were performed and ORs were calculated for the results.
Results: A total of 26 qualified SNPs were ultimately selected as proxies for glucocorticoids. The results of the random effects from IVW revealed that glucocorticoids had no causal effect on sepsis(OR=0.989, 95% CI 0.895-1.093, P=0.833, se=0.051), 28-day mortality of sepsis(OR=0.921, 95% CI 0.768-1.104, P=0.375, se=0.093). These results were confirmed using the MR-Egger, weighted median, simple model, and weighted models.
Conclusions: In a two-sample Mendelian randomization analysis, no strong evidence was found to support a causal effect of glucocorticoids on sepsis, sepsis 28-day mortality, from a genetic perspective.