Causal Effects of Gastroesophageal Reflux on Chronic Rhinosinusitis: A Bidirectional Two-Sample Mendelian Randomization Study.

Abstract

Observational studies have shown a bidirectional association between gastroesophageal reflux (GER) and chronic rhinosinusitis (CRS) or chronic rhinitis (CR), but it is not clear whether this association is causal. This study was to investigate the causality between GER and CRS or CR using bidirectional two-sample Mendelian randomization (MR) analysis. Using pooled data from large genome-wide association studies (GWAS), genetic loci independently associated with GER, CRS and CR in populations of European and American ancestry were selected as instrumental variables (IVs). The inverse variance weighted (IVW) method was used to analyse the random effects model of MR, and the odds ratio (OR) was used as the evaluation index to explore the bidirectional causality between GER and CRS or CR. Single nucleotide polymorphism (SNP) outliers were detected using MR-pleiotropy Residual Sum and Outliers (MR-PRESSO). The MR-Egger intercept test examined the horizontal pleiotropy of SNPs. The "leave-one-out" sensitivity analysis examined whether MR results were affected by a single SNP. The main results of IVW showed that GER increased the risk of CRS (OR = 1.3795, 95% CI = 1.188-1.603, p < 0.0500) and CR (OR = 1.3941, 95% CI = 1.1671-1.6652, p < 0.0500). The obtained SNPs as IVs for GER, CRS and CR had no significant horizontal pleiotropy, heterogeneity or bias. Regarding the reverse directions, no notable associations could be found. This MR analysis revealed that genetically predicted GER had a causal effect on an increased risk of CRS or CR, but not vice versa. These results have great implications for the management of CRS (especially for refractory CRS) or CR in clinical practice.