Obstructive sleep apnea and 19 gastrointestinal diseases: a Mendelian randomization study

Abstract

Background: Alterations gastrointestinal diseases (GDs) were reported in individuals with obstructive sleep apnea (OSA), however, the genetic background between OSA and GDs is still unclear.

Methods: This investigation employed Mendelian randomization (MR) analyses to evaluate the causal effect between OSA and 19 types of GDs (gastroesophageal reflux disease (GERD), ulcerative colitis, celiac disease, Crohn's disease, chronic gastritis, irritable bowel syndrome, primary biliary cholangitis, diverticular disease, gastroduodenal ulcer, acute pancreatitis, non-alcoholic fatty liver disease, primary sclerosing cholangitis, cirrhosis, calculus of bile duct, calculus of gallbladder, pancreatic cancer, gastric cancer, colorectal cancer, and esophageal cancer). The inverse-variance weighted (IVW) method was used to evaluate the main effects model of causality.

Results: This MR study suggests that OSA may play a causal role inflammation-related GDs (GERD, P_IVW=5.94×10^-9; gastroduodenal ulcer, P_IVW=1×10^-4; chronic gastritis, P_IVW=0.0214; ulcerative colitis, P_IVW=0.0296), and gallstones (calculi of the gallbladder, P_IVW=0.0429; calculi of the bile duct, P_IVW=0.0068). After accounting for obesity, type 2 diabetes, smoking, and alcohol consumption, the multivariate MR (MVMR) analysis identified that OSA is an independent risk factor for GERD, gastroduodenal ulcer, and calculus of the bile duct. The reverse MVMR analysis showed a causal effect of GERD on OSA. Besides, we did not find that the predisposition to OSA was associated with 4 cancers.

Conclusion: This MR analysis provides compelling evidence of an independent causal relationship between genetically predicted OSA and an elevated risk of inflammation-related GDs. Besides, no causal association was observed between OSA and cancers. Further studies should be carried out to verify our findings.

Keywords: Mendelian randomization; causal relationship; gastroesophageal reflux; gastrointestinal disease; obstructive sleep apnea.

Figure 1

Study flow diagram. (A) The flow of Two-Sample MR analysis. (B) The flow of MVMR analysis. The dashed lines represent possible pleiotropic or direct causal effects between variables that may violate the MR assumptions. IVW, inverse-variance weighted; WM, weighted median; MR, Mendelian randomization; MR-PRESSO, MR pleiotropy residual sum and outlier; OSA, Obstructive sleep apnea; GDs, gastrointestinal diseases; IVs, instrumental variables; MVMR, Multivariate Mendelian randomization; LASSO, least absolute shrinkage and selection operator.

Figure 2

 A forest plot of the potential causal relationship between OSA and GDs. (A) The forest plot of the potential causal relationship between OSA and GERD; (B) The forest plot of the potential causal relationship between OSA and other GDs. OSA, Obstructive sleep apnea; GERD, gastroesophageal reflux disease; GDs, gastrointestinal diseases; SNP, single nucleotide polymorphism; OR, odds ratio; SE, standard error; CI, confidence interval; IVW, inverse-variance weighted.

Figure 3

 A forest plot of the MVMR results between OSA and GERD. OSA, Obstructive sleep apnea; GERD, gastroesophageal reflux disease; OR, odds ratio; CI, confidence interval; IVW, inverse-variance weighted; LASSO, least absolute shrinkage and selection operator; MVMR, multivariate Mendelian randomization.

Figure 4

 A forest plot of the MVMR results between OSA and other GDs. OSA, Obstructive sleep apnea; GDs, gastrointestinal diseases; OR, odds ratio; CI, confidence interval; IVW, inverse-variance weighted; LASSO, least absolute shrinkage and selection operator; MVMR, multivariate Mendelian randomization.