Causal analysis between gastro-oesophageal reflux disease and obstructive sleep apnoea

Gui Chen; Xin Gong; Shenrong Liu; Junyang Xie; Yingqi Wang; Wucheng Guo; Wenjing Liao; Lijuan Song; Xiaowen Zhang

doi:10.1183/23120541.00127-2023

Causal analysis between gastro-oesophageal reflux disease and obstructive sleep apnoea

ERJ Open Res. 2023 Aug 29;9(4):00127-2023. doi: 10.1183/23120541.00127-2023. eCollection 2023 Jul.

Authors

Gui Chen^{1

2}, Xin Gong^{3

2}, Shenrong Liu⁴, Junyang Xie¹, Yingqi Wang¹, Wucheng Guo⁵, Wenjing Liao¹, Lijuan Song¹, Xiaowen Zhang¹

Affiliations

¹ State Key Laboratory of Respiratory Disease, Department of Otolaryngology - Head and Neck Surgery, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou, China.
² These authors contributed equally.
³ Department of Otolaryngology - Head and Neck Surgery, Wushan County People's Hospital of Chongqing, Wushan, China.
⁴ Department of Cardiology, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou, China.
⁵ Department of Stomatology, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou, China.

Abstract

Background: Based on evidence from existing observational research, clarifying the causal relationship between gastro-oesophageal reflux disease (GORD) and obstructive sleep apnoea (OSA) is challenging. Here, Mendelian randomisation, a method based on genetics, was used to provide new evidence for causality.

Methods: Summary statistics from two publicly available genome-wide association studies were used to evaluate the causal relationship between GORD and OSA (the GORD database was used as an exposure variable and the OSA database as an outcome). Inverse variance weighting was used as the main analytical tool in Mendelian randomisation to estimate causal effects. The robustness of the results was evaluated by sensitivity analysis. Possible mediators were evaluated using multivariate Mendelian randomisation.

Results: A statistically significant causal relationship was observed between GORD and OSA (OR 1.597, 95% CI 1.401-1.821, p<0.001), and similar results were observed in weighted median and Mendelian randomisation-Egger regression analyses. No bias was found in the sensitivity analysis of Mendelian randomisation estimation. Multivariate Mendelian randomisation showed that GORD significantly increased the risk of developing OSA, even when the possible mediator was excluded (OR 1.107, 95% CI 1.101-1.212, p<0.001).

Conclusion: Our study confirmed a causal relationship between GORD and OSA and suggests that intervention measures should be taken for patients with GORD to prevent the occurrence of OSA.