Objective: Both obstructive sleep apnea (OSA) and obesity are highly prevalent worldwide, and are intrinsically linked. Previous studies showed that obesity is one of the major risk factors for OSA, but the causality of the relationship is still unclear. The study was to investigate the causal relationships of overall obesity and abdominal obesity with OSA and its quantitative traits.
Methods: In this case-control study, a total of 7134 participants, including 4335 moderate-to-severe OSA diagnosed by standard polysomnography and 2799 community-based controls were enrolled. Anthropometric and biochemical data were collected. Mendelian randomization (MR) analyses were performed using the genetic risk score, based on 29 body mass index (BMI)- and 11 waist-hip-ratio (WHR)-associated single nucleotide polymorphisms as instrumental variables. The causal associations of these genetic scores with OSA and its quantitative phenotypes were analyzed.
Results: Obesity was strongly correlated with OSA in observational analysis (β= 0.055, P = 3.7 × 10^{− 5}). In MR analysis, each increase by one standard deviation in BMI was associated with increased OSA risk [odds ratio (OR): 2.21, 95% confidence interval (CI): 1.62– 3.02, P = 5.57 × 10^{− 7}] and with 2.72-, 4.68-, and 3.25-fold increases in AHI, ODI, and MAI, respectively (all P < 0.05) in men. However, no causal associations were found between WHR and OSA risk or OSA quantitative traits in men and women.
Conclusion: Compared to abdominal obesity, overall obesity showed a causal relationship with OSA and its quantitative traits, especially in men.
Keywords: body mass index, Mendelian randomization, obesity, obstructive sleep apnea, waist-hip ratio