Abdominal Obesity Is More Strongly Correlated with Obstructive Sleep Apnea than General Obesity in China: Results from Two Separated Observational and Longitudinal Studies

• Published in: Obesity surgery 2019-08, Vol.29 (8), p.2535-2547
• Main Authors: Zhao, Xiaolong, Xu, Huajun, Qian, Yingjun, Liu, Yupu, Zou, Juanjuan, Yi, Hongliang, Guan, Jian, Yin, Shankai
• Format: Article
• Language: English
• Subjects: Abdomen; Adolescent; Adult; Aged; Bariatric Surgery - statistics & numerical data; Body Mass Index; China - epidemiology; Cross-Sectional Studies; Diabetes Mellitus, Type 2 - complications; Diabetes Mellitus, Type 2 - epidemiology; Diabetes Mellitus, Type 2 - surgery; Female; Gastrointestinal surgery; Humans; Longitudinal Studies; Male; Medicine; Medicine & Public Health; Middle Aged; Obesity; Obesity - complications; Obesity - epidemiology; Obesity - surgery; Obesity, Abdominal - complications; Obesity, Abdominal - epidemiology; Obesity, Abdominal - surgery; Original Contributions; Polysomnography; Risk Factors; Sleep apnea; Sleep Apnea, Obstructive - diagnosis; Sleep Apnea, Obstructive - epidemiology; Sleep Apnea, Obstructive - etiology; Surgery; Triglycerides; Waist Circumference; Young Adult
• ISSN: 0960-8923; 1708-0428
• DOI: 10.1007/s11695-019-03870-z

Background Previous studies have reported that obesity can result in or worsen obstructive sleep apnea (OSA). However, whether abdominal or general obesity indices or visceral adiposity indicators have a stronger association with OSA remains unclear. Methods This cross-sectional study included 4344 patients who underwent polysomnography (PSG) due to suspicion of OSA. We also performed a longitudinal study on 86 patients who underwent bariatric surgery to confirm the relationship between OSA and obesity. Data on overnight PSG parameters, biochemical biomarkers, and multiple anthropometric obesity indices were collected. Results In the cross-sectional study, waist circumference (WC) and body mass index (BMI) were independently associated with the apnea-hypopnea index (AHI) after adjusting for potential confounding factors (additional R 2  = 0.232, standardized beta coefficient [Beta] = 0.210; and additional R 2  = 0.015, Beta = 0.183, respectively). Logistic regression analysis showed similar results, as did stratified analysis of adult males aged ≤ 55 years. Restricted cubic spline (RCS) analysis revealed a linear dose-response relationship between OSA and obesity. In the longitudinal study, no significant relationship was found between remission of OSA and improvement in WC and BMI ( r  = 0.252, p  = 0.098; and r  = 0.132, p  = 0.395, respectively), whereas the change in the visceral adiposity indicator (lipid accumulation calculated according to WC and fasting triglycerides) was significantly correlated with ΔAHI ( r  = 0.322, p  = 0.033). Conclusions Abdominal obesity, rather than general obesity, appears to play a more important role in OSA.