Differences in Physiologic Endotypes Between Nonpositional and Positional OSA Results from the Shanghai Sleep Health Study Cohort
Xiaoting Wang,Tianjiao Zhou,Weijun Huang,Jingyu Zhang,Jianyin Zou,Jian Guan,Hongliang Yi,Shankai Yin
DOI: https://doi.org/10.1016/j.chest.2024.01.021
IF: 9.6
2024-01-01
Chest
Abstract:BACKGROUND: Positional OSA (POSA) is a recognized subtype of OSA that exhibits distinct endotypic characteristics when compared with nonpositional OSA (NPOSA). The basis for the disparity in endotypes between these subtypes remains poorly understood. RESEARCH QUESTION: (1) Do individuals with NPOSA and POSA have different underlying OSA endotypes? (2) Which endotypic characteristics are critical in determining NPOSA and POSA severity? STUDY DESIGN AND METHODS: Within the Shanghai Sleep Health Study cohort, individuals with OSA were recruited and classified as having POSA or NPOSA. Endotypes were calculated using polysomnography. RESULTS: Endotype analysis was conducted in 1,036 individuals with OSA. Compared with individuals with NPOSA, those with POSA had lower loop gain calculated during all sleep stages and all sleep positions (0.55; interquartile range [IQR], 0.46-0.66 vs 0.68, IQR, 0.52-0.90; P < .001), lower arousal threshold calculated during all sleep stages and all sleep positions (ArTHAll) (138.67; IQR, 118.94-180.87 percentage of the eupneic ventilation [%Veupnea] vs 189.00; IQR, 129.71-257.76 %Veupnea; P < .001), lower pharyngeal collapsibility calculated during all sleep stages and all sleep positions (Vpassive(All)) (91.85; IQR, 83.13-95.15 %Veupnea vs 76.38; IQR, 23.77-92.08 %Veupnea; P < .001), and higher muscle compensation calculated during all sleep stages and all sleep positions (6.50; IQR, -6.77 to 16.39 %Veupnea vs 3.65; IQR, -10.47 to 12.14 %Veupnea; P = .003). Logistic regression analyses indicated that higher Vpassive(All) was associated with increased odds of POSA vs NPOSA. In NPOSA, fully adjusted linear regression analyses indicated that Vpassive(All) (beta = -0.55; 95% CI, -0.68 to -0.42; P < .001) and lower loop gain calculated during all sleep stages and all sleep positions (beta = 0.19; 95% CI, 0.08-0.30; P < .001) were significant independent predictors of the apnea hypopnea index, with Vpassive(All) being the most critical factor. In contrast, in POSA, collapsibility appeared to be less influential (beta = -0.09; 95% CI, -0.21 to 0.03; P = .138). Nonanatomic endotypic characteristics (LG(All): beta = 0.29; 95% CI, 0.18-0.41; P < .001; arousal threshold in all sleep stages and all sleep positions: beta = 0.15; 95% CI, 0.01-0.28; P = .031; muscle compensation in all sleep stages and all sleep positions: beta = -0.21; 95% CI, -0.29 to -0.12; P < .001) were significant in determining the severity of POSA, with loop gain being the most crucial factor. INTERPRETATION: This study highlights the differences in endotypes between NPOSA and POSA. In Chinese individuals, anatomic factors were more signi fi cant in determining the severity of NPOSA, whereas nonanatomic traits were more likely to determine the severity of POSA. Future research should focus on developing personalized management strategies for individuals with NPOSA and POSA based on their endotypes.