Given the high prevalence of OSA, low rates of clinical consultation, limited accessibility to PSG, and the widespread adoption of advanced portable technologies, the development of novel indices capable of accurately evaluating both the presence and severity of OSA is urgently required. We evaluated novel hypoxia-related indicators, such as the sleep breathing impairment index (SBII) and the percentage of sleep time with the duration of respiratory events causing desaturation (pRED₃p), and compared their predictive efficacy for OSA with traditional indicators. Anthropometric parameters and polysomnographic data from patients diagnosed with OSA were retrospectively collected. SBII and pRED₃p values were computed based on SPO 2 metrics. Logistic regression was utilized to construct predictive models for OSA, and the diagnostic performance of each indicator in determining OSA severity was assessed through receiver operating characteristic curve analysis. In the context of OSA onset and disease stratification, BMI and longest apnea- hypopnea time (L-AHT) demonstrated limited predictive capability. In contrast, sleep-related variables including maximum oxygen desaturation (MOD), lowest oxygen saturation (LSPO 2), and microarousal index (MAI) yielded moderate predictive performance, while oxygen desaturation index (ODI), pRED₃p, and SBII exhibited superior predictive accuracy with enhanced sensitivity and specificity. Notably, when predicting mild-to-moderate OSA, pRED₃p and SBII outperformed ODI, offering heightened sensitivity as hypoxia-related markers, particularly in patients who were non-obese and did not present with pronounced SPO 2 reductions. These novel hypoxia-related indices have demonstrated strong predictive power in identifying OSA and grading its severity, and can detect non-obese, non-severe hypoxic OSA patients at an early stage.
Zhang et al. (Mon,) studied this question.