Effects of Age and Sex on the Performance of the NoSAS Score As a Screening Tool for Obstructive Sleep Apnea: a Hospital-Based Retrospective Study in China.

Zhang Zhigang,Yang Dan,Wang Haiying,Liu Xinmin

DOI: https://doi.org/10.1007/s11325-020-02254-3

2020-01-01

Sleep And Breathing

Abstract:The NoSAS score has been shown to be a reliable screening tool for obstructive sleep apnea (OSA) in overall populations. This study aimed to explore the effects of age and sex on the predicting performance of this score. A retrospective analysis was conducted on 1119 subjects aged ≥ 18 years and with a total sleep time of ≥ 4 h during overnight polysomnography. Discrimination was assessed by using areas under receiver operating characteristic curve (AUCs), while predictive parameters were calculated by using contingency tables. Overall, a NoSAS score of 8 points or higher resulted in sensitivity, specificity, and AUC for predicting an apnea−hypopnea index (AHI) of ≥ 20 events/h of 74%, 36%, and 0.63 (in non-elderly 73%, 46%, and 0.65; in elderly 91%, 17%, and 0.59; in men 85%, 18%, and 0.56; in women 52%, 76%, and 0.71, respectively). The AUCs at all AHI cutoffs were significantly lower in men than in women (all with p < 0.01), while the AUCs at AHI cutoff of 5, 15, and 30 events/h were significantly lower in elderly than in non-elderly (p < 0.01, 0.05, and 0.05, respectively). In non-elderly, a conventional NoSAS with cutoff of 7 or a modified NoSAS with age cutoff of 50 years provided sensitivity and specificity for predicting an AHI of ≥ 20 events/h of 87%, 37% and 80%, 36%, respectively, with comparable AUCs. In women, a conventional NoSAS with cutoff of 6 or a modified NoSAS with neck circumference cutoff of 35 cm provided sensitivity and specificity for predicting an AHI of ≥ 20 events/h of 85%, 39% and 79%, 52%, respectively, with comparable AUCs. NoSAS score has better discrimination but lower sensitivity for predicting OSA in non-elderly and women than in their counterparts. Age- and sex-specific cutoff values reverse this imbalance. Our results underline the preference of age- and sex-specific cutoff values and the need for better age- and sex-specific screening algorithms.