Demographic and clinical differences in patients with positional obstructive sleep apnea and development of a discrimination model
Purpose: Obstructive sleep apnea (OSA) is a highly prevalent disease, and positional OSA (pOSA) is a subgroup whose OSA severity is highly affected by sleeping position. This study investigated differences in demographic and sleep characteristics between patients with and without pOSA and developed a simple discrimination model. Methods: We reviewed polysomnography records of patients admitted to the Sleep Center at Shuang-Ho Hospital between March 2015 and March 2019. They were categorized into pOSA and non-pOSA groups, and their demographic and sleep characteristics were compared. The receiver operating characteristic (ROC) curve was used to estimate the feasibility of discrimination model. Results: Of the patients, 33% received diagnoses of pOSA; they had smaller neck circumference and waistline and lower weight, body mass index (BMI), OSA severity, heart rate, and snoring and respiratory-related limb movement indexes but higher sleep efficiency and mean oxygen saturation compared with patients without pOSA. Sleep stage analysis revealed that as severity increased, the proportion of sleep time spent in N2, N3, and rapid eye movement stages decreased, but the pro-portion of time spent in the N1 stage increased in both populations. Sleep position analysis revealed a higher proportion of sleep time in a supine position among patients with pOSA after adjustment for severity. The corresponding area under the ROC curve of our discrimination model was 0.924. Conclusions: Demographic and sleep characteristics differed significantly between patients with and without pOSA. Our model uses readily available measurements such as BMI and waistline and can aid physicians in the timely identification of patients with pOSA.