Evaluation of pulse oximeter derived photoplethysmographic signals for obstructive sleep apnea diagnosis

Abstract

Abstract High prevalence of obstructive sleep apnea (OSA) has increased the demands for more convenient and accessible diagnostic devices other than standard in-lab polysomnography (PSG). Despite the increasing utility of photoplethysmograph (PPG), it remains understudied in underserved populations. This study aimed to evaluate the reliability of a standard pulse oximeter system with an automated analysis based on the PPG signal for the diagnosis of OSA, as compared with PSG derived measures. Consecutive out-patients with suspect OSA completed a PPG monitoring simultaneous with an overnight in-lab standard PSG. Forty-nine OSA patients (38 males, age 43.5 ± 16.9 years, BMI 26.9 ± 0.5 kg/m2) were included in this study. Automated analyses were based on PPG and oximetry signals only. The PPG calculated measures were compared with PSG derived measures for agreement tests. Respiratory events index derived from PPG significantly correlated with PSG-derived apnea–hypopnea index (r = 0.935, P < .001). The calculation of total sleep time and oxygen desaturation index from PPG and PSG also significantly correlated (r = 0.418, P = .003; r = 0.933, P < .001, respectively). Bland–Altman plots showed good agreement between the PPG and the PSG measures. The overall sensitivity and specificity of PPG are good, especially in moderate and severe OSA groups. The tested PPG approach yielded acceptable results compared to the gold standard PSG among moderate to severe OSA patients. A pulse oximeter system with PPG recording can be used for the diagnosis or screening of OSA in high risk population.