Screening for obstructive sleep apnea using a contact-free system compared with polysomnography.

Abstract

To evaluate the utility of a contact-free device in screening for obstructive sleep apnea. Three hundred fifty-nine participants (mean age 46 ± 13 years, body mass index 26.1 ± 4.2 kg/m², 67.7% male) underwent overnight monitoring using a contact-free device, the OrbSense, and polysomnography (PSG) in the sleep laboratory simultaneously. The OrbSense recordings were analyzed automatically, and PSG was scored based on recommended guidelines. The respiratory event index from the OrbSense was lower than the apnea-hypopnea index (AHI) from PSG (25.5 ± 20.7 vs 27.0 ± 25.2 events/h; P = .007) and was significantly correlated with AHI (Pearson coefficient, 0.92; P < .0001). Bland-Altman analysis showed a mean difference of 1.5 events/h, and the limit of agreement was -18.6 to 21.5 events/h. Use of the OrbSense resulted in larger underestimates of AHI and lower negative predictive values at higher AHI values (especially when AHI ≥ 30 events/h). When we used a PSG diagnostic criterion of AHI > 5 events/h, the optimal diagnostic cutoff value from the OrbSense was 8 events/h, with a sensitivity of 90.4%, a specificity of 77.6%, a 94.6% positive predictive value, and a 65% negative predictive value. For patients with moderate to severe obstructive sleep apnea whose AHI was > 15 events/h, the OrbSense cutoff was 16.6 events/h, with a sensitivity of 87.1% and a specificity of 89.7%. Among the 359 participants, 250 patients (69.6%) had the same obstructive sleep apnea severity division classified by both PSG and the OrbSense. The contact-free device OrbSense can detect respiratory events during sleep and has close agreement with in-laboratory PSG in screening for obstructive sleep apnea. Further studies are warranted to test its utility in community-based settings and at home.