Home-based diagnosis of obstructive sleep apnea by polysomnography type 2: accuracy, reliability, and feasibility

Purpose: Despite being used in large cohort studies, role of polysomnography (PSG) type 2 is still controversy. This study was aimed to determine its accuracy, reliability, and feasibility in diagnosis of obstructive sleep apnea (OSA) compared to gold standard.

Methods: Adult patients with stable medical conditions who complained of snoring or excessive sleepiness and lived around Bangkok were recruited from a sleep clinic. All were asked to fill questionnaires and have PSG done in laboratory (in-Lab PSG) and at home (Home PSG) on separate nights within 2–4 weeks interval.

Conclusions: This is the first report in Asia demonstrating that home-based diagnosis of OSA by PSG type 2 was feasible performing with good reliability, high accuracy, and a low failure rate. However, further studies focusing on its cost-effectiveness are required.

Results: Eighty-six patients, 48 males and 38 females, were included. Mean of total sleep time, sleep efficiency, and stage R were significantly greater in Home PSG than in-Lab PSG (p < 0.05). Apnea–hypopnea index (AHI) was slightly higher in Home PSG (25.7 versus 23.5, p = 0.04), but with excellent reliability, intra-class correlation coefficients of 0.96 (95 % CI; 0.93–0.97), and good agreements (κ = 0.59–0.70) between both tests. The sensitivity, specificity, and accuracy of Home PSG at cut-off point of AHI ≥5, were 0.97, 0.56, and 0.85, respectively, and at AHI ≥15 were 0.95, 0.76, and 0.85, respectively. Sixty-four patients (74.4 %) preferred home-PSG but four patients (4.7 %) needed repeated tests due to significant data loss.