Abstract
Home set-up polysomnography (PSG) has advantages over other portable monitoring devices, but remains unendorsed by professional bodies despite excellent utility in the Sleep Heart Health Study (SHHS). The study aims to determine technical reliability and diagnostic accuracy of unattended, home set-up versus attended laboratory-based PSG in patients with suspected obstructive sleep apnea (OSA). Thirty patients with suspected OSA without significant co-morbidity were recruited. After initial lab-PSG (Compumedics S series), patients underwent home set-up PSG (Compumedics Siesta) and lab-based PSG in random order. Studies were compared for study success, signal loss and likelihood ratio for OSA diagnosis [apnea-hypopnea index (AM) > 10]. Thirty subjects (mean age 49 +/- 13.8 years, body mass index 31 +/- 6.1 kg m(-2)) completed investigations. SHHS technical acceptability criteria were met by all lab-based PSGs and 90% of home-based PSGs (93% clinically acceptable). Signal loss was higher at home (P = 0.008). Sleep efficiency was similar between sites, but more preferred home-based PSG (50%). ANCOVA revealed AHI was significantly different if initial AHI > 26 per h (P = 0.006), with an average underestimate of 5.1 per h at home. In technically acceptable studies the likelihood ratios to 'rule in' and 'rule out' OSA were 8.1 and 0.1, respectively. Unattended, home set-up PSG is technically reliable and achieves excellent diagnostic utility. Signal loss was higher at home but mitigated by multi-channel redundancy. Success rate was similar to SHHS and superior to laboratory set-up home studies. Home set-up PSG is a valid alternative to laboratory-based PSG for suspected OSA.