Comparison of a limited computerized diagnostic system (ResCare Autoset) with polysomnography in the diagnosis of obstructive sleep apnoea syndrome.

The increasing numbers of patients referred for evaluation of suspected obstructive sleep apnoea (OSA) places a growing burden on available sleep laboratory resources. A number of limited diagnostic systems have been developed in an effort to cope with this clinical problem. In this study, the diagnostic capabilities of one limited diagnostic system (ResCare Autoset) were compared with full polysomnography (PSG), using the Oxford SAC computerized system. Thirty six patients with suspected OSA had simultaneous studies performed both with the Autoset and Oxford PSG systems. The apnoea plus hypopnoea index (AHI) (events x h(-1)) scored by the Autoset system was compared with the AHI scored from the PSG raw tracings by an experienced sleep technician. There were highly significant correlations between the Autoset AHI and the AHI scored by the manual PSG scoring method (r=0.92; p<0.001). The positive predictive value for diagnosis of OSA for the Autoset was 86% when compared with manual PSG scoring, based on an AHI threshold for OSA of 15 events x h(-1). However, the agreement between Autoset and PSG was poor in severe cases of OSA, although not sufficiently so as to result in mistaken diagnosis in any of these cases. We conclude that the Autoset system is a sensitive and easy to use system, which facilitates screening for obstructive sleep apnoea with a reasonable degree of accuracy.