Evaluation of a portable recording device (Somnocheck (R)) for use in patients with suspected obstructive sleep apnoea

Background: Portable recording devices for the diagnosis of obstructive sle ep apnoea (OSA) should be carefully evaluated before being used on a routin e basis in outpatients. Objective: This study was designed to evaluate the diagnostic accuracy of the new portable recording device, Somnocheck(R) (SC ). Methods: Nocturnal poly somnographies (PSG) and SC recordings were perfo rmed simultaneously in the sleep laboratory in 51 patients with suspected O SA, and were analysed manually by a blinded investigator. Results: The apno ea/hypopnoea index (AHI) obtained by manual SC analysis correlated closely with that obtained by PSG (r = 0.98), but the correlation with AHI obtained by automatic SC analysis was markedly lower (r = 0.83). When an AHI greate r than or equal to 10 was defined as indicative of OSA, the sensitivity of manual SC analysis was 0.97 and its specificity 1.00. The sensitivity of au tomatic analysis was 0.83 and its specificity 0.95. Conclusions: In this ex perimental setting, the SC showed a very high diagnostic accuracy for the d iagnosis of OSA and was able to define its severity precisely. The diagnost ic accuracy of manual analysis was found to be superior to that of automati c analysis. Considering the results of this technical evaluation, the SC ma y also be expected to work reliably when it is used in an ambulatory settin g. Copyright (C) 2001 S. Karger AG, Basel.