Abstract

BACKGROUND--Increasing referral numbers make the development of simplified accurate methods of diagnosing the sleep apnoea/hypopnoea syndrome highly desirable. The accuracy of one such system--the ResCare Autoset--has been examined. METHODS--Thirty one consecutive patients assessed by polysomnography had simultaneous monitoring of their respiratory pattern using the Autoset system. The Autoset detects episodes of flattening of the flow/time profile using nasal cannulae. RESULTS--There was a good correlation (r = 0.85) between the number of apnoeas+hypopnoeas/hour in bed recorded using polysomnography and the Autoset system. The median difference in such events was 3.1 (95% confidence interval 8.4 to -1.6)/hour in bed. In two patients the Autoset scored 70 apnoeas+hypopnoeas/hour in bed compared with 34 apnoeas+hypopnoeas with 35 arousals/hour in bed by polysomnography; however, this did not alter the diagnostic category of either patient. Autoset gave a sensitivity of 100%, specificity of 92%, positive predictive value of 92%, and negative predictive value of 100%, which was better than oximetry alone. A sleep study using the Autoset system costs 14 pounds compared with 126 pounds for polysomnography. CONCLUSIONS--The Autoset is clinically useful for diagnosing the sleep apnoea/hypopnoea syndrome.