Randomized trial of 6 auto-adapting CPAP devices

Auto-adapting CPAP (aCPAP) is used to manage patients with obstructive sleep apnoea (OSA). Different suppliers use different algorithms. We have compared 6 commercial devices - ResMed S8, Breas PV10, Weinmann Somnosmart, Goodnight 420E, Devilbiss Auto LT, Respironics M-series. Patients had used fixed pressure CPAP for 6+ months before the study and had good compliance. Each aCPAP device was used for 2 weeks in a randomized order. Data from the second week was analyzed and included CPAP use (hours), 7 day sleep diary card (SDC), and patient preference. The SDC included scores for Sleep Quality and Exercise Level (Excellent to Poor), Awakenings (0 to 6+), and a 100mm VAS (0= Alert;100=Drowsy). 23 patients were enrolled, 15 completed. Mean±SD weight - 117.8±29.2kg, age - 57±11yrs and 4% diprate 27±15. If patients could not use an aCPAP device, they used their fixed pressure device.There were significant differences between devices. We conclude that auto-CPAP devices have different outcomes and patient preferences. This may affect adherence to treatment for OSA in short-term trials.