AutoSet nasal CPAP titration: constancy of pressure, compliance and effectiveness at 8 month follow-up

We have previously shown that AutoSet satisfactorily improves sleep-disordered breathing and sleep architecture in subjects with obstructive sleep apnoea (OSA) syndrome. The aim of this study was to determine, in subjects treated with long-term conventional fixed pressure continuous positive airway pressure (CPAP) at the AutoSet recommended pressure, whether: the long-term compliance is satisfactory; the improvement persists once initial rebound is over; the titration pressure is stable with time; and the titration pressure is comparable with manual titration pressure using a similar end-point. Twenty males with OSA, previously studied with full polysomnography on their diagnostic night, at manual and AutoSet titration, and at the AutoSet recommended fixed pressure, were re-studied after a mean of 3 and 8 months of treatment at the recommended fixed pressure. Re-study included home respiratory monitoring (Nellcor EdenTrace), and repeated manual and AutoSet titration with polysomnography. Compliance was assessed with hour-meter readings. Mean (+/-SEM) usage was 5.7 +/- 0.1 h.night-1 at 3 and 8 months. The arousal index remained normalized. Diagnostic respiratory disturbance index (RDI) was 60.3 +/- 5.7 events.h-1. On AutoSet at fixed CPAP, RDI was initially 2.6 +/- 0.7 events.h-1, then rose slightly (p < 0.001) to 4.3 +/- 0.6 events.h-1 at 3 months, and was 3.6 +/- 0.5 events.h-1 at 8 months. AutoSet titration pressure was: 9.9 +/- 0.4 cmH2O initially, 10.6 +/- 0.4 cmH2O at 3 months, and 9.7 +/- 0.5 cmH2O at 8 months (NS). Manual titration pressure at 8 months was 10.4 +/- 0.4 cmH2O. The standard deviation of the discrepancy with AutoSet was 0.84 cmH2O. In conclusion, the AutoSet recommended pressure varies little with time, and closely predicts the final manual titration pressure; the improvement in respiratory disturbance index was largely maintained, and compliance was good, although probably enhanced by close supervision.