[Characteristics of auto-CPAP devices during the simulation of sleep-related breathing flow patterns]

The function of automatic CPAP devices is difficult to investigate using clinical examinations due to the high variability of breathing disorders. With a flow generator, however, identical breathing patterns can be reproduced so that comparative studies on the behaviour of pressure of APAP devices are possible. Because the algorithms of APAP devices based on the experience of users can be modified without much effort, also previously investigated devices should regularly be reviewed with regard to programme changes.

Had changes occurred in the algorithms of 3 selected devices--compared to the previously published benchmark studies? Do the current versions of these investigated devices differentiate between open and closed apnoeas?

With a self-developed respiratory pump, sleep-related breathing patterns and, with the help of a computerised valve, resistances of the upper respiratory tract were simulated. Three different auto-CPAP devices were subjected to a bench test with and without feedback (open/closed loop).

Open loop: the 3 devices showed marked differences in the rate of pressure rise but did not differ from the earlier published results. From an initial pressure of 4 mbar the pressure increased to 10 mbar after a different number of apnoeas (1-6 repetitive apnoeas). Only one device differentiated between closed and open apnoeas. Closed loop: due to the pressure increase, the flow generator simulated reduced obstruction of the upper airways (apnoeas changed to hypopnoeas, hypopnoeas changed to flattening) but different patterns of pressure regulation could still be observed.

By applying bench-testing, the algorithms of auto-CPAP devices can regularly be reviewed to detect changes in the software. The differentiation between open and closed apnoeas should be improved in several APAP devices.