Continuous positive airway pressure (CPAP) is routinely used for respiratory support in neonates. Air leaks are not uncommon and can hinder treatment. This study compared leak flow data obtained using different leak definitions through modeling and in vitro measurements.
Methods: Using a computer simulation of an ideal CPAP system, the relationship between leak flow and three leak definitions was investigated. The leak flow was based on the measured inspiratory, expiratory and averaged breathing flow as commonly used in neonates. The theoretical results were compared with in vitro measurements using a commercial CPAP device for neonates (Leoni, Heinen & Löwenstein, Germany). Spontaneous breathing was simulated using a mechanical lung model, and defined air leaks were simulated using open silicone tubes of different lengths.
Results: Computer simulations showed that leak calculations were affected by leak flow and breathing pattern, and that the relationships were mostly non-linear. There were large differences in the results between the three leak definitions. The in vitro measurements correlated strongly with the theoretical modeling results. The derived numerical volume correction reduced the leak-dependent volume error in the mean (S.D.) to -1.6 (4.6)%.
Conclusion: It is difficult to compare different measuring conditions and different CPAP devices due to the variation in results depending on the leak definition and the breathing pattern. Leak flow displays would provide superior air leak monitoring.