Abstract Rationale Although nasal Continuous Positive Airway Pressure (CPAP) remains the gold-standard treatment for obstructive sleep apnea, excessive leak management is a common challenge. Excessive leak is an important CPAP side effect that can impair CPAP treatment and adherence. CPAP leak is described by different manufacturers using different leak variables and different excessive leak definitions. We hypothesized that the classification of excessive leak within a cohort of nasal CPAP users differs according to the specific criteria of different manufacturers. Methods Thirty-nine patients who underwent in-lab polysomnography (PSG) for nasal CPAP titration were analyzed. For each CPAP titration PSG, total leak was determined. Intentional leak was calculated using the pressure-specific intentional leak for the mask used. Unintentional leak was determined by subtracting intentional leak from total leak. CPAP leak variables used by different manufacturers were also calculated: 95th percentile unintentional leak (ResMed), duration of leak exceeding intentional leak x 2 (Philips Respironics), duration of leak exceeding 60 L/min (Fisher & Paykel) and 90 L/min (DeVilbiss). CPAP leak from each patient was classified according to the manufacturer’s excessive leak definition, except for DeVilbiss that does not define excessive leak (Table 1). Correlation analyses were performed to assess the relationship between leak variables. Results Anthropometric characteristics and CPAP leak variables are described in Table 1. The classification of excessive leak varied widely according to different manufacturers: 40% (ResMed), 4% (Philips Respironics), and 0% (Fisher & Paykel) (p ≤ 0.001). Leak variables used by ResMed, Philips Respironics, and Fisher & Paykel were associated (rho: 0.71-0.81, p ≤ 0.001). DeVilbiss leak variable was not associated with the other 3 manufacturers. Conclusion Excessive leak definition among different manufacturers widely disagrees. The lack of consensus in excessive leak definition impairs the implementation of excessive leak management strategies. This abstract is funded by: CAPES
Rosanelli et al. (Fri,) studied this question.