Abstract
BACKGROUND: Noninvasive ventilation (NIV) is widely used to improve oxygenation and reduce carbon dioxide (CO(2)) retention in patients with respiratory failure. However, it remains unclear whether different types of exhalation valves affect CO(2) rebreathing and ventilator performance during NIV. METHODS: Three noninvasive ventilators (V60, Flexo, and Stellar150) with single-limb circuits and four different exhalation valves (single-arch, whisper swivel, plateau exhalation, and vented mask valves) were separately connected in series to a lung simulator. CO(2) gas was injected from the simulated lung outlet, maintaining the end-expiratory CO(2) (PetCO(2)) at 80 mmHg. Both the CO(2) rebreathing volume (CO(2REB)) and the parameters displayed on the lung simulator and ventilator were recorded under each condition. RESULTS: The mean CO(2REB) values of the four aforementioned valves were 18.51 ± 2.87, 18.25 ± 2.73, 17.78 ± 2.98, and 14.26 ± 0.92 mL/breath, respectively, with no significant differences among the first three types but all significantly higher than that of the mask valve (all p < 0.0001, rate of difference > 10%). Except with the V60 ventilator, some ventilator performance parameters (triggering and control performance) were significantly lower for the plateau valve than for the others, the rate of difference in tidal volume (V(T)) between the ventilator and the simulated lung exceeded 10% for all exhalation valves (all p < 0.01). CONCLUSION: Mask valves showed significantly lower CO(2) rebreathing than circuit-located valves (single-arch, whisper swivel, and plateau exhalation) in this NIV bench study. The different valves influenced ventilator performance differently, particularly the plateau valve. These findings necessitate further clinical validation in vivo.