Effects of different continuous positive airway pressure devices and periodic hyperinflations on respiratory function.

Objective: To compare the effect on respiratory function of different continuous positive airway pressure systems and periodic hyperinflations in patients with respiratory failure. Design: Prospective Setting: Hospital intensive care unit. Patients: Sixteen intubated patients (eight men and eight women, age 54 ± 18 yrs, Pao 2 /Fio 2 277 ± 58 torr, positive end-expiratory pressure 6.2 ± 2.0 cm H 2 O). Interventions: We evaluated continuous flow positive airway pressure systems with high or low flow plus a reservoir bag equipped with spring-loaded mechanical or underwater seal positive end-expiratory pressure valve and a continuous positive airway pressure by a Servo 300 C ventilator with or without periodic hyperinflations (three assisted breaths per minute with constant inspiratory pressure of 30 cm H 2 O over positive end-expiratory pressure). Measurements and Main Results: We measured the respiratory pattern, work of breathing, dyspnea sensation, end-expiratory lung volume, and gas exchange. We found the following: a) Work of breathing and gas exchange were comparable between continuous flow systems; b) the ventilator continuous positive airway pressure was not different compared with continuous flow systems; and c) continuous positive airway pressure with periodic hyperinflations reduced work of breathing (10.7 ± 9.5 vs. 6.3 ± 5.7 J/min, p <.05) and dyspnea sensation (1.6 ± 1.2 vs. 1.1 ± 0.8 cm, p <.05) increased end-expiratory lung volume (1.6 ± 0.8 vs. 2.0 ± 0.9 L, p <.05) and Pao 2 (100 ± 21 vs. 120 ± 25 torr, p <.05) compared with ventilator continuous positive airway pressure. Conclusions: The continuous flow positive airway pressure systems tested are equally efficient; a ventilator can provide satisfactory continuous positive airway pressure; and the use of periodic hyperinflations during continuous positive airway pressure can improve respiratory function and reduce the work of breathing.