Objective: To evaluate the effects of various levels of pressure support (PS) during noninvasive ventilation delivered through a helmet on breathing pattern, inspiratory effort, CO2 rebreathing, and comfort.
Design: Physiologic study.
Setting: University-affiliated hospital.
Patients and participants: Eight healthy volunteers.
Interventions: Volunteers received ventilation through a helmet with four different PS/positive end-expiratory pressure combinations (5/5 cm H2O, 10/5 cm H2O, 15/5 cm H2O, and 10/10 cm H2O) applied in random order.
Measurements and results: The ventilatory respiratory rate, esophageal respiratory rate (RRpes), airway pressure, esophageal pressure tracings, esophageal swing, and pressure-time product (PTP) [PTP per breath, PTP per minute, and PTP per liter] were evaluated. We also measured the partial pressure of inspired CO2 (Pico2) at the airway opening, mean partial pressure of expired CO2 (Peco2), CO2 production (V̇co2), minute ventilation (V̇e) delivered to the helmet (V̇eh), and the true inspired V̇e. By subtracting V̇e from V̇eh, we obtained the V̇e washing the helmet (V̇ewh). A visual analog scale (from 0 to 10) was used to evaluate comfort. Compared to spontaneous breathing, different levels of PS progressively increased tidal volume (Vt) and decreased RRpes, reducing inspiratory effort. The increased levels of assistance did not produce significant changes in Pico2, end-tidal CO2, and V̇co2. Peco2 had a slight decrease when increasing the level of PS from 5 to 10 cm H2O (p < 0.05). Despite the presence of constant values of V̇e, the increase of PS produced an increase in V̇ewh, without significant differences comparing 10 cm H2O and 15 cm H2O of PS. The subjects had a slight but not significant increase in discomfort by augmenting the level of assistance. At the highest level of PS (15 cm H2O), the discomfort was significantly higher (p < 0.001) than at the other levels of assistance.
Conclusion: In volunteers, the helmet is efficient in ventilation, allowing a Vt increase and RRpes reduction. A significant discomfort was present only at the highest level of assistance; however, it did not affect patient/ventilator interaction.