Low versus high positive end expiratory pressure in noninvasive ventilation for hypoxemic respiratory failure: a multicenter randomized controlled trial

Purpose

To assess whether high positive end expiratory pressure (PEEP) reduces the rate of noninvasive ventilation (NIV) failure in hypoxemic patients.

Methods

This multicenter, open-label, randomized controlled trial was conducted across seven ICUs in China. Hypoxemic patients who received NIV via oronasal or nasal mask were randomized 1:1 to either low PEEP (5 cmH₂O) or high PEEP (10–15 cmH₂O) groups, with inspiratory positive airway pressure (IPAP) set at 10–20 cmH₂O and 15–20 cmH₂O, respectively. The primary outcome was NIV failure, defined as intubation, death, or therapy withdrawal (refusal of intubation despite need).

Results

Between January 11, 2022, and August 31, 2024, 380 patients (190 per group) were enrolled in an intention-to-treat analysis. NIV failure occurred in 43% (82/190) of the low PEEP group and 32% (61/190) of the high PEEP group (absolute difference: 11.1%, 95% CI 1.3–20.5%, p = 0.034). Within 72 h post-randomization, the low PEEP group exhibited lower PaO₂/FiO₂ ratios (mean difference: − 31 mmHg, 95% CI − 38 to − 24) and higher tidal volume (0.8 mL/kg predicted body weight, 95% CI 0.5–1.1) than the high PEEP group. However, the low PEEP group required higher support pressure (mean difference: 2.9 cmH₂O, 95% CI 2.7–3.1). Adverse events did not differ between the groups.

Conclusions

High PEEP during NIV may reduce treatment failure in patients with acute hypoxemic respiratory failure, although this benefit may be partially confounded by higher tidal volume observed in the low PEEP group. However, the interpretation of this effect should be carried out with caution as the study has insufficient statistical power to detect a significant difference.

Graphical abstract