Respiratory failure & ventilation
Low versus high positive end expiratory pressure in noninvasive ventilation for hypoxemic respiratory failure: a multicenter randomized controlled trial
Primary result · NIV failureResult — go to the resultAbsolute difference 11.1 % 95% CI 1.3–20.5%Authors' conclusion
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.
Study context
- Population
- Hypoxemic patients receiving noninvasive ventilation (NIV) via oronasal or nasal mask, seven ICUs in Chinaintention-to-treat
- Participants
- 380 patients
- Comparison
- High positive end expiratory pressure (PEEP) vs low PEEP during NIV
- High PEEPn = 190
- Low PEEPn = 190reference
- Primary endpoint
- NIV failure (intubation, death, or therapy withdrawal [refusal of intubation despite need])
Methodological detailsmulticentre, open-label, 1:1 randomisation
- Multicentre
- Open-label
- 1:1 randomisation
Primary result
NIV failure
Absolute difference
11.1 %
95% CI 1.3–20.5%
P = 0.034
Events by treatment group
Low PEEP vs High PEEP
absolute difference 11.1 % (95% CI 1.3–20.5%); P = 0.034
As reported in the abstract
Further findings
| Outcome | Difference (mL/kg predicted body weight) (95% CI) | |
|---|---|---|
| Tidal volume · Within 72 h post-randomization | 0.8 (0.5–1.1) | |
| −0.500.511.5Difference, mL/kg predicted body weight (linear scale) |
| Outcome | Mean difference (cmH2O) (95% CI) | |
|---|---|---|
| Support pressure · Within 72 h post-randomization | 2.9 (2.7–3.1) | |
| −2024Difference, cmH2O (linear scale) |
| Outcome | Mean difference (mmHg) (95% CI) |
|---|---|
| PaO2/FiO2 ratio · Within 72 h post-randomization | - 31 (- 38 to - 24) |
Safety
Adverse events
- Did not differ between the groups
Original publication
Low versus high positive end expiratory pressure in noninvasive ventilation for hypoxemic respiratory failure: a multicenter randomized controlled trial. Intensive Care Med. Published 30 April 2025. doi:10. 1007/ s00134-025-07902-4
Composed by AI from the published abstract and checked against it. How Briefs are made →