Early use of noninvasive positive pressure ventilation for acute lung injury: a multicenter randomized controlled trial.

OBJECTIVE

Noninvasive positive pressure ventilation is beneficial for patients with acute respiratory failure. However, its possible benefit for patients with acute lung injury (200 mm Hg < PaO(2)/FIO(2) ≤300 mm Hg) remains unclear. Our aim was to assess the safety and efficacy of noninvasive positive pressure ventilation for patients with acute lung injury and compare this with high-concentration oxygen therapy.

DESIGN

A multicentered randomized controlled trial.

SETTING

Ten multipurpose intensive care units.

PATIENTS

Forty patients who fulfilled the criteria for acute lung injury were included in this study.

INTERVENTIONS

Patients were randomly allocated to receive either noninvasive positive pressure ventilation (noninvasive positive pressure ventilation group) or high-concentration oxygen therapy through a Venturi mask (control group).

MEASUREMENTS AND MAIN RESULTS

Twenty-one patients were assigned to the noninvasive positive pressure ventilation group and 19 were in the control group. At study entry, the patients' characteristics in the two groups were similar. Noninvasive positive pressure ventilation application decreased the respiratory rate and improved PaO(2)/FIO(2) with time. The proportion of patients requiring intubation and the actual number of intubations in the noninvasive positive pressure ventilation group were significantly less than in the control group (one of 21 vs. seven of 19; p = .02, and one of 21 vs. four of 19; p = .04, respectively). Noninvasive positive pressure ventilation showed a trend for reducing inhospital mortality (one of 21 vs. five of 19; p = .09). The total number of organ failures in the noninvasive positive pressure ventilation group was significantly lower than in the control group (three vs. 14; p < .001).

CONCLUSIONS

Noninvasive positive pressure ventilation is safe for selected patients with acute lung injury. However, a larger randomized trial with need for intubation and mortality as the outcomes of interest is required.