Purpose: Noninvasive ventilation (NIV) had proven benefits in clinical trials that included selected patients admitted to highly skilled centers. Whether these benefits apply to every patient and in everyday practice deserves appraisal. The aim of the study was to assess the use and outcomes of NIV over the last 15 years.
Methods: Multicenter database study of critically ill patients who required ventilatory support for acute respiratory failure between 1997 and 2011. The impact of first-line NIV on 60-day mortality was evaluated using a marginal structural model. Follow-up was censored on day 60.
Results: Of 3,163 patients, 1,232 (39 %) received NIV. Over the study period, first-line NIV increased from 29 to 42 %, and NIV success rates increased from 69 to 84 %. NIV decreased 60-day mortality [adjusted hazard ratio (aHR), 0.75; 95 % confidence interval (95 % CI), 0.68-0.83; P < 0.0001]. This protective effect was observed in patients with acute-on-chronic respiratory failure (aHR, 0.71; 95 % CI, 0.57-0.90; P = 0.004), but not in patients with cardiogenic pulmonary edema (aHR, 0.85; 95 % CI, 0.70-1.03; P = 0.10) or in patients with hypoxemic ARF, either immunocompetent (aHR, 1.18; 95 % CI, 0.87-1.59; P = 0.30) or immunocompromised (aHR, 0.89; 95 % CI, 0.70-1.13; P = 0.35). NIV failure was an independent time-dependent risk factor for mortality (aHR, 4.2; 95 % CI, 2.8-6.2; P < 0.0001).
Conclusions: The use of NIV increased steadily over the study period. First-line NIV was associated with better 60-day survival and fewer ICU-acquired infections compared to first-line intubation. Survival benefits from NIV occurred only in patients with acute-on-chronic respiratory failure and immunocompromised patients.