PURPOSE:

To compare the outcomes of patients hospitalized with pneumonia treated with noninvasive ventilation (NIV) and invasive mechanical ventilation (IMV). MATERIALS AND METHODS:

Using the HealthFacts multihospital electronic medical record database, we included patients hospitalized with a diagnosis of pneumonia and treated with NIV or IMV. We developed a propensity model for receipt of initial NIV and assessed the outcomes in a propensity-matched cohort, and in a covariate adjusted and propensity score weighted models. RESULTS:

Among 3971 ventilated patients, 1109 (27.9%) were initially treated with NIV. Patients treated with NIV were older, had lower acuity of illness score, and were more likely to have congestive heart failure and chronic pulmonary disease. Mortality was 15.8%, 29.8% and 25.9.0% among patients treated with initial NIV, initial IMV and among those with NIV failure. In the propensity matched analysis, the risk of death was lower in patients treated with NIV (relative risk: 0.71, 95% CI: 0.59-0.85). Subgroup analysis showed that NIV was beneficial among patients with cardiopulmonary comorbidities (relative risk 0.59, 95% CI: 0.47-0.75) but not in those without (relative risk 0.96, 95% CI: 0.74-0.1.25)NIV failure was significantly (p=0.002) more common in patients without cardiopulmonary conditions (21.3%) compared to those with these conditions (13.8%). CONCLUSIONS:

Initial NIV was associated with better survival among the subgroup of patients hospitalized with pneumonia who had COPD or heart failure. Patients who failed NIV had high in-hospital mortality, emphasizing the importance of careful patient selection monitoring when managing severe pneumonia with NIV.