Noninvasive ventilation for patients presenting with acute respiratory failure: the randomized controlled trials.

Noninvasive ventilation (NIV) in patients with acute respiratory failure (ARF), which was originally described decades ago, underwent a rebirth after reports of successful use in 1989. Over the following 18 years the literature on NIV has grown substantially. This paper summarizes the randomized controlled trials (RCTs) on NIV for acute respiratory failure. We conducted an extensive literature search and selected RCTs from that search. The results are presented primarily by etiology of respiratory failure, but we also include a short section on NIV for ARF in immunocompromised patients. The latter studies included patients with various etiologies of respiratory failure but with the common comorbidity of immunocompromise. Most of the RCTs have studied NIV for exacerbation of chronic obstructive pulmonary disease (COPD) or cardiogenic pulmonary edema. In general the RCTs have been small and used endotracheal intubation or NIV failure rate as primary outcomes. We conclude that NIV for ARF is supported by strong evidence from patients with COPD, but there is only weak support for NIV in other patient groups, such as immunocompromised patients. For other groups, such as patients with asthma, pneumonia, or acute lung injury, RCT-level evidence is lacking or does not suggest benefit. Clearly, major gaps remain in our evidence base.