Should NPPV Be Used in Extubation Failure?

Abstract & Commentary

Synopsis: In this randomized, controlled trial, when patients developed recurrent respiratory failure following extubation, the use of noninvasive ventilation delayed but did not prevent reintubation, and this delay was associated with a higher mortality rate in the ICU.

Source: Esteban A, et al. N Engl J Med. 2004;350:2452-2460.

In this international, randomized, controlled trial, mechanically ventilated patients in the ICUs in 37 medical centers in 8 countries were observed for development of post-extubation respiratory failure. Adults requiring critical care for a variety of medical and surgical illnesses, who had been ventilated for at least 48 hours and who had been successfully extubated after a trial of spontaneous breathing, were included in the study. After extubation patients were observed for development of signs of recurrent respiratory failure, which Esteban and colleagues defined as respiratory acidosis (pH < 7.35 with arterial PCO2 > 45 mm Hg), sustained tachypnea (respiratory rate > 25/min for > 2 hr), worsening oxygenation (arterial PO2 < 80 mm Hg or saturation < 90% on FIO2 0.50 or more), or signs of excessive work of breathing or respiratory muscle fatigue. If 2 or more of these criteria were met during the 48 hours following extubation, the patients were randomized either to usual care or to a trial of noninvasive positive-pressure ventilation (NPPV) by full-face mask. Standardized criteria for the need for reintubation were used. Mortality was the primary study outcome; secondary outcomes were the need for reintubation, the time to reintubation, and ICU length of stay. Esteban et al calculated that 194 patients would be required in each study group (388 total) to demonstrate a mortality difference based on expected baseline mortality and severity of illness.

The study was stopped early because of the finding on interim analysis of a higher mortality rate among the patients assigned to NPPV: 14% in the usual care group died in the ICU vs 25% in the NPPV group (relative risk, 1.78; 95% CI, 1.03-3.20; P = 0.048). Of 980 ventilated patients screened, 244 met criteria for respiratory failure following extubation. Immediate intubation was required in 23 patients, leaving 221 who were randomized to either usual care of NPPV. The need for reintubation was not different in the 2 groups: 51/107 (48%) with usual care vs 55/114 (48%) with NPPV. However, the median time from the onset of respiratory failure to intubation was longer in the NPPV group (2.5 hr vs 12 hr, P = 0.02). Length of stay in the ICU was not different in the 2 groups. Thus, in this study NPPV delayed but did not reduce the need for reintubation among patients who developed recurrent respiratory failure following extubation, and this delay was associated with an increase in mortality.

Comment by David J. Pierson, MD

As discussed later in this issue’s Special Feature, NPPV has become the standard of care for patients hospitalized for COPD exacerbations, and the available evidence also supports its use in cardiogenic pulmonary edema and certain forms of acute respiratory failure among immunocompromised patients.1 A well-done randomized controlled trial has also shown that extubation to NPPV in intubated patients with COPD complicated by acute respiratory failure may shorten the duration of mechanical ventilation and improve other outcomes—in carefully selected patients who are, among other things, alert and cooperative without serious complicating illness.2

Publication of the present study sounds a note of caution, however. Although it is logical to assume that decreasing the work of breathing with NPPV might enable patients with post-extubation respiratory failure to avoid reintubation, Esteban et al have demonstrated that this is not the case in actual practice. Extubation failure evidently involves more than just ventilatory muscle fatigue. Reintubation was delayed by NPPV in this study, but it was not prevented, and the fact that mortality was actually increased when NPPV was used suggests that delaying the resumption of invasive ventilatory support is a bad thing in this setting.

Only a small proportion of the patients in this study had COPD or cardiogenic pulmonary edema (12% and 7%, respectively, in the NPPV group), and it is possible that NPPV could be effective in extubation failure in such patients, given that they have been shown to benefit from NPPV at the onset of acute respiratory failure. However, whether this might be the case cannot be determined from this study. Based on what we know now, NPPV should not be used in extubation failure. When patients who have required intubation and mechanical ventilation manifest evidence of recurrent respiratory failure during the first 48 hours after extubation, they should be reintubated and ventilatory support reinstituted while the cause of the deterioration is investigated.


1. Liesching T, et al. Chest. 2003;124:699-713.

2. Nava S, et al. Ann Intern Med. 1998;128:721-728.

David J. Pierson, MD, Pulmonary and Critical Care Medicine, Harborview Medical Center, University of Washington, is Editor of CriticalCare Alert.