By Betty Tran, MD, MSc, Editor

SYNOPSIS: In this multicenter, randomized, open-label trial of patients who failed a spontaneous breathing trial, those who were extubated to noninvasive ventilation did not have a shorter time to liberation from any form of mechanical ventilation compared to those who were randomized to protocolized standard weaning.

SOURCE: Perkins GD, et al. Effect of protocolized weaning with early extubation to noninvasive ventilation vs invasive weaning on time to liberation from mechanical ventilation among patients with respiratory failure. The Breathe Randomized Clinical Trial. JAMA 2018;320:1881-1888.

For patients who have passed a spontaneous breathing trial (SBT) but are deemed at high risk for extubation failure (e.g., patients with hypercapnia, chronic obstructive pulmonary disease, congestive heart failure), current guidelines recommend extubation to preventive noninvasive ventilation (NIV). These recommendations are based on evidence that suggests a reduction in ventilator- and ICU-related complications, including improved extubation success and a reduction in ICU length of stay.1 However, approximately 20-30% of intubated patients will require more than one SBT and up to seven days from the first attempt before extubation; subsequently, they are deemed to be a “difficult weaning group.”2

Given these findings, Perkins et al hypothesized that extubating patients who were difficult to wean directly to NIV would reduce the time to liberation from mechanical ventilation overall. The authors created a multicenter, randomized, allocation-concealed, controlled, open-label trial that was executed in 41 general adult ICUs in the United Kingdom. Researchers recruited patients between March 2013 and October 2016, with follow-up ending April 2017. The authors randomized 364 adults who received invasive mechanical ventilation through an endotracheal tube for more than 48 hours and who had failed an SBT to receive a NIV weaning protocol (n = 182) or an invasive ventilation weaning protocol (n = 182). The NIV group underwent extubation directly to NIV via face mask with initially equivalent pressure support settings that were provided via invasive ventilation. The treating physician assessed patients every two hours for distress or fatigue. The treating physician increased the inspiratory positive airway pressure to achieve patient comfort and a respiratory rate less than 30/minute. In the absence of clinical issues, patients were trialed off the mask or was put on a lower level of positive airway pressure (reduction by 2 cm H2O). In the invasive mechanical ventilation group, the patient also was assessed every two hours and their pressure support settings were adjusted down by 2 cm H2O in the absence of distress or fatigue or up if no reversible causes could be corrected and they were showing signs of distress or fatigue.

This cycle continued until patients could pass an SBT to be extubated or underwent tracheostomy. The primary outcome was time from randomization to successful liberation from ventilation, defined as the time point when the patient was alive and free of both invasive and noninvasive ventilator support for more than 48 hours. There were multiple secondary outcomes, including duration of invasive ventilation, total ventilator days, proportion of patients receiving antibiotics, rate of reintubation, and mortality at 30, 90, and 180 days, although the study was not adequately powered to detect differences in any of these.

The primary outcome of time from randomization to liberation from ventilation was not significantly different in the NIV group — 4.3 days (95% confidence interval [CI], 2.63-4.48) vs. 4.5 days (95% CI, 3.46-7.25), compared with the invasive group (adjusted hazard ratio, 1.1; 95% CI, 0.86-1.34). Patients in the NIV group received less invasive ventilation (median 1 day vs. 4 days; incidence rate ratio, 0.6; 95% CI, 0.47-0.87). There were no significant differences in rates of reintubation, tracheostomy, survival, or adverse events between the two groups.


Traditionally, an SBT with inspiratory pressure augmentation (5-8 cm H2O) is recommended as the best test to gauge whether a patient is “ready” to come off the ventilator.1 Interestingly, the results of this trial suggest that despite the results of our best test (i.e., an SBT), patients can be extubated and supported without the need for an endotracheal tube. Although this method was not superior in terms of shortening the time on positive pressure ventilation overall, it was not associated with any significant differences in terms of reintubation, tracheostomy, or survival rates (with the caveat that these were all secondary outcomes that the study was underpowered to evaluate). It is possible that extubation to NIV does not shorten the time to liberation from mechanical ventilation, as patients will come off ventilator support when they are truly “ready.”

A few limitations in this trial need to be noted. First, most of the enrolled patients mainly came from three centers that recruited an average of 1.3 patients/month (compared to 1.5 patients per year in the remaining hospitals). This likely limits generalizability in terms of patient population enrolled and physician expertise and patterns of decision-making overall. Second, the authors of the editorial accompanying this study3 noted that the standard weaning protocol group received quite intensive support and intervention, perhaps more than typical usual care, which resulted in a median duration of ventilator support that was much shorter than anticipated (2.9 days vs. the originally projected 6.4 days). This led to a reduction in the projected sample size to encourage enrollment. Therefore, a lack of a treatment effect in the NIV group could have happened because the control group received more aggressive, better-than-standard care. Finally, minimal details are available on how the NIV group fared while on ventilator support and how/why they were not able to be liberated sooner. It is possible that subsets of patients in this group could have benefited from early extubation to NIV, but this benefit was diluted by a more heterogeneous treatment effect as a whole.

The results of this trial suggest that difficult-to-wean patients potentially could be extubated to NIV despite failing an SBT, although this pathway has not been shown to be superior in terms of time to liberation from all mechanical support. The risk/benefit ratio with either invasive and noninvasive ventilation will need to be evaluated carefully in each case. Further investigation in terms of how NIV is provided in specific patient groups in such situations is warranted before it can be recommended as a viable alternative to continuing invasive support until a patient successfully passes an SBT.


  1. Ouellette DR, et al. Liberation from mechanical ventilation in critically ill adults: An official American College of Chest Physicians/American Thoracic Society clinical practice guideline: Inspiratory pressure augmentation during spontaneous breathing trials, protocols minimizing sedation, and noninvasive ventilation immediately after extubation. Chest 2017;151:166-180.
  2. Peñuelas O, et al. Characteristics and outcomes of ventilated patients according to time to liberation from mechanical ventilation. Am J Respir Crit Care Med 2011;184:430-437.
  3. Munshi L, Ferguson ND. Weaning from mechanical ventilation. What should be done when a patient’s spontaneous breathing trial fails? JAMA 2018;320:1865-1867.