Several studies have examined the use of NIV as an alternative to ongoing MV in patients who are deemed fit enough to start the process of weaning from MV but have failed a spontaneous breathing trial (SBT). The rationale for using NIV in this situation is that the patient will continue to receive mechanical support via NIV as their respiratory muscle strength and coordination recovers without being exposed to the potentially harmful effects of sedative drugs and ongoing endotracheal intubation. It has been suggested that weaning in this way with NIV may reduce the incidence of VAP and accelerate the weaning process compared with conventional methods of invasive weaning.

Several studies have compared NIV with conventional weaning in patients who have failed an SBT and then been randomized to either extubation onto NIV or continued weaning via MV. In two notable studies, NIV weaning was demonstrated to reduce time spent on MV, rates of VAP, and mortality when compared with invasive weaning [7, 10]. These studies predominantly included patients with underlying COPD, and the treatment groups received NIV for prolonged periods at high levels of pressure support.

A large randomized, controlled trial RCT, published in 2011, reported no benefit from NIV weaning compared with continued invasive weaning or extubation onto high flow oxygen [11], leading to debate regarding the merits of using NIV to wean patients from MV. However, a recent Cochrane review, which pooled data from 16 trials and 994 patients – including the VENISE trial mentioned above – concluded that NIV weaning in patients who had failed an SBT reduced mortality, ICU and hospital length of stay, and rates of VAP, and that these benefits were more pronounced in patients with underlying COPD [12].

Use of NIV to wean patients is perhaps underutilized by clinicians, as it requires something of a “leap of faith” to extubate patients who have failed SBTs and therefore do not meet criteria for extubation. There has also been some conflict in the literature as to the benefits to be gained by weaning patients using NIV. However, a number of studies suggest that in COPD patients NIV weaning will reduce rates of VAP, ICU length of stay, and mortality. Therefore, the use of NIV weaning should be considered in all patients with known COPD who are ready to wean from MV but not suitable for extubation.

Extubation of ICU patients who have received MV for a period of time carries the risk of extubation failure and the need for further MV. Although the reported rate of extubation failure in the literature varies, it may be as high as 19 % [13]. It is also widely acknowledged that failing an extubation is associated with worse outcomes and increased risk of morbidity and mortality, although this may be a result of sicker patients with more comorbidities having a higher risk of extubation failure rather than a direct effect of reintubation per se. The use of NIV in recently extubated patients is an attractive treatment option, as it has the potential to provide ongoing respiratory support to recently extubated patients without the attendant risks of endotracheal intubation and MV, and several studies have examined the use of NIV in this setting.

NIV has been assessed as a preventative strategy in ventilated ICU patients who have risk factors for post-extubation failure, such as age greater than 65, poor cough, cardiac and respiratory comorbidity, and hypercapnia (while ventilated or preexisting). The application of NIV immediately post-extubation for periods of up to 48 h was demonstrated to reduce reintubation rates and mortality in one large RCT [8]. NIV used prophylactically has also been demonstrated to reduce the incidence of respiratory failure post-extubation when used for up to 24 h post-extubation [14], and a later study of 106 patients with chronic respiratory disease demonstrated that prophylactic NIV use for 24 h following extubation reduced respiratory failure and 90-day mortality when compared with standard medical therapy [15].

There have been several RCTs examining the use of NIV as a rescue treatment for post-extubation respiratory distress. Early work suggested that application of NIV to patients with premorbid cardiorespiratory disease who developed respiratory failure post-extubation did not reduce reintubation rates, duration of MV (mechanical ventilation), hospital mortality, or length of stay compared with standard therapy [16]. A subsequent multicenter RCT reported that patients who had been extubated following a successful SBT but then developed post-extubation respiratory failure had an increased ICU mortality if then treated with NIV compared with standard medical therapy [17]. There has been some criticism of this trial, and it is important to note that the patients who failed on NIV and went on to require intubation had received long periods of ineffective NIV before reintubation – on average 9 h longer than the controls – which is likely to have contributed to their worse outcomes. Additionally, post hoc analysis of patients with COPD in this study suggested that use of NIV may still be warranted if used judiciously in post-extubation respiratory distress. However, in general, the onset of post-extubation respiratory failure is an ominous development and delaying reintubation by any means risks potential harm and detriment to the patient.