Some randomized controlled trials (RCTs) have investigated the effects of preventive NIV in an unselected population immediately after extubation in comparison with standard treatment. There was no statistical difference in patient outcome, most likely because of the unselected candidates [8]. In early studies of unselected patients, only a few patients had previously diagnosed chronic obstructive pulmonary disease (COPD), and hypercapnia was not an included criterion to define the indication for primary NIV use after extubation. In an RCT comparing NIV versus standard medical treatment in patients with risk factors for weaning failure, the reintubation rate was lower in the NIV group, but there was no difference between groups in ICU and hospital length of stay and survival [9]. Patients with risk factors for respiratory failure after extubation were randomized to receive NIV or oxygen mask. NIV reduced the incidence of respiratory failure and improved the hospital survival rate. However, overall mortality was reduced only in a subgroup of hypercapnic patients. A study of selected hypercapnic patients reported a decreased incidence of respiratory failure, reintubation, and mortality rate at 3 months [10]. This demonstrates that preventive application of NIV in a specific targeted population is beneficial.

Risk factors for post-extubation failure include the following [9]:

NIV was considered a promising therapy after extubation failure to avoid reintubation at a 2001 International Consensus Conference in Intensive Care Medicine. This consideration was mainly based on information from uncontrolled studies or comparison with matched historical patients collective [11]. The results of RCTs addressing the use of NIV in the management of patients with respiratory failure after extubation were not, however, as promising as expected. One trial with more than 300 patients included defined respiratory distress (>30 bpm or respiratory rate >50 % from baseline) in 81 patients. These patients were randomized to NIV or standard treatment. The study showed no significant difference in reintubation, ICU stay, or survival [12]. Another RCT also showed no difference in reintubation rate or ICU stay and an increased mortality rate among the NIV group patients [13]. This was related to a longer time to reintubation in the NIV group. Both of these studies had a patient cohort with only about 10 % of patients having COPD. NIV in both studies was performed with active exhalation valve systems. In one study, the ventilation mode was volume controlled [13].