Non-invasive ventilation alternating with high-flow nasal oxygen versus high-flow nasal oxygen alone after extubation in COPD patients: a post hoc analysis of a randomized controlled trial

Background Several randomized clinical trials have shown that non-invasive ventilation (NIV) applied immediately after extubation may prevent reintubation in patients at high-risk of extubation failure. However, most of studies included patients with chronic respiratory disorders as well as patients without underlying respiratory disease. To date, no study has shown decreased risk of reintubation with prophylactic NIV after extubation among patients with chronic obstructive pulmonary disease (COPD). We hypothesized that prophylactic NIV after extubation may decrease the risk of reintubation in COPD patients as compared with high-flow nasal oxygen. We performed a post hoc subgroup analysis of COPD patients included in a multicenter, randomized, controlled trial comparing prophylactic use of NIV alternating with high-flow nasal oxygen versus high-flow nasal oxygen alone immediately after extubation. Results Among the 651 patients included in the original study, 150 (23%) had underlying COPD including 86 patients treated with NIV alternating with high-flow nasal oxygen and 64 patients treated with high-flow nasal oxygen alone. The reintubation rate was 13% (11 out of 86 patients) with NIV and 27% (17 out of 64 patients) with high-flow nasal oxygen alone [difference, − 14% (95% CI − 27% to − 1%); p = 0.03]. Whereas reintubation rates were significantly lower with NIV than with high-flow nasal oxygen alone at 72 h and until ICU discharge, mortality in ICU did not differ between groups: 6% (5/86) with NIV vs. 9% (6/64) with high-flow nasal oxygen alone [difference − 4% (95% CI − 14% to 5%); p = 0.40]. Conclusions In COPD patients, prophylactic NIV alternating with high-flow nasal oxygen significantly decreased the risk of reintubation compared with high-flow nasal oxygen alone. Trial registration The study was registered at http://www.clinicaltrials.gov with the trial registration number NCT03121482 (20 April 2017)

Background In intensive care units (ICUs), extubation failure rates leading to reintubation approximate 20-30% in patients with chronic respiratory disorders [1][2][3][4][5][6][7]. Several randomized clinical trials have shown that non-invasive ventilation (NIV) applied immediately after extubation may prevent reintubation in patients at high-risk of extubation failure [2,7]. However, the patients in these studies considered at high-risk have been heterogeneous and included patients with chronic respiratory disorders as well as patients without underlying respiratory disease. Whether prophylactic application of NIV after extubation reduces reintubation rate in patients with chronic obstructive pulmonary disease (COPD) has been poorly assessed. Previous clinical trials have shown that NIV could be particularly effective to prevent post-extubation respiratory failure in patients with chronic respiratory disorders, but again, these studies did not specifically focus on COPD patients and showed beneficial effects of NIV only in hypercapnic patients [3,4]. To date, only one small-scale trial has compared NIV with standard oxygen exclusively in COPD patients, and this study failed to highlight the beneficial effects of NIV [8]. Conversely, high-flow nasal oxygen could be as effective as NIV in preventing reintubation in patients at high-risk [9]. Therefore, we performed a post hoc subgroup analysis of a recent large-scale randomized controlled trial showing that prophylactic NIV alternating with high-flow nasal oxygen significantly reduced the risk of reintubation as compared to high-flow nasal oxygen alone in a heterogeneous population of patients at high-risk of extubation failure [7].
We hypothesized that prophylactic NIV alternating with high-flow nasal oxygen after extubation may be particularly effective in COPD patients and may decrease the risk of reintubation in this subgroup of patients as compared with high-flow nasal oxygen alone.

Study design and patients
The present study is a post hoc analysis of a multicenter, randomized, controlled trial comparing prophylactic use of NIV alternating with high-flow nasal oxygen versus high-flow nasal oxygen alone immediately after extubation in 641 patients at high-risk of reintubation in ICUs [7]. The present analysis focused on the subset population of COPD patients. Underlying COPD could be either documented by spirometry or highly suspected by physician team. Patients who were considered as having underlying COPD without previous spirometry needed to have been intubated for acute hypercapnic respiratory failure and for no other reason, and to have a history of smoking with dynamic hyperinflation during mechanical ventilation and/or emphysema on chest X-ray or scanner. Patients with long-term treatment with NIV or continuous positive airway pressure at home for sleep obstructive apnea syndrome were excluded. The original study was approved by the central ethics committee (Ethics Committee Ouest III, Poitiers, France) with registration number 2016-A01078-43. Written informed consent was obtained from all patients or next of kin before inclusion. According to French law, this post hoc analysis did not require further ethics approval.

Treatment groups
All patients were extubated after successful spontaneous breathing trial and received prophylactic NIV alternating with high-flow nasal oxygen or high-flow nasal oxygen alone during the first 48 h following extubation. In the two groups, treatment could be continued beyond the first 48 h following extubation until complete recovery of respiratory status. Patients assigned to the control group were continuously treated by high-flow nasal oxygen alone with a flow of 50 L/min and fraction of inspired oxygen (FiO 2 ) adjusted to obtain adequate oxygenation, with a pulse oximetry (SpO 2 ) at least 92%. Patients assigned to the interventional group were treated by NIV alternating with high-flow nasal oxygen. Non-invasive ventilation was initiated immediately after extubation with a first session of at least 4 h and minimal duration of at least 12 h a day during the 48 h following extubation. NIV was carried out with an ICU ventilator with NIV mode or dedicated bi-level ventilator, in pressuresupport mode with a minimal pressure-support level of 5 cm H 2 O targeting a tidal volume around 6-8 ml/kg of predicted body weight, a positive end-expiratory pressure level between 5 and 10 cmH 2 O and a FiO 2 adjusted to obtain adequate oxygenation (SpO 2 ≥ 92%). Between

Conclusions:
In COPD patients, prophylactic NIV alternating with high-flow nasal oxygen significantly decreased the risk of reintubation compared with high-flow nasal oxygen alone.
Trial registration The study was registered at http://www.clini caltr ials.gov with the trial registration number NCT03121482 (20 April 2017) Keywords: Airway extubation, Weaning, Non-invasive ventilation, High-flow nasal oxygen, Chronic obstructive pulmonary disease non-invasive ventilation sessions, high-flow nasal oxygen was delivered as in the control group.

Outcomes
The main outcome was reintubation rates within the 7 days following extubation according to oxygenation strategy. Secondary outcomes included post-extubation respiratory failure within the 7 days following extubation, reintubation rates at 48 h, 72 h and up until ICU discharge, and mortality in ICU.
Severe respiratory failure leading to reintubation was defined by the presence of at least two criteria among the following: respiratory rate > 35 breaths per minute, clinical signs of respiratory distress, respiratory acidosis defined as pH < 7.25 units and PaCO 2 > 45 mm Hg, hypoxemia defined as FiO 2 ≥ 80% to maintain SpO 2 ≥ 92% or a PaO 2 /FiO 2 ≤ 100 mm Hg.
An episode of post-extubation respiratory failure was defined by the presence of at least two criteria among the following: respiratory rate > 25 breaths per minute, clinical signs of respiratory distress, respiratory acidosis defined as pH < 7.35 units and PaCO 2 > 45 mm Hg, and hypoxemia defined as FiO 2 ≥ 50% to maintain SpO 2 ≥ 92% or a PaO 2 /FiO 2 ≤ 150 mm Hg.

Statistical analysis
Continuous variables were expressed as mean ± standard deviation (SD) or median [interquartile range, 25-75th percentiles], and qualitative variables were expressed as number and percentage. Patients' characteristics were compared between the NIV group and the high-flow nasal oxygen group by means of the χ 2 tests or Fisher exact test for categorical variables and Student's t-test or Wilcoxon rank-sum test for continuous variables as appropriate. Primary and secondary outcomes were compared between the two groups by means of the χ 2 test. Kaplan-Meier curves were plotted to assess the time from extubation to reintubation and were compared by means of the log-rank test at day 7. The results were presented as odds ratio (OR) with 95% confidence interval (95 CI). A two-tailed p value of less than 0.05 was considered statistically significant. We used SAS software, version 9.4 (SAS Institute), for all the analyses.

Results
Among the 651 patients extubated in the 30 participating ICUs, 150 (23%) had underlying COPD including 102 patients (68%) with COPD confirmed by spirometry. Median forced expiratory volume during the first second (FEV1) was 58% [interquartile range, 42-71%] expressed in percentage of predicted value according to sex and age. Among the 86 patients in whom FEV1 was available, 44% (38 of 86 patients) had severe COPD (stage 3 or 4 according to the Gold classification, i.e., FEV1 < 50% of predicted value).
Of the 150 COPD patients, weaning was considered as simple in 102 patients (68%), difficult in 44 (29%), and prolonged in 4 (3%). At time of extubation, 53 patients (35%) had hypercapnia (PaCO 2 > 45 mm Hg). After extubation, 86 patients were treated with prophylactic NIV alternating with high-flow nasal oxygen and 64 with high-flow nasal oxygen alone. The characteristics of the patients at inclusion were similar in the two groups aside from a higher proportion of patients with ineffective cough in the NIV group (Table 1).
The proportion of patients with post-extubation respiratory failure did not significantly differ between groups. The proportion of patients who received NIV to treat post-extubation respiratory failure was 86% (19 of 22 patients) in the NIV group and 26% (6 of 23 patients) in the high-flow oxygen group [difference 60%  Fig. 1 Kaplan-Meier analysis of time from extubation to reintubation according to oxygenation strategy. Reintubation rates were significantly lower in patients treated with non-invasive ventilation (red bars) than in those treated with high-flow nasal oxygen alone (blue bars). The reintubation rate within the 7 days following extubation was 13% (11 out of 86 patients) with NIV and 27% (17 out of 64 patients) with high-flow nasal oxygen alone [difference, − 14% (95% CI − 27% to − 1%); p = 0.033 using log-rank test] (95% CI 32-76%); p < 0.001]. However, the proportion of patients reintubated among those who experienced postextubation respiratory failure did not significantly differ between the 2 groups. Mortality in ICU did not differ between groups: 6% (5/86) with NIV vs. 9% (6/64) with high-flow nasal oxygen alone [difference − 4% (95% CI − 14% to 5%); p = 0.40].

Discussion
In this post hoc analysis of a randomized controlled trial focusing on COPD patients, prophylactic use of NIV alternating with high-flow nasal oxygen immediately after extubation significantly decreased the rate of reintubation as compared with high-flow nasal oxygen alone.
This trial is the largest one reporting the effects of prophylactic NIV after extubation in COPD patients. Previously, only one small-scale trial comparing NIV with standard oxygen exclusively in COPD patients failed to highlight the beneficial effects of NIV [8]. Another clinical trial showing decreased risk of post-extubation respiratory failure in patients with chronic respiratory disorders treated with NIV included a majority of COPD patients (74 of 106 patients) [4]. However, the risk of reintubation was not significantly lower with NIV than with standard oxygen and all patients had hypercapnia at time of extubation. Here, we report for the first time decreased risk of reintubation in COPD patients treated with prophylactic NIV after extubation.

Limitations
The main limitation of this study is the post hoc nature of the analysis. However, characteristics of the patients were similar between the 2 groups and the reintubation rates observed are very close to reintubation rates reported in previous studies (around 10% with NIV and between 20 and 30% with standard oxygen) reinforcing external validity [5,[7][8][9]. The only imbalance between groups was a higher proportion of patients with ineffective cough in the NIV group, situation associated with an increased risk of extubation failure [10,11]. Despite this unfavorable imbalance, prophylactic NIV was associated with a decreased risk of reintubation. This is in keeping with a previous study reporting that NIV may avoid reintubation in patients with weak cough as compared with standard oxygen [12].
Another major limitation is that more than one-third of patients had suspected but not confirmed COPD using spirometry. These patients were admitted for their first acute exacerbation of COPD or never underwent pulmonary function tests planned after hospital discharge, as is often the case. Indeed, several studies have reported the difficulty for follow-up of these patients and their reluctance to come back for pulmonary explorations after hospital discharge [13,14]. However, all these patients were admitted for acute hypercapnic respiratory failure and had common risk factors of COPD. Consequently, it is likely that the majority of patients with suspected obstructive spirometric pattern actually had underlying COPD.
Although NIV was the one additional treatment in the interventional group, the decreased risk of reintubation observed in the NIV group might be due to the combination of NIV with high-flow nasal oxygen between NIV sessions. Although the beneficial effects of NIV on oxygenation, alveolar ventilation, and work of breathing are well-demonstrated [15,16], continuation of high-flow nasal oxygen between NIV sessions may provide further clinical improvement by decreasing work of breathing [17,18]. Whereas prophylactic NIV could be strongly recommended in patients at high-risk of extubation failure and especially in COPD patients, whether NIV alternating with high-flow nasal oxygen is a better oxygenation strategy than NIV alternating with standard oxygen is a question that requires further investigation.
Lastly, the proportion of patients who experienced post-extubation respiratory failure as well as the proportion of patients reintubated among those with respiratory failure did not significantly differ. However, both rates were reduced, resulting in a significant decreased risk of reintubation with NIV as compared with high-flow nasal oxygen alone. Patients receiving prophylactic NIV were more likely to be treated with NIV in case of post-extubation respiratory failure than those receiving high-flow nasal oxygen alone. As a result, NIV may prevent postextubation respiratory failure and subsequently avoid reintubation among patients with respiratory failure. Even though the most recent international clinical practice guidelines suggest that NIV should not be used in the treatment of patients with established post-extubation respiratory failure [19], NIV as rescue therapy may avoid reintubation in a number of cases, especially in patients with chronic obstructive pulmonary disease [3][4][5]20].

Clinical implications
Several studies have suggested that prophylactic NIV after extubation may be particularly effective in hypercapnic patients [3,4,7]. However, less than 40% of patients included in our study had hypercapnia, and the majority of patients had moderate underlying COPD with a mean forced expiratory volume during the first second (FEV1) above 50% of predicted value. It should be emphasized that patients had not necessarily been intubated for acute exacerbation of COPD and that nearly one-third of them had been intubated for another reason than acute respiratory failure (cardiac arrest, shock, coma, or surgery). Prophylactic NIV may decrease the risk of reintubation even in patients with mild or moderate COPD and regardless of the level of PCO 2 before extubation, and thereby, NIV should be applied in all COPD patients to prevent extubation failure in the ICU.

Conclusion
Prophylactic use of NIV alternating with high-flow nasal oxygen immediately after extubation of COPD patients was associated with decreased risk of reintubation compared with high-flow nasal oxygen alone.

Funding
The study was conducted with the support of the "Programme Hospitalier de Recherche Clinique National 2015" from the French Ministry of Health (Grant Number 15-0060). The firm Fisher & Paykel provided to all the participating centers the high-flow nasal oxygen equipment and masks for non-invasive ventilation, but has no other involvement in the study.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate
The study has been approved by the central ethics committee (Ethics Committee Ouest III, Poitiers, France) with the registration number 2016-A01078-43 (06 September 2016).

Consent for publication
Written informed consent was obtained from all patients or next of kin before inclusion. According to French law, this post hoc analysis did not require further ethics approval.