Risk Factors Associated with Late Failure of Noninvasive Ventilation in Patients with Chronic Obstructive Pulmonary Disease

Background Risk factors for noninvasive ventilation (NIV) failure after initial success are not fully clear in patients with acute exacerbation of chronic obstructive pulmonary disease (COPD). Methods Patients who received NIV beyond 48 h due to acute exacerbation of COPD were enrolled. However, we excluded those whose pH was higher than 7.35 or PaCO2 was less than 45 mmHg which was measured before NIV. Late failure of NIV was defined as patients required intubation or died during NIV after initial success. Results We enrolled 291 patients in this study. Of them, 48 (16%) patients experienced late NIV failure (45 received intubation and 3 died during NIV). The median time from initiation of NIV to intubation was 4.8 days (IQR: 3.4–8.1). Compared with the data collected at initiation of NIV, the heart rate, respiratory rate, pH, and PaCO2 significantly improved after 1–2 h of NIV both in the NIV success and late failure of NIV groups. Nosocomial pneumonia (odds ratio (OR) = 75, 95% confidence interval (CI): 11–537), heart rate at initiation of NIV (1.04, 1.01–1.06 beat per min), and pH at 1–2 h of NIV (2.06, 1.41–3.00 per decrease of 0.05 from 7.35) were independent risk factors for late failure of NIV. In addition, the Glasgow coma scale (OR = 0.50, 95% CI: 0.34–0.73 per one unit increase) and PaO2/FiO2 (0.992, 0.986–0.998 per one unit increase) were independent protective factors for late failure of NIV. In addition, patients with late failure of NIV had longer ICU stay (median 9.5 vs. 6.6 days) and higher hospital mortality (92% vs. 3%) compared with those with NIV success. Conclusions Nosocomial pneumonia; heart rate at initiation of NIV; and consciousness, acidosis, and oxygenation at 1–2 h of NIV were associated with late failure of NIV in patients with COPD exacerbation. And, late failure of NIV was associated with increased hospital mortality.


Introduction
Chronic obstructive pulmonary disease (COPD) is the fourth leading cause of death [1]. Acute exacerbations of COPD are responsible for more than 600,000 hospitalizations annually and result in direct costs of more than $20 billion in the United States [2]. Noninvasive ventilation (NIV) as an effective intervention has been used to manage patients with acute exacerbation of COPD for decades. It improves pH, reduces respiratory rate, reduces PaCO 2 , and subsequently reduces intubation rate and mortality [3,4]. Because of these advantages, use of NIV in patients with acute exacerbation of COPD has continuously increased in recent years [5,6]. Moreover, current guidelines strongly recommend NIV to be used in patients with acute exacerbation of COPD [7,8].
In spite of benefits from NIV in patients with acute exacerbation of COPD, late failure of NIV after initial improvement is not rare. It ranges from 8% to 23% [9][10][11][12]. e reasons for early failure of NIV (failure occurred at initial 48 h of NIV) have been widely discussed in patients with acute exacerbation of COPD [13][14][15][16][17][18]. However, only few studies have reported the reasons for late failure of NIV [9,12,19]. Because of small sample sizes, these studies only identified poor sleep, delirium, metabolic complications, and functional limitation were associated with late failure of NIV. us, we aimed to find other potential risk factors for late failure of NIV in patients with acute exacerbation of COPD.

Methods
We performed an observational study in a respiratory ICU of a teaching hospital from January 2012 to December 2015.
e study protocol was approved by our ethics committee and the institutional review board (the First Affiliated Hospital of Chongqing Medical University). Because of the observational nature, the informed consents were waived.
Patients who were admitted to our ICU for NIV as a firstline intervention because of acute exacerbation of COPD were screened for eligibility. COPD was diagnosed based on the guideline developed by our Respiratory Disease Committee, Chinese Medical Association in 2002 [20]. We enrolled the patients whose pH was less than 7.35 and PaCO 2 was more than 45 mmHg which were measured before NIV. However, we excluded those whose NIV was terminated because of clinical improvement, requirement of intubation, or death within 48 h of NIV. Late failure of NIV was defined as intubation or death during NIV after initial success [9].
In our department, NIV was managed by attending physicians, respiratory therapists, and nurses as the protocol reported previously [21]. e face mask (ZS-MZA Face Mask; Shanghai Zhongshan Medical Technology Co., Shanghai, China) was the first choice for NIV (BiPAP Vision or Respironics V60). Patients were positioned at 30°to 45°to avoid aspiration, if there were no contraindications to this positioning. Bi-level positive airway pressure (S/Tmode) was used for all patients. Expiratory positive airway pressure was initially set at 4 cmH 2 O and titrated according to the flow curve to ensure that expiratory flow reached zero prior to inspiration or diminished ineffective efforts. However, it was limited to less than 12 cmH 2 O. Inspiratory positive airway pressure was set at 8 cmH 2 O and increased by increments of 2 cmH 2 O to obtain a tidal volume of more than 6 mL/kg or to the maximum tolerated level for each patient. e inspiratory positive airway pressure was limited to less than 25 cmH 2 O. e fraction of inspired oxygen was set to maintain SpO 2 at around 95%. Humidification was provided by a heated humidifier. If humidification was inadequate, intermittent drinking was allowed. If respiratory failure was reversed, disconnection of NIV equipment was performed per hospital protocol [22].
Intubation was performed referencing the criteria as follows (one major criterion or at least two minor criteria), but it was determined at the discretion of the attending physicians [21]. Major criteria were (1) respiratory arrest, (2) loss of consciousness, (3) hemodynamic instability without response to fluids and vasoactive agents, (4) inability to correct dyspnea, (5) development of conditions necessitating intubation to protect the airway or to manage copious tracheal secretions, and (6) PaO 2 /FiO 2 below 100 mmHg. Minor criteria were (1) respiratory rate more than 35 breaths/min, (2) blood pH less than 7.30, (3) persistent tachycardia, (4) persistent activation of accessory respiratory muscles, and (5) PaO 2 /FiO 2 below 150 mmHg.
Nosocomial pneumonia was diagnosed by the methods we reported previously [23]. It was suspected if a patient had a radiographic infiltrate that was new or progressive, along with clinical findings suggesting infection, including new onset of fever, purulent sputum, leukocytosis, and decline in oxygenation. In patients with suspected pneumonia, respiratory tract culture was performed. Samples were obtained by coughing, nasotracheal suction, a protected specimen brush, or bronchoalveolar lavage. Nosocomial pneumonia was confirmed by positive culture and clinical presentations.
Data were analyzed by statistical software (SPSS 17.0; SPSS, Chicago, IL, USA) and reported as mean and standard deviation or median and interquartile range when appropriate. Normally distributed continuous variables were analyzed with the independent-sample t-test. Abnormally distributed continuous variables were analyzed with the Mann-Whitney U test. Categorical variables were analyzed by the chi-square or Fisher's exact test when appropriate. Within groups, a paired-sample t-test was used to analyze the data collected at NIV initiation and 1-2 h of NIV. Kaplan-Meier curves were used to analyze the proportions of intubation in patients with late failure of NIV. Independent risk factors for late failure of NIV were identified by multivariate logistic regression analysis. p < 0.05 was considered significant.

Results
We enrolled 291 patients in this study. After 48 h of NIV, 45 patients experienced intubation. e median time from initiation of NIV to intubation was 4.8 days (interquartile range (IQR): 3.4-8.1) (Figure 1). In addition, 3 patients reached the criteria of intubation. In spite of attending physicians, they did not benefit from intubation and continuous use of NIV. Finally the 3 patients died during NIV.
us, a total of 48 patients (16%) experienced late NIV failure after initial success.
Outcomes between patients with NIV success and late failure are summarized in Table 3. ere were no differences in duration of NIV and the length of stay in the hospital between the two groups. However, the patients with NIV success had shorter length of stay in the ICU (median 6.6, IQR: 4.9-9.8 vs. 9.5, 5.7-13.8, p � 0.02) and lower hospital mortality (3% vs. 92%, p < 0.01) than those with late failure of NIV.

Discussion
e current study found the incidence of late failure of NIV was 16% in patients with acute exacerbation of COPD with a relatively large sample size. Although some clinical variables improved both in the NIV success and late failure of NIV groups, the variables in the NIV success group improved faster than those in the late failure of NIV group. Nosocomial pneumonia; heart rate at initiation of NIV; and  consciousness, acidosis, and oxygenation at 1-2 h of NIV were associated with late failure of NIV. In addition, late failure of NIV was associated with increased hospital mortality. e mortality in patients with late failure of NIV was 68% in Moretti's study and 80% in Carratu's study [9,12]. In our study, the mortality was 92%, which was higher than the value reported by previous studies. We noted that most of the patients experienced NIV failure within 15 days of NIV in our study. However, some cases experienced NIV failure beyond 30 days. From 15 to 30 days of NIV, there was no NIV failure. It indicates that some patients had significantly impaired respiratory function and required prolonged noninvasive ventilation. In addition, longer exposure in the ICU is associated with a higher incidence of nosocomial pneumonia.
ese reasons contribute much to hospital mortality.
Previous studies reported that patients with late failure of NIV had higher APACHE II score, higher heart rate, lower GCS, and lower blood pressure compared with successful ones [12,19,24]. Our study also found similar results. Different from previous studies, we found nosocomial pneumonia was an independent risk factor for late failure of NIV. It reminds us that nosocomial pneumonia played an important role in late failure of NIV. Among the NIV patients who experienced nosocomial pneumonia in our study, NIV failure occurred in 78% of cases.
us, prevention of nosocomial pneumonia in NIV patients was as important as in those who received invasive mechanical ventilation. Both in the NIV success and late failure of NIV groups, most of the clinical variables significantly improved after 1-2 h of NIV. However, the respiratory rate, pH, PaCO 2 , and PaO 2 /FiO 2 improved faster in the NIV success group than those in the late failure of NIV group. ese results are new findings compared with previous studies [9,12,19,24]. ese data indicate that the patients in the late failure of NIV group responded not so well than those who experienced NIV success.
at may be the reason for initial improvement but later failure in the late failure of NIV group.
Our study has several limitations. We found nosocomial pneumonia was associated with late failure of NIV in a patient with acute exacerbation of COPD. However, we only enrolled 9 patients with nosocomial pneumonia. e small sample size may skew this result. us, the result is required to be validated with a larger sample size. Secondly, this study was only performed in a respiratory ICU. e single-center study also limited the results to extrapolate to other centers.
irdly, patients who received intubation later were  Canadian Respiratory Journal 5 associated with higher mortality [12,25]. erefore, early intubation (e.g., 24 h of NIV) is an alternative to reduce mortality.