|Year : 2022 | Volume
| Issue : 4 | Page : 223-227
A prospective observational study to determine the predictors of failure of non-invasive ventilation in patients with type 2 respiratory failure
Mohammed Fahad Khan, Vasudeva Acharya, Barkur Ananthakrishna Shastry
Department of Internal Medicine, Kasturba Medical College, Manipal, Karnataka, India
|Date of Submission||18-Mar-2022|
|Date of Decision||14-May-2022|
|Date of Acceptance||14-May-2022|
|Date of Web Publication||04-Oct-2022|
Professor, Department of Internal Medicine, Kasturba Hospital, Manipal 576 104, Karnataka
Source of Support: None, Conflict of Interest: None
Background: This study aims to evaluate clinical and biochemical parameters likely to predict the failure of non-invasive ventilation (NIV) in patients with type 2 respiratory failure.
Methods: The study was conducted from September 2016 to August 2018. Clinical and biochemical parameters of 151 patients were analysed after the application of NIV and divided into two groups, success group if they remained on NIV after 48 h and failure group if intubated.
Results: Among 151 patients, 65 patients (43%) were intubated and included in the failure group. The remaining 86 patients (57%) were included in the success group. Of the 65 patients who got intubated, 16 patients died while 49 recovered. Out of the 151 patients enrolled the gender distribution was 65 males and 86 females. There was no statistically significant association of the outcome of NIV usage with the comorbidity of the patient and with the underlying lung pathology of the patient. Seventy-two per cent (47 patients) of the NIV failures were in the first 6 h. Lower pH, higher partial pressure of carbon dioxide, higher heart rate and respiratory rate (RR) at baseline showed a statistically significant association with the failure of NIV usage in patients with type 2 respiratory failure. Failure of clinical and biochemical parameters to improve at various time points was also associated with increased chances of failure of NIV usage.
Conclusions: Linear regression analysis showed that pH and RR at baseline and presence of severe pulmonary artery hypertension were found to be predictors of failure of NIV.
Keywords: Chronic obstructive pulmonary disease, non-invasive ventilation, type 2 respiratory failure
|How to cite this article:|
Khan MF, Acharya V, Shastry BA. A prospective observational study to determine the predictors of failure of non-invasive ventilation in patients with type 2 respiratory failure. J Clin Sci Res 2022;11:223-7
|How to cite this URL:|
Khan MF, Acharya V, Shastry BA. A prospective observational study to determine the predictors of failure of non-invasive ventilation in patients with type 2 respiratory failure. J Clin Sci Res [serial online] 2022 [cited 2022 Dec 5];11:223-7. Available from: https://www.jcsr.co.in/text.asp?2022/11/4/223/357842
| Introduction|| |
Respiratory failure is characterised by the failure of the respiratory system to carry out one or both of the functions of gas exchange: oxygenation and carbon dioxide elimination. Respiratory failure is defined as malfunction of one or more components of the respiratory system leading to inadequate gas exchange.
Mechanical ventilation is the treatment for acute respiratory failure. It can be provided by an invasive method or a non-invasive method. Invasive mode of mechanical ventilation can cause complications such as ventilator-associated pneumonia, sinusitis, especially in patients with comorbidities and who are immunosuppressed.,
These complications can be reduced when ventilation is provided non-invasively. However, non-invasive ventilation (NIV) is not associated with 100% success rate., There are a group of patients who require invasive ventilation. Failure to identify this group of patients leads to increased mortality. Clinical and laboratory parameters can be used as a guide to predict the outcome of NIV so that delay in intubation can be avoided. The present study was designed to evaluate clinical parameters which are likely to predict the failure of NIV in patients with type 2 respiratory failure.
| Material and Methods|| |
The study was a prospective observational study conducted at Kasturba Hospital, Manipal, between September 2016 and August 2018.
Patients with acute hypercapnic respiratory failure aged above 18 years started on NIV in the ICU. Patients with haemodynamic and cardiac instability (hypotension blood pressure [BP] <90/60 mmHg, myocardial infarction and arrhythmia), need for intubation to protect airways, impending respiratory arrest, facial deformity/trauma, claustrophobia, upper airway obstruction and severe gastrointestinal bleed were studied.
A brief clinical history of patients satisfying the inclusion criteria was obtained. Baseline clinical parameters including heart rate (HR) respiratory rate (RR), BP and laboratory parameters like arterial blood gas analysis were recorded before initiating NIV (baseline) and at 1, 6–12, 24 and 48 h after initiation of NIV. Electrocardiography features like P- pulmonale, echocardiography features such as pulmonary artery hypertension, tricuspid regurgitation and right ventricular hypertrophy and chest X-ray of the patient was also analysed and recorded. Change in these parameters and the need for intubation were recorded and analysed. Patients were considered a part of the success group if they were still on NIV after 48 h or weaned off NIV. They were included in the failure group if they were intubated at any point up to 48 h. Clinical parameters that were evaluated are HR, RR, BP and mental status. Biochemical parameters evaluated were pH, arterial carbon dioxide tension (PaCO2), bicarbonate and partial pressure of oxygen.
Mean and standard deviation (SD) were used summarise continuous variables in normal distribution. Frequency and percentage were used to summarise categorical variables. Independent sample t-test was used to compare clinical and biochemical parameters between success and failure groups. Linear regression analysis was used to determine the predictors of failure of NIV. Change in HR, RR, pH and PaCO2 between success and failure groups from baseline, 1 h and 6–12 h was analysed. Chi-square test was used to compare categorical variables among the groups. P < 0.05 was considered statistically significant. Data analysis was done using Statistical Package for the Social Sciences (SPSS) version 23 (IBM Corp., Somers NY, USA).
| Results|| |
In this study, 151 patients who fit into the inclusion criteria were studied. Out of the 151 patients 65 patients (43%) got intubated and were included in the failure group. The remaining 86 patients (57%) were included in the success group. Out of the 65 patients who got intubated, 16 patients died while the remaining 49 recovered. There was no mortality in the patients of the success group. Out of the 65 patients who got intubated, 47 patients were intubated in the first 6 h.
Majority of the patients (n = 58; 38.4%) belonged to 61–70 years age group. Their mean age was 63.5±11.8 years; there were 86 females. The most number of intubations were in the age group of 61–70 years (25 patients). Among the male patients, 25 required intubation while 45 females required intubation in this study.
[Table 1] shows the distribution of underlying respiratory illness in the NIV success and failure groups. Biochemical and clinical parameters of the patients at baseline which is before the application of NIV [Table 2]. Baseline clinical parameters like HR, RR and biochemical parameters like pH, PaCO2 showed a statistically significant association with failure of NIV application. Higher the HR, RR, PaCO2 and lower the pH, greater are the chances of failure of NIV. There was no association of bicarbonate with the success or failure of NIV. Eighty-four patients had a pH of <7.25. Among these 84 patients, 50 patients (59.5%) belong to the failure group. pH of more than 7.25 was seen in 67 patients and 52 (77.6%) of these patients belonged to the success group. Hence, pH of <7.25 had a greater chance of NIV failure. There are 55 patients with PaCO2 more than 80 mm Hg. Out of these 55 patients, 34 (61.8%) belonged to the failure group. Hence, a PaCO2 of more than 80 mm Hg had a greater chance of NIV failure. Biochemical and clinical parameters at 1 h are shown in [Table 3].
|Table 1: Distribution of underlying respiratory illness in the NIV success and failure groups|
Click here to view
Ten patients were intubated with the 1st hour of NIV application. Hence, at 1 h the total number of patients under consideration was 141. Patients in the NIV failure group even after an hour continued to have a lower pH, higher PaCO2, higher HR and a higher RR compared to the success group. As can be seen by comparing with the parameters of the NIV success group at baseline [Table 1], patients in the success group at 1 h showed an improvement in pH, PaCO2, HR and RR.
Sixty-four patients had a pH of less than 7.25, out of which 48 patients (75%) belonged to the NIV failure group. Hence, after 1 h of NIV application, pH of <7.25 had a greater chance of NIV failure. Fifty-one patients had PaCO2 more than 80 mm Hg. Out of these 51 patients, 38 patients (74.5%) had failure of NIV. Hence, after 1 h of NIV application, PaCO2 of more than 80 mmHg had greater chances of NIV failure. Biochemical and clinical parameters at 6–12 h after NIV application are shown in [Table 4].
Since 47 patients were intubated in the first 6 h, the patients included in the failure group at 6–12 h were 18 patients. There was one patient in the success group who no longer required NIV and was also excluded from this time period. Hence, a total number of patients studied and analysed at 6–12 h of NIV application are 103 patients.
Patients in the failure group even after 6–12 h continued to have a lower pH, higher PaCO2, higher HR and a higher RR. Comparing the parameters at baseline [Table 1], at 1h [Table 2], a significant improvement in pH, PaCO2, HR, RR is seen at 6–12 h in the success group.
Twenty-six patients have a pH of <7.25. Out of these 26 patients, 15 patients (57.7%) had a failure of NIV application. Hence, at 6–12 h, after NIV application patients with pH lower than 7.25 had a greater chance of failure. Eighteen patients had a PaCO2 of more than 80 mmHg. Out of these 18 patients, 12 patients (66.67%) belonged to the failure group. Hence, at 6–12 h, after NIV application, a PaCO2 of more than 80 mmHg had a greater chance of failure. The clinical and biochemical parameters were not evaluated at time points of 12–24 h and 48 h as the number of patients in the failure group was only six. Pulmonary arterial hypertension (PAH) and NIV outcome are shown in [Table 5].
Pearson coefficient for PAH was 0.069. Linear by Linear association was present 0.011, which means that with the severity of PAH the chances of failure of NIV also increase.
Predictors of failure of NIV were determined by linear regression analysis as shown in [Table 6].
|Table 6: Comparison of the non-invasive ventilation success and failure group|
Click here to view
| Discussion|| |
The current study was undertaken to identify factors which predict the outcome of NIV in patients with Type 2 respiratory failure. The study consisted of 151 subjects with acute hypercapnic respiratory failure who were initiated on NIV. Based on whether, they required invasive mechanical ventilation with endotracheal intubation they were classified into success and failure groups. Success group involved those patients who continued to remain on NIV after 48 h or were weaned off before 48 h of NIV application. The failure group included those patients who needed invasive ventilation within 48 h of NIV application.
The intubation rate within 48 h of NIV application in our study was 42% which is similar to the results obtained in a meta-analysis where the intubation rate was 40% in the NIV group. It was observed that majority of the patients were in the 50–80 years of age subgroups (mean±SD = 63.5±11.8). This is possible because the most common causes such as chronic obstructive pulmonary disease (COPD) and pneumonia commonly affect populations in these age groups.
In a study the average age of the patients was noticed to be 45.7 years. Similarly, another study had noticed that the mean age in the failure group was 62.3±5 which is very similar to the results we obtained.
Out of 151 patients included in the study, 86 of them were female. There was no statistically significant relation observed between the gender of the patient and NIV failure.
A varied spectrum of lung pathologies was studied in these 151 patient. A Considerable overlap between these pathologies was seen and most patients had more than one underlying lung pathology. We observed no statistically significant association between the underlying lung pathology and the outcome of NIV. This is probably because of the heterogeneous nature of the lung pathologies of these patients and the fact that one patient did not have an exclusive underlying problem.
In the hypercapnic respiratory failure, there was a significant difference in mean of baseline parameters among the success and failure groups. On analysis of pH, PaCO2, HR and RR at baseline had a statistically significant association with the outcome of NIV. Linear regression analysis showed that the baseline clinical and laboratory parameters that can predict the outcome in hypercapnic respiratory failure in our study were pH at baseline, RR at baseline and severity of the pulmonary artery hypertension.
In a study baseline HR and RR was significantly higher in failure patients as compared to successful patients. Other studies have showed that baseline pH predicts outcome among patients with hypercapnic respiratory failure. In a study initial lower pH and higher PaCO2 were associated with a higher chance of failure of NIV.
The fact that a lower pH leads to increased chances of failure of NIV is explained by the fact that even though the patient is kept on NIV, he still uses his respiratory muscles for respiration which can further worsen the acidosis. Hence, intubating patients with a pH < 7.20 would help relax their respiratory muscles and improve the acidosis as the blood would now be shunted away from the respiratory muscles due to decreased demands.
At the end of 1 h, there was a significant improvement in the parameters in the success group showing an increase in pH, a decrease in PaCO2, HR and RR. However, the failure group failed to show such improvement. In a study. initial pH of failure group was significantly lower and did not improve after 1 hour of NIV initiation. In a study it was shown that if the pH measured after 2 hours of NIV support is <7.25, the likelihood of NIV failure is higher.
In our study at the end of 1 hour, 48 patients (75%) out of 64 patients with a pH <7.25 belonged to the NIV failure group. Seventy-four per cent of the patients who had a PaCO2 of greater than 80 mm Hg belonged to the failure group. In several studies,,,,,,, similar results were obtained where the failure group continued to have a lower pH and a higher PaCO2.
At the end of 12 hours, there was a significant reduction in the HR and RR among the success group which was not noticed in the failure group. Those who had a persisting high RR at the end of 12 hours failed to improve on NIV.
The pH showed a significant improvement in the success group. Persistent respiratory acidosis after the initiation of NIV is a predictor of failure. Fifteen out of 26 patients who at this point had a pH <7.25 belonged to the failure group. Twelve out 18 patients at this time point who had a PaCO2 >80 mmHg belonged to the failure group. These results were similar to other reports,,,, where improvement in pH and decrease in PaCO2 over time predicted success of NIV.
An interesting point noted in our study was the linear association seen with the severity of pulmonary artery hypertension and NIV failure. The increased PaCO2 at baseline in these type 2 respiratory failure patients is believed to act as a pulmonary vasoconstrictor and as this comes down with mechanical ventilation the PAH severity is presumed to decrease, however not normalise. The studies reviewed do not mention the same. However, a case report showed a similar result.
In our study, lower pH, higher PaCO2, higher HR and RR showed a statistically significant association with the failure of NIV usage in patients with type 2 respiratory failure. The presence of pulmonary artery hypertension on echocardiography showed a statistically significant association with the failure of NIV usage. Linear regression analysis showed that pH at baseline, RR at baseline and presence of severe pulmonary artery hypertension were found to be predictors of failure of NIV usage.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Pierson DJ. Noninvasive positive pressure ventilation: History and terminology. Respir Care 1997;42:370-9.
Nava S, Fanfulla F. The Interfaces for NIV. Non invasive artificial ventilation - How, When and Why: Springer Milan; 2014. p. 15-9.
Ram FS, Picot J, Lightowler J, Wedzicha JA. Non-invasive positive pressure ventilation for treatment of respiratory failure due to exacerbations of chronic obstructive pulmonary disease. Cochrane Database Syst Rev 2004:CD004104.
Agarwal R, Gupta R, Aggarwal AN, Gupta D. Noninvasive positive pressure ventilation in acute respiratory failure due to COPD vs. other causes: Effectiveness and predictors of failure in a respiratory ICU in North India. Int J Chron Obstruct Pulmon Dis 2008;3:737-43.
Soliman MA, El-Shazly MI, Soliman YM, Mostafa AI. Effectiveness of non invasive positive pressure ventilation in chronic obstructive pulmonary disease patients. Egypt J Chest Dis Tuberc 2014;63:309-12.
Singh VK, Khanna P, Rao BK, Sharma SC, Gupta R. Outcome predictors for non-invasive positive pressure ventilation in acute respiratory failure. J Assoc Physicians India 2006;54:361-5.
Confalonieri M, Garuti G, Cattaruzza MS, Osborn JF, Antonelli M, Conti G, et al.
A chart of failure risk for noninvasive ventilation in patients with COPD exacerbation. Eur Respir J 2005;25:348-55.
Kumar A, Kumar A, Rai K, Ghazal S, Rizvi N, Kumar S, et al
. Factors leading to poor outcome of noninvasive positive pressure ventilation in acute exacerbation of chronic obstructive pulmonary disease. J Acute Dis 2015;4:44-7.
Crummy F, Buchan C, Miller B, Toghill J, Naughton MT. The use of noninvasive mechanical ventilation in COPD with severe hypercapnic acidosis. Respir Med 2007;101:53-61.
Moretti M, Cilione C, Tampieri A, Fracchia C, Marchioni A, Nava S. Incidence and causes of non-invasive mechanical ventilation failure after initial success. Thorax 2000;55:819-25.
Lightowler JV, Elliott MW. Predicting the outcome from NIV for acute exacerbations of COPD. Thorax 2000;55:815-6.
Antón A, Güell R, Gómez J, Serrano J, Castellano A, Carrasco JL, et al.
Predicting the result of noninvasive ventilation in severe acute exacerbations of patients with chronic airflow limitation. Chest 2000;117:828-33.
Girault C, Briel A, Hellot MF, Tamion F, Woinet D, Leroy J, et al.
Noninvasive mechanical ventilation in clinical practice: A 2-year experience in a medical Intensive Care Unit. Crit Care Med 2003;31:552-9.
van Gemert JP, Brijker F, Witten MA, Leenen LP. Intubation after noninvasive ventilation failure in chronic obstructive pulmonary disease: Associated factors at emergency department presentation. Eur J Emerg Med 2015;22:49-54.
Plant PK, Owen JL, Elliott MW. Non-invasive ventilation in acute exacerbations of chronic obstructive pulmonary disease: Long term survival and predictors of in-hospital outcome. Thorax 2001;56:708-12.
Meduri GU, Turner RE, Abou-Shala N, Wunderink R, Tolley E. Noninvasive positive pressure ventilation via face mask. First-line intervention in patients with acute hypercapnic and hypoxemic respiratory failure. Chest 1996;109:179-93.
Carlucci A, Richard JC, Wysocki M, Lepage E, Brochard L; SRLF Collaborative Group on Mechanical Ventilation. Noninvasive versus conventional mechanical ventilation. An epidemiologic survey. Am J Respir Crit Care Med 2001;163:874-80.
Dursunoglu N, Dursunoglu D, Moray A, Gur S, Kavas M. A rapid decrease in pulmonary arterial pressure by noninvasive positive pressure ventilation in a patient with chronic obstructive pulmonary disease. Indian J Crit Care Med 2007;11:90-2. [Full text]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]