Comparison of Nasal bi-level Positive Airway Pressure Versus High-flow Nasal Cannula as a Means of Noninvasive Respiratory Support in Pediatric Cardiac Surgery

Abstract

Background:

Noninvasive respiratory support is often used in preventing postextubation respiratory failure in neonates and infants after cardiac surgery.

Aim:

We compared the efficacy of nasal Bilevel Positive Airway Pressure (N/BiPAP) with that of High- flow Nasal Cannula(HFNC) in prevention of post extubation respiratory failure and maintenance of gas exchange in neonates and infants undergoing cardiac surgery. The incidence of complications related to the use of these modes were also compared.

Settings and Design:

This is a retrospective review of medical records of patients in pediatric cardiac intensive unit of a high-volume center.

Methods:

A total of 100 patients who received noninvasive respiratory support postextubation were divided into N/BiPAP group and HFNC group. The two groups were compared for postextubation respiratory failure, gas exchange in arterial blood gas at 24 h of extubation, and incidence of complications, namely pneumothorax, abdominal distension, and device–interface-related pressure ulcers.

Results:

Fifty patients each received N/BiPAP and HFNC after extubation. Patients who received N/BiPAP were younger (2.68 ± 2.97 months vs. 6.94 ± 4.04 months, P = 0.001) and had longer duration of postoperative ventilation (106.98 ± 79.02 h vs. 62.72 ± 46.14 h, P = 0.001). The reintubation rates were similar (20% [n = 10] in N/BiPAP group vs. 8% [n = 4] in HFNC group, P = 0.074). The mean arterial PO₂ values at 24 h of extubation was 119.17 ± 56.07 mmHg for N/BiPAP group versus 123.32 ± 64.33 mmHg for HFNC group (P = 0.732). Arterial PCO₂ values at 24 h were similar (43.97 ± 43.64 mmHg in N/BiPAP vs. 37.67 ± 4.78 mmHg in HFNC, P = 0.318). N/BiPAP group had higher incidence of abdominal distension (16% [n = 8] vs. nil in HFNC group, P = 0.003) and interface-related pressure ulcers (86% [n = 43] vs. 14% [n = 7] P = 0.006).

Conclusion:

N/BiPAP and HFNC have comparable efficacy in preventing reintubation and maintaining gas exchange. HFNC has fewer complications compared to N/BiPAP.

INTRODUCTION

Neonates and young infants are often susceptible to respiratory failure in the postextubation period after cardiac surgery.[1,2] This is largely attributed to the effects of cardiopulmonary bypass (CPB) on the lung due to the systemic inflammatory response syndrome, increased extravascular lung water, and propensity for atelectasis.[1,2,3] Noninvasive respiratory support is frequently used as a bridging therapy to minimize respiratory failure and postextubation complications such as lung collapse and postextubation stridor. Conventionally, at our center, bi-level positive airway pressure (BiPAP) delivered by a standard pediatric ventilator through an endotracheal tube with its tip placed in the nasopharynx is used to provide noninvasive respiratory support to neonates and young infants following extubation. BiPAP provides two levels of positive airway pressure during the respiratory cycle with a higher level of pressure during inspiration (inspiratory positive airway pressure [IPAP]) and a lower level of pressure during expiration (expiratory positive airway pressure [EPAP]). High- flow nasal cannula (HFNC) is being increasingly used in pediatric patients to prevent respiratory failure.[4,6] The high-flow system offers a form of noninvasive respiratory support via a cannula which allows the delivery of high flow rates up to 25 L/min of humidified air and oxygen. The high gas flow facilitates washout of the pharyngeal dead space, reduction in nasopharyngeal resistance, improved lung compliance, and reduction in the work of breathing.[5] Although HFNC is widely used in neonates and children, there is paucity of data comparing this modality with the commonly used nasal BiPAP (N/BiPAP) in pediatric cardiac care settings. We compared N/BiPAP and HFNC with respect to the ability to prevent postextubation respiratory failure, maintenance of arterial PO₂ and PCO₂ at 24 h of extubation, as well as the incidence of complications, namely abdominal distension, pressure ulcers, and pneumothorax in neonates and infants after cardiac surgery.

METHODS

This was a retrospective study conducted by reviewing medical records of patients in the pediatric cardiac intensive care unit (PCICU) of a high-volume tertiary pediatric cardiac center. The study duration was 8 months (September 2015–April 2016); the retrospective review was conducted after obtaining approval of the institutional review board. A total of 100 consecutive neonates and infants < 1 year who received either form of noninvasive respiratory support using N/BiPAP or HFNC following extubation after congenital heart surgery were included in the data analysis. The need for informed consent was waived as this was a review of standard postoperative practice with respect to noninvasive respiratory support after pediatric cardiac surgery. As this was an early experience with the use of HFNC, choice of the mode and the need to initiate noninvasive support in individual patients was based primarily on the discretion of the attending pediatric cardiac intensivist evaluating the infant at the time of extubation. Patients who had preoperative airway abnormalities were excluded from the study.

All patients who received conventional N/BiPAP after extubation were included in the N/BiPAP group. All patients who received HFNC after extubation were included in the HFNC group. In the group N/BiPAP, N/BiPAP was delivered by a pediatric ventilator (Datex Engstrom ventilator, Datex Ohmeda Inc., Madison, WI, USA) through an endotracheal tube placed into the nose with its tip in the nasopharynx. The BiPAP settings typically included a positive end expiratory pressure (PEEP), equivalent to EPAP of 5 cmH₂O and a pressure support level above PEEP of 2–3 cmH₂O. The pressure support (IPAP) was initiated at 2 cmH₂O and increased depending on the respiratory status and comfort level of the patients. Humidification was provided using a heated humidifier (MR 850, Fisher and Paykel Healthcare Ltd., Auckland, NZ) incorporated in the circuit. The N/BiPAP was provided for a variable duration of 24–72 h depending on the respiratory status of the infant.

In group HFNC, patients received respiratory support through a HFNC (Optiflow™ Junior, Fisher and Paykel Healthcare Ltd.) with an integrated flow generator (AIRVO 2, Fisher and Paykel Healthcare Ltd., Auckland, NZ). The size of the cannula was chosen according to the age of the patient. The flow rate was calculated based on the weight of the infant as 2 L/Kg/min.[6] Humidification and warming of inspired gases was provided by a heated breathing tube and auto-fill humidification chamber.

All patients were nursed according to standard intensive care unit protocol and monitored by continuous electrocardiogram monitoring, pulse oximetry, and arterial blood pressure monitoring. The gas exchange was monitored by arterial blood gas measurements at 1 h after extubation and thereafter at 4^th hourly intervals. Patients with impending respiratory failure as evidenced by the clinical signs of respiratory distress and arterial blood gas analysis, were promptly re-intubated and initiated on mechanical ventilation. All patients had a nasogastric tube placed to facilitate gastric decompression while on noninvasive support. The patients were also monitored for potential adverse effects namely abdominal distension, development of pressure ulcers in the proximity of the nasal interface, and occurrence of pneumothorax. Abdominal distension was defined as an increase in abdominal girth >1 standard deviation from the baseline in the postoperative period.

Data were collected using a pro forma. The preoperative variables included age, sex, weight, diagnosis, and presence of Downs syndrome. Intraoperative parameters included surgical procedure, CPB time, and aortic cross-clamp (ACC) time. The two groups were compared for the efficacy of noninvasive respiratory support modality in preventing postextubation respiratory failure. The primary outcome variable namely postextubation respiratory failure was defined as need for reintubation within 72 h of extubation. Need for reintubation was decided based on the following criteria:

• Apneic episode lasting beyond 1 min and not recovering with bag and mask assistance

• Heart rate <60 beats/min

• Severe and persistent subcostal or intercostal retractions

• Lung collapse in chest radiograph

• Cardiovascular collapse requiring resuscitation

• FiO₂ requirement >70% to maintain SaO₂ >90% in patients with two-ventricle physiology and SaO₂ >75% in patients with single-ventricle physiology

• Respiratory acidosis with pH <7.30.

The efficacy of noninvasive respiratory support mode in terms of oxygenation and ventilation was compared by evaluating arterial PO₂ and PCO₂ at 24 h postextubation. The two groups were also compared with respect to the occurrence of adverse effects namely presence of abdominal distension, pneumothorax, and pressure ulcers related to device interfaces.

Statistical analysis

The data were analyzed using Statistical Package for Social Sciences version 20.0 (SPSS Inc. Chicago, IL, USA). The data were expressed as frequencies and percentages or as mean values and standard deviations as appropriate. To test the statistical significance of the difference in the mean values between the two groups, Student's t-test was applied. To test the statistical significance of the association between the categorical variables between the two groups, the Chi-square test was applied. P ≤ 0.05 was considered to be statistically significant.

RESULTS

A total of 100 consecutive patients who received noninvasive respiratory support by means of N/BiPAP or HFNC were included. Fifty patients who received N/BiPAP after extubation were included in N/BiPAP group. Fifty patients who received HFNC after extubation were designated as the HFNC group. The two groups were compared with respect to demographic data, preoperative variables, intraoperative parameters, and the outcome variables. The characteristics of patients in the two groups are summarized in Table 1. Patients who received N/BiPAP were younger (2.68 ± 2.97 months vs. 6.94 ± 4.04 months, P = 0.001) and had longer course of postoperative mechanical ventilation (106.98 ± 79.02 h vs. 62.72 ± 46.14 h, P = 0.001). The distribution of surgical procedures in the entire study cohort is depicted in Table 2.

Table 1.

Comparison of preoperative and intraoperative characteristics between high-flow nasal cannula and nasal bilevel positive airway pressure groups

Preoperative/intraoperative characteristic	HFNC	N/BiPAP	P
Male, n (%)	25 (50)	26 (52)	0.5
Age (months), mean±SD	6.94±4.04	2.68±2.97	0.001
Weight (kg), mean±SD	6.05±2.37	3.83±1.34	0.001
CPB time (min), mean±SD	144±76.621	153.98±73.97	0.518
ACC time (min), mean±SD	81.06±50.143	91.38±52.38	0.317

CPB=Cardiopulmonary bypass, ACC=Aortic cross clamp, HFNC=Highflow nasal cannula, N/BiPAP=Nasal bilevel positive airway pressure, SD=Standard deviation

Table 2.

Distribution of surgical procedures in the entire study cohort

Surgical procedure	n (%)
Ventricular septal defect repair	29 (29)
Tetralogy of Fallot repair	23 (23)
Arterial switch operation	11 (11)
TAPVC repair	7 (7)
Glenn shunt	6 (6)
Atrioventricular septal defect repair	5 (5)
PDA ligation	5 (5)
Coarctation of aorta repair	4 (4)
Pulmonary artery banding	3 (3)
Systemic to pulmonary artery shunt	2 (2)
ALCAPA repair	2 (2)
Miscellaneous	3 (3)

TAPVC=Total anomalous pulmonary venous connection, PDA=Patent ductus arteriosus, ALCAPA=Anomalous left coronary artery from pulmonary artery

Reintubation

The reintubation rates while on noninvasive respiratory support mode were compared between the two groups. Nearly 20% (n = 10) of the patients in the N/BiPAP group needed reintubation within 72 h compared to 8% (n = 4) in the HFNC group. Though the number of patients who required reintubation in the N/BiPAP group was higher, this was not statistically significant (P = 0.074).

Efficiency of oxygenation and ventilation

Comparing the efficacy of the two modes of noninvasive support in maintaining oxygenation and ventilation, the mean arterial PO₂ values at an FiO₂ ≤60% at 24 h of extubation in patients remaining extubated till this time point was 119.17 ± 56.07 mmHg for N/BiPAP group versus 123.32 ± 64.33 mmHg for HFNC group (P = 0.732).

The two groups were also compared with respect to arterial blood gas PCO₂ at 24 h. It was found that PCO₂ values at 24 h were similar in N/BiPAP and HFNC groups (43.97 ± 43.64 mmHg vs. 37.67 ± 4.78 mmHg, respectively, P = 0.318). The comparison of both groups with respect to postoperative parameters is depicted in Table 3.

Table 3.

Comparison of postoperative parameters between high-flow nasal cannula and nasal bilevel positive airway pressure groups

Parameter	HFNC	N/BiPAP	P
Reintubation, n (%)	4 (8)	10 (20)	0.074
Mean PO2 (mmHg) at 24 h postextubation	123.32±64.33	119.17±56.07	0.732
Mean PCO2 (mmHg) at 24 h postextubation	37.67±4.78	43.97±43.64	0.318
Ventilation hours, mean±SD	62.72±46.14	106.98±79.02	0.001
Pneumothorax, n (%)	1 (2)	1 (2)	0.753
Abdominal distension, n (%)	0 (0)	8 (16)	0.003
Pressure ulcers related to device-interface, n (%)	7 (14)	43 (86)	0.006

HFNC=High-flow nasal cannula, N/BiPAP=Nasal bilevel positive airway pressure, SD=Standard deviation

Adverse effects related to the use of noninvasive respiratory support mode

The incidence rates of pneumothorax were very low in both the groups. N/BiPAP group had a significantly higher incidence of abdominal distension (16% [n = 8]) versus nil in the HFNC group (P = 0.003). Device–interface-related pressure ulcers were also significantly higher in the N/BiPAP group (86% [n = 43]) versus 14% (n = 7) in the HFNC group (P = 0.006). The incidence rates of noninvasive mode-related complications are demonstrated in Table 3.

DISCUSSION

To our knowledge and after literature review, studies comparing the two modes of noninvasive respiratory support, namely N/BiPAP and HFNC in pediatric cardiac surgical patients, are limited. In this study, we have attempted to compare two commonly used modalities of noninvasive respiratory support, namely N/BiPAP and HFNC in postoperative pediatric cardiac surgical population. Neonates and infants were primarily included in our study because this group remains vulnerable to postextubation respiratory failure primarily due to the inherent physiological considerations different from the adult population.[7]

Although the gender distribution was similar between the two groups, it was found that patients in the N/BiPAP group were significantly younger and smaller than the patients who received HFNC. This could be due to the fact that physicians probably preferred conventional N/BiPAP in neonates and young infants as they might have anticipated greater probability of respiratory failure in this group. Longer CPB times and ACC times are considered as potential risk factors for extubation failure after congenital heart surgery. Longer CPB time is associated with increased risk of inflammatory response, increased edema, decreased respiratory compliance, and acute lung injury, all of which can adversely affect the likelihood of successful extubation.[8] In our study, both the groups were comparable with respect to the duration of CPB and the ACC time.

Pediatric patients who remain on mechanical ventilation for longer duration after cardiac surgery are more likely to have extubation failure.[9] In a large multicentric retrospective study which analyzed 1734 mechanical ventilation episodes in PCICUs, 5.8% (n = 100) of the patients had extubation failure and longer duration of mechanical ventilation was significantly associated with extubation failure.[10] Our data also revealed that patients who were transitioned to N/BiPAP had longer duration of postoperative mechanical ventilation than those who received HFNC. This suggests that patients with prolonged mechanical ventilation who are more vulnerable for extubation failure are more likely to receive N/BiPAP therapy after extubation.

The primary objective of our study was to compare the efficacy of N/BiPAP with that of HFNC in preventing postextubation respiratory failure after pediatric cardiac surgery. The outcome which determines failure of the respiratory support strategy, namely reintubation rates, was not significantly different between the two groups. Our results thus confirm that HFNC has comparable efficiency to conventional N/BiPAP in preventing postextubation respiratory failure. In addition, the mean arterial PO₂ and PCO₂ levels at 24 h were comparable, confirming equal efficacy of the two modes in maintaining postextubation gas exchange. Our findings corroborate with the conclusions of Yodder et al.[11] In this randomized trial on neonates, which compared the effects of nasal continuous positive airway pressure (NCPAP) with HFNC, the authors found that there was no difference in the need for reintubation between the two groups (15.5% in HFNC vs. 11.4% in NCPAP, P = 0.344).

Abdominal distension, air leaks resulting in pneumothorax, and pressure ulcers related to the interface of the respiratory support modes are some of the common adverse effects of N/BiPAP or HFNC, that are encountered in critical care settings. The incidence rate of pneumothorax was too low to allow comparison. However, infants in the HFNC group had a significantly lower incidence of interface-related pressure ulcers. Nasal trauma and interface-related pressure ulcers can cause disruption of the mucosal barrier and potentiate infection. In a recent observational study comparing the nasal continuous positive airway pressure with HFNC in preterm infants with respiratory distress, the authors could not find any difference in the occurrence of nasal trauma and interface-related injury between the two modes.[12] In our study, the definite reduction in pressure ulcers in the HFNC group makes it an attractive respiratory support mode, which would be better tolerated with less complications in neonates and smaller infants. In our study, all patients had a nasogastric tube placed for gastric decompression, however significant abdominal distension was noted only in the N/BiPAP mode. This indirectly lowers the risk for aspiration and subsequent pulmonary complications, which might result from abdominal distension.

Limitations of the study

Our study was a retrospective analysis of practices related to the use of various noninvasive respiratory support modes in pediatric patients after congenital heart surgery. The choice of the respiratory support mode was at the discretion of the primary care provider in the intensive care unit, and this might have added bias in the selection of one mode over the other, especially in younger population. Another limitation is the small sample size of the study cohort. We agree that randomized controlled studies with larger sample size are warranted to draw definite conclusions.

CONCLUSION

HFNC has comparable efficacy to nasal BiPAP as a non invasive respiratory support mode for preventing post extubation respiratory failure and maintaining gas exchange in infants after cardiac surgery. HFNC is associated with lower incidence of abdominal distension and device- interface related pressure ulcers compared to nasal BiPAP.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.