Prolonged non-invasive ventilation in extremely low birth weight preterm infants is associated with bronchopulmonary dysplasia

Abstract

AIM:

To evaluate the association between the use of nasal continuous positive airway pressure (nCPAP) or nasal intermittent positive pressure ventilation (NIPPV) with the development of bronchopulmonary dysplasia (BPD).

METHODS:

This is a single center retrospective cohort analysis of infants born at ≤1000 grams and ≤28 weeks gestation with respiratory distress treated with nCPAP or NIPPV. Groups were compared using Student’s t test or chi-square, and associations estimated by logistic regression.

RESULTS:

Compared to nCPAP, infants who received NIPPV had a higher incidence of moderate to severe (M–S) BPD (84.2 vs 65.5%, p = 0.044) and death or severe BPD (75.0 vs 47.6%, p = 0.003). Each day on NIPPV was associated with an increased risk of M-S BPD (OR 1.08, p < 0.001) and an increased risk of death or severe BPD (OR 1.03, p = 0.006). After adjusting for days on oxygen, ventilator days, and days on all respiratory support, the odds of developing M-S BPD increased by 4.9% for each additional week on NIPPV (CI 2.1–7.7%, p = .0001).

CONCLUSION:

In this cohort, use of NIPPV was associated with an increased risk for developing BPD when compared to infants receiving nCPAP, and each additional day on NIPPV carried significant increased risk for developing BPD.

Keywords

Non-invasive ventilation chronic lung disease extremely low birth weight infants preterm neutrally-adjusted ventilatory assist bronchopulmonary dysplasia respiratory distress syndrome

Abbreviations

BPD

Bronchopulmonary Dysplasia

ELBW

Extremely low birth weight

Gestational age

HHFNC

Heated high flow nasal cannula

IVH

Intraventricular hemorrhage

NAVA

Neurally-adjusted ventilatory assist

nCPAP

Nasal continuous positive airway pressure

NIPPV

Non-invasive positive pressure ventilation

PDA

Patent ductus arteriosus

ROP

Retinopathy of prematurity

1. Introduction

Approximately 40% of extremely low birth weight (ELBW) infants develop bronchopulmonary dysplasia (BPD) during their initial hospitalization, and rates of BPD have been increasing throughout the United States, thought to be due in part to increased survival of these high-risk infants [1]. Non-invasive ventilation is frequently used in the neonatal intensive care unit to avoid intubation or as post-extubation support for spontaneously breathing infants with respiratory distress. The use of non-invasive ventilation has also significantly increased over the past several years, as most practitioners believe it to be less injurious to the developing lung than ventilation through an endotracheal tube [2, 3].

Despite the recent increase in the use of non-invasive modes of ventilation, long-term respiratory outcomes in ELBW infants appear to have worsened over the past ten years [4]. This may be in part due to the lack of data on what pressure ranges and duration of use are ideal and safe in this population. Additionally, it is unclear how the different modes compare to each other with regards to optimizing short-and long-term outcomes.

In neonates, the primary modes of non-invasive ventilation that have historically been available include nasal continuous positive airway pressure (nCPAP) and asynchronous nasal ventilation termed nasal intermittent positive pressure ventilation (NIPPV) [5]. More recently, other modalities such as heated high flow nasal cannula (HHFNC) have also become a mainstay of neonatal non-invasive respiratory support, however it is unclear how the distending pressure of HHFNC compares to the other modes due to the primary driver being flow rather than consistent pressure. Neurally-adjusted ventilatory assist (NAVA) is a unique mode of synchronous non-invasive positive pressure support in which both the timing and degree of ventilator assist are controlled by the patient through the use of diaphragmatic electromyography [6]. NAVA has been shown in premature infants to decrease the frequency and length of desaturations, fraction of inspired oxygen, respiratory rate, and peak inspiratory pressure over asynchronous ventilation, without significant adverse effects [5 , 7–9].

Several studies have assessed very short-term outcomes between modes of non-invasive respiratory support, and many have demonstrated improved outcomes with the use of NIPPV over nCPAP. Early NIPPV immediately or soon after birth is reported to be superior to nCPAP for decreasing the need for intubation and endotracheal tube ventilation and reduces the rate of extubation failure among preterm infants with respiratory distress syndrome [10 –14]. The effects of NIPPV and NAVA ventilation on longer-term outcomes such as BPD have not been well evaluated in premature infants.

Within our NICU, we perceived a gap in the comfort and experience of our clinical providers with regard to the use of NIPPV and NAVA, leading to some providers using these modes more commonly while others preferred the use of CPAP in these infants. We sought to take advantage of this pseudo-randomization to evaluate the associations between the use of NIPPV or NAVA and the development of adverse neonatal events such as pneumothorax, intraventricular hemorrhage (IVH), BPD, or death. We hypothesized that there would be no difference in these outcomes between the group of infants receiving NIPPV or NAVA (referred to as the NIPPV group) and the group receiving CPAP.

2. Subjects and methods

This single-center retrospective cohort study was approved by the University of Washington IRB committee. ELBW infants (birth weight ≤1000 g) born at ≤28 weeks gestational age (GA) between January 2011 and December 2015 were included in the study. Exclusion criterion included death before 48 hours age or infants transferred in after one week of age.

Using the electronic health record, we collected demographic data including sex, GA, birth weight, administration of prenatal steroids, and history of surfactant administration, as well as the need for and duration on various respiratory modalities and duration of time on oxygen. The outcomes that were recorded included length of stay and clinical disease outcomes or co-morbidities (BPD, patent ductus arteriosus (PDA), necrotizing enterocolitis, retinopathy of prematurity (ROP), grade 3–4 IVH, culture-positive sepsis, pneumothorax, discharge on oxygen, and death). For BPD grading, the NIH definition was used [15]. Briefly, to diagnose BPD under the NIH definition, infants must have an ongoing requirement for supplemental oxygen for 28 days. At 36 weeks corrected gestational age, infants are then categorized into mild if on room air, moderate if on oxygen support without need for positive pressure and at an FiO₂ < 0.3, or severe if requiring positive pressure and/or FiO₂≥0.3. Positive pressure was defined in this study as ≥2 LPM of HHFNC, any amount of nCPAP or NIPPV/NAVA, or endotracheal ventilation.

During the study period, our NICU started utilizing non-invasive NAVA, significantly decreasing the use of non-synchronized nasal ventilation. We were unable to adequately separate the two modalities using the medical record, so grouped them both under the “NIPPV group”. When comparing nCPAP and NIPPV, only infants that received one mode of support and not the other (e.g. nCPAP but never NIPPV) during their hospital stay were included in each group. There were no standard clinical guidelines that mandated the use of certain modes of respiratory support, the choice of settings, or the method or duration of weaning. In fact, this clinical variability was what allowed this study to be performed.

Statistical analyses were performed using STATA (version 14.2, StataCorp, College Station, TX). Student’s t and Chi-squared tests were performed for continuous and categorical variables, respectively. Logistic regression was performed to evaluate the association between NIPPV and death and/or BPD. A multivariable regression model was created to adjust for days on oxygen and duration of respiratory support. P-values and 95% confidence intervals were evaluated when appropriate, using an alpha level of 0.05 to test for statistical significance.

3. Results

286 infants met initial inclusion criteria of BW < 1000 g and GA≤28 weeks. 18 died at less than 48 hours of life and 29 were transferred in after 1 week of age. 60 infants received both nCPAP and NIPPV and 51 received neither, resulting in 128 infants included in the final analyses. 84 of the 128 infants received CPAP only and 44 received NIPPV only.

There were significantly more outborn infants in the NIPPV group (59.1%) compared to the nCPAP group (17.9%). There were no differences in GA, BW, ventilator days, or days on oxygen between the two groups (Table 1). There was a trend toward more of the infants in the NIPPV group receiving surfactant (63.6 versus 46.4%, p = 0.064). The clinical outcomes of the infants are summarized in Table 2. There was a significantly greater mean length of stay in infants who received NIPPV compared to those who received nCPAP (99.5 versus 83.4 days, p = 0.016). Compared to infants who received nCPAP without NIPPV, infants who received NIPPV without nCPAP had higher incidence of M-S BPD (84.2 versus 65.5%, p = 0.044) as well as death or severe BPD (75.0 versus 47.6%, p = 0.003).

Table 1
Demographics and baseline characteristics

Ncpap (n = 84) NIPPV (n = 44) P value

Birth weight (grams) 758±135 766±136 0.769

Gestational age (weeks) 26.1±1.4 25.5±1.4 0.226

Male 41 (48.8) 18 (40.9) 0.394

White race^a 56 (66.7) 31 (70.5) 0.663

Inborn 69 (82.1) 18 (40.9) <0.001

Surfactant administration 39 (46.4) 28 (63.6) 0.064

Prenatal steroids^a 79 (100) 36 (97.3) 0.142

Length of stay (days) 83.4±34.7 99.5±35.5 0.016

Days intubated 20.1±21.0 27.6±27.0 0.112

Days on oxygen 54.4±40.3 67.1±40.3 0.166

	Ncpap (n = 84)	NIPPV (n = 44)	P value
Birth weight (grams)	758±135	766±136	0.769
Gestational age (weeks)	26.1±1.4	25.5±1.4	0.226
Male	41 (48.8)	18 (40.9)	0.394
White race^a	56 (66.7)	31 (70.5)	0.663
Inborn	69 (82.1)	18 (40.9)	<0.001
Surfactant administration	39 (46.4)	28 (63.6)	0.064
Prenatal steroids^a	79 (100)	36 (97.3)	0.142
Length of stay (days)	83.4±34.7	99.5±35.5	0.016
Days intubated	20.1±21.0	27.6±27.0	0.112
Days on oxygen	54.4±40.3	67.1±40.3	0.166

^aData not available for race in two subjects (both in NIPPV group) and for prenatal steroids in 12 subjects (5 nCPAP and 7 NIPPV); Categorical variables described as N (%), continuous as mean±SD; nCPAP, nasal continuous positive airway pressure; NIPPV, non-invasive positive pressure ventilation.

Table 2

Clinical outcomes

	NCPAP (n = 84)	NIPPV (n = 44)	P value
PDA^a	42 (87.5)	38 (97.4)	0.090
Sepsis	12 (14.3)	2 (4.6)	0.094
Pneumothorax	4 (4.8)	1 (2.3)	0.477
Necrotizing enterocolitis	10 (11.9)	5 (11.4)	0.928
Severe IVH (Grade 3–4)	4 (4.9)	3 (6.8)	0.650
Threshold retinopathy of	6 (7.14)	8 (18.2)	0.057
prematurity
Any BPD	58 (79.5)	38 (92.7)	0.063
Moderate - severe BPD^b	38 (65.5)	32 (84.2)	0.044
Death	2 (1.2)	1 (2.3)	0.969
Death or severe BPD	40 (47.6)	33 (75.0)	0.003

^aData based on echocardiographic confirmation and not available in 41 subjects (36 resident and 5 NIPPV); ^bOnly infants who survived to 36 weeks corrected gestational age included in these calculations; Categorical variables described as N (%), Chi square for categorical variables; BPD, bronchopulmonary dysplasia; IVH, intraventricular hemorrhage; nCPAP, nasal continuous positive airway pressure; NIPPV, non-invasive positive pressure ventilation; PDA, patent ductus arteriosus.

Univariate regression demonstrated an association between each increased day on NIPPV with an increased risk of M-S BPD (OR 1.08, p < 0.001, CI 1.05–1.13) and the combined outcome of death or severe BPD (OR 1.03, p = 0.006, CI 1.01–1.06). After adjusting for days on oxygen, ventilator days, and days on all respiratory support in a multivariate model, the odds of developing M-S BPD increased by 4.9% for each additional week on NIPPV (p = .0001, CI 2.1–7.7%). Because of the increased number of outborn infants in the NIPPV group, a post-hoc analysis was performed that demonstrated that there was no significant difference in the rates of M-S BPD between inborn and outborn infants (57.7% versus 69.4%, respectively, p = 0.231) and adding it to the multivariable model did not change the association between NIPPV duration and M-S BPD.

4. Discussion

Contrary to our hypothesis, this study demonstrated a significant increase in the primary outcomes of BPD and combined BPD or death in ELBW infants receiving NIPPV for non-invasive respiratory support compared to those receiving CPAP. Further investigation demonstrated a significant association between NIPPV and M-S BPD even after adjusting for duration of oxygen, mechanical ventilation, and all respiratory support. These findings further underlie the concern that more data are needed to support the safe and effective use of non-invasive respiratory support in ELBW infants, including optimal pressure ranges and effective methods of rapid and safe weaning.

Although non-invasive respiratory support is thought to be less injurious to the developing lung than endotracheal ventilation, this study adds to the developing literature supporting the need to be careful in the choice of non-invasive respiratory modalities. Several studies have shown that early NIPPV immediately or soon after birth may be superior to nCPAP for decreasing the need for intubation and endotracheal tube ventilation and reducing the rate of extubation failure among preterm infants with respiratory distress syndrome [10 –14]. Longer-term outcomes, however, have not demonstrated consistent benefits of the use of NIPPV. Our study is one of the few published studies to date comparing NIPPV and nCPAP in the ELBW population; most have included infants up to 32 to 36 weeks gestation at birth. These studies that included the larger preterm infants have shown variable effects on BPD, ranging from no difference between the two modalities [16, 17] to a 4–25% decreased in all BPD [18, 19] or a 9% decrease in M-S BPD [20] in the NIPPV group.

One of the few previous studies of NIPPV versus nCPAP restricted to the ELBW or <30 week GA populations did not demonstrate any differences in BPD or the combined outcome of death and BPD, although the study did not assess the effects on different grades of BPD [21]. Our study also demonstrated no significant differences in the incidence of all BPD but did find a difference in M-S grade BPD. Contrary to our findings, Ramanathan, et al. studied a group of ELBW infants and demonstrated a 2.4 times increased odds of clinical BPD and a 6.6 times increased odds of physiological BPD in their nCPAP group compared to NIPPV [22]. While the difference between their outcomes and ours may be due to the benefits of standardized care that is inherent in randomized clinical trials, it is also possible that the discrepancy lies in the different populations and analyses. In the Ramanathan study, they enrolled infants born at 26–30 weeks GA versus ELBW infants ≤28 weeks in ours.

This study is limited by its retrospective nature. Although we were unable to control for initial severity of illness, we found no difference between the two groups with regards to several indicators of respiratory severity such as need for surfactant administration, duration of mechanical ventilation, and days on oxygen. Similarly, we could not capture the clinical indication for starting an infant on NIPPV versus nCPAP, so although the two groups were similar in the indicators of respiratory illness, there remains the possibility that infants on one mode had different pathophysiology or severity than the other. Lastly, we were also unable to distinguish between NAVA and asynchronous NIPPV. Although previous studies have not demonstrated any differences in the rates of BPD between synchronized and asynchronous NIPPV [23], we cannot make claims about each modality individually from these study results. In this study, ELBW infants receiving NIPPV for non-invasive respiratory support demonstrated a significant increase in BPD or death compared to those receiving CPAP. The fact that our findings are different from some previous prospective clinical trials suggests that the way NIPPV is utilized clinically (e.g. the settings used and method of weaning) may have a significant effect on clinical outcomes. These findings therefore do not necessarily suggest the need to decrease the use of NIPPV in this population, but rather highlight the need for the development of standardized guidelines for appropriate and active weaning from NIPPV. Further studies are needed to assess the effects of using varied NIPPV pressures and weaning protocols on long-term outcomes.

Statement of Conflict of Interest

The authors have no conflicts of interest to declare.

Statement of Funding

The authors received no funding for the work described in this manuscript.

Financial Disclosure

None.

Conflict of Interest

None.

Funding

None.

References

Stoll

, Hansen

, Bell

, Walsh

, Carlo

, Shankaran

, et al. Trends in care practices, morbidity, and mortality of extremely preterm neonates, 1993-2012. JAMA. 2015;314(10):1039–51.

Taha

, Kornhauser

, Greenspan

, Dysart

, Aghai

ZH.

High flow nasal cannulause is associated with increased morbidity and length of hospitalization in extremely low birth weight infants. J Pediatr. 2016;173:50–5e1.

Heath Jeffery

, Broom

, Shadbolt

, Todd

DA.

Increased use of heated humidified high flow nasal cannula is associated with longer oxygen requirements. J Paediatr Child Health. 2017.

Doyle

, Carse

, Adams

, Ranganathan

, Opie

, Cheong

JLY

, et al. Ventilation in extremely preterm infants and respiratory function at 8 years. N Engl J Med. 2017;377(4):329–37.

Stein

, Alosh

, Ethington

, White

DB.

Prospective crossover comparison between NAVA and pressure control ventilation in premature neonates less than 1500 grams. J Perinatol. 2013;33(6):452–6.

Stein

, Firestone

Application of neurally adjusted ventilatory assist in neonates. Semin Fetal Neonatal Med. 2014;19(1):60–9.

Gibu

, Cheng

, Ward

, Castro

, Heldt

GP.

Feasibility and physiological effects of noninvasive neurally adjusted ventilatory assist in preterm infants. Pediatr Res. 2017.

Lee

, Kim

, Jung

, Shin

, Choi

, Kim

, et al. Non-invasive neurally adjusted ventilatory assist in preterm infants: A randomised phase II crossover trial. Arch Dis Child Fetal Neonatal Ed. 2015;100(6):F507–13.

Beck

, Reilly

, Grasselli

, Mirabella

, Slutsky

, Dunn

, et al. Patient-ventilator interaction during neu-rally adjusted ventilatory assist in low birth weight infants. Pediatr Res. 2009;65(6):663–8.

10.

Lemyre

, Laughon

, Bose

, Davis

PG.

Early nasal intermittent positive pressure ventilation (NIPPV) versus early nasal continuous positive airway pressure (NCPAP) for preterm infants. Cochrane Database Syst Rev. 2016;12:CD005384.

11.

Lemyre

, Davis

, De Paoli

, Kirpalani

Nasal intermittent positive pressure ventilation (NIPPV) versus nasal continuous positive airway pressure (NCPAP) for preterm neonates after extubation. Cochrane Database Syst Rev. 2017;2:CD003212.

12.

, Long

, Zhangxue

, Jinning

, Shifang

, Juan

, et al. Nasal intermittent positive pressure ventilation versus nasal continuous positive airway pressure for preterm infants with respiratory distress syndrome: A meta-analysis and up-date. Pediatr Pulmonol. 2015;50(4):402–9.

13.

Salvo

, Lista

, Lupo

, Ricotti

, Zimmermann

LJI

, Gavilanes

AWD

, et al. Comparison of three noninvasive ventilation strategies (NSIPPV/BiPAP/NCPAP) for RDS in VLBW infants. J Matern Fetal Neonatal Med. 2018;31(21):2832–8.

14.

Meneses

, Bhandari

, Alves

JG.

Nasal intermittent positive-pressure ventilation vs nasal continuous positive airway pressure for preterm infants with respiratory distress syndrome: A systematic review and meta-analysis. Arch Pediatr Adolesc Med. 2012;166(4):372–6.

15.

Jobe

, Bancalari

Bronchopulmonary dysplasia. Am J Respir Crit Care Med. 2001;163(7):1723–9.

16.

Chen

, Wang

, Li

, Wang

, Shi

Noninvasive ventilation for preterm twin neonates with respiratory distress syndrome: A Randomized Controlled Trial. Sci Rep. 2015;5:14483.

17.

Kahramaner

, Erdemir

, Turkoglu

, Cosar

, Sutcuoglu

, Ozer

EA.

Unsynchronized nasal intermittent positive pressure versus nasal continuous positive airway pressure in preterm infants after extubation. J Matern Fetal Neonatal Med. 2014;27(9):926–9.

18.

Jasani

, Nanavati

, Kabra

, Rajdeo

, Bhandari

Comparison of non-synchronized nasal intermittent positive pressure ventilation versus nasal continuous positive airway pressure as post-extubation respiratory support in preterm infants with respiratory distress syndrome: A randomized controlled trial. J Matern Fetal Neonatal Med. 2016;29(10):1546–51.

19.

Esmaeilnia

, Nayeri

, Taheritafti

, Shariat

, Moghimpour-Bijani

Comparison of Complications and Efficacy of NIPPV and Nasal CPAP in Preterm Infants With RDS. Iran J Pediatr. 2016;26(2):e2352.

20.

Oncel

, Arayici

, Uras

, Alyamac-Dizdar

, Sari

, Karahan

, et al. Nasal continuous positive airway pressure versus nasal intermittent positive-pressure ventilation within the minimally invasive surfactant therapy approach in preterm infants: A randomised controlled trial. Arch Dis Child Fetal Neonatal Ed. 2016;101(4):F323–8.

21.

Kirpalani

, Millar

, Lemyre

, Yoder

, Chiu

, Roberts

RS.

A Trial Comparing Noninvasive Ventilation Strategies in Preterm Infants. N Engl J Med. 2013;369(7):611–20.

22.

Ramanathan

, Sekar

, Rasmussen

, Bhatia

, Soll

RF.

Nasal intermittent positive pressure ventilation after surfactant treatment for respiratory distress syndrome in preterm infants <30 weeks’ gestation: A randomized, controlled trial. J Perinatol. 2012;32(5):336–43.

23.

Dumpa

, Katz

, Northrup

, Bhandari

SNIPPV vs NIPPV: Does synchronization matter?

J Perinatol. 2012;32(6):438–42.