Keep the PEEP: A quality improvement project assessing safety and feasibility of prolonged non-invasive positive end expiratory pressure

Abstract

Introduction

Infants born less than 32-week gestation are at increased risk of respiratory distress syndrome due to underdeveloped pulmonary alveoli. Despite known respiratory benefits associated with positive end expiratory pressure (PEEP), our staff expressed concern that prolonged exposure to PEEP would contribute to increased risk of pneumothorax and delayed initiation of oral feeding. This quality improvement project aimed to standardize PEEP use with a guideline for early initiation, continuation, and subsequent discontinuation in infants born at less than 32-week gestation.

Methods

We developed and implemented a “Keep the PEEP” guideline, which included recommendations for initiation of PEEP at delivery, continuation through 32-week post-menstrual age (PMA), and discontinuation. Data collection occurred between July 2019-July 2021 (pre-intervention) and July 2021-August 2023 (post-intervention). Data analysis included run charts for monthly compliance, and bivariate analysis of pre- and post-intervention data. Interventions included team member education and team engagement through auditing.

Results

We achieved a median compliance of 100% with PEEP use through 32-week PMA. We found no statistically significant differences in a pre- versus post-intervention comparison of pneumothorax incidence [14 versus 13, p = 0.86], bronchopulmonary dysplasia (BPD) incidence [82 versus 68, p = 0.65], median post-menstrual age at first oral feed [34.1 versus 34.1, p = 0.56], or median length of stay [55 versus 43, p = 0.2].

Conclusions

We successfully implemented a new respiratory guideline to standardize PEEP use in our unit during the study period. Despite initial concerns, the incidence of pneumothorax and the median gestational age at initiation of oral feedings remained unchanged.

Keywords

oral feeding pneumothorax positive end expiratory pressure (PEEP)respiratory distress

Introduction

Infants born less than 32-week gestation are at increased risk of respiratory distress syndrome due to underdevelopment of pulmonary alveoli. Providing these infants with positive end expiratory pressure (PEEP) soon after birth maintains functional residual capacity (FRC) and prevents further alveolar collapse, which may lead to worsening respiratory distress.^1,2 Early initiation of continuous positive airway pressure (CPAP) can reduce the need for mechanical ventilation with a trend toward reduction in rates of bronchopulmonary dysplasia (BPD).^2–4

We identified inconsistent use of PEEP in infants born less than 32 weeks, as well as variability in respiratory support practices amongst neonatal providers secondary to different levels of training and experience, varying comfort levels with newer methods of ventilation, and slower adoption to change despite new evidence from ongoing studies. For instance, our staff expressed concern that prolonged exposure to PEEP would contribute to increased risk of pneumothorax as well as delayed initiation of oral feeding which may in turn result in longer length of stay. Practice variability not only affects patient care, resulting in pulmonary co-morbidities, but it can also have negative effects on the hospital system, such as increased length of stay, cost, and resource utilization.^5–7 Furthermore, in an academic setting, variability in practice can lead to an inconsistent learning environment for trainees.⁸ We aimed to improve the equity of care by implementing a standardized guideline, using a patient centered approach targeting eligible infants requiring respiratory support in the delivery room (DR). Using the current evidence from studies and knowledge of preterm lung physiology, we proposed a guideline to standardize the use of PEEP in infants born less than 32-week gestation starting in the delivery room and continuing through 32-week post-menstrual age (PMA). Our primary target was to maintain compliance with continuation of PEEP in clinically appropriate infants through 32-week PMA. Our secondary targets were to improve consistency, reduce practice variation, and monitor for unintended consequences such as pneumothorax or delayed oral feeding.

Methods

Setting and patient population

Our 52-bed, Level IV NICU serves a high-risk birthing center and provides advanced subspecialty and surgical services to the neonatal population. The NICU admits more than 800 patients annually, with approximately 100 inborn patients less than 32-week gestation per year. Patients are cared for by attending physicians, pediatric residents, neonatal-perinatal fellows, advanced practice providers (APPs), respiratory therapists, and neonatal nurses.

Data collection

Our data collection included all inborn infants born less than 32-week gestation. The data points collected to assess our primary outcome include birthweight, gestational age, surfactant administration, and respiratory support in DR through initial discontinuation of PEEP. We also collected data for our unintended consequences which includes the presence of pneumothorax, PMA at initiation of oral feeding and length of stay. In addition, we collected data on respiratory support at 32 weeks, date of initial room air trial, whether subsequent escalation of respiratory support was required, and the presence of BPD (defined as respiratory support at 36-weeks PMA).

Pre-intervention data was retrospectively reviewed for all inborn infants less than 32-week gestation between July 2019 and October 2020 and then prospectively collected between November 2020 and July 2021. There were 255 inborn infants less than 32-week gestation during this timeframe; 11 were excluded due to previable gestation (born less than 22 weeks) or expiry in the DR. Post-intervention data was prospectively collected for all inborn infants less than 32 weeks between July 2021 and May 2023. During this timeframe, there were 245 inborn infants, of which 2 expired in the DR and 1 was excluded for a congenital anomaly which precluded the ability to initiate non-invasive respiratory support.

Analysis

We analyzed monthly compliance on a run chart to assess for shifts or trends using standard run chart interpretation rules to calculate the center line (median).⁹ We compared pre- and post-intervention data including mean birthweight (grams), mean birth gestational age, incidence of pneumothorax and BPD, median gestational age at first oral feed, and median length of stay. We used Student’s t-test for continuous variables and chi-squared test or Fisher’s exact test, as appropriate, for categorical variables. For comparison of the incidence of pneumothorax and the median gestational age at first oral feed, we chose a pre- and post-intervention analysis over traditional run charts due to the low monthly occurrence of events and sample size.

Plan-Do-Study-Act Cycles: We developed a key driver diagram to guide interventions for Plan-Do-Study-Act Cycles (PDSA) (Figure 1). The factors identified influenced subsequent PDSA cycle strategies.

Figure 1.

Key driver diagram.

Cycle 1: Guideline development

Between December 2020 and June 2021, we developed a “Keep the PEEP” guideline, which outlined recommendations for initiation of PEEP at delivery and continuation of PEEP through 32-week PMA in clinically appropriate infants born less than 32-week gestation. The NICU Respiratory Care Collaborative, which consisted of attending neonatologists, fellows and also included representation from respiratory therapy and nursing, led this effort. After an evidence-based review of the literature, guideline content was determined based on attending neonatologist consensus. The target audience of the guideline was primarily neonatal providers including attending neonatologists, fellows, and APPs. The guidelines were reviewed by the unit medical director and clinical faculty before being approved in June 2021.

Cycle 2: Staff education and awareness

In June-July 2021, we introduced the “Keep the PEEP” guideline to relevant staff including nursing, APPs, fellows, attendings, and respiratory therapists. We held multiple educational sessions targeting each role (i.e., nursing, respiratory therapy, and providers) at various times to ensure education of both day and night shift staff. During the sessions, we described the initiative, summarized the evidence, and reviewed the guidelines. In July 2021, signage was posted throughout the unit to generate awareness for the “Keep the PEEP” go-live date of July 19, 2021. Signage included a “Keep the PEEP” logo (Figure 2).

Figure 2.

Keep the PEEP logo.

Cycle 3: Performance awareness

Beginning in September 2021, we provided monthly data on compliance during division QI meetings to NICU staff which included neonatologists, neonatal fellows, advanced practitioners, nurses, and respiratory therapists. This forum is a culturally accepted venue for our NICU frontline staff to raise quality and safety concerns, and to provide feedback on unit-based improvement projects.

Ethical considerations

The University of Maryland Baltimore Institutional Review Board evaluated this quality improvement project and determined that it was not human subjects research.

Results

We identified 486 neonates meeting inclusion criteria, 244 in the pre-intervention group and 242 in the post-intervention group. Figure 3 displays the disposition of neonates in the pre-intervention and the post-intervention groups during the study period.

Figure 3.

(a) Pre-intervention disposition. *Due to post-menstrual age or medical reasons (e.g., respiratory support, inability to protect airway, and surgical patient). (b) Post-intervention disposition. *Due to post-menstrual age or medical reasons (e.g., respiratory support, inability to protect airway, and surgical patient).

Compliance of using PEEP on admission, through the first week of life, and through 32-week PMA, is shown in Figure 4. We demonstrated compliant initiation of PEEP in the delivery room and use of PEEP at NICU admission (Figure 4(a) and 4(b), respectively) with a median compliance of 100%. PEEP use through the first week of life improved from a median of 82%–100% (Figure 4(c)) and PEEP use through 32-week PMA improved from 55% to 100% (Figure 4(d)). Post-implementation, compliance remained 100%.

Figure 4.

(a) Delivery room PEEP. (b) Continuation of PEEP on admission. (c) Continuation of PEEP during first week of life. (d) Continuation of PEEP to 32-week PMA. PEEP, positive end expiratory pressure, PMA, post-menstrual age.

A comparison of patient characteristics and outcomes of interest in the pre-intervention and post-intervention groups is shown in Table 1. There were no statistically significant differences in mean birthweight, mean gestational age, incidence of BPD, pneumothorax, median post-menstrual age at initiation of oral feeds, or median length of stay. Additionally, we found no statistically significant differences among unintended consequences. The incidence of pneumothorax was 14 and 13 in the pre- and post-intervention periods, respectively (p = 0.86); and our neonates initiated oral feeding at a median post-menstrual age of 34 1/7-weeks in both the pre- and post-intervention periods (p = 0.56). We followed the incidence of pneumothorax through 32-week PMA and found no statistically significant difference (p = 0.86).

Table 1.

Comparison of patient characteristics and outcomes of interest in the pre-intervention and post-intervention groups.

Characteristics	Pre-intervention (N)	Post-intervention (N)	p-value
Mean birthweight (grams)	1152 (244)	1120 (242)	0.37
Mean birth GA (weeks)	28.7 (244)	28.5 (242)	0.34
Pneumothorax	14 (244)	13 (242)	0.86
BPD	82 (135)	68 (107)	0.65
Median PMA (weeks) at first oral feed	34.1 (187)	34.1 (165)	0.56
Median LOS (days)	55 (223)	43 (221)	0.2

GA, gestational age, BPD, bronchopulmonary dysplasia, LOS, length of stay, PMA, post-menstrual age.

Discussion

Through a multidisciplinary quality improvement initiative, we successfully standardized the use of PEEP in infants born less than 32-week gestation in our urban Level IV NICU. Our team applied core QI principles, which included establishing an aim statement and key driver diagram, testing changes with PDSA cycles, stakeholder involvement, and measuring data over time. In doing so, we were able to maintain high compliance and improve consistency in practice while avoiding harm. We believe that our comprehensive, interprofessional approach was necessary to reduce variability in practice by promoting a standardized guideline. We successfully maintained a median compliance of 100% for each aim over the 2-year post-implementation period.

Based on the concern that prolonged PEEP exposure may delay initiation of oral feeding, we analyzed the median post-menstrual age at first oral feed in both groups. Oral feeding is typically not initiated before 32-week PMA due to the immature coordination of sucking, swallowing and breathing of neonates at this age. Furthermore, the required rhythmic breathing during feeding is acquired between 34 and 36-weeks PMA.^10,11 Despite the concern from some frontline staff, our neonates initiated oral feeding at a median post-menstrual age of 34 1/7-weeks in both the pre- and post-intervention periods (p = 0.56). Therefore, the lack of statistical significance is reassuring that there was no delay in initiation of oral feeding outside of what is developmentally and physiologically appropriate for gestational age. To the best of our knowledge, this concern has not been previously reported in the literature. Further, a large randomized control trial conducted in Australia “CeasIng Cpap At standard criteriA” (CICADA) who also aimed to standardize CPAP use demonstrated that weaning to room air from CPAP versus gradual weaning to nasal cannula did not adversely affect weight gain or time to reach full enteral feeding.¹² Our guidelines recommend discontinuation of CPAP to room air, which may have prevented a potential tendency to avoid oral feeding trials for infants who may be on higher flow nasal cannula.

Staff expressed an additional concern that early, prolonged, or unnecessary PEEP exposure may increase the incidence of pneumothorax. Previous studies evaluating continuation of CPAP in this patient population have evaluated potential adverse effects including pneumothorax, with only one study showing an increased risk of pneumothorax in the CPAP group.^13–16 Nonetheless, we followed the incidence of pneumothorax through 32-week PMA and found no statistically significant difference (p = 0.86).

Given the evidence to support prolonged PEEP use in this patient population, it is postulated that this may lead to a reduction in rates of BPD.^4,17,18 As one of two level IV NICUs in the state, we frequently transfer stable babies back to their referral hospitals, lower acuity NICUs or pediatric rehabilitation hospitals, thus we were limited in our ability to consistently follow patients through 36 weeks PMA.¹⁹ In the pre-intervention cohort, we were able to follow about 55% of patients through 36-weeks PMA; however, in the post-intervention cohort, we had a higher rate of transfers and consequently followed less than 45% of patients through 36-weeks PMA. As a result, this may have contributed to the lack of observed differences in the rate of BPD in our study.

Prior to the official roll out of the project, we noted improved consistency in monthly compliance as information about the project circulated. We observed rapid acceptance and adoption of the process change from the various stakeholders including nursing, respiratory therapy, APPs, residents, fellows, and attendings. One possible factor that could account for the dampened effect of our intervention was the anticipation of our study and knowledge that our guideline would be evaluated. Providers may have experienced the Hawthorne effect in which knowing their decisions would later be evaluated made them adhere to the guideline prior to its official implementation.^20,21 Yet, the Hawthorne effect has been described as a short-term effect, on the order of weeks to months.²² Consequently, the Hawthorne effect does not completely explain the observed improvement in compliance, as it was maintained over the 2-year post-implementation period.

Another important limitation of this study is the pre-post-intervention analysis, which is fraught with error in quality improvement initiatives, as it is an evolving process.²³ As a result, this limits the ability to adequately assess outcome differences and could potentially explain the lack of observed differences in any of the outcomes of interest. Additionally, this study represents a single institution where we had strong buy-in from stakeholders, which we believe significantly contributed to our high rates of compliance. This stresses the importance of stakeholder considerations when implementing any quality improvement project.²⁴ Furthermore, the single center design limits generalizability of this study. As a result, additional studies, particularly with multicenter involvement, are required to further evaluate these findings.

Conclusion

Formalized implementation of a “Keep the PEEP” guideline improved and maintained high compliance with the use of PEEP in infants born less than 32 weeks in our urban academic Level IV NICU and could be considered safe, without increasing the incidence of pneumothorax, or post-menstrual age at initiation of oral feeding. Our findings help substantiate the literature around the benefits of PEEP and may allay concerns of unintended consequences to support adoption of PEEP in other NICUs.

Footnotes

Authors’ note

Presentations of preliminary data at Hot Topics in Neonatology, Washington, DC, December 4-7, 2022, and Pediatric Academic Societies Meeting, Washington, DC. April 27-May 1, 2023.

Acknowledgments

Assistance with the study: The authors would like to acknowledge Jennifer Dukes, NNP for her original artwork in creating the “Keep the PEEP” logo.

ORCID iDs

E Sangillo

S Patel

K Nyman

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Conflicting interest

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Owen

Manley

Davis

, et al. The evolution of modern respiratory care for preterm infants. Lancet 2017; 389(10079): 1649–1659.

Donn

Attar

. Assisted ventilation of the neonate and its complications. In: Fanaroff and Martin’s Neonatal-Perinatal Medicine: Diseases of the Fetus and Infant. 11th ed. Elsevier, 2020, vol Vol 2, pp. 1174–1199.

Behnke

Lemyre

Czernik

, et al. Non-invasive ventilation in neonatology. Dtsch Arztebl Int 2019; 116(11): 177–183.

Subramaniam

Davis

. Prophylactic or very early initiation of continuous positive airway pressure (CPAP) for preterm infants. Cochrane Database Syst Rev 2021; 2021(10): CD001243.

Youn

Heim

Kumar

, et al. Examining the impacts of clinical practice variation on operational performance. Prod Oper Manag 2021; 30(4): 839–863.

Patole

de Klerk

. Impact of standardised feeding regimens on incidence of neonatal necrotising enterocolitis: a systematic review and meta-analysis of observational studies. Arch Dis Child Fetal Neonatal Ed 2005; 90(2): F147–F151.

Jasani

Patole

. Standardized feeding regimen for reducing necrotizing enterocolitis in preterm infants: an updated systematic review. J Perinatol 2017; 37(7): 827–833.

Shalish

Sant’ Anna

. Respiratory care protocols in neonatal intensive care. In: Aly

Abdel-Hady

(eds). Respiratory Management of Newborns. InTech, 2016.

Perla

Provost

Murray

. The run chart: a simple analytical tool for learning from variation in healthcare processes. BMJ Qual Saf 2011; 20(1): 46–51.

10.

Mathew

. Science of bottle feeding. J Pediatr 1991; 119(4): 511–519.

11.

Mizuno

Ueda

. The maturation and coordination of sucking, swallowing, and respiration in preterm infants. J Pediatr 2003; 142(1): 36–40.

12.

Broom

Ying

Wright

, et al. CeasIng Cpap at standarD criteriA (CICADA): impact on weight gain, time to full feeds and caffeine use. Arch Dis Child Fetal Neonatal Ed 2014; 99(5): F423–F425.

13.

Roberts

Owen

Manley

, et al. Nasal high-flow therapy for primary respiratory support in preterm infants. N Engl J Med 2016; 375(12): 1142–1151.

14.

Murki

Singh

Khant

, et al. High-flow nasal cannula versus nasal continuous positive airway pressure for primary respiratory support in preterm infants with respiratory distress: a randomized controlled trial. Neonatology 2018; 113(3): 235–241.

15.

Manley

Arnolda

GRB

Wright

IMR

, et al. Nasal high-flow therapy for newborn infants in special care nurseries. N Engl J Med 2019; 380(21): 2031–2040.

16.

Abdel-Hady

Shouman

Aly

. Early weaning from CPAP to high flow nasal cannula in preterm infants is associated with prolonged oxygen requirement: a randomized controlled trial. Early Hum Dev 2011; 87(3): 205–208.

17.

Todd

Wright

Broom

, et al. Methods of weaning preterm babies <30 weeks gestation off CPAP: a multicentre randomised controlled trial. Arch Dis Child Fetal Neonatal Ed 2012; 97(4): F236–F240.

18.

Aly

Mohamed

. An experience with a bubble CPAP bundle: is chronic lung disease preventable? Pediatr Res 2020; 88(3): 444–450.

19.

Donohue

Hussey-Gardner

Sulpar

, et al.

Convalescent care of infants in the neonatal intensive care unit in community hospitals: risk or benefit?

Pediatrics 2009; 124(1): 105–111.

20.

Berkhout

Berbra

Favre

, et al. Defining and evaluating the Hawthorne effect in primary care, a systematic review and meta-analysis. Front Med 2022; 9: 1033486.

21.

Haessler

. The Hawthorne effect in measurements of hand hygiene compliance: a definite problem, but also an opportunity. BMJ Qual Saf 2014; 23(12): 965–967.

22.

McCambridge

Witton

Elbourne

. Systematic review of the Hawthorne effect: new concepts are needed to study research participation effects. J Clin Epidemiol 2014; 67(3): 267–277.

23.

Speroff

O’Connor

. Study designs for PDSA quality improvement research. Qual Manag Health Care 2004; 13(1): 17–32.

24.

Kaplan

Brady

Dritz

, et al. The influence of context on quality improvement success in health care: a systematic review of the literature. Milbank Q 2010; 88(4): 500–559.