Early Bubble CPAP and Outcomes in ELBW Preterm Infants

Original ...... Article Early Bubble CPAP and Outcomes in ELBW Preterm Infants

Vivek Narendran, MD, MRCP (UK) study first identified the Columbia approach of using ‘‘bubble Edward F. Donovan, MD continuous positive airway pressure (CPAP)’’ in the delivery room 1 Steven B. Hoath, MD as a possible strategy to reduce the incidence of BPD. A recent Henry T. Akinbi, MD comparison of outcomes between an NICU in Boston and the Columbia NICU, reinforced the apparent benefits of the Columbia Jean J. Steichen, MD 2 Alan H. Jobe, MD, PhD approach. Chronic lung disease, which occurs primarily in infants less than 1000 g (the ‘‘new BPD’’), is thought to have a qualitatively different pathogenesis than the traditional BPD described by 3 OBJECTIVE: Northway et al. Traditional BPD has a pathogenesis dominated by To test whether the introduction of early bubble continuous positive airway hyperoxic baro/volutrauma in surfactant deficient lungs, while the pressure (CPAP) results in improved respiratory outcomes in extremely low new BPD has a pathogenesis dominated by immaturity and birth-weight infants. alveolar hypoplasia. Recent evidence attributes these changes to prenatal (chorioamnionitis) and postnatal proinflammatory STUDY DESIGN: mediators (ventilator-induced lung injury), which are known to Outcomes of all infants between 401 and 1000 g born in a level 3 influence both lung development and lung injury.4 neonatal intensive care units (NICU) between July 2000 and October 2001 Several recent epidemiologic studies have demonstrated a (period 2) were compared using historical controls (period 1). Early decreased incidence of BPD by avoiding intubation in the delivery bubble (CPAP) was prospectively introduced in the NICU during period 1. room and using early (CPAP).5,6 There is little experimental Univariate and adjusted comparisons were made across time periods. information as to the mechanism whereby CPAP might minimize RESULTS: lung injury in a preterm lung. There is one study demonstrating Delivery room intubations, days on mechanical ventilation and use of decreased indicators of acute lung injury using bubble CPAP in a 7 postnatal steroids decreased ( p<0.001) in period 2, while mean days on preterm lamb model. Furthermore, there is a paucity of data CPAP, number of babies on CPAP at 24 hours (p<0.001) and mean regarding differences in gas exchange effectiveness between weight at 36 weeks corrected gestation also increased (p<0.05) after conventional and bubble CPAP systems. Bubble CPAP is a form of introduction of early bubble CPAP. oscillatory pressure delivery in which mechanical vibrations are transmitted into the chest secondary to the nonuniform flow of gas CONCLUSIONS: bubbles across a downstream water seal. This system results in Early bubble CPAP reduced delivery room intubations, days on waveforms similar to those produced by high-frequency ventilation mechanical ventilation, postnatal steroid use and was associated with when recorded by a transducer attached to the infant’s airway.8 increased postnatal weight gain with no increased complications. At present, there are no definitive studies and no clinical Journal of Perinatology (2003) 23, 195–199. doi:10.1038/sj.jp.7210904 consensus about the best approach to minimize ventilator- associated lung injury in extremely low birth-weight (ELBW) infants. Strategies vary from early delivery room surfactant administration and intermittent mandatory ventilation to early INTRODUCTION delivery room CPAP with or without pretreatment with surfactant.9 In 1987, Avery et al.1 reported large site differences in the risk- The primary objective of this study was to test whether the adjusted incidence of bronchopulmonary dysplasia (BPD) in a introduction of early bubble CPAP in the delivery room was safe group of 12 academic neonatal intensive care units (NICUs). This and resulted in improved respiratory outcomes in ELBW. We defined safety as improved respiratory outcomes without an increase in mortality and nonpulmonary morbidities such as

Division of Neonatology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA. IVH and NEC. These outcomes along with growth parameters were monitored on a regular basis. Risk-adjusted comparisons Address correspondence and reprint requests to Vivek Narendran, MD, MRCP (UK), Division of Neonatolgy, 231 Bethesda Avenue, Medical Sciences Building, ML 6158, Cincinnati, of care and outcomes of infants during the two time periods OH 45267-0541, USA. were made.

METHODS In period 2, the neonatologists prospectively introduced routine, The study population comprised all infants with birth weights early, bubble nasal CPAP in the delivery room modeling closely the between 401 and 1000 g, born at a neonatal tertiary care unit Columbia method. Hudson nasal prongs were used to initiate in Cincinnati, OH, between 1 July 2000 and 10th October 2001 bubble CPAP immediately after the initial steps of drying and (period 2). These infants were compared to historical controls stimulation in all spontaneously breathing infants (Figure 1). The (period 1) born between 1 January 1998 and 31st December 1999. bubble CPAP circuit was very similar to conventional CPAP except The same group of neonatologists provided care in both periods. that the expiratory seal is an ‘underwater seal’ in the former The neonatal unit was part of the NICHD Neonatal Research compared to a mechanical device in the latter (Figure 2). The Network, and had extensive data collected prospectively throughout depth of immersion (5 cm) of the expiratory limb in 25% acetic the two time periods. acid determined the level of CPAP provided (5 cm of water). This was maintained constant during the study. The NICHD Neonatal Research Network is a consortium of s tertiary neonatal centers.10 The network maintains a registry of all Infants were electively intubated and given Survanta (Ross very low birth-weight (VLBW) infants (401–1500 g) born and/or Abbott Laboratories, OH), when FiO2 was greater than 60% to admitted to participating centers within 14 days of birth. This maintain saturations between 92 and 96%, pH <7.15, or PCO2 registry was developed to describe the populations at each >65. FiO2 was commonly adjusted using pulse oximetry to participating center, to survey neonatal practice, to assess morbidity maintain saturations between 92 and 96% and indwelling arterial and mortality rates, and to provide information for the planning of catheters were used sparingly. Ventilated infants were weaned randomized clinical trials. Trained research nurses collect aggressively and extubated back to CPAP as early as possible. maternal demographic, pregnancy and delivery data soon after Infants were monitored for neonatal morbidities for a period of 120 birth and infant data until 120 days, discharge, or death. days unless discharged or transferred to another hospital. In period 1, infants were managed with intermittent positive Statistical Analysis pressure ventilation with bag and mask in the delivery room, early Data were retrospectively analyzed by w2-test and continuous t-test intubation, surfactant administration and intermittent mandatory for comparisons between the two time periods and expressed as ventilation for infants with moderate-to-severe respiratory distress proportions. Results were considered to be statistically signiﬁcant syndrome. CPAP was not used in the delivery room and CPAP was for p<0.05. rarely introduced in the early management of RDS compared to intubation and surfactant administration. FiO2 was adjusted based on transcutaneous PaO2 or arterial PaO2 from an indwelling arterial catheter. No new major technologies or clinical practice RESULTS guidelines were introduced in the intensive care units during the Demographic characteristics did not differ during the two study two study periods. periods (Table 1). When compared to period 1, after initiation of

Figure 1. Premature infant in incubator on bubble CPAP.

196 Journal of Perinatology 2003; 23:195–199 Bubble CPAP in the Delivery Boom Narendran et al.

Figure 2. Schematic diagram of bubble CPAP circuit. early bubble CPAP, delivery room intubations, days on mechanical Table 1 Demographics and Non Pulmonary outcomes ventilation and postnatal steroid use decreased (Table 2). Infants Before CPAP After CPAP on CPAP at 24 hours and number of days on CPAP also increased (Table 2). Mean weight at 36 weeks corrected gestational Number of admissions 92 79 age increased (Table 1). Chronic lung disease defined as Mean gestational age (weeks) 26 26 oxygen requirement at 36 weeks corrected gestational age Mean birth weight (g) 763 753 tended to decrease but was not statistically significant. Mean Female (%) 46 54 White (%) 54 58 length and head circumference at 36 weeks corrected gestational Mortality (%) 38 34 age tended to increase. Other nonpulmonary morbidities Age at full enteral feeds (days) 24 18 tended to decrease except PDA (Table 1). There were three ICH (grades 3 & 4) (%) 9.8 6.3 infants with septal erosions, which resolved spontaneously and PDA (%) 33.7 44.3 required no surgical intervention. No infant in period NEC (%) 12 8.9 2 developed a pneumothorax. Mean weight at 36 weeks (g) 1917 2134** Mean length at 36 weeks (cm) 41.5 42.2 Mean head circumference at 36 weeks (cm) 31 31.6 DISCUSSION **P<0.05 univariate analysis for comparisons over two time periods. CPAP has been used primarily to treat surfactant deficiency in preterm infants for many years.11 Particular interest in CPAP focuses on its potential role to reduce ventilator-induced lung The controversy of early vs delayed CPAP continues as many of injury and BPD. The mechanisms responsible for the possible the trials favoring early CPAP were carried out in the larger infants effects of CPAP to decrease BPD have not yet been evaluated. One prior to routine use of antenatal steroids and postnatal surfactant.17 postulated mechanism is the avoidance of aggressive initiation of The evidence suggesting decreased mechanical ventilation and intermittent positive pressure ventilation with high tidal volumes increased exogenous surfactant use with early CPAP and brief and inadvertent hyperventilation/under ventilation that occurs in ventilation in ELBW is from one small randomized clinical trial.18 ventilated infants.12 CPAP also protects the airway from mechanical There are no studies as yet comparing the role of early surfactant injury and bacterial colonization related to the endotracheal tube. administration with brief ventilation and extubation to nasal CPAP CPAP putatively increases both functional residual capacity and vs selective surfactant use with continued mechanical ventilation endogenous respiratory drive leading to decreased delivery room for ELBW preterm infants at risk for RDS. Additional randomized intubations, reintubations and days on mechanical ventilation.12,13 trials are needed and are underway.19 CPAP has been associated with decreased BPD in several clinical There are practical concerns regarding the implementation of reports.14–16 Multivariate regression analysis has shown intubation early CPAP in the delivery room. Preterm infants allowed to breathe and mechanical ventilation of preterm infants to be the single most spontaneously with nasal prongs will not ‘pink up’ as rapidly and 2 important predictor of subsequent BPD. PCO2 values will be higher than tolerated by most clinicians

Journal of Perinatology 2003; 23:195–199 197 Narendran et al. Bubble CPAP in the Delivery Boom

Table 2 Pulmonary Outcomes of respiratory support at 36 weeks gestation there was no difference between the groups. Nonrespiratory outcomes such as NEC, time to Before CPAP After CPAP full feeds, and the average length of stay tended to decrease, Intubated in delivery room (%) 59.8 31.6* without any additional morbidity. Received surfactant (%) 69.6 58.2 The role of bubble CPAP in decreasing postnatal steroid use is Diagnosis of RDS (%) 93.5 88.6 unclear. Whether this represents a true causal effect or an associa- CLD (O2 at 36 weeks) (%) 40.2 34.2 tion is unclear. Emerging literature during the time frame of the Received IMV (%) 82.6 72.2 study increasingly linked postnatal steroids with poor neurodevelop- Mean days on IMV 28 13* Mean days on CPAP 4 16* mental outcome, which could have changed clinical practice. Opponents of bubble CPAP suggest that infants may not gain Mean days on O2 53 46 Antenatal steroids (%) 82.6 81 weight adequately in a relatively ‘hypercarbic’ environment. In Postnatal steroids (%) 42.4 13.9* addition, nonrespiratory outcomes such as weight, height and head CPAP at 24 hours (%) 10.9 46.8* circumference at 36 weeks corrected gestation have not been Death or CLD (%) 78.3 68.3 previously reported for infants managed with bubble CPAP. In our *p<0.05 univariate analysis for comparisons over two time periods. study, infants exhibited significantly increased weight gain following institution of early bubble CPAP. In summary, we found that early bubble CPAP when begun in the delivery room was safe, inexpensive and an effective way to previously. The decision point of when to intervene is imprecise avoid intubations in the delivery room. Randomized control trials and depends on experience and constant vigilance. Secondly, the are needed to definitely demonstrate beneficial effects on BPD and safety of permissive hypercapnia has to be accepted.20 Although neurodevelopmental outcomes. 21 high PCO2 levels may decrease lung injury, safe upper values for PCO2 have not been determined for preterm infants. Long-term follow-up of infants treated with permissive hypercapnia are now 22 References being reported and appear to be safe. Thirdly, the prolonged use 1. Avery ME, Tooley WH, Keller JB, Hurd SS, Bryan MH, Cotton RB. Is chronic of CPAP delivered by nasal prongs can lead to nasal septal erosions lung disease in low birth weight infants preventable? A survey of eight and abnormal head molding that can complicate clinical centers. Pediatrics 1987;79:26–30. management. 2. Van Marter LJ, Allred EN, Pagano M, et al. Do clinical markers of In this study, we evaluated early delivery room CPAP on babies barotraumas and oxygen toxicity explain interhospital variation in rates of at highest risk for developing BPD; that is, infants with birth weight chronic lung disease? Pediatrics 2000;105:1194–201. less than 1000 g. Two time periods were compared where the only 3. Northway WH, Rosan RC, Porter DY. Pulmonary disease following respirator change in the later period was the introduction of early bubble therapy of hyaline-membrane disease. Bronchopulmonary dysplasia. N Engl CPAP. We compared respiratory outcomes using bubble CPAP to J Med 1967;276:357–68. test if results from Columbia were reproducible and if their success 4. Jobe AH, Ikegami M. Prevention of bronchopulmonary dysplasia. Curr opin Pediatr 2001;13:124–9. in reducing the incidence of BPD was related to their use of ‘bubble 5. Lindner W, Vossbeck S, Hummler H, Pohlandt F. Delivery room CPAP’. We used a similar system as used by Dr. Jen-Tien Wung at 23 management of extremely low birth weight infants: spontaneous breathing Columbia. or intubation? Pediatrics 1999;103:961–7. The clinical population studied was demographically stable and 6. Gitterman MK, Fusch C, Gitterman AR, Regazzoni BM, Moessinger AC. well defined. A single group of neonatologists cared for these Early nasal continuous positive airway pressure treatment reduces the need infants during both periods. Our data were collected prospectively for intubation in very low birth weight infants. Eur J Pediatr 1997;156:384–8. as part of the NIH Neonatal Network to monitor safety. Drawbacks 7. Jobe AH, Kramer BW, Moss TJ, Newnham JP, Ikegami M. Decreased of the study were that it was not a randomized controlled trial and indicators of lung injury with continuous positive expiratory pressure in used historical controls. preterm lambs. Pediatr Res 2002;52:387–92. In this study, the use of early bubble CPAP significantly 8. Kyong-Soon Lee, Dunn MS, Fenwick M, Shennan AT. A comparison of decreased delivery room intubations, use of postnatal steroids and underwater bubble CPAP with ventilator derived CPAP in premature days on mechanical ventilation. Similar results were reported in neonates ready for extubation. Biol Neonate 1998;73:69–75. 9. Stevens TP, Blennow M, Soll RF. Early Surfactant Adminstration with Brief larger infants (1000–1499 g) from the group in South Auckland, 24 Ventilation vs Selective Surfactant and Continued Mechanical Ventilation New Zealand after starting nasal CPAP closely modeled after the for Preterm Infants with or at Risk for RDS. The Cochrane Library, Issue 2. Columbia method. When compared to historical controls, the Oxford: Update Software Ltd., 2002. number of infants ventilated (65 vs 14%), median days of 10. Wright LL, McNellis D. National Institute of Child Health and Human ventilation (6 vs 2 days) and chronic lung disease at 28 days of life Development sponsored perinatal research networks. Semin Perinatol (11 vs 0%) decreased. However, when CLD was defined as any form 1995;19( 2 ):112–23.

198 Journal of Perinatology 2003; 23:195–199 Bubble CPAP in the Delivery Boom Narendran et al.

11. Gregory GA, Kitterman JA, Phibbs RH, Tooley WA, Hamilton WK. Treatment 18. Verder H, Albertsen P, Ebbesen F, et al. Nasal continuous positive airway of the idiopathic respiratory distress system with continuous positive airway pressure and early surfactant therapy for respiratory distress syndrome in pressure. N Engl J Med 1971;284:1333–40 newborns of less than 30 weeks gestation. Pediatrics 1999;103( 2 ):E25. 12. Ho JJ, Subramaniam P, Henderson-Smart DJ, Davis PG. Continuous 19. Merran AT (IFDAS study group). Early nasal continuous positive airway Distending Pressure for Respiratory Distress Syndrome in Preterm Infants. pressure with prophylactic surfactant for neonates at risk of RDS. The IFDAS The Cochrane Library, Issue 2. Oxford: Update Software Ltd., 2002. multi-center randomized trial. Pediatr Res 2002;51( 4 ):379A. 13. Davis PG, Henderson-Smart DJ. Nasal continous positive airway pressure 20. Mariani G, Cifuentes J, Carlo WA. Randomised trial of permissive immediately after extubation for preventing morbidity in preterm infants. hypercapnia in preterm infants. Pediatrics 1999;104:1082–8. The Cochrane Library, Issue 2. Oxford: Update Software Ltd., 2002. 21. Shibata K, Cregg N, Engelberts D, Takeuchi A, Fedorko L, Kavanagh BP. 14. Poets CF, Sens B. Changes in intubation rates and outcome of very low birth Hypercapnic acidosis may attenuate acute lung injury by inhibition of weight infants: a population based study. Pediatrics 1996;98:24–7 endogenous xanthine oxidase. Am J Respir Crit Care Med 1998;158: 15. Kamper J, Wulff K, Larsen C, Lindequist S. Early treatment with nasal 1578–84. continuous positive airway pressure in very low-birth weight infants. Acta 22. Sanocka U, Durkin M, Chiroboga C, Kuban K, Hinton V, VanMarter L, Paediatr Scand 1993;82:193–7. Bellinger D. Variations in respiratory management of VLBW infants: Effects 16. Lundstrom KE, Griesen G. Early treatment with nasal-CPAP. Acta Paediatr on long term outcomes. Pediatr Res 2002;51( 4 ):289A. Scand 1993;82:856. 23. Wung JT. Respiratory management for low birth weight infants. Crit Care 17. Ho JJ, Henderson-Smart DJ, Davis PG. Early vs Delayed Initiation of Med 1993;21(9 Suppl):S364–5. Continuous Distending Pressure for Respiratory Distress Syndrome in Preterm 24. Klerk AD, Klerk RD. Nasal continuous positive airway pressure and Infants. The Cochrane Library, Issue 2. Oxford: Update Software Ltd., 2002. outcomes of preterm infants. J Paediatr Child Health 2001;37:161–7.

Journal of Perinatology 2003; 23:195–199 199