Objectives. Early nasal continuous positive airway pressure (ENCPAP) has recently emerged in neonatal units as an acceptable alternative to routine intubation and mandatory ventilation. The risks and benefits of ENCPAP have yet to be established. In this study, we aimed to examine variables that influenced the decision to initiate ENCPAP in the delivery room (DR). We also explored potential harmful effects of early intubation and examined whether unsuccessful ENCPAP attempts might subject infants to any unforeseen morbidity.
Methods. All inborn very low birth weight (VLBW) infants admitted to the NICU since the implementation of the ENCPAP policy were included in this retrospective study. Infants were stratified initially into 2 cohorts according to whether they were intubated in the DR or began ENCPAP. Infants were then stratified into 4 groups according to the respiratory management during their first week of life. Infants in group 1 were supported with ENCPAP in the DR and continued to receive continuous positive airway pressure (CPAP) at least for the entire first week. Infants in group 2 began ENCPAP treatment in the DR but required intubation during the first week of life. Infants in group 3 were intubated in the DR but transitioned successfully to CPAP within the first 48 hours and were treated with CPAP for the first week of life or longer. Infants in group 4 were intubated in the DR and treated with intermittent mandatory ventilation for >48 hours. Univariate analyses compared different groups with the Wilcoxon nonparametric test, Kruskal-Wallis test, and analysis of variance. A multivariate regression model adjusted for differences in birth weights (BWs), gestational ages (GAs), race, and Apgar scores between the groups.
Results. A total of 234 VLBW infants (weight of <1500 g) were admitted to the NICU during the period from August 1997 to December 2003. The mean BW was 977.1 ± 305.8 g, and the mean GA was 27.7 ± 2.7 weeks. The overall mortality rate was 11.1%, and the incidence of bronchopulmonary dysplasia among survivors was 17.4%. ENCPAP was implemented successfully in the DR for 151 (64.5%) infants, whereas 83 (35.5%) infants required intubation. Infants who required intubation had significantly lower GAs, BWs, and 1-minute Apgar scores. The use of ENCPAP in the DR increased significantly over time. The chance of successful maintenance with ENCPAP for >48 hours was not demonstrable at <24 weeks of gestation (10% success). Use of ENCPAP improved significantly by 25 weeks of gestation (45% success). Infants in group 1 required a shorter duration of oxygen use than did infants in group 3 (7.9 ± 18.3 vs 39 ± 32.7 days; regression coefficient [b] = 19 ± 5.3). None of the infants in group 1 developed intraventricular hemorrhage of grade III or IV or retinopathy of prematurity of stage 3 or 4. Infants in group 3 did not show improved outcomes, compared with group 1. Compared with group 4, infants in group 2 had a higher incidence of necrotizing enterocolitis (15.6% vs 7.3%; b = 2.5 ± 1.2).
Conclusions. The success of ENCPAP improved with increased GA and with staff experience over time. Infants treated successfully with ENCPAP were unlikely to develop intraventricular hemorrhage of grade III or IV. Infants who experienced ENCPAP failure were at increased risk for the development of necrotizing enterocolitis. Infants who were intubated briefly in the DR were at increased risk for prolonged oxygen requirement. An individualized approach should be considered for respiratory support of VLBW infants.
- early nasal continuous positive airway pressure
- low birth weight infants
- bronchopulmonary dysplasia
- intraventricular hemorrhage
Initial airway management for premature infants has received considerable attention recently. Early nasal continuous positive airway pressure (ENCPAP) therapy, either in conjunction with or with the elimination of intubation, has emerged as a potential alternative.1–5 The risks and benefits of ENCPAP treatment have yet to be established in a randomized trial, although a recent pilot study reported by the National Institute of Child Health and Human Development Neonatal Research Network confirmed its feasibility, even for infants with extremely low birth weights (BWs).6 Other centers reported anecdotal evidence suggesting potential benefits of this respiratory management strategy.1–5 Despite reported improvements in the rates of bronchopulmonary dysplasia (BPD), days of mechanical ventilation, and daily weight gain associated with ENCPAP,3,7 questions linger regarding the universal applicability of this strategy. A better understanding of variables associated with success and failure in implementing this modality of care, both in its early phase in the delivery room (DR) and later in its maintenance phase in the NICU, requires greater analysis.
This report describes our overall experience using ENCPAP as the preferred and exclusive initial airway management modality. We examined variables that influenced the decision to initiate ENCPAP in the DR. We also explored beneficial and harmful effects associated with early intubation. In addition, we investigated whether unsuccessful ENCPAP attempts might subject infants to unforeseen morbidity.
Data for all inborn very low BW (VLBW) infants admitted to the NICU since the implementation of the ENCPAP policy (August 1997) were reviewed retrospectively. Infants were stratified initially into 2 cohorts according to whether they were intubated in the DR or treated with ENCPAP. Infants were stratified subsequently into 4 groups according to the respiratory management in their first week of life. Infants in group 1 were supported with continuous positive airway pressure (CPAP) therapy in the DR and continued to receive CPAP treatment at least during the entire first week of life. Infants in group 2 began CPAP treatment in the DR but required intubation during the first week of life. Infants in group 3 were intubated in the DR but transitioned successfully to CPAP within the first 48 hours and continued to receive CPAP therapy for the first week of life or longer. Infants in group 4 were intubated in the DR and treated with intermittent mandatory ventilation (IMV) for >48 hours (Fig 1). This retrospective study was approved by the institutional review board at George Washington University and was conducted in compliance with Health Insurance Portability and Accountability Act regulations.
Before August 1997, VLBW infants were intubated in the DR and prophylactic surfactant was administered routinely. After August 1997, premature infants with spontaneous respiratory efforts were offered distending pressure through a CPAP device with a 4-limb connector piece (Carden CPAP device; Dupaco, Oceanside, CA) (Fig 2). One limb of the valve was attached to a face mask, the second to gas tubing, and the third to a pressure manometer; the fourth port was left free to allow leakage of any extra flow. The flow of oxygen was adjusted while the mask was tested against the palm of the physician until a pressure reading of 5 cm H2O was achieved. Infants were supported with facial CPAP in the DR and en route to the NICU. To avoid unnecessary exposure to oxygen, blenders were attached to the oxygen source in the DR and in the transport isolette. On arrival at the NICU, infants were immediately treated with water-seal NCPAP with nasal prongs from Hudson Respiratory Care (Temecula, CA).2 Infants were then handled minimally while being observed closely for any of the 4 preset warning criteria for intubation, ie, severe retraction, frequent apnea, increased oxygen requirement (fraction of inspired oxygen of >60%), or intractable metabolic acidosis (base deficit of more than −10 mEq). In the DR, infants who did not breathe spontaneously were supported with bag-mask ventilation for 30 seconds. If they continued to lack spontaneous breathing during reevaluation, then they were intubated immediately and supported with positive pressure ventilation en route to the NICU. Intratracheal surfactant therapy was offered to intubated infants in an individualized rescue approach, depending on clinical and radiographic findings.
Analyses characterized infants who were intubated in the DR and those who began ENCPAP treatment successfully. Subsequently, 2 hypotheses were tested. The first null hypothesis was that infants who underwent brief intubation in the DR and were extubated to CPAP within 48 hours were not at increased risk of complications, compared with infants who began and were maintained with ENCPAP (group 3 = group 1). The second null hypothesis was that infants who required intubation after failed ENCPAP attempts within the first week of life were not at increased risk, compared with infants who underwent intubation in the DR and were maintained with IMV (group 2 = group 4). Univariate analyses compared different groups with the Wilcoxon nonparametric test, Kruskal-Wallis test, and analysis of variance. A multivariate regression model was used to adjust for differences between the groups in BW, gestational age (GA), race, and Apgar scores. We identified infants with BPD as those who required oxygen until 36 weeks' postconceptional age.
A total of 234 VLBW infants (weight of <1500 g) were admitted to NICU during the period from August 1997 to December 2003. The mean BW was 977.1 ± 305.8 g and the mean GA was 27.7 ± 2.7 weeks. The overall mortality rate was 11.1%, and the incidence of BPD among survivors was 17.4%. ENCPAP was implemented successfully in the DR for 151 (64.5%) infants, whereas 83 (35.5%) infants required intubation. The characteristics of the ENCPAP and intubated groups are shown in Table 1. Infants who required intubation had significantly lower GAs, BWs, and 1-minute Apgar scores (P < .001).
Of the 151 infants who began ENCPAP, 117 (77.5%) infants received CPAP treatment throughout the first week of life (group 1), whereas 34 (22.5%) infants required intubation (group 2). Of the 83 infants intubated in the DR, 28 (33.7%) infants were weaned successfully and extubated to CPAP within the first 48 hours of life and remained extubated until at least the end of the first week of life (group 3). The remaining 55 infants (66.3%) required IMV beyond the first 48 hours of life (group 4) (Fig 1). Mean BWs were 1121 ± 273 g for group 1, 744 ± 225 g for group 2, 926 ± 279 g for group 3, and 896 ± 282 g for group 4. Mean GAs were 28.9 ± 2.6 weeks for group 1, 25.8 ± 2.3 weeks for group 2, 27.2 ± 2.2 weeks for group 3, and 26.6 ± 1.8 weeks for group 4. Description of the characteristics of each group and their outcomes are shown in Table 2.
Factors Influencing ENCPAP Use
There was no exclusionary GA limit for initial treatment with ENCPAP in the DR, because infants at the lowest GA (1 infant at 22 weeks and 8 infants at 23 weeks) began ENCPAP in the DR. The chance of successful maintenance on ENCPAP for >48 hours was not demonstrable for <24 weeks' GA (10% success). Use of ENCPAP improved significantly by 25 weeks' GA (45% success) (Fig 3). Compared with infants treated with ENCPAP in the DR, intubated infants had significantly lower GAs, BWs, and 1-minute Apgar scores.
Use of ENCPAP in the DR increased significantly in our unit over time, which could be a reflection of the level of comfort of the health care team with this new approach to management. Such transition in the DR practice was associated with a reduction in the incidence of BPD from an initial 33% to the current 6% (Fig 4).
Clinical Outcomes Influenced by Initial DR Management (Hypothesis 1)
The comparison between group 1 (never intubated) and group 3 (initially intubated but weaned to CPAP within 48 hours) demonstrated a higher incidence of BPD, longer duration of oxygen use, and longer hospital stay in group 3. In a multivariate regression analysis to control for differences in GAs, BWs, Apgar scores, and race, differences between groups remained significant only for duration of oxygen use (regression coefficient [b] = 19 ± 5.3; P < .001) (Table 3). None of the infants in group 1 developed intraventricular hemorrhage (IVH) of grade III or IV or retinopathy of prematurity (ROP) of stage 3 or 4. Infants in group 3 did not demonstrate any particularly improved outcomes, compared with group 1.
Clinical Outcomes Associated With Unsuccessful ENCPAP Attempts (Hypothesis 2)
In a comparison between group 2 (experienced ENCPAP failure within 1 week) and group 4 (intubated in the DR and treated with IMV for at least 1 week), group 2 had a higher survival rate than did group 4 (94% and 67%, respectively; P = .003). In a multivariate regression model controlling for GAs, BWs, Apgar scores, and race, the difference in survival rates lost significance. In the same model, the incidence of necrotizing enterocolitis (NEC) was significantly higher in group 2 (b = 2.5 ± 1.2; P = .039) (Table 4).
We demonstrated previously the feasibility of implementing ENCPAP in the DR, even for the lowest BW categories, and maintaining it as the preferred form of respiratory support during the early course in the NICU.3 In this study, we examined factors influencing the DR decision to initiate CPAP therapy and factors restricting the maintenance of this form of treatment during the hospital course. Infants supported with ENCPAP had significantly greater GAs. Low GA, however, did not restrict the decision for ENCPAP in the DR, because ENCPAP was attempted even for the lowest GA group. The rate of ENCPAP failure was related inversely to GA. This result should not be surprising, because the immaturity of the respiratory center among neonates of lower GA could have influenced their ability to initiate spontaneous breathing and their ability to maintain respiratory drive over time. Furthermore, high compliance of the chest wall and weakness of the respiratory muscles might have contributed to respiratory failure with NCPAP pressures of 5 cm H2O. It is not clear whether the designated pressure of 5 cm H2O is the maximum allowable in such cases. It would be useful to test the efficacy of higher pressures among these infants, to sustain better chest inflation and airway stenting. Anatomic limitations among these extremely immature infants, most significantly the size of the nasal passage, may present obstacles to ENCPAP treatment and the transmission of delivered pressure to the lower airway passages.
Debate in current clinical practice centers on the need for early intubation for prophylactic surfactant administration for VLBW neonates.8,9 Our practice of preferred ENCPAP precludes us from intubating these infants routinely, leaving us with only a small population (29%) of infants who ever received surfactant, albeit in a rescue mode. It is unclear what added benefit could be obtained if surfactant was administered more routinely.10 None of the earlier randomized surfactant trials used a control group with ENCPAP treatment; therefore, a true risk-benefit analysis remains to be performed. In addition to our report (Fig 4), earlier reports linked avoidance of intubation with the use of ENCPAP in the DR with lower incidences of BPD.1–5 It is plausible to speculate that VLBW infants deprived of administered surfactant because of ENCPAP can maintain their alveoli open and are capable of producing their own surfactant subsequently. This physiologically based theory was proven in a recent animal study.11
In our group of infants, survival rates were lowest among infants intubated initially and treated with IMV for ≥1 week. This group (group 4) had a 67% survival rate and a 65% rate of surfactant administration. Infants intubated initially for <48 hours of life (group 3) had better survival rates (93%) and lower utilization of surfactant (54%). In fact, the latter group had a survival rate similar to that of infants who experienced ENCPAP failure (94% in group 2, with surfactant being administered in 38% of the cases). Survival rates in these 3 groups seemed to be influenced most by the severity of respiratory illness, rather than early intubation and surfactant administration.
In our study population, higher grades of IVH and advanced stages of ROP were limited to the group of infants requiring intubation. Infants who were not exposed to tracheal intubation in the DR or early in the NICU course avoided these 2 complications. This protective effect could be a direct result of their greater GA. However, intubation itself is associated with fluctuations in hemodynamic factors and oxygen saturation,12 and positive pressure ventilation increases intrathoracic pressures and thus impedes venous return among intubated infants.13 It has been suggested that arterial carbon dioxide pressure among spontaneously breathing infants with extremely low BWs might play a role in controlling cerebral perfusion and thus decreasing the risk of IVH.14 The carbon dioxide pressures among infants supported with IMV are not self-regulated but are influenced through manipulation of the respirator settings by the managing physician, which can interfere with this mechanism. Because this was not a prospective trial, we can conclude only that ENCPAP does not constitute a risk for the development of either IVH or ROP.
Infants who were intubated briefly in the DR were more likely to require oxygen supplementation for longer times than were those who were not intubated, controlling for GA, BW, gender, and Apgar score differences. This is an expected finding. Literature reports emphasized the significance of early barotrauma and the damaging effects of early exposure to high partial pressures of ambient oxygen.15–17 In addition, tracheal intubation can impose an increased risk of bacterial colonization that contributes to the already existing inflammation of the alveoli and lower airways.18
It cannot be assumed that ENCPAP application for these VLBW infants is without hazard.19 Infants who experienced failure of ENCPAP treatment had a significantly increased risk of NEC. Several factors might have contributed to this observation. Severe or repeated apneic episodes with desaturation (which are frequently observed with failure of the ENCPAP system) might have compromised perfusion of the bowel and increased susceptibility to NEC among these infants. This risk might have been compounded by the fact that infants who begin ENCPAP therapy are fed earlier and advance more regularly than do ventilation-treated infants. It may be advisable to stop feeding infants who fail ENCPAP until they are stabilized completely on the ventilator and to proceed cautiously with feeding thereafter. It is also important to note that the higher mortality rate among intubated infants (33% in group 4) could have contributed to a deceptively lower incidence of NEC in the surviving group.
Staff experience over time and GA determine the success of ENCPAP. Infants treated successfully with ENCPAP are unlikely to develop IVH of grade III or IV. Infants who experience ENCPAP failure are at increased risk for developing NEC. VLBW infants who are briefly intubated in the DR are at increased risk for prolonged oxygen requirement. These findings emphasize the significant impact of early manipulation of neonatal respiratory physiologic processes through IMV for even short periods. Therefore, this study urges investigators of a planned randomized trial to include a never-intubated control group of infants who are intended to be treated exclusively with ENCPAP in the DR. Until then, an individualized approach should be considered for respiratory support of these vulnerable infants.
We thank Cynthia Poindexter for editorial help and Mary Rivas for compiling the data set.
- Accepted January 20, 2005.
- Reprint requests to (H.A.) George Washington University Hospital, 900 23rd St NW, Suite G-2092, Washington, DC 20037. E-mail:
No conflict of interest declared.
- ↵Lindner W, Vossbeck S, Hummler H, Pohlandt F. Delivery room management of extremely low birth weight infants: spontaneous breathing or intubation? Pediatrics.1999;103 :961– 967
- ↵Aly H. Nasal prongs continuous positive airway pressure: a simple yet powerful tool. Pediatrics.2001;108 :759– 761
- ↵Aly H, Milner JD, Patel K, El-Mohandes AE. Does the experience with the use of nasal continuous positive airway pressure (NCPAP) improve over time in extremely low birth weight (ELBW) infants? Pediatrics.2004;114 :697– 702
- Van Marter LJ, Allred EN, Pagano M, et al. Do clinical markers of barotrauma and oxygen toxicity explain interhospital variation in rates of chronic lung disease? The Neonatology Committee for the Developmental Network. Pediatrics.2000;105 :1194– 1201
- ↵Avery ME, Tooley WH, Keller JB, et al. Is chronic lung disease in low birth weight infants preventable? A survey of eight centers. Pediatrics.1987;79 :26– 30
- ↵Finer NN, Carlo WA, Duara S, et al. Delivery room continuous positive airway pressure/positive end-expiratory pressure in extremely low birth weight infants: a feasibility trial. Pediatrics.2004;114 :651– 657
- ↵Thomson MA. Early nasal CPAP plus prophylactic surfactant for neonates at risk of RDS: the IFADS trial [abstract]. Pediatr Res.2001;50 :304A
- ↵Morley CJ, Davis P, Doyle L. Continuous positive airway pressure: randomized controlled trial in Australia [letter]. Pediatrics.2001;108 :1383
- ↵Bhutada A, Sahni R, Rastogi S, Wung JT. Randomized controlled trial of thiopental for intubation in neonates. Arch Dis Child Fetal Neonatal Ed.2000;82 :F34– F37
- ↵Shekerdemian L, Bohn D. Cardiovascular effects of mechanical ventilation. Arch Dis Child.1999;80 :475– 480
- ↵Nguyen AT, Aly H, Milner J, Patel KM, El-Mohandes A. Partial pressure of carbon dioxide in extremely low birth weight infants supported by nasal prongs continuous positive airway pressure. Pediatrics.2003;112 (3). Available at: www.pediatrics.org/cgi/content/full/112/3/e208
- Tsuno K, Prato P, Kolobow T. Acute lung injury from mechanical ventilation at moderately high airway pressures. J Appl Physiol.1990;69 :956– 961
- ↵Vento M, Asensi M, Sastre J, Garcia-Sala F, Pallardo FV, Vina J. Resuscitation with room air instead of 100% oxygen prevents oxidative stress in moderately asphyxiated term neonates. Pediatrics.2001;107 :642– 647
- ↵Garland JS, Nelson DB, Rice T, Neu J. Increased risk of gastrointestinal perforations in neonates mechanically ventilated with either face mask or nasal prongs. Pediatrics.1985;76 :406– 410
- Copyright © 2005 by the American Academy of Pediatrics