ARTICLE |
Division of Neonatology, Department of Pediatrics, Leiden University Medical Center, Leiden, Netherlands
| ABSTRACT |
|---|
|
|
|---|
METHODS. Two hundred seven very preterm infants were assigned randomly in the delivery room to either a sustained inflation through a nasopharyngeal tube followed by early nasal continuous positive airway pressure (early functional residual capacity intervention) or repeated manual inflations with a self-inflating bag and mask followed by nasal continuous positive airway pressure, if necessary, after arrival at the NICU. The primary outcome measure was intubation <72 hours of age and bronchopulmonary dysplasia at 36 weeks was used as secondary outcome. This trial was registered as an early functional residual capacity intervention trial (ISRCTN 12757724).
RESULTS. In the early functional residual capacity intervention group, fewer infants were intubated at <72 hours of age or received >1 dose of surfactant, and the average duration of ventilatory support was less. Infants in the early functional residual capacity intervention group developed bronchopulmonary dysplasia less frequently.
CONCLUSIONS. A sustained inflation followed by early nasal continuous positive airway pressure, delivered through a nasopharyngeal tube, is a more efficient strategy than repeated manual inflations with a self-inflating bag and mask followed by nasal continuous positive airway pressure on admission to the NICU.
Key Words: early nasal continuous positive airway pressure resuscitation preterm infants respiratory distress syndrome bronchopulmonary dysplasia
Abbreviations: BPD—bronchopulmonary dysplasia NCPAP—nasal continuous positive airway pressure FIO2—fraction of inspired oxygen ENCPAP—early nasal continuous positive airway pressure RDS—respiratory distress syndrome EFURCI—early functional respiratory capacity intervention PIP—peak inspiratory pressure PEEP—positive end-expiratory pressure IVH—intraventricular hemorrhage IQR—interquartile range OR—odds ratio CI—confidence interval
The pathogenesis of bronchopulmonary dysplasia (BPD) or chronic lung disease in very preterm infants is multifactorial, but ventilator-induced lung injury plays a major contributing role.1 Various new ventilation strategies have been introduced, but this has not reduced the incidence of BPD.2 Early respiratory management, that is, ventilatory support from birth during the first days of life, may influence pulmonary outcome, but, because of lack of data, there is no consensus on the early ventilatory management of preterm infants.3 Retrospective cohort and experimental studies suggest that the initial ventilation strategy may play an important role in the development of BPD.4–9 The most effective and least injurious way to recruit the lung in very premature neonates at birth may be a combination of a sustained inflation and early nasal continuous positive airway pressure (ENCPAP). This attempt to avoid intubation and mechanical ventilation may reduce lung injury and BPD in preterm infants as suggested in a retrospective study by Lindner et al10 A trial of ENCPAP at birth seems justified in infants at risk of respiratory distress syndrome (RDS), providing early surfactant rescue is given if required.11,12 We performed a randomized, controlled trial and compared the traditional ventilatory approach with a new method that combined a number of techniques, which theoretically could improve the respiratory outcome of preterm infants. We hypothesized that a sustained inflation followed by ENCPAP, using a pressure-limited mechanical device, is a more effective and less injurious management strategy in preterm infants than conventional intervention.
| PATIENTS AND METHODS |
|---|
|
|
|---|
EFURCI Approach
After oropharyngeal and nasal suctioning (time = 0–30 seconds), and if breathing was insufficient (ie, no signs of spontaneous breathing or spontaneous breathing present, but signs of poor air entry [severe retractions, nasal flaring]), a pressure-controlled (20 cm H2O) inflation was sustained for 10 seconds, using a nasopharyngeal tube and a T-piece ventilator (Neopuff Infant Resuscitator; Fisher and Paykel, Auckland, New Zealand) (time = 30–45 seconds). This T-piece ventilator is a pressure-limited mechanical device that supplies a consistent peak inspiratory pressure (PIP) and positive end-expiratory pressure (PEEP) and is capable of delivering a sustained inflation.13–16 Use of a sustained inflation reduces the need for higher initial airway pressures. To avoid PEEP leakage, a nasopharyngeal tube was used as interface.17 Nasopharyngeal tubes with a diameter of 2.5 to 4.0 mm were used, according to gestational age/birth weight. The length of the tube was cut down to 6 cm. The mouth and other nostril were held closed manually during the inflation. This procedure was repeated (time = 50–65 seconds) with an increased pressure (25 cm H2O) if breathing remained insufficient and/or the heart rate was <100 beats per minute and/or the infant was cyanotic. After the initial inflation, ENCPAP at 5 to 6 cm H2O was started. If breathing was sufficient, the patient was observed in the delivery room before transportation to the NICU. If there was improvement (heart rate >100 beats per minute and pink color, but apnea or insufficient breathing), intermittent ventilation with a PIP of 20 to 25 cm H2O and a rate of 60 per minute was delivered through the nasopharyngeal tube for several minutes until the infant improved (heart rate >100 beats per minute, pink color, and spontaneous breathing). Endotracheal intubation and mechanical ventilation was initiated if the heart rate did not increase above 100 beats per minute, the infant remained cyanosed, breathing was absent, or marked dyspnea occurred (time = 90 seconds to 5 minutes). Patients were transferred to the NICU with ENCPAP or intermittent mandatory ventilation.
Conventional Intervention Group
In this group, a self-inflating bag and mask with a built-in pressure limitation (Ambu Infant R Resuscitators, Ambu, Ballerup, Denmark) and an oxygen reservoir were used after birth. A manometer was attached to monitor the pressures given. The mask and bag deliver inconsistent PIP and minimal PEEP and are unable to deliver a sustained inflation.14,18,19 With this approach, a higher initial pressure is used to open the lung, and ENCPAP was only given on arrival in the NICU if needed. Mask and bag ventilation was administered during 30 seconds if breathing was insufficient after oropharyngeal and nasal suctioning (time = 30–60 seconds). Initial inflation pressures of 30 to 40 cm H2O were used; after that not >20 cm H2O was allowed.3 If breathing remained insufficient, or the heart rate was < 100 beats per minute, or the infant remained cyanotic, or inflation was not possible, endotracheal intubation and mechanical ventilation were performed (time = 60–90 seconds). If bag and mask resuscitation was successful and breathing was sufficient (spontaneous breathing, normal chest movements, no cyanosis, heart rate >100 beats per minute), the infant was observed in the delivery room and transferred to the NICU. Nonintubated infants were transferred with oxygen monitored by measuring oxygen saturation.
In Both Groups
One hundred percent oxygen was initially used and weaned down as quickly as possible depending on the infants' response, color, and heart rate. Pulse oximetry was started immediately after resuscitation. If no respiratory support in the delivery room was needed, the infant was observed and transferred to the NICU. Nasal continuous positive airway pressure (NCPAP) was started if there were signs of respiratory distress or the fraction of inspired oxygen (FIO2) was >0.3. Infants on ENCPAP or NCPAP were placed on an Infant Flow Device (EME Tricomed, Brighton, United Kingdom) or Infant Star ventilator (Infrasonics Inc, San Diego, CA) with Hudson prongs (Hudson-RCI, Temecula, CA). The level of pressure was titrated from 5 to 8 cm H2O according to the degree of respiratory distress, assessed by observing chest retractions, effort of breathing, chest radiograph, and oxygen requirement (PaO2 >50 mm Hg, while pH >7.20 and PaCO2 <60 mm Hg).
Caffeine or theophylline were given as soon as possible after birth to infants <30 weeks' gestation and in more mature infants if they had apnea. Arterial and transcutaneous partial pressures of oxygen and carbon dioxide and oxygen saturation were monitored. RDS was defined in the presence of clinical features (need of supplemental oxygen, sternal retraction, intercostal and subcostal recession, grunting and tachypnea) and radiologic finding of poor lung expansion. Chest radiographs were used to assess the severity of RDS and lung expansion. Chest radiographs were reviewed by a radiologist, and the reading was recorded in the database. The radiologist did not participate in this trial, only the usual clinical information was given, and he/she was not aware of the treatment assignment. Intubation and mechanical ventilation were initiated either when the arterial oxygen saturation values were <88% or PaO2
50 mm Hg while receiving FIO2
0.40 (corresponding with an alveolar-arterial oxygen tension difference of <0.22), or the PaCO2 was >60 mm Hg, with a pH <7.20, or there were >4 apneic episodes in 1 hour or the infant needed >2 episodes of bagging per hour. These criteria were agreed on by participating clinicians before the study started and were applied rigorously. The decision was made by clinicians other than the investigators. Whenever 1 of the investigators was supervising the NICU, 1 of their colleagues (fellows) made the decision to intubate or not intubate the infants included in the study. Surfactant (Curosurf, Chiesi, Italy) was given at 12-hour intervals when on mechanical ventilation with a mean airway pressure x FIO2 ratio >2. All infants intubated in the delivery room received surfactant shortly after arrival in the NICU, if required. All infants intubated later on in the NICU received surfactant shortly after intubation if required. Neonates were extubated as soon as the FIO2 was <0.3 and the mean airway pressure <7 cm H2O. Immediately after extubation, NCPAP was started. NCPAP was discontinued when the neonate remained stable with a capillary PCO2 <60 mm Hg and oxygen saturation >92% without supplementary O2. When taken off NCPAP, infants were given supplemental oxygen using low flow nasal cannulae if saturation was <92%. If oxygen requirements exceeded 30% (effective FIO220), NCPAP was restarted.
All infants had cerebral ultrasounds performed at least 3 times in the first week and weekly thereafter.
The primary outcome measure was the percentage of infants intubated within 72 hours of age. Secondary outcome measures were intubation in the delivery room, the need for mechanical ventilation and surfactant treatment, death during admission or BPD based on the National Institute of Child Health and Human Development definition,21 intraventricular hemorrhage (IVH), periventricular leucomalacia, retinopathy of prematurity, persistent ductus arteriosus, and necrotizing enterocolitis.
Data are reported as means and SDs or as medians and interquartile range (IQR) if appropriate. Sample size analysis showed that to detect a reduction in intubation and mechanical ventilation from 60% to 40%, with a power of 80% and an
error of 5% (2-tailed test), 97 infants were required for each arm of the study. Because our center admits
200 eligible newborns per year and we expected 20% of the parents to refuse consent, the duration of the study was estimated at 1 year and 3 months. All analyses were performed on an intention-to-treat principle. The baseline characteristics and outcome parameters in the 2 treatment groups were compared using Student's t test for parametric and the Mann-Whitney U test for nonparametric comparisons for continuous variables, and the
2 test for categorical variables. Reported P values are 2-sided, and P < .05 was considered statistically significant. The presented odds ratio (OR) with the corresponding 95% confidence interval (CI) is an approximation to the relative risk.
This study was approved by the Leiden University Medical Center Ethics Review Committee.
| RESULTS |
|---|
|
|
|---|
|
|
Secondary outcomes are shown in Table 2. The duration of ventilatory support (including NCPAP) was shorter in infants in the EFURCI group compared with those in the conventional group (median days [IQR]: 2.7 [0.5–10] vs 4.3 [0.5–20]; P = .01) In the subgroup of infants ventilated within 72 hours of age, total time of ventilatory support (including NCPAP) was less in infants in the EFURCI than in the conventional group (median days [IQR]: 10 [4–19.5] vs 15 [5.6–36.3]; P = .04). The first pH, PaCO2, and FIO2 on arrival in the NICU and maximum of FIO2 used were similar in both groups (pH 7.23 ± 0.1 in both groups; PaCO2 6.9 ± 1.4 vs 6.8 ± 1.6 kPa; FIO2 0.32 ± 0.17 vs 0.32 ± 0.19; maximum FIO2 used 0.4 ± 0.25 vs 0.36 ± 0.19). The incidence of RDS was less in infants in the EFURCI group compared with those in the conventional group (39 [38%] of 104 vs 56 [54%] of 103; P = .015; OR: 0.50 [95% CI: 0.29–0.88]). The incidence of pneumothoraces was not significant different between the groups (1 [1%] of 104 vs 7 [7%] of 103; P = .069; OR: 0.13 [95% CI: 0.02–1.10]).
|
|
| DISCUSSION |
|---|
|
|
|---|
This is the first randomized, controlled trial, to our knowledge, in which the EFURCI ventilation strategy is compared with bag and mask ventilation advised by international neonatal resuscitation guidelines.3 Our results are consistent with the findings in the retrospective report of Lindner et al10, who compared the same lung recruitment maneuver and ENCPAP in 1996 with elective intubation as historical control in 1994. Their study group consisted of smaller infants (mean gestational age of 26.9 weeks and birth weight of 739 g in the intervention group), but their intubation rate in the delivery room decreased from 84% to 40% (P < .001), and 7% were never intubated in the 1994 group compared with 25% in 1996 (P < .01). The rate of moderate-to-severe BPD decreased from 32% to 12% (P < .05).10 Consistent with our results, no harmful effects of their new approach were found; for example, there were no increased rates of IVH or pneumothorax.
Our trial showed that the EFURCI approach allows more infants to breathe during the first days with ENCPAP alone. ENCPAP and selective surfactant treatment is an accepted alternative, and retrospective studies have shown less morbidity when ENCPAP is used in the delivery room to avoid intubation, even if ENCPAP fails later on and intubation follows.1,4,22–24 More prospective trials are under way, but there is currently insufficient information to evaluate the effectiveness of prophylactic (early) NCPAP in very preterm infants.25 Finer et al26 found, in a feasibility study among infants <28 weeks' gestation receiving CPAP/PEEP or not in the delivery room, no differences in intubation rate or surfactant use.
Animal studies have shown that an inflammatory process can be initiated with the first large manual breaths during resuscitation and may ultimately lead to BPD.5–9 Very preterm infants may not be able to generate high enough inspiratory pressures to achieve effective lung expansion and, therefore, need ventilatory support.27,28 A prolonged inflation time, used if spontaneous breathing is inadequate, may help the preterm infant to overcome the long time constant of a fluid-filled lung and prevent the use of potentially dangerous high inspiratory pressures.29,30
The beneficial effects of a sustained inflation were not confirmed in recent randomized, controlled trials.31,32 Lindner et al31 compared a 15-second inflation to intermittent mandatory ventilation in infants <29 weeks' gestation. Consistent with our findings, there was no difference regarding the intubation rate <72 hours in this group of infants. Harling et al32 used a different method by comparing a sustained inflation of 5 seconds with a conventional inflation of 2 seconds. Both studies lacked power because of small sample sizes. To maintain an adequate lung volume after inflation and to prevent atelectrauma, application of PEEP/CPAP is necessary.33–35 Starting time of ENCPAP is important, because a noncompliant, surfactant-deficient lung has a tendency to collapse and lung volume is not maintained if CPAP/PEEP is not given immediately to keep the lung open. Self-inflating bags may deliver insufficient or excessively high PIP and minimal PEEP, leading to volutrauma and atelectrauma, even when a manometer is incorporated.14,18,19 A pressure-limited mechanical device with a T-piece delivers more consistent PIP and PEEP and is better able to deliver a sustained inflation compared with a self-inflating bag.13–16 Another advantage is that it is easy to use and thereby increases the chance of effective ventilation, even in the hands of inexperienced physicians.14 It is difficult to deliver PEEP with a face mask because the seal can break very easily.15,36 To avoid this PEEP leakage, a nasopharyngeal tube is used as interface. Data are limited, but a randomized trial showed significantly less intubations in neonates with moderate asphyxia.17 Another advantage of using a nasopharyngeal tube as interface is that CPAP/PEEP can be continued very easily and directly after resuscitation.
There was a significant decrease of RDS in infants in the EFURCI group. A possible explanation for this is that the EFURCI intervention was effective in preserving surfactant. During initial assessment of a very preterm infant in the delivery room, it is difficult to differentiate between respiratory distress attributable to transitional problems or RDS, and a trial of ENCPAP provides time to solve this problem. In the conventional group, some preterm neonates who were intubated and ventilated in the delivery room may have had transitional problems, but developed RDS secondary to lung injury.
Some very preterm infants failed ENCPAP later on, especially infants <28 weeks' gestation with RDS. The maximum level of ENCPAP used was 8 cm H2O, and a higher level of CPAP or a recruitment maneuver during CPAP might have reduced later treatment failure. Another explanation for ENCPAP failure may have been the low threshold for intubation and surfactant treatment at our institution (FIO2 >40% or PaCO2 >8.0 kPa). This low threshold was chosen to prevent the disadvantageous and deleterious effects of a late rescue with surfactant treatment (FIO2
0.6) in infants who are quickly deteriorating because of RDS.11,37 Prophylactic or early surfactant treatment of neonates requiring mechanical ventilation is more effective than late rescue treatment38,39 but has the potential disadvantage that some preterm neonates are treated who are surfactant sufficient and will not develop RDS.
There are limitations of this randomized, controlled trial. To reach the most effective application of an open lung strategy, we combined several techniques (mechanical pressure-limited device, prolonged inflation, nasopharyngeal tube as interface, and direct ENCPAP in the delivery room). This makes it impossible to determine which factor contributes most to the final result. In the conventional method, a higher pressure is used initially to open the lung. Although this technique is in agreement with the international guidelines, this could have contributed to the poorer outcome. Biases in the management could have occurred because the study was not blinded, and the staff performing the study also took care of the infants later on. However, the decision to intubate was made by clinicians other than the investigators. Whenever 1 of the investigators was supervising, 1 of their colleagues (fellow) made the decision to intubate or not in the infants included in the study. We tried to minimize these biases by maintaining strict criteria and definitions during the trial. This trial was performed in a single center with experienced neonatologists, trained in the techniques of EFURCI, and it is possible that others might not get the same results.
Limits of viability in the Netherlands are set at 25 weeks' gestation, thus the results of this study cannot be applied to infants <25 weeks' gestation. We detected little effect of the EFURCI approach in infants <28 weeks' gestation, but this trial was not designed and powered to detect a difference in this subgroup. Whether the EFURCI approach is more efficient and less injurious among the most vulnerable preterm infants, that is, those with a gestational age of 23 to 27 weeks, needs additional evaluation.
| CONCLUSIONS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Address correspondence to Arjan B. te Pas, MD, Department of Pediatrics, Leiden University Medical Center, J6-S, Box 9600, 2300 RC Leiden, Netherlands. E-mail: a.b.te_pas{at}lumc.nl
Both authors have contributed to the manuscript and have seen and approved the final version.
Dr te Pas had primary responsibility for protocol development, patient screening, enrollment, outcome assessment and writing of the manuscript. Dr Walther contributed to protocol development and writing of the manuscript.
The authors have indicated they have no financial relationships relevant to this article to disclose.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. J Morley and P. G Davis Advances in neonatal resuscitation: supporting transition Arch. Dis. Child. Fetal Neonatal Ed., September 1, 2008; 93(5): F334 - F336. [Full Text] [PDF] |
||||
![]() |
A. B. te Pas and F. J. Walther Possible Sequelae of Sustained Lung Inflation in Resuscitation of Preterm Infants: In Reply Pediatrics, November 1, 2007; 120(5): 1222 - 1223. [Full Text] [PDF] |
||||
![]() |
E. J. Molloy Possible Sequelae of Sustained Lung Inflation in Resuscitation of Preterm Infants Pediatrics, November 1, 2007; 120(5): 1221 - 1222. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||