Sustained Lung Inflation Does Not Decrease Bronchopulmonary Dysplasia or Death among Extremely Preterm Infants: The SAIL Randomized Clinical Trial

July 25, 2019

MANUSCRIPT CITATION

Kirpalani H, Ratcliffe SJ, Keszler M, Davis PG, Foglia EE, Te Pas A et al. Effect of Sustained Inflations vs Intermittent Positive  Pressure Ventilation on Bronchopulmonary Dysplasia or Death Among Extremely Preterm Infants: The SAIL Randomized Clinical Trial, JAMA. 2019 Mar 26;321(12):1165-1175. PMID  30912836

REVIEWED BY

Rajesh Pandey, MD
McGovern Medical School, University of Texas Health Science Center at Houston, Houston, TX, United States.

Narmada Pandey Sapkota
Institution: Cleveland State University

TYPE OF INVESTIGATION

Prevention

QUESTION

Do preterm infants, born between 23 to 26 weeks’ gestational age with inadequate respiratory efforts or bradycardia at birth (P) who received sustained lung inflation (I), compared to standard care (intermittent positive pressure ventilation) (C), have a lower rate of bronchopulmonary dysplasia or death (O) at 36 weeks’ postmenstrual age (T)?

METHODS

  • Design: Randomized parallel group trial with permuted block randomization (variable block sizes of 2, 4, or 6) stratified by site and gestational age (23 weeks to 24 weeks and 6 days; and 25 weeks to 26weeks and 6 days). Multiple births were randomized in the same group.
  • Allocation: Allocation concealed with opaque, sealed envelope color-coded by gestational age
  • Blinding: unblinded
  • Follow-up period: 36 weeks’ postmenstrual age
  • Setting: 18 neonatal intensive care units in 9 countries
  • Patients: infants between 23 weeks and 0 days’ to 26 weeks and 6 days’ gestational age who required resuscitation for inadequate respiratory efforts (apneic or gasping) or bradycardia (heart rate < 100 peats per minute) at birth. Infants who were deemed nonviable or with major congenital anomalies were excluded.
  • Intervention: Airway assessment, application of CPAP of 5 to 7 cm H2O and assessment of meeting criteria for the trail was done within 30 seconds of placement of the infant in the resuscitation bed. The intervention group received sustained lung inflation at a peak pressure of 20 cm H2O lasting for 15 sec and second sustained inflation at a peak pressure of 25 cm H2O lasting 15 seconds, if needed, noninvasively, using a facemask or nasopharyngeal tube attached to a T piece resuscitator. Control group received intermittent positive pressure ventilation with positive end-expiratory pressure. Pre-specified criteria was followed for intubation and extubation.
  • Primary outcome: Rate of BPD or death at 36 weeks’ postmenstrual age
  • Secondary outcomes: 27 related to the efficacy of the intervention, 7 related to the safety of the intervention and 5 adverse events were requested by data safety and monitoring committee (DSMC)
    • Secondary outcomes related to the efficacy of the intervention
      • Delivery Room: (1) Heart rate (<60, 60-100, >100) (2) Type of respiratory support (CPAP, PPV) on departure from DR (3) Fraction of Inspired Oxygen (FiO2) on departure from DR (4) pressure-volume characteristics in the DR (5) need for intubation in DR
      • First 48 hours of life (6) use of inotropes (7) pneumothorax (8) need for new chest drains (9) oxygen requirement of more than 40% for 2 hours or more (10) highest FiO2 level recorded during the first 48 hours post DR (11) Area under the hourly FiO2 curve (12) Need for intubation in DR or during first 48 hours of life
      • First 7 days of life (13) Death or need for positive pressure ventilation at 7 days
      • First 10 days of life (14) Need for new chest drains (15) Duration of any chest drain (16) Highest FiO2 level recorded from 48 hours to 10 days of life (17) Air leak
      • 36 weeks PMA (18) Death (19) BPD defined using a standardized oxygen reduction test
      • Discharge: (20) Retinopathy of prematurity (ROP) stage 3 or greater requiring treatment, (21) Death in hospital, (22) Survival to discharge home without BPD, retinopathy of prematurity (grades 3 & 4), or significant brain abnormalities on head ultrasound, (23) Length of hospital stay in days, (24) Use of postnatal steroids for treatment of BPD, (25) Duration of respiratory support (ventilation, CPAP, supplemental oxygen)
      • 22-26 Months corrected gestational age: (26) Neurodevelopmental outcomes and (27) Respiratory outcomes
    • Related to Safety: The primary safety endpoint was: Death or BPD at 36 weeks PMA. Pre-specified secondary safety endpoints were: (1) Death within 48 hours of delivery (2) Oxygen requirement of FiO2 more than 40% for 2 hours or more within the first 48 hours post-delivery (3)Rate of pneumothorax within the first 10 days of life (4) Rate of pulmonary interstitial emphysema (PIE) within the first 10 days of life (5) Rate of pneumopericardium within the first 10 days of life (6) Grade 3 or 4 IVH within the first 10 days of life and (7) Any other serious adverse events that have been adjudicated as potentially relating to the intervention
    • Pre-specified Adverse Events requested by DSMC
      • (1) Epinephrine in the first 48 hours
      • (2) Chest compressions in the first 48 hours
      • (3) Grade 1 or 2 IVH within the first 10 days of life.
      • (4) Infant requiring > 30% Oxygen or NC > 2 LPM on day of life 28.
      • (5) Infant on mechanical support (includes CPAP, NIPPV, invasive ventilation via ETT) on day of life 28.
  • Deferred consent was done where sustained inflation was used commonly and was allowed by the IRB
  • Analysis and Sample Size: Analyzed as randomized. To achieve the absolute risk reduction of death or BPD by 12.5 % with 80% power and 2-sided α = 0.038 (adjusted p-value for two planned interim analyses), 262 per group was required to be enrolled but adjusting for clustering of multiple births, the investigators targeted recruitment of 600 infants (300 in each group). The trial was stopped after recruiting only 426 infants as predefined secondary safety endpoint was met because the DSMC reported suggestion of higher death in the sustained inflation group within the first 48 hours of life compared to control group
  • Patient follow-up: 460 infants randomized with 426 (92.6%) included in the analysis

MAIN RESULTS

Maternal and infant characteristics were comparable between the groups (Table 1). In the sustained inflation group, 137 infants (63.7%) died or survived with BPD compared to 125 infants (59.2%) in the standard resuscitation group (adjusted risk difference [aRD], 4.7% [95%CI, −3.8% to 13.1%]; P = 0.29) (Table 2). Sixteen infants (7.4%) in the sustained inflation group vs 3 infants (1.4%) in the standard resuscitation group died at less than 48 hours of age (aRD, 5.6% [95%CI, 2.1% to 9.1%]; P = 0.002). Early death was more likely to occur with placenta abruption or vaginal breech delivery and more likely to be born at 23 and 24 weeks GA. Twenty-six of 27 secondary efficacy outcomes assessed by 36weeks’ were not significantly different between groups. The study was underpowered by early stopping, but a Bayesian analysis revealed that it was highly unlikely that sustained inflation would be shown to be preferable if the study had continued to complete planned recruitment.

Table 1. Maternal and Infant Characteristics by Randomized Group

Resuscitation, No. (%)
Sustained Inflation (n=215) Standard (n=211)
Maternal Age, median (IQR) y 31.1 (26.5-35.4) 30.5 (26.0-34.9)
Any antenatal steroids 208 (96.7) 205 (97.2)
Full course of antenatal steroids 168 (78.1) 165 (78.2)
Placental abruption 33 (15.3) 28 (13.3)
Chorioamnionitis 81 (37.7) 68 (32.2)
Mode of Delivery
vaginal – vertex 68 (31.7) 62 (29.4)
vaginal – breech 13 (6.0) 15 (7.1)
Cesarean delivery 134 (62.3) 134 (63.5)
Male 119 (55.3) 103 (48.8)
Gestational Age
23- to 24-wk stratum 76 (35.3) 75 (35.5)
25- to 26-wk stratum 139 (64.7) 136 (64.5)
Birth weight, median (IQR), g 725 (620-855) 731 (680-845)
Proportion <10th centile birth weight 28 (13.0) 25 (11.8)
Multiple birth status
Twin 54 (25.1) 54 (25.6)
Triplet 3 (1.4) 4 (1.9)

Abbreviations: IQR= Interquartile range, Y= Years

Table 2: Primary Composite Outcome and Component Secondary Outcomes at 36Weeks’ Postmenstrual Age

Resuscitation, No. (%)
Outcome Sustained Inflation (n=215) Standard (n=211) Adjusted Risk Difference, %(95% CI) Adjusted Relative Risk % (95% CI) P Value
Death or BPD 137 (63.7) 125 (59.2) 4.7 (-3.8-13.1) 1.1 (0.9 to 1.2) 0.29
Death 45 (20.9) 33 (15.6) 5.2 (-2.3 to 12.7) 1.3 (0.9 to 1.9) 0.17
BPD 92 (42.8) 92 (43.6) 0.5 (-8.5 to 9.4) 1.0 (0.8 to 1.2) 0.92

Risk difference of sustained inflation − adjusted for in all models were gestational age, site, infant sex, maternal corticosteroid use, initial heart rate, small for gestational age, and consent type used. Also adjusted for the correlation between infants from multiple births.

P value from comparison of adjusted risk difference.

BPD= bronchopulmonary dysplasia

CONCLUSION:

Sustained lung inflations, compared with standard intermittent positive pressure ventilation, did not reduce the risk of BPD or death at 36 weeks’ postmenstrual age. These findings do not support the use of ventilation with sustained inflations among extremely preterm infants.

COMMENTARY

The CONSORT guidelines allow for meaningful assessment of how randomized controlled trials have been designed, analyzed, and reported, and a thorough evaluation of a RCT within the context of these guidelines allows for fair interpretation of the data presented.(1) Title and abstract sections of the article identifies the trial as a randomized trial and includes a structured summary of trial design, method, result, and conclusion. (2) The introduction section includes the scientific background, rationale as well as specific objectives of whether sustained lung inflation at birth reduces bronchopulmonary dysplasia (BPD) or death which is a relevant topic as the rates of BPD remain high.(3)

Method section describes the trial design (parallel group trial with permuted block randomization), eligibility criteria, settings, and locations of participants. There were no critical changes in the method after starting the trial. Interventions to each group and outcomes, both primary and secondary, were defined in sufficient detail. The sample size was calculated to detect absolute risk reduction of death or BPD by 12.5 % with 80% power and 2-sided α = 0.038 (adjusted p-value for two planned interim analyses). The authors had pre-specified planned interim analysis and stopping rules. Computer generated parallel group randomization was done, and the trial was not blinded. Statistical analysis, for primary and secondary analyses, as well as subgroup analyses and sensitivity analyses were robust. Flow diagram accounting for all enrolled infants was presented.

The investigators planned to recruit 600 infants but trial was stopped early after recruiting 426 infants as predefined secondary safety endpoint was met when the data and safety monitoring committee reported suggestion of higher death in the sustained inflation group within the first 48 hours of life compared to control group (7.4 % vs. 1.4 % aRD, 5.6% [95% CI, 2.1% to 9.1%]; P = .002). Baseline characteristics of the two groups were comparable. Primary outcomes death or BPD at 36 weeks, as well as death and BPD at 36, weeks were not different between the groups. There were no significant differences in secondary outcomes.

The discussion section mentions the limitations of the article. The article does not report the comparative outcomes between the centers where sustained lung inflation was routine versus the centers where the procedure was investigational. These findings would be meaningful because the familiarity of the procedure may change the outcomes. The trial is registered in clinicaltrials.gov, the full protocol is available, and funding information is provided.

Previous trials to evaluate the efficacy and effectiveness of sustained lung inflation on preterm infants that enrolled preterm infants of different gestational age and birth weight have shown the inconsistent result in short term outcomes. (4, 5, 6) Animal studies have shown an inverse relationship between gestational age and lung aeration from sustained lung inflation(7), so the response to sustained inflation may vary with the size and gestational age of an infant. The SAIL trial performed well under CONSORT assessment and concluded that sustained lung inflation should not be used in this patient population.

REFERENCES

  1. Moher D, Hopewell S, Schulz KF, Montori V, Gøtzsche PC, Devereaux PJ, Elbourne D, Egger M, Altman DG, CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials. BMJ. 2010 Mar 23;340:c869
  2. Kirpalani H, Ratcliffe SJ, Keszler M, Davis PG, Foglia EE, Te Pas A et al., Effect of Sustained Inflations vs Intermittent Positive Pressure Ventilation on Bronchopulmonary Dysplasia or Death Among Extremely Preterm Infants: The SAIL Randomized Clinical Trial.JAMA. 2019 Mar 26;321(12):1165-1175
  3. Horbar JD, Edwards EM, Greenberg LT, Morrow KA, Soll RF, Buus-Frank ME et al., Variation in Performance of Neonatal Intensive Care Units in the United States. JAMA Pediatr. 2017 Mar 6;171(3):e164396
  4. Lindner W, Högel J, Pohlandt F. Sustained pressure-controlled inflation or intermittent mandatory ventilation in preterm infants in the delivery room? A randomized, controlled trial on initial respiratory support via nasopharyngeal tube. Acta Paediatr. 2005 Mar; 94(3):303-9
  5. Lista G, Boni L, Scopesi F, Mosca F, Trevisanuto D, Messner H et al., Sustained lung inflation at birth for preterm infants: a randomized clinical trial, Pediatrics. 2015 Feb;135(2):e457-64.
  6. te Pas AB, Walther FJ. A randomized, controlled trial of delivery-room respiratory management in very preterm infants. Pediatrics. 2007 Aug; 120(2):322-9.
  7. McCall KE, Waldmann AD, Pereira-Fantini P, Oakley R, Miedema M, Perkins EJ  Time to lung aeration during a sustained inflation at birth is influenced by gestation in lambs. Pediatr Res. 2017 Oct; 82(4):712-720.
css.php