Manuscript Citation: Roberts KD, Brown R, Lampland AL, Leone TA, Rudser KD, Finer NN, et al. Laryngeal mask airway for surfactant administration in neonates: A randomized, controlled trial, J. Pediatr, 2018 Feb; 193; 40-6.
Name: Captain Andrew J. Groberg, MD, and LTC Jay M. Dintaman, MD, FAAP
Title: Dr. Groberg is a neonatology fellow (PGY-6). Dr. Dintaman is the Neonatology Service Chief
Institution: Walter Reed National Military Medical Center (WRNMMC)
Email: Andrew.firstname.lastname@example.org and Jay.email@example.com
Corresponding author: Andrew J. Groberg
Address: Walter Reed National Military Medical Center, NICU.
8901 Rockville Pike,
Bethesda, MD 20889
Type of Investigation: Treatment
Question: In premature infants 280/7 – 356/7 weeks gestation with moderate respiratory distress syndrome (RDS), does administration of surfactant by laryngeal mask airway (LMA) decrease the rate of intubation and mechanical ventilation during the first 7 days of life, compared to infants maintained on continuous positive airway pressure (CPAP) with no surfactant administration?1
Design: Prospective, multicenter, randomized controlled trial
Allocation: Subjects were randomized to one of two groups:
LMA group: LMA placed, surfactant administered, LMA removed and placed back on CPAP 6 cm H2O.
Control group: Maintained on CPAP 6 cm H2O with no surfactant administered.
Randomization by computer-generated random numbers, stratified by study site and gestational age group (280/7- 316/7 weeks and 320/7-356/7 weeks) using sequentially numbered opaque, sealed envelopes. Eligible multiples were assigned to the same study group.
Blinding: Clinical providers were not masked to the study intervention
Follow-up period: 7 days of life for primary outcome. Other outcomes evaluated during initial hospitalization.
Setting: Seven level III and level IV NICUs in the United States.
Inclusion Criteria: Eligibility criteria included preterm infants 280/7 – 356/7 weeks, weight ≥ 1250g, and age ≤36 hours. Infants were on noninvasive CPAP or BiPAP, requiring FiO2 0.3-0.4 for ≥30 minutes to maintain O2 sats 88%-92%. Chest radiograph and clinical presentation were consistent with RDS. Written, informed, parental consent was obtained before enrollment.
Exclusion Criteria: Exclusion criteria included infants with prior mechanical ventilation or surfactant administration, major congenital anomalies, abnormality of the airway, respiratory distress because of etiology other than RDS, and APGARS <5 at 5minutes.
LMA group: Infant pre-medicated with 24% oral sucrose onto the tip of their tongue and atropine (0.02mg/kg IV) and placed in supine position. An orogastric tube was placed and gastric contents aspirated. An LMA y-piece adapter (Neo-verso airway access adapter CSC200) was attached to a size 1 LMA. A closed suction catheter (Neo-Verso Fluid/Access Catheter CSC205S) with a syringe of surfactant was connected to one arm of the LMA adapter and a PediCap with a self-inflating bag was placed on the other arm. The size 1 LMA was placed into the pharynx and the LMA balloon inflated with 3 mL of air. Placement of the LMA was confirmed with PediCap color change. The suction catheter was advanced to the first red mark and Curosurf (2.5ml/kg) was administered in 2mL aliquots. Bag ventilation was continued for 2 minutes post-surfactant. The LMA was deflated and removed and the infant returned to CPAP 6 cm H2O. A post-procedure gastric aspiration was performed to assess for surfactant in the stomach.
Control group: Infant maintained on CPAP 6 cm H2O
Total number of hours on mechanical ventilation, CPAP and supplemental O2 during the first 7 days of life and for the entire hospitalization
Incidence of pulmonary air leak
Survival until discharge
O2 therapy at 36 weeks Postmenstrual age
O2 therapy at discharge
Severe IVH (grade III/IV)
Analysis and Sample Size: Based on previous data for 1250-2500g neonates, 36% of infants were expected to require intubation2. To detect an absolute risk difference of 23%, from 36% of controls (requiring intubation) to 13% in the LMA group, 124 infants (62 in each group) were needed for 80% power (with type I error of 0.05). With the anticipated enrollment of multiples, the total sample size was increased by 15% to 144 (72 infants per group). However, enrollment terminated after 4 years, with approval by the Data Safety Monitoring Board, based on the assumption that target enrollment would not be reached.
Patient follow-up: First 7 days of life for primary outcome. Additional outcomes assessed during the initial hospitalization with no follow-up after discharge. Time zero was defined as the time of surfactant delivery in the LMA group and the time of randomization in the control group.
Surfactant administration by LMA significantly decreased the need for intubation and mechanical ventilation in the first 7 days of life compared to controls (38% vs 64%, OR 0.30 [95% CI 0.13, 0.70] P = 0.006). Baseline characteristics did not significantly differ between the LMA and control group. There were more infants with no supplemental oxygen requirement in the LMA group, compared to control group, at 5, 15, 30 and 60 minutes post-intervention (P <0.001) for each comparison. The mean FiO2 requirement was also significantly lower for infants in the LMA group at 15 minutes, 30 minutes and 1 hour. However, there was no difference in need for supplemental oxygen between the groups at 4 and 12 hours post-intervention and the total number of hours on mechanical ventilation, CPAP, and supplemental oxygen during the first 7 days of life and for the entire hospitalization were not statistically significant between groups. The investigators reported no serious adverse events including emergent intubations and tension pneumothorax. The incidence of pulmonary air leak and chronic lung disease was similar between groups and all enrolled infants in both groups survived until discharge. In the LMA group, post-procedure gastric aspirates revealed 26% had no surfactant aspirated from the stomach, 46% had <10% of the administered dose and 18% had >50% of the administered dose.
In this prospective, multicenter, randomized controlled trial, the authors concluded that premature infants with moderate respiratory distress syndrome who received surfactant through an LMA had decreased rates of intubation and mechanical ventilation compared to infants on CPAP who received no surfactant.
This multicenter study adds preliminary support for the incorporation of LMAs into neonatal resuscitations for preterm infants with moderate respiratory distress. This project and the preceding pilot study3 demonstrate that LMAs can be placed quickly and administering surfactant via an LMA is feasible.
The investigators note that LMA administration of surfactant may reduce the need for laryngoscopy, sedation and paralysis compared to other listed methods of surfactant administration (Intubation-Surfactant-Rapid Extubation method, Less Invasive Surfactant Administration (LISA), aerosolized, etc.). In comparison with alternative surfactant delivery methods, the authors noted that an LMA coupled with a y-piece adaptor for surfactant administration allowed for continuous PEEP delivery while minimizing lung de-recruitment. In resource limited environments, they suggest surfactant delivery via LMA may be useful at improving treatment of RDS. Future, large-scale studies comparing surfactant delivery via LMA versus established surfactant delivery techniques (i.e. INSURE method) would be useful in clarifying the efficacy and safety of LMA surfactant delivery in clinical practice. Some limitations of study design may have impacted the conclusions of this project; those include limiting LMA delivery of surfactant to a single dose and limiting the level of CPAP to 6 cm H2O. Sixty-three percent of treatment failure in the LMA group occurred >12 hours after the first dose of surfactant. If the study design had permitted a second surfactant dose to be delivered by LMA, then the number of treatment failures in the LMA group may have potentially been reduced. Standardization of CPAP at 6 cm H2O may limit the study's generalizability to centers where higher levels (> 6 cm H2O) of non-invasive respiratory support (i.e. SiPAP, NIPPV, etc.) is commonplace. The investigators did acknowledge that gastric aspiration was an imprecise measure of LMA cuff leak but used this technique to draw conclusions related to distribution of administered surfactant. They reported clinical evidence suggesting surfactant reached the lungs because >50% of neonates weaned to 21% FiO2 within 30 min of receiving surfactant compared to 0% from the control group.
There is substantial practice variation concerning the optimal timing and method for neonatal surfactant delivery. Maternal administration of corticosteroids and applying CPAP at delivery for premature infants at risk for RDS is a reasonable approach to avoid elective intubation5. Previous studies have also demonstrated that early surfactant delivery to infants with RDS requiring intubation results in better outcomes than later administration6. In the present study, it is notable that 19 infants (36%) in the control group avoided intubation and surfactant administration during the first 7 days of life, highlighting the finding that a substantial portion of premature infants born >28 weeks with symptoms of moderate respiratory distress did not require surfactant.
This is a valuable study that adds to the growing body of knowledge focused on alternative surfactant delivery techniques in the premature population. In order to make true comparisons between current surfactant delivery methods and LMA surfactant delivery, additional studies, such as non-inferiority trials, would be useful. Respiratory support strategies will continue to evolve and impact the indications for surfactant, as well as the delivery methods for surfactant.
EBM lesson: Interpreting trials stopped early
There are many reasons to stop a trial early including patient safety concerns, slow participant accrual, and highly statistically significant interim findings7. Early trial termination can potentially save resources and time but may warrant caution when interpreting results. Frequent interim data analysis increases the likelihood of observing random fluctuations above or below the true treatment effect thereby resulting in a higher chance of falsely declaring an effect significant (type I error). Certain techniques can mitigate this type I error, such as limiting the number of, and requiring more conservative thresholds (lower p value) for the interim analyses and/or using alpha spending functions to account for multiple analyses. This practice should be pre-planned and incorporated into protocols that will consider early termination for efficacy. In contrast, stopping a trial early for futility does not require statistical adjustment since there are no subsequent opportunities for additional hypothesis testing, but this early stoppage may ultimately result in a reduction of trial power. The Roberts et al. trial was terminated early for slow patient accrual after 4 years of active enrollment. While stopping a trial for slow participant accrual does not warrant specific practices to protect against type I error7, interpretation of positive findings must account for a likely smaller true positive effect. Roberts et al. detected an OR of 0.3 in their primary outcome but given the early trial termination, the true magnitude of this effect could have been lower.
The Journal Club is a collaboration between the American Academy of Pediatrics - Section of Neonatal Perinatal Medicine and the International Society for Evidence-Based Neonatology (EBNEO.org).
Roberts KD, Brown R, Lampland AL, Leone TA, Rudser KD, Finer NN, et al. Laryngeal mask airway for surfactant administration in neonates: A randomized, controlled trial, J. Pediatr, 2018 Feb; 193; 40-6.
Vermont Oxford Network. Expanded database nightingale internet reporting system. Burlington (VT): Vermont Oxford Network; 2007-2009.
Wanous AA, Wey A, Rudser KD, Roberts KD. Feasibility of laryngeal mask airway device placement in neonates. Neonatology 2016;111:222-7.
Pinheiro JMB, Santana-Rivas Q, Pezzano C. Randomized trial of laryngeal mask airway versus endotracheal intubation for surfactant delivery. J perinatal 2016;36:196-201.
Niemarkt HJ, Hütten MC, Kramer BW. Surfactant for Respiratory Distress Syndrome: New Ideas on a Familiar Drug with Innovative Applications. Neonatology. 2017 May; 111(4): 408–414.
Bahadue FL, Soll R. Early versus delayed selective surfactant treatment for neonatal respiratory distress syndrome. Cochrane Database Syst Rev. 2012;11
Viele, K, McGlothlin, A, Broglio, K. Interpretation of Clinical Trials That Stopped Early. JAMA. 2016 Apr; Vol. 315, No. 15, 1646-1647.