Comparison of Devices for Newborn Ventilation in the Delivery Room

Edgardo Szyld, MD, MSc1, Adriana Aguilar, MD1,2, Gabriel A. Musante, MD, MSc1,3, Nestor Vain, MD1, Luis Prudent, MD1,

Jorge Fabres, MD, MSPH4, and Waldemar A. Carlo, MD5, on behalf of the Delivery Room Ventilation Devices Trial Group*

Objective To evaluate the effectiveness and safety of a T-piece resuscitator compared with a self-inflating bag forproviding mask ventilation to newborns at birth.Study design Newborns at $26 weeks gestational age receiving positive-pressure ventilation at birth wereincluded in this multicenter cluster-randomized 2-period crossover trial. Positive-pressure ventilation was providedwith either a self-inflating bag (self-inflating bag group) with or without a positive end-expiratory pressure valve or aT-piece with a positive end-expiratory pressure valve (T-piece group). Delivery room management followed Amer-ican Academy of Pediatrics and International Liaison Committee on Resuscitation guidelines. The primary outcomewas the proportion of newborns with heart rate (HR) $100 bpm at 2 minutes after birth.Results A total of 1027 newbornswere included. Therewas no statistically significant difference in the incidence ofHR$100 bpm at 2minutes after birth between the T-piece and self-inflating bag groups: 94% (479 of 511) and 90%(466 of 516), respectively (OR, 0.65; 95% CI, 0.41-1.05; P = .08). A total of 86 newborns (17%) in the T-piece groupand 134 newborns (26%) in the self-inflating bag group were intubated in the delivery room (OR, 0.58; 95% CI, 0.4-0.8; P = .002). The mean � SD maximum positive inspiratory pressure was 26 � 2 cm H2O in the T-piece group vs28 � 5 cm H2O in the self-inflating bag group (P < .001). Air leaks, use of drugs/chest compressions, mortality, anddays on mechanical ventilation did not differ significantly between groups.Conclusion There was no difference between the T-piece resuscitator and a self-inflating bag in achieving an HRof $100 bpm at 2 minutes in newborns $26 weeks gestational age resuscitated at birth. However, use of the T-piece decreased the intubation rate and the maximum pressures applied. (J Pediatr 2014;-:---).

See editorial, p ��� andrelated article, p ���

pproximately 10% of infants require some assistance to begin breathing at birth. Of these, roughly 50% need assisted

Aventilation, and less than 10% require extensive resuscitation.1,2 The most important action in the resuscitation of adepressed newborn in the delivery room is to establish effective ventilation.3 To date, there is insufficient evidence

regarding the optimal device for establishing effective ventilation in newborns at birth. Current International Liaison Commit-tee on Resuscitation (ILCOR) and American Academy of Pediatrics (AAP)/American Heart Association (AHA) recommenda-tions state that ventilation of the newborn can be performed effectively with a flow-inflating bag, a self-inflating bag, or apressure-limited T-piece resuscitator.3,4 An increase in heart rate (HR) remains the most sensitive indicator of resuscitationefficacy.4

From the 1Research Department, FUNDASAMIN,Fundaci�on para la Salud Materno Infantil; 2MastersProgram in Clinical Trials, Universidad AbiertaInteramericana; 3Department of Pediatrics, Facultad deCiencias Biom�edicas, Universidad Austral, Buenos Aires,Argentina; 4Division of Pediatrics, School of Medicine,Pontificia Universidad Cat�olica de Chile, Santiago, Chile;and 5Department of Pediatrics, University of Alabama at

Self-inflating bags are the most commonly used manual ventilation devices forproviding positive-pressure ventilation (PPV) at birth.5 In recent years, the T-piece has been included as an alternative device for ventilation in the deliveryroom, and its use is becoming more widespread.5-7 Several studies havecompared the effectiveness of both devices by testing providers with differentlevels of training in simulated neonatal resuscitation on mannequins, and foundthat the resulting pressures are lower and more consistent when using the

Birmingham, Birmingham, AL

*A list of members of the Delivery Room VentilationDevices Trial Group is available at www.jpeds.com(Appendix).

Supported by Fundaci�on para la Salud Materno Infantil(FUNDASAMIN), the National Ministry of Science andTechnology, Argentina (PICT-O- Austral 2008 # 77), theNeonatal Resuscitation Program, Covidien, USA(101409), and Fisher & Paykel. Ambu donated the self-inflating bag and PEEP valves, Fisher & Paykel donatedthe T-piece resuscitators, circuits, and masks, Uniccompany, Masimo donated most of the pulse oximeters,and Medix donated 3 pulse oximeters. The authorsdeclare no conflicts of interest.

Registered with ClinicalTrials.gov: NCT00443118.

0022-3476/$ - see front matter. Copyright ª 2014 Elsevier Inc.

All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2014.02.035

AAP American Academy of Pediatrics

AHA American Heart Association

BPD Bronchopulmonary dysplasia

FiO2 Fraction of inspired oxygen

GA Gestational age

HR Heart rate

ILCOR International Liaison Committee on Resuscitation

PEEP Positive end-expiratory pressure

PPV Positive-pressure ventilation

SpO2 Oxygen saturation

VLBW Very low birth weight

1

THE JOURNAL OF PEDIATRICS � www.jpeds.com Vol. -, No. -

T-piece.8-11 One small study compared the T-piece with apositive end-expiratory pressure (PEEP) valve and a self-inflating bag without a PEEP valve in extremely preterm in-fants, and found no difference in oxygen saturation (SpO2) at5 minutes after birth.12 No randomized controlled study hascompared the effectiveness of these 2 devices in the deliveryroom under standard clinical practice. Thus, the aim of thepresent study was to evaluate the effectiveness and safety ofa T-piece compared with a self-inflating bag for providingventilation through a face mask to newborns at birth.

Methods

This was a cluster-randomized, open-label, 2-period cross-over trial performed in 11 centers from 5 countries(Argentina, Chile, Peru, Italy, and the US) that included pub-lic, private, and university hospitals. The protocol, theinformed consent form, and the parent information materialwere approved by the Institutional Review Board of eachparticipating center. Following Institutional Review Boardguidelines and in compliance with each country’s regula-tions, prenatal consent was requested for each admitted preg-nant woman in 4 hospitals in Argentina and 1 hospital inPeru, and postnatal consent to analyze perinatal andfollow-up data was requested in 3 centers in Chile, 1 centerin Italy, and 1 center in the US. An informed consent waiverusing an opt-out approach was approved by the InstitutionalReview Board in 1 center in the US considering both devicesas standards of care.

Newborn infants at $26 weeks gestational age (GA)receiving PPV through a face mask at birth, in accordancewith current Neonatal Resuscitation Program (AAP/AHA)and ILCOR recommendations,4,13,14 were eligible for enroll-ment. Infants were excluded who required immediate endo-tracheal intubation at birth, presented with a majorcongenital malformation, or were part of a multiple birth.(Because an extra individual was necessary for each recruitedinfant, it was considered logistically impossible to enroll mostof the multiples, and so to avoid a source of bias, we decidedto exclude all of them.) We did not include infants at<26 weeks GA, because many of them would have been intu-bated immediately after birth, thereby increasing the hetero-geneity of the sample. In addition, neonates at <26 weeks GAwere also excluded owing to difficulties in standardizing theirinitial management approach.

InterventionsEach center was randomized to ventilate infants at birth with1 of 2 devices, a T-piece or self-inflating bag. The order ofintervention was assigned at random, and the group assign-ment was not submitted to the centers until study initiation.Before the start of patient enrollment, health care providerswere trained in the use of the device assigned for that period.After 50 patients were enrolled came an interval withoutintervention (“washout period”), during which the healthteam was trained in the use of the alternate device and the

2

necessary changes were made to adapt the logistic aspectsof delivery room management for the second period.Subsequently, each center was crossed over to the alternate

treatment for the period in which a second set of 50 patientswas enrolled. For the self-inflating bag period, each centerwas assigned at random to use the bag either with or withouta PEEP valve, to allow subgroup analysis related to the use ofPEEP.Delivery room management followed AAP/AHA and IL-

COR guidelines,4,13,14 and at least 1 person skilled in resusci-tation attended every delivery. In addition, an assistant timedand recorded the HR, actual maximum inspiratory pressureused during ventilation, and other study variables until theHR was stabilized. The self-inflating bag group receivedPPV with a 300-mL self-inflating bag (Mark IV Baby;Ambu, Ballerup, Denmark) with a manometer attachedvisible to the provider. In centers randomly assigned to itsuse, a PEEP valve was attached (PEEP 10; Ambu). The T-piece group received PPV using a Neopuff T-piece resusci-tator (Fisher & Paykel, Auckland, New Zealand) with aPEEP valve.Patients were initially ventilated with the following target

pressures: a positive inspiratory pressure of 25 cm H2O forboth devices, and a PEEP of 5 cm H2O when a PEEP valvewas used. The initial fraction of inspired oxygen (FiO2) was0.4 for all infants up to September 2011, and 0.21 for term in-fants ($37 weeks GA) and 0.4 for preterm infants(<37 weeks GA) thereafter. FiO2 adjustments were deter-mined by the attending physician. The following maskswere available at each delivery: silicon RD (Fisher & Paykel)of various sizes, and Ambu 0A (Ambu). For both devices,blended oxygen flow rate was set at 8 L/minute, and thepop-off valve was set at 40 cm H2O. If the infant was intu-bated in the delivery room, then PPV was provided withthe device assigned for that period. Each center subsequentlyfollowed its usual practices.Infants were followed until hospital discharge except in

cases of prolonged hospitalization. Infants of >32 weeks GAwere followed until 28 days postnatal age; those of#32 weeks GA, until 36 weeks postmenstrual age.

OutcomesThe primary outcome measure was the proportion of infantswith an HR$100 bpm at 2 minutes after birth. This outcomewas chosen based on published data as a proxy for effectiveresponse to resuscitation in the delivery room.4,15-17 Alltime measurements were made from cord clamping by atrained observer with a stopwatch. All centers used an earlycord clamping strategy throughout the study period. HRwas measured by pulse oximetry when possible or, alterna-tively, with a stethoscope.Each center used the pulse oximeter available in its delivery

room. The probe was placed immediately after birth on theinfant’s right wrist or hand before being connected to the ox-imeter cable.4 Each pulse oximeter was set to the maximumsensitivity and an averaging time of a 2-3 seconds.

Szyld et al

Table I. Baseline characteristics of the enrolled patients

VariableT-piece group(n = 511)

SIB group(n = 516) P value

Maternal age, y, mean � SD 28 � 7 28 � 7 .774Primiparity (n = 998), n (%) 220 (44) 239 (47) .320Clinical chorioamnionitis, n (%) 26 (5) 28 (5) .810Cesarean delivery, n (%) 262 (51) 290 (56) .108Meconium-stained amniotic fluid

(n = 1019), n (%)82 (16) 94 (18) .286

Male sex, n (%) 303 (59) 298 (58) .616GA, y, mean � SD 36 � 4.1 36 � 4.4 .539Antenatal steroids, n (%) 142 (27) 156 (30) .405Birth weight, g, mean � SD 2720 � 1025 2686 � 1069 .619Birth weight >2500 g, n (%) 318 (62) 330 (64) .511Birth weight <1500 g, n (%) 85 (17) 110 (21) .050

SIB, self-inflating bag.

- 2014 ORIGINAL ARTICLES

The secondary outcomes in the delivery room includedelapsed time to achievement of an HR $100 bpm, time toinitiation of spontaneous breathing, SpO2 at 2 minutes, pro-portion of infants who were intubated after failure of PPV bymask, use of chest compressions and/or medications, Apgarscores at 1 and 5 minutes, mortality rate, presence of air leaks(pneumothorax and pneumomediastinum), and maximumventilation pressures and FiO2 administered. Subsequent sec-ondary outcomes included the incidence of air leaks, use andduration of oxygen administration, number of days on me-chanical ventilation and/or continuous positive airway pres-sure, incidence of hypoxic ischemic encephalopathy, andincidence of bronchopulmonary dysplasia (BPD), definedas oxygen supplementation at 36 weeks postmenstrual agefor newborns of <32 weeks GA or for more than 28 days inthose of $32 weeks GA. Mortality before hospital dischargewas also included.

Sample Size and Statistical AnalysesIt was initially estimated that approximately 5% of newbornswould require PPV for resuscitation at birth.1 A clusterdesign was chosen because it was considered ethically unac-ceptable to delay resuscitation to randomize patients individ-ually. In addition, the need for resuscitation cannot always beanticipated, making it impossible to prepare the requiredteam and equipment for each delivery with a potential needfor resuscitation.3 The allocation sequence for the randomi-zation was computer-generated by the study statistician.

The sample size for this cluster-randomized crossoverstudy with binary response was estimated using the formulaproposed by Hughes et al18 and confirmed by computer sim-ulations. Assuming 50 patients for each center–period com-bination and a proportion of newborns with HR $100 bpmat 2 minutes after cord clamping of 0.8, for 80% power and asignificance level of 0.05 (2-sided), at least 10 centers wereneeded to detect a difference of 0.1. This would be equivalentto 70% in the group ventilated with a self-inflating bag and80% in the group ventilated with a T-piece resuscitator basedon Dawson’s data.17

An external Data and Safety Monitoring Committee car-ried out 2 planned interim analyses after the inclusion of25% and 50% of the estimated study population, to ensuresafety and evaluate the need to reestimate the sample size.The principal investigators and clinicians were not informedof partial results until study completion. After 22 months ofrecruitment, to prevent an unnecessary extension of the studywithout compromising its precision and power, the Data andSafety Monitoring Committee recommended a reduction inthe number of subjects per period in 2 centers that hadstarted their recruitment late, because their slow enrollmentrate would have doubled the trial period without substan-tially increasing the power. Data were analyzed using SPSSversion 17.0 (SPSS, Chicago, Illinois) and SAS version 9.2for Windows (SAS Institute, Cary, North Carolina). Logisticand linear regression mixed models were adjusted with thecluster effect as the random effect and treatment and birthweight <1500 g as fixed effects.

Comparison of Devices for Newborn Ventilation in the Delivery R

Data Management and MonitoringData collection was done by an investigator in each centerwho was responsible for completing the case report forms.A monitoring investigator audited and evaluated the study’sprogress through periodic visits to each center throughoutthe recruitment period. Data were entered in an electronicdatabase designed for this study.

Results

The study was initiated with 14 centers. Three centers werewithdrawn because of difficulties in the recruitment strategy;none of these completed the first stage of the study, and theirpatients were not included in the analysis. Between December2009 and August 2012, 1449 infants were eligible out of 54 891born at the participating centers (2.6%). Of these 1449 infants,1032 were enrolled, and after 5 exclusions, 511 infants in the T-piece group and 516 infants in the self-inflating bag groupwereanalyzedby intention to treat (Figure1; available atwww.jpeds.com). Six centers were allocated to use the self-inflating bagwith a PEEP valve and 5 centers used the self-inflating bagwithout a PEEP valve. The duration of the washout periodranged from 14 days to 118 days.Baseline characteristics of the 2 groups were comparable

except for the distribution of very low birth weight(VLBW) infants (Table I). Thus, we adjusted the resultsnot only by center, but also by birth weight <1500 g. Theproportion of VLBW infants also differed among centers(Figure 2; available at www.jpeds.com).There were no significant between-group differences in the

primary outcome; an HR$100 at 2 minutes was achieved by479 infants (93.7%) in the T-piece group and by 466 infants(90.3%) in the self-inflating bag group (OR, 0.65; 95% CI,0.41-1.05; P = .08). In 69% of the entire study population,HR was measured with a pulse oximeter; this proportionwas similar in both groups. Themain outcome was consistentamong centers (Figure 3; available at www.jpeds.com).Eighty-six infants (17%) from the T-piece group and 134

infants (26%) from the self-inflating bag group were intu-bated in the delivery room (OR, 0.58; 95% CI, 0.4-0.8;P = .002). The mean � SD maximum positive inspiratory

oom 3

Table II. Secondary outcomes

Outcome measure T-piece group (n = 511) SIB group (n = 516) OR (95% CI)* P value*

Outcomes in the delivery roomIntubation for ventilatory support in the delivery room, n (%) 86 (17) 134 (26) 0.58 (0.4-0.8) .0021-min Apgar score #3, n (%) 153 (30) 177 (34) 1.3 (0.9-1.7) .0705-min Apgar score #5, n %) 30 (6) 47 (9) 1.5 (0.9-2.5) .080Drugs/chest compressions, n (%) 8 (1.6) 17 (3.3) 0.5 (0.2-1.1) .090Time to spontaneous breathing, min, mean � SD† 2.7 � 36 3.05 � 3.9 - .100Time elapsed until HR $ 100 bpm, min, median (IQR) 1 (0.5-1.6) 1 (0.5-1.8) - .068Maximum PIP, mean � SDz variability 25.58 � 1.9 28 � 4.9 - <.001PIP >25 cm H2O, n (%) 52 (10) 184 (37) 5.0 (3.6-7.0) <.001Mean maximum FiO2 in delivery room, mean � SD 0.46 � 0.19 0.50 � 0.21 - .001

Outcomes after the delivery roomMortality, n (%) 11 (2.2) 15 (2.9) 1.1 (0.5-2.5) .810Air leaks (pneumothorax and/or pneumomediastinum), n (%) 13 (2.5) 8 (1.6) 0.6 (0.2-1.4) .250Mechanical ventilation, n (%) 116 (22.7) 147 (28.5) 1.3 (0.9-1.8) .160Days on mechanical ventilation, mean � SD 5.0 � 7.6 8.3 � 13.3 - .007Days on CPAP, mean � SD 7.83 � 11.3 7.96 � 10.1 - .901Hypoxic ischemic encephalopathy, n (%) 21 (4.1) 28 (5.4) 1.3 (0.7-2.4) .330Use of oxygen, n (%) 208 (40.7) 222 (43.0) 1.1 (0.8-1.5) .411Days on oxygen, mean � SD 13.8 � 17 22.8 � 25 - <.001

CPAP, continuous positive airway pressure; PIP, peak inspiratory pressure.*Corrected by center and birth weight <1500 g.†990 subjects.z1007 subjects.

THE JOURNAL OF PEDIATRICS � www.jpeds.com Vol. -, No. -

pressure delivered was 26� 1.9 cmH2O in the T-piece groupvs 28� 4.9 cmH2O in the self-inflating bag group (P < .001).Air leaks, drug administration or chest compressions, mortal-ity, and the use of mechanical ventilation did not differ signif-icantly between the 2 groups (Table II). No deaths or air leaksrequiring treatment occurred in the delivery room. In the 704infants (69%) with reliable pulse oximetry data at 2 minutes,the mean SpO2 was 82% � 14% for the T-piece group and78% � 20% for the self-inflating bag group (P = .004). TheFiO2 at that precise time was not registered.

After observing a trend toward an increased incidence ofBPD in the self-inflating bag group, we performed an explor-atory analysis of the main outcomes in the subgroup ofVLBW infants. We found that 88% of VLBW infants in theT-piece group and 76% of VLBW infants in the self-inflating bag group achieved an HR $100 at 2 minutes(P = .04). In addition, 25% of the T-piece group and 40%of the self-inflating bag group developed BPD (P = .04)(Table III).

We performed a subgroup analysis of the centers random-ized to the self-inflating bag with a PEEP valve and those as-

Table III. Analysis in the subgroup of VLBW infants

Outcome measureT-piece gro

(n = 85)

HR $100 bpm at 2 min, n (%) 75 (88.2)Intubation for ventilatory support, n (%) 45 (52.9)Drugs/chest compressions, n (%) 3 (3.5)Mechanical ventilation, n (%) 62 (72.9)BPD, n (%) 21 (24.7)Air leaks (pneumothorax and/or neumomediastinum), n (%) 3 (3.5)Use of oxygen, n (%) 71 (83)Days on oxygen, mean � SD 21 � 20

*Corrected by center.

4

signed to the use of self-inflating bag without a PEEP valveand found results comparable to those from our analysis ofthe whole cohort (Table IV). The distribution of therecruited subjects by center is presented in Table V(available at www.jpeds.com).

Discussion

In this large, pragmatic, multicenter study, there was no dif-ference between the T-piece resuscitator and a self-inflatingbag for administering PPV with a face mask to achieve anHR of $100 bpm at birth in newborns at $26 weeks GA.However, the use of a T-piece significantly decreased theintubation rate and the maximum pressure applied, as wellas the variability of maximum pressure.In this trial, most infants responded appropriately with

either device, consistent with previous studies.2 This responsehas been attributed to the Head paradoxical reflex, in whichinflation of the lung triggers an inspiratory effort.19 However,a higher proportion of infants (both term and preterm

up SIB group(n = 110) OR (95% CI)* P value*

84 (76.4) 0.43 (0.19-0.95) .03776 (69.1) 2.01 (1.12-3.60) .0195 (4.6) 1.30 (0.30-5.61) .72385 (77.3) 1.26 (0.66-2.43) .48744 (40.0) 2.03 (1.09-3.79) .0362 (1.8) 0.51 (0.08-3.1) .461

101 (92) 2.2 (0.9-5.5) .08235 � 27 - .0007

Szyld et al

Table IV. Subgroup analysis comparing the patients from centers using SIB with PEEP and centers using SIB withoutPEEP vs T-piece

Outcome measureT-piece(n = 226)

SIB without PEEP valve(n = 226) P value

T-piece(n = 285)

SIB with PEEP valve(n = 290) P value

Outcomes in the delivery roomHR $100 bpm at 2 min, n (%) 218 (96) 210 (93) .099 261 (92) 256 (88) .192Intubation for ventilatory support in the delivery room, n (%) 31 (14) 53 (23) .008 55 (19) 81 (28) .0121-min Apgar score #3, n (%) 51 (23) 66 (29) .092 102 (36) 111 (39) .3405-min Apgar score #5, n (%) 12 (5) 23 (10) .055 18 (6) 24 (8) .364Drugs/chest compressions, n (%) 2 (1) 6 (3) .175 6 (2) 11 (4) .239Time to spontaneous breathing, min, mean � SD† 2.7 � 4.4 2.9 � 3.7 .668 2.7 � 2.8 3.2 � 4.1 .114Time elapsed until HR $100 bpm, min, median (IQR) 1 (0.5-2) 1 (0.5-1.9) .109 1 (0.5-1.5) 1 (0.5-1.8) .247Maximum PIP, mean � SDz variability 25.3 � 1.2 27.3 � 3.9 <.001 25.8 � 2.3 28.7 � 5.5 <.001PIP >25 cm H2O, n (%) 11 (5) 73 (33) <.001 41 (15) 111 (40) <.001Maximum FiO2 in delivery room, mean � SD 0.47 � 0.2 0.53 � 0.2 .005 0.46 � 0.2 0.48 � 0.2 .118

Outcomes after the delivery roomMortality, n (%) 7 (3) 7 (3) .999 4 (1) 8 (3) .265Air leaks (pneumothorax and/or pneumomediastinum), n (%) 6 (3) 2 (1) .175 7 (2) 6 (2) .759Mechanical ventilation, n (%) 37 (16) 43 (19) .449 79 (28) 104 (36) .028Days on mechanical ventilation, mean � SD 6 � 5 12 � 17 .014 5 � 8 7 � 11 .148Days on CPAP, mean � SD 41 � 22 51 � 31 .030 88 � 49 86 � 37 .161Hypoxic ischemic encephalopathy, n (%) 8 (4) 7 (3) .793 13 (5) 21 (7) .169Use of oxygen, n (%) 93 (41) 96 (42) .079 115 (40) 126 (43) .321Days on oxygen, mean � SD 13 � 15 23 � 22 .014 14 � 19 23 � 27 .012Birth weight <1500 g, n (%) 30 (13) 41 (18) .146 55 (19) 69 (24) .180BPD, n (%) 9 (30) 19 (46) .153 12 (22) 25 (36) .090

*Corrected by center and birth weight <1500 g.†990 subjects.z1007 subjects.

- 2014 ORIGINAL ARTICLES

infants) in the self-inflating bag group was intubated for res-piratory support. Animal studies have demonstrated that theuse of PEEP during resuscitation reduces the time to estab-lishment of a functional residual capacity and improves gasdistribution in the lungs.20,21 Our findings may be partiallyexplained by the fact that a T-piece provides PEEP moreeffectively than a self-inflating bag with a PEEP valve,enhancing the clearance of fluid from the lungs and establish-ment of an effective functional residual capacity.22,23 Analysisof the VLBW infant subgroup revealed a higher incidence ofBPD in the self-inflating bag group.

Recognizing the limitations of the subgroup analysis,24 weconsider that some other differences observed between thegroups could contribute to these findings. BPD is a complexdisease associated with volutrauma and atelectrauma, amongother factors.25-28 In the self-inflating bag group, the averagemaximum ventilation pressure applied to infants was signif-icantly higher and highly variable; in addition, a greater pro-portion of these infants received a pressure >25 cm of H2O.These results are consistent with the findings of Dawsonet al12 and with those reported by us and others in studiesin mannequins.8,10,11 Furthermore, inconsistent PEEP orthe use of a self-inflating bag without PEEP also could havecontributed to a higher incidence of BPD in this subgroupof VLBW infants. Finally, the mean number of days on oxy-gen and days on mechanical ventilation were greater in theself-inflating bag group.

These results should be interpreted with caution becausethis was a secondary outcome with an insufficient samplesize to allow us to draw conclusions. Future studies targetedto this subgroup of vulnerable infants may help better eluci-date these findings.

Comparison of Devices for Newborn Ventilation in the Delivery R

Our study has several limitations. The trial could not beblinded because its design precludedmasking of the interven-tion and the outcome evaluation. In addition, actual PEEP,lung volumes, tidal volumes, and leaks around the maskwere not measured as recommended by some authors,because this trial tested interventions in routine clinical prac-tice.29,30 Moreover, use of the self-inflating bag with orwithout a PEEP valve was not crossed over among centers,making the analysis of its effect more difficult and less reli-able. Also, we standardized the management only of theinitial minutes after birth, and thus some secondary out-comes could have been influenced by the different modalitiesof care provided by each center. Finally, we did not stratify bybirth weight, leading to differing proportions of infants withbirth weight <1500 g among the centers. However, this po-tential source of bias is compensated for, at least in part, bythe nature of the crossover design, in which each centerincluded a similar number of subjects per group in consecu-tive periods providing the same standard of care.In summary, we found no difference between the T-piece

resuscitator and a self-inflating bag in administering PPVwith a face mask to achieve an HR of $100 bpm at birth innewborns at $26 weeks GA. However, the use of a T-piecesignificantly decreased the intubation rate and the maximumpressure applied, as well as the variability of maximum pres-sure. Our data support the use of T-piece resuscitators in thedelivery room when compressed gases are available. n

The authors gratefully acknowledge and thank all the patient’s fam-ilies, the Chairs, and the staff of the participating centers. We are alsograteful to Susana Rodriguez, MD, Eduardo Bancalari, MD (holdspatent and license agreements with CareFusion), Fernando

oom 5

THE JOURNAL OF PEDIATRICS � www.jpeds.com Vol. -, No. -

Rubinstein, MD, and Cecilia Rabasa, MD, for their participation asmembers of the Data and Safety Monitoring Committee, and toEduardo Bergel, MD, Santiago Perez Lloret, MD, and Leandro Kova-levsky, MSc, for statistical advice.

Submitted for publication Jul 30, 2013; last revision received Feb 3, 2014;

accepted Feb 18, 2014.

Reprint requests: Edgardo Szyld, MD, MSc, Fundaci�on para la Salud Materno

Infantil, Gavilan 1086, 1406 Buenos Aires, Argentina. E-mail: eszyld@gmail.com

References

1. Barber CA, Wyckoff MH. Use and efficacy of endotracheal versus intra-

venous epinephrine during neonatal cardiopulmonary resuscitation in

the delivery room. Pediatrics 2006;118:1028-34.

2. Ersdal HL, Mduma E, Svensen E, Perlman JM. Early initiation of basic

resuscitation interventions including face mask ventilation may reduce

birth asphyxia related mortality in low-income countries: a prospective

descriptive observational study. Resuscitation 2012;83:869-73.

3. Kattwinkel J, Perlman JM, Aziz K, Colby C, Fairchild K, Gallagher J, et al.

Neonatal resuscitation: 2010 American Heart Association Guidelines for

Cardiopulmonary Resuscitation and Emergency Cardiovascular Care.

Pediatrics 2010;126:e1400-13.

4. Perlman JM, Wyllie J, Kattwinkel J, Atkins DL, Chameides L,

Goldsmith JP, et al. Part 11: neonatal resuscitation: 2010 International

Consensus on Cardiopulmonary Resuscitation and Emergency Cardio-

vascular Care Science with Treatment Recommendations. Circulation

2010;122:S516-38.

5. O’Donnell CP, Davis PG, Lau R, Dargaville PA, Doyle LW, Morley CJ.

Neonatal resuscitation 2: an evaluation of manual ventilation devices

and face masks. Arch Dis Child Fetal Neonatal Ed 2005;90:F392-6.

6. Leone TA, Rich W, Finer NN. A survey of delivery room resuscitation

practices in the United States. Pediatrics 2006;117:e164-75.

7. International Liaison Committee on Resuscitation. The International

Liaison Committee on Resuscitation (ILCOR) consensus on science

with treatment recommendations for pediatric and neonatal patients:

neonatal resuscitation. Pediatrics 2006;117:e978-88.

8. Szyld EG, Aguilar AM, Musante GA, Vain NE, Guerrero MN, Serra ME,

et al. Newborn ventilation: comparison between a T-piece resuscitator

and self-infating bags in a neonatal preterm simulator. Arch Argent Pe-

diatr 2012;110:106-12.

9. Bennett S, Finer NN, Rich W, Vaucher Y. A comparison of three

neonatal resuscitation devices. Resuscitation 2005;67:113-8.

10. Finer NN, Rich W, Craft A, Henderson C. Comparison of methods of

bag and mask ventilation for neonatal resuscitation. Resuscitation

2001;49:299-305.

11. Oddie S, Wyllie J, Scally A. Use of self-inflating bags for neonatal resus-

citation. Resuscitation 2005;67:109-12.

12. Dawson JA, Schmolzer GM, Kamlin CO, Te Pas AB, O’Donnell CP,

Donath SM, et al. Oxygenation with T-piece versus self-inflating bag

for ventilation of extremely preterm infants at birth: a randomized

controlled trial. J Pediatr 2011;158:912-8.e1-2.

13. Kattwinkel J, Short J, eds. Neonatal resuscitation. 5th ed. Elk Grove

Village (IL): American Academy of Pediatrics and American Heart Asso-

ciation; 2006.

6

14. Kattwinkel J, ed. Neonatal resuscitation. 6th ed. Elk Grove Village (IL):

American Academy of Pediatrics and American Heart Association;

2011.

15. Yam CH, Dawson JA, Schmolzer GM, Morley CJ, Davis PG. Heart rate

changes during resuscitation of newly born infants <30 weeks gesta-

tion: an observational study. Arch Dis Child Fetal Neonatal Ed 2011;

96:F102-7.

16. Saugstad OD, Rootwelt T, Aalen O. Resuscitation of asphyxiated

newborn infants with room air or oxygen: an international controlled

trial. The Resair 2 study. Pediatrics 1998;102:e1.

17. Dawson JA, Kamlin CO, Wong C, te Pas AB, Vento M, Cole TJ, et al.

Changes in heart rate in the first minutes after birth. Arch Dis Child Fetal

Neonatal Ed 2010;95:F177-81.

18. Hughes J, Goldenberg RL, Wilfert CM, Valentine M, Mwinga KG, Guay

LA, et al. Design of theHIV Prevention Trials Network (HPTN) protocol

054: a cluster randomized crossover trial to evaluate combined access to

Nevirapine in developing countries. Available from: http://www.bepress.

com/uwbiostat/paper195. Accessed February 1, 2014.

19. Wyllie J. Applied physiology of newborn resuscitation. Curr Paediatr

2005;15:143-50.

20. Jobe AH,HillmanN, Polglase G, Kramer BW, Kallapur S, Pillow J. Injury

and inflammation from resuscitation of the preterm infant. Neonatology

2008;94:190-6.

21. Polglase GR, Morley CJ, Crossley KJ, Dargaville P, Harding R,

Morgan DL, et al. Positive end-expiratory pressure differentially alters

pulmonary hemodynamics and oxygenation in ventilated, very prema-

ture lambs. J Appl Physiol 2005;99:1453-61.

22. Kelm M, Proquitte H, Schmalisch G, Roehr CC. Reliability of two com-

mon PEEP-generating devices used in neonatal resuscitation. Klin Pa-

diatr 2009;221:415-8.

23. Morley CJ, Dawson JA, Stewart MJ, Hussain F, Davis PG. The effect of a

PEEP valve on a Laerdal neonatal self-inflating resuscitation bag. J Pae-

diatr Child Health 2010;46:51-6.

24. Wang R, Lagakos SW, Ware JH, Hunter DJ, Drazen JM. Statistics in

medicine: reporting of subgroup analyses in clinical trials. N Engl J

Med 2007;357:2189-94.

25. Miller JD, Carlo WA. Pulmonary complications of mechanical ventila-

tion in neonates. Clin Perinatol 2008;35:273-81.

26. Auten RL, Vozzelli M, Clark RH. Volutrauma: what is it, and how do we

avoid it? Clin Perinatol 2001;28:505-15.

27. Hillman NH, Nitsos I, Berry C, Pillow JJ, Kallapur SG, Jobe AH. Positive

end-expiratory pressure and surfactant decrease lung injury during initi-

ation of ventilation in fetal sheep. Am J Physiol Lung Cell Mol Physiol

2011;301:L712-20.

28. Hillman NH, Moss TJ, Kallapur SG, Bachurski C, Pillow JJ,

Polglase GR, et al. Brief, large tidal volume ventilation initiates lung

injury and a systemic response in fetal sheep. Am J Respir Crit Care

Med 2007;176:575-81.

29. Schmolzer GM, Kamlin OC, O’Donnell CP, Dawson JA, Morley CJ,

Davis PG. Assessment of tidal volume and gas leak during mask ventila-

tion of preterm infants in the delivery room. Arch Dis Child Fetal

Neonatal Ed 2010;95:F393-7.

30. Kattwinkel J, Stewart C, Walsh B, GurkaM, Paget-Brown A. Responding

to compliance changes in a lung model during manual ventilation:

perhaps volume, rather than pressure, should be displayed. Pediatrics

2009;123:e465-70.

Szyld et al

Appendix

Members of the Delivery Room Ventilation Devices TrialGroup include:

Argentina–Daniela Satragno, MD (Sanatorio de los Arcos,Ciudad de Buenos Aires); Etelvina Soria, MD (Hospital Ma-terno Infantil de Salta, Salta); Luis Ahumada, MD,Mirta Fer-reyra, MD, and Adriana Mitrano, MD) HospitalMisericordia, C�ordoba); Mar�ıa Carola Capelli, MD, and Ce-cilia Cocucci, MD (Hospital Universitario Austral, Pcia deBuenos Aires).

Chile–Paulina Toso, MD, and Miriam Faunes, RN (Uni-versidad Cat�olica, Santiago); Viviana Veas Y�a~nez, MD (Hos-pital Dr. Gustavo Fricke, Vi~na del Mar); Andr�es Rom�anNavarro, MD (Hospital Dr. Hern�an Henr�ıquez Aravena, Te-muco).

Peru–Tania Paredes Quiliche, MD, Wilber G�omez, MD,Mar�ıa Mur, MD, and Juana Molina, RN (Instituto NacionalMaterno Perinatal, Lima).

Italy–Daniele Trevisanuto, MD, Nicoletta Doglioni, MD(Universit�a di Padova, Azienda Ospedaliera di Padova).

US–Beau J. Batton, MD, Kara Weigler, RN (St. John’sChildren’s Hospital, Springfield, Illinois); Ricardo Rodri-guez, MD, Firas Saker, MD, Lory Lewis, CNP, and John Dick-son, RRT-NPS (Hillcrest Hospital, Cleveland Clinic,Cleveland, OH).

- 2014 ORIGINAL ARTICLES

Comparison of Devices for Newborn Ventilation in the Delivery Room 6.e1

Figure 1. Flow diagram of the progress of patients through the study. SIB, self-inflating bag.

Figure 2. Distribution of birth weight <1500 g by center and period (n = 195; overall 19%).

THE JOURNAL OF PEDIATRICS � www.jpeds.com Vol. -, No. -

6.e2 Szyld et al

Table V. Distribution of enrolled subjects by center(n = 1032)

Center

Number of subjects per center

Period 1 Period 2 Total

1 50 50 1002 51 50 1013 51 50 1014 54 51 1055 53 50 1036 50 51 1017 51 50 1018 34 34 689 51 50 10110 51 50 10111 25 25 50

Figure 3. Proportion of infants with an HR$100 at 2minutes.Distribution by center, differentiating those assigned to theself-inflating bag group with a PEEP valve or without a PEEPvalve.

- 2014 ORIGINAL ARTICLES

Comparison of Devices for Newborn Ventilation in the Delivery Room 6.e3