BACKGROUND:

International guidelines for resuscitation recommend using positive end-expiratory pressure (PEEP) during ventilation of preterm newborns. Reliable PEEP-valves for self-inflating bags have been lacking, and effects of PEEP during resuscitation of term newborns are insufficiently studied. The objective was to determine if adding a new PEEP valve to the bag-mask during resuscitation of term and near-term newborns could improve heart rate response.

METHODS:

This randomized controlled trial was performed at Haydom Lutheran Hospital in Tanzania (September 2016 to June 2018). Helping Babies Breathe–trained midwives performed newborn resuscitation using self-inflating bags with or without a new, integrated PEEP valve. All live-born newborns who received bag-mask ventilation at birth were eligible. Heart rate response measured by ECG was the primary outcome, and clinical outcome and ventilation data were recorded.

RESULTS:

Among 417 included newborns (median birth weight 3200 g), 206 were ventilated without and 211 with PEEP. We found no difference in heart rate response. Median (interquartile range) measured PEEP in the PEEP group was 4.7 (2.0–5.6) millibar. The PEEP group received lower tidal volumes (4.9 [1.9–8.2] vs 6.3 [3.9–10.5] mL/kg; P = .02) and had borderline lower expired CO2 (2.9 [1.5–4.3] vs 3.3 [1.9–5.0] %; P = .05). Twenty four-hour mortality was 9% in both groups.

CONCLUSIONS:

We found no evidence for improved heart rate response during bag-mask ventilation with PEEP compared with no PEEP. The PEEP valve delivered a median PEEP within the intended range. The findings do not support routine use of PEEP during resuscitation of newborns around term.

What’s Known on This Subject:

Positive end-expiratory pressure (PEEP) may facilitate lung liquid clearance and help establish functional residual capacity, and is recommended in international guidelines for resuscitation of preterm newborns. PEEP is also commonly used for newborns around term, but evidence for beneficial effects is sparse.

What This Study Adds:

Term and near-term newborns who received bag-mask ventilation with PEEP had no better heart rate response or survival than newborns ventilated without PEEP. Adding a PEEP valve to the bag-mask increased leak, and reduced tidal volumes and expired CO2.

At birth, newborn lungs are liquid-filled, and a successful lung aeration is necessary to establish effective pulmonary gas exchange. Between 3% and 8% of newborns receive positive pressure ventilation (PPV), and the needs are highest where antenatal care is poor.15  Each year, ∼700 000 newborns die of intrapartum events, 99% in low-income countries.6  Positive end-expiratory pressure (PEEP) during PPV may facilitate lung liquid clearance and help establish functional residual capacity.7  Results from animal studies and studies in premature newborns indicate several beneficial effects: PEEP improves oxygenation, reduces alveolar collapse, lowers expiratory resistance and increases lung volume, surface area, and compliance.810  In 2010, the International Liaison Committee on Resuscitation stated that PEEP is likely to be beneficial during newborn resuscitation and should be used if suitable equipment is available.11  In the 2015 revision, the committee concluded that for term newborns, no recommendation could be given because of insufficient data.4  The report was, however, ambiguous at this point, because it still suggested to use PEEP “when the facilities and equipment permit it to be given reliably.” T-piece resuscitators may deliver PEEP if compressed air is available but are prone to user errors in unexperienced hands, and unintentional variation in pressures have been observed.1218  PEEP valves for self-inflating bags (SIBs) exist, but serious concerns have been raised about the ability of these to deliver reliable PEEP.1922  Researchers of 4 randomized controlled trials (RCTs) have investigated PEEP delivered by different devices at birth.2326  Two studies included term in addition to preterm newborns but did not report the effects for term newborns alone.23,25  Despite low level of evidence from clinical trials on term newborns, use of PEEP is common during resuscitation.2731 

The objective with this RCT was to study if adding a new PEEP valve32  to the bag-mask during resuscitation of term and near-term newborns could improve heart rate (HR) response. Secondary outcomes included ventilation parameters and 24-hours mortality.

This study was a nonblinded RCT performed at Haydom Lutheran Hospital, a rural Tanzanian referral hospital with ∼4000 deliveries annually. The study was conducted by Safer Births, a research consortium on labor surveillance and newborn resuscitation in low-income settings.3,33,34 

All live-born bag-mask ventilated–newborns without major deformities were eligible for inclusion. Midwives approached mothers for informed oral consent at admission for labor. If predelivery consent was considered inappropriate, research nurses sought deferred consent the day after delivery.

Laerdal Upright Resuscitators (320 mL; Laerdal Medicals, Stavanger, Norway) with (catalog no. 846060) or without PEEP valve (catalog no. 846050) were used for newborn resuscitation. Except for the PEEP valve (Supplemental Fig 4), the 2 bags were identical. The first and last author made a randomization schedule using www.randomizer.org to assign choice of SIB per week in the study period. Research nurses changed all SIBs weekly so that only the SIB according to randomization was placed at the resuscitation tables. Randomization per patient was not feasible in this setting in which mothers frequently arrive in late stage labor. Blinding was impossible because the presence of PEEP valve was visible on the bags.

The local procedure for newborn resuscitation followed the Helping Babies Breathe (HBB) guidelines, introduced at the study site in 2009, emphasizing stimulation and early initiation of bag-mask ventilation (BMV).35,36  Newborn resuscitation was mainly the responsibility of midwives. Cord clamping was done before BMV in all cases. Before and during the study, all midwives participated in 3 one-day HBB-training courses including demonstration of the PEEP valve. The midwives were encouraged to train in ventilation skills frequently.36  There were 6 labor rooms, each equipped with a resuscitation table. After cesarean delivery, newborn resuscitation was performed in a room adjacent to the operating theater. After resuscitation, the midwives decided on the basis of the clinical condition whether to keep the newborn with the mother or transfer to the neonatal ward. Respiratory support available in the neonatal ward was supplemental oxygen by nasal cannula; CPAP was unavailable in most of the study period and no ventilators were used.

Laerdal Newborn Resuscitation Monitors available at all resuscitation tables automatically started data collection when bag-mask or ECG-sensor were used. They had dry-electrode ECG sensors to measure HR and sensors for flow (Acutronic Medical Systems, Hirzel, Switzerland), pressure (MPX5010 sensor; Freescale Semiconductor, Austin, TX), and side stream CO2 (ISA; Masimo, Irvine, XA) connected between the bag and the mask; the attachment device added a dead space of 1 mL.37  The monitors provided HR feedback during resuscitation but did not display ventilation parameters. Trained nonmedical research assistants documented time intervals from birth and recorded observations during resuscitation, perinatal information, and outcomes.33 

Primary outcome per protocol was change in HR per ventilation sequence. We defined a ventilation sequence as continuous BMV ≥5 seconds with <5 seconds pause between ventilations. We analyzed up to 60 seconds per sequence and a maximum of 5 sequences within the first 10 minutes of BMV per newborn. To account for potential differences within groups, we performed stratified analyses for newborns with first detected HR <100 vs ≥100 beats per minute and birth weight (BW) <2500 vs ≥2500 g. Because of imprecise estimation of gestational age (GA) in the study population, we used BW ≥2500 g as a proxy for term and near-term newborns. For sensitivity analyses of the primary end point, we compared mean HR in the first 1 and 5 minutes of BMV and time from first ventilation until HR ≥100, ≥120, and ≥140 beats per minute in the 2 groups. We also studied HR at 2 minutes age and proportion of newborns with HR >100 beats per minute at 2 minutes age.23 

Predefined secondary outcomes were ventilation factors including PEEP, mean airway pressure, peak inflation pressure (PIP), expiratory tidal volumes (VTEs), expired carbon dioxide (ECO2) (mean ECO2 in first minute of BMV and time to ECO2 >2% and >4%) and leak. Clinical outcomes included Apgar scores at 1 and 5 minutes, total duration of BMV and death or admission to the neonatal unit at 24 hours.

Sample size calculation was based on HR changes found in a previous study.38  To detect a clinical relevant difference of 25% with a 2-sided significance level of 0.05 and power 0.80, we needed a minimum of 882 ventilation sequences from 300 newborns. Accounting for 10% missing data, we aimed to enroll a minimum of 330 newborns, 165 in each group. Because of frequently missing HR data at the start of BMV, we expanded the data collection period to 4 months after reaching the target of 330 included newborns. Power calculation was done for primary outcomes, and secondary analyses may be considered as exploratory.

For the primary analysis of change in HR, we fitted a 3-level linear random effects model with robust SEs taking repeated observations per newborn and dependency between ventilation sequences into account and allowing for individual slopes for HR change within ventilation sequences. The analysis was performed with splines for time (3 knots) to optimize model fit. The effect of PEEP was tested as a binary variable.

For baseline characteristics and secondary outcomes, we used Pearson χ2 or Wilcoxon rank tests as appropriate. For outcomes with repeated measurements within each newborn, we used linear random effects models. For variables showing systematic increase or decrease, we graphed predicted marginal values by time.

To evaluate reliability of PEEP, we performed post hoc analyses within newborns ventilated with PEEP valve to quantify ventilations with measured PEEP within the desired range of 4 to 8 millibar (mbar). Because ventilations against blocked airway or with high leak will not be effective with any equipment,3941  we repeated the quantification excluding ventilations with inspired volumes <2 mL or leak >80%.42  We also compared groups by median PEEP < vs ≥ 4 and 8 mbar to assess clinical characteristics associated with low and high PEEP.

Initial data processing was done by using Matlab (MathWorks, Natick, MA). The data were analyzed by using Stata version 16 (StataCorp 2019, College Station, TX). A 2-sided P < .05 defined statistical significance.

Upright bag and PEEP valves were thoroughly tested and approved for medical use before study start.32,34,43  In weeks assigned to PEEP during the study period, research nurses inspected the PEEP valves and tested all bags daily with a manometer to ensure that delivered PEEP kept unchanged within 4 to 8 mbar. All PEEP valves worked as intended and none needed replacement during the study period. An External Data and Safety Monitoring Committee conducted 2 preplanned interim analyses with stopping rules for the clinical end point death at 24 hours after enrollment of ∼100 and 200 newborns.

All mothers of included newborns gave informed oral consent either before or after delivery. Ethical approval was granted by the National Institute for Medical Research in Tanzania (Ref. NIMR/HQ/R.8a/Vol.IX/1434 and NIMR/HQ/R.8c/Vol.I/325) and the Regional Committee for Medical and Health Research Ethics for Western Norway (Ref. 2013/110).

Among 6012 live-born newborns between September 26, 2016, and June 30, 2018, 473 (7.9%) received BMV (Fig 1). After exclusions due to missing consent or wrong bag used according to randomization, 417 newborns were included; 206 received BMV without and 211 with PEEP. Baseline characteristic were mainly similar in the 2 groups, with slightly lower median BW in the PEEP group (Table 1).

FIGURE 1

Flowchart. Consent was obtained at admission in most cases; 80 mothers were approached for deferred consent, among whom none refused. Among 365 newborns with both ventilation parameters and HR data, 308 had available HR data within the first 10 seconds of BMV.

FIGURE 1

Flowchart. Consent was obtained at admission in most cases; 80 mothers were approached for deferred consent, among whom none refused. Among 365 newborns with both ventilation parameters and HR data, 308 had available HR data within the first 10 seconds of BMV.

Close modal
TABLE 1

Perinatal Characteristics of the Enrolled Newborns

No PEEPPEEPPa
nValuesNValues
Cesarean delivery, n (%) 206 93 (45) 211 92 (44) .75 
Birth wt, g, median (IQR) 205 3250 (2927–3600) 211 3150 (2590–3500) .02 
Birth wt <2500 g, n (%) 205 24 (12) 211 42 (20) .02 
GA, wk, median (IQR) 194 38 (37–40) 194 39 (37–40) .88 
Female sex, n (%) 206 94 (46) 211 84 (40) .23 
Multiple pregnancies, n (%)b 206 13 (6) 211 16 (8) .76 
Maternal preeclampsia or eclampsia, n (%)c 206 0 (0) 211 7 (3) .01 
Labor complications, n (%)d 206 27 (13) 211 26 (12) .81 
Meconium stained amniotic fluid, n (%) 206 95 (46) 211 79 (37) .07 
Time from birth to cord clamp, s, median (IQR) 206 26 (14–50) 206 28 (14–53) .97 
First recorded HR, median (IQR), beats per min 189 104 (65–150) 176 97 (66–151) .83 
Newborns with first recorded HR <100 beats per min, n (%) 189 96 (51) 176 86 (49) .71 
Time from birth to first ventilation, s, median (IQR) 206 100 (74–152) 208 96 (72–154) .44 
No PEEPPEEPPa
nValuesNValues
Cesarean delivery, n (%) 206 93 (45) 211 92 (44) .75 
Birth wt, g, median (IQR) 205 3250 (2927–3600) 211 3150 (2590–3500) .02 
Birth wt <2500 g, n (%) 205 24 (12) 211 42 (20) .02 
GA, wk, median (IQR) 194 38 (37–40) 194 39 (37–40) .88 
Female sex, n (%) 206 94 (46) 211 84 (40) .23 
Multiple pregnancies, n (%)b 206 13 (6) 211 16 (8) .76 
Maternal preeclampsia or eclampsia, n (%)c 206 0 (0) 211 7 (3) .01 
Labor complications, n (%)d 206 27 (13) 211 26 (12) .81 
Meconium stained amniotic fluid, n (%) 206 95 (46) 211 79 (37) .07 
Time from birth to cord clamp, s, median (IQR) 206 26 (14–50) 206 28 (14–53) .97 
First recorded HR, median (IQR), beats per min 189 104 (65–150) 176 97 (66–151) .83 
Newborns with first recorded HR <100 beats per min, n (%) 189 96 (51) 176 86 (49) .71 
Time from birth to first ventilation, s, median (IQR) 206 100 (74–152) 208 96 (72–154) .44 

IQR, interquartile range.

a

χ2 test or Wilcoxon rank sum test as appropriate.

b

One set of quadruplets in the PEEP group, the rest were twins.

c

All but 1 of newborns born by mothers with registered preecplampsia or eclampsia had birth wt <2500 g.

d

Includes number of deliveries with the following complications registered: obstructed labor, uterine rupture, cord prolapse, shoulder dystocia, or bleeding.

HR per time per ventilation sequence was similar, or nonsignificantly lower, in the PEEP group compared with newborns ventilated without PEEP (Fig 2, Supplemental Fig 5A). For newborns with first recorded HR <100 beats per minute, we observed increasing HR during the first 3 ventilation sequences (Supplemental Fig 5B). Among newborns with first HR ≥100 beats per minute, the HR remained nearly unchanged (Supplemental Fig 5C). We found no significant PEEP effect in any of the groups, either in first or later ventilation sequences (Supplemental Fig 5 A–C). HR changes for newborns with BW <2500 vs ≥2500 g were similar (Supplemental Fig 6). Mean HR in the first 1 and 5 minutes of ventilation; time to HR>100, >120, and >140 beats per minute; HR at 2 minutes of age; and the proportion of newborns with HR >100 at 2 minutes of age were similar for newborns ventilated with or without PEEP (Table 2).

FIGURE 2

HR response with or without PEEP valve. Mean HR (with 95% confidence intervals) by time in 902 ventilation sequences. A ventilation sequence was defined as continuous ventilation lasting for ≥5 seconds with <5-second pause between ventilations, ≤5 sequences were included per newborn.

FIGURE 2

HR response with or without PEEP valve. Mean HR (with 95% confidence intervals) by time in 902 ventilation sequences. A ventilation sequence was defined as continuous ventilation lasting for ≥5 seconds with <5-second pause between ventilations, ≤5 sequences were included per newborn.

Close modal
TABLE 2

Sensitivity Analyses of the Primary Outcome and Secondary Outcomes

No PEEPPEEPP
nValuesnValues
Sensitivity analyses for HR      
 HR (beats per min) in the first min after start of BMVa 185 126 (120–131) 175 124 (119–130) .62 
 HR (beats per min) in the first 5 min after start of BMVa 189 147 (142–151) 176 142 (138–147) .19 
 Proportion of newborns with HR ≥100 at 2 min ageb 152 103 (68%) 131 87 (66%) .81 
 HR at 2 min of age (or first registered HR within 30 s after 2 min)c 152 141 (77–169) 131 134 (79–162) .50 
 Time (s) from first ventilation until HR ≥100 beats per minc,d 87 27 (20–47) 89 30 (21–55) .75 
 Time (s) from first ventilation until HR ≥120 beats per minc,e 104 29 (18–53) 95 32 (22–78) .16 
 Time (s) from first ventilation until HR ≥140 beats per minc,f 124 33 (22–73) 114 40 (24–88) .19 
Ventilation parameters      
 PIP, mbarg 191 39 (37–40) 181 39 (37–41) .56 
 Mean inflation pressure, mbarg 191 18 (15–20) 181 20 (18–22) <.001 
 PEEP, mbarg 191 0.1 (0.1–0.2) 181 4.7 (2.1–5.6) <.001 
 VTE, mL/kgg,i 190 6.3 (3.0–10.5) 181 4.9 (1.9–8.2) .02 
 Leak, %g,i 175 40 (18–60) 178 46 (22–64) .06 
 ECO2, % of expired airg,i 191 3.3 (1.9–5.0) 181 2.9 (1.5–4.3) .05 
 Time to ECO2 >2%, sc,j 126 7.2 (3.6–19.5) 139 8.6 (3.6–23.1) .27 
 Time to ECO2 >4%, sc,k 152 7.9 (3.8–27.4) 158 10.5 (3.7–40.5) .29 
Clinical outcomes      
 1-min Apgar score ≤6, No.(%)b 206 92 (45%) 211 85 (40%) .37 
 5-min Apgar score ≤6, No.(%)b 206 44 (21%) 211 37 (18%) .32 
 Time from first to last ventilation, se,h 191 132 (66–277) 181 112 (68–263) .68 
Outcome at 24 h, No. (%)b      
 Normal 205 130 (63%) 211 137 (65%) .92 
 Admitted — 57 (28%) — 55 (26%) — 
 Dead — 18 (9%) — 19 (9%) — 
No PEEPPEEPP
nValuesnValues
Sensitivity analyses for HR      
 HR (beats per min) in the first min after start of BMVa 185 126 (120–131) 175 124 (119–130) .62 
 HR (beats per min) in the first 5 min after start of BMVa 189 147 (142–151) 176 142 (138–147) .19 
 Proportion of newborns with HR ≥100 at 2 min ageb 152 103 (68%) 131 87 (66%) .81 
 HR at 2 min of age (or first registered HR within 30 s after 2 min)c 152 141 (77–169) 131 134 (79–162) .50 
 Time (s) from first ventilation until HR ≥100 beats per minc,d 87 27 (20–47) 89 30 (21–55) .75 
 Time (s) from first ventilation until HR ≥120 beats per minc,e 104 29 (18–53) 95 32 (22–78) .16 
 Time (s) from first ventilation until HR ≥140 beats per minc,f 124 33 (22–73) 114 40 (24–88) .19 
Ventilation parameters      
 PIP, mbarg 191 39 (37–40) 181 39 (37–41) .56 
 Mean inflation pressure, mbarg 191 18 (15–20) 181 20 (18–22) <.001 
 PEEP, mbarg 191 0.1 (0.1–0.2) 181 4.7 (2.1–5.6) <.001 
 VTE, mL/kgg,i 190 6.3 (3.0–10.5) 181 4.9 (1.9–8.2) .02 
 Leak, %g,i 175 40 (18–60) 178 46 (22–64) .06 
 ECO2, % of expired airg,i 191 3.3 (1.9–5.0) 181 2.9 (1.5–4.3) .05 
 Time to ECO2 >2%, sc,j 126 7.2 (3.6–19.5) 139 8.6 (3.6–23.1) .27 
 Time to ECO2 >4%, sc,k 152 7.9 (3.8–27.4) 158 10.5 (3.7–40.5) .29 
Clinical outcomes      
 1-min Apgar score ≤6, No.(%)b 206 92 (45%) 211 85 (40%) .37 
 5-min Apgar score ≤6, No.(%)b 206 44 (21%) 211 37 (18%) .32 
 Time from first to last ventilation, se,h 191 132 (66–277) 181 112 (68–263) .68 
Outcome at 24 h, No. (%)b      
 Normal 205 130 (63%) 211 137 (65%) .92 
 Admitted — 57 (28%) — 55 (26%) — 
 Dead — 18 (9%) — 19 (9%) — 

IQR, interquartile range. —, not applicable.

a

Linear random intercept model adjusted for repeated observations per newborn, results given as mean (95% confidence interval).

b

Pearson χ2 test, results given as n (%).

c

Wilcoxon rank test, results given as median (IQR).

d

Time from start BMV to HR ≥100: only newborns with registered HR <100 after start of BMV were included; 82 of 87 in the no-PEEP and 83 and 89 in the PEEP group reached HR ≥100. Cox regression, hazard ratio, 0.96 (0.71–1.30), P = .80.

e

Time from start BMV to HR ≥ 120: only newborns with registered HR <120 after start of BMV were included. Ninety-eight of 105 in the no-PEEP and 86 of 95 in the PEEP group reached HR ≥120. Cox regression: hazard ratio, 0.81 (0.61–1.09), P = .16.

f

Time from start BMV to HR ≥ 140: only newborns with registered HR <140 after start BMV included. One hundred eleven of 124 in the no-PEEP and 99 of 114 in the PEEP group reached HR ≥140. Cox regression: hazard ratio, 0.85 (0.65–1.11), P = .23.

g

Wilcoxon rank test on medians per newborn for all ventilations within first 10 min of BMV per newborn; results are given as median per medians (IQR).

h

Cox regression for time from first to last ventilation: hazard ratio, 1.12 (0.91–1.37), P = .30 for PEEP compared with no PEEP.

i

Measures increased or decreased by time. Means with 95% confidence intervals by time analyzed in linear random effects model are displayed in Fig 3A–C.

j

Only newborns with registered ECO2 <2% after start of BMV are included. One hundred twenty-two of 126 newborns in the no-PEEP and 136 of 139 in the PEEP group reached ECO2 >2% within 5 min. Cox regression for time to CO2 >2%: hazard ratio, 0.93 (0.73–1.18), P = .54.

k

Only newborns with registered ECO2 <4% after start of BMV are included. One hundred forty-one of 152 newborns in the no-PEEP and 144 of 158 in the PEEP group reached ECO2 >4% within 5 min. Cox regression for time to CO2 > 4%: hazard ratio, 0.88 (0.70–1.12), P = .30.

Newborns ventilated with PEEP valve received a median PEEP of 4.7 mbar per ventilation (Table 2). Median PIP during first 10 minutes of BMV was similar, but mean airway pressure was higher in the PEEP group. Means for VTE and ECO2 increased by time, and leak decreased (Fig 3 A–C). The PEEP group received significantly lower VTE than newborns ventilated without PEEP (Table 2). ECO2 was lower and leak was higher in the PEEP group, but the differences were not significant. Median ventilation time per ventilation sequence was 26 seconds with no significant differences between groups (Supplemental Fig 5). Apgar scores at 1 and 5 minutes and total duration of BMV were similar. We found no differences in morbidity or mortality at 24 hours. Secondary outcomes were similar for newborns with BW <2500 vs ≥2500 g (Supplemental Table 3).

FIGURE 3

VTE, ECO2, and leak by time. Means for secondary outcome parameters (with 95% confidence intervals) by time in the first 60 seconds of BMV. A, VTE. B, ECO2. C, Leak.

FIGURE 3

VTE, ECO2, and leak by time. Means for secondary outcome parameters (with 95% confidence intervals) by time in the first 60 seconds of BMV. A, VTE. B, ECO2. C, Leak.

Close modal

Adjusting for differences in VTE by adding VTE as a covariate shifted HR and ECO2 upwards in the PEEP group compared with no PEEP (Supplemental Fig 7), but still with no significant differences between the groups. Adjusting for BW, which was by chance slightly different between groups (Table 1), in the primary analysis did not change the results (data not shown).

Among newborns ventilated with PEEP, 52% of ventilations in first 10 minutes of BMV had a PEEP within the desired range of 4 to 8 mbar. Excluding ventilations with likely obstructed airway (inspired volume <2 mL or ≥80% leak), the proportion with desired PEEP increased to 62%. Supplemental Fig 8 shows distribution of measured PEEP and medians per newborn. Newborns who received a median PEEP <4 mbar had significantly higher Apgar scores and first recorded HR compared with newborns who received ≥4 mbar (Supplemental Table 4).

Only 2 newborns had median PEEP ≥8 mbar.

In this RCT of BMV with or without PEEP during resuscitation of mainly term and near-term newborns, we found no significant differences in HR response to ventilation. The median PEEP in the PEEP group was within the target of 4 to 8 mbar. Clinical outcomes were similar between the groups.

HR recovery indicates effective ventilation during resuscitation.4  ECO2 is closely related to end-inflation lung volumes in the immediate newborn period.44  Before study start, we hypothesized that HR and ECO2 would rise faster in newborns ventilated with than without PEEP because of improved lung aeration. The direction of observed differences, though not significant, was opposite of expected. A contributing explanation may be the significantly lower VTE in the PEEP group. Results from recent studies indicate that sufficient tidal volumes are critical for HR recovery and ECO2.10,34,4447  Lower VTE measured in the PEEP group could partly be due to accumulation of functional residual capacity and thus less expired air. However, because expected effects on HR and ECO2 were not seen, we speculate that other mechanisms were more important. A reduced pressure difference between PIP and PEEP when adding a PEEP valve to the SIB without increasing PIP reduces the driving force for flow. This will reduce VTE if the lung compliance is unchanged. Other factors, like an increase in mask leak, dilution of CO2 in dead space, reduced venous return and cardiac output in hearts compromised by asphyxia,48  increasing effects of reflex-mechanisms and increased incidence of pneumothorax28  related to a higher mean airway pressure,49  may also have affected HR and ECO2.

Sicker newborns received higher and more reliable PEEP. Sensitivity analyses by levels of measured PEEP would be confounded by differences in clinical condition and were not performed. Higher initial HR and 1-minute Apgar scores among newborns with median PEEP <4 compared with ≥4 mbar may indicate difficulties keeping a good mask seal in more vigorous newborns.

Researchers of most previous human studies of PEEP during newborn resuscitation have focused on preterm newborns.2326,50  Mature lungs are less susceptible to injury, with better compliance, larger volume capacity, and a stiffer chest wall.51,52  Asphyxia is the most common reason why term newborns need PPV,33  and for asphyxiated term newborns a fast reoxygenation of vital organs is likely more crucial for the outcome than lung protective ventilation strategies.3,53  Authors of recent studies indicate an increased risk for pneumothorax with increasing GA using continuous positive airway pressure (CPAP) during newborn stabilization.28,30  Thus, findings from studies of preterm newborns cannot always be extrapolated to term newborns with different underlying physiology, anatomy, and pathologies.

We are aware of only 2 previous RCTs in which researchers assess use of PEEP when ventilating term newborns at birth, and in both studies, T-piece devices were used to deliver PEEP.23,25  Szyld et al23  also included a group ventilated with a SIB with a PEEP valve, but did not report on delivered PEEP. Because T-piece provides CPAP, which may facilitate spontaneous breathing, it may have additional positive effects compared with adding a PEEP valve during BMV. Our finding of similar HR recovery for newborns ventilated with or without PEEP is in line with Szyld et al,23  who found no difference in proportion of newborns with HR>100 beats per minute at 2 minutes of age. Thakur et al25  found a significant lower duration of ventilation for newborns who received PEEP. We could not replicate this. The researchers also found differences in intubation rates and use of oxygen (not available at our study site), but both studies included newborns with GA down to 26 weeks and did not display results for term and near-term newborns separately.

This is the first RCT performed to study clinical effects of ventilations delivered by SIBs with or without PEEP during resuscitation of newborns around term. Except for the PEEP valve, the SIBs were identical in both groups. The continuous accurate signal recording of ECG, ECO2, and ventilation parameters was another major strength in this trial.

The study was performed in a rural setting with long transport and potential delay for complicated deliveries to be assisted, which may partly explain the high frequency of BMV. Antenatal care was insufficient, and information on maternal morbidity is incomplete. Local procedure for newborn resuscitation followed the HBB algorithm,35  in line with International Liaison Committee on Resuscitation recommendations, but does not mention chest compressions, medication, or intubation. More advanced neonatal care and respiratory support, including CPAP, was not available. This could affect outcome for the most compromised newborns in both groups and reduce the generalizability; along with the fact that the study was performed in a single center, we could not randomize individuals, and there was a imprecise estimation of GA and a higher proportion of low BW in the PEEP group. The setting also hindered investigation of more advanced medical interventions.2326,50  We cannot report on the incidence of pneumothorax or neurodevelopmental outcomes due to lack of radiographs and long-term follow-up.50  The rural, low-income study site is, however, representative for where most neonatal deaths occur54  and where a SIB with a PEEP valve is the only feasible alternative to deliver PEEP.4  Approximately half of newborns who received BMV had first detected HR >100 beats per minute; thus the group of newborns with potential for a significant HR increase was smaller than anticipated. The study was not powered to detect small differences in subgroups, the use of multiple ventilation sequences per newborn may have increased the risk for type 2 errors, and false-positive findings in exploratory analyses may have occurred by chance.

Using PEEP during resuscitation is recommended for premature newborns.4  Because the new PEEP valve did deliver PEEP, and no clear harms were detected, the equipment may be useful during ventilation of premature newborns in settings where devices requiring pressurized air are not available.

The lower VTE in the PEEP group likely affected the results of this trial. The optimal VTE during BMV at birth necessary to balance the need to establish an effective gas exchange as soon as possible against the risk for lung-injury is unknown. Further trials with a larger sample size adding chest radiographs to look for air leak, long-term-follow-up to evaluate neurodevelopmental outcome, and, in a different setting, assessment of lung gas volumes, arterial blood gases, cardiac function, and brain perfusion could add useful information.

We found no evidence for improved HR response to BMV with PEEP compared with BMV without PEEP, although delivered PEEP was within the intended range. Clinical outcomes were similar among groups. Our findings do not support routine use of PEEP ∼5 mbar during resuscitation of newborns around term.

We thank the participants, mothers and children, and all midwives, research assistants, research nurses, and other personnel at Haydom Lutheran Hospital for making this study possible. Special thanks to Sara Lyanga, head-midwife at the hospital, for your cooperation, and to Paul Mejan and Domitila Augustino for all your efforts during data collection.

Dr Holte conceptualized and designed the study, coordinated and supervised data collection, conducted the analyses, drafted the initial manuscript, and revised the manuscript; Drs Størdal and Ersdal conceptualized and designed the study, coordinated and supervised data collection, helped analyze the data, and thoroughly reviewed and revised the manuscript; Dr Eilevstjønn and Mr Gomo designed the positive end-expiratory pressure valve and the data collection instruments, gave technical support, extracted and processed data, and reviewed and revised the manuscript; Dr Klingenberg contributed considerably to conceptualization and design of the study, during data analysis, and in the writing process and critically reviewed and revised the manuscript; Drs Thallinger, Linde, and Kidanto took part in the planning of the study and training of local personnel, facilitated data collection, and reviewed and revised the manuscript; Dr Stigum was the study statistician providing supervision and quality control of the statistical analyses and reviewed and revised the manuscript; Ms Yeconia had the daily responsibility for data collection and reviewed and revised the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

This trial has been registered at www.clinicaltrials.gov (identifier NCT 02971553).

Deidentified individual participant data will be made available to researchers whose methodologically sound proposal has been approved by the Scientific Steering Comitee for Safer Births Study Group. Proposals may be submitted up to 36 months following article publication to [email protected].

FUNDING: All phases of this study were supported by Global Health and Vaccines Research program, Norwegian Research Council. Dr Holte was supported by South-Eastern Norway Regional Health Authority, and Drs Thallinger and Linde were supperted by unrestricted grants from Laerdal Foundation for Acute Medicine.

BMV

bag-mask ventilation

BW

birth weight

CPAP

continuous positive airway pressure

GA

gestational age

ECO2

expired carbon dioxide

HR

heart rate

Mbar

millibar

PEEP

positive end-expiratory pressure

PIP

peak inflation pressure

PPV

positive pressure ventilation

RCT

randomized controlled trial

SIB

self-inflating bag

VTE

expiratory tidal volume

1
O’Donnell
CP
,
Davis
PG
,
Morley
CJ
.
Positive pressure ventilation at neonatal resuscitation: review of equipment and international survey of practice
.
Acta Paediatr
.
2004
;
93
(
5
):
583
588
2
Wall
S
,
Lee
ACC
,
Niermeyer
S
, et al
.
Neonatal resuscitation in low-resource settings: What, who, and how to overcome challenges to scale up?
Int J Gynecol Obstet
.
2009
;
107
(
suppl 1
):
S47
S62, S63–S64
3
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
(
7
):
869
873
4
Perlman
JM
,
Wyllie
J
,
Kattwinkel
J
, et al;
Neonatal Resuscitation Chapter Collaborators
.
Part 7: neonatal resuscitation: 2015 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations
.
Circulation
.
2015
;
132
(
16
suppl 1
):
S204
S241
5
World Health Organization
.
Guidelines on Basic Newborn Resuscitation
.
Geneva, Switzerland
:
World Health Organization
;
2012
6
Liu
L
,
Oza
S
,
Hogan
D
, et al
.
Global, regional, and national causes of under-5 mortality in 2000-15: an updated systematic analysis with implications for the Sustainable Development Goals
.
Lancet
.
2016
;
388
(
10063
):
3027
3035
7
Hooper
SB
,
Siew
ML
,
Kitchen
MJ
,
te Pas
AB
.
Establishing functional residual capacity in the non-breathing infant
.
Semin Fetal Neonatal Med
.
2013
;
18
(
6
):
336
343
8
Dinger
J
,
Töpfer
A
,
Schaller
P
,
Schwarze
R
.
Effect of positive end expiratory pressure on functional residual capacity and compliance in surfactant-treated preterm infants
.
J Perinat Med
.
2001
;
29
(
2
):
137
143
9
Siew
ML
,
Te Pas
AB
,
Wallace
MJ
, et al
.
Positive end-expiratory pressure enhances development of a functional residual capacity in preterm rabbits ventilated from birth
.
J Appl Physiol (1985)
.
2009
;
106
(
5
):
1487
1493
10
Wheeler
K
,
Wallace
M
,
Kitchen
M
, et al
.
Establishing lung gas volumes at birth: interaction between positive end-expiratory pressures and tidal volumes in preterm rabbits
.
Pediatr Res
.
2013
;
73
(
6
):
734
741
11
Perlman
JM
,
Wyllie
J
,
Kattwinkel
J
, et al;
Neonatal Resuscitation Chapter Collaborators
.
Part 11: neonatal resuscitation: 2010 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations
.
Circulation
.
2010
;
122
(
16
suppl 2
):
S516
S538
12
Hawkes
CP
,
Oni
OA
,
Dempsey
EM
,
Ryan
CA
.
Potential hazard of the Neopuff T-piece resuscitator in the absence of flow limitation
.
Arch Dis Child Fetal Neonatal Ed
.
2009
;
94
(
6
):
F461
F463
13
Hawkes
CP
,
Oni
OA
,
Dempsey
EM
,
Ryan
CA
.
Should the Neopuff T-piece resuscitator be restricted to frequent users?
Acta Paediatr
.
2010
;
99
(
3
):
452
453
14
Bennett
S
,
Finer
NN
,
Rich
W
,
Vaucher
Y
.
A comparison of three neonatal resuscitation devices
.
Resuscitation
.
2005
;
67
(
1
):
113
118
15
Finer
NN
,
Rich
WD
.
Unintentional variation in positive end expiratory pressure during resuscitation with a T-piece resuscitator
.
Resuscitation
.
2011
;
82
(
6
):
717
719
16
McHale
S
,
Thomas
M
,
Hayden
E
,
Bergin
K
,
McCallion
N
,
Molloy
EJ
.
Variation in inspiratory time and tidal volume with T-piece neonatal resuscitator: association with operator experience and distraction
.
Resuscitation
.
2008
;
79
(
2
):
230
233
17
Hinder
M
,
McEwan
A
,
Drevhammer
T
,
Donaldson
S
,
Tracy
MB
.
T-piece resuscitators: how do they compare?
Arch Dis Child Fetal Neonatal Ed
.
2019
;
104
(
2
):
F122
F127
18
Guinsburg
R
,
Miyoshi
MH
,
de Almeida
MFB
.
T-piece resuscitators: a warning sign
.
Arch Dis Child Fetal Neonatal Ed
.
2019
;
104
(
2
):
F116
F117
19
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 Paediatr Child Health
.
2010
;
46
(
1–2
):
51
56
20
Hartung
JC
,
Wilitzki
S
,
Thio-Lluch
M
,
te Pas
AB
,
Schmalisch
G
,
Roehr
CC
.
Reliability of single-use PEEP-valves attached to self-inflating bags during manual ventilation of neonates--an in vitro study
.
PLoS One
.
2016
;
11
(
2
):
e0150224
21
Thio
M
,
Dawson
JA
,
Crossley
KJ
, et al
.
Delivery of positive end-expiratory pressure to preterm lambs using common resuscitation devices
.
Arch Dis Child Fetal Neonatal Ed
.
2019
;
104
(
1
):
F83
F88
22
Kelm
M
,
Proquitté
H
,
Schmalisch
G
,
Roehr
CC
.
Reliability of two common PEEP-generating devices used in neonatal resuscitation
.
Klin Padiatr
.
2009
;
221
(
7
):
415
418
23
Szyld
E
,
Aguilar
A
,
Musante
GA
, et al
.
Comparison of devices for newborn ventilation in the delivery room
.
J Pediatr
.
2014
;
165
(
2
):
234
239.e3
24
Dawson
JA
,
Schmolzer
GM
,
Kamlin
CO
, 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
(
6
):
912
918–2
25
Thakur
A
,
Saluja
S
,
Modi
M
, et al
.
T-piece or self inflating bag for positive pressure ventilation during delivery room resuscitation: an RCT
.
Resuscitation
.
2015
;
90
:
21
24
26
Kitsommart
R
,
Nakornchai
K
,
Yangthara
B
,
Jiraprasertwong
R
,
Paes
B
.
Positive end-expiratory pressure during resuscitation at birth in very-low birth weight infants: a randomized-controlled pilot trial
.
Pediatr Neonatol
.
2018
;
59
(
5
):
448
454
27
Poets
CF
,
Rüdiger
M
.
Mask CPAP during neonatal transition: too much of a good thing for some term infants?
Arch Dis Child Fetal Neonatal Ed
.
2015
;
100
(
5
):
F378
F379
28
Hishikawa
K
,
Goishi
K
,
Fujiwara
T
,
Kaneshige
M
,
Ito
Y
,
Sago
H
.
Pulmonary air leak associated with CPAP at term birth resuscitation
.
Arch Dis Child Fetal Neonatal Ed
.
2015
;
100
(
5
):
F382
F387
29
Murthy
V
,
Rao
N
,
Fox
GF
,
Milner
AD
,
Campbell
M
,
Greenough
A
.
Survey of UK newborn resuscitation practices
.
Arch Dis Child Fetal Neonatal Ed
.
2012
;
97
(
2
):
F154
F155
30
Smithhart
W
,
Wyckoff
MH
,
Kapadia
V
, et al
.
Delivery room continuous positive airway pressure and pneumothorax
.
Pediatrics
.
2019
;
144
(
3
):
e20190756
31
Australian and New Zealand Committee on Resuscitation
.
ANZCOR guideline 13.4 – airway management and mask ventilation of the newborn infant
. East
Melbourne, Australia
:
Australian and New Zealand Committee on Resuscitation
;
2016
32
Thallinger
M
,
Ersdal
HL
,
Morley
C
, et al
.
Neonatal ventilation with a manikin model and two novel PEEP valves without an external gas source
.
Arch Dis Child Fetal Neonatal Ed
.
2017
;
102
(
3
):
F208
F213
33
Ersdal
HL
,
Mduma
E
,
Svensen
E
,
Perlman
J
.
Birth asphyxia: a major cause of early neonatal mortality in a Tanzanian rural hospital
.
Pediatrics
.
2012
;
129
(
5
).
34
Thallinger
M
,
Ersdal
HL
,
Francis
F
, et al
.
Born not breathing: a randomised trial comparing two self-inflating bag-masks during newborn resuscitation in Tanzania
.
Resuscitation
.
2017
;
116
:
66
72
35
American Academy of Pediatrics
.
Helping Babies Breathe
, second edition. Available at: https://internationalresources.aap.org/Resource/ShowFile?documentName=HBB_Wall%20Poster_Second_Edition_20-00372_Rev_D.pdf. Accessed August 12, 2020
36
Mduma
E
,
Ersdal
H
,
Svensen
E
,
Kidanto
H
,
Auestad
B
,
Perlman
J
.
Frequent brief on-site simulation training and reduction in 24-h neonatal mortality--an educational intervention study
.
Resuscitation
.
2015
;
93
:
1
7
37
Linde
JE
,
Eilevstjønn
J
,
Øymar
K
,
Ersdal
HL
.
Feasibility of a prototype newborn resuscitation monitor to study transition at birth, measuring heart rate and ventilator parameters, an animal experimental study
.
BMC Res Notes
.
2017
;
10
(
1
):
235
38
Linde
JE
,
Schulz
J
,
Perlman
JM
, et al
.
Normal newborn heart rate in the first five minutes of life assessed by dry-electrode electrocardiography
.
Neonatology
.
2016
;
110
(
3
):
231
237
39
Schmölzer
GM
,
Kamlin
OC
,
O’Donnell
CP
,
Dawson
JA
,
Morley
CJ
,
Davis
PG
.
Assessment of tidal volume and gas leak during mask ventilation of preterm infants in the delivery room
.
Arch Dis Child Fetal Neonatal Ed
.
2010
;
95
(
6
):
F393
F397
40
van Os
S
,
Cheung
PY
,
Pichler
G
,
Aziz
K
,
O’Reilly
M
,
Schmölzer
GM
.
Exhaled carbon dioxide can be used to guide respiratory support in the delivery room
.
Acta Paediatr
.
2014
;
103
(
8
):
796
806
41
Schmölzer
GM
,
Kamlin
OC
,
Dawson
JA
,
te Pas
AB
,
Morley
CJ
,
Davis
PG
.
Respiratory monitoring of neonatal resuscitation
.
Arch Dis Child Fetal Neonatal Ed
.
2010
;
95
(
4
):
F295
F303
42
O’Donnell
CP
,
Kamlin
CO
,
Davis
PG
,
Morley
CJ
.
Neonatal resuscitation 1: a model to measure inspired and expired tidal volumes and assess leakage at the face mask
.
Arch Dis Child Fetal Neonatal Ed
.
2005
;
90
(
5
):
F388
F391
43
Thallinger
M
,
Ersdal
HL
,
Ombay
C
,
Eilevstjønn
J
,
Størdal
K
.
Randomised comparison of two neonatal resuscitation bags in manikin ventilation
.
Arch Dis Child Fetal Neonatal Ed
.
2016
;
101
(
4
):
F299
F303
44
Hooper
SB
,
Fouras
A
,
Siew
ML
, et al
.
Expired CO2 levels indicate degree of lung aeration at birth
.
PLoS One
.
2013
;
8
(
8
):
e70895
45
Linde
JE
,
Schulz
J
,
Perlman
JM
, et al
.
The relation between given volume and heart rate during newborn resuscitation
.
Resuscitation
.
2017
;
117
:
80
86
46
Holte
KEH
,
Eilevstjønn
J
,
Thallinger
M
, et al
.
Predictors for expired CO2 in neonatal bag-mask ventilation at birth: observational study
.
BMJ Paediatr Open
.
2019
;
3
:
e000544
47
van Vonderen
JJ
,
Roest
AA
,
Siew
ML
,
Walther
FJ
,
Hooper
SB
,
te Pas
AB
.
Measuring physiological changes during the transition to life after birth
.
Neonatology
.
2014
;
105
(
3
):
230
242
48
Probyn
ME
,
Hooper
SB
,
Dargaville
PA
, et al
.
Positive end expiratory pressure during resuscitation of premature lambs rapidly improves blood gases without adversely affecting arterial pressure
.
Pediatr Res
.
2004
;
56
(
2
):
198
204
49
Kuypers
KLAM
,
Lamberska
T
,
Martherus
T
, et al
.
The effect of a face mask for respiratory support on breathing in preterm infants at birth
.
Resuscitation
.
2019
;
144
:
178
184
50
Guinsburg
R
,
de Almeida
MFB
,
de Castro
JS
, et al
.
T-piece versus self-inflating bag ventilation in preterm neonates at birth
.
Arch Dis Child Fetal Neonatal Ed
.
2018
;
103
(
1
):
F49
F55
51
Schmölzer
GM
,
Te Pas
AB
,
Davis
PG
,
Morley
CJ
.
Reducing lung injury during neonatal resuscitation of preterm infants
.
J Pediatr
.
2008
;
153
(
6
):
741
745
52
Tremblay
LN
,
Slutsky
AS
.
Ventilator-induced injury: from barotrauma to biotrauma
.
Proc Assoc Am Physicians
.
1998
;
110
(
6
):
482
488
53
Niermeyer
S
.
Ventilation remains the key to neonatal resuscitation
.
J Pediatr
.
2016
;
171
:
8
10
54
GBD 2015 Child Mortality Collaborators
.
Global, regional, national, and selected subnational levels of stillbirths, neonatal, infant, and under-5 mortality, 1980-2015: a systematic analysis for the Global Burden of Disease Study 2015
.
Lancet
.
2016
;
388
(
10053
):
1725
1774

Competing Interests

POTENTIAL CONFLICT OF INTEREST: Dr Eilevstjønn and Mr Gomo are employees at Laerdal Medical. Jørgen Linde is married to an employee at Laerdal Global Health; the other authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose. The external funding sources had no role in study design, data collection, data analysis, data interpretation, writing of the report, or in the decision to submit the paper for publication.

Supplementary data