Article Text
Abstract
Importance The current neonatal resuscitation guidelines recommend positive pressure ventilation via face mask or nasal prongs at birth. Using a nasal interface may have the potential to improve outcomes for newborn infants.
Objective To determine whether nasal prong/nasopharyngeal tube versus face mask during positive pressure ventilation of infants born <37 weeks’ gestation in the delivery room reduces in-hospital mortality and morbidity.
Data sources MEDLINE (through PubMed), Google Scholar and EMBASE, Clinical Trials.gov and the Cochrane Central Register of Controlled Trials through August 2019.
Study selection Randomised controlled trials comparing nasal prong/nasopharyngeal tube versus face mask during positive pressure ventilation of infants born <37 weeks’ gestation in the delivery room.
Data analysis Risk of bias was assessed using the Covidence Collaboration Tool, results were pooled into a meta-analysis using a random effects model.
Main outcome In-hospital mortality.
Results Five RCTs enrolling 873 infants were combined into a meta-analysis. There was no statistical difference in in-hospital mortality (risk ratio (RR 0.98, 95% CI 0.63 to 1.52, p=0.92, I2=11%), rate of chest compressions in the delivery room (RR 0.37, 95% CI 0.10 to 1.33, p=0.13, I2=28%), rate of intraventricular haemorrhage (RR 1.54, 95% CI 0.88 to 2.70, p=0.13, I2=0%) or delivery room intubations in infants ventilated with a nasal prong/tube (RR 0.63, 95% CI 0.39,1.02, p=0.06, I2=52%).
Conclusion In infants born <37 weeks’ gestation, in-hospital mortality and morbidity were similar following positive pressure ventilation during initial stabilisation with a nasal prong/tube or a face mask.
- neonatology
- resuscitation
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Statistics from Altmetric.com
Introduction
When a newborn infant fails to initiate spontaneous breathing at birth, the current neonatal resuscitation guidelines recommend positive pressure ventilation (PPV) using a face mask or nasal prongs and a ventilation device.1–3 To provide adequate mask ventilation, operator must correctly position the mask onto the infants face by rolling the mask onto the face, holding it with the two-point hold, and keep a tight seal between the mask and face.4 5 However, mask ventilation is complicated by airway obstruction and mask leak, which often remains undetected by the resuscitator.6–8 Several studies have shown substantial mask leaks ranging from 24% to 59% during mask ventilation,9 10 and mask leaks occurred in 51% of recordings within just the first 2 min of PPV.6 Furthermore, mask ventilation is complicated by airway obstruction.8 Finer et al reported airway obstruction rates as high as 75% in low birth weight infants while Schmölzer et al observed 25% of recordings having airway obstructions within the first 2 min of PPV.6 11 Airway obstruction and mask leak could compromise tidal volume delivery, which will results in delay in achieving lung aeration or an increase in heart rate.6
Furthermore, placement of a mask onto an infant can cause apnoea or bradycardia due to the trigeminocardiac reflex.12 Placing a face mask over the mouth and nose activates the trigeminal nerve.12 The activation of this nerve leads to the trigeminocardiac reflex, which is characterised by apnoea, reduction of heart rate and changes in blood pressure.12 A recent cohort study reported that 54% of preterm infants had a reduction in spontaneous breathing after face mask application, which suggests that placement of a face mask could trigger the trigeminocardiac reflex.12 In comparison, the activation of the trigeminocardiac reflex after nasal prongs/tubes placement remains unclear. However, a study by van Vonderen et al reported no difference in breathing rates and heart rate between nasal tube and face mask.13
An alternative approach to face masks is using a nasal prong or nasopharyngeal tube positioned into one or both newborn nostrils.14 With the use of a nasal prong/tube, an infant’s face and mouth can be more easily observed.14 Also, the infant can be more easily moved without the loss of distending pressure.14 A retrospective delivery room study reported that infants who received sustained inflation via nasal cannula had less intubations than the group with sustained inflation via a face mask.15 Similarly, continuous positive airway pressure (CPAP) ventilation via a nasopharyngeal tube in the delivery room resulted in lower rates of endotracheal intubation and mechanical ventilation in extremely low birth weight infants compared with bag and mask ventilation.16 There might be a potential advantage of a nasal interfaces compared with a face mask during PPV in the delivery room.
Therefore, we aimed to compare if a nasal prong or nasopharyngeal tube versus a face mask during PPV in the delivery room reduces in-hospital mortality in infants born <37 weeks’ gestation.
Methods
This review was conducted with the standard methods of Cochrane Handbook for Systematic Reviews of Interventions V.5.3.17 Reporting was in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).18 The review has been submitted to the International Prospective Register of Systematic Reviews (PROSPERO CRD42020178250).
We searched the following electronic databases: MEDLINE (through PubMed), Google Scholar, EMBASE, the Cumulative Index of Nursing and Allied Health Literature, Clinical Trials.gov and the Cochrane Central Register of Controlled Trials using a predefined algorithm (see online supplemental appendix A), with the included search terms “infant”, “newborn”, “resuscitation”, “facemask” and “nasal prong”. We also reviewed abstracts from annual meetings of the Pediatric Academic Society (2000–19), and performed a manual search of references in articles identified by our search strategy (online supplemental appendix A). No language or publication period restrictions were applied. Studies conducted in older children or adults or in settings outside the delivery room were excluded. We also excluded studies performed in a simulation setting.
Supplemental material
Study selection
Two reviewers (AM and GMS) independently assessed the title and abstract of studies for eligibility. Then full texts were retrieved and were included based on the eligibility criteria. Any disagreement was resolved through discussion. Only randomised controlled trials (RCTs) comparing the use of face mask with a nasal prong or nasopharyngeal tube in newborn infants receiving neonatal resuscitation in the delivery room and reporting in-hospital mortality were included. Outcomes were chosen according to the consensus outcome rating for international neonatal resuscitation guidelines.19 Our primary outcome measure was in-hospital mortality in infants. Secondary outcomes included intubation in the delivery room or in the neonatal intensive care unit (NICU) <72 hours or both, chest compressions in the delivery room, air leaks (ie, pneumothorax), intraventricular haemorrhage (grade III/IV or described as severe), bronchopulmonary dysplasia or mortality and necrotising enterocolitis (stage 2 or higher).20 21 The review team resolved any discrepancies regarding inclusion through consensus.
Data extraction
Data extraction was performed using a standardised collection form which included study design and methods, patient characteristics, interventions and outcomes. We documented the mode of randomisation, allocation concealment, blinding and compliance with intention-to-treat analysis. Data extraction was performed using the Covidence Collaboration Tool. Two investigators (AM/GMS) independently extracted data and resolved discrepancies in consultation with another member of the review team (MB). We also contacted the authors of any studies that had missing outcome data.
Assessment of methodological quality
The methodological quality of the included trials and evaluated risk of bias was assessed using elements of the Cochrane Collaboration tool. The domains used included randomisation and allocation concealment (selection bias), blinding (performance and detection bias) and adherence to the intention to treat principle (attrition bias). Similarly, two authors (AM/GMS) assessed the certainty of evidence (confidence in the estimate of effect) for each outcome based on the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework, including the calculation of the optimal information size to assess imprecision (GRADEpro Guideline Development Tool; McMaster University, Hamilton, Canada).
Statistical analysis
Covidence, GRADEpro and Review Manager software V.5.3 were used to abstract, summarise and analyse the data, respectively. The principal summary measures were the weighted mean difference for continuous outcomes and risk ratio (RR) for dichotomous outcomes. For each trial, we retrieved or calculated the crude RR estimates and corresponding 95% CIs for the assessed outcomes. Heterogeneity was assessed using a χ2 test and the I2 statistic. We used a random effect models to summarise RR estimates. All p values are two tailed. The study is reported according to the PRISMA checklist.
Results
Our search obtained 1140 records with two being removed as duplicates. A total of 1118 records were removed after titles and abstracts screening and 20 full texts were reviewed. Fifteen studies were removed due to wrong study design, intervention, patient population, setting, outcomes and comparator. Consequently, five RCTs were included in this review. The PRISMA flow diagram is presented in figure 1, and the GRADE Assessment of Evidence Table for key prespecified outcomes is shown in table 1. One relevant ongoing study was found in the Cochrane Central Register of Controlled Trials but no data were available from this study.22
Characteristics of included studies
Five RCTs from Australia, Netherlands, Ireland, Italy and Sweden were included in this review. A total of 873 newborn infants born <37 weeks’ gestation receiving PPV in the delivery room were included in this review. Two studies compared the effect of a single nasal prong with a face mask during PPV and/or CPAP using a T-piece resuscitator.23 24 One study investigated nasopharyngeal tube and T-piece resuscitator with sustained inflation versus face mask and self-inflating bag.25 The study by Capasso et al had both preterm and term infants in the study population and compared the effect of face mask with nasal cannula during self-inflating bag resuscitation.26 But after contacting the authors we were able to receive data for preterm infants which we included in our analysis.26 The study by Donaldsson et al compared three groups: T-piece systems with a face mask, a new system with lower work of breathing with a face mask and with short binasal prongs. We contacted the authors to obtain the missing outcome data from some of the included trials, but the authors stated that said data were not collected in their respective studies.26 27 Characteristics of the included studies are shown in table 2.
Quality of individual studies
Assessment of potential sources of bias are presented in figure 2. Risk of bias of the included studies was assessed using the Cochrane Collaboration Tool. Three studies reported the adequate method of allocation concealment.23–25 None of the studies reported blinding of participants, personnel and outcomes assessors for the interventions. For incomplete outcome reporting and selective outcome reporting, three out of the five studies were at low risk of bias,23–25 while the risk was unclear for the other two studies.26 27 Other sources of risk were not identified in the included studies.
Meta-analysis of primary outcome
In-hospital mortality
Four of the included studies reported infant mortality.23–26 The pooled data suggest similar infant mortality between nasal prong/tube versus face mask during PPV (RR 0.98, 95% CI 0.63 to 1.52, p=0.92, I2=11%) (figure 3). The use of a nasal prong/tube during ventilation could result in 2 fewer per 1000 (from 39 fewer to 55 more) infant deaths.
Meta-analysis of secondary outcomes
Intubation in the delivery room
All five studies reported rates of intubation in the delivery room.23–27 The pooled data suggest a lower intubation rates with nasal prong/tube use versus face mask during PPV (RR 0.63, 95% CI 0.39 to 1.02, p=0.06, I2=52%) (figure 4A). The use of a nasal prong/tube during ventilation could result in 98 fewer per 1000 (from 162 fewer to 5 more) intubations in the delivery room.
Intubation in the NICU <72 hours,
Three of the studies reported on intubation in the NICU <72 hours.23–25 The pooled data found similar incidences of intubation in the NICU <72 hours between nasal prong and versus face mask during PPV (RR 0.99, 95% CI 0.73 to 1.38, p=0.97, I2=71%) (figure 4B). The use of a nasal prong/tube during ventilation could result in 4 fewer per 1000 (from 110 fewer to 145 more) intubations in the NICU <72 hours.
Air leaks
Four of the studies reported on air leaks (ie, pneumothorax).23 25–27 The pooled data found no difference in air leaks between nasal prong and versus face mask during PPV (RR 0.98,95% CI 0.23 to 4.23, p=0.98, I2=52%) (figure 4C). The use of a nasal prong/tube during ventilation could result in 2 fewer per 1000 (from 39 fewer to 55 more) air leaks in the delivery room.
Bronchopulmonary dysplasia
Three of the studies reported bronchopulmonary dysplasia.23–25 The pooled data found no difference in bronchopulmonary dysplasia between nasal prong/tube versus face mask during PPV (RR 0.91,95% CI 0.66 to 1.27, p=0.59, I2=42%) (figure 4D). The use of a nasal prong/tube during ventilation could result in 26 more per 1000 (from 60 fewer to 149 more) bronchopulmonary dysplasia incidences.
Chest compressions in the delivery room
Three of the studies reported on the rates of chest compression in the delivery room.23 24 26 The pooled data found a lower incidence of infants receiving chest compressions in the nasal prong/tube group (RR 0.37, 95% CI 0.10 to 1.33, p=0.13, I2=28%) (figure 4E). The use of a nasal prong/tube during ventilation could result in 31 fewer per 1000 infants (from 44 fewer to 16 more) chest compressions in the delivery room.
Intraventricular hemorrhage (grade III/IV or described as severe)
Three of the included studies reported on intraventricular haemorrhage.23–25 The pooled data found a lower incidence of intraventricular haemorrhage in the face mask group (RR 1.54, 95% CI 0.88 to 2.70, p=0.13, I2=0%) (figure 4F). The use of a face mask during ventilation could result in 18 fewer per 1000 (from 33 fewer to seven more) intraventricular haemorrhage.
Necrotising enterocolitis (stage 2 or higher)
Three of the included studies reported on the incidence of necrotising enterocolitis.23–25 The pooled data found no difference in necrotising enterocolitis incidences between groups (RR 0.74, 95% CI 0.29 to 1.88, p=0.53, I2=0%) (figure 4G). The use of a nasal prong/tube during ventilation could result in 10 more per 1000 (from 13 fewer to 67 more) necrotising enterocolitis incidences in the delivery room.
Discussion
This systematic review and meta-analysis compared nasal prong or nasopharyngeal tube versus face mask during PPV in the delivery room in preterm infants born <37 week’s gestation. There was no difference in the primary outcome of in-hospital mortality between nasal prong/tube and face mask. There was a reduction in rate of intubation and chest compression in the delivery room when using nasal prongs/tubes and a reduction in incidence of intraventricular haemorrhage when using a face mask.
The included studies were conducted over a 12-year period (2005–2017), in which several changes to neonatal resuscitation guidelines occurred.1–3 28 Most importantly, change from 100% to lower oxygen concentrations for respiratory support, titration of oxygen and delayed cord clamping.1 2 29 Capasso et al and te Pas et al used 100% oxygen while the other studies used oxygen concentrations between 21% and 30%. The varying oxygen concentrations and different criteria for intubation might have introduced heterogeneity in our analysis. Overall, there was a 37% reduction in rate of intubation with a nasal prong/tube compared with face mask ventilation (73/422 vs 120/451; p=0.06; figure 4A), with no difference in rate of intubation after admission to the NICU. The lower rate of intubation in the delivery room might be associated with the higher oxygen concentration. This is supported observational data reporting that a brief period of up to 100% oxygen results in improved higher respiratory effort, improved oxygenation and shorter duration of mask ventilation.30 31 It is possible that studies with 100% oxygen during initial for respiratory support had lower rates of intubation as well as chest compressions due to improved respiratory stability and lower risk of apnoea.30 31
Infants receiving PPV with nasal prong/tube had lower rates of chest compressions compared with mask PPV (5/306 vs 16/324; p=0.13, figure 4E). The number of infants receiving chest compression was higher in the study by Capasso et al, which could be attributed to starting chest compressions with a heart rate <80/min while all other studies used <60/min to start chest compressions.1 3
Nasal prong/tube might reduces an infants’ needs for intubation and chest compression, which are associated with an increased neonatal morbidity and mortality.16 32 A nasal interface might result in less respiratory support due to improved spontaneous breathing compared with a face masks, which can trigger the trigeminocardiac reflex and thereby causing reduced spontaneous breathing efforts and/or apnoea.12 Similar to our meta-analysis, previous studies reported lower rates of chest compressions and intubations with nasal prong use.15 33 One study reported significantly more air leaks and airway obstructions with a nasopharyngeal tube compared with face mask ventilation, however our meta-analysis did not identify an increased risk of air leaks.13 Further studies reported that lower gestational age is a risk factor for increased CPAP failure and nasal injury during nasal prong/tube use.34 35
Limitations
Our meta-analysis may have been affected by heterogeneity between interventions, small sample sizes and bias from personnel and outcome assessors. The study by te Pas et al compared a T-piece resuscitator with nasopharyngeal tube to a self-inflating bag with a face mask.25 Additionally, there was no consistency in the type of nasal prong or nasopharyngeal tube used. Combined with the effect of personnel’s experience in applying these interventions, this might have had a synergistic effect with the interface.23 The sample sizes for many of the outcomes were small due to missing data or low recruitment in individual studies such as Donaldsson et al only included 36 infants.27 There was also no mention of blinding personnel and outcome assessors in any of the studies which introduces bias. Unfortunately, none of the studies examined long-term neurodevelopment outcomes.
Clinical applications
Our results might have implications for the delivery room in high-resource and resource-limited settings. The use of a nasal interfaces during PPV might reduce number of intubation in the delivery room, but not within the first 72 hours after birth. There were a lower number of chest compressions and in infants receiving PPV with a nasal prong. However, the incidence of intraventricular haemorrhage was lower in infants receiving PPV with a face mask. Furthermore, paramedics, emergency room personal or in resource-limited settings might not have the proper neonatal face masks available and therefore could use a cut endotracheal tube, which might be an easier solution to provide PPV.
Conclusions
There was no difference in in-hospital mortality in infants born <37 weeks’ gestation receiving PPV with either a nasal prong/tube or a face mask. Our results suggest that newborn infants receiving respiratory support at birth with a nasal interface might have lower rates of intubations and chest compressions in the delivery room but higher rates of intraventricular haemorrhage. Large clinical trials to assess whether nasal prong/tubes or face mask should be sued for mask ventilation in the delivery room are warranted.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Acknowledgments
The authors would like to thank the public for donating money to the funding agencies.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors Conception: GMS, AM. Data acquisition: GMS, AM, MB. Data analysis: GMS, AM, MB. Interpreting of results: GMS, AM, MB. Drafting of the manuscript: GMS, AM, MB. Critical revision of the manuscript: GMS, AM, MB. Final approval of the manuscript: GMS, AM, MB.
Funding GMS is a recipient of the Heart and Stroke Foundation/University of Alberta Professorship of Neonatal Resuscitation, a National New Investigator of the Heart and Stroke Foundation Canada and an Alberta New Investigator of the Heart and Stroke Foundation Alberta.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
Linked Articles
- Fantoms