Available via license: CC BY 4.0
Content may be subject to copyright.
RESEARCH ARTICLE
Effect of a short training on neonatal face-
mask ventilation performance in a low
resource setting
Alessandro Mazza
1
, Francesco Cavallin
2
, Anita Cappellari
1
, Antuan Divisic
1
, Ivana Grbin
1
,
Jean Akakpo
3
, Abdou Razak Moukaila
3
, Daniele Trevisanuto
1
*
1Department of Women’s and Children’s Health, University of Padua, Azienda Ospedaliera di Padova,
Padova, Italy, 2Independent Statistician, Padova, Italy, 3Maternitè, Marie Mère de la Providence, Centre
Me
´dico-Social, Kouvè, Togo, Africa
*daniele.trevisanuto@gmail.com
Abstract
Background
We assessed whether a short training, effective in a high resource country, was able to
improve the quality of face-mask ventilation (FMV) in a low resource setting.
Methods
Local healthcare providers at the Centre Me
´dico-Social, Kouvè, Togo were asked to venti-
late a neonatal leak-free manikin before (time—t
1
) and after (t
2
) a two-minute training ses-
sion. Immediately after this section, a further two-minute training with participants aware of
the data monitor was offered. Finally, a third 1-minute FMV round (t
3
) was performed by
each participant. Ventilatory parameters were recorded using a computerized system. Pri-
mary outcome was the percentage of breaths with relevant mask leak (>25%). Secondary
outcomes were percentages of breaths with a low peak inspiratory pressure (PIP<20 cm
H
2
O), within the recommended PIP (20–35 cm H
2
O) and with a high PIP (>35 cm H
2
O).
Results
Twenty-six subjects participated in the study. The percentage of relevant mask leak signifi-
cantly decreased (p<0.0001; β= -0.76, SE = 0.10) from 89.7% (SD 21.5%) at t
1
to 45.4%
(SD 27.2%) at t
2
and to 18.3% (SD 20.1%) at t
3
. The percentage of breaths within the rec-
ommended PIP significantly increased (p<0.0001; β= +0.54, SE = 0.12). The percentage
of breaths with PIP>35 cm H2O was 19.5% (SD 32.8%) at t
1
and 39.2% (SD 37.7%) at t
2
(padj = 0.27; β= +0.61, SE = 0.36) and significantly decreased (padj = 0.01; β= -1.61, SE =
0.55) to 6.0% (SD 15.4%) at t
3
.
Conclusions
A 2-minute training on FMV, effective in a high resource country, had a positive effect also in
a low resource setting. FMV performance further improved after an extra 2-minute verbal
recall plus real time feedback. Although the training was extended, it still does not cost much
PLOS ONE | https://doi.org/10.1371/journal.pone.0186731 October 26, 2017 1 / 9
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
OPEN ACCESS
Citation: Mazza A, Cavallin F, Cappellari A, Divisic
A, Grbin I, Akakpo J, et al. (2017) Effect of a short
training on neonatal face-mask ventilation
performance in a low resource setting. PLoS ONE
12(10): e0186731. https://doi.org/10.1371/journal.
pone.0186731
Editor: Francesco Staffieri, University of Bari,
ITALY
Received: May 18, 2017
Accepted: October 8, 2017
Published: October 26, 2017
Copyright: ©2017 Mazza et al. This is an open
access article distributed under the terms of the
Creative Commons Attribution License, which
permits unrestricted use, distribution, and
reproduction in any medium, provided the original
author and source are credited.
Data Availability Statement: All relevant data are
within the paper and its Supporting Information
files.
Funding: Autonomous Province of Trento and
Autonomous Region Trentino-Alto Adige, Italy,
supported the project. The funders had no role in
study design, data collection and analysis, decision
to publish, or preparation of the manuscript.
Competing interests: The authors have declared
that no competing interests exist.
time and effort. Further studies are needed to establish if these basic skills are transferred in
real patients and if they are maintained over time.
Introduction
Every year around 6,6 million children worldwide under 5 year die. Of these, 44% are in the
neonatal period. Intrapartum-related events (“birth asphyxia”), account for a quarter of neona-
tal deaths suggesting that basic skill training of those involved in the care of neonates at deliv-
ery is a crucial investment [1–3].
Recently, neonatal resuscitation is receiving increasing attention as a missed opportunity to
improve morbidity and mortality outcomes. Newton and English reviewed the evidence for
neonatal resuscitation and concluded that effective resuscitation in low-resource settings was
possible with basic equipment and skills [4]. Training health care providers in neonatal resus-
citation may prevent 30% of deaths of full-term babies experiencing adverse intrapartum
events, as well as 5%–10% of deaths among infants born preterm [5,6].
Effective positive pressure ventilation (PPV) is the most important intervention for success-
ful resuscitation of the newborn [7,8]. In the settings where continuous gas flow is not avail-
able, PPV is administered by using a self-inflating bag (SIB) and mask. However, achieving
effective face-mask ventilation (FMV) can be difficult [9–12]. Specific key points of the proce-
dure, such as reduce leak around the mask, avoid the airway block and administer adequate
pressures, need to be well known by healthcare providers [7,8].
A previous study conducted in a high resource setting showed that a structured two-minute
training based on 6 key-points significantly improved the quality of FMV in a manikin model.
The authors suggested that this training could be incorporated into any educational program
[13].
In addition, Kelm et al. demonstrated that a training scheme of FMV with a SIB, including
a simple respiratory function monitor to feed back the level of the applied ventilator parame-
ters to the individual operator significantly reduced the occurrence of excessive pulmonary
pressures and volumes [14].
However, the impact of these training interventions in a low-resource setting remains
unknown.
The aim of this study was to assess whether a short training, effective in high resource coun-
tries, was able to improve the quality of FMV in a low resource setting.
Methods
Setting
This study was conducted at the Centre Me
´dico-Social, Kouvè, Pre
´fecture de Yoto, Togo. This
is a rural hospital where about 600 deliveries occur every year.
Study design
Participants consisted of physicians, midwives and nurses. Most of them attended a neonatal
theoretical and practical resuscitation course two years before. All participants were asked to
administer FMV for a minute to a manikin. A neonatal manikin (Laerdal Resusci Baby, Laer-
dal, Stavanger, Norway) was modified for obtaining a leak free system with a 50 ml test lung as
previously described [13]. A 240 ml-self inflating bag and a size 1 round mask (Laerdal,
Face-mask ventilation performance in a low resource setting
PLOS ONE | https://doi.org/10.1371/journal.pone.0186731 October 26, 2017 2 / 9
Abbreviations: FMV, face-mask ventilation; PIP,
peak inspiratory pressure; PPV, positive pressure
ventilation; SIB, self-inflating bag; VR, Ventilatory
rate.
Stavanger, Norway) were used to administer PPV. The bag had a pressure release valve at 35
cmH
2
O without end expiratory pressure (PEEP) valve.
PPV parameters (Ventilatory rate–VR, Peak Inspiratory Pressure—PIP, Flow, Inspiratory
Volume -Vti-, Expiratory Volume -Vte-, Leak) were measured using a training computerized
system for neonatal mask ventilation (NewLifebox-T—Neonatal Resuscitation Trainer,
Advanced Life Diagnostics UG, Weener, Germany). It measures pressure and air flow through
the resuscitation mask using a flow/pressure probe which is placed between the face-mask and
the resuscitation device (dead space of 0,7 ml). Data were analyzed by a software installed on a
standard computer (NewLifebox Training Center software, Advanced Life Diagnostics UG,
Weener, Germany). All signals were digitized and recorded with a data acquisition program in
the software. After installing the software, the Newlifebox-T device was connected to the com-
puter using the USB-cable. The flow sensor was re-calibrated before starting each measure-
ment. In this way, the sensors do not drift away from zero. The software indicates it. For this
baseline test, the flow probe was connected to the NewLifebox-T, while absolutely no air flo-
wed to the probe and no pressure was applied to the probe.
The air flow was integrated to provide inspired—and expired tidal volumes (Vti and Vte).
Leak at the face-mask was calculated as the difference between the inspired and expired tidal
volumes, expressed as a percentage of the inspired tidal volume (leak percentage = [(inspira-
tory tidal volume–expiratory tidal volume) / inspiratory tidal volume] x 100) [13].
The verbal instruction and the demonstration were decided and standardized before start-
ing the study.
The experiment was subdivided into 5 steps, according to the following protocol (Fig 1):
1. At the beginning, participants ventilated the manikin without any instruction for 1 minute.
Mask leak, PIP and VR were recorded during performance, but participants were unaware
of these data. Each participant was allowed to ventilate once.
2. After this first round, participants received a verbal instruction and demonstration by one
instructor. In agreement with a previous study; the training was based on 6 key points and
the duration was approximately 2 minutes [10].
Fig 1. Study design. FMV—face-mask ventilation.
https://doi.org/10.1371/journal.pone.0186731.g001
Face-mask ventilation performance in a low resource setting
PLOS ONE | https://doi.org/10.1371/journal.pone.0186731 October 26, 2017 3 / 9
3. Immediately after the instruction, participants were asked to repeat the procedure for 1
minute once. Also during this round, participants were unaware of the ventilatory parame-
ters recorded by the software.
4. At this point, a second two-minute training, including the verbal recall of the 6-points
scheme and a short ventilatory trial with participant aware of the data monitor, was offered
to each participant.
5. Immediately after this second training, a third FMV round lasting 1 minute was performed
by each participant, who remained blind to the data monitor.
Key points of the face-mask ventilation. All participants received two verbal instructions
by one instructor. The following key points were explained and recommended:
1. positioning of the head in sniffing position;
2. positioning mask on the tip of chin, mouth and the nose, but not on the eyes;
3. appling mask to face using mild downward pressure and lifting the mandible up toward the
mask;
4. gentle squeeze of the bag aimed to obtain effective chest movements that correspond to a
peak inspiratory pressures between 20–35 cmH20;
5. beware that when the pop off valve of the SIB releases (at 35 cm H2O) uncontrolled pres-
sures can be given;
6. mask ventilation at a rate of 40–60 inflations per minute.
Outcomes
Primary outcome was the percentage of breaths per minute with relevant mask leak (higher
than 25%). Secondary outcomes were the number of inflations per minute (VR), the percent-
age of breaths per minute with a low PIP (below 20 cm H
2
O), the percentage of breaths per
minute within the recommended PIP (between 20 and 35 cm H
2
O) and the percentage of
breaths per minute with a high PIP (higher than 35 cm H
2
O).
Ethics statement
The study was approved by the institutional review board (IRB) of the Centre Me
´dico-Social,
Kouvè, Togo. All staff members gave verbal consent to participate in the study; it was docu-
mented in an excel sheet. Consent procedure was approved by the local IRB.
Statistics
Continuous data were expressed as mean and standard deviation (SD) or median and inter-
quartile range (IQR). VR (number of breaths per minute) was expressed as count number,
whereas the number of breaths per minute with relevant mask leak, the number of breaths per
minute with a low peak inspiratory pressure, the number of breaths per minute with a recom-
mended peak inspiratory pressure and the number of breaths per minute with a high peak
inspiratory pressure were expressed as percentages on total number of breaths per minute.
These variables were recorded at 3 time points (before the training t
1
, after the training t
2
and after the recall training t
3
) for each subject, thus they were evaluated using a Poisson
regression model for repeated measurements. The models for the primary and secondary
Face-mask ventilation performance in a low resource setting
PLOS ONE | https://doi.org/10.1371/journal.pone.0186731 October 26, 2017 4 / 9
outcomes included also the logarithm of ventilatory rate as offset. The effect of possible con-
founders (previous participation to a course on neonatal resuscitation and age) were assessed
by including such effects in the models. Bonferroni’s adjustment for multiple comparisons was
used when appropriate.
Statistical analysis will be performed using R 3.2.2 software (R Foundation for Statistical
Computing, Vienna, Austria) [15].
A p-value less than 0.05 was considered statistically significant.
Results
Participants
Twenty-six healthcare providers attended the training and were included in the study. There
were 14 males and 12 females with a median age of 36 years (IQR 29–41). Five participants
were physicians, 6 midwives and 15 nurses. Most of them (21 out of 26, 80.8%) had already
attended a course on neonatal resuscitation in the previous 2 years.
Primary outcome
Mean percentage of breaths per minute with relevant mask leaks (Fig 2A) significantly
decreased (p<0.0001; β= -0.76, SE = 0.10) from 89.7% (SD 21.5%) at t
1
to 45.4% (SD 27.2%) at
t
2
and to 18.3% (SD 20.1%) at t
3
. The effects of a previous course (p = 0.15), and age (p = 0.80)
were not statistically significant.
Secondary outcomes
Mean VR significantly decreased (p = 0.02; β= -0.17, SE = 0.05) from 58.7/min (SD 21.5) at t
1
to 49.9/min (SD 14.7) at t
2
and to 51.1/min (SD 11.4) at t
3,
with a reduced effect if the partici-
pants had already attended a course (p = 0.08; β= 0.12, SE = 0.07). The effect of age (p = 0.82)
Fig 2. Percentage of breaths per minute (bpm) with (a) relevant mask leak (>25%), (b) low peak inspiratory
pressure (PIP<20 cm H2O), (c) peak inspiratory pressure in the recommended range (PIP = 20–35 cm H2O),
and (d) high peak inspiratory pressure (PIP>35 cm H2O). Data are expressed as mean (95%CI).
https://doi.org/10.1371/journal.pone.0186731.g002
Face-mask ventilation performance in a low resource setting
PLOS ONE | https://doi.org/10.1371/journal.pone.0186731 October 26, 2017 5 / 9
was not statistically significant. Mean percentage of breaths per minute with a low PIP (Fig 2B)
significantly decreased (p
adj
<0.0001; β= -1.38, SE = 0.31) from 73.3% (SD 36.5%) at t
1
to
19.8% (SD 31.8%) at t
2
then leveled (p
adj
= 0.18; β= +0.60, SE = 0.32) to 32.9% (SD 30.4%) at
t
3
. The effect of a previous neonatal course (p = 0.53) and age (p = 0.22) were not statistically
significant.
Mean percentage of breaths per minute within the recommended PIP range (Fig 2C) signif-
icantly increased (p<0.0001; β= +0.54, SE = 0.12) from 7.1% (SD 12.3%) at t
1
to 41.0% (SD
33.7%) at t
2
and to 61.1% (SD 29.3%) at t
3
. This increment was enhanced if the participant had
already attended a previous neonatal course (p = 0.04, β= +0.34, SE = 0.17), whereas the effect
of age (p = 0.10) was not statistically significant.
Mean percentage of breaths per minute with a high PIP (Fig 2D) was 19.5% (SD 32.8%) at
t
1
and 39.2% (SD 37.7%) at t
2
(p
adj
= 0.27; β= +0.61, SE = 0.36) and significantly decreased
(p
adj
= 0.01; β= -1.61, SE = 0.55) to 6.0% (SD 15.4%) at t
3
. The effects of a previous neonatal
course (p = 0.63) and age (p = 0.72) were not statistically significant.
Discussion
Worldwide an estimated 3 to 6% of newborn infants need assisted PPV at birth [8]. It has been
hypothesized that basic neonatal training programs improve neonatal survival [11,12]. As
effective FMV is the most important intervention during neonatal resuscitation, all healthcare
providers involved in the delivery room management of neonates have to be capable to per-
form this procedure [7,8]. Repetition of ventilation skills by a simple and short training may
contribute to improve the basic neonatal resuscitation [5].
In this study, we assessed the efficacy of a short training program on the quality of FMV in
a low resource setting. We found that the quality of manual ventilation, defined as mask leak
and adequate PIP, significantly improved after the training.
A previous study, conducted in a high resource setting, showed that the quality of FMV in a
manikin model improved significantly by using a structured 2-minute training consisting of 6
key-points [13]. In the first part of this study by using the same training program [13], we
found a significant improvement on FMV performance, but the effectiveness of the perfor-
mance was amplified when participants was offered a respiratory function monitor to feed
back the level of the applied ventilatory parameters and the mask leak.
In comparison to the manikin study of Vonderen et al. [13], we modified the training a bit
by adding an extra 2 minutes with verbal recall plus with monitor visible. This decision to use
a respiratory function monitor was based on the positive results of a previous bench study con-
ducted in a high resource setting [14]. These findings show that it only takes a few minutes to
teach caregivers adequate mask ventilation, when using a monitor, and thus should be incor-
porated in every training.
Although the mask leak and the percentage of ventilatory breaths in the recommended PIP
(between 20 and 35 cmH2O) improved after the initial training, this change was limited. In
contrast with van Vonderen et al’s study [13], we noted that the percentage of ventilatory
breaths in the “dangerous zone” (PIP>35cmH2O) significantly increased after the initial train-
ing. As an “aggressive” ventilation can give air-leak and pneumothorax that can be fatal in a
setting where mechanical ventilation is not available, it is important to underline this risk fol-
lowing the training. In other words, the pre-training performance indicated the risk of hypo-
ventilation of the manikin, but immediately after the first training, there was a risk of
barotrauma. Both these situations should be avoided.
The quality of FMV significantly improved after the second section of 2-minute training,
including the verbal recall of 6-key points and a short ventilatory trial allowing the participant
Face-mask ventilation performance in a low resource setting
PLOS ONE | https://doi.org/10.1371/journal.pone.0186731 October 26, 2017 6 / 9
to have a real time feedback of their performance. These results show that, in addition to the
short training program suggested by previous work [13], an additional intervention was
needed in a low resource setting to reach a good performance.
There are several reasons that need to be considered to explain the different impact of the
same training on two groups of healthcare providers, including participants’ experience and
setting. As the study was performed in a rural hospital with a low number of deliveries per year
(about 600), it has to be recognized that caregivers had a low exposure to the procedure. How-
ever, we believe that our results can be translated to the majority of low-resource delivery cen-
ters where there is a limited number of births/year.
Our results cannot demonstrate whether the improvement registered after the second edu-
cational intervention was due to the verbal recall of the 6-key points or to the short ventilatory
trial in which the participant was aware of the ventilatory parameters.
A further finding of this study was that participants who previously attended a neonatal the-
oretical and practical resuscitation course showed a better performance in comparison with
those who did not suggesting that basic neonatal resuscitation skill may be maintined over
time through repeated training [5,12].
The strength of this study is that it assessed the effect of a structured short training on the
quality of FMV in a low resource setting. Nevertheless, it has some limitations that should be
considered when interpreting the results. We enrolled a limited number of participants, but
they represent the entire staff involved in the care of the newborns born in a typical organiza-
tional and cultural environment of a rural African delivery setting; our results could be differ-
ent in other contexts. We evaluated the short-term effect of the intervention in a manikin
model; it would be relevant to investigate the consequences of this training on the manage-
ment of real patients. As previous work showed that low dose high frequency training pro-
grams reach the best results [12], it remains to be demonstrated if repetition of our
intervention can further improve healthcare providers’ performance.
Conclusions
A 2 minute structured training on FMV effective in a high resource country improved FMV
performance in a low resource setting, but adding an extra session of 2 minutes with monitor
visible led to further improvement. Although the training was extended, it still does not cost
much time and effort. Further studies are needed to demonstrate whether participants’ skills
are maintained over the time and whether our training model can improve FMV in clinical
practice in a low-resource setting.
Supporting information
S1 Table. Supplementary data.
(PDF)
Acknowledgments
We thank the Autonomous Province of Trento and Autonomous Region Trentino-Alto
Adige, Italy that supported the project.
Author Contributions
Conceptualization: Alessandro Mazza, Anita Cappellari, Antuan Divisic, Ivana Grbin, Jean
Akakpo, Abdou Razak Moukaila, Daniele Trevisanuto.
Face-mask ventilation performance in a low resource setting
PLOS ONE | https://doi.org/10.1371/journal.pone.0186731 October 26, 2017 7 / 9
Data curation: Alessandro Mazza, Francesco Cavallin, Anita Cappellari, Antuan Divisic,
Ivana Grbin, Jean Akakpo, Abdou Razak Moukaila.
Formal analysis: Alessandro Mazza, Francesco Cavallin.
Funding acquisition: Daniele Trevisanuto.
Methodology: Jean Akakpo.
Supervision: Daniele Trevisanuto.
Validation: Abdou Razak Moukaila.
Writing – original draft: Daniele Trevisanuto.
Writing – review & editing: Francesco Cavallin, Anita Cappellari, Antuan Divisic, Ivana
Grbin, Jean Akakpo, Abdou Razak Moukaila, Daniele Trevisanuto.
References
1. WHO. Health-related Millennium Development Goals. World Health Statistics. 2014. Available: http://
www.who.int/gho/publications/world_health_statistics/2014/en/.
2. Unicef, Levels & Trends in Child Mortality. Report 2013. New York, USA, 2013. Available:http://www.
data.unicef.org/fckimages/uploads/1410869227_Child_Mortality_Report_2014.pdf.
3. Lawn JE, Blencowe H, Oza S, You D, Lee AC, Waiswa P, et al. Lancet Every Newborn Study Group.
Progress, priorities, and potential beyond survival. Lancet. 2014; 384(9938):189–205. https://doi.org/
10.1016/S0140-6736(14)60496-7 Epub 2014 May 19. PMID: 24853593.
4. Newton O, English M. Newborn resuscitation: defining best practice for low-income settings. Trans R
Soc Trop Med Hyg. 2006 Oct; 100(10):899–908. Epub 2006 Jun 6. https://doi.org/10.1016/j.trstmh.
2006.02.012 PMID: 16757009.
5. Singhal N, Niermeyer S. Neonatal resuscitation where resources are limited. Clin Perinatol 2006;
33;219–28, x-xi. https://doi.org/10.1016/j.clp.2005.11.009 PMID: 16533646.
6. Lee A, Cousens S, Wall SN, Niermeyer S, Darmstadt GL, Carlo WA, et al. Neonatal resuscitation and
immediate newborn assessment and stimulation for the prevention of neonatal deaths: a systematic
review, meta-analysis and Delphi estimation of mortality effect. BMC Public Health 2011; 11(suppl 3):
S12. https://doi.org/10.1186/1471-2458-11-S3-S12 PMID: 21501429.
7. Wyckoff MH, Aziz K, Escobedo MB, Kapadia VS, Kattwikel J, Perlman JM, et al. Part 13: Neonatal
Resuscitation: 2015 American Heart Association Guidelines Update for Cardiopulmonary Resuscitation
and Emergency Cardiovascular Care. Circulation. 2015 Nov 3; 132(18 Suppl 2):S543–60. https://doi.
org/10.1161/CIR.0000000000000267 PMID: 26473001.
8. Wyllie J, Bruinenberg J, Roehr CC, Ru¨diger M, Trevisanuto D, Urlesberger B. European Resuscitation
Council Guidelines for Resuscitation 2015: Section 7. Resuscitation and support of transition of babies
at birth. Resuscitation. 2015 Oct; 95:249–63. https://doi.org/10.1016/j.resuscitation.2015.07.029 Epub
2015 Oct 15. PMID: 26477415.
9. Trevisanuto D, Bertuola F, Lanzoni P, Cavallin F, Matediana E, Manzungu OW, et al. Effect of a neona-
tal resuscitation course on healthcare providers’ performances assessed by video recording in a low-
resource setting. PLoS One. 2015 Dec 11; 10(12):e0144443. https://doi.org/10.1371/journal.pone.
0144443 eCollection 2015. PMID: 26659661.
10. Wood FE, Morley CJ, Dawson JA, Kamlin CO, Owen LS, Donath S, et al. Improved techniques reduce
face-mask leak during simulated neonatal resuscitation: study 2. Arch Dis Child Fetal Neonatal Ed.
2008 May; 93(3):F230–4. Epub 2007 Nov 26. https://doi.org/10.1136/adc.2007.117788 PMID:
18039750.
11. Msemo G, Massawe A, Mmbando D, Rusibamayila N, Manji K, Kidanto HL et al. Newborn mortality and
fresh stillbirth rates in Tanzania after helping babies breathe training. Pediatrics. 2013 Feb; 131(2):
e353–60. https://doi.org/10.1542/peds.2012-1795 PMID: 23339223.
12. 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 Aug; 93:1–7. https://doi.org/10.1016/j.resuscitation.2015.04.019 Epub 2015 May 6. PMID:
25957942.
Face-mask ventilation performance in a low resource setting
PLOS ONE | https://doi.org/10.1371/journal.pone.0186731 October 26, 2017 8 / 9
13. Van Vonderen JJ, Witlox RS, Kraaij S, te Pas AB. Two-minute training for improving neonatal bag and
mask ventilation. PLoS One. 2014 Oct 3; 9(10):e109049. https://doi.org/10.1371/journal.pone.0109049
eCollection 2014. PMID: 25279800.
14. Kelm M, Dold SK, Hartung J, Breckwoldt J, Schmalisch G, Roehr CC. Manual neonatal ventilation train-
ing: a respiratory function monitor helps to reduce peak inspiratory pressures and tidal volumes during
resuscitation. J Perinat Med. 2012 Sep; 40(5):583–6. PMID: 23120761.
15. R Core Team (2015). R: A language and environment for statistical computing. R Foundation for Statis-
tical Computing, Vienna, Austria. URL https://www.R-project.org/.
Face-mask ventilation performance in a low resource setting
PLOS ONE | https://doi.org/10.1371/journal.pone.0186731 October 26, 2017 9 / 9
