Article Text
PDF
Abstract
Objective Mask positive pressure ventilation (PPV) in the delivery room is routinely delivered with set peak inflation pressures. To aid mask PPV, stand-alone respiratory function monitors (RFMs) have been used in the delivery room, while ventilator-based, volume-targeted ventilation (VTV) is routinely used in the neonatal intensive care unit (NICU).
Design This is a prospective, randomised, crossover simulation study. Participants were briefly trained to use a neonatal ventilator for volume-targeted mask ventilation (VTV-PPV), then performed mask ventilation on a manikin in a randomised order using VTV-PPV, T-piece PPV or T-piece PPV with RFM visible.
Setting In situ in a neonatal resuscitation room within a level 3 NICU.
Participants Healthcare professionals (HCPs) trained in neonatal resuscitation with experience as team leaders.
Interventions Semiautomated, ventilator-based VTV-PPV using two-hand hold versus manual PPV via a T-piece device (T-piece, RFM masked) versus manual PPV with RFM visible using one-hand hold.
Main outcome measures Respiratory characteristics including % mask leak, tidal volume (VT) and peak inflation pressure (PIP).
Results Thirty-two HCPs (23 (72%) female and 9 (28%) male) participated. The median mask leak was significantly lower with ‘VTV-PPV’ (11%, IQR 0%–14%) compared with both ‘T-piece, RFM visible’ (82%, IQR 30%–91%) and ‘T-piece, RFM masked’ (81%, IQR 47%–91%) (p<0.0001). The median delivered VT was 4.1 mL/kg (IQR 3.9–4.4) with VTV-PPV compared with 2.1 mL/kg (IQR 1.2–9) with T-piece, RFM visible and 1.8 mL/kg (IQR 1.1–5.8) with T-piece, RFM masked (p=0.0496). PIP was also significantly lower with VTV-PPV.
Conclusion During neonatal simulation, VTV-PPV reduced mask leak and allowed for consistent VT delivery compared with T-piece with and without RFM guidance.
- 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
WHAT IS ALREADY KNOWN ON THIS TOPIC
The delivered tidal volume (VT) during positive pressure ventilation (PPV) may lead to lung and brain injury.
Volume-targeted ventilation is routinely used in the neonatal intensive care unit and delivers targeted VT.
Volume-targeted mask ventilation was easily adapted by healthcare professionals into existing neonatal resuscitation algorithms.
WHAT THIS STUDY ADDS
A neonatal ventilator can be used with mask ventilation in a two-hand hold to provide targeted VT during simulated PPV.
A T-piece resuscitator, with a single healthcare professional using one-hand hold, with or without a respiratory function monitor delivers a wide range of VT.
Mask leak was significantly lower during ventilator PPV, which might be due to a focus on mask hold.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
A neonatal ventilator can be used to provide PPV during simulation with reduced mask leak and more stable VT.
This promising result could be partly due to the ability of a single healthcare professional to provide PPV using a two-hand hold.
Clinical testing of ventilator-based mask ventilation is warranted.
Background
For non-vigorous neonates immediately after birth, positive pressure ventilation (PPV) serves to create functional residual capacity, deliver adequate tidal volume (VT) to facilitate gas exchange and stimulate breathing.1 In the delivery room, PPV is routinely provided with pressure-limited devices (called T-piece resuscitators) via a face mask, where a peak inflation pressure (PIP) is chosen with the assumption that an adequate VT will be delivered.2 In reality, delivered VT is constantly changing as a result of the complex interplay between dynamic lung compliance, mask leak, airway obstruction, PIP and inspiratory time (Ti).3 Unfortunately, delivered VTs are rarely measured and therefore PIPs are rarely adjusted to optimise VT.2 Although resuscitation guidelines recommend judging the adequacy of PPV by assessing heart rate and chest wall movements,4 5 delivery room studies demonstrated that up to 75% of healthcare professionals (HCPs) are unable to accurately assess chest wall movement.2 A further problem is the large and variable gas leak between the mask and the face that can occur during mask ventilation, to which most resuscitators are often unaware.2 As VT and leak change during mask ventilation, HCPs might deliver a low, appropriate or excessive VT with the same PIP.6 A respiratory function monitor (RFM) can help to assess mask leak and VT delivery during mask PPV.7 However, while randomised trials demonstrated that RFM use can reduce high VT delivery, RFM use did not result in a reduction in bronchopulmonary dysplasia.8
Thus, better volume-targeting strategies for mask ventilation are required. In the neonatal intensive care unit (NICU), invasive volume-targeted ventilation (VTV) with a set VT is routinely used9 and has been associated with lower rates of combined death or bronchopulmonary dysplasia (typical relative risk (RR) 0.73, 95% CI 0.57 to 0.93) or periventricular leucomalacia or grade 3–4 intraventricual hemorrhage (IVH) (typical RR 0.48, 95% CI 0.28 to 0.84).9 However, clinicians have not adopted VTV for mask ventilation.
We have recently demonstrated that VTV-PPV is feasible and acceptable to HCPs and can be integrated into existing neonatal resuscitation algorithms.8 No study has yet compared mask PPV using a ventilator to provide VTV (VTV-PPV) to PPV with a T-piece with and without RFM. We hypothesised that VTV-PPV will significantly reduce mask leak compared with using a T-piece with either an RFM visible or masked during simulated mask ventilation in an infant manikin model.
Methods
This was a prospective, randomised, crossover simulation study carried out at the Royal Alexandra Hospital, Edmonton, a tertiary perinatal centre admitting ~350 infants with a birth weight of <1500 g to the NICU annually. HCPs were eligible to participate if they were members of our neonatal resuscitation team, all of whom are (1) Neonatal Resuscitation Program (NRP) instructors and/or providers of NRP with NRP certification, (2) experienced with mask ventilation in the delivery room and (3) experienced with VTV in the NICU.
Interventions
There were three interventions: ‘VTV-PPV’, ‘T-piece, RFM visible’ and ‘T-piece, RFM masked’. In all, HCPs provided mask PPV with a size 0 round silicone face mask (Laerdal, Stavanger, Norway). Flow sensors were placed between the mask and the ventilation device to record respiratory characteristics, including gas flow, VT and airway pressures.
For VTV-PPV, the mask was connected to a VN500 neonatal ventilator (Dräger Medical, Lübeck, Germany) capable of VTV. The following were the default settings: VT of 4 mL/kg, ventilation rate of 50 per minute, and a max PIP and positive end expiratory pressure (PEEP) of 40 cmH2O and 6 cmH2O. Respiratory variables were measured via a hot-wire anemometer from the VN500 ventilator. The participants assessed mask leak and VT delivery on the ventilator screen and adjusted their mask position to reduce mask leak. For troubleshooting, the participants increased the VT to 5 mL/kg in response to the ‘P’ (increase pressure) within the MR.SOPA mnemonic (M: open mouth of baby, R: reposition head, S: suction mouth and nose, O: open mouth, P: increase pressure).
For T-piece, RFM visible and for T-piece, RFM masked interventions, the mask was connected to a Neopuff T-piece (GE Healthcare, Chicago, Illinois), a continuous-flow, peak pressure-limited device with a manometer. The following were the default settings: PIP and PEEP of 24 cmH2O and 6 cmH2O, flow rate of 10 L/min, and ventilation rate of 40–60 per minute. The RFM used is the Respironics NM3 (Philips Healthcare, Philips Electronics, Markham, Ontario, Canada), which was also used to record gas flow, VT and airway pressure. In T-piece, RFM visible, the participants were instructed to observe the RFM and adjust their mask position to reduce mask leak (as deduced from a continuous expiratory flow and low VT) or to adjust the PIP to target a VT of 5 mL/kg. In T-piece, RFM masked, RFM was only used for data collection; the screen was masked from the participants, and the participants used a manometer and the sensation of their hands to assess mask leak.
Simulation protocol
We used a modified infant manikin (Resusci Baby, Laerdal Medical, Armonk, New York) fitted with a test lung and made leak-free. To fit the test lung, the manikin’s airway assembly tubing was directly connected to a neonatal test lung. The test lung was then fitted within the manikin’s chest cavity. The system was tested to ensure that 0% mask leak and 5 mL VT could be achieved with mask ventilation with the manikin using the ventilator. This confirms that the system is leak-free. Under ideal conditions (tight mask seal and sniffing neck position), this system achieves a VT of 5 mL using PIP of 17 and PEEP of 5, and Ti of 0.3. This manikin provided only chest rise as feedback to the participants. Other clinical signs (heart rate, respiratory effort, oxygen saturation) were provided verbally by the research assistants.
After providing written informed consent, the participants were randomly allocated to perform VTV-PPV, T-piece with RFM visible and T-piece with RFM masked in a set order. Randomisation was done using a computer-generated, block randomised sequence with variable-sized blocks. A sequentially numbered, sealed brown envelope containing the allocation was opened prior to the start of simulation.
The participants had up to 10 min to familiarise themselves with the ventilation devices and RFM. They then completed three similar scenarios, selected from the RETAIN Board Game,10 of preterm infants between 26 and 28 weeks’ gestation born via vaginal delivery. After deferred cord clamping for 60 s, the participants were expected to perform the initial steps while the baby was spontaneously breathing and supported the infant with continuous positive airway pressure (CPAP). Thirty seconds after starting CPAP, the infant became apnoeic and bradycardic and the participants were expected to start PPV. They were then expected to complete the first five steps of MR.SOPA, which would result in an increase in heart rate to 120–140 per minute. At this time, the participants provided CPAP for another 30 s before each scenario ended.
After completing all scenarios, the participants completed a questionnaire about their experience with VTV-PPV using a Likert scale (1=strongly disagree, 2=disagree, 3=neither agree nor disagree, 4=agree, 5=strongly agree).
Sample size calculation and data analysis
The primary outcome was % mask leak during PPV. Our previous data showed that the mean (SD) mask leak was 51% (20) using a T-piece. A sample size of 32 was sufficient to detect a clinically important (20%) reduction in mask leak from 51% to 31%, with 80% power, a two-tailed α error of 0.05 and an effect size of 1. Sample calculation was performed using the sample size calculation function in Stata V.12.
For all respiratory variables, the median value for each recording was calculated, followed by the mean or median of those medians. The results are presented as mean (SD) for normally distributed continuous variables and as median (IQR) when the distribution was skewed. The respiratory variables were compared using Kruskal-Wallis rank test. VT and mask leak were also compared by using post-hoc Wilcoxon matched-pairs signed-rank tests to compare each group pairing. In case only two groups were available (ie, peak inspiratory and expiratory flow), a two-sample Wilcoxon rank-sum test was used. All p values were two-sided, and p<0.05 was considered statistically significant. All statistical analyses were performed using Stata V.17.
Results
Data collection occurred in June 2022. Thirty-two HCPs (23 (72%) female and 9 (28%) male) participated in the study. The HCPs had their last NRP certification a median of 12 (IQR 4–14) months prior to participation and a median of 11 (IQR 5–20) years of experience.
The respiratory variables are presented in table 1 and figure 1. The median mask leak was significantly lower with VTV-PPV (11%, IQR 0%–14%) compared with both T-piece, RFM visible (82%, IQR 30%–91%) and T-piece, RFM masked (81%, IQR 47%–91%) (p<0.0001, Kruskal-Wallis); Wilcoxon matched-pairs signed-rank tests revealed a significant mask leak difference between VTV-PPV and T-piece, RFM visible (p<0.0001) and T-piece, RFM masked (p<0.0001), but no difference between T-piece with RFM visible and T-piece with RFM masked (p=0.7718) (figure 1 and table 1).
Respiratory variables
Mask leak (as % of expired tidal volume) and tidal volume (mL/kg). Box plots show median (solid bar), IQR (margins of the box) and 95% CI. PPV, positive pressure ventilation; RFM, respiratory function monitor; VTV, volume-targeted ventilation.
The median delivered VT was 4.1 mL/kg (IQR 3.9–4.4) with VTV-PPV compared with 2.1 mL/kg (IQR 1.2–9) with T-piece, RFM visible and 1.8 mL/kg (IQR 1.1–5.8) with T-piece, RFM masked (p=0.0496, Kruskal-Wallis); Wilcoxon matched-pairs signed-rank tests revealed a significantly different delivered VT between VTV-PPV and T-piece, RFM visible (p=0.032) and T-piece, RFM masked (p=0.0345), but no difference between T-piece with RFM visible and T-piece with RFM masked (p=0.9795).
The following were the proportions of VT delivery (within ±10% of 4 mL/kg (3.6–4.4), <3.6 mL/kg and >4.4 mL/kg): VTV-PPV, 62%, 18% and 20%; T-piece, RFM visible, 7%, 51% and 42%; and T-piece, RFM masked, 16%, 42% and 42%, respectively. The median PIP was also significantly lower with VTV-PPV (16 cmH2O compared with 27 cmH2O and 26 cmH2O, p<0.0001) (figure 1 and table 1).
Thirty participants provided survey responses (table 2). Most were neutral, comfortable or very comfortable with switching between CPAP and PPV during VTV-PPV (24/30, 80%), not directly managing PIP during VTV-PPV (23, 76%) and changing the MR.SOPA approach with VTV-PPV (22/30, 73%). However, in the free-text comments, the participants identified challenges to adapting corrective steps (MR.SOPA), required more training prior to clinical use and frustration with ventilator display (it takes an average of 15 s to see changes in mask leak on display).
Postsimulation survey responses
Discussion
To our knowledge, this is the first study to compare mask ventilation characteristics during simulated neonatal resuscitation using ventilator-based volume targeting (VTV-PPV) versus T-piece with an RFM either visible or masked. With VTV-PPV, (1) mask leak was significantly reduced, (2) the delivered VT was more likely within the target range of 4 mL/kg (figure 1), (3) the delivered VT was significantly higher and (4) the PIP was significantly lower (table 1).
A major problem of mask PPV is the often large and variable mask leak.2 Mask leak occurs frequently, even for experienced operators, and seriously affects the delivered VT.2 Although RFM use has been postulated to reduce mask leak, only two of three randomised trials comparing PPV with an RFM visible or masked in the delivery room were able to show this.8 In the current study, we did not observe any change in mask leak with an RFM visible compared with an RFM masked, while with VTV-PPV the mask leak was significantly reduced (figure 1). However, participants used a two-hand mask ventilation technique during VTV-PPV, which has previously shown to significantly reduce mask leak11; therefore, we cannot rule out that the lower mask leak in VTV-PPV was due to the two-hand technique alone.
Relying on manually set PIP and subjective chest rise assessments may result in harm by either underventilation or overventilation. Indeed, delivery room studies reported that VT during mask PPV ranges between 0 mL/kg and 31 mL/kg.2 This is concerning as animal studies reported that lung injury was predominantly caused by high VT and occurred within 2 min of starting ventilation.12 13 Therefore, it is prudent to closely target delivered VT during PPV. Studies using RFMs reported that the proportion of high VT can be reduced when an RFM is visible.8 However, this did not translate to a reduction in bronchopulmonary dysplasia (the pooled RR was 0.93, 95% CI 0.76 to 1.15, p=0.50). Using VTV-PPV might improve outcomes by providing tighter VT control than achieved by RFM use alone. Encouragingly, our study demonstrated that not only was VT more on target, PIPs were also lower. Our manikin/test lung model required only a PIP of 17 under ideal conditions to achieve a VT of 5 mL; therefore, if mask leak and VT targeting is performed well, we would expect a lower PIP. Tighter VT and lower PIP together suggests that VTV-PPV might reduce both volutrauma and barotrauma for the preterm lung right from birth.
In contrast, in situations where lung aeration is poor after a few initial PPV breaths, VTV-PPV may have the additional advantage of automatically delivering higher pressures required to achieve lung aeration, thus avoiding delays associated with HCPs’ manual decision-making and PIP adjustments. Simultaneously, VTV-PPV would maintain VT targeting and protection from barotrauma through the upper limits the ventilator imposes on PIPmax and through automatic downtitration of PIP as lung aeration improves.
Similar to our previous study focusing on HCP opinions and subjective workload of VTV-PPV versus T-piece,8 most participants in this study were neutral or comfortable with the technical aspects of VTV-PPV, including switching between mask ventilation and CPAP, integrating VTV-PPV with corrective steps and managing the VT rather than PIP. This provides further reassurance that HCP comfort and training will not be a barrier to clinical testing of VTV-PPV.
This study has several limitations. Our participants were experienced professionals with knowledge of VTV using a neonatal ventilator, limiting generalisability. Our simple manikin model did not replicate changing lung compliance that occurs after birth; dynamic lung compliance may further favour VTV-PPV as PIPs are automatically adjusted. Our model also did not test the interaction of spontaneous breathing effort, airway secretions and vocal cord movement; these factors may further influence VT delivery due to asynchrony with spontaneous breaths and airway obstruction. Clinical studies are thus warranted.
Conclusion
In a neonatal manikin model, VTV-PPV using a two-hand hold reduced mask leak and allowed for consistent VT delivery compared with T-piece with and without RFM guidance.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by the Human Research Ethics Board of the University of Alberta (Pro00118447). Participants gave informed consent to participate in the study before taking part.
References
Footnotes
Twitter @Research4Babies
Contributors GMS and BHYL were responsible for conception and design, collection and assembly of data, analysis and interpretation of data, drafting of the article, critical revision of the article for important intellectual content, and final approval of the article.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.