Article Text
PDF
Abstract
Objective We sought to determine the effect of stimulation during positive pressure ventilation (PPV) on the number of spontaneous breaths, exhaled tidal volume (VTe), mask leak and obstruction.
Design Secondary analysis of a prospective, randomised trial comparing two face masks.
Setting Single-centre delivery room study.
Patients Newborn infants ≥34 weeks’ gestation at birth.
Methods Resuscitations were video recorded. Tactile stimulations during PPV were noted and the timing, duration and surface area of applied stimulus were recorded. Respiratory flow waveforms were evaluated to determine the number of spontaneous breaths, VTe, leak and obstruction. Variables were recorded throughout each tactile stimulation episode and compared with those recorded in the same time period immediately before stimulation.
Results Twenty of 40 infants received tactile stimulation during PPV and we recorded 57 stimulations during PPV. During stimulation, the number of spontaneous breaths increased (median difference (IQR): 1 breath (0–3); padj<0.001) and VTe increased (0.5 mL/kg (−0.5 to 1.7), padj=0.028), whereas mask leak (0% (−20 to 1), padj=0.12) and percentage of obstructed inflations (0% (0–0), padj=0.14) did not change, compared with the period immediately prior to stimulation. Increased duration of stimulation (padj<0.001) and surface area of applied stimulus (padj=0.026) were associated with a larger increase in spontaneous breaths in response to tactile stimulation.
Conclusions Tactile stimulation during PPV was associated with an increase in the number of spontaneous breaths compared with immediately before stimulation without a change in mask leak and obstruction. These data inform the discussion on continuing stimulation during PPV in term infants.
Trial registration number Australian and New Zealand Clinical Trial Registry (ACTRN12616000768493).
- intensive care units
- neonatal
- neonatology
- resuscitation
Data availability statement
Data are available upon reasonable request. De-identified individual participant data are available upon reasonable request from the corresponding author to researchers who provide a methodologically sound proposal, with approval by an independent review committee ('learned intermediary').
Statistics from Altmetric.com
What is already known on this topic?
Tactile stimulation is part of the newborn resuscitation algorithm prior to initiating positive pressure ventilation (PPV).
Tactile stimulation may assist in establishing spontaneous breathing, but objective data are lacking.
There is considerable variation between units in duration, and site of stimulation and their influence on the effectiveness of tactile stimulation are unclear.
What this study adds?
Infants are less likely to be stimulated during PPV compared with before PPV.
Tactile stimulation during PPV may improve spontaneous breathing and this effect may be stronger in longer stimulations covering a larger surface area.
Mask leak and airway obstruction do not appear to be affected by tactile stimulation during mask ventilation.
Introduction
Approximately 10% of newly born infants receive some support during the transition to extrauterine life and 3%–5% receive positive pressure ventilation (PPV).1–3 Current neonatal resuscitation guidelines recommend that infants with apnoea should be stimulated within the first minute after birth and in case of persistent apnoea or bradycardia, PPV should be initiated.2 3 If the infant is unresponsive, the Neonatal Resuscitation Program (NRP) guidelines suggest corrective steps in the ‘MRSOPA’ mnemonic (Mask adjustment, Reposition airway, Suction mouth and nose, Open mouth, Pressure increase, Alternate airway).4 Tactile stimulation is neither encouraged nor discouraged during PPV.2–4 Previous studies report that 50%–90% of infants are stimulated at least once after birth,5–8 but it remains unclear whether stimulation during PPV is beneficial or harmful.
Tactile stimulation may increase respiratory efforts5 and repetitive stimulation has been shown to improve oxygen saturation.9 Conversely, tactile stimulation during the provision of PPV may distract clinicians or induce infant movement, thus potentially increasing mask leak which could hinder effective ventilation.10 However, objective data on the benefits and adverse effects of stimulation during PPV are lacking.
Longer stimulations covering a larger surface area of the infant’s body may activate more mechanoreceptors, thus improving the response to tactile stimulation.11 There is considerable variation between units in duration, site and timing of stimulation,12 but their influence on the efficacy of tactile stimulation remains unclear.
Our primary aim was to assess the effect of tactile stimulations during PPV on the number of spontaneous breaths, as well as on expired tidal volume, mask leak and airway obstruction. A secondary aim was to evaluate the influence of duration of stimulation and the surface area to which stimulation was applied on the efficacy of tactile stimulation.
Methods
Population
This is a secondary analysis of a prospective, randomised clinical trial conducted at the Royal Women’s Hospital, Melbourne, comparing the effect of two face masks on mask leak during PPV in newborn infants ≥34 weeks’ gestation.13
Data collection
Resuscitations were video recorded from above, providing a view of the infant’s body and the operator’s hands.14 We reviewed the video recordings for tactile stimulations during the provision of PPV. For each stimulation, we recorded duration, site and onset of tactile stimulation. The initial act of drying the infant was included as a tactile stimulation as it often serves two purposes, that is, drying and stimulation.15 If there was a gap of at least 2 s between two stimulations, we recorded them as separate stimulations.5 To assess the overall area of stimulation, we divided the infant’s body surface into regions based on burn charts (ie, head, neck, chest, back, arms, legs, feet, genitalia).16 Since the thorax is one of the most stimulated regions during newborn transition,5 17 we further divided the thorax into four regions (back, chest and two sides of the thorax instead of only ventral and dorsal aspects). We counted any stimulation to a specific site as if the complete body part was stimulated. We then calculated the surface area over which the stimulus was applied based on the WHO burn chart.16 Corrective steps during face mask ventilation (ie, ‘MRSOPA’)4 were assessed between the onset of PPV and the first stimulation.
A Neopuff Infant Resuscitator (Fisher & Paykel Healthcare, Auckland, New Zealand) was used to provide PPV. Airway flow and pressure were measured using a flow sensor with an accuracy of ±5% placed between the T-piece ventilation device and the face mask. Respiratory function parameters were recorded at 200 Hz using the New Life Box recording system (Advanced Life Diagnostics, Weener, Germany). We assessed flow and pressure curves visually using Pulmochart software (Advanced Life Diagnostics) in a breath-by-breath analysis.
Intervention and control periods
For stimulations occurring during PPV, we assessed all outcomes during the entire duration of each stimulation plus 1 s after the stimulation (to be able to assess the effect of short stimulations) and compared it with the same time period immediately before the stimulation (ie, during stimulation vs before stimulation). If the control period of a stimulation overlapped with the intervention period of the preceding stimulation, overlapping outcome data were assessed in both periods (ie, preceding intervention period and current control period). If flow data were not assessable in the entire control period (eg, mask not placed on the face), the time frame for outcome evaluation was reduced to the same length in the intervention period to minimise systematic bias.
Outcomes
We compared the number of spontaneous breaths, respiratory rate (RR), exhaled tidal volume (VTe), leak and obstruction during stimulation with before stimulation. We counted a spontaneous breath if there was inspiratory and expiratory flow with corresponding opposing pressure waves.18 We calculated leak as the difference between inspired and expired tidal volume (VTi and VTe), divided by VTi,19 20 and compared leak before versus during stimulation. An obstructed inflation was defined as occurring if leak was <30% and VTe was <2 mL/kg.19 We then divided the number of obstructed inflations by the total number of inflations before and during stimulation, respectively, to account for different durations of stimulation (per cent of obstructed inflations). For VTe, we also calculated the coefficient of variation (SD of VTe divided by the mean) to assess the variability of tidal volumes before and during stimulation.
Effect modifiers
For an exploratory analysis, we prespecified the following potential effect modifiers: duration of stimulation, surface area of the applied stimulation, duration of PPV prior to the start of stimulation and gestational age, and we calculated their effect on the change in the number of spontaneous breaths as well as on the change in VTe, leak and obstruction.
Statistical analysis
Non-parametric data are presented as median and IQR. Differences in proportions were assessed using Χ2 test and differences in median values before versus during stimulation were analysed using a paired Wilcoxon test. The influence of potential effect modifiers was assessed by performing a linear or logistic regression (depending on the measurement scale of the dependent variable), accounting for within-subjects variance by using generalised estimating equations.21 As this is an exploratory analysis, tests were adjusted for multiple testing for each hypothesis separately using the Bonferroni-Holm method and adjusted p values of <0.05 were considered statistically significant.22 Data were analysed using R statistics, V.3.6.2.23
Results
Population
Forty infants had assessable video recordings. Infants were less likely to be stimulated during PPV compared with before PPV (20 of 40 infants (50 %) vs 39 of 40 infants (98%), p<0.001; figure 1). The median (IQR) number of stimulations before initiation of PPV was 2 (1–3). Table 1 displays patient characteristics of all infants who were stimulated during PPV.
Flow chart of analysed stimulations. PPV, positive pressure ventilation.
Population demographics of infants who were stimulated during PPV
Description of stimulations during PPV
Fifty-seven tactile stimulation episodes with evaluable flow data during PPV were analysed (figure 1). Median time between onset of PPV and first stimulation was 43 s (31–70). In 13 of 20 infants (65%), corrective steps for mask ventilation were performed before the first stimulation (mask readjustment and head repositioning in 13 of 20 infants (65%); suctioning and opening of the mouth in 2 of 20 infants (10%); pressure adjustment in 1 of 20 infants (5%)). All stimulations during PPV were performed by a second person not involved in mask ventilation. Stimulation episodes during PPV had a median (IQR) duration of 4 (3–9) s and 13% (8–24) of the total body surface area was covered. The most commonly included site of stimulation during PPV was the lateral thorax (44%), followed by anterior chest (40%), legs (37%), arms (30%), feet (25%), back (11%) and head (5%).
Effect of stimulations on spontaneous breathing efforts and tidal volumes
The number of spontaneous breaths increased significantly during stimulation (median 1 breath (IQR: 0–4) before stimulation vs 3 breaths (1–5) during stimulation; median paired difference: 1 (IQR: 0–3); padj<0.001, figure 2). Correspondingly, RR increased during stimulation (median 20 breaths/min (IQR: 0–38) before stimulation vs 36 breaths/min (20–60) during stimulation; median paired difference 13 (IQR: 0–30); padj<0.001). VTe also increased during stimulation (median 4.2 mL/kg (IQR: 2.7–6.2) before stimulation vs 5.0 mL/kg (3.8–7.3) during stimulation; paired difference: 0.5 mL/kg (−0.5 to 1.7), padj=0.021) and variability of VTe was greater during versus before stimulation (median coefficient of variation: 33% (IQR: 13–52) before stimulation vs 51% (38–78) during stimulation; paired difference: 18% (−6 to 48); padj=0.008). An example of an infant who started to breathe spontaneously during stimulation is shown in figure 3.
Number of spontaneous breaths before and during stimulation. Grey lines show individual changes before versus during stimulation.
Example recording of an infant who started to breathe spontaneously during tactile stimulation. Black vertical arrows indicate spontaneous breaths, the dotted vertical line indicates the time stimulation was initiated. PPV, positive pressure ventilation.
Effect of stimulations on mask leak and airway obstruction
Tactile stimulation during PPV affected neither leak (median 31% (IQR: 0–57) before vs 26% (0–51) during stimulation; paired difference: 0% (20% decrease to 1% increase), padj=0.124) nor the percentage of obstructed inflations (median 0% (IQR: 0–11) before vs 0% (0–0) during stimulation, paired difference: 0% (0–0), padj=0.14).
Influence of potential effect modifiers
A longer duration of stimulation (estimate=0.38, padj<0.001) and a larger surface area covered (estimate=0.10, padj=0.026) were associated with a greater increase in the number of spontaneous breaths. Duration of PPV prior to stimulation (estimate=0.0, padj=0.99) and gestational age (estimate=0.05, padj=0.99) did not affect the number of spontaneous breaths and none of the potential effect modifiers had any influence on the change in VTe, leak or obstruction (all p>0.05).
Discussion
In this retrospective observational study, we demonstrated that there was an increase in the number of spontaneous breaths and in VTe during stimulation compared with before stimulation without any effect on mask leak and airway obstruction. The increase in the number of spontaneous breaths was larger in longer stimulations and stimulations covering a larger surface area. There were no discernible adverse effects on mask leak.
We showed that once PPV was started, fewer infants received tactile stimulations compared with before PPV. Possibly, healthcare providers are primarily focused on establishing adequate ventilation and obtaining accurate monitoring during the provision of PPV. We speculate that the number of stimulations during PPV may be even lower in case of newborn stabilisation without a second person present. While international guidelines on newborn resuscitation recommend stimulation of infants with apnoea immediately after birth, stimulation during PPV is neither encouraged nor discouraged. This is likely due to a lack of data,2 3 leading to a degree of uncertainty, reflected by a decrease in stimulations.
In our study population, tactile stimulation during PPV was associated with an increase in the number of spontaneous breaths and an increase in respiratory rate compared with the period immediately before stimulation. Previously, tactile stimulation had been shown to increase the number of cries elicited during newborn transition, but mostly in infants without respiratory support.5 A randomised controlled trial showed a benefit of repetitive tactile stimulation on oxygen saturation and oxygen need in very preterm infants receiving respiratory support even though the minute volume was not improved.9 Our study is the first to analyse the direct effect of each tactile stimulation on spontaneous breaths during PPV in term infants. An increase of only one spontaneous breath per stimulation may have limited clinical relevance and may be due to the short duration of each stimulation. However, our findings may also reflect a true positive effect of tactile stimulation on the establishment of spontaneous breathing efforts, indicated by the concurrent increase in VTe. The higher variability of VTe during stimulation may be explained by the increasing number of spontaneous breaths during stimulation and their timing in relation to the inflation (ie, whether they are coinciding with higher peak inspiratory pressures or lower end-expiratory pressures). While excessive tidal volumes may be harmful in preterm infants,24 the effect of high tidal volume variability is unclear in term infants. It seems prudent to discontinue PPV once spontaneous breathing has commenced. Also, our study suggests that further investigation of repetitive stimulations in term infants during PPV is warranted as the combination of several stimulations may lead to an improved clinical outcome.
We demonstrated that duration of stimulation as well as surface area covered during each stimulation contribute to an increased effectiveness of tactile stimulation without negatively affecting mask leak. Previous studies in the delivery room had restricted their analyses to known areas of stimulation (ie, chest, back, side thorax and feet),5–8 while a recent study on tactile stimulation in the neonatal intensive care unit included all body parts in their report.25 As mechanoreceptors from any body part may contribute to stimulatory nerve signals,11 26 27 any stimulation may assist the onset of spontaneous breaths. Matching our findings, a larger covered surface area and a longer duration may activate a greater number of mechanoreceptors whose signals could be combined by temporal and spatial summation which may ultimately improve the clinical response.28 29 Additionally, high-threshold mechanoreceptors require longer or stronger stimuli to elicit a response.11 30 The effect of a larger surface area covered during stimulation is in line with recent findings that stimulation of the trunk is more effective in eliciting cry or movement compared with stimulating the foot.5 These exploratory analyses of potential effect modifiers need to be interpreted with caution but are important for generating new hypotheses which could then be evaluated prospectively. Due to the retrospective nature of this study, intensity of stimulation could not be assessed. Based on our findings, we speculate that more intense stimulations would be associated with a greater response, similar to duration and surface area covered.
Reassuringly, we demonstrated that there was no effect of tactile stimulation on mask leak during the provision of PPV if the stimulation is performed by a second person not involved in mask ventilation. Our findings could inform the discussion on continuing tactile stimulation during the provision of PPV. If a beneficial effect is demonstrated in prospective studies evaluating relevant clinical outcomes, the NRP mnemonic could be changed to include stimulation (ie, ‘MRS SOPA’ instead of ‘MR SOPA’). However, prospective studies are needed to evaluate the impact on relevant clinical outcomes.
Our study has various limitations. First, it is a retrospective observational study. However, using objective data from flow recordings improved reliability and validity of our study. Second, our study comprised only a small number of infants and stimulation episodes for analysis. Third, it is difficult to assess the clinician’s intentions using video recordings. Our data demonstrate that independent of the intention, stimulation may assist in improving spontaneous breathing in term-born infants. Fourth, due to the short duration of stimulations, we only assessed a short period of time and accordingly, only a small number of inflations and spontaneous breaths. Despite the short time frame, we were able to record an increase in the number of spontaneous breaths. Fifth, this was a single-centre study in a high-resource setting where sufficient staff was available for all births and results may differ in case of staff unavailability where tactile stimulation may hinder effective PPV. Sixth, this is a post-hoc analysis of a trial investigating mask leak and the awareness of the study aim may have led to an overall reduced mask leak during the study. However, since clinicians were blinded to the respiratory function monitor and this is a pre/post-analysis, this should not impact generalisability of the results. Our study provides new insight into the potential beneficial effect of tactile stimulation on spontaneous breathing during PPV using objective respiratory function monitor data and thus provides the foundation for future avenues of research, ideally to be conducted as larger prospective studies.
Conclusion
This retrospective observational study demonstrated that infants are less likely to be stimulated once PPV was initiated. Tactile stimulations performed by a second person during PPV of near-term and term-born infants were associated with an increase in the number of spontaneous breaths and VTe without an increase in mask leak. Longer stimulations and stimulations covering a larger surface area may contribute to a greater effect of tactile stimulations. Our findings may inform the discussion on continuing tactile stimulation during the provision of PPV in newly born infants.
Data availability statement
Data are available upon reasonable request. De-identified individual participant data are available upon reasonable request from the corresponding author to researchers who provide a methodologically sound proposal, with approval by an independent review committee ('learned intermediary').
Ethics statements
Patient consent for publication
Ethics approval
The original trial was approved by the local ethics committee (ref no 16/08). All parents provided written informed consent prior to commencement of the study.
Acknowledgments
We thank the patients, their parents and the staff of the Royal Women’s Hospital for their assistance in conducting the initial study.
References
Footnotes
Contributors All authors were involved in planning, conducting and reporting of the work. VDG and LS conceptualised and designed the study. VDG watched the videos and analysed the flow data, performed data analyses and wrote the first version of the manuscript. LS, CMR and EO collected data for the original trial. CMR, DB, EO, COFK, SBH and PGD provided essential intellectual input to data interpretation as well as manuscript writing. LS supervised the project. VDG is acting as the guarantor of the study. All authors approved the final version of the manuscript.
Funding This study was funded by the NHMRC Programme Grant 2017–2021 (App 1113902), (App ID 1059111, to PGD), (App ID 1073533, to COFK). VDG received an Endeavour Research Fellowship (Australia) (ERF_RDDH_5276_2016) and a Start-Up grant by the European Society for Paediatric Research. LS received a research fellowship from the German Research Society (DFG-grant LO 2162/1-1). CMR received an early Postdoc Mobility fellowship from the Swiss National Science Foundation (P2ZHP3_161749).
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- Highlights from this issue