You are here

Article Text

Article menu
Download PDFPDF
Original research
Lung volume changes during apnoeas in preterm infants
  1. Vincent D Gaertner1,
  2. Andreas D Waldmann2,
  3. Peter G Davis3,4,5,
  4. Dirk Bassler1,
  5. Laila Springer6,
  6. David Gerald Tingay4,5,7,
  7. Christoph Martin Rüegger1
  1. 1 Newborn Research, Department of Neonatology, University Hospital and University of Zurich, Zurich, Switzerland
  2. 2 Department of Anesthesiology and Intensive Care Medicine, Rostock University Medical Center, Rostock, Germany
  3. 3 Newborn Research Centre and Neonatal Services, The Royal Women's Hospital, Melbourne, Victoria, Australia
  4. 4 The University of Melbourne, Melbourne, Victoria, Australia
  5. 5 Murdoch Children's Research Institute, Melbourne, Victoria, Australia
  6. 6 Department of Neonatology, University Children’s Hospital, Tübingen, Germany
  7. 7 Department of Neonatology, The Royal Children's Hospital, Parkville, Victoria, Australia
  1. Correspondence to Dr Vincent D Gaertner, Department of Neonatology, University Hospital Zurich, Zurich, Switzerland; vincent.gaertner{at}usz.ch

Abstract

Objective Mechanisms of non-invasive high-frequency oscillatory ventilation (nHFOV) in preterm infants are unclear. We aimed to compare lung volume changes during apnoeas in preterm infants on nHFOV and nasal continuous positive airway pressure (nCPAP).

Methods Analysis of electrical impedance tomography (EIT) data from a randomised crossover trial comparing nHFOV with nCPAP in preterm infants at 26–34 weeks postmenstrual age. EIT data were screened by two reviewers to identify apnoeas ≥10 s. End-expiratory lung impedance (EELI) and tidal volumes (VT) were calculated before and after apnoeas. Oxygen saturation (SpO2) and heart rate (HR) were extracted for 60 s after apnoeas.

Results In 30 preterm infants, 213 apnoeas were identified. During apnoeas, oscillatory volumes were detectable during nHFOV. EELI decreased significantly during apnoeas (∆EELI nCPAP: −8.0 (−11.9 to −4.1) AU/kg, p<0.001; ∆EELI nHFOV: −3.4 (−6.5 to −0.3), p=0.03) but recovered over the first five breaths after apnoeas. Compared with before apnoeas, VT was increased for the first breath after apnoeas during nCPAP (∆VT: 7.5 (3.1 to 11.2) AU/kg, p=0.001). Falls in SpO2 and HR after apnoeas were greater during nCPAP than nHFOV (mean difference (95% CI): SpO2: 3.6% (2.7 to 4.6), p<0.001; HR: 15.9 bpm (13.4 to 18.5), p<0.001).

Conclusion Apnoeas were characterised by a significant decrease in EELI which was regained over the first breaths after apnoeas, partly mediated by a larger VT. Apnoeas were followed by a considerable drop in SpO2 and HR, particularly during nCPAP, leading to longer episodes of hypoxemia during nCPAP. Transmitted oscillations during nHFOV may explain these benefits.

Trial registration number ACTRN12616001516471.

  • intensive care units, neonatal
  • neonatology
  • respiratory medicine

Data availability statement

Data are available on reasonable request. De-identified individual participant data and statistical analysis codes are available from 3 months to 2 years following article publication to researchers who provide a methodologically sound proposal, with approval by an independent review committee (‘learned intermediary’). Proposals should be directed to christoph.rueegger@usz.ch to gain access. Data requestors will need to sign a data access or material transfer agreement approved by USZ.

http://dx.doi.org/10.1136/archdischild-2022-324282

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    What is already known on this topic

    • Apnoea of prematurity is a developmental phenomenon in preterm infants which is associated with subsequent hypoxemia and bradycardia.

    • A longer time spent in hypoxemia is associated with worse neurodevelopmental outcomes at 18–24 months corrected age.

    • Non-invasive high-frequency oscillatory ventilation (nHFOV) decreases the number of desaturations and bradycardia in preterm infants when compared with nasal continuous positive airway pressure (nCPAP), but the underlying mechanisms are unclear.

    What this study adds

    • Functional residual capacity dropped during apnoeas but was subsequently regained over the first five breaths after the apnoea, partly mediated by a larger VT.

    • Apnoeas were followed by a considerable drop in SpO2 and HR in both modes of ventilation with a longer time spent in hypoxemia during nCPAP.

    • Oscillatory volumes transmitted to the lung level during nHFOV may alleviate lung volume losses and subsequent adverse clinical effects.

    How this study might affect research, practice or policy

    • Escalating respiratory support to nHFOV may be sensible in selected cases with a high number of apnoeas before considering endotracheal intubation.

    • The role of different levels of positive distending pressures and frequencies needs to be evaluated in further studies.

    Introduction

    Apnea of prematurity is a developmental phenomenon in preterm infants due to immaturity of breathing control and is often associated with subsequent hypoxemia and bradycardia.1 2 A longer time spent in hypoxemia is associated with worse neurodevelopmental outcomes at 18–24 months corrected age.3

    Common measures to decrease apnoeas and concomitant episodes of desaturations and bradycardia include stimulation, postural adjustments, nasal respiratory support and the administration of caffeine.4–6 Non-invasive high-frequency oscillatory ventilation (nHFOV) was recently shown to decrease the number of desaturations and bradycardia in preterm infants,7 but the underlying mechanisms are unclear.

    Previously, we demonstrated that oscillatory volumes (VOsc) are detectable at lung level during nHFOV.8 During apnoeas, this may alleviate lung volume losses and subsequent adverse clinical effects. However, this remains incompletely understood.

    Using electrical impedance tomography (EIT) data from a previous randomised crossover trial comparing nHFOV with nasal continuous positive airway pressure (nCPAP), we aimed to (1) establish whether oscillations generated by nHFOV are detectable at lung level during apnoeas, and (2) assess intrapulmonary volume changes as well as (3) cardiorespiratory parameters during and after apnoeas.

    Materials and methods

    Study design and data collection

    Setup of the original study and methods of data collection and extraction were described previously and are provided in the online supplemental material.7–9 A textile electrode belt with 32 electrodes was fastened at nipple level.10 During each intervention period, four 10 min EIT sequences were recorded using the SenTec BB2 EIT device (SenTec AG, Landquart, Switzerland) at a frame rate of 48 Hz in a custom-built infant imaging package.10 11 The EIT belt remained on the infant’s thorax throughout the entire study.8 Recordings were excluded from analysis if more than three electrodes had insufficient skin contact. Oxygen saturation (SpO2) and heart rate (HR) were recorded using a Masimo pulse oximeter with a 2 s averaging time (Masimo Radical 7; Masimo Cooperation, Irvine, CA, USA).

    Data extraction and analysis

    EIT signals inside predefined anatomical lung regions (based on the vendor-provided human model chest atlas) were identified for the entire EIT sequences.12 13 Apnoeas were defined as episodes without a spontaneous breath for more than 10 s and the following steps were taken for data extraction: (1) An automated algorithm was programmed in Matlab (V.2019a; MathWorks, Natick, MA, USA) to detect and extract all episodes without a spontaneous breath for at least 5 s within each 10 min EIT sequence. This approach was chosen in order not to miss episodes of more than 10 s duration in case of temporal inaccuracy of the algorithm. (2) All extracted sequences were then visually screened by two independent reviewers for correct identification of apnoeic episodes (VDG and CMR) and disagreements were resolved by consensus. Episodes without a spontaneous breath for more than 10 s were extracted for final analysis. (3) After visual identification and manual selection of beginning and end of each apnoea, a bandpass filter was applied at 8 and 16 Hz (corresponding to the set frequency during nHFOV and its second harmonic) to assess the magnitude of oscillatory volumes (VOsc) during the apnoeic episode. (4) The last five breaths before apnoea (=baseline) and the first five breaths after apnoea were selected without a bandpass filter to assess the spontaneous breathing signal before and after the apnoea. End-expiratory as well as end-inspiratory lung volumes (EELI and EILI, respectively) were measured using arbitrary units (AU) and tidal volumes (VT) were calculated by subtracting EELI from EILI. (6) EIT variables (ie, VOsc, EELI, EILI, VT) were normalised for body weight. (7) Changes from baseline to the first five breaths after apnoea as well as changes from the first to fifth breath after apnoea were calculated for EELI and VT, respectively.

    Finally, corresponding cardiorespiratory data were extracted for each apnoea separately. Mean SpO2 and HR over the last 30 s before apnoea (=baseline) as well as minimum SpO2 and HR within 60 s after the end of each apnoea were noted and changes to baseline were calculated. Time spent below 80% SpO2 and time with HR below 80 bpm were recorded within 60 s after each apnoea.

    Statistical analysis

    Normally distributed data are presented as mean with SD or 95% CI. Non-parametric data are presented as median and IQR. Changes in EIT data or cardiorespiratory data during apnoeas were analysed using a paired Wilcoxon or t-test, depending on data distribution. Since more than one apnoea could be noted per infant and mode of ventilation, comparisons between nHFOV and nCPAP were performed using a mixed-model analysis of variance controlling for within-subjects variance (using the ‘afex’ package in R statistics, V.3.6.2).14 This means that statistical findings from this analysis relate to within-infants differences and thereby to differences between the two modes of respiratory support. P values <0.05 were considered statistically significant.

    Results

    Population

    Of 40 infants enrolled in the original trial, 10 had no or insufficient EIT data available, leaving 30 infants for this analysis. In 228 EIT recordings (112 during nHFOV and 116 during nCPAP), 3447 apnoeas were identified by the algorithm. After visual screening, 213 apnoeas were eligible for final analysis, 89 during nCPAP and 124 during nHFOV (figure 1). Demographic characteristics of the included infants are provided in table 1.

    Figure 1
    Figure 1

    Flowchart of extracted apnoeas. nCPAP, nasal continuous positive airway pressure; nHFOV, non-invasive high-frequency oscillatory ventilation.

    View this table:
    Table 1

    Baseline demographics and clinical characteristics (N=30 infants)

    Transmission of oscillatory volumes during apnoeas

    Mean (SD) oscillatory amplitude was significantly larger during nHFOV than during nCPAP (1.6 (1.0) AU/kg vs 0.1 (0.06) AU/kg; F=184, p<0.001). Figure 2 shows a typical example.

    Figure 2
    Figure 2

    Example of raw impedance signal and oscillatory signal during nCPAP and nHFOV. One exemplary apnoeic phase during both modes of respiratory support and the corresponding oscillatory signal after application of a bandpass filter at 8 and 16 Hz. Colours also correspond to colours in figure 3: blue=baseline before apnoea, red=apnoea, yellow=first breath after apnoea, green=fifth breath after apnoea. nCPAP, nasal continuous positive airway pressure; nHFOV, non-invasive high-frequency oscillatory ventilation; ∆Z, raw impedance changes; ∆ZOsc, oscillatory signal.

    Lung volumes during and after apnoeas

    Baseline EELI before apnoeas was comparable between nHFOV and nCPAP (mean difference (95% CI): 0.4 (−0.2 to 1.1) AU/kg, p=0.19). EELI dropped during apnoeas in both modes of ventilation but recovered over the first five breaths after the episode. There was a non-significant trend towards a larger drop during nCPAP compared with nHFOV (figure 3A and online supplemental table 1).

    Figure 3
    Figure 3

    Comparison of intrapulmonary volume changes during nCPAP and nHFOV. Part A: changes in end-expiratory lung impedance (∆EELI) compared with baseline EELI before apnoea. Part B: changes in tidal volumes (∆VT) compared with baseline VT before apnoea. Colours correspond to colours in figure 2: blue=baseline before apnoea, red=apnoea, yellow=first breath after apnoea, green=fifth breath after apnoea. Asterisks mark significant differences: ***p<0.001, **p<0.01, *p<0.05. Exact values and mean differences are shown in online supplemental tables 1 and 2, respectively. AU/kg, arbitrary units per kilogram body weight; nCPAP, nasal continuous positive airway pressure; nHFOV, non-invasive high-frequency oscillatory ventilation.

    The VT of the first breath after apnoea was increased during nCPAP but not during nHFOV (mean difference nHFOV minus nCPAP −7.0 (−12.7 to −1.4), p=0.017). For the subsequent breaths after apnoea, VT were similar to before apnoea during both modes (figure 3B and online supplemental table 2).

    Cardiorespiratory parameters after apnoeas

    During both modes of respiratory support, SpO2 and HR dropped significantly within 60 s after the apnoea, but the fall was greater during nCPAP (mean difference nHFOV minus nCPAP (95% CI) ∆SpO2: 3.6% (2.7% to 4.6%), p<0.001; ∆HR: 15.9 (13.4 to 18.5) beats/min, p<0.001; table 2).

    View this table:
    Table 2

    Changes in cardiorespiratory parameters within 60 s after the apnoeic episode

    SpO2 dropped below 80% after 21/89 apnoeas (24%) during nCPAP and 12/124 apnoeas (10%) during nHFOV. The time spent below 80% SpO2 was longer during nCPAP compared with nHFOV (mean difference (95% CI): 3.3 (1.4 to 5.2) s, p=0.001). Heart rate was lower than 80 bpm after 5/89 apnoeas (6%) during nCPAP and 0/124 apnoeas (0%) during nHFOV.

    Discussion

    Using EIT data, we were able to detect oscillatory volumes at lung level during apnoeas in preterm infants supported on nHFOV. Apneas were characterised by a decrease in EELI which was subsequently regained over the first five breaths after the apnoea, partly mediated by a larger VT of the first breath after apnoea. This entire succession of events was more pronounced during nCPAP and accordingly, the drop in SpO2 and HR as well as time spent in hypoxemia after apnoeas were greater during nCPAP. We speculate that transmission of the oscillatory waveform during nHFOV may play a role in reducing episodes of desaturations and bradycardia after apnoeas in preterm infants.

    In the current study, we could show that oscillations generated during nHFOV were detectable at lung level during apnoeas. The magnitude of transmitted VOsc to the lung level was slightly smaller than previously reported during spontaneous breathing.8 While many apnoeas in preterm infants are of central origin, most show an obstructive component due to the high compliance of the nasopharynx,15 16 and increased supraglottic resistance during and after apnoeas.17 However, providing positive distending pressure may stabilise the nasopharynx, thereby reducing the obstructive component of apnoeas and allowing some gas flow (including oscillations) to pass the glottis to the infant’s lungs,18 potentially explaining the slightly reduced magnitude of VOsc during apnoeas. While the number of apnoeas was even increased during nHFOV in this study, we speculate that transmission of oscillations during apnoeas may have improved cardiorespiratory stability.

    There was a significant loss of EELI (corresponding to functional residual capacity; FRC) during apnoeas in both modes of respiratory support. Preterm infants have a high chest wall compliance with a strong influence of the physiological lung recoil.19 20 This may lead to alveolar collapse in the absence of own respiratory efforts and subsequently to a loss in FRC with corresponding desaturations and bradycardia. Interestingly, EELI dropped slightly more during nCPAP compared with nHFOV (approximately half a regular VT during nCPAP and one-fifth of a regular VT during nHFOV). While this comparison did not reach statistical significance, it may partly explain the findings of the original clinical trial where nHFOV was associated with less desaturations and bradycardia compared with nCPAP.7 In the current study, FRC was quickly regained over the first five breaths after the apnoea, similar to the recovery seen in a previous study evaluating lung volumes after extubation in preterm infants.21 In infants without respiratory support, FRC seems to remain low after apnoeas,22 and thus, we speculate that positive distending pressure stabilises the nasopharynx, thereby assisting in rapid FRC gain. However, the role of applied pressure levels on apnoea needs to be evaluated in future studies, ideally also including other types of respiratory support such as non-synchronised and synchronised non-invasive positive pressure ventilation.

    After apnoeas, spontaneous breaths during nCPAP were often characterised by an initial large breath or sigh. This is unsurprising as one of the functions of sighs is the restoration of lost FRC.22 It is possible that the larger tidal volumes may be partly caused by the Hering-Breuer deflation reflex, which leads to increased VT after FRC loss and has been shown to be important in both term and preterm infants.23 24 However, these initial studies had been performed using an inflatable jacket to achieve expiration which is different to spontaneously breathing infants.23 24 In addition, oscillations during nHFOV may have contributed to carbon dioxide (CO2) clearance during apnoeas. In contrast, no ventilation occurred during nCPAP and consequently, increased VT may be needed for CO2 clearance. In a previous study in infants without respiratory support, VT remained increased for the first five breaths after apnoeas.17 In contrast, VT in our study returned to pre-apnoea levels in both modes of respiratory support after the first breath. We speculate that applying positive distending pressure alleviates the need for a higher VT. It is important to note that our data were collected in relatively stable preterm infants and it remains unclear whether less mature infants are able to regain FRC as quickly. However, we speculate that the same pathophysiological mechanism could be important in infants with a larger disease burden, but this needs to be investigated in subsequent physiological and clinical investigations.

    Apnoea of prematurity is an important problem for preterm infants as it may impact their long-term neurodevelopment.3 Common measures to reduce concomitant desaturations and bradycardia such as non-invasive respiratory support or caffeine are already widely implemented into clinical practice.6 25–27 However, the use of other measures including the use of nHFOV is still debated and mainly guided by clinicians’ personal preference.28 29 In this study, we demonstrated a pronounced decrease in FRC which was followed by a decrease in SpO2 and HR during both modes of respiratory support. This is not surprising as there is a known correlation between FRC loss and desaturations in preterm infants.30 More importantly, the decreases in SpO2 and HR in the current study were more pronounced during nCPAP even though baseline EELI was largely comparable between the two modes of respiratory support. It is important to note that FiO2 levels were higher during nHFOV than during nCPAP (0.31 vs 0.28), as reported in the original publication from this dataset.7 This may have influenced baseline SpO2 levels as well as hypoxemic spells during the two modes of respiratory support. Still, there were more apnoeas followed by desaturations <80% and bradycardia <80 bpm, leading to a longer time spent in hypoxemia during nCPAP compared with nHFOV despite more apnoeic episodes during nHFOV. This is in line with the findings from the original trial.7 Thus, our results may provide rationale for escalating respiratory support to nHFOV in selected cases with a high number of apnoeas before considering endotracheal intubation. However, these findings need to be interpreted with caution because they are based on stable infants and SpO2 and HR remained above 80% and 80 bpm for most apnoeas. As the time spent in hypoxemia is particularly relevant for long-term neurodevelopment,3 the clinical importance of this finding needs to be investigated in adequately powered clinical trials.

    This study has several limitations. First, due to the crossover nature of the study, we were unable to evaluate longer-term clinical effects of nHFOV. In addition, the study population consisted of stable preterm infants who experienced few apnoeas leading to pronounced hypoxemia or bradycardia. We speculate that the direction of results would be similarly favourable for nHFOV in less stable infants. In fact, a recent large multicentre trial demonstrated that nHFOV used as post-extubation respiratory support reduced duration of mechanical ventilation compared with nCPAP or non-invasive intermittent positive pressure ventilation.31 Second, EIT data only show relative changes rather than absolute lung volumes. However, differences are independent of absolute values and may more accurately predict clinical benefits. Furthermore, EIT measurements correlate well with measured flow data in animal studies,32 33 and, in preterm infants, data are representative for the whole lung.34 Third, we studied a small sample of 30 infants. However, we identified a large number of apnoeas across these infants and the consistent results strengthens the validity of our study. Fourth, we were not able to differentiate between central, obstructive or mixed apnoeas. Future studies should evaluate the impact of nHFOV in these different types of apnoeas. Fifth, reviewers of the EIT recordings were unblinded to the type of respiratory support. Finally, this was a single-centre study with fixed ventilator settings, and all infants remained in prone position throughout the study. Using other devices, settings or body positions may lead to different findings.35

    Conclusion

    In this study, we used EIT to demonstrate the development of lung volumes during and after apnoeas in preterm infants supported on nHFOV and nCPAP. In both modes of respiratory support, FRC dropped during apnoeas and was subsequently regained over the first five breaths after the apnoea, partly mediated by a larger VT. Finally, apnoeas were followed by a considerable drop in SpO2 and HR. This entire succession of events was more pronounced during nCPAP compared with nHFOV, including a longer time spent in hypoxemia during nCPAP. Transmitted oscillations during nHFOV may partly explain these benefits.

    Data availability statement

    Data are available on reasonable request. De-identified individual participant data and statistical analysis codes are available from 3 months to 2 years following article publication to researchers who provide a methodologically sound proposal, with approval by an independent review committee (‘learned intermediary’). Proposals should be directed to christoph.rueegger@usz.ch to gain access. Data requestors will need to sign a data access or material transfer agreement approved by USZ.

    Ethics statements

    Patient consent for publication

    Ethics approval

    The original trial was registered with the Australian and New Zealand Clinical Trials Registry. This study involves human participants and was approved by the local ethics committee of the Royal Women’s Hospital Melbourne (Ref No 16/32). All parents provided written informed consent prior to the study.

    Acknowledgments

    We thank all the parents and infants who participated in the study and the staff at the neonatal intensive care unit of The Royal Women’s Hospital, Melbourne, Australia.

    References

    1. ↵
      2. Martin RJ ,
      3. Wang K ,
      4. Köroğlu O , et al
      . Intermittent hypoxic episodes in preterm infants: do they matter? Neonatology 2011;100:303–10.doi:10.1159/000329922 pmid:http://www.ncbi.nlm.nih.gov/pubmed/21986336
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      2. Eichenwald EC ,
      3. Watterberg KL ,
      4. Aucott S , et al
      . Apnea of prematurity. Pediatrics 2016;137.doi:10.1542/peds.2015-3757
    3. ↵
      2. Poets CF ,
      3. Roberts RS ,
      4. Schmidt B , et al
      . Association between intermittent hypoxemia or bradycardia and late death or disability in extremely preterm infants. JAMA 2015;314:595–603.doi:10.1001/jama.2015.8841 pmid:http://www.ncbi.nlm.nih.gov/pubmed/26262797
      OpenUrlCrossRefPubMed
    4. ↵
      2. Jenni OG ,
      3. von Siebenthal K ,
      4. Wolf M , et al
      . Effect of nursing in the head elevated tilt position (15 degrees) on the incidence of bradycardic and hypoxemic episodes in preterm infants. Pediatrics 1997;100:622–5.doi:10.1542/peds.100.4.622 pmid:http://www.ncbi.nlm.nih.gov/pubmed/9310515
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      2. Bader D ,
      3. Tirosh E ,
      4. Hodgins H , et al
      . Effect of increased environmental temperature on breathing patterns in preterm and term infants. J Perinatol 1998;18:5–8.pmid:http://www.ncbi.nlm.nih.gov/pubmed/9527936
      OpenUrlPubMed
    6. ↵
      2. Schmidt B ,
      3. Roberts RS ,
      4. Davis P , et al
      . Caffeine therapy for apnea of prematurity. N Engl J Med 2006;354:2112–21.doi:10.1056/NEJMoa054065 pmid:http://www.ncbi.nlm.nih.gov/pubmed/16707748
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      2. Rüegger CM ,
      3. Lorenz L ,
      4. Kamlin COF , et al
      . The effect of noninvasive high-frequency oscillatory ventilation on desaturations and bradycardia in very preterm infants: a randomized crossover trial. J Pediatr 2018;201:269–73.doi:10.1016/j.jpeds.2018.05.029 pmid:http://www.ncbi.nlm.nih.gov/pubmed/29954606
      OpenUrlPubMed
    8. ↵
      2. Gaertner VD ,
      3. Waldmann AD ,
      4. Davis PG , et al
      . Transmission of oscillatory volumes into the preterm lung during noninvasive high-frequency ventilation. Am J Respir Crit Care Med 2021;203:998–1005.doi:10.1164/rccm.202007-2701OC pmid:http://www.ncbi.nlm.nih.gov/pubmed/33095994
      OpenUrlCrossRefPubMed
    9. ↵
      2. Gaertner VD ,
      3. Waldmann AD ,
      4. Davis PG , et al
      . Lung volume distribution in preterm infants on non-invasive high-frequency ventilation. Arch Dis Child Fetal Neonatal Ed 2022;107:551–7.doi:10.1136/archdischild-2021-322990 pmid:http://www.ncbi.nlm.nih.gov/pubmed/35101993
      OpenUrlAbstract/FREE Full Text
    10. ↵
      2. Sophocleous L ,
      3. Frerichs I ,
      4. Miedema M , et al
      . Clinical performance of a novel textile interface for neonatal chest electrical impedance tomography. Physiol Meas 2018;39:044004.doi:10.1088/1361-6579/aab513 pmid:http://www.ncbi.nlm.nih.gov/pubmed/29516865
      OpenUrlCrossRefPubMed
    11. ↵
      2. Adler A ,
      3. Lionheart WRB
      . Uses and abuses of EIDORS: an extensible software base for EIT. Physiol Meas 2006;27:S25–42.doi:10.1088/0967-3334/27/5/S03
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      2. Ukere A ,
      3. März A ,
      4. Wodack KH , et al
      . Perioperative assessment of regional ventilation during changing body positions and ventilation conditions by electrical impedance tomography. Br J Anaesth 2016;117:228–35.doi:10.1093/bja/aew188 pmid:http://www.ncbi.nlm.nih.gov/pubmed/27440635
      OpenUrlCrossRefPubMed
    13. ↵
      2. Tingay DG ,
      3. Waldmann AD ,
      4. Frerichs I , et al
      . Electrical impedance tomography can identify ventilation and perfusion defects: a neonatal case. Am J Respir Crit Care Med 2019;199:384–6.doi:10.1164/rccm.201808-1551LE pmid:http://www.ncbi.nlm.nih.gov/pubmed/30365365
      OpenUrlCrossRefPubMed
    14. ↵
      1. R Core Team
      . R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing, 2013. http://www.r-project.org/
    15. ↵
      2. Finer NN ,
      3. Barrington KJ ,
      4. Hayes BJ , et al
      . Obstructive, mixed, and central apnea in the neonate: physiologic correlates. J Pediatr 1992;121:943–50.doi:10.1016/S0022-3476(05)80349-X pmid:http://www.ncbi.nlm.nih.gov/pubmed/1447664
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      2. Mathew OP ,
      3. Roberts JL ,
      4. Thach BT
      . Pharyngeal airway obstruction in preterm infants during mixed and obstructive apnea. J Pediatr 1982;100:964–8.doi:10.1016/S0022-3476(82)80529-5 pmid:http://www.ncbi.nlm.nih.gov/pubmed/7086601
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      2. Miller MJ ,
      3. Petrie TG ,
      4. Difiore JM
      . Changes in resistance and ventilatory timing that accompany apnea in premature infants. J Appl Physiol 1993;75:720–3.doi:10.1152/jappl.1993.75.2.720 pmid:http://www.ncbi.nlm.nih.gov/pubmed/8226474
      OpenUrlPubMed
    18. ↵
      2. Miller MJ ,
      3. Carlo WA ,
      4. Martin RJ
      . Continuous positive airway pressure selectively reduces obstructive apnea in preterm infants. J Pediatr 1985;106:91–4.doi:10.1016/S0022-3476(85)80475-3 pmid:http://www.ncbi.nlm.nih.gov/pubmed/3917503
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      2. Gerhardt T ,
      3. Bancalari E
      . Chestwall compliance in full-term and premature infants. Acta Paediatr Scand 1980;69:359–64.doi:10.1111/j.1651-2227.1980.tb07093.x pmid:http://www.ncbi.nlm.nih.gov/pubmed/7376862
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      2. Heldt GP ,
      3. McIlroy MB
      . Dynamics of chest wall in preterm infants. J Appl Physiol 1987;62:170–4.doi:10.1152/jappl.1987.62.1.170 pmid:http://www.ncbi.nlm.nih.gov/pubmed/3558177
      OpenUrlPubMedWeb of Science
    21. ↵
      2. Plastina L ,
      3. Gaertner VD ,
      4. Waldmann AD , et al
      . The DELUX study: development of lung volumes during extubation of preterm infants. Pediatr Res 2021. doi:doi:10.1038/s41390-021-01699-w. [Epub ahead of print: 31 Aug 2021].pmid:http://www.ncbi.nlm.nih.gov/pubmed/34465873
    22. ↵
      2. Poets CF ,
      3. Rau GA ,
      4. Neuber K , et al
      . Determinants of lung volume in spontaneously breathing preterm infants. Am J Respir Crit Care Med 1997;155:649–53.doi:10.1164/ajrccm.155.2.9032208 pmid:http://www.ncbi.nlm.nih.gov/pubmed/9032208
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      2. Marsh MJ ,
      3. Fox GF ,
      4. Hoskyns EW , et al
      . The Hering-Breuer deflationary reflex in the newborn infant. Pediatr Pulmonol 1994;18:163–9.doi:10.1002/ppul.1950180308 pmid:http://www.ncbi.nlm.nih.gov/pubmed/7800433
      OpenUrlPubMed
    24. ↵
      2. Hannam S ,
      3. Ingram DM ,
      4. Milner AD
      . A possible role for the Hering-Breuer deflation reflex in apnea of prematurity. J Pediatr 1998;132:35–9.doi:10.1016/S0022-3476(98)70481-0 pmid:http://www.ncbi.nlm.nih.gov/pubmed/9469997
      OpenUrlPubMed
    25. ↵
      2. Hatch LD ,
      3. Clark RH ,
      4. Carlo WA , et al
      . Changes in use of respiratory support for preterm infants in the US, 2008–2018. JAMA Pediatr 2021;175:1017–24.doi:10.1001/jamapediatrics.2021.1921 pmid:http://www.ncbi.nlm.nih.gov/pubmed/34228110
      OpenUrlPubMed
    26. ↵
      2. Patel RM ,
      3. Zimmerman K ,
      4. Carlton DP , et al
      . Early caffeine prophylaxis and risk of failure of initial continuous positive airway pressure in very low birth weight infants. J Pediatr 2017;190:108–11.doi:10.1016/j.jpeds.2017.07.006 pmid:http://www.ncbi.nlm.nih.gov/pubmed/28890204
      OpenUrlPubMed
    27. ↵
      2. Sand L ,
      3. Szatkowski L ,
      4. Kwok T'ng Chang , et al
      . Observational cohort study of changing trends in non-invasive ventilation in very preterm infants and associations with clinical outcomes. Arch Dis Child Fetal Neonatal Ed 2022;107:150–5.doi:10.1136/archdischild-2021-322390 pmid:http://www.ncbi.nlm.nih.gov/pubmed/34413093
      OpenUrlAbstract/FREE Full Text
    28. ↵
      2. Fischer HS ,
      3. Bohlin K ,
      4. Bührer C , et al
      . Nasal high-frequency oscillation ventilation in neonates: a survey in five European countries. Eur J Pediatr 2015;174:465–71.doi:10.1007/s00431-014-2419-y pmid:http://www.ncbi.nlm.nih.gov/pubmed/25227281
      OpenUrlCrossRefPubMed
    29. ↵
      2. De Luca D ,
      3. Dell'Orto V
      . Non-Invasive high-frequency oscillatory ventilation in neonates: review of physiology, biology and clinical data. Arch Dis Child Fetal Neonatal Ed 2016;101:F565–70.doi:10.1136/archdischild-2016-310664 pmid:http://www.ncbi.nlm.nih.gov/pubmed/27354382
      OpenUrlAbstract/FREE Full Text
    30. ↵
      2. Tourneux P ,
      3. Léké A ,
      4. Kongolo G , et al
      . Relationship between functional residual capacity and oxygen desaturation during short central apneic events during sleep in "late preterm" infants. Pediatr Res 2008;64:171–6.doi:10.1203/PDR.0b013e318179951d pmid:http://www.ncbi.nlm.nih.gov/pubmed/18414144
      OpenUrlCrossRefPubMedWeb of Science
    31. ↵
      2. Zhu X ,
      3. Qi H ,
      4. Feng Z , et al
      . Noninvasive high-frequency oscillatory ventilation vs nasal continuous positive airway pressure vs nasal intermittent positive pressure ventilation as postextubation support for preterm neonates in China: a randomized clinical trial. JAMA Pediatr 2022;176:551–9.doi:10.1001/jamapediatrics.2022.0710 pmid:http://www.ncbi.nlm.nih.gov/pubmed/35467744
      OpenUrlPubMed
    32. ↵
      2. Adler A ,
      3. Amyot R ,
      4. Guardo R , et al
      . Monitoring changes in lung air and liquid volumes with electrical impedance tomography. J Appl Physiol 1997;83:1762–7.doi:10.1152/jappl.1997.83.5.1762 pmid:http://www.ncbi.nlm.nih.gov/pubmed/9375349
      OpenUrlPubMedWeb of Science
    33. ↵
      2. Mosing M ,
      3. Waldmann AD ,
      4. Raisis A , et al
      . Monitoring of tidal ventilation by electrical impedance tomography in anaesthetised horses. Equine Vet J 2019;51:222–6.doi:10.1111/evj.12998 pmid:http://www.ncbi.nlm.nih.gov/pubmed/30035329
      OpenUrlPubMed
    34. ↵
      2. van der Burg PS ,
      3. Miedema M ,
      4. de Jongh FH , et al
      . Cross-Sectional changes in lung volume measured by electrical impedance tomography are representative for the whole lung in ventilated preterm infants. Crit Care Med 2014;42:1524–30.doi:10.1097/CCM.0000000000000230 pmid:http://www.ncbi.nlm.nih.gov/pubmed/24561568
      OpenUrlCrossRefPubMed
    35. ↵
      2. Tingay DG ,
      3. John J ,
      4. Harcourt ER , et al
      . Are all oscillators created equal? In vitro performance characteristics of eight high-frequency oscillatory ventilators. Neonatology 2015;108:220–8.doi:10.1159/000431216 pmid:http://www.ncbi.nlm.nih.gov/pubmed/26304262
      OpenUrlCrossRefPubMed

    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 VDG and CMR conceptualised and designed this post hoc analysis. VDG analysed the data and wrote the first draft of the manuscript. VDG acts as guarantor of the article. ADW developed the electrical impedance tomography (EIT) analysis software and performed EIT data extraction. VDG and CMR reviewed all apnoea sequences. PGD, DB, LS, DGT and CMR developed the concept and design of the initial study. LS and CMR were involved in patient recruitment and conducted the EIT measurements. CMR supervised the project. All authors participated in data interpretation and revised the manuscript for important intellectual content.

    • Funding VDG received an Endeavour Research Fellowship by the Australian Government (ERF_RDDH_5276_2016), PGD is supported by the Victorian Government Operational Infrastructure Support Programme (Melbourne, Australia) and the National Health and Medical Research Council (Practitioner Fellowship GNT 1059111), LS received a grant from the German Research Society (LO 2162/1-1), DGT received Career Development Fellowships by the NHMRC (GNT 11123859 and 1057514), and CMR was supported by the Swiss National Science Foundation (Early Postdoctoral Mobility fellowship P2ZHP3_161749) and the Swiss Society of Neonatology (Milupa Fellowship Award).

    • Competing interests VDG and CMR declare that they received an EIT monitor free of charge for a different research project by SenTec AG. All other authors declare that they have no conflict of interest.

    • 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.