Article Text

Assessing the influence of abdominal compression on time to return of circulation during resuscitation of asphyxiated newborn lambs: a randomised preclinical study
  1. Graeme R Polglase1,2,
  2. Colin Hwang1,
  3. Douglas A Blank1,3,4,
  4. Shiraz Badurdeen1,5,
  5. Kelly J Crossley1,
  6. Martin Kluckow6,
  7. Andrew W Gill7,
  8. Emily Camm1,
  9. Robert Galinsky1,
  10. Yoveena Brian1,
  11. Stuart B Hooper1,2,
  12. Calum T Roberts1,3,4
  1. 1 The Ritchie Centre, Hudson Institute of Medical Research, Clayton, Victoria, Australia
  2. 2 Department of Obstetrics and Gynaecology, Monash University, Melbourne, Victoria, Australia
  3. 3 Department of Paediatrics, Monash University, Clayton, Victoria, Australia
  4. 4 Monash Newborn, Monash Children's Hospital, Clayton, Victoria, Australia
  5. 5 Newborn Research Centre, The Royal Women's Hospital, Parkville, Victoria, Australia
  6. 6 Department of Neonatology, Royal North Shore Hospital, St Leonards, New South Wales, Australia
  7. 7 Centre for Neonatal Research and Education, University of Western Australia, Perth, Western Australia, Australia
  1. Correspondence to Dr Calum T Roberts, Department of Paediatrics, Monash University, Clayton, Victoria 3168, Australia; calum.roberts{at}


Objective During neonatal resuscitation, the return of spontaneous circulation (ROSC) can be achieved using epinephrine which optimises coronary perfusion by increasing diastolic pressure. Abdominal compression (AC) applied during resuscitation could potentially increase diastolic pressure and therefore help achieve ROSC. We assessed the use of AC during resuscitation of asystolic newborn lambs, with and without epinephrine.

Methods Near-term fetal lambs were instrumented for physiological monitoring and after delivery, asphyxiated until asystole. Resuscitation was commenced with ventilation followed by chest compressions. Lambs were randomly allocated to: intravenous epinephrine (20 µg/kg, n=9), intravenous epinephrine+continuous AC (n=8), intravenous saline placebo (5 mL/kg, n=6) and intravenous saline+AC (n=9). After three allocated treatment doses, rescue intravenous epinephrine was administered if ROSC had not occurred. Time to achieve ROSC was the primary outcome. Lambs achieving ROSC were ventilated and monitored for 60 min before euthanasia. Brain histology was assessed for micro-haemorrhage.

Results Use of AC did not influence mean time to achieve ROSC (epinephrine lambs 177 s vs epinephrine+AC lambs 179 s, saline lambs 602 s vs saline+AC lambs 585 s) or rate of ROSC (nine of nine lambs, eight of eight lambs, one of six lambs and two of eight lambs, respectively). Application of AC was associated with higher diastolic blood pressure (mean value >10 mm Hg), mean and systolic blood pressure and carotid blood flow during resuscitation. Cortex and deep grey matter micro-haemorrhage was more frequent in AC lambs.

Conclusion Use of AC during resuscitation increased diastolic blood pressure, but did not impact time to ROSC.

  • resuscitation
  • neonatology
  • physiology

Data availability statement

Data are available upon reasonable request.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

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.


  • Infants exposed to severe asphyxia may be asystolic at birth and require cardiopulmonary resuscitation including chest compressions.

  • During resuscitation, successfully achieving return of spontaneous circulation is thought to be dependent on elevation of diastolic blood pressure to achieve coronary perfusion.

  • Epinephrine can effectively increase diastolic blood pressure through intense vasoconstriction, mediated by its action on α-adrenergic receptors.

  • Abdominal compression has the potential to increase diastolic pressure and impact the return of spontaneous circulation, but has not been investigated during neonatal resuscitation.


  • In asystolic newborn lambs receiving cardiopulmonary resuscitation, continuous abdominal compression did not influence time to achieve return of spontaneous circulation or rate of success.

  • Abdominal compression did produce a mean increase in diastolic blood pressure of >10 mm Hg during resuscitation.

  • Abdominal compression resulted in increased rates of brain micro-haemorrhage in the cortex and deep grey matter.


  • Abdominal compression cannot currently be recommended for clinical practice.

  • Future research should focus on identification of strategies that can increase diastolic blood pressure by a greater magnitude than can be achieved with abdominal compression.


Most newborn infants do not require resuscitation at birth, and among those who do, respiratory support is sufficient in most cases. Infants exposed to more severe asphyxia may require both respiratory and circulatory support, known as cardiopulmonary resuscitation (CPR). Current consensus guidelines advise initiation of CPR for those infants who are asystolic or bradycardic (heart rate below 60 beats per minute (bpm)), despite effective respiratory support.1–3 Epinephrine administration is recommended if heart rate remains <60 bpm.

Use of CPR, with or without epinephrine, is relatively infrequent, occurring in approximately 0.1% of term births, or 6–9% of more preterm infants.4 5 However, the consequences for infants requiring these measures are significant, with high rates of mortality and neurodisability in survivors.6 7 Improving the effectiveness of newborn CPR could potentially improve outcomes.8 It is currently understood that coronary perfusion during CPR occurs during the relaxation phase of the cardiac cycle (diastole), due to the pressure gradient between aorta and right ventricle.9 10 This is believed to be the principal mode of action of epinephrine during resuscitation, which produces systemic vasoconstriction via α-adrenergic receptors, resulting in increased diastolic arterial pressure.11 Strategies targeting increased diastolic pressure could increase coronary perfusion during CPR and enhance the likelihood of return of spontaneous circulation (ROSC).

Abdominal compression (AC) is one strategy that could be used to increase diastolic pressure.12 Despite intermittent periods of interest, particularly in the military setting (use of ‘military anti-shock trousers’ for major haemorrhage),13 this approach is not widely used during resuscitation, in any age group. To our knowledge, AC has not been examined in the newborn period, in preclinical or clinical studies. We aimed to assess the impact of AC on response to CPR, in asphyxiated newborn lambs. We hypothesised that application of AC would raise diastolic blood pressure and reduce the time needed to achieve ROSC.


Instrumentation and delivery

Pregnant Border-Leicester ewes (Ovis aries) at 140±2 days’ gestation (mean±SD; term ~148 days)14 were anaesthetised with intravenous thiopentone sodium (20 mg/kg) and intubated for delivery of inhaled anaesthesia (isofluorane 1.5–2.5%) in air/oxygen to maintain maternal peripheral oxygen saturation >95%. Fetal instrumentation was performed, as described previously and in the online supplemental methodology statement.15

Immediately before surgery, lambs were randomly allocated, using a web-based random sequence generator (, to one of four treatment groups:

  1. ‘Epinephrine’ (n=9), treated with intravenous epinephrine (20 µg/kg) according to neonatal resuscitation guidelines, followed by 0.9% saline flush (5 mL).

  2. ‘Saline’ (n=6), treated with intravenous 0.9% saline placebo (5 mL).

  3. ‘Epinephrine+AC’ (n=8), treated as per group 1, with the addition of AC.

  4. ‘Saline+AC’ (n=9), treated as per group 2, with the addition of AC.

The researcher directing resuscitation and determining time of ROSC was blinded to allocation of epinephrine or saline placebo. AC could not be blinded. ROSC was defined as diastolic blood pressure >20 mm Hg and rising, and spontaneous heart rate >100 bpm as determined by the researcher leading resuscitation.

The umbilical cord was clamped, and the lamb weighed for drug dosing. Asphyxia was allowed to progress until asystole, defined as mean blood pressure of ~0 mm Hg and no discernible activity visible on the blood pressure/flow traces.15

Resuscitation was initiated with positive pressure ventilation, in air. After 1 min, chest compressions were commenced, and fraction of inspired oxygen was increased to 1.00.2 For groups 3 and 4, AC was applied at commencement of chest compressions, and maintained until ROSC was achieved or resuscitation was ceased. AC was achieved by flexing the lower limbs onto the abdomen. An oval pressure pad held in place by a polypropylene strap encircling the abdomen was used to hold the limbs in the flexed position, with additional downward pressure applied manually by a researcher, with the aim of providing pressure continuously through resuscitation.

The first treatment dose of either intravenous epinephrine or saline was administered after 1 min of chest compressions and every 3 min thereafter, until ROSC was achieved. Lambs allocated to epinephrine could receive a maximum of five doses. Lambs allocated to saline could receive three doses, after which two ‘rescue’ epinephrine doses (20 µg/kg), starting at 11 min, could be administered. Resuscitation was ceased at 15 min after ventilation onset if ROSC had not been achieved. At achievement of ROSC, AC was ceased immediately.

After ROSC, lambs were supported and monitored for 60 min before humane euthanasia, as described in the online supplemental methodology statement.

Brain histology

Postmortem brain histology was assessed for micro-haemorrhage, as described in the online supplemental methodology statement .

Statistical analysis

All lambs were analysed for the assessments made during CPR. Only those lambs achieving ROSC were included in post-ROSC analyses of physiology and histology. Baseline fetal and physiological data were compared with one-way analysis of variance (ANOVA) (GraphPad Prism; GraphPad Software, California, USA). Change in mean abdominal and intrapleural pressures, recorded during chest compressions, was compared with a paired t-test. Two-way ANOVA with ‘epinephrine’ and ‘AC’ as the factors was used to compare groups during CPR. Two-way repeated measures ANOVA was used to compare serial physiological data. Rates of ROSC and micro-haemorrhage were compared using a Fisher’s exact test. Statistical significance was accepted at p<0.05. The primary outcome was time to achieve ROSC. We have previously reported a mean (±SD) time from onset of CPR to ROSC of 2.4±0.4 min after epinephrine and 11.2±1.2 min after saline, in asystolic near-term lambs.16 We calculated that seven lambs per group would be sufficient, with 80% power and significance at 5%, to detect a 20% relative reduction in time to ROSC with use of AC. We allocated eight to nine lambs per group to allow for failure to achieve ROSC in one to two lambs per group. Recruitment to the ‘saline’ group was restricted to six lambs, as investigators regarded it as unethical to allocate further lambs to this group given the low rate of ROSC observed in this and a previous study.16


Lamb characteristics

Lamb clinical characteristics and blood gas parameters were similar in all treatment groups (table 1). The epinephrine+AC group had more second-delivered twins than the epinephrine and saline groups.

Table 1

Lamb characteristics

Time to ROSC and achievement of ROSC

The primary outcome, time to achieve ROSC, was significantly shorter in the epinephrine (2.9±0.3 min) and epinephrine+AC groups (3.0±0.2 min), versus the saline (10.0±3.7 min) and saline+AC lambs (9.7±2.5 min) (p<0.001, figure 1). No effect of AC was noted on primary outcome, either with or without epinephrine. Achievement of ROSC in response to allocated treatment, without rescue epinephrine, was more common in epinephrine-treated lambs than saline-treated lambs: nine of nine epinephrine lambs, eight of eight epinephrine+AC lambs, one of six saline lambs and two of eight saline+AC lambs (p<0.05). Including use of rescue epinephrine, ROSC occurred in four of six saline lambs and five of eight saline+AC lambs. AC did not appear to influence the rate of ROSC. All lambs in the epinephrine groups achieved ROSC after one treatment dose. One lamb in the saline group and two lambs in the saline+AC group achieved ROSC without rescue epinephrine. The other six lambs that achieved ROSC in either saline group required rescue epinephrine.

Figure 1

Time from initiation of resuscitation to return of spontaneous circulation (ROSC). Time taken to achieve ROSC (min) in epinephrine (red), epinephrine+abdominal compression (AC) (purple), saline (dark blue) and saline+AC (light blue) lambs. Black dots indicate lambs which failed to achieve ROSC. Data are mean±SD. *** indicates p<0.01 AC versus no AC.

Physiology during chest compressions

The AC manoeuvre increased intra-abdominal pressure from 4.4±4.6 mm Hg to 20.7±8.4 mm Hg, and intrapleural pressure from 0.0±2.6 mm Hg to 7.9±6.1 mm Hg (figure 2).

Figure 2

Intra-abdominal and intrapleural pressure. Mean intra-abdominal pressure (left panel) and intrapleural pressure (right panel) in lambs measured immediately prior to abdominal compression (AC) (control, open circles) and during the first minute of chest compressions (CCs) with AC (CC+AC, closed circles), prior to administration of epinephrine. Mean 1 min values for each individual lamb plotted with group mean±SD. n=14 due to equipment failure in two lambs.

Lambs in both AC groups had significantly higher mean, systolic and diastolic blood pressure, and mean and peak systolic carotid blood flow, than lambs that did not receive AC (figure 3). Pulmonary blood flow during chest compressions was similar in all four groups.

Figure 3

Physiological response to chest compressions (CCs) and abdominal compression (AC). (A) Mean, (B) systolic and (C) diastolic blood pressure (BP) measured from the carotid artery (n=15), (D) mean and (E) peak systolic carotid arterial blood flow (CBF) and (F) pulmonary blood flow (n=13; PBF) in lambs measured immediately prior to AC (control) and during the first minute of CCs and AC (CC+AC), prior to administration of epinephrine. Individual animals plotted with mean±SD. ***P<0.01, ****p<0.001.

The change in diastolic blood pressure during resuscitation differed in AC and non-AC groups, as shown in figure 4. Lambs that did not receive AC had little change in diastolic pressure from baseline (end-asphyxia) during the first minute of chest compressions. Pressure increased only after epinephrine administration, with significant increase from baseline occurring only in the final 20 s before ROSC. Lambs receiving AC had an immediate significant increase in diastolic pressure from baseline, on commencement of chest compressions+AC, to >10 mm Hg (p<0.0001), which persisted throughout resuscitation. A further increase occurred after epinephrine administration, shortly before ROSC. A similar significant increase in diastolic pressure to >10 mm Hg occurred in the saline+AC lambs (not shown in figure), which persisted for the duration of the resuscitation (range 5–14 min of chest compressions), with a greater increase seen shortly prior to ROSC, if achieved.

Figure 4

Diastolic blood pressure (BP) changes over time, during resuscitation. Diastolic BP measured from the carotid artery in epinephrine (red) and epinephrine+AC (purple) lambs, at end-asphyxia, during the first 60 s of chest compressions, after epinephrine administration in the 60 s prior to ROSC, at ROSC and during recovery after ROSC. Each time point is an average of 20 s of recording. Dashed line separates chest compressions prior to or after epinephrine administration. Data are mean±SD. * indicates p<0.0001 from end-asphyxia. AC, abdominal compression; ROSC, return of spontaneous circulation.

Physiology after ROSC

Ventilation parameters (data not shown) and blood gas parameters were similar in the four groups after ROSC (online supplemental figure 1). There was no difference in arterial or cerebral saturation after ROSC.

Systolic blood pressure was higher in saline versus saline+AC lambs between 3 and 4 min; no other differences were observed (online supplemental figure 2). Diastolic blood pressure was lower in epinephrine lambs compared with saline lambs after ROSC; no other differences were observed. Heart rate, pulmonary blood flow and carotid blood flow were not different between groups.

Brain micro-haemorrhage

Among survivors to study end, occurrence of white matter micro-haemorrhage was similar in non-AC lambs (5 of 11 total, 5 of 8 epinephrine and 0 of 3 saline) and AC lambs (4 of 12 total, 2 of 7 epinephrine+AC and 2/5 saline+AC). Cortex micro-haemorrhage was more frequent in AC lambs (10 of 12 total, 5 of 7 epinephrine+AC and 5 of 5 saline+AC) than non-AC lambs (4 of 11 total, 3 of 8 epinephrine and 1 of 3 saline) (p=0.04). Micro-haemorrhage in deep grey matter was more frequent in AC lambs (5 of 12 total, 3 of 7 epinephrine+AC and 2 of 5 saline+AC) than non-AC lambs (0 of 11 total, 0 of 8 epinephrine and 0 of 3 saline) (p=0.04). No difference was observed in the hippocampus, with haemorrhage detected in one lamb from the saline+AC group. No effect of epinephrine or saline was observed on haemorrhage rates.


In this study conducted in asphyxiated newborn lambs, we found that the use of AC as an adjunct to neonatal resuscitation did not influence time to achievement of ROSC. AC resulted in significant changes in some physiological variables during CPR. As hypothesised, lambs receiving AC had higher mean intra-abdominal and intrapleural pressure (by 16.3 and 7.9 mm Hg, respectively) during CPR, with significant increases in mean, systolic and diastolic blood pressure, and carotid blood flow. Despite an increase in diastolic blood pressure of >10 mm Hg, seen from onset of AC, we observed no difference in the rate of ROSC in animals receiving AC, with or without epinephrine.

Epinephrine has agonistic effects on both β-adrenergic receptors, causing myocardial stimulation, and on α-adrenergic receptors, causing peripheral vasoconstriction in vascular beds where α receptors predominate, resulting in increased diastolic blood pressure.11 Preclinical studies assessing rates of ROSC after resuscitation of dogs receiving epinephrine in conjunction with either α-blockade or β-blockade suggested that the α-adrenoceptor action of epinephrine is the critical component.17 We have shown that AC can increase diastolic pressure. However, the failure to improve time to achieve ROSC, or rate of ROSC, may indicate that the pressure increase resulting from AC is insufficient to adequately mimic epinephrine treatment. Administration of epinephrine provoked an increase in diastolic pressure during CPR, in the AC and non-AC groups (figure 4). In the final 20 s before ROSC, values of 15–20 mm Hg were achieved. Data in adult patients who had a cardiac arrest suggest 15–25 mm Hg is the critical threshold in that population.18

The α-adrenergic effects of epinephrine can prevent arterial wall collapse,19 which, by maintaining patency of the coronary arteries during CPR, may be an important effect that AC cannot reproduce. Alternatively, our findings may indicate that, in the asphyxiated newborn, the β-adrenergic effects of epinephrine have a role in establishing ROSC. The aforementioned study of α-blockade and β-blockade was conducted in adult animals,17 and did not account for important differences in newborn physiology, such as the need for aeration of the fluid-filled lung and the presence of a transitional circulation with pulmonary–systemic shunts. The neonatal cardiomyocyte has significant structural and metabolic differences in comparison with the adult cardiomyocyte, with impaired excitation–contraction coupling.20 21 A severe hypoxic insult is required to produce asystole, resulting in internal cellular disorganisation and energy depletion. The response of the neonatal myocardium to treatment after severe hypoxia may differ from that seen in adults, and it is conceivable that elevation of diastolic pressure alone will not produce ROSC in some asphyxiated newborns.

Although the increase in blood pressure produced by AC did not influence ROSC, it may influence perfusion to organs other than the heart. Mean carotid blood flow in the AC groups was more than double that seen in the non-AC groups. Increased cerebral perfusion during CPR could reduce hypoxic-ischaemic damage during the period between asystole and ROSC. However, cerebral vasculature, already maximally dilated in response to severe asphyxia, is potentially susceptible to haemorrhage secondary to higher flows and pressures.22 We found a greater proportion of lambs with micro-haemorrhage in the cortex, and the deep grey matter, after AC. It is unclear how these preclinical histological findings would translate into clinical neurodevelopmental outcomes in resuscitated infants.

We observed very few differences between the four treatment groups during the 60-minute period after resuscitation, suggesting that AC does not impact physiology beyond initial resuscitation. Limitations of this study include the potential influence of the anatomical differences between lambs and humans, and that the study was not powered to assess differences in rates of ROSC between AC and non-AC animals.

Previous research into the use of AC during CPR has focused on adults or preclinical studies of adult animals. Two principal approaches have been described: continuous static pressure or interposed AC, in which AC is applied intermittently in the relaxation phase between chest compressions.23 Continuous static pressure has the advantage of being simpler to apply, using binding or a mechanical device. Studies in adult dogs undergoing CPR for asphyxia or ventricular fibrillation-induced arrest showed high rates of ROSC with this approach. Interposed AC positively influenced blood pressure and cerebral perfusion in adult animal studies.24 Although increased rates of ROSC and survival to discharge were seen in one randomised clinical trial,25 this study had a limited sample size and a particularly low rate of survival in the control group. Other trials have shown no advantage.26 Advocates of this technique have suggested that the variability in compression technique described in the different trials may be the explanation for differing rates of efficacy.27 Application of AC by either method presents practical challenges that may influence clinical feasibility.


We have shown that the use of AC in asystolic newborn lambs undergoing resuscitation did not reduce the time to achievement of ROSC, and that use of AC may increase the risk of brain micro-haemorrhage. The actions by which epinephrine promotes ROSC in asphyxic newborns undergoing CPR may be different from those that are most important in adult resuscitation. AC cannot be recommended for use during neonatal resuscitation.

Data availability statement

Data are available upon reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

Experimental procedures were approved by Monash Medical Centre Animal Ethics Committee A, Monash University (MMCA2020/04) and conducted in accordance with the National Health and Medical Research Council of Australia and ARRIVE guidelines.27


The authors would like to thank Alison Moxham, Valerie Zahra and Karyn Rodgers for their technical support.


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.


  • X @calumtheroberts

  • Contributors All named authors contributed to one or more of: conception and design of the study, data acquisition, analysis and interpretation of the data. GRP and CTR cowrote the first draft of the manuscript and approved it prior to submission. GRP and CTR accept full responsibility for the work and act as guarantors.

  • Funding This research was supported by the National Health and Medical Research Council (NHMRC) Project Grant (APP1158494) and Fellowships (GRP: APP1173731, SBH: APP545921, CTR: APP1175634), a National Heart Foundation of Australia Vanguard Grant (103022), the Rebecca L Cooper Medical Research Foundation and the Victorian Government's Operational Infrastructure Support Program.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.