Article Text
PDF
Abstract
Background The 2015 neonatal resuscitation guidelines added ECG as a recommended method of assessment of an infant’s heart rate (HR) when determining the need for resuscitation at birth. However, a recent case report raised concerns about this technique in the delivery room.
Objectives To compare accuracy of ECG with auscultation to assess asystole in asphyxiated piglets.
Methods Neonatal piglets had the right common carotid artery exposed and enclosed with a real-time ultrasonic flow probe and HR was continuously measured and recorded using ECG. This set-up allowed simultaneous monitoring of HR via ECG and carotid blood flow (CBF). The piglets were exposed to 30 min normocapnic alveolar hypoxia followed by asphyxia until asystole, achieved by disconnecting the ventilator and clamping the endotracheal tube. Asystole was defined as zero carotid blood flow and was compared with ECG traces and auscultation for heart sounds using a neonatal/infant stethoscope.
Results Overall, 54 piglets were studied with a median (IQR) duration of asphyxia of 325 (200-491) s. In 14 (26%) piglets, CBF, ECG and auscultation identified asystole. In 23 (43%) piglets, we observed no CBF and no audible heart sounds, while ECG displayed an HR ranging from 15 to 80/min. Sixteen (30%) piglets remained bradycardic (defined as HR of <100/min) after 10 min of asphyxia, identified by CBF, ECG and auscultation.
Conclusion Clinicians should be aware of the potential inaccuracy of ECG assessment during asphyxia in newborn infants and should rather rely on assessment using a combination of auscultation, palpation, pulse oximetry and ECG.
- newborn
- auscultation
- electrocardiography
- heart rate
- neonatal resuscitation
Statistics from Altmetric.com
What is already known on this topic?
The 2015 neonatal resuscitation guidelines recommended ECG as a method to assess an infant’s heart rate when determining the need for resuscitation at birth.
A recent case report raised concerns about the reliability of ECG in a newborn infant with hydrops fetalis requiring extensive resuscitation.
What this study adds?
While electrocardiography usually provides a rapid and accurate display of the heart’s electrical activity, it does not prove the presence of effective cardiac function.
In this study in asphyxiated newborn piglets, pulseless electrical activity was observed in 43% of cases with no carotid blood flow or audible heart sounds but ECG displayed a heart rate ranging from 15 to 80/min.
Presence of ECG activity does not always imply adequate cardiac output; one should not rely solely on ECG, but include other assessment tools in the management of the compromised infant.
Introduction
Heart rate (HR) is the most important clinical indicator to evaluate the status of a compromised newborn, and it is used to guide resuscitation efforts in the delivery room.1 Indeed, an increase in the newborn’s HR is considered the most reliable indicator of adequate ventilation.1 Until recently, the HR in newborn infants at birth was assessed using either (i) palpation of the umbilical cord, (ii) auscultation of the precordium and/or (iii) pulse oximetry.2 The 2015 neonatal resuscitation guidelines suggested that ECG could be used in this setting. This addition was based on observational data and small randomised controlled trials showing that ECG displays a reliable HR faster than pulse oximetry.1 2 However, most of the included infants in these studies did not require resuscitation, and very few required chest compressions.1 Interventions are recommended if an infant’s HR is <100/min and it seems prudent to use the fastest and most accurate technique to assess HR.1 This is of clinical importance as underestimation of HR can result in unnecessary interventions, whereas overestimation may result in a failure to recognise the need for resuscitation. A recent case report raised concerns about the reliability of ECG in a newborn infant with hydrops fetalis requiring extensive resuscitation, potentially due to thoracic subcutaneous oedema.3 Pulseless electrical activity (PEA) is a preterminal rhythm characterised by cardiac electrical activity but the absence of a palpable pulse is commonly seen following cardiac arrest in adults. The phenomenon is not widely recognised in neonates. Therefore, we aimed to compare the accuracy of ECG with that of auscultation during asystole in a porcine model of cardiac arrest.
Methods
This is a secondary analysis of two recently completed but unpublished randomised controlled animal studies. For this secondary analysis, we included 54 newborn mixed breed piglets (1–3 days of age, weighting 1.7–2.3 kg) from the original two studies.
Piglets were instrumented following the induction of anaesthesia using isoflurane, and were then intubated via a tracheotomy and pressure-controlled ventilation (Sechrist infant ventilator, model IV-100; Sechrist Industries, Anaheim, CA) was initiated at a respiratory rate of 16–20 breaths/min and pressure of 20/5 cm H2O. Oxygen saturation was kept within 90%–100%, and glucose levels and hydration were maintained with an intravenous infusion of 5% dextrose at 10 mL/kg/hour. The piglet’s body temperature was maintained in the normal range of 38.5–39.5°C using an overhead warmer and water-heated pads. During the experiment, anaesthesia was maintained with intravenous propofol 5–10 mg/kg/hour and morphine 0.1 mg/kg/hour. Additional doses of propofol (1–2 mg/kg) and morphine (0.05–0.1 mg/kg) were administered as needed.
Study protocol
All piglets had the right common carotid artery exposed and enclosed with a real-time ultrasonic flow probe (2 mm; Transonic Systems, Ithaca, New York, USA), and HR was continuously measured and recorded using ECG (Hewlett Packard 78833B monitor, Hewlett Packard, Palo Alto, California, USA). This set-up allowed us to simultaneously monitor HR via ECG and carotid blood flow (CBF) during the experiments.4 Hypoxia was induced by exposing piglets to 30 min of normocapnic alveolar hypoxia at a fractional inspired oxygen concentration of 0.10. Hypoxia was then followed by asphyxia until asystole, achieved by disconnecting the ventilator and clamping the endotracheal tube. The study protocol further specified that either after 10 min of asphyxia or after asystole chest compression will be initiated.
Cardiac asystole (arrest) was defined as zero CBF. We compared CBF with ECG and auscultation of the piglet’s chest using a neonatal/infant stethoscope (3M Littmann Classic II Infant Stethoscope, USA) (performed by GMS (n=28) or P-YC (n=26)).
Data collection and analysis
Markers for asystole were placed within the LabChart program (ADInstruments, Dunedin, New Zealand) to indicate time of cardiac arrest before initiation of the resuscitation protocol. This marker was then used to compare timing of onset of asystole as determined by auscultation, ECG and CBF. The data are presented as mean±SD for normally distributed continuous variables and median (IQR) when the distribution is skewed. The data were tested for normality and analysed using Stata (Intercooled 10, StataCorp).
Results
We studied 54 piglets; the median (IQR) duration of asphyxia was 325 (200–491) s. Of these, 37 became asystolic. In 14 (26%) piglets, CBF, ECG and auscultation identified asystole (figure 1A). In 23 (43%) piglets, we observed no CBF and no audible heart sounds, while ECG displayed an HR ranging from 15 to 80/min (figure 1B). Sixteen (30%) piglets remained bradycardic (defined as HR of <100/min) after 10 min of asphyxia, identified by CBF, ECG and auscultation (figure 1C). There was also one piglet with ventricular tachycardia/fibrillation on the ECG with absence of CBF and no audible heart sounds (figure 1D). The overall accuracy of ECG was 56% and of auscultation 100%, and predictive values, sensitivity and specificity are presented in table 1.
Waveforms of carotid blood flow (CBF) and ECG: (A) asystole correctly assessed with absence of CBF, ECG and no audible heart sound; (B) ECG clearly showing bradycardia in the absence of CBF and no audible sound; (C) bradycardia correctly assessed with the presence of CBF, ECG and audible sound; (D) ventricular tachycardia/fibrillation with the absence of CBF and no audible heart sound.
Accuracy of ECG and auscultation compared with the gold standard carotid blood flow
Discussion
In this translational study using a newborn piglet model equivalent to a human infant at 36–38 weeks gestation, we found that ECG was in agreement with CBF in 37% of cases. However, ECG displayed an HR (15–80/min) in 43% of cases where the CBF was absent and the HR inaudible. This phenomenon of electrical activity in the absence of a detectable pulse is called pulseless electrical activity, a common phenomenon during resuscitation after cardiac arrest in adults. Previously, neonatal healthcare providers were aware of bradycardia and asystole, but other rhythm disturbances (eg, ventricular fibrillation or PEA) are not described in the delivery room.1 Our results suggest that PEA is more common than isolated bradycardia or electrical asystole in asphyxiated piglets. Potential causes of PEA in adults include hypoxia, severe volume loss, sepsis or tension pneumothorax. In paediatric cardiopulmonary resuscitation, PEA is considered a non-perfusing, non-shockable rhythm, and chest compression and epinephrine should be provided.5 In the delivery room, hypoxia and hypovolaemia are the two principal causes of PEA. In these cases, assessing the need for ongoing resuscitation using ECG alone may mislead clinicians that cardiac output is adequate, when in fact it is either absent or insufficient.
The 2015 neonatal resuscitation guidelines added HR assessment via ECG to identify if resuscitation measures are required in newborn infants.1 The guidelines further emphasise that ECG assessment does not replace the need for other assessment methods of an infant’s HR.1 However, no clinical study has identified potential pitfalls when HR is assessed using ECG, but our observations demonstrate potential limitations of ECG technology during severe asphyxia, at least in term infants. In particular, in cases with PEA and an ECG HR of 60–100/min the clinical team might postpone initiation of chest compressions. Accuracy of ECG is of clinical importance as underestimation of HR can result in unnecessary interventions, whereas overestimation may result in a failure to recognise the need for resuscitation. No clinical study has systematically examined the limitations of ECG measurement of HR, but reports are starting to emerge regarding the potential limitations of ECG technology in subsets of infants.3 In our controlled laboratory experiment, auscultation was 100% accurate in determining the presence of effective cardiac activity; in a delivery room situation with a newborn infant undergoing resuscitation procedures this may not be the case. Our results raise concerns regarding the use of ECG in the delivery room.
Conclusion
These piglet data suggest a potential for inaccuracy of ECG assessment during the stabilisation of asphyxiated newborn infants. Clinicians should rather rely on assessment using a combination of auscultation, palpation, pulse oximetry and ECG. Future studies should evaluate the benefits and limitations of ECG during neonatal cardiopulmonary resuscitation.
Acknowledgments
The authors thank the public for donation of money to their funding agencies: GMS is a recipient of the Heart and Stroke Foundation/University of Alberta Professorship of Neonatal Resuscitation, a National New Investigator of the Heart and Stroke Foundation Canada and an Alberta New Investigator of the Heart and Stroke Foundation Alberta. The authors acknowledge the Women and Children’s Health Research Institute, University of Alberta, for supporting the study.
Footnotes
Contributors Conception and design: KJB, PGD and GMS. Data collection: SP, P-YC, ALS and GMS. Data analysis and interpretation, drafting of the article, critical revision of the article for important intellectual content and final approval of the manuscript: all authors.
Funding PGD is supported by an NHMRC Practitioner Fellowship #1059111. The study was supported by a Grant-in-Aid from the Heart and Stroke Foundation Canada (grant number: G-15-0009284).
Disclaimer The sponsors of the study had no role in study design, data collection, data analysis, data interpretation or writing of the report.
Competing interests None declared.
Patient consent Not required.
Ethics approval The original studies were conducted in accordance with Animal Research Reporting of In Vivo Experiments guidelines and approved by the Animal Care and Use Committee (Health Sciences) University of Alberta (AUP00001764, AUP00002151)
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- Fantoms