The accuracy of human senses in the detection of neonatal heart rate during standardized simulated resuscitation: Implications for delivery of care, training and technology design

Abstract

Aim

Auscultation and palpation are recommended methods of determining heart rate (HR) during neonatal resuscitation. We hypothesized that: (a) detection of HR by auscultation or palpation will vary by more than ±15 BPM from actual HR; and (b) the inability to accurately determine HR will be associated with errors in management of the neonate during simulated resuscitation.

Subjects and methods

Using a prospective, randomized, controlled study design, 64 subjects participated in three simulated neonatal resuscitation scenarios. Subjects were randomized to technique used to determine HR (auscultation or palpation) and scenario order. Subjects verbalized their numeric assessment of HR at the onset of the scenario and after any intervention. Accuracy of HR determination and errors in resuscitation were recorded. Errors were classified as errors of omission (lack of appropriate interventions) or errors of commission (inappropriate interventions). Cochran's Q and chi square test were used to compare HR detection by method and across scenarios.

Results

Errors in HR determination occurred in 26–48% of initial assessments and 26–52% of subsequent assessments overall. There were neither statistically significant differences in accuracy between the two techniques of HR assessment (auscultation vs palpation) nor across the three scenarios. Of the 90 errors in resuscitation, 43 (48%) occurred in association with errors in HR determination.

Conclusions

Determination of heart rate via auscultation and palpation by experienced healthcare professionals in a neonatal patient simulator with standardized cues is not reliable. Inaccuracy in HR determination is associated with errors of omission and commission. More reliable methods for HR assessment during neonatal resuscitation are required.

Introduction

According to the current guidelines of the Neonatal Resuscitation Program (NRP) based on the consensus on science published by the International Liaison Committee on Resuscitation, a newborn's heart rate (HR) may be determined by either listening to the precordium with a stethoscope or feeling pulsations at the base of the umbilical cord.1, 2, 3 Interventions are either administered or withheld based upon the numeric HR value determined by the healthcare professionals (HCPs) at the bedside. There exist defined HR cut-offs (100 beats per minute [BPM], 60 BPM) below which certain interventions (positive pressure ventilation [PPV], chest compressions [CC], epinephrine administration) are recommended; failure to do so may result in cardiac arrest and death. Similarly, underestimation of the true HR and inappropriately applied interventions may also result in harm. Thus if HR is not accurately determined, certain therapeutic interventions may be inappropriately withheld or administered, potentially resulting in serious injury or death.

This study sought to determine the accuracy of auscultation of the precordium (with a stethoscope) and palpation of the umbilical cord in the detection of HR during simulated neonatal resuscitation. To provide appropriate clinical context, the frequency of errors of omission (failure to perform appropriate interventions) and errors of commission (performance of inappropriate interventions) was also assessed. Using a neonatal patient simulator capable of generating a HR that is fixed in rate, volume, tone and location, and umbilical cord pulsations that are fixed in frequency, amplitude and location, we hypothesized that: (1) the detection of HR by auscultation or palpation will vary by more than ±15 BPM from actual HR, and (2) the inability to accurately determine HR will be associated with errors of omission and/or commission.