PhD Candidate McGill University Pointe-Claire, Quebec, Canada
Background: In the Neonatal Intensive Care Unit (NICU), vital signs including heart rate (HR), oxygen saturation and respiratory rate, are monitored continuously, as a standard of care. Traditionally, vital sign monitoring has been performed using bedside devices connected to sensors via wires. However, this system creates numerous challenges for physicians and nurses. These wires can tangle around the infant’s body, increase the risk of skin infection, and interfere with the parents’ ability to bond with their child through kangaroo care. In recent years, novel wireless monitoring solutions have begun to emerge, but these platforms still require extensive testing prior to widespread deployment. Objective: The Montreal Children’s Hospital (MCH) is conducting an ongoing study comparing the feasibility of wireless vital sign monitoring to the wired standard of care. We previously demonstrated strong agreement between the two systems for HR monitoring, but we now want to explore the differences at extrema. Thus, current work will examine the accuracy of the wireless platform in identifying bradycardia and tachycardia of clinical significance across neonatal HR recordings. Design/Methods: Twenty-five neonates were monitored on 4 consecutive days each, 8 hours at a time in the NICU of the MCH, using both a wired reference device (Philips Intellivue MX450, Philips Healthcare, Best, Netherlands) and a wireless alternative (ANNE One, Sibel Health, Niles, Michigan, USA). The study was approved by the Research Ethics Board at the McGill University Health Centre (MUHC) and informed consent was obtained from parents prior to infant enrolment. Following the resampling and alignment of corresponding HR signals from the two devices, bradycardia ( < 100 bpm for 5 seconds) and tachycardia (>180 bpm for 5 seconds) will be identified across all patient recordings. Fleiss’ kapa will be used to assess the agreement between the two systems, and both thresholds and durations will be later parametrized to evaluate performance under a range of bradycardia and tachycardia definitions.
