Detection of Postoperative Vital Signs Abnormalities on a Surgical Ward using Conventional and Remote Automated Monitoring

Background: The true incidence of abnormal vital signs on post-surgical wards may be seriously underestimated based on nurse obtained conventional measurement. We sought to determine the incidence and severity of postoperative tachycardia, bradycardia and hypoxemia detected by continuous remote automated monitoring (RAM) versus the incidence of these vital sign abnormalities detected during routine nursing care. Methods: We conducted a prospective cohort proof-of-concept study of 121 patients aged ≥45 years recovering from orthopedic surgery. Eligible patients were at risk of postoperative myocardial injury and had a planned hospital stay ≥48 hours. Philips’ IntelliVue MX40 wearable RAM technology was used to continuously monitor patients’ heart rate and pulse oximetry up to 72 hours following surgery. In addition, study personnel obtained vital signs collected during routine nursing care from participants’ medical charts. Clinically meaningful tachycardia, bradycardia and hypoxemia were defined as heart rates >100, <55, and blood oxyhemoglobin saturation (SpO 2 ) of <90% for >15 contiguous minutes, respectively. Results: Continuous RAM identified clinically meaningful episodes of tachycardia in 42 of 121 patients [34.7%] versus 7 patients [5.8%] identified by routine nursing care, for an absolute difference 28.9% (95% confidence interval [CI] 20.8, 37.0; p=0.001). RAM also detected bradycardia in 14 of 121 patients [11.6%] versus 6 patients [5.0%] detected by routine care, for an absolute difference 6.6% (95% CI 2.2, 11.0; p=0.07). RAM detected hypoxemia in 25 of 107 patients [23.3%] compared with 1 patient [0.9%] detected through routine monitoring, for an absolute difference of 22.4% (95% CI 14.5, 30.3; p=0.001). Conclusion: Most clinically meaningful episodes of vital signs abnormalities detected by continuous RAM were missed by nurses through conventional periodic monitoring. Continuous RAM technologies have the potential to improve patient outcomes through early identification of physiological abnormalities on surgical wards.