Continuous vs Intermittent Postoperative Vital Sign Monitoring: A Cluster Randomized Crossover Trial.

Abstract

Continuous postoperative monitoring on general nursing units detects vital sign abnormalities that are missed with conventional intermittent monitoring. Early recognition may prompt interventions to limit potentially harmful blood pressure, heart rate, and oxygen saturation perturbations. To test the hypothesis that continuous unblinded monitoring of vital signs would decrease blood pressure, heart rate, and oxygen saturation abnormalities compared with intermittent monitoring in patients recovering from noncardiac surgery. This multiple crossover randomized cluster trial in 2 postoperative hospital wards included patients recovering from noncardiac surgery and was conducted from October 7, 2020, to October 7, 2021. Data were analyzed from January to July in 2025. One participating ward was randomized to an initial month of either unblinded or clinician-blinded continuous vital sign monitoring and thereafter alternated at 4-week intervals for 1 year; the other ward always used the alternate monitoring approach. All patients had intermittent vital signs always available, whereas continuous vital signs were only available to clinicians during unblinded weeks. When continuous vital signs were unblinded, alerts at a mean arterial pressure less than 65 mm Hg, an oxygen saturation less than 90%, and a heart rate more than 110 per minute were transmitted to treating teams. The primary outcomes were durations of hypotension, hypoxemia, and tachycardia exceeding alert thresholds during the initial 48 hours of ward admission or until the third postoperative morning. Secondary outcomes were interventions categorized, with increasing severity, as none, independent nursing intervention, physician notification, or activation of the hospital emergency or rapid response systems. Among 798 total patients (median [IQR] age, 70.7 [62.5-78.7] years; 440 females [55%]), who included 404 assigned to continuous monitoring and 394 to intermittent monitoring, 561 (70%) with the American Society of Anesthesiologists' physical status III were included in the final analysis. The duration of oxygen saturation less than 90% was reduced by approximately 30 minutes over 48 hours of monitoring time, in continuously monitored patients (unblinded continuous monitoring group median [IQR], 70.8 [23.1-154.2] minutes vs blinded intermittent group median [IQR], 103.5 [29.5-249.7] minutes; geometric means ratio, 0.86 [95% CI, 0.78-0.95]; P = .002). Hypotension and tachycardia durations were not significantly reduced. The number of interventions was similar, with approximately half in each group receiving new oxygen therapy (unblinded continuous monitoring: 224 [55%] and blinded intermittent monitoring: 214 [54%]). In this cluster randomized crossover trial of continuous compared with intermittent vital sign monitoring, continuous monitoring reduced the duration of desaturation in patients recovering from noncardiac surgery on general hospital wards. Reducing the duration of vital sign abnormalities may decrease the risk of more serious complications, and robust patient-centric outcomes trials seem warranted. ClinicalTrials.gov Identifier: NCT04574908.