Poster W5 - Assessing the Efficacy and Utility of Wearables for Perioperative Patient Monitoring

Introduction: With improved perioperative care, hospital stays after major operations have decreased in duration. The average hospital length of stay (LOS) after colectomy surgery has declined while readmissions rates for this procedure have increased. As a result, for patients undergoing such surgeries, there remains a persistently high rate of costly readmissions due to infection, prolonged ileus, hemorrhage, thromboembolic events, anastomotic leak, and dehydration). As implementation of Enhanced Recovery After Surgery (ERAS) protocols to reduce post-operative LOS continues to increase, complications which once were noticed while a patient was still in the hospital are now presenting when the patient is home. Increasingly post-operative care is being transitioned from hospital to home, which creates a void in capturing physiologic data for complications after discharge. Complications and readmissions can be the result of minor, correctable physiologic derangements, (e.g., acute kidney injury (AKI) which could be prevented with early management of dehydration from an ileostomy). Therefore, there is a clear need to determine whether an unobtrusive, inexpensive, and accurate outpatient monitoring strategy can be developed to alert providers of evolving complications and allow for early clinical rescue.

Recent improvements in digital health technologies (DHTs) provide an opportunity to address these gaps to better define postoperative complications and revolutionize their detection. The central hypothesis is that DHTs can be used to predict and detect post-operative complications and to passively screen for events on an outpatient basis following hospital discharge.

Materials and

Methods: In our study, we seek to monitor up to 50 participants with the Oura Gen 2 Ring and Garmin Vivosmart 4 smartwatch and assess whether they can provide continuous monitoring of surgery patients in the perioperative setting. Specifically, participants are asked to wear both devices up to 30 days prior to a planned surgical procedure and up to 90 days post-surgery. Participants also receive daily online surveys throughout the study to assess severity and location of pain (Figure 1). Participant data is then processed to determine 1) the adherence to wearing the wearable devices and 2) the ability of the wearable device data and derived digital biomarkers to predict post-operative morbidity and mortality via statistical and machine learning methods.

Results, Conclusions, and Discussions: To date we have enrolled 32 participants, all of whom have undergone surgery. By October 2024, we expect to complete enrollment with at least 50 participants. Of the 28 participants who have completed the study, seven have experienced outpatient complications. An example of a participant's wearable data revealed deviations that could potentially be linked to such exhibited complications can be seen in Figure 2. Originally, participants were monitored for only 30 days post-surgery. After an analysis of step count data over time, it was realized that participants did not achieve pre-surgery activity levels (baseline) within 30 days of surgery (Figure 3). This led to a protocol amendment where participants are now monitored for 90 days post surgery instead of for 30 days. Initial analysis of participant wearable device wear-time (i.e., adherence) indicates that the Oura Ring is worn more consistently (17% longer on average) by participants when compared to the Garmin (Figure 4). This was as expected given the less obtrusive form factor of a ring vs a wrist band watch. Participants also reported a more comfortable form factor with the Oura Ring.

Acknowledgements (Optional):