Using Consumer Wearable Devices to Profile Postoperative Complications After Pediatric Appendectomy.

INTRODUCTION: Markers of postoperative recovery in pediatric patients are difficult for parents to evaluate after hospital discharge, who use subjective proxies to assess recovery and the onset of complications. Consumer-grade wearable devices (e.g., Fitbit) generate objective recovery data in near real time and thus may provide an opportunity to remotely monitor postoperative patients and identify complications beyond the initial hospitalization. The aim of this study was to use daily step counts from a Fitbit to compare recovery in patients with complications to those without complications after undergoing appendectomy for complicated appendicitis. METHODS: Children ages 3-17 years old undergoing laparoscopic appendectomy for complicated appendicitis were recruited. Patients wore a Fitbit device for 21 d after operation. After collection, patient data were included in the analysis if minimum wear-time criteria were achieved. Postoperative complications were identified through chart review, and step count trajectories for patients recovering with and without complications were compared. Additionally, to account for the patients experiencing a complication on different postoperative days, median daily step count for pre- and post-complication were analyzed. RESULTS: Eighty-six patients with complicated appendicitis were enrolled in the study, and fourteen children developed a postoperative complication. Three patients were excluded because they did not meet the minimum wear time requirements. Complications were divided into abscesses (n = 7, 64%), surgical site infections (n = 2, 18%), and other, which included small bowel obstruction and Clostridioides difficile infection (n = 2, 18%). Patients presented with a complication on mean postoperative day 8, while deviation from the normative recovery trajectory was evident 4 d prior. When compared to children with normative recovery, the patients with surgical complications experienced a slower increase in step count postoperatively, but the recovery trajectory was specific to each complication type. When corrected for day of presentation with complication, step count remained low prior to the discovery of the complication and increased after treatment resembling the normative recovery trajectory. CONCLUSIONS: This study profiled variations from the normative recovery trajectory in patients with complication after appendectomy for complicated appendicitis, with distinct trajectory patterns by complication type. Our findings have potentially profound clinical implications for monitoring pediatric patients postoperatively, particularly in the outpatient setting, thus providing objective data for potentially earlier identification of complications after hospital discharge.

Using Wearable Devices to Profile Demographic-Specific Recovery After Pediatric Appendectomy.

INTRODUCTION: Counseling patients and parents about the postoperative recovery expectations for physical activity after pediatric appendectomy varies significantly and is not specific to patients' demographic characteristics. Consumer wearable devices (CWD) can be used to objectively assess patients' normative postoperative recovery of physical activity. This study aimed to develop demographic-specific normative physical activity recovery trajectories using CWD in pediatric patients undergoing appendectomy. METHODS: Children ages 3-18 y old undergoing laparoscopic appendectomy for acute appendicitis were recruited. Patients wore a Fitbit device for 21 d postoperatively and daily step counts were measured. Patients with postoperative complications were excluded. Segmented regression models were fitted and time-to-plateau was estimated for patients with simple and complicated appendicitis separately for each age group, sex, race/ethnicity, and body mass index category. RESULTS: Among 147 eligible patients; 76 (51.7%) were female, 86 (58.5%) were in the younger group, and 79 (53.7%) had complicated appendicitis. Patients 3-11 y old demonstrated a faster trajectory to a physical activity plateau compared to those 12-18 in both simple (postoperative day [POD] 9 versus POD 17) and complicated appendicitis (POD 17 versus POD 21). Males and females had a similar postoperative recovery trajectory in simple and complicated appendicitis. There was no clear pattern differentiating trajectories based on race/ethnicity. Overweight/obese patients demonstrated a slower recovery trajectory in simple appendicitis. CONCLUSIONS: This study demonstrates that factors other than the disease itself, such as age, may affect recovery, suggesting the need for more tailored discharge instructions. CWDs can improve our understanding of recovery and allow for better data-driven counseling perioperatively.

Consumer Wearables Define a Normative Recovery Trajectory Following the Modified Nuss Procedure.

INTRODUCTION: The modified Nuss procedure is an elective procedure associated with a lengthy recovery, uncontrolled pain, and risk of infrequent, yet life-threatening complications. The absence of objective measures of normative postoperative recovery creates uncertainty about the postdischarge period, which remains highly dependent on the patients' and their caregivers' expectations and management of recovery. We aimed to describe an objective-normative, physical activity recovery trajectory after the modified Nuss procedure, using step counts from the Fitbit. METHODS: This observational study enrolled children ≤18 y with pectus excavatum who underwent the modified Nuss procedure from 2021 to 2022. The Fitbit was worn for 21 postoperative days. Postdischarge outcomes and health-care utilization were evaluated. For patients without postoperative complications, piecewise linear regression analysis was conducted to generate a normative recovery trajectory model of daily step counts. RESULTS: Of 80 patients enrolled, 66 (86%) met eligibility criteria (mean age, 15.1 ± 1.3 y; 89.4% male, 62.1% non-Hispanic White). The mean number of telephone and electronic message encounters regarding concerns with the patient's recovery within 30 d postoperatively was 2.1 (standard deviation = 2.7). Ten patients (15.2%) returned to the emergency department (ED) within the 30-d postoperative period. Seven patients (10.6%) presented to the ED one time, and three patients (4.5%) presented to the ED twice. Thirty-day readmission rate was four patients (6.0%). Piecewise regression model showed that patients without complications steadily increased their daily step count on each postoperative day and plateaued on day 18. CONCLUSIONS: We have developed a normative recovery trajectory following the modified Nuss procedure using step count data collected by a consumer wearable device. This offers the potential to inform preoperative patient expectations and reduce avoidable health-care utilization through informed preoperative counseling, thus laying the ground work for the use of consumer wearable devices as a postdischarge remote monitoring tool.

Applying machine learning to consumer wearable data for the early detection of complications after pediatric appendectomy.

When children are discharged from the hospital after surgery, their caregivers often rely on subjective assessments (e.g., appetite, fatigue) to monitor postoperative recovery as objective assessment tools are scarce at home. Such imprecise and one-dimensional evaluations can result in unwarranted emergency department visits or delayed care. To address this gap in postoperative monitoring, we evaluated the ability of a consumer-grade wearable device, Fitbit, which records multimodal data about daily physical activity, heart rate, and sleep, in detecting abnormal recovery early in children recovering after appendectomy. One hundred and sixty-two children, ages 3-17 years old, who underwent an appendectomy (86 complicated and 76 simple cases of appendicitis) wore a Fitbit device on their wrist for 21 days postoperatively. Abnormal recovery events (i.e., abnormal symptoms or confirmed postoperative complications) that arose during this period were gathered from medical records and patient reports. Fitbit-derived measures, as well as demographic and clinical characteristics, were used to train machine learning models to retrospectively detect abnormal recovery in the two days leading up to the event for patients with complicated and simple appendicitis. A balanced random forest classifier accurately detected 83% of these abnormal recovery days in complicated appendicitis and 70% of abnormal recovery days in simple appendicitis prior to the true report of a symptom/complication. These results support the development of machine learning algorithms to predict onset of abnormal symptoms and complications in children undergoing surgery, and the use of consumer wearables as monitoring tools for early detection of postoperative events.

Discordance between Subjective and Objective Assessments of Activity after Pediatric Appendectomy.

BACKGROUND: Physical activity recovery after pediatric surgery can be assessed using objective measures such as step counts, but practice currently relies on subjective assessment by proxy. It is unclear how subjective and objective assessments of activity relate. We compared caregiver assessment of return to normal physical activity after pediatric appendectomy to step count recovery measured by a Fitbit. METHODS: Pediatric patients who underwent appendectomy were recruited between 2020 and 2022 to be monitored for 21 d with a Fitbit. Patients were grouped by the postoperative day (POD) (7, 14, or 21) their caregiver first reported their activity was "back to normal." Objective return to normal step count was estimated for each group by modeling the inflection point from increasing steps to a plateau. These measures were determined discordant if the subjective report remained outside the modeled 95% confidence interval (CI) for the day the group plateaued. RESULTS: Thirty-nine simple appendicitis and 40 complicated appendicitis patients were recruited. Among simple appendicitis patients, daily steps plateaued on POD 10.8 (95% CI 7.4-14.3), POD 14.0 (95% CI 11.0-17.1), and POD 11.1 (95% CI 6.9-15.3) for the day 7, day 14, and day 21 groups, respectively. Complicated appendicitis groups plateaued on POD 12.8 (95% CI 8.7-16.9), POD 15.2 (95% CI 11.1-19.3), and POD 16.7 (95% CI 12.3-21.0), respectively. Significant discordance was observed between subjective and objective assessments for the day 7 and day 14 simple groups and for the day 7 complicated group. CONCLUSIONS: There was significant discordance between caregiver and accelerometer-assessed activity recovery after pediatric surgery. Development of objective measures of recovery could help standardize assessment of children's recovery after surgery.

Effect of consumer-grade wearable device data on clinician decision making during post-discharge telephone calls after pediatric surgery.

BACKGROUND: During post-discharge telephone calls after pediatric surgery, clinicians must rely on parents/caregivers' assessment of symptoms, which can be inaccurate and often lead to unnecessary emergency department (ED) visits. Physiology (heart rate and physical activity) data from consumer-grade wearables, e.g., Fitbit™, may inform clinical decision making, yet there has been little study of clinician interpretation of this data. This study assessed whether wearable data availability, during simulated telephone calls about postoperative, post-discharge pediatric patients, affects clinician decision making. METHODS: Three simulated telephone call scenarios were presented to a diverse group of pediatric surgery clinicians. The scenarios were based on actual postoperative patients (scenarios 1 and 3 have worrisome symptoms and scenario 2 has non-worrisome symptoms) who had worn a Fitbit™ postoperatively. Each scenario was presented to clinicians (1) without any wearable data; (2) with "concerning" wearable data; and (3) with "reassuring" wearable data. Clinicians rated their likelihood, on a scale of 1-10, of recommending an emergency department (ED) visit for the three instances of each scenario, 10 being definitely ED. RESULTS: Twenty-four (24) clinicians participated in the study. When presented with "reassuring" wearable data, clinicians' likelihood of recommending an ED visit decreased from a median score of 6 to 1 (p < 0.001) for scenario 1 and from 9 to 3 (p < 0.001) for scenario 3. When presented with "concerning" wearable data, the median likelihood of recommending an ED visit increased from 1 to 6 (p = 0.003) for scenario 2. CONCLUSION: This study showed that wearable data affect clinicians' decision making and may be useful in triaging postoperative, post-discharge pediatric patients. LEVEL OF EVIDENCE: V.

Utility of Wearable Sensors to Assess Postoperative Recovery in Pediatric Patients After Appendectomy.

BACKGROUND: Despite more than two million pediatric operations performed in the United States annually, normal postoperative recovery remains difficult to define. Wearable sensors that assess physical activity and vital signs in real time represent a tool to assess postoperative recovery. This study examined the use of a wearable, the FitBit Inspire HR, to describe recovery in children after appendectomy and to determine the sensitivity of wearable data to distinguish disease severity. MATERIALS AND METHODS: Children 3-18 y old undergoing appendectomy in a tertiary children's hospital were invited to participate. Participants wore the FitBit Inpire HR after surgery for 21 d. t-tests compared daily step counts, and piecewise linear regression models were fit to examine recovery trajectories for patients with simple and complicated appendicitis. RESULTS: Thirty-two patients were enrolled, and 26 met eligibility criteria. Nine (35%) children had complicated appendicitis, and 14 (54%) were female; the mean age was 9.1 y (standard deviation: 2.9). Four hundred nineteen postoperative days were captured (range: 8-22 d; median: 16 d). Step counts increased after surgery; however, piecewise models showed that patients with simple appendicitis had a more rapid increase (P < 0.01) and reached a plateau (approximately 8000 steps/d) on postoperative day 9, whereas patients with complicated appendicitis did not reach a plateau and had lower step counts during the entire 21-postoperative day period (P < 0.01). CONCLUSIONS: Recovery in children after surgery can be characterized using wearables, which can also distinguish recovery trajectories based on disease severity. Establishing such "normative" recovery patterns may lead to earlier detection of complications.