Insights from Wearable Biometrics After Adult Spinal Deformity Surgery: Does Wearable Biometric Data Correlate to Complication and 30-Day Readmission Better Than Traditional Patient Reported Outcomes?
Chief, Spine Division Washington University in St. Louis Washington University in St. Louis
Disclosure(s):
Brian Neuman, MD: No relevant disclosure to display
Introduction: Adult spinal deformity (ASD) surgery has been shown to provide substantial long-term benefit to patients, however many patients ask about short-term recovery in the 4-6 weeks after surgery. Traditional patient reported outcomes (PROs) have failed to reliably predict post-operative course, with studies showing inconsistent results. This may reflect PROs capturing a single moment in time, whereas pain, function, and activity are dynamic processes. Wearable biometric data (WBD), including motion trackers like Fitbit, provide a more comprehensive picture of physical activity. We aim to evaluate the relationship between WBD, Patient-Reported Outcomes Measurement Information System (PROMIS) Physical Function, and postoperative course after ASD surgery.
Methods: ASD patients were enrolled at their preoperative visit. PROMIS scores were collected, and patients were provided a Fitbit to wear prior to surgery. Preoperatively, various activity metrics were recorded for a minimum of 1 week. Perioperative course and complications were recorded for 30 days after discharge, such as reoperation, deep vein thrombosis, dehiscence, infection, and others. Parametric and non-parametric analyses were performed to assess significance.
Results: A total of 23 ASD patients were enrolled. Average fusions levels were 10.4 (SD=3.6), all patients were fused to pelvis. All-posterior surgery was received by 8 patients and 15 received anterior-posterior surgery. Perioperative complications were experienced by 8 patients (35%) and 3 (13%) patients were readmitted within 30-days of discharge. Patients with greater activity measured by WBD were less likely to experience complications. Between complication and non-complication groups, number of steps per activity bout was 44.5 and 81.0 (p=0.017), active time per bout was 1.78 and 2.59 (p=0.028), and number of steps per minute of activity was 23.5 and 29.1 (p=0.028) respectively. No difference was found for PROMIS Physical Function between complication groups (p=0.104).
Conclusion : Increased preoperative activity determined by WBD was associated with decreased complications, however no relationship was found with self-reported activity questionnaires. These preliminary findings suggest that WBD is superior to traditional PROs in assessing activity levels and may have utility in predicting perioperative course.