2023 ISAKOS Biennial Congress Paper
Smartphone-Based Step-Count Measures Correlate with KOOS-12 Function and UCLA Activity Proms During Early TKA Recovery
Corey E. Ponder, MD, Edmond, OK UNITED STATES
Eric M. Slotkin, DO
Paramjeet Gill, MD, Fresno, CA UNITED STATES
Simon Coffey, MBBS, FRACS(Orth), Beecroft, NSW AUSTRALIA
Stephen McMahon, FRACS, Melbourne, VIC AUSTRALIA
Edgar A Wakelin, PhD, Raynham, Massachusetts UNITED STATES
Christopher Plaskos, PhD, Raynham, MA UNITED STATES
Alexander Orsi, PhD UNITED STATES
Jeffrey Michael Lawrence, MD, Viroqua, Wisconsin UNITED STATES
John M. Keggi, MD, Middlebury, CT UNITED STATES
Corin Group, Raynham, MA, UNITED STATES
FDA Status Cleared
Summary
High step count led to improved PROMs scores compared to low step-count across all time points.
Abstract
Introduction
Passive smartphone-based apps are becoming more common for measuring patient progress after total knee arthroplasty (TKA). Optimum activity levels during early TKA recovery haven’t been well documented. Correlations between step-count and patient reported outcome measures (PROMs) during early recovery were explored. This study also investigated how demographics impact step-count during early post-operative recovery.
Methods
Smartphone capture step-count data from 456 TKA patients was retrospectively reviewed. Mean age was 68±8years. 61% were female. Mean BMI was 31±6kg/m2. Mean daily step count was calculated over three time-windows: 60 days prior to surgery (preop), 5-6 weeks postop (6wk), and 11-12 weeks postop (12wk).
Linear correlations between step-count and KOOS12-function and UCLA activity scores were performed. Patients were separated into three step-count levels: low (<1500steps/day), medium (1500-4000steps/day), and high (>4000steps/day). Age >65years, BMI >30kg/m2, and sex were used for demographic comparisons.
Student’s t-tests determined significant differences in mean step-counts between demographic groups, and in mean PROMs between step-count groups.
Results
UCLA correlated with step-count at all time-windows (p<0.001). KOOS12-Function correlated with step-count at 6wk and 12wk (p<0.05). High step-count individuals had improved PROMs scores compared to low step-count individuals preoperatively (UCLA: delta=1 [p<0.001]), at 6wk (UCLA: delta=0.8 [p<0.01], KOOS12-Function: delta=6 [p<0.05]), and at 12wk (UCLA: delta=0.8 [p<0.01], KOOS12-Function: delta=6.5 [p<0.05]).
Younger patients had greater step-count preoperatively (3.6±2.8k vs. 2.6±2.4k, p<0.001), and at 12wk (3.8±2.6k vs. 2.7±2.3k, p<0.01). Males had greater step-count preoperatively (3.7±2.7k vs. 2.5±2.5k, p<0.001), at 6wk (3.5±2.6k vs. 2±2.3k, p<0.001), and at 12wk (3.7±2.1 vs. 2.7±2.6k, p<0.01). No differences in step-count were observed between low and high BMI patients preoperatively, at 6wk, or at 12wk.
Conclusion
High step count led to improved PROMs scores compared to low step-count across all time points. Early post-operative step-count was significantly impacted by age and sex. Generic recovery profiles may not be appropriate across a diverse population.