054 - TIBIAL ACCELERATION WHILE WALKING IS NOT ASSOCIATED WITH 2-YEAR CHANGES IN KNEE PAIN: THE MULTICENTER OSTEOARTHRITIS STUDY

Samantha Price; Joshua Stefanik; David Felson; Nene Ukonu; Michael LaValley; Michael C. Nevitt; Cora E. Lewis; James Torner; Patrick Corrigan

doi:10.1016/j.joca.2026.01.066

Purpose (the aim of the study): Walking biomechanics play an important role in the development and progression of knee osteoarthritis (OA). However, because of equipment costs, time, and complexity, most biomechanical metrics are inaccessible outside of research settings. With recent advancements in wearable technology, it is now possible to collect and analyze accelerometer data in clinical settings with little burden. Peak acceleration of the tibia (i.e., maximal acceleration along the long axis of the tibia) may be a simple metric that clinicians can use to characterize knee joint forces and identify individuals at risk for disease progression. A recent study showed that increased peak tibial acceleration during walking is cross-sectionally associated with increased pain in adults with knee OA. Yet, it remains unknown if peak tibial acceleration can be used to predict longitudinal changes in knee pain. The purpose of this study was to determine if peak tibial acceleration is associated with 2-year changes in knee pain. We hypothesized that greater peak tibial acceleration while walking would be associated with worsening knee pain. Methods: This study used data from the 12-year visit of the Multicenter Osteoarthritis (MOST) Study. MOST is a NIH-funded, observational, cohort study of >3000 adults with or at risk for knee OA. Participants completed two 20-meter walking trials, instructed to walk as fast as possible, with inertial measurement units (OpalTM, APDM Inc., Portland, OR) secured to their low back and bilaterally to each tibia, immediately proximal to the lateral malleoli. Three-dimensional acceleration data were continuously collected at 128 Hz. Using a custom-written MATLAB script (R2024b, Natick, MA, USA), peak tibial acceleration was extracted for each step (Figure 1) and averaged within limbs. To avoid inclusion of data during acceleration and deceleration phases of the 20-meter walk, 5 strides were removed from the beginning and end of each trial. The outcome of interest was worsening knee pain from the 12- to 14-year visit. A 2-point or greater increase (MCID) in the Western Ontario and McMaster Universities Arthritis Index (WOMAC, 0-20) pain subscale was the primary outcome definition. A 20-point or greater increase (MCID) in the Visual Analog Scale (VAS, 0-100) for knee pain over the past month was a secondary outcome definition. To characterize the relation of peak tibial acceleration to worsening knee pain, we constructed logistic regression models with generalized estimating equations to account for bilateral data from each participant. Covariates included age, sex, body mass index, clinic site, race, baseline pain, WOMAC function subscale (0-68), depression (CES-D≥16), and pain catastrophizing (I feel my pain is terrible and never going to get better, ≥1). Separate models were used for each outcome definition. Knees were excluded if they had a history of a replacement at the 12-year visit, missing tibial acceleration data, missing outcome data, missing covariate data at the 12-year visit, or history of a neurological condition at the 12-year visit. Results: 3,043 knees from 1,591 participants (924(58%) F; age=65(10) years; BMI=29.5(6.6) kg/m2) were included in the WOMAC-P model, and 2,805 knees from 1,462 participants (849(58%) F; age=64(10) years; BMI=29.4(5.6) kg/m2) were included in the VAS model. Mean (SD) peak tibial acceleration was 2.5 (0.5) g. Using the WOMAC pain definition, knee pain worsened for 605 knees (20%) between the 12- and 14-year visits. Using the VAS pain definition, knee pain worsened for 298 knees (11%) between the 12-year and 14-year visits. Peak tibial acceleration was not significantly associated with worsening knee pain using WOMAC-P (OR=1.09, 95% CI=0.88-1.34) or VAS (OR=1.14, 95% CI=0.91-1.44). Conclusions: Peak tibial acceleration was not significantly associated with worsening knee pain over 2-years. Alternative, clinically accessible metrics that are associated with knee pain in adults with or at risk for knee OA need to be identified.

054 - TIBIAL ACCELERATION WHILE WALKING IS NOT ASSOCIATED WITH 2-YEAR CHANGES IN KNEE PAIN: THE MULTICENTER OSTEOARTHRITIS STUDY

View Online

Abstract

Details

Metrics