Study at a glance

In 71 individuals awaiting partial or total knee replacement, markerless motion capture found no differences in single-speed walking or sit-to-stand performance between the two groups, but revealed that those awaiting partial replacement adapted their gait substantially more than those awaiting total replacement when asked to walk faster, a difference only multi-speed testing could detect.

Summary

Disclaimer: The following summary details independent academic research. Theia3D is intended for research, sports performance and educational use only. It is not cleared as a medical device and is not intended to be used for the diagnosis, treatment, mitigation or prevention of any disease, injury or medical condition.

Editor's note: Theia3D version v2023.1.0.3161, as stated in the preprint.

Study Overview

Design

A cross-sectional, between-groups comparison of pre-operative walking and sit-to-stand biomechanics in individuals awaiting partial versus total knee arthroplasty, using markerless motion capture.

Participants

Seventy-one individuals with end-stage knee osteoarthritis took part: 15 awaiting partial knee arthroplasty (average age 60, SD 7) and 56 awaiting total knee arthroplasty (average age 68, SD 9), a significant age difference between groups (p < .001). No other baseline measures, including sex distribution, body mass index, Oxford Knee Score, pain measures, quality of life, or depression screening scores, differed significantly between groups. Participants were recruited ahead of their scheduled surgery; 104 individuals were initially recruited, with 30 excluded for incomplete tasks and 3 further removed for unusable movement data.

Functional Tasks

Each participant completed 30 seconds of quiet standing, 60 seconds of walking at a self-selected pace, 30 seconds of walking at a faster self-selected pace, and five repeated sit-to-stand transitions, along a 7.6-meter walkway.

Data Collection and Analysis

Ten synchronized Sony RX0 II cameras recorded each task, with video processed through Theia3D to generate three-dimensional whole-body kinematics. Linear mixed models, with surgery type, walking condition, and their interaction as fixed effects and participant identity as a random effect, were used to analyze the walking trials; sit-to-stand measures were compared using t-tests or Mann-Whitney tests with Cohen's d effect sizes. The analysis focused on discrete joint-angle measures at the hip, knee, and ankle rather than full movement waveforms, a deliberate scope choice by the authors.

Key Findings

  • At a single walking speed, and during sit-to-stand, the two surgical groups performed similarly. No significant surgery-type differences emerged when comparing preferred-pace walking alone or the five sit-to-stand transitions, across the joint-angle measures examined.
  • Walking faster revealed a substantial difference the single-speed comparison missed. When moving from preferred-pace to faster-pace walking, those awaiting PKA showed significantly larger increases than those awaiting TKA in stride length (interaction effect, p = 0.003), knee flexion at peak stance (p = 0.004) and peak swing (p = 0.02), knee range of motion through the stride (p < .001), hip flexion at peak stance (p = 0.02), and hip range of motion (p = 0.02).
  • No ankle measures showed a significant difference between groups at any speed. Across both walking speeds, ankle-joint measures did not distinguish the PKA and TKA groups, suggesting the group difference was concentrated at the knee and hip rather than distributed across the whole limb.
  • The authors interpret the pattern as reflecting greater baseline functional capacity in the PKA group. Because PKA is typically reserved for less extensive joint damage, the authors suggest their larger increase in movement range when challenged to walk faster reflects a greater reserve of adaptability going into surgery, one that a single comfortable walking speed was not demanding enough to reveal.

What This Means for Orthopedic and Movement Researchers

For teams designing pre-operative functional assessments ahead of knee replacement, this preprint's central message is about test sensitivity: a single walking speed, however standard, may understate real differences between surgical candidate groups that only emerge under a more demanding task. Building a faster-paced walking condition, and not just a comfortable one, into pre-operative testing protocols could surface functional differences that would otherwise be missed. The authors are careful to frame their interpretation, that PKA candidates carry greater baseline adaptability into surgery, as consistent with their data rather than definitively proven, and note that confirming it would require follow-up work tracking the same individuals through and beyond their respective procedures.

The Role of Theia3D in Movement Analysis

Theia3D processed footage from a ten-camera array to generate the joint-level kinematics this comparison relied on, allowing the research team to capture walking and sit-to-stand biomechanics across a real-world population awaiting knee replacement without the setup burden that marker-based capture can add to an already demanding assessment visit. That efficiency supported testing participants across three distinct tasks and two walking speeds in a single session, which is what ultimately allowed the speed-dependent difference between groups to surface at all.

Study Limitations and Future Directions

The authors note several limitations to how these findings should be read. The analysis examined discrete joint-angle values at specific points in the movement cycle rather than full waveform patterns across the entire stride, a deliberate scope choice that leaves room for future work using more comprehensive movement-pattern analysis. The PKA and TKA groups were unevenly sized (15 versus 56), reflecting how these procedures are distributed in real-world surgical populations, but this imbalance limits statistical power to detect smaller differences between groups. The PKA group was also significantly younger, so age-related effects on movement capacity cannot be fully separated from differences tied to the surgical group itself. This preprint reports pre-operative data only, with no post-operative follow-up or analysis of surgical parameters, so how these pre-operative differences relate to recovery or long-term outcomes after either procedure remains an open question. The authors also note that this preprint has not yet been certified by peer review and, per standard preprint practice, should not be used on its own to guide care decisions ahead of formal publication.

Citation: Shishis, M. R., Gafoor, F., Ruder, M. C., Di Bacco, V. E., Madden, K., Adili, A., & Kobsar, D. (2025). Investigating pre-operative joint-level biomechanics in partial versus total knee arthroplasty. medRxiv. https://doi.org/10.1101/2025.09.21.25336261

Link: Read the full preprint

Does a Single Walking Speed Tell You Enough?

This preprint suggests that a comfortable, single-speed walk can hide real functional differences between surgical candidate groups that only appear under more demanding movement, a finding with implications well beyond knee arthroplasty. If your work involves pre-operative or outcome-tracking assessments and you want to understand how markerless motion capture could help build more sensitive, multi-speed testing into your protocols, talk to our team about what that could look like for your population.

Recent Posts