Knee OA Severity and Hip Movement Quality — C.R.A.F.T. Scores Across Kellgren–Lawrence Grades in 177 Patients
All Conferences
London, United Kingdom · 2026

09 · Motion Analysis

Knee OA Severity and Hip Movement Quality

C.R.A.F.T. Scores Across Kellgren–Lawrence Grades in 177 Patients

Hip movement quality deteriorated as radiographic knee osteoarthritis worsened — in patients who had no hip osteoarthritis at all. Three of five C.R.A.F.T. hip scores differed significantly across Kellgren–Lawrence grades.

Presentation Details

Presented, and open to scrutiny.

Each presentation connects surgical decision-making, markerless movement measurement and quantitative imaging — bringing MSK Doctors Research to an international clinical and scientific audience.

Event

BOA Annual Congress 2026, Olympia London

Format

Scientific e-poster

Authors

Feza Korkusuz · Selin Demirel — Hacettepe University, Sports Medicine, Ankara, Türkiye; Paul Lee (presenting) · Tanvi Verma — MSK Doctors, Sleaford, UK

Themes

Kinetic chainC.R.A.F.T. frameworkMarkerless motion analysisKellgren–Lawrence grading

Scientific Focus

Knee osteoarthritis affects not only the knee but proximal kinetic-chain biomechanics, including hip movement patterns. This study assessed 177 patients with knee osteoarthritis and no radiographic hip osteoarthritis, graded by Kellgren–Lawrence — control (n=21), grade 1 (n=32), grade 2 (n=63), grade 3 (n=40) and grade 4 (n=21). Hip movement was captured during routine outpatient evaluation using MAI Motion, an AI-based markerless platform, and scored on the C.R.A.F.T. framework: control, repetition, asymmetry, flow and rotational dynamics (twist). Analysis used Kruskal–Wallis testing with Dunn–Bonferroni post-hoc comparison. Ethical approval: Hacettepe University Health Sciences Research Ethics Committee, Decision No. 2025/21-22, Research No. SBA 25/369.

Why It Matters

Three of the five hip scores showed significant overall group differences — Hip-Flow (p=0.011), Hip-Twist (p=0.032) and Hip-Overall (p=0.002) — while Hip-Repetition (p=0.182) and Hip-Asymmetry (p=0.615) did not. Significant pairwise comparisons were Twist control versus grade 4 (p=0.012), and Overall control versus grades 2, 3 and 4 (p=0.002, 0.04 and 0.032). The finding is that the hip is already moving differently before it shows any radiographic disease of its own, which suggests early detection and treatment of knee osteoarthritis — particularly at lower grades — may help prevent secondary changes in hip biomechanics. Only p values were reported; individual score means and dispersion were not available.

Next in the programme

The Human Algorithm

Designing Knee Surgery through Robotics, Motion and Biology

View Conference
Privacy & Cookies Policy