Usefulness of 3D joint space width on weight-bearing CT in comparison with 2D joint space width on radiographs for predicting 24-month worsening of knee osteoarthritis pain and function in the MOST study.

OBJECTIVE

This study compared the predictive validity of three-dimensional joint space width (3D-JSW) on weight-bearing computed tomography (WBCT) versus two-dimensional joint space width at predetermined mediolateral locations (2D-JSWx) on radiographs for worsening knee pain and physical function over 24 months.

DESIGN

Data from 302 participants (425 knees) in the Multicenter Osteoarthritis Study (MOST) were analyzed. Baseline assessments included bilateral standing radiographs for 2D-JSWx and WBCT for 3D-JSW. Minimal clinically important worsening (MCIW) of pain (Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain subscale) and function (20-m walk, sit-to-stand, WOMAC function subscale) over 24 months was assessed. Logistic regression with generalized estimating equations (GEE) accounted for within-person correlation, and predictive validity was evaluated by comparing receiver operating characteristic (ROC) areas under the curves (AUC).

RESULTS

Participants had a mean age of 63.2±8.9 years and body mass index (BMI) of 28.2±4.8 kg/m². Over 24 months, 38 knees exhibited worsening WOMAC pain, with no significant association for either 3D-JSW or 2D-JSWx. For WOMAC pain, 3D-JSW (AUC=0.578) was not superior to 2D-JSWx (AUC=0.511; difference: 0.067 (95% confidence interval (CI): -0.040, 0.175)). Similar results were noted for the 20-m walk (difference: 0.003 (95% CI: -0.049, 0.055)), sit-to-stand test (difference: 0.014 (95% CI: -0.034, 0.062)), and WOMAC function (difference: -0.020 (95% CI: -0.091, 0.050)).

CONCLUSIONS

3D-JSW on WBCT did not outperform 2D-JSWx on radiography for predicting knee pain and functional worsening. However, WBCT offers several advantages, including improved imaging capabilities, that may facilitate assessment of osteoarthritis in clinical practice and research.