Background/Purpose : Responsive measures of radiographic joint space width (JSW) in hip osteoarthritis (OA) are a major requirement for the evaluation of treatment interventions. The primary purpose of this study was to validate a semi-automated quantitative software tool for hip JSW by measuring  responsiveness at 3 different fixed   locations.

Methods: We used data from the Osteoarthritis Initiative (OAI), a longitudinal cohort study of knee OA. All OAI participants had standing AP radiographs at baseline and 48 month visits using a standardized protocol.

We used a nested case-control design, and had two case definitions. First, subjects who had a total hip replacement (THR) sometime after the 48 month visit (at 60 and 72 months) and available AP pelvic films at 0 and 48 months (n=27) were selected and matched (1:1) on sex and age (2 years) to subjects without a THR and no hip pain during the study period.

Second, a larger sample of cases was selected that included all subjects who had a THR at any point after baseline and had AP pelvis films at 0 and 48 months (n=72). The contralateral (CL) hip from the THR was designated the case hip, and subjects were matched (1:1) as above.

Measurements of superior hip JSW were made at 3 fixed locations (10°, 30° and 50° medial to a line from the femoral head centre to the outer edge of the acetabular roof – lateral line in Figure 1) and were facilitated by software that delineated the femoral head and found the acetabular margin along each of the 3 lines. A reader used software to correct margins if needed.

Statistical analysis. Sensitivity to change was estimated by the standardized response mean for change from baseline to 48 months. Paired t-tests were used to test statistical significance between cases and controls.

Results: The overall sample was 47% male, 91% Caucasian had a mean age of 64.2 and BMI of 27.9.  The results are given in Table 1 and Figure 2.  Significant differences were observed between cases and controls in both case-control groups. The superior medial location (position 3) was the most responsive.

Conclusion: Location 3 (sup-med) was the most responsive in all 4 groups and warrants further study. Location specific measures of JSW are potentially an improved method to assess hip OA.

We acknowledge NIH/NIAMS R01AR056664.