Session Information
Session Type: Abstract Submissions (ACR)
Background/Purpose: We investigated hip shape by active shape modeling (ASM) as a potential predictor of incident radiographic and symptomatic hip OA (rHOA and srHOA) in a community-based study that includes African American and Caucasian men and women.
Methods: All hips developing rHOA from baseline (Kellgren-Lawrence grade [KLG] 0 or 1, 1991-7) to follow up (KLG >=2, 1999-2004, mean 6 years follow up, 190 hips), and 1:1 control hips (KLG 0 or 1 at both timepoints, 192 hips) were selected in approximately equal numbers from 4 race-by-gender strata. The shape of the proximal femur was defined on a baseline AP pelvis radiograph for all hips by a trained reader (AEN), and 60 landmark points were input into an ASM. The ASM produced a mean shape, plus continuous variables representing independent modes of variation in that shape. The scores for modes which together explained 95% of shape variance (n=14) were simultaneously included in logistic regression models as independent predictors, with incident 1) rHOA and 2) srHOA (defined as rHOA plus symptoms in the case hip at follow up) as the dependent variables, and adjusted for intra-person correlations. Analyses were adjusted for sex, race, body mass index (BMI), and baseline KLG and/or symptoms. Stratified analyses for sex, baseline KLG and symptoms were performed.
Results: We evaluated 382 hips (Figure) from 342 individuals: 61% women, 83% Caucasians, with a mean age of 62 years and BMI of 29 kg/m2. No ASM modes were associated with incident rHOA in the sample, or in stratified analyses among women. However, among men only, modes 1 and 2 (53% of total shape variance) were significantly associated (for a 1-SD decrease in mode 1 score, OR 1.7 [95% CI 1.1, 2.5], and for a 1-SD increase in mode 2 score, OR 1.5 [95% CI 1.0, 2.2]) with incident rHOA (Figure, left).
For incident srHOA, modes 2 and 3 (representing 16 and 13% of total shape variance, respectively) were significantly associated, with a 1-SD decrease in either of these modes increasing the odds of srHOA by 50% (mode 2: OR 1.5 [95% CI 1.0, 2.1], mode 3: figure, right). The presence of baseline hip symptoms increased the odds (OR 3.2 [95% CI 1.7, 5.9]) of incident srHOA, while African Americans compared to Caucasians had 70% lower odds of incident srHOA (OR 0.3, [95% CI 0.1, 0.8]); no other covariates were associated. Analyses stratified by the presence of baseline symptoms showed a consistent association between mode 3 and srHOA. However, among those without baseline symptoms, an increase in mode 6 (3% of variance) was associated with srHOA (OR 1.9 [95% CI 1.2, 3.1]), while among those with symptoms the association was with a decrease in mode 6 (OR 2.1 [95% CI 1.3, 3.5]).
Conclusion: Variations in shape modes 1 and 2, derived from the ASM, were associated with incident rHOA in men only. Shape modes 2 and 3 were associated with srHOA overall, and mode 6 with srHOA depending upon baseline symptom presence. Such shape variations may contribute to hip OA risk.
Disclosure:
A. E. Nelson,
None;
F. Liu,
None;
J. A. Lynch,
None;
J. B. Renner,
None;
T. A. Schwartz,
None;
N. E. Lane,
None;
J. M. Jordan,
None.
« Back to 2012 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/incident-symptomatic-hip-osteoarthritis-is-associated-with-differences-in-hip-shape-by-active-shape-modeling-the-johnston-county-osteoarthritis-project/