High bone mass and cam morphology are independently related to hip osteoarthritis: findings from the High Bone Mass cohort

Abstract

Background High bone mass (HBM, BMD Z-score ≥ + 3.2) and cam morphology (bulging of lateral femoral head) are associated with greater odds of prevalent radiographic hip osteoarthritis (rHOA). As cam morphology is itself a manifestation of increased bone deposition around the femoral head, it is conceivable that cam morphology may mediate the relationship between HBM and rHOA. We therefore aimed to determine if individuals with HBM have increased odds of prevalent cam morphology. In addition, we investigated whether the relationship between cam and prevalent and incident osteoarthritis was preserved in a HBM population. Methods In the HBM study, a UK based cohort of adults with unexplained HBM and their relatives and spouses (controls), we determined the presence of cam morphology using semi-automatic methods of alpha angle derivation from pelvic radiographs. Associations between HBM status and presence of cam morphology, and between cam morphology and presence of rHOA (or its subphenotypes: osteophytes, joint space narrowing, cysts, and subchondral sclerosis) were determined using multivariable logistic regression, adjusting for age, sex, height, weight, and adolescent physical activity levels. The association between cam at baseline and incidence of rHOA after an average of 8 years was determined. Generalised estimating equations accounted for individual-level clustering. Results The study included 352 individuals, of whom 235 (66.7%) were female and 234 (66.5%) had HBM. Included individuals contributed 694 hips, of which 143 had a cam deformity (20.6%). There was no evidence of an association between HBM and cam morphology (OR = 0.97 [95% CI: 0.63–1.51], p = 0.90) but a strong relationship was observed between cam morphology and rHOA (OR = 3.96 [2.63–5.98], p = 5.46 × 10–11) and rHOA subphenotypes joint space narrowing (OR = 3.70 [2.48–5.54], p = 1.76 × 10–10), subchondral sclerosis (OR = 3.28 [1.60–6.60], p = 9.57 × 10–4) and osteophytes (OR = 3.01 [1.87–4.87], p = 6.37 × 10–6). Cam morphology was not associated with incident osteoarthritis (OR = 0.76 [0.16–3.49], p = 0.72). Conclusions The relationship between cam morphology and rHOA seen in other studies is preserved in a HBM population. This study suggests that the risk of OA conferred by high BMD and by cam morphology are mediated via distinct pathways.

Associations between life course longitudinal growth and hip shapes at ages 60 to 64 years: evidence from the MRC National Survey of Health and Development

“…Similarly, in the Johnston County Osteoarthritis Project, variations with hip shape were associated with prevalent knee radiographic osteoarthritis [19], and in a high bone mass cohort, there was strong evidence of an association between cam morphology (bulging of the lateral femoral head) and radiographic hip osteoarthritis, consistent with other studies [20]. However, there was no evidence of an association between high bone mass and cam morphology, suggestive of distinct pathways by which the risk of osteoarthritis is conferred by these [20]. Therefore, despite the majority of osteoarthritis research focussing on the impact on articular cartilage, these existing studies together with our novel data suggest that bone related changes in shape should be investigated further, especially in the early stages of disease and when considering disease predisposition.…”

“…The study found that significant changes in the shape of the proximal femur occurred within those individuals who developed osteoarthritis [3]. Similarly, in the Johnston County Osteoarthritis Project, variations with hip shape were associated with prevalent knee radiographic osteoarthritis [19], and in a high bone mass cohort, there was strong evidence of an association between cam morphology (bulging of the lateral femoral head) and radiographic hip osteoarthritis, consistent with other studies [20]. However, there was no evidence of an association between high bone mass and cam morphology, suggestive of distinct pathways by which the risk of osteoarthritis is conferred by these [20].…”

Deriving alpha angle from anterior-posterior dual-energy x-ray absorptiometry scans: an automated and validated approach

“…In addition, this code was developed using DXAs of only left hips which might be systematically different in shape to right hips. In an independent study, we have applied these methods to both right and left hip AP radiographs (the right hip images were mirrored before applying the outline points and code), validating the results with manual measures from two clinical annotators 31 . The automated methods performed well against manual measures in this study but more work is needed to assess the reproducibility of these results in external cohorts and clinical environments.…”

Osteoporosis, Osteoarthritis, and Subchondral Insufficiency Fracture: Recent Insights

“…Despite numerous reports linking OP to OA and SIF, a consensus remains elusive. Furthermore, several reports have associated high BMD with OA [20,[106][107][108][109]. However, these findings may be indicative of subchondral sclerosis in late-stage OA, with decreased rigidity reportedly caused by insufficient mineralization in the sclerotic subchondral bone [214].…”

“…Several studies have reported on the association between high BMD and lowerextremity OA [20,106,107]. In the UK Biobank study, evidence of the causality of all OA, KOA, and HOA was observed for high femoral neck BMD [18].…”