Objective To investigate whether subchondral bone mediates the relationship between knee malalignment and osteophyte progression. Design Data were obtained from the Osteoarthritis Initiative (OAI) and two ancillary imaging studies. Two analytic cohorts were defined by mediator type: a DXA cohort (periarticular bone mineral density [paBMD]) and an MRI cohort (trabecular morphometry: aBVF, aTb.N, aTb.Sp, aTb.Th). Alignment (hip–knee–ankle [HKA] angle) was assessed at either 12 or 24 months under the Lower Limb Alignment study; bone measures were obtained at either 30 or 36 months through the Bone Ancillary Study; and radiographic osteophyte scores were evaluated at baseline and 48 months. Knees were categorized into five alignment groups (neutral, mild/severe varus, mild/severe valgus). Analyses used generalized estimating equations and multivariable linear regression with non-parametric bootstrapping. Results A total of 826 knees (DXA cohort) [ (mean age 61.9 years (SD 9.1), BMI 29.8 kg/m2 (SD 4.5); 60.5% female; 79.8% KL grade ≥ 2)] and 401 knees (MRI cohort) were analyzed. Only severe varus and valgus were associated with osteophyte progression. For severe varus, medial progression was mediated by higher medial paBMD and a higher medial-to-lateral (M:L) paBMD ratio, accounting for 20.3% and 70.7% of the total effect for the femur, and 23.2% and 65.4% for the tibia. Medial trabecular morphometry contributed to medial tibial progression, with indirect effects via higher aBVF (27.2%) and aTb.N (35.4%), and via lower aTb.Sp (12.2%). For severe valgus, lateral progression was mediated by higher lateral paBMD and a lower M:L paBMD ratio, accounting for 14.2% and 67.9% for the femur, and 9.3% and 40.7% for the tibia. Conclusions Compartment-specific bone measures partly mediate the association between static malalignment and osteophyte progression, suggesting that targeting bone may modify this longitudinal relationship. <Table Float="No" ID="Taba"> <tgroup cols="2"> <colspec align="left" colname="c1" colnum="1" /> <colspec align="left" colname="c2" colnum="2" /> <tbody> <row> <entry align="left" nameend="c2" namest="c1"> Key Points • Compartment-specific paBMD mediates the relationship between malalignment and osteophyte progression, with the M:L BMD ratio strengthening this effect more than either medial or lateral paBMD alone. • Medial trabecular morphometry partially mediates the varus alignment-osteophyte progression relationship. • Subchondral bone remodeling is locally driven by mechanical stress, as systemic measures like femoral neck BMD do not mediate the malalignment-osteophyte link. </entry> </row> </tbody> </tgroup> </Table>

Subchondral bone density and trabecular morphometry mediate the association between knee malalignment and osteophyte progression in osteoarthritis: data from the osteoarthritis initiative

期刊论文
发表日期：

2026年4月9日

查看全文

Shiwen Yuan,
Guangfeng Ruan,
Xiaodong Wang,
Grace H. Lo,
Cuicui Wang,
Zhaohua Zhu,
Siqi Xu,
Weinian Li,
Xiaoyan Cai,
David J. Hunter

摘要

Objective

To investigate whether subchondral bone mediates the relationship between knee malalignment and osteophyte progression.

Design

Data were obtained from the Osteoarthritis Initiative (OAI) and two ancillary imaging studies. Two analytic cohorts were defined by mediator type: a DXA cohort (periarticular bone mineral density [paBMD]) and an MRI cohort (trabecular morphometry: aBVF, aTb.N, aTb.Sp, aTb.Th). Alignment (hip–knee–ankle [HKA] angle) was assessed at either 12 or 24 months under the Lower Limb Alignment study; bone measures were obtained at either 30 or 36 months through the Bone Ancillary Study; and radiographic osteophyte scores were evaluated at baseline and 48 months. Knees were categorized into five alignment groups (neutral, mild/severe varus, mild/severe valgus). Analyses used generalized estimating equations and multivariable linear regression with non-parametric bootstrapping.

Results

A total of 826 knees (DXA cohort) [ (mean age 61.9 years (SD 9.1), BMI 29.8 kg/m² (SD 4.5); 60.5% female; 79.8% KL grade ≥ 2)] and 401 knees (MRI cohort) were analyzed. Only severe varus and valgus were associated with osteophyte progression. For severe varus, medial progression was mediated by higher medial paBMD and a higher medial-to-lateral (M:L) paBMD ratio, accounting for 20.3% and 70.7% of the total effect for the femur, and 23.2% and 65.4% for the tibia. Medial trabecular morphometry contributed to medial tibial progression, with indirect effects via higher aBVF (27.2%) and aTb.N (35.4%), and via lower aTb.Sp (12.2%). For severe valgus, lateral progression was mediated by higher lateral paBMD and a lower M:L paBMD ratio, accounting for 14.2% and 67.9% for the femur, and 9.3% and 40.7% for the tibia.

Conclusions

Compartment-specific bone measures partly mediate the association between static malalignment and osteophyte progression, suggesting that targeting bone may modify this longitudinal relationship.

Key Points

• Compartment-specific paBMD mediates the relationship between malalignment and osteophyte progression, with the M:L BMD ratio strengthening this effect more than either medial or lateral paBMD alone.

• Medial trabecular morphometry partially mediates the varus alignment-osteophyte progression relationship.

• Subchondral bone remodeling is locally driven by mechanical stress, as systemic measures like femoral neck BMD do not mediate the malalignment-osteophyte link.