Gait mechanics at the hip may affect medial joint loading at the knee in persons with knee OA. Greater external hip adduction moment (reflecting torque generation by hip abductor muscles) during gait may limit excessive hip drop or trunk lean. This, in turn, may prevent excessive medial knee joint loading and eventual progression of knee OA. Little is known about the relationship between hip adduction moment and progression of medial tibiofemoral OA by MRI. We hypothesized that greater baseline external hip adduction moment is associated with reduced risk of baseline-to-2-year progression of medial tibiofemoral cartilage damage and bone marrow lesions in persons with knee OA. Methods

Participants with knee OA, defined by osteophyte presence in at least one knee, underwent quantitative gait analysis at baseline. Both knees underwent 3.0T MRI at baseline and two year follow-up using double oblique coronal and axial FLASHwe, coronal T1weighted spinecho, and sagittal, axial and coronal fat suppressed turbo spin echo sequences. Cartilage damage and bone marrow lesions assessed using WORMS, blinding readers to hypotheses and all other data. Logistic regression with GEE used to account for correlation between limbs, assess the association between baseline hip adduction moment and baseline-to-2-year cartilage damage progression and bone marrow lesion progression, defined as any worsening of score in medial compartment or joint surface. Analyses adjusted for gait speed, age, gender, and disease severity (K/L grade). Results reported as ORs and 95% CIs. Results

204 persons [64.2 years (±9.9), BMI 28.5 kg/m² (±5.7), 76.5% women] contributing 391 knees comprised the study sample. Mean hip adduction moment was 4.43 (±0.9) %BW*HT. Table 1 shows a greater hip adduction moment was significantly associated with reduced likelihood of cartilage damage progression in the medial tibiofemoral compartment and medial femoral surface. After adjustment for covariables, these associations were no longer significant. Table 2 shows means (SD) of the hip adduction moment at baseline for knees with/without cartilage damage and bone marrow lesion progression in specific subregions. Conclusion

In persons with knee OA, greater hip adduction moment was associated with reduced likelihood of 2-year medial tibiofemoral cartilage damage progression and bone marrow lesion progression, but findings were no longer significant in adjusted analyses. The consistently protective direction of findings at compartment, joint surface, and subregional levels warrants further evaluation at longer follow-up.