Background/Purpose: With an annual incidence rate of 2-5% in high-risk populations, the use of established knee OA as an outcome challenges the feasibility of preventive OA research. Therefore, valid surrogate outcomes, for which short-term changes capture long-term OA incidence, are urgently needed. Therefore, this study aimed to assess the association of 2-year’s semi-quantitative scores for BMLs, cartilage defects, osteophytes, meniscus pathologies, meniscus extrusion, and effusion/synovitis worsening on MRI and 11-year’s incidence of radiographic OA, among subjects with an ACL-rupture.

Methods: 154 individuals (18-45 years) with an ACL-rupture confirmed by physical examination and MRI, free of radiographic features of knee OA (KLG = 0), were enrolled in the study within 6 months of their injury. At baseline and at two years, multi-sequential MRIs were obtained (sag. and cor. proton density–weighted turbo spin echo (slice thickness, 3 mm; TR/TE 2700/27 ms), cor. T2-weighted TSE with fat saturation (slice thickness, 3 mm; TR/TE 5030/71 ms), axial PD (TR/TE 3500/25 ms) and T2-weighted (TR/TE 3500/74 ms) TSE dual echo (slice thickness, 3 mm), and sag. T2-weighted 3-dimensional DESS (slice thickness, 1.5 mm; TR/TE 21.35/7.97 ms) and scored using MOAKS. After 11 years, weight-bearing semi-flexed AP-radiographs were obtained and scored for radiographic OA incidence (KLG ≥2). Two-year’s worsening of BMLs, cartilage defects, osteophytes (all in PF, medial and lateral TF compartments), medial and lateral meniscus pathology and meniscus extrusion, and of effusion/synovitis were determined, using established criteria. Features showing worsening in ≥10% of the knees were related to OA incidence after 11 years, using logistic regression analysis.

Results: Follow-up data after 11.7 ± 0.7 years was available for 99 individuals (baseline age 27.8 ± 7.2 years, 68% men). Over the first two years, 48 individuals (48%) underwent ACL-reconstruction surgery. After 11 years, 41 individuals (41%) developed radiographic OA in their injured knee. Worsening of lateral cartilage defects (23%), medial (24%) and lateral (28%) meniscus pathology, and medial meniscus extrusion (17%) reached the pre-defined threshold of ≥10% within two years. Despite increased post-test probabilities for meniscus pathology, two-year’s worsening of selected features was not significantly associated to long-term radiographic knee OA incidence (see Table).

Conclusion: Most knee OA MRI features showed little progression (< 10%) over two years, across compartments, in ACL-injured knees. Lateral TF cartilage defects, meniscus pathology, and medial meniscus extrusion showed high progression rates (18-26%), but these changes showed no statistical significant association to radiographic knee OA incidence after 11 years. Of the assessed features, only progression of meniscus pathology seemed to have potential as a surrogate outcome for long-term radiographic knee OA incidence, which should be confirmed in a larger sample.

« Back to ACR Convergence 2024

ACR Meeting Abstracts - https://acrabstracts.org/abstract/two-years-worsening-of-semi-quantitative-mri-features-as-surrogate-outcomes-for-long-term-incident-radiographic-knee-osteoarthritis-after-acl-rupture/