Background/Purpose: Structural progression of knee OA has been associated with several radiographic and MRI-measured pathological features of OA, including malalignment, meniscal pathology and bone marrow changes but studies examining these risk factors generally examine one of them and not their constellation although these features coexist. To date no study has been done to assess if baseline MRI-based scores of various knee OA features predict progression of knee OA as determined by quantitative cartilage thickness measurement. We aimed to determine which semiquantitative MRI-based OA features were most predictive of subsequent cartilage thickness loss measured with quantitative MRI.

Methods: From the Multicenter OA Study subcohort, participants (one knee per person) who volunteered for a longitudinal substudy in which MRI measurement of cartilage thickness and volume were done were selected: These subjects also had conventional MRI’s allowing Whole Organ MRI Score (WORMS) based scoring of cartilage lesions, bone marrow lesions (BML), meniscal pathology, effusion synovitis, and Hoffa synovitis at both time points. Progression in the medial or lateral femorotibial compartment (MFTC/LFTC) was defined as cartilage thickness loss exceeding the change observed in OAI control cohort knees (mean ± 2xSD, MFTC/LFTC: -162µm/-145µm). All MRI predictors were dichotomized into “present” (score≥2 for cartilage, ≥0 for others) or “absent”. Differences in baseline scores of ipsi-compartmental predictor variables were compared between progressor and non-progressor knees by multivariable logistic regression, adjusting for age, sex, body mass index, and anatomical alignment axis (degrees). We combined MFTC and LFTC to calculate ajusted odds ratio (aOR) and 95% CI of ipsi-compartmental cartilage thickness loss across compartments, using Generalized Estimating Equations. Also, ORs and 95%CIs were calculated for MFTC and LFTC cartilage thickness loss, separately.

Results: 196 participants were included and their mean age was 59.8±6.3 years, mean BMI 29.5±4.6 kg/m², and 62% were women. 46 knees had radiographic knee OA at baseline. In the MFTC/LFTC, there were 35/29 progressors and 161/167 non-progressors. For analysis combining MFTC and LFTC, predictors of cartilage thickness loss were baseline cartilage lesions (aOR 2.6 [1.4-5.0]), BML (aOR 1.9 [1.1-3.3]), meniscal damage (aOR 4.5 [2.4-8.4]) and meniscal extrusion (aOR 3.3 [1.9-5.8]), all in the ipsilateral compartment, but not effusion synovitis or Hoffa synovitis. In MFTC-only analysis, MFTC progressors had higher aOR for having baseline medial meniscal damage (aOR 2.4, [95%CI 1.1-5.6]), medial meniscal extrusion (aOR 2.6 [1.1-5.8]), but not cartilage lesions, BML, effusion synovitis and Hoffa synovitis. In the LFTC-only analysis, baseline lateral cartilage lesions (aOR 3.4 [1.3-9.3]), lateral meniscal damage (aOR 13.9 [3.3-9.0]) and lateral meniscal extrusion (aOR5.0 [1.4-18.0]) predicted LFTC progression.

Conclusion: Amongst MRI features, the presence of cartilage lesions, BML, meniscal damage and extrusion in the ipsilateral FTC predict quantitatively assessed cartilage thickness loss over 30-months, but not effusion synovitis or Hoffa synovitis.

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/semiquantitative-mri-features-of-knee-osteoarthritis-show-compartment-specific-relationships-with-quantitative-cartilage-thickness-loss-the-multicenter-osteoarthritis-study/