INTERLIMB DIFFERENCES IN CARTILAGE STRAIN 6 MONTHS POST-ACLR: A PRE-OSTEOARTHRITIS BIOMARKER

Introduction: Knee osteoarthritis (OA) is a degenerative joint disease that is a leading cause of joint pain, disability, and health care costs worldwide. OA affects the whole joint and is characterized by a gradual loss of articular cartilage often due to repeated microtraumas from injury, overuse, or aging. Diagnosing early-stage post-traumatic OA can be difficult due to a lack of validated outcome measures [1]. The most common diagnosis metric, morphological changes to the cartilage, is limited to patients with moderate to severe OA, after irreversible changes to the joint have occurred. Therefore, non-invasively quantifying mechanical and biochemical changes to cartilage in vivo is of substantial interest. In recent years, considerable emphasis has been placed on the use of imaging methods and biomarkers for the early detection of OA-related biomechanical and biochemical changes. Previously, we coupled spiral displacement encoding with stimulated echoes (spiral DENSE) MRI to displacement under applied loading MRI (dualMRI) to quantify mechanical changes to the tibiofemoral cartilage. This technique was applied in cartilage explants, bovine and porcine knee joints, and in healthy in vivo data. In this work, we applied the same technique to capture in vivo mechanical strain and displacement maps of the human knee articular cartilage during varus loading. We imaged a cohort of patients who underwent anterior cruciate ligament reconstruction (ACLR) six months prior to the scan date. We then examined differences in strain and displacement in the reconstructed knee and healthy contralateral knee.

Materials and

Methods: Twelve individuals underwent MRI 6±2 months following ACLR. All subjects had no prior history of symptomatic knee pathologies on either knee. An MRI-compatible loading apparatus capable of providing a varus load to the tibiofemoral cartilage of the knee, leading to compression of the medial femoral condyle, was used to load the cartilage during scanning [3-5]. Moment conservation was used to calculate the load applied on the foot to be equivalent to 0.5 times body weight of load applied at the medial condyle. Patients were subjected to a 0.5 Hz cyclic loading regime (1s load & 1s unload) to mimic a walking cadence. The knee imaging protocol consisted of 3D double echo steady state (DESS) acquisition and DENSE MRI during cyclic varus loading on the knee joint. All imaging was performed using a clinical MRI system (3-T; Siemens Prismafit) with a 15-channel quadrature knee coil (Tx/Rx Knee 15 Flare Coil, QED, LLC). The articular cartilage was segmented via a semi-automatic segmentation algorithm into two regions of interest (ROI) based on the anatomical regions of the knee (i.e., medial femur, medial tibia). This was repeated on all 27 time frames to obtain raw displacement values. The displacement values were later smoothed and in-plane Green-Lagrange strains (Exx, Exy, Eyy) were calculated [2]. The spatial average (SA) of the pixels within each ROI (femoral and tibial) was then calculated. A paired t-test was used to compare the ACLR and uninjured knee.

Results, Conclusions, and Discussions: Both tibial and femoral displacement in the x direction were significantly different between the ACLR and contralateral limbs (p=0.006 and p=0.004). On average, displacement was less in the ACLR limb compared to the contralateral limb. This may suggest that the cartilage is more stiff in the injured limb. Additionally, shear strain (Exy) was significantly higher in the reconstructed limb, 0.035 +/- 0.035, versus 0.01+/- 0.031 in the contralateral limb (p=0.027). The increased strain indicates the injured limb experienced more compression during the cyclic loading. In other recent work, higher shear strain has been found to correlate with worse patient reported outcome scores [6]. Imaging methods that are capable of detecting pre-osteoarthritic cartilage changes are critical to assess diseased tissue before irreversible changes to the knee joint occur. This work demonstrates that both strains and displacements quantified by dualMRI could be associated with interlimb differences, and further demonstrates that mechanical imaging biomarkers could be sensitive to early OA.

Acknowledgements (Optional): This work was supported by NIH grant 2 R01 AR063712.

[1] MacKary J.W., et al., Osteoarthr Cartil., 2018. [2] Chan D. et al., Sci. Rep. 2016. [3] Lee W. et al., MRM 90(3), 2023. [4] Lee W. et al., MRM 89(2), 2023. [5]. Zhu H. et al., SSRN 4569548. [6] Miller E.Y. et al., MedRxiv, 2024