Integrating MR imaging with full-surface indentation mapping of femoral cartilage in an ex vivo porcine stifle

Abstract

The potential of MRI to predict cartilage mechanical properties across an entire cartilage surface in an ex vivo model would enable novel perspectives in modeling cartilage tolerance and predicting disease progression. The purpose of this study was to integrate MR imaging with full-surface indentation mapping to determine the relationship between femoral cartilage thickness and T2 relaxation change following loading, and cartilage mechanical properties in an ex vivo porcine stifle model. Matched-pairs of stifle joints from the same pig were randomized into either 1) an imaging protocol where stifles were imaged at baseline and after 35 min minutes of static axial loading; and 2) full surface mapping of the instantaneous modulus (IM) and an electromechanical property named quantitative parameter (QP). The femur and femoral cartilage were segmented from baseline and post-intervention scans, then meshes were generated. Coordinate locations of the indentation mapping points were rigidly registered to the femur. Multiple linear regressions were performed at each voxel testing the relationship between cartilage outcomes (thickness change, T2 change) and mechanical properties (IM, QP) after accounting for covariates. Statistical Parametric Mapping was used to determine significance of clusters. No significant clusters were identified; however, this integrative method shows promise for future work in ex vivo modeling by identifying spatial relationships among variables.

Results

24 matched pairs were acquired for testing. The 24 femoral specimens randomized to undergo indentation were probed and indented at Biomomentum, Inc. For the matched pair undergoing loading and imaging, 19 specimens were tested. Facility closures due to the COVID-19 pandemic precluded testing of the remaining five specimens. One loaded specimen was removed from analysis due to issues with post-intervention image reconstruction, therefore 18 specimens were used in the final analysis.

Acknowledgements

Thank you to Imaging Research Centre director Norm Konyer for your help with imaging. Thank you to Ariana Fraschetti, Cooper Gluek, Guillaume Hesketh, Greg Noseworthy, Jeff Rice, and Kevin Perera for their help with the MRI-compatible loading apparatus. We would also like to extend thanks to Drs Eric Quenneville and Sotcheadt Sim at Biomomentum, Inc (Laval, QC) for their assistance in indentation mapping.