Validation of quantitative magnetic resonance imaging-based apparent bone volume fraction in peri-articular tibial bone of cadaveric knees

Background: In the knee, high-resolution magnetic resonance (MR) imaging has demonstrated that increased apparent bone volume fraction (trabecular bone volume per total volume; BV/TV) in the peri-articular proximal medial tibia is associated with joint space narrowing and the presence of bone marrow lesions. However, despite evidence of construct validity, MR-based apparent BV/TV has not yet been cross-validated in the proximal medial tibia by comparison with a gold standard (e.g., micro-computed tomography [microCT]). In this cadaveric validation study we explored the association between MR-based apparent BV/TV and microCT-based BV/TV in the proximal peri-articular medial tibia. Methods: Fresh cadaveric whole knee specimens were obtained from individuals 51 to 80 years of age with no knee pathology other than osteoarthritis. Ten knees were collected from five cadavers within 10 hours of death and underwent a 3-Tesla MR exam including a coronal-oblique 3-dimensional fast imaging with steady state precession (3D FISP) sequence within 36 hours of death. The specimens were placed in a 4% paraformaldehyde in phosphate buffer within 58 hours of death. After preservation, a subchondral region from the tibial plateau was collected and underwent microCT imaging with a voxel size of 9 mu m x 9 mu m x 9 mu m. A single reader analyzed the microCT images in a similar volume of interest as selected in the MR measures. A different reader analyzed the MR-based trabecular morphometry using a custom analysis tool. To analyze the MR-based trabecular morphometry, a rectangular region of interest (ROI) was positioned on the 20 central images in the proximal medial tibial subchondral bone. The primary outcome measures were MR-based and microCT-based trabecular BV/TV in the proximal medial tibia. Results: The MR-based apparent BV/TV was strongly correlated with microCT-based BV/TV (r = 0.83, confidence interval = 0.42 to 0.96), despite the MR-based apparent BV/TV being systematically lower than measured using microCT. Conclusions: MR-based apparent BV/TV in the proximal peri-articular medial tibia has good construct validity and may represent an alternative for CT-based BV/TV.