Articular cartilage: in vivo diffusion-tensor imaging

To investigate technical feasibility, test-retest reproducibility, and the ability to differentiate healthy subjects from subjects with osteoarthritis (OA) with diffusion-tensor (DT) imaging parameters and T2 relaxation time.

This study was approved by the institutional review board and was HIPAA compliant. All subjects provided written informed consent. DT imaging parameters and T2 (resolution=0.6×0.6×2 mm) of patellar cartilage were measured at 7.0 T in 16 healthy volunteers and 10 patients with OA with subtle inhomogeneous signal intensity but no signs of cartilage erosion at clinical magnetic resonance (MR) imaging. Ten volunteers were imaged twice to determine test-retest reproducibility. After cartilage segmentation, maps of mean apparent diffusion coefficient (ADC), fractional anisotropy (FA), and T2 relaxation time were calculated. Differences for ADC, FA, and T2 between the healthy and OA populations were assessed with nonparametric tests. The ability of each MR imaging parameter to help discriminate healthy subjects from subjects with OA was assessed by using receiver operating characteristic curve analysis.

Test-retest reproducibility was better than 10% for mean ADC (8.1%), FA (9.7%), and T2 (5.9%). Mean ADC and FA differed significantly (P<.01) between the OA and healthy populations, but T2 did not. For ADC, the optimal threshold to differentiate both populations was 1.2×10(-3) mm2/sec, achieving specificity of 1.0 (16 of 16) and sensitivity of 0.80 (eight of 10). For FA, the optimal threshold was 0.25, yielding specificity of 0.88 (14 of 16) and sensitivity of 0.80 (eight of 10). T2 showed poor differentiation between groups (optimal threshold=22.9 msec, specificity=0.69 [11 of 16], sensitivity=0.60 [six of 10]).

In vivo DT imaging of patellar cartilage is feasible, has good test-retest reproducibility, and may be accurate in discriminating healthy subjects from subjects with OA. ADC and FA are two promising biomarkers for early OA.