The objective of this study was to compare articular cartilage thickness observed in the different 2D and 3D sequences to the cartilage thickness of the equine stifle in cadavers to determine the accuracy of each sequence. The study was conducted as a blinded laboratory study using seven equine stifle specimens. The 2D (T2W TSE) and 3D (3D VIEW T2W HR, T2 3D mFFE, T1W VISTA SPAIR, 3D PDW SPAIR) 3-tesla MRI sequences of each stifle were obtained. Cartilage thickness was measured at 30 locations on MRI and on gross pathology. Thickness measurements were compared using a Bland-Altman plot and post hoc analysis tests. The 3D sequences were found to be generally more accurate than the 2D sequence (p < 0.001). The smallest difference to macroscopic measurements was observed in the 3D VIEW T2W HR and T1W VISTA SPAIR sequences with no statistical difference between each other. Knowing the accuracy of different sequences will improve the evaluation of equine cartilage and the early detection of cartilage pathologies. This would promote MRI as a noninvasive imaging modality for horses suffering from stifle lameness with no findings in conventional imaging methods. Furthermore, since 3D sequences seem to have better accuracy in depicting cartilage, they may replace 2D sequences, thereby shortening scanning times.
Keywords: 2D sequences; 3D sequences; MRI; articular cartilage; horse; stifle joint.