PURPOSE: Evaluation of TSE- and T 1 -3D-GRE-sequences for focal cartilage lesions in vitro in comparison to ultrahigh resolution multi-slice CT. MATERIALS AND METHODS: Forty artificial cartilage lesions in ten bovine patellae were immersed in a solution of iodinated contrast medium and assessed with ultrahigh resolution multi-slice CT. Fat-suppressed TSE images with intermediate- and T 2 -weighting at a slice thickness of 2, 3 and 4 mm as well as fat-suppressed T 1 -weighted 3D-FLASH images with an effective slice thickness of 1, 2 and 3 mm were acquired at 1.5 T. After adding Gd-DTPA to the saline solution containing the patellae, the T 1 -weighted 3D-FLASH imaging was repeated. RESULTS: All cartilage lesions were visualised and graded with ultrahigh resolution multi-slice CT. The TSE images had a higher sensitivity and a higher inter- and intraobserver kappa compared to the FLASH-sequences (TSE: 70 - 95 %; 0.82 - 0.83; 0.85 - 0.9; FLASH: 57.5 - 85 %; 0.53 - 0.72; 0.73 - 0.82, respectively). An increase in slice thickness decreased the sensitivity, whereby deep lesions were even reliably depicted on TSE images at a slice thickness of 3 and 4 mm. Adding Gd-DTPA to the saline solution increased the sensitivity by 10 % with no detectable advantage over the T 2 -weighted TSE images. CONCLUSION: TSE sequences and application of Gd-DTPA seemed to be superior to T 1 -weighted 3D-FLASH sequences without Gd-DTPA in the detection of focal cartilage lesions. The ultrahigh resolution multi-slice CT can serve as in vitro reference standard for focal cartilage lesions.
