IMAGING ARTICULAR-CARTILAGE DEFECTS WITH 3D FAT-SUPPRESSED ECHO-PLANAR IMAGING - COMPARISON WITH CONVENTIONAL 3D FAT-SUPPRESSED GRADIENT-ECHO SEQUENCE AND CORRELATION WITH HISTOLOGY

Purpose: Our goal was to shorten examination time in articular cartila ge imaging by use of a recently developed 3D multishot echo planar ima ging (EPI) sequence with fat suppression (FS). We performed comparison s with 3D FS GE sequence using histology as the standard of reference. Method: Twenty patients with severe gonarthrosis who were scheduled f or total knee replacement underwent MRI prior to surgery. Hyaline cart ilage was imaged with a 3D FS EPI and a 3D FS GE sequence. Signal inte nsities of articular structures were measured, and contrast-to-noise ( C/N) ratios were calculated. Each knee was subdivided into 10 cartilag e surfaces. From a total of 188 (3D EPI sequence) and 198 (3D GE seque nce) cartilage surfaces, 73 and 79 histologic specimens could be obtai ned and analyzed. MR grading of cartilage lesions on both sequences wa s based on a five grade classification scheme and compared with histol ogic grading. Results: The 3D FS EPI sequence provided a high C/N rati o between cartilage and subchondral bone similar to that of the 3D FS GE sequence. The C/N ratio between cartilage and effusion was signific antly lower on the 3D EPI sequence due to higher signal intensity of f luid. MR grading of cartilage abnormalities using 3D FS EPI and 3D GE sequence correlated well with histologic grading. 3D FS EPI sequence a greed within one grade in 69 of 73 (94.5%) histologically proven carti lage lesions and 3D FS GE sequence agreed within one grade in 76 of 79 (96.2%) lesions. The gradings were identical in 38 of 73 (52.1%) and in 46 of 79 (58.3%) cases, respectively. The difference between the se nsitivities was statistically not significant. Conclusion: The 3D FS E PI sequence is comparable with the 3D FS GE sequence in the noninvasiv e evaluation of advanced cartilage abnormalities but reduces scan time by a factor of 4.