SPINAL-CORD LESIONS IN PATIENTS WITH MULTIPLE-SCLEROSIS - COMPARISON OF MR PULSE SEQUENCES

PURPOSE: To compare T2-weighted conventional spin-echo (CSE), fast spi n-echo (FSE), short-tau inversion recovery (STIR) FSE, and fluid-atten uated inversion recovery (FLAIR) FSE sequences in the assessment of ce rvical multiple sclerosis plaques. METHODS: Twenty patients with clini cally confirmed multiple sclerosis and signs of cervical cord involvem ent were examined on a 1.5-7 MR system, Sagittal images of T2-weighted and proton density-weighted CSE sequences, T2-weighted FSE sequences with two different sets of sequence parameters, STIR-FSE sequences, an d FLAIR-FSE sequences were compared by two independent observers. In a ddition, contrast to-noise measurements were obtained. RESULTS: Spinal multiple sclerosis plaques were seen best on STIR-FSE images, which y ielded the highest lesion contrast. Among the T2-weighted sequences, t he FSE technique provided better image quality than did the CSE techni que, but lesion visibility was improved only with a repetition time/ec ho time of 2500/90; parameters of 3000/150 provided poor lesion contra st but the best myelographic effect and overall image quality. CSE ima ges were degraded by prominent image noise; FLAIR-FSE: images showed p oor lesion contrast and strong cerebrospinal fluid pulsation artifacts . CONCLUSIONS: The STIR-FSE sequence is the best choice for assessment of spinal multiple sclerosis plaques. For T2-weighted FSE sequences, shorter echo times are advantageous for spinal cord imaging, long echo times are superior for extramedullary and extradural disease. FLAIR-F SE sequences do not contribute much to spinal imaging for multiple scl erosis detection.