[Fast spin echo imaging of vertebral metastasis: comparison of fat suppression techniques (FSE-CHESS, STIR-FSE)]

To investigate the capabilities of fast spin echo (FSE) sequences in diagnosing spinal metastases and to compare two fat-suppression techniques: CHESS (chemical shift selective saturation) and STIR (short T1 inversion recovery). Fat suppression is recommended with FSE sequences because on them, different from conventional spin echo (SE) sequences, fat has high signal intensity in both T1 and T2 weighting, masking such high-signal bone lesions as metastases.

Ninety metastatic lesions in 32 patients were studied with T1-weighted SE and T2-weighted FSE sequences with and without fat suppression (FSE-CHESS and STIR-FSE). Quantitative analyses (metastasis size, signal intensity, margins and conspicuity) and qualitative analyses (artifacts and fat saturation homogeneity) were carried out of both fat-suppression sequences. Signal-to-noise (S/N) and contrast-to-noise (C/N) ratios were calculated in 20 metastatic lesions. The results were analyzed with the chi 2 method and the paired t-test.

Both fat-suppression FSE sequences were more sensitive to focal lesions (100%) than T1-weighted SE sequences (96.6%). Lesion signal was higher than that of surrounding bone in 95.5% of fat-suppressed images, which facilitated the recognition of spinal metastases. On the contrary, conspicuity was higher (73.3%) and the margins more definite (68.8%) on T1-weighted SE than on T2-weighted FSE sequences. Fat suppression was more apparent on STIR-FSE than on FSE-CHESS sequences, even with no significant difference in C/N ratio.

Both STIR-FSE and FSE-CHESS sequences are rapid and useful techniques to obtain fat saturation in FSE images. CHESS saturation is selective on fat signal, while STIR suppresses the signal of all the substances with the same short T1 as fat.