Comparison of STIR and spin-echo MR imaging at 1.5 T in 90 lesions of the chest, liver, and pelvis.

Abstract

Short TI inversion recovery (STIR) produces both fat suppression and the additive effect of T1 and T2 mechanisms on tissue brightening, in contrast to the subtractive effect of these two mechanisms on spin-echo sequences. In order to compare STIR and spin-echo imaging, we reviewed 90 lesions detected in 76 consecutive MR studies of the chest, liver, or pelvis performed at 1.5 T with both STIR and double-echo spin-echo techniques. Images were compared for the number of individual lesions detected. Lesion conspicuity was scored by using a subjective scale for each sequence. Lesion size was measured with hand-held calipers, and volume was calculated assuming a prolate ellipse. Because of inherent error in such calculations, lesions were judged to be similar in size (within 20%) or dissimilar (more than 20% difference). The presence of a lesion was proved by direct biopsy in 36 (40%), by tissue pathology from some other focus plus follow-up of the lesion in 37 (41%), or by other imaging plus follow-up in 12 (13%). STIR images detected five (6%) more lesions than spin echo and did not miss any of the lesions detected by spin echo. Conspicuity was greater on STIR images than on spin-echo images in 82 (91%) of the lesions. Twenty-six (29%) of the lesions appeared larger on STIR images than on spin-echo images. For these reasons, STIR may be a useful adjunct to spin echo for body MR in some cases. However, STIR images typically display lower signal-to-noise than spin-echo images do, and all abnormalities (tumor or edema) may appear equally bright on STIR.