Comparison of HASTE and segmented-HASTE sequences with a T2-weighted fast spin-echo sequence in the screening evaluation of the brain.

OBJECTIVE The purpose of this study was to evaluate the neuroradiologic application of half-Fourier acquisition single-shot turbo spin-echo (HASTE) and segmented-HASTE (s-HASTE) sequences in comparison with a T2-weighted fast spin-echo sequence. MATERIALS AND METHODS First, HASTE, s-HASTE, and fast spin-echo sequences were evaluated for blurring artifacts with a stationary phantom and for motion artifacts with a moving phantom, which repeated constant or intermittent to-and-fro motions at variable intervals. Second, 30 consecutive patients with various intracranial diseases were prospectively examined with the three sequences. Lesions were classified into four groups according to size and signal intensity on fast spin-echo MR images as follows: large hyperintense, small hyperintense, small markedly hyperintense, and hypointense lesions. Signal intensities of the lesion, putamen, and gray matter were compared with the signal intensity of white matter, and contrast-to-noise ratios were calculated. Overall image quality, conspicuity of lesions, delineation of the junction between gray matter and white matter, conspicuity of the putamen, and certain types of artifacts were evaluated qualitatively. RESULTS In the phantom study, the HASTE sequence was least affected by motion artifacts and the fast spin-echo sequence was most affected although the images of the HASTE sequence were most degraded by blurring artifacts. In the clinical study, we found no significant differences among the three sequences for contrast-to-noise ratios or conspicuity of large hyperintense and small markedly hyperintense lesions. However, the contrast-to-noise ratios of hypointense lesions and gray matter, and the conspicuity of hypointense lesions were significantly poorer for the HASTE sequence than for the fast spin-echo sequence. The contrast-to-noise ratios of small hyperintense lesions and the putamen, conspicuity of small hyperintense lesions and putamen, and delineation of the junction between gray matter and white matter were significantly poorer for HASTE and s-HASTE sequences than for the fast spin-echo sequence. Ghost artifacts, which were observed during the s-HASTE sequence, were sometimes superimposed on the image. CONCLUSION The HASTE and s-HASTE sequences afford substantial time reduction and also decrease motion artifacts and thus have potential advantages for neuroradiologic application, especially in uncooperative or unsedated children. The s-HASTE sequence may be preferable to the HASTE sequence because of fewer blurring artifacts and higher T2 contrast. However, small hyperintense and hypointense lesions may be overlooked when HASTE and s-HASTE sequences are used.