Partially saturated fluid attenuated inversion recovery (FLAIR) sequences in multiple sclerosis: comparison with fully relaxed FLAIR and conventional spin-echo.

Abstract

Fluid attenuated inversion recovery (FLAIR) sequences produce selective cerebrospinal fluid (CSF) suppression by employing a very long inversion time (TI). We used the FLAIR sequence to study patients with multiple sclerosis (MS) at 0.6 T. So far, a very long TR (and long acquisition time) has been used in a fully relaxed (FR-FLAIR) system. To speed up the FLAIR sequences, we used a shorter TR, and demonstrated that complete CSF suppression can be maintained with partial saturation (PS-FLAIR) by reducing TI at the same time. The introduction of partial saturation, however, reduced the contrast between lesions and normal appearing white matter (NAWM). Suboptimal CSF suppression therefore had to be accepted to maintain sufficient lesion to NAWM contrast. Using a TE of 60 ms, the PS-FLAIR and FR-FLAIR performed equally well in the detection of MS-lesions, although the former provided poorer CSF suppression. Both FLAIR sequences, however, provided poorer constrast between lesions and NAWM compared to conventional spin-echo sequences. Although the long acquisition time of the FLAIR sequence can be reduced by using partial saturation, complete CSF suppression and good lesion to NAWM contrast are incompatible at short TRs. Using a TE of 60 ms, conventional spin-echo sequences detect more lesions and provide better contrast between lesions and NAWM than FLAIR sequences in MS patients.