T 2 relaxometry measurements in low spatial frequency brain regions differ between fast spin-echo and multiple-echo spin-echo sequences

Abstract

Dual-echo fast spin-echo (FSE) sequences are used in T2 relaxometry studies of neurological disorders because of shorter clinical scanning times and protocol simplicity. However, FSE sequences have possible spatial frequency-dependent effects, and derived T2 values may include errors that depend on the spatial frequency characteristics of the brain region of interest. Dual-echo FSE and multi-echo spin-echo (MESE) sequences were acquired in nine subjects. The T2 decay curves for FSE and MESE sequences were estimated and percent error maps were generated. T2 error values were obtained along each patient's corticospinal tract (CST). Whole-brain white matter (WM) and gray matter (GM) T2 error values were also obtained. The paired t test was performed to evaluate differences in T2 values in the CST between FSE and MESE sequences. Histograms of error values in CST and in whole-brain WM and GM structures revealed systematic errors in FSE sequences. Significant differences (P < 0.001) in CST T2 values were also observed between FSE and MESE sequences. Our findings indicate that T2 values derived from FSE sequences are prone to large errors, even in low spatial frequency regions such as the CST, when compared to MESE sequences. Future studies should be aware of this limitation of FSE sequences.