Implementation of 3 T lactate-edited 3D 1H MR spectroscopic imaging with flyback echo-planar readout for gliomas patients

Abstract

The purpose of this study was to implement a new lactate-edited 3D 1H magnetic resonance spectroscopic imaging (MRSI) sequence at 3 T and demonstrate the feasibility of using this sequence for measuring lactate in patients with gliomas. A 3D PRESS MRSI sequence incorporating shortened, high bandwidth 180° pulses, new dual BASING lactate-editing pulses, high bandwidth very selective suppression (VSS) pulses and a flyback echo-planar readout was implemented at 3 T. Over-prescription factor of PRESS voxels was optimized using phantom to minimize chemical shift artifacts. The lactate-edited flyback sequence was compared with lactate-edited MRSI using conventional elliptical k-space sampling in a phantom and volunteers, and then applied to patients with gliomas. The results demonstrated the feasibility of detecting lactate within a short scan time of 9.5 min in both phantoms and patients. Over-prescription of voxels gave less chemical shift artifacts allowing detection of lactate on the majority of the selected volume. The normalized SNR of brain metabolites using the flyback encoding were comparable to the SNR of brain metabolites using conventional phase encoding MRSI. The specialized lactate-edited 3D MRSI sequence was able to detect lactate in brain tumor patients at 3 T. The implementation of this technique means that brain lactate can be evaluated in a routine clinical setting to study its potential as a marker for prognosis and response to therapy.

Figure 1

The new BASING pulse waveform designed for 3 T (a) and its inversion profile (b). Carrier frequency of the second cycle was shifted 198 Hz from the first cycle so that the lactate methine quartet was placed in either the passband or the stopband. As a result, lactate doublet is upright in one cycle and inverted in other cycle, so that summing the two data sets produces uncoupled resonances while subtracting them produces lactate resonance (c)

Figure 2

Lactate-edited 3D PRESS MRSI was applied to study the effect of chemical shift artifacts on lactate signal and other brain metabolites. The mean value of lactate SNR and choline to NAA ratio were calculated from the right, left, middle, anterior, and posterior voxels of the central two slices

Figure 3

The illustration of chemical shift mis-registration artifact for 1.2 and 1.5 over-PRESS factors. The 1.2 over-PRESS factor produces shifted excitation profiles for lactate methyl (dashed line) and methine (dotted line) resonances from the excitation profile of carrier frequency (solid line), resulting in a 73 × 82 mm region with the simultaneous excitation of two lactate resonances (a). Similarly, the 1.5 over-PRESS factor produces a 91 × 102 mm region with the simultaneous excitation of two lactate resonances (b) and therefore smaller chemical shift artifacts compared to the 1.2 over-PRESS factor

Figure 4

The effect of chemical shift artifacts on brain metabolite signal was evaluated using a phantom. Mean lactate SNR (a) and choline to NAA ratio (Cho/NAA) (b) were compared between the right (R), left (L), middle (M), anterior (A), and posterior (P) voxels of the excited volume. As the over-PRESS factor increased from 1.2 to 1.7, the chemical shift artifact lessened

Figure 5

The distribution of lactate methyl doublet from MRSI data using phantom. The top images show the color maps of lactate SNR for the over-PRESS factors of 1.2 (a), 1.5 (b), and 1.7 (c). The corresponding spectra zoomed-in around lactate doublet are shown in the bottom for the over-PRESS factors of 1.2 (d), 1.5 (e), and 1.7 (f). The lactate doublet in the left column (highlight) had lower intensity compared to other regions of the PRESS volume for the over-PRESS factors of 1.2 and 1.5. The overall uniformity of lactate intensity increased with increases in the over-PRESS factor

Figure 6

Lactate spectra from the first (edit-on) and second (edit-off) cycles acquired from phantom using the over-PRESS factor of 1.5. The use of higher-bandwidth 180° refocusing pulses in combination with over-prescription and high bandwidth VSS pulses minimized signal cancellation across the excited volume and provided relatively uniform lactate signal for both data sets

Figure 7

Lactate-edited spectra with a flyback gradient in a phantom (a, b) and a volunteer (c, d). Lactate is detected in the subtracted spectrum of phantom (b), and uncoupled resonances are produced in summed spectra in both phantom (a) and volunteer (c). T1-weighted image from the volunteer (e) shows the PRESS box and the voxel chosen for the spectra

Figure 8

Comparison of MRSI data between the flyback-gradient method (a) and the conventional elliptical MRSI method (c) from a volunteer. T1-weighted SPGR (b) shows PRESS box and the spectral array

Figure 9

An example of lactate-edited spectra using the flyback-gradient method from a patient. The voxels in contrast enhancing lesions (a, b) contain high choline (Cho) as well as lipid in summed spectra (top) and lactate in subtracted spectra (bottom). The normal tissue from contra-lateral hemisphere (c) shows normal display of brain metabolite level (top in c) without lactate (bottom in c). The numbers indicate raw SNR values

Figure 10

An example of lactate-edited spectra using the flyback-gradient method from a patient, showing the full coverage of a slice from the PRESS volume. In the summed spectra (a), the highlighted voxels show abnormal levels of metabolites with highly elevated choline around lesions in T1 post-contrast image (top, b) and T2-weighted FLAIR image (bottom, b). The highlighted voxels in the subtracted spectra (c) show lactate doublet at 1.3 ppm