Introduction

Diffusion tensor imaging (DTI) plays an important role in neurosurgical planning of brain tumor resection by visualizing the intricate structural network of white matter fibers in the surrounding brain tissue. However, limitations of standard DTI (s-DTI) using single shot EPI include its long scanning time, limited spatial resolution, and limited coverage. Simultaneous multi-slice excitation has been proposed to accelerate image acquisition times and increase slice coverage.1,2 Multi-band DTI (mb-DTI) is a new technique that utilizes simultaneous multi-slice excitation, but assessment of its clinical feasibility remains limited to a few studies involving healthy controls and a small number of clinical cases.3,4 Further research involving mb-DTI in patient populations remains critical in assessing its potential advantages in terms of scanning time and image quality for presurgical planning. The purpose of this study is to qualitatively and quantitatively compare the tractography results from s-DTI and mb-DTI in patients with brain tumor for the arcuate fasciculus (AF) and corticospinal tract (CST), which are major white tracts for language and motor function, respectively.

Methods

Forty-two patients with brain tumor (mean age=56±14(SD), male/female=18/24) who underwent a presurgical planning MRI protocol involving both whole brain s-DTI and mb-DTI in the same session were analyzed. All MR images were acquired using a GE 3T (750w) clinical scanner and a GE 24 channel neurovascular head coil. The imaging parameters for s-DTI were: TR/TE=11000/74.8 ms; slice thickness=3 mm; number of slices=48; field of view (FOV)=240×240 mm2; matrix size=128×128; b=1000 s/mm2; 25 directions. The imaging parameters for mb-DTI were selected to be similar to the s-DTI: TR/TE = 6500/75.7 ms; number of multi-bands=2; slice thickness=3 mm; number of slices=60; FOV=240×240 mm2, matrix size=128×128; b=1000 s/mm2; 25 directions. The acquisition times were 5 min 25 sec for s-DTI and 3 min 2 sec for mb-DTI. In addition, 3-dimensional T1-weighted images were acquired for anatomical information. The distortion caused by eddy currents was corrected using FSL. Data was corrected for image distortion induced by local field inhomogeneity in sinuses near the orbital frontal cortex and area near the ear canals. DSIstudio was used for tensor analysis and tractography analysis. Tract-specific analysis to determine the fractional anisotropy (FA), tract volume, and tract length was applied into the AF and CST. Seed ROIs were set at the superior temporal gyrus and inferior frontal gyrus for the AF and at the pons and foot motor area for the CST tract. Two-tailed, paired t-tests were used to assess the differences in parameters between s-DTI and mb-DTI.

Results

Scan time was reduced by approximately 40% when using mb-DTI instead of s-DTI. Moreover, mb-DTI was able to generate images with very similar appearance to s-DTI for whole-brain FA maps (Fig 1), whole-brain tractography (Fig 2), AF tractography (Fig 3), and CST tractography (Fig 4), which was qualitatively validated by a neuroradiologist. When comparing the quantitative diffusion parameters of fiber length, fiber volume, and FA between mb-DTI and s-DTI in the whole-brain, AF, and CST, these parameters were slightly higher for mb-DTI than in s-DTI for nearly all subjects. Moreover, these small differences at the individual level resulted in a statistically significant difference in FA values between mb-DTI and s-DTI for the whole-brain (mb-DTI/s-DTI=0.48±0.02/0.47±0.02), AF (0.55±0.04/0.53±0.04), and CST (0.60±0.03/0.58±0.03) as well as in fiber length (72±14/58±13) and fiber volume (2760±1280/2080±970) in the AF (p<0.0001 for all).

Discussion

Our data shows that s-DTI using a single shot EPI and mb-DTI using simultaneous multi-slice acquisition with 2 multi-bands produced very similar results for the fiber tractography of the AF and CST, two white matter tracts of critical importance during presurgical planning. These results suggest that mb-DTI would be beneficial for clinical implementation because of its considerable scan time reduction compared to s-DTI. Moreover, the significant increase in quantitative parameters in mb-DTI may be the result of higher signal-to-noise ratio compared to s-DTI, which may suggest that mb-DTI can allow for more accurate and sensitive presurgical planning for white matter tracts compared to s-DTI.

Conclusion

As a result, this study demonstrated the clinical feasibility of utilizing mb-DTI for presurgical planning in patients with brain tumor. The reduced scan time of mb-DTI along with its similar image quality and quantitative diffusion parameters to s-DTI for analyzing two white matter tracts particularly important during presurgical planning suggest clinical advantages for utilizing mb-DTI over current techniques.

Acknowledgements

No acknowledgement found.

References

1) Maudsley AA. Multiple‐line‐scanning spin‐density imaging. J Magn Reson 1980; 41:112–126. 2) Müller S. Multifrequency Selective RF Pulses for Multislice MR Imaging. Magn Reson Med 1988; 6:364–371. 3) Mitsuda et al., Feasibility of Diffusion Tensor Imaging at 1.5T Using Multi-Band Echo Planar Acquisition. Magn Reson Med Sci. 2017; 16(2): 169–175. 4) Wen Q et al., Clinically feasible NODDI characterization of glioma using multiband EPI at 7 T. Neuroimage Clin. 2015 Sep 3;9:291-9