Introduction

High resolution susceptibility map-weighted images (SMWI) has recently been developed to improve visualization of the nigrosome-1 region of the substantia nigra (SN) on 3T imaging.¹ SMWI utilizes susceptibility weighting mask derived from quantitative susceptibility mapping (QSM) to enhance the contrast for the SWI magnitude images. In this study, we explore the utility of SMWI to enhance visual identification of the subthalamic nucleus (STN) at 3T, which may be useful to aid in neuro-navigational procedures requiring precise and accurate identification of the STN such as deep brain stimulation (DBS) surgery.

Methods

This feasibility study was performed on three patients with Parkinson’s Disease (PD). Written consent was obtained from all subjects. MRI was performed on a 3T (Siemens Skyra) scanner.
A 3D high resolution T2* susceptibility weighted imaging was performed using a multi-echo GRE sequence with TR = 48 ms, TE = 13.77, 26.39 and 39 ms, FA = 20°, in-plane resolution = 0.5 x 0.5 x 1 mm3, number of slices = 44, scan duration = 5:42 sec. A coronal imaging plane oriented perpendicular to the anterior-posterior bi-commissural line was chosen for acquisition. For co-registration, a 3D T1-weighted MPRAGE sequence was also performed with the following parameters: TR/TE=1900/2.3ms, TI=925ms, FA=9°, in-plane resolution = 0.9 x 0.9 x 1 mm3, number of slices = 208, scan time = 4:24 sec.
QSM images were reconstructed from the multi-echo GRE images with STI Suite (University of Berkeley, California, USA).² The magnitude images of the three echoes were combined into a single image by root sum of squares. Phase images of different TEs were unwrapped using a Laplacian unwrapping algorithm with a combined frequency calculated in each voxel. From the frequency images, the background field was removed using the harmonic background phase removal using the Laplacian operator (HARPERELLA) method.QSM was reconstructed using the improved sparse linear equation and least-squares (iLSQR) method. SMWI images were generated from the combined magnitude image and QSM image with SMWI software written in MATLAB (Laboratory for Imaging Science & Technology, Seoul National University, Seoul, South Korea).¹

Results

QSM and SMWI images showed clear distinction between the STN and SN (Figure 1). SMWI image of subject #2 showed enhanced visualization of the STN as compared to the QSM image (Figure 1, middle row).

Dicussion

Nam et al has showed that SMWI improved the contrast-to-noise ratio between region with high iron content such as SN with its surrounding anatomical structures.¹ SMWI was originally developed to improve the visualization of nigrosome-1 inside the SN but here we showed that it could also be used to enhance differentiation between SN and STN. Precise localization of the STN for DBS is important as it impacts treatment outcomes. However, the STN is small, deeply situated and difficult to be reliably or reproducibly imaged on routine clinical 1.5T and 3T scanners. Traditionally, STN targeting requires additional information from brain atlas coordinates. Subject #3 underwent DBS two weeks after MRI brain scan following traditional neurosurgical approaches, supplemented by additional pre-DBS SMWI imaging (Figure 2) with patient consent. Patient was well after the DBS procedure with good motor outcome. Figure 2 shows the final position of the DBS electrodes very closely situated to localization of the STN on the pre-DBS SMWI image.

Conclusion

SMWI may offer useful supplementary visualization aid to identify the STN in surgical neuro-navigation. Further study is needed to evaluate the full potential of SMWI in aiding safe, accurate and faster DBS surgical planning.

Acknowledgements

No acknowledgement found.

References

1. Nam YH, Gho SM, Kim DH, Kim EY, Lee JH. Imaging of nigrosome-1 in substantia nigra at 3T using multi-echo susceptibility map-weighted imaging (SMWI). J Magn Reson Imaging. 2017 Aug;46(2):528-536. 2. Li W, Wu B, Liu C. STI Suite: a Software Package for Quantitative Susceptibility Imaging. ISMRM 2014 Proc. #3265.