Multimodal Imaging to Support Vim Targeting in Magnetic Resonance-Guided Focused Ultrasound Thalamotomy: Two Technically and Anatomically Challenging Cases

Mohammad Ali AKBAR; Tomoko HANADA; Kazumichi YAMADA; Takuichiro HIGASHI; Tsuneo MAKIUCHI; Ryosuke HANAYA

doi:10.2176/jns-nmc.2025-0390

Abstract

Essential tremor is the most common adult movement disorder, typically presenting with kinetic and postural tremor that interferes with daily activities. For patients unresponsive to medications, surgical options such as radiofrequency thalamotomy, deep brain stimulation, and magnetic resonance-guided focused ultrasound can provide therapeutic benefit.

Magnetic resonance-guided focused ultrasound, an incisionless lesioning modality, has gained clinical use; however, accurate targeting of the ventral intermediate nucleus remains challenging because the nucleus cannot be directly visualized on routine magnetic resonance imaging and exhibits substantial individual variability.

This report describes 2 cases of medication-refractory essential tremor treated with magnetic resonance-guided focused ultrasound ventral intermediate nucleus thalamotomy under structurally complex conditions. One patient had a thick skull with a low skull density ratio, and the other exhibited marked thalamic distortion. In both cases, individualized ventral intermediate nucleus targeting was performed using multimodal imaging that combined a stereotactic planning platform, dentatorubrothalamic tract tractography, and Fast Gray Matter Acquisition T1 Inversion Recovery. The Fast Gray Matter Acquisition T1 Inversion Recovery sequence, which enhances gray-white matter contrast, provided relative contrast of intrathalamic structures, including signal patterns corresponding to the internal medullary lamina. When integrated with connectivity-based dentatorubrothalamic tract tractography, this approach provided complementary information to support anatomically guided targeting.

Stepwise sonication with intraoperative thermal monitoring resulted in substantial tremor reduction without new neurological deficits. These observations suggest that combining Fast Gray Matter Acquisition T1 Inversion Recovery with tractography offers practical, complementary guidance for ventral intermediate nucleus targeting, including in settings without access to advanced commercial integration software. Although limited to 2 patients, this work suggests the value of accessible multimodal imaging for improving confidence in anatomy-informed targeting in magnetic resonance-guided focused ultrasound and potentially other lesioning procedures.

Introduction

Essential tremor (ET) is among the most common movement disorders and may require surgical intervention when pharmacologic therapy fails.¹^,²⁾ Magnetic resonance-guided focused ultrasound (MRgFUS) has emerged as a noninvasive treatment option that complements conventional approaches such as radiofrequency thalamotomy and deep brain stimulation (DBS). However, precise targeting of the ventral intermediate nucleus (Vim) remains challenging because the nucleus is not directly visualized on standard magnetic resonance imaging (MRI) and shows substantial interindividual variability.³⁾

Commercial stereotactic planning platforms are widely used to assist Vim localization by integrating structural imaging, tractography, and atlas information. Although these systems have improved stereotactic precision, they are not available in all clinical settings. Recent imaging advances-such as Fast Gray Matter Acquisition T1 Inversion Recovery (FGATIR) sequences that enhance gray-white matter contrast⁴^,⁵⁾ and dentatorubrothalamic tract (DRT) tractography that delineates cerebello-thalamo-cortical connectivity⁶^,⁷⁾-offer accessible imaging markers that support anatomy-informed targeting. These modalities may provide additional complementary information when advanced integration systems are unavailable. In this report, we present 2 anatomically complex cases of medication-refractory ET treated with MRgFUS Vim thalamotomy. We demonstrate that FGATIR and tractography can support individualized targeting decisions, including in settings without commercial integration software. Rather than proposing a novel imaging technique, this report aims to illustrate the practical integration of established imaging modalities to support anatomy-informed targeting in technically and anatomically challenging MRgFUS cases.

Case Presentation

Case 1

A 50-year-old male chef had experienced tremor since childhood, with gradual worsening over the past few years. His symptoms were refractory to medications such as arotinolol and clonazepam. Although radiofrequency ablation was a potential option, he strongly preferred MRgFUS to avoid burr-hole surgery. Preoperative evaluation revealed a skull density ratio of 0.39 and a markedly thick calvarium, raising concern regarding limited thermal accumulation.

Preoperative targeting was performed using a stereotactic planning platform that enabled multimodal image fusion and atlas-based anatomical mapping, along with integrated DRT tractography (Figure 1A-C). This allowed FGATIR-derived intrinsic anatomy to be interpreted in parallel with atlas-defined thalamic landmarks. On FGATIR, a hypointense band consistent with the DRT was visible, whereas Vim borders remained indistinct. Therefore, Vim localization was estimated by integrating FGATIR-based thalamic architecture, particularly features corresponding to the internal medullary lamina. The spatial relationship provided by DRT tractography offered additional complementary information, while a more posterior hypointense area was presumed to represent the ventral caudal nucleus (Vc) (Figure 2).

Figure 1

Case 1-Multiplanar MR images with coordinate adjustment.

(A-C) FGATIR images fused with DRT tractography using an image-fusion-based stereotactic planning system. The Vim (pink), Vc (green), and red nucleus (red) are shown. The asterisk marks the planned target.

(D-F) Postoperative T2-weighted MRI fused with DRT tractography shows a well-defined lesion (3.7 × 3.6 × 3.7 mm) corresponding to the planned target.

A thin dotted line represents the anterior commissure-posterior commissure (AC-PC) line (also shown in Figures 2 and 3).

AC: anterior commissure; DRT: dentatorubrothalamic tract; FGATIR: Fast Gray Matter Acquisition T1 Inversion Recovery; MR: magnetic resonance; MRI: magnetic resonance imaging; PC: posterior commissure; Vc: ventral caudal nucleus; Vim: ventral intermediate nucleus

Figure 2

FGATIR-based visualization of intrathalamic architecture.

(A-C) Multiplanar FGATIR images demonstrating the thalamus and adjacent structures. The lenticular nucleus border is outlined by a white dotted line, and blue boxes indicate areas of interest.

(D-F) Magnified views of the boxed regions

(G-I) Anatomical parcellation of thalamic nuclei and adjacent structures, based on intrathalamic laminae (white dashed lines).

FGATIR: Fast Gray Matter Acquisition T1 Inversion Recovery

The final target was positioned near the inferior margin of the presumed Vim. During sonication, initial thermal rise was slow, and careful parameter adjustment was required to achieve an effective lesion. The postoperative lesion corresponded well to the planned target (Figure 1D-F). Final coordinates were 7.0 mm anterior to the posterior commissure (PC), 13.5 mm lateral to the midline, and 1.5 mm superior to the anterior-posterior commissural (AC-PC) plane (AC-PC length = 23.9 mm). The mean maximum temperature achieved was 54°C. Further technical details are provided in Supplementary Table S1. The Clinical Rating Scale for Tremor (CRST) score improved from 37 (A: 20, B: 14, C: 3) preoperatively to 14 (A: 6, B: 7, C: 1) postoperatively. No complications occurred, and tremor suppression remained stable at the 6-month follow-up, with high patient satisfaction.

Case 2

A man in his 60s had experienced right-predominant tremor for over a decade, and he had a maternal family history of tremor. Arotinolol and clonazepam provided no benefit. The skull density ratio was 0.46, and MRI demonstrated marked thalamic distortion with asymmetric posterior displacement of intrathalamic structures.

Preoperative targeting was performed using the same stereotactic planning platform, which enabled multimodal image fusion and atlas-based anatomical mapping, together with integrated DRT tractography (Figure 3A-C). FGATIR imaging provided additional structural contrast that assisted in the interpretation of intrathalamic anatomy. A posteriorly displaced hypointense band corresponding to the DRT suggested posterior shifting of the Vim region, while another posterior band was interpreted as likely representing the Vc. Based on this concordant anatomical information, the target was adjusted posteriorly, and thermal energy was increased gradually to avoid sensory side effects.

Figure 3

Case 2-Multiplanar MR images with coordinate adjustment.

(A-C) FGATIR images fused with DRT tractography and stereotactic overlays, demonstrating posterior displacement of the Vim. The Vim (pink), Vc (green), and red nucleus (red) are shown. The asterisk marks the planned target.

(D-F) Postoperative T2-weighted MRI fused with DRT tractography demonstrates a focal lesion (3.3 × 3.3 × 3.2 mm) corresponding to the target.

DRT: dentatorubrothalamic tract; FGATIR: Fast Gray Matter Acquisition T1 Inversion Recovery; MR: magnetic resonance; Vc: ventral caudal nucleus

The postoperative lesion corresponded well to the planned target (Figure 3D-F). Final coordinates were 2.7 mm anterior to the PC, 14.0 mm lateral to the midline, and 1.0 mm superior to the AC-PC plane (AC-PC length = 21.4 mm). The mean maximum temperature reached was 56°C. Further technical details are provided in Supplementary Table S1. CRST improved from 56 (A: 32, B: 21, C: 3) to 15 (A: 7, B: 7, C: 1). No adverse events occurred. At 6 months, a mild goal-directed tremor re-emerged, but overall function and patient satisfaction remained improved.

FGATIR images were acquired on a 3 T MRI using parameters optimized for intrinsic white-matter suppression, and DRT tractography was reconstructed using deterministic algorithms, with regions of interest placed at the dentate nucleus, red nucleus, and precentral gyrus. Further technical details are provided in Supplementary Table S2 and Supplementary Note.

Discussion

Precise Vim targeting has long been a challenge in tremor surgery, including MRgFUS thalamotomy, because the nucleus is not directly visible on conventional MRI and shows marked anatomical variability.³^,⁶^,⁷⁾ Traditional coordinate-based or indirect landmark methods also have limitations, particularly in anatomically complex cases in which individualized imaging guidance is required.⁸^,⁹⁾

FGATIR provides high gray-white matter contrast and has been explored for thalamic parcellation in DBS planning. Although prior reports describe improved visualization of structures within the ventrolateral thalamus, direct delineation of the Vim remains limited.⁴^,⁵^,¹⁰⁾ In parallel, other high-contrast sequences, such as white-matter-nulled Magnetization-Prepared Rapid Gradient-Echo, have also been explored for Vim visualization in MRgFUS.¹¹⁾ Although our present cases used FGATIR rather than Magnetization-Prepared Rapid Gradient-Echo, both techniques enhance thalamic contrast and complement tractography for patient-specific targeting.

In our practice, targeting is typically performed using multimodal image fusion combined with deterministic tractography and atlas information, often supported by commercial planning software. These systems provide practical and reliable guidance in most cases; however, their academic basis for defining individual Vim borders is inherently limited by the lack of direct nuclear visualization. In the present cases, FGATIR was incorporated as an additional structural reference to address this gap, particularly under anatomically challenging conditions. We also aimed to explore whether this multimodal approach could serve as complementary guidance in settings in which such commercial platforms are not available.

On FGATIR, a hypointense band corresponding to the DRT identified on tractography was observed just beneath the presumed Vim. Although the Vim itself was not sharply delineated, its inferior margin could be inferred from the overall thalamic contrast pattern (Figure 2).

Within the Vim, the DRT-related signal showed gradual attenuation, a finding that may indirectly relate to the mixed neuronal composition and multidirectional fiber dispersion characteristic of this region. Histological analyses by Hirai et al.¹²⁾ demonstrated that, although the Vim is often described as relatively "cell-sparse," it contains numerous large and medium-sized relay neurons with complex dendritic arborization. These cytoarchitectural features, together with dense crossing fibers, likely contribute to reduced FGATIR contrast compared with adjacent nuclei such as Vc. The final target in each case was determined by integrating FGATIR findings with tractographic and atlas-based information. The combined use of FGATIR and tractography appeared to enhance confidence in anatomy-informed targeting, particularly in patients with atypical thalamic configuration. Although limited to 2 cases, these observations suggest that accessible multimodal imaging may complement existing planning systems and broaden the applicability of imaging-assisted, anatomy-informed targeting. Further validation in larger cohorts is warranted.

Conclusion

These cases illustrate that integrating FGATIR with tractography and multimodal image fusion can support individualized Vim targeting during MRgFUS thalamotomy, even under anatomically complex conditions. Although FGATIR alone does not sharply define nuclear borders, its combined use with tractography and atlas-based anatomical mapping provided complementary structural information that improved targeting confidence.

Our experience suggests that such multimodal imaging may offer practical and reproducible complementary guidance for Vim targeting, including in settings where commercial integration platforms are not available, although in our cases these platforms were used routinely. Taken together, these observations indicate that accessible multimodal imaging can serve as a useful adjunct to existing targeting strategies.

Acknowledgments

We thank the radiology technologists and clinical staff for their assistance with MRI acquisition and perioperative care.

Disclaimer

Author Ryosuke Hanaya is one of the Editorial Board members of the Journal. This author was not involved in the peer-review or decision-making process for this paper.

Conflicts of Interest Disclosure

All authors have no conflict of interest.

Ethical Compliance

This report is based on 2 cases extracted from the "Epidemiological Study-Rev. 1 (amended as Rev. 2) " cohort, which had been approved by the institutional review board of our institution. All procedures involving human participants were conducted in accordance with the principles of the Declaration of Helsinki and its later revisions, as well as the "Ethical Guidelines for Medical and Health Research Involving Human Subjects (Provisional Translation as of March 2015) " and subsequent amendments. Written informed consent for publication of this case report and accompanying images was individually obtained from both patients.

References

責任著者(Corresponding author)

J-STAGEへの登録はこちら（無料）