The ventral intermediate nucleus (VIM) of the thalamus is currently established as the premiere target for magnetic resonance-guided focused ultrasound (FUS) thalamotomy in essential tremor (ET). Over the past decade, the efficacy of FUS VIM ablation has been proven in multiple international studies (1–3) and recent systematic reviews, which demonstrated pooled tremor suppression of 56.7%, 62.4%, and 61.5%, respectively (4–7). Although other tremor-specific targets such as the cerebellothalamic tract (8) or even a combination of targets such as the VIM with the posterior subthalamic area (PSA) (9, 10) have been explored, the VIM alone remains the most frequently used target in FUS treatment for ET (11, 12). Several alternative targets have been considered in Parkinson's disease (PD) (13–16), but the VIM remains the target of choice for FUS treatment of tremor-dominant PD (17). The success of VIM ablation can be readily seen with its global adoption and the growth in the number of FUS centers performing thalamotomy for tremors (18).

The VIM, a motor nucleus within the lateral thalamic subgroup of nuclei, is a proven tremor-sensitive nucleus within the cerebello-thalamo-cortical network (19, 20). In FUS thalamotomy, accurate targeting of the VIM is crucial to ensure adequate tremor suppression while avoiding the erroneous ablation of adjacent structures, risking motor and sensory adverse effects. The VIM is predominantly bordered anteriorly by the ventral oralis posterior (VOP), a motor nucleus in the pallidothalamocortical pathway (21, 22), and posteriorly by the ventralis caudalis (VC), a large sensory nucleus (21, 23). The medial border of the VIM is less well defined, but on the Schaltenbrand–Wahren (S-W) atlas (21), it includes the ventro-oralis internus, the lamella medialis interpolaris, and the nuclei centrales thalami. The VIM's lateral border is the nucleus reticularis, a thin strip of tissue separating the VIM from the internal capsule (IC), which contains the pyramidal tracts. On the S-W atlas, the superior border of VIM is predominantly formed by the nucleus zentrolateralis intermedius. The fasciculus interstitio-thalamicus, the zona incerta (ZI), and the prelemniscal radiation (RAPRL) form the inferior border of the VIM and are included on the coronal plates of the S-W atlas. The ZI and RAPRL are often considered together as the PSA. Although some centers deliberately target PSA, a cautious approach should also be taken to inferior lesioning as there is a risk of chorea (9, 10). Consideration of the posterior and lateral borders is also crucial in VIM targeting to minimize the risk of ablation of key adjacent structures and associated adverse sensory and motor effects.

Unfortunately, current clinical MRI scanners at 1.5 and 3 Tesla (T) cannot delineate the VIM on conventional MRI pulse sequences. Although post-processing techniques have allowed the demarcation of thalamic nuclei subgroups on 3T MRI (20), individual nuclei cannot be determined. Therefore, VIM targeting in FUS traditionally relies on anatomical landmarks to infer the VIM position. The key structures used are demonstrated across all brain cross-sectional imaging modalities and include the third ventricle and internal capsule (IC). The cerebrospinal fluid-filled third ventricle borders the thalamus medially. The IC is a large confluence of white matter tracts that, on CT and MRI, form a distinct lateral border to the thalamus, which itself is a large confluent region of gray matter. The anterior commissure (AC) and posterior commissures (PC) are relatively thin white matter tracts that cross the cerebral hemispheres, which can be visualized on specific MRI sequences, including the Fast Gray Matter Acquisition T1 Inversion Recovery (FGATIR) sequence (24) or the Magnetization Prepared RApid Gradient Echo sequence (MP-RAGE) (25). In the midline, the AC–PC line or intercommissural line (ICL) is an imaginary line joining these two structures and is widely used as an imaging plane and as the baseline for stereotactic neurosurgery. As the inferior border of the VIM lies close to the axial plane projected at the level of the ICL, it can readily be inferred on an MRI.

The traditional approach to VIM targeting in FUS thalamotomy utilizes the ICL to set the superior-inferior (SI) position with pre-determined measurements in anterior-posterior (AP) and medial-lateral (ML) positions along this trajectory. FUS treatment allows the target to be adjusted according to patient response, in sub-millimeter increments, before a permanent ablation is performed. This technique has been well described in the literature (26) and allows the tailoring of treatments to individual neuroanatomy. The most commonly published or “traditional” VIM targeting approach (VIM-TA) utilizes the following trajectories: (AP) 25% of ICL length, anterior to PC; (ML) 14–16 mm lateral; and (SI) on the ICL plane. However, as clinical experience grows, FUS centers naturally adapt their VIM-TA. For example, our centre's approach has evolved over 7 years of practice from the described traditional VIM-TA to (AP) 3–5 mm posterior to MCP; (ML) 3–5 mm medial to IC; and (SI) 2 mm above ICL in 2023. At this site, we achieve better tremor control with minimal adverse effects and are completing treatments in fewer sonications with a shorter procedural time. This learning from experience, or “evolution”, will have occurred at every FUS center; however, there is currently no published data describing the various VIM-TAs used internationally, and an update is vital.

Diffusion tensor imaging (DTI) or “tractography” is an imaging technique that utilizes the anisotropic diffusion properties of water in the white matter tracts to create three-dimensional maps of neural pathways and provide information on directionality. As the VIM lies between several large white matter tracts, the medial lemniscus, the pyramidal tract, and the dentatorubothalamic tract, some FUS centers use tractography to infer the VIM position (27–30). Tractography promises highly individualized VIM-TA; however, it is not yet universally adopted. Of note, tractography does not directly visualize the VIM, but an ultrahigh-field strength MRI at 7T can provide enough contrast between thalamic nuclei to delineate the VIM (31). Current clinical scanners operate at 1.5T and 3T; unfortunately, 7T MRIs are not readily available in healthcare institutions, so direct VIM visualization remains within the research space. DTI and ultrahigh-field MRI offer patient-specific targeting, and future developments in these and other advanced imaging techniques may lead to higher adoption.

Early FUS publications focused on the safety and efficacy of this novel treatment for tremor, with a paucity of data on the technique itself. With safety and efficacy established (1, 2), there has been a notable trend in scientific output from FUS centers with more detailed technical methodology, including closer reporting of their approach to VIM targeting (32, 33). There has also been further enquiry into improving FUS treatments from a technical perspective (34), including specific consideration of skull factors (29, 35–37), thermal dose and lesion size (38–41), and imaging aspects (42–45). However, there has not yet been a review or consensus on the optimum location for FUS VIM ablation. To establish this, an evaluation of VIM-TAs utilized internationally and documentation of the evolution of FUS centres' practice are the natural first step. Furthermore, sharing any such analysis based on clinical experience gained with the optimal FUS technique is vital to ensure improved tremor suppression and minimization of adverse effects for ET patients treated with FUS.

This article aims to ascertain the various international approaches to targeting the VIM (VIM-TA) in magnetic resonance-guided focused ultrasound (FUS) thalamotomy for essential tremor (ET) and consider how targeting has evolved internationally as experience develops.

Between July 2019 and July 2021, all 39 MRgFUS centers from the Insightec Limited (Haifa, Israel) international FUS tremor database were invited to participate in this study and share their VIM targeting approach (VIM-TA). Each FUS center was contacted at least three times via email. Invitations included reassurance that participating FUS centres' contributions would be acknowledged in any subsequent academic output from this study, but individual VIM-TAs would remain anonymous. Where possible, the system operator (neurosurgeon and/or neuroradiologist) was contacted directly. At many centers, correspondence was first conducted through clinical or research administrators before reaching the appropriate clinician. The best efforts were made to ensure the system operator provided the VIM-TA where initial contact was via a third party. Written informed consent from the participants was not required to participate in this study in accordance with national legislation and institutional requirements. Ethical review and approval were not required for the study in accordance with the local legislation and institutional requirements.

Participants were invited to share their VIM-TA for 2019 and 2021 with open correspondence rather than a rigid questionnaire to encourage the sharing of information and discussion.

Where the VIM-TA was anatomical, coordinates were calculated with respect to the mid-commissural point (MCP) to allow 3D modeling and mapping graphically. Coordinates were determined in three planes, namely, anterior-posterior (AP), medial-lateral (ML), and superior-inferior (SI), with the MCP considered coordinate 0 in all three axes. Positive numbers were assigned to anterior, lateral, and superior movements. Negative numbers were assigned for medial, posterior, and inferior movements. This method was chosen to accommodate various ICL lengths. Where centers provided a range (mm) for a specific plane, the mid-point was taken.

Where tractography-based targeting was reported, technical details of the methodology were requested. Where tractography was used to complement anatomical-based targeting, the primary anatomical-targeting method was mapped.

VIM-TAs were mapped graphically and on a 3D model created from the S-W atlas (21). The model was created from Brain LXXVIII, Axial Plates 53–55. Images were stacked in MATLAB (R2021a, MathWorks Inc.) to obtain uniform resolution in three dimensions, and using the 3D slicer (v4.11.2021022), key thalamic nuclei were segmented, including VIM, VOP, and VC, and the model surfaces were smoothed. The model ICL length was scaled to a modern average of 27.8 mm (based on the last 30 MRgFUS patients at our center), and the model coordinates were scaled accordingly prior to mapping.

The results were first analyzed with regard to VIM-TA, whether anatomical and/or using tractography. Further analysis considered VIM-TA coordinates in relation to MCP, both graphically and within key nuclei on the 3D model. Finally, any change, trend, or evolution in VIM-TA across the study period was determined.

In 2022, all participating centers were invited to share further details on their experience with FUS thalamotomy, the rationale for their VIM-TA, and any change in practice. Further analysis of the centres' years of experience, number of treatments performed, and geography was conducted to determine whether any correlation to the trends in VIM-TA is ascertained.

Across the study period, a total of 30 participants from the database of 39 centers participated (Table 1), with a response rate of 76.9%. Complete VIM-TA was reported by 26 centers for 2019 (Appendix 1), 23 centers for 2021 (Appendix 2), and 20 centers provided data for both years. For each year cohort (2019 and 2021), the results are provided as percentages of the number of participating centers in that year. The majority of the centers replied with direct answers describing VIM-TA, but some provided presentation slides, unpublished data summaries, and published papers. Where appropriate, replies were clarified with responders directly before converting VIM-TAs to coordinates for mapping. Two centers were excluded from the coordinate analysis as complete VIM-TAs could not be determined.

The vast majority of FUS centers used anatomical VIM-TAs: 96.2% in 2019 (n = 25) and 95.7% in 2021 (n = 22), with only one center using primarily tractography-based VIM-TA. However, across the study period, more centers incorporated tractography in conjunction with or as an adjunct to their anatomical targeting (Figure 1). In 2019, only 30.7% were utilizing tractography in their clinical practice, and in 2021, this doubled to 60.8% (total groups T1–T3). Participating centers using tractography only for research increased more than four-fold across the study period, from only 3.8% in 2019 to 17.4% in 2021 (Group T-5). Furthermore, the percentage of centers not using tractography in any role decreased from 65.4% to 21.7%. The one center with a tractography-based VIM-TA shared its published papers, which included a well-described methodology (27, 28). As the S-W atlas does not delineate individual white matter tracts, this centre's VIM-TA was not mapped onto the 3D model.

All anatomical VIM-TAs were calculated in relation to the MCP (considering ICL length) prior to analysis. The distribution of anatomical VIM-TA coordinates in the AP, ML, and SI planes was tabulated and graphically demonstrated (Figures 2A–C). All VIM-TAs were mapped onto the axial and sagittal graphs (Figures 3A, B) and the 3D thalamic nucleic model (Figures 4A–C); full coordinates are listed in Appendices 1, 2.

Across the study period, the VIM-TAs evolved more superiorly to 2 mm above the ICL. This movement occurred independently across autonomous international FUS centers, with a combined experience exceeding 1,800 treatments. This change developed as experience accrued, presumably reflecting the view that this superior target provides better tremor suppression and/or minimizes adverse effects. The S-W 3D model also supports this concept, demonstrating the natural inferior tapering of VIM in the AP and ML planes (Figure 4). As VIM-TAs at 2 mm above the ICL lie more centrally within the VIM, sonications here will ablate more VIM tissue than at the ICL. Interestingly, centers with the least experience were most likely to move their VIM-TA across the study period, suggesting that the evolution of VIM-TAs tends to occur within the first 100 FUS treatments (Table 2B).

In 2022, 63.3% of participating centers (n = 19) provided the rationale for their VIM-TAs and key demographics, including the number of treatments performed (to date, 1 January 2022); a detailed analysis of this data is provided in Appendix 5. In summary, in 2021, the centers were more likely to target 2 mm above the ICL if they had increased experience (more than 100 treatments) (Table 2A) and/or if they were North American (rather than European or Asian) (Table 3, Figure 5). The reported rationales for VIM-TAs (Table 4, Appendix 4) included improved tremor suppression and a reduction in adverse effects or safety. Some centers reported that their VIM-TAs were influenced by their prior experience with deep brain stimulation (DBS) or gamma knife (GK). Others discussed the size, shape, and risk of cranial-caudal extension of the FUS sonication spot. Many centers reported that moving superiorly allowed them to perform a second, more inferior lesion in the same FUS procedure (which is also the practice at our centre at Imperial). The solitary center that moved inferiorly from 2019 to 2021 reported in their rationale high sensory adverse effects and a possible second ablation below ICL (Table 4, Appendices 4, 5). Interestingly, of the three centers targeting ICL who provided their rationale, all reported performing a second target (in the same FUS procedure) if tremor suppression was inadequate, either superior or inferior to ICL (Table 4B). Although these findings are of interest, not all participating centers provided a rationale for their VIM-TA, reducing the significance of these summations, and thus, clinical conclusions can only be drawn to a limited extent.

This study demonstrates that anatomical-based targeting of VIM is the most widely utilized methodology internationally for FUS thalamotomy despite recent advances in tractography. Over the study period, there was a statistically significant superior movement to target the VIM 2 mm above the intercommissural line. This superior evolution of VIM targeting has occurred independently across autonomous international centers, suggesting that it is an optimized site for FUS thalamotomy in the treatment of tremors.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

AJ: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing—original draft, Writing—review & editing. SA: Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing—review & editing. NY: Conceptualization, Data curation, Formal analysis, Methodology, Software, Validation, Visualization, Writing—review & editing. JS: Conceptualization, Methodology, Software, Visualization, Writing—review & editing, Formal analysis. BJ: Supervision, Methodology, Validation, Writing—review & editing. DN: Resources, Supervision, Writing—review & editing, Formal analysis, Methodology. PB: Resources, Supervision, Writing—review & editing, Formal Analysis, Methodology, Validation. WG: Conceptualization, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing—review & editing, Data curation, Formal analysis.