Abstract
Objective
Focused ultrasound thalamotomy is an effective treatment for tremor; however, side effects may occur. The purpose of the present study was to investigate the spatial relationship between thalamotomies and specific sensory side effects and their functional connectivity with somatosensory cortex and relationship to the medial lemniscus (ML).
Methods
Sensory adverse effects were categorized into 4 groups based on the location of the disturbance: face/mouth/tongue numbness/paresthesia, hand-only paresthesia, hemibody/limb paresthesia, and dysgeusia. Then, areas of significant risk (ASRs) for each category were defined using voxel-wise mass univariate analysis and overlaid on corresponding odds ratio maps. The ASR associated with the maximum risk was used as a region of interest in a normative functional connectome to determine side effect–specific functional connectivity. Finally, each ASR was overlaid on the ML derived from normative template.
Results
Of 103 patients, 17 developed sensory side effects after thalamotomy persisting 3 months after the procedures. Lesions producing sensory side effects extended posteriorly into the principle sensory nucleus of the thalamus or below the thalamus in the ML. The topography of sensory adverse effects followed the known somatotopy of the ML and the sensory nucleus. Functional connectivity patterns between each sensory-specific thalamic seed and the primary somatosensory areas supported the role of the middle insula in processing of gustatory information and in multisensory integration.
Conclusions
Distinct regions in the sensory thalamus and its afferent connections rise to specific sensory disturbances. These findings demonstrate the relationship between the sensory thalamus, ML, and bilateral sensory cortical areas.
Magnetic resonance–guided focused ultrasound (MRgFUS) thalamotomy is an effective therapy for tremor due to essential tremor (ET),1,2 tremor-predominant Parkinson disease,3,4 and dystonic tremor.5 Reported improvements in the clinical rating scale for tremor (CRST) range from 41% to 62% at 3 months of follow-up.1–3 Adverse effects are related to either the thalamotomy lesion or perilesional edema and include gait disturbance, ataxia, paresthesias, dysgeusia, and hemiparesis.1,3,4
Gait disturbance and/or ataxia have been reported in 5%–24% of patients at 1–3 months after MRgFUS.1,3,4,6 Focal paresthesias involving the face, hand, and/or tongue are also common, with a reported incidence approximately 25%.1,3 Taste disturbance (dysgeusia) has been reported in 3.6%–13.3% of patients,1,4 and hemiparesis has been reported in 3.7%–10% of patients.1,3 Prior work conducted at the authors' institution has helped to clarify the topography of these thalamotomy-related side effects.6,7 Areas of the thalamus known to elicit ataxia and gait disturbance when lesioned have been shown to overlap with the areas associated with maximal tremor reduction.6 This suggests that some degree of gait disturbance and ataxia may be unavoidable if the goal is to achieve an optimal therapeutic response. Conversely, lesions associated with sensory side effects extend into the ventral caudal nucleus (Vc) of the thalamus and the neighboring medial lemniscus (ML).6,7 Building on these findings,6,7 the goal of the present work was to investigate the relationship between MRgFUS lesions and specific sensory symptoms in terms of lesion location, functional network engagement, and association with ML.
Methods
Standard Protocol Approvals, Registrations, and Patient Consents
Institutional research ethics board approval to perform this retrospective study was granted at Sunnybrooke Health Science Centre (SHSC RED #059-2015) and Toronto Western Hospital (TWH REB #19-6194). As patient risk, including the risk of a breach of confidentiality, was deemed to be minimal, a waiver of individual patient informed consent was granted by the research ethics board at both centers. All clinical and neuroimaging data were anonymized on collection, and no patient identifiers were used during the analysis.
Patients
On attaining institutional research ethics board approval, we retrospectively reviewed the medical records of patients treated for tremor with MRgFUS thalamotomy at Sunnybrook Health Science Centre and Toronto Western Hospital between May 2012 and May 2018. Patients were seen for follow-up at 1 week, 1 month, 3 months, and 1 year. We identified patients with persistent sensory side effects at 3 months of follow-up. Although in some cases, side effects may persist at 1 year, we did not use data from this time point due to a higher likelihood of patients becoming lost to follow-up or delaying their appointments. Sensory side effects were separated into 4 groups based on the location/nature of the disturbance: face/mouth/tongue numbness/paresthesia, hand-only paresthesia, hemibody/limb paresthesia, and dysgeusia.
Lesion Segmentation
Lesions were segmented according to a previously published method.7 Briefly, postoperative axial 3D T1 fast spoiled gradient echo (FSPGR) images were obtained 1 day after MRgFUS treatment at 2 different sites: Sunnybrook Health & Science Centre (GE 3T MR650 Discovery repetition time: 8 ms; echo time; 3 ms; flip angle: 12; isotropic voxel: 1 mm) and Toronto Western Hospital (GE 3T HDx scanner repetition time: 8 ms; echo time: 3 ms; flip angle: 12°; isotropic voxel: 1 mm). Thalamotomy lesions, specifically zones I and II,8 visualized on axial FSPGR images were manually segmented using Display (MNI Display, Montreal Neurological Institute [MNI]; bic.mni.mcgill.ca/software/Display/Display.html).
The native postoperative 3D T1 MRI was coregistered with the native preoperative MRI using rigid transformation. Next, the native preoperative 3D T1 MRI scans were normalized to a standard brain space (MNI ICBM 152b Nonlinear Asymmetric) to enable group comparison and analysis. Nonlinear low variance normalization (ANTS, stnava.github.io/ANTs/, open source) was used as this has previously been shown to optimally normalize subcortical structures relevant to deep brain stimulation (DBS) and other functional neuromodulation techniques.9 The resulting transformations were sequentially applied to the MRgFUS lesion label to warp it to MNI space. The majority of lesions (84/103) were located in the left thalamus. Right thalamic lesions were nonlinearly flipped to the left side to allow for comparison of all lesions.
Lesion Analysis
Manually segmented thalamotomy lesion masks were categorized according to the type of sensory disturbance they were associated with: no sensory disturbance; paresthesias affecting the mouth, face or tongue; paresthesias affecting the hand only; paresthesias involving the limbs or hemibody; and dysgeusia. Each thalamotomy lesion could belong to more than 1 category depending on the types of side effects the patient experienced.
To investigate the relationship between thalamotomy location and symptom-specific sensory side effects (face/mouth/tongue numbness, hand numbness, hemibody/limb numbness, and dysgeusia), a voxel-wise mass univariate analysis was performed on the voxels included in the thalamotomies. The mass univariate t test compared thalamotomies associated with a specific symptom with those with no sensory side effects. The resultant t-maps were then thresholded at p false discovery rate (FDR) <0.05 (R 3.4.4: r-project.org/; and RMINC: github.com/Mouse-Imaging-Centre/RMINC) yielding areas of statistically significant risk of the corresponding side effect. Next, the significant voxels were binarized, producing lesion masks of the areas associated with statistically significant risk of each specific sensory symptom. We will refer to these masks as areas of significant risk (ASRs).
To estimate the magnitude of risk associated with the ASR for each type of sensory side effect, we computed symptom-specific voxel-wise odds ratio maps (VOR).6,10 OR maps display a relative risk at each voxel of the specific sensory side effect when this voxel is overlapped by a thalamotomy lesion. VOR maps were calculated as follows:
 |
Where Np = number of patients having a particular sensory symptom, Nc = the number of patients without the side effect, Vp = number of patients with the side effect with a lesion at a specific voxel, and Vc = number of patients without the side effect with a lesion at a specific voxel. These VOR maps were then overlaid on the corresponding ASR, providing a likelihood at each ASR voxel of being associated with specific sensory side effects.
Functional Networks Engaged With Symptom-Specific Areas
For each type of sensory disturbance, the voxel(s) of the ARS maps associated with the maximal odds of causing the sensory disturbance was chosen as a seed for measuring functional connectivity. A 2 × 2 × 2 mm cube was centered on the maximal odds ratio voxel (or the center of gravity of all the maximal odds ratio voxels in the case that multiple voxels shared the highest value) within the ARS map as a region of interest. These regions of interests (ROIs) were used for functional connectomic mapping (in-house MATLAB script, Version R2018a; The MathWorks, Inc., Natick, MA). This analysis used a normative data set compiled from resting-state functional MRI scans of 1,000 healthy subjects, as reported previously.11 Whole-brain connectivity r-maps were generated for each individual seed using the entire normative data set. For each seed, a connectivity r-map describing the correlation between the ROI seed and every voxel in the brain was obtained. These r-maps were converted to t-maps. We focused on connectivity between the thalamic ROIs and the cortical areas known to be involved in processing of somatosensory and gustatory information.12–15
Structural Networks Engaged With Symptom-Specific Areas
To visualize which portions of the ML interacted with each of the ASR maps, each map was overlaid on an aggregate ML tract constructed from a normative structural template. This template was derived using previously reported methods16 from a 985-subject multishell diffusion-weighted MRI data set sourced from the Human Connectome Project (humanconnectomeproject.org/). The ascending somatosensory fiber streamlines (here, presumed to represent ML in addition to spinothalamic tract streamlines) within this tractogram were then isolated using inclusion ROIs in the dorsal pontine tegmentum and postcentral gyrus (MI-Brain; imeka.ca/mi-brain). The streamlines within this ML tract that intersected each of the ASR maps were identified. All steps of the neuroimaging analysis are summarized in figure e-1 (links.lww.com/CPJ/A228).
Data Availability
The data and custom code that support the central findings of this study are available from the corresponding author.
Results
Patients
Patient demographics are summarized in table 1. The entire patient cohort consisted of 70 males and 33 females, with a mean baseline CRST total score of 20.1 ± 5.4 and a median postoperative improvement of 56% (40.3%–75%) at 3 months of follow-up. Of the 103 patients included in the analysis, 17 (17%) experienced sustained sensory side effects at 3 months of follow-up after MRgFUS thalamotomy. Among patients with persistent sensory disturbances, 12 patients had numbness of the face, mouth, or tongue. Seven patients had numbness or paresthesias of the hand, 3 patients had paresthesias of the limb or hemibody, and 4 patients had dysgeusia.
Table 1.
Patient Characteristics (Total N = 103)
Lesion Analysis
The subtraction maps produced by subtracting the summed map of the lesions associated with any type of sensory disturbance from the summed maps of the lesions associated with no sensory side effects showed that in general, lesions associated with sensory side effects tended to extend more laterally and posteriorly into the region of the Vc and ML (figure 1).
Figure 1. Location of Sensory Side Effects.
Subtraction maps were created by subtracting each summed map of the lesions associated with each type of sensory side effect (n = 17) from the summed map of all the lesions associated with no sensory side effects (n = 86). Shown are areas covered by the thalamotomy lesions associated with all sensory side effects combined. The color bar indicates the number of subtraction maps that overlap at each voxel. (A) Sagittal and (B) axial T1-weighted MRI (Montreal Neurological Institute) of the brain demonstrates that the lesions associated with sensory side effects tended to extend more posteriorly from the ventral intermediate nucleus (green) into the ventral caudal nucleus (pink) (Hassler/Schaltenbrand and Wahren nomenclature).
The ASR maps—indicating voxels associated with a statistically significant risk of each specific sensory side effect (p < 0.05, FDR-corrected for multiple comparisons)—were visualized in standard MNI space in relation to the sensory regions of the thalamus (figure 2). The dysgeusia ASR was located the most medial. According to the Hassler/Schaltenbrand and Wahren nomenclature, 14.0% of the volume of the dysgeusia ASR occupied the Vc. Using the Hirai and Jones nomenclature, 61.2% resided within the ventral posterior medial (VPM) nucleus, and 20.9% extended into the ventral posterior lateral (VPL) nucleus. Approximately 6% of the dysgeusia ASR overlapped with fibers of the ML. The MNI coordinates of the center of gravity of the voxels associated with the highest odds of dysgeusia were X(−10.6), Y(20.5), and Z(2.1).
Figure 2. Sensory Side Effect ASRs.
Binary ASR maps overlaid on thalamic nuclei, indicating areas associated with a statistically significant risk of the sensory side effect (p < 0.05, FDR-corrected) on T1-weighted MRI (Montreal Neurological Institute) of the brain. (A) Coronal image of the sensory ASRs in relation to the ML (yellow), Vim (green), and Vc (pink). (B) Axial image of the sensory ASRs in relation to the ML, Vim, and Vc. (C) Coronal image of the sensory ASRs in relation to the VPM (blue) and VPL (red) thalamic nuclei. (D) Axial image of the sensory ASRs in relation to the VPM and VPL thalamic nuclei. Blue: dysgeusia; red: hand; turquoise: face; green: limb/hemibody, A and B: pink: Vc; light green: Vim, C and D: blue: VMP; red: VPL. ASR = area of significant risk; FDR = false discovery rate; ML = medial lemniscus; Vc = ventral caudal nucleus; Vim = ventral intermediate nucleus; VPL = ventral posterior lateral; VPM = ventral posterior medial.
Approximately 60% of the ASR maps for hand and face paresthesias overlapped with one another within the Vc, with the face ASR extending more medially. Areas associated with the highest odds of face paresthesias were located adjacent to those areas associated with the highest odds of hand paresthesias. The MNI coordinates of the maximal risk of face paresthesias were X(−18.8), Y(−22.0), and Z(−1.5). Those associated with the highest risk of hand paresthesias were X(−17.5), Y(−22.7), and Z(−3.1). Approximately, 61% of the face ASR was located within the Vc, and 56% overlapped with the VPL. Surprisingly, the face ASR was not observed to overlap with the VPM. Approximately 39% of the face ASR overlapped with ML fibers. Approximately 68% of the volume of the hand ASR overlapped with the Vc, and 45% overlapped with the VPL. Thirty-eight percent of the hand ASR overlapped with the ML.
The voxels comprising the limb ASR were more scattered and extended more laterally compared with the other ASRs. The voxels associated with increased odds of limb paresthesias were located outside and inferior to the Vc and VPL nuclei. This suggests that limb and hemibody paresthesias were likely related to inclusion of ML fibers in the lesions. Approximately 30% of the total volume of the limb/hemibody ASR overlapped with the ML, whereas 20.3% overlapped with the Vc. Approximately 5% of the limb ASR occupied the area of the VPM and another 5% occupied the region of the VPL. The MNI coordinates associated with the highest risk of limb paresthesias were X(−15.5), Y(−19.6), and Z(−8.1).
Overlap between the sensory ASRs and thalamic nuclei/ML is summarized in table e-1 (links.lww.com/CPJ/A229). The odds ratios and maximal risk coordinates for each sensory side effect are provided in table e-2 (links.lww.com/CPJ/A229).
Functional Connectivity With Primary Somatosensory Cortical Regions
The ASR voxels associated with the maximal likelihood of each sensory side effect were used as seeds in a normative functional connectivity template to explore underlying networks engaged with the thalamotomies. Connectivity between these thalamic seeds and the primary somatosensory cortex and insula was examined. Each seed exhibited bilateral positive cortical connectivity, although the areas of positive connectivity were more expansive in the left hemisphere compared with the right hemisphere.
The areas of strongest functional connectivity with the dysgeusia seed were found within the middle insular cortex, known to serve as the primary gustatory processing area.13,15,17 Another focus of strong connectivity was found within the central sulcus at the convexity of the postcentral gyrus. Positive functional connectivity between the hand seed and the primary somatosensory cortex was localized to the parietal operculum. The hand seed also demonstrated strong connectivity with the anterior inferior insular cortex. The strongest connectivity between the face seed and the somatosensory cortex was distributed along the length of the anterior postcentral gyrus, possibly reflecting the extensive representation of the face, mouth, and tongue. There was strong connectivity between the face seed and the middle insula as well. Positive connectivity between the limb/hemibody seed and somatosensory cortex was localized to a portion of the medial postcentral gyrus and also more laterally at the operculum. There was also strong connectivity found between the limb/hemibody seed and the anterior inferior insula. Connectivity profiles are visually depicted in figure 3.
Figure 3. Cortical Connectivity Maps of Thalamic Seed ROIs Associated With the Highest Risk of Sensory Side Effects.
Primary somatosensory cortex (postcentral gyrus) and insular cortex ROIs were isolated from the Harvard-Oxford atlas.26 These ROIs were used to visualize the functional connectivity between each seed and the corresponding primary sensory areas. The strength of connectivity is expressed by t values. Note that although all thalamic seed ROIs demonstrated positive connectivity with portions of the insula, the strongest was with the dysgeusia seed ROI, which is reflected in this figure. Blue: dysgeusia; turquoise: face; red: hand; green: limb/hemibody. ROI = region of interest.
Involvement of ML Fibers
Each ASR map was overlaid on the ML to visualize the extent and somatotopy of ML involvement in each thalamotomy-related sensory side effect. The streamlines contacted by the hand, face, and limb/hemibody ASRs demonstrated a higher degree of overlap and were located more laterally within the ML. Axial and radial views are used to visualize the somatotopy within the ML just inferior to the thalamus (figure 4A) and at the level of the midbrain (figure 4B). These maps show the fibers of the gustatory tract located at the medial extent of the ML followed by fibers associated with the face, hand, and limb/hemibody ASRs.
Figure 4. Somatotopy of the Medial Lemniscus.
Streamlines of the ML touched by the ASRs overlaid on a T1-weighted MRI (Montreal Neurological Institute) of the brain. (A) Just inferior to the level of the thalamus and (B) at the level of the midbrain. ML fibers touched by the dysgeusia ASR are located medially, whereas fibers touched by the limb/hemibody ASR are located at the lateral margin of the ML. Blue: dysgeusia; turquoise: face; red: hand; green: limb/hemibody. ASR = area of significant risk; ML = medial lemniscus.
Discussion
The purpose of this work was to identify the topography of MRgFUS thalamic lesions associated with specific sensory disturbances. We found that the thalamic regions associated with a statistically significant risk of tactile sensory side effects were distributed along the known somatotopic organization of the sensory thalamus.14,18–20 Voxels carrying the highest risk of dysgeusia were located the most medial within the thalamus and contacted fibers at the medial margin of the ML. This is in agreement with the findings of Sajonz et al.,21 in which patients who developed dysgeusia from thalamic DBS were found to have stimulation volumes significantly closer to the medial margin of the ML compared with those of patients without dysgeusia.
The ASRs associated with hand and face/mouth paresthesias were found to occupy a more lateral position within the Vc. The contiguity of these regions is supported by the finding that over half of the patients with hand paresthesias also had paresthesias of the face and mouth. Voxels associated with the highest risk of limb and hemibody paresthesias were located inferior to the Vc, indicating that limb and hemibody paresthesias are related to the inclusion of ML fibers in the thalamotomy lesion.
There are several limitations of the present study that should be noted. As this was a retrospective study, we had to rely on the description of sensory side effects in the patients' medical records, and this comprises a source of potential error. In addition, there were relatively few patients with persistent side effects at 3 months, and therefore, patients often belonged to more than 1 group. Ideally, with a larger study group and in a prospective manner, patients might better be grouped according to their most predominant sensory side effect.
Microelectrode recordings performed in humans and nonhuman primates have demonstrated that neurons responding to stimulation of distinct anatomic regions are organized in concentric parasagittal lamellae with medial lamellae processing sensory input from the mouth and face, and the more lateral lamellae receiving sensory input from the hand, followed by the arms and legs.14,18,19 The VPM has been identified by anatomists as containing neurons with receptive fields on the face, mouth, and intraoral structures, whereas VPL, which is separated from the VPM by the arcuate lamina, contains neurons with receptive fields in the hand and extremities.14,18 Experimental lesion studies in animal models have demonstrated that fibers carrying sensory information from the lower limbs run laterally within the ML and terminate in the lateral ventral thalamus.22 Similarly, fibers carrying information from the upper limbs and fibers projecting from the spinal nucleus of V have been shown to run in the more medial portions of the ML and terminate in the middle and medial regions of the ventral thalamus, respectively.22
The thalamic area responsible for processing of gustatory information has been identified anatomically in animal models as the parvocellular component of the VPM (VPMpc).23 Located at the medial tip of the VPM, the VPMpc receives second-order projections from the nucleus of the solitary tract.17 Lesion cases suggest that the human gustatory tract enters the thalamus adjacent to the medial margin of the ML.24 This organization is supported by the observations of the present study in which streamlines running at the extreme medial edge of the ML are seen to interact with the ASR for dysgeusia. From the thalamus, fibers conveying taste information travel to the ipsilateral insula and frontal operculum, which is believed to be the location of the primary gustatory center.17 A second projection terminating in areas 3a, 3b, 2, and 1 of the somatosensory cortex has also been identified in nonhuman primates.17 Functional MRI studies have supported that the anterior and middle insula are primary receiving areas of gustatory information and are functionally connected with the thalamus.13,15,25 In line with this, we observed positive functional connectivity between the ASR for dysgeusia and the middle insula and focal areas of the somatosensory cortex.
MRgFUS is an effective modality for performing thalamotomy to treat tremor. Although unwanted side effects may occur, the majority are self-limiting and resolve over time. Group-wise imaging analysis of patients with persistent side effects may identify thalamic regions within the thalamus and neighboring white matter associated with increased risks of specific side effects. This information may in turn be implemented to refine targeting for the purpose of avoiding adverse effects. Other strategies for achieving optimal response while minimizing side effects include use of tractography to visualize the ML, pyramidal tract, and dentate‐rubro‐thalamic tract. Incorporation of multiple imaging strategies to help guide intraoperative clinical decision making will likely yield the most optimal patient outcomes.
Appendix. Authors
Study Funding
This work was supported by the RR Tasker Chair in Functional Neurosurgery at University Health Network.
Disclosure
M. Paff, A. Boutet, J. Germann, G.J.B. Elias, C.T. Chow, A. Loh, and W. Kucharczyk have no conflicts of interest to disclose. A. Fasano reports grants, personal fees, and nonfinancial support from AbbVie; grants, personal fees, and nonfinancial support from Medtronic; grants and personal fees from Boston Scientific; personal fees from Sunovion; personal fees from Chiesi Farmaceutici; personal fees from UCB; and grants and personal fees from Ipsen, all unrelated to the present study. M.L. Schwartz has no conflicts of interest to disclose. A.M. Lozano is a consultant for Insightec and reports personal fees from Medtronic, personal fees from St. Jude, personal fees from Boston Scientific, and personal fees from Functional Neuromodulation and grants from GE Healthcare unrelated to the activities of the present study. All other authors have no conflicts of interest to disclose. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/cp.
References
- 1.Elias WJ, Lipsman N, Ondo WG, et al. A randomized trial of focused ultrasound thalamotomy for essential tremor. N Engl J Med 2016;375:730–739. [DOI] [PubMed] [Google Scholar]
- 2.Lipsman N, Schwartz ML, Huang Y, et al. MR-guided focused ultrasound thalamotomy for essential tremor: a proof-of-concept study. Lancet Neurol 2013;12:462–468. [DOI] [PubMed] [Google Scholar]
- 3.Bond AE, Shah BB, Huss DS, et al. Safety and efficacy of focused ultrasound thalamotomy for patients with medication-refractory, tremor-dominant Parkinson disease: a randomized clinical trial. JAMA Neurol 2017;74:1412–1418. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Zaaroor M, Sinai A, Goldsher D, Eran A, Nassar M, Schlesinger I. Magnetic resonance-guided focused ultrasound thalamotomy for tremor: a report of 30 Parkinson's disease and essential tremor cases. J Neurosurg 2018;128:202–210. [DOI] [PubMed] [Google Scholar]
- 5.Fasano A, Llinas M, Munhoz RP, Hlasny E, Kucharczyk W, Lozano AM. MRI-guided focused ultrasound thalamotomy in non-ET tremor syndromes. Neurology 2017;89:771–775. [DOI] [PubMed] [Google Scholar]
- 6.Boutet A, Ranjan M, Zhong J, et al. Focused ultrasound thalamotomy location determines clinical benefits in patients with essential tremor. Brain 2018;141:3405–3414. [DOI] [PubMed] [Google Scholar]
- 7.Ranjan M, Elias GJB, Boutet A, et al. Tractography-based targeting of the ventral intermediate nucleus: accuracy and clinical utility in MRgFUS thalamotomy. J Neurosurg 2019:1–8. [DOI] [PubMed] [Google Scholar]
- 8.Wintermark M, Druzgal J, Huss DS, et al. Imaging findings in MR imaging-guided focused ultrasound treatment for patients with essential tremor. AJNR Am J Neuroradiol 2014;35:891–896. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ewert S, Horn A, Finkel F, Li N, Kuhn AA, Herrington TM. Optimization and comparative evaluation of nonlinear deformation algorithms for atlas-based segmentation of DBS target nuclei. Neuroimage 2019;184:586–598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sprenger T, Seifert CL, Valet M, et al. Assessing the risk of central post-stroke pain of thalamic origin by lesion mapping. Brain 2012;135:2536–2545. [DOI] [PubMed] [Google Scholar]
- 11.Smith SM, Beckmann CF, Andersson J, et al. Resting-state fMRI in the human connectome project. Neuroimage 2013;80:144–168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Ruben J, Schwiemann J, Deuchert M, et al. Somatotopic organization of human secondary somatosensory cortex. Cereb Cortex 2001;11:463–473. [DOI] [PubMed] [Google Scholar]
- 13.Iannilli E, Noennig N, Hummel T, Schoenfeld AM. Spatio-temporal correlates of taste processing in the human primary gustatory cortex. Neuroscience 2014;273:92–99. [DOI] [PubMed] [Google Scholar]
- 14.Kaas JH, Nelson RJ, Sur M, Dykes RW, Merzenich MM. The somatotopic organization of the ventroposterior thalamus of the squirrel monkey, Saimiri sciureus. J Comp Neurol 1984;226:111–140. [DOI] [PubMed] [Google Scholar]
- 15.Cauda F, D'Agata F, Sacco K, Duca S, Geminiani G, Vercelli A. Functional connectivity of the insula in the resting brain. Neuroimage 2011;55:8–23. [DOI] [PubMed] [Google Scholar]
- 16.Horn A, Reich M, Vorwerk J, et al. Connectivity Predicts deep brain stimulation outcome in Parkinson disease. Ann Neurol 2017;82:67–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Small DM. Central gustatory processing in humans. Adv Otorhinolaryngol 2006;63:191–220. [DOI] [PubMed] [Google Scholar]
- 18.Jones EG, Friedman DP. Projection pattern of functional components of thalamic ventrobasal complex on monkey somatosensory cortex. J Neurophysiol 1982;48:521–544. [DOI] [PubMed] [Google Scholar]
- 19.Lenz FA, Dostrovsky JO, Tasker RR, Yamashiro K, Kwan HC, Murphy JT. Single-unit analysis of the human ventral thalamic nuclear group: somatosensory responses. J Neurophysiol 1988;59:299–316. [DOI] [PubMed] [Google Scholar]
- 20.Lenz FA, Seike M, Richardson RT, et al. Thermal and pain sensations evoked by microstimulation in the area of human ventrocaudal nucleus. J Neurophysiol 1993;70:200–212. [DOI] [PubMed] [Google Scholar]
- 21.Sajonz BE, Madler B, Herberhold S, Paus S, Allert N, Coenen VA. Stimulation induced hypogeusia in thalamic deep brain stimulation for tremor: an underestimated yet common side effect. J Neurol Neurosurg Psychiatry 2016;87:565–567. [DOI] [PubMed] [Google Scholar]
- 22.le Gros Clark WE. The termination of ascending tracts in the thalamus of the Macaque monkey. J Anat 1936;71:7–40. [PMC free article] [PubMed] [Google Scholar]
- 23.Verhagen JV, Giza BK, Scott TR. Responses to taste stimulation in the ventroposteromedial nucleus of the thalamus in rats. J Neurophysiol 2003;89:265–275. [DOI] [PubMed] [Google Scholar]
- 24.Lee BC, Hwang SH, Rison R, Chang GY. Central pathway of taste: clinical and MRI study. Eur Neurol 1998;39:200–203. [DOI] [PubMed] [Google Scholar]
- 25.Iannilli E, Singh PB, Schuster B, Gerber J, Hummel T. Taste laterality studied by means of umami and salt stimuli: an fMRI study. Neuroimage 2012;60:426–435. [DOI] [PubMed] [Google Scholar]
- 26.Desikan RS, Segonne F, Fischl B, et al. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. Neuroimage 2006;31:968–980. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data and custom code that support the central findings of this study are available from the corresponding author.