Differences in intrinsic functional networks in patients with essential tremor who had good and poor long-term responses after thalamotomy performed using MR-guided ultrasound

Chongwon Pae Center for Systems and Translational Brain Sciences, Institute of Human Complexity and Systems Science, Yonsei University, Seoul;
Department of Nuclear Medicine, Yonsei University College of Medicine, Seoul;
Department of Psychiatry, Bundang CHA Medical Center, CHA University School of Medicine, Seongnam, Republic of Korea

Search for other papers by Chongwon Pae in
Current site
Google Scholar
PubMed
Close
 PhD
,
Myung Ji Kim Department of Neurosurgery, Korea University College of Medicine, Korea University Medical Center, Ansan Hospital, Gyeonggi-do;

Search for other papers by Myung Ji Kim in
Current site
Google Scholar
PubMed
Close
 MD
,
Won Seok Chang Department of Neurosurgery, Yonsei University College of Medicine, Seoul;
Center for Innovative Functional Neurosurgery, Brain Research Institute, Seoul;

Search for other papers by Won Seok Chang in
Current site
Google Scholar
PubMed
Close
 MD, PhD
,
Hyun Ho Jung Department of Neurosurgery, Yonsei University College of Medicine, Seoul;
Center for Innovative Functional Neurosurgery, Brain Research Institute, Seoul;

Search for other papers by Hyun Ho Jung in
Current site
Google Scholar
PubMed
Close
 MD, PhD
,
Kyung Won Chang Department of Neurosurgery, Yonsei University College of Medicine, Seoul;

Search for other papers by Kyung Won Chang in
Current site
Google Scholar
PubMed
Close
 MD
,
Jinseok Eo Center for Systems and Translational Brain Sciences, Institute of Human Complexity and Systems Science, Yonsei University, Seoul;
Department of Nuclear Medicine, Yonsei University College of Medicine, Seoul;
Graduate School of Medical Science, Brain Korea 21 Project, Yonsei University College of Medicine, Seoul;

Search for other papers by Jinseok Eo in
Current site
Google Scholar
PubMed
Close
,
Hae-Jeong Park Center for Systems and Translational Brain Sciences, Institute of Human Complexity and Systems Science, Yonsei University, Seoul;
Department of Nuclear Medicine, Yonsei University College of Medicine, Seoul;
Graduate School of Medical Science, Brain Korea 21 Project, Yonsei University College of Medicine, Seoul;
Department of Cognitive Science, Yonsei University, Seoul; and

Search for other papers by Hae-Jeong Park in
Current site
Google Scholar
PubMed
Close
 PhD
, and
Jin Woo Chang Department of Neurosurgery, Yonsei University College of Medicine, Seoul;
Center for Innovative Functional Neurosurgery, Brain Research Institute, Seoul;

Search for other papers by Jin Woo Chang in
Current site
Google Scholar
PubMed
Close
 MD, PhD
Free access

OBJECTIVE

Thalamotomy at the nucleus ventralis intermedius using MR-guided focused ultrasound has been an effective treatment method for essential tremor (ET). However, this is not true for all cases, even for successful ablation. How the brain differs in patients with ET between those with long-term good and poor outcomes is not clear. To analyze the functional connectivity difference between patients in whom thalamotomy was effective and those in whom thalamotomy was ineffective and its prognostic role in ET treatment, the authors evaluated preoperative resting-state functional MRI in thalamotomy-treated patients.

METHODS

Preoperative resting-state functional MRI data in 85 patients with ET, who were experiencing tremor relief at the time of treatment and were followed up for a minimum of 6 months after the procedure, were collected for the study. The authors conducted a graph independent component analysis of the functional connectivity matrices of tremor-related networks. The patients were divided into thalamotomy-effective and thalamotomy-ineffective groups (thalamotomy-effective group, ≥ 50% motor symptom reduction; thalamotomy-ineffective group, < 50% motor symptom reduction at 6 months after treatment) and the authors compared network components between groups.

RESULTS

Seventy-two (84.7%) of the 85 patients showed ≥ 50% tremor reduction from baseline at 6 months after thalamotomy. The network analysis shows significant suppression of functional network components with connections between the areas of the cerebellum and the basal ganglia and thalamus, but enhancement of those between the premotor cortex and supplementary motor area in the noneffective group compared to the effective group.

CONCLUSIONS

The present study demonstrates that patients in the noneffective group have suppressed functional subnetworks in the cerebellum and subcortex regions and have enhanced functional subnetworks among motor-sensory cortical networks compared to the thalamotomy-effective group. Therefore, the authors suggest that the functional connectivity pattern might be a possible predictive factor for outcomes of MR-guided focused ultrasound thalamotomy.

ABBREVIATIONS

CRST = clinical rating scale for tremor; ET = essential tremor; ET-NR = ET nonresponder; ET-R = ET responder; FC = functional connectivity; graph-ICA = graph independent component analysis; HFS = hemifacial spasm; IC = independent component; ICL = intercommissural line; MCP = midcommissural point; MRgFUS = MR-guided focused ultrasound; NC = non-ET control; PCA = principal component analysis; rsfMRI = resting-state functional MRI; VIM = ventral intermediate nucleus.

OBJECTIVE

Thalamotomy at the nucleus ventralis intermedius using MR-guided focused ultrasound has been an effective treatment method for essential tremor (ET). However, this is not true for all cases, even for successful ablation. How the brain differs in patients with ET between those with long-term good and poor outcomes is not clear. To analyze the functional connectivity difference between patients in whom thalamotomy was effective and those in whom thalamotomy was ineffective and its prognostic role in ET treatment, the authors evaluated preoperative resting-state functional MRI in thalamotomy-treated patients.

METHODS

Preoperative resting-state functional MRI data in 85 patients with ET, who were experiencing tremor relief at the time of treatment and were followed up for a minimum of 6 months after the procedure, were collected for the study. The authors conducted a graph independent component analysis of the functional connectivity matrices of tremor-related networks. The patients were divided into thalamotomy-effective and thalamotomy-ineffective groups (thalamotomy-effective group, ≥ 50% motor symptom reduction; thalamotomy-ineffective group, < 50% motor symptom reduction at 6 months after treatment) and the authors compared network components between groups.

RESULTS

Seventy-two (84.7%) of the 85 patients showed ≥ 50% tremor reduction from baseline at 6 months after thalamotomy. The network analysis shows significant suppression of functional network components with connections between the areas of the cerebellum and the basal ganglia and thalamus, but enhancement of those between the premotor cortex and supplementary motor area in the noneffective group compared to the effective group.

CONCLUSIONS

The present study demonstrates that patients in the noneffective group have suppressed functional subnetworks in the cerebellum and subcortex regions and have enhanced functional subnetworks among motor-sensory cortical networks compared to the thalamotomy-effective group. Therefore, the authors suggest that the functional connectivity pattern might be a possible predictive factor for outcomes of MR-guided focused ultrasound thalamotomy.

In Brief

To analyze the brain connectivity difference between patients with essential tremors treated with MRI-guided focused ultrasound that resulted in long-term thalamotomy effectiveness and noneffectiveness, the authors evaluated preoperative functional connectivity in patients who were successfully treated but who had good and poor long-term outcomes. Patients in the thalamotomy-noneffective group have suppressed functional subnetworks in the cerebellum and subcortex, and have enhanced functional subnetworks among motor-sensory cortical networks compared to thalamotomy-effective group. Functional connectivity patterns could be a possible predictive factor for outcomes of thalamotomy.

Essential tremor (ET) is the most prevalent movement disorder, affecting up to 4% of the population. It is characterized by a postural and intentional tremor that can range from mild to disabling high-amplitude tremor.1 Pharmacotherapy is the first-line treatment for ET but has drug-induced adverse effects. Approximately 50% of patients with ET experience recurrent tremors, indicating that the medication is ineffective in controlling the tremor.2 MR-guided focused ultrasound (MRgFUS) thalamotomy of the ventral intermediate nucleus (VIM) of the thalamus has recently been proven to be an effective treatment option for medication-refractory ET after receiving US Food and Drug Administration approval.3 It allows for real-time monitoring of patients to assess tremor control while identifying any possible side effects. Most treated patients experience reduced and sustained tremor symptoms and have minimal or no disability due to tremors. However, recurrence occurs in some patients following MRgFUS thalamotomies. The prognostic factors for thalamotomy have been discussed in terms of maximal temperature and energy delivery in many studies.4,5 However, recurrence may not be explained purely by nonbrain factors. It has been reported that approximately 23% of successfully treated patients have recurrence of symptoms at 6 months or 1 year later.3 In addition to an improper location or inadequate size of the thermal lesion, there may be neurobiological factors that are associated with treatment outcomes. According to a perspective on the brain as a network,6 we hypothesized that the nonresponder’s brain network may influence the poor outcome of MRgFUS thalamotomy. Thus, we explored which brain networks in ET predisposed patients to recurrence after MRgFUS thalamotomy, even with proper thermal treatment of the lesion.

To construct functional brain networks, we analyzed resting-state functional MRI (rsfMRI), which detects low-frequency endogenous synchrony (called functional connectivity [FC]) in the brain at rest.6 Despite the widespread use of the Pearson correlation as an FC measure, the reliability of the rsfMRI FC has been an issue.7 Given that multivariate connectivity analysis is advantageous in terms of reliability,8 we used a novel graph independent component analysis (graph-ICA),9 which decomposes independent components (ICs) from cross-sectional group networks. The ICs correspond to brain subnetworks embedded in resting-state spontaneous fluctuations. By comparing each subnetwork’s contribution at every time point, we investigated the tremor-related brain networks of patients with ET and identified brain network organization that differentiates among ET patients with thalamotomy-effective (ET responders [ET-Rs]), thalamotomy-ineffective (ET nonresponders [ET-NRs]), and non-ET control (NC) groups.

Methods

Study Subjects

Patients

A total of 118 patients with medication-refractory ET who underwent unilateral MRgFUS thalamotomy were enrolled in the present study. Twenty-one patients were excluded because their follow-up periods were less than 6 months or they did not undergo a 6-month follow-up clinical evaluation. Two patients with artifact rsfMRI and 10 patients with maximal temperature less than 54°C, which may be related to insufficient or inadequate thermal lesion,10 were also excluded. Ultimately, 85 patients who were in tremor relief at the time of treatment were included in the study. The patient demographics and treatment parameters are described in Table 1. The mean age of the 85 patients was 65.3 ± 8.7 years, ranging from 45 to 86 years. Among them, 67 patients were men (78.8%), and 18 were women (21.2%). The mean duration of disease was 13.8 ± 10.6 years, ranging from 1 to 40 years. We used patients with hemifacial spasm (HFS) (n = 14) and trigeminal neuralgia (n = 2) as a control group (6 men and 10 women), both of which were caused by benign compression of the facial and trigeminal nerves. None of the patients in the control group presented with hand tremors or had structural abnormalities on their MR images. The mean age of the control group was 58.8 ± 9.0 years, which is significantly different from the age of ET patients (p = 0.044). A significant sex difference exists among the 3 groups, and is attributable to the female predominance of the patients with HFS used as a control group.11 Patients who were on antitremor medication stopped taking the drug 1 day before the procedure in order not to suppress tremor activity.

TABLE 1.

Demographic characteristics, CRST, and treatment-related parameters, locations, and volume sizes of the treated lesions

Value
Age, yrs65.3 ± 8.7 (45–86)
Sex
 Male67 (78.8%)
 Female18 (21.2%)
Medical comorbidity
 Hypertension17 (20.0%)
 Diabetes mellitus9 (10.6%)
 Arrhythmia3 (3.5%)
 Angina1 (1.2%)
Lt side treated85 (100%)
Disease duration, yrs13.8 ± 10.6 (1–40)
Baseline CRST subscore, treated18.2 ± 5.3 (7–31)
Baseline CRST subscore, untreated11.4 ± 6.0 (0–24)
Combined resting tremor33 (38.8%)
 Hand20 (23.5%)
 Head8 (9.4%)
 Hand & head5 (5.9%)
1-mo CRST subscore, treated3.7 ± 4.0 (0–19)
1-mo CRST subscore, untreated11.3 ± 6.1 (0–26)
1-mo tremor improvement in %, treated81.7 ± 17.6 (20–100)
3-mo CRST subscore, treated3.9 ± 4.3 (0–16)
3-mo CRST subscore, untreated9.2 ± 4.9 (0–25)
3-mo tremor improvement in %, treated86.9 ± 20.1 (26–100)
6-mo CRST subscore, treated5.1 ± 5.5 (0–25)
6-mo CRST subscore, untreated10.5 ± 6.4 (0–34)
6-mo tremor improvement in %, treated74.3 ± 25.5 (15–100)
Transient adverse events16 (18.8%)
 Balance9 (10.6%)
 Sensory3 (3.5%)
 Speech3 (3.5%)
 Strength1 (1.2%)
Serious permanent adverse event0 (0%)
Skull density ratio0.54 ± 0.08 (0.39–0.78)
Skull vol, cm3336.9 ± 41.4 (236–419)
Max energy, J20,686.2 ± 7,923.0 (6,700–36,000)
Max energy delivered, J20,166.6 ± 7,885.8 (6,700–36,501)
Max power, W1,014.7 ± 1,714.7 (455–16,550)
Max temperature, °C57.2 ± 2.0 (54–64)
No. of sonications14.1 ± 2.3 (9–25)
No. of therapeutic sonications7.9 ± 1.8 (2–12)
Length of AC-PC, mm24.2 ± 1.6 (21.5–27.0)
1/2 length of 3rd ventricle, mm3.8 ± 1.1 (2.0–6.1)
Location of necrotic core; center of zone I
 Lateral to midline, mm14.4 ± 0.7 (12.8–15.4)
 Ant to PC as ratio of AC-PC length0.30 ± 0.07 (0.23–0.36)
 Sup to ICL, mm0.15 ± 0.80 (−1.20 to 1.15)*
Vol of zone I, cm30.011 ± 0.004 (0.005–0.017)
Vol of zone II, cm30.058 ± 0.031 (0.032–0.114)
Vol of zone III, cm30.173 ± 0.064 (0.082–0.326)
Total vol, cm30.242 ± 0.091 (0.123–0.455)
Diam of zone I, mm2.35 ± 0.34 (1.70–2.80)
Diam of zone I & II, mm5.16 ± 0.99 (3.00–6.40)
Diam of zone I–III, mm8.34 ± 0.75 (6.50–9.50)

AC = anterior commissure; ant = anterior; diam = diameter; max = maximal; PC = posterior commissure; sup = superior.

Descriptive statistics are presented as means ± SDs (ranges).

Negative values indicate coordinates of necrotic core inferior to ICL.

All patients with ET underwent MRgFUS thalamotomy of the left VIM. The MRgFUS thalamotomy procedure was previously described in detail.3,12 Briefly, all procedures were performed using a 3.0-T MRI system (GE Medical Systems) with an ExAblate 4000 device (InSightec). A Radionics frame (Integra Radionics) was used to fix the patient’s head in place. Unilateral lesioning was conducted with a series of ablative sonication after refinement between the transducer focus and the target with low-dose sublethal energy sonication. Temperature elevation was measured using real-time MR thermometry during the delivery of acoustic energy, and the patient was evaluated for tremor improvement and adverse events. Thalamotomy-related adverse events were categorized according to previous studies.5 This retrospective study was approved by the Severance Hospital Institutional Review Board.

Tremor Outcome

To evaluate tremor symptoms, we used the clinical rating scale for tremor (CRST) scores.13 Summing the observed and performance-based scores from CRST parts A and B determined the upper-extremity tremor subscore (CRST subscore) for the treated hand. A movement disorder specialist assessed tremors at baseline and at 1, 3, and 6 months posttreatment to identify the severity of symptoms and functional impairment. In this study, tremor outcome was defined as a reduction in tremor at 6 months posttreatment, with clinical symptom relief. Patients in the current study were divided into two groups: responders and nonresponders, based on a threshold value of 50% improvement in the CRST subscore at the 6-month follow-up.3,14,15

The mean CRST subscore of all patients at baseline was 18.2 ± 5.3. Combined resting tremor was present in 33 patients (38.8%). The patients’ tremors improved by 74.3% ± 25.5% at 6 months posttreatment. The responder group consisted of 72 patients (84.7%), and the nonresponder group consisted of 13 patients (15.3%) (Table 2). The mean 1-month and 3-month CRST subscore in the nonresponders tended to be higher than in the responders, but there was no statistical significance (p > 0.05). All patients enrolled in the present study experienced tremor relief right after the procedures.

TABLE 2.

Comparison between responders and nonresponders

Responders, n = 72Nonresponders, n = 13p Value
Age, yrs65.1 ± 8.266.5 ± 12.00.6044
Sex
 Male57 (79.2%)10 (76.9%)>0.99
 Female15 (20.8%)3 (23.1%)
Hypertension14 (19.4%)3 (23.1%)0.846
Diabetes mellitus8 (11.1%)1 (7.7%)0.668
Disease duration, yrs14.2 ± 10.611.7 ± 10.60.419
Baseline CRST subscore, treated18.0 ± 5.319.6 ± 5.30.310
Combined resting tremor28 (38.9%)5 (38.5%)0.852
Baseline CRST subscore, untreated11.4 ± 5.611.1 ± 7.90.880
1-mo CRST subscore, treated2.6 ± 3.12.0 ± 2.80.165
1-mo tremor improvement in %, treated86.9 ± 14.384.6 ± 21.80.102
3-mo CRST subscore, treated2.9 ± 4.04.5 ± 3.50.075
3-mo tremor improvement in %, treated85.5 ± 19.065.4 ± 27.20.085
6-mo CRST subscore, treated3.5 ± 3.913.6 ±4.6<0.0001
6-mo tremor improvement in %, treated82.7 ± 17.830.5 ± 9.4<0.0001
Transient adverse events
 Balance630.137
 Sensory300.440
 Speech210.408
 Strength100.660
Skull density ratio0.55 ± 0.080.50 ± 0.080.072
Skull vol, cm3339.3 ± 37.8324.1 ± 56.90.212
Max energy, J20,821.9 ± 8,019.719,978.6 ± 7,643.70.712
Max energy delivered, J23,645.6 ± 7,301.014,735.0 ± 11,363.20.544
Max power, W855.4 ± 172.5802.0 ± 144.30.801
Max temperature, °C57.0 ± 2.057.9 ± 1.60.081
No. of sonications14.1 ± 2.114.4 ± 3.30.782
No. of therapeutic sonications7.6 ± 2.08.0 ± 1.40.653
Length of AC-PC, mm24.9 ± 0.924.2 ± 1.90.329
1/2 length of 3rd ventricle, mm3.3 ± 0.44.5 ± 0.90.177
Lateral to midline, mm14.5 ± 0.314.6 ± 1.20.429
Ant to PC as ratio of AC-PC length0.30 ± 0.100.31 ± 0.070.690
Sup to ICL, mm0.20 ± 0.490.08 ± 0.970.413
Vol of zone I, cm30.014 ± 0.0040.011 ± 0.0030.537
Vol of zone II, cm30.071 ± 0.0290.053 ± 0.0340.537
Vol of zone III, cm30.149 ± 0.0560.197 ± 0.0800.662
Total vol, cm30.233 ± 0.0840.260 ± 0.120>0.99
Diam of zone I, mm2.6 ± 0.32.2 ± 0.30.177
Diam of zone I & II, mm5.8 ± 0.55.0 ± 1.20.429
Diam of zone I–III, mm8.3 ± 1.18.5 ± 0.40.662

Descriptive statistics are presented as means ± SDs. Boldface type indicates statistical significance.

Statistically significant differences were detected between the two groups with respect to the 6-month CRST subscore and tremor improvement. Tremors in the nonresponders showed recurrence of < 50% improvement between 3 and 6 months. There were no significant differences between the two groups in terms of demographics, presence of resting tremor, and treatment parameters (Table 2). No serious and permanent complication occurred following thalamotomy. Transient complications causing a minor inconvenience that did not affect daily activities reappeared within 3 months after procedures in 16 patients (18.8%) (Table 1). There was no significant difference in the occurrence of adverse events between the responders and the nonresponders (Table 2).

Location of Necrotic Core and Volumetric Analysis

Standard stereotactic coordinates for MRgFUS were approximately 10–11 mm lateral to the lateral wall of the third ventricle and approximately 26%–28.5% anterior to the posterior commissure in the intercommissural plane, slightly more anterior than the target of deep brain stimulation in our center. The vertical coordinate started 2 mm above the intercommissural line (ICL) and lowered while monitoring patients’ responses. The location of the necrotic core was determined on the immediate posttreatment T2-weighted MRI with Leksell SurgiPlan (Elekta) (Fig. 1C left). The coordinates of the center of the necrotic core were measured relative to the midcommissural point (MCP) as the reference point (X = 0, Y = 0, Z = 0). Positive Z values indicate coordinates superior to the MCP, and negative Z values indicate coordinates inferior to the MCP (Table 1). There was no significant difference in the location of the necrotic core between the two groups (Table 2). The volumetric analysis with Leksell GammaPlan (Elekta) on the immediate posttreatment MRI was divided into the following 3 volumes as described by Wintermark et al.12 (Fig. 1C right): zone I (necrotic core); zone II (cytotoxic edema); and zone III (vasogenic edema). The volumetric differences at those zones between groups were not statistically significant (Table 2).

FIG. 1.
FIG. 1.

Motor-tremor subnetwork (A) and lesion locations on T2-weighted images in 3 patients (B). A: The tremor-related network comprises 90 brain regions, including the motor-related frontal and parietal lobe, globus pallidum, putamen, ventrolateral (VL) and non-VL thalamus, red nucleus, subthalamic nucleus, substantia nigra, and cerebellum at both left and right hemispheres. Only the left hemispheric regions are shown. B: The target lesion is in the left VIM of the thalamus (arrows). C: Location of necrotic core and volumetric analysis (left). Location of the necrotic core relative to the ICL (right). Blue, green, and red lines represent the border of zone I (necrotic core), zone II (cytotoxic edema), and zone III (vasogenic edema), respectively. D: Procedures of functional network analyses in the tremor-related network. The time series at each node of a tremor-related brain atlas in the preoperative rsfMR image was used to compose tremor-related FC matrices for each subject using Pearson correlation coefficients among pairs of the nodes. The graph-ICA extracted IC subnetworks from the tremor-related subnetworks of all individuals. Statistical evaluations were performed on the involvement of each of the IC subnetworks between groups. Figure is available in color online only.

Data Acquisition and Processing

Resting-state fMRI data were acquired axially using T2*-weighted single-shot echo planar imaging sequences using a 3.0-T GE 750 MRI scanner with a 16-channel Head Neck Spine array coil with the following parameters: voxel size 1.88 × 1.88 × 4.5 mm3; slice number 30 (interleaved); matrix 128 × 128; slice thickness 4 mm; TR 2000 msec; TE 30 msec; flip angle 90°; and field of view 240 × 240 mm2. Two hundred fMRI images for 400 seconds were scanned while patients were instructed to keep their eyes closed without sleeping or thinking. A high-resolution structural MRI was also obtained using a 3D T1-BRAVO sequence with the following parameters: voxel size 0.43 × 0.43 × 1 mm3; matrix size 516 × 516; field of view 220 × 220 mm2; TR 8.59 msec; TE 3.32 msec; flip angle 12°.

Image processing and network analysis were conducted using SPM12 (http://www.fil.ion.ucl.ac.uk/spm/) and in-house software for multimodal network analysis (http://neuroimage.yonsei.ac.kr/mnet). All fMRI data underwent standard preprocessing steps, including correction of acquisition time delays between different slices, correction for head motion by realigning all consecutive volumes to the first image of the session, and nonlinear coregistration of the T1-weighted image to the first fMRI data. A coregistered T1-weighted image was used to spatially normalize fMRI into the Montreal Neurological Institute template space by using nonlinear transformation in SPM12. All registration steps were visually confirmed and semiautomatically adjusted in the case of misregistration.

The Brainnetome atlas16 was used to parcellate the cortex. To supplement the cerebellar regions closely related to motor function, we used the SUIT cerebellar lobule atlas (http://www.diedrichsenlab.org/imaging/suit.htm)17 containing 34 cerebellar regions. fMRI time series in a total of 280 whole brain regions were extracted from the normalized fMRI data in the Montreal Neurological Institute template space. The time series of eigenvalues corresponding to the first eigenvector—that is, the mode—of the time series for multiple voxels in each region (extracted using principal component analysis [PCA]) was used as a representative activity of the region rather than the average of the voxels.18 After discarding the first 5 scans to address stability issues, we preprocessed the fMRI time series by regressing out the effects of 6 rigid motions and their derivatives, and 3 principal components of the white matter and the CSF masks.19 fMRI time series were again detrended by linear and quadratic repressors, followed by high-pass filtering up to 0.009 Hz. Comparison of group differences in motion artifacts by using framewise displacement19 revealed no significant group differences (F[1,2] = 0.52, p = 0.5989).

Construction of Functional Networks for the Whole Brain and Tremor-Related Motor Networks

Among 280 nodes in the Brainnetome atlas, we chose 90 brain regions as the tremor-related subnetwork according to Buijink et al.20 based on the action category in the behavioral domain identified as motor areas in the atlas. The motor-tremor subnetwork includes regions of the frontal and parietal lobe with motor-related function, globus pallidum, putamen, thalamus, red nucleus, subthalamic nucleus, substantia nigra, and cerebellum (Fig. 1). A list of detailed regions is shown in Supplemental Table 1. Note that the names of the cortical regions in the Brainnetome atlas follow those of the Brodmann atlas (e.g., A4 corresponds to Brodmann area 4). Functional networks were constructed using Pearson correlation coefficients of time series from all pairs of 90 regions.21 The FC matrix for each individual was created by converting the correlation coefficients to Z values using Fisher’s z-transformation. All procedures are summarized in Fig. 1.

Graph-ICA

Graph-ICA was applied to the tremor-related subnetwork FC matrices of all patients. The optimal number of ICs constituting the functional network was estimated using Laplace PCA, a default method in the FSL toolbox (https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/). The process of extracting the number of ICs through FastICA-ICASSO22 was performed for 100 iterations. ICs with a quality index of less than 0.8 were removed.

Using 1-way ANOVA, we compared the weights of the ICs extracted from graph-ICA after z-transformation. The statistical significance was adjusted for multiple comparisons using the Bonferroni correction with the number of reliable ICs. In the post hoc analysis of ANOVA, we used Tukey’s honestly significant difference (HSD) test.

Results

Graph-ICA of the resting-state functional network among the tremor-related brain regions identified 22 ICs according to Laplace PCA. Among them, only 12 ICs satisfied the reliability criteria of 0.8 (Fig. 2). Based on this result, statistical significance for the ANOVA after Bonferroni correction was set to p < 0.05/12 (p = 0.0041). Node names of the 12 ICs are presented in Supplemental Table 2. It should be noted that all the motor-related subnetworks were decomposed from the resting-state data without any voluntary movements. ANOVA of weights of the ICs, representing contributions or usages of a subnetwork to the entire network, showed statistical significance in the 6 ICs, as presented in Fig. 3. A significant reduction in the ET-R compared to the NC group was found in the subnetwork (IC 5) between the areas that are related to the movement in the head, face, and upper and lower limbs, constituting the frontal and parietal lobes (F[2,100] = 7.31, p = 0.0011). ET-R showed reduced IC weight compared to NC (NC 0.12 ± 0.25, ET-R −0.03 ± 0.13, p = 0.0007), but NC was not significantly different from ET-NR.

FIG. 2.
FIG. 2.

Tremor-related subnetworks identified by graph-ICA. Edges of each group IC were thresholded by Z = 3. The line width indicates the weight of the edge. The size of a sphere is proportional to the node degree (the number of connections) at the node. A1/2/3tonla = area 1/2/3 (tongue and larynx region); A1/2/3ulhf = area 1/2/3 (upper-limb, head, and face region); A2 = area 2; A4 = area 4; A4tl = area 4 tongue and larynx region; A6 = area 6; A6cdl = caudal area 6; A6vl = ventrolateral area 6; A7 = area 7; A40c = caudal area 40; A40rv = rostroventral area 40; CBL = cerebellum; CBL.X = cerebellum area X; PUT = putamen; RN = red nucleus; SNg = substantia nigra; Tha = thalamus. Figure is available in color online only.

FIG. 3.
FIG. 3.

Differential involvement of specific motor-related subnetworks among groups. The ICs that showed significant differences in the statistical results are shown. The 3D map indicates IC subnetworks (thresholded by Z = 3). In the circular map of each subnetwork, the colors of edges represent the weight of connectivity in the IC subnetworks. The group-level comparisons of IC weights of functional motor-related subnetworks using ANOVA. Adjusted p value summary: *p < 0.05, **p < 0.01, ***p < 0.005, ****p < 0.0005. ns = not significant. Figure is available in color online only.

The subnetwork (IC 7), which covers the motor area from the bilateral precentral gyrus for the upper-limb parts, shows a significant reduction (F[2,100] = 6.79, p = 0.0017) in the ET-R group compared to the NC group (NC 0.11 ± 0.13; ET-R −0.02 ± 0.14; p = 0.0017) and ET-NR (ET-NR −0.03 ± 0.14, p = 0.0149).

The subnetwork (IC 9) that links the cerebellum to the basal ganglia region, including the thalamus, differed among all groups (F[2,100] = 9.88, p = 0.0001). Reduced IC weights in the ET-R (p = 0.0046) and ET-NR (p < 0.0001) were found (NC 0.12 ± 0.12, ET-R −0.01 ± 0.17, ET-NR −0.13 ± 0.11) compared to the NC group, and the ET-NR group had lower weights than the ET-R group (p = 0.0438).

Both groups of patients with ET showed reduced weights for the subnetwork (IC 10) (F[2,100] = 8.21, p = 0.0005). This subnetwork component consists of the connectivity of the subthalamic area that acts as a hub between most of the target basal ganglia area, the cerebellum, and some motor areas. Reduced IC weights of the ET-R group compared to NC were found (NC 0.10 ± 0.09, ET-R −0.02 ± 0.13, p = 0.0003). ET-NR also showed reduced weights for IC 10 compared to NC (ET-NR −0.01 ± 0.13, p = 0.0378).

Significant group differences were found between all 3 groups in the weight of the IC 11 consisting of Brodmann area 6, which connects the surrounding motor areas (IC 11) (F[2,100] = 29.13, p < 0.0001). ET-R had reduced IC weights compared to NC (NC 0.20 ± 0.16, ET-R −0.06 ± 0.13, p < 0.0001), but ET-NR showed an increased weight compared to ET-R (ET-NR 0.06 ± 0.11, p = 0.0101). It was found that ET-NR had lower weights than the NC group (p = 0.0119).

The subnetwork (IC 12) was mainly based on the interactions within the cerebellum, and showed reduced weights in both ET groups compared to the NC group (F[2,100] = 9.66, p = 0.0001). Post hoc comparison results showed a significant reduction in ET-R (NC 0.14 ± 0.18, ET-R −0.03 ± 0.14, p = 0.0001) and ET-NR (ET-NR −0.04 ± 0.17, p = 0.0048) compared to NC. Table 3 summarizes all these results.

TABLE 3.

Statistical results of IC weights

IC No.Names of Network Nodes (relative node degree)1-Way ANOVA, F(2,100)Study GroupPost Hoc (Tukey’s HSD)
NCET-RET-NR
IC 5A4hf.R(1), A1/2/3ulhf.R(0.76), A4ll.R(0.67), A4ul.R(0.62), A4ll.L(0.62), A2.L(0.62)F = 7.31

p = 0.0011
0.12 (0.25)−0.03 (0.13)0.00 (0.13)NC > ET-R: p = 0.0007
IC 7A4tl.R(1), A40rv.R(0.83), A40rv.L(0.71), A1/2/3tonla.R(0.58), A1/2/3tonla.L(0.42)F = 6.79

p = 0.0017
0.11 (0.13)−0.02 (0.14)−0.03 (0.14)NC > ET-R: p = 0.0017

NC > ET-NR: p = 0.0149
IC 9Left.VIIIa(1), Vermis.VIIb(1), SNg.L(0.91), mPMtha.R(0.82), Right.Crus.II(0.82), Left.VIIIb(0.82)F = 9.88

p = 0.0001
0.12 (0.12)−0.01 (0.17)−0.13 (0.11)NC > ET-R: p = 0.0046

NC > ET-NR: p < 0.0001

ET-R > ET-NR: p = 0.0438
IC 10RN.L(1), SNg.L(0.17), RN.R(0.11), Stha.L(0.09), mPMtha.L(0.07), PPtha.R(0.07), lPFtha.L(0.07)F = 8.21

p = 0.0005
0.10 (0.09)−0.02 (0.13)−0.01 (0.13)NC > ET-R: p = 0.0003

NC > ET-NR: p < 0.0378
IC 11A6vl.R(1), A6dl.R(0.61), A6vl.L(0.54), A6dl.L(0.5), A6cdl.R(0.43), A6cvl.R(0.43)F = 29.13

p < 0.0001
0.20 (0.16)−0.06 (0.13)0.06 (0.11)NC > ET-R: p < 0.0001

NC > ET-NR: p = 0.0119

ET-R < ET-NR: p = 0.0101
IC 12Left.VIIIb(1), Left.IX(1), Vermis(1), Right.VIIIb(0.95), Left.X(0.67), Left.VIIIa(0.67)F = 9.66

p = 0.0001
0.14 (0.18)−0.03 (0.14)−0.04 (0.17)NC > ET-R: p = 0.0001

NC > ET-NR: p = 0.0048

A1/2/3/tonla = area 1/2/3 (tongue and larynx region); A1/2/3ulhf = area 1/2/3 (upper-limb, head, and face region); A2 = area 2; A4hf = area 4 (head and face region); A4ll = area 4 (lower-limb region); A4tl = area 4 (tongue and larynx region); A4ul = area 4 (upper-limb region); A6cdl = caudal dorsolateral area 6; A6cvl = caudal ventrolateral area 6; A6dl = dorsolateral area 6; A6vl = ventrolateral area 6; A40rv = rostroventral area 40; HSD = honestly significant difference; lPFtha = lateral prefrontal thalamus; mPMtha = premotor thalamus; PPtha = posterior thalamus; RN = red nucleus; SNg = substantia nigra; Stha = sensory thalamus.

Values in the NC, ET-R, and ET-NR columns are expressed as the mean (SD).

Discussion

We investigated the macroscopic neural circuitry differences between thalamotomy-effective and thalamotomy-ineffective groups before the treatment of ET using a minimally invasive MRgFUS technique. Despite the efficacy of MRgFUS thalamotomy of the VIM of the thalamus for medication-refractory ET,3,14,23,24 some failures have also been reported.10 The factors of failure have been primarily researched in terms of insufficient or inadequate thermal lesion of the target. However, to the best of our knowledge, no direct evidence has been reported for the neurobiological factors concerning the ineffectiveness or recurrence after thalamotomy, which was up to 23%.3,25,26

Previous studies on the neural mechanism of ET revealed the importance of motor-related cortical-subcortical-cerebellar circuitry, including the motor cortex,27 thalamus,28 red nucleus,29 and cerebellum.30 These regions may work as a system in fMRI studies of ET, implying altered cerebello-thalamo-cortical signaling.20 Accordingly, we noticed the reduction of the usage of the 6 tremor-related subnetworks in patients in the ET (regardless of treatment outcomes) compared to the control group, which is generally in line with previous studies.31,32 In a previous study, patients with ET had resting-state FC alterations in motor areas such as the premotor and supplementary motor areas and the somatosensory cortex, and showed differences in connectivity between the thalamus and cerebellum and networks within the cerebellum.31 In another study, ET had an altered subnetwork composed of the thalamus, red nucleus, and cerebellum.32 Given that the subnetworks in this study were derived from rsfMRI, these subnetwork components are intrinsic, implying an altered baseline circuit for motor executions.

Because the ablation target was the VIM of the thalamus, we found alterations in the usage of subnetworks with the thalamus as hubs (IC 9 and IC 10). The connection between the cerebellum and thalamus (IC 9 and IC 10) is located before the VIM in the pathway from the cerebellum to the motor cortex. Thus, the treatment effect at the VIM may be explained by the networked nature of the brain. As a node participating in the entire motor network, ablation of the VIM of the thalamus appeared to have an effect at the circuitry level of the motor network, particularly the cerebellum-thalamus connection.33 Note that IC 3 contains a hub at the thalamus. Nevertheless, IC 3 was mainly associated with the orofacial part of the precentral gyrus and did not differ between the groups. Significantly reduced usage weight was found in IC 7 and IC 11 of ET, which are mainly associated with connectivity within the sensory-motor area.

The FC pattern before MRgFUS thalamotomy was associated with this motor symptom level months after thalamotomy. There was no significant difference in the baseline CRST subscore and the maximal temperature and energy delivered during treatment between patients in the thalamotomy-effective and -noneffective groups; however, there was a considerable difference in the CRST subscore 6 months after thalamotomy. At the pretreatment time, patients in the thalamotomy-noneffective (i.e., nonresponder) group have differential network usage from those in the thalamotomy-effective group in the two tremor-related subnetworks. First, patients in the thalamotomy-noneffective group have suppressed functional subnetworks in the cerebellum and subcortex regions, including the thalamus and subthalamic areas (IC 9). Second, patients in the thalamotomy-noneffective group have enhanced functional subnetworks among motor-sensory cortical networks (IC 11) compared to those in the thalamotomy-effective group. Both subnetworks are less used at rest in patients with ET regardless of thalamotomy-effective and -noneffective compared to control groups (Fig. 3).

The more suppressed functional subnetwork between the cerebellum and subcortical regions in the patients in the thalamotomy-noneffective group indicates their more severe cerebellar pathology, previously described in other studies.34,35 In ET, the reduction of Purkinje cells in the cerebellum and their morphological changes have been reported.36 In addition, the number of γ-aminobutyric acid receptors in the dentate nucleus is reduced, resulting in disinhibition of cerebellar output activity.37 Neurodegeneration in the cerebellum may be indicative of disease progression in ET.

Discussing a possible mechanism of enhanced functional subnetwork among motor-sensory cortical networks in patients in the thalamotomy-noneffective group compared with the thalamotomy-effective group, such as a loss of inhibition due to cerebellar pathology leading to increased motor output is speculative. Gallea and colleagues38 reported the structural and functional changes in the supplementary motor area in patients with ET, suggesting that these changes could be a direct consequence of cerebellar defects in ET. The secondary motor areas are among the multiple targets of the cerebellum, which are involved in the oscillatory network through the corticospinal tract or their connection to the primary motor cortex.39,40 Restuccia and colleagues41 revealed the abnormal gating of somatosensory inputs in ET in the primary somatosensory cortex. Increased FC among the somatosensory cortex in the present study might be a compensatory mechanism for less effective connectivity with the cerebellum. Tremor severity might be associated with functional changes in the motor-sensory cortical network.

Previous studies have reported the FC differences in patients with ET with normal control by rsfMRI,42 but the present study went a step further and compared patients with ET in whom treatment (MRgFUS thalamotomy) was effective and noneffective. The current study has significance and novelty in that it revealed that patients in the thalamotomy-noneffective group showed significantly decreased functional subnetworks between the cerebellum and the thalamus and subthalamic area (IC 9) and significantly increased functional subnetworks among motor-sensory cortical networks (IC 11) compared to those in the thalamotomy-effective group. These findings suggest that changes in the intrinsic usage of the cerebello-thalamo-cortical pathway could be correlated with the progression of ET. Therefore, FC patterns can be a possible predictive factor for outcomes of MRgFUS thalamotomy.

The graph-ICA used in this study is a novel multivariate method to decompose subnetworks from the cross-sectional brain networks (graphs).9 Graph-ICA assumes that individual brain networks are linearly weighted mixtures of independent subnetworks. The method searches the inverse of the weight matrix to make the estimated subnetworks independent across them. The previous study9 has shown graph-ICA to successfully identify the underlying subcomponents for a set of graphs from both simulation and real-world human data. In the current study, graph-ICA segregated subnetworks composed of motor-related brain areas from the resting-state networks. However, the independent subnetworks derived from graph-ICA may not necessarily correspond to actual brain subnetworks. Among the decomposed subnetworks, we visually identified tremor-related subnetworks. Thus, the current results should be interpreted in terms of the principle and assumption of graph-ICA.

The current study has several limitations. We only explored intrinsic functional network differences without evaluating FC during task performance or tremor movement. Although a tight relationship between task and resting-state connectivity has been reported, the prognostic factor for thalamotomy should be further researched during motor performance. Although the current study primarily focused on the differential FC between thalamotomy-effective and -noneffective groups, there were significant differences in the mean age and sex distribution between the patients with ET and controls due to the nature of HFS (control group), which occurs most often in middle-aged women. Alterations in connectivity have recently been reported in HFS. Compared to healthy controls, HFS has increased resting-state FC of the anterior cerebellum with the secondary and associative visual cortex, but decreased FC with the orbitofrontal and subgenual area and dorsal entorhinal cortex.43 Structural connectivity analysis shows that HFS increased structural connectivity in temporal and parietal cortices and decreased connectivity in the frontal cortex.44 All these altered connectivity findings, however, are not directly associated with the motor pathways that we focused on in the current study. No significant difference was reported in static and dynamic low-frequency fluctuations between patients with HFS and healthy controls.45 Nevertheless, to interpret the ET-specific circuitry differences, healthy controls with age and sex matching are needed. Due to the high efficacy of thalamotomy treatment, we have only a small number of nonresponders. Because of the small sample size and high variability of ET, we did not apply machine learning to predict the outcome. Finally, the current result does not specify whether the current finding of differential usage of subnetworks is the cause or effect of thalamotomy ineffectiveness. Thus, more data would be necessary to use the pretreatment functional organization to predict the clinical outcome of thalamotomy in patients with ET.

Conclusions

We demonstrated that patients in the thalamotomy-noneffective group have suppressed functional subnetworks in the cerebellum and subcortex regions and have enhanced functional subnetworks among motor-sensory cortical networks compared to those in the thalamotomy-effective group. These findings support the hypothesis that patients in whom thalamotomy is noneffective have more severe cerebellar pathology and change the motor-sensory-cortical network, which is compensatory for less effective connectivity with the cerebellum. This study suggests that the FC pattern is a possible predictive factor for outcomes of MRgFUS thalamotomy. We expect that our study can make a significant contribution to patient selection, improve the efficacy and efficiency of MRgFUS thalamotomy for ET, and may provide more clues to understanding the pathogenesis of ET.

Acknowledgments

We thank Eun Jung Kweon and Maeng Keun Oh for their help in the data acquisition and data processing. This research was supported by the Brain Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT (grant no. NRF-2017M3C7A1049051 to Dr. Park).

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Park, JW Chang. Acquisition of data: Kim, WS Chang, Jung, KW Chang. Analysis and interpretation of data: Pae, Kim, Eo, Park. Drafting the article: Pae, Kim, Park. Critically revising the article: Park, JW Chang. Reviewed submitted version of manuscript: JW Chang, Pae, Kim, Park. Approved the final version of the manuscript on behalf of all authors: JW Chang. Statistical analysis: Pae. Administrative/technical/material support: JW Chang, Pae, WS Chang, Jung, KW Chang, Park. Study supervision: JW Chang, Park.

Supplemental Information

Online-Only Content

Supplemental material is available with the online version of the article.

References

  • 1

    Louis ED, Ottman R, Hauser WA. How common is the most common adult movement disorder? estimates of the prevalence of essential tremor throughout the world. Mov Disord. 1998;13(1):510.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Zappia M, Albanese A, Bruno E, et al. Treatment of essential tremor: a systematic review of evidence and recommendations from the Italian Movement Disorders Association. J Neurol. 2013;260(3):714740.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    Elias WJ, Lipsman N, Ondo WG, et al. A randomized trial of focused ultrasound thalamotomy for essential tremor. N Engl J Med. 2016;375(8):730739.

  • 4

    Krishna V, Sammartino F, Cosgrove R, et al. Predictors of outcomes after focused ultrasound thalamotomy. Neurosurgery. 2020;87(2):229237.

  • 5

    Kim MJ, Park SH, Chang KW, et al. Technical and operative factors affecting magnetic resonance imaging-guided focused ultrasound thalamotomy for essential tremor: experience from 250 treatments. J Neurosurg. 2021;135(6):17801788.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6

    Biswal B, Yetkin FZ, Haughton VM, Hyde JS. Functional connectivity in the motor cortex of resting human brain using echo-planar MRI. Magn Reson Med. 1995;34(4):537541.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Noble S, Scheinost D, Constable RT. A decade of test-retest reliability of functional connectivity: A systematic review and meta-analysis. Neuroimage. 2019;203:116157.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    Yoo K, Rosenberg MD, Noble S, Scheinost D, Constable RT, Chun MM. Multivariate approaches improve the reliability and validity of functional connectivity and prediction of individual behaviors. Neuroimage. 2019;197:212223.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9

    Park B, Kim DS, Park HJ. Graph independent component analysis reveals repertoires of intrinsic network components in the human brain. PLoS One. 2014;9(1):e82873.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10

    Chang WS, Jung HH, Zadicario E, et al. Factors associated with successful magnetic resonance-guided focused ultrasound treatment: efficiency of acoustic energy delivery through the skull. J Neurosurg. 2016;124(2):411416.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11

    Yaltho TC, Jankovic J. The many faces of hemifacial spasm: differential diagnosis of unilateral facial spasms. Mov Disord. 2011;26(9):15821592.

  • 12

    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(5):891896.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13

    Stacy MA, Elble RJ, Ondo WG, Wu SC, Hulihan J. Assessment of interrater and intrarater reliability of the Fahn-Tolosa-Marin Tremor Rating Scale in essential tremor. Mov Disord. 2007;22(6):833838.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Chang JW, Park CK, Lipsman N, et al. A prospective trial of magnetic resonance-guided focused ultrasound thalamotomy for essential tremor: results at the 2-year follow-up. Ann Neurol. 2018;83(1):107114.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15

    D’Souza M, Chen KS, Rosenberg J, et al. Impact of skull density ratio on efficacy and safety of magnetic resonance-guided focused ultrasound treatment of essential tremor. J Neurosurg. 2019;132(5):13921397.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Fan L, Li H, Zhuo J, et al. The Human Brainnetome Atlas: a new brain atlas based on connectional architecture. Cereb Cortex. 2016;26(8):35083526.

  • 17

    Diedrichsen J. A spatially unbiased atlas template of the human cerebellum. Neuroimage. 2006;33(1):127138.

  • 18

    Friston KJ, Rotshtein P, Geng JJ, Sterzer P, Henson RN. A critique of functional localisers. Neuroimage. 2006;30(4):10771087.

  • 19

    Power JD, Barnes KA, Snyder AZ, Schlaggar BL, Petersen SE. Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. Neuroimage. 2012;59(3):21422154.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Buijink AW, van der Stouwe AM, Broersma M, et al. Motor network disruption in essential tremor: a functional and effective connectivity study. Brain. 2015;138(Pt 10):29342947.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21

    Smith SM, Vidaurre D, Beckmann CF, et al. Functional connectomics from resting-state fMRI. Trends Cogn Sci. 2013;17(12):666682.

  • 22

    Himberg J, Hyvarinen A. Icasso: software for investigating the reliability of ICA estimates by clustering and visualization. In: 2003 IEEE XIII Workshop on Neural Networks for Signal Processing (IEEE Cat No03TH8718). IEEE; 2003:259268.Accessed June 3, 2022. https://ieeexplore.ieee.org/document/1318025

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23

    Halpern CH, Santini V, Lipsman N, et al. Three-year follow-up of prospective trial of focused ultrasound thalamotomy for essential tremor. Neurology. 2019;93(24):e2284e2293.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24

    Park YS, Jung NY, Na YC, Chang JW. Four-year follow-up results of magnetic resonance-guided focused ultrasound thalamotomy for essential tremor. Mov Disord. 2019;34(5):727734.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 25

    Levi V, Eleopra R, Franzini A, Romito L. Is Deep Brain Stimulation still an option for tremor recurrence after Focused Ultrasound thalamotomy? A case report. J Clin Neurosci. 2019;68:344346.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26

    Weidman EK, Kaplitt MG, Strybing K, Chazen JL. Repeat magnetic resonance imaging-guided focused ultrasound thalamotomy for recurrent essential tremor: case report and review of MRI findings. J Neurosurg. 132(1):211216.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27

    Halliday DM, Conway BA, Farmer SF, Shahani U, Russell AJ, Rosenberg JR. Coherence between low-frequency activation of the motor cortex and tremor in patients with essential tremor. Lancet. 2000;355(9210):11491153.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28

    Hua SE, Lenz FA, Zirh TA, Reich SG, Dougherty PM. Thalamic neuronal activity correlated with essential tremor. J Neurol Neurosurg Psychiatry. 1998;64(2):273276.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29

    Wills AJ, Jenkins IH, Thompson PD, Findley LJ, Brooks DJ. Red nuclear and cerebellar but no olivary activation associated with essential tremor: a positron emission tomographic study. Ann Neurol. 1994;36(4):636642.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 30

    Applegate LM, Louis ED. Essential tremor: mild olfactory dysfunction in a cerebellar disorder. Parkinsonism Relat Disord. 2005;11(6):399402.

  • 31

    Nicoletti V, Cecchi P, Pesaresi I, Frosini D, Cosottini M, Ceravolo R. Cerebello-thalamo-cortical network is intrinsically altered in essential tremor: evidence from a resting state functional MRI study. Sci Rep. 2020;10(1):16661.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 32

    Tuleasca C, Régis J, Najdenovska E, et al. Pretherapeutic resting-state fMRI profiles are associated with MR signature volumes after stereotactic radiosurgical thalamotomy for essential tremor. J Neurosurg. 2018;129(suppl 1):6371.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 33

    Jang C, Park HJ, Chang WS, Pae C, Chang JW. Immediate and longitudinal alterations of functional networks after thalamotomy in essential tremor. Front Neurol. 2016;7:184.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 34

    Hanajima R, Tsutsumi R, Shirota Y, Shimizu T, Tanaka N, Ugawa Y. Cerebellar dysfunction in essential tremor. Mov Disord. 2016;31(8):12301234.

  • 35

    Louis ED. Linking essential tremor to the cerebellum: neuropathological evidence. Cerebellum. 2016;15(3):235242.

  • 36

    Louis ED, Yi H, Erickson-Davis C, Vonsattel JP, Faust PL. Structural study of Purkinje cell axonal torpedoes in essential tremor. Neurosci Lett. 2009;450(3):287291.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 37

    Paris-Robidas S, Brochu E, Sintes M, et al. Defective dentate nucleus GABA receptors in essential tremor. Brain. 2012;135(Pt 1):105116.

  • 38

    Gallea C, Popa T, García-Lorenzo D, et al. Intrinsic signature of essential tremor in the cerebello-frontal network. Brain. 2015;138(Pt 10):29202933.

  • 39

    Dum RP, Strick PL. Frontal lobe inputs to the digit representations of the motor areas on the lateral surface of the hemisphere. J Neurosci. 2005;25(6):13751386.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 40

    Maier MA, Armand J, Kirkwood PA, Yang HW, Davis JN, Lemon RN. Differences in the corticospinal projection from primary motor cortex and supplementary motor area to macaque upper limb motoneurons: an anatomical and electrophysiological study. Cereb Cortex. 2002;12(3):281296.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 41

    Restuccia D, Valeriani M, Barba C, et al. Abnormal gating of somatosensory inputs in essential tremor. Clin Neurophysiol. 2003;114(1):120129.

  • 42

    Lan H, Suo X, Li W, et al. Abnormalities of intrinsic brain activity in essential tremor: a meta-analysis of resting-state functional imaging. Hum Brain Mapp. 2021;42(10):31563167.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 43

    Fang TC, Chen CM, Chang MH, Wu CH, Guo YJ. Altered functional connectivity and sensory processing in blepharospasm and hemifacial spasm: coexistence and difference. Front Neurol. 2021;12:759869.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 44

    Chirumamilla VC, Dresel C, Koirala N, et al. Structural brain network fingerprints of focal dystonia. Ther Adv Neurol Disord. 2019;12:1756286419880664.

  • 45

    Luo Y, Guo Y, Zhong L, et al. Abnormal dynamic brain activity and functional connectivity of primary motor cortex in blepharospasm. Eur J Neurol. 2022;29(4):10351043.

    • PubMed
    • Search Google Scholar
    • Export Citation

Supplementary Materials

  • Collapse
  • Expand
Images from Holland et al. (pp 329–336).
  • FIG. 1.

    Motor-tremor subnetwork (A) and lesion locations on T2-weighted images in 3 patients (B). A: The tremor-related network comprises 90 brain regions, including the motor-related frontal and parietal lobe, globus pallidum, putamen, ventrolateral (VL) and non-VL thalamus, red nucleus, subthalamic nucleus, substantia nigra, and cerebellum at both left and right hemispheres. Only the left hemispheric regions are shown. B: The target lesion is in the left VIM of the thalamus (arrows). C: Location of necrotic core and volumetric analysis (left). Location of the necrotic core relative to the ICL (right). Blue, green, and red lines represent the border of zone I (necrotic core), zone II (cytotoxic edema), and zone III (vasogenic edema), respectively. D: Procedures of functional network analyses in the tremor-related network. The time series at each node of a tremor-related brain atlas in the preoperative rsfMR image was used to compose tremor-related FC matrices for each subject using Pearson correlation coefficients among pairs of the nodes. The graph-ICA extracted IC subnetworks from the tremor-related subnetworks of all individuals. Statistical evaluations were performed on the involvement of each of the IC subnetworks between groups. Figure is available in color online only.

  • FIG. 2.

    Tremor-related subnetworks identified by graph-ICA. Edges of each group IC were thresholded by Z = 3. The line width indicates the weight of the edge. The size of a sphere is proportional to the node degree (the number of connections) at the node. A1/2/3tonla = area 1/2/3 (tongue and larynx region); A1/2/3ulhf = area 1/2/3 (upper-limb, head, and face region); A2 = area 2; A4 = area 4; A4tl = area 4 tongue and larynx region; A6 = area 6; A6cdl = caudal area 6; A6vl = ventrolateral area 6; A7 = area 7; A40c = caudal area 40; A40rv = rostroventral area 40; CBL = cerebellum; CBL.X = cerebellum area X; PUT = putamen; RN = red nucleus; SNg = substantia nigra; Tha = thalamus. Figure is available in color online only.

  • FIG. 3.

    Differential involvement of specific motor-related subnetworks among groups. The ICs that showed significant differences in the statistical results are shown. The 3D map indicates IC subnetworks (thresholded by Z = 3). In the circular map of each subnetwork, the colors of edges represent the weight of connectivity in the IC subnetworks. The group-level comparisons of IC weights of functional motor-related subnetworks using ANOVA. Adjusted p value summary: *p < 0.05, **p < 0.01, ***p < 0.005, ****p < 0.0005. ns = not significant. Figure is available in color online only.

  • 1

    Louis ED, Ottman R, Hauser WA. How common is the most common adult movement disorder? estimates of the prevalence of essential tremor throughout the world. Mov Disord. 1998;13(1):510.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Zappia M, Albanese A, Bruno E, et al. Treatment of essential tremor: a systematic review of evidence and recommendations from the Italian Movement Disorders Association. J Neurol. 2013;260(3):714740.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    Elias WJ, Lipsman N, Ondo WG, et al. A randomized trial of focused ultrasound thalamotomy for essential tremor. N Engl J Med. 2016;375(8):730739.

  • 4

    Krishna V, Sammartino F, Cosgrove R, et al. Predictors of outcomes after focused ultrasound thalamotomy. Neurosurgery. 2020;87(2):229237.

  • 5

    Kim MJ, Park SH, Chang KW, et al. Technical and operative factors affecting magnetic resonance imaging-guided focused ultrasound thalamotomy for essential tremor: experience from 250 treatments. J Neurosurg. 2021;135(6):17801788.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6

    Biswal B, Yetkin FZ, Haughton VM, Hyde JS. Functional connectivity in the motor cortex of resting human brain using echo-planar MRI. Magn Reson Med. 1995;34(4):537541.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Noble S, Scheinost D, Constable RT. A decade of test-retest reliability of functional connectivity: A systematic review and meta-analysis. Neuroimage. 2019;203:116157.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    Yoo K, Rosenberg MD, Noble S, Scheinost D, Constable RT, Chun MM. Multivariate approaches improve the reliability and validity of functional connectivity and prediction of individual behaviors. Neuroimage. 2019;197:212223.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9

    Park B, Kim DS, Park HJ. Graph independent component analysis reveals repertoires of intrinsic network components in the human brain. PLoS One. 2014;9(1):e82873.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10

    Chang WS, Jung HH, Zadicario E, et al. Factors associated with successful magnetic resonance-guided focused ultrasound treatment: efficiency of acoustic energy delivery through the skull. J Neurosurg. 2016;124(2):411416.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11

    Yaltho TC, Jankovic J. The many faces of hemifacial spasm: differential diagnosis of unilateral facial spasms. Mov Disord. 2011;26(9):15821592.

  • 12

    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(5):891896.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13

    Stacy MA, Elble RJ, Ondo WG, Wu SC, Hulihan J. Assessment of interrater and intrarater reliability of the Fahn-Tolosa-Marin Tremor Rating Scale in essential tremor. Mov Disord. 2007;22(6):833838.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Chang JW, Park CK, Lipsman N, et al. A prospective trial of magnetic resonance-guided focused ultrasound thalamotomy for essential tremor: results at the 2-year follow-up. Ann Neurol. 2018;83(1):107114.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15

    D’Souza M, Chen KS, Rosenberg J, et al. Impact of skull density ratio on efficacy and safety of magnetic resonance-guided focused ultrasound treatment of essential tremor. J Neurosurg. 2019;132(5):13921397.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Fan L, Li H, Zhuo J, et al. The Human Brainnetome Atlas: a new brain atlas based on connectional architecture. Cereb Cortex. 2016;26(8):35083526.

  • 17

    Diedrichsen J. A spatially unbiased atlas template of the human cerebellum. Neuroimage. 2006;33(1):127138.

  • 18

    Friston KJ, Rotshtein P, Geng JJ, Sterzer P, Henson RN. A critique of functional localisers. Neuroimage. 2006;30(4):10771087.

  • 19

    Power JD, Barnes KA, Snyder AZ, Schlaggar BL, Petersen SE. Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. Neuroimage. 2012;59(3):21422154.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Buijink AW, van der Stouwe AM, Broersma M, et al. Motor network disruption in essential tremor: a functional and effective connectivity study. Brain. 2015;138(Pt 10):29342947.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21

    Smith SM, Vidaurre D, Beckmann CF, et al. Functional connectomics from resting-state fMRI. Trends Cogn Sci. 2013;17(12):666682.

  • 22

    Himberg J, Hyvarinen A. Icasso: software for investigating the reliability of ICA estimates by clustering and visualization. In: 2003 IEEE XIII Workshop on Neural Networks for Signal Processing (IEEE Cat No03TH8718). IEEE; 2003:259268.Accessed June 3, 2022. https://ieeexplore.ieee.org/document/1318025

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23

    Halpern CH, Santini V, Lipsman N, et al. Three-year follow-up of prospective trial of focused ultrasound thalamotomy for essential tremor. Neurology. 2019;93(24):e2284e2293.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24

    Park YS, Jung NY, Na YC, Chang JW. Four-year follow-up results of magnetic resonance-guided focused ultrasound thalamotomy for essential tremor. Mov Disord. 2019;34(5):727734.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 25

    Levi V, Eleopra R, Franzini A, Romito L. Is Deep Brain Stimulation still an option for tremor recurrence after Focused Ultrasound thalamotomy? A case report. J Clin Neurosci. 2019;68:344346.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26

    Weidman EK, Kaplitt MG, Strybing K, Chazen JL. Repeat magnetic resonance imaging-guided focused ultrasound thalamotomy for recurrent essential tremor: case report and review of MRI findings. J Neurosurg. 132(1):211216.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27

    Halliday DM, Conway BA, Farmer SF, Shahani U, Russell AJ, Rosenberg JR. Coherence between low-frequency activation of the motor cortex and tremor in patients with essential tremor. Lancet. 2000;355(9210):11491153.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28

    Hua SE, Lenz FA, Zirh TA, Reich SG, Dougherty PM. Thalamic neuronal activity correlated with essential tremor. J Neurol Neurosurg Psychiatry. 1998;64(2):273276.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29

    Wills AJ, Jenkins IH, Thompson PD, Findley LJ, Brooks DJ. Red nuclear and cerebellar but no olivary activation associated with essential tremor: a positron emission tomographic study. Ann Neurol. 1994;36(4):636642.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 30

    Applegate LM, Louis ED. Essential tremor: mild olfactory dysfunction in a cerebellar disorder. Parkinsonism Relat Disord. 2005;11(6):399402.

  • 31

    Nicoletti V, Cecchi P, Pesaresi I, Frosini D, Cosottini M, Ceravolo R. Cerebello-thalamo-cortical network is intrinsically altered in essential tremor: evidence from a resting state functional MRI study. Sci Rep. 2020;10(1):16661.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 32

    Tuleasca C, Régis J, Najdenovska E, et al. Pretherapeutic resting-state fMRI profiles are associated with MR signature volumes after stereotactic radiosurgical thalamotomy for essential tremor. J Neurosurg. 2018;129(suppl 1):6371.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 33

    Jang C, Park HJ, Chang WS, Pae C, Chang JW. Immediate and longitudinal alterations of functional networks after thalamotomy in essential tremor. Front Neurol. 2016;7:184.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 34

    Hanajima R, Tsutsumi R, Shirota Y, Shimizu T, Tanaka N, Ugawa Y. Cerebellar dysfunction in essential tremor. Mov Disord. 2016;31(8):12301234.

  • 35

    Louis ED. Linking essential tremor to the cerebellum: neuropathological evidence. Cerebellum. 2016;15(3):235242.

  • 36

    Louis ED, Yi H, Erickson-Davis C, Vonsattel JP, Faust PL. Structural study of Purkinje cell axonal torpedoes in essential tremor. Neurosci Lett. 2009;450(3):287291.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 37

    Paris-Robidas S, Brochu E, Sintes M, et al. Defective dentate nucleus GABA receptors in essential tremor. Brain. 2012;135(Pt 1):105116.

  • 38

    Gallea C, Popa T, García-Lorenzo D, et al. Intrinsic signature of essential tremor in the cerebello-frontal network. Brain. 2015;138(Pt 10):29202933.

  • 39

    Dum RP, Strick PL. Frontal lobe inputs to the digit representations of the motor areas on the lateral surface of the hemisphere. J Neurosci. 2005;25(6):13751386.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 40

    Maier MA, Armand J, Kirkwood PA, Yang HW, Davis JN, Lemon RN. Differences in the corticospinal projection from primary motor cortex and supplementary motor area to macaque upper limb motoneurons: an anatomical and electrophysiological study. Cereb Cortex. 2002;12(3):281296.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 41

    Restuccia D, Valeriani M, Barba C, et al. Abnormal gating of somatosensory inputs in essential tremor. Clin Neurophysiol. 2003;114(1):120129.

  • 42

    Lan H, Suo X, Li W, et al. Abnormalities of intrinsic brain activity in essential tremor: a meta-analysis of resting-state functional imaging. Hum Brain Mapp. 2021;42(10):31563167.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 43

    Fang TC, Chen CM, Chang MH, Wu CH, Guo YJ. Altered functional connectivity and sensory processing in blepharospasm and hemifacial spasm: coexistence and difference. Front Neurol. 2021;12:759869.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 44

    Chirumamilla VC, Dresel C, Koirala N, et al. Structural brain network fingerprints of focal dystonia. Ther Adv Neurol Disord. 2019;12:1756286419880664.

  • 45

    Luo Y, Guo Y, Zhong L, et al. Abnormal dynamic brain activity and functional connectivity of primary motor cortex in blepharospasm. Eur J Neurol. 2022;29(4):10351043.

    • PubMed
    • Search Google Scholar
    • Export Citation

Metrics

All Time Past Year Past 30 Days
Abstract Views 1006 134 0
Full Text Views 554 296 24
PDF Downloads 611 313 28
EPUB Downloads 0 0 0