Bilateral Radiofrequency Ventral Intermediate Thalamotomy for Essential Tremor Free

Essential tremor (ET) is the most common involuntary movement disorder, with tremors in both upper extremities as the primary symptom [1]. Medications, such as clonazepam, propranolol, and primidone, are prescribed for ET treatment [2]. However, neurosurgical treatment is available for drug-resistant ET. Neurosurgical treatments include deep brain stimulation (DBS) and coagulation using focused ultrasound (FUS), gamma knife (GK), and radiofrequency (RF), all of which generally target the thalamic ventral intermediate (Vim) nucleus [2, 3].

Most surgeons do not recommend bilateral Vim thalamotomies in ET patients owing to concerns about the high incidence of dysarthria, dysphonia, and dysphagia [4, 5]. In such cases, DBS is the preferred treatment of choice [6, 7]. However, numerous complications of bilateral thalamotomy were reported around the 1960s, and coagulation of the ventrolateral nucleus encompassing the Vim and ventro-oral nuclei was the target. Moreover, this target was set by ventriculography in the absence of magnetic resonance imaging (MRI) or computed tomography (CT) [5, 8, 9]. The volume of the lesion and targeting manner may have contributed to high complications associated with bilateral thalamotomy. Subsequently, researchers established the concept of selective Vim thalamotomy, which selectively destroys the Vim nucleus [10, 11]. This enabled accurate targeting by MRI and CT, thus facilitating a more precise thalamotomy with reduced risk.

The recently introduced Vim thalamotomy with MR-guided FUS has proven effective for the treatment of ET of the unilateral upper extremity, thereby making ET the best indication for FUS [12]. In addition, there are few reports of FUS Vim thalamotomy being performed in a staged manner on the bilateral thalamic Vim nuclei to treat ET of the bilateral upper extremities [13‒15]. No serious complications have been reported in either case; good tremor suppression was obtained in these cases. Our previous report on the quality of life in ET following bilateral Vim thalamotomy was a novel study on the surgical technique for ET using RF with MRI targeting [16]. We performed stepwise bilateral thalamic Vim thalamotomies with RF thermal coagulation for ET of the bilateral upper extremities. In this report, we aimed to retrospectively review the results and complications of our treatment for ET in both upper extremities.

This retrospective study was approved by the Ethics Committee of our institution (No. 2021-016) which waived the requirement for informed patient consent, considering the observational study design. Informed written consent for the surgical treatment was obtained from each patient and/or their families.

We included 17 patients who underwent bilateral Vim thalamotomies with RF thermal coagulation at the Tokyo Women’s Medical University from 2015 to 2020. All eligible patients were diagnosed with ET by a neurologist or neurosurgeon specializing in involuntary movements and underwent staged Vim thalamotomy for tremors in both upper extremities. The interval between the first and second thalamotomies was 21.3 ± 14.7 months (range: 12–75 months).

A Leksell frame was placed on the patient’s head under local anesthesia, and T1-weighted axial and T2-weighted coronal MRI (1 mm slice) and CT scans (1 mm slice) were used to determine the tentative target. The surgical planning was done using Leksell SurgiPlan (Elekta AB, Stockholm, Sweden) and Brainlab Elements (Brainlab, Munich, Germany). RF thermal coagulation was performed using a monopolar RF probe (1.0 mm diameter tip with a 4.0 mm uninsulated length) and a Leksell Neuro Generator (Elekta AB, Stockholm, Sweden). Test stimulation was done using 100 µs, 133 Hz, and 1.5–3.7 mA. After confirming tremor improvement and the absence of numbness, dysarthria, and capsular response, we performed thermal coagulation at 70°C for 40 s.

The upper extremity that the patient wanted to be treated was determined to be the first treatment side. The Vim nucleus was targeted 5–7 mm anterior to the posterior commissure, 11–12 mm lateral to the third ventricular wall, and 0–1 mm superior to the anterior commissure-posterior commissure plane. We performed thermal coagulation after confirming its safety using test stimuli. The electrode was withdrawn in 3 mm increments to increase the lesion size, thus producing two contiguous lesions. Similarly, two contiguous lesions were created 3 mm antero-medially to the tentative target, which resulted in a total of four lesions (Fig. 1a, b).

The opposite side of the first thalamotomy was selected as the treatment side. The target selection was similar to that for the first thalamotomy. Thermocoagulation was performed only at the target and the site where the electrode was withdrawn in 3 mm increments, which resulted in two contiguous lesions (Fig. 1c, d).

Immediately following surgery, we performed T1- and T2-MRI and CT to confirm the presence of bleeding and the location of lesions. T1, T2, and susceptibility-weighted imaging were performed in the postoperative 3 months.

We used the Clinical Rating Scale for Tremor (CRST) part A and part B before treatment and at the final evaluation after the first and second treatments. It rates the severity of tremor from zero (none) to 4 (severe). The scale is divided into three parts. Part A (with a maximum score of 80) quantifies the tremor at rest, during posture, and during intentional maneuvers for nine body parts, including the face, tongue, voice, head, trunk, and bilateral upper and lower extremities. Part B (with a maximum score of 36) rates the action tremor of the upper limbs, particularly while writing and pouring liquids.

We evaluated the complications before treatment, after the first treatment, and after the second treatment. Moreover, we reviewed the complications for their timing of onset and severity. Complication severity was assessed using the Common Terminology Criteria for Adverse Events (CTCAE) v5.0. Moreover, we considered the number of patients without complications at the final evaluation. Online supplementary Table 1 summarizes the details of CTCAE v5.0 for adverse events (for all online suppl. material, see www.karger.com/doi/10.1159/000528825).

Different classifications of the thalamic nuclei have been proposed. In this paper, the terminologies used for referring to the thalamic nuclei are those used in the Brainlab Elements: Vim nucleus (Hassler’s classification) and ventroposterolateral (VPL)/ventroposteromedial (VPM) nucleus (Hirai/Jones’s classification). Based on the pre-treatment thin-slice T1 MRI images, we performed anatomical mapping using Brainlab Elements to delineate the thalamic Vim nucleus, VPL nucleus, VPM nucleus, and internal capsule (IC). In the postoperative T1 images, we measured the center position of the lesion and its distance from the PC. Moreover, we examined the lesion encroachment on the VPL nucleus, VPM nucleus, and IC. We measured the volume of the postoperative T1 image using a 3-D slicer.

For non-normally distributed data, we performed the Wilcoxon signed rank test to compare the preoperative CRST parts A, B, and A + B scores, with those at the evaluation of the post-first and post-second thalamotomy. All statistical analyses were performed using SPSS (version 25.0; SPSS Inc., Chicago, IL, USA). All statistical tests were two-tailed, and significance was set at p < 0.05.

Seventeen patients with ET underwent staged bilateral Vim thalamotomy. Table 1 summarizes the patient characteristics. The mean age at the onset of tremor was 36.6 ± 16.9 years. Ten patients had a significant family history of tremors. All patients were right-handed. Table 2 summarizes the patient characteristics. The CRST A + B scores significantly improved from 34.6 ± 9.7 preoperatively to 7.4 ± 6.8 after second thalamotomy, with 29.3 ± 15.1 months of follow-up (p < 0.0001, 78.6% improvement, Table 2).

The surgical side of the first thalamotomy was 16 and 1 on the left and right vim nuclei, respectively. The CRST part A + B scores significantly improved from 34.6 ± 9.7 preoperatively to 20.8 ± 7.0 at 21.3 months following the first thalamotomy (p < 0.0001, 39.9% improvement). The surgical side of the second thalamotomy was 16 and 1 on the right and left vim nuclei, respectively. The CRST part A + B scores significantly improved from 20.8 ± 7.0 before the thalamotomy to 7.4 ± 6.8 at 29.3 months following the second thalamotomy (p < 0.0001, 64.4%).

After the first thalamotomy, 7 patients suffered from eleven transient adverse events, including dysarthria, tongue/finger numbness, dysgeusia, and unsteady gait. Three patients had three prolonged adverse events, including foot weakness and mild dysarthria. All transient and prolonged adverse events were classified as CTCAE grade 1 (Table 3). After the second thalamotomy, 5 patients presented seven transient adverse events, including dysarthria, tongue numbness, and unsteady gait. Nine patients presented with 13 prolonged adverse events, including dysarthria, dysgeusia, dysphagia, tongue numbness, unsteady gait, and postural instability. We confirmed CTCAE grade 2 in 2 patients with dysgeusia and 1 patient with dysarthria. All other adverse events were classified as CTCAE grade 1 (Table 3). In 2 patients, microbleeding occurred postoperatively after right-side thalamotomy. One patient was asymptomatic, and the other developed temporary gait instability that resolved completely within 6 months. There was no surgical site infection.

Table 4 summarizes the detailed lesion locations and volumes. The mean volumes of the first thalamotomy and second thalamotomy were 50.0 ± 18.2 mm³ and 43.9 ± 28.2 mm³, respectively. The Brainlab Elements anatomical mapping tool retrospectively confirmed representative postoperative MRI images with successful lesions (Fig. 2a) and lesion encroachment on the surrounding structures, including the IC, VPM nucleus, and VPL nucleus (Fig. 2b–d). The lesion was infero-lateral to the thalamic Vim nucleus in 1 and 2 patients with left and right thalamotomies, respectively. However, these patients presented significant improvements in the tremors.

In this study, 17 patients with ET underwent staged Vim thalamotomy and demonstrated 78.6% improvement over a follow-up period of 29.3 months (the period after the second thalamotomy), as evaluated by the CRST part A + B scores. At the final evaluation, 9 patients displayed prolonged complications.

Bilateral Vim thalamotomy reportedly improves bilateral upper extremity tremor in ET [13‒19]. We reviewed articles on bilateral Vim thalamotomy for ET using the localization technique with MRI (Table 5). There are four reports of bilateral Vim thalamotomy using MRgFUS, all with a short follow-up period (3–6 months) [13‒15, 17]. Fukutome et al. [13] performed staged bilateral Vim thalamotomy using MRgFUS with a mean interval of 27.8 months in 5 patients with ET. Moreover, CRST improved from 63.6 preoperatively to 21.8 bilaterally postoperatively (65.9% improvement). Fenrandez et al. [13] performed staged bilateral Vim thalamotomy with a mean interval of 24 months in 9 patients with ET, with the CRST A + B assessment improving from 22.3 preoperatively to 10.8 bilaterally postoperatively (66.6% improvement) [15]. A study of staged Vim thalamotomy with GK evaluated midline tremors in 68 patients with ET [19]. The midline tremor improved from 8.7 points preoperatively to 3.4 points following bilateral treatment (60.9% improvement). Gallay et al. [20] reported on ET treatment with bilateral MRgFUS targeting the posterior subthalamic area (cerebellothalamic tract). Two patients with ET underwent staged bilateral cerebellothalamic tractotomy and displayed 81.3% improvement in CRST part B posttreatment. Our previous study on an improvement in the quality of life is the only report of Vim thalamotomy by RF thermocoagulation with MRI localization. There are no reports on CRST-based evaluation [16]. In the present study, 17 patients with ET underwent staged bilateral Vim thalamotomy with a mean interval of 21.3 months and displayed 78.6% improvement in CRST A + B following 29.3 months of bilateral treatment. All previous studies reported on the favorable improvement of tremor, thereby suggesting bilateral Vim thalamotomy is effective for ET treatment.

Bilateral thalamotomy was performed between 1960 and 1970, primarily as a coagulation of the thalamic ventrolateral nuclei for Parkinson’s disease. The surgeons performed ventriculography to determine the therapeutic target. There are several reports of frequent dysarthria, ranging from 23.8% to 60% [5, 8, 9, 21, 22]. Reduced voice volume, balance problems, and swallowing problems reportedly occurred in 36%, 39.6%, and 24% of the patients, respectively [8, 23]. This high complication rate resulted in the recognition of bilateral thalamus coagulation as an unrecommended treatment. The target disease, the volume of destruction, and targeting by ventriculography may have contributed to the high complication rate of pioneer thalamic coagulation procedures. Dysarthria, dysphagia, decreased voice volume, and balance problems are the frequently reported complications of bilateral thalamotomy. These complications are covered by the symptoms of PD. Almost all of the diseases for which bilateral thalamotomy had been reported between the 1960s and 80s were PD. A recent review article on speech problems associated with thalamotomy and thalamic DBS reported that speech problems in bilateral surgery occur in 42.5% and 13.9% of the patients with PD and ET, respectively [24]. PD is a risky condition for treatment by bilateral thalamotomy, whereas ET is more acceptable. In addition, MRI or CT could not confirm the volume of destruction; therefore, this volume was confirmed only at postmortem autopsy. Krayenbuhl et al. [9] reported on a lesion volume ranging from 200 mm³ to 250 mm³. They performed bilateral VL thalamotomy in 23 patients with PD and observed worsening speech, psychiatric disturbances, worsening gait, and mortality in 30.4%, 34.8%, 17.4%, and 8.7% of the cases, respectively. Hariz et al. [25, 26] reported that a lesion volume of 100 mm³ was sufficient for thalamotomy following the advent of MRI, with selective Vim coagulation. In this study, the lesion volume was <100 mm³. However, the rate of complications was higher than those reported in recent years. Reports of frequencies regarding complications are variable and difficult to interpret. The primary reason is the lack of uniformity in assessing the presence and severity of complications. Research studies may miss minor complications, and it is necessary to consider the possibility of an underestimation. Iorio-Morin et al. [14] performed the only study that provided a detailed evaluation of the complications of bilateral thalamotomy. Ten patients with ET underwent staged Vim thalamotomy with FUS. We assessed the severity of complications using the CTCAE v5.0. Using this method, complications at 3 months following bilateral surgery included 1 case each of gait impairment (grade 2) and dysarthria (grade 1), 2 cases each of dysphagia (grade 1: 1, grade 2: 1) and dysesthesia (grade 1: 2), and three cases of dysgeusia (grade 1: 2, grade 2: 1). Grade 1 and 2 complications at the 3-month evaluation were observed in 58.3% and 41.7% of the patients, respectively. However, this study had a short follow-up period (3 months postoperatively), and the aforementioned complications may improve with time. We identified only grade 2 complications at the final evaluation in 2 of 4 patients with dysgeusia; whereas, the other had grade 1 complications. There are no reports on serious complications of bilateral thalamotomy, which will likely be of an acceptable severity even with the use of RF.

The targeting manner of bilateral thalamic Vim nuclei may reduce postoperative complications. There are reports of both asymmetrical and symmetrical targeting of the bilateral Vim thalamotomy. Fukutome et al., Iorio-Morin et al., and Bruno et al. set the position of the second lesion dorsal to the first lesion (asymmetrical targeting) [13, 14, 17]. We set the second lesion at an identical position as the first lesion (symmetrical targeting) but halved the target coagulation volume for lowering the complications of bilateral thalamic Vim thalamotomy. However, the complication rate in our report was higher than that in previous reports. Reducing the volume of the second lesion was not beneficial. By contrast, Fukutome et al. and Iorio-Morin et al. reported a lower incidence of more permanent complications for an asymmetrical target. Fukutome et al. [13] mentioned that only one in 5 patients had mild dysarthria at the final evaluation; disorders of gait, balance, swallowing, or speech were absent. Iorio-Morin et al. [14] observed one case each of gait disturbance and dysarthria and 2 cases of dysphagia at the final evaluation; however, they selected a short follow-up period (3 months), and these symptoms may improve upon additional follow-up. Our previous report suggested the possibility of using different targets during bilateral coagulation to reduce complications in the treatment of dystonia. Bilateral pallidotomy resulted in the occurrence of Parkinsonism, such as gait and balance deficits in 25% of the patients [27], whereas unilateral pallidotomy and contralateral pallidothalamic tractotomy did not cause gait or balance deficits [28]. Krayenbuhl et al. [9] reported concerns about destroying the similar structure bilaterally, quoting Cooper as follows: “I feel like hesitating to sacrifice the same structure bilaterally, no matter what that structure may be.” Other treatment targets besides the Vim nucleus in ET include the cerebellothalamic tract, as reported by Gallay et al. [20]. The ablation of the unilateral Vim and contralateral CTT may supposedly result in less complicated treatment.

Complications associated with bilateral thalamotomy may develop later. In the present study, 35% of the complications following bilateral thalamotomy developed 1 month post-surgery. No complications developed after 1 month or later, following surgery on the first side. Few studies have examined the timing of the onset of complications; regarding speech problems, Iorio-Morin et al. [14] reported that patient self-reports did not reveal apparent changes at 2 h posttreatment, a decrease at 1 month, and an improvement to pre-treatment level at 3 months [14]. One patient reported a slurring of speech after 3 months. Two patients reported gait disturbance following 1 month of treatment and reported an improvement at 3 months [14]. We performed a test stimulation and evaluation of dysarthria and numbness before coagulation to ensure the absence of these symptoms. Nevertheless, 18% of the patients developed dysarthria, and 50% displayed some complications at the final evaluation. The absence of dysarthria or numbness caused by an intraoperative test stimulation is not a perfect predictor of its occurrence as a postoperative complication. Postural instability developed in the present study in the first month following the second surgery to an extent that caused an easy loss of balance with the posterior traction by pull test. However, it partially decreased in the third month following surgery and subsequently improved to a level that did not interfere with daily activities 1 year later. In our previous study on bilateral pallidotomy for dystonia, all gait and balance problems occurred approximately 1 month following the second surgery [27]. Gait and balance impairments can significantly impair daily activities; nonetheless, it is difficult to confirm these impairments with an intraoperative test stimulation. This is equally difficult in FUS-based treatment. Notably, complications following bilateral thalamotomy may develop later and require a careful follow-up.

Vim thalamotomy for tremors can be performed using a variety of treatment modalities, including RF, GK, laser interstitial thermal therapy (LITT), MRgFUS, and DBS. In terms of bilateral thalamotomy for ET, RF, DBS, MRgFUS, and GK have been reported to be available for ET treatment. DBS is a safe treatment modality that can improve tremors without an irreversible lesion if the patient is receptive to device implantation. DBS also has the advantage of allowing treatment to be completed in a single procedure. LITT has recently been reported to be effective for Vim thalamotomy for tremors, including ET [29]. The safety and efficacy of bilateral thalamotomy using LITT are not well established. Since the procedure is performed under general anesthesia, the safety and efficacy cannot be confirmed by a physiological response by stimulation before lesioning [29]. GK use has the same concern as that of LITT because of the inability to confirm safety and efficacy by physiological response before lesioning. Additionally, unpredictable radiation complication is another concern [30]. Although treatment of tremor by bilateral GK thalamotomy has been reported, its safety profile (complication rate: 2.9%) raises questions about whether robust clinical evaluation has been conducted [19]. GK is the only treatment that can be used for patients with bleeding tendencies; therefore, it is indicated for patients who cannot discontinue antiplatelet agents and for patients on dialysis. We believe that bilateral GKT should be indicated only for patients with bleeding tendencies, such as those on antiplatelet/anticoagulant agents or hemodialysis, and those in whom RF, DBS, LITT, and MRgFUS are not indicated. MRgFUS has been reported in the past several years for bilateral Vim thalamotomy for ET without the risk of complications related to surgical procedures, such as bleeding, which can happen in DBS, RF, and LITT. In addition, the area of the lesion can be visualized by MR thermometry. We believe that MRgFUS is the most preferred treatment technique for bilateral Vim thalamotomy for ET. RF is a treatment that is mostly available in developing countries due to the low cost of medical supplies. We consider that the safety and efficacy confirmed in this study are acceptable. In patients who are reluctant to undergo device implantation and whose bone density is not an indication for MRgFUS, RF may be a viable treatment modality.

This study had several limitations. In the retrospective analysis, subtle adverse events were likely overlooked. Due to the open-label nature of this study, the possible influence of rater bias cannot be ruled out. The major concerns in this study are the small number of cohorts and the short-term follow-up. Long-term efficacy could not be concluded in this study. Lastly, due to the lack of evaluations of cognitive and mood states, the safety profile of psychiatric or behavioral function is unknown.

In conclusion, staged thalamic Vim thalamotomy with RF thermal coagulation was effective for ET treatment in both upper extremities. Despite the majority of possible complications being minor, additional studies with a larger sample size are required to ensure patient safety.

This study was approved by the Ethics Committee of Tokyo Women’s Medical University (No. 2021-016). The Ethics Committee waived the requirement for informed patient consent, considering the observational study design and the retrospective nature of the study. Informed written consent for the surgical treatment was obtained from each patient and/or their families.

The authors have no conflicts of interest to declare.

This work was supported by JSPS KAKENHI Grant No. JP21K09113.

SH: conception and development of the study design, acquisition and analysis of data, and drafting of a significant portion of the manuscript, figures, and tables, TN, KK, and TM: acquisition of data. TK: study supervision. TT: conception and development of the study design, acquisition of data, and study supervision.

All data generated or analyzed during this study are included in this article and its online supplementary materials. Further inquiries can be directed to the corresponding author.