Deep Brain Stimulation for Tremor Recurrence after Focused Ultrasound Thalamotomy: A Retrospective Case Series and Literature Review

Abstract

Introduction

MRI-guided focused ultrasound (MRgFUS) thalamotomy has demonstrated efficacy in tremor relief; however, tremor recurrence may occur. For patients experiencing such recurrence, deep brain stimulation (DBS) presents a viable alternative. This study reports our findings on the application of DBS following MRgFUS.

Methods

We evaluated four right-handed patients diagnosed with essential tremor (ET), with a mean age of 65 years (range 47–76); three were male and one was female. Tremor was assessed using the Fahn-Tolosa-Marin score (CRST) and hemi-CRST scores, while quality of life was assessed using quality of life in the essential tremor questionnaire (QUEST). A responder analysis using ≥50% improvement cut-off was used for evaluation. These patients exhibited tremor recurrence at a mean time of 5 months post-MRgFUS. DBS electrodes were implanted bilaterally traversing the ventro-intermediate (VIM) nucleus of the thalamus and the posterior subthalamic area (PSA) in 2 patients, 1 patient had electrodes implanted to the VIM alone and another had the PSA as the sole target.

Results

Responder analysis of hemi-CRST scores for the treated hand post-DBS showed improvement in patient 3 and a borderline positive response in patient 1 (0.44), whereas patients 2 and 4 had negative values, suggesting worsening post-DBS. Total CRST responder analysis after DBS showed improvement in patients 1, 3, and 4, while patient 2 showed a negative response. Quality-of-life outcomes were heterogeneous: patient 1 demonstrated sustained QUEST improvement after both MRgFUS and DBS, patients 2 and 4 showed worsening, and patient 3 showed improvement after MRgFUS but worsened after DBS. Adverse events included 3 patients who developed stimulation-induced gait disturbance, two of them also had stimulation-induced dysarthria.

Conclusion

The effects of DBS following MRgFUS in ET patients with tremor recurrence are not well characterized. Our single-center experience suggests that DBS can be effective in select patients with manageable adverse events, highlighting the importance of careful patient selection. Further studies are warranted to comprehensively evaluate this sequential treatment approach.

Plain Language Summary

MRI-guided focused ultrasound (MRgFUS) is a treatment that can reduce tremor in people with essential tremor. However, in some patients, tremor may return after treatment. Deep brain stimulation (DBS), a surgical treatment in which electrodes are implanted in specific brain regions to control symptoms, may be an option for these patients. In this study, we describe 4 patients with essential tremor whose tremor returned after MRgFUS, on average about 5 months after treatment. All patients later underwent DBS. The electrodes were placed in different tremor-related brain targets, including the ventro-intermediate nucleus of the thalamus (VIM), the posterior subthalamic area, or both. The response to DBS varied between patients. When focusing on tremor in the hand previously treated with MRgFUS, 1 patient clearly improved after DBS, one showed borderline improvement, and two showed worsening. When looking at overall tremor scores, 3 patients improved and one worsened. Quality-of-life results were also mixed: 1 patient improved after both treatments, two worsened, and one improved after MRgFUS but worsened after DBS. Side effects were generally related to stimulation. Three patients developed walking difficulties, and two of these also developed speech problems. Overall, our experience suggests that DBS after MRgFUS may help selected patients whose tremor returns, but the benefit is not consistent. Careful patient selection and programming are important, and larger studies are needed to better understand who is most likely to benefit from this treatment sequence.

Introduction

Essential tremor (ET) is a common movement disorder with a global prevalence of approximately 1.3% [1]. ET increases in prevalence reaching up to 5.8% in individuals aged 65 and older [1]. As ET progresses patients may become disabled and require medical treatment. However, up to 55% of cases show insufficient response to medications over time [2].

MRI-guided focused ultrasound (MRgFUS) targeting the ventral intermediate nucleus (VIM) of the thalamus has emerged as an effective treatment for refractory ET, offering tremor relief with a favorable safety profile [3–5]. Despite its promising outcomes, studies have reported tremor recurrence in an average of 11% within the first 6 months post-MRgFUS [6]. For these patients, deep brain stimulation (DBS) surgery may be considered. DBS has long been the standard surgical treatment for refractory ET, demonstrating both safety and efficacy. However, DBS is increasingly being considered less frequently as a first-line surgical intervention due to its invasiveness and adverse events profile compared to MRgFUS [7, 8].

There are limited data on the safety and efficacy of performing DBS after failure of MRgFUS. To date, only three case reports of DBS following MRgFUS have been published showing promising results (see Table 1). An additional case of DBS after thalamotomy was published by Lee and Jung in 2022, yet it did not specify the method used for the thalamotomy [9].

Table 1.

Case reports of DBS post-MRgFUS

Paper	Patient age at MRgFUS (sex)	Diagnosis	Baseline tremor score	Sonications	Time to DBS	DBS target	Post-DBS tremor score (time from DBS)	DBS long-term complications
Wang et al. (2018) [10]	81 (F)	ET	Treated hand CRST: 25	Peak temperature: 63°C, 2/4 therapeutic sonications >60°C	10 months	Ipsilateral VIM	Treated hand CRST: 10 (6 months)	N/A
Levi et al. (2019) [11]	73 (M)	Mixed tremor	TETRAS: 11	No data	12 months	Ipsilateral VIM+cZI	TETRAS: 3 (12 months)	N/A
Saluja et al. (2020) [12]	70 (F)	ET	CRST part A: 30	Peak temperature: 61°C, 4 therapeutic sonications >55°C	24 months	Bilateral VIM	CRST part A: 7 (16 months)	N/A

The cases were reported as follows.

• 1.An 81-year-old woman with severe refractory ET and a baseline Clinical Rating Scale for Tremor (CRST) score of 25 in the treated-hand. She underwent left MRgFUS thalamotomy (peak temperature: 63°C, 2/4 therapeutic sonications >60°C), achieving initial significant tremor reduction with a gradual tremor recurrence which started 2 weeks post procedure (treated-hand CRST score of 16 at 1 month post-MRgFUS). By 6 months, symptoms recurred to baseline severity (treated-hand CRST score of 28). Transient gait instability post-MRgFUS resolved. Ipsilateral VIM-DBS performed 10 months post-MRgFUS showed reduction in tremor (treated-hand CRST score of 10) that was sustained at 6 months. There were no long-term adverse effects following DBS. The MRgFUS lesion was undetectable on MRI at the time of DBS, suggesting incomplete lesioning or edema resolution as an explanation for MRgFUS treatment failure [10].
• 2.A 73-year-old man with hereditary hemochromatosis and a 7-year history of right-hand tremor, diagnosed as mixed tremor, underwent left VIM MRgFUS thalamotomies twice (maximal temperature not specified), achieving transient complete relief before recurrence 1 year after the second procedure. DBS was performed 1 year post-second MRgFUS with insertion of the electrode in the left VIM and caudal zona incerta (cZI) with reduction in tremor by 80% at 1-year follow-up (The Essential Tremor Rating Assessment Scale - TETRAS score in treated hand reduced from 11 before DBS to 3 following DBS). No long-term adverse effects were reported. The authors hypothesized MRgFUS failure stemmed from insufficient lesion size, edema resolution, targeting inaccuracy or disease progression [11].
• 3.A 70-year-old woman with medically refractory ET for 30 years (baseline CRST part A score 30) underwent MRgFUS thalamotomy targeting the left VIM nucleus. The procedure, halted due to dysarthria after 21 sonications (peak temperature: 61°C, 4 therapeutic sonications >55°C), achieved initial tremor reduction (CRST part A score at 1 month 18). Symptoms recurred within 6 months and gradually worsened (CRST part A score at 24 months 28). Bilateral VIM DBS performed 24 months post-MRgFUS yielded sustained tremor control (CRST part A score at 16 months 7). Tractography revealed the MRgFUS lesion (20.28 mm³) was smaller and ventromedially displaced compared to the DBS volume of activated tissue (233.16 mm³). Long-standing adverse reactions included transient dysarthria post-MRgFUS, which resolved post-DBS. The authors hypothesized MRgFUS failure resulted from suboptimal lesion size/placement, insufficient corticothalamic tract engagement, and procedural interruption [12].

This case series describes our experience in 4 patients who underwent DBS following MRgFUS for tremor recurrence. We present the outcomes and challenges in our patients.

Methods

Study Design and Cohort

This retrospective study analyzed data from medical records of 4 ET patients with significant tremor recurrence following MRgFUS treatment at Rambam Health Care Campus, Haifa, Israel, who were subsequently treated with DBS. Three were treated with DBS at Rambam Health Care Campus between 11/2022 and 8/2023, and a single patient underwent DBS at Hadassah Medical Center, Jerusalem in 7/2022 and continued follow-up at Rambam Health Care Campus. Patients’ demographic, clinical, and neurological data were collected from medical charts. Evaluations were conducted at three time points: prior to MRgFUS, at the last follow-up before undergoing DBS, and during follow-up visits following DBS, with data from the last available follow-up visit used for analysis.

Patients

The cohort included 4 ET patients with bilateral tremor, three males, with a mean age of 65 ± 12.7 years (range 47–76) at the time of DBS implantation. For individual patients’ data, see Tables 2–4. All patients were right-handed and initially treated with unilateral MRgFUS for the right side. Notably, two patients had a peak temperature of 53°^C during treatment, one due to severe headache and another due to cavitations. In the other two patients no clear explanation for tremor recurrence was found as the target and peak temperature were optimal. VIM lesions on SWI sequence on MRI were visible for all patients during DBS planning. The preoperative planning included lesion assessment with tractography. It showed lesions in the VIM partially disrupting the dentatorubrothalamic tract (DRTT); therefore, it was decided to perform DBS instead of repeating MRgFUS. The posterior subthalamic area (PSA) was chosen as the primary target with the VIM as secondary target when trajectory allowed it. An additional consideration at the time of the procedure, was that staged bilateral MRgFUS was not yet approved. The mean duration between MRgFUS and tremor recurrence was 5 ± 5.2 months (range 1–12 months). The average interval between MRgFUS and DBS was 29.5 ± 21.6 months (range 12–59 months). The mean follow-up duration after DBS was 18.5 ± 5.6 months (range 13–24 months). Following MRgFUS, 1 patient has had transient difficulties with tandem gait while the other patients experienced no adverse events.

Table 2.

Patients’ characteristics

Patient’s serial number, year of DBS procedure	Tremor score before MRgFUS	Sonications	Tremor score before DBS (time from MRgFUS)	Time to DBS	DBS target	Post-DBS tremor score (time from DBS)	DBS long-term complications
1 (2023)	Hemi-CRST: 13	2/8 therapeutic sonications >52°C	Hemi-CRST: 4 (24 months)	31 months	Bilateral PSA	Hemi-CRST: 0 (13 months)	None
2 (2023)	Hemi-CRST: 15	4/11 therapeutic sonications ≥55°C	Hemi-CRST: 7 (12 months)	14 months	Bilateral VIM+PSA	Hemi-CRST: 8 (21 months)	Stimulation-induced dysarthria and gait disorder
3 (2023)	Hemi-CRST: 14	4/6 therapeutic sonications ≥55°C	Hemi-CRST: 10 (52 months)	59 months	Bilateral VIM+PSA	Hemi-CRST: 6 (13 months)	Stimulation-induced gait disorder
4 (2022)	Hemi-CRST: 17	2/12 therapeutic sonications ≥50°C	Hemi-CRST: 11 (3 months)	12 months	Bilateral VIM	Hemi-CRST: 14 (24 months)	Stimulation-induced dysarthria and gait disorder

Table 4.

Last follow-up DBS stimulation parameters

​	Contact	Amplitude (mA)	Frequency (Hz)	Pulse width (µs)
Patient 1	Boston Scientific – directional leads	​	​	​
Left	2–7%	2.7	174	60
	3–86%
	4–7%
Right	1–30%	1.9	174	100
	2–35%
	4–35%
Patient 2	Boston scientific – directional leads	​	​	​
Left	5–	3.3	170	50
Right	2–2%	​	​	​
	3–26%
	4–2%
	5–5%
	6–60%
	7–5%
Patient 3	Boston scientific – directional leads	​	​	​
Left	6–40%	2.0	185	30
	7–40%
	8–20%
Right	2–12%	1.3	185	50
	3–12%
	5–38%
	6–38%
Patient 4	Medtronic – ring electrode	​	​	​
Left	0–33%	0.3	125	80
	1–33%	1.1
	2–33%	0.5
Right	0–1+	2.7	125	60

Assessments

CRST [13] (scores ranging from 0 to 156) and, for the treated hand, the hemi-CRST (items 5/6, 8/9, 11–15; scores ranging from 0 to 28) [14] with higher scores indicating worse tremor were performed. Functional status was assessed using the self-reported Quality of Life in Essential Tremor Questionnaire (QUEST) [15] (range 0–100) with lower scores indicating better quality of life. Raters were not blinded to treatment.

Tractography

Deterministic tractography was performed using Brainlab Elements software (Brainlab AG, Munich, Germany), utilizing the diffusion tensor imaging method (fractional anisotropy threshold, 0.1; 30° maximum deviation). For each tract, regions of interest (ROIs) were marked using the fractional anisotropy map for tracking. To delineate the DRTT, ROIs were placed over the red nucleus region and the ipsilateral primary motor cortex. In addition, only fibers projecting to the cerebellum through the superior cerebellar peduncle were kept. The pyramidal tract was tracked by ROIs in the cerebral peduncle and the ipsilateral primary motor cortex. ROIs for tracing the medial lemniscus were placed over the dorsal column at the brain stem and the ipsilateral primary sensory cortex. The obtained fibers were validated utilizing a priori anatomical knowledge, and irrelevant fibers were excluded.

Diffusion-tensor MRI images were acquired before the procedure on a 3.0-T system (Discovery MR750; GE Healthcare, Milwaukee, WI). Imaging parameters were as follows: 9,538/82; Δ/δ, 33/26; b value, 1,000 s/mm²; 60 directions of diffusion gradients and five b₀; 2 × 2 × 2 mm³ voxel dimensions.

MRgFUS Procedure

Targeting methods included the canonical anterior commissure – posterior commissure (AC-PC) landmark marking and measurement with respect to these anatomical locations with some corrections. The initial target was set as 25% of the AC-PC length from the PC, laterally to the midpoint between 14 mm from the midline and 11.5 mm from the 3rd ventricle wall at the axial AC-PC plane. From this point, corrections were made based on clinical outcome and adverse reactions. The initial direction was toward the dentatorubrothalamic tract interception with the thalamus in steps of 0.3–0.5 mm as calculated by tractography. The exact number of sonications and the parameters used were highly individualized and depended on skull density ratio (SDR) – a parameter calculated by the system. Also, other personal factors such as the skull shape and size were taken into consideration.

Surgical Procedure

Before DBS surgery all patients underwent 3 Tesla MRI in order to demarcate the MRgFUS lesion and in order to plan the DBS. Patients underwent a CT scan on the day of the procedure. The target for the DBS electrode was initially determined by direct targeting using T1 MRI sequence and tractography based images. The electrode tract was selected using Medtronic S9 planning system in a direction which enabled safe approach to the target without crossing the sulci while avoiding visible blood vessels. Leksell Vantage Stereotactic frame (Elekta, Stockholm Sweden) was attached to the patients’ head and adjusted according to the planned coordinates. Alpha-omega system (Nazareth, Israel) was used for microelectrode insertion and recording. The permanent electrode position was determined after inspecting the microelectrode insertion and recording for the desired localization and evaluating the awake patient for clinical effect on tremor reduction and possible adverse effects. The neurostimulator (Vercise, Boston Scientific, USA, in 3 patients and Medtronic Percept PC, USA, in 1 patient) was implanted in the right chest and connected to the brain implanted electrodes. Two weeks after surgery the stimulator was activated. Stimulation parameters were adjusted to optimize the clinical results in terms of tremor reduction while avoiding adverse effects (Table 4). Repeated stimulation adjustments were performed when necessary.

Statistical Analysis

Descriptive statistics were used to evaluate patients’ clinical and demographic characteristics. A responder analysis for hemi-CRST and CRST was conducted. Positive response to treatment was defined as ≥50% improvement and defined as hemi-CRST or CRST score calculated to account Pre-MRgFUS (baseline) scores using the following formula:

$$\frac{\left(Baseline-postDBS\right)-\left(Baseline-preDBS\right)}{\left(Baseline-preDBS\right)}\ge 0.5.$$

Results

DBS procedure was performed bilaterally in all 4 patients (Tables 2–3). In two patients, the electrodes were inserted in the PSA traversing the VIM, in 1 patient the electrodes were inserted in the PSA and in the fourth patient the electrodes were inserted in the VIM (see Fig. 1–2, for the spatial relationship of DBS electrodes and the MRgFUS VIM lesion for patient 3).

Table 3.

Additional imaging and surgical data

Patient’s serial number, year of DBS procedure	MRgFUS lesion volume, cm3	Lesion size, mm	Lesion-electrode distance, mm	Orientation	Available skull area, mm2	SDR	Sonications, n	Maximal energy, Joule	Maximal temperature, °C	Maximal sonication time, seconds	Total energy, Joule
1 (2023)	0.08	6.4 × 4.3 × 4.9	4.3 (electrode tip medial, inferior and posterior to lesion)	Antero-lateral	356	0.47	14	24,070	53	25	140,520
2 (2023)	0.09	7.4 × 3.7 × 4.6	0 (within lesion)	Anterior	363	0.66	14	12,550	59	12	89,000
3 (2023)	0.13	8.0 × 6.3 × 6.2	0 (within lesion)	Antero-medial	299	0.45	11	6,600	60	13	49,200
4 (2022)	0.05	3.1 × 3.7 × 5.1	0 (within lesion)	Anterior	338	0.4	12	19,000	53	25	89,800

Fig. 1.

3D simulation of DBS electrode placement and VIM lesion for patient 3. An oblique view of the electrodes implanted in patient 3 – the upper contact of the left electrode is seen to pass through the VIM lesion while the lower contacts are outside the lesion.

Fig. 2.

Double oblique views of DBS electrode placement and VIM lesion for patient 3. Two oblique images demonstrating electrode and lesion location for patient 3 – images show brain MRI for reference with the lower contacts of the electrodes at midbrain level.

The hemi-CRST scores for the treated hand as calculated by responder analysis post-DBS were improved for patient 3 (bilateral VIM+PSA electrodes), and had a positive borderline value of 0.44 for patient 1 (bilateral PSA electrodes); negative values for patients 2 (bilateral VIM+PSA electrodes) and 4 (bilateral VIM electrodes) suggested a trend of worsening following DBS (Fig. 3a). Notably, all patients had a sustained improvement compared to baseline following MRgFUS (Fig. 3b).

Fig. 3.

Hemi-CRST responder analysis and individual hemi-CRST scores. Panel a: a responder analysis with a 50% response cut-off line for hemi-CRST post-DBS bar chart. Patient 3 is above the line, patient 1 is near the line at 44% response, and patients 2 and 4 show negative response, worsening post-DBS. Panel b: individual patients’ hemi-CRST scores bar chart at baseline, latest follow-up pre-DBS, and post-DBS. Patients 1 and 3 show a trend of tremor reduction, while patient 2 and 4 show tremor reduction pre-DBS compared to baseline, yet worsening post-DBS. Notably, scores are still better than baseline.

The total CRST scores as calculated by responder analysis post-DBS were improved for patients 1, 3 and 4; negative value for patients 2 suggested a trend of worsening following DBS (Fig. 4). Quality of life metrics results were variable. Patient 1 showed a consistent trend of QUEST scores improvement following both treatments, while patients 2 and 4 showed a trend of worsening. Patient 3 showed QUEST improvement pre-DBS compared to baseline with worsening following DBS (Fig. 5).

Fig. 4.

CRST responder analysis and individual CRST scores. Panel a: a responder analysis with a 50% response cut-off line for total CRST post-DBS bar chart. Patients 1, 3, and 4 are above the line, while patient 2 shows negative response, worsening post-DBS. Panel b: individual patients’ CRST scores bar chart at baseline, latest follow-up pre-DBS, and Post-DBS. Patients 1, 3, and 4 show a trend of tremor reduction, while patient 2 shows tremor reduction pre-DBS compared to baseline, yet worsening post-DBS. Notably, post-DBS scores are still better than baseline.

Fig. 5.

Individual patients’ QUEST scores. Lower scores mean a better quality of life. Patient 1 is showing a trend of score reduction, patients 2 and 4 show a trend of score increase, while patient 3 shows a lower QUEST score pre-DBS compared to baseline, yet worsening post-DBS. Notably, the post-DBS score is still better than baseline.

Safety and Adverse Events

Post-DBS, 3 patients developed stimulation-induced gait disturbance, two of them also had stimulation-induced dysarthria. One patient experienced no adverse effects. In patient number 4, programming did not help alleviate the adverse reactions. In patients 2 and 3, programming helped, with a partial improvement; yet, to maintain efficacy for tremor control, adverse reactions were not resolved completely.

Discussion

Our single-center experience of DBS following MRgFUS showed mixed results. Following DBS, the tremor score in the hand that had previously been treated with MRgFUS was improved in only two of the four patients treated. Furthermore, adverse events following DBS included stimulation-induced gait disturbance in three of the four patients, and stimulation-induced dysarthria in two of them. Only one patient improved with no adverse effects. This patient was unique in that he had both electrodes implanted in the PSA and his electrodes did not traverse the VIM.

We demonstrated improvement in total CRST scores with DBS following MRgFUS in 3 out of 4 patients. This reduction in total tremor score is likely due to a reduction in tremor in the hand that had not been previously treated with MRgFUS, rather than improvements in ipsilateral tremor control.

Following DBS, quality of life assessment did not show improvement in 3 out of 4 patients in our cohort. This may be attributable to lack of significant improvement following DBS, the effect of stimulation-induced adverse events, general comorbidities, disease progression, or other life circumstances unrelated to the procedure. The mean follow-up period of 18 months post-DBS allowed for a comprehensive evaluation of long-term outcomes but may also have introduced confounding effects such as age-related decline and non-DBS-related factors influencing quality of life. In particular, 1 of the patients had end stage cancer and died within 25 months of surgery.

In comparison to prior cases documented in the literature [9–12], our cohort exhibited a higher incidence of stimulation-induced adverse reactions, with 3 out of 4 patients affected. This discrepancy may be attributed to the bilateral intervention employed in our cohort, which has historically been associated with elevated adverse event rates [16]. Target selection in our cohort aligned with established approaches, focusing on VIM and PSA. Notably, two patients underwent a dual-target strategy (VIM + PSA), a method increasingly adopted in recent years due to emerging evidence supporting its safety and efficacy [17, 18]. The rationale for dual targeting is that VIM is considered a classical target for tremor control; when dual targeting can be achieved, it is preferred to a single target strategy, even in cases where the VIM was treated. Using directional leads in DBS can stimulate a large area outside the lesion within the tract and has the potential to add to tremor control. Our results show considerable heterogeneity in terms of efficacy and adverse reactions and, due to small patient numbers, are not indicative for optimizing DBS target selection. The optimal relation between electrode placement and the MRgFUS lesion remains inconclusive, yet placing the electrodes in the PSA, outside the MRgFUS lesion area showed better results in our patients.

The peak temperature during MRgFUS represents a key technical determinant for treatment success. It should be noted that two of our patients had reached low peak temperatures (53°C), below the conventional therapeutic threshold of 54–55°C [19, 20], although contemporary studies propose that lower temperatures (e.g., 52.5°C) may suffice for adequate lesioning [21, 22].

The rationale for proceeding with DBS in these cases rested on the premise that unsatisfactory MRgFUS outcomes for any reason may be improved by a sequential DBS procedure as these procedures are the current standard of care for refractory ET patients. These procedures present markedly different approaches – lesioning vs. neuro-stimulation. Furthermore, it was suggested that patients who underwent technically suboptimal MRgFUS would be likely to respond to DBS similarly to MRgFUS-naïve patients. However, this hypothesis was not substantiated in our cohort. Only one of the patients with low peak temperature demonstrated meaningful tremor reduction post-DBS without complications, whereas the other experienced both lack of tremor reduction and persistent stimulation-induced adverse effects.

This study is the first case series to analyze DBS outcomes in MRgFUS patients with tremor recurrence. It increases the existing knowledge regarding outcomes of DBS following MRgFUS. Despite its limitations, our findings provide valuable insights into this novel treatment pathway.

Our results suggest that bilateral DBS is a feasible option for managing tremor recurrence following MRgFUS in patients with refractory ET. However, several important considerations arise, including common stimulation-induced adverse outcomes, lack of tremor reduction in some cases, and importantly lack of improvement in quality of life in most patients following the procedure. This shows the critical need for careful patient selection. Further research is needed to fully assess the safety, efficacy, and patient outcomes associated with this sequential treatment approach.

Statement of Ethics

This study was conducted retrospectively in accordance with the ethical standards of the Institutional Research Committee of Rambam Health Care Campus and with the 1975 Helsinki Declaration and its later amendments or comparable ethical standards. This study protocol was reviewed and approved by Rambam Health Care Campus IRB, Approval No. 0404-17-RMB. The study has been granted an exemption from requiring written informed consent by the IRB Committee.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This study was not supported by any sponsor or funder.

Author Contributions

Mark Katson – acquisition, analysis, and interpretation of data and drafting and reviewing the manuscript. Lior Lev-Tov – conception of study, acquisition and interpretation of data, and reviewing the manuscript. Maria Nassar, Inna Senderova, Libby Brants, Alon Sinai, Ilana Erikh, and Noam Shalem – data acquisition and manuscript review. Marius Constantinescu – data acquisition. Ilana Schlesinger – conception of study, acquisition and interpretation of data, reviewing the manuscript, and final approval of the version to be published.

Funding Statement

This study was not supported by any sponsor or funder.

Data Availability Statement

The data that support the findings of this study are not publicly available due to their containing information that could compromise the privacy of research participants but are available from the corresponding author (M.K.).