Motor and non-motor outcome in tremor dominant Parkinson’s disease after MR-guided focused ultrasound thalamotomy

Background and objectives Magnetic Resonance-guided Focused Ultrasound (MRgFUS) is an emerging technique for the treatment of severe, medication-refractory tremor syndromes. We here report motor and non-motor outcomes 6 and 12 months after unilateral MRgFUS thalamotomy in tremor-dominant Parkinson’s disease (tdPD). Methods 25 patients with tdPD underwent neuropsychological evaluation including standardized questionnaires of disability, quality of life (QoL), mood, anxiety, apathy, sleep disturbances, and cognition at baseline, 6 and 12 months after MRgFUS. Motor outcome was evaluated using the Clinical Rating Scale for Tremor (CRST) and Movement Disorder Society–Unified Parkinson’s Disease Rating Scale (MDS-UPDRS). In addition, side effects and QoL of family caregivers were assessed. Results 12 months after MRgFUS significant improvements were evident in the tremor subscores. Patients with concomitant rest and postural tremor showed better tremor outcomes compared to patients with predominant rest tremor. There were no differences in the non-motor assessments. No cognitive decline was observed. Side effects were mostly transient (54%) and classified as mild (62%). No changes in the caregivers' QoL could be observed. Conclusion We found no changes in mood, anxiety, apathy, sleep, cognition or persistent worsening of gait disturbances after unilateral MRgFUS thalamotomy in tdPD. Concomitant postural tremors responded better to treatment than predominant rest tremors. Supplementary Information The online version contains supplementary material available at 10.1007/s00415-024-12469-z.


Introduction
Parkinson's disease (PD) is characterized clinically by a variety of motor and non-motor symptoms.PD is increasingly considered a heterogeneous syndrome rather than a clear-cut entity, which is evident also with regard to the main clinical subtypes, i.e. tremor dominant (tdPD), akinetic-rigid or postural instability and gait difficulty type [1].The prevalent tremor in PD is an asymmetric resting tremor suppressed upon movement initiation.This rest tremor is often accompanied by kinetic and/or postural tremor and the underlying tremor generator(s) remain ill-defined [2].TdPD patients may experience slower disease progression [1], but the tremor as such can be disabling [3] and medical treatment is often ineffective [4].Thus, stereotactic treatments, in particular deep brain stimulation (DBS) and thalamotomy, have been successfully used in medication-refractory tremors for many years [5].Herein, common targets include the posteroventral part of the ventrolateral nucleus of the thalamus (VLpv; corresponding to the ventral intermediate nucleus (VIM) in the Schaltenbrand atlas [6]).
The Magnetic Resonance-guided Focused Ultrasound (MRgFUS), which enables interventional neuromodulation by selective thermal lesions of defined structures in the central nervous system without opening the skull has re-awakened the interest in lesional approaches and several studies demonstrated the efficacy and safety of the procedure in tdPD [7][8][9].While motor efficacy and side effects such as paraesthesias are observed immediately or within the first few days after treatment, waning effects or effects on cognition and other neuropsychological symptoms might occur subsequently.Previous studies mainly focused on disability, quality of life and cognition, and neuropsychological changes have been less studied extensively [7,8,10,11].
Herein, we analyzed the clinical outcome of 25 tdPD patients treated with MRgFUS after 6 and 12 months with a special emphasis on safety, neuropsychological measurements and quality of life (QoL) in patients and family caregivers.Furthermore, differences in tremor outcomes in patients with different tremor manifestations were evaluated.

Patients
From December 2019 to October 2022, we enrolled 25 patients with tdPD according to the IPMDS consensus criteria [2].The pre-existing diagnosis was confirmed in our outpatient department by two movement disorder neurologists (UW and VP).All patients showed a tremor-dominant subtype classified based on established methods using the Movement Disorder Society-Unified Parkinson's Disease Rating Scale (MDS-UPDRS) III [12] (see Suppl.Methods S1 and [13]).A tremor-dominant subtype was defined by a tremor/non-tremor ratio equal or greater than 1.15.Tremor had to be moderate to severe (score of ≥ 2 in the dominant hand on the Clinical Rating Scale for Tremor (CRST) [14]) and to affect daily activities and/or QoL (score > 2 in the disability suspicion of the CRST or ≥ 30% self-rated reduction of QoL caused by the tremor).Reports of at least two previous medication trials (dopaminergic, anticholinergics, amantadine, clozapine, propranolol, budipine) that failed to achieve satisfactory tremor control were required.Current dopaminergic and other tremor-reducing drugs had to be stable for at least 30 days at the timepoint of enrollment and had to be discontinued prior to treatment (at least 12 h overnight).Exclusion criteria involved structural brain damage, epilepsy, coagulopathies, severe cardiac conditions, history of psychiatric disorders or substance abuse, reported cognitive impairment, or a skull density ratio < 0.3.All patients were also evaluated for and informed about DBS.The study was performed according to the Declaration of Helsinki and approved by the local Ethics Committee (314/18).All participants provided written informed consent.
25 patients were included in the primary outcome analysis at 6 months.6 patients were lost to follow-up at 12 months (2 patients with an initial good tremor effect denied the follow-up because of waning effects, unfortunately, further information on the extent of worsening have not become available, 1 patient with stable tremor effect refused followup because of health problems, 3 patients because of travelling restrictions caused by the COVID-19 pandemic).

Magnetic resonance-guided focused ultrasound thalamotomy
Treatment was performed in a 3-Tesla MRI system (Discovery MR750w, GE Healthcare, Chicago, IL, USA) and the focused ultrasound system (ExAblate 4000 System, InSightec, Haifa, Israel).The VLpv of the thalamus contralateral to the more affected hand was selected as the target.Localization of the VLpv followed standard stereotactic coordinates (on the level of anterior commissure to posterior commissure (AC-PC) line, 14 mm lateral to midline or 11 mm lateral to the third ventricle wall, 25% of the AC-PC line anterior to PC).Adjustment of the target was based on intraoperative clinical evaluation, i.e. the patient's feedback regarding tremor reduction and possible side effects and the diffusion tensor imaging (DTI)-derived coordinates of the cerebello-thalamic tract (CTT) and selected 'no-go areas', the corticospinal tract (CST) and lemniscus medialis (ML), which had been calculated in advance (details in [15]).For further treatment details, we refer to previous descriptions [16,17].

Efficacy
For assessment of the motor outcome, the CRST and MDS-UPDRS III. were used.Higher values indicated more severe symptoms.The CRST consists of three parts: tremor severity in different body parts (part A), motor task performance for both hands (part B), and subjective functional disability related to the tremor (part C).Tremor outcome was measured using a hand-specific subscore combining parts A and B of the treated upper extremity (CRST mod , ranging from 0 to 28) and calculated using the Weber-Fechner equation [18], which accounts for the logarithmic relationship between tremor rating R and tremor amplitude T = ((T f -T i )/ T i = 10 (α/N)*ΔR -1).Hereby, T is a function of the change in tremor ratings ΔR, α is a coefficient for the 0-4 scale and equals to 0.5, and N is the number of items included in the scale.
According to the clinical rating of rest and postural tremor of the upper extremity on the CRST part A, patients were divided into the predominant rest tremor group (rest/ postural ratio ≥ 1.5; n = 11) and the concomitant rest and postural tremor group (rest/postural ratio < 1.5; n = 14).
To further analyze the target-effect relationship, we divided patients according to their individual outcome into the higher tremor improvement group with T ≥ 0.7 (n = 12) and the lower tremor improvement group with T < 0.7 n = 13) according to the CRST mod score after 6 months (T2).
MDS-UPDRS subscales for tremor, rigidity, bradykinesia, and axial symptoms were calculated in total and separately for the treated and untreated side.Details on the subscale calculation are provided in the Supplementary Material.Clinical evaluation was conducted in an off-medication state (minimum of 4-h withdrawal of antiparkinsonian drugs).
To determine changes in tremor-reducing drugs, levodopa equivalent doses (LED) (in milligrams per day) were assessed according to established conversion rates; anticholinergics and propranolol were recorded in addition [19].

Safety
As safety parameters, the incidence and severity of the reported and observed side effects were monitored.Side effects that were only subjective and not clearly related to MRgFUS treatment without impact on daily activities and QoL were classified as mild.Side effects associated with mild or major restrictions in everyday life were recorded as moderate or severe, respectively.A worsening of motor complications was assessed with the MDS-UPDRS IV.As unsteadiness or gait disturbances are common side effects after MRgFUS treatment, we used the scale for the assessment and rating of ataxia (SARA) [31] and extracted the subitems measuring gait (item 1) and stance (item 2) separately.

Statistical analysis
Data analysis was performed using IBM SPSS Statistics for Windows, version 25 (IBM Corp., Armonk, N.Y., USA).Evaluation of the normal distribution of the data was performed using the Shapiro-Wilk test.Since most of the data were not normally distributed, non-parametric tests were used for further analysis.Friedman's test was used to assess within-group changes of the test variables among all timepoints.Post hoc analysis with Wilcoxon signed-rank tests with Bonferroni corrections were used for pairwise comparisons.Differences in tremor outcome between patients with concomitant rest and postural tremor or predominant rest tremor subtype and subgroup analysis of side effects were assessed using the Mann-Whitney U test.The same test was used to evaluate whether changes on the neuropsychological assessments differ between patients with high and low tremor improvement.The changes in the different scales were calculated by subtracting the baseline value from the value at T3.
Correlations between demographic data and the baseline motor and non-motor test scores were assessed using Spearman's correlation coefficient r.P-values < 0.05 was considered statistically significant.

Patients' demographics
Detailed demographic and clinical characteristics of the 25 patients are given in Table 1.
The mean age was 63.7 ± 11.3 years (range 35 to 82) and the mean disease duration 5.4 ± 2.6 years (range 1 to 10).23 (92%) patients were right-handed and in 17 (68%) patients the targeted VLpv was located in the left hemisphere.The mean tremor/non-tremor ratio was 2.0 ± 2.0 (range 1.0 to 10.7).Comparing the more to the less severely affected side on the MDS-UPDRS scale, asymmetric tremor (ratio ≥ 1.5) was found in 22 (88%) patients.1 patient with the initial diagnosis of essential tremor developed (additional) PD after years.

Tremor improvement
Table 2 summarizes the motor outcomes after MRgFUS.At all follow-up timepoints, a significant reduction in tremor scores and subscores of the treated side was found, with the most marked reduction at T1. Tremor was found to be reduced by 78% at T1, 59% at T2, and 60% at T3.At T1, a significant reduction in rigidity and bradykinesia of the treated side was also observed but was no longer significant at T2 and T3.There were no significant changes in the subscores of the untreated side and the subscore of the MDS-UPDRS III.

Medication
Medical adjustments were made based on parkinsonian symptoms.12 months after MRgFUS thalamotomy reductions in medications were observed in 6 (32%) patients while an increase was observed in 11 (58%) patients.A significant increase of LED was observed between timepoint T1 and T3 (Z = -1.37,p = 0.007).

Non-motor outcomes
The non-motor outcomes are given in Table 3.Taken together, only at T1, a significant improvement in the patient's disability, measured by part C of the CRST, was found which no longer reached significance at the other follow-up visits.No changes were observed in the MDS-UPDRS I and II, and NMSQuest.Furthermore, no differences were observed in the neuropsychological assessments measuring general and PD-specific QoL, mood, anxiety, apathy, and sleep disturbances.Cognitive testing (MoCA) and relatives' interview (FAQ) revealed no cognitive decline at 12 months.The MoCA even showed higher values, corresponding to better cognitive performance.
In addition to the patient's outcome, no changes in the caregivers' QoL (PQoL Carers) could be observed (Table 3).

Safety parameters
Table 4 and Supplementary Table S5 summarize the side effects related to the MRgFUS procedure and during follow-up.Procedure-related side effects included flipping sensations (68%), sensation of heat, pressure or pain during sonication (56%), transient numbness of the skull (16%), and pin-site bleeding (12%).Side effects related to thalamotomy were gait disturbances (84%), paraesthesias (40%), weakness of the contralateral limb (32%), taste disturbances (28%), reported involuntary movements of the contralateral limb, most likely corresponding to "thalamic" dystonia (20%).dysmetria (16%) or dysarthria (12%).All cases of reported involuntary movements and 2 of the taste disturbances occurred with delay and were reported at the 6 months visit.Gait disturbances could be divided into clear-cut ataxia on examination (76%) and subjective unsteadiness (24%).Paraesthesias mostly involved the face, lips, and tongue (80%), or the fingers of the contralateral hand (20%).The number of side effects and lesion volume at T1 correlated significantly (r = 0.509, p = 0.009).Furthermore, in patients with more than 2 side effects larger target adjustments were mademainly in lateral, anterior and inferior direction-but the differences did not reach statistical significance (Suppl.Fig. S1).Further subgroup analyses between patients with and without gait disturbances, paraesthesias or weakness also revealed no significant differences with regard to target adjustments.Lesion volumes at T1 were significantly larger in patients with weakness of the contralateral limb (Z = − 2.15, p = 0.03), however, this was no longer evident after 6 months.
Overall, most side effects were initially classified as mild (62%), 28% as moderate, and 10% as severe.54% of the side effects were transient, 32% resolved partially within 12 months, and 14% remained unchanged.

Discussion
Our study summarizes motor and non-motor outcomes 12 months after unilateral MRgFUS thalamotomy in patients with medication-refractory tdPD.A tremor improvement of 60% on the CRST mod of the treated side was present 12 months after unilateral MRgFUS thalamotomy, comparable to previous studies, reporting the effects of MRgFUS in PD patients [7-9, 32, 33].
Bradykinesia and rigidity of the treated side improved initially but this effect faded over time-either because of disease progression or more likely edema regression.Previous studies targeting the VLpv nucleus, reported diminishing effects on rigidity, too [32].The adjacent anterior (VLa) and posterodorsal part (VLpd) of the ventralolateral nucleus (corresponding to the ventri-oralis anterior (Voa) and posterior (Vop) nucleus in the Schaltenbrand atlas [6]), which receive pallidal projections, are considered to be associated with rigidity [34] and thus might be affected by perilesional edema after VLpv ablation.
The average tremor recurrence rate six months after MRgFUS is estimated at around 11% overall, with a worse outcome in PD patients compared to ET [33,35].We thus evaluated differences in tremor outcomes with regard to different tremor manifestations and found that 12 months after MRgFUS, patients with concomitant rest and postural tremor showed tremor improvement of 84% on the CRST subscale of the treated side (CRST mod ), whereas patients with predominant rest tremor had only 38% tremor improvement.Although tremor in PD is typically described as asymmetric rest tremor, accompanying postural tremor is common, including the more frequently observed "re-emergent" tremor and the less common "pure postural tremor" [36].Although rest and postural tremor in PD may share pathophysiological mechanisms involving interaction of cerebellothalamocortical and basal ganglia-thalamocortical circuits ("dimmer-switch-hypothesis") [37], different mechanistic hypothesis need to be considered.Transcranial magnetic stimulation in ten PD patients ameliorated rest tremor by stimulation of the primary motor cortex but not by stimulation of the cerebellum, whereas postural tremor was reset by both, suggesting an involvement of the cerebello-thalamocortical tract (CTT) particularly in postural tremor [38].More precise delineation of lesion localization with regard to different responses of rest and postural tremor might help to decipher the different pathophysiological mechanism.
No significant reduction of the LED or dose of anticholinergics and propranolol was observed after MRgFUS treatment.Previous studies reported prevention of increased LED or delayed need for levodopa on the one hand [8,39] or unchanged or even increased LED after MRgFUS thalamotomy [9].Nevertheless, treatment of other symptoms, such as bradykinesia and rigidity in progressive disease, surely contributed to altered medications.
Gait disturbances and paraesthesias were the most common side effects [32,40].Consistent with observations in essential tremor patients [17,41], we also observed taste disturbances and delayed occurrence of involuntary movements, most likely corresponding to "thalamic" dystonia.
Most side effects were mild and improved over time, indicating a consequence of perilesional edema.Recovery in the SARA subitems "gait" and "stance" was evident at the 6-month visit already.Nevertheless, considering that 46% did not fully recover (32% partially recovered), and given the impact of lesion location and target adjustment on the occurrence of side effects, highlights the importance of individualized and precise targeting.
No changes in disability, mood, anxiety, apathy, sleep disturbances or cognition became evident.Previous studies examining the safety of MRgFUS thalamotomy generally used briefer screening batteries to assess changes in disability, mood, or cognition [7][8][9].Sperling et al. evaluated non-motor outcomes with an emphasis on cognition in 20 tdPD patients using a comprehensive test battery.They found improvements in QoL and activities of daily living (ADL), but no decline in cognitive function, mood, and behavior 12 months after thalamotomy [10].Studies on non-motor symptoms following VIM-DBS are also rare, but improvement of ADL, depression, anxiety, and emotional well-being has been reported42.Surprisingly, in our study, tremor severity did not correlate significantly with QoL, while scores measuring disability, mood, anxiety, apathy, sleep, and cognition highly correlated with PDQ-39 and SF-36.Sperling et al. reported similar findings and hypothesized that non-motor symptoms might have a greater impact on QoL than tremor severity [10].
Although we used a comprehensive test battery, most of the tests were based on patient's self-assessment rather than objective clinical testing.Moreover, the MoCA measures global cognitive function and does not refer to specific cognitive domains.
Further limitations of our study are the small sample size as well as the lack of stable medications.As many patients were followed up by their local neurologist in the meantime, there was a lack of meaningful data regarding the reasons for medication adjustments in many cases.While we tried to overcome this disadvantage by conducting the patient's examination in the off-medication state, this was feasible in the outpatient condition only for a 4-h withdrawal, as many patients had to manage a longdistance journey.These limitations may indeed have a considerable impact on the motor outcome as well.Hence, stable dopaminergic therapies are recommended for future investigations.
In conclusion, our results confirm that unilateral MRg-FUS thalamotomy in tdPD is reasonably safe and efficient.Using a comprehensive screening test battery, no declines in mood, anxiety, apathy, sleep or cognition were observed.Furthermore, the SARA scores indicated no persistent worsening of gait disturbances after MRgFUS.Interestingly, the motor outcome tended to be better in patients with concomitant postural tremor.

#
Post hoc analysis was conducted for all follow-up time points compared to baseline using the Wilcoxon signed-rank test and Bonferroni correction for multiple comparisons (p' value < 0 0.017) ‡ The modified score was derived from the CRST, part A (3 items) and part B (4 items) for the treated and untreated upper extremity (range 0-28) MRgFUS Magnetic Resonance-guided Focused Ultrasound, T0 Baseline, T1 1-3 days post-MRgFUS, T2 6 months post-MRgFUS, T3 12 months post-MRgFUS, CRST Clinical Rating Scale for Tremor, CRST mod modified score of the Clinical Rating Scale for Tremor, MDS-UPDRS Movement Disorder Society-Unified Parkinson's Disease Rating Scale Timepoint T0 (n = 25) Friedman test value* T1 (n = 25) # T2 (n = 25) #

Fig. 1
Fig.1Tremor improvement after MRI-guided focused ultrasound (MRgFUS) thalamotomy in patients with concomitant rest and postural and predominant rest tremor subtype.12 months post-MRgFUS, significant tremor improvement was observed in the treated extrem-

*
Values are given for 1-3 days (T1), 6 months (T2) and 12 months (T3) after MRgFUS # Side effects related to sonication occurred only during the 10 to 20 s of sonication ‡ P values are based on Friedman test (p < 0.05) and post hoc analysis using the Wilcoxon signed-rank test and Bonferroni correction for multiple comparisons (p' < 0.017).Patients with completed 12-months follow-up were included (n = 19)MRgFUS Magnetic Resonance-guided Focused Ultrasound, T1 1-3 days post-MRgFUS, T2 6 months post-MRgFUS, T3 12 months post-MRgFUS, SARA Scale for the assessment and rating of ataxia to sonication # Sensation of "heat", "pressure" or "pain" 14

Table 1
Demographic and clinical characteristics of the study participants (n = 25) * Values are means ± SD

Table 2
Motor outcome after MRgFUS Values are means ± SD * P values are based on the Friedman Test (p < 0.05).Patients with completed 12-months follow-up were included (n = 19)

Table 3
Non-motor outcome after MRgFUS Post hoc analysis was conducted for all follow-up time points compared to baseline using the Wilcoxon signed-rank test and Bonferroni correction for multiple comparisons (p' value < 0 0.017) ‡ The Wilcoxon signed-rank test was used to compare the baseline and post-12-months score (p < 0.05) MRgFUS Magnetic Resonance-guided Focused Ultrasound, T0 Baseline, T1 1-3 days post-MRgFUS, T2 6 months post-MRgFUS, T3 12 months post-MRgFUS, CRST Clinical Rating Scale for Tremor, MDS-UPDRS Movement Disorder Society-Unified Parkinson's Disease Rating Scale, ADL Activities of daily living, NMSQuest non-motor Symptoms Questionnaire, SF-36 Short-Form-36 questionnaire, PCS Physical Component Scale, MCS Mental Component Scale, PDQ-39 39-item Parkinson's Disease Questionnaire, BDI Beck Depression Questionnaire, GDS Geriatric Depression Scale, STAIT State-Trait Anxiety Scale-"trait" anxiety, AES Apathy evaluation scale, ESS Epworth Sleepiness Scale, RBDSQ REM Sleep Behavious Disorder Questionnaire, MoCA Montreal Assessment of Cognition, FAQ Functional Activities Questionnaire, PQoL Carers Parkinsonism Carers Quality of Life Questionnaire * P values are based on the Friedman Test (p < 0.05).Patients with completed 12-months follow-up were included (n = 19)#

Table 4
Side effects after MRgFUS