Emerging strategies in the management of essential tremor

Abstract

Currently available therapies for essential tremor (ET) provide sufficient control only for less than a half of patients and many unmet needs exist. This is in part due to the empiric nature of existing treatment options and persisting uncertainties about the pathogenesis of ET. The emerging concept of ET as a possible neurodegenerative disorder, better understanding of associated biochemical changes, including alterations in the γ-aminobutyric acid (GABA)-ergic system and gap junctions, and the identification of the role of the leucine-rich repeat and immunoglobulin-like domain-containing 1 (LINGO-1) gene in ET pathogenesis suggest new avenues for more targeted therapies. Here we review the most promising new approaches to treating ET, including allosteric modulation of GABA receptors and modifications of the LINGO-1 pathway. Medically refractory tremor can be successfully treated by high-frequency deep brain stimulation (DBS) of the ventral intermediate nucleus, but surgical therapies are also fraught with limitations due to adverse effects of stimulation and the loss of therapeutic response. The selection of additional thalamic and extrathalamic targets for electrode placements and the development of a closed-loop DBS system enabling automatic adjustment of stimulation parameters in response to changes in electrophysiologic brain activity are also reviewed. Tremor cancellation methods using exoskeleton and external hand-held devices are also briefly discussed.

Introduction

Essential tremor (ET), despite being one of the most common movement disorders, remains a poorly understood entity. The available pharmacologic and nonpharmacologic treatments are purely symptomatic and empiric [Hedera et al. 2013]. The historical term ‘benign tremor’ has been largely abandoned because of undisputable evidence of severe functional and psychological disabilities seen in the majority of patients with ET. It has been suggested that ET behaves as a neurodegenerative disorder and many patients develop cerebellar dysfunction and cognitive decline in addition to the hallmark postural and action arm tremor of ET [Benito-León and Louis, 2006]. However, the concept of ET as another neurodegenerative disorder has not been fully accepted [Rajput et al. 2012].

More than 90% of patients with ET report tremor interfering with their activities of daily living and pharmacologic therapy is a mainstay initial approach [Elble, 2006]. In spite of many attempts to develop other agents, primidone and propranolol remain the only first-line medications [Hedera et al. 2013]. Moreover, all medications used for the reduction of tremor have initially been developed and approved for other indications and their use in ET was established empirically. The rate of responders varies among different studies but more that 50% patients do not experience clinically meaningful benefits from medications [Zesiewicz et al. 2005]. Even patients who report a significant tremor improvement do not achieve a tremor-free state. Adherence to therapy is further reduced by common adverse effects induced by anti-tremor medications, including somnolence and other cognitive side effects. This narrow therapeutic window results in many patients taking subtherapeutic doses or opting to stop medication entirely [Louis, 2015].

Medically refractory tremor is suitable for surgical management and targeting the ventral intermediate nucleus (ViM) of the thalamus has proved a very effective therapy for an advanced ET [Benabid et al. 1991]. Deep brain stimulation (DBS) is currently the preferred method of disrupting the abnormal cerebellar-thalamic outflow that is hypothesized to be involved in the pathogenesis of ET [Ilinsky and Kultas-Ilinsky, 2002; Lorenz and Deuschl, 2007]. Thalamotomy is another option in the control of medically refractory tremor and recent improvements in lesioning techniques and improved targeting have led to a renaissance of ViM thalamotomies [Kluger et al. 2009]. Focused ultrasound lesioning with magnetic resonance imaging (MRI) guidance is a safe and effective lesioning procedure but at present has limited availability [Elias et al. 2016].

In contrast to thalamotomy, DBS allows bilateral thalamic stimulation and stimulation fields may be adjusted based on a patient’s response or adverse effects from the electric current. Some patients after several years of excellent response experience a loss of efficacy requiring a gradual increase in DBS output to regain control of tremor. Unfortunately there is a high risk of circumferential current spread into structures near the ViM, such as the internal capsule and the nucleus ventralis caudalis, causing poorly tolerated adverse effects of muscle contractures and paresthesias, respectively. This can limit the therapeutic use of DBS in patients with severe tremor and requires additional approaches to surgical therapy of ET.

Current limitations in both pharmacologic and surgical approaches to ET with their many unmet needs for effective, safe and sustained therapies have generated a lot of activity in the ET field. Here we review the most promising new pharmacologic agents and novel surgical methods to compensate for shortcomings of traditional ViM DBS. Lastly, we will also briefly review other tactics utilizing muscle co-contractions and other tremor-cancelling methods.

Emerging strategies in pharmacologic therapy of essential tremor

Current treatment options for ET have been extensively summarized and we will not review specific details of existing first, second, and third lines of therapy [Hedera et al. 2013]. Available ET medications have been tried empirically for tremor control and there are still uncertainties about their mechanism of action due to our limited understanding of ET pathogenesis. The mechanism of action is best understood for the first-line agent propranolol, a nonselective β-adrenergic receptor antagonist. Blocking of peripheral noncardiac beta-2 receptors located in the muscle spindles is most likely responsible for the tremor control of propranolol [Abila et al. 1985]. However, there is a little doubt that the pathogenesis of ET is of a central origin and other existing ET medications do not display any significant peripheral pharmacologic action. Aberrant oscillations in the olivocerebellar circuit have been implicated in the generation of abnormal rhythmic motor activity [Ilinsky and Kultas-Ilinsky, 2002; Lorenz and Deuschl, 2007]. Neurodegenerative changes in patients with ET are also most pronounced in the cerebellum with a selective loss of Purkinje cells and signs of their axonal swelling and degeneration with formation of axonal torpedoes found in some patients [Louis et al. 2007]. These findings were not always replicated and some authors have suggested that the degree of Purkinje-cell loss and axonal swelling does not differ between patients with ET and age-matched controls [Rajput et al. 2012].

Purkinje cells, together with cerebellar Golgi interneurons and basket cells, are γ-aminobutyric acid (GABA)-ergic neurons and their intracerebellar output is inhibitory. Furthermore, the loss of the GABA_A receptors and GABA levels are also seen in the dentate nucleus where the cerebellar-thalamic pathway originates [Paris-Robidas et al. 2012]. Thus, diminished cerebellar GABA-ergic tone may cause reduced inhibitory cerebellar outflow with resulting overactivity of the cerebellar output.

The proposed mechanism of action of other available ET medications, such as the first-line agent primidone, second-line agents gabapentin, pregabalin, topiramate, or benzodiazepines, and third-line agent sodium oxybate, is the enhancement of GABA-ergic activity in the central nervous system. It remains unknown whether activation of synaptic GABA_A receptors, which are benzodiazepine sensitive and facilitate phasic inhibition, or extrasynaptic GABA_A receptors, which are insensitive to benzodiazepines but are activated by ethanol and GABA and facilitate tonic inhibition, are more important for tremor control [Farrant and Nusser, 2005]. GABA_A receptors are heteropentameric chloride-conducting ion channels that mediate fast inhibition of synaptic transmission via reduction of membrane excitability. GABA_A receptors are assembled with two alpha and two beta subunits [Olsen and Sieghart, 2009]. Synaptic receptors also contain a gamma subunit while extrasynaptic receptors have an additional delta subunit [Belelli et al. 2009]. Alpha-1 subunit knockout mice develop a tremor that has been proposed to represent a model of ET, but there is no evidence that mutations or normal variants in this subunit gene are associated with the disease in studied cohorts of patients with ET [Kralic et al. 2005; Deng et al. 2006]. Alpha-4 and alpha-6 subunits are especially enriched in the cerebellum but there is also no genetic evidence that variants in these genes are associated with ET [Pirker et al. 2000]. GABA_B receptors that are presynaptic metabotropic G-protein-coupled receptors are also reduced in the dentate cerebellar nucleus of patients with ET [Paris-Robidas et al. 2012]. Furthermore, the dentate nucleus expresses only GABA_B(1a+b), but not GABA_B(2) subunits, suggesting that GABA_B receptors in the dentate nucleus may be specific to this nucleus because typically the majority of GABA_B receptor heterodimers contain both a GABA_B(1a+b) and a GABA_B(2) subunit, co-expressed by the same cells. Undoubtedly, indiscriminately enhanced inhibitory action by increased GABA levels in the brain has not been sufficient to control tremor and may be associated with significant adverse effects, especially in the cognitive domains.

Different GABA_A receptor subtypes have been associated with diverse effects triggered by their agonists. Alpha-1 unit activation may be responsible for sedation, alpha-2 activation muscle relaxation, alpha-2 and alpha-3 may induce anxiolytic effects, alpha-5 memory impairment, and alpha-6 subunit may affect sensitivity to ethanol [Schuckit et al. 1999; Smith et al. 2012]. Nonselective activation of various types of GABA_A receptors using currently available medications, such as benzodiazepines, may result in undesirable adverse effects. Allosteric modulation or the development of GABA_A partial subtype selective agonists may optimize the adverse effect profile compared with nonselective GABA_A agonists.

Allopregnanolone (SAGE-547)

Allopregnanolone, an endogenous neuroactive steroid, is a principal metabolite of progesterone [Schumacher et al. 2014]. It exhibits a robust positive allosteric modulation on both synaptic and extrasynaptic GABA_A receptors, enhancing GABA-mediated currents [Lambert et al. 2003]. Positive allosteric modulation occurs when the binding of one ligand enhances the attraction between substrate molecules and other binding sites. The administration of allopregnanolone has a well-supported anxiolytic, sedative, and especially anticonvulsant effect. Allopregnanolone was initially developed and tested for intractable status epilepticus [Bialer et al. 2015]. Adverse effects observed after intravenous administration of allopregnanolone are mild and less than 5% of treated patients experienced sedation and dizziness. It is currently under investigation for ET [ClinicalTrials.gov identifier: NCT02277106].

Octanol

The eight-chain alcohol octanol has been explored as a treatment option based on the previous observations that almost three quarters of patients with ET report a reduction in tremor after moderate alcohol consumption [Lou and Jankovic, 1991; Deuschl et al. 2000]. Octanol is considered safe for human use and is actually approved by the US Food and Drug Administration as a food flavoring additive. Administration of long-chain alcohols in humans is not associated with an intoxication effect, which is a limitation of ethanol use for ET control. This is a contrast to alcohol that has a narrow therapeutic window when used for tremor control. Mean blood alcohol levels of 0.35 and 0.50 g/l had tremor-reducing effects but the effect of alcohol on tremor is most commonly observed at approximately 0.8 g/l blood alcohol levels, and signs of intoxication may be observed at similar doses [Koller and Biary, 1984; Knudsen et al. 2011]. The effect of alcohol is followed by a severe rebound effect and long-term consumption of alcohol is associated with many negative consequences.

Alcohol is an indirect agonist of the GABA_A receptors and its effect on tremor is likely induced by reduction of aberrant synchronization of inferior olive oscillations [Boecker et al. 1996]. Additional possible mechanisms of action include the activation of extrasynaptic GABA_A receptors, especially alpha-6/delta types that are enriched in the cerebellar granule cell neurons [Hanchar et al. 2005; Glykys et al. 2007]. An antagonistic action on low threshold calcium channels resulting in increasing T-type currents in the inferior olive may be another mechanism of action of ethyl alcohol in ET [Mu et al. 2003].

The eight-chain alcohol octanol is another potent blocker of low threshold calcium channels in neurons isolated from the inferior olive [Llinas and Yarom, 1986]. Using the harmaline-induced rodent model of ET, octanol showed a robust reduction of tremor in a dose-dependent manner after intraperitoneal injections [Sinton et al. 1989]. Moreover, studying the same tremor model, octanol had a longer duration of tremor-controlling effect than another long-chain alcohol, heptanol, at the same doses.

The first clinical trial of octanol for ET was a placebo-controlled study with a single dose of 1 mg/kg in 12 subjects [Bushara et al. 2004]. Octanol was well tolerated and tremor reduction was observed up to 1.5 h after the single dose. An open-label dose escalation study showed good tolerability and tremor control at doses up to 64 mg/kg [Shill et al. 2004]. There was a trend to dose-dependent response and no signs of intoxication were observed, even though a mild sedation was rarely reported.

Plasma levels of octanol remain low after ingestion but the metabolite octanoic acid displays dose-dependent blood levels, suggesting that this metabolite is responsible for tremor control [Nahab et al. 2011]. Octanoic acid is a fatty acid that is also used as a food adjuvant and cosmetic additive. It has been classified as ‘generally recognized as safe’. In addition, it has been used as part of a ketogenic diet for intractable epilepsy and is generally well tolerated. The most common adverse effects were diarrhea and abdominal pain.

Given the possibility that octanoic acid is an active metabolite responsible for the observed tremor control, it was studied in a preselected population of patients with ET who responded to alcohol in a double-blind, placebo-controlled, crossover study using a single dose of 4 mg/kg [Haubenberger et al. 2013]. It did not show a significant difference between the active drug and placebo using the primary outcome of objective change of postural tremor or secondary outcomes, including digital spiral analysis. Secondary outcome points showed a trend toward better tremor control. The safety data did not identify any major concerns with toxicity or tolerability. Dose escalation of octanoic acid did not reach dose-limiting toxicity levels [Voller et al. 2016]. Secondary efficacy measures suggested a dose-dependent reduction of tremor and additional studies are needed to explore safety at higher dose ranges and to confirm dose-dependent efficacy.

Even though a large, placebo-controlled study of octanol has not been performed, based on available open-label studies, the required amounts to effectively control tremor are likely to be unpractically high. However, octanoic acid remains an interesting option and further studies using higher and escalating doses are warranted. In addition, the target population may be expanded beyond patients with ET with alcohol sensitivity because it remains unknown whether the effects of long-chain alcohols and their active metabolites are similar to ethanol.

TPA023

TPA023 is a selective partial agonist of GABA_A alpha-2 and alpha-3 receptor subunits [de Haas et al. 2012]. Preclinical studies showed anxiolytic efficacy that was comparable to lorazepam without any adverse effects on alertness and balance. Additional benefits may include muscle relaxant action. This nonsedating agent has been tried in a small clinical trial to control ET given the published positive results of several benzodiazepines in short-term tremor control. The effects of 2 mg TPA023 on ET were compared with the effects of a stable alcohol level (0.6 g/l) and placebo in nine patients with ET who also reported tremor-suppressing effects of moderate alcohol consumption. TPA023 was inferior to alcohol and displayed a nonsignificant trend for tremor control compared with placebo [de Haas et al. 2012]. It was well tolerated overall and mild headaches and dizziness were the most common adverse effects in the TPA023 arm. Even though this study was negative, more selective modulation of GABA-ergic circuits remains a promising avenue to control tremor in a safe and effective manner.

Leucine-rich repeat and immunoglobulin-like domain-containing 1 gene modulation

It remains unknown whether observed alterations in the GABA-ergic neurotransmission are primary or secondary phenomena in ET. The concept of ET as a neurodegenerative disease, although not universally accepted, has recently gained more traction [Louis et al. 2007]. Advances in the genetics of ET indirectly support the degenerative nature of ET. Genome-wide association studies have identified single nucleotide polymorphisms (SNPs) in the leucine-rich repeat and immunoglobulin-like domain-containing 1 (LINGO-1) gene [Stefansson et al. 2009]. This was replicated in several independent cohorts and two SNPs located in the gene promoter, rs9652490 and rs11856808, show a strong association with ET [Jimenez-Jimenez et al. 2012]. The functional significance of these LINGO-1 SNPs associated with ET remains unclear but increased LINGO-1 expression has been reported in the cerebellar cortex of patients with ET [Kuo et al. 2013]. This appears to be relatively specific for ET as patients with Parkinson’s disease were found to have normal cerebellar levels of this protein.

LINGO-1 is a negative regulator of neuronal survival, oligodendrocyte differentiation, and axonal outgrowth and regeneration [Mi et al. 2005; Inoue et al. 2007; Lee et al. 2014]. It binds to the Nogo-66 receptor and this inhibits axonal regeneration [Mi et al. 2004]. Additional binding partners include epidermal growth factor receptor and serine/threonine kinase WNK3, activating apoptotic pathways. LINGO-1 is involved in injury response and its expression increases after neuronal injury, further emphasizing its role as a negative regulator of axonal myelination and neurite extension.

Reduction of LINGO-1 activity may enhance neurologic recovery with improved neuronal survival and neuritic outgrowth [Mi et al. 2007]. Multiple sclerosis (MS) and traumatic spinal cord injury are the conditions that have been most studied with suppression of LINGO-1 function. The antibody, BIIB033, a fully human anti-LINGO-1 monoclonal antibody blocking epitopes in the LINGO-1 immunoglobulin G domain, is currently under study for MS [Tran et al. 2014]. Undifferentiated oligodendrocytes express LINGO-1 and its upregulation after an acute demyelinating attack may interfere with oligodendrocyte differentiation and subsequent remyelination. Phase I studies found that BIIB033 is safe and effectively penetrates the blood–brain barrier after intravenous or subcutaneous administration. Recent data also support the findings that BIIB033 may indeed promote remyelination in MS after optic neuritis.

The growing evidence of LINGO-1 abnormalities in ET and the existence of safe and effective means to eliminate or downregulate its expression using monoclonal antibodies, RNA interference, or small molecules may represent a potential therapeutic target in ET. However, more preliminary data using animals models with gene dosage changes of LINGO-1 are needed before anti-LINGO-1 therapies can be tested in humans.

Blockade of gap junctions

Better understanding of ET pathophysiology will certainly widen the potential therapeutic targets. The olivocerebellar circuits generating abnormal synchronizations in the inferior olive nucleus are hypothesized to be an important part of tremor origin. Hypersynchronization is mediated through gap junctions, electrically coupling dendritic spines in synaptic glomeruli within the interior olive, causing abnormal oscillatory activity [Placantonakis et al. 2004]. Gap-junction blockers, such as mefloquine or other inhibitors of connexin 36, a major gap-junction protein, may represent another exciting class of potentially novel agents for ET [Martin and Handforth, 2006].

Emerging strategies in surgical and stimulation therapy of essential tremor

DBS

Patients with ET who fail adequate therapeutic trials of both first-line and one or two second-line medications should be considered medically refractory and potential candidates for surgical management of tremor [Zesiewicz et al. 2005; Hedera et al. 2013]. Functional neurosurgery using DBS allows adjustments of stimulation fields based on patients’ response or adverse effects from the electric current. High-frequency stimulation affects the firing rate and eliminates abnormal rhythmic oscillation between the cortex, basal ganglia, and cerebellum [Hammond et al. 2008]. Stimulation delivered through an electrode placed within a nuclear region affects neuronal bodies, axons, and fibers passing within the area of therapy. This generates an inhibitory synaptic effect on the neuronal cells and a concurrent high-frequency effect on efferent axons and fibers [Anderson et al. 2004]. The most likely mode of action is a network-wide modulatory effect and the disruption of the abnormal cerebellar-thalamic outflow.

DBS targeting the ViM thalamic nucleus is an established therapy for patients with medically refractory and debilitating tremor. Published trials report a clinically meaningful tremor control of over 80% in patients with ET and most patients enjoy a sustained benefit from DBS [Flora et al. 2010]. Complications can be related to surgical procedure with hemorrhage and infections being most common and on average they are seen in about 5–7% of all patients [Tolleson et al. 2014; Verla et al. 2015]. However, the vast majority of adverse effects from stimulation is induced by the current and disappears when the stimulation is off. Bilateral ViM stimulation is associated with more common side effects than unilateral procedures. Dysarthria is the most commonly observed adverse effect and it can be problematic for some patients. Adjustments of stimulation parameters may ameliorate this problem but this may also reduce the degree of tremor control [Sydow et al. 2003; Baizabal-Carvallo et al. 2014]. Disequilibrium or even frank ataxia can be observed with the spread of current to the inferior border of the ViM into the posterior subthalamic region affecting the dento-thalamic tract [Groppa et al. 2014]. Finally paresthesias may be common and are caused by stimulation of the nucleus ventralis caudalis. Low threshold paresthesias are seen in the posteriorly placed leads and reposition of the implanted lead may be necessary.

Moreover, some patients after several years of an excellent tremor control experience gradual reduction in tremor control that cannot be accounted for by lead migration, suboptimal lead placement, or hardware failure [Favilla et al. 2012]. The loss of efficacy may be due to the development of ‘brain tolerance’ that may be offset by stimulation holiday. However, disease progression appears to be the more common explanation as tremor scores in these patients are worse when DBS is off than at their presurgical baseline [Favilla et al. 2012]. Furthermore, frequency of arm tremor tends to decrease and the amplitude of tremor increases with longer disease duration, making it more difficult to control with electrical stimulation [Deuschl et al. 2000]. Worsening of tremor under these circumstances often requires adjustment of stimulation parameters with increased amplitude and/or increased pulse width. However, a higher DBS output is commonly associated with more pronounced adverse effects of stimulation, resulting in dysarthria, paresthesias, tonic motor contractures, and ataxia.

The traditional DBS lead delivers current flow along the axis of its trajectory with circumferential current spread. Thus, these adverse effects result from the current reaching structures surrounding the ViM nucleus, including the somatic sensory structure of the nucleus ventralis caudalis that is posterior to an optimal lead position and the internal capsule that lies medial to the ViM. Modification of the field geometry of the stimulation area may improve tremor control and reduce adverse effects caused by current spillover to surrounding nuclei and white matter tracts. This can be achieved by placing two parallel leads in close proximity allowing current flow in an orthogonal direction to the leads and diverting current away from structures responsible for undesirable effects [Yu et al. 2009].

Alternatively, dual electrode placement can widen the area of stimulation beyond the ViM. Stimulation encompassing the ViM nucleus with the thalamic ventro-oral (Vo) complex has been used with increasing rates in patients with severe ET or secondary causes of tremor, such as posttraumatic or MS-related tremors. In the thalamic Vo complex, the Vo anterior (VoA) nucleus is associated with the pallido-thalamic pathway and the Vo posterior (VoP) nucleus with the cerebello-thalamic pathway. The placement of two parallel leads, one at the ViM/VoP border and one at the VoP/VoA border includes the cerebellar and pallidal projections in the field of therapy, or simply broadens the thalamic somatotopy because it allows for a larger volume of stimulation [Foote and Okun, 2005]. Even though the published data look promising, the results are based on a small series of patients and more rigorous trials are needed to confirm the superiority of this approach.

Advanced ET, affecting proximal segments, tends to have a prominent action component resembling cerebellar outflow tremor and many of these patients achieve only partial control using the previously described approaches. Suboptimal tremor control with ViM or Vo complex targeting prompted the search for additional surgical targets. Zona incerta (ZI) and prelemniscal radiation show the most promising results [Xie et al. 2012]. ZI and prelemniscal radiation are a part of the posterior subthalamic area and are the extension of the reticular thalamic nucleus. It receives multiple afferents, including those from the globus pallidus, pars interna, substantia nigra, pars reticularis, ascending reticular activating system, and the interpositus nucleus of the cerebellum. The ZI provides a link between the basal ganglia output nuclei and the cerebello-thalamo cortical loop, and high-frequency stimulation of the ZI likely suppresses tremor by overriding the oscillations in these areas [Plaha et al. 2008]. The adverse effect profile of the ZI DBS is typically mild and transient, without lasting or severe dysarthria, disequilibrium, or tolerance, as would have been seen in the bilateral ViM DBS. However, more studies are needed to confirm the long-term efficacy of this approach.

The contacts on the available DBS leads are cylindrical providing circumferential current spread. There is increasing interest in the development of a DBS system capable of steering stimulation, especially laterally, allowing guidance of current towards the area generating the tremor and away from structures responsible for adverse effects. Leads capable of lateral current steering are currently under development by several companies [Bour et al. 2013; McIntyre et al. 2015]. Furthermore, additional features include a conversion of the open DBS system into closed-loop DBS. Current open-loop DBS systems cannot record neuronal activity and deliver stimulation based on a programmed set of stimulation parameters that do not respond to patients’ actual symptoms [Shukla et al. 2014]. Constant stimulation can also disrupt normal brain signals embedded in pathological activity, possibly inducing adverse effects of stimulation. Stimulation targeting only abnormal oscillations may be applied intermittently based on the abnormal brain activity and would allow normal signals to re-emerge. These are the potential advantages of a closed-loop DBS system that would be able automatically to adjust stimulation parameters with changes in electrophysiologic brain activity providing real-time feedback [Hosain et al. 2014]. This technology must be capable of recording local field potentials reflecting the electrical oscillatory activity from deep brain structures. Oscillations in the range 8–27 Hz recorded from the ViM are significantly coherent with tremor frequency recorded from surface electromyography [Marsden et al. 2000]. Patients with Et also exhibited enhanced coherence of 5–15 Hz when the electrodes were located in the ViM and VoP nuclei. However, the precise relationship between electrophysiologic recordings and DBS stimulation parameters in ET is not fully understood yet [Pedrosa et al. 2014]. Outside sensors, such as accelerometers can also measure tremor and this information may be used as real-time feedback for the adjustment of DBS parameters.

Lesioning procedures

Ablative procedures produce a permanent lesion within a selected target and thalamotomy is one of the surgical methods used for the treatment of refractory tremor. Lesioning surgery is still suitable for patients who are unwilling to undergo DBS, are at a very high risk of infection, for whom frequent DBS programming sessions are logistically difficult, and unilateral tremor control is sufficient to improve quality of life. Thermocoagulation of the ViM has similar perioperative risk and is performed only rarely. New targeting and lesioning methods have reinvigorated interest in thalamotomy. Stereotactic radiosurgery is one possible method of performing thalamotomy using precisely focused external radiation beams [Campbell et al. 2015]. However, it has latent effects occurring months after treatment and it is not possible to assess the degree of tremor control during the procedure. The ViM can also be effectively targeted with transcranial MRI-guided focused ultrasound [Elias et al. 2013]. High-intensity ultrasound and MRI allows for precise intracerebral targeting with real-time clinical and radiographic monitoring of the treatment location and intensity with the use of thermal imagery. The procedure starts with low-power sonications producing temperatures of 40–45°C. This can confirm accurate targeting using magnetic resonance thermography and the patients are assessed during the procedure for efficacy and the absence of adverse effects, such as dysarthria, paresthesias, or weakness. Therapeutic sonications are then implemented by gradually escalating the power and monitoring the temperature. Final sonication temperatures range from 55°C to 63°C at the maximal voxel measured by means of magnetic resonance thermography.

The results of a randomized trial using either focused ultrasound thalamotomy or a sham procedure in 76 patients have been reported recently [Elias et al. 2016]. Patients treated with thalamotomy experienced 50% improvement in tremor rating and this was maintained during the 12-month follow-up period. Side effects of thalamotomy included gait difficulties in 36% of treated patients and paresthesias in 38% of treated patients. They persisted in 9% and 14% of patients, respectively during the 12-month follow-up period [Elias et al. 2016]. More long-term follow-up data are needed to establish clearly the role of this method in ET therapy.

Figure 1.

Diagram of the major structures important for tremor pathogenesis with the most promising new therapies and procedures. DBS, deep brain stimulation; GABA, γ-aminobutyric acid; LINGO-1, leucine-rich repeat and immunoglobulin-like domain-containing 1 gene; rTMS, repetitive transcranial magnetic stimulation; SBS, superficial brain stimulation.

Repetitive transcranial magnetic stimulation and superficial brain stimulation

Abnormal oscillations in the olivocerebellar circuit are projected to the whole motor network through the cerebello-thalamo-cortical pathways. Repetitive transcranial magnetic stimulation (rTMS) applied to the cerebellar cortex has been suggested as another approach to modulate cerebellar output and thereby disrupt the tremor [Koch et al. 2008; Popa et al. 2010]. A single session of low-frequency (1 Hz) rTMS of the cerebellum produced an improvement of tremor with short-term benefit that disappeared within 5 min after the end of the stimulation [Gironell et al. 2002]. A significant and positive effect sustained for up to 3 weeks after the cerebellar rTMS was observed in 11 patients with ET using an open-label design [Popa et al. 2013]. Improvement of tremor was also associated with normalized functional connectivity in the cerebello-thalamo-cortical pathway using functional MRI analysis.

Imaging studies also detected abnormal activation in primary and supplementary motor cortical areas that are part of the cerebello-thalamo-cortical network playing a crucial role in tremor pathogenesis [Boecker and Brooks, 1998; Schnitzler et al. 2009]. rTMS applied to the primary motor cortex resulted in tremor reduction but this was detectable only by accelerometric analysis and clinical rating did not improve with active stimulation [Hellriegel et al. 2012]. Direct superficial brain stimulation using an implanted subdural electrode also exhibits inhibitory effects on the cortex when applied to the motor or premotor cortex of patients with ET. Continuous theta burst stimulation effectively controlled tremor and was superior to either high-frequency stimulation or sham stimulation [Picillo et al. 2015]. However, the same type of stimulation applied to the cerebellar cortex was not effective [Gironell et al. 2014].

Emerging strategies in biomechanical loading therapy of essential tremor

Reduction of tremor after the application of mechanical forces or inertial (mass) loads to the tremulous limb is a well-known phenomenon [Heroux et al. 2009]. Biomechanical loading has emerged as a potential alternative for tremor management without the need for pharmacotherapy or surgical procedures [Rocon et al. 2012]. A number of devices have been developed and ambulatory prosthetic devices may be most suitable for patients with ET because other devices are wheel-chair mounted or fixed to an external frame.

An orthosis is a wearable device that acts in parallel to the affected limb applying an inertial load to a selected set of limb articulations to reduce tremor. The wearable orthosis for tremor assessment and suppression (WOTAS) exoskeleton is an active exoskeleton that can apply intersegment forces when attached to the patient’s upper limb [Rocon et al. 2007]. It is equipped with kinematic (angular velocity) and kinetic (interaction force between limb and orthosis) sensors. The system generates an equal but opposite motion to the tremor, actively compensating and effectively cancelling tremor from the overall motion. This device achieved a consistent 40% reduction of tremor power for all users and patients with severe tremor achieved almost 80% tremor reduction. The disadvantages include a feeling of resistance with every voluntary movement. In addition, the device did not meet users’ expectations in terms of cosmetics and aesthetics.

Another biomechanical loading approach is a neuroprosthesis that applies forces to the tremulous limb through transcutaneous neurostimulation [Rocon et al. 2012]. The co-contraction of the affected muscles reduces tremor without affecting the concomitant voluntary movement as was observed for the WOTAS device. Loads are applied concurrently at a pair of antagonist muscles and the intensity of co-contraction is continuously adapted to the ongoing severity of the tremor, which is monitored with gyroscopes. An increase in tremor amplitude is compensated by an increase in the level of muscle co-contraction and tremor is reduced when muscle contractions in active muscles are sufficiently amplified.

A similar approach has been applied to unattached external devices, such as pens or utensils. One example is the Liftware stabilizing handle using active cancellation of tremor technology. It detects the limb motion with the separation of tremor from the intended movement of the hand. The sensors have a tri-axial accelerometer embedded in the handle base to sense the direction of tremor in the horizontal and vertical directions. It then directs the motors to move the utensil in the opposite direction of the detected tremor. Preliminary data suggests that this is an effective and safe method to improve various activities of daily living but this concept may be suitable for a mild and moderate degree of tremor only [Pathak et al. 2014].