Computational modeling to improve treatments for essential tremor

Abstract

Essential tremor (ET) is a neurological disorder of unknown etiology that is typically characterized by an involuntary periodic movement of the upper limbs. No longer considered monosymptomatic, ET patients often have additional motor and even cognitive impairments. Although there are several pharmacological treatments, no drugs have been developed specifically for ET [1], and 30–70% of patients are medication-refractory [2]. A subset of medication-refractory patients may benefit from electrical deep brain stimulation (DBS) of the ventral intermediate nucleus of the thalamus (VIM), which receives cerebellar inputs. Abnormal cerebellar input to VIM is presumed to be a major contributor to tremor symptoms, which is alleviated by DBS. Computational modeling of the effects of DBS in VIM has been a powerful tool to design DBS protocols to reduce tremor activity. However, far less is known about how these therapies affect non-tremor symptoms, and more experimental and computational modeling work is required to address these growing considerations. Models capable of addressing multiple facets of ET will lead to novel, more efficient treatment.

Biophysical modeling of VIM to assess contributions of oscillatory activity to essential tremor

“…To understand effects of non-tremor activity in ET [32], we simulated Poisson-distributed synaptic inputs from cerebellum or possibly cortex that are related to features such as motor adaptation. Our results showed that non-tremor inputs potentiated intrinsic tremor and depressed extrinsic tremor ( Fig.…”

“…Aside from tonic currents and Poisson-distributed inputs applied to individual cells in the model (see above), externally sourced synaptic inputs were also simulated. To model ongoing, non-tremor inputs that might arise from cerebellum or cortex [32], we simulated additional Poisson-distributed inputs that terminated on AMPA-ergic synapses on the dendrites of the VIM cells. Each event time of the Poisson input consisted of 25 Gaussian-distributed inputs with a 3 ms standard deviation.…”

“…Essential Tremor (ET) is the most common movement disorder in adults, affecting 5-6% of the population [35,36]. ET is dominated by a low frequency (4-10 Hz), involuntary "intention" tremor, often of the upper limbs [13,36], but is no longer considered a monosymptomatic disorder [32], as both motor and cognitive symptoms have been observed [31,37,56]. Potential cerebellar dysfunction has been implicated in the etiology of ET [9,13,14,37,58], though the pathophysiology is heterogeneous and not fully understood [6,17,37,48].…”

“…An improved understanding of the cellular and circuit mechanisms generating tremor frequency neural activity in ET may lead to more effective pharmacological or electrical treatments. Prior computational modeling has investigated multiple aspects of the timing of deep brain stimulation on thalamic cells, including rebound bursting [3,5,32,59], and recent work investigated mean field tremor activity in thalamus [65]. Yet to our knowledge, within the VIM, whether tremor frequency neural activity is intrinsically or extrinsically modulated has not been fully established and is essential to understand mechanisms of generation toward improving treatment strategies.…”

Modeling mechanisms of tremor reduction for essential tremor using symmetric biphasic DBS

“…Prior computational modeling has investigated multiple aspects of the timing of DBS on thalamic cells, including rebound bursting [8,9,26,48], and recent modeling has been focused on understanding tremor frequency oscillations (TFOs) in the VIM and their relation to DBS [27,51]. We recently developed a detailed computational model of a tremor frequency generating network within the VIM to show how network biophysical properties affected tremor and non-tremor activity [27].…”