P 73 Shorter pulse width reduces gait ataxia associated with VIM-DBS

Abstract

Background Limb and gait ataxia following chronic VIM-DBS is a frequent and clinically relevant phenomenon ( Hwynn et al., 2011 ), which is likely caused by disruption of physiological cerebello-thalamic signaling and inadvertent cerebellar activation ( Reich et al., 2016 ). Ataxia dramatically improves after a pause of stimulation for more than 72 h. At the same time, tremor will reoccur, leaving the patient disabled. To date, no effective therapeutic option for stimulation-induced gait ataxia has been identified. Shorter pulse width than 60 μs have been suggested to broaden the therapeutic windows as fiber tracts involved in tremor suppression and induction of ataxia have differential electrophysiological properties ( Reich et al., 2015 , Groppa et al., 2014 ). Here, we explore the therapeutic value of 40 μs pulse width in treatment of stimulation-induced gait ataxia. Methods 5 patients received special access codes that allowed their neurostimulators to be programmed beyond regular limits. Spatiotemporal gait parameters were measured using an accelerometer-based kinematic motion analysis system (Mobility Lab, APDM). Measurements were repeated (1) ON best stimulation for tremor suppression (2) after 4 days of withdrawal of stimulation (3) after more than 6 weeks of adjusted stimulation with 40 μs pulse width. Clinical assessment of gait ataxia and tremor was done using Score for Assessment and Rating of Ataxia (SARA) items 1–3 and the Tremor-Rating-Score (TRS). Patients were asked to self-report their global impression of change (GIC-S). Results 4 of 5 patients self-reported an improvement of gait performance off stimulation and at DBS at short pulse width. Tremor scores were reduced with VIM-DBS similar for 40 μs (9.2 ± 3.8) and 60 μs (9.6 ± 4.0) pulse width compared to OFF DBS (21.4 ± 5.5; p = 0.065) and SARA axial items (1–3) decreased with stimulation OFF (2.8 ± 0.86) and 40 μs VIM-DBS (2.4 ± 0.74;) compared to 60 μs VIM-DBS (6.0 ± 0.89; p = 0.065). Coefficient of variability (CoV) of stride length and of shank range of motion were significantly reduced both OFF DBS and with 40 μs VIM-DBS (p Conclusions Our results add further evidence to the growing number of studies pointing to gait ataxia as a severe yet reversible side effect of thalamic stimulation rather than disease progression. Shorter pulsewidths may therefore be a promising approach to target stimulation-induced side effects while maintaining desired clinical efficacy of tremor suppression.