First steps towards adaptive deep brain stimulation in Parkinson’s disease

Introduction: Deep brain stimulation (DBS) is used to treat Parkinson's disease (PD), when medical therapies are not efficient enough [1]. DBS requires the implantation of stimulation electrodes reaching into deep brain areas, commonly into the subthalamic nucleus (STN) that is one of the relay of the extrapyramidal circuit. Electrical pulses are sent on a fixed regular pace. Despite the proven efficacy of DBS on rigidity, bradykinesia and tremor, the principle clinical features of PD, its underlying principles and mechanisms are still not completely clear. Moreover, stimulation side effects such as speech problems, heat or numbness sensations, and dyskinesia can be caused by the stimulation. Current DBS systems are still in their early stages and can be compared to the early generation of heart pacemaker systems. Stimulation parameters (amplitude, frequencies or pulse-width) are typically fixed shortly after the implantation and remain unchanged. Altogether, the current practice –providing continuous stimulation with fixed parameters– may be suboptimal since it does not take into account any symptoms fluctuations. Adaptive DBS, where stimulation is triggered by certain neurophysiological markers, has emerged as an improvement over conventional systems. Such systems would reduce the amount of stimulation provided, thus extending battery life and potentially reducing side effects [2]. An adaptive DBS should be able to record the local field activity via the same electrodes as used for stimulation, recognize pathological brain activity, and finally drive the stimulation based on the specific needs of the patient [3]. Here we present our initial results on the development of a closed-loop adaptive DBS system.