Deep brain stimulation (DBS) of the globus pallidus (GP) is an effective treatment for patients with advanced Parkinson's disease (PD). Despite recent successes, the precise site(s) within the pallidum and physiological mechanism(s) of the therapy remain uncertain due in part to our limited understanding of the neural response to DBS. We hypothesize that parkinsonian motor symptoms have therapeutically distinct targets within the GP. By identifying the mechanisms of GP-DBS, we can then develop new stimulation patterns and electrodes to selectively target the cell groups implicated in the therapeutic benefit while minimizing stimulation induced side-effects. In this study, we propose to develop anatomically and biophysically accurate models of DBS in the globus pallidus externus (GPe) and globus pallidus internus (GPi) for four MPTP-treated, hemi-parkinsonian rhesus macaques. Each animal has been or will be implanted with a monkey-scaled version of a clinical DBS lead such that the four electrode contacts span the sensorimotor regions of both GPe and GPi. The computational models will be applied retrospectively to determine the neural response during therapeutic and non-therapeutic DBS in two monkeys. Models developed in two additional monkeys will then be used to prospectively evaluate the effects of targeted stimulation of specific anatomical territories. Our working hypothesis is that direct stimulation of the posteroventral sensorimotor GPi will primarily improve rigidity and levodopa-induced dyskinesias, whereas targeted stimulation of the sensorimotor aspects of dorsal GPi and ventral GPe will primarily improve bradykinesia. If the results of this study support this hypothesis with both retrospective and prospective evaluation, it will provide two important contributions to the field: 1) substantiate the technique of using detailed computational models to guide DBS parameter selection, and more importantly 2) provide anatomical and electrical guidelines for the clinical programming of GP-DBS implants.
