Prospective Connectomic-Based Deep Brain Stimulation Programming for Parkinson's Disease

Background: Efficacy of deep brain stimulation (DBS) relies on accurate lead placement as well as optimization of the stimulation parameters. Although clinical software tools are now available, programming still largely relies on a monopolar review, a tedious process for both patients and programmers. Objective: This study investigates the safety and feasibility of prospective automated connectomic DBS programming (automated connectomic programming [ACP]), focusing on the recruitment of specific white matter pathways. Methods: After DBS implantation, a detailed connectomic DBS model in patient-specific space was developed for each study participant. A driving-force model was used to quantify pathway recruitment across 2400 different DBS settings. Optimization algorithms maximized recruitment of therapeutic pathways while minimizing recruitment of side-effect pathways. Thirteen subjects were enrolled in two study phases that compared DBS settings derived from ACP to standard clinical DBS settings. Results: Nine patients underwent reprogramming with ACP (5 globus pallidus interna [GPi], 4 subthalamic nucleus [STN]). Four patients underwent initial programming with ACP (3 GPi, 1 STN). All patients tolerated ACP without persistent side effects. In the reprogramming cohort, 3 patients preferred their ACP program, and 1 patient felt it was comparable to their clinical program. Unified Parkinson's Disease Rating Scale, Part III, scores for the initial ACP cohort (3 GPi, 1 STN) improved by an average of 43.5% (40.4-52.6 +/- 5.6%). Conclusions: ACP appeared clinically safe and feasible. It provided reasonable motor improvement, which can be further optimized with subsequent clinical adjustment. Additional investigation is required to refine the optimization algorithm and to quantify the clinical benefit of ACP in a larger cohort. (c) 2024 International Parkinson and Movement Disorder Society.