Mechanisms and Clinical Applications of Deep Brain Stimulation Local Evoked Potentials

Abstract

Deep brain stimulation (DBS) is an effective surgical intervention for movement disorders such as Parkinson’s disease (PD) and essential tremor. While DBS treats the motor symptoms of PD, and thus improves the quality of life of patients, several side effects and refractory symptoms limit its effectiveness. Further, the mechanism of action of DBs for PD is not fully known, making improvements to DBS implementation difficult. DBS local evoked potentials (DLEPs) reflect neural activity in response to DBS, providing insights into the mechanisms underlying DBS. For example, DLEPs change over time in response to continuous DBS, suggesting a plasticity mechanisms is at play in DBS. However, the mechanisms mediating this change in DLEPs over time are unknown. DLEPs also have potential utility as a DBS biomarker for parameter selection, lead localization, and adaptive DBS. The goals of this dissertation were to investigate the mechanistic underpinnings of DLEP dynamics, and to assess the viability of DLEPs as a DBS programming tool.First, we used a computational model of DLEPs to investigate the mechanisms underlying DLEP dynamics. The results of the computational model suggested short-term synaptic depression (STSD) of the hyperdirect pathway mediates DLEP dynamics in STN DBS. To test the computational model prediction, we developed a rat model of DLEPs – the only reported small animal model of DLEPs to our knowledge. We overexpressed endophilin A1 and alpha-synuclein, two proteins implicated in synaptic vesicle recycling, in the rat model and found DLEP dynamics changed, agreeing with the computational model prediction. With the well-known link between alpha-synuclein and PD in mind, the rat model results inspired a retrospective analysis of human motor symptoms in subjects we gathered intraoperative DLEPs from. We found motor assessments using Unified Parkinson’s Disease Scale Rating Scale (UPDRS) part III correlated with DLEP dynamics. This result suggests DLEP dynamics may have utility as a biomarker of PD disease state. Second, we investigated the viability of DLEPs as a tool for DBS programming. We recorded intraoperative DLEPs in participants undergoing DBS lead implantation and used that data to predict effective and ineffective DBS settings. We gathered DLEPs under many different DBS configurations using a rapid probing approach in which we trialed different DBS settings in 1 s epochs. The DLEPs were then used to predict effective and ineffective settings. Participants returned for motor assessment follow-up months postoperatively. We found a significant difference in motor scores between DLEP effective and DLEP ineffective settings. We did not detect a difference between clinician settings and DLEP effective settings. These results suggest DLEPs may be a viable DBS programming tool. The findings in this dissertation contribute to our understanding of DBS mechanisms and provide evidence of the clinical value of DLEPs.