Characterization of the electrode-tissue interface during long-term deep brain stimulation in the 6-OHDA rat model of Parkinson's disease.

Abstract

<i>Objective.</i>Deep brain stimulation is an established therapy for neurological disorders such as Parkinson's disease, but its underlying mechanisms and tissue effects remain incompletely understood. A particular challenge arises from the foreign body response to implanted electrodes, which leads to scar formation and encapsulation, altering the electrical properties of the surrounding tissue. This study aims to characterize the electrode-tissue interface during long-term stimulation and to improve model-based estimation of the stimulation volume using subject-specific dielectric properties.<i>Approach.</i>Continuous deep brain stimulation was delivered for six weeks using a fully implantable stimulator in a unilateral 6-hydroxydopamine Parkinson rat model.<i>In vivo</i>impedance spectroscopy and<i>post mortem</i>histology were performed to assess encapsulation tissue properties. Principal component analysis was applied to identify group differences in the impedance spectra. Encapsulation layer thickness was quantified histologically and incorporated into subject-specific numerical models to resolve the non-identifiability between layer thickness and dielectric parameters. Dielectric properties were estimated by fitting simulated spectra to experimental measurements.<i>Main results.</i>Impedance spectra differed significantly between stimulated and non-stimulated animals, indicating that impedance spectroscopy can distinguish tissue responses to chronic stimulation. Incorporating subject-specific encapsulation parameters substantially altered estimates of the stimulation volume compared to conventional assumptions.<i>Significance.</i>By integrating<i>in vivo</i>measurements, histology, and computational modeling, this study replaces generic interface assumptions with subject-specific electrode-tissue properties. The results identify measurable markers of encapsulation tissue and demonstrate that conventional modeling approaches overestimate the stimulation volume when generic interface assumptions are used. Incorporating subject-specific electrode-tissue properties improves the reliability of deep brain stimulation simulations and supports individualized stimulation strategies.