P 35 Local Field Potential Guided Determination of Effective Deep Brain Stimulation Contacts from an Implantable Pulse Generator – A Pilot Study

Abstract

Question : Local field potential (LFP) activity in the beta band (13-35 Hz) has been suggested as a biomarker for motor symptoms in Parkinson’s disease (PD). Up to date, intraoperative recordings remain the only predictor of clinically effective deep brain stimulation (DBS) contacts. With the introduction of novel systems, however, LFP activity can be recorded from chronically implanted DBS leads. Consecutively, we aim to investigate how postoperative LFP recordings can be used as a reliable predictor for clinically effective stimulation contacts. Methods : Here, we present LFP recordings from an ongoing analysis. A sample of 6 PD patients (11 hemispheres) was recorded in the nucleus subthalamicus. 5 patients received newly implanted directional DBS leads for the treatment of PD (9 hemispheres), whereas 1 patient (2 hemispheres) underwent pulse generator exchange on chronically implanted non–directional DBS leads. All recordings were obtained while the patient being at rest. LFPs were recorded in a bipolar fashion from all possible contact configurations and then assigned to the respective single contacts using a weighted average. Normalised power around beta peak activity (±2 Hz) of each contact was correlated to its clinical stimulation results determined during a standardised monopolar review performed 3 months postoperatively. If no clear visual beta peak was present, total power in the low beta band (13-20 Hz) was used. Results : Preliminary results from newly implanted patients (n = 5) indicate that clinically determined effective contacts correspond to those contacts that exhibit the highest ß-activity in (5/9) hemispheres investigated. Strikingly, recordings that revealed a clear visual beta peak were the most reliable predictor of DBS outcome (2/5 hemispheres), while total power in the low beta band resulted in poorer prediction of effective stimulation contacts (3/5 hemispheres). Finally, recordings from 1 patient (2 hemispheres) that underwent IPG exchange revealed poor recording quality in terms of signal-to-noise ratio and predictive power of effective stimulation contacts. Conclusion : Our preliminary results indicate that postoperative bipolar recordings from chronically implanted electrodes provide a useful tool to guide selection of effective simulation contacts. In the future, this approach may be useful for clinical decision-making and offers determination of effective stimulation contacts without time-consuming and burdensome systematic stimulation testing.