154 Comparison of Chronic and Intraoperative LFP and ECoG Recordings in Humans With Parkinson’s Disease

Abstract

INTRODUCTION: Advances in implantable neurotechnologies have enabled chronic invasive long-term recordings of field potentials from depth leads or paddle-type electrocorticography (ECoG) leads. The relationship between these long-term recordings and those obtained acutely in the operating room has not been studied. METHODS: Human subjects (n=6) with movement disorders (Parkinson’s disease, dystonia) were permanently implanted with quadripolar DBS leads, and quadripolar ECoG leads over primary motor and sensory cortices, attached to a Medtronic Summit RC+S investigational device in the chest. Recordings were acquired intraoperatively using a NeuroOmega (Alpha Omega, Inc) recording system customized for research use. RC+S recordings were acquired between 1 and 345 days post implantation. The same processing pipeline was applied to both data sets, and average power spectra in canonical frequency bands (delta, theta, alpha, beta, gamma) were compared using rank sum tests within each subject. RESULTS: Power spectra show similar power law scaling phenomena between intraoperative and chronic recordings. Disease specific frequency band characteristics such as beta peaks could be seen in both chronic and intraoperative recordings, however, within subjects these were not always consistently seen. CONCLUSIONS: We present within-patient comparisons of signal characteristics (power spectra) of intraoperative recordings from a research grade recording system, with wirelessly streamed data from a chronic implantable device, using the same brain leads. Power spectra have similar characteristics between the intraoperative and long term recordings, however, there are distinct differences between the intraoperative and longer term clinic recordings, reinforcing the utility of tracking neural signals over time after the effects of lead insertion on the motor circuit have diminished.