Electrophysiological changes in the acute phase after deep brain stimulation surgery

With the introduction of sensing-enabled deep brain stimulation devices, characterization of long-term biomarker dynamics is of growing importance for treatment optimization. The microlesion effect is a well-known phenomenon of transient clinical improvement in the acute post-operative phase. While beta band activity is confirmed as a reliable biomarker for bradykinesia using chronic recordings, little is known about the ideal time point for initial electrophysiology-based programming.

To investigate the microlesion effect impact in chronic biomarker recordings.

Subthalamic peak biomarker power was continuously recorded during the first 40 post-operative days in 12 Parkinson's disease patients implanted with a sensing-enabled neurostimulator. Daily change in mean peak power and complexity was analyzed. Additionally, power spectral density at rest was compared between immediate postoperative period and three-months-follow-up. We additionally present continuous pallidal recordings in 2 dystonia patients.

Mean peak power increased postoperatively, and the rate of change stabilized at 22-29 days. Peak power complexity showed a decrease in the number of recurrence states and laminarity, stabilizing around the same time point. Biomarker activity showed a significant increase at 3-month-follow up compared to the early post-operative phase. The microlesion effect was clinically reflected as a decrease in pre-vs. postoperative medication before setting of chronic stimulation parameters.

The transient postoperative microlesional effect is characterized by reduced beta band power and reduced neural signal complexity that gradually stabilize towards the end of the first month after surgery and most likely reflect local neuronal adaptation. These findings are important for the timing of electrophysiology-supported DBS programming, such as contact selection or adaptive algorithms.