(63) Short-latency potentials evoked by subthalamic DBS demonstrate activation of prefrontal hyperdirect pathway in humans

Introduction: Many individuals with Parkinson’s disease (PD) experience non-motor impairments affecting behavioral regulation, such as impulsivity and difficulty inhibiting responses. These deficits are worsened by dopaminergic medications or deep brain stimulation (DBS). While studies suggest that movement inhibition is mediated by a hyperdirect pathway between the prefrontal cortex—particularly the inferior frontal gyrus (IFG)—and the subthalamic nucleus (STN), direct evidence for this pathway in humans remains limited. Our goal in this study was to demonstrate a direct projection from the prefrontal cortex to the STN using in vivo physiological recordings in PD patients.

Methods: We measured prefrontal and motor cortex evoked potentials in response to STN DBS in PD patients undergoing awake surgery. Cortical responses were recorded using two temporary subdural electrocorticography (ECoG) strips (28 or 6 contacts), inserted through the same burr hole used for DBS implantation. The strips targeted the arm area of the primary motor cortex (M1) and the pars triangularis of the IFG. ECoG electrode locations were determined using intraoperative CT coregistered to preoperative MRI, with the cortical surface extracted using FreeSurfer or FastSurfer software and visualized in Slicer. Contacts over caudal- and rostral middle frontal gyri were considered as located over dorsolateral prefrontal cortex (DLPFC), compared to those directly over the IFG. ECoG signals were re-referenced using either a narrow or wide bipolar montage, aligned to stimulus onset, and averaged to generate cortical evoked potentials (cEP) from approximately 120 pulses per DBS setting. On average, 35 DBS settings were tested per patient.

Results: We recorded cortical responses to STN DBS in 10 PD patients. In nine patients, we detected a short-latency (~2–10 ms) cEP in both the DLPFC and IFG. In three patients, the response in the IFG occurred ~ 2-5 ms earlier than in the DLPFC. Moreover, the cEP amplitude was higher when more ventral DBS contacts were stimulated. In M1, all patients demonstrated a short-latency cEP (~2–10 ms) in response to the same stimulation settings as the DLPFC and IFG, but M1 cEP had higher peak amplitude compared to the IFG and DLPFC responses.

Conclusion: The identification of a short-latency prefrontal cEP in response to STN DBS likely represents antidromic activation of the prefrontal hyperdirect pathway, which has been suggested to play a crucial role in regulating behavior in humans. Incorporating the activation of this pathway into DBS computational modeling could improve the management of impulsivity in neurologic and psychiatric disorders.