Electrode Reconstruction Assists Postoperative Contact Selection in Deep Brain Stimulation.

Abstract

Postoperative contact selection of deep brain stimulation (DBS), testing one contact at a time, is a clinically time-consuming procedure being challenged by incoming applications of more complex DBS leads. The objective of this study is to guide clinicians to select the optimal contact by neuroanatomic information derived from electrode reconstruction. We reviewed 33 patients with Parkinson disease who underwent bilateral subthalamic nucleus (STN) DBS with preoperative magnetic resonance imaging and postoperative computed tomography scan. All electrodes were reconstructed in standardized Montreal Neurological Institute and Hospital space, and the volume of tissue activated (VTA) was estimated for each contact. According to the DISTAL atlas, we calculated 6 neuroanatomic parameters: the distances of each contact to the STN and the motor part of the STN (M-STN), the volumes of the overlapping areas of the VTA and STN/M-STN, and the number and ratio of fiber tracts through both the VTA and motor areas (primary motor cortex and supplementary motor area). All 6 parameters showed significant group differences between clinical optimal and nonoptimal contacts by permutation test. Moreover, the possibility of being the optimal contact was estimated for all contacts under the combination of these 6 parameters using a generalized linear model, which resulted in accuracy of 72.7% for prediction of contact selection. Electrode reconstruction could assist in selecting the clinical optimal contact and improving its efficiency.