Spatial Topographies of Unilateral Subthalamic Nucleus Deep Brain Stimulation Efficacy for Ipsilateral, Contralateral, Midline, and Total Parkinson Disease Motor Symptoms

Abstract

Subthalamic nucleus (STN) deep brain stimulation is a successful intervention for medically refractory Parkinson disease, although its efficacy depends on optimal electrode placement. Even though the predominant effect is observed contralaterally, modest improvements in ipsilateral and midline symptoms are also observed. To elucidate the role of contact location of unilateral deep brain stimulation on contralateral, ipsilateral, and axial subscores of Parkinson disease motor symptoms. Eighty-six patients receiving first deep brain stimulation STN electrode placements were identified, yielding 73 patients with 3-month follow-up. Total preoperative and postoperative Unified Parkinson Disease Rating Scale Part III scores were obtained and divided into contralateral, ipsilateral, and midline subscores. Contact location was determined on immediate postoperative magnetic resonance imaging. A 3-dimensional ordinary "kriging" algorithm generated spatial interpolations for total, ipsilateral, contralateral, and midline symptom categories. Interpolative reconstructions were performed in the axial planes (z = -0.5, -1.0, -1.5, -3.5, -4.5, -6.0) and a sagittal plane (x = 12.0). Interpolation error and significance were quantified by use of a cross-validation technique and quantile-quantile analysis. There was an overall reduction in Unified Parkinson Disease Rating Scale Part III symptoms: total = 37.0 ± 24.11% (P < .05), ipsilateral = 15.9 ± 51.8%, contralateral = 56.2 ± 26.8% (P < .05), and midline = 26.5 ± 34.7%. Kriging interpolation was performed and cross-validated with quantile-quantile analysis with high correlation (R2 > 0.92) and demonstrated regions of efficacy for each symptom category. Contralateral symptoms demonstrated broad regions of efficacy across the peri-STN area. The ipsilateral and midline regions of efficacy were constrained and located along the dorsal STN and caudal zona incerta. We provide evidence for a unique functional topographic window in which contralateral, ipsilateral, and midline structures may achieve the best efficacy. Although there are overlapping regions, laterality demonstrates distinct topographies. Surgical optimization should target the intersection of optimal regions for these symptom categories.