Assessing accuracy of the magnetic resonance imaging-computed tomography fusion images to evaluate the electrode positions in subthalamic nucleus after deep-brain stimulation.

Abstract

Accurate evaluation of the electrode position in subthalamic nucleus (STN) after deep-brain stimulation (DBS) has rarely been verified before. We superimposed the postoperative thin-sliced computed tomography (CT) on the preoperative magnetic resonance imaging (MRI), and the positional information of the implanted electrodes with electrophysiological data was reflected on MRI. The accuracy of this method was assessed by comparing STN demarcated on this method (M-STN) with the electrophysiologically defined STN (E-STN) and with STN on the Schaltenbrand-Wahlen atlas (SW-STN). In 16 patients, T2-weighted MRI was obtained before surgery and STN boundary was demarcated. Seven days after surgery, the position of the implanted electrodes was localized on thin-sliced CT and superimposed on the preoperative MRI. Their accordance with E-STN and compatibility of M-STN or SW-STN with E-STN were statistically assessed. In all patients, postoperative CT corresponded well with the preoperative MRI. Between inside and outside the boundaries of M-STN, the mean amplitude levels of multiunit neuronal activities were significantly different on both the rostral and caudal sides (P < .0001), and the marginal errors between M- and E-STN were 0.388 +/- 0.755 mm (mean +/- standard deviation) at the rostral margin and 0.271 +/- 0.738 mm at the caudal margin. Statistical comparison disclosed that M-STN was more similar to E-STN than SW-STN on the axial and coronal images. M-STN corresponded well with the high-amplitude area on the electrophysiological data, and the MRI-CT fusion method allowed sufficiently accurate assessment of the electrode position after DBS surgery.