EP 79. Comparative study of microrecording-based STN location and MRI-based STN location

Abstract

Question STN-DBS could be improved by specifically targeting the dorsolateral part of the STN. However, the correspondence between the location of the STN on T2 MRI images used for targeting and the location of the neurophysiological STN remains unclear. In this study, we compared the STN borders on T2 MRI images with the STN borders identified by microelectrode recordings (MER), measured during DBS surgery. We study the differences between MER-based STN and MRI-based STN in 1.5 T, 3.0 T and 7.0 T images. Methods An atlas-based 3D model of the STN was automatically fitted onto the exact stereotactic locations of the MER points using their classifications as inside or outside the STN. The agreement between STN models and the MRI images of the STN was evaluated by an experienced neurosurgeon. T2 MRI images of 1.5 T (n = 14), 3.0 T (n = 10) and 7.0 T (n = 8) scans were compared. Results The dorsal border of the STN was identified more dorsal on the MRI than it was by MER. This was significant in the 3.0 T and 7.0 T images, with similar mismatch patterns decreasing towards the medial side of the dorsal border ( Fig. 1 ). The lateral border was identified significantly more medial on the MRI than it was by MER, for both 1.5 T and 3.0 T images. The STN borders on 7.0 T images showed no significant differences with MER ( Fig. 2 ). Conclusions MER helps to refine the delineation of the dorsal and lateral border of the STN in MRI-based targeting. Increasing MRI field strength to 7.0 T may improve identification of the lateral, but not the dorsal STN border.