Elucidating the systemic effects of deep brain stimulation in Parkinson's disease with diffusion tensor imaging (DTI)

Objective: While deep brain stimulation (DBS) of the subthalamic nucleus (STN) significantly improved the therapeutic options in Parkinson's disease, the direct mechanisms of DBS are poorly understood. To elucidate the systemic mechanisms of DBS, we pursued a multidisciplinary approach using preoperative high resolution DTI imaging and postoperative analysis of electrode positions, clinical outcome and stimulation parameters. Methods: Fifteen idiopathic PD patients (11 males, age 63.3±8.2,H&Y=3.5±0.8) were selected for DBS. Preoperative T1 and DTI at 3T and postoperative T1 images at 1.5T were recorded. Probabilistic tractography was run from subcortical-(Gl. pallidus internus and externus) and cortical seeds (primary motor-, M1; dorsal-(PMd) and ventral-(PMv) premotor- and supplementary motor cortex, SMA) to targets represented by active DBS contacts. The microstructural properties, as derived from diffusion parameters in regions of interest at active contacts were analysed in relation to postoperative DBS parameters. Results: Our analysis revealed negative linear correlations of the postoperative DBS current intensity at the active contacts with the connectivity profiles from M1 (r=–0.40,p<0.05) and SMA (r=–0.41,p<0.05). There was no consistent dependency of the analysed clinical data neither on the connectivity of the other studied pathways nor on the microstructure of the active electrode regions. Conversely the white matter integrity represented by fractional anisotropy- and probability of major fibber direction values in the stimulated regions predicted the postoperative amendment of DBS-current strength. Conclusions: Our data suggests that STN-DBS might neuromodulatory interfere with pathways connecting M1 and SMA with stimulated areas. Microstructural integrity of these regions could predict the postoperative adjustment of stimulation parameters. DTI and tractography are promising tools that might refine DBS targeting and elucidate the achieved effects