Objective: To investigate whether neuroimaging can help in choosing effective electrodes in PD patients with deep brain stimulation (DBS) of the subthalamic nucleus (STN).

Background: Clinical determination of the best electrode for stimulation in PD patients with STN‑DBS can be a time‑consuming process. Alternative strategies, which for example guide electrode selection on imaging‑based information are thus warranted.

Method: Using the Lead-DBS toolbox (www.lead-dbs.org), we determined electrode locations in 30 patients who had received bilateral STN DBS with DBS leads comprising of either 4 or 8 levels of electrodes. We then investigated whether the active levels, chosen for stimulation 6 months after implantation, were inside or outside an atlas‑based STN, its sensorimotor subpart, as well as a previously published, cohort‑based, probabilistic ‘sweetspot’. We also investigated the distances of the active levels to the centers of gravity of  the STN, the sensorimotor STN, and the ‘sweetspot’ and analyzed whether the levels closest to the centers of gravity were chosen for stimulation. Lastly, we determined how much of each simulated volume of activation (VTA) lay inside the respective structures.

Results: 94 % of active levels lay inside the STN while only 55 % lay inside its sensorimotor part. 89 % of active levels lay inside the ‘sweetspot’. In only 18% of investigated DBS leads, the electrode level closest to the center of gravity of the STN was selected for stimulation. A similar result (17 %) was observed for the sensorimotor STN. However, when investigating the ‘sweetspot’, the level closest to its center of gravity was chosen for stimulation in 47 % of the DBS leads. When analyzing VTAs, an average of 38 % of each stimulation volume lay inside the STN, while only 19 % of each stimulation lay inside the sensorimotor STN. On the other hand, an average of 62 % of each VTA lay inside the ‘sweetspot’.

Conclusion: These preliminary results highlight that selecting electrodes for stimulation based on their relation to the STN presents many challenges. Our previously published probabilistic sweetspot, seems to be superior in predicting electrode selection but it is still unclear, whether such strategies as demonstrated here could one day assist clinicians in programming DBS devices.

« Back to 2019 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/predicting-active-electrode-selection-from-atlas-based-targets-and-probabilistic-sweetspots-in-stn-dbs/