Deep brain stimulation (DBS) is widely used for treatment of advanced, medication-refractory Parkinson's disease (PD). However, a significant proportion of patients may suffer adverse effects; up to 10% will present one or more transient or permanent neurobehavioral events.
In our case study, a 44-year-old woman diagnosed with PD 6 years previously who was suffering from motor fluctuations, dyskinesia, and freezing of gait episodes was submitted for DBS and implanted with directional electrodes. Intraoperative local field potentials (LFPs) were recorded. After surgery, conventional monopolar revision was performed. Preoperative 3T MRI studies and postoperative 3D and X-ray data were integrated using the Guide DTI software application (Brainlab), and diffusion tensor imaging tractography traced from cortical areas to each subthalamic nucleus (STN) using Elements software (Brainlab).
We observed that left STN stimulation in the ring mode significantly improved motor symptoms, but the patient presented uncontrollable mirthful laughter. Stimulation was then switched to the directional mode; laughter remained when using the more posteromedial contact (3-C+) but not 2-C+ or 4-C+ at the same parameters. Interestingly, LFP recordings showed the highest beta-band activity over contacts 4 and 2, and very scarce beta power over contact 3. The orientation of the directional leads was selected based on the 3D postoperative X-rays. Associative fibers showed the shortest distance to contact number 3.
Stimulation of the STN can affect motor and associative loops. The use of directional electrodes is a good option to avoid not only undesirable capsular or lemniscal effects, but also limbic/associative events. Oscillatory activity in the beta range that preferentially takes place over the somatomotor STN region and is closely related to motor improvement, provides a reliable guide for optimizing the DBS programming. The importance of the exact location of electrical stimulation to determine the non-motor symptoms such as mood, apathy, attention, and memory, as well as the usefulness of biological markers such as LFP for optimal programming, is discussed in relation to this case.