Deep brain stimulation (DBS) is often used in refractory cases of Parkinson’s disease as well as in a number of other movement disorders. Electrodes are placed in deep nuclei of the brain, and continuous stimulation is typically used. The stimulation can be adjusted periodically by the physician in order to improve the efficacy of the treatment. Brain-computer interfaces are increasingly being explored in order to modulate stimulation in a variety of settings. In a recent exploratory study, Little and colleagues (2013) aimed to test whether such an interface could be used in DBS to control stimulation settings at an individual patient level on a moment-by-moment basis; this type of adaptive DBS would potentially stand as a major advance in the therapy of movement disorders.
The authors developed a system in which local field potentials from the DBS electrodes could be recorded and then used by a computer for feedback to quickly adjust the stimulation parameters of the electrode to finely control stimulation. Patients were tested, in a blind and random fashion, with the following paradigms: no stimulation, typical continuous DBS, adaptive DBS, and a random stimulation paradigm in which bursts of stimulation occurred and were not triggered by the local field potentials. Eight patients with advanced Parkinson’s disease were included in the study, and all underwent DBS implantation in the subthalamic nucleus. Clinical assessments were made by both unblinded raters and those blind to the stimulation paradigm (via videotaped assessments) using the standard Unified Parkinson’s Disease Rating Scale. Although DBS leads were implanted bilaterally, only unilateral stimulation was assessed.
The authors found that both routine and adaptive DBS improved the motor scores of the patients compared with baseline. However, in the blind assessments, the improvement with adaptive DBS was 30.5% greater than that seen with standard, continuous DBS (p = .005). This effect was maintained if rigidity was excluded from the scored assessments, as this examination finding can be difficult to judge via videotape. Random stimulation also was found to be significantly inferior to the adaptive DBS protocol.
Adaptive DBS also resulted in a significant energy savings, potentially extending battery life and reducing side effects of stimulation. The mean total energy delivered with adaptive DBS was significantly less than that needed for continuous DBS (p < .0001), and over the entire block of stimulation, adaptive DBS was only “on” for 44.2% of the time. As the paradigm progressed, less and less “on” stimulation time was needed.
This intriguing study follows a recent trend where brain-computer interfaces are being used to more finely tune settings of various types of interventions and tailor them to an individual patient’s physiology. The results in this very small proof-of-concept study are impressive. Improved motor outcomes were observed while at the same time the strategy used substantially less energy. Although other confirmatory studies are certainly needed, the future of DBS brain-computer interface for treatment of Parkinson’s disease appears to be bright.
AccessMedicine from McGraw-Hill © 2013 The McGraw-Hill Companies