Until now, however, there has been little understanding of precisely why and how DBS works. And this lack of knowledge has held back efforts to further improve the therapy. Now, scientists at the University of California, San Francisco (UCSF) have discovered a possible mechanism for how DBS acts on brain circuits to exert its therapeutic effects (Nature Neuroscience doi: 10.1038/nn.3997).

DBS mechanisms

The new research reveals that DBS keeps Parkinson's disease symptoms in check by reducing excessive synchronization of brain activity in the motor cortex, a region on the outer surface of the brain that governs movements of the body.

"This therapy is becoming widespread for many brain disorders aside from movement disorders, including psychiatric conditions such as depression, but no one knows how it works," said senior author Philip Starr. "This is a significant step in answering this question on the level of brain networks, not just addressing where you're actually applying the stimulation in the brain."

Previous research led by Coralie de Hemptinne, a postdoctoral fellow in Starr's laboratory, showed that a measure of synchronized rhythmic activity in the brain is excessively high in the motor cortex in Parkinson's disease (Proc. Natl. Acad. Sci. USA 110 4780). In that paper, the team hypothesized that this increased synchronization thwarts the flexibility that the brain needs to plan and execute movements, and that DBS might work by decoupling activity patterns in the motor cortex.

The researchers next investigated the relationship between this excessive synchrony and symptoms, and whether synchrony lessens when symptoms are improved by DBS. To do this, they measured synchrony in the motor area of the brains of 23 patients with Parkinson's disease, before, during and after DBS, and while the patient was resting or engaged in a movement task.

The data were recorded during DBS implantation surgery, in which the end of the DBS stimulating lead is placed in the patient's subthalamic nucleus, a structure deep in the brain that's part of a "loop" of neural circuitry including the motor cortex on the brain's surface.

During surgery, the UCSF team slid a temporary strip of six recording electrodes under the skull and over the motor cortex. As seen previously, recordings of neural activity showed excessive synchronization of activity rhythms. When the DBS device was activated, the effect of the stimulation reached the motor cortex, where over-synchronization rapidly diminished. If the device was turned off, excessive synchrony re-emerged.

DBS surgery generally takes about six hours, and in the middle of the procedure patients are awakened for device testing and to ensure that the stimulating lead is correctly. During this period, 12 patients were asked to perform a reaching task in which they had to touch a blue dot appearing on a screen. Importantly, recordings revealed that DBS eliminated excessive synchrony of motor cortex activity and facilitated movement without altering the accompanying normal changes in brain activity.

"Our 2013 paper showed how Parkinson's disease affects the motor cortex, and this paper shows how DBS affects the motor cortex," said Starr. "With these two pieces of information in hand, we can begin to think of new ways for stimulators to be automatically controlled by brain activity, which is the next innovation in the treatment of movement disorders."

Looking ahead

In these experiments, the recording strip had to be removed before the end of surgery, thus data were collected over a relatively short time. Starr and his team have now collaborated with medical device company Medtronic on a new generation of permanently implantable DBS devices that can record activity in the motor cortex while delivering stimulation to the subthalamic nucleus.

Five UCSF patients have received treatment with these new devices. All of the data collected can be uploaded for research during follow-up visits, further improving the understanding of how DBS reshapes brain activity. "Now we can try to find even better correlations between DBS and symptoms, and we can even look at the effects of medications," said de Hemptinne, first author of the Nature Neuroscience paper. "This new ability to collect data over a longer time course will be very powerful in driving new research."

Related stories

• Machine learning diagnoses Parkinson's
• Optical scanner maps brain activity
• Deep brain stimulation of benefit in early stages of Parkinson's
• 'Sound' science underpins brain treatment