Deep-brain stimulation has been used for the last 15 years to treat Parkinson's disease (PD). However, researchers have long wondered about the best target. A clinical trial that compares three-year outcomes of targeting the subthalamic nucleus versus the globus pallidus interna in PD patients finds that motor outcomes are equivalent between the two approaches. Some subtle differences seem to favor stimulating the globus pallidus, but should be cautiously interpreted, say the authors, led by Kenneth Follett, University of Nebraska, Omaha. The results appeared online in the June 20 Neurology. In related news, scientists report that a miniaturized version of transcranial magnetic stimulation, which uses large magnetic coils outside the head, could be on the horizon. In the June 26 Nature Communications, researchers co-led by Shelley Fried, Harvard Medical School, describe teeny copper coils less than a millimeter in length that generate electrical fields capable of activating neurons in culture. Though the technology is still preliminary, the coils are small enough to implant into the living brain.

The globus pallidus interna (GPi) was the ideal target for surgical lesion treatment of late-stage PD in the 1990s. But when researchers turned to deep-brain stimulation (DBS), the focus moved to the subthalamic nucleus (STN) (see Bergman et al., 1990). Many physicians hopped on the bandwagon and zeroed in on that region. But "no one had done a randomized, controlled trial of STN versus GPi stimulation," said Frances Weaver, Hines Veterans Affairs Hospital, Illinois, first author on the Neurology paper. She co-led the first randomized, blinded Phase 2 trial to compare stimulation of either the GPi or STN in PD patients. The trial recruited 299 volunteers with an average age of about 62, who had electrodes implanted bilaterally into either region. After two years, both groups' motor skills improved equally when they were off medication and on DBS, as shown by the Unified Parkinson’s Disease Rating Scale motor subscale (UPDRS III) (see ARF related news story on Follett et al., 2010).

But what would happen in the longer term? Weaver and colleagues looked at 89 GPi-targeted and 70 STN-targeted patients from the same study in a three-year follow-up. As in the two-year study, motor skills were equal between the two cohorts in the off medication/on stimulation condition. However, three differences jumped out, all seemingly favoring the GPi. First, cognitive scores dropped for the STN group, while they stayed the same in the GPi cohort, suggesting a benefit for GPi targeting. The STN cohort also declined in the usually beneficial on medication/on stimulation condition, according to UPDRS III scores, while the GPi group retained and even improved motor function. Further, when measured in the absence of drugs or when electrical pulses were turned off, STN patients’ motor scores declined over the three years, while GPi subjects’ scores remained stable, suggesting a lack of disease progression in the GPi group.

"This study seems to suggest that even though they are equivalent in treating motor symptoms over the long term, the GPi seems to gain an advantage over the STN in some critical aspects of the disease that have to do with balance, cognition, and response to medications," said Michele Tagliati, Cedars-Sinai Medical Center, Los Angeles, California, who authored an accompanying Neurology editorial. "Those are really disabling in Parkinson's disease," he added. He is among those doctors who now offer GPi stimulation to patients based on these and similar results, whereas he used to target the STN exclusively. One reason GPi may be slightly superior to STN DBS is because the patients in the STN group reduce their medication, while medication in GPi patients remains higher, he speculated. "Maybe the brain needs dopamine."

Differences should be cautiously interpreted, said Weaver. Dropout was high in this trial over the three years, and as Tagliati noted, medication dose was reduced more in the STN group, meaning that less levodopa rather than a difference in DBS targeting could account for the slightly worsening cognition and greater disease progression. STN subjects also had lower cognitive scores to begin with, suggesting that the STN group only appeared to be at a long-term disadvantage, said Weaver. Stepping back, though, the bottom line is that after three years, the main motor benefit remains equal between the two DBS targets, Weaver said, so more doctors should be open to targeting the GPi with DBS.

Certain baseline conditions might steer a doctor toward either GPi or STN, Weaver suggested. For instance, if a patient has trouble with medication, the doctor may plan to target the STN, which would allow more of a reduction in levodopa dose. Cognitive impairments, on the other hand, might point to a GPi target as the better of the two. "We still don't know which patients do better with one versus the other, but this study starts to point us in that direction," Weaver said. Her colleagues plan to follow these patients in clinical trials for an even longer term—up to nine years after surgery—and Weaver will also conduct a study to tease apart the cost effectiveness of each target.

Regardless of the ideal target, other types of brain stimulation are now under investigation. Getting electricity into the brain by inducing currents with magnetic coils could have its advantages, suggest researchers co-led by Fried. In DBS, electrode tips heat up in magnetic resonance imaging (MRI), limiting use of imaging in those patients. Wires can also cause inflammation and deliver somewhat unfocused signals, according to the authors. Perhaps tiny magnetic coils implanted in the brain could skirt these problems, while still eliciting neuronal action potentials, they suggest.

First authors Giorgio Bonmassar and Seung Woo Lee put tiny coils a few hundred microns above rabbit retinas. When they passed current through the coils to create a magnetic field, it induced an electric field large enough to generate action potentials in the neurons. The bigger the magnetic field, the more spikes were emitted. Interestingly, the orientation of the coil axis (parallel or perpendicular to the neuron axon) could elicit a specific spike response. The coil's small size may allow better focus of electrical activity, suggested the authors. At the same time, the coated coils will not heat up in MRI, and may also prevent glial scarring and inflammation.

"It's an exciting development, one that could be a great research and clinical tool," said Emad Eskandar, Massachusetts General Hospital, Boston. He wonders how much power will be needed to give neuronal stimulation comparable to DBS, but realizes it is still early in the process. Fried says he and colleagues are experimenting with the size and shape of these coils, and are collaborating to test coils in animal models.—Gwyneth Dickey Zakaib


  1. Magnetic induction of current, if done transcranially, is anti-inflammatory. We have studied this for six years in both human and animal models of a number of diseases. The implanted coils are interesting and may prove to be a more successful application for PD.


    . Mitochondria of mice and men: moderate magnetic fields in obesity and fatty liver. Med Hypotheses. 2012 Sep;79(3):287-93. Epub 2012 Jun 26 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. DBS Double Update: Call for Trial Registry, Two Targets Work for PD

Paper Citations

  1. . Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science. 1990 Sep 21;249(4975):1436-8. PubMed.
  2. . Pallidal versus subthalamic deep-brain stimulation for Parkinson's disease. N Engl J Med. 2010 Jun 3;362(22):2077-91. PubMed.

External Citations

  1. Phase 2 trial
  2. clinical trials

Further Reading

Primary Papers

  1. . Randomized trial of deep brain stimulation for Parkinson disease: Thirty-six-month outcomes. Neurology. 2012 Jun 20;
  2. . Turning tables: Should GPi become the preferred DBS target for Parkinson disease?. Neurology. 2012 Jun 20; PubMed.
  3. . Microscopic magnetic stimulation of neural tissue. Nat Commun. 2012;3:921. PubMed.