Deep brain stimulation (DBS) for the treatment of Parkinson’s disease and other movement disorders

Research to understand the brain circuitry underlying movement-related symptoms of PD and other disorders contributed significantly to the development of DBS and can be traced back several decades.

Early surgical treatments

In the 1950s, various movement disorders including PD, essential tremor, and dystonia, were treated by surgically inactivating, or lesioning, brain regions involved in motor control. Surgeons used a frame called a stereotactic apparatus¹, which had been developed in 1947, to target specific areas based on knowledge of brain anatomy and function at the time and refined through trial and error. Surgical targets for PD and essential tremor included parts of a central brain structure called the thalamus as well as the globus pallidus (GP), which was serendipitously identified as an effective target when it was damaged by accident during a procedure to lesion a different area^2,3. Overall, surgical lesions improved motor symptoms for many patients, though sometimes at the expense of irreversible deficits in other functions.

Neurosurgeons performing these operations used electrodes for brain stimulation in awake patients to define the region to inactivate and limit unintended side effects. As early as the 1960s, a number of reports noted that high frequency stimulation of target regions tended to mimic surgical lesions, while lower frequency stimulation tended to make motor symptoms worse⁴. Neurophysiologist Natalia Bechtereva of the former Soviet Union was among the first to suggest in the 1970s, that brain stimulation might itself be used as a treatment for movement disorders, instead of permanent lesions^4,5. However, another decade would pass before technological advances would make chronic stimulation suitable for broad clinical application. Moreover, medical treatment with L-DOPA⁶ soon eclipsed neurosurgery as the preferred therapy for PD.

Modeling brain circuits for movement

Although interest in ablative surgery for PD waned, researchers were learning more about brain circuits for motor control through studies in animal models. In the 1970s, Mahlon DeLong, then a research fellow at NIH working with scientists at NIMH and NINDS, used electrical stimulation to meticulously characterize the functions of neurons in different brain areas as animals performed movements^7,8,9,10. Over a period of several years and continuing after DeLong moved to Johns Hopkins University School of Medicine, these NIH and NINDS-supported studies led to a model of basal ganglia circuit organization, with parallel, functionally segregated circuits involved in movement and other complex functions^11,12. As part of the movement-related circuit, the subthalamic nucleus (STN) modulates the GP, which sends signals through the thalamus to the supplementary motor area (SMA), a brain area important for planning and coordinating movement. By this time, researchers knew that the loss of dopamine-producing neurons in the basal ganglia contributed to PD, but how such changes disrupted motor circuit activity and function was still not clear, in part because no animal model of the disease was available.

That changed in the 1980s, after reports described how a few people succumbed to Parkinson’s-like symptoms after taking an illicit drug. NIH scientists published an initial case¹³, and investigations into more cases by J. William Langston and colleagues in California¹⁴ determined that the cause was an impurity known as MPTP (1-methyl-4-phenyl-1,2,5,6-tetrahydropyridine), which selectively killed some of the same neurons lost in PD. Out of this unfortunate event came the opportunity to develop an MPTP-induced animal model of PD, described by NIH scientists¹⁵ in 1983. DeLong and others used this model to understand how the disease leads to motor symptoms and to probe the effects of inactivating parts of the basal ganglia circuit^16,17. Guided by the circuitry map he helped define, DeLong lesioned the STN in MPTP-treated monkeys and achieved dramatic results: a rapid reduction in PD-like motor symptoms, including tremor, rigidity, and reduced voluntary movement¹⁸. Studies by others produced similar results, using either surgical lesions or high frequency stimulation in the STN^19,20. Together, the findings altered prevailing assumptions about how circuit abnormalities might drive motor symptoms, and importantly, provided scientific rationale for therapy targeting the STN.

Meanwhile, medical device technology had advanced rapidly in the years since L-DOPA’s introduction, and improved electrical pulse generators had developed alongside the first implantable cardiac pacemakers. The young device company Medtronic began a neurological division²¹ in 1975, building on early uses of modified pacemakers for neurological indications. For example, NINDS-supported investigators were among the first to use an implanted device for deep brain stimulation (DBS) in the thalamus as a treatment for chronic pain²². Around this time, the limitations of L-DOPA were becoming apparent, and as neurosurgeons sought to improve on past surgical approaches for PD and other motor disorders, some also began exploring the use of DBS.

DBS becomes a clinical success

During a series of surgeries to treat patients with PD and other motor disorders who were not helped by L-DOPA, French neurosurgeon Alim-Louis Benabid observed that applying high frequency stimulation to the ventral intermediate nucleus of the thalamus (VIM) caused the patients’ tremor to cease²³. Additional small trials^24,25 and Medtronic-funded multisite trials in the US and Europe confirmed the treatment’s effectiveness and showed that thalamic stimulation was safer than surgical inactivation. In 1997, the FDA granted approval for Medtronic’s device for VIM-DBS to treat essential tremor and tremor associated with PD²⁶.

While DBS in the thalamus reduced tremor, it did not reverse other symptoms reported by patients as even more debilitating, including rigidity and slowed or reduced voluntary movement, also called akinesia. Neurosurgeons therefore began to try DBS in other areas. In 1994, Swiss neurosurgeons Jean Siegfried and Bodo Lippitz reported that chronic stimulation in the internal segment of the GP (GPi) improved symptoms in three patients with advanced PD²⁷. Prompted by DeLong’s basal ganglia circuit models and observations on inactivating the STN in MPTP-treated monkeys, Benabid’s team applied DBS to the STN in patients with PD. Stimulating the STN improved tremor, rigidity, and akinesia, yielding their best results in patients yet^28,29, and successfully bridging basic and clinical neuroscience research. Based on results from a large, multisite clinical trial³⁰, the FDA approved DBS in the STN or GPi to treat motor symptoms in advanced PD in 2002³¹. A year later, the FDA granted a Humanitarian Device Exemption (HDE) allowing the use of DBS in the GPi and STN for dystonia³².

Subsequent studies have continued to assess long term outcomes of DBS, optimize targets for different symptom profiles, and compare the effectiveness of DBS with medical treatment. In 2009, a pivotal clinical trial jointly supported by NINDS, the Department of Veterans Affairs, and Medtronic demonstrated that DBS for PD was superior to treatment with L-DOPA³³. With growing evidence for the safety of DBS and results suggesting earlier intervention in PD may be beneficial, the FDA expanded approval beyond advanced stages of PD in 2016, to allow treatment in patients diagnosed for at least four years who experience troublesome “off” periods or dyskinesia³⁴.

In recognition of their pioneering work to develop DBS for PD, Benabid and DeLong shared the 2014 Lasker-Debakey Clinical Medical Research Award³⁵. Besides motor control, DeLong’s work to understand the functional organization of the basal ganglia also pointed to separate parallel circuits involved in cognitive and emotional functions. Accordingly, researchers are examining the use of DBS to alter basal ganglia circuit activity to treat psychiatric disorders, such as obsessive-compulsive disorder, for which the FDA approved DBS through an HDE in 2009³⁶. DBS devices that target other brain areas are also approved and under development for additional neurological disorders, including epilepsy and chronic pain. Moreover, further innovations are emerging with advances in neuroscience and technology. For example, while traditional DBS delivers constant stimulation, newer adaptive devices can self-tune stimulation in response to certain features of a person’s brain activity or behavior. One such closed-loop device is approved for the treatment of medically refractory epilepsy, and research is underway to apply this approach to DBS for PD and other disorders as well³⁷. More broadly, questions remain about exactly how DBS works, and new directions are likely to emerge through research on the mechanisms that underlie its benefits.