Surgical neuromodulation has emerged as the primary method to treat the medically refractory symptoms of essential tremor and Parkinson disease. With reversible manipulation of CNS neurons, neuromodulation can be used to intraoperatively localize and verify a stereotactic target, and to chronically treat movement disorders. This article discusses the historical advances in stereotactic surgery using various modalities of neuromodulation leading to contemporary treatment. Electrical neuromodulation, or deep brain stimulation, is emphasized as the major surgical intervention with a discussion of the technique, surgical targets, and clinical outcomes. A comparison of neuromodulation techniques is presented.
Key points
- •
Since the advent of stereotactic neurosurgery, various neuromodulation modalities have been used to confirm target localization.
- •
Electrical stimulation is the most common form of neuromodulation and is highly effective in treating essential tremor and Parkinson disease.
- •
Further experimental refinement of chemical and ultrasound neuromodulation may lead to highly selective and minimally invasive treatments of movement disorders.
- •
Magnetic and ultrasound neuromodulation have the potential for neuromodulation without open cranial surgery.
Introduction
The treatment of movement disorders represents an ideal application for neuromodulation of the central nervous system. Essential tremor (ET) and Parkinson disease (PD) are the most common movement disorders that are treated surgically. Although there are no curative therapies, the symptoms of both diseases are managed initially and effectively with medicine, but severe and refractory impairments occur during disease progression, requiring surgery to maintain quality of life. Over the past 2 decades, surgical neuromodulation has seen a dramatic resurgence in the treatment of ET and PD with the popularization and refinement of high-frequency, deep brain stimulation (DBS).
The critical element for successful DBS, or any stereotactic intervention, is precise anatomic localization and method for physiologic verification. This has significantly improved with modern neuroimaging, stereotactic equipment, and electrophysiology. However, even during the pioneering days of stereotactic neurosurgery when pallidotomy and thalamotomy were common procedures for the treatment of involuntary movements, surgeons recognized the need for precise localization and reversible means to confirm their target. This ensured the best outcomes. Surgeons creatively modulated neural circuitry with a variety of methods, including chemical inhibition, thermoregulation, focused ultrasound, and electrical stimulation.
Modern movement disorder surgery uses neuromodulation in 2 ways. First, it is used intraoperatively to verify the stereotactic target in the thalamus or basal ganglia by measuring the clinical or electrophysiologic responses to electrical stimulation. Second, neuromodulation is used chronically with DBS for treatment ( Figs. 1 and 2 ).
Chemical Neuromodulation
Chemical neuromodulation refers to injection or infusion of a pharmacologically active substance within the nervous system. This application has the theoretical benefit of targeting specific receptors with precise effects. While chemical neuromodulation is principally used in animal models for experimental purposes, there are several studies that have reported chemical neuromodulation in the human brain for movement disorders. In 1955, Cooper published a series of 5 patients in whom he injected procaine into the globus pallidus before creating a permanent lesion with ethanol for tremor in advanced PD. He used roentgenography for initial catheter placement with minor modifications of catheter position based on small volume tests of procaine until the “physiologic landmark” was identified by reduced tremor and rigidity in the contralateral limbs without evidence of motor weakness. He reported improved tremor and rigidity in 6 months of follow up for 3 of these patients. Similarly, Narabayashi and colleagues used early stereotactic methods to inject procaine into the pallidum before permanent lesioning in patients with choreoathetosis. As with Cooper, small volumes of local anesthetic were used to determine whether the site of putative lesioning would be safe. In the series of 80 patients, improvement in athetosis was reported in approximately 60% of patients.
Infusion of local anesthetic was also applied to the thalamus during treatment of tremor. During radiofrequency thalamotomy, Parrent and colleagues first infused 1 to 2 μL of lidocaine in 10 patients with tremor. They observed a transient suppression of tremor with a mean onset of 69 seconds and duration of 171 seconds. Interestingly, the lidocaine infusions correlated with microstimulation effects in 67% of cases.
As the understanding of neurotransmitter systems advanced with the use of selective antagonists and agonists, so did its application to movement disorders. In 1984, Penn and Kroin used intrathecal baclofen, a GABA-B receptor agonist, to alleviate spasticity of spinal origin. Shortly thereafter, they targeted the globus pallidum with muscimol, a GABA-A receptor agonist, during pallidotomy surgery for a patient with PD. Within 20 minutes, bradykinetic movements increased and rigidity resolved, although tremor worsened.
Intraoperative microinjections of muscimol into deep brain nuclei of patients with PD have been reported to transiently inhibit either the globus pallidus internus (GPi) or subthalamic nucleus (STN) neurons. The effect of muscimol infusion elicits a temporary clinical effect that is similar to stimulation or lesioning. Levy and colleagues targeted the STN with muscimol in 7 patients with PD. Modern microelectrode recording techniques were used to confirm target location, and small doses of lidocaine and muscimol were injected into the STN with simultaneous microelectrode recording in 2 patients. Injection of lidocaine blocked nearby neural firing within minutes and improved contralateral limb rigidity with peak effect 10 to 20 minutes after injection. Dyskinesias were noted while blocking the STN with lidocaine. In 2 of 3 patients, these effects wore off during the procedure. Muscimol injection had a similar effect by decreasing tremor in the contralateral limb of both patients tested, and altered the spectrum of a single neuron oscillatory frequency. No adverse events were noted with injection of muscimol. The clinical improvements with muscimol infusion were correlated closely with successful final treatment effects.
Pahapill and colleagues reported infusion of muscimol into the Vim nucleus of patients with ET. Similarly, microelectrode recordings were used to confirm tremor-synchronous neurons in the lateral thalamus, and microelectrode stimulation ceased tremor. Subsequent microinfusion of muscimol reliably reduced tremor with a latency of 7 minutes and for a mean duration of 9 minutes.
Although chemical neuromodulation is principally used in experimental models, these examples in humans are important advances in understanding the basic pathophysiology of movement disorders and its potential application as a therapeutic tool ( Table 1 ).
| Study | n | Target | Drug | Concentration | Dose/Infusion |
|---|---|---|---|---|---|
| Cooper, 1955 | 5 | GPi | Procaine | NS | <250 μL |
| Narabayashi et al, 1960 | 80 | GPi | Procaine | NS | 1–2 μL |
| Parrent et al, 1993 | 10 | Thalamus | Lidocaine | 2% | 1–2 μL |
| Penn et al, 1998 | 1 | GPi | Muscimol | 8.8 mM | 2.5 μL |
| Levy et al, 2001 | 4 | STN | Lidocaine | 2% | 3.5–23 μL |
| 2 | STN | Muscimol | 8.8 mM | 5–10 μL | |
| Pahapill et al, 1999 | 6 | Thalamus | Muscimol | 8.8 mM | 1–5 μL |
| 3 | Thalamus | Saline |
Cryogenic Neuromodulation
During the 1950s, scientists and surgeons were examining the effects of cooling on nervous tissues and function. Results from these experiments showed that cooling various structures of the brain to 0 to 10°C produced a reversible inhibition of neural activity, and that cooling below –20° could create a permanent lesion. In 1961, Mark and colleagues used a refrigeration probe to cool the region of the third nerve nucleus in cats and demonstrated reversible pupillary dilation. Rowbotham and colleagues applied this concept to humans for the treatment of glioma. Cooper published a report of 100 cryothalamotomies for parkinsonism and concluded that the procedure was the ideal technique for movement disorder surgery, as it provides a reversible, physiologic test before the creation of a stable lesion. Although the study does not specifically cite examples of neuromodulation during the course of the target localization, subsequent commentary notes, “I have had the pleasure of seeing Dr Cooper turn a Babinski on and off by adjustment of a valve. This is truly impressive (Cambell JB).”
Thermal Neuromodulation
During the same period as cryogenic thalamotomies, radiofrequency waves were also being used to produce thermal lesions within the brain. Experimental models demonstrated reversible inhibition of neural activity. Using a model similar to Mark and colleagues, Brodkey and colleagues used radiofrequency stimulation to heat the Edinger-Westphal (EW) nucleus in cats. They found that heating the EW nucleus to 44 to 49°C produced a reversible dilation of the pupil that returned to baseline size within 20 minutes of heating. The premise that low-temperature heating can produce reversible lesions in the brain is based on these studies. It is now recognized that low-temperature heating can cause thermal injury depending on the duration of exposure such that tissue ablation even occurs at approximately 43°C when exposed for a duration of 240 minutes.
Ultrasound Neuromodulation
Interestingly, even before the publication of cryothalamotomy and radiofrequency lesioning in the basal ganglia, Fry and colleagues reported the use of high-intensity ultrasound to create destructive lesions of the internal capsule in cats. The goal of their research was to provide neurosurgeons with a tool to perform functional neurosurgery in the treatment of movement disorders. They sonicated the feline lateral geniculate nucleus with lower doses of acoustic energy and temporarily suppressed visual evoked responses recorded at the cortex. These experiments necessitated craniotomy because the skull reflected and absorbed the ultrasound waves. The recent decade has led to advances in ultrasound transducer design so that transcranial delivery of high-intensity ultrasound is possible and precise in humans. Our group observed neuromodulation of the sensory, ventrolateral thalamus in 5 of 15 patients with ET undergoing focused ultrasound thalamotomy although likely from thermal mechanisms. In the laboratory, ultrasound neuromodulation has been demonstrated in vivo in rodents with low-intensity, pulsed parameters and without heating.
Magnetic Neuromodulation
Transcranial magnetic stimulation (TMS) is a noninvasive technique used for measuring and modulating cortical plasticity introduced by Barker and colleagues in 1985. TMS is delivered via an electrical coil placed on the scalp, which generates a magnetic field that traverses the cranium and induces an electrical field in the cortex. This electrical field depolarizes neurons and has been used extensively to measure cortical plasticity in a variety of neurologic disorders. This contrasts with transcranial electrical stimulation where current flow is achieved directly through the skull via leads injected into the scalp. Repetitive pulsing of TMS, known as repetitive transcranial magnetic stimulation (rTMS), has been used in the past 2 decades to modulate cortical excitability in ways that treat neurologic and psychiatric disease. rTMS is currently approved for use in medication-refractory depression in the United States and Canada. It has been studied in neurologic diseases such as PD, tremor, dystonia, tics, spasticity, and epilepsy. High-frequency rTMS (>1 Hz) increases cortical excitability and low-frequency rTMS (<1 Hz) reduces cortical excitability. Paradigms of stimulation based on these observations have driven the design of studies investigating rTMS in movement disorders.
rTMS has been studied extensively for PD motor features with the hypothesis that high-frequency, cortically excitatory stimulation can overcome decreased output from the basal ganglia via the thalamus. Elahi and colleagues performed a meta-analysis of studies with high-frequency and low-frequency rTMS to the motor cortex on PD motor scores. All of the studies had sham-controlled arms and compared the sham group with active groups (either low-frequency or high-frequency rTMS). In the pooled effect, they found a significant reduction in the UPDRS part III (motor) of 6.68 points (95% confidence interval = –9.66 to –3.69) in the high-frequency studies and no significant change on UPDRS part III in the low-frequency studies. It should be noted that the power of the analysis in both paradigms was low. Recent studies have looked at intermittent bursts of very high frequency stimulation to the motor cortex, stimulation of the supplementary motor area, and of the cerebellum and have shown promising results. This is tempered, however, by other studies showing no significant motor benefit of high-frequency stimulation of the motor cortex. Further studies are needed to identify the paradigms and sites of stimulation that may be effective in treating motor features of PD.
Levodopa-induced dyskinesia is a common, disabling feature of PD characterized by excessive, often uncontrollable, movements in the medicated PD state. Use of low-frequency and high-frequency rTMS over the motor cortex, supplemental motor area, and cerebellum have demonstrated mixed results on dyskinesia and UPDRS motor scores.
In essential tremor, the role of excitation of the motor cortex seems promising. Studies have shown that DBS to the Vim increases motor cortex excitability and subdural motor cortex stimulation has shown benefit in ET. Hellriegel and colleagues investigated very high frequency rTMS (50 Hz) over the motor cortex in ET and found significant reduction in tremor as measured by accelerometry. However, this benefit was not appreciated by the study subjects and patient ratings of change were no different between active and sham stimulation. The role of the cerebellum as part of the cerebello-thalamo-cortical pathway in essential tremor has been explored via cerebellar rTMS. Popa and colleagues used 1-Hz bilateral cerebellar stimulation and found significant improvement in tremor amplitude and functional disability due to tremor that was persistent for 3 weeks.
Electrical Neuromodulation: DBS
Shortly after the development of a stereotactic frame applicable to the human skull by Spiegel and Wycis in 1947, many teams of neurosurgeons and neurophysiologists began electrical recordings and stimulation of subcortical structures in the human brain. The principal investigations in these studies were patients with psychiatric disease; however, they quickly moved to movement disorders. Spiegel and Wycis published reports using stereotactic surgery to treat Huntington disease, choreoathetosis, and PD shortly after their description of the stereoencephatome. In these operations, electrical stimulation was used to ensure the electrode was not in an eloquent structure, such as the internal capsule. During these operations, the surgical conditions made it difficult to assess symptoms; however, subsequent surgeries used electrical stimulation of the target to monitor clinical symptoms. It was noted that electrical stimulation of the target could mimic the effects of a lesion. In 1961, Alberts and colleagues found that stimulation of the ventrolateral thalamus or internal segment of the globus pallidus at 60 Hz could evoke or abate tremor. Chronic electrical stimulation of the thalamus and pallidum was also described to locate targets for subsequent lesioning. In 1965, Sem-Jacobsen reported chronic stimulation of the thalamus with multiple implanted electrodes to determine the optimal target for lesioning. In his description, he noted that electrodes could be kept in place for months without complication. In 1972, Bechtereva and colleagues reported chronic electrode placement in the ventrolateral thalamus with intermittent high-frequency stimulation, the results of which were used for later ablative procedures. These and many other studies set the stage for modern DBS. In 1987, Benabid and colleagues published their results of unilateral Vim thalamotomy and unilateral continuous, high frequency Vim stimulation with an implanted electrode in patients with PD. The principal benefits of stimulation compared with lesion are well noted, but include the ability for neuromodulation. These studies demonstrated that tremor was optimally suppressed with higher frequency (>130 Hz) stimulation, and that this suppression could be maintained chronically with implanted neurostimulator devices. This opened the door to electrical neuromodulation of several different subcortical structures previously targeted by lesioning for the treatment of movement disorders.
Introduction
The treatment of movement disorders represents an ideal application for neuromodulation of the central nervous system. Essential tremor (ET) and Parkinson disease (PD) are the most common movement disorders that are treated surgically. Although there are no curative therapies, the symptoms of both diseases are managed initially and effectively with medicine, but severe and refractory impairments occur during disease progression, requiring surgery to maintain quality of life. Over the past 2 decades, surgical neuromodulation has seen a dramatic resurgence in the treatment of ET and PD with the popularization and refinement of high-frequency, deep brain stimulation (DBS).
The critical element for successful DBS, or any stereotactic intervention, is precise anatomic localization and method for physiologic verification. This has significantly improved with modern neuroimaging, stereotactic equipment, and electrophysiology. However, even during the pioneering days of stereotactic neurosurgery when pallidotomy and thalamotomy were common procedures for the treatment of involuntary movements, surgeons recognized the need for precise localization and reversible means to confirm their target. This ensured the best outcomes. Surgeons creatively modulated neural circuitry with a variety of methods, including chemical inhibition, thermoregulation, focused ultrasound, and electrical stimulation.
Modern movement disorder surgery uses neuromodulation in 2 ways. First, it is used intraoperatively to verify the stereotactic target in the thalamus or basal ganglia by measuring the clinical or electrophysiologic responses to electrical stimulation. Second, neuromodulation is used chronically with DBS for treatment ( Figs. 1 and 2 ).
Chemical Neuromodulation
Chemical neuromodulation refers to injection or infusion of a pharmacologically active substance within the nervous system. This application has the theoretical benefit of targeting specific receptors with precise effects. While chemical neuromodulation is principally used in animal models for experimental purposes, there are several studies that have reported chemical neuromodulation in the human brain for movement disorders. In 1955, Cooper published a series of 5 patients in whom he injected procaine into the globus pallidus before creating a permanent lesion with ethanol for tremor in advanced PD. He used roentgenography for initial catheter placement with minor modifications of catheter position based on small volume tests of procaine until the “physiologic landmark” was identified by reduced tremor and rigidity in the contralateral limbs without evidence of motor weakness. He reported improved tremor and rigidity in 6 months of follow up for 3 of these patients. Similarly, Narabayashi and colleagues used early stereotactic methods to inject procaine into the pallidum before permanent lesioning in patients with choreoathetosis. As with Cooper, small volumes of local anesthetic were used to determine whether the site of putative lesioning would be safe. In the series of 80 patients, improvement in athetosis was reported in approximately 60% of patients.
Infusion of local anesthetic was also applied to the thalamus during treatment of tremor. During radiofrequency thalamotomy, Parrent and colleagues first infused 1 to 2 μL of lidocaine in 10 patients with tremor. They observed a transient suppression of tremor with a mean onset of 69 seconds and duration of 171 seconds. Interestingly, the lidocaine infusions correlated with microstimulation effects in 67% of cases.
As the understanding of neurotransmitter systems advanced with the use of selective antagonists and agonists, so did its application to movement disorders. In 1984, Penn and Kroin used intrathecal baclofen, a GABA-B receptor agonist, to alleviate spasticity of spinal origin. Shortly thereafter, they targeted the globus pallidum with muscimol, a GABA-A receptor agonist, during pallidotomy surgery for a patient with PD. Within 20 minutes, bradykinetic movements increased and rigidity resolved, although tremor worsened.
Intraoperative microinjections of muscimol into deep brain nuclei of patients with PD have been reported to transiently inhibit either the globus pallidus internus (GPi) or subthalamic nucleus (STN) neurons. The effect of muscimol infusion elicits a temporary clinical effect that is similar to stimulation or lesioning. Levy and colleagues targeted the STN with muscimol in 7 patients with PD. Modern microelectrode recording techniques were used to confirm target location, and small doses of lidocaine and muscimol were injected into the STN with simultaneous microelectrode recording in 2 patients. Injection of lidocaine blocked nearby neural firing within minutes and improved contralateral limb rigidity with peak effect 10 to 20 minutes after injection. Dyskinesias were noted while blocking the STN with lidocaine. In 2 of 3 patients, these effects wore off during the procedure. Muscimol injection had a similar effect by decreasing tremor in the contralateral limb of both patients tested, and altered the spectrum of a single neuron oscillatory frequency. No adverse events were noted with injection of muscimol. The clinical improvements with muscimol infusion were correlated closely with successful final treatment effects.
Pahapill and colleagues reported infusion of muscimol into the Vim nucleus of patients with ET. Similarly, microelectrode recordings were used to confirm tremor-synchronous neurons in the lateral thalamus, and microelectrode stimulation ceased tremor. Subsequent microinfusion of muscimol reliably reduced tremor with a latency of 7 minutes and for a mean duration of 9 minutes.
Although chemical neuromodulation is principally used in experimental models, these examples in humans are important advances in understanding the basic pathophysiology of movement disorders and its potential application as a therapeutic tool ( Table 1 ).
| Study | n | Target | Drug | Concentration | Dose/Infusion |
|---|---|---|---|---|---|
| Cooper, 1955 | 5 | GPi | Procaine | NS | <250 μL |
| Narabayashi et al, 1960 | 80 | GPi | Procaine | NS | 1–2 μL |
| Parrent et al, 1993 | 10 | Thalamus | Lidocaine | 2% | 1–2 μL |
| Penn et al, 1998 | 1 | GPi | Muscimol | 8.8 mM | 2.5 μL |
| Levy et al, 2001 | 4 | STN | Lidocaine | 2% | 3.5–23 μL |
| 2 | STN | Muscimol | 8.8 mM | 5–10 μL | |
| Pahapill et al, 1999 | 6 | Thalamus | Muscimol | 8.8 mM | 1–5 μL |
| 3 | Thalamus | Saline |
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
