Parkinson’s Disease Application

Charles B. Mikell, Bradley Ashcroft

Abstract

Deep brain stimulation (DBS) can be used to alleviate the motor and cognitive symptoms of Parkinson’s disease (PD). The therapeutic benefit of DBS is equivalent to both the maximal “ontime” and symptom reduction conferred by classic dopamine drugs, and it allows a reduction in the use of medication to manage symptoms. Several structures have been studied and used as targets, chief among them being the subthalamic nucleus and the globus pallidus pars interna. Mutliple techniques can be utilized to target these structures. Frame-based and frameless approaches can be used to guide the implant, while different magnatic resonance imaging modalities can be combined with microelectrode recording to identify and confirm the correct target and electrode placement. DBS offers a therapeautic benefit and should be considered for PD patients whose symptoms are no longer well controlled by medication.

Keywords: Parkinson’s disease, deep brain stimulation, subthalamic nucleus, globus pallidus interna, frame-based, frameless, targeting

“The thing about Parkinson’s disease is, it gets worse.”

– Guy Schwartz, MD

8.1 Introduction

Parkinson’s disease (PD) is a devastating degenerative illness that affects movement, gait, and thinking of a patient. As of 2010, more than 630,000 Americans suffered from PD; the number is projected to double by 2040.1 Neurologists usually treat PD patients with drugs that augment dopamine transmission; levodopa-based agents are generally first-line treament for most patients.² Although these agents are effective in a majority of patients with PD, the response to these agents starts declining with time or the benefit is tempered due to intolerable side effects associated with increased dosing requirements. The major motor symptoms of PD include rigidity, tremors, bradykinesia, and gait and posture abnormalities. In addition, many nonmotor symptoms can manifest, such as depression/anxiety, fatigue, worsening sense of smell, and cognitive changes. Patients whose symptoms are responsive to dopamine drugs yet are not well controlled or patients experiencing worsening on/off periods and/or receiving ineffective doses should be considered as surgical candidates.

Although originally considered for patients with severe PD, recent evidence suggests that movement disorder surgery should be considered relatively early in the course of the disease. The Medtronic Deep Brain Stimulation (DBS) system is now Food and Drug Administration (FDA) labeled for patients who have had 4 years of symptoms with motor fluctuations that are not controlled, in addition to patients with longer-duration disease. This labeling was updated in response to data from the EARLYSTIM trial in which patients with 4 years of disease were randomized to surgery or best medical therapy.³ Patients who underwent surgery had gains in quality of life, as assessed by the Parkinson’s Disease Questionnaire (PDQ-39),^3,4 whereas patients in the medical therapy arm remained the same or their disease progressed. Because of this improvement in quality of life, DBS should be considered as soon as movement symptoms are not fully controlled with medication; it is only going to get worse.

8.2 Patient Selection

PD patients are generally older and may suffer from multiple comorbidities. Some younger patients are also surgical candidates, but these are the exceptions, given the increasing incidence of PD with age.5 In general, DBS is not a highly morbid surgery, and some very elderly patients are candidates for surgery. There is no age cutoff in our opinion, though treatment decisions have to be individualized based on discussions between the patient, his or her family, and a multidisciplinary team including the neurologist, surgeon, and other caregivers (i.e., neuropsychologist, physical therapist). Treatment decisions should be informed by Dr. Schwartz’s truism: the disease is only going to get worse.

The most important consideration in determining candidacy for DBS is whether the patient’s symptoms respond to dopaminergic medication. Symptoms that respond to carbidopa/levodopa, or dopamine agonists can generally be expected to improve with surgery. Cardinal symptoms of PD, including rigidity, tremor, and bradykinesia, almost always respond to dopamine equivalents and can, therefore, be counted upon to respond to surgery.⁶ Tremor can be a relative exception to this general rule as tremor is often quite responsive to DBS even when resistant to medical therapy. However, gait symptoms are more complicated. According to early reports, stride length and velocity improve with DBS without benefit for freezing.^7,8 More recent evidence suggest that high-frequency DBS may worsen freezing, while low-frequency (60 Hz) DBS improves it.^9,10 We have found that patients whose gait symptoms respond to medication will generally improve with stimulation, though we inform the patient of the uncertainty during the informed consent discussion. We relate our experience, while acknowledging that literature on the topic does not show a consistent benefit for gait with surgery. Finally, voice symptoms are a special case; we have seen both amelioration and aggravation of hypophonia. The response to DBS in this case is not clearly linked to prior response to medication. One report used cluster analysis to identify subtypes of voice dysfunction after DBS; while some patients have dysarthria related to corticobulbar stimulation, patients with low volume actually benefit from subthalamic nucleus (STN) DBS.¹¹ One group has hypothesized that this variability has to do with variations in the location of the dentatorubrothalamic tracts in DBS patients.¹² As a general rule, symptoms that get better with medication also get better with surgery.

Due to the high efficacy of DBS in the treatment of PD, DBS should be considered in dopamine-responsive PD patients as soon as motor symptoms are not fully controlled with medication or side effects become bothersome, because the disease is only going to get worse. To determine responsiveness to dopaminergic medication, most centers score the patient with the United Parkinson’s Disease Rating Scale (UPDRS) when the patient is both on and off medication; a decrease of 30% in the UPDRS-III is considered to be appropriate for consideration of surgery.13 Of note, tremor symptoms are often well controlled by stimulation, but may respond to levodopa only at very high or poorly tolerated doses, and thus some tremor-predominant PD patients may be excellent candidates for DBS, without a 30% decrease in the UPDRS.¹⁴

According to the FDA-approved labeling of both of the commercially available devices, patients should have experienced 4 years of symptoms prior to surgery. However, at present it is more common that a patient is referred for surgery after living with the disease for many years. Though it would be unusual to consider surgery prior to 4 years of symptoms, we would consider offering surgery after explaining to the patient that this operation would constitute off-label therapy. We would also mention the caveat that Parkinson’s-plus syndromes, distinguished by motor symptoms similar to PD that are unresponsive to dopamine drugs, as well as prominent autonomic or cognitive dysfunction are often only obvious after a few years of disease, and these syndromes do not respond well to DBS therapy. In addition, we do not think there is current evidence that DBS modifies the course of PD or is neuroprotective in human subjects, despite promising animal data.^15,16 Nonetheless, most patients with motor symptoms that were previously well controlled with medications but are not fully controlled with medicine now or those who are experiencing on/off fluctuations would benefit from surgery.

Despite the general success of DBS in PD, a number of absolute and relative contraindications exist. The only truly absolute contraindication is the presence of symptoms that do not respond to dopamine equivalents. This is a hallmark of Parkinson’s-plus syndromes such as multiple system atrophy (MSA) and progressive supranuclear palsy. To establish whether and to what extent carbidopa/levodopa is helpful for the patient being considered for surgery, the treating physicians should judge whether the patient has been followed by a neurologist long enough to establish the diagnosis. Prominent dementia and autonomic symptoms (characteristic of MSA) may take several years to declare themselves. However, fainting and orthostatic hypotension, which may be present in MSA, are also common in PD, and may represent a side effect of carbidopa/levodopa more than a true symptom. In these cases, we do not hesitate to implant, though we may allow the disease to “play out” a bit longer to see if other symptoms of Parkinson’s-plus syndromes appear. Nonetheless, for a first approximation, a weak or no response to dopamine should be considered an absolute contraindication to DBS.

Relative contraindications to DBS include cognitive impairment, medical comorbidities, and frailty. Cognitive impairment may be exacerbated by DBS. Some prior reports suggested globus pallidus pars interna (GPi) DBS is preferred in patients with cognitive problems; however, a recent meta-analysis suggested that although declines occur across a broader spectrum of cognitive functions with STN DBS, the treatments are largely similar in their cognitive effects.¹⁷ Many centers obtain preoperative neuropsychological assessments to identify patients at serious risk for cognitive decline. Serious medical comorbidities may make the surgery itself dangerous. Finally, general frailty, which can be defined as general poor health, vulnerability to poor wound healing and infection, and limited physiological reserve aside from a disease process,^18,19 may limit the patient’s ability to tolerate a lengthy procedure. Frail patients tend to have a more challenging and prolonged postoperative course, and they may require discharge to inpatient rehabilitation. Notably, independent of frailty, we do not consider age a contraindication as both young and old PD patients can be frail; thus, it is more useful to consider frailty than age. Frailty predicts surgical morbidity, mortality, and discharge institutionalization.^20,21,22 These are relative and not absolute contraindications, and should be discussed with the patient and his or her caregivers as part of the informed consent discussion. An important consideration in this discussion should also be the lack of other effective treatment modalities.

8.3 Goals of Treatment

After it has been determined that a patient is a suitable candidate for DBS, clinicians should communicate clearly the goals of surgery. We explain that the most important goal is prolonging the “on time,” in which the patient can do the things he or she needs to do. In large, randomized trials, DBS improves “on time” by 4 hours a day.23,24 With STN surgery, medication doses can usually be decreased by 30 to 50% on average, while GPi surgery allows for a lesser decrease;^24,25 however, there is still a marked increase in “on time” and an equivalent UPDRS benefit. With all surgical targets, patients should expect improvements in all of the symptom domains that are treated by medication, including rigidity, bradykinesia, and tremor. An important caveat for patient expectations is that the patients are typically as good after surgery as their “best on.” Surgery does not cure PD, and it is important to emphasize that the patient may still expect significant disability. In addition, the nonmotor symptoms of PD are typically not well treated with DBS. In sum, we try to underpromise and overdeliver.

A common question is how long the treatment effect is expected to last. In long-term follow-up studies, patients continue to benefit from DBS for many years, and are generally doing better at 5 or even 10 years of follow-up than patients treated with medical therapy.^26,27 However, the disease does continue to progress (“it gets worse”) and the severity of motor symptoms will inevitably increase. Non-dopamine-responsive symptoms, such as cognitive and autonomic problems, may contribute to disability more significantly in the later course of the disease. In general, the goal of treatment is to treat dopamine-responsive symptoms for a prolonged period, even though the disease progresses. However, each patient will have his or her own goals and these must be taken into consideration in the surgical plan.

8.4 Target Selection

Selection of a surgical target depends on the surgeon’s and the patient’s personal goals for treatment. The best-characterized targets for DBS in PD are STN and GPi. Though randomized trials have not demonstrated superiority of one target over the other, many neurologists favor STN implantation simply because many patients prefer to take less medication. GPi implantation, however, does have several advantages. Some trials have suggested that GPi implantation has less cognitive morbidity than STN implantation,24,28 though this evidence has been called into question.¹⁷ In addition, GPi is surrounded by fewer critical structures than STN, thus making implantation and programming more forgiving. Also, unilateral GPi implants are more beneficial than unilateral STN implants, so frail patients who may not very well tolerate bilateral surgery may benefit from staged (or unilateral) GPi implantation.²⁹ Other relative advantages and disadvantages of STN and GPi are summarized in Table 8.1.

Table 8.1 Relative advantages and disadvantages of the STN and GPi as implantation targets for DBS treatment of PD 30

Target	Advantages	Disadvantages
Subthalamic nucleus	Surgically safe with proper targeting	Surrounded by more critical structures
	Robust visualization via 3 Tesla MRI and microelectrode recording	More difficult to program
	Greater reduction of dosing of concurrent medication	Smaller size compared to GPi
	More commonly targeted due to slightly greater improvement in motor output	Generally requires bilateral implantation to achieve therapeutic advantage
	Surgeons are typically more familiar with the STN	Risk of adverse neurocognitive effects
	Very clear intraoperative effects	Risk of voice impairment
Globus pallidus pars interna	Equivalent of on-time benefit compared to STN	Significant motor improvement, though slightly better in STN vs. GPi
	Surrounded by fewer critical structures	Less familiar to some surgeons
	Larger size and easier to target for programming as compared to STN	Lesser reduction in dosing of medication
	Unilateral GPi implantation may offer greater therapeutic outcome compared to unilateral STN implantation	Evidence for superior cognitive outcome relative to STN implantation may be questionable and limited
	May be more appropriate for the elderly and frail patients

Abbreviations: DBS, deep brain stimulation; GPi, globus pallidus pars interna; MRI, magnetic resonance imaging; PD, Parkinson’s disease; STN, subthalamic nucleus.

8.4.1 Less Frequently Used Targets

VIM thalamus

Tremor control is the chief benefit of stimulation of the ventral intermediate (VIM) nucleus of thalamus.31,32 However, STN and GPi surgery also control tremor in addition to rigidity and bradykinesia, so VIM surgery for Parkinson’s is second line.

Zona incerta/posterior subthalamic area

This area lies between the posterior STN and the red nucleus, and stimulating it has excellent benefit for control of tremor;³³ at least one randomized trial comparing zona incerta stimulation with VIM stimulation is ongoing.³⁴ However, several groups have reported benefit in bradykinesia and rigidity from stimulation of this area.^35,36,37 A larger randomized trial is required to confirm efficacy.

Centromedian/parafascicular nucleus

In early studies, lesions of the central nuclei of the thalamus lead to improvements in rigidity and tremor.³⁸ DBS of the centromedian nucleus improves freezing of gait³⁹ and may also treat dyskinesia.⁴⁰ Experience is limited to very small series, however.

Pedunculopontine nucleus

The pedunculopontine nucleus is a cholinergic brainstem nucleus that has been proposed as a DBS target in PD for reducing gait-related symptoms, especially freezing.⁴¹ Early reports of unilateral and/or bilateral stimulation of this region, some using ostensibly stimulatory low-frequency settings, describe improvement in freezing.^42,43,44 However, a recent systematic review described wide variation in implantation techniques and settings, as well as variability of response, and ultimately expressed skepticism about proposed benefit.^45,46 Further studies and/or registries of implanted patients may be useful in understanding the role of this target.

Nucleus basalis of Meynert

Recently, a small clinical trial explored stimulation of the nucleus basalis of Meynert for the alleviation of dementia symptoms in PD.⁴⁷ Despite a slight improvement in the patients’ Neuropsychiatric Inventory, the study found no significant difference in cognitive improvement. Nonetheless, implantation surgery and stimulation were both well tolerated in all patients in this and other trials. Cognitive decline is otherwise untreatable, so further studies on this target will probably be conducted in PD and other dementing illnesses.^47,48

8.5 Benefits of DBS

As mentioned above, the benefit of DBS mostly consists of improvements in “on time.” Patients should be counseled that DBS will only make them as good as their “best on,” and that they should still expect significant disability from PD. There are also benefits for other dopamine-responsive symptoms of the disease, such as tremor, rigidity, and bradykinesia. Medication-induced dyskinesias, as well as other medication side effects (such as orthostatic hypotension and constipation), can also improve after DBS as medication doses go down. Some general benefits of DBS are listed in Table 8.2.

Table 8.2 General benefits for the use of DBS in the treatment of PD

Benefits of DBS treatmentfor PD
Provides significant improvement to combined motor scores	Allows a reduction in concurrent medications: reduction of side effects, reduction of variable drug efficacy
Increases mobility	Less drug-induced dyskinesia
Rapidly and reliably achieves the threshold of maximum benefit conferred by dopamine drugs	Electrical stimulation works synergistically with chemical stimulation from drugs to enhance therapeutic effect
Improves the length of maximum functional benefit (on-time) compared to dopamine drugs	Direct electrical stimulation may be neuroprotective, though data is limited

Abbreviations: DBS, deep brain stimulation; PD, Parkinson’s disease.

8.6 Risks of DBS

The risks associated with DBS can be grouped into immediate surgical risks and long-term issues. Immediate risks of DBS include bleeding with the electrode pass that may result in permanent neurological deficits (0.7–7%),49 infection (2.2–8%),^50,51 perioperative seizure, and a confusional state characterized by low arousal, poor cognitive function, and balance problems for the perioperative period (2–20%).^24,52 We counsel patients that the first three risks are on the order of low, single-digit percentages in the literature. There is some evidence that number of microelectrode passes correlates with hemorrhage, but the data are equivocal.⁵³ Frail patients and patients with significant cortical atrophy should be counseled that surgery may require a longer recovery period relative to other patients, and they may require discharge to acute rehabilitation. In the long term, these patients generally do well.

Long-term risks of DBS include hardware complications, such as infection and erosions, and side effects from stimulation. The leads and lead extensions do occasionally fracture and require revision.^54,55 Reported risks include: lead fracture (1.4–3%), lead erosion/infection (1.7–10%), and lead migration (3–12%).^55,56,57 Implanted pulse generator (IPG) malfunctions are distinctly rare. More commonly, prominent scars can develop around the leads and cause a restriction of head movement termed as “bowstringing.”⁵⁸ This condition is often managed conservatively but can also require revision. Finally, stimulation can cause untoward side effects, such as face pulling (if the electrode is near the internal capsule or other critical structures), and these can be dose limiting. As the disease progresses, patients’ stimulation needs may increase and dose-limiting capsular effects can become a problem, even if they were not present in the starting. In rare instances, the electrode may require revision to a more medial or posterior position. Tantalizingly, the availability of directional stimulation may prevent the need for such revisions, but no data are available on this yet.⁵⁹

8.7 Techniques

DBS implantation techniques may be frame-based or frameless, and may or may not use microelectrode recording (MER). At our institution, we combine a frameless technique (STarFix) with MER, and we do simultaneous bilateral recording and implantation. However, surgeon preferences vary widely, and there are advantages and disadvantages of each technique. We have described all techniques in the subsequent text.

8.7.1 Frame-based Implantation

Most neurosurgeons use a frame-based implantation technique, utilizing either the Cosman-Robert-Wells (CRW) or Leksell frames. Both frames must be applied preoperatively, and then a localizer scan must be obtained. There is some discomfort with both head frames which varies widely between patients. Additionally, some patients want that the frame be applied under general anesthesia, which may prolong the OR time. Nonetheless, head frames are simple to use, accurate, and reliable.

8.7.2 Frameless Implantation

Several frameless implantation techniques are available. The Medtronic Nexframe™ system is a bone-fiducial based system in which a plastic tower attaches to bone anchors which are also used for stereotactic registration. Nexframe can be combined with MER,60 intraoperative MRI,⁶¹ or CT⁶² to optimize lead location. While the Nexframe was used for the original intraoperative MRI-based implantations, the newer system Clear-Point™ (MRI Interventions, Irvine, CA) has taken over the bulk of MRI-guided DBS placements. ClearPoint is also a skull-mounted device that is designed for intraoperative imaging inside the MRI environment. It can be used in both “intraoperative” MRIs, as well as regular, closed-bore MRIs.⁶¹ Finally, the STarFix™ platform (FHC Inc., Bowdoin, ME) is another bonefiducial based system which uses 3D-printed single-use frames for targeting.⁶³ We use STarFix, for several reasons: low upfront capital investment, ease of use, and capacity to perform simultaneous bilateral implantation.

8.7.3 Stereotactic Targeting: Subthalamic nucleus

The STN is the most common DBS target in PD. Surgery depends on precise placement of the electrode in the dorsolateral portion of the STN, which is usually accomplished with a combination of image-based targeting and MER. Several methods are available for targeting.

Direct (image-guided) targeting

Modern, 3-Tesla MRI scanners are usually able to visualize the STN as an almond-shaped, T2-hypointense structure, just lateral to the red nucleus ( Fig. 8.1).⁶⁴ Reports of targeting STN with magnetic susceptibility-weighted imaging are also available as it is high in iron content.^65,66 In both of these reports, as well as another description of direct targeting,⁶⁷ the authors also used MER. This suggested that the targeting was reasonably robust since the correct location could be confirmed through two reliable methods.

AC-PC coordinate-based targeting

Many neurosurgeons use coordinate-based targeting from a line drawn between the anterior and posterior commissures. Generally, the STN is targeted 11 to 12 mm lateral to the midcommissural point, 3 mm posterior, and 4 mm inferior.⁶⁸ This approach is time-honored but dates back to the time before advanced neuroimaging was available. This approach is generally combined with MER,⁶⁸ though it may also be combined with impedance measurement and macrostimulation to confirm targeting.^69,70