Neuromodulation Options and Patient Selection for Parkinson's Disease
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.302473
Source of Support: None, Conflict of Interest: None
Keywords: Parkinson's disease; neuromodulation; deep brain stimulation; infusion therapies; lesioning
The Global Burden of Disease study (2018) estimated that Parkinson's disease (PD) prevalence has risen exponentially over the past two decades, from 2.5 million patients in 1990 to 6.1 million patients in 2016. Dopaminergic medications offer considerable symptomatic benefits in the early stages of PD. As PD progresses, motor complications related to levodopa replacement hamper the sustained benefit and quality of life for patients. Neuromodulation therapies, including Deep Brain Stimulation (DBS) and pump therapies, are currently the standard of care for PD patients with advanced disease and motor complications that are difficult to control with medical management alone., The quest for alternate approaches led to the development of several novel strategies for continuous dopaminergic stimulation, including intrajejunal levodopa infusions (IJLI) and continuous subcutaneous apomorphine infusions (CSAI) [Figure 1]. More recently, innovative methods for non-invasive neurostimulation were developed and applied in PD, leading to FDA approval for magnetic resonance-guided focused ultrasound (MRgFUS) ablation to treat tremor in PD. While the traditional non-invasive cranial stimulation strategies like Transcranial Magnetic Stimulation (TMS) are largely limited to research use to understand the physiological processes involving neuromodulation, promising short-term results have been reported following these modalities on specific domains of PD. Furthermore, the past few years have witnessed exciting progress with spinal cord stimulation, vagal nerve stimulation and near-infrared stimulation in various stages of development. For the practicing clinician, these have opened up a repertoire of strategies that may potentially benefit the patient. In this review, we will provide an overview of the neuromodulation strategies currently available for PD, emphasizing on patient selection and choosing among the various strategies.
Deep brain stimulation
DBS involves delivering electrical stimulation through intracranial electrodes implanted into a target nucleus and connected to an implanted pulse generator (IPG) located extracranially. The implanted electrodes (one on either side) have several contacts interspersed at regular intervals at their tips. During post-operative programming, each contact can be activated independently and programmed to deliver current of required pulse width, frequency, and amplitude. Accurate implantation of electrodes into the target nucleus is achieved by a three pronged strategy consisting of (1) Direct or indirect visualization of the target nucleus on cranial MRI and stereotactic localization, (2) Analysis of microelectrode recordings (MER) obtained intra-operatively to identify the borders of the target nucleus and (3) macrostimulation and clinical assessment of benefits and adverse events during surgery. The STN and GPi are the nuclei that are commonly targeted for DBS in PD, with majority of the centers targeting the STN.,
Pre-operative levodopa responsiveness is the strongest predictor of good outcomes following DBS, and in general it is the levodopa responsive symptoms that improve after surgery. Tremor is an exception and that may respond to DBS even if levodopa unresponsive. Gait and axial symptoms are generally considered to be significantly unaltered by DBS; levodopa responsive gait disturbances including freezing of gait may respond to STN-DBS, though the effects maybe ill-sustained beyond the initial years. The benefits of DBS on the quality of life maybe lesser in patients who are severely disabled in the ON time (other than disability attributable to dyskinesias or tremor), those with cognitive impairment, advanced age, levodopa unresponsive gait and axial symptoms, significant autonomic dysfunction and poor social support., In general, these factors are therefore considered to be relative contraindications to DBS [Table 1]. It is important to note that DBS is only a symptomatic therapy aimed at controlling the motor symptoms and improving the patient's quality of life without any proven effect on the progression of neurodegeneration.,
Advances in imaging and lesioning technologies in the last decade have rekindled interest in lesioning for the treatment of many movement disorders. The limited resources in some centers to afford DBS costs and the difficulty for some patients to travel from remote areas to be monitored regularly makes lesioning therapy a valuable option.
Lesioning surgery can be done using several techniques:
The commonly targeted structures for lesioning in PD include the thalamus (VIM), pallidum (GPi) and STN. For thalamotomy, the VIM nucleus is considered the best target, with excellent short- and long-term tremor suppression in 80-90% of patients with PD. Especially with bilateral interventions, the rate of adverse events is higher, they are associated with a higher risk of permanent dysarthria and imbalance. For pallidotomy, the antidyskinetic posteroventrolateral part of the GPi is targeted most frequently. Pallidotomy studies have demonstrated significant improvements in the cardinal symptoms of PD (tremor, rigidity, bradykinesia), as well as a significant reduction in dyskinesia. Both RFA and Gamma-knife thalamotomy seems to result in comparable effects on both dyskinesia (83.3% and 86.6% improvement, respectively) and bradykinesia/rigidity (63.6% and 65.5% improvement, respectively). The most serious and frequent (3.6%) adverse effect of pallidotomy is a scotoma in the contralateral lower-central visual field. Less frequent complications include injury to the internal capsule, facial paresis, and intracerebral hemorrhage (1-2%). Abnormalities of speech, swallowing, and cognition may also be observed. Unilateral pallidotomy is safe and well tolerated. Bilateral lesions are not recommended because there is a high risk of severe adverse effects, such as corticobulbar syndrome with dysarthria and dysphagia. Subthalamotomy involves destruction of a part of the STN, although it is commonly avoided because of the concern of producing hemiballismus. Subthalamotomy has been typically performed unilaterally because of the higher risk of neurological side effects such as speech disturbances, ataxia, or generalized chorea associated with bilateral procedures.
Pump therapies are an integral part of neuromodulatory therapies for advanced Parkinson's disease. Currently there are 2 pump systems available – a) continuous subcutaneous apomorphine infusion (CSAI) and b) Levodopa/carbidopa intestinal gel (LCIG) pump. Major indication for both these pump therapies is advanced PD with motor fluctuations not controlled by oral or transdermal dopaminergic medications [Table 2]. Although both of them compete with DBS in these clinical settings, there are no randomized controlled comparative studies directly evaluating any of these systems and comparison is possible only based on smaller open-label studies.,
Apomorphine is a potent direct dopamine agonist (DA) with very fast onset of effect and short plasma half-life. Unlike other DA, it activates all dopamine receptors, including D1-like and D2-like receptors, similar to levodopa. Apomorphine is currently available as a pen (for rescue injections in unpredictable OFF episodes) and as CSAI when continuous dopaminergic stimulation is desirable. CSAI is delivered via a subcutaneous catheter connected to a small portable pump, typically during waking hours.
Single apomorphine injection has a rapid onset of effect (typically within 4-12 min) with a rapid clearance half-life and a mean duration of anti-parkinsonian action lasting about 45-60 min. As apomorphine was put in clinical use long before rigorous requirements for drug registration, there is only a very limited number of randomized trials for single injections and a single RCT for CSAI. Despite this perceived “evidence disadvantage” compared to DBS and LCIG, a number of open-label trials report very consistent and substantial improvement of OFF time with a mean OFF time reduction of 59%. Reported improvements in dyskinesia are variable and seem to be more pronounced in patients who achieve a more substantial reduction of levodopa after CSAI initiation. Several open-label trials report improvement of non-motor symptoms (NMS) and quality of life (QoL).,
Spectrum of CSAI side-effects include most commonly autonomic and vegetative symptoms, such as nausea, peripheral edema, orthostatic hypotension as well as neuropsychiatric symptoms typical for DAs. It is, however, important to note that CSAI has a significantly lower propensity to induce both hallucinations and impulse control disorders compared to other currently used DAs such as ropinirole, pramipexole and rotigotine. Mild to moderate skin nodules at the site of infusion are relatively common, may be bothersome and may lead to discontinuation of treatment if infection or necrosis occurs.
LCIG is administered via jejunal extension tube-percutaneous endoscopic gastrostomy (JET-PEG) connected to a portable pump directly into the proximal segment of jejunum bypassing the stomach with its erratic emptying. This approach allows delivery of levodopa directly to the site of its absorption and leads to very stable plasma levels of levodopa compared to oral administration. The infusion is typically administered during waking hours, although 24-hour infusions may be delivered in case of severe night-time problems. In many cases, the LCIG may be administered as a monotherapy, although combination with other medications such as COMT inhibitors, amantadine, dopamine agonists in specific indications or oral levodopa for night-time problems may be beneficial.
Several randomized trials have shown significant improvement of motor scores as well as a significant reduction of OFF time and increase in ON time without bothersome dyskinesia after LCIG compared to oral levodopa in patients with motor fluctuations., Reduction of dyskinesia severity and duration has been indirectly shown in previous trials, as well as in clinical practice. LCIG was shown to significantly improve QoL and NMS, especially in the sleep/fatigue, mood/cognition and gastrointestinal domains.
Side-effects or treatment-related complications are relatively common, but in most cases well manageable. The most common complications are device- or stoma-related that occur basically in all patients over a long course of treatment but rarely lead to treatment discontinuation., These include e.g., JET-PEG or jejunal tube displacement or occlusion, leakage, unintentional removal, connection issues, infections or granulomas around stoma. Good multidisciplinary management involving a dedicated gastroenterologist and PD nurse is key to successful management of these issues. Medication side-effects are similar to oral levodopa, although an increased prevalence of polyneuropathy (PNP) has been noted in LCIG. This may present as worsening of pre-existing PNP or onset of acute PNP (Guillian–Barre syndrome-like) requiring discontinuation of treatment. Polyneuropathy may be associated with decrease in vitamins B or folate levels and increase in homocysteine and methylmalonic acid in approximately 1/3 of patients and is commonly associated with significantly increased LCIG daily levodopa equivalents (LEDD) compared to LEDD on previous oral medication.
Spinal cord stimulation
Spinal cord stimulation (SCS) is a quite common and safe procedure. It has been used for many years for the treatment of medically refractory neuropathic pain of the lower extremities and trunk and failed back surgery syndrome.[36–38] As we started to know about the central effects of SCS, a very helpful insight was obtained in central modulation due to spinal stimulation. Our study in pain patients showed that analgesia offered by SCS is not only by closing the gate but by decreased connectivity between certain central neural circuits like that of perception (somatosensory) and affect (limbic). This knowledge about the ability of SCS to modulate central circuits and accidental findings in patients with Parkinson's disease (PD) and chronic pain, led to the use of SCS for the treatment of disorders of movement.
The first report about use of SCS in a 6OHDA rat model of PD was published in Science in 2009. They not only documented functional recovery but demonstrated disruption of anti-kinetic low-frequency synchronous cortico-striatal oscillations in animals with SCS. They also proposed the prokinetic effects of SCS are a result of antidromic or afferent stimulation of brainstem nuclei involved in initiation of locomotion. After this multiple case reports and small studies have documented the benefit of SCS in patients with PD and chronic pain, for sensory symptoms of PD and in camptocormia in PD.[42–46]
Recent studies with long term follow up (3 years) have shown persistent benefits in the gait dysfunction in PD. Some recent studies have also proposed neuro-protective effect of cervical spinal cord stimulation in animal models of PD.,
While selecting patients for SCS for PD, the following factors could be considered:
The implantation of spinal cord leads should be at T8 to T10 levels (paddle or percutaneous leads) and programming should be at or around pulse-width of 400 microseconds with rates of 60 Hz. The stimulation should be continuous and amplitudes at the tolerable level.
Transcranial magnetic stimulation (TMS) involves using an external magnetic stimulator to stimulate different areas of the cortex non-invasively to induce changes in cortical excitability. Depending on the frequency, intensity and duration of stimulus applied, the effects may be excitatory or inhibitory on the cortex. Single-pulse and paired-pulse TMS have been used extensively as a research tool to understand the pathophysiology of movement disorders, including PD. Repetitive TMS (rTMS) is the most commonly used protocol in a therapeutic setting. In PD, excitatory rTMS (≥5 Hz) delivered to the primary motor cortex (M1) and inhibitory rTMS (≤1 Hz) delivered to the supplementary motor area (SMA) have been shown to improve motor symptoms like rigidity and bradykinesia., Some studies have also suggested efficacy of M1, SMA or cerebellar stimulation on levodopa-induced dyskinesias.[54–56] Among the non-motor symptoms, high-frequency rTMS of the left DLPFC is “probably efficacious” in depression associated with PD. However, it is important to note that not all studies assessing rTMS in PD have shown positive or consistent results. The plasticity induction attributed to rTMS is transient, lasting about 30 min after a session. Hence, continued stimulation over multiple days is necessary for lasting effects, and heterogeneity in protocols could be a factor contributing to the inconsistent reports from studies., As of now, other than as a treatment for co-morbid depression, rTMS is generally offered in a clinical trial setting in PD.
Non-invasive vagus nerve stimulation (VNS) is postulated to modulate the activity of the locus coeruleus via the nucleus of the solitary tract, and hence of clinical interest particularly for locomotor disturbances in PD. In an open-label pilot study, VNS improved freezing of gait in PD patients. Another smaller randomized trial suggested beneficial effects on gastrointestinal symptoms in PD. Although there is promising early evidence, further studies are required prior to clinical use.
Experimental neuromodulation strategies for the treatment of PD include biological, genetic, stimulation and combined modalities. Optogenetics offers the ability for cell type-specific stimulation and is being studied in conjunction with DBS for delivery. It acts by utilizing light-sensitive proteins that control particular cellular functions and can be used to selectively control cells that respond to “stop” and “go” lights delivered through fiber optics through different colored orders.
Stem cell therapy has attempted disease modification or cure for PD. Initial studies of fetal nigral transplant failed to show long term benefits and were complicated by severe dyskinesias. Newer approaches have looked at development of neural progenitor cells and miniature SN-like structures (mini-SNLSs) for transplant. These remain in pre-clinical development while ongoing clinical trials are assessing the efficacy of fetal derived ventral mesencephalic tissue, embryonic stem cell-derived dopaminergic progenitor cells and induced pleuripotent stem cells. Pending the results of these and future larger studies, stem cell therapy is not yet to be offered outside of a clinical trial setting in PD. Among novel stimulation strategies being explored, photobiomodulation using both extracranial and intracranial near-infrared stimulation is currently undergoing clinical testing.,
The idea to consider an interventional therapy should be introduced early on at the time of diagnosis of PD when presenting the patient with options that are available for PD management. This broadens the options for the patient and gives time for the patient to process these options when the time comes to make a choice.
The neurologist should be aware of when not to offer an invasive surgery like in the presence of significant cognitive impairment or medically refractory psychiatric conditions. The provider should make a point to identify the features that are not expected to improve with neuromodulation like axial motor symptoms including freezing of gait (FOG), postural instability, camptocormia and Pisa syndrome, dysphagia, dysarthria and memory complaints. The motor complications that are expected to improve with neuromodulation should be emphasized and also the hope that it will address medication-resistant tremor in PD.
In general, the patient's age and cognitive functioning should be considered in parallel to the presence of complications like dyskinesia. Levodopa related motor complications including wearing off, delayed ON or failed ON, ON/OFF fluctuations, peak dose and biphasic dyskinesias and OFF dystonia should be recognized. Additionally, non-motor symptoms like pain, sleep disruption should also be considered. When a patient has significant cognitive impairment or language problems, then an interventional therapy especially DBS is less likely to be offered, for concerns of further decline.
A systematic approach to patient characteristics, age and comorbidities can help decide what technique best suits a particular phenotype of PD patient [Figure 2]. Since PD is a progressive neurodegenerative condition, the ability to adapt to the changing disease would make that intervention superior. If there is medication-resistant tremor or severe dyskinesia in a young and cognitively intact patient, DBS would be a good option. If the patient prefers not to have brain surgery or has significant cognitive impairment or poorly controlled psychiatric comorbidities, then levodopa-carbidopa intestinal gel (LCIG) or subcutaneous apomorphine injection (SCAI) can be offered. It may be best to consider different patient profiles in an individualized manner and the match them to an interventional therapy.
The ideal DBS candidate is a patient with established PD, whose main disability is attributable to off-time motor symptoms like tremor, rigidity and bradykinesia which are present for considerable periods of time during awake hours accompanied by troublesome dyskinesias during the medication-ON time. Criteria for patient selection are broadly based on the inclusion and exclusion criteria for randomized controlled trials comparing STN-DBS to best medical therapy and GPi-DBS to STN-DBS.,,,,,, In addition, long-term prospective follow-up studies are available for both STN and GPi DBS.,
Emerging data promotes the idea of early intervention especially in the field of DBS. While the previous practice was to offer DBS for patients with a mean duration of illness of 12-15 years, evidence from the EARLYSTIM trial reported that “early DBS” (duration of PD ≈ 7.5 years) was superior to medical therapy., Clinical studies are now assessing DBS offered even before the onset of motor fluctuations- ‘very early DBS’. At five years, patients receiving ‘very early DBS’ required lesser medication and had less tremor compared to those receiving optimal drug therapy. A larger multicenter study comparing DBS and medical therapy for early-stage PD is ongoing. With these recent advancements in mind, it is reasonable to start moving the post of when to offer DBS and to consider offering it earlier in the course of the illness.
In comparison to lesioning techniques, advantages of DBS include reversibility of the effects/side effects; flexibility of the stimulation settings; easier identification of new targets and feasibility of bilateral surgeries that make DBS currently the first treatment option in terms of PD surgery. However, a sub-population of PD patients may significantly benefit from lesioning procedures despite not being suitable candidates for DBS due to higher age, presence of neuropsychiatric symptoms (e.g., dementia, psychotic symptoms), higher risk of complications due to surgery, long term stimulation or implanted device (e.g., allergy to metals, uncontrolled diabetes mellitus with increased risk of device infections, etc.). Also, lesioning procedures present a significantly lower financial burden for low-income patients and healthcare systems, making them more widely available. As one-time procedures, lesioning also does not require regular follow-ups and this can be more practical for patients from remote regions who have problems with regular access to tertiary movement disorder centers.
Both CSAI and LCIG significantly reduce OFF time, increase ON time, improve NMS and QoL. Although direct comparative studies are lacking, current limited evidence and clinical experience seems to favor LCIG over CSAI especially in patients with severe motor fluctuations and dyskinesia. LCIG has also less potential caveats compared to CSAI in terms of neurocognitive profile of patients (mild to moderate dementia, presence of hallucinations and impulse control disorder). Advantages of CSAI include especially its less invasive character and thus it may be favorable in patients with milder motor fluctuations or those who have not decided yet for DBS or LCIG. In both pump systems the major potential limitation is patient non-cooperation and inappropriate social background due to technical aspects of the therapy. Both pumps are typically administered only during waking hours and thus in patients with severe night-time problems DBS seems to be more favorable, although 24-hour infusions may still be an option.
The landscape of PD management has continuously evolved over the years and currently, there are a host of options that can be offered to the patient. Neuromodulatory strategies including DBS, lesioning and pump therapies can all contribute to improved quality of life in carefully selected patients. Patient education, comprehensive, multidisciplinary evaluation and technical expertise are equally important to achieve good outcomes.
Financial support and sponsorship
No targeted funding.
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2]
[Table 1], [Table 2]