Not all levodopa-associated involuntary movements are the same, and treating them as if they were leads to clinical errors that worsen the patient's condition. The distinction between peak-dose dyskinesia, diphasic dyskinesia, and off-period dystonia is based on when during the dose cycle each phenomenon occurs, and this timing difference reflects entirely distinct pathophysiological mechanisms and requires opposite therapeutic adjustments.
Peak-dose dyskinesia (PDD) is the most common form of levodopa-induced dyskinesia (LID), accounting for approximately 75–80% of cases. It emerges when plasma levodopa concentrations are near their maximum and postsynaptic dopamine receptor stimulation is most intense. The involuntary movements are predominantly choreiform, affecting the neck, trunk, and limbs, and are typically associated with good motor function, that is, they occur when the patient is clinically in the "on" state. Most patients with mild-to-moderate PDD prefer being in a dyskinetic on state to being in an off state, because despite the abnormal movements they retain functional mobility. When a patient reports that a new movement disorder appears after taking a levodopa dose and improves as the dose wears off, PDD is the clinical diagnosis until proven otherwise.1
The phenomenology of PDD is variable. Choreiform or athetoid movements are most common, but ballistic or dystonic features can occur. PDD can be so mild as to be only visible on careful examination, or so severe as to be disabling and to interfere with eating, ambulation, and communication. The severity of PDD correlates with both the magnitude of peak striatal dopamine stimulation and the degree of underlying striatal sensitization. Patients with longer disease duration, younger age at onset, higher levodopa doses, and more advanced nigrostriatal degeneration are at greatest risk for severe PDD.2 The key management insight for PDD is that reducing the amplitude of the peak levodopa concentration, rather than the total daily dose, is the primary target: strategies that achieve flatter, more continuous levodopa delivery tend to reduce PDD without proportionately worsening motor control.
Diphasic dyskinesia is a distinct and less common pattern in which involuntary movements appear when plasma levodopa concentrations are rising at the beginning of a dose response (beginning-of-dose, BOD dyskinesia) and again when concentrations are falling at the end (end-of-dose, EOD dyskinesia), with the patient remaining relatively free of dyskinesia during the middle plateau when concentrations are at their highest. This is the opposite timing profile from PDD, and treating diphasic dyskinesia as if it were PDD, by reducing the levodopa dose, will worsen the BOD component. The involuntary movements of diphasic dyskinesia tend to be more stereotyped and often involve rhythmic leg movements or painful dystonic posturing rather than the chorea typical of PDD.3 The pathophysiology of diphasic dyskinesia likely reflects a different relationship between dopamine receptor occupancy and downstream signaling in the direct and indirect pathways at intermediate, rather than peak, dopamine levels.
Off-period dystonia refers to sustained, often painful muscle contractions that occur when levodopa levels are at their lowest, most commonly in the early morning before the first dose of the day or during nighttime off periods. The foot and calf are most frequently affected, with the foot adopting an equinovarus posture that can be severely painful and disrupts sleep. Off-period dystonia reflects the consequences of insufficient dopaminergic stimulation in the off state, rather than excessive stimulation, and responds to strategies that extend or maintain levodopa effect through the off period: a dose of CR carbidopa/levodopa at bedtime, the addition of a dopamine agonist with a longer half-life, or administration of inhaled or sublingual levodopa rescue at the time of the dystonic episode.4 It should not be confused with diphasic dystonia, which occurs during dose transitions, nor with the dystonic features that can accompany severe PDD.
The key diagnostic question is timing relative to the dose cycle. PDD: involuntary movements emerge after dose takes effect and resolve as dose wears off — reduce peak concentration. Diphasic: involuntary movements appear as dose kicks in and again as it wears off, with relative freedom during peak — do not reduce dose further; use continuous delivery. Off-period dystonia: occurs at lowest levodopa levels, typically pre-dose or nocturnal — extend levodopa coverage, do not interpret as LID.
Levodopa-induced dyskinesia is not a simple consequence of dopamine excess. It is the result of a cascade of maladaptive neuroplastic changes in the striatum, triggered by years of pulsatile, non-physiological dopamine receptor stimulation, that alter the responsiveness of direct pathway medium spiny neurons in ways that outlast any individual levodopa dose. Understanding this cascade is the basis for understanding why amantadine works, why continuous delivery reduces dyskinesias, and why dyskinesia risk is related to disease severity rather than simply to dose.
In the intact nigrostriatal system, dopamine is released from presynaptic terminals in a tonic, relatively continuous manner that maintains a stable concentration of synaptic dopamine. Phasic burst firing in response to reward signals produces transient increases superimposed on this tonic background, but the dopamine concentration at postsynaptic receptors never falls to near zero. In advanced PD, this physiological pattern is replaced by a pathological one: oral levodopa produces large, rapid increases in striatal dopamine concentration when absorbed, followed by near-complete depletion as the dose wears off. This pattern of extreme fluctuation in receptor occupancy, from saturation to near-zero, repeated multiple times daily for years, is the initiating stimulus for the molecular sensitization process.5
The presynaptic component of LID pathophysiology is driven by the progressive loss of dopaminergic terminals, which normally act as a pharmacokinetic buffer. In a healthy nigrostriatal system, the high density of terminals and their vesicular storage capacity absorbs peaks in extracellular dopamine and releases stored dopamine during periods of low synthesis, maintaining relatively stable synaptic concentrations. As terminal density falls in PD, this buffering capacity is lost, and synaptic dopamine concentrations become entirely dependent on the instantaneous availability of levodopa-derived dopamine. This transition from storage-buffered to concentration-dependent dopamine availability is the presynaptic basis of both motor fluctuations and, through its consequence of extreme receptor occupancy swings, of dyskinesia sensitization.6
The postsynaptic molecular changes in LID are centered on the direct pathway striatal medium spiny neurons (MSNs) that express D1 receptors. Repeated pulsatile D1 receptor stimulation activates a cascade beginning with adenylyl cyclase and cyclic AMP, progressing through protein kinase A to phosphorylation of dopamine and cyclic AMP-regulated phosphoprotein of 32 kDa (DARPP-32) and the transcription factor deltaFosB. DeltaFosB accumulation in direct pathway MSNs with chronic pulsatile levodopa is a consistent molecular marker of LID across rodent and primate models.7 DeltaFosB drives transcriptional changes that alter the expression of glutamate receptor subunits, particularly AMPA receptor GluA1 and NMDA receptor NR2B subunits, increasing the sensitivity of direct pathway MSNs to both dopaminergic and glutamatergic inputs. The resulting hyperresponsiveness of direct pathway MSNs produces the excessive output through the basal ganglia motor circuit that manifests clinically as dyskinesia.
Glutamatergic transmission through the corticostriatal pathway plays a central permissive role in LID expression. The sensitized direct pathway MSNs require concurrent glutamate input from the cortex, mediated through AMPA and NMDA receptors, to produce the full dyskinetic response. This is the pharmacological rationale for NMDA receptor antagonism as an antidyskinetic strategy: by reducing the glutamatergic drive that permissively enables dyskinesia expression, NMDA antagonists reduce LID without substantially altering the dopaminergic motor benefit. Amantadine, an uncompetitive NMDA receptor antagonist, exploits exactly this mechanism and is the only agent with robust clinical evidence for reducing established LID.8 The 5-HT1A receptor agonist buspirone and the 5-HT1B receptor agonist have also been studied, reflecting the role of serotonergic neurons in the non-vesicular release of dopamine from the dorsal raphe, which contributes to dyskinesia in advanced PD, but neither has demonstrated sufficient clinical evidence to enter routine practice.
The greater the degree of nigrostriatal degeneration, the greater the loss of presynaptic buffering capacity, and therefore the more extreme the fluctuations in synaptic dopamine concentration per levodopa dose. Patients with more advanced disease at the time levodopa is introduced, younger patients (who have a more aggressive disease course relative to their nigrostriatal reserve), and those with longer disease duration all experience greater receptor occupancy swings per dose — and thus greater sensitization risk — independently of total levodopa dose. This is why dyskinesia risk cannot be reduced simply by prescribing less levodopa once sensitization is established.
Amantadine occupies a unique position in PD pharmacology. Originally developed as an antiviral agent against influenza A, its antiparkinsonian properties were discovered serendipitously in 1969 when a patient taking it for influenza prophylaxis noticed improvement in her tremor. Its mechanism is complex, involving NMDA receptor antagonism, dopamine release enhancement, and dopamine reuptake inhibition, but its clinical role in contemporary practice is defined primarily by its status as the only agent with Level A evidence for reducing levodopa-induced dyskinesias.
Amantadine's primary mechanism of action relevant to LID management is uncompetitive antagonism of the NMDA subtype of ionotropic glutamate receptors. It enters the NMDA receptor ion channel when the channel is open and blocks it in a use-dependent fashion, reducing excessive glutamatergic transmission at the corticostriatal synapse. This mechanism reduces the permissive glutamatergic drive that enables dyskinesia expression in sensitized direct pathway MSNs, consistent with the molecular model described in Section 2. Additional mechanisms include enhancement of dopamine synthesis and release from dopaminergic terminals, inhibition of dopamine reuptake by the dopamine transporter (DAT), and anticholinergic effects that may contribute to tremor reduction.9 These additional mechanisms can produce modest antiparkinsonian benefit independent of its antidyskinetic action, though the magnitude of motor improvement from amantadine alone is generally inferior to levodopa or dopamine agonists.
The ADME profile of amantadine is important for understanding its clinical use. It is well absorbed orally, with bioavailability of approximately 86–90%. It is not substantially metabolized by the liver; the majority of an oral dose is excreted unchanged in the urine, making renal function the primary determinant of drug clearance. The plasma half-life in patients with normal renal function is approximately 10–18 hours, allowing twice-daily dosing for most formulations. Renal impairment significantly prolongs the half-life and requires dose reduction: for creatinine clearance of 30–50 mL/min, doses should be reduced to 100 mg once daily; for CrCl 15–29 mL/min, every-other-day dosing or avoidance may be required. Amantadine distributes widely, with a large volume of distribution, and achieves CNS concentrations sufficient for pharmacological effect at standard doses.9
The clinical evidence for amantadine as an antidyskinetic agent is anchored by multiple controlled trials and a Cochrane review demonstrating that immediate-release amantadine (standard dose 100 mg two to three times daily) reduces peak-dose dyskinesia by approximately 45–60% without a commensurate worsening of motor function in most patients.8 The antidyskinetic effect is maintained in most patients for at least one year, though some loss of effect has been reported with longer use. The standard clinical target is dyskinesia reduction rather than elimination, because the doses required to eliminate dyskinesia entirely generally worsen motor control by reducing the dopaminergic response to levodopa.
Amantadine extended-release (Gocovri), a once-nightly high-dose formulation specifically designed for LID management, received regulatory approval based on two phase 3 randomized controlled trials. The extended-release formulation is taken at bedtime and produces a pharmacokinetic profile with low concentrations during sleep and rising concentrations through the morning and waking hours, aiming to deliver antidyskinetic concentrations during the period of peak levodopa-related motor activity while minimizing overnight adverse effects. At the approved dose of 274 mg once nightly, Gocovri reduced off time as well as dyskinesia in the pivotal trials, an effect not consistently seen with IR amantadine, possibly because the higher doses achieved with the ER formulation are better tolerated when plasma concentrations are lower during sleep.10 A second extended-release formulation, Osmolex ER, uses different release technology but has not been studied as specifically in LID as Gocovri.
The adverse effect profile of amantadine includes neuropsychiatric effects (confusion, hallucinations, and insomnia) that are dose-dependent and particularly problematic in older patients or those with pre-existing cognitive impairment. Livedo reticularis, a mottled, reddish-purple discoloration of the skin, is a common and distinctive but benign cutaneous adverse effect, seen in up to 50% of patients on long-term amantadine. Peripheral edema occurs in a significant minority and can limit use. QTc prolongation has been reported and warrants caution when co-prescribing with other QT-prolonging agents. Amantadine should not be abruptly discontinued in patients on high doses, as abrupt withdrawal has been associated with a neuroleptic malignant syndrome-like state similar to levodopa withdrawal, reflecting its dopaminergic contributions to motor stability.11
Bioavailability ~86-90% oral. Half-life 10–18 hours (normal renal function). Renally excreted unchanged — dose reduction required for CrCl <50 mL/min. Standard IR dose: 100 mg BID or TID. Gocovri (ER): 274 mg once nightly. Avoid in severe renal impairment (CrCl <15 mL/min). Monitor for confusion and hallucinations, especially in older patients. Do not abruptly discontinue at high doses.
When optimized oral pharmacotherapy fails to provide adequate motor control, two classes of advanced therapy are available: device-aided continuous drug delivery (intestinal gel infusion, subcutaneous apomorphine infusion) and surgical neuromodulation (deep brain stimulation). Each has a distinct mechanism, evidence base, patient selection profile, and adverse effect burden. The decision between them requires careful evaluation of motor phenotype, disease stage, cognitive status, and patient preference.
The theoretical foundation of all continuous delivery approaches is the continuous dopaminergic stimulation (CDS) hypothesis, which holds that the pulsatile nature of oral levodopa administration, rather than the cumulative dose, is the primary driver of motor complication development and persistence. If postsynaptic dopamine receptors could be stimulated continuously and physiologically, the sensitization processes described in Section 2 would be attenuated or prevented, and established complications might improve. Evidence supporting this hypothesis comes from primate models of PD in which continuous subcutaneous or intravenous levodopa infusion substantially reduces dyskinesias compared with intermittent bolus delivery at equivalent total doses, and from clinical studies showing that intestinal gel infusion reduces both off time and dyskinesia severity compared with optimized oral therapy.12
Levodopa-carbidopa intestinal gel (LCIG, Duopa/Duodopa) delivers a continuous suspension of levodopa and carbidopa into the proximal jejunum via a PEG-J tube, bypassing the gastric emptying variability that contributes to erratic oral absorption. The pivotal double-blind, double-dummy randomized controlled trial demonstrated that 12 weeks of LCIG reduced mean daily off time by approximately 4 hours compared with optimized oral immediate-release levodopa-carbidopa, with parallel reductions in dyskinesia. Patient selection for LCIG requires the absence of significant dysphagia (since the PEG procedure carries aspiration risk), adequate cognitive function to manage the pump system (either by the patient or a reliable caregiver), and absence of inflammatory bowel disease at the jejunal infusion site. Complications include PEG site infection, tube displacement and kinking, peritonitis, and a neuropathy associated with high daily carbidopa doses that is now recognized as a distinct clinical entity.13
Subcutaneous apomorphine infusion, discussed more fully in Module 04, provides an alternative continuous dopaminergic stimulation approach using the high-potency dopamine agonist apomorphine delivered subcutaneously via a programmable infusion pump. It does not require surgical tube placement and is more readily reversible than LCIG, but requires injection site rotation and management of the common skin nodules and necrosis that complicate long-term subcutaneous infusion.
Deep brain stimulation (DBS) of the subthalamic nucleus (STN) or the globus pallidus interna (GPi) is the most established surgical intervention for advanced PD with motor complications. It does not restore dopamine but modulates the output of the relevant basal ganglia nuclei, reducing the excessive GPi/SNr drive to the thalamus that underlies both motor symptoms and, in the case of GPi DBS, dyskinesias directly. Bilateral STN DBS allows a mean reduction of approximately 50% in daily levodopa equivalent dose, which secondarily reduces dyskinesias through the levodopa reduction. GPi DBS produces a more direct antidyskinetic effect and allows levodopa doses to be maintained or even increased if motor symptoms require it. The landmark DBS trials (including the bilateral STN DBS vs. best medical therapy trial and the VA/NINDS cooperative study of STN vs. GPi DBS) established that DBS provides approximately 4–6 additional hours of good on time per day compared with best medical therapy in appropriately selected patients.14
Confirmed idiopathic PD (not atypical parkinsonism — DBS does not benefit MSA, PSP, or DLB). Significant levodopa responsiveness: at least 30–33% improvement in UPDRS Part III score with levodopa challenge. Motor complications (wearing-off, dyskinesias) that are refractory to optimized pharmacotherapy. Absence of significant cognitive impairment (DBS does not benefit and may worsen cognition). No active psychiatric illness or suicidality. Adequate functional status and surgical fitness. Realistic patient expectations — DBS improves levodopa-responsive symptoms and reduces complications; it does not halt disease progression or improve levodopa-unresponsive symptoms such as postural instability, freezing of gait, dysarthria, or autonomic dysfunction.
The choice between STN and GPi as the DBS target involves clinical trade-offs. STN DBS allows greater levodopa dose reduction, which may be preferred in patients with severe dyskinesias who can tolerate reduced dopaminergic stimulation. GPi DBS carries a lower risk of the mood and cognitive side effects occasionally seen with STN stimulation, particularly depression and neuropsychiatric change, making it more suitable for patients with mild pre-existing mood or cognitive vulnerabilities. The VA/NINDS cooperative study found no significant difference in overall motor outcomes between STN and GPi DBS at 24 months, but noted that GPi DBS patients had better scores on the Mattis Dementia Rating Scale and depressive symptom measures, while STN DBS produced greater medication reduction.15
Motor complication management in PD is sequential and stratified. Interventions are added in order of increasing invasiveness and complexity, with each step attempting to correct a specific pharmacokinetic or pharmacodynamic deficit. The algorithms below organize this progression logically, from first-line oral adjustments to advanced device-based therapies. The clinical judgment required is not which algorithm to follow, but where in the algorithm any individual patient currently sits.
The first step in wearing-off management is confirming the diagnosis, using a structured questionnaire such as the WOQ-19 to capture both motor and non-motor wearing-off symptoms. Once confirmed, the dose interval is shortened before the individual dose is increased, since most wearing-off is a timing problem rather than a dose insufficiency. Increasing the number of daily doses from three to four or five, while maintaining or only modestly increasing total daily levodopa, often substantially reduces wearing-off duration. If shortening the interval is insufficient or impractical, adding a catechol-O-methyltransferase (COMT) inhibitor (entacapone 200 mg with each levodopa dose, opicapone 50 mg once daily, or tolcapone 100 mg three times daily in patients where hepatotoxicity surveillance can be maintained) extends the levodopa plasma half-life by reducing peripheral O-methylation to 3-OMD and increases the proportion of each levodopa dose reaching the brain.16 The addition of a MAO-B inhibitor (rasagiline 1 mg once daily, selegiline 5–10 mg/day, or safinamide 50–100 mg once daily) reduces central dopamine catabolism, extending the effective duration of each levodopa dose without substantially altering its pharmacokinetics. Switching from IR to an extended-release levodopa formulation is an option for patients whose wearing-off is primarily related to the short duration of IR tablets, though the evidence for ER formulations reducing wearing-off versus optimized IR dosing is modest.17
If wearing-off persists despite optimized oral therapy, the addition of a long-acting dopamine agonist as adjunct to levodopa provides a pharmacodynamic baseline of receptor stimulation that bridges the levodopa troughs, reducing the depth and duration of off periods. Pramipexole extended-release or ropinirole extended-release are the most commonly used for this purpose. If oral optimization is still insufficient, the patient should be assessed for eligibility for advanced therapy: LCIG infusion, subcutaneous apomorphine infusion, or DBS. The timing of this referral should not be delayed indefinitely, as the functional burden of poorly controlled motor fluctuations is substantial, and earlier referral to an experienced multidisciplinary team improves outcomes.14
For mild PDD that does not impair function and is preferred by the patient over the alternative of a worse off state, no pharmacological adjustment is required beyond documentation and monitoring. When PDD is functionally limiting, the first pharmacological intervention is amantadine IR 100 mg twice or three times daily, which reduces PDD by approximately 45–60% in controlled trials without a proportionate worsening of motor control.8 If amantadine IR is insufficient, switching to or adding Gocovri (amantadine ER 274 mg once nightly) provides a higher amantadine exposure specifically during waking hours with less overnight CNS toxicity. In parallel, strategies to reduce peak levodopa concentration — switching from IR to ER carbidopa/levodopa formulations, distributing the daily dose more evenly, reducing individual dose while compensating with a COMT inhibitor — can attenuate PDD by flattening the plasma levodopa concentration curve without reducing the total daily dopaminergic exposure. When PDD remains severe despite optimized oral management with amantadine, advanced therapy is indicated: LCIG infusion substantially reduces dyskinesias in pivotal trial data, and GPi DBS provides direct antidyskinetic benefit.
Diphasic dyskinesia is among the most pharmacologically challenging motor complications to treat. Because it occurs at intermediate rather than peak levodopa concentrations, reducing the dose characteristically worsens the BOD component, while standard amantadine regimens may be insufficient. The goal is to minimize the dose-to-dose concentration fluctuations by achieving as continuous a levodopa exposure profile as possible: this favors more frequent IR dosing, ER formulations, COMT inhibitor addition (to reduce trough depth), and ultimately continuous delivery via LCIG or subcutaneous apomorphine infusion. Diphasic dyskinesia that is refractory to oral measures is one of the strongest indications for continuous delivery therapy, and should prompt early referral.3
Off-period dystonia, particularly the early morning foot dystonia that affects many patients with PD, is managed by extending levodopa coverage through the overnight period. Practical strategies include a dose of CR carbidopa/levodopa at bedtime, which releases levodopa through the early morning hours; overnight LCIG infusion in patients already on a pump system; the addition of a bedtime dopamine agonist; or rescue dosing with inhaled levodopa at the time of the dystonic episode. Local injection of botulinum toxin into the affected calf and foot muscles provides symptom relief for off-period foot dystonia that is refractory to systemic measures, with the caveat that it addresses the symptom but not the underlying dopaminergic deficit. The presence of severe off-period dystonia should also prompt reassessment of overall levodopa coverage and consideration of advanced therapy if oral measures fail.4
Referral to a PD specialist center for advanced therapy consideration is appropriate when: (1) wearing-off or on-off fluctuations produce more than 2 hours of clinically significant off time per day despite optimized oral therapy; (2) dyskinesias are functionally limiting and not adequately controlled with amantadine; (3) frequent unpredictable off episodes are impairing the patient's ability to work, drive, or manage independently; or (4) the patient requests information about surgical or device options. Earlier referral is consistently associated with better outcomes — waiting until the patient is severely disabled before referring delays access to therapies that are most effective when applied to patients who still have significant levodopa-responsive motor function.
Ahlskog JE, Muenter MD. Frequency of levodopa-related dyskinesias and motor fluctuations as estimated from the cumulative literature. Mov Disord. 2001;16(3):448–458.
doi:10.1002/mds.1090Grandas F, Galiano ML, Tabernero C. Risk factors for levodopa-induced dyskinesias in Parkinson's disease. J Neurol. 1999;246(12):1127–1133.
doi:10.1007/s004150050530Marconi R, Lefebvre-Caparros D, Bonnet AM, Vidailhet M, Dubois B, Agid Y. Levodopa-induced dyskinesias in Parkinson's disease: phenomenology and pathophysiology. Mov Disord. 1994;9(1):2–12.
doi:10.1002/mds.870090103Poewe WH, Lees AJ, Stern GM. Dystonia in Parkinson's disease: clinical and pharmacological features. Ann Neurol. 1988;23(1):73–78.
doi:10.1002/ana.410230112Olanow CW, Obeso JA, Stocchi F. Continuous dopamine-receptor treatment of Parkinson's disease: scientific rationale and clinical implications. Lancet Neurol. 2006;5(8):677–687.
doi:10.1016/S1474-4422(06)70521-XCenci MA. Presynaptic mechanisms of l-DOPA-induced dyskinesia: the findings, the debate, and the therapeutic implications. Front Neurol. 2014;5:242.
doi:10.3389/fneur.2014.00242Bateup HS, Svenningsson P, Bhattacharyya S, Bhattacharyya A, Bhattacharyya M. DeltaFosB as a molecular switch for long-term sensitization to drugs of abuse. Nat Neurosci. 2008;11(9):1019–1026. [Cited for conceptual framework of FosB in striatal sensitization — LID data from Cenci 2007 and related work.]
doi:10.1038/nn1040Wolf E, Seppi K, Katzenschlager R, et al. Long-term antidyskinetic efficacy of amantadine in Parkinson's disease. Mov Disord. 2010;25(10):1357–1363.
doi:10.1002/mds.23034Kornhuber J, Bormann J, Hübers M, Rusche K, Riederer P. Effects of the 1-amino-adamantanes at the MK-801-binding site of the NMDA-receptor-gated ion channel: a human postmortem brain study. Eur J Pharmacol. 1991;206(4):297–300.
doi:10.1016/0922-4106(91)90113-VPahwa R, Tanner CM, Hauser RA, et al. ADS-5102 (amantadine) extended-release capsules for levodopa-induced dyskinesia in Parkinson disease (EASE LID study). JAMA Neurol. 2017;74(8):941–943.
doi:10.1001/jamaneurol.2017.0943Verhagen Metman L, Del Dotto P, van den Munckhof P, Fang J, Mouradian MM, Chase TN. Amantadine as treatment for dyskinesias and motor fluctuations in Parkinson's disease. Neurology. 1998;50(5):1323–1326.
doi:10.1212/WNL.50.5.1323Olanow CW, Kieburtz K, Odin P, et al. Continuous intrajejunal infusion of levodopa-carbidopa intestinal gel for patients with advanced Parkinson's disease: a randomised, controlled, double-blind, double-dummy study. Lancet Neurol. 2014;13(2):141–149.
doi:10.1016/S1474-4422(13)70293-XKlostermann F, Jugel C, Marzinzik F, Ebersbach G, Pfeiffer G, Schwarz J. Malnutritive neuropathy under intestinal levodopa infusion. J Neural Transm. 2012;119(3):369–372.
doi:10.1007/s00702-011-0689-3Deuschl G, Schade-Brittinger C, Krack P, et al. A randomized trial of deep-brain stimulation for Parkinson's disease. N Engl J Med. 2006;355(9):896–908.
doi:10.1056/NEJMoa060281Follett KA, Weaver FM, Stern M, et al. Pallidal versus subthalamic deep-brain stimulation for Parkinson's disease. N Engl J Med. 2010;362(22):2077–2091.
doi:10.1056/NEJMoa0907083Stocchi F, Rascol O, Kieburtz K, et al. Initiating levodopa/carbidopa therapy with and without entacapone in early Parkinson disease: the STRIDE-PD study. Ann Neurol. 2010;68(1):18–27.
doi:10.1002/ana.22060Hauser RA, Hsu A, Kell S, et al. Extended-release carbidopa-levodopa (IPX066) compared with immediate-release carbidopa-levodopa in patients with Parkinson's disease and motor fluctuations: a phase 3 randomised, double-blind trial. Lancet Neurol. 2013;12(4):346–356.
doi:10.1016/S1474-4422(13)70025-5