Motor Complications in Parkinson's Disease
Motor complications — wearing off, dyskinesia, and on-off fluctuations — are an inevitable consequence of long-term levodopa therapy in Parkinson's disease. Within 5 years of levodopa initiation, approximately 40–50% of patients develop motor fluctuations; by 10 years, this rises to 70–80%. These complications represent a fundamental shift in disease management, from straightforward dopaminergic titration to a complex balancing act between motor benefit, OFF time, and dyskinesia. This article reviews the pathophysiology, pharmacologic management, surgical interventions, and rescue therapies for motor complications, anchored in the MDS Evidence-Based Medicine Reviews (Fox et al., 2018; de Bie et al., 2025) and landmark DBS trials.
🔹 Bottom Line: Motor Complications
- Wearing off is the earliest and most common motor complication. First-line management: optimize levodopa dosing (more frequent, smaller doses), then add a MAO-B inhibitor (preferred over COMT inhibitor per PD MED). Rescue therapies (inhaled levodopa, apomorphine SQ) address individual OFF episodes.
- Levodopa-induced dyskinesia: Amantadine (IR or ER/Gocovri) is the only pharmacologic agent with established antidyskinetic efficacy. Gocovri is the first FDA-approved drug specifically for dyskinesia.
- DBS is highly effective for motor complications: EARLYSTIM showed bilateral STN-DBS improved quality of life by 8.0 points on PDQ-39 (P=0.002) in patients with early motor complications (mean disease duration 7.5 years).
- GPi vs STN: Both targets produce similar motor improvement (CSP 468: −11.8 vs −10.7 UPDRS-III, P=0.50). STN allows greater medication reduction; GPi may be preferred when cognitive/mood concerns exist.
- Motor complications are primarily driven by disease progression (loss of presynaptic dopaminergic buffering capacity), not cumulative levodopa exposure — reinforcing that early levodopa use is appropriate.
Pathophysiology of Motor Complications
In early PD, surviving nigrostriatal neurons buffer intermittent oral levodopa dosing — storing and releasing dopamine in a relatively tonic fashion. As disease progresses and more dopaminergic terminals are lost, this buffering capacity diminishes. The result is that striatal dopamine levels increasingly mirror the pulsatile plasma pharmacokinetics of oral levodopa, producing cycles of excessive (peak-dose dyskinesia) and insufficient (wearing off/OFF state) dopaminergic stimulation.
Three key mechanisms underlie motor complications: Presynaptic loss: Reduced dopaminergic terminal density eliminates the ability to store and tonically release dopamine from exogenous levodopa. Postsynaptic sensitization: Chronic pulsatile stimulation induces maladaptive changes in striatal medium spiny neurons, including altered D1/D2 receptor signaling, changes in NMDA and AMPA receptor expression, and aberrant plasticity at corticostriatal synapses — the molecular substrate of dyskinesia. Non-dopaminergic contributions: Serotonergic neurons in the raphe nuclei can convert levodopa to dopamine but lack the feedback mechanisms for regulated release, contributing to erratic striatal dopamine levels in advanced disease.
The clinical spectrum evolves in a predictable sequence: wearing off (predictable end-of-dose deterioration) → peak-dose dyskinesia (involuntary choreiform movements at time of maximum levodopa effect) → on-off fluctuations (unpredictable, rapid transitions between ON and OFF states) → diphasic dyskinesia (stereotyped, often dystonic movements occurring at the beginning and end of levodopa effect — the rarest and most difficult to treat).
Managing Wearing Off / OFF Episodes
Levodopa Optimization
Before adding adjunctive therapy, levodopa dosing should be optimized. Strategies include: increasing dosing frequency (from TID to QID or 5×/day) while reducing individual dose size to maintain total daily dose; ensuring consistent timing relative to meals (protein competition for intestinal absorption); switching from IR to ER formulations (Rytary/IPX066) for more sustained levodopa delivery; and protein redistribution diets (concentrating dietary protein at the evening meal) for patients with marked absorption-related fluctuations.
Pharmacologic Adjuncts
When levodopa optimization is insufficient, adjunctive therapy is the next step. Based on the MDS EBM Reviews (Fox et al., 2018; de Bie et al., 2025), the following agents reduce OFF time(1,2):
MAO-B inhibitors: Rasagiline (1 mg/day) and safinamide (50–100 mg/day) each reduce daily OFF time by approximately 0.5–1.0 hour. PD MED showed MAO-B inhibitors were superior to COMT inhibitors as first adjunct in terms of patient-rated mobility and quality of life (PDQ-39 mobility +4.2 points, P=0.03)(3). Safinamide's dual mechanism (MAO-B inhibition + glutamate modulation) may provide additional benefit for both fluctuations and dyskinesia.
COMT inhibitors: Entacapone (200 mg with each levodopa dose) and opicapone (50 mg once daily at bedtime) reduce OFF time by approximately 1.0–1.5 hours/day. Opicapone's once-daily dosing and lack of hepatotoxicity concerns make it increasingly preferred over entacapone. Tolcapone is the most potent but requires LFT monitoring. All COMT inhibitors may worsen dyskinesia (requiring levodopa dose reduction).
Dopamine agonists: Adding pramipexole, ropinirole, or rotigotine reduces OFF time, but with the neuropsychiatric risks discussed in the Motor Treatment article (ICDs, hallucinations, somnolence).
Istradefylline: The adenosine A2A antagonist provides a non-dopaminergic mechanism for reducing OFF time (~0.5–0.7 h/day) and is particularly useful when further dopaminergic augmentation is limited by side effects.
Rescue Therapies for Acute OFF Episodes
Rescue therapies provide rapid, on-demand relief for individual OFF episodes and represent an important advance in motor complication management:
| Agent | Trade Name | Route | Onset | Key Efficacy | Key Concerns |
|---|---|---|---|---|---|
| Apomorphine SC injection | Apokyn | Subcutaneous | ~10–20 min | Potent, reliable OFF reversal; UPDRS-III improvement ~20 points | Nausea (requires trimethobenzamide pre-treatment), injection site reactions, orthostatic hypotension, somnolence |
| Apomorphine SL film | Kynmobi* | Sublingual | ~15–30 min | Effective OFF rescue; needle-free alternative to SC | *Discontinued (2024) due to business reasons despite clinical efficacy. Oropharyngeal swelling risk |
| Inhaled levodopa | Inbrija | Oral inhalation | ~10–15 min | UPDRS-III improvement ~9.8 points at 30 min (SPAN-PD) | Cough, upper respiratory infection, discolored sputum. Not for use in patients with asthma/COPD |
🔹 Clinical Relevance: When to Add Rescue Therapy
- Rescue therapies are indicated for patients with breakthrough OFF episodes despite optimized oral therapy — particularly those with unpredictable OFF or delayed onset of oral levodopa effect.
- Apomorphine SC injection remains the most potent and reliable rescue agent. Pre-treatment with trimethobenzamide (starting 3 days before) and careful dose titration in-office are essential.
- Inhaled levodopa (Inbrija) offers a non-invasive alternative with rapid onset but more modest efficacy than apomorphine.
- Rescue therapies do not replace baseline oral dopaminergic optimization — they supplement it for episodic breakthroughs.
Levodopa-Induced Dyskinesia
Levodopa-induced dyskinesia (LID) most commonly manifests as peak-dose dyskinesia — involuntary choreiform or choreoathetoid movements of the limbs, trunk, and head occurring at the time of maximum levodopa plasma concentration. Less commonly, diphasic dyskinesia produces stereotyped, often dystonic movements at the beginning and end of the levodopa cycle (as levels rise and fall through a "threshold zone"). Off-period dystonia — typically painful foot/leg dystonia in the early morning before the first levodopa dose — is managed differently (by ensuring adequate dopaminergic coverage).
General strategies for peak-dose dyskinesia include: reducing individual levodopa doses (at the cost of potentially increasing OFF time), switching from IR to ER levodopa (smoother kinetics), reducing or eliminating COMT inhibitors or DA (which may amplify peak levels), and adding amantadine.
Amantadine
Amantadine is the only pharmacologic agent with established antidyskinetic efficacy. Its mechanism involves NMDA receptor antagonism, which attenuates the maladaptive corticostriatal plasticity underlying dyskinesia. Amantadine IR (Symmetrel; 100 mg BID–TID) has been used for decades for this indication, though its benefit may attenuate over time in some patients. The MDS EBM Review rates amantadine as "clinically useful" for LID treatment(1).
Amantadine ER (Gocovri/ADS-5102) is the first and only FDA-approved drug specifically indicated for the treatment of LID. Dosed at 274 mg at bedtime, the extended-release formulation provides peak amantadine concentrations in the morning when dyskinesia is typically most troublesome. The EASE LID and EASE LID 3 trials demonstrated significant reduction in dyskinesia (UDysRS score reduction ~20–37%) and OFF time reduction (~0.5–1 hour/day) compared to placebo. Adverse effects include hallucinations, dizziness, dry mouth, peripheral edema, constipation, falls, and livedo reticularis(1).
🔴 Do Not Abruptly Discontinue Amantadine
- Abrupt withdrawal of amantadine can precipitate a neuroleptic malignant-like syndrome (parkinsonism-hyperthermia syndrome) with hyperthermia, rigidity, altered consciousness, and rhabdomyolysis.
- Always taper amantadine gradually over 1–2 weeks, even if the patient is hospitalized for unrelated reasons.
- This risk applies to both IR and ER formulations.
Deep Brain Stimulation for Motor Complications
Deep brain stimulation (DBS) is the most established surgical intervention for PD and is the standard of care for patients with medically refractory motor complications. DBS delivers continuous electrical stimulation to target nuclei — primarily the subthalamic nucleus (STN) or globus pallidus internus (GPi) — modulating pathologic basal ganglia circuit activity. Four landmark RCTs form the evidence base.
EARLYSTIM: DBS in Early Motor Complications
EARLYSTIM (Schuepbach et al., 2013) was the first RCT to evaluate DBS in patients with early motor complications rather than advanced disease. A total of 251 patients (age 18–60, disease duration ≥4 years, early motor complications with mean duration ~1.7 years, no dementia) were randomized to bilateral STN-DBS plus medical therapy versus medical therapy alone for 24 months(4).
STN-DBS produced a significant improvement in quality of life: PDQ-39 summary index improved 7.8 points in the DBS group versus worsening of 0.2 points with medical therapy (between-group difference 8.0 points; 95% CI 4.8–11.9; P=0.002). Secondary outcomes overwhelmingly favored DBS: off-medication UPDRS-III improved by 16.4 points more in the DBS group (P<0.001), levodopa-induced complications (UPDRS-IV), activities of daily living, and time with good mobility all significantly improved. Serious adverse events related to surgery occurred in 17.7% (lead revision, infection, intracranial hemorrhage), with no deaths related to surgery(4).
EARLYSTIM shifted the paradigm for DBS candidacy: patients no longer need to be severely disabled to benefit. The trial supported DBS consideration as soon as motor complications become disabling — within the first 3 years of complication onset in appropriately selected patients (age <60, good levodopa response, no dementia).
CSP 468: GPi vs STN
CSP 468 (Follett et al., 2010) was a VA Cooperative Study randomizing 299 patients with advanced PD and motor complications to bilateral GPi or bilateral STN DBS, with 24-month follow-up(5).
The primary outcome — change in UPDRS-III motor score in the off-medication/on-stimulation state — showed no significant difference between targets (GPi: −11.8 points vs STN: −10.7 points; difference −1.1; P=0.50). However, important secondary differences emerged: STN-DBS allowed significantly greater dopaminergic medication reduction (−408 vs −243 mg levodopa equivalents, P=0.02); STN-DBS was associated with greater decline in visuomotor processing speed (P=0.03) and worsening depression (P=0.02), while GPi-DBS showed stable or improved mood; and GPi required more reoperations for programming or hardware issues(5).
PD SURG: DBS vs Medical Therapy
PD SURG (Williams et al., 2010) randomized 366 patients with PD and inadequate symptom control to DBS (primarily STN, some GPi) plus medical therapy versus medical therapy alone for 1 year. DBS improved PDQ-39 summary index by −5.0 points versus −0.3 for medical therapy (difference −4.7, P=0.001), with significant improvements in mobility (P=0.0004), ADLs (P<0.0001), and off-state motor score (−16.8 point difference, P<0.0001). Serious surgery-related adverse events occurred in 19%(6).
NSTAPS: 3-Year GPi vs STN Outcomes
The NSTAPS 3-Year follow-up (Odekerken et al., 2016) of 128 patients randomized to bilateral GPi or STN DBS found that at 3 years, STN-DBS showed significantly greater improvement in off-drug UPDRS-ME scores (median STN 28 vs GPi 33, P=0.04). There was no between-group difference in the composite score for cognitive, mood, and behavioral effects (GPi 83% vs STN 86% with negative outcome, P=0.69). STN-DBS allowed significantly greater medication reduction, while GPi required more reoperations(7).
🔹 Clinical Relevance: GPi vs STN Target Selection
- STN is the most commonly chosen target due to greater medication reduction, cost advantages (lower stimulation energy), and modestly greater off-state motor improvement at 3 years (NSTAPS).
- GPi may be preferred when: pre-existing depression or mood instability (GPi showed mood stability vs STN worsening in CSP 468), cognitive borderline status (GPi had less cognitive decline), severe disabling dyskinesia as the dominant problem (GPi provides direct antidyskinetic effect without requiring medication reduction), or mild cognitive impairment.
- DBS candidacy criteria: Confirmed levodopa-responsive PD (≥30% UPDRS-III improvement on levodopa), disabling motor complications despite optimized medical therapy, no dementia (Mattis DRS >130 or equivalent), no untreated psychiatric disorders, and appropriate surgical risk profile.
- Key discussion points with patients: DBS improves symptoms but does not slow disease progression; levodopa-responsive symptoms respond best; speech, balance, and cognition may not improve and can worsen; hardware complications (infection ~3–5%, lead migration, battery replacement) are expected over time.
Continuous Drug Delivery Systems
For patients who are not DBS candidates or who prefer non-surgical options, continuous drug delivery systems aim to provide more physiologic, steady-state dopaminergic stimulation:
Levodopa-carbidopa intestinal gel (LCIG/Duopa): Delivered via percutaneous endoscopic gastrojejunostomy (PEG-J) tube and a portable infusion pump. Provides continuous jejunal levodopa delivery, reducing OFF time by ~4 hours/day and increasing ON time without troublesome dyskinesia. MDS EBM rated "clinically useful." Limitations include device complications (tube dislocation, kinking, stoma-site infection in ~30%), cost, and the need for a permanent abdominal stoma. Polyneuropathy has been reported with long-term use (possibly related to vitamin B12/homocysteine effects).
Subcutaneous levodopa infusion (foslevodopa-foscarbidopa / Produodopa / Vyalev): A newer approach using subcutaneous continuous infusion, avoiding the need for intestinal tubes. Approved in Europe (2024) and under FDA review. Provides 24-hour subcutaneous levodopa with similar efficacy to LCIG. Infusion site nodules and reactions are the main concern.
Apomorphine continuous subcutaneous infusion: Used widely in Europe and Australia (less in the US) for advanced motor fluctuations. Delivered via a small ambulatory pump, it provides continuous D1/D2 agonist stimulation, reducing OFF time and often allowing significant levodopa dose reduction. Skin nodules at the infusion site are common and can be minimized with site rotation and ultrasound-guided injection.
Clinical Decision Algorithm
| Stage of Motor Complications | Management Steps | Specific Interventions |
|---|---|---|
| Early wearing off | Optimize levodopa dosing → Add oral adjunct | Increase dose frequency; MAO-B inhibitor (preferred first adjunct per PD MED) → COMT inhibitor → DA → istradefylline |
| Persistent wearing off despite oral adjuncts | Add rescue therapy → Consider advanced therapy referral | Inhaled levodopa (Inbrija) or apomorphine SC (Apokyn) for breakthrough OFFs. ER levodopa (Rytary) for smoother delivery |
| Peak-dose dyskinesia | Reduce levodopa peak → Add amantadine | Lower individual levodopa doses, increase frequency; discontinue COMT-i if present; add amantadine IR or Gocovri (ER, only FDA-approved agent) |
| Combined fluctuations + dyskinesia refractory to oral therapy | Advanced therapies | DBS (if good levodopa response, no dementia, age-appropriate), LCIG/Duopa, or SC levodopa infusion. Target selection: STN (most common) vs GPi (mood/cognitive concerns) |
| Diphasic dyskinesia | Most difficult to treat | Higher individual levodopa doses (paradoxically) to push through threshold zone; consider DBS or continuous infusion |
| Off-period dystonia | Ensure adequate dopaminergic coverage overnight | CR levodopa at bedtime, rotigotine patch, early morning rescue therapy; botulinum toxin for focal dystonia if refractory |
Trial Comparison Table
| Trial | Year | N | Population | Comparison | Primary Outcome | Key Secondary / Safety |
|---|---|---|---|---|---|---|
| EARLYSTIM | 2013 | 251 | PD with early motor complications (age <60, disease ≥4 yr, H&Y ≤3) | STN-DBS + medical vs medical alone | PDQ-39: improved 7.8 vs worsened 0.2 (P=0.002) | UPDRS-III (off): −16.4 point advantage for DBS. Surgery-related SAE 17.7% |
| CSP 468 | 2010 | 299 | Advanced PD, disabling motor complications | GPi DBS vs STN DBS | UPDRS-III (off/on-stim): GPi −11.8 vs STN −10.7 (P=0.50) | STN: greater med reduction (P=0.02), worse depression (P=0.02), worse visuomotor speed (P=0.03) |
| PD SURG | 2010 | 366 | PD with inadequate symptom control | DBS + medical vs medical alone | PDQ-39: −5.0 vs −0.3 (P=0.001) | UPDRS motor (off): −16.8 difference (P<0.0001). Surgery SAE 19% |
| NSTAPS 3-Year | 2016 | 128 | Advanced PD, 3-yr DBS follow-up | GPi DBS vs STN DBS | UPDRS-ME (off-drug): STN 28 vs GPi 33 (P=0.04, favoring STN) | No cognitive/mood difference. STN: greater med reduction. GPi: more reoperations |
| PD MED (adjunct) | 2022 | 500 | PD with motor complications on levodopa | DA vs MAO-Bi vs COMT-i (as first adjunct) | MAO-Bi > COMT-i: PDQ-39 mobility +4.2 (P=0.03) | DA vs MAO-Bi: NS. Trends for less dementia with MAO-Bi |
| EASE LID | 2017 | 126 | PD with troublesome LID on stable levodopa | Gocovri 274 mg vs placebo | UDysRS: ~20% reduction in LID scores (P<0.001) | Also reduced OFF time. AE: hallucinations, edema, livedo reticularis |
References
- Fox SH, Katzenschlager R, Lim SY, et al. International Parkinson and movement disorder society evidence-based medicine review: update on treatments for the motor symptoms of Parkinson's disease. Mov Disord. 2018;33(8):1248–1266.
- de Bie RMA, Katzenschlager R, Swinnen BEKS, et al. Update on treatments for Parkinson's disease motor fluctuations — an International Parkinson and Movement Disorder Society evidence-based medicine review. Mov Disord. 2025;40(5):776–794.
- Gray R, Ives N, Rick C, et al. Long-term effectiveness of adjuvant treatment with catechol-O-methyltransferase or monoamine oxidase B inhibitors compared with dopamine agonists among patients with Parkinson disease uncontrolled by levodopa therapy (PD MED). JAMA Neurol. 2022;79(2):131–140.
- Schuepbach WMM, Rau J, Knudsen K, et al. Neurostimulation for Parkinson's disease with early motor complications (EARLYSTIM). N Engl J Med. 2013;368(7):610–622.
- Follett KA, Weaver FM, Stern M, et al. Pallidal versus subthalamic deep-brain stimulation for Parkinson's disease (CSP 468). N Engl J Med. 2010;362(22):2077–2091.
- Williams A, Gill S, Varma T, et al. Deep brain stimulation plus best medical therapy versus best medical therapy alone for advanced Parkinson's disease (PD SURG). Lancet Neurol. 2010;9(6):581–591.
- Odekerken VJJ, Boel JA, Schmand BA, et al. GPi vs STN deep brain stimulation for Parkinson disease: three-year follow-up (NSTAPS). Neurology. 2016;86(8):755–761.