MDS EBM Review: Treatment of Motor Fluctuations in Parkinson's Disease (2025 Update)
This is a comprehensive update of the International Parkinson and Movement Disorder Society (MDS) Evidence-Based Medicine (EBM) review on treatments for motor fluctuations in Parkinson's disease (PD). Published by de Bie et al. in Movement Disorders (2025), it uses a revised methodology incorporating the Cochrane Risk of Bias tool and a modified GRADE framework. A total of 102 RCTs were included following a literature search through January 2023.
Bottom Line
- Efficacious (highest designation): Levodopa ER (IPX066), pramipexole IR & ER, ropinirole IR, rotigotine, opicapone, safinamide, bilateral STN DBS
- Likely efficacious: Levodopa intestinal gel, subcutaneous foslevodopa, ropinirole PR, ropinirole patch, continuous SC apomorphine, entacapone, rasagiline, istradefylline, amantadine CR, zonisamide, bilateral GPi DBS, unilateral pallidotomy
- Key methodology change: Efficacy conclusions now based on clinical relevance of outcomes (using MCID thresholds) rather than just effect size — leading to several reclassifications from the prior 2018 review
- Removed agents: Pergolide, bromocriptine, cabergoline, and tolcapone are no longer included due to insufficient evidence or market withdrawal
Methodology
The MDS EBM Committee used the following PICOT framework:
- Population: PD patients on oral levodopa experiencing motor fluctuations (early or persistent)
- Intervention: Any commercially available treatment
- Comparator: Placebo or active comparator including best medical treatment (BMT)
- Outcomes: OFF-time duration, ON-time duration, off-medication severity (UPDRS/MDS-UPDRS), disability, HR-QoL, adverse effects
- Timepoints: ≥3 months follow-up from RCTs only
Key Methodological Changes from 2018 Review
- Adopted modified GRADE framework — evidence quality assessed on risk of bias, imprecision, inconsistency, indirectness, and publication bias
- Used Cochrane Risk of Bias tool (8 items) instead of the previous Rating Scale for Quality of Evidence
- Efficacy conclusions now consider clinical relevance via MCID thresholds (e.g., 1 h/day for ON/OFF-time, 4 points for MDS-UPDRS-III) rather than just statistical significance
- Both low and very low quality evidence now result in "Insufficient Evidence" regardless of effect size
- Entirely new literature search (not supplementing prior reviews)
Summary of Efficacy Conclusions
| Class | Intervention | Efficacy Conclusion | Safety |
|---|---|---|---|
| Levodopa/PDI | Controlled release | Insufficient evidence | Acceptable risk with specialized monitoring (infusion/SC) |
| IPX066 (ER) | Efficacious | ||
| Intestinal infusion | Likely efficacious | ||
| Subcutaneous infusion | Likely efficacious | ||
| Dopamine Agonists | Pramipexole IR | Efficacious | — |
| Pramipexole ER | Efficacious | ||
| Ropinirole IR | Efficacious | ||
| Ropinirole PR | Likely efficacious | ||
| Ropinirole patch | Likely efficacious | ||
| Rotigotine | Efficacious | ||
| Apomorphine continuous SC | Likely efficacious | Acceptable risk with specialized monitoring | |
| Apomorphine intermittent | Insufficient evidence | — | |
| COMT Inhibitors | Entacapone | Likely efficacious | — |
| Opicapone | Efficacious | ||
| MAO-B Inhibitors | Rasagiline | Likely efficacious | — |
| Selegiline | Insufficient evidence | ||
| MAO-B + Channel Blocker | Safinamide | Efficacious | — |
| Others | Istradefylline | Likely efficacious | — |
| Amantadine CR | Likely efficacious | ||
| Zonisamide | Likely efficacious | ||
| Amantadine IR | Insufficient evidence | ||
| Surgery | Bilateral STN DBS | Efficacious | Acceptable risk with specialized monitoring |
| Bilateral GPi DBS | Likely efficacious | ||
| Unilateral pallidotomy | Likely efficacious | ||
| Subthalamotomy | Insufficient evidence |
GRADE Efficacy Framework
| Evidence Level | Important Benefit (MCID met) | Intermediate Benefit (uncertain clinical relevance) | No Important Benefit (MCID not met) |
|---|---|---|---|
| High | Efficacious | Likely efficacious | Not efficacious |
| Moderate | Likely efficacious | Likely efficacious | Unlikely efficacious |
| Low / Very Low | Insufficient evidence | Insufficient evidence | Insufficient evidence |
MCID thresholds used: 1 hour/day for ON-time and OFF-time changes; 4 points for MDS-UPDRS-Part III and UPDRS-Part III; 3 points for UPDRS-Part II; 5 points for PDQ-39.
Levodopa Formulations
Levodopa Carbidopa Controlled Release
Three trials evaluated standard controlled-release levodopa (Sinemet CR). All had significant design limitations and raised serious risk of bias concerns. No difference was found versus immediate release in reducing OFF-time or improving disability scores. Conclusion: Insufficient evidence.
IPX066 (Levodopa Carbidopa Extended Release)
One large phase 3 trial (n=393) showed IPX066 reduced OFF-time by 1.2 hours (P < 0.0001) versus immediate release, with increased ON-time without troublesome dyskinesia (0.9 h, P < 0.001) and ON-time without any dyskinesia (0.7 h, P < 0.05). Mean dosing frequency was 3.6 doses/day for IPX066 versus 5.0 doses/day for IR. IPX066 did not reach clinical relevance for disability improvement. Conclusion: Efficacious.
Levodopa Carbidopa Intestinal Gel (LCIG)
Two trials evaluated LCIG (~16 h/day infusion). One placebo-controlled study showed OFF-time reduction of 1.9 hours (P < 0.01), increased ON-time without troublesome dyskinesia (1.9 h, P < 0.01), and ON-time without any dyskinesia (2.3 h, P = 0.01). Disability improved by 3 points on UPDRS-II (P < 0.001) with a 7-point PDQ-39 improvement (P < 0.05). A second open-label study versus BMT showed greater disability improvement but no significant motor impairment or HR-QoL differences. Conclusion: Likely efficacious (limited by small sample size and short follow-up).
Continuous Subcutaneous Foslevodopa Foscarbidopa
One placebo-controlled phase 3 trial showed OFF-time reduction of 1.8 hours (P = 0.002) with increased ON-time without troublesome dyskinesia (1.8 h, P < 0.01). Insufficient evidence for effects on disability and quality of life. Conclusion: Likely efficacious.
Dopamine Agonists
Pramipexole
Six trials evaluated pramipexole IR and ER. Five studies versus placebo showed statistically significant and clinically relevant reduction in OFF-time, improvement in MDS-UPDRS-II off-medication state, and one study showed clinically meaningful increase in ON-time without troublesome dyskinesia and quality of life. Both IR and ER formulations provided clinically meaningful improvement of OFF-time duration, off-medication state MDS-UPDRS-II, and total PDQ-39. Pramipexole was non-inferior to rotigotine and comparable to bromocriptine. Conclusion: Efficacious.
Ropinirole
Fourteen trials were included. Ropinirole IR: four trials showed statistically significant borderline clinically meaningful OFF-time reduction. Ropinirole PR: three studies showed OFF-time reduction; likely efficacious but high risk of bias in two of three studies. Ropinirole patch: one study showed clinically relevant OFF-time reduction but no relevant change in ON-time duration or disability. Efficacy was comparable between ropinirole IR, ropinirole PR, and rotigotine and bromocriptine.
- Ropinirole IR: Efficacious
- Ropinirole PR: Likely efficacious
- Ropinirole patch: Likely efficacious
Rotigotine
Six trials versus placebo showed statistically significant and clinically meaningful OFF-time reduction of 0.9–1.8 hours. Treatment difference versus placebo was >1 h/day in five of six studies. All six showed improvement in disability with statistical significance in four. Evidence for effects on quality of life was limited. Conclusion: Efficacious.
Apomorphine
Continuous subcutaneous infusion (~16 h/day): One trial showed clinically meaningful OFF-time reduction of 1.9 hours (P = 0.0025) and ON-time without troublesome dyskinesia improvement of 2 hours (P = 0.0008). Concerns about unblinding due to skin nodules. Likely efficacious.
Intermittent (sublingual film and SC pen): Both improved off-medication state within 15–40 minutes but neither showed sustained improvement over the 12-week treatment period. MDS-UPDRS-II scores did not change. Insufficient evidence.
COMT Inhibitors
Entacapone
Six RCTs compared entacapone with placebo. Changes in ON-time duration were assessed in five and OFF-time in five. Concerns regarding allocation concealment led to moderate evidence quality. Entacapone did not show a statistically significant improvement in quality of life. Insufficient evidence for superiority over cabergoline or for switching immediately versus delayed to levodopa-carbidopa-entacapone. Conclusion: Likely efficacious.
Opicapone
Three trials showed statistically significant and clinically meaningful OFF-time reduction (~120 min in two trials, ~40 min in a third). Corresponding ON-time increase of ~60 min. None of the three trials showed quality of life improvement on PDQ-39. Two studies assessed disability (UPDRS-II): one found no significant improvement, the other reported statistically significant but minimal improvement (~1 point). Conclusion: Efficacious.
MAO-B Inhibitors
Rasagiline
Six trials evaluated rasagiline at 0.5 and 1.0 mg/day. Reduction in OFF-time ranged from 0.5–0.9 h/day — statistically significant but not clinically meaningful by MCID threshold. Four trials showed statistically significant but not clinically meaningful motor improvement (MDS-UPDRS-III or UPDRS-III). ON-time without troublesome dyskinesia increased by 0.5–1.2 hours. One trial noted a dose-dependent increased risk of dyskinesia. Three trials assessed QoL via PDQ-39: statistically significant but not clinically meaningful. Conclusion: Likely efficacious.
Selegiline
Five trials met criteria. Main concerns were low quality of evidence, small sample sizes, imprecision, and heterogeneous results. Only one of four studies reported a meaningful OFF-time reduction. Mixed results for disability. Conclusion: Insufficient evidence.
Safinamide
Four trials showed clinically meaningful OFF-time reduction (~1 h vs placebo) and increased ON-time without troublesome dyskinesia (~1 h vs placebo). Motor impairment in off-medication state was not evaluated. QoL results were mixed — only one study reached the 5-point PDQ-39 MCID threshold, but with a large confidence interval. Conclusion: Efficacious.
Other Pharmacological Agents
Istradefylline (Adenosine A2A Receptor Antagonist)
Eight trials evaluated doses of 10, 20, 40, and 60 mg daily. OFF-time reduction was inconsistent and of unclear clinical relevance (12–114 min depending on study size). ON-time without troublesome dyskinesia increase was similarly inconsistent (12–48 min). No improvement in UPDRS-III on- or off-medication. Conclusion: Likely efficacious.
Amantadine
Extended release (ADS-5102): Two trials found clinically meaningful increase in ON-time (without troublesome dyskinesia, with troublesome dyskinesia, and overall). OFF-time reduction was not clinically meaningful but washout data suggested a significant treatment effect. Neither trial assessed off-medication motor scores or quality of life. Likely efficacious.
Immediate release: Excluded after screening due to short study duration and high dropout. Insufficient evidence.
Zonisamide
Two trials evaluated doses of 25, 50, and 100 mg daily. OFF-time reduction ranged from ~40–85 min with a dose-dependent gradient. On-medication UPDRS-III improvement was 1.5–3.8 points versus placebo. One study showed zonisamide 50 mg improved UPDRS-II by ~1 point. Dropout rate was considerable (~20%). Conclusion: Likely efficacious.
Agents with Insufficient Evidence
- Terguride: 1 trial — no meaningful improvement on motor scales
- Nicotine: 1 open-label trial — no significant difference in MDS-UPDRS-III; cannot exclude placebo effect
- Perampanel: 2 trials — failed to reduce OFF-time or improve ON-time, disability, or motor impairment
- Helicobacter pylori eradication: 1 trial (n=34) — improved UPDRS-III after 3 months, suggesting effect on levodopa absorption
- Coenzyme Q10: 1 trial — Not efficacious (only agent with this designation)
- Exenatide: 2 small trials — ~4-point MDS-UPDRS-III improvement (at MCID threshold) but no differences in other outcomes; effect limited to a single measure
Surgical Interventions
Bilateral Subthalamic Nucleus (STN) DBS
Eight RCTs (the most of any intervention in this review) compared STN DBS to BMT or delayed stimulation. Seven studies showed improvement in off-medication UPDRS-III (average improvement 18.6 points). One sham-controlled trial showed 3.0-hour increase in ON-time without troublesome dyskinesia (P < 0.0001). Average OFF-time reduction was 2.6 hours. Four studies showed moderate evidence for disability improvement (UPDRS-II: 4.5–8.8 points). Four studies showed QoL improvement (PDQ-39: 6.5–10.0 points), though evidence was downgraded to moderate due to blinding concerns. Conclusion: Efficacious.
Bilateral GPi DBS
Five studies compared bilateral GPi and STN DBS, with two having longer follow-up periods. Neither study clearly separated GPi DBS results from STN DBS. The Veterans Affairs CSP 468 study (n=121 DBS, n=134 BMT) compared 61 GPi and 60 STN DBS patients. At the 36-month follow-up, there was a trend toward better outcomes with GPi versus STN in secondary outcomes including ON-time without troublesome dyskinesia and PDQ-39, though Mattis Dementia Rating Scale scores declined faster in the STN group. Conclusion: Likely efficacious.
Unilateral Pallidotomy
Four trials compared pallidotomy to medical therapy. Both studies with blinded assessors showed a statistically significant and clinically meaningful UPDRS-III improvement of ~15 points in the pallidotomy group at 6 months, with no significant change in the medical therapy group. One trial showed unilateral pallidotomy was inferior to STN DBS at 6 and 12 months. Evidence for OFF-time, ON-time, and QoL effects was limited to one trial each. Conclusion: Likely efficacious.
Subthalamotomy
Two small trials (n=10 and n=16) met criteria. All procedures improved off-medication motor impairment, but without significant differences between groups. Severe hemiballismus was reported in 20–50% of patients. Conclusion: Insufficient evidence.
Additional Surgical Agents with Insufficient Evidence
- Bilateral zona incerta DBS: 1 trial (n=19) — 41% improvement in off-medication UPDRS-III with strong tremor response, but high risk of bias and small sample
- GDNF (intraputamenal infusion): 1 trial — failed to show superiority in UPDRS-III; no benefits in disability, dyskinesia, or quality of life
Changes from the 2018 MDS EBM Review
| Intervention | 2018 Conclusion | 2025 Conclusion | Reason for Change |
|---|---|---|---|
| Levodopa intestinal infusion | Efficacious | Likely efficacious | New methodology requires longer follow-up; limited by small sample and 3-month duration |
| Entacapone | Efficacious | Likely efficacious | Concerns about allocation concealment reduced evidence quality |
| Rasagiline | Efficacious | Likely efficacious | OFF-time reduction (0.5–0.9 h) did not meet 1 h/day MCID threshold |
| Zonisamide | Efficacious | Likely efficacious | Concerns about inconsistency and high dropout rates |
| Bilateral GPi DBS | Efficacious | Likely efficacious | Difficulty separating GPi results from STN in combined studies |
| Pallidotomy | Efficacious | Likely efficacious | No sham control; small sample sizes |
| Intermittent apomorphine | Efficacious | Insufficient evidence | No sustained improvement over treatment period; new minimum 3-month follow-up requirement |
| Pergolide, bromocriptine, cabergoline | Included | Removed | Insufficient evidence or no longer commercially available |
| Tolcapone | Included | Removed | More alternatives now available |
Discussion & Clinical Implications
- This review reflects evidence, not treatment priorities — a higher efficacy designation (e.g., STN DBS) does not mean it should be used before oral medications with a "likely efficacious" designation
- In practice, the first step for motor fluctuations remains adjusting the levodopa dosing schedule (more frequent doses, shorter intervals, empty stomach administration, treating constipation)
- If insufficient, consider oral or transdermal adjunctive therapies first, then infusion therapies or surgery
- The EARLYSTIM trial showed that STN DBS was superior to medical therapy in PD patients with early motor complications, suggesting earlier surgical referral may be beneficial
- The PD MED RCT (n=500) found that dopamine agonists had better PDQ-39 mobility scores than MAO-B inhibitors and COMT inhibitors at 4.5-year follow-up, though the difference with COMT inhibitors was not statistically significant
- Few head-to-head comparisons exist between treatments (e.g., DBS vs infusion, safinamide vs rasagiline), making it difficult to establish a treatment hierarchy
- Different outcome measures across studies complicate meta-analysis
- High-intensity focused ultrasound (FUS) RCTs were published after the literature search and will be included in future updates
References
- 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. doi:10.1002/mds.30162
- 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.
- Katzenschlager R, de Bie RMA, Costa J, Sampaio C. MDS Evidence Based Medicine Committee: Revision of the methodological process for Systematic Reviews: Adoption of the Modified GRADE system. 2021.