Answers are grounded in this paper and cite their sources.
Background: Whether the initial choice between levodopa (L-Dopa) and a dopamine agonist (DA) modifies the biological course of Parkinson's disease (PD), rather than merely its symptomatic expression, remains one of the most persistent controversies in movement-disorder therapeutics. The historical hypotheses of L-Dopa-related oxidative toxicity and DA-mediated neuroprotection have never been reconciled with long-term clinical outcomes. Objective: To synthesise randomised and non-randomised comparative evidence published within the last 15 years on the effects of L-Dopa versus DAs on neuroimaging and clinical markers of progression, motor complications, long-term disease milestones, patient-reported quality of life (QoL) and safety. Methods: This review followed PRISMA 2020. PubMed/MEDLINE, Europe PMC, Semantic Scholar and ClinicalTrials.gov were searched from 1 January 2011 to 13 November 2026. Randomised controlled trials (RCTs), comparative studies, cohorts and case series directly comparing L-Dopa-based with DA-based strategies were eligible. Risk of bias was assessed with RoB 2 and ROBINS-I. Random-effects (DerSimonian–Laird) models with Hartung–Knapp–Sidik–Jonkman (HKSJ) sensitivity analyses pooled hazard ratios (HR), risk ratios (RR), mean differences (MD) and Hedges g; certainty was rated with GRADE. Results: Of 2,009 records, 11 studies (6 RCTs, 5 cohorts; 8,423 participants in relevant comparison arms) were included and 9 contributed to 10 quantitative syntheses. DA-first therapy reduced the hazard of dyskinesia/motor complications (HR 0.45, 95% CI 0.28–0.73; I² = 0%; low certainty), although the HKSJ interval crossed unity (0.16–1.28). No disease-modifying signal emerged: neuroimaging progression markers (g −0.16, −1.00 to 0.68), UPDRS-III change (MD −1.02, −3.77 to 1.74) and Hoehn–Yahr change did not differ. In the largest pragmatic RCT, levodopa-sparing therapy did not lower dementia (HR 1.23, 0.93–1.64), institutionalisation (HR 1.16, 0.85–1.59) or death (HR 1.18, 0.95–1.46). QoL was slightly worse with DAs (g 0.14, 0.04–0.24), discontinuation or failure of allocated therapy was more frequent (RR 3.15, 0.88–11.33; 2.14, 1.22–3.77 after excluding the dominant trial) and impulse control and related disorders were four-fold more frequent (ratio 4.15, 2.55–6.75). Conclusion: Contemporary evidence does not support a neuroprotective advantage of DAs over L-Dopa, nor a neurotoxic effect of L-Dopa, on the progression pattern of PD. The benefit of DA-first therapy is confined to delayed dyskinesia and is offset by poorer tolerability, neuropsychiatric harm and marginally lower QoL. Initial therapy should be individualised on symptomatic, not disease-modifying, grounds.
No takes yet. Share an insight, caveat, or question.
Delhi et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: