We report the first large-scale double-blind, randomly assigned study to compare two active dopaminergic therapies for Restless Legs Syndrome (RLS), the dopamine agonist cabergoline (CAB) and levodopa/benserazide (levodopa). Patients with idiopathic RLS were treated with fixed daily doses of 2 or 3 mg CAB or 200 or 300 mg levodopa for 30 weeks. Efficacy was assessed by changes in the IRLS (International RLS Severity Scale) and by time to discontinuation of treatment due to loss of efficacy or augmentation. 361 of 418 screened patients (age 58 +/- 12 years, 71% females) were randomly assigned and treated (CAB: n = 178; levodopa: n = 183) in 51 centers of four European countries. Baseline IRLS total score was 25.7 +/- 6.8. The baseline-adjusted mean change from baseline to week 6 in IRLS sum score was d = -16.1 in the CAB group and d = -9.5 in the levodopa group (d = -6.6, P < 0.0001). More patients in the levodopa group (24.0%) than in the CAB group (11.9%, P = 0.0029, log-rank test) discontinued because of loss of efficacy (14.2% vs. 7.9%, P = 0.0290) or augmentation (9.8% vs. 4.0%, P = 0.0412). Adverse events (AEs) occurred in 83.1% of the CAB group and in 77.6% of the levodopa group. In both groups, most frequent AEs were gastrointestinal symptoms (CAB: 55.6%, levodopa: 30.6%, P < 0.0001). This first large-scale active controlled study in RLS showed superior efficacy of cabergoline versus levodopa after a 30-week long-term therapy. Tolerability was found more favorable with levodopa than with cabergoline.