Background: Although "uncomfortable and unpleasant" limb sensations are a core symptom of restless legs syndrome (RLS), change in sensory symptomatology is usually not evaluated as a treatment outcome. Methods: In two double-blind trials, patients with idiopathic RLS (n = 357 in trial 615 and 398 in trial 604) were randomized to placebo or pramipexole (optimized at 0.125, 0.25, 0.50, or 0.75 mg/day). For entry, trial 604 also required at least moderate mood disturbance. In both trials, 12-week change in RLS-related limb pain was assessed using a 100-mm visual analogue scale (VAS). Results: At baseline, approximately 75% of patients had limb-pain scores > 30. Treatment with pramipexole yielded significant score reduction as early as day 5. At week 12, median score reduction for pramipexole relative to placebo was -33.5 vs. -11.0(p < 0.0001)in trial 615 and -31.0 vs. -11.0(p < 0.0001)in trial 604. Conclusions: Painful sensations may be more frequent in RLS than has previously been appreciated, and their amelioration may be a facet of pramipexole's benefit even in patients with concurrent mood disturbance. Limb pain assessment, e.g., by a VAS, is a useful measure of change in RLS symptom severity. (C) 2010 Elsevier B.V. All rights reserved.