Our aim was to evaluate the safety and efficacy of botulinum toxin type A (BoNTA; BOTOX) for prophylaxis of episodic migraine. In this double-blind, placebo-controlled study, patients were randomized to 225, 150 or 75 U of BoNTA or placebo after a 30-day placebo run-in for three 90-day treatment cycles. The primary efficacy end-point was the mean reduction from baseline in the frequency of migraine episodes at day 180 in the placebo non-responder stratum. All groups (N = 495) improved, with no significant differences. At day 180, the frequency of migraine episodes was reduced from baseline means of 4.3, 4.7, 4.7 and 4.4 by 1.6, 1.7, 1.5 and 1.4 for BoNTA 225 U, 150 U and 75 U and placebo, respectively. The primary end-point was not met. Treatment-related adverse events were transient and mild to moderate. BoNTA treatment was safe and well tolerated but did not result in significantly greater improvement than placebo in this study. Several factors may have confounded the results.