Valproate Versus Ketorolac Versus Metoclopramide

One million patients present to US emergency departments (ED) annually for treatment of acute migraine headache. A variety of parenteral medications are used to treat acute migraine, but none offer rapid and complete headache relief without side effects. Preliminary studies have suggested that intravenous valproate, an anti-epileptic medication with established efficacy as a migraine preventive, may be useful for the treatment of acute migraine. We propose a randomized, comparative efficacy trial in which intravenous valproate is compared to two standard parenteral therapies for acute migraine. There will not be a placebo control. Included subjects will be adults 64 years and younger who meet International Headache Society criteria for acute migraine, who do not have clinical evidence of secondary (organic) headache, and who do not have allergy or contra-indication to the investigational medications. The investigational medications are 1gm of valproate, 10mg of metoclopramide, and 30mg of ketorolac. Patients will be approached for participation and randomized at the time of presentation to the ED. Pain will be assessed on an 11 point (0 to 10) verbal pain scale, validated for use in acute pain trials. Medication will be infused as a slow intravenous drip. The primary outcome will be an improvement in headache intensity one hour after initiation of the intravenous drip. Secondary outcomes include assessments of pain, functional disability, adverse events, and satisfaction with the investigational medication one, two and 24 hours after initiation of the intravenous drip. The primary analysis will use a Student's t-test for independent samples and involve three pair-wise comparisons. Using an alpha of .017 (to account for the three pairwise comparisons), a standard beta, and a validated minimum clinically significant difference on the verbal pain scale, we calculated the need for 330 subjects. An interim analysis will be conducted to determine lack of efficacy of valproate.