To compare the efficacy of divalproex sodium (Depakote) with that of propranolol hydrochloride (and placebo) for the prophylaxis of migraine without aura.
Single-investigator, randomized, single-blind, placebo-controlled study with 5 phases: baseline (weeks 1-4); placebo (weeks 5-8); first treatment, 1 agent (weeks 9-20); washout (weeks 21-24); and second treatment, crossover to other agent (weeks 25-36).
Private practice of a general neurologist with a special interest in headache disorders.
Of 37 patients (30 women and 7 men) selected, 32 completed the study. All received placebo, after which half were randomized to receive divalproex or propranolol, then crossed over after washout.
Divalproex and propranolol doses were titrated during the initial 8 weeks of each 12-week treatment cycle. For divalproex, doses were titrated to 1500 mg/d in 23 patients, to 2000 mg/d in 2, and downward in 7; the mean valproate sodium trough level was 68.5 mg/L. Propranolol was titrated to 180 mg/d in 28 patients, to 240 mg/d in 1, and downward in 3.
Migraine frequency was reduced in 19% (6/32) of placebo-treated, 66% (21/32) of divalproextreated, and 63% (20/32) of propranolol-treated patients. Assessment of migraine-days per month revealed significant response to placebo in 22% (7/32) of patients, to divalproex in 66% (21/32), and to propranolol in 69% (22/32). When results were limited to the third month of each active-agent treatment phase, 75% (24/32) of patients receiving divalproex and 78% (25/32) of those receiving propranolol had reduction in migraine frequency.
No significant difference was identified between divalproex and propranolol for the prophylaxis of migraine without aura.