Randomised controlled trials (RCTs) that compared a parenterally administered phenothiazine (chlorpromazine, prochlorperazine and methotrimeprazine) with placebo or an active parenterally administered comparison, for the treatment of patients with acute migraine, were eligible for inclusion. Acute migraine had to be defined using the criteria established by the International Classification of Headache Disorders, or with a reasonable attempt made to specifically include migraine headaches rather than all benign headaches. Trials had to report data on headache intensity/clinical outcome within two hours of treatment.
The primary review outcome was relief of headache. Secondary outcomes were clinical success as defined by trial authors. If clinical success was not reported, use of rescue medication was used as a proxy.
Included trials compared chlorpromazine (0.04 to 0.1mg/kg or 12.5 to 25mg), prochlorperazine (10mg) or methotrimeprazine (37.5mg) with placebo, metoclopramide (0.1mg/kg or 10 to 20mg), sumatriptan (6mg), meperidine (0.4 mg/kg or 75mg), ketorolac (30 to 60 mg) or sodium valproate (500mg). The majority of included trials compared interventions to active comparators. Phenothiazines were administered intravenously or intramuscularly. None of the trials included co-administration of other agents. Headaches were either diagnosed according to International Headache Society criteria, defined criteria or were physician diagnosed. Some trials also administered diphenhydramine or metoclopramide in addition to the studied drugs.
Abstracts identified by the searches were screened by one reviewer; full papers were ordered for potentially relevant abstracts and screened by two reviewers for inclusion. Disagreements were resolved through consensus or referral to a third reviewer.