Randomised placebo-controlled trials of the addition of dexamethasone to standard treatment of acute migraine headache were eligible for inclusion, provided the therapy was initiated in an emergency department setting and that both participants and treating physicians were blinded. A required outcome was self-reported moderate or severe headache at 24 to 72 hour follow up (primary review outcome). The outcome could be defined by published severity scales or could be reported as a dichotomous variable.
The included studies were conducted in the emergency departments of community, urban and tertiary care hospitals. Average or median participant age was 35 to 40 years. Most participants were women (range 78 per cent to 86 per cent). Where reported, mean or median pain score on presentation was 8 to 9 (out of 10). Median headache duration was 12 to 48 hours. All studies used International Headache Society (IHS) criteria to diagnose migraine headache (where specified). In the included studies, dexamethasone was given as a single intravenous injection in most cases. The median dose used was 15 mg (range 8 mg to 24 mg). Concurrent anti-migraine therapies varied across the studies (for example, antiemetics, ketorolac and/or opioid); these were either pre-specified or were at the discretion of the treating physician. The studies used a categorical headache scale or a dichotomous measure (headache present/absent) to ascertain outcomes.
Two authors independently selected potentially eligible studies, from which all three authors selected studies for inclusion. Discrepancies were resolved by consensus.