| Methods |
Double‐blind, placebo‐controlled, randomised trial |
| Participants |
47 patients, 41 female, 6 male, mean age 36 years, mean duration of headache before admission 75 hours. Inclusion criteria: diagnosis of migraine according to the IHSCC. Exclusion criteria: long QT‐interval, usage of drugs prolonging QT‐interval, hepatic disease, epilepsy or history of seizures, hyperthyreosis, parkinsonism, chronic psychiatric disease, other neuroleptic medication, and intoxication |
| Interventions |
Random assignment to either 5 mg haloperidol in 500 ml of normal saline or 500 ml of normal saline alone as a 20 to 30 minute infusion |
| Outcomes |
Pain estimation by a VAS between 1 hour and 3 hours after the infusion
Marked or almost total pain relief |
| Notes |
Type of pain: acute migraine headache; 44 patients included in the placebo‐controlled arm of the trial. Of these, 4 were rejected: 2 were included in the study twice, 1 did not fulfil the inclusion criteria and 1 was pain free before the infusion. Of the remaining 40, 36 were female and 4 male. The mean duration of the headache for these 40 patients was 67 h. 80% of patients reported side effects, mainly motor agitation (53%) and sedation (53%). |
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Random sequence generation (selection bias) |
Low risk |
Described as "...patients were randomized by envelope selection." |
| Allocation concealment (selection bias) |
Unclear risk |
Not reported |
| Blinding of participants and personnel (performance bias)
All outcomes |
Low risk |
Described as "...treatment was carried out by an attending nurse, who prepared and blinded the infusion." |
| Blinding of outcome assessment (detection bias)
All outcomes |
Low risk |
Described as "...treatment was carried out by an attending nurse, who prepared and blinded the infusion." |
| Incomplete outcome data (attrition bias)
All outcomes |
Unclear risk |
Imputation method not described |
| Size |
High risk |
Fewer than 50 participants/treatment arm |
