Eleven RCTs were included (n=625). Four trials had Jadad scores of five points (the maximum); two scored 4 points, two scored 3, and three scored 2.
Meperidine versus active controls
Headache relief: A dihydroergotamine regimen was significantly more effective than meperidine (odds ratio 0.30, 95% CI: 0.09, 0.97, four RCTs, n=254), but statistical heterogeneity was high (I2=73%). There was no statistically significant difference in effectiveness between meperidine and anti-emetics, though there was a trend in favour of antiemetics (odds ratio 0.46, 95% CI: 0.19, 1.11, four RCTs, n=248); with considerable statistical heterogeneity (I2=51%). There was no statistically significant difference in effectiveness between meperidine and ketorolac (three RCTs, n=123). There was a trend in the data for larger and better quality studies to have less extreme odds ratios.
Adverse effects: Meperidine caused significantly more dizziness than dihydroergotamine regimens (odds ratio 8.67, 95% CI 2.66, 28.23, three RCTs). Meperidine caused less akathisia than antiemetics (odds ratio 0.10, 95% CI: 0.02, 0.57, three RCTs). There were no statistically significant differences between the groups for other side effects, but there was a trend for meperidine to cause more sedation than dihydroergotamine regimens (odds ratio 3.52, 95% CI: 0.87, 14.19, three RCTs).
Butorphanol versus dihydroergotamine regimen
No statistically significant differences were found for any outcome (one RCT).
Other results were reported in the review.