A Randomized, Double-Dummy, Emergency Department-Based Study of Greater Occipital Nerve Block With Bupivacaine vs Intravenous Metoclopramide for Treatment of Migraine

Background: Greater occipital nerve blocks (GONB) are used increasingly to treat acute migraine. Objective: We conducted a randomized controlled trial to determine whether GONB was as effective as intravenous metoclopramide for migraine. Methods: This was a double-dummy, double-blind, parallel-arm, non-inferiority study conducted in 2 emergency departments (EDs). Patients with migraine of moderate or severe intensity were randomized to receive bilateral GONB with each side administered 3 mL of bupivacaine 0.5% or metoclopramide 10 mg IV, the putative standard of care. The primary outcome was improvement in pain on a 0-10 scale between time 0 and 1 hour later. To reject the null hypothesis that metoclopramide would be more efficacious in relieving pain, we required that the lower limit of the 95% CI for the difference in pain improvement between those randomized to GONB vs those randomized to metoclopramide be >−1.3, a validated minimum clinically important difference. Secondary outcomes included sustained headache relief, defined as achieving and maintaining for 48 hours a headache level of mild or none without the use of additional analgesic medication, and the use of rescue medication in the ED. Results: Over a 2.5-year study period, 1358 patients were screened for participation and 99 were randomized, 51 to GONB and 48 to metoclopramide. All of these patients were included in the primary analysis. Patients who received the GONB reported mean improvement of 5.0 (95% CI: 4.1, 5.8) while those who received metoclopramide reported a larger mean improvement of 6.1 (95% CI: 5.2, 6.9). The 95% CI for the between group difference of −1.1 was −2.3, 0.1. Sustained headache relief was reported by 11/51 (22%) GONB and 18/47 (38%) metoclopramide patients (95% CI for rounded difference of 17%: −1, 35%). Of the 51 GONB patients, 17 (33%) required rescue medication in the ED vs 8/48 (17%) metoclopramide patients (95% CI for rounded difference of 17%: 0, 33%). An adverse event was reported by 16/51 (31%) GONB patients and 18/48 (38%) metoclopramide patients (95% CI for (rounded) difference of 6%: −13, 25%). Conclusion: GONB with bupivacaine was not as efficacious as IV metoclopramide for the first-line treatment of migraine in the ED.