Results and lessons learnt from a randomized controlled trial: prophylactic treatment of vestibular migraine with metoprolol (PROVEMIG)

Background: Vestibular migraine (VM) is the most frequent cause of recurrent spontaneous attacks of vertigo causally related to migraine. The objective of the Prophylactic treatment of vestibular migraine with metoprolol (PROVEMIG) trial was to demonstrate that metoprolol succinate is superior to placebo in the prevention of episodic vertigo- and migraine-related symptoms in patients with VM. Methods: This phase III, two-arm, parallel-group, double-blind, randomized placebo-controlled trial was designed to be conducted at tertiary referral centres at neurology and ear, nose and throat departments of eight German university hospitals. The planned sample size was a total of 266 patients to be allocated. Adults aged 18 years or above diagnosed with probable or definitive VM according to the Neuhauser criteria 2001 were randomly assigned 1:1 to 6 months blinded metoprolol (maintenance dosage of 95 mg daily) or placebo. The primary efficacy outcome was the self-reported number of vertiginous attacks per 30 days documented by means of a paper-based daily symptom diary. The pre-specified time period of primary interest was defined as months 4 to 6. Secondary outcomes included the patient-reported number of migraine days and vertigo days, the Dizziness Handicap Inventory, and clinical assessments. Adverse events were reported throughout the whole 9-month study period. Results: At the time of trial termination, no evidence for a difference in the incidence of vertiginous attacks between groups was detected. For the full analysis set, the incidence rate ratio was 0.983 (95% confidence interval (CI) 0.902-1.071) for metoprolol versus placebo. In both groups, there was a significant decline over time in the overall monthly vertigo attacks by a factor of 0.830 (95% CI 0.776-0.887). Results were consistent for all subjective and objective key measures of efficacy. The treatment was well tolerated with no unexpected safety findings. Conclusions: After randomizing 130 patients PROVEMIG had to be discontinued because of poor participant accrual not related to the tolerability of the study medication or safety concerns; no treatment benefit of metoprolol over placebo could be established. Additional preparatory work is much needed in the development, psychometric evaluation and interpretation of clinically meaningful end points in trials on episodic syndromes like VM taking into consideration the complexity of this disease entity comprising two domains (vertigo- and headache-related disability).