Abstract
BACKGROUND: Vestibular migraine causes recurrent vertigo attacks that significantly impact quality of life. While various preventive medications are used, their comparative effectiveness was unknown. Previous systematic reviews have been limited by pairwise comparisons only or exclusion of newer treatments. The lack of head-to-head trials comparing all available treatments further complicates evidence-based decision-making. This evidence gap has real-world consequences, has also substantial economic burden of Vestibular migraine. There is a need for direct comparison studies between the most promising treatments to provide clearer guidance for clinical practice. OBJECTIVE: To determine the comparative effectiveness and safety of preventive treatments for vestibular migraine through systematic review and network meta-analysis. Given the lack of head-to-head randomized trials, a network meta-analysis (NMA) provides the most appropriate method to compare available treatments by combining both direct and indirect evidence. METHODS: We searched Embase, Scopus, PubMed, and Cochrane Library from inception to January 15, 2025. We included randomized controlled trials (RCTs) and prospective observational studies (n ≥ 30 for CGRP antagonists) comparing preventive treatments for vestibular migraine diagnosed according to either Bárány Society/International Headache Society criteria (post-2012) or Neuhauser criteria (pre-2012). Primary outcomes were monthly vertigo frequency and quality of life (DHI scores). We conducted frequentist network meta-analysis and assessed certainty using GRADE. RESULTS: From 340 identified records, nine studies met inclusion criteria. Five RCTs (419 patients) comparing seven treatments were included in the network meta-analysis. All treatments significantly reduced monthly vertigo attacks versus control. Propranolol ranked highest (P-score: 0.794; -7.04 attacks/month, 95% CI -12.77 to -1.31), followed by valproic acid (-5.95, 95% CI -9.01 to -2.89) and venlafaxine (-5.94, 95% CI -8.98 to -2.90). Galcanezumab showed moderate efficacy (-5.80, 95% CI -10.61 to -0.99) with zero discontinuations. Network heterogeneity was negligible (τ²<0.001). Evidence certainty was moderate for galcanezumab and low to very low for other treatments. CONCLUSIONS: All evaluated treatments effectively reduce vertigo frequency in vestibular migraine. While propranolol showed the largest effect, this relied on indirect evidence. Galcanezumab offers the best balance of efficacy, tolerability, and evidence quality. Head-to-head trials are urgently needed. PROSPERO REGISTRATION: CRD420251089507.