Unveiling the comparative efficacy and tolerability of comprehensive treatments for migraine: a protocol of systematic review and Bayesian network-meta analysis

Abstract

Background Migraine is a chronic paroxysmal incapacitating neurological disorder, which endangers the health of human worldwide ranking as the third most prevalent medical condition. There are no comprehensive estimates of treatments for migraine. We will conduct this systematic review and Bayesian network meta-analysis (NMA) to synthesis quantitative and comparative evidence on the efficacy and tolerability of all the known pharmacological and non-pharmacological interventions for migraine. Method We will perform the systematic electronic search of the literature utilizing MEDLINE, Embase, Cochrane Central Register of Controlled Trials (CENTRAL), Cumulative Index to Nursing & Allied Health (CINAHL), and PsycINFO. We will only include randomized controlled trials (RCTs) of high quality which appraise the efficacy or safety of any potential pharmacological or non-pharmacological interventions in the treatment of patients with migraine. The traditional pairwise meta-analyses will be performed to anticipate the heterogeneities and publication bias and the NMA will be conducted within a Bayesian hierarchical model framework to obtain estimates for all valuable treatments for migraine. The entire heterogeneity will be quantified by Q statistic and I2 index. Other analyses included sensitivity analyses, meta-regression, and subgroup analyses will also be conducted. The whole process will be conducted using in R-3.6.0 software. Results This study will obtain the efficacy and tolerability of all potential treatments for migraine, aiming at providing consolidated evidence to help make the best choice of interventions. The results will be published in a peer-reviewed journal. Discussion This Bayesian network meta-analysis may be the first attempt to quantitatively synthesize the efficacy and tolerability of all potential treatments for migraine. And this method can ensure us to fully utilize both the direct and indirect evidence as well as gain the comparative estimates displayed in the derived hierarchies. Besides, we have registered this protocol on the international prospective register of systematic review (PROSPERO) (CRD42020157278).