ABSTRACT
We performed a network meta-analysis (NMA) to enhance the corresponding evidence of the relative efficacy and safety of different antiplatelet agents in cerebral ischemic disease. PubMed and EMBASE were searched systematically for relevant studies. Outcomes were compared using odds ratios and 95% credible intervals. Each agent was ranked according to the value of surface under the cumulative ranking curve (SUCRA). Publication bias was evaluated by funnel plots, while consistency between direct and indirect comparison was analyzed by node-splitting and heat plots. Besides, the clustering technique was used to categorize similar agents. A number of 44 eligible studies with 148 578 patients were included in this NMA. In terms of efficacy (including mortality, recurrent stroke, and vascular event), all six interventions were better than placebo. clopidogrel (Clop) and aspirin (ASA)+Clop were the best two interventions from SUCRA. However, the performance of ASA+Clop declined significantly when considering safety (including myocardial infarction, all-cause withdrawal, and intracranial hemorrhage), especially worse in intracranial hemorrhage. In conclusion, Clop was potentially the most preferable treatment for preventing cerebral ischemic in terms of efficacy and safety. However, the addition of ASA was associated with a potential increase in intracranial hemorrhage, therefore, combination therapy of ASA and Clop should be introduced with caution although it may be more effective than the monotherapy of ASA.