Comparative Efficacy And Security Of Antiplatelet Agents In The Secondary Prevention Of Ischemic Stroke Or Transient Ischemic Attack: A Network Meta-analysis

Objective:The purpose of this study was to conduct a network meta-analysis todetermine which antiplatelet or combination of different antiplatelet agents ismost efficacious and tolerable in patients with prior ischemic stroke or transientischemic attack (TIA).Methods:A comprehensive literature search was conducted in Pubmed and Embase (upto October2013) to identify randomized trials evaluating the role of variousantiplatelet agents and combinations for the secondary prevention of ischemicstroke or TIA. Data were screened and evaluated to generate direct and indirectcomparisons for serious vascular events, recurrent stroke, recurrent ischemicstroke and overall hemorrhagic events. Meta-analysis was performed to estimatethe relative efficacy and security of different antiplatelet regimens. Odds ratios(OR) and95%confidence intervals (CI) were calculated.Results:Forty-three trials involving83799ischemic stroke or TIA patients wereincluded. Forty trails reported the events of serious vascular event. In the networkmeta-analysis, seven antiplatelet regimens (aspirin, aspirin plus dipyridamole, clopidogrel, ticlopidine, cilostazol, triflusal, and combination of aspirin andclopidogrel) were significantly more effective than placebo. Four antiplateletregimens (aspirin plus dipyridamole, clopidogrel, cilostazol, and combination ofaspirin and clopidogrel) were significantly more effective than aspirin. Cilostazolwas significantly more effective than aspirin (OR=0.69,95%CI=0.55-0.85),clopidogrel (OR=0.78,95%CI=0.61-0.98), ticlopidine (OR=0.70,95%CI=0.55-0.89), and triflusal (OR=0.69,95%CI=0.51-0.96). Cilostazol was alsomarginally significantly more effective than aspirin plus dipyridamole (OR=0.80,95%CI=0.63-1.00). Direct comparisons showed the same results as the networkmeta-analysis.Thirty-two trails reported the events of hemorrhage. In the networkmeta-analysis, aspirin (OR=2.14,95%CI=1.73-2.72), aspirin plus dipyridamole(OR=1.86,95%CI=1.40-2.58), clopidogrel (OR=1.68,95%CI=1.20-2.59),ticlopidine (OR=1.94,95%CI=1.38-2.75), and combination of aspirin andclopidogrel (OR=4.96,95%CI=3.73-6.92) were associated with morehemorrhagic events than placebo. Cilostazol and triflusal were associated withsignificantly less hemorrhagic events than aspirin, aspirin plus dipyridamole,clopidogrel, ticlopidine, and combination of clopidogrel and aspirin. Directcomparisons showed the same results as the network meta-analysis.Conclusions:(1) Cilostazol was more safe and effective than other antiplatelet regimens.(2) Aspirin plus dipyridamole and clopidogrel were also more effective than aspirin without increasing the risk of bleeding.(3) Triflusal showed the similar efficacy as clopidogrel, aspirin plusdipyridamole and aspirin with lower bleeding risk.(4) Ticlopidine showed the similar efficacy as clopidogrel with poor tolerance.(5) Aspirin was the only antiplatelet regimen that has been compared totriflusal and cilostazol.