Background and Purpose- Two large-scale randomized controlled trials of recurrent
stroke prevention suggest that dual antiplatelet
therapy with
clopidogrel plus
aspirin is beneficial for prevention of subsequent ischemic events. There is a paucity of data, however, on the efficacy or effectiveness of such an approach in the treatment of
stroke patients with symptomatic large artery atherosclerotic occlusive disease. Methods- We used a multicenter
stroke registry database (Clinical Research Collaboration for
Stroke in Korea) to analyze
acute ischemic stroke patients due to large artery atherosclerotic occlusive disease who were treated with
aspirin alone or combination of
clopidogrel and
aspirin from May 2008 to May 2015. The results were analyzed by intention-to-treat, per-protocol, and as-treated methodologies. The primary end point was the 1-year composite outcome of
stroke recurrence,
myocardial infarction, and all-cause death. To balance the differences between groups, a
frailty model using propensity scores and inverse probability of treatment weighting was used. Results- A total of 5934 patients with symptomatic large artery atherosclerotic occlusive disease were treated either with
clopidogrel plus
aspirin (n=2903, 49%) or
aspirin (n=3031, 51%). The frequency of the primary outcome was 12% (n=353) in the
clopidogrel-
aspirin group and 14% (n=410) in the
aspirin group. The hazards of the primary outcome with combination over
aspirin only were significantly reduced in the per-protocol and as-treated analyses (hazard ratio, 0.71; 95% CI, 0.57-0.88; P=0.002 and hazard ratio, 0.81; 95% CI, 0.69-0.96; P=0.02, respectively), but there was borderline significance in the intention-to-treat analysis (hazard ratio, 0.86; 95% CI, 0.74-1.01; P=0.06). Combination
therapy was beneficial for all-cause death in all analyses but did not reduce recurrent
stroke. Conclusions- Compared with patients receiving
aspirin monotherapy, the primary outcome seemed to occur less frequently in patients receiving dual antiplatelet
therapy, which is explained mainly by the decrease of all-cause death. Since this is a nonrandomized, retrospective, observational study, our study should be cautiously interpreted.