Background: We examined the baseline characteristics and the rates of serious vascular events in randomized clinical trials after cerebral ischemia of presumed arterial origin to identify study features that predict vascular event incidence and to identify whether event rates have declined over time. Methods: We performed a systematic review to identify all published randomized controlled trials of patients with stroke or transient ischemic attack of presumed arterial origin that included an aspirin arm. We performed metaregression to analyze whether baseline features of included patients, design features and the year of study start and publication influenced vascular event rates under aspirin treatment. Results: Included studies comprised 23,247 patients who had 3,615 serious vascular events. Baseline characteristics of patients included in the different trials varied substantially over time. In multivariate metaregression analysis a shorter maximum allowed delay from symptom onset to randomization (p = 0.012) and a double-blind design were associated with higher annual event rates (p = 0.014). Event rates did not significantly decrease over time (p = 0.242). Conclusion: The baseline characteristics of patients included in trials that study antiplatelet agents have changed over time. The strongest predictors for higher event rates are a double-blind design and a short delay between symptom onset and randomization. An important reduction in event rates over time could not be demonstrated.