Comparative effectiveness of cilostazol and aspirin in ischemic stroke: a retrospective cohort study.

OBJECTIVE

To compare clinical characteristics and outcomes associated with cilostazol versus aspirin in patients with ischemic stroke, with attention to carotid plaque progression, lipid metabolism, platelet aggregation, adverse events, and 12-month stroke recurrence.

METHODS

This retrospective study included 250 patients with ischemic stroke treated between August 2019 and April 2023, of whom 100 received cilostazol and 150 received aspirin according to routine clinical practice. Cilostazol was more commonly prescribed to patients with aspirin intolerance or perceived bleeding risk. Outcomes included changes in carotid plaque area and stenosis, lipid parameters, platelet function, adverse reactions, and recurrent ischemic stroke within 12 months.

RESULTS

At 12-month follow-up, patients receiving cilostazol showed greater reductions in carotid plaque area and stenosis rate compared with those receiving aspirin (P < 0.05). Differences were also observed in lipid profiles and platelet aggregation assays, with patterns consistent with the known pharmacological actions of cilostazol (P < 0.05). Adverse reactions, including bleeding events, pruritus, myalgia, and liver enzyme elevations, occurred less frequently in the cilostazol group (P < 0.05). Recurrent ischemic stroke was documented in 6.0% of cilostazol-treated patients and 20.0% of aspirin-treated patients (P < 0.05). These findings reflect associations observed in this real-world cohort and should be interpreted in light of underlying baseline differences and the non-randomized study design.

CONCLUSION

In this retrospective cohort, cilostazol use was associated with more favorable safety outcomes and lower recurrence rates compared with aspirin. Because treatment allocation was clinically driven and residual confounding cannot be excluded, the results indicate associations rather than causal effects. Prospective randomized studies are needed to determine whether these observed differences represent true treatment benefits.