Background and purpose: Cerebral vasospasm is a major cause of morbidity and mortality in patients with subarachnoid hemorrhage (SAH). Cilostazol, a selective inhibitor of phosphodiesterase 3, is a peripheral vasodilator, an anti-inflammatory, and causes antiplatelet aggregation. We investigated these effects on cerebral vasospasm after rat SAH. Methods: Thirty-eight Sprague-Dawley rats were randomly divided into three groups: SAH + normal feed (SAH group; n=14), SAH + feed containing 0.1% cilostazol (cilostazol group; n=12) and sham-operated rats (sham group; n=12). The basilar arteries (BA) of all groups were analysed by measuring wall thickness, internal luminal perimeter and cross-sectional area on day 7. Immunohistochemical study with RM-4, an anti-rat macrophage/dendritic cells monoclonal antibody and ultrastructural study with transmission electron microscopy were performed. Results: Although most animals in the SAH group presented with typical vasospasm, the means of inner perimeter and cross-section area of the BA in the cilostazol group were significantly greater than the SAH group (836 +/- 134 mu m versus 771 +/- 125 mu m and 39 177 +/- 15 405 mu m(2) versus 33 098 +/- 13 871 mu m(2), respectively). Wall thickness of the BA in the cilostazol group demonstrated significant decrease, compared with the SAH group (17.4 +/- 2.3 versus 21.0 +/- 2.7 mu m). In immunohistological study, SAH induced an obvious increase in mean perivascular RM-4-positive cell count, whereas cilostazol significantly reduced it by 59%. Ultrastructural study depicted cilostazol markedly attenuating structural deterioration of the vascular wall due to SAH. Conclusions: This work demonstrates that cilostazol attenuates cerebral vasospasm after SAH in rat, possibly in part due to the anti-inflammatory effect.