In the case of oral administration of cilostazol in in vivo experiments, antiplatelet activities of its active metabolites, OPC-13015 and OPC-13213, could also contribute to inhibition of platelet thrombus formation. OPC-13015 has 3 times more potent antiplatelet activity than cilostazol, whereas OPC-13213 has 3 times less potent activity than cilostazol. Cmax of cilostazol, OPC-13015 and OPC-13213 were 2.4, 1.4 and 9.1 mM, respectively respectively, after a single oral administration of cilostazol at a dose of 300 mg/kg in non-fasting male SD rats. On the other hand, the results of the healthy male single dose study showed that in human after administration of 100 mg of cilostazol, Cmax of cilostazol, OPC-13015 and OPC-13213 were about 625, 122 and 64 mg/L, respectively, and AUC0�C72 h were about 8087, 2423 and 617 mg/L?h, respectively. On the basis of these results, we concluded that the dosage of cilostazol used in the rat photothrombotic stroke model in our study could give sufficient plasma concentrations of cilostazol and its metabolites compared with their therapeutic plasma concentrations in human. In contrast to the evidence from clinical trials and guidelines supporting antiplatelet therapies for secondary stroke prevention, the benefit of antiplatelet agents for primary stroke HEAT hydrochloride prevention has not been satisfactorily proven in patients with diabetes. Moreover, anticoagulation therapy with warfarin is substantially more efficacious than antiplatelet therapy with aspirin for primary stroke prevention in patients who have nonvalvular atrial fibrillation. For these reasons, we assume that K-134 is expected to be developed for secondary stroke prevention than for primary prevention. In summary, K-134 significantly prevented brain damage by inhibiting thrombus formation in the rat cerebral infarction model. This effect was attributable to potent antiplatelet activity of K-134. Recurrence of ischemic stroke is involved in platelet activation. Therefore, K-134 is a promising drug for secondary prevention of ischemic stroke due to its potent inhibitory activity on platelet thrombus formation. The rationale of a delayed-start design is that under the null hypothesis, when the active drug has a purely symptomatic effect and has no effect on neuropathologic process, a delay in administration should have no lasting effect on patients. Thus the delayed-start patients are expected to catch up to early-start patients. However, if the effect were not purely symptomatic and had altered the underlying progression, the delayed-start patients could not possibly overcome the losses sustained during the delay period. In other words, if the delayed-start patients do not ��catch up�� with the early-start patients, a conclusion can be drawn that the treatment is not purely symptomatic and has modified the underlying course of the GNTI dihydrochloride disease, representing a disease-modifying effect.