RS-5186, which inhibits thromboxane A2 (TXA2) synthetase activity, ameliorated delayed cerebral vasospasm in a canine two-hemorrhage model. Subarachnoid hemorrhage was induced in 15 dogs, which were divided into two groups. In the RS-5186-treated group (9 dogs), 50 mg kgof RS-5186 was administered twice a day for seven days. The remaining six dogs without administration of RS-5186 were used as a control group. In the RS-5186-treated group, the angiographic diameter of the basilar artery on Day 7 after subarachnoid hemorrhage was constricted to 60.9% ± 11.6% fn = 9, mean±SD) of that on Day 0, before subarachnoid hemorrhage. The corresponding value was 42.8%±6.1% (n = 6) in the control group. There was a statistically significant difference between these percentages. In the RS-5186-treated group, plasma thromboxane B2 level on Day 7 was 144.3 ± 28.1 pg mh1 fn = 4), which was lower than the 815.5± 162.0 pg ml–1 fn = 4) in the control group (p < 0.0005). The plasma 6-keto-prostaglandin F1a level on Day 7 was 180.5 ± 66.5 pg ml*1 fn = 4J in the RS-5186-treated group, and higher than 107.3± 12.4 pg mh1 fn = 4) in the control group fp = 0.0734). Thus, administration of RS-5186 reduced TXA2 plasma level and had a beneficial effect on angiographically-detected delayed vasospasm. [Neurol Res 1999; 21: 513–516]