1. The continuing doubt concerning the value of vasoconstrictive therapy for gastrointestinal bleeding may be related to the complexity of clinical trials in such a situation.
2. The effect of SM 201–995, a somatostatin analogue, was investigated in conscious cirrhotic rats before, during and after experimental bleeding from the portal territory.
3. Before haemorrhage, somatostatin analogue (8 μg h−1 kg−1 body weight, intravenously) produced a significant decrease in portal pressure (17%), whereas a placebo (saline) lacked significant effect.
4. The rats were then subjected to spontaneous bleeding, by disconnection of the portal catheter from the pressure gauge, for a 5 min period. At the end of the haemorrhage, haemodynamic parameters did not significantly differ between the group receiving somatostatin analogue and that receiving placebo.
5. Fifteen minutes after the end of the bleeding period, portal pressure was significantly lower in rats receiving somatostatin analogue [8.5 ± 0.5 mmHg (1.13 ± 0.07 kPa), mean ± sem] than in rats receiving placebo [11.0 ± 1.1 mmHg (1.50 ± 0.15 kPa)].
6. The volume of blood lost and mortality were significantly lower in the group treated with somatostatin analogue (2.3 ± 0.1 ml/100 g body weight and 8%, respectively) than in the group receiving placebo (3.0 ± 0.1 ml/100 g body weight and 50%, respectively).
7. These results demonstrate, in an experimental model, the beneficial effect of somatostatin analogue for the treatment of gastrointestinal bleeding due to portal hypertension. They suggest that administration of this substance should be started as soon as possible after the beginning of haemorrhage and continued after the cessation of bleeding.