A selective cyclo-oxygenase-2 inhibitor, NS-398, may improve portal hypertension without inducing gastric mucosal injury

BACKGROUND AND AIMS: Prostacyclin has been shown to play a role in hyperdynamic circulation in portal hypertension. Recently, a new subtype of cyclo-oxygenase (COX), COX-2, which acts as an inducible synthase in response to various stimuli. The aim of this study was to investigate whether COX-2 contributes to portal hypertension and whether a COX-2 blockade induces the same sort of gastric mucosal injury as a COX-1 blockade. METHODS: Portal hypertension (PHT) in rats was induced by a two-step ligation of the portal vein. The mean arterial pressure (MAP), portal pressure (PP), visceral blood flow volume (BFV), serum levels of 6-keto-prostaglandin F1alpha (PGF1alpha), thromboxane B2 (TXB2) and gastric mucosal injury induced by pure ethanol were all measured in PHT rats receiving different inhibitors (indomethacin, a highly selective COX-1 inhibitor; NS-398, a highly selective COX-2 inhibitor). Control rats treated by a sham operation were also studied. RESULTS: The NS-398 administration significantly decreased PP to the same extent as indomethacin at doses of 5 and 10 mg/kg in PHT rats after a 60 min administration, while neither inhibitor affected the control rats. Both inhibitors significantly increased PP after a 30 min administration in the PHT and control rats at a dose of 5 mg/kg while both inhibitors significantly decreased PP after 60 min administration only in the PHT rats. Portal vein ligation treatment induced a significant increase in PP and BFV of the portal vein, gastric mucosa, oesophageal mucosa and the serum levels of 6-keto-PGF1alpha and TXB2, while portal vein ligation treatment induced a significant decrease in BFV of the liver. Both blockades increased MAP and decreased PP and BFV in the splanchnic area and decreased the serum level of 6-keto-PGF1alpha and TXB2 in the PHT rats, while neither blockade modified any parameters in the control rats, except that indomethacin administration significantly decreased the BFV of the gastric mucosa. Indomethacin administration significantly increased the ulcer index (UI). The NS-398 had no effect on UI in either the PHT or control rats. Only indomethacin significantly increased the number of rats demonstrating gastric mucosal long lesions (> 2 cm) in the PHT rats. CONCLUSION: In the PHT rats, prostaglandin seemed to contribute to portal hypertension. Both COX blockades reduced PP and BFV of the portal vein and gastric mucosa. NS-398, a selective COX-2 inhibitor, may, therefore, improve portal hypertension without inducing gastric mucosal injury.