Background/aims: Recent improvements in the treatment of gastric varices have resulted in better rates of hemostasis and eradication. This prospective study was conducted in order to examine the natural history of gastric varices and whether prophylactic treatment is required.
Methodology: Between January 1991 and December 1996, 52 patients with cirrhosis and no history of prior variceal treatment were enrolled.
Results: Bleeding from gastric varices occurred in 4 patients. Hemorrhage was stopped successfully in all of these patients. The cumulative bleeding rates at 1, 3, and 5 years were 3.8%, 9.4%, and 9.4%, respectively. Erosive gastritis or gastric ulcer was not detected in 3 of these 4 patients at the time of entry into the study, although an ulcer or erosion was found at the point of gastric variceal hemorrhage in all four patients at the time of variceal rupture. There were no significant differences in patient characteristics with ruptured versus non-ruptured varices at the time of their study entry.
Conclusions: Factors which could be used to predict bleeding from gastric varices were not identified. Furthermore, the overall incidence of hemorrhage from gastric varices was found to be low. Once gastric variceal hemorrhage did occur, bleeding from these varices was successfully stopped in all cases. Therefore, prophylactic treatment of gastric varices is not recommended.