Endoscopic diagnosis of gastric varices.

In this retrospective study, the presence and appearance of gastric varices were analyzed among patients presenting at Kasr El Aini Hospital with variceal bleeding during the period from 1984 to 1989. Two groups of patients were studied. The first group included 970 patients with documented variceal bleeding with no history of sclerotherapy. Of these patients 6.7%, had concomitant gastric varices, and 27.1% had direct gastric extension of esophageal varices (mostly of grade 3). Gastric varices alone, without esophageal varices, were found in five cases (0.5%). The second group was a subgroup, and included 376 patients who underwent complete sclerotherapeutic eradication of varices restricted to the esophagus and were without gastric varices at the first presentation. Eleven of these patients (2.9%) developed secondary gastric varices. There was no correlation between the risk of bleeding from primary gastric varices (seen at the first presentation) and the grade of the esophageal varices. There were more bleeding episodes from large gastric varices (35.4% for the cauliflower and 16.9% for the cystic forms) than from the small varices, i.e. ruga-like and network-like gastric varices. The study also shows that the secondary development of gastric varices after endoscopic variceal sclerotherapy is a rare event, and that there was no risk of bleeding from the secondary gastric varices. Patients with esophageal varices with gastric extension had a small risk of bleeding from their gastric varices (5.2%).