Gastric antral vascular ectasia (GAVE) is a rare cause of chronic bleeding in cirrhotic patients. It has been suggested that these gastric lesions might be related to portal hypertension, hepatic insufficiency, or both parameters. We report two cases of cirrhotic patients in whom GAVE was the source of recurrent bleeding. These patients also had complete portal vein thrombosis. Liver transplantation was performed and an end-to-end cavoportal anastomosis was performed, leaving patients with persistent portal hypertension after surgery. We observed complete disappearance of the antral lesions several weeks after transplantation, which shows that the GAVE is not related to portal hypertension but is rather a direct consequence of liver failure. Possible pathophysiologic mechanisms are discussed. (Liver Transpl 2002;8:717-720.) G astric antral vascular ectasia (GAVE) is now a well-recognized complication of cirrhosis and its estimated prevalence is around 3%. 1 The endoscopic distinction between GAVE and portal hypertensive gastropathy (PHG), the other gastric lesion commonly found in cirrhosis, is relatively well defined and in most cases can be performed easily by experienced endoscopists.GAVE lesions are described as longitudinal rows of ectatic, sacculated, mucosal vessels (red spots) mainly located in the antrum. These red spots may be aggregated in linear stripes (watermelon stomach) but may be confluent (honeycomb pattern). 2 By contrast, PHG is observed mostly in the fundus and body of the stomach and is characterized by superficial erythema of an edematous mucosa with cherry-red spots arranged in a mosaic pattern, with or without diffuse bleeding. Both lesions also can be differentiated by pathologic criteria, with a combination of fibrohyalinosis, spindle cell proliferation, vascular ectasia, and thrombi, being highly suggestive of GAVE. 3,4 On the other hand, the two gastric mucosal lesions can occur simultaneously in the same patient.The role of portal hypertension in the pathophysiology of PHG is well shown. However, the relationship between GAVE and both portal hypertension and liver insufficiency is still unclear. In the present report, we describe 2 cases of cirrhotic patients with GAVE and portal vein thrombosis. Persistence of the portal hypertension after transplantation provides an unique opportunity to delineate the respective roles of liver failure and portal hypertension in the pathophysiology of these lesions.
Case 1A 60-year-old man with alcoholic cirrhosis (abstinent since 1984) was repeatedly admitted since 1998 for upper gastrointestinal bleeding secondary to typical GAVE lesions, as shown by successive upper gastrointestinal endoscopies. He required serial blood transfusions in the ensuing 2 years to maintain stable hemoglobin levels. Small, nonbleeding esophageal varices also were noted. In addition, he presented a deep calf thrombophlebitis secondary to hypercoagulation associated with a methylentetrahydrofolate reductase (MTHFR) prothrombin mutation and was anticoagulated. He was refe...