THE SKEPTICISM, originally entertained in some quarters6 regarding the efficacy of surgical portacaval shunt decompression for the complication of bleeding from esophageal varices in the patient with Laennec's cirrhosis, is disappearing rapidly. That massive hemorrhage in the patient with cirrhosis is a surgical problem has become an acceptable proposition among internists and surgeons alike.11 Portalsystemic shunting in these patients with bleeding varices significantly improves the prospect for overall five-year survival in this group.3 There has been concern with the timing of operation for the patient with cirrhosis who is hospitalized with massive hemorrhage from varices. One cannot help but be troubled by the extremely poor prognosis implicit in the episode of massive esophageal hemorrhage in those patients who have been managed by conservative measures alone. It has been estimated that the mortality in this group is well above 60% and it has been reliably reported to be