Immediate endoscopic sclerosis of bleeding esophageal varices. A prospective evaluation over five years.

From January 1982 to January 1987, a prospective, nonrandomized sclerotherapy study was conducted at the Heinz-Kalk Hospital, Bad Kissingen, with 232 consecutive patients receiving immediate endoscopic injection sclerosis (IEIS) for variceal bleeding during emergency endoscopy. No patient was excluded from this study. All patients were classified according to the Child-Pugh criteria: 53 (23%) were in category A; 70 (30%) in category B; and 109 (47%) in category C. More than 93% had liver cirrhosis, 60% of alcoholic origin. IEIS was performed during emergency fiberoptic endoscopy applying 5-40 1-ml injections of 0.5% polidocanol by the paravariceal, "free" injection technique. If this was not successful, an intravariceal approach was used, and after 15 min of injection and persisting hemorrhage a Linton-Nachlas tube was inserted for 6-12 h. If hemorrhaging recurred, a second emergency endoscopy with IEIS was performed and, if this was not successful, a gastroesophageal disconnection followed directly. During the bleeding-free interval, Child-Pugh A and B patients were selected, using special criteria, for a shunt operation. An elective, semiselective, or selective shunt was recommended and carried out. In 56 patients (24%), a gastroesophageal disconnection or a shunt operation was performed on an emergency or elective basis. All sclerotherapy patients were controlled after 4 months and thereafter every 6, 9, or 12 months and reinjected if necessary. Bleeding was controlled in 93% of the cases with IEIS, and, in 97%, with a combination of IEIS and Linton-Nachlas tube. Definitive control of hemorrhaging was accomplished in 94% of the patients. Thirty-five patients died during the first 30 days of admission (15.1%).(ABSTRACT TRUNCATED AT 250 WORDS)