Prospective multicenter randomized trial comparing banding ligation with sclerotherapy of esophageal varices

BACKGROUND/AIMS: Endoscopic variceal banding ligation (EVL), first describe d by Stiegmann in 1988, is now an accepted alternative to sclerotherapy (EV S) for esophageal varices with previous bleeding. However, results are conf licting in terms of complications, eradication, recurrence, rebleeding and mortality rate. We aimed to compare EVL with EVS in a prospective randomize d trial in patients with previous esophageal bleeding proved by endoscopy. End points were rebleeding rate and death during a short (eradication perio d) or long-term follow-up (>1 year). METHODOLOGY: One hundred patients (50 EVL, 50 EVS) were enrolled. Eradicati on rate, number of treatments needed to achieve eradication, recurrence of varices, rebleeding and complications were recorded and analyzed. RESULTS: No differences were observed between the two groups regarding age, sex and Child class. One patient dropped out in the EVL group and 6 in the EVS group. Eradication was obtained in 44 (88%) with EVL and 41 (82%) with EVS with a mean of 3.41 and 5.29 treatments (p<0.001), respectively. Reble eding occurred during eradication in 6 patients (12%) in the EVL group and 21 (42%) in the EVS group (p=0.001); after eradication, 7 patients (14%) re bled in the EVL group and 4 (8%) in the EVS group (not significant). Non-va riceal bleeding was observed in 5 patients (2 EVL and 3 EVS) during follow- up. Two patients in the EVL group died after variceal rebleeding; 3 died of gastric bleeding; and, 15 from non-hemorrhagic events (8 EVL and 7 EVS). I n the EVL group 14 patients had recurrent varices and 7 rebled; in the EVS group 11 recurred, with rebleeding in 5. Major complications were fewer in the EVL group (1 stenosis, 4 chronic ulcers) compared to 18 patients in the EVS group (9 stenosis and 9 chronic ulcers) (p<0.005). CONCLUSIONS: EVL might be preferable to EVS for faster reduction and oblite ration of varices, with a lower rate of complications and rebleeding before eradication. No differences were observed in recurrence.