Current issues of ACP Journal Club are published in Annals of Internal Medicine


Endoscopic ligation was better than propranolol for prevention of bleeding from esophageal varices


ACP J Club. 2000 Jan-Feb;132:21. doi:10.7326/ACPJC-2000-132-1-021

Source Citation

Sarin SK, Lamba GS, Kumar M, Misra A, Murthy NS. Comparison of endoscopic ligation and propranolol for the primary prevention of variceal bleeding. N Engl J Med. 1999;340:988-93. [PubMed ID: 10099140]



Is endoscopic ligation as efficacious and safe as propranolol therapy for primary prevention of bleeding from esophageal varices?


Randomized, unblinded,* controlled trial with 18-month follow-up.




90 patients (mean age 42 y, 73% men) with portal hypertension who had never had bleeding from varices and had large (grade 3 or 4) varices but no history of hematemesis or melena. Patients were excluded if they were receiving antiviral therapy or if they had hepatoma or another tumor, severe cardiopulmonary or renal disease, heart failure, peripheral vascular disease, bradycardia, bronchial asthma, diabetes mellitus, prostatic hypertrophy, glaucoma, or a psychiatric disorder. Follow-up was 99%.


46 patients were allocated to ligation, which was done every week until the varices were obliterated or reduced to a size of grade 1. If varices recurred and became grade 2 or larger, ligation was repeated. 44 patients were allocated to oral propranolol, 40 mg/d, with dose increments of 20 to 40 mg/d until a 25% decrease in baseline heart rate was achieved. Propranolol therapy was stopped if the patient had systolic blood pressure < 80 mm Hg, heart rate < 55 beats/min, or other serious side effects.

Main outcome measures

Outcomes included variceal bleeding, need for blood transfusion, overall deaths, and deaths from variceal bleeding.

Main results

Analysis was by intention to treat. At 18 months, the cumulative probability of variceal bleeding was lower in the ligation group than in the propranolol group (15% vs 43%, P = 0.04). The hazard ratio for variceal bleeding in the propranolol group was 3.0 (95% CI 1.3 to 9.3); the age-adjusted hazard ratio was 2.6 (CI 1.0 to 8.2). Fewer patients in the ligation group required blood transfusions (1 vs 7, P = 0.03) than in the propranolol group. The groups did not differ for overall deaths (11% in each group, P = 0.77) or bleeding-related deaths (7% vs 9%, {P = 0.67}†).


Among patients with high-risk esophageal varices, endoscopic ligation was more efficacious than propranolol for primary prevention of bleeding.

*See Glossary.

P value calculated from data in article.

Source of funding: Not stated.

For correspondence: Dr. S.K. Sarin, Department of Gastroenterology, G.B. Pant Hospital, New Delhi 110 002, India. FAX 91-11-6426896.

Revised Commentary 2006

This randomized trial by Sarin and colleagues showed that endoscopic sclerotherapy was superior to propranolol in the prevention of a first variceal hemorrhage in patients with portal hypertension who were at high risk for bleeding. Despite an impressive reduction in the hazard ratio (0.33), both groups had similar mortality rates.

A later meta-analysis (1) analyzing this and 3 other studies comparing variceal band ligation with β-blockade came essentially to the same conclusion as the study by Sarin and colleagues: Variceal ligation reduces bleeding more efficiently than b-blockers (relative risk 0.48, 95% CI 0.24 to 0.96; number needed to treat 13). However, as in the study by Sarin and colleagues, neither bleeding-related mortality (RR 0.61, CI 0.20 to 1.88) or overall mortality (RR 0.95, CI 0.56 to1.62) were affected by ligation treatment as compared with pharmacologic therapy.

Given the cost and invasiveness of ligation treatment, pharmacotherapy remains the treatment of first choice in the prevention of variceal bleeding. However, for the many patients who cannot tolerate β-blockers are not tolerated, it is reasonable to use prophylactic ligation treatment as treatment of first choice. Larger studies (only n = 283 were analyzed in the meta-analysis by Imperiale and Chasalani) should address the question of which treatment is more effective and whether one or the other has a favorable effect on bleeding-related and overall mortality.

Juerg Reichen, MD
University of Berne
Berne, Switzerland


1. Imperiale TF, Chalasani N. A meta-analysis of endoscopic variceal ligation for primary prophylaxis of esophageal variceal bleeding. Hepatology. 2001;33:802-7. [PubMed ID: 11283842]