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


Prophylactic sclerotherapy increased mortality and did not prevent variceal esophageal hemorrhage in men with alcoholic liver disease

ACP J Club. 1991 Nov-Dec;115:73. doi:10.7326/ACPJC-1991-115-3-073

Source Citation

The Veterans Affairs Cooperative Variceal Sclerotherapy Group. Prophylactic sclerotherapy for esophageal varices in men with alcoholic liver disease. A randomized, single-blind, multicenter clinical trial. N Engl J Med. 1991 Jun 20;324:1779-84.



To determine if sclerotherapy prevents variceal esophageal bleeding and improves survival in men with alcoholic liver disease.


Randomized, single-blind, placebo-controlled trial.


12 Veterans Affairs medical centers; screening and enrollment were from February 1985 to October 1986.


2200 patients with alcoholic liver disease were screened. 433 met the inclusion criteria of male sex, consumption of > 48 g/d of alcohol for > 1 year, alcoholic liver disease, no upper gastrointestinal bleeding in the previous 4 years from esophageal varices, and ≥ 3 esophageal variceal channels proven by endoscopy. 65 patients were excluded, and 87 refused to participate. Exclusion criteria included contraindications to upper endoscopy, a positive test for hepatitis B, or previous sclerotherapy. The 2 study groups were similar for age, proportion of patients with > 15-year history of alcohol consumption, number and distribution by size of esophageal varices, and Child score.


Patients were assigned to receive endoscopy and either sclerotherapy (n = 143) or sham therapy (n = 138) at the time of randomization, after 4 to 6 days, 9 to 11 days, 1month, 3 months, then every 3 months to 2 years. All treatment was discontinued 22.5 months after the start of the study because of a higher death rate in the treatment group.

Main outcome measures

Death and upper gastrointestinal bleeding. Bleeding episodes were defined by hematemesis, bloody nasogastric aspirate, bleeding observed at endoscopy, or melena; hospitalization for bleeding; and the need for blood transfusions.

Main results

More deaths occurred during treatment among the sclerotherapy group than among the sham therapy group (32% vs 17%, P = 0.004 ). {This absolute risk difference of 15% means that 1 additional death occurred for every 7 patients who received sclerotherapy (rather than sham therapy), 95% CI 4 to 21; the relative risk increase was 85%, CI 21% to 186%.}* No difference occurred in the number of deaths when therapy was stopped (35% in the sclerotherapy group vs 37% in the sham therapy group) {absolute risk reduction 2%, CI -11% to 15%, P = 0.78 }.* Causes of death both during and after treatment were similar for the 2 groups. No association existed between the time sclerotherapy was done and time of death. The number of patients with upper gastrointestinal bleeding during treatment was similar in the 2 study groups (22% in the sclerotherapy group vs {17%}* in the sham therapy group, P = 0.30).


Prophylactic sclerotherapy in men with alcoholic liver disease and esophageal varices resulted in increased mortality and no improvement in controlling upper gastrointestinal bleeding.

Sources of funding: Department of Veterans Affairs and Elkins Sinn, Inc.

Address for article reprint: Dr. P.B. Gregory, Office of the Vice President and Dean, School of Medicine, Room M-121, Stanford University Medical Center, Stanford, CA 94305-5302, USA.

*Numbers calculated from data in article.


The management of esophageal varices can be broadly classified as follows: emergent, treatment of actively bleeding esophageal varices ; therapeutic, treatment of patients with a known previous episode of variceal bleeding, with the aim of preventing rebleeding; and prophylactic, management of patients with esophageal varices who have never bled from varices, with the aim of preventing bleeding.

The value of injection sclerotherapy as an emergent treatment is relatively firmly established (1, 2). Use of sclerotherapy for the prevention of variceal rebleeding is more controversial, but generally accepted as effective (2). The situation is different, however, for prophylactic sclerotherapy. Because sclerotherapy does nothing to alter the underlying pathophysiology of portal hypertension, it has been predicted that a long-term effect of sclerotherapy will be bleeding from nonesophageal varices (3). Further, it should be noted that only 33% of patients with cirrhosis actually bleed from esophageal varices. Thus, it is not surprising that this large study showed that prophylactic sclerotherapy increases mortality.

During the treatment period, mortality for the sclerotherapy group was higher (32%) than for the sham therapy group (17%). During the follow-up period, the mortality rate in the sclerotherapy group was 35%; in the sham group it was 37%, a marked increase. Thus, there appears to be an increase in the mortality rate in the sham group during the follow-up period, not a decrease in mortality in the sclerotherapy group as stated by the authors ("After the termination of treatment, the excess mortality rate in the sclerotherapy group promptly declined."). Unfortunately, the authors did not discuss this finding of a rise in mortality in the sham group. Readers should compare the results of this study with the results of other studies (4-7) that have been completed since this study was published.

Nirmal S. Mann, MD, DSc
Texas A & M UniversityTemple, Texas, USA.


1. Westaby D, Hayes PC, Gimson AE, et al. Controlled clinical trial of injection sclerotherapy for active variceal bleeding. Hepatology. 1989;9:274-7.

2. Terblanche J, Krige JE, Bornman PC. Endoscopic sclerotherapy. Surg Clin North Am. 1990;70:341-59.

3. Mann NS. Injection sclerosis of esophageal varices is without sound physiologic basis. Gastrointest Endosc. 1982;28:116-7.

4. The PROVA Study Group. Prophylaxis of first hemorrhage from esophageal varices by sclerotherapy, propranolol or both in cirrhotic patients: a randomized multicenter trial. Hepatology. 1991;14:1016-24.

5. Vinel JP, Lamouliatte H, Cales P, et al. Propranolol reduces the rebleeding rate during endoscopic sclerotherapy before variceal obliteration. Gastroenterology. 1992;102:1760-3.

6. Van Ruiswyk J, Byrd JC. Efficacy of prophylactic sclerotherapy for prevention of a first variceal hemorrhage. Gastroenterology. 1992;102:587-97.

7. Triger DR, Johnson AG, Brazier JE, et al. A prospective trial of endoscopic sclerotherapy V oesophageal transection and gastric devascularisation in the long ter management of bleeding oesophageal varices. GUT. 1992;33:1553-8.