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


Nadolol plus isosorbide was better than sclerotherapy for preventing variceal rebleeding

ACP J Club. 1996 Nov-Dec;125:65. doi:10.7326/ACPJC-1996-125-3-065

Source Citation

Villanueva C, Balanzó J, Novella MT, et al. Nadolol plus isosorbide mononitrate compared with sclerotherapy for the prevention of variceal rebleeding. N Engl J Med. 1996 Jun 20;334:1624-9.



To compare the safety and efficacy of nadolol plus isosorbide mononitrate with those of endoscopic sclerotherapy (ES) for the prevention of variceal rebleeding.


Randomized controlled trial with mean 18.5-month follow-up.


Hospital in Spain.


86 patients (mean age 59 y, 67% men) who were hospitalized for upper gastrointestinal bleeding. Inclusion criteria were cirrhosis and endoscopic evidence of hemorrhage from esophageal varices. Emergency endoscopy was done within the first 4 hours after admission. Exclusion criteria were age < 18 years, Child-Pugh score > 12 points, advanced hepatocellular carcinoma, lung cancer, previous ES, or failure of medication to stop index bleeding. Follow-up was 98% at 18 months.


Patients were stratified by severity of liver failure and history of variceal bleeding and were allocated to nadolol plus isosorbide mononitrate (n = 43) or ES (n = 43). Nadolol was given orally, 80 mg once a day; the dose was adjusted for 5 days until the resting heart rate was 55 beats/min. Oral isosorbide mononitrate was then started and progressively increased for 1 week to 40 mg twice a day. ES was done by injecting 5% ethanolamine into esophageal varices up to a total dose of 10 to 20 mL. ES was done on days 0, 4, 10, and 30 and then monthly until varices were eradicated. Compliance with medication was assessed by patient interview.

Main outcome measures

Rebleeding, complications, and death.

Main results

Nadolol plus isosorbide mononitrate led to fewer rebleedings at a mean of 18 months than did sclerotherapy (26% vs 53%, P = 0.001). {This absolute risk reduction (ARR) of 27% means that 4 patients would need to be treated (NNT) with nadolol plus isosorbide (rather than ES) for 18 months to prevent 1 rebleeding, 95% CI 2 to 14; the relative risk reduction (RRR) was 52%, CI 17% to 74%.}* Treatment with nadolol plus isosorbide also led to fewer treatment-related complications than did ES (16% vs 37%, P = 0.03) {ARR 21%; NNT 5, CI 3 to 44; RRR 56%, CI 8% to 80%}*. The groups did not differ for death: 4 patients who received nadolol plus isosorbide died compared with 9 who received ES (9% vs 21%, P = 0.07) {CI for the 12% ARR -4% to 27%}*.


Compared with endoscopic sclerotherapy, nadolol plus isosorbide mononitrate decreased rebleeding and was associated with fewer complications.

Source of funding: In part, Fundació Investigaciö Sant Pau.

For article reprint: Dr. J. Balanzó, Servei de Patologia Digestiva, Hospital de la Santa Creu i Sant Pau, Sant Antoni M. Claret, 167, 08025 Barcelona, Spain. FAX 34-3-221-3237.

*Numbers calculated from data in article.


Varices usually bleed when they are large, when portal pressure is > 12 mm Hg, and during decompensation of liver disease. A reduction in portal pressure to < 12 mm Hg, achieved by β-blocking drugs, prevents bleeding in only about one third of patients and causes disabling side effects in those with advanced liver disease. ES reduces rebleeding by obliterating submucosal varices in the esophagus and proximal stomach without affecting portal pressure, and rebleeding rates remain low even when portal pressure increases after obliteration (1).

The addition of isosorbide mononitrate to β-blocking drugs leads to a mean reduction of 25% in portal pressure, which is the basis for combination therapy in this study. In each patient, the optimal therapeutic dose was reached within 17 days. Medication compliance was 98%, which is a phenomenal feat considering that > 50% of patients had alcohol-related liver disease. The results of ES are less impressive; 53% of patients with relatively good liver function (84% Child-Pugh class A and B) had rebleeding, and 37% had complications. Could treatments given more often than at weekly intervals cause early complications and rebleeding (2)? How soon were varices obliterated?

Comparison of the efficacy of drug therapy (not dependent on technique) with that of endoscopic therapy (highly dependent on technique and expertise) is difficult. Compliance is important for both treatments but is crucial for the success of drug therapy. When varices remain obliterated with ES, surveillance is needed only at yearly intervals (3); this reduces dependence on compliance.

The results of this study indicate that combined drug therapy can be used for the tolerant, compliant, well-compensated patient who is stable after bleeding from varices. Endoscopic therapy (ES or preferably ligation) may be a better choice for physicians with technical skill and for the less compliant patient with alcoholism.

Jacob Korula, MD
University of Southern California School of MedicineLos Angeles, California, USA


1. Korula J, Ralls P. Gastroenterology. 1991; 101:800-5.

2. Akriviadis E, Korula J, Gupta S, Ko Y, Yamada S. Dig Dis Sci. 1989;34:1068-74.

3. Waked I, Korula J. Hepatology. 1994; 20:104A.