The Effectiveness of Current Acute Variceal Bleed Treatments in Unselected Cirrhotic Patients

Refining Short-Term Prognosis and Risk Factors

Lucio Amitrano MD; Maria Anna Guardascione MD; Francesco Manguso MD; Raffaele Bennato MD; Antonio Bove MD; Claudio DeNucci MD; Giovanni Lombardi MD; Rossana Martino MD; Antonella Menchise MD; Luigi Orsini MD; Salvatore Picascia MD; Elisabetta Riccio MD


Am J Gastroenterol. 2012;107(12):1872-1878. 

In This Article

Abstract and Introduction


Objectives: The mortality from esophageal variceal hemorrhage in liver cirrhosis patients remains approximately 15–20%. Predictors of short-term outcomes, such as the hepatic venous pressure gradient, are often unavailable in the acute setting. Clinical variables seem to have a similar predictive performance, but some variables including active bleeding during endoscopy have not been reevaluated after the utilization of endoscopic banding as endoscopic procedure. In addition, patients with severe liver failure are often excluded from clinical trials. The aim of this study was to prospectively reevaluate the risk factors affecting a 5-day failure after acute variceal bleeding in unselected cirrhotic patients, managed with the current standard treatment using vasoactive drugs, band ligation, and antibiotics.

Methods: One hundred and eighty five patients with liver cirrhosis and variceal bleeding admitted from January 2010 to July 2011were evaluated.

Results: Hepatocellular carcinoma was present in 28.1% of cases and portal vein thrombosis (PVT) was present in 17.3% of cases. Band ligation was feasible in 92.4% of cases. Five-day failure occurred in 16.8% of cases; 12 patients (6.5%) experienced failure to control bleeding or early rebleeding, and 66.7% of patients died within 5 days. The overall 5-day mortality rate was 14.6%. By multivariate analysis, we determined that Child-Pugh class C, a white blood cell count over 10×109/l, and the presence of PVT were the only independent predictors of the 5-day failure.

Conclusions: The prognosis of a consistent group of liver cirrhosis patients with variceal bleeding remains poor. The current treatment is highly effective in controlling variceal bleeding, but mortality is related mainly to the severity of liver failure.


The mortality of patients with cirrhosis and esophageal variceal bleeding (EVB) has greatly improved in the past 30 years.[1–5] In 1981, the seminal paper by Graham and Smith[6] reported a mortality rate of 42% at 6 weeks, with rebleeding as the cause of death in 60% of cases. The implementation of intensive-care management of these patients and the widespread use of a combination of vasoactive drugs, endoscopic therapies, and systemic antibiotics have contributed to the amelioration of the overall prognosis.[7]

More recent studies have focused on the identification of patients at a high risk of adverse outcomes requiring alternative or more aggressive therapies within a few days of the index EVB.[8] The hepatic venous pressure gradient is considered the strongest predictor of both rebleeding and mortality after EBV,[9] but it is not widely available in the acute setting. Recently, Abraldes et al.[10] found a correlation between the hepatic venous pressure gradient and some easily detectable clinical parameters. They drew up a simple score based on the etiology of cirrhosis, Child-Pugh class, and the level of blood pressure on admission, which identified patients at different risk levels of a 5-day failure.[10] The majority of patients in this cohort (74%) were treated with sclerotherapy together with vasoactive drugs, whereas the current standard of care is based on band ligation, which has been proven to be more effective.[11,12]

Moreover, some of the risk factors of early rebleeding have not been reevaluated after the utilization of banding. Difficult patients, such as those with advanced liver disease, multifocal hepatocellular carcinoma (HCC), or portal thrombosis, are usually excluded from such studies.

The aims of this study were as follows: (i) to prospectively reevaluate the prognostic indicators for 5-day failure in patients treated with the current standard of care based on vasoactive therapy, antibiotics, and band ligation, and (ii) to evaluate the outcomes of EVB in an all-comers series of patients, including patients usually considered ineligible for liver diseases.