A predictive model for failure to control bleeding during acute variceal haemorrhage

Area of Science:

• Hepatology
• Gastroenterology
• Clinical Medicine

Background:

• Variceal bleeding is a common, high-risk complication of cirrhosis.
• Early rebleeding in variceal bleeding is associated with increased mortality.
• No predictive models currently exist for acute variceal bleeding outcomes.

Purpose of the Study:

• To identify predictors of failure to control bleeding (continued or early rebleeding) in cirrhotic patients.
• To develop a model for predicting 30-day mortality in patients with variceal bleeding.
• To compare bleeding control and outcomes between different treatment strategies.

Main Methods:

• Prospective evaluation of 695 consecutive cirrhotic patients with hematemesis/melaena.
• Defined criteria for failure to control bleeding (continued bleeding or rebleeding within 5 days).
• Compared two treatment groups: initial transfusion/vasopressors followed by sclerotherapy vs. immediate injection sclerotherapy; also included a non-variceal bleeding group.

Main Results:

• Failure to control bleeding occurred in 44% (group 1) and 38% (group 2) of patients with variceal bleeding.
• Independent predictors of early rebleeding included active bleeding at endoscopy, encephalopathy, platelet count, alcoholism history, hematemesis presentation, log urea, and shorter admission interval.
• Predictors of 30-day mortality were early bleeding, bilirubin, encephalopathy, shorter admission interval, and log urea.

Conclusions:

• Active variceal bleeding at diagnostic endoscopy predicts failure to control bleeding (rebleeding within 5 days).
• Failure to control bleeding is a significant predictor of 30-day mortality in cirrhotic patients.
• Sclerotherapy may independently influence bleeding risk in some patients.