Risk of persistent or recurrent and intractable upper gastrointestinal bleeding in the era of therapeutic endoscopy.

OBJECTIVES: Patients who present to the emergency department with upper gastrointestinal bleeding can have persistent or recurrent (further) bleeding or self-limited bleeding. We performed a study to determine the frequency, risks factors, and impact on outcome of further bleeding.
METHODS: Clinical predictors of further bleeding were retrospectively identified in 137 consecutive patients presenting to our institution with upper gastrointestinal bleeding in 1994-1995.
RESULTS: Persistent or recurrent bleeding occurred in 30.7% of the cases, bleeding intractable to endoscopic therapy occurred in 15.3%. Hematemesis (odds ratio [OR] 5.7; 95% confidence interval [CI], 2.4-13.1, p = 0.0001) and a initial hemoglobin (OR, 0.8; 95% CI, 0.7-0.96; p = 0.01) were independent risk factors for persistent or recurrent bleeding, whereas liver disease (OR, 6.0; 95% CI, 2.0-18.4; p = 0.002) and hematemesis were independent risk factors for intractable bleeding. The mortality rate was 14.3 and 1%, respectively, in patients with and without further bleeding. In patients who did not present with hematemesis, liver disease, coagulopathy, hypotension, and initial hemoglobin < 11 g/dl, the frequency of further bleeding and mortality was 0%.
CONCLUSIONS: Persistent, recurrent, and intractable bleeding occurs in a substantial proportion of patients admitted with upper gastrointestinal bleeding. The risk of further bleeding can be estimated on the basis of clinical presentation. Further bleeding is associated with a worse outcome.