Epidemiological and prognostic factors involved in upper gastrointestinal bleeding: results of a French prospective multicenter study.

BACKGROUND AND STUDY AIMS: The mortality rate from upper gastrointestinal bleeding (UGIB) remains high, at 5 % - 10 %. The aim of the current study was to describe the epidemiological characteristics, prognostic factors, and actual practice in a cohort of patients with UGIB admitted to French general hospitals.

METHODS: From March 2005 to February 2006, a prospective multicenter study was conducted at 53 French hospitals. A total of 3298 patients admitted for UGIB were enrolled consecutively. Patient data were collected up to the date of discharge from hospital.

RESULTS: Data were available for 2130 men and 1073 women (mean age 63 ± 18 years), one-third of whom were taking drugs that would increase the risk of UGIB. The two main causes of bleeding were peptic ulcers (38 %) and esophagogastric varices (EGV) or portal hypertensive gastropathy (24.5 %). Mean Rockall score was 5.0 ± 2.3. Endoscopy was performed on 96 % of patients (within 24 hours in 79 %), and 66 % of those with ulcers and 62.5 % of the EGV patients underwent hemostatic therapy when indicated. Rebleeding occurred in 9.9 % of the patients, and 8.3 % died. Independent predictors of rebleeding were: need for transfusion (odds ratio [OR] 19.1; 95 % confidence interval [95 %CI] 10.1 - 35.9); hemoglobin < 10 g/dL (OR: 1.7; 95 %CI 1.1 - 3.3); Rockall score (OR: 1.4 for each 1 point score increase; 95 %CI 1.0 - 1.9), systolic blood pressure < 100 mmHg (OR: 1.9; 95 %CI 1.4 - 2.5), and signs of recent bleeding (OR: 2.4; 95 %CI 1.7 - 3.5). Independent predictors of mortality were: Rockall score (OR: 2.8; 95 %CI 2.0 - 4.0), co-morbidities (OR: 3.6 for each additional co-morbidity; 95 %CI 2.0 - 6.3), and systolic blood pressure < 100 mmHg (OR: 2.1; 95 %CI 1.8 - 2.8). Rockall score, blood pressure and co-morbidities were taken as continuous variables meaning that the OR was 1.4 for every point increase, it was the same for blood pressure.

CONCLUSION: UGIB still occurs mainly as a result of peptic ulcers and portal hypertension in France, and causes significant rates of mortality. There is scope for improvement via better prevention (better use of UGIB-facilitating drugs), endoscopic therapy, and management of co-morbidities.