UNLABELLED: Upper gastrointestinal variceal bleeding is one of the most serious complications in patients with chronic liver disease. The aim of this trial is to identify in hospital mortality predictors in this illness. MATERIAL AND METHODS: 106 hospitalizations due to this disease from October 2001 to April 2006 in cohort design. In hospital mortality was confronted with age, sex, liver disease etiology, Apache II score at admission, variceal severity, severity of the hepatic failure evaluated according to the Child Pugh's classes, a history of variceal bleeding, initial endoscopic treatment failure, haemostatic failure within the first 48 hours, and rebleeding after 48 hours from admission. In order to identify independent mortality predictors, all the variables correlated significantly with hospital mortality were selected. For the quantitative variable APACHE II score at admission, the best mortality discrimination value was chosen. RESULTS: independent in hospital mortality predictors were: initial endoscopic treatment failure (p = 0.005), haemostatic failure in the first 48 hours (p = 0.012), and Child Pugh C class (p = 0.024). Although male sex, Apache II score at admission and rebleeding after 48 hours were also significantly related to mortality by univaried model, they did not qualify as independent predictors. CONCLUSIONS: the independent predictors of intrahospitalary mortality in patients with variceal bleeding due to chronic liver disease, and first-line of endoscopic treatment were: 1) Initial haemostatic endoscopic treatment failure, 2) Haemostatic failure in the first 48 hs, and 3) Child Pugh C class hepatic failure at admission.
UNLABELLED: Upper gastrointestinal variceal bleeding is one of the most serious complications in patients with chronic liver disease. The aim of this trial is to identify in hospital mortality predictors in this illness. MATERIAL AND METHODS: 106 hospitalizations due to this disease from October 2001 to April 2006 in cohort design. In hospital mortality was confronted with age, sex, liver disease etiology, Apache II score at admission, variceal severity, severity of the hepatic failure evaluated according to the Child Pugh's classes, a history of variceal bleeding, initial endoscopic treatment failure, haemostatic failure within the first 48 hours, and rebleeding after 48 hours from admission. In order to identify independent mortality predictors, all the variables correlated significantly with hospital mortality were selected. For the quantitative variable APACHE II score at admission, the best mortality discrimination value was chosen. RESULTS: independent in hospital mortality predictors were: initial endoscopic treatment failure (p = 0.005), haemostatic failure in the first 48 hours (p = 0.012), and Child Pugh C class (p = 0.024). Although male sex, Apache II score at admission and rebleeding after 48 hours were also significantly related to mortality by univaried model, they did not qualify as independent predictors. CONCLUSIONS: the independent predictors of intrahospitalary mortality in patients with variceal bleeding due to chronic liver disease, and first-line of endoscopic treatment were: 1) Initial haemostatic endoscopic treatment failure, 2) Haemostatic failure in the first 48 hs, and 3) Child Pugh C class hepatic failure at admission.