Renal failure after upper gastrointestinal bleeding in cirrhosis: Incidence, clinical course, predictive factors, and short-term prognosis
Article first published online: 30 DEC 2003
Copyright © 2001 American Association for the Study of Liver Diseases
Volume 34, Issue 4, pages 671–676, October 2001
How to Cite
Cárdenas, A., Ginès, P., Uriz, J., Bessa, X., Salmerón, J. M., Mas, A., Ortega, R., Calahorra, B., De Las Heras, D., Bosch, J., Arroyo, V. and Rodés, J. (2001), Renal failure after upper gastrointestinal bleeding in cirrhosis: Incidence, clinical course, predictive factors, and short-term prognosis. Hepatology, 34: 671–676. doi: 10.1053/jhep.2001.27830
- Issue published online: 30 DEC 2003
- Article first published online: 30 DEC 2003
- Manuscript Accepted: 13 JUL 2001
- Manuscript Received: 15 DEC 2000
- Fondo de Investigación Sanitaria. Grant Number: FIS/00/0616
- Dirección General de Investigación Científica y Técnica. Grant Number: SAF99-0016
- Instituto Reina Sofia de Investigacion Nefrológica and Fundació Clínic per la Recerca Biomèdica, respectively.
To assess the incidence, clinical course, predictive factors, and prognosis of renal failure in patients with cirrhosis and gastrointestinal bleeding, 175 consecutive episodes of gastrointestinal bleeding in 161 patients were analyzed. Renal failure occurred in 20 (11%) episodes and was transient in 8 episodes and nontransient in 12. Renal failure was more common in patients with cirrhosis than in a control population of bleeding patients without cirrhosis matched by age and severity of the bleeding episode. Among 39 clinical and laboratory variables obtained at admission or during hospitalization related with the bleeding episode or with liver and renal function, the presence of hypovolemic shock, number of packed red blood cells transfused, Child-Pugh class at admission, and baseline platelet count were independent predictors of renal failure. The development of renal failure and hypovolemic shock was the only independent predictors of in-hospital mortality. Mortality rate among the 20 episodes with renal failure was 55% (11 deaths) as compared with only 3% (5 deaths) in the 155 episodes without renal failure (P < .01). The development of nontransient renal failure entailed a much greater mortality as compared with transient renal failure (10 of 12 [83%] vs. 1 of 8 [12%]; P < .01). In conclusion, renal failure is a common event in patients with cirrhosis and gastrointestinal bleeding, the occurrence of which is mainly related to the severity of bleeding and baseline liver function. Renal failure is a strong predictor of mortality in patients with cirrhosis and gastrointestinal bleeding.