Objectives: To reassess the prognosis of patients with cirrhosis admitted to the intensive care unit.
Design: A retrospective study in a medical intensive care unit in a teaching hospital in France.
Patients: All patients with cirrhosis without previous liver transplantation admitted in the period from 2005 to 2008.
Main results: One hundred thirty-eight patients were studied. Survival rates in the intensive care unit, in hospital, and at 6 months were 59% (95% confidence interval, 50%-67%), 46% (95% confidence interval, 38%-54%), and 38% (95% confidence interval, 30%-47%), respectively. In-hospital survival rates for patients requiring vasopressors, mechanical ventilation, or renal replacement therapy were 20%, 33%, and 31%, respectively. On day 1, independent risk factors for in-hospital mortality were age, albuminemia, international normalized ratio, and the Sequential Organ Failure Assessment score computed after discarding points for hematologic failure (modified Sequential Organ Failure Assessment score). Liver disease severity, assessed using a clinical classification, did not correlate with in-hospital mortality. In patients still alive after 3 days, the only prognostic factor was the modified Sequential Organ Failure Assessment score computed after 3 days. To predict in-hospital mortality, the modified Sequential Organ Failure Assessment score on day 1 had a greater area under the receiver operating characteristic curve (0.84) than the Simplified Acute Physiology Score II (0.78), the Child-Pugh score (0.76), the model for end-stage liver disease score (0.77), or the model for end-stage liver disease-natremia score (0.75). The in-hospital mortality rate with three or four nonhematologic organ failures on day 1 was not >70%, whereas it was 89% with three nonhematologic organ failures after 3 days spent in the intensive care unit.
Conclusion: In-hospital survival rate of intensive care unit-admitted cirrhotic patients seemed acceptable, even in patients requiring life-sustaining treatments and/or with multiple organ failure on admission. The most important risk factor for in-hospital mortality was the severity of nonhematologic organ failure, as best assessed after 3 days. A trial of unrestricted intensive care for a few days could be proposed for select critically ill cirrhotic patients.