[Evaluation of the short-term prognosis of cirrhotic patients admitted to intensive care, using 4 scoring systems]

A prospective evaluation of 4 scoring systems routinely used as prognostic indexes: 2 in hepatology units (Classification of Child-Turcotte, Classification of Pugh et al.) (specialized systems) and 2 in intensive care units (Acute Physiology Score, Simplified Acute Physiology Score) (non-specialized systems) was undertaken in 61 cirrhotic patients. Overall accuracy, defined as the percentage of survivors and non-survivors correctly classified, was not significantly different whatever the specialized system used (range: 59.0 p. 100 to 75.4 p. 100). Similarly, no difference was observed between the two non-specialized systems (overall accuracy of 81.9 p. 100 for the Acute Physiology Score and of 90.0 p. 100 for the Simplified Acute Physiology Score). However, the overall accuracy given by the Simplified Acute Physiology Score was significantly higher than that obtained with the specialized systems (p less than 0.05). Furthermore, this nonspecialized system was as easy to calculate and more reproducible than the specialized systems. These results suggest that the Simplified Acute Physiology Score may be substituted for the Child-Turcotte and Pugh classifications as a prognostic index in cirrhotic patients admitted to intensive care units.