The aim of this study was to assess the value of the monoethylglicinexylidide assay, a dynamic liver function test based on the determination of the serum concentration of lidocaine major metabolite, as a predictor of survival in cirrhosis. For this purpose, the predictive value of monoethylglicinexylidide was evaluated in comparison with the Pugh score, ascites, encephalopathy and a number of different biochemical parameters as collected from the prospective follow-up of 118 patients with cirrhosis. A stepwise regression analysis was performed on the variables of prognostic value according to the Cox model and with respect to 1-yr survival; because Pugh score and monoethylglicinexylidide were the sole variables selected, they were proved to supply independent prognostic information. The most reliable cutoff values for discrimination between death and survival were 25 ng/ml or less for monoethylglicinexylidide and less than 9 for the Pugh score. In 74 patients without overt signs of liver failure (i.e., Pugh < or = 9), monoethylglicinexylidide provided a wide range of results (i.e., 4 to 77 ng/ml), namely values ranging from very low to elevated. Of the 38 patients with satisfactory Pugh scores (< or = 9) but poor monoethylglicinexylidide values (< or = 25), 11 died during follow-up and 3 underwent liver transplantation, despite having shown no clinical signs of liver failure at entry. On the bases of discriminant levels, the monoethylglicinexylidide test is suitable for adoption as a reliable and sensitive indicator of survival in patients with cirrhosis because it supplies more accurate prognostic information compared with the Pugh score.(ABSTRACT TRUNCATED AT 250 WORDS)