We evaluated the adequacy of the trimethadione (TMO) tolerance test (the method of estimation from the serum dimethadione [DMO]/TMO ratio, DMO is only one metabolite of TMO, at 4 hours after oral administration of TMO) for estimating the severity of liver damage in 40 cirrhotic patients with and without hepatic encephalopathy. Serum dimethadione (DMO)/TMO ratios in a single blood sample after oral administration of TMO were significantly lower in cirrhotic patients with (0.07 +/- 0.02, p less than 0.05) or without (0.29 +/- 0.12, p less than 0.05) hepatic encephalopathy than in normal subjects (0.63 +/- 0.04). Serum DMO/TMO rations showed a good correlation with the following laboratory data: plasma indocyanine green retention at 15 minutes (r = -0.857, p less than 0.001), serum choline-esterase activity (r = 0.844, p less than 0.001), and albumin (r = 0.736, p less than 0.001). In cirrhotic patients with hepatic encephalopathy, the serum DMO/TMO ratio was found to be below 0.10, which was 16% of the normal level, by the TMO tolerance test. These findings suggest that the TMO tolerance test is a useful indicator of the hepatic parenchymal function in cirrhotic patients.