Of the 692 patients with hepatocellular carcinoma (HCC) who were admitted to our hospital between 1976 and 1990, 60 (8.8%) had small HCC with a maximal diameter of below 2 cm. The outcome of these 60 cases was analyzed after they had been divided into 4 groups based on the therapeutic method used: operation group (17 cases), percutaneous ethanol injection therapy (PEIT) group (20 cases), transcatheter arterial embolization (TAE) group (13 cases), and oral anticancer drug therapy (per os) group (10 cases). The 1-, 2-, 3-, 4-, and 5-year survival values obtained for the operation group (100%, 87.5%, 87.5%, 87.5%, and 87.5%, respectively) were significantly higher than those found for the per os group (P < 0.01). The best therapeutic results were achieved in the operation group. Although the follow-up period for the PEIT group was short, the 2-year survival of this group was nearly equal to that of the operation group. Whereas the duration of survival tended to increase in inverse proportion to the severity of the underlying liver cirrhosis, the survival values did not differ between solitary and multiple tumors or among the different histological grades of HCC. In this series, 20 patients died; 9 deaths (45.0%) were due to progressive disease and 3 deaths (15.0%) were attributed to hepatic failure. Because the operation group included many patients who displayed relatively good liver function, we cannot rule out the possibility that their excellent outcome may have been associated with this background factor. Therefore, further prospective investigation is necessary to compare the efficacy of various therapies in patient groups with a homogeneous background.