The aim of this study was to identify any significant variables in the prognosis of 547 cases with hepatocellular carcinoma (HCC), and simultaneously confirm the survival among the different surveillance modalities for the initial detection of HCC in a closely followed-up group (regular periodic follow-up with monthly alpha-fetoprotein (AFP) and ultrasonography at least every 4 months), a not closely followed-up group (neither performed with AFP nor ultrasonography regularly) and an incidental group (incidentally discovered due to related symptoms). Five hundred and forty-seven consecutive patients with HCC diagnosed at the Internal Medicine Department of Saga Prefectural Hospital Koseikan from January 1989 to December 1998 were retrospectively analyzed. The 1-, 3- and 5-year survivals in all 547 cases were 69.7, 42.4 and 26.9%, respectively. The 1-, 3- and 5-year survivals in the cases found to have solitary HCC measuring 2 cm or less in diameter at the time of diagnosis were 97.3, 76.2 and 52.3%, respectively. Forty-seven point one percent of the closely followed-up group, which was the high-risk group were found to have solitary HCC measuring 2 cm or less in diameter (48 out of the 102 followed-up cases), while only 18.5 and 11.8% were found in the not closely followed-up group (46 out of 248 cases) and the incidental group (22 out of 186 cases), respectively. The 5-year survival in the closely followed-up, the not closely followed-up and the incidental groups were 42.9, 26.1 and 15.3%, respectively. The significant factors obtained in the closely followed-up group compared to those from the not closely followed-up group included AFP, tumor size, tumor number and portal thrombosis. These findings indicate the importance of a close follow-up for high-risk groups in order to identify HCC at an early stage, and thereby have a positive influence on survival.