The significance of preoperative chemoembolization using ethiodized oil, cisplatin and gelatin sponge (Sandwich therapy) for resectable hepatocellular carcinoma (HCC) was evaluated. One hundred and thirteen patients with solitary and less than 10 cm sized HCC who underwent radical hepatic resection were chosen for this study. Fifty-three patients received Sandwich therapy before surgery (Group A), and the remaining 60 patients under-went surgery without any preoperative treatments (Group B). Any background factors between two groups were not significantly different. The anticancer effects of this therapy were evaluated by histologic examination in 31 patients who had preoperative Sandwich therapy. In 22 of 31 patients (71%), the main nodules were completely necrotic. The ratios of patients with complete necrosis in daughter nodules were 7/12 (58%), in portal vein tumor emboli, 7/10 (70%), in intracapsular invasions, 11/21 (52%), in extracapsular invasions, 4/11 (36%). The 4-year disease-free survival rates in Group A and Group B were 56% and 27% respectively, and the rate of the former was significantly higher than that of the latter (p less than 0.05). The 4-year survival rates in Group A and Group B were 83% and 53% respectively. The rate of Group A was also significantly higher than that of Group B (p less than 0.01). We concluded that preoperative Sandwich therapy was very significant to obtain successful long-term disease-free survival and survival in regard to relatively early stage HCC.