PURPOSE: The prognosis for primary gallbladder cancer is very poor. Because of the lack of specific symptoms, late diagnosis, advanced stage, and ineffective treatment are typical. The purpose of the present study was to investigate the early diagnostic factors that improve the survival rate, by comparing a preoperatively diagnosed group with a postoperatively diagnosed group. METHODS: This study was a clinical analysis of our surgical experience with primary gallbladder carcinomas during the 6 years from 1992 to 1997 at Chung-Ang University in Korea. RESULTS: The incidence of gallbladder carcinomas was 0.6% of the 6132 biliary-tract operations performed during the 6 years (suspected group: unsuspected group=15:23). The sex ratio of males to females was 1 to 2.46 with female predominance. Primary GB cancer was seen most commonly in patients in their 50s and 60s, and this group accounted for 68.4% of the entire group and had a median age of 65 years. The most common clinical manifestation was right upper abdominal pain in 27 cases (71.1%). The duration of the illness in preoperatively suspected patients was much shorter than that for intraoperatively or only pathologically diagnosed patients (p=0.046) As for diagnostic methods, ultrasound and abdominal CT scans were performed, and the preoperative diagnostic accuracies were 31.6% and 69.3%, respectively. The patients were grouped according to the staging system of Nevin et al. The five stages were classified as follows: stage I (suspected group-0 cases; unsuspected group-2 cases), stage II (1 case; 5 cases), stage III (1 case; 2 cases), stage IV (7 cases; 9 cases), and stage V (6 cases; 5 cases). Adenocarcinoma was the most common type of gallbladder cancer (32 cases: suspected group-10 cases; unsuspected group-22 cases), followed by squamoadenocarcinoma (3 cases: 2 cases/1 case), neurogenic carcinoma (1 case: 1 case/0 cases), histiocytoma (1 case: 1 case/0 cases), and papillary carcinoma (1 case: 1 case/0 cases). The total cumulative 1-year, 2-year, and 3-year survival rates were 67.7%, 37.6%, and 12.1%, respectively (suspected: 46.2%, 15.4%, and 0%; unsuspected: 71.4%, 52.9%, and 23.2%). The cumulative survival rates for patients in the unsuspected group were higher than those for patients in the suspected group (p=0.008). CONCLUSION: The survival rate of the preoperatively unsuspected group was better than that of the suspected group. The survival improvement depended the invasion by the gallbladder carcinoma. We suggest that the survival rate should be improved by early detection of the gallbladder carcinoma and on extended operation. Conclusively, further evaluation of patients with etiologic factors should be done, more specific exams (CT or MRI) should be evaluated, and an acceptable operation should be done.