OBJECTIVE: This study was aimed at comparison of neoadjuvant chemotherapy with primary debulking surgery in advanced ovarian carcinoma. METHODS: Retrospective analysis of 285 patients with advanced ovarian carcinoma treated between 1980 and 1997 was performed. RESULTS: In the period 1980-1988 all patients underwent primary debulking surgery and 82% were cytoreduced to less than 0.5 cm largest residual tumor mass (n = 112). Analysis of this group of patients showed that some subgroups of patients (e.g., Stage IV disease or a total metastatic tumor load of more than 1000 g prior to debulking surgery) had a poor survival despite cytoreduction to no or less than 1 g of total residual tumor load. The complication rate was high especially in the group with unfavorable prognosis (postoperative mortality, 6%). In the period 1989-1997 (n = 173) the patients were surgically evaluated to receive primary chemotherapy (43%) or primary debulking surgery (57%). Prognostic variables were similar for both treatment periods. The actuarial crude survival was higher in the second time period (3-year crude survival of 26% +/- 4. 3 and 42% +/- 4.6 for the first and second time periods, respectively; P = 0.0001). The postoperative mortality was 0% during the second time period. From 1993 on, the decision to give neoadjuvant chemotherapy or to perform primary debulking surgery in patients with clinically obvious metastatic disease was made with the help of an open laparoscopy (n = 77). Median duration of the laparoscopy, blood loss, and hospital stay due to this procedure were 25 min, 10 ml, and 2 days. Primary and interval debulking surgery was performed in 36 and 63% of this subgroup of patients, respectively. CONCLUSION: In this retrospective analysis over two different time periods, crude survival was higher when treating about half of the patients with advanced ovarian carcinoma with primary chemotherapy instead of primary debulking surgery. The role of neoadjuvant chemotherapy should be evaluated in a prospective randomized study.