Between 1989 and 1991, 150 patients with ovarian cancer were treated with chemotherapy, including cisplatin, in the Tokai Ovarian Tumor Study Group. Of these patients, 25 underwent cytoreductive surgery with lymphadenectomy, including removal of either pelvic or para-aortic lymph nodes, and 36 underwent both lymphadenectomies. A significant difference was observed between survival curves of the groups with positive and negative lymph nodes, respectively (P = 0.0049). The overall survival was longer in the lymphadenectomy group than in the nonlymphadenectomy group (P = 0.0842), and a significantly longer survival time was noted for stage III patients who underwent lymphadenectomy compared with those who did not (P = 0.0185). Multivariate analysis demonstrated that lymphadenectomy is a positive prognostic factor. The authors conclude that both pelvic and para-aortic lymph nodes should be resected to improve survival as well as to assess exact staging in patients with ovarian cancer.