Objective Borderline ovarian tumors (BOTs) are treated surgically like malignant ovarian tumors with hysterectomy, salpingectomy, omentectomy, and multiple random peritoneal biopsies in addition to removal of the ovaries. It is, however, unknown how often removal of macroscopically normal-appearing tissues leads to the finding of microscopic disease. To evaluate the value of random biopsies, omentectomy, and hysterectomy in operations for BOT, the macroscopic and microscopic findings in a cohort of these patients were reviewed retrospectively.

Materials Women treated for BOT at Odense University Hospital from 2007 to 2011 were eligible for this study. Data were extracted from electronic records. Intraoperative assessment of tumor spread (macroscopic disease) and the microscopic evaluation of removed tissues were the main outcome measures.

Results The study included 75 patients, 59 (78.7%) in International Federation of Gynecology and Obstetrics stage I, 9 (12%) in stage II, and 7 (9.3%) in stage III. The histologic subtypes were serous (68%), mucinous (30.7%), and Brenner type (1.3%). Macroscopically radical surgery was performed in 62 patients (82.7%), and 46 (61.3%) received complete staging. The surgeon’s identification of macroscopic tumor spread to the contralateral ovary and the peritoneum had a sensitivity of 88% and 69.2% and a specificity of 90.2% and 92.5%, respectively. The macroscopic assessment of the uterine surface, the omentum, and the pelvic and para-aortal lymph nodes was not a good predictor of microscopic disease. During follow-up, 4 patients (5.3%) relapsed with no relation to surgical radicality or the extent of staging procedures.

Conclusions Ovaries and peritoneal surfaces with a macroscopically normal appearance rarely contain a microscopic focus of BOT.