Ovarian cancer surgical resectability: relative impact of disease, patient status, and surgeon.

OBJECTIVES: Currently, we are unable to predict which patients are most likely to undergo successful debulking of ovarian cancer. We investigated the impact of clinical and surgical-pathologic factors at the time of initial exploration on the ability to achieve optimal cytoreduction.
METHODS: All consecutive patients with IIIC epithelial ovarian cancer operated at Mayo Clinic between 1994 and 1998 were included. The following pre- and intraoperative factors were included as dichotomous variables: age, ASA, CA125, ascites volume, carcinomatosis, diaphragm and mesentery involvement, and tendency of the operating surgeon (defined by the performance of radical procedures in more vs. less than 50% of patients operated). Pearson chi(2) test and logistic regression analysis were used for statistical analysis.
RESULTS: ASA, ascites, carcinomatosis, diaphragmatic tumor, mesentery involvement, and surgeon tendency all significantly correlated with residual disease (RD) in univariate analysis. However, only ASA, carcinomatosis and surgeon were independently associated with optimal RD. The subset of patients having ASA 3 or 4 and carcinomatosis comprised a high-risk group with just 46% achieving optimal RD overall. Even within this high-risk group, the rate of optimal cytoreduction ranged from 67% to 42% dependent upon surgeon tendency to employ radical procedures.
CONCLUSIONS: High-risk factors such as patient condition and extent of disease impact the ability to achieve optimal RD. However, this is greatly influenced by surgical effort. Models to predict optimal surgical outcomes based only on tumor and patient characteristics will be highly practice-dependent: thus, their utility in selecting patient for non-traditional primary approach to ovarian cancer must be looked at cautiously.