The most important determinant of survival in patients with advanced ovarian cancer is the amount of postoperative residual tumor. Several models based on clinical and/or imaging techniques have failed to predict optimal resectability of advanced-stage ovarian tumors, and thereby failed to identify patients with a high likelihood of an optimal or complete cytoreduction. The recent report of a laparoscopy-based score model with an accuracy of 74% for identifying patients with residual tumors after debulking surgery has not been confirmed. The present study evaluates that model and presents a simplified laparoscopy-based score.
A total of 55 women with stage III/IV ovarian cancer who had both diagnostic laparoscopy and laparotomy for cytoreductive surgery were evaluated retrospectively. The laparoscopy-based score was based on the assessment of seven predictive index parameters: omental cake, peritoneal carcinosis, diaphragmatic carcinosis, mesenteric retraction, bowel infiltration, stomach infiltration, and liver metastases. Each parameter was given a value of 2 points if present and 0 if not, based on the calculation of sensitivity, specificity, positive (PPV) and negative (NPV) predictive values, and overall accuracy. Receiver operating characteristic (ROC) curve analysis was used to identify patients who were most likely to have an optimal or suboptimal surgical outcome.
As in a previous study, a laparoscopy-based score of <8 was associated with optimal cytoreduction (P ≤ .009). Sensitivity, specificity, PPV, NPV, and accuracy were 46%, 89%, 89%, 44%, and 60% respectively. The Area Under the Curve (AUC) value in ROC curve analysis was 0.74. Of the seven parameters examined, four met inclusion criteria for the simplified laparoscopy-based score: diaphragmatic carcinosis, mesenteric retraction, stomach infiltration, and liver metastases. Thirteen patients (12%) with a modified laparoscopy-based score of ≥4 had suboptimal cytoreduction, whereas all 42 patients (88%) with a score of <4 had optimal cytoreduction (P ≤ .002). Sensitivity, specificity, PPV, NPV, and accuracy for the modified laparoscopy-based score of ≥4 for suboptimal surgery were 35% (95% confidence interval [CI], 22–51), 100% (95% CI, 79–100), 100% (95% CI, 73–100), 43% (95% CI, 29–58), and 56% (95% CI, 43–69), respectively. ROC curve analysis showed an AUC of 0.68.
These data confirm a previous study that a laparoscopy-based score can predict the resectability of advanced ovarian tumors. A modified laparoscopy-based score based on four of the parameters was as accurate as a score based on seven parameters in predicting the surgical outcome.