Introduction: Preoperative diagnostic modalities include serum tumor marker level (CA 125) and imaging, both of which suffer from low sensitivity and specificity. The serum tumor marker CA 125 is not specific. Preoperative diagnosis of an ovarian tumor remains problematic. Suspected early ovarian tumors, unlike other malignancies, are seldom diagnosed via needle or punch biopsy. The greatest concern about ovarian tumor biopsy is that this procedure results in tumor leakage or rupture, with subsequent intraperitoneal spreading of cancer cells if the tumor is malignant, leading to upstaging. Hence, intraoperative frozen section examination plays a crucial role in the management of complex ovarian masses. Aim and Objective: Primary objectives are (1) to estimate the diagnostic accuracy of frozen section in diagnosing ovarian cancer in women presenting with suspicious pelvic mass as verified by paraffin section. (2) To estimate the accuracy of the final diagnosis of malignancy in a subgroup of women with a frozen section result of either borderline or cancer. The secondary objective is to determine whether the RMI 2 (Risk of Malignancy Index 2) can avoid frozen section. Results: The overall accuracy of frozen sections in the present study was 88.23%. Intraoperative consultation was concordant in 45 cases and discordant in 6 cases. Out of six discordant cases, four lesions were underdiagnosed, two lesions were overdiagnosed. The overall underdiagnosis rate was 7.84%, and overdiagnosis rate was 3.92%. Of the 25 patients with a benign diagnosis at frozen section, 96% (24/25) of patients had benign diagnoses, and 4% (1/25) patients had borderline ovarian lesions at final diagnosis. Of the 15 patients with borderline ovarian lesion on frozen sections, 33.3% (5/15) of cases were discordant with the final histopathology. 13.3% (2/15) of patients were diagnosed to have benign lesion at the final diagnosis; both patients were overtreated and underwent staging laparotomy in the primary setting. 20% (3/15) of patients were diagnosed to have malignant lesion at final diagnosis. All the intraoperative malignant diagnoses on frozen section were confirmed by the final histopathology. Sensitivity and specificity of the frozen section analysis for benign, borderline, and malignant ovarian masses were 92.3% and 96%, 90.9%, and 87.5% and 78.5% and 100%, respectively. The sensitivity, specificity, positive predictive value, and negative predictive value of the RMI 2 score in diagnosing malignant ovarian lesions were 60%, 61.54%, 60%, and 61.54%, respectively. Conclusion: Intraoperative frozen section diagnostic accuracy rates were high in diagnosing benign and malignant ovarian lesions and were relatively low for borderline ovarian neoplasms. RMI 2 is a reliable preoperative tool in differentiating complex ovarian masses, but it cannot completely replace intraoperative frozen section.
Reddy et al. (Tue,) studied this question.