Institutional accuracy matters : does decision of lymph node dissection in endometrial adenocarcinoma depend on how the institution correlates frozen section and final pathology?

Purpose: The frozen section analysis results help determine the appropriate surgery and treatment for patients with endometrial adenocarcinoma. This study investigates the degree of compliance between the results of frozen section analysis and final pathology reports in patients with endometrial cancer. Material and Methods: This study included 357 patients with endometrial adenocarcinoma who underwent operation, follow-up, and treatment at our hospital. The patients' demographic, clinical, surgical, and pathological data were retrospectively analyzed. We compared the results of the frozen section and pathological specimens in terms of final pathology, tumor grade, myometrial invasion, tumor size, and lymphovascular system involvement (LVSI). Results: The frozen section analysis and final pathology results for tumor size and LVSI were significantly correlated when patients were divided by tumor type (tumor size: P=0.006, LVSI: P=0.024) or by risk for lymph node involvement (P=0.000). Overall, the frozen section analysis had an accuracy of 70% for tumor grades. The histological results of the frozen section analysis had an accuracy of 77.1% for type 1 tumors and 72.7% for type 2 tumors. Overall, the frozen section analysis had an accuracy of 95% for myometrial invasion. Conclusion: Intraoperative frozen section analysis can prevent unnecessary lymph node dissection when performed at qualified institutions.