Intra-operative frozen section results reliably predict final pathology in endometrial cancer

Abstract

Objectives: Typically, complete surgical staging is necessary for patients with high-risk endometrial cancer. However, patients with low-risk disease may be able to avoid lymphadenectomy and its associated morbidity. We sought to evaluate the agreement rates between the intra-operative frozen sections (FSs) and the final paraffin sections (PSs) at our institution, and to determine if this was a reliable method for guiding our intra-operative decision-making with regard to the necessity of lymphadenectomy.

Materials and methods: 116 patients with a pre-operative diagnosis of endometrioid adenocarcinoma of the uterus or complex atypical hyperplasia (CAH) underwent surgery at our institution. Demographic data, as well as information on stage, grade, histology and depth of invasion determined at FS and on PS were collected. Cohen's kappa statistic was used to assess the agreement rate between FS and final PS with regard to depth of invasion, grade, and histology.

Results: Our correlation rate between FS and final PS for histologic subtype, grade, and depth of myometrial invasion was 97.5%, 88%, and 98.2% respectively. Seven cases identified as complex atypical hyperplasia on FS were later determined to be cancerous on final PS, resulting in two patients being undertreated.

Conclusions: Our results support the use of FS analysis as a means to guide intra-operative decisions regarding lymphadenectomy. Determination of histologic subtype, depth of invasion and grade is reliable at our institution, and demonstrates high concordance rates between FS and PS. These factors should be used to guide intra-operative decision-making regarding the necessity of a lymphadenectomy in patients with endometrial cancer.

Keywords: Endometrial cancer; Frozen section; Lymph node dissection.