Concordance and Discordance of Endometrial Biopsy vs Hysterectomy Specimen Findings for the Diagnosis of Endometrial Cancer

Abstract

Background: Endometrial cancer (EC) is the most common malignancy of uterus in the Western World. The literature shows a varied association between biopsy versus hysterectomy specimen findings. The objective of the study is to describe the concordance and discordance between endometrial biopsy vs hysterectomy specimens for the detection and grading of EC.

Methods: This is a retrospective cross-sectional analytical study. Data were reviewed from 2018 to 2021 from the departmental archive.

Results: The mean±SD of age was 55.4±9.6 years. The most common clinical feature was postmenopausal bleeding (45.4%). The overall concordance for diagnosis of EC was 59.1% between biopsy and hysterectomy. The highest concordance found for endometrioid carcinoma-NOS was 67.6%. Concordance for overall grading of endometrial cancer was 52.3%, and it was maximum for G1 (78.6%). The strength of association was mild to moderate (C= 0.379 to 0.592) between biopsy vs hysterectomy was significant with the Chi² test. A biopsy had sensitivity, specificity, positive predictive value, and negative predictive value of 67.6%, 57.1%, 89.3%, and 25%, respectively, compared to hysterectomy for diagnosis of EC. Cohen’s Kappa test showed agreement between biopsy and hysterectomy was significant and moderate for the diagnosis EC (59.1%, k=0.1101) and for grading of EC (overall grading-52.2%, k=0.2925, architectural grading- 45.5% k=0.1156, nuclear grading 40.9%, k=0.1746).

Conclusion: Concordance/agreement between endometrial biopsy and hysterectomy was moderate for the diagnosis and grading of EC. The accuracy of biopsy is moderate in diagnosis of EC. We recommend endometrial biopsy as a minimally invasive surgical and cost-effective approach in resource-poor countries.