ABSTRACT
Purpose The main goal of this study is to assess the diagnostic agreement between preoperative biopsy and definitive histology of the surgical specimen to determine which sampling method is most suitable for diagnosis of early-stage endometrial cancer. Methods We performed a retrospective multicentric study to assess the correlation between three endometrial sampling methods (hysteroscopy, pipelle and D&C) in patients who had undergone preoperative endometrial biopsy and received primary surgical treatment for endometrial cancer. The primary objective was the agreement rate between hysteroscopy (HSC), endometrial biopsy (pipelle) and dilatation and curettage (D&C). Results A total of 1833 women who underwent preoperative sampling at 15 centers were included: 1042 biopsies were performed by HSC, 703 by pipelle and 88 by D&C. All three methods presented a moderate diagnostic concordance (κ = 0.400.61) with the definitive specimens histology: HSC (κ = 0.47), pipelle sampling (κ = 0.48) and D&C (κ = 0.48). Likewise, a subgroup analysis was performed by histological subtype comparing HSC and endometrial biopsy, showing that neither is superior as a diagnostic method. Conclusions According to this study, the use of pipelle sampling could become an adequate diagnostic method in endometrial cancer due to its similar agreement to HSC, ease of use and affordability (AU)