Objective This study aimed to identify the correlation between histology tumor grade of the preoperative biopsy using dilatation and curettage (D&C), Pipelle, or hysteroscopy and final surgical specimen in women with endometrioid endometrial cancer.

Materials and Methods Patients on whom a preoperative biopsy was performed between 2009 and 2016 were reviewed and cases with apparent early-stage endometrioid endometrial cancer were included in the study. The accuracy of preoperative biopsy performed before hysterectomy using D&C, Pipelle, or hysteroscopy was compared.

Results A total of 332 patients were included. The diagnostic method was D&C in 43 cases (13%), Pipelle in 102 (31%), and hysteroscopy in 187 (56%). The preoperative diagnosis included G1 tumors in 177 cases (53.3%), G2 in 103 (31%), and G3 in 52 (15.6%). The surgical specimen confirmed endometrioid endometrial tumor in 309 patients (93%).

The accuracy rates of preoperative biopsy and surgical specimen were 74.69%, 73.19%, and 89.75% for G1, G2, and G3, respectively. Hysteroscopy showed better κ index (κ = 0.551) than did D&C (κ = 0.392) and Pipelle (κ = 0.430). Tumor diameter greater than 30 mm was the only factor independently associated with absence of correlation between preoperative and postoperative tumor grade (odds ratio [95% confidence interval], 1.959 [1.096–3.504], P = 0.023).

Conclusions Preoperative biopsy, regardless of the method, has its limitations in predicting the tumor grade compared with final surgical specimen in women with endometrioid endometrial cancer at an apparent early stage. Concordance between the biopsy and hysterectomy specimen is less likely to happen in the case of preoperative G1 or G2 tumors, as well as in big tumors. Although hysteroscopy was associated with the highest tumor grade agreement, no differences in correlation between the 3 methods (D&C, Pipelle, and hysteroscopy) were found.