The purpose of our research was to determine the reliability and accuracy of the Pipelle device in acquiring an adequate and representative endometrial sample and to compare it with D&C histology.A total of 127 women were prospectively included in the study. After Pipelle endometrial sampling, classic dilatation and curettage (D&C) was performed. The histological results of both techniques were compared to assess the accuracy of Pipelle sampling. Prior to endometrial sampling, transvaginal ultrasonography was performed to determine endometrial thickness and pathological lesions, and to exclude other pathologies confined to the pelvis. The outcome measures were specimen adequacy and comparability of histological diagnosis between the Pipelle specimen and the D&C specimen.Although no patient had a failed biopsy procedure, 13 patients had insufficient tissue in the D&C group, and 29 patients in the Pipelle group. In 100 of 127 patients (79%), the Pipelle endometrial histology results were in agreement with the D&C histology results. Pipelle biopsy was unable to diagnose 1 of 5 endometrial hyperplasia cases.The Pipelle device is a limited endometrial sampling technique for obtaining an adequate and representative endometrial sample. It should be reserved for those patients in whom there is only a minimal risk for endometrial carcinoma, hyperplasia and polyps. Any failure to obtain an endometrial specimen would suggest performing a full endometrial curettage, especially in anamnestically- and sonographically-determined patients at high risk for endometrial cancer.