Endometrial hyperplasia - the dilemma of management remains: a retrospective observational study of 280 women.

Objective: To quantify the rate of inconsistency in histopathological reporting between endometrial biopsy specimens (obtained by Pipelle endometrial sampler or curettage) and hysterectomy specimens using the World Health Organization classification criteria.

Study Design: A retrospective review of the records of 280 women with a histopathological diagnosis of endometrial hyperplasia treated in Ipswich Hospital NHS Trust, UK from 1 January 1998 to 31 May 2009.

Results: Discrepancy was found between the histopathological results of endometrial samples and hysterectomy specimens. The discrepancy was doubled for specimens obtained using a Pipelle endometrial sampler, with false-positive (i.e. overdiagnosis when the hysterectomy specimen showed a better diagnosis) and false-negative (i.e. underdiagnosis when the hysterectomy specimen showed a worse diagnosis) rates of 5.3% and 22.6%, respectively. For curettage specimens, the false-positive and false-negative rates were 1.8% and 13.2%, respectively. All cases of curettage were performed under general or regional anaesthesia, and were preceded by hysteroscopy. Apart from age, no risk factors were associated with a worse diagnosis. The association of age differed between types of endometrial hyperplasia and cancer; the strongest association was seen for cancer and the weakest association was seen for simple hyperplasia.

Conclusion: Hysteroscopy and curettage may be considered when simple or complex hyperplasia is diagnosed from a specimen obtained with a Pipelle endometrial sampler. When a diagnosis of atypical hyperplasia is made, irrespective of the method of endometrial sampling, the gynaecologist must be concerned that endometrial carcinoma exists concomitantly within the uterus.