Endometrial biopsy under direct hysteroscopic visualisation versus blind endometrial sampling for the diagnosis of endometrial hyperplasia and cancer: Systematic review and meta-analysis.

Background: Endometrial cancer is the most common gynaecological neoplasia in western countries. Diagnosis of endometrial cancer requires an endometrial biopsy. A good quality endometrial biopsy allows not only the identification of the pathology, but also preoperative histologic subtype classification. Endometrial biopsy can be performed under direct hysteroscopic visualisation, but also using blind sampling techniques. Objectives: To compare endometrial biopsy performed under direct hysteroscopic visualisation versus blind sampling for the diagnosis of endometrial hyperplasia and cancer. Materials and Methods: Systematic review and meta-analysis. Electronic databases were searched from their inception until March 2022.We included all studies comparing endometrial biopsy performed under direct hysteroscopic visualisation versus blind endometrial sampling. Main outcome measures: Sample adequacy, failure rate to detect endometrial cancer or endometrial hyperplasia, and rate of detection of endometrial cancer. The summary measures were reported as relative risk (RR) with 95% of confidence interval (CI). Results: Four studies with a total of 1,295 patients were included. Endometrial biopsy under direct hysteroscopic visualisation was associated with a significantly higher rate of sample adequacy (RR 1.13, 95% CI 1.10 to 1.17), and significantly lower risk of failure to detect endometrial cancer or endometrial hyperplasia (RR 0.16, 95% CI 0.03 to 0.92) compared to blind endometrial sampling. However, there was no significant difference between endometrial biopsies taken under direct hysteroscopic visualisation or blindly, with or without a preceding diagnostic hysteroscopy, in the rate of detection of endometrial cancer (RR 0.18, 95% CI 0.03 to 1.06). Conclusion: Hysteroscopic endometrial biopsy under direct visualisation is associated with significantly higher rate of sample adequacy and is comparable to blind endometrial sampling for the diagnosis of endometrial cancer and precancer. What is new?: Hysteroscopic endometrial biopsy under direct visualisation would be expected to reduce diagnostic failure for endometrial cancer compared to blind endometrial sampling.