Hysteroscopy-Guided Endometrial Sampling Diagnostic Performance in Endometrial Intraepithelial Neoplasia Patients.

Objective: To compare the diagnostic performance of hysteroscopy-guided versus blind sampling in detecting concurrent endometrial carcinoma in patients with endometrial intraepithelial neoplasia (EIN) and to identify factors associated with missing cancer diagnosis.

Design: This is a retrospective cohort study.

Setting: Integrated academic and community healthcare system in Minnesota and Wisconsin, USA, January 1, 2018, and January 1, 2023.

Participants: This included 151 patients diagnosed with EIN during endometrial sampling who underwent a hysterectomy within 3 months. Patients with concurrent cancer diagnoses were excluded.

Interventions: Patients diagnosed with EIN using hysteroscopy-directed biopsy were compared to those diagnosed with blind-sampling methods using the pathology results of the subsequent hysterectomy specimen as the gold standard comparator to analyze rates of missed endometrial cancer (EC) diagnosis.

Measurements and main results: The primary outcome was a reduced risk of unanticipated concurrent EC on the final hysterectomy pathology result for patients diagnosed with endometrial intraepithelial hyperplasia via a hysteroscopy-directed biopsy (odds ratios [OR] = 0.44, 95% confidence intervals [CI] = 0.20-0.95, p = .033). In multivariate analysis, body mass index ≥30 and patient age >60 were associated with an elevated risk of EC on final pathology (OR = 4.17, 95% CI = 1.51-11.51, p = .004; OR = 5.56, 95% CI = 1.22-35.21, p < .001), respectively, and diabetes mellitus was the only independent variable associated with a higher risk of EIN on final hysterectomy pathology (OR = 7.01, 95% CI = 1.40-35.04, p = .018). Age, body mass index, and endometrial thickness on pre-biopsy ultrasound were not associated with an increased risk of overlooking concurrent endometrial carcinoma on final hysterectomy pathology on univariate and multivariate analyses.

Conclusion: Hysteroscopy-directed biopsy may reduce the risk of missing a concurrent endometrial malignancy during endometrial sampling in women with EIN. The results affirm the superior diagnostic accuracy of hysteroscopy-directed endometrial evaluation.