Sentinel Lymph Node Assessment in Endometrial Intraepithelial Neoplasia/Atypical Endometrial Hyperplasia: An Updated Systematic Review and Meta-Analysis.

Background/

Objectives

Endometrial intraepithelial neoplasia (EIN)/atypical endometrial hyperplasia (AEH) carries a substantial risk of concurrent endometrial carcinoma; however, the role of sentinel lymph node (SLN) assessment during hysterectomy remains controversial. We performed an updated systematic review and meta-analysis to quantify concurrent carcinoma and evaluate SLN/nodal outcomes in women with preoperative EIN/AEH.

Methods

PubMed, Embase, and the Cochrane Central Register of Controlled Trials were searched for studies reporting concurrent carcinoma and/or SLN/nodal outcomes in women with preoperative EIN/AEH undergoing hysterectomy. Proportions were pooled using a random-effects DerSimonian-Laird model with Freeman-Tukey double-arcsine transformation, with subgroup analyses by histological terminology. Risk of bias was assessed using the QUIPS tool.

Results

Nineteen studies, including 13,553 women, were included; 14 studies comprising 1174 women contributed to the pooled nodal-positivity estimate. The pooled rate of concurrent carcinoma was 39.6% (95% CI 34.9-44.3; I2 = 86%), with rates of 43.8% in AEH/atypical hyperplasia cohorts and 33.4% in EIN cohorts. Pooled nodal positivity was 3.0% (95% CI 2.1-4.0; I2 = 0%) and was predominantly low-volume disease. The pooled bilateral SLN detection rate was 71.6% (95% CI 67.3-75.7), and SLN was not associated with excess surgical complications. SLN findings influenced adjuvant treatment decisions in a minority of patients.

Conclusion

Concurrent carcinoma occurs in approximately two in five women with preoperative EIN/AEH, whereas nodal metastasis is uncommon and generally low-volume. SLN mapping appears feasible and may inform adjuvant management in selected patients. However, because this estimate was derived largely from patients already selected for higher-risk features, it does not, by itself, establish that risk-based selection would substantially outperform routine assessment. Taken together with preliminary prospective data, it is more appropriately viewed as hypothesis-generating support for a selective, risk-stratified approach, pending larger prospective validation.