Survival benefit of extended lymphadenectomy in endometrial cancer: a meta-analysis with risk-stratified subgroup analysis.

Objective: To systematically evaluate the effect of combined pelvic and para-aortic lymphadenectomy (PPaLND) versus pelvic lymphadenectomy alone (PLND) or no lymphadenectomy on overall survival (OS) in patients with endometrial cancer, with a focus on risk-stratified subgroup analyses to clarify whether the survival benefit varies across different recurrence risk populations, thereby providing evidence-based guidance for individualized surgical decision-making. Methods: In accordance with the PRISMA guidelines, a systematic search was conducted in PubMed, Embase, Cochrane Library, Web of Science from database inception to December 2025. Randomized controlled trials (RCTs) and cohort studies comparing no lymphadenectomy, PLND, and PPaLND that reported hazard ratios (HR) with 95% confidence intervals (CI) for OS were included. Study quality was assessed using the Newcastle-Ottawa Scale (NOS). Meta-analyses were performed using a random-effects model with Review Manager 5.4 and Stata 17.0 software. Sensitivity analysis and funnel plots were employed to assess result robustness and publication bias. Results: Fourteen studies (1 RCT, 13 retrospective studies) comprising 25,432 patients were included. Given the observational nature of the evidence, adjusted estimates were designated as primary. In unadjusted models, the effect of PPaLND on OS was inconsistent, reflecting confounding by indication. After multivariate adjustment, PPaLND was associated with a significant improvement in OS compared to no lymphadenectomy (HR = 0.61, 95% CI: 0.49-0.75). Compared to PLND, PPaLND showed a survival benefit in the overall population (HR = 0.60, 95% CI: 0.44-0.81). However, in the PPaLND versus PLND comparison, no statistically significant differences were observed in the intermediate-to-high-risk subgroup (HR = 0.62, 95% CI: 0.31-1.24), the intermediate-risk group (HR = 0.65, 95% CI: 0.01-44.47), or the high-risk group (HR = 0.43, 95% CI: 0.05-4.04); the extremely wide confidence intervals in the latter two subgroups, each based on only two studies, indicate very low statistical power, and these results should be considered exploratory. Sensitivity analysis revealed significant changes in the intermediate-to-high-risk subgroup results upon exclusion of specific studies, indicating limited robustness. Funnel plots suggested the presence of potential publication bias. Conclusion: PPaLND may confer an overall survival benefit for patients with endometrial cancer, but its value varies across different recurrence risk strata, with the survival advantage in intermediate-to-high-risk patients remaining unconfirmed. Clinical decision-making requires a careful balance between potential survival benefits and surgical complication risks. Prospective randomized controlled trials are urgently needed to validate the optimal surgical strategy.