Objective: This research was undertaken to identify risk factors for the involvement of sentinel lymph nodes (SLNs) in cases of endometrial cancer.

Methods: From February 2016 to April 2021, the cases of 874 women with endometrial cancer treated with the SLN algorithm at 11 institutions were analyzed in this retrospective study. Clinical and pathologic data were reviewed, and logistic regression was applied to identify predictive factors for SLN involvement.

Results: After the exclusion of 81 patients, the remaining cohort of 793 patients was analyzed. The involvement of SLNs occurred in 9.2% of these cases (n = 73). In univariate analysis, the risk of SLN involvement was seen to be significantly higher among patients aged >60 years and those with high-grade tumors, non-endometrioid histology, lymphovascular space invasion, deep myometrial invasion, tumor diameters of ≥2 cm, and cervical stromal invasion. Multivariate analysis identified the occurrence of deep myometrial invasion (OR 2.42, 95% CI 1.29 to 4.56; p = .006), cervical stromal invasion (OR 2.18, 95% CI 1.13 to 4.21; p = .020), and lymphovascular space invasion (OR 7.27, 95% CI 3.82 to 13.81; p < .001) as risk factors independently predictive of SLN involvement in the treatment of endometrial cancer.

Conclusion: Deep myometrial invasion, cervical stromal invasion, and lymphovascular space invasion were found to be independently predictive of the involvement of SLNs in cases of endometrial cancer. For cases in which SLN dissection was not or could not be performed, the identified independent risk factors are crucial for guiding adjuvant therapy.