IntroductionIt is unclear if sentinel node (SLN) mapping can replace pelvic- (PLD) and paraaortic lymphadenectomy (PALD) for high-risk endometrial cancer (EC). A diagnostically safe surgical algorithm, taking failed mapping cases into account, is not defined. We aimed to investigate the diagnostic accuracy of SLN mapping algorithms in women with exclusively high-risk EC. MethodsWe undertook a prospective national diagnostic cohort study of SLN mapping in women with high-risk EC from March 2017 to January 2023. The power calculation was based on the negative predictive value (NPV). Women underwent SLN mapping, PLD and PALD besides removal of suspicious and any FDG/PET–positive lymph nodes. Accuracy analyses were performed for five algorithms. Results170/216 included women underwent SLN mapping, PLD and PALD and were included in accuracy analyses. 42/170 (24.7%) had nodal metastasis. The algorithm SLN and PLD in case of failed mapping, demonstrated a sensitivity of 86% (95% CI 74–100) and an NPV of 96% (95% CI 91–100). The sensitivity increased to 93% (95% CI 83–100) and the NPV to 98% (95% CI 94–100) if PLD was combined with removal of any PET-positive lymph nodes. Equivalent results were obtained if PLD and PALD were performed in non-mapping cases; sensitivity 93% (95% CI 83–100) and NPV 98% (95% CI 95–100). ConclusionSLN-mapping is a safe staging procedure in women with high-risk EC if strictly adhering to a surgical algorithm including removal of any PET-positive lymph nodes independent of location and PLD or PLD and PALD in case of failed mapping.