Objective: To evaluate the reliability of sagittal abdominal diameter (SAD)—a surrogate of visceral obesity—in MRI, and its accuracy to predict the surgical morbidity of aortic lymphadenectomy. Design: A multicentre reliability (phase 1) and accuracy (phase 2) cohort study. Setting: Three Spanish referral hospitals. Population: High-risk endometrial cancer patients undergoing minimally invasive surgical staging. Patients were classified into subgroups: conventional vs. robotic-assisted laparoscopy, and transperitoneal vs. extraperitoneal technique. Methods: Retrospective analysis of data from the STELLA-2 randomized controlled trial. In the first phase, we measured the agreement of three SAD measurements (at the umbilicus, the renal vein, and the inferior mesenteric artery) and selected the most reliable one. In phase two, we evaluated the diagnostic accuracy of SAD to predict surgical morbidity. Main Outcome Measures: surgical morbidity was defined by a core outcome set including variables related to blood loss, operative time, surgical complications, and para-aortic lymphadenectomy difficulty. Results: In phase one, all measurements showed good inter-rater and intra-rater agreement. Umbilical SAD was the most reliable one. In phase two, we included 136 patients. Umbilical SAD had a good diagnostic accuracy to predict surgical morbidity in patients undergoing transperitoneal laparoscopic lymphadenectomy (0.73 in ROC curve). It performed better than BMI and other anthropometric measurements. We calculated a cut-off point of 246 mm (sensitivity: 0.56 and specificity: 0.80). Conclusions: Umbilical SAD is a simple, reliable, and potentially useful measurement to predict surgical morbidity in endometrial cancer patients undergoing minimally invasive surgical staging, especially when facing transperitoneal aortic lymphadenectomy.