© 2016, Society of Surgical Oncology. Background: There is an ongoing debate on which approach, transperitoneal or extraperitoneal, is superior for the performance of laparoscopic aortic lymphadenectomy (LPA-LND) for the surgical staging of gynecologic cancer. A prospective randomized trial (STELLA trial) was designed to compare the perioperative outcomes and node retrieval of extraperitoneal versus transperitoneal aortic lymphadenectomy by laparoscopy or robot-assisted laparoscopy. Methods: Patients with endometrial or ovarian carcinoma requiring aortic lymphadenectomy for surgical staging were randomized to an extraperitoneal or transperitoneal approach by laparoscopy or robot-assisted laparoscopy between June 2012 and July 2014. Results: A total of 60 patients were entered into the study, 48 with endometrial cancer (80 %) and 12 with ovarian cancer (20 %). Thirty-one patients (51.6 %) were randomly assigned to the extraperitoneal group and 29 to the transperitoneal group (48.3 %). The means LPA-LND operating time was 90 min in both group (p = 0.343). The mean (range) blood loss was 105 (10–400) mL for extraperitoneal versus 100 (5–1000) mL for transperitoneal group (p = 0.541). There were no differences in the number of collected lymph nodes between the two groups [median (range) for extraperitoneal 12 (4–41) vs. 13 (4–29) for transperitoneal (p = 0.719)]. Conclusions: The extraperitoneal and transperitoneal approaches for laparoscopic and robotic aortic lymphadenectomy provide similar perioperative outcomes and nodal yields. Trial registration: The STELLA trial is registered at the US National Institutes of Health (ClinicalTrials.gov) #NCT01810874.