Even if vesicoureteral reflux is a common condition in children, there are no guidelines about the best therapeutic approach. This study aims to compare the results of endoscopic injection and ureteral reimplantation in children with grade III, IV and V VUR. A multicenter retrospective study included children with grade III, IV and V VUR treated from 2003 to 2018 at three Departments of Pediatric Surgery. Patients were divided into Group A (endoscopic injections) and Group B (anti-reflux surgery), B1 (open, OUR), B2 (laparoscopic, LUR) and B3 (robot-assisted laparoscopic RALUR). Follow-up was at least 5 years. 400 patients were included, 232 (58%) in group A and 168 (42%) in group B. Mean age at surgery was 38.6 months [3.1–218.7]. Mean follow-up was 177.8 months [60–240]. Group A had shorter operative time than group B (P < 0.01); lower analgesic requirement (p < 0.05), shorter hospital stay (P < 0.05) and lower overall costs (p < 0.05), but higher postoperative PNPs (p < 0.01), lower success rate (p < 0.01) and higher redo-surgery percentage (p < 0.01). No differences in terms of postoperative complications, success rate and mean radiation exposure between the two groups. Endoscopy is associated with shorter operative time, shorter hospitalization and lower cost, also in case of multiple injections. Recurrence rate after surgery is lower meaning lower rate of re-hospitalization and radiation exposure for children.