Purpose New techniques should be compared with the gold standard of classical open surgery to confirm their true effectiveness and potential advantages. Material and methods In a retrospective review between 1999 and 2005, 45 patients with a mean age of 22 months (range 1 to 168) who underwent lumbotomy pyeloplasty were compared to 45 patients with a mean age of 71 months (range 4 to 172) who underwent a retroperitoneoscopic pyeloplasty. All the patients were operated by the same two surgerons (JSV-HS) well trained in laparoscopic techniques. We analysed operative time, hospital stay, results. Results The respective indication for each technique has changed during this 7 years period : for the first 20 patients open surgery was used in 80% of cases and for the last 20 patients minimally invasive surgery was used in 90% of case. Operative time was longer for the retroperitoneal laparoscopic group (155 min versus 98) ; mean hospital stay was nearly similar (4,5 days versus 5,5). No intraoperative complications occured in either group. The incidence of post operative complication, mainly urine leakage, was 11% in the open group (of which 4 cases needed another general anaesthetic for ureteralcatheter placement) and 17% in the laparoscopic group (of which 6 patients needed a new general anesthesia). Mean follow up was 41 months and 16 months in the open and laparoscopy group, respectively ; 2 patients in the open group and 1 patient in the laparoscopic group needed to be reoperated later for restenosis (success rate 95,5% versus 97,7%). Conclusions Retroperitoneoscopic dismembered pyeloplasty in children remains a difficult operation even after a learning curve ; however this technique is as safe and effective as open pyeloplasty ; it is feasable after 3 months of age ; the main advantage in our study is cosmetic ; the other potential advantages (less pain, fast recovery) seem more clear in adolescents. New techniques should be compared with the gold standard of classical open surgery to confirm their true effectiveness and potential advantages. In a retrospective review between 1999 and 2005, 45 patients with a mean age of 22 months (range 1 to 168) who underwent lumbotomy pyeloplasty were compared to 45 patients with a mean age of 71 months (range 4 to 172) who underwent a retroperitoneoscopic pyeloplasty. All the patients were operated by the same two surgerons (JSV-HS) well trained in laparoscopic techniques. We analysed operative time, hospital stay, results. The respective indication for each technique has changed during this 7 years period : for the first 20 patients open surgery was used in 80% of cases and for the last 20 patients minimally invasive surgery was used in 90% of case. Operative time was longer for the retroperitoneal laparoscopic group (155 min versus 98) ; mean hospital stay was nearly similar (4,5 days versus 5,5). No intraoperative complications occured in either group. The incidence of post operative complication, mainly urine leakage, was 11% in the open group (of which 4 cases needed another general anaesthetic for ureteralcatheter placement) and 17% in the laparoscopic group (of which 6 patients needed a new general anesthesia). Mean follow up was 41 months and 16 months in the open and laparoscopy group, respectively ; 2 patients in the open group and 1 patient in the laparoscopic group needed to be reoperated later for restenosis (success rate 95,5% versus 97,7%). Retroperitoneoscopic dismembered pyeloplasty in children remains a difficult operation even after a learning curve ; however this technique is as safe and effective as open pyeloplasty ; it is feasable after 3 months of age ; the main advantage in our study is cosmetic ; the other potential advantages (less pain, fast recovery) seem more clear in adolescents.