Vol. 48 (6): 961-968, November - December, 2022 ORIGINAL ARTICLE doi: 10.1590/S1677-5538.IBJU.2022.0194 Multicenter comparative study of open, laparoscopic, and robotic pyeloplasty in the pediatric population for the treatment of ureteropelvic junction obstruction (UPJO) _______________________________________________ Sebastian Tobía González 1, 2, Tiago E. Rosito 3, 4, Anna Bujons Tur 5, Javier Ruiz 6, Rafael Gozalbez 7, Anabella Maiolo 2, Patric M. Tavares 3, 4, Antonio Rebello Horta Gorgen 3, Erika Llorens de Kencht 5, Yesica Quiroz Madarriaga 5, Santiago Weller 6, Ignacio Pablo Tobia 1, Miguel Castellan 7, Juan Pablo Corbetta 6 1 Department of Urology, Universidad Nacional de la Plata – UNLP, La Plata, Buenos Aires; 2 Department of Urology, Hospital Interzonal Especializado en Pediatría “Sor María Ludovica”, La Plata, Buenos Aires, Argentina; 3 Departamento de Urologia, Grupo de Urologia Reconstrutiva e Infantil (GURI), Hospital de Clínicas de Porto Alegre, Porto Alegre, RS, Brasil; 4 Programa de Pós-graduação em Ginecologia e Obstetrícia, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brasil; 5 Department Pediatric Urology, Fundació Puigvert, Barcelona, Catalunya, Spain; 6 Department of Urology, Hospital Pediatric Prof. Dr. Juan P. Garrahan, Ciudad Autonoma de Buenos Aires, Buenos Aires, Argentina; 7 Division of Pediatric Urology, Nicklaus Children’s Hospital, Miami, FL, United States ABSTRACT ARTICLE INFO Introduction: Dismembered open pyeloplasty described by Anderson and Hynes is the Juan Corbetta “gold standard” for the treatment of ureteropelvic junction obstruction. The aim of our https://orcid.org/0000-0003-1094-338X study was to compare the results of open (OP) vs laparoscopic (LP) vs robotic (RALP) pyeloplasty. Keywords: Material and Methods: A multicenter prospective review was conducted of pyeloplasty Ureteral Obstruction; surgeries performed at five high-volume centers between 2014 and 2018. Demogra- Laparoscopy; Robotics phic data, history of prenatal hydronephrosis, access type, MAG3 renogram and di- Int Braz J Urol. 2022; 48: 961-8 fferential renal function, surgery time, length of hospital stay, and complication rate (Clavien-Dindo) were recorded. Access type was compared using the Kruskal-Wallis, _____________________ Chi-square, or Fisher’s exact tests. Submitted for publication: Results: A total of 322 patients were included: 62 OP, 86 LP, and 174 RALP. The mean April 05, 2022 age was 8.13 (r: 1-16) years, with a statistically significant lower age (mean 5 years) in OP (p < 0.001). There were no significant differences in the distribution of the side _____________________ affected. Operative time was 110.5 min for OP, 140 min for LP, and 179 min for RALP Accepted after revision: (p < 0.0001). Hospital stay was significantly shorter in the RALP group than in the June 13, 2022 other groups (p < 0.0001). There were no differences in postoperative complications and reoperations between the three groups. _____________________ Conclusions: Minimally invasive surgery for the management of UPJO in children is Published as Ahead of Print: gaining more acceptance, even in patients younger than 1-year-old. Operative time August 20, 2022 continues to be significantly shorter in OP than in LP and RALP. Hospital stay was shorter in RALP compared to the other techniques. No differences were found in com- plication rates, type of complications, and reoperation rate. 961 IBJU | OPEN, LAPAROSCOPIC AND ROBOTIC PYELOPLASTY INTRODUCTION After IRB approval, medical records of all patients with a history of UPJO were reviewed Open dismembered pyeloplasty (OP), evaluating the following characteristics: age at originally described by Anderson and Hynes, is surgery, sex, affected side, history of urinary tract the most commonly performed surgical proce- infection (UTI), flank pain, operative time (skin dure for the treatment of ureteropelvic junction incision to skin closure -port removal in LP and obstruction (UPJO), with long-term success rate RALP -, in minutes), type of stent placement du- of around 95% (1, 2). Over the last two decades, ring the surgery and length of hospital stay (de- however, minimally invasive surgery (MIS) te- termined by the computerized time recorded on chniques such as laparoscopic pyeloplasty (LP) hospital admission and discharge records). and robot-assisted laparoscopic pyeloplasty Indications for surgical intervention were (RALP) have been developed and popularized impairment of differential renal function (< 40%) as a standard of care in common practice. by nuclear scan or a decrease in split renal func- Since its first description by Peters et al., tion of > 10% on subsequent studies, obstructi- laparoscopic pyeloplasty has not been as po- ve drainage curve on diuretic renogram, symp- pular in pediatric urology as in the adult po- tomatic obstruction (recurrent flank pain, UTI), pulation possibly due to its technical difficulty and progressive worsening of hydronephrosis on and long learning curve, and better recovery in ultrasound images (urinary tract dilation - UTD children compared to adults (3, 4). - classification system) (14). Each institution’s re- Although MIS techniques are associated gistered technique is routinely performed outside with a longer operative time compared to open the learning curve. procedures, they have shown benefits in terms Complications were assessed using the of shorter postoperative hospital stay and lower Clavien-Dindo grading system (15). Post-surgical morbidity (4-6). Nevertheless, LP is technically patients were followed-up with ultrasonography demanding and has a long learning curve (7, 8). every 3 months during the first year of surgery, RALP has reduced the technical difficulty of this every 6 months in the second year, and then on procedure with a shorter learning curve compa- an annual basis. Patients were defined as having red to LP (9-13). Therefore, we believe that the a good outcome when SFU grade hydronephro- three techniques should have similar outcomes. sis improved without symptom recurrence. In pa- The aim of this study was to compare tients in whom SFU grade decreased but was still outcomes in safety and effectiveness between greater than 2, we performed a MAG3 renal scan OP, LP, and RALP in a large, multicenter cohort to rule out obstruction. of pediatric patients with UPJO. Continuous variables are expressed as mean, or median and range and categorical varia- MATERIAL AND METHODS bles are expressed as absolute value and/or per- centage. For comparison between techniques, the The study was approved by the Kruskal-Wallis, Chi-square, or Fisher’s exact tests Ethics Committee of each Hospital (IRB: were used depending on the case. In the case of 18967819.8.0000.5327). multiple comparisons, they were adjusted using A multicenter retrospective cohort stu- the Bonferroni correction method. In all cases, a dy was conducted with prospectively collected p-value of less than 0.05 was considered as a cut- data. We included pyeloplasty surgeries (inclu- -off for a significant value. SPSS 22.0™ software ding redo cases) performed in children younger was used. than 16 years of age between 2014 and 2018. The choice of OP, LP, and RALP were due to the Surgical technique surgeon and center preference, and the tech- niques were compared in terms of safety and a. OP: a subcostal flank or dorsal lumbotomy effectiveness. approach was used. Conventional dismem- 962 IBJU | OPEN, LAPAROSCOPIC AND ROBOTIC PYELOPLASTY bered pyeloplasty was performed with a RALP (3 trocars): 3 midline trocars (8 mm) running polydioxanone (6/0 PDS® II) anas- were used, one for the camera at the umbilicus tomotic suture. The bladder catheter was and the others at least 4 cm above and below the removed 24 hours later. A ureteral stent or umbilicus, respectively. nephrostomy tube was used only in select The bladder catheter was removed 24 cases and a perirenal drain was left in place hours later. Ureteral stents, when placed, were in all patients. removed approximately 4 weeks postoperatively b. LP: transperitoneal dismembered pyelo- via cystoscopy using short-acting anesthesia. In plasty was performed using 3 ports (one a group of patients in whom the stent was pla- 5- or 10-mm port for the 30-degree optics ced in a retrograde fashion and a string was left and two 3- or 5-mm ports for the instru- in the genitalia, the stent was removed in the ments). A running (5/0 PDS® II) anasto- clinic 2-3 weeks postoperatively. motic suture was placed, and a ureteral double J stent was placed in an antegrade RESULTS fashion in all patients. The Foley catheter was removed on postoperative day 1 or on A total of 322 patients were included in the next morning. the study: 62 OP, 86 LP, and 174 RALP. The c. RALP (4 trocars): an infraumbilical port was mean age was 8.1 years (range 1-16). Children used for the optics (8.5–12 mm), an 8-mm that underwent OP were younger than those working port in the epigastric region, ano- undergoing procedures with the other techni- ther 8-mm working port below the optic ques (p < 0.001 – Table-1). Demographic data port laterally to the midline in the right analysis showed that in the RALP group fema- / left iliac fossa, and an assistant port (5 le patients prevailed (p < 0.001). There were mm) was placed between the optic port and no significant differences regarding the distri- the lower working port (left pyeloplasty) or bution of the affected side between the three between the optics and the upper working groups (Table-1). port to lift the liver (right pyeloplasty) for The majority of the patients were older the introduction of the suture, retraction, than 6 years (64%, 206 patients). Table-1 sho- or suction. ws the most used technique according to age Table 1 - Sex, Age and Side of UPJO of patients submitted to open, laparoscopic and robotic pyeloplasty. OP LP RALP p (n=62) (n=87) (n= 161) Sex: Female (%) 18 (29) 31 (36) 95 (54,6) 0.0001 Age: mean (range) 5 (1 – 15) 10.5 (1-16) 8.9 (1 – 16) 0.0001 < 1 year (%) 20 (32.3) 2 (2.3) 9 (5.2) 1-6 years (%) 22 (35.5) 13 (15.1) 50 (28.7) > 6 years (%) 20 (32.3) 71 (82.6) 115 (66.1) Side (%) 0.469 Right 27 (43.5) 29 (33.7) 74 (42.5) Left 35 (56.5) 57 (66.3) 98 (56.3) Bilateral 0 0 2 (1.1) OP = Open Pyeloplasty; LP = Laparoscopic Pyeloplasty; RALP = Robotic-assisted laparoscopic pyeloplasty 963 IBJU | OPEN, LAPAROSCOPIC AND ROBOTIC PYELOPLASTY group; OP was more commonly used in children Another meta-analysis by Huang et al. com- younger than 1 year and LP in patients older pared the outcomes of LP and OP in children. than 6 years. In the intermediate group, OP and One randomized controlled trial and 15 com- RALP were more commonly used than LP. parative studies were included in the analysis. Operative time was shorter in the OP Success rates of both procedures were between group (110.48 min vs 140 min for LP and 179 83 and 100% and were found to be comparable min of RALP). On the other hand, hospital stay (19). In our series, the success rates were higher was shorter in the RALP group in comparison than 95% and no statistically significant diffe- with the LP and OP groups (p < 0.0001). There rence was observed regarding success between were no significant differences in postoperati- the procedures. ve complications or reoperation rate. A urete- In infants, the operation may be more ral stent was placed during surgery in almost challenging due to the limited space; (19) ho- all patients undergoing MIS, exceeding the rate wever, RALP has recently been successfully per- of stent placement in OP (p <0.0001) (Figure-1 formed in infants, with an operative time that and Table-2). was even shorter than that of OP (20, 21). Other In only 5 patients (1.55%) UPJO recur- techniques such as the Flexdex® articulating red after a median follow-up of 42.3 months. needle driver may facilitate the procedure (22). In these patients, SFU grade decreased but was Postoperative hospital stays of patients still greater than 2. MAG3 renal scan confirmed undergoing RALP and LP were shorter com- the obstruction (Table-2). pared to open surgery. Since the children had When analyzing complication rate no underlying conditions (comorbidities), the (n=51) according to initial vs redo procedures, length of hospital stay was a reasonably good Clavien-Dindo grades IIIb to V were found to measure of the trauma caused by the surgery. be mostly associated with redo surgeries with a A reduced length of hospital stay with RALP/ statistically significant difference. LP has previously been shown and is of major importance, both for the patient and caregivers DISCUSSION and for the hospital (23). No statistical difference was found betwe- In our study, RALP and LP proved to be en the three groups regarding complication rate. safe and successful methods, with several ad- Compared to other studies reporting complication vantages over OP. While operative time was rates of 0–33% for RALP (24), our results were in longer, hospital stay was shorter. As shown in the lower range (0-18%). In addition, most com- previous studies (16, 17) operative time was plications were stent-related and transient, with longer in RALP than in LP and OP. A contri- only 14/322 (4.3%) patients requiring a second buting factor may have been older age, which surgical intervention. It seemed that postoperative has been shown to be associated with operative complications, especially those related to pyelo- time (16) in the RALP group. nephritis and stent morbidity, increase the risk of The first important outcome to be con- needing secondary repair. On the other hand, it is sidered after minimally invasive pyeloplasty is possible that not the complications per se increa- the success of the procedure. In a meta-analy- sed the risk for reoperation, but that the compli- sis, Cundy et al. compared RALP, LP, and OP cations were an indication of an already distor- in children. Twelve articles were included of ted healing process related to either the surgery which five were cohort and seven case-control- or factors inherent to the patient. Complications led studies. Success rates of higher than 95% rates were higher in the redo surgeries group, but were reported for RALP and LP, whereas in the some studies showed that other techniques such as comparative studies of RALP and OP included, endopielolitotomy have a greater failure rate (25). success rates were higher than 87% and the In a prospective comparative case-control procedures were found to be comparable (18). study (OP vs LP), Piaggio et al. (26) described four 964 IBJU | OPEN, LAPAROSCOPIC AND ROBOTIC PYELOPLASTY Figure 1 - Comparison of surgical time and hospital stay in open, laparoscopic and robotic pyeloplasty. Figure 1A. Surgical Time. Figure 1B. Hospitay Stay OP = Open Pyeloplasty; LP = Laparoscopic Pyeloplasty; RALP = Robotic-assisted laparoscopic pyeloplasty 965 IBJU | OPEN, LAPAROSCOPIC AND ROBOTIC PYELOPLASTY Table 2 - Surgery outcomes and complications’ rate of patients submitted to open, laparoscopic and robotic pyeloplasty. OP LP RALP p (n=62) (n=86) (n=174) Transfusion (%) 0 0 0 NS¹ Operative time, median (min.) 110 (60-210) 140 (60-316) 179 (108-361) p = 0.0001 Stent (%) 22 (35.5%) 83 (96.50%) 174 (100%) p = 0.0001 Length of hospital stay, median (days) 5 (3 – 13) 3 (1-12) 1 (1-11) p = 0.0001 UPJO recurrence (n) 1 2 2 NS Follow-up months median (range) 57.8 (7-136) 41.1 (3-87) 44.6 (3-140) OP = Open Pyeloplasty; LP = Laparoscopic Pyeloplasty; RALP = Robotic-assisted laparoscopic pyeloplasty; UPJO = Ureteropelvic Junction Obstruction NS = non-significant complications in the LP group: febrile UTI (two), tients showing good effectiveness of the pro- double J stent disruption, and meatal stenosis cedure regardless of the technique (OP, LP, or (one each); and three in the OP group: febrile RALP) or age. No significant differences were UTI, flank pain due to stent displacement, and found when comparing with reoperations and/ persistent gross hematuria (one each). Two pa- or complications according to age. tients in the LP group and one in the OP group This study has some limitations, espe- needed additional procedures (p > 0.05): meato- cially due to the retrospective design, but this is a plasty and ureteroscopy for stent removal (LP) multicenter study in the Ibero-american scenario and stent repositioning (OP). with a significant number of patients. The diffe- The optimal age is one of the concerns for rence in age between the groups could be a se- the use of LP in children. It is generally believed lection bias that could affect the results. Also, in that it is difficult to perform LP in small children pediatric urology controversy remains regarding due to the risk of pelvic perforation when introdu- what patients with hydronephrosis require sur- cing the trocars. Excellent intracorporeal laparos- gery and what clinical and imaging methodology copic suturing skills are imperative in limited spa- should be selected to define recurrence of UPJO. ces (27). In our series, no significant differences were found when comparing reoperations and/or CONCLUSIONS complications according to age. In a systematic review, only two studies Open, laparoscopic, and robotic-assisted presented the age of the patient as mean +- SD pyeloplasty in the pediatric population have (28). The analysis showed that patients who similar success rates and are safe to perform, underwent LP were significantly older (50.90 even in children less than 1-year-old where tra- months) than those who underwent OP. The au- ditionally open pyeloplasty is more performed. thors of the seven remaining studies also found No differences were found in complication ra- that patients in the LP group were older than tes, type of complications, and reoperations those in the OP group. Other authors, however, between the three groups. Operative time was reported similar or even younger ages in the LP shorter in the OP. Severe complications requi- group compared to the OP group (29, 30). red a greater number of reoperations. The main strength of this international More studies are needed to evaluate multicenter study is the large number of pa- RALP, especially in infants. In addition, aspects, 966 IBJU | OPEN, LAPAROSCOPIC AND ROBOTIC PYELOPLASTY such as surgical costs and patient satisfaction, 14. Nguyen HT, Benson CB, Bromley B, Campbell JB, Chow should be further assessed. J, Coleman B, et al. Multidisciplinary consensus on the classification of prenatal and postnatal urinary tract CONFLICT OF INTEREST dilation (UTD classification system). J Pediatr Urol. 2014; 10:982-98. None declared. 15. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation REFERENCES in a cohort of 6336 patients and results of a survey. Ann Surg. 2004; 240:205-13. 1. Anderson JC, Hynes W. Retrocaval ureter; a case 16. Lee RS, Retik AB, Borer JG, Peters CA. Pediatric robot diagnosed pre-operatively and treated successfully by a assisted laparoscopic dismembered pyeloplasty: plastic operation. Br J Urol. 1949; 21:209-14. comparison with a cohort of open surgery. J Urol. 2006; 2. O’Reilly PH, Brooman PJ, Mak S, Jones M, Pickup C, 175:683-7. Atkinson C, et al. The long-term results of Anderson- 17. Yee DS, Shanberg AM, Duel BP, Rodriguez E, Eichel Hynes pyeloplasty. BJU Int. 2001; 87:287-9. L, Rajpoot D. Initial comparison of robotic-assisted 3. Peters CA, Schlussel RN, Retik AB. Pediatric laparoscopic laparoscopic versus open pyeloplasty in children. dismembered pyeloplasty. J Urol. 1995; 153:1962-5. Urology. 2006; 67:599-602. 4. Mei H, Pu J, Yang C, Zhang H, Zheng L, Tong Q. 18. Cundy TP, Harling L, Hughes-Hallett A, Mayer EK, Laparoscopic versus open pyeloplasty for ureteropelvic Najmaldin AS, Athanasiou T, et al. Meta-analysis of junction obstruction in children: a systematic review and robot-assisted vs conventional laparoscopic and open meta-analysis. J Endourol. 2011; 25:727-36. pyeloplasty in children. BJU Int. 2014; 114:582-94. 5. Tan HL, Roberts JP. Laparoscopic dismembered 19. Huang Y, Wu Y, Shan W, Zeng L, Huang L. An updated pyeloplasty in children: preliminary results. Br J Urol. meta-analysis of laparoscopic versus open pyeloplasty 1996; 77:909-13. for ureteropelvic junction obstruction in children. Int J 6. Leonardo CR, Muzzi A, Tavora JE, Soares RQ. The Clin Exp Med. 2015; 8:4922-31. outcomes of mini-laparoscopic pyeloplasty in children - 20. Casale P. Robotic pyeloplasty in the pediatric population. brazilian experience. Int Braz J Urol. 2020; 46:253-9. Curr Opin Urol. 2009; 19:97-101. 7. Casale P. Robotic pyeloplasty in the pediatric population. 21. Bansal D, Cost NG, DeFoor WR Jr, Reddy PP, Minevich Curr Opin Urol. 2009; 19:97-101. EA, Vanderbrink BA, et al. Infant robotic pyeloplasty: 8. Palese MA, Munver R, Phillips CK, Dinlenc C, Stifelman M, comparison with an open cohort. J Pediatr Urol. 2014; DelPizzo JJ. Robot-assisted laparoscopic dismembered 10:380-5. pyeloplasty. JSLS. 2005; 9:252-7. 22. Gorgen ARH, Araldi M, de Oliveira Paludo A, da Silva AGT, 9. Gettman MT, Peschel R, Neururer R, Bartsch G. A Ghissi AJ, Fernandes AS, et al. Laparoscopic pediatric comparison of laparoscopic pyeloplasty performed with pyeloplasty using the Flexdex® articulating needle driver: the daVinci robotic system versus standard laparoscopic step-by-step video. J Pediatr Urol. 2019; 15:421-2. techniques: initial clinical results. Eur Urol. 2002; 23. Behan JW, Kim SS, Dorey F, De Filippo RE, Chang AY, 42:453-7. Hardy BE, et al. Human capital gains associated with 10. Atug F, Woods M, Burgess SV, Castle EP, Thomas R. robotic assisted laparoscopic pyeloplasty in children Robotic assisted laparoscopic pyeloplasty in children. J compared to open pyeloplasty. J Urol. 2011; 186 (4 Urol. 2005; 174 (4 Pt 1): 1440-2. Suppl): 1663-7. 11. Casale P. Robotic pediatric urology. Expert Rev Med 24. Olsen LH, Rawashdeh YF, Jorgensen TM. Pediatric robot Devices. 2008; 5:59-64. assisted retroperitoneoscopic pyeloplasty: a 5-year 12. Passerotti C, Peters CA. Pediatric robotic-assisted experience. J Urol. 2007; 178:2137-41. laparoscopy: a description of the principle procedures. 25. Crivelli JJ, Johnson BA, Steinberg RL, Gahan JC, Antonelli ScientificWorldJournal. 2006; 6:2581-8. JA, Morey AF, et al. Clinical and radiographic outcomes 13. Fuchs ME, DaJusta DG. Robotics in Pediatric Urology. Int following salvage intervention for ureteropelvic junction Braz J Urol. 2020; 46:322-7. obstruction. Int Braz J Urol. 2021; 47:1209-18. 967 IBJU | OPEN, LAPAROSCOPIC AND ROBOTIC PYELOPLASTY 26. Piaggio LA, Corbetta JP, Weller S, Dingevan RA, Duran 29. Braga LH, Lorenzo AJ, Bägli DJ, Mahdi M, Salle JL, Khoury V, Ruiz J, et al. Comparative, Prospective, Case-Control AE, et al. Comparison of flank, dorsal lumbotomy and Study of Open versus Laparoscopic Pyeloplasty in laparoscopic approaches for dismembered pyeloplasty in Children with Ureteropelvic Junction Obstruction: Long- children older than 3 years with ureteropelvic junction term Results. Front Pediatr. 2017; 5:10. obstruction. J Urol. 2010; 183:306-11. 27. Valla JS, Breaud J, Griffin SJ, Sautot-Vial N, Beretta 30. Wang L, Qin W, Tian F, Zhang G, Yuan J, Wang H. Cytokine F, Guana R, et al. Retroperitoneoscopic vs open responses following laparoscopic or open pyeloplasty in dismembered pyeloplasty for ureteropelvic junction children. Surg Endosc. 2009; 23:544-9. obstruction in children. J Pediatr Urol. 2009; 5:368-73. 28. Vemulakonda VM, Cowan CA, Lendvay TS, Joyner _______________________ BD, Grady RW. Surgical management of congenital Correspondence address: ureteropelvic junction obstruction: a Pediatric Health Antonio Rebello Horta Gorgen, MD Information System database study. J Urol. 2008; 180 (4 Departamento de Urologia, Suppl): 1689-92. Grupo de Urologia Reconstrutiva e Infantil (GURI), Hospital de Clinicas de Porto Alegre Rua Ramiro Barcelos, 2350 Porto Alegre, RS, 90035-903, Brasil E-mail: antoniogorgen@gmail.com 968