Robotic-Assisted Laparoscopic Pyeloplasty in Challenging Cases of Ureteropelvic Junction Obstruction in the Pediatric Population: A Multicenter Review

Introduction

Ureteropelvic junction obstruction (UPJO) is one of the most common congenital obstructive anomalies of the upper urinary tract. ¹ Open dismembered pyeloplasty, originally described by Anderson and Hynes, has historically been the treatment of choice with success rates higher than 90%. While the open technique is practiced worldwide, laparoscopic and now robotic techniques first described in 1999 are gaining popularity, due to the advantages of the minimally invasive surgery that these offer, including early ambulation, reduced postoperative pain, short hospital stay, and better cosmetic outcome. ² Robotic technique enables magnified three-dimensional and stable vision, precise tremor-free movements, and better maneuverability. The technical evolution of new models has allowed the performance of robotic procedures in small patients, including infants.^3–5 Currently, robotic-assisted laparoscopic pyeloplasty (RALP) is well established for standard cases, showing good outcomes. Recently it has been shown that RALP is feasible also in cases of complex clinical scenario and anatomical malformation.^6–10

In this report, we describe a multicenter review of outcomes and complications after RALP in pediatric patients with complex UPJO.

Materials and Methods

This is a multi-institutional retrospective study, including pediatric RALP performed in complex cases between 2013 and 2021. Institutional review board approval was obtained before this study. All surgical procedures were performed by 3 senior surgeons or by their supervision. The surgeons performed the robotic procedures with the same robotic team, represented by assistant surgeon and nurses, in each institution. RALP was defined complex in the following cases of UPJO: Anatomic variations, including ectopic, malrotated, horseshoe or duplex kidney, recurrent UPJO after failed open pyeloplasty, huge hydronephrosis defines as one crossing midline abdomen, low weight infants (<6 kg), and bilateral cases. Preoperative evaluation included a detailed history, physical examination, urine analysis, culture, renal ultrasound (US), and radionuclide diuretic renal scan mercaptuacetyltriglycine (MAG3).

Indications for surgery included the following: Progressive hydronephrosis on serial US; loss of split renal function <45% on initial or serial MAG3 scans; and worsening/persistence of the obstructed drainage on serial MAG3 scans and symptoms such as recurrent flank pain or urinary tract infections (UTIs).

All procedures were performed by a transperitoneal approach and with the aid of a Da Vinci platform (Si or Xi; Intuitive Surgical Systems). In selected left sided cases, a transmesenteric approach was utilized when the renal pelvis was visualized through the mesentery. The patients are secured in flank position and a urinary catheter is inserted. The contralateral arm is flexed and secured to a padded arm board, while the ipsilateral arm is allowed to rest on the patient's side. The patient is then secured to the table using The Pink Pad (Xodus Medical, USA), folded blue towels, and tape across the chest and hip. Great care is taken to pad all pressure points and protect the child against iatrogenic injury from the robot or working arms.

With an open Hasson technique, a 12-mm camera port of SI and 8 mm of XI is secured to the umbilicus, and two 8-mm working ports are placed on the midline, equidistant from the camera port. We have used an additional 5-mm port placed on the contralateral side of the operated kidney to deliver and cut the sutures during anastomosis and shorten the overall operating time. To increase the operating space, we have employed a burping technique of port placement pulling the abdominal wall with the robotic trocars up, while docking the robotic arms. Following port placement and docking of the Da Vinci Si or Xi, the procedure commences by identifying the dilated renal pelvis.

During dissection, the surgeon uses a bipolar microdissector forceps in the left hand and monopolar scissors in the right. For both sides, the colon is mobilized by incising the peritoneum along the white line and deflecting the colon medially. Next, the proximal ureter and renal pelvis are carefully dissected, while taking care to preserve the ureteral blood supply. If an aberrant vessel is identified, it is dissected free from the UPJ/upper ureteral area. At this stage, a trans-flank holding stitch through the renal pelvis is placed to facilitate pelvic dissection and to stabilize the renal pelvis during intracorporeal anastomosis. The ureter is dismembered proximal to the UPJ, at the level of the renal pelvis. At this point, the ureter is completely detached from the renal pelvis, which is now open. Stones if present were retrieved through an accessory 12-mm trocar.

Through this trocar, it is easy to introduce a 14 Fr flexible cystoscope or ureteroscope and to perform a good inspection of the collecting system, including lower pole calyces, and to proceed with holmium or thulium laser lithotripsy when required, followed by extraction of all stones fragments. The excess tissue of the renal pelvis is used for grasping and manipulating the ureter. The ureter is then spatulated utilizing either monopolar scissors or Potts scissors depending on the surgeon preference and the diameter of the ureter. Laparoscopic anastomosis is done in an identical way to that during open surgery. If a large renal pelvis was encountered, pelvic reduction was performed. The posterior wall of the pelvic ureteral anastomosis is performed using 6/0 Maxon (Medtronic) interrupted sutures with the help of a Black Diamond Needle holder.

We fill that when utilizing interrupted sutures, we could reach a straight line of the anastomosis, further helping us to hold better and probably to decrease urinary leaks. Then, the posterior wall of the anastomosis is completed and the 12/3F JJ ureteral stent is inserted in an antegrade manner over a hydrophilic guidewire through angiographic catheter G14, which is introduced percutaneously, or through an additional 5-mm port. We have utilized the formula (age +10) for choosing the stent length. To assess correct stent placement, the bladder is filled with indigo carmine-stained fluid using a retrograde route through the placed catheter. Presence of the indigo carmine-stained urine from the ureter or stent indicates the correct position of the distal part of the stent in the urinary bladder.

The anterior wall anastomosis is completed following positioning of the proximal end of JJ stent inside the renal pelvis. The use of excess renal pelvic tissue facilitates manipulation of the ureter and avoids injury to the ureteric tissue in the anastomotic area. Following anastomosis completion, the excess renal pelvic tissue is excised and the renal pelvis is closed using running 6/0 Maxon suture. No drain is left alongside the anastomosis. Leaving the camera port in place after removing the remaining ports allows the fascia of each port to be closed under direct vision, thereby minimizing the risk of damage to bowel or postoperative hernias. Infiltration with local anesthesia (0.5–1 mg/kg of levo/bupivacaine) was done at the access sites, to better control postoperative pain.

Data were retrieved from medical records and included the following: operative time, blood loss, stone clearance, conversion rate, and hospital stay. The stent was removed after 4–6 weeks. Postoperative follow-up included US at 1, 3, 6, and 12 months. MAG 3 renal scan was performed when US showed no improvement in hydronephrosis or when clinical symptoms returned. Improvement of hydronephrosis, relief of obstruction on diuretic MAG3 scan, and resolution of symptoms were considered a surgical success. Postoperative complications were classified according to the Clavien–Dindo grading (CDG) system. Categorical variables were analyzed using the Fisher's exact test or chi-square test. P-value of <.05 was considered statistically significant. All statistical analysis was performed on commercially available software Graph Pad Prism 6.01 (Graph Pad prism, Prism 6 for Windows, version 6).

Results

Patients' demographics and outcome parameters are presented in Tables 1 and 2, respectively. Over an 8-year period between 2013 and 2021, a total of 700 pediatric patients underwent RALP in 3 medical centers; of them, 36 (0.05%) children (62% males and 38% females) were defined as complex cases and constitute the study group. Fifteen (41%) patients underwent RALP for recurrent UPJO after failed open pyeloplasty; among them, 2 children required simultaneous surgery for renal stones. There were 3 (8%) cases of lower pole UPJO in double collecting system, 3 (8%) cases of pelvic and horseshoe kidneys, 10 (27%) cases of infants weighing <6 kg, 3 (8%) cases of huge hydronephrosis crossing abdominal midline, and 2 (5%) cases of bilateral UPJO, which were performed in the same setting. The UPJO was left sided in 25/36 (69%) and right sided in 11/36 (31%) children. Crossing vessels were identified in 4/36 children (11%).

Anastomosis in all cases was performed anterior to the vessel. In 4 cases, the orthotopic UPJ was severely scarred or in a position causing kinking or pressure (3 high insertion ureters and 1 posterior insertion); therefore, the anastomosis was re-sited to a healthy dependent aspect of the pelvis. Dismembered Anderson–Hynes pyeloplasty was performed in all patients. The median age and weight were 36 months (interquartile range [IQR] 14–84) and 12 kg (IQR 10–20.5), respectively. All robotic cases were completed successfully, with no conversion to an open procedure. No intraoperative complication occurred. The median operative time (including docking and console time) was 120 minutes (IQR 90–135). A DJ stent and urinary catheter were inserted in all cases. The mean length of hospital stay was 2.6 days. Urinary catheter was removed on postoperative day 1.

The DJ stent was removed 4–6 weeks postoperatively under general anesthesia. Four patients (11%) had postoperative complications—1 ileus (CDG I) and 3 UTIs (CDG II). All complications were treated conservatively. No CDG III or higher complication was encountered. There was no case of port site infection. At a median follow-up of 3 years (IQR 1.5–3.3), the success rate was 95%, with those patients showing no symptom and radiographic evidence of normal drainage on US and MAG 3 when performed. One patient undergoing re-do RALP after failed open surgery for huge hydronephrosis required another re-do procedure due to recurrent intrinsic obstruction identified on follow-up imaging. The re-do procedure was performed utilizing the robotic platform.

Discussion

One of the most common surgical procedures in pediatric urology armamentarium is Anderson–Hynes dismembered pyeloplasty, which was classically utilized through an open surgical approach. ¹¹ This surgery has usually enjoyed high success rate of ∼95%–98%. In the last decade laparoscopic- and robotic-assisted surgery was suggested as an alternative procedure for the open technique, offering at the same time the benefits of minimally invasive surgery, such as smaller incision, a decrease in the length of hospitalization, and decrease in the need for narcotics, while maintaining almost the same success rate. One of us recently published data on 100 children who underwent RALP, showing results similar to those of an open surgery. ⁷ However, the majority of patients in this study were with normal kidney anatomy and operated by a senior surgeon.

The literature on utilizing robotic-assisted approach on challenging cases of UPJO is very scant. One of the reasons for that is the low number of challenging cases in single surgeon practice. In this study, we present a multi-institutional experience with difficult cases of UPJO. Our results add to a growing body of evidence that supports RALP as the primary approach to correct UPJO when indicated although the complexity of the pathology, a step forward for the implication of robotic surgery as an ultimate solution in this population. The overall success rate of the surgery in our study was similar to primary cases. Even with the most challenging cases such as recurrent UPJO after previous correction, we have reached high success with minimal complications. The re-do pyeloplasty is one the most challenging procedures in the pediatric urology/surgery spectrum.

In the large multi-institutional case series, Esposito et al. demonstrated the success rate of 100% (13/13) in children who had Redo RALP. ¹² Also, in our series, the overall success in these patients was similar to the primary cases. The utilization of the transperitoneal approach allows avoiding the area of extensive scarring after open retroperitoneal surgery. Tenfold robotic magnification permits release of the ureter from the scarred tissue, ureteral dissection in both ends, and finally performing a tension free anastomosis. Nowadays the majority of patients is diagnosed with UPJO following antenatal diagnosis of hydronephrosis; therefore, the pediatric urologist should proceed with surgical decision making already at the early stages of the child's life. During initial cases of robotic RALP, majority of surgeons did not recommend to proceed with the robotic approach in children weighing <6 kg.

In this series, all three centers adopted robotic approach in all children, disregarding the age and weight. Twenty-seven percent of all children in this series were with the weight of 6 kg and less, and 8% were with huge hydronephrosis, which crossed the midline. All completed RALP and no conversion was needed. We have to point out that all senior doctors in this study (S.C., B.C., and A.N.) have had an extensive experience with laparoscopic surgery even in small infants; therefore, the initiation of the robotic approach was significantly easier for us. We do believe that it should not be a routine practice and only experienced pediatric urologists may proceed with these cases.

As it was suggested, we utilize a “burping” technique of the trocar placement in these patients, while pulling the trocar with the abdominal wall during docking, which helps to increase operative space and allows easier manipulation of instruments during surgery. We also recommend meticulous inspection of the mesentery of the descending colon on the left side before mobilization of the colon, following opening of the Toldt line and entrance to the retroperitoneum. In the majority of small children, hydronephrotic pelvis might be reached through a small mesenteric window and the use of the black diamond robotic needle holder originally designed for cardiac surgery allows to perform watertight anastomosis utilizing 6/0 monofilament thread without jeopardizing principles of the open surgery.

Open dismembered pyeloplasty enjoys the high success rate not only in patients with UPJO in normal kidney anatomy but also in different anatomical variants such as duplex kidney, horse kidney, or in variants of renal positioning. As we have shown in this study, utilization of the robotic approach allows to achieve the same outcome as after open correction. A well-known fact is that the robotic instruments allow surgical correction in small cavities, and in the zone of interest, their surgical manipulation is difficult, also during open surgery. Robot employment allows meticulous kidney dissection, while preserving ureteral blood supply and possibly hilum dissection.

In some cases, engagement of the fourth arm is essential to ease kidney and UPJ dissection, and further a tension-free anastomosis. In some case of intrarenal pelvis, the use of fourth arm allows to turn the kidney over and insure full dissection of renal pelvis. This maneuver could be also very useful in cases of re-do surgery. It is a well-known fact that there is an increase in the incidence of pediatric urolithiasis. ¹³ In some cases, we have to deal with stones secondary to urinary stasis due to UPJO and in other cases, patients have a simultaneous disease of UPJO and urolithiasis due to a metabolic disorder. Open surgery provides direct, although not always easy access to all calyces of the collecting system. Minimal invasive surgery utilizing laparoscopic approach, requires certain manual skills to extract or proceed with lithotripsy during the operation.

We have found it particularly easy to do so with the robotic approach. One of the advantages of robotic surgery is a relatively easy learning curve in the well-organized academic setting.^14,15 Also, in our study, the junior doctors followed the steps of the senior staff relatively quickly; as a matter of fact, at least 40% of all surgeries were performed by the junior staff independently. In some cases, the senior staff scrubbed in and assisted bed side. In other cases, the senior physician overlooked the procedure and gave his input when needed.

There are a number of limitations to this study, which should be mentioned. The retrospective nature of the study exposes the study to potential bias. The cohort is heterogeneous; however, due to the rarity of such complex cases, we have collaborated with different institutions and focused on including as many cases as possible for better data interpretation. The distribution between the three medical centers was equal regarding the number of cases. We did not find any difference in the results when comparing between the three institutions participating in the study This supports our conclusion regarding feasibility of the robotic surgery in this challenging group of patients.

Conclusions

Our data show that RALP is feasible, reproducible, and safe with good outcomes and low complication rate in complex cases of UPJO in the pediatric population. The robotic platform allows adaptivity to patients' anatomies, obtaining surgical success in small weight infants and providing surgically technical advantages, which allowed us to overcome obstacles encountered in these difficult and challenging cases.