Comparison of Robotic Pyeloplasty and Standard Laparoscopic Pyeloplasty in Infants: A Bi-Institutional Study

Introduction

Ureteropelvic junction (UPJ) obstruction is one of the most common pediatric genitourinary anomalies. The major concern regarding this obstructive condition is the progressive deterioration of kidney function with the potential loss of the renal unit. In their original article from 1949, Anderson and Hynes ¹ described their open dismembered pyeloplasty technique for the treatment of ureteral obstruction, which later became the gold standard for the management of UPJ obstruction. In the last two decades, the minimally invasive approach, first described as laparoscopic pyeloplasty (LP)^2,3 and later on with robotic pyeloplasty (RP) has become the new standard of care after demonstrating similar success rates to the open approach.^4,5 In comparison to the laparoscopic approach, which requires a long learning curve and technical expertise for the performance of reconstructive surgery, the robotic approach has advantages such as three-dimensional visualization, enhanced dexterity, and surgeon comfort. In the pediatric population, RP has been associated with shorter hospital stays and decreased narcotic requirements at the expense of longer operative times in early reports.^5,6 The aim of this study was to compare and assess outcomes between RP and LP for the treatment of UPJ obstruction in the infant population.

Materials and Methods

Results

During the study period a total of 34 patients under the age of 1 year underwent unilateral minimally invasive dismembered pyeloplasty for UPJ obstruction. Patient characteristics are shown in Table 1. Thirteen patients underwent LP and 21 patients underwent RP, by 1 surgeon at each medical center. All operations were performed on an inpatient basis. There were no statistically significant demographic differences between the 2 groups. Median age and median weight at time of operation for the whole cohort were 6.1 months and 7.9 kg. There were no accessory incisions or conversions to open surgery in either group. There were no intraoperative complications or significant blood loss in either group. Patient operative data are summarized in Table 2. Of the cases, 100% (13/13) in the laparoscopic group and 71.4% (15/21) in the robotic group had a ureteral stent. There was no significant difference in operative times between the 2 groups, with a median time of 156 minutes for RP (range 125–249) and 192 minutes (range 98–229) for LP (P = .35). Median length of hospital stay was significantly shorter in the robotic group with a median stay of 1 day (range 1–3) and 7 days (range 7–12) for LP (P < .0001). Drains or nephrostomy tubes were used more often in the laparoscopic group (100%, 13/13) as opposed to RP (9.5% 2/21, P < .0001). All patients returned for follow-up visits. Median follow-up was 5.5 months (range 3.7–8.1) LP and 11.5 months (1.2–39.7) for RP (P < .0001). Ultrasonography was routinely performed to monitor postoperative hydronephrosis. All but 1 patient in each group had a successful outcome (32/34 patients) with improvement in the degree of hydronephrosis during follow-up. Two patients (1 in each group) required a reoperative pyeloplasty due to persistent hydronephrosis or abnormal diuretic renogram. Postoperative diuretic renal scans were not routinely performed. While the use of renal diuretic scan can provide important information on postoperative renal function, we use this tool according to the surgeon's clinical judgment and on a case appropriate decision. When there is clear improvement on follow-up ultrasound we do not recommend performing a diuretic scan routinely.

The complication rate was 30.8% (4/13 cases) for LP and 23.8% (5/21 cases) for RP (P = .65). We included in our review surgical complications according to the Clavien-Dindo grading and also nonsurgical complications. All complications were Clavien-Dindo grade I–II or nonsurgical complications, except for one case of postoperative urinary extravasation which resolved after an insertion of a nephrostomy tube in the robotic group. Complications for LP included a port site infection, two febrile UTIs, and a failure. Complications for RP included one temporary urinary leak requiring intervention, febrile UTI, postoperative ileus, postoperative umbilical hernia, and a failure.

Discussion

UPJ obstruction is a common pediatric genitourinary anomaly and is the leading cause of unresolved hydronephrosis in infants. In the past decade, there is a trend toward earlier reconstructive operation in grade 3 and 4 hydronephrosis as a result of kidney function deterioration concerns. ⁸ Although considered the gold standard for the treatment of UPJ obstruction, open dismembered pyeloplasty as introduced by Anderson and Hynes ¹ has been largely replaced in the past two decades by the minimally invasive approach. LP as was introduced by Peters et al. ³ in 1995 has shown comparable success rates to the open approach while providing early recovery, shorter hospital stay, and better cosmetic results.^4,9 Surgical instruments have improved and reduced in caliber, thus cosmetic results especially of the 3-mm ports are excellent. However, the laparoscopic approach has been shown to be a demanding and technically difficult procedure with a longer operating time and a lengthy learning curve. These issues stem, in part, due to the laparoscopic suturing technique and small operating space in children.^10,11 The emergence of the robotic surgery has shown advantages, including increased visualization due to three-dimensional viewing, increased precision with better articulation, and an increase in surgeon comfort. The demanding aspect of laparoscopic intracorporeal suturing has been made easier with the robotic approach.^12,13 However, issues such as operating cost, concern about the technical difficulty to utilize the robot in a small space, bigger trocars, dedicated team training, and potential for long setup time in the operating room still raise questions regarding the advantage in RP over LP in very young children. The issues of safety, efficacy, and operating time in the robotic approach have been addressed in earlier studies. Both studies by Lee et al. ⁵ and Yee et al., ⁶ which compared RP with the open approach, have shown similar efficacy and complication rate with a decreased hospital stay and narcotic use. However, both studies showed longer operating times in comparison to the open approach. Bansal et al. ¹⁴ compared their cohort of infants operated in the robotic approach (n = 9) and the open approach (n = 61) and showed advantages with shorter operative time, shorter hospital stay, and less narcotic utilization with RP.

A few studies have compared dismembered pyeloplasty between the robotic and standard laparoscopic approach in children. None of them focused on the infant population to the best of our knowledge. Most of these studies show advantages in utilization of RP. Riachy et al. ¹⁵ compared 18 patients (median age 8.1 years) who underwent standard laparoscopic with 46 patients (8.8 years) who underwent RP and showed similar outcomes in terms of success and complication rates with a shorter operating time in the robotic approach. In a recently published multicenter European study from 15 academic centers, ¹⁶ 185 patients under the age of 18 years who underwent RP were compared to 390 patients who were operated in the laparoscopic approach. Hospital stay and time for stenting were found to be shorter for RP (P < .05 for both). Success rates were similar between RP and LP (99.5% versus 97.3%, P = .11). The intraoperative complication rate was comparable between RP and LP (3.8% versus 7.4%, P = .06). However, the postoperative complication rate was significantly higher in LP (3.2% for RP and 7.7% for LP, P = .02).

The results of our study which examined infants undergoing minimally invasive pyeloplasty demonstrated the safety and efficacy of this treatment approach. Surgical success rates between RP and LP were comparable (95% and 92%, respectively) with no difference in the complication rate. This cohort compared patients that were operated in two different academic centers in two different countries and as such show differences in surgical planning, considerations, and hospitalization policies. There is a statistically significant difference in the use of ureteral stents between the groups. This is a point of interest since many factors influence surgeons preference to perform a stentless pyeloplasty. Although not measured in this study, we believe that the anastomosis achieved by RP is potentially more precise than with LP. Performing a stentless pyeloplasty facilitates the ability to perform pyeloplasty as day surgery. Another potential advantage of the stentless pyeloplasty is the avoidance of additional anesthesia and procedure to extract the stent and potentially prevent stent complications with a reduction in the need for home pain management. While the use of drains or nephrostomy tubes and median hospital stay were significantly different between the 2 groups, it is our belief that these data reflect surgeon's preferences and hospital medical care policy. Since utilization of the robotic system demands technical knowledge for patient positioning and robot docking, a dedicated team with experience can shorten operating room time.

Additional aspects of minimally invasive surgery in infants should be taken into consideration. The issue of increased heat loss in infants in comparison to older children ¹⁷ should be kept in mind and addressed appropriately during surgery. Increased room temperature and an insufflation warmer can decrease heat lost and help avoid hypothermia. A fluctuation in blood pressure and difficulty in maintaining normal oxygenation during mechanical ventilation could arise from increased intra-abdominal pressure from pneumoperitoneum.^18,19 Intra-abdominal target pressure should be balanced for effective visualization and maneuverability with minimal cardiopulmonary effect. The renal effects of increased intra-abdominal pressure have also been documented and could potentially cause oliguria in the infant patient. ²⁰ In our study cohort, we did not encounter any complications associated with anesthesia, patient positioning, port placement, or pneumoperitoneum, but adjustment of the surgical plan and appropriate measures should be taken to avoid unnecessary morbidity.

The limitations of this study include its retrospective nature, small cohort size, and a comparison between two academic centers with a different health system policy. Another limitation is a relative short follow-up period which could potentially impede assessment of late surgery failure. To the best of our knowledge, this is the first comparison between RP and LP in the infant population. Further study is needed to better define the role of minimally invasive approaches in infants.

Conclusions

Minimally invasive dismembered pyeloplasty with RP and LP is safe and effective in infants with comparable success, complication rates, and operative time.