Initial comparison of robotic-assisted laparoscopic versus open pyeloplasty in children

PEDIATRIC UROLOGY

INITIAL COMPARISON OF ROBOTIC-ASSISTED LAPAROSCOPIC VERSUS OPEN PYELOPLASTY IN CHILDREN DAVID S. YEE, ALLAN M. SHANBERG, BARRY P. DUEL, ESEQUIEL RODRIGUEZ, LOUIS EICHEL, AND DEEPAK RAJPOOT

ABSTRACT Objectives. To compare the initial results of robotic-assisted laparoscopic versus open pyeloplasty in children with ureteropelvic junction obstruction. Methods. From June 2002 to July 2004, 8 pediatric patients underwent robotic-assisted laparoscopic pyeloplasty and were matched by age group with patients undergoing conventional open pyeloplasty. The mean age was 11.5 years (range 6.4 to 16.5) in the robotic-assisted group and 9.8 years (range 6.0 to 15.6) in the open group. A four-port transperitoneal technique was used to perform the Anderson-Hynes pyeloplasty with the da Vinci Surgical System. Results. The mean operative time and estimated blood loss was 363 minutes (range 255 to 522) and 13.1 mL (range 5 to 25) in the robotic-assisted group versus 248 minutes (range 144 to 375) and 53.8 mL (range 5 to 200) in the open group, respectively. The mean length of hospitalization and pain medication use was 2.4 days (range 1 to 5) and 7.4 mg morphine (range 0 to 23) in the robotic-assisted group compared with 3.3 days (range 1 to 8) and 22.0 mg morphine (range 0 to 100) in the open group, respectively. At a mean follow-up of 14.7 months (range 2 to 24), all robotic procedures were successful as determined by subjective data using pain scales and radiologic data. Conclusions. Robotic-assisted laparoscopic pyeloplasty appears to decrease the length of hospitalization and use of pain medication, but has a longer operative time. Additional clinical experience is required to determine the long-term efficacy of this method. UROLOGY 67: 599–602, 2006. © 2006 Elsevier Inc.

L

aparoscopic pyeloplasty has recently been described as a feasible reconstructive procedure in children with ureteropelvic junction obstruction (UPJO). It was first reported by Peters et al.1 in 1995. The ureteropelvic anastomosis in laparoscopic pyeloplasty, however, is particularly timeconsuming and technically challenging. The introduction of the da Vinci Surgical System (Intuitive Surgical, Mountainview, Calif) has facilitated intracorporeal suturing and shortened the learning curve, as demonstrated in adult pyeloplasty series.2– 4 To our knowledge, only one pediatric series From the Department of Urology, Antoci Center for Pediatric Urology and Nephrology, University of California, Irvine, School of Medicine, Orange, California Reprint requests: David Yee, M.D., M.P.H., Department of Urology, University of California, Irvine, School of Medicine, 101 The City Drive, Building 26, Room 24, Route 81, Orange, CA 92868. E-mail: [email protected] Submitted: May 31, 2005, accepted (with revisions): September 15, 2005 © 2006 ELSEVIER INC. ALL RIGHTS RESERVED

has been reported using a retroperitoneal approach in robotic-assisted laparoscopic pyeloplasty.5 We describe the use of the da Vinci Surgical System in performing transperitoneal laparoscopic pyeloplasty in children and provide an initial comparison of robotic-assisted laparoscopic versus traditional open pyeloplasty. MATERIAL AND METHODS From June 2002 to July 2004, 8 pediatric patients underwent laparoscopic Anderson-Hynes pyeloplasty using the da Vinci system. Each patient undergoing robotic-assisted laparoscopic pyeloplasty was matched by age group to a patient undergoing conventional open pyeloplasty from July 1997 to October 2002. Patient data were taken from the University of California, Irvine, Medical Center and Long Beach Memorial Medical Center after receiving institutional review board approval. The mean age was 11.5 years (range 6.4 to 16.5) in the robotic-assisted group and 9.8 years (range 6.0 to 15.6) in the open group. Table I summarizes other demographic characteristics, including sex and UPJO side and type. Perioperative outcomes between the two groups were analyzed with 0090-4295/06/$32.00 doi:10.1016/j.urology.2005.09.021 599

TABLE I. Patient demographic data Demographic Variable Sex (n) Male Female Age (yr) Side (n) Right Left UPJO type (n) Primary Secondary

Robot (n ⴝ 8)

Open (n ⴝ 8)

7 1 11.5 (6.4–16.5)

4 4 9.8 (6.0–15.6)

4 4

3 5

7 1

8 0

KEY: UPJO ⫽ ureteropelvic junction obstruction. Data in parentheses are ranges.

FIGURE 2. Intracorporeal continuous suturing of anterior ureteropelvic anastomosis performed with 8-mm robotic arms.

FIGURE 1. Patient position and port placement for a right-sided laparoscopic pyeloplasty with da Vinci robotic system.

GraphPad software using paired t tests. The level of statistical significance was set at P ⬍0.05 (two-sided). The preoperative evaluation and diagnosis of UPJO was determined from the presenting symptoms and radiologic imaging findings (ultrasonography, intravenous urography, and/or diuretic renography). The da Vinci system is composed of two interactive robotic arms, a camera arm, a three-dimensional imaging system, and a remote control station. The base of the robot was placed just lateral to the operating room table on the side contralateral to the UPJO. The patient was initially placed in the lithotomy position for cystoscopy and retrograde pyelography through a 4F open-ended ureteral catheter. The catheter was then secured to a Foley catheter, and the patient was repositioned in a flank position. Robotic-assisted laparoscopic pyeloplasty was performed using a four-port transperitoneal technique. A non-bladed 12-mm umbilical port was placed, followed by two 5-mm ports 8 cm from the umbilicus in an equilateral triangle configuration (Fig. 1). A 5-mm lateral “assistant’s” port was also placed above the anterior superior iliac spine for retraction, suction, and introduction of suture. The initial dissection began with a standard laparoscopic technique to take the colon down at the line of Toldt. Gerota’s fascia was then incised, and the UPJO was dissected. A “hitch” stitch was occasionally used to stabilize the renal pelvis to the 600

FIGURE 3. Completed ureteropelvic anastomosis after closure of renal pelvis.

anterior abdominal wall, as described by Tan and Roberts.6 The UPJO was then transected, leaving a segment of pelvis attached to the ureter that could initially be used for traction and then discarded before the anastomosis. Next, the 8-mm robotic arms were docked at the previously placed 5-mm parallel ports. The ureter was spatulated on the posterolateral side, and the redundant renal pelvis was excised. The ureteropelvic anastomosis was performed with three interrupted 4-0 Vicryl sutures placed at the apex of the spatulated ureter to the most dependent portion of the renal pelvis. Then, 4-0 Vicryl running sutures were placed for the anterior and posterior anastomosis (Fig. 2). The renal pelvis was then closed with an additional running 4-0 Vicryl suture (Fig. 3). Before renal pelvis closure, a 4.7F double-pigtail catheter was inserted in a retrograde or antegrade fashion over a guidewire through the “assistant’s” port. After completing the anastomosis, indigo carmine and furosemide was administered to check for urine leakage. Finally, a Jackson-Pratt drain was brought out through the “assistant’s” port. Bupivacaine was injected at the port sites before closure. Conventional open pyeloplasty was performed using a modified flank incision from the tip of the 12th rib laterally and posteriorly for an average 6 to 7-cm incision. Each patient underwent Anderson-Hynes pyeloplasty using 6-0 UROLOGY 67 (3), 2006

TABLE II. Perioperative outcomes Outcome Parameter Mean operative time (min) Mean estimated EBL (mL) Mean length of hospitalization (days) Mean pain medication usage (mg MS) Complications (n) Mean follow-up (mo) Success* (%)

Robot (n ⴝ 8)

Open (n ⴝ 8)

P Value

363 (255–522) 13.1 (5–25) 2.4 (1–5) 7.4 (0–23) 1 14.7 (2–24) 100

248 (144–375) 58.8 (5–200) 3.3 (1–8) 22.0 (0–100) 0 53.2 (32–86) 86

0.03 0.17 0.47 0.31

KEY: EBL ⫽ estimated blood loss; MS ⫽ morphine equivalents. Data in parentheses are ranges. * According to pain score improvement and radiologic data as determined by diuretic renal scan and/or ultrasonography.

PDS suture and placement of a double-pigtail catheter, Jackson-Pratt drain, and Foley catheter. Bupivacaine was also injected into the incision to obtain both intercostal and local nerve blocks. Both groups received intravenous morphine or meperidine (1 mg morphine ⫽ 7.5 mg meperidine) as needed for pain; no epidural or patient-assisted analgesia was given. Each patient had the Foley catheter and Jackson-Pratt drain removed before discharge. The double-pigtail catheter was removed approximately 4 weeks postoperatively. At 3 months of follow-up, patients underwent diuretic renography and/or ultrasonography. The radiologic imaging findings and pain scale scores were used to determine success.

RESULTS The perioperative outcomes evaluated were operative time, estimated blood loss, hospital stay, pain medication use, complications, and success rates (Table II). Only the mean operative time was statistically significant. The mean operative time was 363 minutes (range 255 to 522) in the roboticassisted group versus 248 minutes (range 144 to 375) in the open group. No open conversions were required in the robotic series. The estimated blood loss was 13.1 mL (range 5 to 25) in the roboticassisted group versus 53.8 mL (range 5 to 200) in the open group. The mean length of hospitalization and pain medication use was 2.4 days (range 1 to 5) and 7.4 mg morphine (range 0 to 23) in the robotic-assisted group compared with 3.3 days (range 1 to 8) and 22.0 mg morphine (range 0 to 100) in the open group, respectively. No intraoperative complications occurred in either group. One patient in the robotic-assisted group, however, developed postoperative ileus that extended the patient’s hospital stay before being discharged home on postoperative day 5. At a mean follow-up of 14.7 months (range 2 to 24), all robotic procedures were successful. One failure occurred in the open group within the first 6 months of surgery due to uteropelvic junction stricture. The mean follow-up was 53.2 months in the open group. UROLOGY 67 (3), 2006

COMMENT Laparoscopic pyeloplasty in adults has proved to be an effective minimally invasive treatment for UPJO.7,8 Although pediatric laparoscopic pyeloplasty was first described in 1995, few reports of this operation successfully performed in children have been published.6,9 –12 Its application has been limited, despite refinements in laparoscopic instrumentation with needlescopic 3-mm instruments.13 The laparoscopic mentorship training model described by Farhat et al.14 confirmed that reconstructive laparoscopic procedures require training beyond 10 months of mentorship. The introduction of the da Vinci system to perform precise laparoscopic manipulations offers an opportunity to spread reconstructive laparoscopy among pediatric surgeons. Robotic-assisted laparoscopy appears to reduce the learning curve of intracorporeal suturing.2– 4 The da Vinci system offers a magnified three-dimensional view of the operating field, wrist-like movements of the instrument arms with 6° of freedom, and hand tremor elimination. Another advantage is the ergonomic working environment and surgeon comfort during these time-consuming operations. Previous studies have reported decreased operative times with robotic-assisted pyeloplasty compared with standard laparoscopic pyeloplasty, presumably from the improved efficiency in completing the ureteropelvic anastomosis.3,15 Our mean operative time was longer than that of published series of pediatric laparoscopic transperitoneal pyeloplasty, which reported an average operative time of 90 to 270 minutes.6,9,10 This result, however, was partially affected by 1 case in which a 2-hour delay occurred owing to technical problems with the da Vinci system. Excluding this case, the mean operative time was 340 minutes, and the difference was no longer statistically significant compared with the open group (P ⫽ 0.18). In addition, the robotic operative time also included cystoscopy, patient repositioning, and docking of the robot. Robot docking and fine intraoperative adjustments before 601

its use require about 20 minutes for trained staff. Furthermore, the robotic-assisted group had a learning curve and the open pyeloplasty group did not. Although the first case took 410 minutes, the last 2 cases were completed within 260 minutes. We found this encouraging and believe it can be a feasible operation as additional experience with the da Vinci robot is obtained. Laparoscopic pyeloplasty in children has been described with both a transperitoneal and a retroperitoneal approach.6,9 –12,16 –19 Our experience with conventional transperitoneal laparoscopic pyeloplasty was limited to 3 patients, with one failure requiring endopyelotomy. We found the learning curve to be longer for conventional laparoscopic surgery, as evidenced by our 67% success rate and longer mean operative time of 407 minutes (range 323 to 525). We have not had experience with the retroperitoneal approach and recognize that space is a limitation. Tan10 reported the first pediatric series of transperitoneal laparoscopic dismembered pyeloplasty and did not recommend the procedure for children younger than 6 months of age because of the technical difficulty associated with a small ureteral caliber. Yeung et al.,11 however, did not find age a deterring factor in their series of retroperitoneal laparoscopic dismembered pyeloplasty in children. In our current practice, children younger than 2 years old undergo pyeloplasty through a posterior lumbotomy incision. Given the use of 8-mm robotic instrument arm ports in our series, we do not recommend use of the da Vinci system in children younger than 6 years of age because of body size. The recent introduction of 5-mm robotic arm ports, however, may broaden its application to younger pediatric age groups. The “assistant’s” port is also not absolutely necessary, because suture may be introduced and cut through the robotic instrument ports. It does save time, however, and allows the assistant surgeon to suction and retract. A disadvantage of robotic-assisted laparoscopy was the loss of tactile sensation, as manifested by occasional suture breakage. This deficit required increased attention to the visual cues provided by the improved three-dimensional image control. To reduce suture breakage, 4-0 Vicryl suture was used instead of the 6-0 PDS suture used in the open pyeloplasty cases. Another disadvantage was the infrequent collision of the robotic arms that occurred because of the small patient size. Moreover, the expense of obtaining and operating the da Vinci system limits its use to centers at which highcase volumes can reduce its financial impact through efficiencies of scale.16 At our institution, robotic pyeloplasty costs an average of $5466 compared with $2410 for the open procedure. 602

CONCLUSIONS In this initial comparison, robotic-assisted laparoscopic pyeloplasty in children provided equal efficacy to that of the open procedure. The benefits of the approach appear to include decreased blood loss, length of hospitalization, and use of pain medication. Additional clinical experience and long-term follow-up is required to determine the true efficacy of this method. REFERENCES 1. Peters CA, Schlussel RN, and Retik AB: Pediatric laparoscopic dismembered pyeloplasty. J Urol 153: 1962–1965, 1995. 2. Gettman MT, Neururer R, Bartsch G, et al: AndersonHynes dismembered pyeloplasty performed using the da Vinci robotic system. Urology 60: 509 –513, 2002. 3. Gettman MT, Peschel R, Neururer R, et al: A comparison of laparoscopic pyeloplasty performed with the da Vinci robotic system versus standard laparoscopic techniques: initial clinical results. Eur Urol 42: 453– 457, 2002. 4. Bentas W, Wolfram M, Brautigam R, et al: Da Vinci robot assisted Anderson-Hynes dismembered pyeloplasty: technique and 1 year follow-up. World J Urol 21: 133–138, 2003. 5. Olsen LH, and Jorgensen TM: Computer assisted pyeloplasty in children: the retroperitoneal approach. J Urol 171: 2629 –2631, 2004. 6. Tan HL, and Roberts JP: Laparoscopic dismembered pyeloplasty in children: preliminary results. Br J Urol 77: 909 – 913, 1996. 7. Eden CG, Cahill D, and Allen JD: Laparoscopic dismembered pyeloplasty: 50 consecutive cases. BJU Int 88: 526 – 531, 2001. 8. Turk IA, Davis JW, Winkelmann B, et al: Laparoscopic dismembered pyeloplasty—the method of choice in the presence of an enlarged renal pelvis and crossing vessels. Eur Urol 42: 268 –275, 2002. 9. Schier F: Laparoscopic Anderson-Hynes pyeloplasty in children. Pediat Surg Int 13: 497–500, 1998. 10. Tan HL: Laparoscopic Anderson-Hynes dismembered pyeloplasty in children. J Urol 162: 1045–1047, 1999. 11. Yeung CK, Tam YH, Sihoe JD, et al: Retroperitoneoscopic dismembered pyeloplasty for pelvi-ureteric junction obstruction in infants and children. BJU Int 87: 509 –513, 2001. 12. El-Ghoneimi A, Farhat W, Bolduc S, et al: Laparoscopic dismembered pyeloplasty by a retroperitoneal approach in children. BJU Int 92: 104 –108, 2003. 13. Tan HL: Laparoscopic Anderson-Hynes dismembered pyeloplasty in children using needlescopic instrumentation. Urol Clin North Am 28: 43–51, 2001. 14. Farhat W, Khoury A, Bagli D, et al: Mentored retroperitoneal laparoscopic renal surgery in children: a safe approach to learning. BJU Int 92: 617– 620, 2003. 15. Olsen H, and Jorgensen T: Robotic vs. standard retroperitoneoscopic pyeloplasty in children. BJU Int 91: 73, 2003. 16. Peters CA: Robotically assisted paediatric pyeloplasty: cutting edge or expensive toy? BJU Int 94: 1214 –1215, 2004. 17. Peters CA: Laparoscopy in pediatric urology. Curr Opin Urol 14: 67–73, 2004. 18. El-Ghoneimi A: Laparoscopic management of hydronephrosis in children. World J Urol 22: 415– 417, 2004. 19. El-Ghoneimi A: Paediatric laparoscopic surgery. Curr Opin Urol 13: 329 –335, 2003. UROLOGY 67 (3), 2006