- Email: [email protected]
RETROPERITONEAL LAPAROSCOPIC VERSUS OPEN PYELOPLASTY IN CHILDREN
0022-5347/05/1735-1710/0 THE JOURNAL OF UROLOGY® Copyright © 2005 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 173, 1710 –1713, May 2005 Printed in U.S.A.
DOI: 10.1097/01.ju.0000154169.74458.32
RETROPERITONEAL LAPAROSCOPIC VERSUS OPEN PYELOPLASTY IN CHILDREN ARNAUD BONNARD, VIRGINIE FOUQUET, ELISABETH CARRICABURU, YVES AIGRAIN AND ALAA EL-GHONEIMI* From the Department of Pediatric Surgery and Urology, Robert Debre´ Hospital, AP-HP, University of Paris VII, Paris, France
ABSTRACT
Purpose: The indications for laparoscopy in pediatric urology are expanding and yet the advantages over open surgery remain unclear. We compared the results of retroperitoneal laparoscopic vs open pyeloplasty for pyeloureteral junction obstruction in children. Materials and Methods: A total of 22 children with a mean age of 88 months (range 25 to 192) underwent laparoscopic dismembered pyeloplasty via the retroperitoneal approach. An additional 17 children with a mean age of 103 months (range 37 to 206) underwent similar procedures via open surgery through a flank incision. We retrospectively analyzed and compared operative time, the use of analgesics (acetaminophen or morphine derivatives) and hospital stay. Results: The 2 groups were similar in mean age and weight at surgery. Mean operative time was significantly shorter in the open surgery vs the laparoscopy group (96 minutes, range 50 to 150 vs 219, range 140 to 310, p ⬍0.0001). Mean postoperative use of acetaminophen (1.9 vs 3.22 days, p ⫽ 0.03) and morphine derivatives (1.9 vs 3.06 days, p not significant) was less in the laparoscopy group. Mean hospital stay was shorter in the laparoscopy group than in the open surgery group (2.4 days, range 1 to 5 vs 5, range 3 to 7, p ⬍0.0001). Mean followup was 21 (range 12 to 51) and 24 months (range 12 to 60) in the open and laparoscopy groups, respectively. Conclusions: The operative time of laparoscopic pyeloplasty remains significantly longer than that of the open procedure in children. The main advantage of the laparoscopic approach is that it significantly decreases hospital stay compared with that after an open procedure. Although in our study analgesic use was less after laparoscopy, our results should be confirmed by a prospective, randomized study. KEY WORDS: laparoscopy, kidney, retroperitoneal space
Open dismembered pyeloplasty via a retroperitoneal approach remains the reference standard for correcting ureteropelvic (UPJ) obstruction in children. Although laparoscopic pyeloplasty is well described in adults,1, 2 experience in children remains limited.3, 4 We have already reported that the laparoscopic approach to partial and total nephrectomies significantly decreased the hospital stay compared with that of an open procedure.5, 6 In adults comparative studies show that the laparoscopic approach has the same success rate as open surgery but morbidity and complications are significantly decreased.7 Because we have already observed the feasibility and efficiency of retroperitoneal laparoscopic pyeloplasty in children,8 it seemed necessary to us to compare the results of the laparoscopic approach with those of open surgery. To our knowledge such a study has not been yet done in children. PATIENTS AND METHODS
Between 1999 and 2003 laparoscopic retroperitoneal dismembered pyeloplasty was performed in 22 children with a median age at surgery of 88 months (range 25 to 192). Between 1998 and 2001 an additional 17 consecutive children underwent a similar procedure by conventional open surgery through a flank incision, including a bilateral type done at 2 sessions after a delay of 2 months in 1. Median age at surgery was 103 months (range 37 to 206). Because in our current practice we limit the
laparoscopic approach to children older than 2 years, we excluded from our study children younger than 2 years who underwent open surgery. The indications in the 2 groups were repeat urinary tract infections, hydronephrosis on prenatal diagnosis with postnatal followup showing progressively increasing hydronephrosis on ultrasonography and signs of obstruction on renal scan or abdominal pain related to UPJ obstruction. Evaluation before surgery included renal ultrasonography, voiding cystourethrography, renal scintigraphy (mercaptoacetyltriglycine or dimercapto-succinic acid) and in few cases excretory urography. The retroperitoneal approach was performed as previously described.8 Briefly, the child is placed lateral and 3 trocars (1, 5 mm and 2, 3 mm) are inserted (fig. 1). The kidney is approached posteriorly and the renal pelvis is identified first. The UPJ is identified and minimal dissection is used to free the UPJ from connective tissue. Small vessels are divided after bipolar electrocoagulation. A stay suture of 5-zero polydioxanone is placed at the UPJ for traction and stabilization. The ureteropelvic anastomosis is performed using 6-zero polydioxanone sutures and a tapered 3/8 circular needle. When the UPJ is obstructed by aberrant crossing vessel, dismembered pyeloplasty enables ureteral transposition. A 4.7Fr polyurethane Double-J stent (Medical Engineering Corp., New York, New York) was routinely inserted and remained indwelling for 4 to 6 weeks. In the last 2 cases in this study we inserted a pyelostomy transanastomotic stent, which was removed at the outpatient clinic without anesthesia on postoperative day 10. Perirenal suction drainage was used in the beginning but in the last 8 cases no perirenal
Submitted for publication August 22, 2004. * Correspondence: Department of Pediatric Surgery and Urology, Hoˆpital Robert Debre´, 48, boulevard Se´rurier, 75019 Paris, France (telephone: (33) 1 40 03 2159; FAX: (33) 1 4003 4762; e-mail: [email protected]). 1710
1711
RETROPERITONEAL LAPAROSCOPIC VERSUS OPEN PYELOPLASTY IN CHILDREN
groups for operative time, operative incidents, hospital stay, and postoperative use of analgesic and complications. Results were compared using the Mann-Whitney test and they were considered significant at p ⬍0.05. All patients were assessed clinically and by renal ultrasound 1 month after surgery and every 6 months thereafter. Furosemide renal scintigraphy or excretory urography was done only if there was no decrease in hydronephrosis on ultrasonography or if symptoms recurred. A good result was considered to be a decrease in hydronephrosis on imaging and the resolution of preoperative symptoms.
RESULTS
FIG. 1. Trocar placement for left retroperitoneal laparoscopic pyeloplasty. Child is placed lateral and retroperitoneal access is achieved through first trocar (1) incision 10 to 15 mm long and 1 cm from lower border of 12th rib tip. Second 3 mm trocar (2) is inserted posterior near costovertebral angle, while third 3 mm trocar (3) is inserted 1 cm above top of iliac crest at anterior axillary line. In this case transanastomotic pyelostomy stent was used.
drainage was placed. A Foley catheter remained in situ in all patients for 24 hours after surgery. In the group of children who underwent open surgery the same procedure was performed via an anterolateral, extraperitoneal flank approach. A nephrostomy stent was placed for drainage and perirenal suction drainage was also placed depending on operator convenience. Prophylactic antibiotics (preoperative single dose of third-generation cephalosporin) were routinely prescribed in each group. Pain management was done according to our current pain control protocol. A specialized nurse using a pain score adapted to patient age assessed postoperative pain every 3 hours. The Objective Pain Scale (OPS) was used in patients younger than 6 years. It is based on the evaluation of 5 items, namely blood pressure variation, crying, movements, touch and leg position, and verbalization, with a maximum score of 10. When the OPS score was greater than 4, the child was given analgesics intravenously. A visual analog scale (VAS) graded from 0 to 100 (minimum to maximum individual pain estimation) was used in children 6 years or older. The lower limit for analgesic administration was 40. If the pain score was above the limit with the OPS or the VAS, patients received 15 mg/kg acetaminophen intravenously, limited to 60 mg/kg daily. Alternatively the patient received 0.2 mg/kg nalbuphine (morphine derivative) per dose every 3 hours if pain persisted. Charts were reviewed retrospectively to compare the 2
Two patients (9%) in the laparoscopy group who required conversion to open surgery early in the experience were excluded from study. The table lists the results in the 2 groups. Mean operative time was 95.6 (range 50 to 150) and 219 minutes (range 140 to 310) in open and laparoscopy groups, respectively (p ⬍0.0001). The presence of aberrant crossing vessel did not change mean operative time. The mean postoperative number of days of acetaminophen use was significantly different between the laparoscopic and open surgery groups (1.9 vs 3.22, respectively, p ⫽ 0.03). Mean OPS and VAS scores were less in the laparoscopic group on days 1 and 2 than in the open surgery group but the difference was without statistical significance. One child in the laparoscopic group and 3 in the open surgery group had a postoperative febrile urinary tract infection. Two children (10%) in the laparoscopic group had postoperative leakage. One child did not initially have a stent and leakage was managed by a Double-J stent. The other child was treated with bladder drainage with a Foley catheter. Hospital stay in these children was 9 and 5 days, respectively. Two children (11%) in the open surgery group had postoperative leakage and a Double-J stent was placed. Hospital stay in these children was 11 and 18 days, respectively. All of these children were excluded from hospital stay analysis. Mean hospital stay was 2.4 (range 1 to 5) and 5 days (range 3 to 7) in the laparoscopy and open surgery groups, respectively (p ⬍0.0001). Eight patients were discharged home 2 days after the laparoscopic procedure and 2 were discharged home the day after the procedure. Mean followup was 21 (range 12 to 51) and 24 months (range 12 to 60) in the open and laparoscopy groups, respectively. The failure rate in the laparoscopy group was 4%. A 13-year-old child had increasing hydronephrosis after laparoscopic pyeloplasty. Early reoperation was elected because of increasing dilatation of the pelvis and calices, although a Double-J stent was in situ. We found an inflammatory granuloma on the suture line obstructing the anastomosis and stent.
Results in laparoscopic retroperitoneal and open surgery pyeloplasty groups No. pts Mean age (mos) Mean wt (kg) Mean operative time (mins) Mean postop acetaminophen use (days) Mean postop nalbuphin use (days) Objective pain scale: Day 1 Day 2 Visual analog scale: Day 1 Day 2 Mean hospital stay (days)
Laparoscopy
Open Surgery
p Value
20 88 25.4 218.6 1.85 1.85
17 103 30.4 95.6 3.22 3.06
Not significant Not significant ⬍0.001 0.03 Not significant
0.66 0.41
1.36 0.94
Not significant Not significant
26.6 21 2.9
34.5 33 5.4
Not significant Not significant ⬍0.001
1712
RETROPERITONEAL LAPAROSCOPIC VERSUS OPEN PYELOPLASTY IN CHILDREN DISCUSSION
Open pyeloplasty remained the standard treatment in adult and pediatric patients until the mid 1980s, when morbidity associated with the flank incision led urologists to explore less invasive alternatives. In the 1990s the endourological management of UPJ obstruction became the treatment of choice in the adult population. Although postoperative morbidity is significantly less after the endoscopic procedures, their success rate does not exceed 80%.9 Laparoscopic pyeloplasty was introduced in adults in 1993.10, 11 In the initial report of 5 cases operative time was 3 to 7 hours. The procedure has gained in popularity and more recent series have shown a success rate of greater than 95%.1 Laparoscopic pyeloplasty in children has followed the same evolution as nephrectomy. The procedure was first described through a transperitoneal approach. Tan reported the first pediatric series of transperitoneal laparoscopic dismembered pyeloplasty in 18 children 3 months to 15 years old.12 Mean operative time was 89 minutes. Two patients with persistent obstruction underwent repeat laparoscopic pyeloplasty. There is still controversy concerning which approach to chose, that is transperitoneal or retroperitoneal. Arguments to advocate 1 approach are more theoretical than true objective criteria. The longer time needed for the retroperitoneal approach is probably related to the limited working space, which makes suturing more difficult. In our experience we started our renal laparoscopic surgery using the retroperitoneal approach and we have naturally extended the indications to pyeloplasty without changing our habits. The transperitoneal approach theoretically increases the risk of abdominal organ injury. It requires more dissection to reach the kidney and the colon must be reflected. We believe that the golden standard of pediatric open renal surgery is the retroperitoneal approach and minimal invasive surgery should follow the same rules. We do not believe that starting experience with pyeloplasty through the transperitoneal approach would facilitate the procedure. The only measure that may facilitate the procedure is to respect progressive experience and learn perfectly the 2 approaches. Although urine leakage has not yet been reported, this complication would be better tolerated in the retroperitoneal space than in the intraperitoneal cavity. In our practice a horseshoe kidney is the only anatomical indication for the transperitoneal approach, which we have used in 4 cases. Yeung et al reported their initial experience with retroperitoneal dismembered pyeloplasty in 13 patients, of whom 1 required open conversion.4 Mean operative time was 143 minutes (range 103 to 235). The longer time needed for the retroperitoneal approach is probably related to the limited working space, which renders suturing more difficult. Our experience with retroperitoneal pyeloplasty since 1999 is now approaching 50 children (fig. 2). All of these cases were done by one of us (AE-G) but some were done elsewhere. The cases included in our comparative study were restricted to our center to avoid any bias concerning pain management and hospital stay. We have already reported our initial experience and midterm results in a group of 21 children and confirmed that dismembered retroperitoneal laparoscopic pyeloplasty is a safe and feasible approach in children.8 In our current study the learning curve was significant, in that mean operative time was decreased from 4 to 3 hours. Two patients in the laparoscopy group required conversion to open surgery at the beginning of our experience and they were excluded from study. Conversion in the beginning of our experience was due to difficult suturing and we did not need to convert any of the last 35 cases of our global experience with pyeloplasty (fig. 2). In our current practice children younger than 2 years undergo a posterior lumbar incision to avoid the technical difficulties of suturing in young children
FIG. 2. Retroperitoneal laparoscopic pyeloplasty learning curve, representing global experience with retroperitoneal laparoscopic experience. Total of 50 cases were consecutive from 1999 to 2004. Each point indicates 1 child but 4 (8%) needing conversion to open surgery are not shown. 1 redo, 1 patient (2%) required reoperation.
via the retroperitoneal approach. In infants the postoperative course is usually uneventful and the cosmetic appearance of the posterior approach is excellent compared with that of the anterior or lateral lumbar incision needed in older children. Soulie et al compared retroperitoneal laparoscopic vs open pyeloplasty with a minimal incision in 53 consecutive, nonrandomized adults.7 Mean operative time was similar in the 2 groups (165 and 145 minutes, respectively). The incidence of complications, hospital stay and functional results were equivalent in the 2 groups but the return to painless activity was more rapid with laparoscopy in younger patients. Bauer et al did a similar study, which showed no difference in postoperative outcomes between laparoscopic and open pyeloplasty.13 Klingler et al observed in adults that dismembered laparoscopic pyeloplasty had the same success rate as open surgery but morbidity and complications were significantly decreased.14 In contrast to the adult population, no comparative study has yet been done in pediatric patients.15 The current technique was modified many times during our experience. First we used an interrupted 5-zero suture for the anastomosis but with the development of 3 mm instruments, which made it possible to use 6-zero suturing, we were confident to reproduce the principles of open pyeloplasty and use the usual 6-zero running suture. Since then, we have not used a suction drain around the anastomosis. Another change was that 3 ports were used and not 4, principally to decrease the number of instruments used in the limited retroperitoneal space. Leaving a Double-J stent is a major concern in children because repeat general anesthesia is necessary for its removal. For this reason we currently use a pyelostomy stent to intubate the anastomosis. The stent is clamped the day after surgery and removed at the outpatient clinic without anesthesia on postoperative day 10. The advantage of the laparoscopic approach over the open procedure is the decreased the postoperative pain and hospital stay but a prospective, randomized study is mandatory to confirm it. Although our current study was retrospective, the same pain management protocol was followed in all children and treatment by pain scores was strictly documented in the charts. However, these results must be confirmed by a prospective study. Cosmetically it seems to us that with the 3 mm ports the results are better than open surgery scars. One of our main concerns is about teaching laparoscopic pyeloplasty to surgeons in training or qualified surgeons but with recent experience with advanced laparoscopic surgery. In our study all cases of laparoscopy was done by the same surgeon except 1, in which the same surgeon assisted an-
RETROPERITONEAL LAPAROSCOPIC VERSUS OPEN PYELOPLASTY IN CHILDREN
other qualified surgeon experienced with laparoscopic procedures. In contrast to ablative surgery, pyeloplasty remains a difficult laparoscopic procedure to teach.16 Pyeloplasty should be the best demonstration of the advantage of robot assisted surgery to overcome the difficulties observed during laparoscopic suturing. Peters presented the first preliminary results of robotic assisted laparoscopic transperitoneal pyeloplasty in children.17 Olsen and Jorgensen reported their unique experience with retroperitoneal robotic assisted pyeloplasty.18 The first impression is that the robotic assisted technique makes suturing easier. Other than improving the duration of the learning curve, the robot assisted technique may also facilitate the teaching process and the expansion of laparoscopic pyeloplasty in children to more centers.
8. 9. 10. 11. 12. 13.
CONCLUSIONS
Laparoscopy reproduces the advantages of open pyeloplasty, including the mucosa-to-mucosa anastomosis, and excision of any redundant renal pelvis and diseased ureter. Retroperitoneal laparoscopic dismembered pyeloplasty represents an attractive alternative to conventional open pyeloplasty but it remains a technically challenging procedure. The laparoscopic approach significantly decreases hospital stay compared with the open surgery approach. The main disadvantages of the laparoscopic approach are longer operative time and the long learning curve, making the teaching process more difficult and problematic at teaching institutions. These disadvantages might be improved by the new robot assisted equipment.
14.
15. 16. 17.
18.
1713
pyeloplasty with a minimal incision: comparison of two surgical approaches. Urology, 57: 443, 2001 El-Ghoneimi, A., Farhat, W., Bolduc, S., Bagli, D., McLorie, G., Aigrain, Y. et al: Laparoscopic dismembered pyeloplasty by a retroperitoneal approach in children. BJU Int, 92: 104, 2003 Bernardo, N. and Smith, A. D.: Endopyelotomy review. Arch Esp Urol, 52: 541, 1999 Kavoussi, L. R. and Peters, C. A.: Laparoscopic pyeloplasty. J Urol, 150: 189, 1993 Schuessler, W. W., Grune, M. T., Tecuanhuey, L. V. and Preminger, G. M.: Laparoscopic dismembered pyeloplasty. J Urol, 150: 1795, 1993 Tan, H. L.: Laparoscopic Anderson-Hynes dismembered pyeloplasty in children. J Urol, 162: 1045, 1999 Bauer, J. J., Bishoff, J. T., Moore, R. G., Chen, R. N., Iverson, A. J. and Kavoussi, L. R.: Laparoscopic versus open pyeloplasty: assessment of objective and subjective outcome. J Urol, 162: 692, 1999 Klingler, H. C., Remzi, M., Janetschek, G., Kratzik, C. and Marberger, M. J.: Comparison of open versus laparoscopic pyeloplasty techniques in treatment of uretero-pelvic junction obstruction. Eur Urol, 44: 340, 2003 El-Ghoneimi, A.: Paediatric laparoscopic surgery. Curr Opin Urol, 13: 329, 2003 Farhat, W., Khoury, A., Bagli, D., McLorie, G. and El-Ghoneimi, A.: Mentored retroperitoneal laparoscopic renal surgery in children: a safe approach to learning. BJU Int, 92: 617, 2003 Peters, C. A., Cilento, B. G., Borer, J. G. and Retik, A. B.: Robotically assisted laparoscopic surgery in pediatric urology. Presented at American Academy of Pediatrics, Boston, Massachusetts, October 19 –21, 2002 Olsen, L. H. and Jorgensen, T. M.: Computer assisted pyeloplasty in children: the retroperitoneal approach. J Urol, part 2, 171: 2629, 2004
REFERENCES
1. Ben Slama, M. R., Salomon, L., Hoznek, A., Cicco, A., Saint, F., Alame, W. et al: Extraperitoneal laparoscopic repair of ureteropelvic junction obstruction: initial experience in 15 cases. Urology 56: 45, 2000 2. Soulie´, M., Salomon, L., Patard, J.-J., Mouly, P., Manunta, A., Antiphon, P. et al: Extraperitoneal laparoscopic pyeloplasty: a multicenter study of 55 procedures. J Urol, 166: 48, 2001 3. Peters, C.: Laparoendoscopic renal surgery in children. J Endourol, 14: 841, 2000 4. Yeung, C. K., Tam, Y. H., Sihoe, J. D., Lee, K. H. and Liu, K. W.: Retroperitoneoscopic dismembered pyeloplasty for pelviureteric junction obstruction in infants and children. BJU Int, 87: 509, 2001 5. El-Ghoneimi, A., Sauty, L., Maintenant, J., Macher, M.-A., Lottmann, H. and Aigrain, Y.: Laparoscopic retroperitoneal nephrectomy in high risk children. J Urol, 164: 1076, 2000 6. El-Ghoneimi, A., Farhat, W., Bolduc, S., Bagli, D., McLorie, G. and Khoury, A.: Retroperitoneal laparoscopic vs open partial nephroureterectomy in children. BJU Int, 91: 532, 2003 7. Soulie, M., Thoulouzan, M., Seguin, P., Mouly, P., Vazzoler, N., Pontonnier, F. et al: Retroperitoneal laparoscopic versus open
EDITORIAL COMMENT The authors report the practical usefulness of retroperitoneal laparoscopic pyeloplasty in children with a clear indication of decreased postoperative morbidity, as evidenced by reduced analgesic requirements and shorter hospital stays in their health care system. The authors illustrate technical capability, persistence and diligence in developing this technique and in performing a direct comparative evaluation to permit the accurate assessment of this new technology. While it seems that this will be a major part of our future surgical armamentarium, I am concerned about the limited diffusion of pediatric laparoscopic pyeloplasty. There are few pediatric urologists performing laparoscopic pyeloplasties in the world today despite its having been described almost 10 years ago. The real challenge for the future lies in being able to pass this skill on to our trainees along with the possible development of enhanced technologies to facilitate that dispersion of knowledge. Craig Peters Division of Urology Children’s Hospital Boston, Massachusetts
Vol. 173, 1710 –1713, May 2005 Printed in U.S.A.
DOI: 10.1097/01.ju.0000154169.74458.32
RETROPERITONEAL LAPAROSCOPIC VERSUS OPEN PYELOPLASTY IN CHILDREN ARNAUD BONNARD, VIRGINIE FOUQUET, ELISABETH CARRICABURU, YVES AIGRAIN AND ALAA EL-GHONEIMI* From the Department of Pediatric Surgery and Urology, Robert Debre´ Hospital, AP-HP, University of Paris VII, Paris, France
ABSTRACT
Purpose: The indications for laparoscopy in pediatric urology are expanding and yet the advantages over open surgery remain unclear. We compared the results of retroperitoneal laparoscopic vs open pyeloplasty for pyeloureteral junction obstruction in children. Materials and Methods: A total of 22 children with a mean age of 88 months (range 25 to 192) underwent laparoscopic dismembered pyeloplasty via the retroperitoneal approach. An additional 17 children with a mean age of 103 months (range 37 to 206) underwent similar procedures via open surgery through a flank incision. We retrospectively analyzed and compared operative time, the use of analgesics (acetaminophen or morphine derivatives) and hospital stay. Results: The 2 groups were similar in mean age and weight at surgery. Mean operative time was significantly shorter in the open surgery vs the laparoscopy group (96 minutes, range 50 to 150 vs 219, range 140 to 310, p ⬍0.0001). Mean postoperative use of acetaminophen (1.9 vs 3.22 days, p ⫽ 0.03) and morphine derivatives (1.9 vs 3.06 days, p not significant) was less in the laparoscopy group. Mean hospital stay was shorter in the laparoscopy group than in the open surgery group (2.4 days, range 1 to 5 vs 5, range 3 to 7, p ⬍0.0001). Mean followup was 21 (range 12 to 51) and 24 months (range 12 to 60) in the open and laparoscopy groups, respectively. Conclusions: The operative time of laparoscopic pyeloplasty remains significantly longer than that of the open procedure in children. The main advantage of the laparoscopic approach is that it significantly decreases hospital stay compared with that after an open procedure. Although in our study analgesic use was less after laparoscopy, our results should be confirmed by a prospective, randomized study. KEY WORDS: laparoscopy, kidney, retroperitoneal space
Open dismembered pyeloplasty via a retroperitoneal approach remains the reference standard for correcting ureteropelvic (UPJ) obstruction in children. Although laparoscopic pyeloplasty is well described in adults,1, 2 experience in children remains limited.3, 4 We have already reported that the laparoscopic approach to partial and total nephrectomies significantly decreased the hospital stay compared with that of an open procedure.5, 6 In adults comparative studies show that the laparoscopic approach has the same success rate as open surgery but morbidity and complications are significantly decreased.7 Because we have already observed the feasibility and efficiency of retroperitoneal laparoscopic pyeloplasty in children,8 it seemed necessary to us to compare the results of the laparoscopic approach with those of open surgery. To our knowledge such a study has not been yet done in children. PATIENTS AND METHODS
Between 1999 and 2003 laparoscopic retroperitoneal dismembered pyeloplasty was performed in 22 children with a median age at surgery of 88 months (range 25 to 192). Between 1998 and 2001 an additional 17 consecutive children underwent a similar procedure by conventional open surgery through a flank incision, including a bilateral type done at 2 sessions after a delay of 2 months in 1. Median age at surgery was 103 months (range 37 to 206). Because in our current practice we limit the
laparoscopic approach to children older than 2 years, we excluded from our study children younger than 2 years who underwent open surgery. The indications in the 2 groups were repeat urinary tract infections, hydronephrosis on prenatal diagnosis with postnatal followup showing progressively increasing hydronephrosis on ultrasonography and signs of obstruction on renal scan or abdominal pain related to UPJ obstruction. Evaluation before surgery included renal ultrasonography, voiding cystourethrography, renal scintigraphy (mercaptoacetyltriglycine or dimercapto-succinic acid) and in few cases excretory urography. The retroperitoneal approach was performed as previously described.8 Briefly, the child is placed lateral and 3 trocars (1, 5 mm and 2, 3 mm) are inserted (fig. 1). The kidney is approached posteriorly and the renal pelvis is identified first. The UPJ is identified and minimal dissection is used to free the UPJ from connective tissue. Small vessels are divided after bipolar electrocoagulation. A stay suture of 5-zero polydioxanone is placed at the UPJ for traction and stabilization. The ureteropelvic anastomosis is performed using 6-zero polydioxanone sutures and a tapered 3/8 circular needle. When the UPJ is obstructed by aberrant crossing vessel, dismembered pyeloplasty enables ureteral transposition. A 4.7Fr polyurethane Double-J stent (Medical Engineering Corp., New York, New York) was routinely inserted and remained indwelling for 4 to 6 weeks. In the last 2 cases in this study we inserted a pyelostomy transanastomotic stent, which was removed at the outpatient clinic without anesthesia on postoperative day 10. Perirenal suction drainage was used in the beginning but in the last 8 cases no perirenal
Submitted for publication August 22, 2004. * Correspondence: Department of Pediatric Surgery and Urology, Hoˆpital Robert Debre´, 48, boulevard Se´rurier, 75019 Paris, France (telephone: (33) 1 40 03 2159; FAX: (33) 1 4003 4762; e-mail: [email protected]). 1710
1711
RETROPERITONEAL LAPAROSCOPIC VERSUS OPEN PYELOPLASTY IN CHILDREN
groups for operative time, operative incidents, hospital stay, and postoperative use of analgesic and complications. Results were compared using the Mann-Whitney test and they were considered significant at p ⬍0.05. All patients were assessed clinically and by renal ultrasound 1 month after surgery and every 6 months thereafter. Furosemide renal scintigraphy or excretory urography was done only if there was no decrease in hydronephrosis on ultrasonography or if symptoms recurred. A good result was considered to be a decrease in hydronephrosis on imaging and the resolution of preoperative symptoms.
RESULTS
FIG. 1. Trocar placement for left retroperitoneal laparoscopic pyeloplasty. Child is placed lateral and retroperitoneal access is achieved through first trocar (1) incision 10 to 15 mm long and 1 cm from lower border of 12th rib tip. Second 3 mm trocar (2) is inserted posterior near costovertebral angle, while third 3 mm trocar (3) is inserted 1 cm above top of iliac crest at anterior axillary line. In this case transanastomotic pyelostomy stent was used.
drainage was placed. A Foley catheter remained in situ in all patients for 24 hours after surgery. In the group of children who underwent open surgery the same procedure was performed via an anterolateral, extraperitoneal flank approach. A nephrostomy stent was placed for drainage and perirenal suction drainage was also placed depending on operator convenience. Prophylactic antibiotics (preoperative single dose of third-generation cephalosporin) were routinely prescribed in each group. Pain management was done according to our current pain control protocol. A specialized nurse using a pain score adapted to patient age assessed postoperative pain every 3 hours. The Objective Pain Scale (OPS) was used in patients younger than 6 years. It is based on the evaluation of 5 items, namely blood pressure variation, crying, movements, touch and leg position, and verbalization, with a maximum score of 10. When the OPS score was greater than 4, the child was given analgesics intravenously. A visual analog scale (VAS) graded from 0 to 100 (minimum to maximum individual pain estimation) was used in children 6 years or older. The lower limit for analgesic administration was 40. If the pain score was above the limit with the OPS or the VAS, patients received 15 mg/kg acetaminophen intravenously, limited to 60 mg/kg daily. Alternatively the patient received 0.2 mg/kg nalbuphine (morphine derivative) per dose every 3 hours if pain persisted. Charts were reviewed retrospectively to compare the 2
Two patients (9%) in the laparoscopy group who required conversion to open surgery early in the experience were excluded from study. The table lists the results in the 2 groups. Mean operative time was 95.6 (range 50 to 150) and 219 minutes (range 140 to 310) in open and laparoscopy groups, respectively (p ⬍0.0001). The presence of aberrant crossing vessel did not change mean operative time. The mean postoperative number of days of acetaminophen use was significantly different between the laparoscopic and open surgery groups (1.9 vs 3.22, respectively, p ⫽ 0.03). Mean OPS and VAS scores were less in the laparoscopic group on days 1 and 2 than in the open surgery group but the difference was without statistical significance. One child in the laparoscopic group and 3 in the open surgery group had a postoperative febrile urinary tract infection. Two children (10%) in the laparoscopic group had postoperative leakage. One child did not initially have a stent and leakage was managed by a Double-J stent. The other child was treated with bladder drainage with a Foley catheter. Hospital stay in these children was 9 and 5 days, respectively. Two children (11%) in the open surgery group had postoperative leakage and a Double-J stent was placed. Hospital stay in these children was 11 and 18 days, respectively. All of these children were excluded from hospital stay analysis. Mean hospital stay was 2.4 (range 1 to 5) and 5 days (range 3 to 7) in the laparoscopy and open surgery groups, respectively (p ⬍0.0001). Eight patients were discharged home 2 days after the laparoscopic procedure and 2 were discharged home the day after the procedure. Mean followup was 21 (range 12 to 51) and 24 months (range 12 to 60) in the open and laparoscopy groups, respectively. The failure rate in the laparoscopy group was 4%. A 13-year-old child had increasing hydronephrosis after laparoscopic pyeloplasty. Early reoperation was elected because of increasing dilatation of the pelvis and calices, although a Double-J stent was in situ. We found an inflammatory granuloma on the suture line obstructing the anastomosis and stent.
Results in laparoscopic retroperitoneal and open surgery pyeloplasty groups No. pts Mean age (mos) Mean wt (kg) Mean operative time (mins) Mean postop acetaminophen use (days) Mean postop nalbuphin use (days) Objective pain scale: Day 1 Day 2 Visual analog scale: Day 1 Day 2 Mean hospital stay (days)
Laparoscopy
Open Surgery
p Value
20 88 25.4 218.6 1.85 1.85
17 103 30.4 95.6 3.22 3.06
Not significant Not significant ⬍0.001 0.03 Not significant
0.66 0.41
1.36 0.94
Not significant Not significant
26.6 21 2.9
34.5 33 5.4
Not significant Not significant ⬍0.001
1712
RETROPERITONEAL LAPAROSCOPIC VERSUS OPEN PYELOPLASTY IN CHILDREN DISCUSSION
Open pyeloplasty remained the standard treatment in adult and pediatric patients until the mid 1980s, when morbidity associated with the flank incision led urologists to explore less invasive alternatives. In the 1990s the endourological management of UPJ obstruction became the treatment of choice in the adult population. Although postoperative morbidity is significantly less after the endoscopic procedures, their success rate does not exceed 80%.9 Laparoscopic pyeloplasty was introduced in adults in 1993.10, 11 In the initial report of 5 cases operative time was 3 to 7 hours. The procedure has gained in popularity and more recent series have shown a success rate of greater than 95%.1 Laparoscopic pyeloplasty in children has followed the same evolution as nephrectomy. The procedure was first described through a transperitoneal approach. Tan reported the first pediatric series of transperitoneal laparoscopic dismembered pyeloplasty in 18 children 3 months to 15 years old.12 Mean operative time was 89 minutes. Two patients with persistent obstruction underwent repeat laparoscopic pyeloplasty. There is still controversy concerning which approach to chose, that is transperitoneal or retroperitoneal. Arguments to advocate 1 approach are more theoretical than true objective criteria. The longer time needed for the retroperitoneal approach is probably related to the limited working space, which makes suturing more difficult. In our experience we started our renal laparoscopic surgery using the retroperitoneal approach and we have naturally extended the indications to pyeloplasty without changing our habits. The transperitoneal approach theoretically increases the risk of abdominal organ injury. It requires more dissection to reach the kidney and the colon must be reflected. We believe that the golden standard of pediatric open renal surgery is the retroperitoneal approach and minimal invasive surgery should follow the same rules. We do not believe that starting experience with pyeloplasty through the transperitoneal approach would facilitate the procedure. The only measure that may facilitate the procedure is to respect progressive experience and learn perfectly the 2 approaches. Although urine leakage has not yet been reported, this complication would be better tolerated in the retroperitoneal space than in the intraperitoneal cavity. In our practice a horseshoe kidney is the only anatomical indication for the transperitoneal approach, which we have used in 4 cases. Yeung et al reported their initial experience with retroperitoneal dismembered pyeloplasty in 13 patients, of whom 1 required open conversion.4 Mean operative time was 143 minutes (range 103 to 235). The longer time needed for the retroperitoneal approach is probably related to the limited working space, which renders suturing more difficult. Our experience with retroperitoneal pyeloplasty since 1999 is now approaching 50 children (fig. 2). All of these cases were done by one of us (AE-G) but some were done elsewhere. The cases included in our comparative study were restricted to our center to avoid any bias concerning pain management and hospital stay. We have already reported our initial experience and midterm results in a group of 21 children and confirmed that dismembered retroperitoneal laparoscopic pyeloplasty is a safe and feasible approach in children.8 In our current study the learning curve was significant, in that mean operative time was decreased from 4 to 3 hours. Two patients in the laparoscopy group required conversion to open surgery at the beginning of our experience and they were excluded from study. Conversion in the beginning of our experience was due to difficult suturing and we did not need to convert any of the last 35 cases of our global experience with pyeloplasty (fig. 2). In our current practice children younger than 2 years undergo a posterior lumbar incision to avoid the technical difficulties of suturing in young children
FIG. 2. Retroperitoneal laparoscopic pyeloplasty learning curve, representing global experience with retroperitoneal laparoscopic experience. Total of 50 cases were consecutive from 1999 to 2004. Each point indicates 1 child but 4 (8%) needing conversion to open surgery are not shown. 1 redo, 1 patient (2%) required reoperation.
via the retroperitoneal approach. In infants the postoperative course is usually uneventful and the cosmetic appearance of the posterior approach is excellent compared with that of the anterior or lateral lumbar incision needed in older children. Soulie et al compared retroperitoneal laparoscopic vs open pyeloplasty with a minimal incision in 53 consecutive, nonrandomized adults.7 Mean operative time was similar in the 2 groups (165 and 145 minutes, respectively). The incidence of complications, hospital stay and functional results were equivalent in the 2 groups but the return to painless activity was more rapid with laparoscopy in younger patients. Bauer et al did a similar study, which showed no difference in postoperative outcomes between laparoscopic and open pyeloplasty.13 Klingler et al observed in adults that dismembered laparoscopic pyeloplasty had the same success rate as open surgery but morbidity and complications were significantly decreased.14 In contrast to the adult population, no comparative study has yet been done in pediatric patients.15 The current technique was modified many times during our experience. First we used an interrupted 5-zero suture for the anastomosis but with the development of 3 mm instruments, which made it possible to use 6-zero suturing, we were confident to reproduce the principles of open pyeloplasty and use the usual 6-zero running suture. Since then, we have not used a suction drain around the anastomosis. Another change was that 3 ports were used and not 4, principally to decrease the number of instruments used in the limited retroperitoneal space. Leaving a Double-J stent is a major concern in children because repeat general anesthesia is necessary for its removal. For this reason we currently use a pyelostomy stent to intubate the anastomosis. The stent is clamped the day after surgery and removed at the outpatient clinic without anesthesia on postoperative day 10. The advantage of the laparoscopic approach over the open procedure is the decreased the postoperative pain and hospital stay but a prospective, randomized study is mandatory to confirm it. Although our current study was retrospective, the same pain management protocol was followed in all children and treatment by pain scores was strictly documented in the charts. However, these results must be confirmed by a prospective study. Cosmetically it seems to us that with the 3 mm ports the results are better than open surgery scars. One of our main concerns is about teaching laparoscopic pyeloplasty to surgeons in training or qualified surgeons but with recent experience with advanced laparoscopic surgery. In our study all cases of laparoscopy was done by the same surgeon except 1, in which the same surgeon assisted an-
RETROPERITONEAL LAPAROSCOPIC VERSUS OPEN PYELOPLASTY IN CHILDREN
other qualified surgeon experienced with laparoscopic procedures. In contrast to ablative surgery, pyeloplasty remains a difficult laparoscopic procedure to teach.16 Pyeloplasty should be the best demonstration of the advantage of robot assisted surgery to overcome the difficulties observed during laparoscopic suturing. Peters presented the first preliminary results of robotic assisted laparoscopic transperitoneal pyeloplasty in children.17 Olsen and Jorgensen reported their unique experience with retroperitoneal robotic assisted pyeloplasty.18 The first impression is that the robotic assisted technique makes suturing easier. Other than improving the duration of the learning curve, the robot assisted technique may also facilitate the teaching process and the expansion of laparoscopic pyeloplasty in children to more centers.
8. 9. 10. 11. 12. 13.
CONCLUSIONS
Laparoscopy reproduces the advantages of open pyeloplasty, including the mucosa-to-mucosa anastomosis, and excision of any redundant renal pelvis and diseased ureter. Retroperitoneal laparoscopic dismembered pyeloplasty represents an attractive alternative to conventional open pyeloplasty but it remains a technically challenging procedure. The laparoscopic approach significantly decreases hospital stay compared with the open surgery approach. The main disadvantages of the laparoscopic approach are longer operative time and the long learning curve, making the teaching process more difficult and problematic at teaching institutions. These disadvantages might be improved by the new robot assisted equipment.
14.
15. 16. 17.
18.
1713
pyeloplasty with a minimal incision: comparison of two surgical approaches. Urology, 57: 443, 2001 El-Ghoneimi, A., Farhat, W., Bolduc, S., Bagli, D., McLorie, G., Aigrain, Y. et al: Laparoscopic dismembered pyeloplasty by a retroperitoneal approach in children. BJU Int, 92: 104, 2003 Bernardo, N. and Smith, A. D.: Endopyelotomy review. Arch Esp Urol, 52: 541, 1999 Kavoussi, L. R. and Peters, C. A.: Laparoscopic pyeloplasty. J Urol, 150: 189, 1993 Schuessler, W. W., Grune, M. T., Tecuanhuey, L. V. and Preminger, G. M.: Laparoscopic dismembered pyeloplasty. J Urol, 150: 1795, 1993 Tan, H. L.: Laparoscopic Anderson-Hynes dismembered pyeloplasty in children. J Urol, 162: 1045, 1999 Bauer, J. J., Bishoff, J. T., Moore, R. G., Chen, R. N., Iverson, A. J. and Kavoussi, L. R.: Laparoscopic versus open pyeloplasty: assessment of objective and subjective outcome. J Urol, 162: 692, 1999 Klingler, H. C., Remzi, M., Janetschek, G., Kratzik, C. and Marberger, M. J.: Comparison of open versus laparoscopic pyeloplasty techniques in treatment of uretero-pelvic junction obstruction. Eur Urol, 44: 340, 2003 El-Ghoneimi, A.: Paediatric laparoscopic surgery. Curr Opin Urol, 13: 329, 2003 Farhat, W., Khoury, A., Bagli, D., McLorie, G. and El-Ghoneimi, A.: Mentored retroperitoneal laparoscopic renal surgery in children: a safe approach to learning. BJU Int, 92: 617, 2003 Peters, C. A., Cilento, B. G., Borer, J. G. and Retik, A. B.: Robotically assisted laparoscopic surgery in pediatric urology. Presented at American Academy of Pediatrics, Boston, Massachusetts, October 19 –21, 2002 Olsen, L. H. and Jorgensen, T. M.: Computer assisted pyeloplasty in children: the retroperitoneal approach. J Urol, part 2, 171: 2629, 2004
REFERENCES
1. Ben Slama, M. R., Salomon, L., Hoznek, A., Cicco, A., Saint, F., Alame, W. et al: Extraperitoneal laparoscopic repair of ureteropelvic junction obstruction: initial experience in 15 cases. Urology 56: 45, 2000 2. Soulie´, M., Salomon, L., Patard, J.-J., Mouly, P., Manunta, A., Antiphon, P. et al: Extraperitoneal laparoscopic pyeloplasty: a multicenter study of 55 procedures. J Urol, 166: 48, 2001 3. Peters, C.: Laparoendoscopic renal surgery in children. J Endourol, 14: 841, 2000 4. Yeung, C. K., Tam, Y. H., Sihoe, J. D., Lee, K. H. and Liu, K. W.: Retroperitoneoscopic dismembered pyeloplasty for pelviureteric junction obstruction in infants and children. BJU Int, 87: 509, 2001 5. El-Ghoneimi, A., Sauty, L., Maintenant, J., Macher, M.-A., Lottmann, H. and Aigrain, Y.: Laparoscopic retroperitoneal nephrectomy in high risk children. J Urol, 164: 1076, 2000 6. El-Ghoneimi, A., Farhat, W., Bolduc, S., Bagli, D., McLorie, G. and Khoury, A.: Retroperitoneal laparoscopic vs open partial nephroureterectomy in children. BJU Int, 91: 532, 2003 7. Soulie, M., Thoulouzan, M., Seguin, P., Mouly, P., Vazzoler, N., Pontonnier, F. et al: Retroperitoneal laparoscopic versus open
EDITORIAL COMMENT The authors report the practical usefulness of retroperitoneal laparoscopic pyeloplasty in children with a clear indication of decreased postoperative morbidity, as evidenced by reduced analgesic requirements and shorter hospital stays in their health care system. The authors illustrate technical capability, persistence and diligence in developing this technique and in performing a direct comparative evaluation to permit the accurate assessment of this new technology. While it seems that this will be a major part of our future surgical armamentarium, I am concerned about the limited diffusion of pediatric laparoscopic pyeloplasty. There are few pediatric urologists performing laparoscopic pyeloplasties in the world today despite its having been described almost 10 years ago. The real challenge for the future lies in being able to pass this skill on to our trainees along with the possible development of enhanced technologies to facilitate that dispersion of knowledge. Craig Peters Division of Urology Children’s Hospital Boston, Massachusetts