+
MODEL
Journal of Pediatric Urology (2015) xx, 1.e1e1.e6
Transperitoneal versus retroperitoneal laparoscopic pyeloplasty in children: Randomized clinical trial Haytham Badawy a, Amr Zoaier b, Tamer Ghoneim c, Ahmed Hanno d a
Unit of Pediatric Urology, Department of Urology, University of Alexandria, Egypt
b
Unit of Pediatric Urology, Department of Urology, Sporting Children Insurance Hospital, Egypt
c Department of Anaesthesia, University of Alexandria, Egypt
d
Department of Urology, University of Alexandria, Egypt Correspondence to: H. Badawy, Department of Urology, University of Alexandria, Egypt, Tel.: þ20 35758575, þ20 1223690597 (mobile) hithamalmetwale@ yahoo.com (H. Badawy) Keywords Pyeloplasty; Laparoscopy; Retroperitonoscopic; Children Received 24 January 2014 Accepted 26 November 2014 Available online xxx
Summary Introduction Laparoscopic pyeloplasty achieves good cosmetic and functional outcomes. Both transperitoneal and retroperitoneal approaches are used. No single study to date has compared the two approaches in a prospective randomized design. Objective We present a prospective randomized comparison between both approaches in children in a trial to define which technique is better with regard to multiple factors including operative time, hospital stay, recovery of bowel movement, analgesic requirement and complication rate. Study design In the period from June 2010 to September 2012, 38 children (25 boys and 13 girls) were operated laparoscopically. Children were randomized into Group I (19 children) operated by the transperitoneal approach, and Group II (19 children) operated by the retroperitoneal approach. Both groups were compared as regards to the operative time, anesthetic changes, and postoperative recovery. A minimum sample size required was calculated to be 19 for each arm based on previous studies of laparoscopic pyeloplasty, using a mean difference in operative time Z 40 min, effect size Z 0.95, an alpha of 0.05 and power 80% and an online sample size calculator. Statistical analysis was performed using SPSS software using the Fischer exact test, chi square test and ManneWhitney U test. The operative time was the primary endpoint for comparison between both approaches. Results
Transperitoneal Retroperitoneal P value
transperitoneal and retroperitoneal laparoscopic pyeloplasty in children. Shouma et al. is the only prospective randomized study to compare both techniques in adult pyeloplasty. They had a significantly shorter operative time in the transperitoneal group however, the author in the discussion mentioned that he was at the start of the learning curve for retroperitonoscopic pyeloplasty when he conducted his study, which affected the result of the operative time. Hence, as mentioned above, we stressed the importance of a single surgeon with adequate equal experience in both techniques. The recovery of the intestinal motility and start of oral feeding were significantly faster in the retroperitoneal group compared to the transperitoneal group. In our opinion this can be explained by the absence of intraperitoneal manipulations and urine leakage in the peritoneal space. In their series of retroperitoneal pyeloplasty, El Ghoneimi et al. reported feeding after a mean of 1.4 days, however, in our series there was even earlier oral feeding. Shouma et al. reported no significant difference in the start of oral feeding in their adult series. The limitations of our study are: the choice of the 40 min difference created a statistically significant difference in operative time between the groups which might not be considered a truly clinically important difference. In addition, the single author operating for both approaches, which might create a bias, however the author has sufficient experience in both approaches. Moreover, although there were significant differences in hospital stay and intestinal movement between the two groups, it is not clear if these were of clinical significance. Conclusion Both transperitoneal and retroperitoneal approaches have high success rate. The shorter operative time,
Median age (years)
Median operative time (minutes)
Median hospital stay (hours)
Start oral feeding (hours)
6 5 0.437
150 129 0.010
48 24 0.002
16 10 0.000
Discussion Our series is the first in the literature that compares in a prospective randomized design the
shorter hospital stay, rapid recovery of intestinal movement and early resumption of oral feeding are in favor with the retroperitoneal approach.
http://dx.doi.org/10.1016/j.jpurol.2014.11.019 1477-5131/Published by Elsevier Ltd on behalf of Journal of Pediatric Urology Company.
Please cite this article in press as: Badawy H, et al., Transperitoneal versus retroperitoneal laparoscopic pyeloplasty in children: Randomized clinical trial, Journal of Pediatric Urology (2015), http://dx.doi.org/10.1016/j.jpurol.2014.11.019
+
MODEL
1.e2
Introduction Laparoscopic pyeloplasty achieves equal success rates to open pyeloplasty with regard to minimal morbidity, rapid recovery, and better cosmesis [1]. Both transperitoneal and retroperitoneal approaches are used with equal success rates [2e4]. However, there is a lack of randomized studies with a prospective design comparing the two approaches [5]. The present study was a prospective, randomized trial comparing the two approaches in children to test if the retroperitoneal approach, compared to the transperitoneal approach, has a shorter operative time, a more rapid recovery of intestinal movement, a shorter length of hospital stay and a comparable overall success rate.
Materials and methods In the present, prospective randomized study, 38 children aged 2 years or older (25 boys, 13 girls) were enrolled between June 2010 and September 2012. Approval was obtained from the University of Alexandria’s ethical committee and parental informed consent was received. Children were randomized using closed envelope randomization into two equal arms: transperitoneal approach and retroperitoneal approach. Malrotated kidneys, renal fusion anomalies and redo cases were excluded from the study. A stented dismembered pyeloplasty was performed on all of the children by the same surgeon (first surgeon). Anterior transposition of the UPJ was performed in the presence of a crossing vessel. Follow-up was undertaken every 3 months by ultrasound, and DTPA renogram in persistent hydronephrosis. A standard anesthesia protocol was used in all children as follows: all children received premedication; midazolam at a dose of 0.05 mg/kg bodyweight through an intravenous catheter; propofol at a dose of 2.0e2.5 mg/kg bodyweight; a muscle relaxant in the form of cisatracurium at a dose of 0.15 mg/kg bodyweight; and analgesia in the form of fentanyl at a dose of 2.0 Ugm/kg bodyweight. All children were mechanically ventilated after insertion of an appropriately sized endotracheal tube. A nasogastric tube was inserted, and intraoperative monitoring was performed with a pulse oximeter, non-invasive blood pressure monitor, and an electrocardiogram; end tidal carbon dioxide (ETCO2) was monitored through a capnogram. In both approaches, the children were positioned in the lateral flank position. In the transperitoneal group, open access was established through the umbilicus; a 5-mm port was inserted and insufflation was maintained at 12 mmHg, then two 3-mm trocars were inserted under vision, one midway between the anterior superior iliac spine and the umbilicus, and the other midway between the xiphisternum and the umbilicus. In the retroperitoneal group, open access was established by a 0.5e1.0 cm incision in the midaxillary line below the last rib; gerota fascia was grasped and opened under vision; a 5-mm trocar was inserted and secured in place by a stitch in the sheath; insufflation was maintained at a pressure of 12 mm Hg; then, two 3-mm trocars were inserted, one in the costovertebral angle and the other in the anterior axillary line one finger above the iliac crest.
H. Badawy et al. In both approaches, the basic principles of pyeloplasty were followed: dismembering of the UPJ, trimming of the renal pelvis, spatulation of the ureter after excision of the aperistalsis segment, anastomosis of the ureter to the pelvis by a 6/0 vicryl suture, and antegrade insertion of the double J (DJ) in all cases. The drain was left inside, through one of the ports, until it stopped leaking urine. Both groups were compared regarding the following intraoperative anesthetic changes: heart rate, blood pressure, ETCO2 changes, PH changes, O2 saturation, and urine output. The operative time was calculated from the insertion of the first optic trocar until the end of the procedure. The recovery of bowel movement and toleration of oral feeding were compared in the two arms. Length of hospital stay (in hours) was calculated from the day of admission until discharge. Intraoperative and postoperative complications, as well as overall success rates (disappearance of the symptoms, regression of hydronephrosis, and increase in the renal parenchyma) in the follow-up period were also noted. Complications were reported according to the Clavien-Dindo classification system [6]. On the first operative day, an anesthetist assessed postoperative pain and analgesia every 4 h, and on the second day this was assessed every 6 h. Analgesia was established by giving all children paracetamol (Perfalgan) 15 mg/kg every 6 h, plus rescue analgesia in the form of the non-steroidal anti-inflammatory drug (NSAIDS) diclofenac sodium in a dose of 0.5e1.5 mg/kg every 8 h. Pain was assessed according to the pain score (Visual Analogue Scale in children older than 6 years and the Face, Legs, Activity, Cry, Consolability scale (FLACC) for younger children. Opioid analgesia was not required for any of the children. Based on previous studies of transperitoneal and retroperitoneal laparoscopic pyeloplasty, a mean difference in operative time of 40 min, an effect size of 0.95, an alpha of 0.05 and power of 80% were used [7]. Using an online sample size calculator, statistical analysis was performed using SPSS software using the Fischer exact test, Chisquared test and ManneWhitney U test. The minimum sample size that was required was calculated to be 19 for each arm. The operative time was deemed to be the primary endpoint for comparison between both approaches.
Results The two groups were comparable, with no significant differences between them for the mean age, weight, gender distribution and clinical presentation (Table 1). There was no significant difference between the two groups concerning the intraoperative anesthetic changes and urine output (as shown in the Supplementary Data). There were no significant differences in the amount of rescue NSAIDS analgesia used in the first and second postoperative days between the two groups (Table 2). The retroperitonoscopic approach was significantly shorter in operative time than the transperitoneal approach (P Z 0.010) (Table 3). The presence or absence of crossing vessels did not significantly affect the operative time, regardless of the approach utilized (P Z 0.961) (Table 3). The length of hospital stay was significantly shorter in the retroperitonoscopic group than the transperitoneal group
Please cite this article in press as: Badawy H, et al., Transperitoneal versus retroperitoneal laparoscopic pyeloplasty in children: Randomized clinical trial, Journal of Pediatric Urology (2015), http://dx.doi.org/10.1016/j.jpurol.2014.11.019
+
MODEL
Comparison between laparoscopic transperitoneal and retroperitoneal pyeloplasty in children. Table 1
Demographic data and clinical presentation.
Age (years) Median Weight (kg) Median
Sex Male Female Presentation Loin pain Febrile UTI Incidentally discovered Prenatal diagnosis with hydronephrosis
Table 2
Transperitoneal
Retroperitoneal
ManneWhitney
6.00
5.00
24.00 Transperitoneal N %
18.00 Retroperitoneal N %
14 5
73.7 26.3
11 8
57.9 42.1
X2 Z 1.05 P Z 0.305
15 2 2 0
78.9 10.5 10.5 0
10 5 1 3
52.6 26.3 5.3 15.8
X2 Z 5.619 P Z 0.132
Z Z 0.778 P Z 0.437 Z Z 0.556 P Z 0.578 Tests of significance
Non-steroidal anti-inflammatory drugs in the first two postoperative days.
NSAIDS first day in mg/d Median NSAIDS second day in mg/d Median
Transperitoneal
Retroperitoneal
25.00
25.00
37.50
37.50
ManneWhitney P Z 0.213 P Z 0.222
(P Z 0.002) (Table 3). A significantly more rapid recovery of bowel movement and toleration of oral feeding were found in the retroperitoneal group, compared to the transperitoneal group (Table 3). Significant urine leakage and, hence, delayed drain removal was present in the transperitoneal group, compared to the retroperitoneal group. The median duration of a drain in situ in the transperitoneal group was 48.00 h while in the retroperitoneal group it was 24.00 h (P Z 0.000).
Table 3
1.e3
No intraoperative complications were encountered, and no conversion to open surgery occurred. Postoperative complications (Grade III b according to the Clavien-Dindo classification) [6] in the form of urinomas occurred in two children in the transperitoneal group: (1) after 1 week, a perirenal urinoma that was percutaneously aspirated; and (2) after 10 days of discharge, one child presented with a fever, intraperitoneal collection and sepsis, which was
Operative time, length of hospital stay, and oral feeding.
Operative time (minutes) Median Minemax range Length of hospital stay (hours) Median Minemax range Recovery of bowel movement (hours) Median Minemax range Start of oral feeding (hours) Median Minemax range Operative time (minutes) Median Minemax range
Transperitoneal
Retroperitoneal
150.00 120e240 Transperitoneal
129.00 120e189 Retroperitoneal
48.00 24e72
24.00 24e72
14.00 9e20
7.00 6e12
16.00 12e22
10.00 7e14
Crossing vessels present
Crossing vessels absent
138.00 120e240
133.2 120e189
ManneWhitney Z Z 2.586 P Z 0.010 ManneWhitney P Z 0.002
Z Z 4.869 P Z 0.000 Z Z 4.945 P Z 0.000 ManneWhitney Z Z 0.51 P Z 0.961
Please cite this article in press as: Badawy H, et al., Transperitoneal versus retroperitoneal laparoscopic pyeloplasty in children: Randomized clinical trial, Journal of Pediatric Urology (2015), http://dx.doi.org/10.1016/j.jpurol.2014.11.019
+
MODEL
1.e4 treated with exploration and open drainage of an infected intrabdominal urine collection. The mean follow-up period was 10 months (range 6e24). The overall success rate was 94.7%. There were two redoes, one in each group: in the transperitoneal group, after 3 months, which was repaired by open pyeloplasty; in the retroperitoneal group, after 6 months, which was repaired by antegrade endopyelotomy.
Discussion The present study is the first prospective, randomized design in the literature that compares the transperitoneal and retroperitoneal laparoscopic pyeloplasty in children. In the present study, a single surgeon (the first author) performed both the transperitoneal and retroperitoneal laparoscopic pyeloplasties. The surgeon has adequate and equal experience in both transperitoneal and retroperitoneal approaches. This is an important feature in the study, as it alleviates any bias related to surgeon experience or learning curve. Shouma et al. conducted the only prospective, randomized study that compares both techniques in adult pyeloplasty. They had a significantly shorter operative time in the transperitoneal group; however, in the discussion, the author mentioned that he was at the start of his learning curve for retroperitonoscopic pyeloplasty when he conducted the study, which affected the results of the operative time for the retroperitoneal approach, making it longer. Hence, as mentioned above, the importance of a single surgeon with adequate equal experience in both techniques is very important [7]. In the present study, the operative time was significantly shorter in the retroperitoneal group (median 129.00 min) compared to the transperitoneal group (median 150.00 min) (P Z 0.010). Although these results are different to many other studies in the literature, the shorter operative time in the retroperitoneal approach in the present study could be explained by the direct access to the UPJ offered by this approach, in contrast to the transperitoneal approach where the colon must be reflected and the intestines must be manipulated to reach the target organ [6,8e10]. We decided not to use the transmesocolic approach which easily performed on the left side pyeloplasty but not on the right side pyeloplasty not to create any internal bias in the study groups. However, the use of the transmesocolic approach in further studies might create shorter operative times in the transperitoneal group e this needs to be explored in a further well-designed study. In their study, Shouma et al. pointed out that the presence of a crossing vessel is significantly related to increased operative times, as anastomosis is more difficult, especially in the retroperitoneal approach [7]. Contrary to their experience, in the present study, there was no increased difficulty in the presence of a crossing vessel in the retroperitoneal group. After dissection of the pelviureteric junction obstruction (PUJO) from the crossing vessel, the pelvis and ureter were anteriorly transposed to the crossing vessel, the renal pelvis was attached to the psoas muscle by a 5/0 vicryl suture, which stabilizes the renal pelvis and
H. Badawy et al. takes the crossing vessel upward and away from the surgeon, thus making the anastomosis easier and quicker. El-Ghoneimi et al. [4] encountered crossing vessels in nine children: the anterior transposition of the PUJ was performed and no significant prolongation of operative time was reported. In the present study, the DJ was inserted antegrade after finishing the anastomosis of the posterior wall. It is better to insert the DJ in a retrograde way because it decompresses the renal pelvis, making its dissection more difficult; moreover, it sometimes hinders the ureteral spatulation, especially in a narrow ureter. In a series of twenty-four children, Mandhani et al. pointed out that laparoscopic stenting is a simple technique, which obviates the need for any other procedure [11]. In a prospective study of 120 patients, Arumainayagam et al. reported shorter operative time in antegrade DJ stenting versus retrograde stenting (P < 0.0001) [12]. In the present series, oxygen saturation was not significantly different between the two approaches; however, in their retrospective series of seventy-three children operated by the retroperitoneal approach, Halachmi et al. reported a decrease in oxygen saturation recorded after CO2 insufflation [13]. In the present study, hemodynamic and blood gas changes were not significantly different between the transperitoneal and retroperitoneal groups. However, Halachmi et al. reported a significant decrease in heart rate in the transperitoneal group (113 18) versus (116 19) beat per minute (P Z 0.019) after CO2 insufflation [13]. In the present study, there was no difference in the mean ETCO2 between the transperitoneal and retroperitoneal groups. This is in contrast to Karsli et al. who had a significant difference between the transperitoneal and retroperitoneal groups [14]. The analgesic requirement in the present study was not significantly different between the two groups. Paracetamol (Perfalgan) was given in a fixed dose and frequency, according to bodyweight, in the postoperative period. The only factor that could have made a difference between the two groups, concerning the overall dose of Perfalgan, would have been a significant difference in the weight of the children between the groups, which was not the case in the present study where there was a comparable weight between the groups. Rescue analgesia, in the form of NSAIDS in a dose of 0.5e1.5 mg/kg, was given according to the scheduled regular evaluation of pain using validated scores (VAS for children older than 6 years and FLACC for younger children); the anesthetist continuously evaluated and adjusted the doses. Because NSAIDS were used according to the pain scale, the two groups could be compared using an objective parameter of pain intensity rather than the weight of the children, however, no statistically significant difference could be found between the groups. The recovery of bowel movement and start of oral feeding were significantly faster in the retroperitoneal group, compared to the transperitoneal group. It is believed that this is due to the absence of intraperitoneal manipulations and urine leakage in the peritoneal space. El-Ghoneimi et al. reported quicker feeding in their group of retroperitoneal children after a mean of 1.4 days [4]; however, in the present study this was faster. Shouma et al.
Please cite this article in press as: Badawy H, et al., Transperitoneal versus retroperitoneal laparoscopic pyeloplasty in children: Randomized clinical trial, Journal of Pediatric Urology (2015), http://dx.doi.org/10.1016/j.jpurol.2014.11.019
+
MODEL
Comparison between laparoscopic transperitoneal and retroperitoneal pyeloplasty in children. reported no significant differences in the start of oral feeding in their adult series [7]. In the present study, drain removal was significantly quicker in the retroperitoneal versus the transperitoneal group. This might be explained by the closed retroperitoneal space, which allows the suture line to seal after desufflation and prevents urine leakage. Shouma et al. found no differences in the time of removal of the drain between the two groups [7]. In the present study, the length of hospital stay was significantly shorter in the retroperitoneal group, compared to the transperitoneal group. This might be attributable to the rapid recovery of intestinal movement, early start of oral feeding, and earlier removal of the drain. The median hospital stay was shorter compared to that of the series by Singh et al. [15], and Bonnart et al. [16]. In the present study, neither intraoperative complications nor conversion to open surgery occurred. In their retrospective series of 83 patients, Davenport et al. reported 17 retroperitonoscopic pyeloplasties, two of which required open surgery due to failure to progress [1]. In a series of 49 children by Canon et al., 27 underwent retroperitonoscopic pyeloplasty and 19 underwent transperitoneal laparoscopic pyeloplasty. Three cases of transient prolonged urine leakage were reported e two in the retroperitoneal group and one in the transperitoneal group, and six cases undergoing balloon dilatation were reported e five in the retroperitoneal group and one in the transperitoneal group [8]. In the present study, there were no intraoperative complications encountered, and no conversion to open surgery occurred. Two children in the transperitoneal group had postoperative complications in the form of urinomas, one after 1 week, and the other after 10 days of discharge. The study had several limitations: the choice of the 40 min difference created a statistically significant difference in operative time between both groups, which might not be considered to be a clinical difference; the single surgeon operating with both approaches, which might have created bias, however, it was previously stated that he had sufficient experience in both approaches. Moreover, although the length of hospital stay and intestinal movement were statistically significant different between the two approaches, it is unclear as to whether there is any clinical significance.
Conclusion Both the transperitoneal and retroperitoneal approaches had a high success rate. However, the shorter operative time, shorter length of hospital stay, rapid recovery of intestinal movement, and early resumption of oral feeding are in favor of the retroperitoneal approach.
Ethical approval Number of IRB acceptance: 0101570.
Source of funding None declared.
1.e5
Conflict of interest None declared.
Acknowledgment Dr Moataza Abdelwahab (Department of statistics, medical research institute, University of Alexandria, Egypt) for her outstanding effort in performing the statistics and analysis of the study.
Appendix A. Supplementary data Supplementary data related to this article can be found online at http://dx.doi.org/10.1016/j.jpurol.2014.11.019.
References [1] Davenport K, Minervini A, Timoney AG. Our experience with retroperitoneal and transperitoneal laparoscopic pyeloplasty for pelvi-ureteric junction obstruction. Eur Urol 2005;48(6):973e7. [2] Tan HL. Laparoscopic Anderson-Hynes dismembered pyeloplasty in children. J Urol 1999;162(3 Pt 2):1045e7. [3] Yeung CK, Tam YH, Sihoe JD, Lee KH, Liu KW. Retroperitoneoscopic dismembered pyeloplasty for pelvi-ureteric junction obstruction in infants and children. BJU Int 2001; 87(6):509e13. [4] El-Ghoneimi A, Farhat W, Bolduc S, Bagli D, McLorie G, Aigrain Y, et al. Laparoscopic pyeloplasty by retroperitoneal approach in children. BJU Int 2003;92:104e8. [5] Wu Y, Dong Q, Han P, Liu L, Wang L, Wei Q. Meta-analysis of transperitoneal versus retroperitoneal approaches of laparoscopic pyeloplasty for ureteropelvic junction obstruction. J Laparoendosc Adv Surg Tech A 2012;22(7):658e62. [6] Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240(2): 205e13. [7] Shoma MA, El Nahas AR, Bazeed MA. Laparoscopic pyeloplasty: a prospective randomized comparison between the transperitoneal approach and retroperitoneoscopy. J Urol 2007; 178:2020e4. [8] Canon JS, Jaynathi VR, Lowe GJ. Which is betterretroperitoneoscopic or laparoscopic dismembered pyeloplasty in children? J Urol 2007;178(4):1791e5. [9] Inagaki T, Rha KH, Ong AM, Kavoussi LR, Jarrett TW. Laparoscopic pyeloplasty: current status. BJU Int 2005;95:102e6. [10] Janetschek G, Peschel R, Franscher F. Laparoscopic pyeloplasty. Urol Clin North Am 2000;27:695e704. [11] Mandhani A, Bhandari SM. Is antegrade stenting superior to retrograde stenting in laparoscopic pyeloplasty? J Urol 2004; 177:1440e2. [12] Arumainayagam N, Minervini A, Davenport K, Kumar V, Masieri L, Serni S, et al. Antegrade versus retrograde stenting in laparoscopic pyeloplasty. J Endourol 2008;22(4):671e4. [13] Halachmi S, El-Ghoneimi A, Bissonnette B, Zaarour C, Bagli DJ, McLorie GA, et al. Hemodynamic and respiratory effect of pediatric urological laproscopic surgery: a retrospective analysis. J Urol 2003;170:1651e4. [14] Karsli C, El-Hout Y, Lorenzo AJ, Langer JC, Ba ¨gli DJ, Pippi Salle JL, et al. Physiological changes in transperitoneal versus retroperitoneal laparoscopy in children: a prospective analysis. J Urol 2011;69:1649e52.
Please cite this article in press as: Badawy H, et al., Transperitoneal versus retroperitoneal laparoscopic pyeloplasty in children: Randomized clinical trial, Journal of Pediatric Urology (2015), http://dx.doi.org/10.1016/j.jpurol.2014.11.019
+
MODEL
1.e6 [15] Singh H, Ganpule A, Malhotra T, Manohar T, Muthu V, Desai M. Transperitoneal laparoscopic pyeloplasty in children. J Endourol 2007;21(12):1461e7.
H. Badawy et al. [16] Bonnart A, Fouquet V, Carricaburu E, Aigrain Y, El-Ghoneimi A. Retroperitoneal laparosopic versus open pyeloplasty in children. J Urol 2005;173:1710e4.
Please cite this article in press as: Badawy H, et al., Transperitoneal versus retroperitoneal laparoscopic pyeloplasty in children: Randomized clinical trial, Journal of Pediatric Urology (2015), http://dx.doi.org/10.1016/j.jpurol.2014.11.019
MODEL
Journal of Pediatric Urology (2015) xx, 1.e1e1.e6
Transperitoneal versus retroperitoneal laparoscopic pyeloplasty in children: Randomized clinical trial Haytham Badawy a, Amr Zoaier b, Tamer Ghoneim c, Ahmed Hanno d a
Unit of Pediatric Urology, Department of Urology, University of Alexandria, Egypt
b
Unit of Pediatric Urology, Department of Urology, Sporting Children Insurance Hospital, Egypt
c Department of Anaesthesia, University of Alexandria, Egypt
d
Department of Urology, University of Alexandria, Egypt Correspondence to: H. Badawy, Department of Urology, University of Alexandria, Egypt, Tel.: þ20 35758575, þ20 1223690597 (mobile) hithamalmetwale@ yahoo.com (H. Badawy) Keywords Pyeloplasty; Laparoscopy; Retroperitonoscopic; Children Received 24 January 2014 Accepted 26 November 2014 Available online xxx
Summary Introduction Laparoscopic pyeloplasty achieves good cosmetic and functional outcomes. Both transperitoneal and retroperitoneal approaches are used. No single study to date has compared the two approaches in a prospective randomized design. Objective We present a prospective randomized comparison between both approaches in children in a trial to define which technique is better with regard to multiple factors including operative time, hospital stay, recovery of bowel movement, analgesic requirement and complication rate. Study design In the period from June 2010 to September 2012, 38 children (25 boys and 13 girls) were operated laparoscopically. Children were randomized into Group I (19 children) operated by the transperitoneal approach, and Group II (19 children) operated by the retroperitoneal approach. Both groups were compared as regards to the operative time, anesthetic changes, and postoperative recovery. A minimum sample size required was calculated to be 19 for each arm based on previous studies of laparoscopic pyeloplasty, using a mean difference in operative time Z 40 min, effect size Z 0.95, an alpha of 0.05 and power 80% and an online sample size calculator. Statistical analysis was performed using SPSS software using the Fischer exact test, chi square test and ManneWhitney U test. The operative time was the primary endpoint for comparison between both approaches. Results
Transperitoneal Retroperitoneal P value
transperitoneal and retroperitoneal laparoscopic pyeloplasty in children. Shouma et al. is the only prospective randomized study to compare both techniques in adult pyeloplasty. They had a significantly shorter operative time in the transperitoneal group however, the author in the discussion mentioned that he was at the start of the learning curve for retroperitonoscopic pyeloplasty when he conducted his study, which affected the result of the operative time. Hence, as mentioned above, we stressed the importance of a single surgeon with adequate equal experience in both techniques. The recovery of the intestinal motility and start of oral feeding were significantly faster in the retroperitoneal group compared to the transperitoneal group. In our opinion this can be explained by the absence of intraperitoneal manipulations and urine leakage in the peritoneal space. In their series of retroperitoneal pyeloplasty, El Ghoneimi et al. reported feeding after a mean of 1.4 days, however, in our series there was even earlier oral feeding. Shouma et al. reported no significant difference in the start of oral feeding in their adult series. The limitations of our study are: the choice of the 40 min difference created a statistically significant difference in operative time between the groups which might not be considered a truly clinically important difference. In addition, the single author operating for both approaches, which might create a bias, however the author has sufficient experience in both approaches. Moreover, although there were significant differences in hospital stay and intestinal movement between the two groups, it is not clear if these were of clinical significance. Conclusion Both transperitoneal and retroperitoneal approaches have high success rate. The shorter operative time,
Median age (years)
Median operative time (minutes)
Median hospital stay (hours)
Start oral feeding (hours)
6 5 0.437
150 129 0.010
48 24 0.002
16 10 0.000
Discussion Our series is the first in the literature that compares in a prospective randomized design the
shorter hospital stay, rapid recovery of intestinal movement and early resumption of oral feeding are in favor with the retroperitoneal approach.
http://dx.doi.org/10.1016/j.jpurol.2014.11.019 1477-5131/Published by Elsevier Ltd on behalf of Journal of Pediatric Urology Company.
Please cite this article in press as: Badawy H, et al., Transperitoneal versus retroperitoneal laparoscopic pyeloplasty in children: Randomized clinical trial, Journal of Pediatric Urology (2015), http://dx.doi.org/10.1016/j.jpurol.2014.11.019
+
MODEL
1.e2
Introduction Laparoscopic pyeloplasty achieves equal success rates to open pyeloplasty with regard to minimal morbidity, rapid recovery, and better cosmesis [1]. Both transperitoneal and retroperitoneal approaches are used with equal success rates [2e4]. However, there is a lack of randomized studies with a prospective design comparing the two approaches [5]. The present study was a prospective, randomized trial comparing the two approaches in children to test if the retroperitoneal approach, compared to the transperitoneal approach, has a shorter operative time, a more rapid recovery of intestinal movement, a shorter length of hospital stay and a comparable overall success rate.
Materials and methods In the present, prospective randomized study, 38 children aged 2 years or older (25 boys, 13 girls) were enrolled between June 2010 and September 2012. Approval was obtained from the University of Alexandria’s ethical committee and parental informed consent was received. Children were randomized using closed envelope randomization into two equal arms: transperitoneal approach and retroperitoneal approach. Malrotated kidneys, renal fusion anomalies and redo cases were excluded from the study. A stented dismembered pyeloplasty was performed on all of the children by the same surgeon (first surgeon). Anterior transposition of the UPJ was performed in the presence of a crossing vessel. Follow-up was undertaken every 3 months by ultrasound, and DTPA renogram in persistent hydronephrosis. A standard anesthesia protocol was used in all children as follows: all children received premedication; midazolam at a dose of 0.05 mg/kg bodyweight through an intravenous catheter; propofol at a dose of 2.0e2.5 mg/kg bodyweight; a muscle relaxant in the form of cisatracurium at a dose of 0.15 mg/kg bodyweight; and analgesia in the form of fentanyl at a dose of 2.0 Ugm/kg bodyweight. All children were mechanically ventilated after insertion of an appropriately sized endotracheal tube. A nasogastric tube was inserted, and intraoperative monitoring was performed with a pulse oximeter, non-invasive blood pressure monitor, and an electrocardiogram; end tidal carbon dioxide (ETCO2) was monitored through a capnogram. In both approaches, the children were positioned in the lateral flank position. In the transperitoneal group, open access was established through the umbilicus; a 5-mm port was inserted and insufflation was maintained at 12 mmHg, then two 3-mm trocars were inserted under vision, one midway between the anterior superior iliac spine and the umbilicus, and the other midway between the xiphisternum and the umbilicus. In the retroperitoneal group, open access was established by a 0.5e1.0 cm incision in the midaxillary line below the last rib; gerota fascia was grasped and opened under vision; a 5-mm trocar was inserted and secured in place by a stitch in the sheath; insufflation was maintained at a pressure of 12 mm Hg; then, two 3-mm trocars were inserted, one in the costovertebral angle and the other in the anterior axillary line one finger above the iliac crest.
H. Badawy et al. In both approaches, the basic principles of pyeloplasty were followed: dismembering of the UPJ, trimming of the renal pelvis, spatulation of the ureter after excision of the aperistalsis segment, anastomosis of the ureter to the pelvis by a 6/0 vicryl suture, and antegrade insertion of the double J (DJ) in all cases. The drain was left inside, through one of the ports, until it stopped leaking urine. Both groups were compared regarding the following intraoperative anesthetic changes: heart rate, blood pressure, ETCO2 changes, PH changes, O2 saturation, and urine output. The operative time was calculated from the insertion of the first optic trocar until the end of the procedure. The recovery of bowel movement and toleration of oral feeding were compared in the two arms. Length of hospital stay (in hours) was calculated from the day of admission until discharge. Intraoperative and postoperative complications, as well as overall success rates (disappearance of the symptoms, regression of hydronephrosis, and increase in the renal parenchyma) in the follow-up period were also noted. Complications were reported according to the Clavien-Dindo classification system [6]. On the first operative day, an anesthetist assessed postoperative pain and analgesia every 4 h, and on the second day this was assessed every 6 h. Analgesia was established by giving all children paracetamol (Perfalgan) 15 mg/kg every 6 h, plus rescue analgesia in the form of the non-steroidal anti-inflammatory drug (NSAIDS) diclofenac sodium in a dose of 0.5e1.5 mg/kg every 8 h. Pain was assessed according to the pain score (Visual Analogue Scale in children older than 6 years and the Face, Legs, Activity, Cry, Consolability scale (FLACC) for younger children. Opioid analgesia was not required for any of the children. Based on previous studies of transperitoneal and retroperitoneal laparoscopic pyeloplasty, a mean difference in operative time of 40 min, an effect size of 0.95, an alpha of 0.05 and power of 80% were used [7]. Using an online sample size calculator, statistical analysis was performed using SPSS software using the Fischer exact test, Chisquared test and ManneWhitney U test. The minimum sample size that was required was calculated to be 19 for each arm. The operative time was deemed to be the primary endpoint for comparison between both approaches.
Results The two groups were comparable, with no significant differences between them for the mean age, weight, gender distribution and clinical presentation (Table 1). There was no significant difference between the two groups concerning the intraoperative anesthetic changes and urine output (as shown in the Supplementary Data). There were no significant differences in the amount of rescue NSAIDS analgesia used in the first and second postoperative days between the two groups (Table 2). The retroperitonoscopic approach was significantly shorter in operative time than the transperitoneal approach (P Z 0.010) (Table 3). The presence or absence of crossing vessels did not significantly affect the operative time, regardless of the approach utilized (P Z 0.961) (Table 3). The length of hospital stay was significantly shorter in the retroperitonoscopic group than the transperitoneal group
Please cite this article in press as: Badawy H, et al., Transperitoneal versus retroperitoneal laparoscopic pyeloplasty in children: Randomized clinical trial, Journal of Pediatric Urology (2015), http://dx.doi.org/10.1016/j.jpurol.2014.11.019
+
MODEL
Comparison between laparoscopic transperitoneal and retroperitoneal pyeloplasty in children. Table 1
Demographic data and clinical presentation.
Age (years) Median Weight (kg) Median
Sex Male Female Presentation Loin pain Febrile UTI Incidentally discovered Prenatal diagnosis with hydronephrosis
Table 2
Transperitoneal
Retroperitoneal
ManneWhitney
6.00
5.00
24.00 Transperitoneal N %
18.00 Retroperitoneal N %
14 5
73.7 26.3
11 8
57.9 42.1
X2 Z 1.05 P Z 0.305
15 2 2 0
78.9 10.5 10.5 0
10 5 1 3
52.6 26.3 5.3 15.8
X2 Z 5.619 P Z 0.132
Z Z 0.778 P Z 0.437 Z Z 0.556 P Z 0.578 Tests of significance
Non-steroidal anti-inflammatory drugs in the first two postoperative days.
NSAIDS first day in mg/d Median NSAIDS second day in mg/d Median
Transperitoneal
Retroperitoneal
25.00
25.00
37.50
37.50
ManneWhitney P Z 0.213 P Z 0.222
(P Z 0.002) (Table 3). A significantly more rapid recovery of bowel movement and toleration of oral feeding were found in the retroperitoneal group, compared to the transperitoneal group (Table 3). Significant urine leakage and, hence, delayed drain removal was present in the transperitoneal group, compared to the retroperitoneal group. The median duration of a drain in situ in the transperitoneal group was 48.00 h while in the retroperitoneal group it was 24.00 h (P Z 0.000).
Table 3
1.e3
No intraoperative complications were encountered, and no conversion to open surgery occurred. Postoperative complications (Grade III b according to the Clavien-Dindo classification) [6] in the form of urinomas occurred in two children in the transperitoneal group: (1) after 1 week, a perirenal urinoma that was percutaneously aspirated; and (2) after 10 days of discharge, one child presented with a fever, intraperitoneal collection and sepsis, which was
Operative time, length of hospital stay, and oral feeding.
Operative time (minutes) Median Minemax range Length of hospital stay (hours) Median Minemax range Recovery of bowel movement (hours) Median Minemax range Start of oral feeding (hours) Median Minemax range Operative time (minutes) Median Minemax range
Transperitoneal
Retroperitoneal
150.00 120e240 Transperitoneal
129.00 120e189 Retroperitoneal
48.00 24e72
24.00 24e72
14.00 9e20
7.00 6e12
16.00 12e22
10.00 7e14
Crossing vessels present
Crossing vessels absent
138.00 120e240
133.2 120e189
ManneWhitney Z Z 2.586 P Z 0.010 ManneWhitney P Z 0.002
Z Z 4.869 P Z 0.000 Z Z 4.945 P Z 0.000 ManneWhitney Z Z 0.51 P Z 0.961
Please cite this article in press as: Badawy H, et al., Transperitoneal versus retroperitoneal laparoscopic pyeloplasty in children: Randomized clinical trial, Journal of Pediatric Urology (2015), http://dx.doi.org/10.1016/j.jpurol.2014.11.019
+
MODEL
1.e4 treated with exploration and open drainage of an infected intrabdominal urine collection. The mean follow-up period was 10 months (range 6e24). The overall success rate was 94.7%. There were two redoes, one in each group: in the transperitoneal group, after 3 months, which was repaired by open pyeloplasty; in the retroperitoneal group, after 6 months, which was repaired by antegrade endopyelotomy.
Discussion The present study is the first prospective, randomized design in the literature that compares the transperitoneal and retroperitoneal laparoscopic pyeloplasty in children. In the present study, a single surgeon (the first author) performed both the transperitoneal and retroperitoneal laparoscopic pyeloplasties. The surgeon has adequate and equal experience in both transperitoneal and retroperitoneal approaches. This is an important feature in the study, as it alleviates any bias related to surgeon experience or learning curve. Shouma et al. conducted the only prospective, randomized study that compares both techniques in adult pyeloplasty. They had a significantly shorter operative time in the transperitoneal group; however, in the discussion, the author mentioned that he was at the start of his learning curve for retroperitonoscopic pyeloplasty when he conducted the study, which affected the results of the operative time for the retroperitoneal approach, making it longer. Hence, as mentioned above, the importance of a single surgeon with adequate equal experience in both techniques is very important [7]. In the present study, the operative time was significantly shorter in the retroperitoneal group (median 129.00 min) compared to the transperitoneal group (median 150.00 min) (P Z 0.010). Although these results are different to many other studies in the literature, the shorter operative time in the retroperitoneal approach in the present study could be explained by the direct access to the UPJ offered by this approach, in contrast to the transperitoneal approach where the colon must be reflected and the intestines must be manipulated to reach the target organ [6,8e10]. We decided not to use the transmesocolic approach which easily performed on the left side pyeloplasty but not on the right side pyeloplasty not to create any internal bias in the study groups. However, the use of the transmesocolic approach in further studies might create shorter operative times in the transperitoneal group e this needs to be explored in a further well-designed study. In their study, Shouma et al. pointed out that the presence of a crossing vessel is significantly related to increased operative times, as anastomosis is more difficult, especially in the retroperitoneal approach [7]. Contrary to their experience, in the present study, there was no increased difficulty in the presence of a crossing vessel in the retroperitoneal group. After dissection of the pelviureteric junction obstruction (PUJO) from the crossing vessel, the pelvis and ureter were anteriorly transposed to the crossing vessel, the renal pelvis was attached to the psoas muscle by a 5/0 vicryl suture, which stabilizes the renal pelvis and
H. Badawy et al. takes the crossing vessel upward and away from the surgeon, thus making the anastomosis easier and quicker. El-Ghoneimi et al. [4] encountered crossing vessels in nine children: the anterior transposition of the PUJ was performed and no significant prolongation of operative time was reported. In the present study, the DJ was inserted antegrade after finishing the anastomosis of the posterior wall. It is better to insert the DJ in a retrograde way because it decompresses the renal pelvis, making its dissection more difficult; moreover, it sometimes hinders the ureteral spatulation, especially in a narrow ureter. In a series of twenty-four children, Mandhani et al. pointed out that laparoscopic stenting is a simple technique, which obviates the need for any other procedure [11]. In a prospective study of 120 patients, Arumainayagam et al. reported shorter operative time in antegrade DJ stenting versus retrograde stenting (P < 0.0001) [12]. In the present series, oxygen saturation was not significantly different between the two approaches; however, in their retrospective series of seventy-three children operated by the retroperitoneal approach, Halachmi et al. reported a decrease in oxygen saturation recorded after CO2 insufflation [13]. In the present study, hemodynamic and blood gas changes were not significantly different between the transperitoneal and retroperitoneal groups. However, Halachmi et al. reported a significant decrease in heart rate in the transperitoneal group (113 18) versus (116 19) beat per minute (P Z 0.019) after CO2 insufflation [13]. In the present study, there was no difference in the mean ETCO2 between the transperitoneal and retroperitoneal groups. This is in contrast to Karsli et al. who had a significant difference between the transperitoneal and retroperitoneal groups [14]. The analgesic requirement in the present study was not significantly different between the two groups. Paracetamol (Perfalgan) was given in a fixed dose and frequency, according to bodyweight, in the postoperative period. The only factor that could have made a difference between the two groups, concerning the overall dose of Perfalgan, would have been a significant difference in the weight of the children between the groups, which was not the case in the present study where there was a comparable weight between the groups. Rescue analgesia, in the form of NSAIDS in a dose of 0.5e1.5 mg/kg, was given according to the scheduled regular evaluation of pain using validated scores (VAS for children older than 6 years and FLACC for younger children); the anesthetist continuously evaluated and adjusted the doses. Because NSAIDS were used according to the pain scale, the two groups could be compared using an objective parameter of pain intensity rather than the weight of the children, however, no statistically significant difference could be found between the groups. The recovery of bowel movement and start of oral feeding were significantly faster in the retroperitoneal group, compared to the transperitoneal group. It is believed that this is due to the absence of intraperitoneal manipulations and urine leakage in the peritoneal space. El-Ghoneimi et al. reported quicker feeding in their group of retroperitoneal children after a mean of 1.4 days [4]; however, in the present study this was faster. Shouma et al.
Please cite this article in press as: Badawy H, et al., Transperitoneal versus retroperitoneal laparoscopic pyeloplasty in children: Randomized clinical trial, Journal of Pediatric Urology (2015), http://dx.doi.org/10.1016/j.jpurol.2014.11.019
+
MODEL
Comparison between laparoscopic transperitoneal and retroperitoneal pyeloplasty in children. reported no significant differences in the start of oral feeding in their adult series [7]. In the present study, drain removal was significantly quicker in the retroperitoneal versus the transperitoneal group. This might be explained by the closed retroperitoneal space, which allows the suture line to seal after desufflation and prevents urine leakage. Shouma et al. found no differences in the time of removal of the drain between the two groups [7]. In the present study, the length of hospital stay was significantly shorter in the retroperitoneal group, compared to the transperitoneal group. This might be attributable to the rapid recovery of intestinal movement, early start of oral feeding, and earlier removal of the drain. The median hospital stay was shorter compared to that of the series by Singh et al. [15], and Bonnart et al. [16]. In the present study, neither intraoperative complications nor conversion to open surgery occurred. In their retrospective series of 83 patients, Davenport et al. reported 17 retroperitonoscopic pyeloplasties, two of which required open surgery due to failure to progress [1]. In a series of 49 children by Canon et al., 27 underwent retroperitonoscopic pyeloplasty and 19 underwent transperitoneal laparoscopic pyeloplasty. Three cases of transient prolonged urine leakage were reported e two in the retroperitoneal group and one in the transperitoneal group, and six cases undergoing balloon dilatation were reported e five in the retroperitoneal group and one in the transperitoneal group [8]. In the present study, there were no intraoperative complications encountered, and no conversion to open surgery occurred. Two children in the transperitoneal group had postoperative complications in the form of urinomas, one after 1 week, and the other after 10 days of discharge. The study had several limitations: the choice of the 40 min difference created a statistically significant difference in operative time between both groups, which might not be considered to be a clinical difference; the single surgeon operating with both approaches, which might have created bias, however, it was previously stated that he had sufficient experience in both approaches. Moreover, although the length of hospital stay and intestinal movement were statistically significant different between the two approaches, it is unclear as to whether there is any clinical significance.
Conclusion Both the transperitoneal and retroperitoneal approaches had a high success rate. However, the shorter operative time, shorter length of hospital stay, rapid recovery of intestinal movement, and early resumption of oral feeding are in favor of the retroperitoneal approach.
Ethical approval Number of IRB acceptance: 0101570.
Source of funding None declared.
1.e5
Conflict of interest None declared.
Acknowledgment Dr Moataza Abdelwahab (Department of statistics, medical research institute, University of Alexandria, Egypt) for her outstanding effort in performing the statistics and analysis of the study.
Appendix A. Supplementary data Supplementary data related to this article can be found online at http://dx.doi.org/10.1016/j.jpurol.2014.11.019.
References [1] Davenport K, Minervini A, Timoney AG. Our experience with retroperitoneal and transperitoneal laparoscopic pyeloplasty for pelvi-ureteric junction obstruction. Eur Urol 2005;48(6):973e7. [2] Tan HL. Laparoscopic Anderson-Hynes dismembered pyeloplasty in children. J Urol 1999;162(3 Pt 2):1045e7. [3] Yeung CK, Tam YH, Sihoe JD, Lee KH, Liu KW. Retroperitoneoscopic dismembered pyeloplasty for pelvi-ureteric junction obstruction in infants and children. BJU Int 2001; 87(6):509e13. [4] El-Ghoneimi A, Farhat W, Bolduc S, Bagli D, McLorie G, Aigrain Y, et al. Laparoscopic pyeloplasty by retroperitoneal approach in children. BJU Int 2003;92:104e8. [5] Wu Y, Dong Q, Han P, Liu L, Wang L, Wei Q. Meta-analysis of transperitoneal versus retroperitoneal approaches of laparoscopic pyeloplasty for ureteropelvic junction obstruction. J Laparoendosc Adv Surg Tech A 2012;22(7):658e62. [6] Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240(2): 205e13. [7] Shoma MA, El Nahas AR, Bazeed MA. Laparoscopic pyeloplasty: a prospective randomized comparison between the transperitoneal approach and retroperitoneoscopy. J Urol 2007; 178:2020e4. [8] Canon JS, Jaynathi VR, Lowe GJ. Which is betterretroperitoneoscopic or laparoscopic dismembered pyeloplasty in children? J Urol 2007;178(4):1791e5. [9] Inagaki T, Rha KH, Ong AM, Kavoussi LR, Jarrett TW. Laparoscopic pyeloplasty: current status. BJU Int 2005;95:102e6. [10] Janetschek G, Peschel R, Franscher F. Laparoscopic pyeloplasty. Urol Clin North Am 2000;27:695e704. [11] Mandhani A, Bhandari SM. Is antegrade stenting superior to retrograde stenting in laparoscopic pyeloplasty? J Urol 2004; 177:1440e2. [12] Arumainayagam N, Minervini A, Davenport K, Kumar V, Masieri L, Serni S, et al. Antegrade versus retrograde stenting in laparoscopic pyeloplasty. J Endourol 2008;22(4):671e4. [13] Halachmi S, El-Ghoneimi A, Bissonnette B, Zaarour C, Bagli DJ, McLorie GA, et al. Hemodynamic and respiratory effect of pediatric urological laproscopic surgery: a retrospective analysis. J Urol 2003;170:1651e4. [14] Karsli C, El-Hout Y, Lorenzo AJ, Langer JC, Ba ¨gli DJ, Pippi Salle JL, et al. Physiological changes in transperitoneal versus retroperitoneal laparoscopy in children: a prospective analysis. J Urol 2011;69:1649e52.
Please cite this article in press as: Badawy H, et al., Transperitoneal versus retroperitoneal laparoscopic pyeloplasty in children: Randomized clinical trial, Journal of Pediatric Urology (2015), http://dx.doi.org/10.1016/j.jpurol.2014.11.019
+
MODEL
1.e6 [15] Singh H, Ganpule A, Malhotra T, Manohar T, Muthu V, Desai M. Transperitoneal laparoscopic pyeloplasty in children. J Endourol 2007;21(12):1461e7.
H. Badawy et al. [16] Bonnart A, Fouquet V, Carricaburu E, Aigrain Y, El-Ghoneimi A. Retroperitoneal laparosopic versus open pyeloplasty in children. J Urol 2005;173:1710e4.
Please cite this article in press as: Badawy H, et al., Transperitoneal versus retroperitoneal laparoscopic pyeloplasty in children: Randomized clinical trial, Journal of Pediatric Urology (2015), http://dx.doi.org/10.1016/j.jpurol.2014.11.019
