+ MODEL

Journal of Pediatric Urology (2014) xx, 1e5

Robot-assisted laparoscopic ureteralreimplantation: A single surgeon comparisonto open surgery

John L. Schomburg, Ken Haberman,Katie H. Willihnganz-Lawson, Aseem R. Shukla*

Section of Pediatric Urology, University of Minnesota Amplatz Children’s Hospital, Minneapolis,MN, USA

Received 30 August 2013; accepted 25 February 2014

KEYWORDSVesicoureteral reflux;Robot-assistedlaparoscopy;Ureteralreimplantation

* Corresponding author. The Children215 590 7889; fax: þ1 215 590 3985.

E-mail address: shuklaa@email.cho

Please cite this article in press as: Schto open surgery, Journal of Pediatric

http://dx.doi.org/10.1016/j.jpurol.201477-5131/ª 2014 Published by Elsevi

Abstract Objective: The aim was to report a single surgeon’s experience comparing openand robot-assisted laparoscopic extravesical ureteral reimplantation (RALUR) to treat vesi-coureteral reflux (VUR).Subjects and methods: We retrospectively reviewed the outcomes of RALUR and open extrave-sical ureteral reimplantations consecutively performed by a single surgeon between January2008 and December 2010 using the da Vinci� Surgical System. Both groups of patients were sub-jected to identical pre- and postoperative care protocols.Results: During the defined study interval, 20 open and 20 RALUR procedures were completedby a single surgeon at our institution. Gender and VUR grade were similar in both cohorts.Operative times were longer in the RALUR group, but postoperative opioid use (morphineequivalents) was significantly lower in the RALUR group (RALUR: 0.14 mg/kg, open:0.25 mg/kg, p Z 0.021). There was no significant difference in estimated blood loss (EBL) orlength of hospitalization (LOH). The overall rate of surgical complications was similar; howev-er, the complications in the open group tended to be less severe than those occurring in theRALUR group. On follow-up, after a median of 52 months for open surgery and 39 monthsfor RALUR, two children had developed a febrile urinary tract infection in both groups, ofwhich one in the open group had persistent VUR.Conclusion: This single-surgeon experience of open and initial experience with RALUR per-formed with the same surgical technique on consecutive cohorts with identical post-surgical

’s Hospital of Philadelphia, 34th Street and Civic Center Blvd., Philadelphia, PA 19104, USA. Tel.: þ1

p.edu (A.R. Shukla).

omburg JL, et al., Robot-assisted laparoscopic ureteral reimplantation: A single surgeon comparisonUrology (2014), http://dx.doi.org/10.1016/j.jpurol.2014.02.013

14.02.013er Ltd on behalf of Journal of Pediatric Urology Company.

2 J.L. Schomburg et al.

+ MODEL

Please cite this article in press as: Schto open surgery, Journal of Pediatric

care protocol allows a comparative analysis of outcomes for a surgeon transitioning to RALUR.The RALUR reduces postoperative analgesic requirements while yielding similar clinical out-comes as the open technique.ª 2014 Published by Elsevier Ltd on behalf of Journal of Pediatric Urology Company.

Introduction

It is estimated that 1e3% of children, and up to 30% of thosewith urinary tract infections, suffer from vesicoureteralreflux (VUR) [1]. Left untreated, symptomatic VUR can be arisk for recurrent pyelonephritis and its associatedmorbidity. While several treatment alternatives exist forVUR, the gold standard surgical alternative for clinicallysignificant VUR remains open surgery with either an intra-vesical or extravesical ureteral reimplantation. With thegoal of reducing the morbidity associated with open sur-gery, laparoscopic and robot-assisted laparoscopic (RAL)procedures in children have been described as viableminimally invasive alternatives to open surgery for manypediatric urologic conditions. For various procedures, pa-tients undergoing RAL are experiencing shorter hospitalstays, decreased pain medication requirements, andimproved cosmesis when compared to patients who un-derwent open surgery [2,3].

The laparoscopic extravesical ureteral reimplantationwas first described in 1994 [4], and was later shown toproduce similar results as the open technique whilereducing postoperative discomfort [5]. However, thistechnique was not widely adopted as the approach istechnically challenging [6]. The same technique was lateradapted for use with the surgical robot, overcoming manyof those challenges. Initial experience with the robot-assisted laparoscopic ureteral reimplantation (RALUR)showed high rates of VUR resolution with few complica-tions, including a low rate of urinary retention [2,7].However, these early feasibility studies did not include acontrol group.

By comparing a single surgeon’s experience with theopen and RAL approaches to the extravesical ureteralreimplantation, we sought to compare treatment outcomesand the feasibility of transitioning to RALUR in a pediatricurology practice.

Materials and methods

Patient selection

A total of 20 patients underwent RALUR by a single surgeon(AS) at our institution during the defined study period(January 2008 to December 2010). A group of 23 consecu-tive patients undergoing extravesical ureteral reimplanta-tion by the same surgeon during the same time period wasidentified; three patients with previous or concomitantintra-abdominal surgery were excluded. All children un-dergoing unilateral or bilateral intravesical ureteral reim-plants during the same time period were also excluded fromanalysis.

omburg JL, et al., Robot-assistedUrology (2014), http://dx.doi.or

Study design and data collection

After obtaining approval from the University of MinnesotaInstitutional Review Board, patient data were obtainedretrospectively from the electronic medical record. Pre-operative evaluation included voiding cystourethrogram(VCUG) to assess the grade of VUR as well as renal ultra-sound or nuclear medicine scan to evaluate for renal scar-ring, atrophy and differential function. While nuclearrenography was obtained preoperatively in some patients,there was no absolute criteria for consideration for surgery.Parent preference played a key role in decision makingprior to surgery. Every child in the open and RALUR groupshad experienced at least one febrile urinary tract infectionprior to surgery, and all children had VUR present on VCUG.While adjunctive studies such as a DMSA scan were obtainedfor patients preoperatively, there were no absolute criteriaby DMSA prior to consideration for surgery. Parent prefer-ence played a key role in decision-making process prior tosurgery. All parents were appropriately counseled regardingall management options, including observation with orwithout prophylactic antibiotics, endoscopic injection ofbulking agents, or ureteral reimplantation.

Since the adoption of the RALUR at our institution in2008, the procedure is offered to all patients consideredcandidates for ureteral reimplantation over 18 months ofage. Children under that age are typically offered only theopen approach due to surgeon preference. An openapproach is also preferred in cases of a ureterocele or likelyneed for ureteral tapering, but those patients were notincluded in this analysis.

Operative times represent total in-room time for bothopen and RALUR groups, including port placement, robotdocking, and console times for the RALUR group. EBL was asrecorded in the operative reports. Pain medication datawere collected from inpatient medication administrationreports. The majority of patients were given hydrocodoneorally as needed and acetaminophen. Rare patients weregiven intravenous morphine, intravenous dilaudid, oribuprofen orally. Total narcotic administered was calcu-lated and was converted using a correction factor (0.5 �mgof intravenous hydrocodone or 6.66 � mg of intravenousdilaudid) to IV morphine equivalents, which was thenstandardized for patient weight. Non-narcotic pain medi-cation was also included in the analysis. As all pain medi-cations were prescribed as needed, the total number ofpain medication dosages of any type was also totaled foreach patient. No patients received an epidural or patientcontrolled analgesia.

Follow-up data were obtained from the electronicmedical record. While the only way to definitively confirmresolution of reflux is with postoperative VCUG, the retro-spective nature of this study precludes this as it is not our

laparoscopic ureteral reimplantation: A single surgeon comparisong/10.1016/j.jpurol.2014.02.013

Table 1 Patient demographics.

Robot Open p Value

No. patients 20 20M:F ratio 1:9 1:9 1Mean age at surgery (mo) 74 52 0.034Mean weight (kg) 27.22 18.22 0.043Unilat:bilat ratio 1:3 3:7 0.330Mean VUR grade:

Overall 2.84 3.13 0.230Unilateral 3.07 3.21 0.613Bilateral 2.50 3.04 0.200

No. with duplicated system 5 6 0.716No. with previous treatment 1 1 1

Robot-assisted laparoscopic ureteral reimplantation 3

+ MODEL

standard practice to obtain VCUG in each patient followingsurgery. Since the success of the open and RALUR tech-niques we utilized were previously confirmed both inter-nally and in previous reports [7] with routine postoperativeVCUGs, a VCUG was offered to all families, but obtainedpostoperatively only upon family preference or in instanceswhere patients developed a febrile urinary tract infectionfollowing surgery. Operative success was therefore definedas absence of febrile urinary tract infections; where VCUGwas obtained, success was defined by radiographic evi-dence of VUR resolution. Complications were identifiedfrom the medical record and graded with the Clav-ieneDindo classification system [8].

Statistical analysis was performed using Microsoft Excel.Data were analyzed using paired, two-tailed t tests. Sta-tistical significance was defined as p < 0.05.

Surgical technique

Our RALUR technique is similar to what has previously beendescribed in the literature [7], using the same fundamentalprinciples utilized for an open extravesical reimplantation,with the exception that a transperitoneal approach is usedto access the ureter and bladder. Cystoscopy is performedonly when ectopic ureteral insertion is suspected. For therobotic camera port, an 8 or 10 mm port is inserted justinferior to the umbilicus. Two 5 mm robotic ports areplaced on the right and left midclavicular line approxi-mately 1 cm inferior to the umbilicus. No fourth, assistantport, is placed. The da Vinci� Surgical System is docked instandard fashion.

The ureters are identified distal to the round ligament infemales and vas deferens in males, and mobilized to theirinsertion in the bladder. Care is taken to minimize the useof electrocautery lateral to this insertion point.

The detrusor is incised and separated using sharpdissection and intermittent monopolar cautery with theendowrist cautery hook, leaving the mucosa intact. Adetrusor incision to ureteral diameter ratio of 4 or 5:1 ismaintained. To aid in elongating the detrusor trough duringdetrusorrhaphy, a 2.0 polypropylene suture on a Keithneedle (or straightened SH needle for smaller patients) ispassed suprapubically into the peritoneal space and used tohitch the bladder anteriorly. A series of simple interrupted5-0 polydioxanone stitches are used to close the detrusorover the ureter, after which the retraction suture isrelaxed.

Postoperative course

Patients in both groups had identical access to ad libitumanalgesia as needed throughout the inpatient postoperativeperiod. All patients left the operating room with anindwelling urinary catheter. The catheter was removed onthe morning of postoperative day 1. Post-void residualbladder volume was measured in all patients and ensured tobe less than 20% of expected based on bladder capacityprior to discharge. All patients underwent renal ultrasoundat 1 month. Patients remained on antibiotic prophylaxis 1month postoperatively until the renal ultrasound wascompleted.

Please cite this article in press as: Schomburg JL, et al., Robot-assistedto open surgery, Journal of Pediatric Urology (2014), http://dx.doi.or

Results

Patient demographics and clinical features are presented inTable 1. Five patients in the RALUR and six in the opengroup underwent bilateral reimplantation. Gender distri-bution was identical between groups. There was a signifi-cant difference in age at surgery (RALUR: 74 months, Open:52 months, p Z 0.034) and perioperative weight (RALUR:27.2 kg, Open: 18.2 kg, p Z 0.042). VUR grade and inci-dence of duplicated systems were similar between groups.One patient in each group had previously failed treatmentwith endoscopic injection.

Operative details are presented in Table 2. There wereno intraoperative complications in either group. Operativetimes were longer for both unilateral and bilateral opera-tions in the RALUR groups than in the open groups (Unilat-eral: RALUR: 165 min, Open: 109 min, p < 0.001, Bilateral:RALUR: 227 min, Bilateral: 135 min, p < 0.001) (Fig. 1). Themean actual in-console operative time for the RALUR wasrecorded as 55 min for unilateral, and 94 min for bilateralRALUR. EBL, days with Foley catheter, and length of hos-pital stay were similar between all groups.

There was a significantly lower postoperative narcoticanalgesic (RALUR: 0.14 mg/kg, Open: 0.25 mg/kg,p Z 0.021) and total postoperative as needed analgesicdose (RALUR: 4.0, Open: 7.5, pZ 0.020) requirement in theRALUR group as presented in Table 3 and Fig. 2. One patientin the RALUR group and seven patients in the open groupreceived non-narcotic pain medication in the form ofketorolac. To account for these, the total number of asneeded doses of any analgesic was calculated. The use ofnon-narcotics was significantly lower in the RALUR group(RALUR: 4.00 doses, Open 7.50 doses, p Z 0.020).

Follow up data are presented in Table 4. There were twofebrile urinary tract infections in each group post-operatively. A single case of persistent VUR was diagnosedin the open group, and 12 patients (60%) undergoing VCUGin the RALUR did not have persistent VUR, though the pa-tients in the open group have been observed over a longermedian time period (RALUR: 39 months, Open: 52 months,p < 0.001). Surgical complications as classified by theClavieneDindo classification system [8] are presented inTable 5. While there was no significant difference in theoverall rate of surgical complications, the complications in

laparoscopic ureteral reimplantation: A single surgeon comparisong/10.1016/j.jpurol.2014.02.013

Table 2 Operative and postoperative outcomes.

Robot Open p Value

Mean procedure time (min)Unilateral 165 109 <0.001Bilateral 227 134.5 <0.001

Estimated blood loss (mL) 4.5 8.1 0.144Unilateral 4.3 4.6 0.367Bilateral 5.2 16.2 0.093

Days with urinary catheter 0.85 1.05 0.297Length of stay (days) 1.05 1.4 0.308Postoperative urinary retention 1 3 0.330Persistent VUR 0 1 0.165Mean follow-up (mo) 13 27 <0.001

Table 3 Postoperative analgesic requirements.

Robot Open p value

Intravenous morphineequivalent dose (mg)

3.30 4.02 0.207

Perioperative weight 27.22 18.22 0.043Morphine equivalent

per kg body weight0.14 0.25 0.021

Total as needed doses of analgesia 4.00 7.50 0.020

4 J.L. Schomburg et al.

+ MODEL

the open group tended to be less severe than thoseoccurring in the RALUR group.

Discussion

While open surgery remains the gold standard definitivetherapy for VUR, there has been increased interest inminimally invasive therapy for VUR, including endoscopicsubureteral injections and minimally invasive surgery [6].While the traditional laparoscopic approach has beenshown to be as effective as open surgery [5], it has not beenwidely adopted due to the degree of technical challenge.With improved dexterity, motion scaling, and otherinherent benefits, RAL has taken a larger role in pediatricsand has renewed interest in minimally invasive surgery forthe correction of VUR.

Early experience with the technique did show promisingresults [2,7], which have been supported by recent com-parison studies [6,9].

Our results are consistent with those of Smith et al. [9],demonstrating decreased postoperative analgesic re-quirements in the RALUR group. In contrast to that study,and perhaps reflecting a practice preference at our insti-tution, we found no significant difference in length ofpostoperative hospital stay. However, Smith et al. did notcompare identical operations as the open procedure wasperformed intravesically and the RALUR procedure was

Figure 1 Operative times were longer for both unilateral andbilateral operations in the RALUR groups than the open groups(Unilateral: RALUR: 165 min, Open: 109 min, p < 0.001,Bilateral: RALUR: 227 min, Bilateral: 135 min, p < 0.001).

Please cite this article in press as: Schomburg JL, et al., Robot-assistedto open surgery, Journal of Pediatric Urology (2014), http://dx.doi.or

performed extravesically. And, as such, potentially othercontributing variables beyond the varied surgical approachcontributed to different outcome metrics.

Marchini et al. [6] compared matched cohorts of bothintravesical and extravesical RAL and open procedures intheir study. They demonstrated similar clinical results interms of VUR resolution and operative times, and did note ashorter duration of hospital stay and fewer bladder spasmsin the RALUR group, but only for the intravesical procedure.No differences in outcomes or hospital parameters in theextravesical cohorts were observeddcorroborated by ourown experience. Marchini et al. also did not report post-operative analgesic requirements for either group, whileour study interrogated our electronic medical record tocalculate analgesic equivalents administered.

While our study specifically examined postoperativenarcotic use in our postoperative pathway, ketorolac is anoptional adjunct which we feel would likely decreasenarcotic use for both open and RALUR groups. Futurepostoperative pain protocols will include ketorolac forselect cases without renal dysfunction.

The present study allows an isolated comparison of RALand open techniques for the extravesical ureteral reimplantas we compared technically identical procedures per-formed using either approach by the same surgeon. Similarto previous studies, we have shown that operative timeswith RALUR are significantly longer than those for the openprocedure. In our experience, the bilateral procedure re-quires a nearly twofold robotic console time differential(Unilateral: 55 min, Bilateral: 94 min), suggesting that the

Figure 2 There was a significantly lower postoperativenarcotic analgesic (RALUR: 0.14 mg/kg, Open: 0.25 mg/kg,p Z 0.021) and total postoperative as needed analgesic dose(RALUR: 4.0, Open: 7.5, p Z 0.020) requirement in the RALURgroup.

laparoscopic ureteral reimplantation: A single surgeon comparisong/10.1016/j.jpurol.2014.02.013

Table 4 Follow-up data.

Robot Open p value

Median time since surgery (ms) 39 52 <0.001Postoperative febrile urinary

tract infection2 2 1

Persistent reflux 0 1 0.165

Robot-assisted laparoscopic ureteral reimplantation 5

+ MODEL

bulk of the time difference occurs as a result of time toperform the operation itself and cannot be attributedsolely to robot set-up time. While it is likely that there willbe an improvement in operative time with increasedexperience with the RALUR technique, the transitioningRALUR surgeon should expect to invest significantly moretime for the procedure than the open technique. Further-more, at our institution, and as reported before [10], wehave instituted a resident robotic training curriculum wherea stepwise progression to intracorporeal suturing is fol-lowed. As such, we find that actual surgical console trainingtimes remain static over time as incoming residents on thepediatric urology service continue to begin surgical trainingat the same point on the learning curve.

Notably, our study did not show any difference in EBL,days with urinary catheter, length of hospitalization, orrates of urinary retention between groups. It should also benoted that one of the suggested advantages of minimallyinvasive surgery is improved cosmesis, which is difficult toassess and was not specifically addressed in this study.

There was no difference in the overall rate of compli-cations between RALUR and open groups. However, theseverity of complications in the open group tended to beless acute than those occurring in the RALUR group. Bothhigh-grade (Clavien III) complications in the RALUR group,at an earlier point in our experience, were likely related tomanipulation of the distal ureter as one patient developeda urine leak and the second developed ureteral stenosis. Asa result, the RALUR operative technique was modified toeliminate the percutaneous ureteral hitch stitch for pa-tients later in the series as well as incorporation of ureteraladventitia with each detrusorrhaphy suture pass. Whilefurther experience at our institution confirms that suchmajor complications are rare and have not recurred, it isimportant to note that the risk of complications with RALURnot only includes those associated with the open technique,but also the added risks of peritoneal access andinsufflation.

Our study is limited by its retrospective design and re-mains vulnerable to observer bias. The operative surgeon inthis study counseled each family undergoing the RALUR oropen procedure, and the selection criteria for eitherapproach were not randomized or prospectively

Table 5 Postoperative complications.

Robot Open p value

Complications 2 7 0.083Clavien grade

IeII 0 4 0.042III 2 1 0.330

Please cite this article in press as: Schomburg JL, et al., Robot-assistedto open surgery, Journal of Pediatric Urology (2014), http://dx.doi.or

determined. The lower median age for patients undergoingthe open reimplant suggests a bias towards utilizing theRALUR for older childrendan age preference seen early inour RALUR experience. And while the surgeon confirmedthe feasibility and successful outcomes in eradicating VURwith the RALUR in a pilot institutional study prior to thestudy initiation, the absence of standard VCUGs being ob-tained in all patients necessarily limits absolute efficacycomparisons.

Conclusions

Our study demonstrates that RAL extravesical ureteralreimplantation reduces postoperative analgesic re-quirements while yielding similar clinical outcomescompared to the open extravesical ureteral reimplantation.In this single-surgeon study, the technique utilized andpostoperative care protocol were uniform for two consec-utive cohort series, allowing an isolated comparison ofoutcomes as related to surgical technique. Future studiesmust include randomized protocols with larger cohorts,analysis of patient and family preferences as well as out-comes, and extensive cost comparisons of the twotechniques.

Conflict of interest

None.

Funding

None.

References

[1] Stansfeld JM. Clinical observations relating to the incidenceand aetiology of urinary-tract infections in children. Br Med J1966;1:631.

[2] Peters CA. Robotically assisted surgery in pediatric urology.Urol Clin N Am 2004;31:743e52.

[3] Lee D, Kim P, Koh C. Current trends in pediatric minimallyinvasive urologic surgery. Korean J Urol 2010;51:80e7.

[4] Ehrlich R, Gershman A, Fuchs G. Laparoscopic vesicoureter-oplasty in children: initial case reports. Urology 1994;43:255e61.

[5] Lakshmanan Y, Fung L. Laparoscopic extravesicular ureteralreimplantation for vesicoureteral reflux: recent technicaladvances. J Endourol 2000;14:589e93.

[6] Marchini G, Hong Y, Minnillo B, Diamond D, Houck C, Meier P,et al. Robotic assisted laparoscopic ureteral reimplantation inchildren: case matched comparative study with open surgicalapprroach. J Urol 2011;185:1870.

[7] Casale P, Patel R, Kolon T. Nerve sparing robotic extravesicalureteral reimplantation. J Urol 2008;179:1987e90.

[8] Dindo D, Demartines N, Clavien P. Classification of surgicalcomplications. Ann Surg 2004;240:205e13.

[9] Smith R, Oliver J, Peters C. Pediatric robotic extravesicalureteral reimplantation: comparison with open surgery. J Urol2011;185:1876.

[10] O’Brien S, Shukla A. Transition from open to robotic-assistedpediatric pyeloplasty: a feasibility and outcome study. JPediatr Urol 2012;8:276e81.

laparoscopic ureteral reimplantation: A single surgeon comparisong/10.1016/j.jpurol.2014.02.013