1

Extravesical Ureteroneocystostomy With and Without Internalized Ureteric Stents in Pediatric Renal Transplantation

Christopher G FrenchPhilip D Acott

John FS CrockerHinrich Bitter-Suermann

Joseph G Lawen

Department of Urology and PediatricsIWK-Grace Health Center

Dalhousie University Halifax, Nova Scotia

Canada

2

French CG, Acott PD, Crocker JFS, Bitter-Suermann H and Lawen JG. Extravesical

ureteroneocystostomy with and without internalized ureteric stents in pediatric renal

transplantation.

Pediatr Transplantation

Abstract: The use of ureteric double-J stents and the Lich-Gregoir (extravesical)

technique of ureteroneocystotomy have both been shown to decrease the rate of urologic

complications in adult kidney transplantation. There are few studies of systematic use of

stents in the pediatric renal transplantation. Between 1991 and 1997 thirty-two

consecutive pediatric renal transplants recipients routinely received a 6F-12cm

indwelling double-J stent and were studied prospectively. These patients were compared

to 32 consecutive immediately-prior pediatric recipients in whom a stent was not used.

The latter were transplanted between 1987 and 1991 and formed the control group. All

patients had a Lich-Gregoir ureteroneocystotomy. Stents were removed with general-

anesthetic cystoscopy 2-3 weeks after transplantation. Immunosuppression for stented

patients was polyclonal antibody induction, delayed (7-10days) cyclosporine,

azathioprine and prednisone. The control group received the same triple drug regimen

with no induction in 29/32. All patients were followed with at least 1 ultrasound

3

evaluation in the first month and repeat ultrasound and renal scan were performed if there

was any rise in serum creatinine.

In the stented group there were 2 patients with urinary leak and no obstructions. In the

non-stent group there were no leaks and 1 obstruction. There was no graft loss due to

urologic complications in either group. There were 3 cases of stent expulsion (all in girls)

and one case of stent migration in the posterior urethra (boy). The 1-year graft survival

rate was 90.6% in the stented group and 65.6% in the non-stent group.

The prophylactic use of an indwelling ureteral stent in pediatric renal transplantation did

not reduce the risk of urinary leakage or obstruction. Stent migration is a common

phenomena and while not a serious complication is traumatic to children. Furthermore

removal of internalized double J stent requires a general anesthetic. We recommend using

a stent for selected cases only.

Key Words: stenting ureter- pediatric renal transplantation- urinary leakage-

complications

Corresponding Author:

Dr. Joseph G. Lawen MD, FRCSC Associate Professor and Chair of Renal Transplantation Dept. of Urology, Faculty of MedicineDalhousie UniversityRm.294 5 South, Victoria Building1278 Tower Road, HalifaxNova Scotia, CanadaB3H 2Y9Email: lawen@is.dal.caFax (902) 473-5850 Tel. (902) 473-5469

4

5

GLOSSARY OF ABBREVIATIONS

CsA Cyclosporine

ATG Anti-thymocyte Globulin

ALG Anti-lymphocyte Globulin

HLA-DR Human Leukocyte Antigen – DR locus: Class II MHC antigen

DTPA Diethylenetriaminepentaacetic acid

MAG-3 Mercaptoacetyltriglycine

ATN Acute Tubular Necrosis

6

INTRODUCTION

At our institution we have performed over 140 pediatric renal transplants since 1972. The

preferred method for ureterovesical anastomosis has been the modified Lich-Gregoir

extravesical technique 1. In the early 1990’s with cyclosporine (CsA) the mainstay of

immunosuppression there remained in our institution as elsewhere a sizable discrepancy

between adult and pediatric renal graft survival 2,3,4. Against this background we

implemented a number of proposals in an effort to improve renal graft function in

pediatric recipients. These included 1) designation of 2 surgeons with special interest in

pediatric kidney transplantation, 2) refusal of single-kidneys from cadaveric donors under

the age of six, 3) adoption of sequential quadruple induction immunosuppression with

ALG/ATG, 4) refusal of cadaveric kidneys from donors that did not have at least 1 HLA-

DR match, 5) adoption of routine prophylactic use of double-J ureteric stent. The

outcomes of these changes in our program have been previously reported 5. Here we

evaluate the impact of routine ureteric stenting in pediatric kidney transplantation.

Ureteric complications, leaks or obstructions, make up more than 90% of urologic

problems after transplantation and can lead to significant morbidity and mortality6.

Ureteric obstructions may occur because of twists, kinks and technical difficulties with

the anastomosis, while leaks occur because of ischemic or mechanical injury to the ureter

as well as technical difficulties with the anastomosis. Excessive ureteric length may

contribute to the complication rate in so much that it predisposes the anastomosis to

ischemia and increases the risk of mechanical mishaps. Ureteric complication rates

7

ranged between 2% - 21% in series of adult kidney transplantation 7,8,9,10,11. Studies

comparing extravesical and intravesical ureteroneocystostomy have shown that there is a

significant reduction (9.4% vs. 3%) in urologic complications when the extravesical

technique is used. Intravesical ureteroneocystostomy results in ureterovesical junction

obstruction in 3.7% compared with 0.6% in the extravesical group 12,13. Extravesical

ureteroneocystostomy had long been the preferred method of ureteric implantation both

in adult and pediatric transplantation.

In the early 1990s in our adult kidney transplant program the non-stented ureteric

complication rate remained relatively steady at about 6% and seemed resistant to further

reductions (unpublished data). On the other hand, in the non-transplant setting the

prophylactic use of stents in ureteric surgery has been shown to reduce complications and

was relatively well established. Following early reports showing decreased urologic

complications in the transplant population, prophylactic stenting with deliberate ureteric

trimming was adopted in our adult and pediatric renal transplant programs. The aim of

this analysis was to identify the advantages and disadvantages of prophylactic ureteric

stenting unique to pediatric kidney transplantation. We prospectively analyzed the

urologic complications and graft survival of pediatric kidney transplant recipients

managed with a stented extravesical ureteroneocystostomy compared to those

transplanted immediately before the change in technique and had not received a stent.

8

PATIENTS AND METHODS

This prospective historical control study included 64 consecutive pediatric renal

transplant recipients from January 1987 to September 1997. The study started in October

of 1991.

Stented Group

After 1991 two designated surgeons performed nearly all kidney transplants.

Transplanted kidneys were from live donors or cadavers over the age of 6 with at least 1

HLA-DR match. All pediatric renal transplant recipients received a 6F-12cm double-J

indwelling ureteric stent (Cook, Indiana). Ureters were trimmed to provide for a tension-

free non-redundant anastomosis. No other change in surgical technique was effected

compared to the non-stent era. Our intention was to remove the stent between 1-2 weeks

after transplantation. This was based on the assumption that maximal tissue edema occurs

within 72 hrs postoperatively and ischemic ureteric necrosis occurs most frequently

within the first 2 weeks. However the stent was left in longer with complications such as

delayed graft function or medical conditions making administration of general anesthetic

inappropriate. Alternatively, if bothersome irritative urinary symptoms or infection

occurred then the stents were removed as soon as possible beyond the first week.

Non-stented Group:

9

Thirty-two consecutive patients receiving a kidney transplant immediately before

the implementation of routine stenting and deliberate trimming of excess ureter. They

were transplanted between 1987 and 1991 and formed the historical control group. These

patients were transplanted by a group of four surgeons, received some kidneys from

donors under the age of 6 years and from cadaveric donors with less than 1 HLA-DR

match.

Standard Procedures:

A modified Lich-Gregoir extravesical anastomosis was used in all cases and the

anastomosis was performed with 5-0 or 6-0 running absorbable suture. Perioperative

antibiotics were used in all patients. All patients had a renal scan within the first month

and again within 6 months post operatively. If there was any rise in serum creatinine then

a renal scan and biopsy were performed. Foley catheter was routinely removed on the

first post-operative day or delayed if the urinary volumes were difficult to manage by

normal voiding. The incidence of urinary infection was not compared between the two

groups. Pyuria and urinary symptoms in the presence of an indwelling ureteric stent is

very common, thus rendering these criteria less useful in the diagnosis of a urinary tract

infection and complicating the comparison of the two groups.

10

Immunosuppression:

The stented group received sequential quadruple induction therapy with ATG/

ALG, with CsA being started on day 7, azathioprine and prednisone. The non-stented

group received CsA, azathioprine and prednisone induction, although patients also

received polyclonal anti-lymphocyte induction

Ureteric complications:

Ureteric complications were defined as leak or obstruction. Urinary leak was

defined as extravasation of radionuclide on renal scan, or perirenal / periureteral fluid

collection with a creatinine value >3 times that of serum, or peritoneal fluid >3 times that

of serum. Obstruction was defined as hydronephrosis on ultrasound in association with

perfusion, excretion and delayed drainage on furosemide DTPA or MAG-3 renal

scan.

11

RESULTS

Demographics:

The mean ages in months were 125 and 129 for the stented and non-stented

groups respectively (Table 1). The youngest was 21 months in the stented group and 9

months in the control group. . Our stented group had a higher number (19/32) of living

related donors (LRD) than the control (12/32). The causes of renal failure were equally

distributed in both groups. Medical causes consisted of glomerulonephritis, hemolytic

uremic syndrome and renal vein thrombosis. Congenital or structural causes included

dysplasia, Bardet-Biedel-Moon syndrome, prune belly syndrome, posterior urethral

valves, vesicoureteral reflux and megacystis-megaureter. The mean length of time of

stenting was 17 days. The one-year patient and graft survival were calculated using

actuarial data.

Ureteric and stent complications:

In the stented group there were 2 ureteric leaks (table 3). The first was in a 14-

year-old girl with megacystis megaureter syndrome. She received of a pair of 4 antigen

mismatched pediatric en-bloc kidneys from a 23-month-old donor who died of Klebsiella

pneumonia sepsis. Initial graft function was good and serum creatinine fell to 0.55 mg/dl.

She developed wound dehiscence at 12 days post-op and the wound fluid had a creatinine

12

level of 7.1 mg/dl, indicating urinary contamination. An ultrasound showed a perirenal

collection of nearly 400cc. Open exploration was performed due to the need for wound

closure. On exploration of the transplant no leak was found by inspection after methylene

blue instillation intravesically, confirming an intact bladder. Leakage was presumed to be

anastomotic and stented ureteral reimplantations were performed for both ureters. The

stents were removed 18 days after the second operation and the serum creatinine

remained stable at 0.68 mg/dl six months later. The second patient was a 14-year-old boy

with Bardet-Biedel-Moon syndrome who received a 5 antigen mismatched cadaveric

renal allograft and suffered latex-allergy anaphylaxis intraoperatively. The initial graft

function was poor and there was a drop in urine output at day 7. Renal ultrasound was

normal and no leak was detected on MAG-3 renal scan. The peritoneal dialysis catheter

fluid had a creatinine level >10 mg/ml associated with a serum creatinine of 1.9 mg/ml,

which indicated that urine had leaked into the peritoneal cavity. Renal biopsy failed to

demonstrate rejection and the patient was treated conservatively. The stent was removed

on the 18th postoperative day. The serum creatinine fell to 0.9 mg/dl after one month and

0.8 mg/dl after two months postoperatively. The patient developed recurrent fungal

urinary tract infections that were treated and eradicated prior to discharge from hospital.

The patient died of untreated gram-negative sepsis en route to hospital with a functioning

graft 9 months later.

There were 4 stent-related complications. Three girls had severe urgency,

frequency and dysuria related to stent migration. Two girls had expelled the stent partially

13

and one developed hematuria with the stent visible at the urethral meatus at time of

cystoscopy. One boy developed gross hematuria and at cystoscopy the stent had migrated

into the posterior urethra. There were no major complications related to these episodes.

One patient had transient obstruction immediately after stent removal. This occurred in a

6 year-old boy who received a 2 antigen mismatched cadaveric transplant with the stent

being removed 10 days post-op. He developed anuria and a renal ultrasound showed

hydronephrosis without ureteric dilatation consistent with ureteropelvic junction

obstruction. This was managed conservatively and diuresis ensued 24 hours later with no

evidence of hydronephrosis on subsequent scans. He had a stable functioning graft 3years

later.

In the non-stented group there were no leaks and only 1 obstruction. The

obstruction was in 13-year-old boy with congenital renal disease secondary to prune belly

syndrome who received a 1-antigen mismatched kidney from his sister. There was good

initial function and his serum creatinine dropped to 0.55 mg/dl. At 2 weeks post-op his

serum creatinine rose to 1.5 mg/dl and a renal ultrasound showed moderate

hydronephrosis while renal scan demonstrated poor emptying. He was treated

conservatively and the hydronephrosis improved, but continued to show mild

hydronephrosis. His serum creatinine at 1 year was 1.2 mg/dl and his graft is still

functioning at 7 years.

14

In the non-stented group early graft loss was significantly greater than in the

stented group, 34.4% and 9.4% respectively at one year (table 2). This might best be

explained by surgical, tissue matching and immunosuppressive strategies implemented

concurrently with the stenting change 5.

15

Discussion

Urologic complications (ureteric leak and obstruction) have significant morbidity

in renal transplantation. The major factors influencing the success of

ureteroneocystostomy include 1) the vascular integrity of the donor ureter and 2) the

anastomotic technique employed and 3) the avoidance of technical mishaps such as kinks

and twists of the ureter.

Pediatric transplantation differs from that in adults in primary cause of renal

failure and outcomes of cadaveric transplantation 8,11. Acute rejection is more common in

children and sequential quadruple induction therapy with anti-lymphocyte preparations is

routine. Vascular and ureteric problems remain a significant cause of morbidity. Children

tend to receive a higher proportion of live related donor kidneys in effort to offset some

of these increased risks. When urologic complications occur in children, auxiliary

instrumentation of the urinary tract has technical boundaries of size compared to adults,

especially in very young recipients.

Our results did not reveal a significant benefit of routinely placing an internalized

transanastomotic ureteric stent in pediatric renal transplant recipients. Given our low rate

of complications in the non-stented group this review could hardly show improvements.

These results contradict those of Bergmeijer and associates 14. They retrospectively

16

studied 36 pediatric renal transplant recipients in whom 18 had stents. All patients had a

Lich-Gregoir anastomosis with percutaneous ureteral stenting. They report 3 leaks and 3

obstructions in the non-stent group (33%) and one complication in the stented group, a

necrotic strictured ureter. They rationalized the complications in the non-stent group to a

number of causes: technical failure to create a non-obstructed water tight anastomosis,

failure to preserve the ureteric blood supply at the level of the renal hilum, dissection of

the periureteral tissues or ureteric injury. Initially the authors used an indwelling double-

J stent, but changed to a transvesical percutaneous stent principally for its ease of

removal. The small numbers and unusually high complication rate (33%) in the non-stent

group may explain the discrepancy between their findings and ours.

In our study the 1-year graft survival in the stented and non-stented groups were

90.6% and 65.6% respectively. Immunologic failures were the principle cause of graft

loss in both groups with 3/32 in the stented and 10/32 occurring in the non-stented

groups. This is related to improved immunosuppression with antibody induction (ATG/

ALG) in the stented group in addition to other factors 5. There were no graft losses due to

urologic complications.

There are possible explanations for the lower ureteric complication rate in the

pediatric age group; the relatively shorter ureter required compared to adults or the higher

number of living related donors in children. Distal ureteral perfusion is dependant on

hilar blood flow. Therefore, if there is marginal perfusion a shorter ureter is less likely to

17

become ischemic. The vascular anatomy in living related donors is well defined

preoperatively such that one can avoid kidneys with accessory lower pole arteries that

may jeopardize ureteric perfusion. The short cold ischemia time of living related

transplants might improve distal ureteric healing.

We have been using the modified Lich-Gregoir anastomosis for over 20 years in

both adult and pediatric renal transplants at our center. Recent studies confirm that this

technique has less morbidity (leak and obstruction) than other pull-through techniques 3.

Therefore the technique of anastomosis cannot be ignored when comparing complications

of stented and non-stented ureteroneocystostomy. This technique is versatile, easy to

perform, does not require a large cystostomy and requires a shorter ureter when compared

to transvesical approaches.

There may be advantages to stenting that do not show up in the analysis of

complications. There is no doubt that the primary benefit is the ease and precision of

performing the anastomosis. We have found that using stents assures mucosa-to-mucosa

anastomosis without causing inadvertent obstruction (fig.1). When trainees are learning

this technique, stents provide a more confident anastomosis. Additionally, an indwelling

stent excludes leak, obstruction or transient edema as a cause of early postoperative

oliguria. We have found that our threshold to do renal scans in children in the early

postoperative period is lowered. Other studies have shown that stents decrease the rate of

18

other urologic complications in adults but there remains a paucity of data in the pediatric

age group.

There are some disadvantages to stenting. These include the potential for an

increase in urinary tract infections. We did not quantify this, but one adult study showed

no difference 4. Although there is no added operative time in stenting, in children stent

removal requires a general anesthetic. This adds significant stress to the child and

additional cost in the provision of care. The physiological effects of a stent on the

transplant ureter are unclear. In-vivo studies using color Doppler ultrasound have

demonstrated diminished ureteral peristalsis associated with a stent. The degree of

hypoperistalsis is related to the length of time the ureter is stented 15. We theorize that this

explains the case of transient obstruction previously mentioned. Other complications of

ureteric stenting include ureteral wall musculature thickening and submucosal edema and

insertional trauma from ureteral perforation may occur when using guide wires. Stent

migration is common in the first two weeks post transplant manifest by irritative voiding

symptoms and hematuria 16.

We have decided not to use routine ureteric stents in all pediatric renal transplant

patients. Stenting was initiated in our study in an attempt to improve outcomes in our

pediatric recipients not because of urologic complications. Our experience suggests that

when the rate of complications is low such as in our historical controls, stenting provides

for no additional benefit. Stents will be placed when there are technical difficulties with

19

the anastomosis, any possibility of vascular compromise to the ureter. We recognize that

other series show improvement in urologic complications when stents are used. This may

be due to the technical superiority of the stented anastomosis itself not that the stent

improves healing or patency of the anastomosis. In conclusion, we believe that the

drawbacks of routinely placing an internalized double-J ureteric stent for an extravesical

ureteroneocystostomy in pediatric renal transplantation outweigh the benefits.

20

References:

1. Gibbons WS, Barry JM, Hefty TR. Complications following unstented parallel incision extravesical ureteroneocystostomy in 1000 kidney transplants. J Urol 1992: 148: 38-40

2. McEnery PT, Alexander SR, Sullivan K, Tejani A. Renal transplantation in children and adolescent: the 1992 report of the North American Pediatric Renal Transplant Cooperative Study (NAPRTCS). Pediatr Nephrol 1993: 7: 711-720

3. Hirata M, Terasaki PI. Pediatric Renal Transplantation. Clin Transplant 1994: 32: 395-402.

4. Tejani, AH, Fine RF. Pediatric Renal Transplantation. Wiley-Liss Inc. 1994: 337-348.

5. Crocker JF, Wade AW, McDonald AT, McLellan DH, Lawen JG, Bitter-Suermann H, Acott PD. Kidney graft loss in children: implications for program development. CMAJ 1998 Aug 11: 159(3): 229-35

6. Tan EC, Lim SM, Rauff A. Techniques of ureteric reimplantation in kidney transplantation and its related urological complications. Ann Acad Med Singapore 1991: 20: 524-528

7. Butterworth PC, Horsburgh T, Veitch PS, Bell PR, Nicholson ML. Urological complications in renal transplantation: impact of a change of technique. Br J Urol 1997: 79: 499-502

8. Eschwege P, Blanchet P, Bellamy J, Charpentier B, Jardin A, Benoit G. Does the use of double J ureteral stents reduce stenosis and fistulas in renal transplantation. Transplant Proc, 1995: 27: 2436

9. Pleass HC, Clark KR, Rigg KM, Reddy KS, Forsythe JL, Proud G, Taylor, RM. Urologic complications after renal transplantation: a prospective randomized trial comparing different techniques of ureteric anastomosis and the use of prophylactic ureteric stents. Transplant Proc 1995: 27: 1091-1092

10. Benoit G, Blanchet P, Eschwege P, Alexandre L, Bensadoun H, Charpentier B. Insertion of a double pigtail ureteral stent for the prevention of urological complications in renal transplantation: a prospective randomized study. J Urol 1996: 156: 881-884

21

11. Bassiri A, Amiransari B, Yazdani M, Sesavar Y, Gol S. Renal transplantation using ureteral stents. Transplant Proc 1995: 27: 2593-2594

12. Thrasher JB, Temple DR, Spees EK. Extravesical versus Leadbetter-Politano ureteroneocystostomy: A comparison of urologic complications in 320 renal transplants. J Urol 1990 144: 1105-1109

13. Hooghe L, Kinnaert P, Schulman CC, et al. Ureteroneocystostomy in renal transplantation: Comparison of endo- and extravesical anastomosis. World J Surg 1977: 1: 231-255

14. Bergmeijer JH, Nijman R, Kalkman E, Nauta J, Wolff ED and Molenaar JC. Stenting of the ureterovesical anastomosis in pediatric renal transplantation. Transplant Int 1990: 3: 146-148

15. Patel U, Kellett MJ. Ureteric drainage and peristalsis after stenting studied using color Doppler ultrasound. Br J Urol 1996: 77: 530-535

16. Culkin DJ. Complications of Ureteral Stents. Infect Urol 1996: 9: 141-146

22

Tables and Figures

Table 1: Demographics of Stented and Control Groups

Controls StentedNumber 32 32

Age(mo) Mean 129 125Range 212 174

Minimum 9 21

Sex F:M 10:22 18:14

Cause Of Renal Failure Medical 9 9Congenital/Structural 23 23

Stent Removed (days) Mean 17

23

Table 2: Patient and Graft survival and cause of failures in Stented and Controls

Controls Stented

Number 32 32

Donor Type LRD 12 19CAD 20 12

CAD-EnBloc 0 1

Patient survival (1 year) 30(93.8%) 31(96.9%)

Graft survival (1year) 21(65.6%) 29(90.6%)

Failures Immunological/Medical 9 2Technical/Vascular 2 1

Urologic 0 0____________________________________________________________________

24

Table 3: Major urologic complications in the stented and controls

Controls Stented

Complications Leak 0 2Obstruction 1 0Lymphocele 1 1

25

Fig1: Stented Lich-Gregoir ureteroneocystostomy. The stent greatly facilitates the

anastomosis.

26

Top

French CG, Lawen JG

Acott PD, Crocker JFS.

27

Dr. Joseph G Lawen MD FRCSC June 5, 2000Dept. of UrologyRm.294 5 South, Victoria Building1278 Tower Road, HalifaxNova Scotia, CanadaB3H 2Y9Email: Lawen@is.dal.caFax (902) 473-5850 Tel. (902) 473-5469

Pediatric Transplantation Manuscript # 00-08 Richard N. Fine, M.D.Editor in ChiefDepartment of PediatricsSUNY at Stony BrookStony Brook, New York 11794-8111, USA

Dear Dr Fine,

Thank you for reviewing our manuscript titled “Extravesical Neoureterocystostomy With and Without Internalized Ureteric Stents in Pediatric Renal Transplantation”. We appreciate your interest and the comments by the reviewers and have extensively revised the manuscript according to your comments.

These revisions include: a glossary of abbreviations, a comment in the methods about the decision process of how long to stent, a comment on why infections were not quantified, clarification on the clinical course of the 2 patients with leaks, correction of the timing of stent removal, revisions of grammar, Neoureterocystostomy was changed to Ureteroneocystostomy, CyA was changed to CsA, imuran was changed to azathioprine and the layout was completely revised for clarity. We also reviewed all of our data and there are 2 minor changes related to graft survival and we have included patient survival.

Enclosed are 3 copies of the revised manuscript and a diskette with the manuscript that has changes highlighted in MSWord.

In consideration of the Editors of the Pediatric Transplantation taking action in reviewing and editing this submission, the author’s undersigned hereby transfers, assigns, or otherwise conveys all copyright ownership to Pediatric Transplantation, in the event that such work is published in that Journal. All authors have read and comply with the requirements set forth in Information for Authors.

Sincerely,

Christopher G. French M.D. Joseph G. Lawen M.D. FRCSC