OtBstMtmatacRrsdo1Cor

Recrevte

FA

CD

2

ADULT UROLOGY

9

COMPARISON OF MODIFIED TAGUCHI AND BRICKERURETERAL REIMPLANTATION TECHNIQUES AFTER

RADICAL CYSTECTOMY

CHERYL T. LEE, BERT T. CHEN, EDWARD GONG, KHALED S. HAFEZ,JACQUELINE H. SHEFFIELD, AND JAMES E. MONTIE

ABSTRACTbjectives. To present our experience with the modified Taguchi “single-stitch” ureteral reimplantation

echnique in patients undergoing radical cystectomy with urinary diversion compared with a traditionalricker reimplantation technique. Improved techniques are continually sought for ureteroenteric anastomo-es during urinary diversion. The modified Taguchi “single-stitch” ureteral reimplantation is reportedly aime-efficient technique that preserves anastomotic integrity.ethods. We retrospectively examined 75 consecutive patients with bladder cancer who underwent cys-

ectomy and urinary diversion between October 1, 1999 and March 31, 2001. The ureteroenteric anasto-osis was performed using a reinforced single-stitch modified Taguchi technique in the first 36 patients andn interrupted two-layer Bricker technique in the subsequent 39 patients during the creation of 47 ortho-opic neobladders and 28 ileal conduit diversions. Ureteral stents were not routinely used. The demographicnd perioperative clinical parameters were evaluated in each cohort, with particular attention to ureteralomplications.esults. Modified Taguchi and Bricker ureteral anastomoses were performed in 48% and 52% of patients,espectively. Patient age, sex, and body mass indexes were similar between groups. Apart from pathologictage, univariate analysis did not demonstrate statistically significant differences between the groups in theemographic, intraoperative (estimated blood loss, diversion type, operative time) or postoperative (lengthf stay, rate of complications) parameters. Ureteral complications occurred in 8% of the Bricker group and5% of the modified Taguchi group (P � 0.23).onclusions. Either technique can be performed safely and in a timely fashion. However, the increased numberf ureteral leaks in the modified Taguchi cohort combined with no advantage in procedure time, prompted oureturn to the Bricker technique exclusively. UROLOGY 64: 940–944, 2004. © 2004 Elsevier Inc.

tr

stTtamuTdclrc

adical cystectomy is the most definitive treat-ment for high-risk bladder cancer. However,

xcellent local cancer control is accompanied by aomplication rate of 20% to 40%, with 2% to 10%elated to ureteroenteric reconstruction.1–5 In anffort to reduce the morbidity of the procedure,arying techniques in ureteroenteric reconstruc-ion have been explored to increase the ease andfficiency of ureteral reimplantation while main-

rom the Department of Urology, University of Michigan, Annrbor, MichiganReprint requests: Cheryl T. Lee, M.D., Department of Urology,

omprehensive Cancer Center 7303, 1500 East Medical Centerrive, Ann Arbor, MI 48109-0946Submitted: March 8, 2004, accepted (with revisions): July 6,

r004

© 2004 ELSEVIER INC.40 ALL RIGHTS RESERVED

aining anastomotic patency and integrity compa-able with established techniques.6,7

MacKinnon et al.8 described a double-mattresstitch technique for extravesical ureteroneocystos-omy in cadaveric renal transplantation in 1968.his technique, and a single-stitch (SS) modifica-

ion by Schanfield9 in 1972, have proved to be valu-ble methods for transplant ureterovesical anasto-osis, resulting in low rates of urinary leakage,reteral stricture, or hematuria.8–12 The modifiedaguchi SS technique has been applied to urinaryiversion after radical cystectomy for bladder car-inoma in only a small number of patients.13,14 Pre-iminary data have suggested that the technique iseliable without added complications, although noomparisons with traditional techniques have been

eported in this population.

0090-4295/04/$30.00doi:10.1016/j.urology.2004.07.005

peinTtn

d(wpmrmisgfittawue

mbptoaas

stp

au5arpsiaoipvVis1utbp

cAcrlvtc

t(dapr

FsS ia to

U

In the current pilot study, we compared our ex-erience with a modified SS technique with ourxperience with a traditional Bricker anastomosisn patients undergoing radical cystectomy and uri-ary diversion. We hypothesized that the modifiedaguchi technique would have a similar complica-

ion profile as that of the traditional Bricker tech-ique, but would be easier to perform.

MATERIAL AND METHODS

During a 9-month period, radical cystectomy and urinaryiversion were performed by one experienced surgeonJ.E.M.) in 75 consecutive patients (61 men and 14 women)ith carcinoma of the bladder. The patient demographics anderioperative parameters were retrospectively reviewed. Theean age and follow-up time was 64.8 years and 21.5 months,

espectively. The indications for radical cystectomy includeduscle invasive cancer, recurrent superficial or early-stage

nvasive tumors, and refractory carcinoma in situ. The deci-ion to perform urinary diversion was determined by the sur-eon. An SS type of ureteral anastomosis was performed in therst 36 patients and a Bricker anastomosis was performed inhe following 39 patients, without randomization. Postopera-ively, patients were surveyed every 4 months for the first yearnd every 6 months for the next 4 years. Upper tract imagingas obtained 4 to 6 weeks after cystectomy and then annuallynless clinical signs and symptoms dictated more frequentvaluation.

The clinical parameters evaluated were age, sex, race, bodyass index, operative time, length of hospital stay, estimated

lood loss, diversion type, pathologic stage, and ureteral com-lication rate. The organ-confined pathologic stage includedumors with a final stage of Ta, Tis, T1, or T2 without evidencef nodal or metastatic spread; non-organ-confined tumors hadfinal stage of T3, T4, N�, or M�. All tumors were staged

ccording to the 2002 American Joint Committee on Cancertaging system.15

In both techniques, the left ureter is tunneled under theigmoid mesentery during conduit construction and throughhe mesentery for orthotopic reconstruction, taking care to

IGURE 1. Traditional Bricker ureteroenteric anastomopatulated ureter to corresponding enterotomy. Rougecond layer of anastomosis apposes ureteral adventit

reserve the sigmoid and ureteral blood supply. Both ureters i

ROLOGY 64 (5), 2004

re laid in close proximity to the ileal segment, trimmed, spat-lated medially, and opposed to the ileal segment using two-0 Prolene sutures. In the Bricker technique, the ureteroilealnastomosis is performed in an end-to-side fashion with inter-upted 5-0 Monocryl suture (Fig. 1).16 A 5F feeding tube isassed proximally in each ureter to identify any proximal ob-truction or kinking and is subsequently removed. In the mod-fied SS technique, a double-armed 3-0 Vicryl suture is placedt the distal aspect of the spatulated ureter (Fig. 2A). Each armf the suture is then passed through a small enterotomy in annside-to-outside fashion approximately 1 to 2 cm from theroximal enterotomy apex (Fig. 2A). The ureter is then ad-anced into the enterotomy and fixed in position with 3-0icryl suture (Fig. 2B). In both techniques, an additional re-

nforcing second layer is created using interrupted 5-0 Proleneuture to oppose the ureteral adventitia and ileal serosa (Figs.B and 2C). The ileal segment is irrigated to ensure watertightreteroileal anastomoses. Ureteral stents are only used rou-inely for solitary kidneys or periureteral fibrosis after externaleam radiotherapy. Transabdominal Jackson-Pratt drains areositioned in the pelvis and can detect anastomotic leakage.Statistical analysis was performed using standard, commer-

ially available computer software (Statview, SAS Institute).17

ssociations between the type of ureteral anastomosis andategorical (sex, race, pathologic stage, ureteral complicationate, diversion type) and continuous (age, estimated bloodoss, body mass index, operative time, length of hospital stay)ariables were assessed using the Student t test and chi-squareest, respectively. Tests with a P value of less than 0.05 wereonsidered statistically significant.

RESULTS

The Bricker and SS ureteroileal anastomoticechnique was performed in 39 (52%) and 3648%) patients, respectively. Orthotopic neoblad-er reconstruction and ileal conduit diversion wasccomplished in both the Bricker (BR) (23 and 16atients, respectively) and SS (24 and 12 patients,espectively) cohorts. The demographic character-

(A) End-to-side anastomosis created by approximatingight sutures used to complete this anastomosis. (B)seromuscular ileum with 5-0 Prolene suture.

sis.hly e

stics of the entire population and each anasto-

941

mplaqhdh

(msg

att2auetupfestbmlruthFSi

ma

FieP

P

MS

MMR

M

MD

P

MU

K

9

otic group are provided in Table I. The meanatient age, sex distribution, body mass index, fol-

ow-up time, and race were similar between the BRnd SS groups. Neoadjuvant chemotherapy was re-uired in 3 patients (2 BR and 1 SS); 2 patients hadad prior chemotherapy in combination with ra-iotherapy (1 BR and 1 SS). Although the SS group

IGURE 2. Modified SS Taguchi ureteroenteric anastomng spatulated ureter to corresponding enterotomy usinnterotomy. (C) Reinforcing second layer of anastomosisrolene suture.

TABLE I. Demographics andperioperative parameters

arameterBricker(n � 39)

Single Stitch(n � 36)

PValue

ean age (yr) 64.4 65.3 0.69ex (n)Male 33 28 0.45Female 6 8ean BMI 27.1 27.8 0.47ean follow-up (mo) 20.8 22.2 0.50aceWhite 32 29 0.65Other 7 7ean operative time(min)

284 278 0.65

ean EBL (mL) 1087 1251 0.38iversion type (n)Ileal conduit 16 12 0.54Orthotopic neobladder 23 24

athologic stage (n)Organ confined 28 20 0.03Non-organ confined 11 16ean LOS (days) 7.8 8.4 0.63reteral complications (n)Ureteral stricture 3 4 0.23Ureteral leak 0 2

EY: BMI � body mass index; EBL � estimated blood loss; LOS � length of stay.

ad a greater rate of non-organ-confined disease t

42

44%) than the BR group (28%; P � 0.03), theean operative time and estimated blood loss were

imilar between the two cohorts (Table I). Bothroups had a mean length of hospital stay of 8 days.We observed ureteral complications in 3 BR (8%)

nd 6 SS (15%) patients, respectively. Patients inhe SS group experienced 4 cases of ureteral stric-ure within 1.6 to 4.0 months after cystectomy andcases of ureteral anastomotic leaks, occurring at 4nd 11 days after cystectomy. Three of the distalreteral strictures were managed successfully withndoscopic balloon dilation with good patency af-er stent removal; one required open revision. Bothrine leaks were managed with percutaneous ne-hrostomy drainage and each resolved withouturther incident. Patients in the BR group experi-nced no ureteral leaks but had 3 cases of ureteraltricture within 0.5 to 3.0 months after cystec-omy. One stricture was managed successfully withalloon dilation, one failed endoscopic manage-ent requiring open revision, and one underwent

eft nephroureterectomy for upper tract recur-ence. No statistically significant difference in thereteral complication rate was observed betweenhe SS and BR cohorts (P � 0.23). Both groups alsoad a similar overall complication rate (P � 0.14).ive (83%) of the six ureteral complications in theS group and two (67%) of three in the BR groupnvolved the left ureteroileal anastomosis.

COMMENT

Radical cystectomy remains the standard treat-ent for high-risk bladder cancer.1–4 The associ-

ted urinary diversion requires ureteral reimplan-

. (A) End-to-side anastomosis created by approximat-e double-armed suture. (B) Distal ureter advanced intooses ureteral adventitia to seromuscular ileum with 5-0

osisg onapp

ation, which is most often accomplished with a

UROLOGY 64 (5), 2004

utmabuptmwbpnwrmb

mohcLnpap8TsBittqs

tdnarotugtubt

tawicwt

tmitstoru(usvatsotsast

tplaiteurtftpucloi

uwtqutuetel

rmw

U

reteroenteric anastomosis. The two-layered anas-omosis popularized by Bricker has been the soleethod of ureteral anastomosis used by the senior

uthor (J.E.M.) for more than 20 years, in partecause ureteral leaks with this method are quitencommon.5 Although reliable, potential im-rovements exist. Efforts to diminish the operativeime related to our standard Bricker-style anasto-osis led to a trial of an alternative technique thatould be dependable, easy to perform, and haveroad applicability. Owing to the encouraging ex-erience from Europe, the modified Taguchi tech-ique was explored because of its SS approach,hich offered technical ease at a potentially more

apid pace without added complications. An addedodification (reinforcing sutures) was used to sta-

ilize the distal ureter (Fig. 2C).The modified Taguchi technique has been usedost often for ureteroneocystostomy in the setting

f renal transplantation in which large case seriesave demonstrated little difference in ureteralomplication rates compared with the Politano-eadbetter and multiple-stitch extravesical tech-iques, all with rates of 6% to 8%.11 Likewise, com-arable rates were observed between the Taguchind Lich-Gregoir techniques in a similar patientopulation with ureteral complication rates of.8%.10 To our knowledge, the use of the modifiedaguchi SS technique for ureteroenteric anastomo-es after cystectomy has been limited. Gutierrezanos and colleagues14 reported their experience

n a cohort of 20 patients undergoing Studer ortho-opic ileal reconstruction. Their ureteral complica-ion profile included one ureteral leak that re-uired reimplantation and an episode of ureteraltenosis that resulted in renal atrophy.The current study compared a modification of

his technique with a standard anastomotic proce-ure. Although we appreciate the limitations of aonrandomized and retrospective study, we wereble to examine two cystectomy cohorts that wereeasonably similar in their demographic and peri-perative parameters. These groups included pa-ients who underwent continent and incontinentrinary diversions. The increased rate of non-or-an-confined pathologic stages in the SS popula-ion was unlikely to have had any impact on thereteral complication rate or the operative time,ecause the increased stage was not appreciated byhe surgeon intraoperatively.In our experience, the rate of ureteral complica-

ions with the SS technique of 15% did not offerny advantage over that of the Bricker technique,hich had a rate of 8%. We remain cautious in our

nterpretation of the lack of a statistically signifi-ant difference between these small groups. Stentsere not used routinely with either ureteral anas-

omotic technique in this study, per the usual prac- r

ROLOGY 64 (5), 2004

ice of the operating surgeon. Whether stentsight have altered the results is speculative. There

s little evidence in published studies to suggesthat stents significantly decrease the rate of ureteraltricture, which ranges from 2.4% in stented pa-ients to 4.7% in unstented patients.18 The impactf stenting on ureteral leakage is unclear. Olderetrospective evidence has implied that stentedreters may have a greater incidence of leakage11%) after cystectomy compared with unstentedreteroenteric anastomoses (5%).19 Others haveuggested a reduced rate of leakage in stented (0%)ersus unstented (2.5%) patients after ureteroilealnastomoses.18 Still, a recent randomized clinicalrial demonstrated that routine stenting did notignificantly reduce the rate of ureteral obstructionr leakage in patients undergoing renal transplan-ation and ureteroneocystostomy.20 In the lattertudy, the rate of ureteral obstruction (stricture)nd leakage was 2.1% and 1.4% in those routinelytented and 2.3% and 4.4% in those without rou-ine stenting, respectively.In the current study, two ureteral leaks were de-

ected in the SS group and none in the BR cohort,ossibly owing to the nature of the technique. Botheaks resolved with percutaneous nephrostomynd urinoma drainage without the need for openntervention or stenting. It was not clear whetherhe reinforcing sutures used in the SS had any ben-fit in reducing ureteral leakage. The incidence ofreteral obstruction was similar in both groups,equiring endoscopic balloon dilation and even-ual open revision in 1 patient in each cohort. Ef-orts to preserve renal function and ureteral pa-ency were successful in most cases, although 1atient did lose a renal unit owing to malignantpper tract disease. Ultimately, the rates of ureteralomplications did not result in a significantly pro-onged hospital stay for the SS group, because fivef the six ureteral complications occurred after thenitial hospital discharge.

Most ureteral complications occurred at the leftreteroileal anastomosis, which may be consistentith general experience. Typically, adequate ure-

eral mobilization and mesenteric windowing is re-uired to maintain a sufficient length of the leftreter for anastomosis without tension. The SSechnique may necessitate a greater length of freereter to allow advancement of the ureter into thenterotomy. A lack of additional length might con-ribute to added tension during anastomosis andxplain the increased rate of complications at theeft ureter.

The perceived advantage in operative time expe-ienced by others was not realized in our series, aseasured by the overall procedure times, whichere similar between both groups. Secin et al.10

eported the Taguchi technique was significantly

943

fmIettbpoldpttcidtfpp

anetttdfect

tS

ii

t

d1

tt

t

mdt

p1

dJ

t6

Tn

ot

Mk1

Ucn

Sn

C3

t

ul

tJ

u

9

aster than the Lich-Gregoir ureterovesical anasto-osis, averaging 14.2 minutes versus 29 minutes.

n the current study, we did not record the uret-roenteric anastomotic time; instead, we recordedhe total procedure time. This limited specific in-erpretation of the anastomotic performance time,ut focused on a more practical endpoint, the totalrocedure time. Unlike renal transplantation, theperative time for radical cystectomy remainsengthy because of the multiple steps required, in-ependent of ureteral reconstruction. With a totalrocedure time averaging more than 4 hours, theime required for ureteral reimplantation is cer-ainly not the rate-limiting step for open radicalystectomy. Still, the timesaving benefit of one stepn the overall procedure will have less utility if itoes not translate into a benefit in the total opera-ive time. One must also be cautious about saving aew minutes during a very important portion of therocedure that might translate into a serious com-lication.

CONCLUSIONS

In our experience, a modified SS ureteroentericnastomosis can be performed in an effective man-er in the radical cystectomy population. How-ver, we did not observe an improvement over theraditional Bricker ureteral anastomosis and, withhe potentially greater rate of ureteral complica-ions, we concluded that the modified SS techniqueoes not provide a significant advantage. There-ore, we have returned to the Bricker techniquexclusively for postcystectomy ureteroenteric re-onstruction because of its reliability and ex-remely low rate of anastomotic leakage.

REFERENCES1. Bhojwani AG, and Mellon JK: Contemporary cystec-

omy combined with ileal conduit or bladder substitution.urg Oncol 11: 65–75, 2002.

2. Stein JP, Lieskovsky G, Cote R, et al: Radical cystectomyn the treatment of invasive bladder cancer: long-term resultsn 1,054 patients. J Clin Oncol 19: 666–675, 2001.

3. Figueroa AJ, Stein JP, Dickinson M, et al: Radical cys-

ectomy for elderly patients with bladder carcinoma: an up- d

44

ated experience with 404 patients. Cancer 83: 141–147,998.

4. Amling CL, Thrasher JB, Frazier HA, et al: Radical cys-ectomy for stages Ta, Tis and T1 transitional cell carcinoma ofhe bladder. J Urol 151: 31–35, 1994.

5. Montie JE, and Wood DP Jr: The risk of radical cystec-omy. Br J Urol 63: 483–486, 1989.

6. Muraishi O, Yamashita T, Ishikawa S, et al: Improve-ent of ureteroileal anastomosis in orthotopic ileal neoblad-

er with modified Le Duc procedure: short submucosal tunnelechnique. J Urol 165: 798–801, 2001.

7. Sagalowsky AI: Further experience with split-cuff nip-le ureteral reimplantation in urinary diversion. J Urol 159:843–1844, 1998.

8. MacKinnon KJ, Oliver JA, Morehouse DD, et al: Ca-aver renal transplantation: emphasis on urological aspects.Urol 99: 486–490, 1968.

9. Schanfield I: New experimental methods for implanta-ion of the ureter in bladder and conduit. Transplant Proc 4:37–638, 1972.10. Secin FP, Rovegno AR, Marrugat RE, et al: Comparing

aguchi and Lich-Gregoir ureterovesical reimplantation tech-iques for kidney transplants. J Urol 168: 926–930, 2002.11. Hakim NS, Benedetti E, Pirenne J, et al: Complications

f ureterovesical anastomosis in kidney transplant patients:he Minnesota experience. Clin Transplant 8: 504–507, 1994.

12. Gutierrez Banos JL, Portillo Martin JA, Correas GomezA, et al: Single-stitch extravesical ureteroneocystostomy in

idney transplantation (Taguchi’s technique). Acta Urol Esp8: 569–572, 1994.13. Gutierrez Banos JL, Portillo Martin JA, Martin GB, et al:

reteral reimplantation with single point on tubularized ilealhimney in Studer’s substitution enterocystoplasty: prelimi-ary results in 12 cases. Arch Esp Urol 47: 122–127, 1994.14. Gutierrez Banos JL, Martin GB, Hernandez RR, et al:

ubstitutive orthotopic ileocystoplasty: the Studer’s tech-ique. Arch Esp Urol 48: 783–790, 1995.15. Greene FL, Page DL, Fleming ID, et al (Eds). AJCC

ancer Staging Manual. New York, Springer-Verlag, 2002, pp67–373.16. Bricker EM: Bladder substitution after pelvic exentera-

ion. Surg Clin North Am 30: 1511–1521, 1950.17. Statview. SAS Institute Inc., version 5.0.1, 1992–1998.18. Regan JB, and Barrett DM. Stented versus nonstented

reteroileal anastomoses: is there a difference with regard toeak and stricture? J Urol 134: 1101–1103, 1985.

19. Richie JP, Skinner DG, and Kaufman JJ: Radical cystec-omy for carcinoma of the bladder: 16 years of experience.Urol 113: 186–189, 1975.20. Dominguez J, Clase CM, Mahalati K, et al: Is routine

reteric stenting needed in kidney transplantation? A ran-

omized trial. Transplantation 70: 597–601, 2000.

UROLOGY 64 (5), 2004