RISK OF DETRUSOR DENERVATION IN ANTIREFLUX SURGERYDEMONSTRATED IN A NEUROPHYSIOLOGICAL ANIMAL MODEL

FRANCISCO J. MARTINEZ PORTILLO,* CHRISTOPH SEIF, PETER M. BRAUN, GEORG BOHLER,DANIAR K. OSMONOV, JOACHIM LEISSNER, RUDOLF HOHENFELLNER, PETER ALKEN

AND KLAUS P. JUENEMANNFrom the Departments of Urology, University Hospital Kiel (FJMP, CS, PMB, GB, DKO, KPJ), Kiel, University Hospital Magdeburg (JL),

Magdeburg, University Hospital Mannheim, Mannheim (PA) and University Hospital Mainz (RH), Mainz, Germany

ABSTRACT

Purpose: Earlier anatomical studies have shown a close connection between the ureterovesicaljunction and detrusor innervation. It prompted us to develop an animal model to demonstrate therisk of partial or complete impairment of this neuronal connection during antireflux surgery.

Materials and Methods: Six female Gottinger minipigs were anesthetized and laminectomized.After placement of the S3 sacral nerves into separate electrode compartments of a modifiedBrindley electrode the lower urinary tract was exposed by an abdominal midline incision. Afterbladder instillation with 150 ml NaCl 1 bilateral and 2 unilateral stimulations (left and rightsides) were performed and intravesical pressure was recorded urodynamically. The left ureterwas then prepared circularly in 3 steps 10, 5 and 1 cm, respectively, proximal to the ureteroves-ical junction. After each preparation step bilateral and unilateral stimulation was repeated.Results were recorded urodynamically and video documented.

Results: Bilateral stimulation before preparation of the left ureter led to a concentric detrusorcontraction with an average maximum detrusor pressure of 51 cm H2O. Unilateral stimulationresulted in ipsilateralbound bladder tilting with an intravesical pressure of 18 and 19 cm H2O onthe right and left sides, respectively. After preparation of the left ureter 10, 5 and 1 cm from theureterovesical junction a maximum detrusor pressure of 17, 10 and 1 cm H2O was documented,respectively. While there was almost no stimulation response of the bladder after the lastpreparation step at 1 cm on the left ureter, the initial bladder pressure of 18 cm H2O could bereproduced under stimulation on the right side.

Conclusions: Analogous to human cadaver studies, we were able to prove neurophysiologicallystrictly unilateral detrusor innervation, drawing from the pelvic plexus dorsomedial to theureterovesical junction into the bladder. Preparation of this ureterovesical junction duringantireflex surgery, coagulating measures in this area or the affixation of anchor sutures after aVest suture involves the risk of unilateral or bilateral detrusor decentralization.

KEY WORDS: ureter, swine, denervation, vesico-ureteral reflux

In the 1997 American Urological Association PediatricVesicoureteral Reflux Guidelines Panel summary reportElder et al stated that many research areas urgently deservefurther investigation.1 They demanded that particular prior-ity should be given to the role of complications in antirefluxsurgery. In this context they suggested the development ofminimally invasive techniques for antireflux surgery.

In the current study we attempted to make a significantcontribution toward a solution to problems that can resultafter bilateral extravesical antireflux surgery, from the well-known phenomenon of bladder voiding dysfunction to com-plete urinary retention, which is a severe if rare complica-tion.2, 3 The described complications are attributable topossible intraoperative impairment of bladder innervation.Indications for the correctness of this hypothesis have beenfound in the earlier anatomical investigations of Leissner etal, who were able to show a close connection between theureterovesical junction and nerve fibers emanating from theinferior pelvic plexus, which innervate the bladder.4 Theyreenacted operative techniques of different standard antire-

flux operations in human cadavers, which led them to pos-tulate that the avoidance of nerve lesions depends on theselected operative procedure.

To investigate this issue further we developed a valid an-imal model that would allow us to demonstrate the neuro-physiological and neuropathophysiological phenomena ofpartial or complete impairment of the vesical innervation,which can be caused by bladder denervation during antire-flux surgery in the small pelvis, especially in antireflux sur-gery. We chose the minipig because of its great affinity withhuman neurophysiology of the lower urinary tract to ensurea high degree of comparability of results.5

MATERIALS AND METHODS

We present the steps of the trial using images of and dataon 1 of 6 isoflurane intubation anesthetized, adult, 11 to13-month-old female Gottinger minipigs weighing 27.3 to32.4 kg. As step 1 a dorsomedian incision was made above thespinal segments of L4 to S4. After laminectomy of thesesegments the dura mater beneath was exposed and opened inthe longitudinal direction until finally the cauda equina wasexposed (fig. 1).

After identification of the spinal nerves of S3, which pro-vided the optimal detrusor response in test stimulation, the

Accepted for publication March 14, 2003.Presented at annual meeting of American Urological Association,

Orlando, Florida, May 25–30, 2002.* Corresponding author: Department of Urology, University of

Kiel, Arnold-Heller Strasse 7, 24105 Kiel, Germany.

0022-5347/03/1702-0570/0 Vol. 170, 570–574, August 2003THE JOURNAL OF UROLOGY® Printed in U.S.A.Copyright © 2003 by AMERICAN UROLOGICAL ASSOCIATION DOI: 10.1097/01.ju.0000077446.49441.a9

570

nerve fibers were inserted into separate compartments of asize modified Brindley electrode (fig. 1). Recording detrusorpressure (pdet) responses to test stimulations (bilateral, uni-lateral left and unilateral right) was performed by a trans-vesically inserted catheter connected to a Medtronic Multi P(Dantec Medical, Santa Clara, California) urodynamics unit.As the next step we performed watertight closure of the duramater above the Brindley electrode using a muscle fascialpatch, which had been previously extracted (fig. 1). Aftersupine repositioning of the animal and median laparotomyonly the left kidney, left renal pelvis and proximal ureterwere prepared retroperineally. For better visual evaluationthe bladder was filled with 150 ml NaCl and surrounded withdark cloth without any manipulation of the lower urinarysystem (fig. 2).

To control correct electrode placement bilateral test stim-ulation with 2.0 V (impulse 200 �seconds at 20 Hz) was done.The bladder midline was marked with a tissue marker for abetter evaluation of bladder tilting. Bilateral stimulation ofthe bladder led to concentric contraction. Unilateral stimu-lation on the left and then the right side caused the bladderto tilt toward the ipsilateral side.

The left ureter was then prepared by freeing the ureter incircular fashion from its surrounding tissue. There was nofurther manipulation of the ureter or bladder. To simulateintraoperative dissection of the distal ureter and surround-ing tissue to assess the risk of potential injury to the pelvicplexus we prepared the ureter in 10, 5 and 1 cm steps,respectively, from the ureterovesical junction, starting fromthe renal pelvis (fig. 3). Subsequently 1 bilateral and 2 uni-lateral (left and right) stimulations were performed, as de-scribed. All results were recorded urodynamically (fig. 4). Onthe right ureter no preparation was done, so that this sidecould serve as a reference.

RESULTS

Prior to manipulation of the left ureter urodynamic record-ings showed a detrusor contraction after bilateral stimula-

tion with a mean maximum pressure of 51 cm H2O. Duringsubsequent unilateral stimulation of S3 (left side) significantbladder tilting to the ipsilateral side was observed (fig. 5).

FIG. 5. Results after bladder stimulation. A, ipsilateral tilting af-ter left unilateral bladder stimulation. B, representative urodynam-ics curve from 1 trial shows pdet during bladder stimulation. 1,bilateral. 2, unilateral left side. 3, unilateral right side.

FIG. 1. Opening of dura mater, cauda equina exposure, intraduralBrindley electrode insertion and subsequent watertight closure withfascial patch.

FIG. 2. Bladder surrounded by cloth with midline marked withtissue marker.

FIG. 3. Retroperitoneal preparation of left kidney, renal pelvisand proximal ureter in steps 10, 5 and 1 cm, respectively, fromureterovesical junction.

FIG. 4. Original urodynamics curve shows detrusor stimulation(stim.) at beginning of trial and after ureteral preparation 1 cmbefore ureterovesical junction. Unilat., unilateral.

DETRUSOR DECENTRALIZATION AFTER ANTIREFLUX SURGERY 571

Simultaneously recorded pdet showed an average maximumof 19 cm H2O (figs. 5 and 6). Analogous to these results thebladder tilted to the right after stimulation on the right side.Mean pdet was 18 cm H2O (fig. 6).

After documentation of these reference values circularpreparation of the left ureter 10, 5 and 1 cm, respectively,proximal to the ureterovesical junction was performed. Eachpreparation step was followed by 1 bilateral and 2 unilateral(left and right) stimulations. After preparation 10, 5 and 1 cmproximal to the ureterovesical junction stimulation of the leftspinal nerves led to a mean pdet of 17, 9.5 and 1 cm H2O,respectively (see table and fig. 6).

To verify correct electrode placement and exclude the pos-sibility that the acquired results were caused by detrusormuscle fatigue each preparation step was accompanied bycontralateral control stimulation on the right side. Thesecontrol stimulations resulted in bladder contractions with aconstant and reproducible pdet of almost 18 cm H2O at allpreparation steps (figs. 6 and 7). Nerve fiber bundles thathad been dissected at the ureterovesical junction (with theresult of ipsilateral detrusor acontractility) were clearly lo-calized dorsomedial of the ureterovesical junction by histo-logical analysis (fig. 7).

DISCUSSION

No relevant difference occurred between pdet documentedafter circular preparation 10 cm from the ureterovesical junc-tion of the left ureter with unilateral stimulation of the leftsacral nerves and pdet recorded after unilateral stimulationon the right side. After preparation of the left ureter directlybefore the intersection with the hypogastric artery at about 5cm proximal of the ureterovesical junction and unilateral leftS3 stimulation a mean 4.7% decrease in bladder pressureresulted in relation to pressure values acquired from stimu-lation on the unprepared right side resulted (10 vs 17 cmH2O, fig. 6). Preparation of the left ureter 1 cm before theureterovesical junction decreased this value further, so thatafter left stimulation there was almost no detrusor response(only 5.3% of the initial pressure value) (fig. 6). It is evenpossible to consider unilateral detrusor decentralization atthis site.

Transferability of these experimental data to the clinicalsituation becomes clear when considering the complications

after antireflux surgery described by Schaefer2 or Lipski etal,3 who reported serious neurogenic bladder voiding dys-function after 1-stage bilateral operations with extensivebladder neck mobilization. Unilateral and bilateral impair-ment of the pelvic plexus are known in the field of urogyne-cologic or colorectal malignancy surgery. Ralph et al reportedthat after radical abdominal hysterectomy, in which the re-section area extends up to the pelvis wall, bladder dysfunc-tion accompanied by urinary tract infection, decreased blad-der compliance and increased post-void residual urineoccurred postoperatively.6 Urodynamic investigations afterproctological operations, such as abdominal rectal resection,have also shown that up to 54% of the patients examined haddetrusor acontractility and 52% showed increased post-voidresidual urine, which did not exist before the operation.7

In contrast, after unilateral antireflux surgery no seriousbladder voiding dysfunctions have been reported regardlesswhether the chosen procedure was extramural or intramural.However, in 1-stage bilateral antireflux surgery the choicebetween an extravesical and an intravesical proceduremakes a difference since extravesical bilateral antireflux sur-gery has led to post-void residual urine problems in signifi-cantly more patients, of whom 78% required suprapubic uri-nary drainage, than intravesical procedures.8 Zaontz et alreported that after detrusorraphy using anchor sutures afterthe Vest method 3.8% of operated patients had a neuropathicbladder.9 Barrieras et al noted that after bilateral antireflux-plasty bladder voiding dysfunction occurred in 10% of cas-es.10 Houle et al even reported inefficient bladder voiding in26% of patients.11 In contrast to bilateral antireflux surgeryaccording to the Cohen or Politano-Leadbetter technique, nopostoperative bladder voiding dysfunction has been reportedafter unilateral antireflux surgery, such as the unilateralpsoas hitch technique or the Lich-Gregoir operation in 900patients when myotomy ran to the bladder roof coming fromthe ventral side.9, 12–16

These data indicate that operative methods involving po-tential nerve damage in the dorsomedial area of the uret-erovesical junction carry an increased risk of partial dener-

FIG. 6. Unilateral stimulation before and after ureteral prepara-tion (prep) 10, 5 and 1 cm from ureterovesical junction, respectively.A, pdet. B, bladder position after stimulation and preparation 1 cmproximal to ureterovesical junction on left ureter.

Unilateral stimulation values in 6 minipigs according to left pdet

Animal No.pdet (cm H2O)

Preop 10 Cm 5 Cm 1 Cm

1 14 11 7 02 31 29 15 13 18 16 10 24 12 12 5 05 21 18 9 16 20 18 11 1

Mean 19.3 17.3 9.5 0.8

FIG. 7. Exposure of nerve bundle at ureterovesical junction andhistological analysis.

DETRUSOR DECENTRALIZATION AFTER ANTIREFLUX SURGERY572

vation with consecutive bladder voiding dysfunction.Whether complete axotomy or impairment caused by pres-sure or traction on the nerve is responsible for these postop-erative symptoms depends on the chosen operative method,possible complications such as bleeding and the extent ofintervention as well as on surgeon expertise. To our knowl-edge our study represents the first animal model providing aneurophysiological explanation of the phenomenon of postop-erative bladder voiding dysfunction, which has often beendescribed after antireflux surgery or other operation tech-niques affecting the ureterovesical junction. It has been pos-sible to show that in the dorsomedial area directly before theureterovesical junction ipsilateral innervation of the detrusoris highly sensitive to impairment.

CONCLUSIONS

This neurophysiological model leads us to the hypothesis ofunilateral innervation of the detrusor vesicae and provides apossible explanation of why a hypocontractile or acontractiledetrusor can result from 1-stage bilateral antireflux surgery.High comparability of the results with the clinical situationcan be assumed since the urinary tract of minipigs bearsgreat neurophysiological resemblance to that of humans.Analogous to the investigations of Leissner et al,4 the currentdata could provide valid neurophysiological evidence of thedetrusor innervation running dorsomedial of the ureteroves-ical junction to the bladder (fig. 7). Exact knowledge of theseneuroanatomical positions is of essential clinical significancebecause preparation of the ureter in the area distal to theintersection of the hypogastric artery and about 1 cm abovethe ureterovesical junction carries the risk of unilateral orbilateral detrusor decentralization and consecutive detrusorhypocontractility or acontractility, whether during antirefluxsurgery, coagulatory measures in this area, second look op-erations or trigonal anchor suture placement after the Vesttechnique.

REFERENCES

1. Elder, J. S., Peters, C. A., Arant, B. S., Jr., Ewalt, D. H.,Hawtrey, C. E., Hurwitz, R. S. et al: Pediatric VesicoureteralReflux Guidelines Panel summary report on the managementof primary vesicoureteral reflux in children. J Urol, 157: 1846,1997

2. Schaefer, M., Bruhl, P. and Jankowski, A.: Complications andpseudocomplications after Lich-Gregoir antireflux plasty. IntUrol Nephrol, 22: 537, 1990

3. Lipski, B. A., Mitchell, M. E. and Burns, M. W.: Voiding dysfunc-tion after bilateral extravesical ureteral reimplantation.J Urol, 159: 1019, 1998

4. Leissner, J., Allhoff, E. P., Wolff, W., Feja, C., Hockel, M., Black,P. et al: The pelvic plexus and antireflux surgery: topograph-ical findings and clinical consequences. J Urol, 165: 1652, 2001

5. Scheepe, J. R., Braun, P. M., Bross, S., Weiss, J., Zendler, S.,Gramatte, T. et al: Ein standardisiertes In-vivo-Modell zurEvaluierung anticholinerger Effekte auf die Blasenkontrak-tion, den Speichelfluss und auf das Elektromyogramm derHarnblase. Akt Urol, 31: 311, 2000

6. Ralph, G., Winter, R. and Michelitsch, L.: Effect of radical ap-proach in surgical therapy of cervix cancer on the dynamics ofthe lower urinary tract. Gynakol Rundsch, suppl., 30: 99, 1990

7. Zanolla, R., Campo, B., Ordesi, G. and Martino, G.: Bladderurethral dysfunction after abdominoperineal resection of therectum for ano-rectal cancer. Tumori, 70: 555, 1984

8. Fung, L. C. T., McLorie, G. A., Jain, U., Khoury, A. E. andChurchill, B. M.: Voiding efficiency after ureteral reimplanta-tion: a comparison of extravesical and intravesical techniques.J Urol, 153: 1972, 1995

9. Zaontz, M. R., Maizels, M., Sugar, E. C. and Firlit, C. F.: Detru-sorrhaphy: extravesical ureteral advancement to correct vesi-coureteral reflux in children. J Urol, 138: 947, 1987

10. Barrieras, D., Lapointe, S., Reddy, P. P., Williot, P., McLorie,G. A., Bagli, D. et al: Urinary retention after bilateral ex-travesical ureteral reimplantation: does dissection distal to

the ureteral orifice have a role? J Urol, 162: 1197, 199911. Houle, A. M., Vernet, O., Jednak, R., Pippi Salle, J. L. and

Farmer, J. P.: Bladder function before and after selective dor-sal rhizotomy in children with cerebral palsy. J Urol, 160:1088, 1998

12. Cohen, S. J.: Ureterozystoneostemie: eine neue Antirefluxtech-nik. Akt Urol, 6: 1, 1975

13. Burbige, K. A.: Ureteral reimplantation: a comparison of resultswith the cross-trigonal and Politano-Leadbetter techniques in120 patients. J Urol, 146: 1352, 1991

14. Arap, S., Abrao, E. G. and Menezes de Goes, G.: Treatment andprevention of complications after extravesical antireflux tech-nique. Eur Urol, 7: 263, 1981

15. Gregoir, W.: Le reflux vesico-ureteral congenital. Acta Urol Belg,30: 286, 1962

16. Politano, V. A. and Leadbetter, W. F.: An operative technique forthe correction of vesicoureteral reflux. J Urol, 79: 932, 1958

EDITORIAL COMMENT

The authors describe a novel minipig model to determine theneurophysiological effect on the bladder by periureteral dissection invarious locations away from the bladder without actually dissectingthe bladder itself. Their findings indicate that there are nerve fibersin the dorsomedial region just outside of the ureteral entrance to thebladder. If this neural network were injured during dissection, itwould lead to decreased contractility on the affected detrusor side.The authors observed in the 10 minipigs studied that unilateraldissection periureterally 5 cm from the bladder caused an almost50% decrease in bladder pressure during detrusor stimulation. Whenthe same dissection was performed within 1 cm of the bladder, therewas little if any contraction of the affected detrusor side. While it istruly impressive and the photos graphically attest to the authorfindings, I am not convinced that clinically it matters. Most intra-vesical techniques invariably dissect in the vicinity of the nerves inquestion and rarely is there an inability to void. It has also beenshown in the study of Minevich et al (as corroborated by my experi-ence) that after performing bilateral detrusorrhaphies voiding dys-function occurred in a small percent of these children.1 In fact, I havepersonally found that if a child is not an infrequent voider and has anormal voiding pattern, they are unlikely to have voiding dysfunc-tion, such as urinary retention postoperatively, after bilateral detru-sorrhaphy. Hence, patient selection would obviously be important inthese children. Interestingly figure 7 demonstrates that the nervefibers in question are not directly attached to the ureter and accord-ing to the authors they are generally not cut even during a detruso-rrhaphy procedure. However, no 2 surgeons operate the same exactway, and the degree of lateral and medial dissection during reim-plantation surgery varies in this regard.

The take-home message of this article is that there is clearly aneurophysiological explanation for why some of these children havepostoperative voiding dysfunction. However, I would recommend tothe authors that their next experiment should include detrusor dis-section alone and with nerve denervation to paint a more accuratepicture of what actually happens during ureteral reimplantationsurgery.

Mark R. ZaontzUrology for Children, L.L.C.Voorhees, New Jersey

1. Minevich, E., Wacksman, J., Lewis, A. G. and Sheldon, C. A.:Incidence of contralateral vesicoureteral reflux following uni-lateral extravesical destrusorrhaphy (ureteroneocystostomy).J Urol, suppl., 157: 36A, abstract 141, 1997

REPLY BY AUTHORS

We agree with the comments and are conscious of the fact thatbilateral detrusorrhaphy by an experienced surgeon can produceexcellent results, especially if the dissection for laying open theureter extravesically is performed in line with the natural course ofthe ureter. Moreover, it would be possible to avoid the main portionof the pelvic plexus by staying in this line, as described by Leissneret al.1 On the other hand, there are various studies reporting tran-sient urinary retention after bilateral detrusorrhaphy as describedin our article. As demonstrated in our neurophysiological animalmodel, the circumferential myotomy near the dorsomedial region ofthe vesicoureteral junction caused postoperative bladder voiding

DETRUSOR DECENTRALIZATION AFTER ANTIREFLUX SURGERY 573

dysfunction. It remains unclear as to why urinary retention developsafter bilateral detrusorrhaphy in some but not all patients. Webelieve that the nerve fibers may be destroyed only in cases withintraoperative complications such as bleeding, forcing the surgeon toperform coagulation or ligation in the dorsomedial region of thevesicoureteral junction. Another possible explanation is that theaffected nerve fibers at the vesicoureteral junction in humans are notsolely responsible for the voiding function. As described by Zaontz,probably only children with preoperative abnormal voiding patternsdepend on innervation via the pelvic plexus while the others are able

compensate for the impairment of the nerve fibers of the pelvicplexus. However, there is no doubt about the necessity to includedetrusor dissection with and without nerve denervation in the sameanimal model if we want to clarify what really happens duringdetrusorrhaphy.

1. Leissner, J., Allhoff, E. P., Wolff, W., Fseja, C., Hockel, M. andBlack, P.: The pelvic plexus and antireflux surgery: topo-graphical findings and clinical consequences. J Urol, 165:1652, 2001

DETRUSOR DECENTRALIZATION AFTER ANTIREFLUX SURGERY574