- Email: [email protected]
BLADDER NECK FISTULA AFTER THE COMPLETE PRIMARY REPAIR OF EXSTROPHY: A MULTI-INSTITUTIONAL EXPERIENCE
0022-5347/05/1744-1687/0 THE JOURNAL OF UROLOGY® Copyright © 2005 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 174, 1687–1690, October 2005 Printed in U.S.A.
DOI: 10.1097/01.ju.0000176621.99922.35
BLADDER NECK FISTULA AFTER THE COMPLETE PRIMARY REPAIR OF EXSTROPHY: A MULTI-INSTITUTIONAL EXPERIENCE SETH A. ALPERT,* EARL Y. CHENG, WILLIAM E. KAPLAN, WARREN T. SNODGRASS, DUNCAN T. WILCOX AND BRADLEY P. KROPP From the Division of Pediatric Urology, Children’s Memorial Hospital, Chicago, Illinois (SAA, EYC, WEK), Division of Pediatric Urology, Children’s Medical Center and UT Southwestern Medical Center, Dallas, Texas (WTS), Paediatric Urology, Great Ormond Street Hospital for Children, London, United Kingdom (DTW), and Section of Pediatric Urology, Children’s Hospital of Oklahoma, Oklahoma City, Oklahoma (BPK)
ABSTRACT
Purpose: The major goals of complete primary repair of exstrophy (CPRE) are the re-creation of normal anatomy which allows bladder cycling and to reduce the number of future procedures necessary to achieve continence. It is unclear whether CPRE is associated with a higher bladder neck fistula rate than the traditional staged repair. We review a multi-institutional experience with the CPRE technique to evaluate the rate of bladder neck fistula. Materials and Methods: A retrospective review of 18 boys and 4 girls who underwent CPRE at 4 tertiary pediatric urology referral centers during the last 6 years was performed. All pertinent technical aspects were reviewed, including timing of procedure, whether osteotomies were performed, and number of layers used to reconstruct the bladder neck and urethra. Complications were noted, especially that of bladder neck fistula. Results: Mean followup was 22.6 months. Of the patients 14 (64%) underwent primary closure within the first 48 hours of life and only 1 required osteotomies. The remaining 8 patients underwent closure between 5 days and 3 months of age (mean 24.6 days) and all required osteotomies. Bladder neck fistula occurred postoperatively at the pubic junction in 9 males (41%). Four cases had a 2 layer closure that was covered with a single layer small intestinal submucosa onlay (Surgisis®) and no patient had a fistula. Fistulas developed in 62.5% of patients with delayed closure vs 29% of those with immediate closure (p ⫽ 0.135). Two fistulas closed spontaneously and 7 required surgical closure at a mean of 7.5 months after the fistula occurred. Conclusions: This multi-institutional study demonstrates that bladder neck fistulas occur in almost half of patients following CPRE by experienced pediatric urologists. While spontaneous closure is possible, most will eventually require repair. The long-term implications of this finding with regard to continence and the need for additional bladder neck procedures remain to be seen. We are encouraged by the preliminary results of small intestinal submucosa coverage and will continue to evaluate its use at the time of primary exstrophy closure. KEY WORDS: bladder exstrophy, fistula
The complete penile disassembly technique for epispadias repair was pioneered by Mitchell and Bagli in 1996.1 Based on the same surgical principles, the complete primary repair of exstrophy (CPRE) technique was introduced in 1999 as an anatomically based approach to closure of bladder exstrophy.2 The potential advantages of this technique include allowing the bladder to cycle during the first year of life, minimizing the number of surgical procedures required to achieve urinary continence and placing the closed bladder in a more anatomic location deep in the pelvis. One of the most common potential postoperative complications is bladder neck fistula at the pubic junction.2, 3 The published rate of immediate bladder neck fistula at 1 institution is 14%. Initial experience with CPRE at other institutions has suggested a higher rate of bladder neck fistula. We sought to determine the bladder neck fistula rate after CPRE by performing a review at 4 institutions. MATERIALS AND METHODS
The records of all patients who underwent CPRE at 4 tertiary pediatric urology referral centers between 1998 and * Correspondence: Division of Pediatric Urology, Children’s Memorial Hospital, 2300 Children’s Plaza, Box 24, Chicago, Illinois 60614 (telephone: 773– 880-4428; FAX: 773-880-3339; e-mail: [email protected]).
2004 were reviewed. CPRE was performed as previously described by Grady and Mitchell.2 All patients had classic bladder exstrophy. Gender, age at repair and length of followup were noted. All technical aspects of the repair were recorded, including performance of osteotomies and the number of layers used to reconstruct the urethra (see table). All postoperative complications were noted, especially that of bladder neck fistula. We performed statistical evaluation using GB-STAT software, version 8.0 (Dynamic Microsystems, Silver Spring, Maryland). Data are reported as mean ⫾ standard deviation. The chi-square test or Fisher’s exact test (1-tail) was used for statistical comparison with p ⱕ0.05 considered significant. RESULTS
Between 1998 and 2004, 18 males and 4 females underwent CPRE for primary reconstruction of classic bladder exstrophy. The procedure was done within 2 days of life in 14 cases (64%) and only 1 (7.1%) required innominate osteotomies. All of the remaining 8 cases were closed after 2 days of life (mean 24.6 ⫾ 29.2 days, range 5 to 90) and all required innominate osteotomies during CPRE. All cases underwent closure of the pubis at the time of surgery. Mean length of
1687
1688
FISTULA AFTER PRIMARY REPAIR OF EXSTROPHY Technical Aspects of Closure by CPRE
Pt No. — Sex 1—M 2—M 3—M 4—M 5—M 6—M 7—M 8—M 9 — M* 10 — M 11 — M* 12 — F 13 — F 14 — M 15 — M 16 — M 17 — M 18 — M 19 — F 20 — F 21 — M* 22 — M* * SIS onlay used.
Age at CPRE (days)
Osteotomies Done
1 or 2-Layer Urethral Closure
Penopubic Fistula
Spontaneous Closure
0 45 1 2 1 2 12 2 8 1 2 2 2 5 9 14 14 90 1 2 2 2
No Yes No No No No Yes No Yes No No No No Yes Yes Yes Yes Yes No Yes No No
2 2 1 1 1 1 2 2 2 2 2 2 2 1 1 2 2 2 2 2 2 2
Yes No No Yes No Yes Yes Yes No No No No No Yes Yes Yes Yes No No No No No
No
followup for all cases was 22.6 ⫾ 17.9 months (range 2 to 73). Urethral tubularization was done in 2 layers in 16 cases (73%) and 1 layer in 6 cases. The urethral meatus was hypospadiac in 13 cases (72% of males) after the primary repair. Bladder neck fistula occurred postoperatively at the pubic junction in 9 male patients (41%). No female patients had a postoperative fistula. Fistula occurred at a mean of 9.71 ⫾ 12 weeks after surgery (range 1 to 32). Fistulas developed in 62.5% of patients who underwent delayed closure (after 48 hours of life) compared to 29% of those treated with immediate closure (p⫽0.135). Of the 16 patients whose urethra was closed in 2 layers 5 (31%) had fistulas compared to 4 fistulas (67%) in the 6 patients whose urethra was closed in 1 layer (p ⫽ 0.155). Two fistulas (both in cases with delayed closure and 1 layer urethral closure) closed spontaneously. The remaining 7 fistulas required surgical closure at a mean of 7.53 ⫾ 2.9 months after fistula occurrence (1 patient is presently awaiting surgical closure of a persistent fistula). Four patients (immediate closure in 3, delayed closure in 1) underwent a 2-layer urethral closure with an anterior urethral onlay of single-layer small intestinal submucosa (SIS, Surgisis®) and no patient had a fistula. In comparison, 9 of 18 patients (50%) had a bladder neck fistula when SIS was not used to supplement the urethral closure (p ⫽ 0.098). Other postoperative complications were seen in addition to bladder neck fistula. Two patients had systemic fungal infection, which required antifungal treatment, and 1 patient each had bilateral renal calculi, urethral obstruction associated with retained polydioxanone monofilament suture, which necessitated cystoscopy for removal, unsatisfactory penile skin quality, and distal urethral stenosis requiring dilation and temporary catheterization. DISCUSSION
Bladder exstrophy is one of the most challenging conditions pediatric urologists manage. Presently, controversy exists regarding the optimal method of bladder reconstruction, that is traditional staged repair vs complete primary repair. Advocates of CPRE maintain that this procedure provides a complete anatomical reconstruction of the bladder and, therefore, optimizes early bladder cycling and assists bladder development.2 Also, there is the hope that by performing bladder closure, bladder neck repair and urethral closure in 1 procedure at birth, the patient will require fewer surgeries overall to achieve urinary continence.4 The rate of urinary
No No No No
Yes Yes No No
continence has been determined in those patients old enough to be evaluated. In one of the largest series of patients treated with CPRE daytime urinary continence was defined as 2-hour dry intervals with volitional voiding.3 Although this definition is not universally accepted, most recent studies define daytime continence in exstrophy as 2 to 3 hours of dryness. Using this definition, Grady et al found 76% of boys and girls with exstrophy were continent after CPRE, although only 29% were able to achieve continence without an additional bladder neck reconstruction procedure.3 However, 1 child in their series required bladder augmentation for inadequate compliance and capacity to attain urinary continence. The most frequent complication of CPRE seen previously has been bladder neck fistula at the pubic junction, with a rate of 14% of 35 patients studied in 1 large single institutional series.1, 2 Approximately half of these fistulas closed spontaneously and the remainder required surgical repair. One of the fistulas that required surgical repair occurred in a female. In another recent large series 33 patients with classic bladder exstrophy were treated with CPRE (initial closure in 29 and after previous failed initial closure in 4).5 Bladder neck fistula was reported in 2 children (6%), including 1 female, while distal urethral fistula was recorded in 1 boy. All 3 cases required surgery for fistula closure. Caoine and Capozza reported a high fistula rate (36%) in 41 exstrophy or epispadias repairs done in male patients using a variety of techniques that did not include complete penile disassembly.6 A separate group of 17 patients (exstrophy in 10, epispadias in 6 and previously failed repair in 1) underwent complete penile disassembly. Complications included a urethral fistula at the glans in 1 case and distal meatal stenosis in another. The overall low complication rate was attributed to the addition of perineal muscular complex reassembly to the complete penile disassembly technique. Of our 22 patients a bladder neck fistula developed at the pubic junction postoperatively in 41%, of whom 78% required an additional procedure to close the fistula. All fistulas were seen in males, and were located at the region where the corporal bodies diverge and the urethra passes under the symphysis pubis. When the timing of closure and the number of layers of urethral closure were examined there was no statistically significant factor that predicted fistula formation. Although there was no statistical correlation between the number of layers used for urethral closure and fistula
1689
FISTULA AFTER PRIMARY REPAIR OF EXSTROPHY
formation, there was an apparent increased rate of fistulas in the group treated with a 1-layer closure compared to those treated with 2 layers (67% vs 31%). One patient in our series who had a fistula underwent cystoscopy shortly after the fistula became apparent. A polydioxanone monofilament suture from the pubic closure was seen protruding into the lumen, which may explain why the fistula formed. Another explanation could be the relative lack of barrier layers between the skin and urethral closure, in contrast to the multiple layers of coverage usually used in hypospadias repairs. The use of SIS as an anterior urethral onlay may help provide an additional layer during closure, although there were not enough cases in our series to achieve statistical significance. In another series of 19 cases which included original closures and failed initial closures 10 (52%) had temporary suprapubic urinary drainage after removal of the cystostomy tube but all fistulas closed within a mean of 5 days and none had a persistent fistula.7 Only 4 males in this series underwent closure within the first 72 hours of life, whereas the other patients underwent closure later in life. All patients in this subset have achieved 60 to 90-minute intervals of urinary continence. In a large study of patients who underwent a modified Cantwell-Ransley epispadias repair Surer et al reported an overall postoperative fistula rate of 24% in 93 patients.8 Of these fistulas 82% were at the base of the penis (19% fistula rate total). In addition, 10% of patients had a urethral stricture at the proximal anastomosis, which has not been reported with the CPRE. Whereas the previously reported bladder neck fistula rate of 14% for CPRE is not statistically different from the 19% rate with the staged repair (chisquare p⫽0.505), the 41% rate of bladder neck fistula after CPRE in this study did show a statistically significant difference compared to the staged repair (chi-square p ⫽ 0.032). However, it may not be fair to compare the fistula rates between these 2 procedures since patients in the staged group were older (mean age 3.4 years)8 and, by definition, the bladder repair is not done concurrently with urethral closure. Gearhart has reported additional complications after complete repair done without osteotomies at other institutions, including wound dehiscence, loss of penile skin, bladder prolapse and partial loss of a hemiglans.9 The penile disassembly technique used for penile repair during CPRE has been associated with complete or partial loss of bilateral or unilateral hemiglans, loss of a corporal body, penile urethra or penile shaft skin.10 The purpose of our study was to investigate the bladder neck fistula rate after CPRE at 4 tertiary pediatric urology centers. We found a 41% bladder neck fistula rate after CPRE performed by pediatric urologists experienced with exstrophy surgery. It is too early to determine the effect this may have on long-term continence rates. The overall number of patients available for analysis for this study is low, in part due to the decreased incidence of bladder exstrophy in the United States during the last decade, as well as the relative newness of the CPRE technique in the armamentarium of exstrophy surgery. Four patients in our series had the original urethral closure reinforced with a single layer SIS onlay and none of them had fistulas. Although this technique has promise to possibly prevent fistula formation, it was used in only a small number of patients and cannot be generalized to larger numbers without further study. CONCLUSIONS
This multi-institutional study demonstrates that bladder neck fistulas occur in almost half of patients after CPRE performed by experienced pediatric urologists. While spontaneous closure is possible, most cases will eventually require repair. The long-term implications of this finding with regard
to continence and the need for additional bladder neck procedures remain to be seen. We are encouraged by the preliminary results of SIS coverage and will continue to evaluate its use at the time of primary exstrophy closure.
REFERENCES
1. Mitchell, M. E. and Bagli, D. J.: Complete penile disassembly for epispadias repair: the Mitchell technique. J Urol, 155: 300, 1996 2. Grady, R. W. and Mitchell, M. E.: Complete primary repair of exstrophy. J Urol, 162: 1415, 1999 3. Grady, R. W., Joyner, B. D., Roedel, M., Feng, W., Casale, P. and Mitchell, M. E.: The complete primary repair of exstrophy (CPRE) technique for repair of bladder exstrophy and epispadias: long-term follow-up. Presented at the Section on Urology, American Academy of Pediatrics, New Orleans, Louisiana, November 2, 2003 4. Grady, R. W. and Mitchell, M. E.: Complete primary repair of exstrophy. Surgical technique. Urol Clin North Am, 27: 569, 2000 5. Hammouda, H. M. and Kotb, H.: Complete primary repair of bladder exstrophy: initial experience with 33 cases. J Urol, 172: 1441, 2004 6. Caione, P. and Capozza, N.: Evolution of male epispadias repair: 16-year experience. J Urol, 165: 2410, 2001 7. El-Sherbiny, M. T., Hafez, A. T. and Ghoneim, M. A.: Complete repair of exstrophy: further experience with neonates and children after failed initial closure. J Urol, 168: 1692, 2002 8. Surer, I., Baker, L. A., Jeffs, R. D. and Gearhart, J. P.: The modified Cantwell-Ransley repair for exstrophy and epispadias: 10-year experience. J Urol, 164: 1040, 2000 9. Gearhart, J. P.: Complete repair of bladder exstrophy in the newborn: complications and management. J Urol, 165: 2431, 2001 10. Husmann, D. A. and Gearhart, J. P.: Loss of the penile glans and/or corpora following primary repair of the bladder exstrophy using the complete penile disassembly technique. Presented at Section on Urology, American Academy of Pediatrics, New Orleans, Louisiana, November 1, 2003 EDITORIAL COMMENT The “complete” repair using the Mitchell technique has been increasingly performed for reconstruction for bladder exstrophy. As more centers perform this procedure in infants, complications are increasingly being identified. This report from 4 respected institutions in the management of complex pediatric urological cases indicates the difficulties associated with reconstruction of this most major of urological abnormalities. The authors indicate a 41% incidence of fistulas, all of which occurred in males and the majority required later operative management. The authors also noted a higher incidence in those cases treated with later closure, and suggest that the use of SIS may prevent development of fistulas. However, SIS to prevent fistula at the penopubic junction was not statistically significant in their series. Since many of these patients will need later bladder neck reconstruction, it will be interesting to see if the location of the SIS at the bladder neck will present an impediment to dissection, bladder neck mobilization and successful reconstruction. As the number of children treated with CPRE increases, the results must be compared to those of the current standard management, ie modern staged reconstruction. Furthermore, when complete reconstruction is explained to parents, we should indicate that many children will require later surgical procedures, such as ureteral reimplantation, hypospadias repair and bladder neck reconstruction. Additionally, since bladder exstrophy is a relatively rare anomaly, most pediatric urologists will perform exstrophy closure every 1 to 2 years. Given this infrequent surgical exposure, being conservative is perhaps in the best interest of our patients, and pediatric urologists lacking expertise in CPRE should be forewarned. Ranjiv Mathews Division of Pediatric Urology Johns Hopkins Hospital Baltimore, Maryland
1690
FISTULA AFTER PRIMARY REPAIR OF EXSTROPHY
DISCUSSION Dr. Joao L. Pippi-Salle. Why do you think you had such a high incidence of fistula that did not close spontaneously? In my opinion you have to really dissect proximally the bladder neck and the proximal urethra, and place down and cover the proximal urethra with corpora cavernosa rotated on top, and then you usually do not have a fistula. Do you think that could be a factor? Dr. Seth A. Alpert. That is a possibility. In our study we did not find any 1 factor that was associated with this increased fistula rate. The fact that this area is thinner than the other areas of the urethra and it is more difficult to get other multiple layers of coverage could be possibilities. A couple of our patients were wrapped with SIS and they did not have a fistula, and so we thought that deserves future study. The patients who underwent earlier repairs and did not have osteotomies also had fewer fistulas.
Vol. 174, 1687–1690, October 2005 Printed in U.S.A.
DOI: 10.1097/01.ju.0000176621.99922.35
BLADDER NECK FISTULA AFTER THE COMPLETE PRIMARY REPAIR OF EXSTROPHY: A MULTI-INSTITUTIONAL EXPERIENCE SETH A. ALPERT,* EARL Y. CHENG, WILLIAM E. KAPLAN, WARREN T. SNODGRASS, DUNCAN T. WILCOX AND BRADLEY P. KROPP From the Division of Pediatric Urology, Children’s Memorial Hospital, Chicago, Illinois (SAA, EYC, WEK), Division of Pediatric Urology, Children’s Medical Center and UT Southwestern Medical Center, Dallas, Texas (WTS), Paediatric Urology, Great Ormond Street Hospital for Children, London, United Kingdom (DTW), and Section of Pediatric Urology, Children’s Hospital of Oklahoma, Oklahoma City, Oklahoma (BPK)
ABSTRACT
Purpose: The major goals of complete primary repair of exstrophy (CPRE) are the re-creation of normal anatomy which allows bladder cycling and to reduce the number of future procedures necessary to achieve continence. It is unclear whether CPRE is associated with a higher bladder neck fistula rate than the traditional staged repair. We review a multi-institutional experience with the CPRE technique to evaluate the rate of bladder neck fistula. Materials and Methods: A retrospective review of 18 boys and 4 girls who underwent CPRE at 4 tertiary pediatric urology referral centers during the last 6 years was performed. All pertinent technical aspects were reviewed, including timing of procedure, whether osteotomies were performed, and number of layers used to reconstruct the bladder neck and urethra. Complications were noted, especially that of bladder neck fistula. Results: Mean followup was 22.6 months. Of the patients 14 (64%) underwent primary closure within the first 48 hours of life and only 1 required osteotomies. The remaining 8 patients underwent closure between 5 days and 3 months of age (mean 24.6 days) and all required osteotomies. Bladder neck fistula occurred postoperatively at the pubic junction in 9 males (41%). Four cases had a 2 layer closure that was covered with a single layer small intestinal submucosa onlay (Surgisis®) and no patient had a fistula. Fistulas developed in 62.5% of patients with delayed closure vs 29% of those with immediate closure (p ⫽ 0.135). Two fistulas closed spontaneously and 7 required surgical closure at a mean of 7.5 months after the fistula occurred. Conclusions: This multi-institutional study demonstrates that bladder neck fistulas occur in almost half of patients following CPRE by experienced pediatric urologists. While spontaneous closure is possible, most will eventually require repair. The long-term implications of this finding with regard to continence and the need for additional bladder neck procedures remain to be seen. We are encouraged by the preliminary results of small intestinal submucosa coverage and will continue to evaluate its use at the time of primary exstrophy closure. KEY WORDS: bladder exstrophy, fistula
The complete penile disassembly technique for epispadias repair was pioneered by Mitchell and Bagli in 1996.1 Based on the same surgical principles, the complete primary repair of exstrophy (CPRE) technique was introduced in 1999 as an anatomically based approach to closure of bladder exstrophy.2 The potential advantages of this technique include allowing the bladder to cycle during the first year of life, minimizing the number of surgical procedures required to achieve urinary continence and placing the closed bladder in a more anatomic location deep in the pelvis. One of the most common potential postoperative complications is bladder neck fistula at the pubic junction.2, 3 The published rate of immediate bladder neck fistula at 1 institution is 14%. Initial experience with CPRE at other institutions has suggested a higher rate of bladder neck fistula. We sought to determine the bladder neck fistula rate after CPRE by performing a review at 4 institutions. MATERIALS AND METHODS
The records of all patients who underwent CPRE at 4 tertiary pediatric urology referral centers between 1998 and * Correspondence: Division of Pediatric Urology, Children’s Memorial Hospital, 2300 Children’s Plaza, Box 24, Chicago, Illinois 60614 (telephone: 773– 880-4428; FAX: 773-880-3339; e-mail: [email protected]).
2004 were reviewed. CPRE was performed as previously described by Grady and Mitchell.2 All patients had classic bladder exstrophy. Gender, age at repair and length of followup were noted. All technical aspects of the repair were recorded, including performance of osteotomies and the number of layers used to reconstruct the urethra (see table). All postoperative complications were noted, especially that of bladder neck fistula. We performed statistical evaluation using GB-STAT software, version 8.0 (Dynamic Microsystems, Silver Spring, Maryland). Data are reported as mean ⫾ standard deviation. The chi-square test or Fisher’s exact test (1-tail) was used for statistical comparison with p ⱕ0.05 considered significant. RESULTS
Between 1998 and 2004, 18 males and 4 females underwent CPRE for primary reconstruction of classic bladder exstrophy. The procedure was done within 2 days of life in 14 cases (64%) and only 1 (7.1%) required innominate osteotomies. All of the remaining 8 cases were closed after 2 days of life (mean 24.6 ⫾ 29.2 days, range 5 to 90) and all required innominate osteotomies during CPRE. All cases underwent closure of the pubis at the time of surgery. Mean length of
1687
1688
FISTULA AFTER PRIMARY REPAIR OF EXSTROPHY Technical Aspects of Closure by CPRE
Pt No. — Sex 1—M 2—M 3—M 4—M 5—M 6—M 7—M 8—M 9 — M* 10 — M 11 — M* 12 — F 13 — F 14 — M 15 — M 16 — M 17 — M 18 — M 19 — F 20 — F 21 — M* 22 — M* * SIS onlay used.
Age at CPRE (days)
Osteotomies Done
1 or 2-Layer Urethral Closure
Penopubic Fistula
Spontaneous Closure
0 45 1 2 1 2 12 2 8 1 2 2 2 5 9 14 14 90 1 2 2 2
No Yes No No No No Yes No Yes No No No No Yes Yes Yes Yes Yes No Yes No No
2 2 1 1 1 1 2 2 2 2 2 2 2 1 1 2 2 2 2 2 2 2
Yes No No Yes No Yes Yes Yes No No No No No Yes Yes Yes Yes No No No No No
No
followup for all cases was 22.6 ⫾ 17.9 months (range 2 to 73). Urethral tubularization was done in 2 layers in 16 cases (73%) and 1 layer in 6 cases. The urethral meatus was hypospadiac in 13 cases (72% of males) after the primary repair. Bladder neck fistula occurred postoperatively at the pubic junction in 9 male patients (41%). No female patients had a postoperative fistula. Fistula occurred at a mean of 9.71 ⫾ 12 weeks after surgery (range 1 to 32). Fistulas developed in 62.5% of patients who underwent delayed closure (after 48 hours of life) compared to 29% of those treated with immediate closure (p⫽0.135). Of the 16 patients whose urethra was closed in 2 layers 5 (31%) had fistulas compared to 4 fistulas (67%) in the 6 patients whose urethra was closed in 1 layer (p ⫽ 0.155). Two fistulas (both in cases with delayed closure and 1 layer urethral closure) closed spontaneously. The remaining 7 fistulas required surgical closure at a mean of 7.53 ⫾ 2.9 months after fistula occurrence (1 patient is presently awaiting surgical closure of a persistent fistula). Four patients (immediate closure in 3, delayed closure in 1) underwent a 2-layer urethral closure with an anterior urethral onlay of single-layer small intestinal submucosa (SIS, Surgisis®) and no patient had a fistula. In comparison, 9 of 18 patients (50%) had a bladder neck fistula when SIS was not used to supplement the urethral closure (p ⫽ 0.098). Other postoperative complications were seen in addition to bladder neck fistula. Two patients had systemic fungal infection, which required antifungal treatment, and 1 patient each had bilateral renal calculi, urethral obstruction associated with retained polydioxanone monofilament suture, which necessitated cystoscopy for removal, unsatisfactory penile skin quality, and distal urethral stenosis requiring dilation and temporary catheterization. DISCUSSION
Bladder exstrophy is one of the most challenging conditions pediatric urologists manage. Presently, controversy exists regarding the optimal method of bladder reconstruction, that is traditional staged repair vs complete primary repair. Advocates of CPRE maintain that this procedure provides a complete anatomical reconstruction of the bladder and, therefore, optimizes early bladder cycling and assists bladder development.2 Also, there is the hope that by performing bladder closure, bladder neck repair and urethral closure in 1 procedure at birth, the patient will require fewer surgeries overall to achieve urinary continence.4 The rate of urinary
No No No No
Yes Yes No No
continence has been determined in those patients old enough to be evaluated. In one of the largest series of patients treated with CPRE daytime urinary continence was defined as 2-hour dry intervals with volitional voiding.3 Although this definition is not universally accepted, most recent studies define daytime continence in exstrophy as 2 to 3 hours of dryness. Using this definition, Grady et al found 76% of boys and girls with exstrophy were continent after CPRE, although only 29% were able to achieve continence without an additional bladder neck reconstruction procedure.3 However, 1 child in their series required bladder augmentation for inadequate compliance and capacity to attain urinary continence. The most frequent complication of CPRE seen previously has been bladder neck fistula at the pubic junction, with a rate of 14% of 35 patients studied in 1 large single institutional series.1, 2 Approximately half of these fistulas closed spontaneously and the remainder required surgical repair. One of the fistulas that required surgical repair occurred in a female. In another recent large series 33 patients with classic bladder exstrophy were treated with CPRE (initial closure in 29 and after previous failed initial closure in 4).5 Bladder neck fistula was reported in 2 children (6%), including 1 female, while distal urethral fistula was recorded in 1 boy. All 3 cases required surgery for fistula closure. Caoine and Capozza reported a high fistula rate (36%) in 41 exstrophy or epispadias repairs done in male patients using a variety of techniques that did not include complete penile disassembly.6 A separate group of 17 patients (exstrophy in 10, epispadias in 6 and previously failed repair in 1) underwent complete penile disassembly. Complications included a urethral fistula at the glans in 1 case and distal meatal stenosis in another. The overall low complication rate was attributed to the addition of perineal muscular complex reassembly to the complete penile disassembly technique. Of our 22 patients a bladder neck fistula developed at the pubic junction postoperatively in 41%, of whom 78% required an additional procedure to close the fistula. All fistulas were seen in males, and were located at the region where the corporal bodies diverge and the urethra passes under the symphysis pubis. When the timing of closure and the number of layers of urethral closure were examined there was no statistically significant factor that predicted fistula formation. Although there was no statistical correlation between the number of layers used for urethral closure and fistula
1689
FISTULA AFTER PRIMARY REPAIR OF EXSTROPHY
formation, there was an apparent increased rate of fistulas in the group treated with a 1-layer closure compared to those treated with 2 layers (67% vs 31%). One patient in our series who had a fistula underwent cystoscopy shortly after the fistula became apparent. A polydioxanone monofilament suture from the pubic closure was seen protruding into the lumen, which may explain why the fistula formed. Another explanation could be the relative lack of barrier layers between the skin and urethral closure, in contrast to the multiple layers of coverage usually used in hypospadias repairs. The use of SIS as an anterior urethral onlay may help provide an additional layer during closure, although there were not enough cases in our series to achieve statistical significance. In another series of 19 cases which included original closures and failed initial closures 10 (52%) had temporary suprapubic urinary drainage after removal of the cystostomy tube but all fistulas closed within a mean of 5 days and none had a persistent fistula.7 Only 4 males in this series underwent closure within the first 72 hours of life, whereas the other patients underwent closure later in life. All patients in this subset have achieved 60 to 90-minute intervals of urinary continence. In a large study of patients who underwent a modified Cantwell-Ransley epispadias repair Surer et al reported an overall postoperative fistula rate of 24% in 93 patients.8 Of these fistulas 82% were at the base of the penis (19% fistula rate total). In addition, 10% of patients had a urethral stricture at the proximal anastomosis, which has not been reported with the CPRE. Whereas the previously reported bladder neck fistula rate of 14% for CPRE is not statistically different from the 19% rate with the staged repair (chisquare p⫽0.505), the 41% rate of bladder neck fistula after CPRE in this study did show a statistically significant difference compared to the staged repair (chi-square p ⫽ 0.032). However, it may not be fair to compare the fistula rates between these 2 procedures since patients in the staged group were older (mean age 3.4 years)8 and, by definition, the bladder repair is not done concurrently with urethral closure. Gearhart has reported additional complications after complete repair done without osteotomies at other institutions, including wound dehiscence, loss of penile skin, bladder prolapse and partial loss of a hemiglans.9 The penile disassembly technique used for penile repair during CPRE has been associated with complete or partial loss of bilateral or unilateral hemiglans, loss of a corporal body, penile urethra or penile shaft skin.10 The purpose of our study was to investigate the bladder neck fistula rate after CPRE at 4 tertiary pediatric urology centers. We found a 41% bladder neck fistula rate after CPRE performed by pediatric urologists experienced with exstrophy surgery. It is too early to determine the effect this may have on long-term continence rates. The overall number of patients available for analysis for this study is low, in part due to the decreased incidence of bladder exstrophy in the United States during the last decade, as well as the relative newness of the CPRE technique in the armamentarium of exstrophy surgery. Four patients in our series had the original urethral closure reinforced with a single layer SIS onlay and none of them had fistulas. Although this technique has promise to possibly prevent fistula formation, it was used in only a small number of patients and cannot be generalized to larger numbers without further study. CONCLUSIONS
This multi-institutional study demonstrates that bladder neck fistulas occur in almost half of patients after CPRE performed by experienced pediatric urologists. While spontaneous closure is possible, most cases will eventually require repair. The long-term implications of this finding with regard
to continence and the need for additional bladder neck procedures remain to be seen. We are encouraged by the preliminary results of SIS coverage and will continue to evaluate its use at the time of primary exstrophy closure.
REFERENCES
1. Mitchell, M. E. and Bagli, D. J.: Complete penile disassembly for epispadias repair: the Mitchell technique. J Urol, 155: 300, 1996 2. Grady, R. W. and Mitchell, M. E.: Complete primary repair of exstrophy. J Urol, 162: 1415, 1999 3. Grady, R. W., Joyner, B. D., Roedel, M., Feng, W., Casale, P. and Mitchell, M. E.: The complete primary repair of exstrophy (CPRE) technique for repair of bladder exstrophy and epispadias: long-term follow-up. Presented at the Section on Urology, American Academy of Pediatrics, New Orleans, Louisiana, November 2, 2003 4. Grady, R. W. and Mitchell, M. E.: Complete primary repair of exstrophy. Surgical technique. Urol Clin North Am, 27: 569, 2000 5. Hammouda, H. M. and Kotb, H.: Complete primary repair of bladder exstrophy: initial experience with 33 cases. J Urol, 172: 1441, 2004 6. Caione, P. and Capozza, N.: Evolution of male epispadias repair: 16-year experience. J Urol, 165: 2410, 2001 7. El-Sherbiny, M. T., Hafez, A. T. and Ghoneim, M. A.: Complete repair of exstrophy: further experience with neonates and children after failed initial closure. J Urol, 168: 1692, 2002 8. Surer, I., Baker, L. A., Jeffs, R. D. and Gearhart, J. P.: The modified Cantwell-Ransley repair for exstrophy and epispadias: 10-year experience. J Urol, 164: 1040, 2000 9. Gearhart, J. P.: Complete repair of bladder exstrophy in the newborn: complications and management. J Urol, 165: 2431, 2001 10. Husmann, D. A. and Gearhart, J. P.: Loss of the penile glans and/or corpora following primary repair of the bladder exstrophy using the complete penile disassembly technique. Presented at Section on Urology, American Academy of Pediatrics, New Orleans, Louisiana, November 1, 2003 EDITORIAL COMMENT The “complete” repair using the Mitchell technique has been increasingly performed for reconstruction for bladder exstrophy. As more centers perform this procedure in infants, complications are increasingly being identified. This report from 4 respected institutions in the management of complex pediatric urological cases indicates the difficulties associated with reconstruction of this most major of urological abnormalities. The authors indicate a 41% incidence of fistulas, all of which occurred in males and the majority required later operative management. The authors also noted a higher incidence in those cases treated with later closure, and suggest that the use of SIS may prevent development of fistulas. However, SIS to prevent fistula at the penopubic junction was not statistically significant in their series. Since many of these patients will need later bladder neck reconstruction, it will be interesting to see if the location of the SIS at the bladder neck will present an impediment to dissection, bladder neck mobilization and successful reconstruction. As the number of children treated with CPRE increases, the results must be compared to those of the current standard management, ie modern staged reconstruction. Furthermore, when complete reconstruction is explained to parents, we should indicate that many children will require later surgical procedures, such as ureteral reimplantation, hypospadias repair and bladder neck reconstruction. Additionally, since bladder exstrophy is a relatively rare anomaly, most pediatric urologists will perform exstrophy closure every 1 to 2 years. Given this infrequent surgical exposure, being conservative is perhaps in the best interest of our patients, and pediatric urologists lacking expertise in CPRE should be forewarned. Ranjiv Mathews Division of Pediatric Urology Johns Hopkins Hospital Baltimore, Maryland
1690
FISTULA AFTER PRIMARY REPAIR OF EXSTROPHY
DISCUSSION Dr. Joao L. Pippi-Salle. Why do you think you had such a high incidence of fistula that did not close spontaneously? In my opinion you have to really dissect proximally the bladder neck and the proximal urethra, and place down and cover the proximal urethra with corpora cavernosa rotated on top, and then you usually do not have a fistula. Do you think that could be a factor? Dr. Seth A. Alpert. That is a possibility. In our study we did not find any 1 factor that was associated with this increased fistula rate. The fact that this area is thinner than the other areas of the urethra and it is more difficult to get other multiple layers of coverage could be possibilities. A couple of our patients were wrapped with SIS and they did not have a fistula, and so we thought that deserves future study. The patients who underwent earlier repairs and did not have osteotomies also had fewer fistulas.