- Email: [email protected]
Long-Term Comparative Outcomes Between 2 Common Ureteroneocystostomy Techniques for Renal Transplantation
Long-Term Comparative Outcomes Between 2 Common Ureteroneocystostomy Techniques for Renal Transplantation Jeffrey L. Veale,* Jay Yew, David W. Gjertson, Craig V. Smith, Jennifer S. Singer, J. Thomas Rosenthal and H. Albin Gritsch From the Departments of Urology (JLV, JSS, JTT, HAG) and Biostatistics (DWG), David Geffen School of Medicine, University of California-Los Angeles, Los Angeles, Department of Urology, Sharp Rees-Stealy Medical Group (JY), San Diego and Department of Surgery, City of Hope National Medical Center (CVS), Duarte, California
Purpose: We compared the incidence of ureteral complications between the classic (Lich-Gregoir) technique and the recently popularized single stitch (Shanfield) technique in renal transplantation. Materials and Methods: The charts of 721 consecutive transplant recipients from May 1999 to July 2002 were retrospectively reviewed. Ureteral and nonureteral complications were reviewed at 3 to 5-year followup. Results: Of the 721 recipients evaluated 713 were included in the study. There were 360 recipients in the Lich-Gregoir group and 353 in the Shanfield group. A significantly higher rate of ureteral complications occurred in the Shanfield group compared to the Lich-Gregoir group (15.6% vs 3.9%, p ⬍0.0001). The Shanfield group consisted of 20 patients with ureteral leakage, 21 with hematuria, 11 with strictures and 3 who had ureteral stones. The Lich-Gregoir group had 8 patients with ureteral leakage, 5 with hematuria and 1 with a stricture. In comparison, urinary tract infections, delayed graft function and rejection rates were not significantly different between the 2 groups (p ⫽ 0.76, 0.12 and 0.19, respectively). Conclusions: In contrast to other reports, the Shanfield group had significantly more ureteral complications. In particular the Shanfield technique may predispose patients to higher rates of hematuria and stone formation. Based on this large series and published meta-analyses we believe that the stented Lich-Gregoir anastomosis is the superior ureteroneocystostomy technique in renal transplantation. Key Words: transplantation, kidney, ureter, complications
associated with its more intrusive bladder dissection.3,4 Today the most common technique used in renal transplantation is the Lich-Gregoir method.5,6 This is an extravesical approach that was first presented by Gregoir at the German Congress of Surgery in April 1961. The technique was later popularized in America by Lich, who published the technique in the Journal of Urology® in November 1961.5 The LichGregoir technique uses multiple interrupted absorbable sutures to create the ureteral-mucosal anastomosis (fig. 1). Other methods of creating extravesical ureteroneocystostomy are the Shanfield,7 Taguchi8 and Barry9 techniques. The Shanfield single stitch technique has gained recent popularity in part due to its speed and ease of completion. The creation of this form of ureteroneocystostomy was first described in 1972 by Shanfield.7 This technique uses only 1 stitch for the mucosal anastomosis (fig. 2). It was incorrectly credited to Taguchi by different groups in the past.4,10,11 Although it is similar to the Shanfield technique, the Taguchi technique uses 2 stitches (fig. 3). Some groups credited the Shanfield single stitch technique with having ureteral complications equivalent to those of the Lich-Gregoir technique,4,12 whereas others reported increased ureteral complications when comparing the Shanfield single stitch and Lich-Gregoir techniques.10,13 We compared long-term ureteral complication rates between the Lich-Gregoir and Shanfield single stitch ureteroneocystostomy techniques in renal transplant recipients at a single center.
uring the early years of experimental renal transplantation kidneys were placed in the thigh with cutaneous ureterostomy drainage or in the renal fossa with drainage by ureteroureterostomy to the recipient native ureter. In January 1951 Kuss and Teinturier placed a donor kidney into the iliopelvic region of the recipient with cutaneous ureterostomy.1 Soon thereafter other Parisian surgical teams established the concept that an allograft placed in the pelvic region could also accommodate a short ureteral segment for bladder drainage. Although these early attempts at kidney transplantation resulted in failure, in 1954 Murray and Harrison completed the first successful renal transplantation.2 This successful transplant between monozygotic twins reinforced the notion that the pelvic location with bladder drainage was the most physiological and natural position for the renal allograft. Although much of the transplant operation has remained the same, a number of techniques have been used for the ureterovesical anastomosis in the last 50 years. The Politano-Leadbetter technique of intravesical ureteroneocystostomy was commonly used during the 1960s. However, this method is now rarely used due to increased complications
D
Submitted for publication March 14, 2006. * Correspondence: Renal Transplantation, Department of Urology, David Geffen School of Medicine, University of California-Los Angeles, 10833 Le Conte Ave., 63-276 CHS, Los Angeles, California 90095 (telephone: 310-794-7152; FAX: 310-206-5343; e-mail: [email protected]).
0022-5347/07/1772-0632/0 THE JOURNAL OF UROLOGY® Copyright © 2007 by AMERICAN UROLOGICAL ASSOCIATION
632
Vol. 177, 632-636, February 2007 Printed in U.S.A. DOI:10.1016/j.juro.2006.09.042
URETERONEOCYSTOSTOMY TECHNIQUES FOR RENAL TRANSPLANTATION
633
used for hemostasis of the ureteral vein and the distal ureter was spatulated. The bladder was then distended to approximately 75% capacity by hanging a 3 l normal saline bag containing 50,000 U bacitracin and instilling this solution through a standard Foley catheter (fig. 4). This system accommodates a urethral catheter of any size and allows repeat bladder filling and emptying to adjust intravesical pressure. An anterior position near the bladder dome was chosen for the ureteral anastomosis. The detrusor muscle was dissected in longitudinal fashion toward the bladder neck using electrocautery. The ureteral tunnel was approximately 5 times the width of the ureter and angled toward the bladder neck for ease of postoperative stent removal. Ureteral stent placement was left to surgeon discretion, including cases of previous bladder augmentation, ileal conduit, detrusor fibrosis and en bloc or dual transplantation. These indications for stent placement were constant throughout the study period (table 1). For the Shanfield technique each end of a double armed absorbable monofilament suture (5-zero PDS) was placed in full-thickness fashion through the distal ureter and then inside to outside through the detrusor. This detrusor stitch was placed approximately 2 cm distal to the apex of the mucosal dissection (fig. 2). The Lich-Gregoir technique involves suturing the bladder mucosa to the ureter in circumferential fashion using 6 interrupted 5-zero PDS sutures (fig. 1). A distal full-thickness anchoring suture was used to
FIG. 1. Lich-Gregoir technique. Detrusor muscle is opened down to mucosal layer using electrocautery (A). Small caudal opening is created in mucosal layer using tenotomy scissors (B). Apical stitch is placed in spatulated ureter (C) and passed inside out through most caudal portion of mucosal opening (D). Circumferential sutures are placed to complete ureterovesical anastomosis (E and F). Detrusor muscle is closed to create antireflux mechanism (G to I).
MATERIALS AND METHODS The charts of 721 consecutive transplant recipients from May 17, 1999 to July 31, 2002 were retrospectively reviewed. We excluded 8 patients from analysis due to insufficient records, leaving 713 remaining in the study. The Lich-Gregoir technique was used by all surgeons from May 17, 1999 to May 17, 2000 and then again from November 13, 2001 to July 31, 2002. Two surgeons elected to use the Shanfield technique from May 18, 2000 to November 12, 2001. This yielded 360 patients in the Lich-Gregoir group and 353 in the Shanfield group. Followup was 3 to 5 years and the 2 groups were equally matched for demographic features (table 1). We focused only on surgical complications related to the ureteral anastomosis, namely urinary leakage, ureteral stricture, hematuria and stone formation. Other complications that were reviewed for comparison were urinary tract infections, delayed graft function and rejection rates. Twosided p values were derived via Fisher’s exact test using STATA™, version 6 statistical software. Technique The technique of ureteral and bladder exposure was the same in the 2 groups. The redundant ureteral segment was transected, a 5-zero chromic tie and/or electrocautery was
FIG. 2. Shanfield technique. Double armed U-stitch is placed in distal aspect of spatulated ureter (A). Each suture is passed through bladder opening and taken out through anterior bladder wall at least 2 cm beyond edge of bladder opening (B). This brings ureteral end through bladder opening, where it is anchored down to bladder wall (C).
634
URETERONEOCYSTOSTOMY TECHNIQUES FOR RENAL TRANSPLANTATION
FIG. 3. Taguchi technique. Two double armed U-stitches are placed in spatulated ureter (A). Ureteral end follows stitches placed through mucosal opening and out of anterior bladder wall (B). Ureteral end is anchored to inner bladder wall (C).
prevent proximal migration of the ureter in the submucosal tunnel. In each technique the mentioned stent is placed intravesically after opening the mucosa with tenotomy scissors and the detrusor muscle is subsequently closed over the anastomosis using 5-zero PDS in interrupted or running fashion to create the antireflux mechanism. RESULTS
FIG. 4. Y connector. By clamping various arms of Y connector bladder can be filled and maintained with antibiotic solution or drained. Advantage of this setup over 3-way catheter is that any size of catheter can be accommodated, which is helpful for clot irrigation or in pediatric recipients.
group ureteral complication rate of 3.9% (p ⬍0.0001). Table 2 shows that urinary leakage was the most common complication in the Lich-Gregoir group, occurring in 2.2% of patients, followed by hematuria in 1.4% and ureteral stricture 0.3%. Hematuria, occurring in 5.9% of patients, was the most common complication in the Shanfield group, followed by urinary leakage in 5.7%, ureteral strictures in 3.1% and
The Shanfield group had a ureteral complication rate of 15.6%, which was significantly higher than the Lich-Gregoir TABLE 2. Complications according to procedure TABLE 1. Patient demographics
No. pts Mean ⫾ SD age % Male % Previous transplant % Oliguria preop % Previous bladder disease % Previous bladder surgery % Cadaveric donor % Living unrelated donor % Living related donor % Simultaneous with pancreas % Simultaneous with liver % Simultaneous with heart % Pediatric en bloc donor % Dual transplant Mean ⫾ SD cadaveric cold ischemia time (mins) % Ureteral stent
Lich-Gregoir
Shanfield
360 42 ⫾ 16 58 11 46 12 6 57 9 24 7 3 Less than 1 2 Less than 1 1,171 ⫾ 467
353 42 ⫾ 16 57 11 41 14 7 53 12 24 8 3 0 5 Less than 1 1,258 ⫾ 489
20
24
Any complication:* Urinary leakage Hematuria Ureteral stricture Stones Urinary tract infection Delayed graft function Any rejection: Less than 3 mos 3 Mos—1 yr Greater than 1 yr
p Value (2-sided Fisher’s exact test)
No. Lich-Gregoir (%)
No. Shanfield (%)
14 (3.9) 8 5 1 0 157 (43.6)
55 (15.6) 20 21 11 3 158 (44.8)
⬍0.0001
83 (23.0)
64 (18.1)
0.12
100 (27.8) 50 20 30
114 (32.3) 54 29 31
0.19
0.76
* For the Lich technique 14 procedures had 1 anastomotic complication each and for the Shanfield technique 50 had a total of 55 anastomotic complications, ie 5 had 2 complications each.
URETERONEOCYSTOSTOMY TECHNIQUES FOR RENAL TRANSPLANTATION stones in 0.8%. Of note 5 patients in the Shanfield group had 2 ureteral complications each. No patient in the Lich-Gregoir group had multiple ureteral complications. Other nonureteral complications were compared between the Lich-Gregoir and Shanfield techniques. There were no significant differences between urinary tract infections, delayed graft function and rejection rates between the 2 groups (p ⫽ 0.76, 0.12 and 0.19, respectively). Urinary tract infections occurred in 43.6% of patients in the Lich-Gregoir group compared to 44.8% in the Shanfield group. Delayed graft function was present in 23.0% of patients in the Lich-Gregoir group compared to 18.1% in the Shanfield group. Of the patients 27.8% had rejection in the Lich-Gregoir group compared to 32.3% in the Shanfield group. DISCUSSION A significant number of patients in the Shanfield group had gross hematuria and this association is also well documented in the literature.4,10,12,13 Secin et al commented that this complication could be minimized with meticulous stump hemostasis.4 This was not our experience because, even when the ureteral stump vessels were carefully ligated and cauterized to ensure hemostasis, hematuria continued to persist in some patients. A possible explanation of why this complication is more prominent with the Shanfield technique is that, by laying the distal ureter intravesically, the ureteral stump is exposed to urine. This exposure to urine and subsequent urokinase could theoretically dissolve the hemostatic fibrin clot. Although hematuria is considered a relatively minor complication, it resulted in some patients having to undergo the inconvenience of multiple bladder irrigations and cystoscopic procedures. Eight patients required fulguration and 1 underwent open repair for bladder rupture that occurred during manual clot irrigation. Another complication that seems to occur significantly more frequently with the Shanfield technique is bladder calculi. The origin of these stones appears to be at the ureteral stump and it may be related to the periureteral tissue serving as a foreign body or nidus for stone formation. With direct mucosa-to-mucosa apposition no patient in the Lich-Gregoir group had calculi. Other anastomotic complications, such as urinary leakage and ureteral strictures, occurred significantly more frequently with the Shanfield technique. These complications may have been more easily recognized in our study compared to others in the past due to our extended followup of 5 years and large study population of 713 patients. Depending on patient populations and immunosuppression protocols the published 1-year rejection rate is approximately 10% to 30%.14 Our calculated 1-year rejection rate in this study was 15%. Of note our 5-year rejection rate approached 30%. Although our study population included patients with increased risk factors for rejection, such as preformed alloantibody, previous transplantation, expanded criteria donor kidneys and pediatric en bloc donor kidneys, the ongoing rejection demonstrated in this patient population stresses the importance of continued followup and research to develop methods to decrease the loss of allograft function with time. There was only 1 stricture in the Lich-Gregoir group, which occurred 82 weeks after transplantation and was managed by reimplantation. In the Shanfield group 11 patients had strictures, of which 8 occurred within the first 3
635
months after transplantation. Six of these strictures were successfully balloon dilated, 4 were managed by reimplantation and 1 required ureteropyelostomy with the native ureter. We did not place stents in retrograde fashion. Rather, endoscopic management of strictures was completed by antegrade percutaneous access with antegrade balloon dilation and stent insertion.15 In a review by Goldfischer and Glenn they concluded that most post-transplantation strictures could be treated initially with percutaneous nephrostomy and antegrade balloon dilation.16 We agree with this statement. However, if the stricture was greater than 2 cm on percutaneous nephrostogram, we proceeded with open surgical repair. Six patients in the Lich-Gregoir group and 7 in the Shanfield group were diagnosed with BK virus, although only 1 had a stricture. Interestingly 71% of the patients in the Shanfield group and 100% of those in the Lich-Gregoir group who had leakage or stricture were nonstented. This is consistent with data from the Mangus and Haag meta-analysis showing that a stented anastomosis has a lower complication rate.17 Since the use of stents was not controlled in our study, we cannot determine whether the high rate of leakage in the Shanfield group could have been prevented. Urine leakage was managed by immediate exploration, distal ureter débridement and reimplantation over a ureteral stent. A weakness of the study was that voiding cystourethrograms were not routinely done after transplantation at our institution. Therefore, the incidence of reflux in the 2 groups in this series is unknown. Although the Shanfield technique has a significantly higher ureteral complication rate than the Lich-Gregoir technique, the Shanfield method remains useful when a recipient is unstable in the operating room and time is of the essence. Some surgeons at our institution recently modified the Lich-Gregoir using a running anastomosis. This modification seems to have increased the speed with which the technique can be completed without compromising the low ureteral complication rate. CONCLUSIONS Although it is rapid and technically easy to perform, Shanfield single stitch ureteroneocystostomy carries a significantly higher risk of surgical complications related to the ureteral anastomosis, most notably increased hematuria and stone formation. As such, we have adopted the LichGregoir technique of ureteroneocystostomy in our renal transplant recipients. ACKNOWLEDGMENTS Michelle Moeck provided the illustrations.
Abbreviations and Acronyms PDS ⫽ polydioxanone REFERENCES 1.
Kuss R: Human renal transplantation memories 1951 to 1981. In: History of Transplantation: Thirty-Five Recollections. Edited by P Terasaki. Los Angeles: University of California-
636
2. 3. 4.
5. 6. 7. 8.
9. 10.
URETERONEOCYSTOSTOMY TECHNIQUES FOR RENAL TRANSPLANTATION Los Angeles Tissue Typing Laboratory, Regents of the University of California 1991; pp 3-4 and 39-40. Harrison JH, Merrill JP and Murray JE: Renal homotransplantation in identical twins. Surg Forum 1956; 6: 432. Politano VA and Leadbetter WF: An operative technique for the correction of vesicoureteral reflux. J Urol 1958; 79: 932. Secin FP, Rovegno AR, Marrugat RE, Virasoro R, Lautersztein GA and Fernandez H: Comparing Taguchi and Lich-Gregoir ureterovesical reimplantation techniques for kidney transplants. J Urol 2002; 168: 926. Lich R, Howerton LW and Davis LA: Recurrent urosepsis in children. J Urol 1961; 86: 554. Gregoir W: Le reflux vesico-ureteral congenital. Acta Urol Belg 1962; 30: 286. Shanfield I: New experimental methods for implantation of the ureter in bladder and conduit. Transplant Proc 1972; 4: 637. MacKinnon KJ, Oliver JA, Morehouse DD and Taguchi Y: Cadaver renal transplantation: Emphasis on urological aspects. J Urol 1968; 99: 486. Barry JM: Unstented extravesical ureteroneocystostomy in kidney transplantation. J Urol 1983; 129: 918. Zargar MA, Shahrokh H, Mohammadi Fallah MR and Zargar H: Comparing Taguchi and anterior Lich-Gregoir ureterovesical reimplantation techniques for kidney transplantation. Transplant Proc 2005; 37: 3077.
11.
12.
13.
14.
15.
16. 17.
Caparros J, Regalado RI, Sanchez-Martin F and Villavicencio H: A simplified technique for ureteroneocystostomy in renal transplantation. World J Urol 1996; 14: 236. Modlin CS, Flechner SM, Boparai N, Goldfarb DA and Novick AC: U-stitch ureteroneocystostomy: a new renal transplantation ureteral reimplantation technique associated with reduced urological complications. Tech Urol 2000; 6: 1. Tzimas GN, Hayati H, Tchervenkov JI and Metrakos PP: Ureteral implantation technique and urologic complications in adult kidney transplantation. Transplant Proc 2003; 35: 2420. Hariharan S, Johnson CP, Bresnahan BA, Taranto SE, McIntosh MJ and Stablein D: Improved graft survival after renal transplantation in the United States, 1988 –1996. N Engl J Med 2000; 342: 605. Kristo B, Phelan MW, Gritsch HA and Schulam PG: Treatment of renal transplant ureterovesical anastomotic strictures using antegrade balloon dilation with or without holmium:YAG laser endoureterotomy. Urology 2003; 62: 831. Goldfischer ER and Glenn SG: Endoscopic management of ureteral strictures. J Urol 1997; 157: 770. Mangus RS and Haag BW: Stented versus nonstented extravesical ureteroneocystostomy in renal transplantation: a meta-analysis. Am J Transplant 2004; 4: 1889.
Purpose: We compared the incidence of ureteral complications between the classic (Lich-Gregoir) technique and the recently popularized single stitch (Shanfield) technique in renal transplantation. Materials and Methods: The charts of 721 consecutive transplant recipients from May 1999 to July 2002 were retrospectively reviewed. Ureteral and nonureteral complications were reviewed at 3 to 5-year followup. Results: Of the 721 recipients evaluated 713 were included in the study. There were 360 recipients in the Lich-Gregoir group and 353 in the Shanfield group. A significantly higher rate of ureteral complications occurred in the Shanfield group compared to the Lich-Gregoir group (15.6% vs 3.9%, p ⬍0.0001). The Shanfield group consisted of 20 patients with ureteral leakage, 21 with hematuria, 11 with strictures and 3 who had ureteral stones. The Lich-Gregoir group had 8 patients with ureteral leakage, 5 with hematuria and 1 with a stricture. In comparison, urinary tract infections, delayed graft function and rejection rates were not significantly different between the 2 groups (p ⫽ 0.76, 0.12 and 0.19, respectively). Conclusions: In contrast to other reports, the Shanfield group had significantly more ureteral complications. In particular the Shanfield technique may predispose patients to higher rates of hematuria and stone formation. Based on this large series and published meta-analyses we believe that the stented Lich-Gregoir anastomosis is the superior ureteroneocystostomy technique in renal transplantation. Key Words: transplantation, kidney, ureter, complications
associated with its more intrusive bladder dissection.3,4 Today the most common technique used in renal transplantation is the Lich-Gregoir method.5,6 This is an extravesical approach that was first presented by Gregoir at the German Congress of Surgery in April 1961. The technique was later popularized in America by Lich, who published the technique in the Journal of Urology® in November 1961.5 The LichGregoir technique uses multiple interrupted absorbable sutures to create the ureteral-mucosal anastomosis (fig. 1). Other methods of creating extravesical ureteroneocystostomy are the Shanfield,7 Taguchi8 and Barry9 techniques. The Shanfield single stitch technique has gained recent popularity in part due to its speed and ease of completion. The creation of this form of ureteroneocystostomy was first described in 1972 by Shanfield.7 This technique uses only 1 stitch for the mucosal anastomosis (fig. 2). It was incorrectly credited to Taguchi by different groups in the past.4,10,11 Although it is similar to the Shanfield technique, the Taguchi technique uses 2 stitches (fig. 3). Some groups credited the Shanfield single stitch technique with having ureteral complications equivalent to those of the Lich-Gregoir technique,4,12 whereas others reported increased ureteral complications when comparing the Shanfield single stitch and Lich-Gregoir techniques.10,13 We compared long-term ureteral complication rates between the Lich-Gregoir and Shanfield single stitch ureteroneocystostomy techniques in renal transplant recipients at a single center.
uring the early years of experimental renal transplantation kidneys were placed in the thigh with cutaneous ureterostomy drainage or in the renal fossa with drainage by ureteroureterostomy to the recipient native ureter. In January 1951 Kuss and Teinturier placed a donor kidney into the iliopelvic region of the recipient with cutaneous ureterostomy.1 Soon thereafter other Parisian surgical teams established the concept that an allograft placed in the pelvic region could also accommodate a short ureteral segment for bladder drainage. Although these early attempts at kidney transplantation resulted in failure, in 1954 Murray and Harrison completed the first successful renal transplantation.2 This successful transplant between monozygotic twins reinforced the notion that the pelvic location with bladder drainage was the most physiological and natural position for the renal allograft. Although much of the transplant operation has remained the same, a number of techniques have been used for the ureterovesical anastomosis in the last 50 years. The Politano-Leadbetter technique of intravesical ureteroneocystostomy was commonly used during the 1960s. However, this method is now rarely used due to increased complications
D
Submitted for publication March 14, 2006. * Correspondence: Renal Transplantation, Department of Urology, David Geffen School of Medicine, University of California-Los Angeles, 10833 Le Conte Ave., 63-276 CHS, Los Angeles, California 90095 (telephone: 310-794-7152; FAX: 310-206-5343; e-mail: [email protected]).
0022-5347/07/1772-0632/0 THE JOURNAL OF UROLOGY® Copyright © 2007 by AMERICAN UROLOGICAL ASSOCIATION
632
Vol. 177, 632-636, February 2007 Printed in U.S.A. DOI:10.1016/j.juro.2006.09.042
URETERONEOCYSTOSTOMY TECHNIQUES FOR RENAL TRANSPLANTATION
633
used for hemostasis of the ureteral vein and the distal ureter was spatulated. The bladder was then distended to approximately 75% capacity by hanging a 3 l normal saline bag containing 50,000 U bacitracin and instilling this solution through a standard Foley catheter (fig. 4). This system accommodates a urethral catheter of any size and allows repeat bladder filling and emptying to adjust intravesical pressure. An anterior position near the bladder dome was chosen for the ureteral anastomosis. The detrusor muscle was dissected in longitudinal fashion toward the bladder neck using electrocautery. The ureteral tunnel was approximately 5 times the width of the ureter and angled toward the bladder neck for ease of postoperative stent removal. Ureteral stent placement was left to surgeon discretion, including cases of previous bladder augmentation, ileal conduit, detrusor fibrosis and en bloc or dual transplantation. These indications for stent placement were constant throughout the study period (table 1). For the Shanfield technique each end of a double armed absorbable monofilament suture (5-zero PDS) was placed in full-thickness fashion through the distal ureter and then inside to outside through the detrusor. This detrusor stitch was placed approximately 2 cm distal to the apex of the mucosal dissection (fig. 2). The Lich-Gregoir technique involves suturing the bladder mucosa to the ureter in circumferential fashion using 6 interrupted 5-zero PDS sutures (fig. 1). A distal full-thickness anchoring suture was used to
FIG. 1. Lich-Gregoir technique. Detrusor muscle is opened down to mucosal layer using electrocautery (A). Small caudal opening is created in mucosal layer using tenotomy scissors (B). Apical stitch is placed in spatulated ureter (C) and passed inside out through most caudal portion of mucosal opening (D). Circumferential sutures are placed to complete ureterovesical anastomosis (E and F). Detrusor muscle is closed to create antireflux mechanism (G to I).
MATERIALS AND METHODS The charts of 721 consecutive transplant recipients from May 17, 1999 to July 31, 2002 were retrospectively reviewed. We excluded 8 patients from analysis due to insufficient records, leaving 713 remaining in the study. The Lich-Gregoir technique was used by all surgeons from May 17, 1999 to May 17, 2000 and then again from November 13, 2001 to July 31, 2002. Two surgeons elected to use the Shanfield technique from May 18, 2000 to November 12, 2001. This yielded 360 patients in the Lich-Gregoir group and 353 in the Shanfield group. Followup was 3 to 5 years and the 2 groups were equally matched for demographic features (table 1). We focused only on surgical complications related to the ureteral anastomosis, namely urinary leakage, ureteral stricture, hematuria and stone formation. Other complications that were reviewed for comparison were urinary tract infections, delayed graft function and rejection rates. Twosided p values were derived via Fisher’s exact test using STATA™, version 6 statistical software. Technique The technique of ureteral and bladder exposure was the same in the 2 groups. The redundant ureteral segment was transected, a 5-zero chromic tie and/or electrocautery was
FIG. 2. Shanfield technique. Double armed U-stitch is placed in distal aspect of spatulated ureter (A). Each suture is passed through bladder opening and taken out through anterior bladder wall at least 2 cm beyond edge of bladder opening (B). This brings ureteral end through bladder opening, where it is anchored down to bladder wall (C).
634
URETERONEOCYSTOSTOMY TECHNIQUES FOR RENAL TRANSPLANTATION
FIG. 3. Taguchi technique. Two double armed U-stitches are placed in spatulated ureter (A). Ureteral end follows stitches placed through mucosal opening and out of anterior bladder wall (B). Ureteral end is anchored to inner bladder wall (C).
prevent proximal migration of the ureter in the submucosal tunnel. In each technique the mentioned stent is placed intravesically after opening the mucosa with tenotomy scissors and the detrusor muscle is subsequently closed over the anastomosis using 5-zero PDS in interrupted or running fashion to create the antireflux mechanism. RESULTS
FIG. 4. Y connector. By clamping various arms of Y connector bladder can be filled and maintained with antibiotic solution or drained. Advantage of this setup over 3-way catheter is that any size of catheter can be accommodated, which is helpful for clot irrigation or in pediatric recipients.
group ureteral complication rate of 3.9% (p ⬍0.0001). Table 2 shows that urinary leakage was the most common complication in the Lich-Gregoir group, occurring in 2.2% of patients, followed by hematuria in 1.4% and ureteral stricture 0.3%. Hematuria, occurring in 5.9% of patients, was the most common complication in the Shanfield group, followed by urinary leakage in 5.7%, ureteral strictures in 3.1% and
The Shanfield group had a ureteral complication rate of 15.6%, which was significantly higher than the Lich-Gregoir TABLE 2. Complications according to procedure TABLE 1. Patient demographics
No. pts Mean ⫾ SD age % Male % Previous transplant % Oliguria preop % Previous bladder disease % Previous bladder surgery % Cadaveric donor % Living unrelated donor % Living related donor % Simultaneous with pancreas % Simultaneous with liver % Simultaneous with heart % Pediatric en bloc donor % Dual transplant Mean ⫾ SD cadaveric cold ischemia time (mins) % Ureteral stent
Lich-Gregoir
Shanfield
360 42 ⫾ 16 58 11 46 12 6 57 9 24 7 3 Less than 1 2 Less than 1 1,171 ⫾ 467
353 42 ⫾ 16 57 11 41 14 7 53 12 24 8 3 0 5 Less than 1 1,258 ⫾ 489
20
24
Any complication:* Urinary leakage Hematuria Ureteral stricture Stones Urinary tract infection Delayed graft function Any rejection: Less than 3 mos 3 Mos—1 yr Greater than 1 yr
p Value (2-sided Fisher’s exact test)
No. Lich-Gregoir (%)
No. Shanfield (%)
14 (3.9) 8 5 1 0 157 (43.6)
55 (15.6) 20 21 11 3 158 (44.8)
⬍0.0001
83 (23.0)
64 (18.1)
0.12
100 (27.8) 50 20 30
114 (32.3) 54 29 31
0.19
0.76
* For the Lich technique 14 procedures had 1 anastomotic complication each and for the Shanfield technique 50 had a total of 55 anastomotic complications, ie 5 had 2 complications each.
URETERONEOCYSTOSTOMY TECHNIQUES FOR RENAL TRANSPLANTATION stones in 0.8%. Of note 5 patients in the Shanfield group had 2 ureteral complications each. No patient in the Lich-Gregoir group had multiple ureteral complications. Other nonureteral complications were compared between the Lich-Gregoir and Shanfield techniques. There were no significant differences between urinary tract infections, delayed graft function and rejection rates between the 2 groups (p ⫽ 0.76, 0.12 and 0.19, respectively). Urinary tract infections occurred in 43.6% of patients in the Lich-Gregoir group compared to 44.8% in the Shanfield group. Delayed graft function was present in 23.0% of patients in the Lich-Gregoir group compared to 18.1% in the Shanfield group. Of the patients 27.8% had rejection in the Lich-Gregoir group compared to 32.3% in the Shanfield group. DISCUSSION A significant number of patients in the Shanfield group had gross hematuria and this association is also well documented in the literature.4,10,12,13 Secin et al commented that this complication could be minimized with meticulous stump hemostasis.4 This was not our experience because, even when the ureteral stump vessels were carefully ligated and cauterized to ensure hemostasis, hematuria continued to persist in some patients. A possible explanation of why this complication is more prominent with the Shanfield technique is that, by laying the distal ureter intravesically, the ureteral stump is exposed to urine. This exposure to urine and subsequent urokinase could theoretically dissolve the hemostatic fibrin clot. Although hematuria is considered a relatively minor complication, it resulted in some patients having to undergo the inconvenience of multiple bladder irrigations and cystoscopic procedures. Eight patients required fulguration and 1 underwent open repair for bladder rupture that occurred during manual clot irrigation. Another complication that seems to occur significantly more frequently with the Shanfield technique is bladder calculi. The origin of these stones appears to be at the ureteral stump and it may be related to the periureteral tissue serving as a foreign body or nidus for stone formation. With direct mucosa-to-mucosa apposition no patient in the Lich-Gregoir group had calculi. Other anastomotic complications, such as urinary leakage and ureteral strictures, occurred significantly more frequently with the Shanfield technique. These complications may have been more easily recognized in our study compared to others in the past due to our extended followup of 5 years and large study population of 713 patients. Depending on patient populations and immunosuppression protocols the published 1-year rejection rate is approximately 10% to 30%.14 Our calculated 1-year rejection rate in this study was 15%. Of note our 5-year rejection rate approached 30%. Although our study population included patients with increased risk factors for rejection, such as preformed alloantibody, previous transplantation, expanded criteria donor kidneys and pediatric en bloc donor kidneys, the ongoing rejection demonstrated in this patient population stresses the importance of continued followup and research to develop methods to decrease the loss of allograft function with time. There was only 1 stricture in the Lich-Gregoir group, which occurred 82 weeks after transplantation and was managed by reimplantation. In the Shanfield group 11 patients had strictures, of which 8 occurred within the first 3
635
months after transplantation. Six of these strictures were successfully balloon dilated, 4 were managed by reimplantation and 1 required ureteropyelostomy with the native ureter. We did not place stents in retrograde fashion. Rather, endoscopic management of strictures was completed by antegrade percutaneous access with antegrade balloon dilation and stent insertion.15 In a review by Goldfischer and Glenn they concluded that most post-transplantation strictures could be treated initially with percutaneous nephrostomy and antegrade balloon dilation.16 We agree with this statement. However, if the stricture was greater than 2 cm on percutaneous nephrostogram, we proceeded with open surgical repair. Six patients in the Lich-Gregoir group and 7 in the Shanfield group were diagnosed with BK virus, although only 1 had a stricture. Interestingly 71% of the patients in the Shanfield group and 100% of those in the Lich-Gregoir group who had leakage or stricture were nonstented. This is consistent with data from the Mangus and Haag meta-analysis showing that a stented anastomosis has a lower complication rate.17 Since the use of stents was not controlled in our study, we cannot determine whether the high rate of leakage in the Shanfield group could have been prevented. Urine leakage was managed by immediate exploration, distal ureter débridement and reimplantation over a ureteral stent. A weakness of the study was that voiding cystourethrograms were not routinely done after transplantation at our institution. Therefore, the incidence of reflux in the 2 groups in this series is unknown. Although the Shanfield technique has a significantly higher ureteral complication rate than the Lich-Gregoir technique, the Shanfield method remains useful when a recipient is unstable in the operating room and time is of the essence. Some surgeons at our institution recently modified the Lich-Gregoir using a running anastomosis. This modification seems to have increased the speed with which the technique can be completed without compromising the low ureteral complication rate. CONCLUSIONS Although it is rapid and technically easy to perform, Shanfield single stitch ureteroneocystostomy carries a significantly higher risk of surgical complications related to the ureteral anastomosis, most notably increased hematuria and stone formation. As such, we have adopted the LichGregoir technique of ureteroneocystostomy in our renal transplant recipients. ACKNOWLEDGMENTS Michelle Moeck provided the illustrations.
Abbreviations and Acronyms PDS ⫽ polydioxanone REFERENCES 1.
Kuss R: Human renal transplantation memories 1951 to 1981. In: History of Transplantation: Thirty-Five Recollections. Edited by P Terasaki. Los Angeles: University of California-
636
2. 3. 4.
5. 6. 7. 8.
9. 10.
URETERONEOCYSTOSTOMY TECHNIQUES FOR RENAL TRANSPLANTATION Los Angeles Tissue Typing Laboratory, Regents of the University of California 1991; pp 3-4 and 39-40. Harrison JH, Merrill JP and Murray JE: Renal homotransplantation in identical twins. Surg Forum 1956; 6: 432. Politano VA and Leadbetter WF: An operative technique for the correction of vesicoureteral reflux. J Urol 1958; 79: 932. Secin FP, Rovegno AR, Marrugat RE, Virasoro R, Lautersztein GA and Fernandez H: Comparing Taguchi and Lich-Gregoir ureterovesical reimplantation techniques for kidney transplants. J Urol 2002; 168: 926. Lich R, Howerton LW and Davis LA: Recurrent urosepsis in children. J Urol 1961; 86: 554. Gregoir W: Le reflux vesico-ureteral congenital. Acta Urol Belg 1962; 30: 286. Shanfield I: New experimental methods for implantation of the ureter in bladder and conduit. Transplant Proc 1972; 4: 637. MacKinnon KJ, Oliver JA, Morehouse DD and Taguchi Y: Cadaver renal transplantation: Emphasis on urological aspects. J Urol 1968; 99: 486. Barry JM: Unstented extravesical ureteroneocystostomy in kidney transplantation. J Urol 1983; 129: 918. Zargar MA, Shahrokh H, Mohammadi Fallah MR and Zargar H: Comparing Taguchi and anterior Lich-Gregoir ureterovesical reimplantation techniques for kidney transplantation. Transplant Proc 2005; 37: 3077.
11.
12.
13.
14.
15.
16. 17.
Caparros J, Regalado RI, Sanchez-Martin F and Villavicencio H: A simplified technique for ureteroneocystostomy in renal transplantation. World J Urol 1996; 14: 236. Modlin CS, Flechner SM, Boparai N, Goldfarb DA and Novick AC: U-stitch ureteroneocystostomy: a new renal transplantation ureteral reimplantation technique associated with reduced urological complications. Tech Urol 2000; 6: 1. Tzimas GN, Hayati H, Tchervenkov JI and Metrakos PP: Ureteral implantation technique and urologic complications in adult kidney transplantation. Transplant Proc 2003; 35: 2420. Hariharan S, Johnson CP, Bresnahan BA, Taranto SE, McIntosh MJ and Stablein D: Improved graft survival after renal transplantation in the United States, 1988 –1996. N Engl J Med 2000; 342: 605. Kristo B, Phelan MW, Gritsch HA and Schulam PG: Treatment of renal transplant ureterovesical anastomotic strictures using antegrade balloon dilation with or without holmium:YAG laser endoureterotomy. Urology 2003; 62: 831. Goldfischer ER and Glenn SG: Endoscopic management of ureteral strictures. J Urol 1997; 157: 770. Mangus RS and Haag BW: Stented versus nonstented extravesical ureteroneocystostomy in renal transplantation: a meta-analysis. Am J Transplant 2004; 4: 1889.