EXPERIMENTAL AND CLINICAL TRANSPLANTATION

Objectives: Transplant surgeons use a myriad of ureteral anastomotic techniques in renal transplant. Although the Lich-Gregoir extravesical anastomosis is the most common, ureteroureterostomy also is used. In this meta-analysis, our objective was to compare the complication rates of these 2 techniques as reported in the literature.

Materials and Methods: A systematic review of the literature revealed 44 articles, 6 of which met our inclusion criteria. Studies were compiled using Review Manager (RevMan version 5.3, Nordic Cochrane Centre, Cochrane Collaboration, Copenhagen, Denmark). Forest plots were generated to assess relative risk. A fixed-effects model was used for low heterogeneity, and a random-effects model was used for high heterogeneity.

Results: Overall complications were similar for both procedures (relative risk, 1.22; 95% confidence interval, 0.9-1.65), as were rates of urine leak and fistula (relative risk, 0.79; 95% confidence interval, 0.17-3.64) and hematuria (relative risk, 0.24; 95% confidence interval, 0.001-4.84). Stricture, obstruction, and stone formation were more common after ureteroureterostomy (relative risk, 0.63; 95% confidence interval, 0.45-0.88), whereas vesicoureteral reflux (relative risk, 6.82; 95% confidence interval, 1.68-27.61) and urinary tract infection (relative risk, 2.29; 95% confidence interval, 1.3-4.03) were more common after ureteroneocystostomy.

Conclusions: With similar overall complication rates, both procedures can be viewed as being acceptable primary anastomotic techniques. In light of differing individual complication rates and the scarcity of data comparing the 2 methods, no specific recommendation regarding that technique should be used can currently be elucidated. We believe that further prospective studies comparing ureteroneocystostomy and primary ureteroureterostomy may reveal which is superior regarding complication rates.

Key words : Kidney, Complications, Ureteral anastomosis

The incidence of urologic complications has decreased since the inception of kidney transplant, with present complications rarely threatening patient or graft survival. Improvements in suture material and surgical procedure techniques have been largely responsible for this decrease. Historically, Politano-Leadbetter internal ureteroneocystostomy has been used as the primary ureteral reconstruction technique in renal transplant; however, it fell out of favor due to high complication and patient mortality rates.^1-3

External ureteroneocystostomy techniques, which avoid creation of a large cystotomy and have shorter operative times, have led to reduced complication rates from 10% to 6.3%.^1,4 The most common of these is the Lich-Gregoir anastomosis technique, which uses a 3- to 4-cm detrusor myotomy to expose bladder mucosa. A small cystotomy is created in the mucosa to which the spatulated ureter is anastomosed, which results in closure of the detrusor over the ureter and thus creates a nonrefluxing anastomosis. The full-thickness method forgoes the extended detrusor myotomy and uses a small cystotomy and anastomosis with large full-thickness bites in both the ureter and bladder.

Ureteroureterostomy was not commonly used in primary transplant procedures but is typically reserved for complicated cases, repeat anastomoses for stenosis, and cases with a short graft ureter. Recently, it was noted that, after ureteroureterostomy, patients seemed to have fewer complications, with the surgical technique no more technically challenging than an external ureteroneocystostomy.^5,6Ureteroureterostomy confers the added benefit of early Foley catheter removal, maintenance of natural antirefluxing anatomy, and maintenance of an orthotopic orifice for endoscopic access if needed.

Common complications associated with ureteral reconstruction in renal transplant include stricture, obstruction, urolithiasis, hematuria, urinary tract infection, and fistula. Ureteral implant methods are not tracked by the United Network of Organ Sharing or Scientific Registry of Transplant Recipients databases, and few papers have directly compared the complication rates of ureteroneocystostomy versus ureteroureterostomy. Here, our aim was to review the literature to elucidate whether ureteroureterostomy confers complication rates similar to the more commonly practiced ureteroneocystostomy.

Materials and Methods

All studies designed to evaluate ureteroureterostomy in kidney transplant were included in our meta-analysis. Studies were included regardless of type of kidney transplant (living or deceased donor), age, or other comorbidities. Case reports were excluded. Outcome measures were urologic complications after ureteroureterostomy. Urologic complications also were evaluated for ureterovesical anastomosis when it was studied in the original paper. Complications were as follows: (1) ureteral stricture, including stone formation and obstruction, (2) urine leak, including urinary fistula and major extravasation, (3) hematuria, (4) vesicoureteral reflux (VUR), and (5) urinary tract infection.

Search methods
One investigator (JO) performed a systematic literature search in the Cochrane Database of systematic reviews and PubMed Medline (November 1979 to October 2011). Databases were queried for papers that discussed complications of ureteral anastomosis following kidney transplant, in which the type of anastomotic technique was discussed. Search terms used were renal transplant, anastomosis, complications, ureteroureterostomy, ureteroneocystostomy, intravesical, and extravesical. No authors were contacted. Titles and abstracts were screened independently by ZB and TS for collection of studies relevant to the meta-analysis. If the article met selection criteria, the full text was retrieved when possible and analyzed. Each article was assessed independently by ZB and TS, and data extraction was performed by ZB using a standard data extraction form. Studies were included in the meta-analysis after discussion among the authors and consensus was reached.

Surgical technique
Ureteroureterostomy uses a small incision in the recipient’s ureter to pass a double J stent ureteral stent to confirm patency of distal ureter. The donor ureter is then spatulated, and the length is adjusted to ensure a tension-free anastomosis. The stent is then passed proximally to the donor ureter, at which point the anastomosis is completed with a running suture.

Ureteroneocystostomy includes several techniques. In the Lich-Gregoir anastomosis technique, the bladder mucosa is reached via a single cystotomy, and the anastomosis is performed between the distal ureter and the mucosa. The detrusor muscle is oversewn to provide an antireflux mechanism. The Politano-Leadbetter technique is an intravesical technique that uses a cranial cystotomy to access the interior of the bladder and a second cystotomy to introduce the ureter into the bladder. Once introduced, the ureter is tunneled under the mucosa for several millimeters before it is anastomosed. The detrusor muscle is then approximated to provide an antireflux mechanism. In the full-thickness technique, the ureter is anastomosed to the bladder wall, with full thickness stitches without tunneling. Finally, the U-stitch technique is performed by placing 1 or 2 absorbable sutures at the distal tip of the ureter. These are then brought through the bladder wall.

Statistical analyses
Data were analyzed using Review Manager (RevMan version 5.3, Nordic Cochrane Centre, Cochrane Collaboration, Copenhagen, Denmark). Heterogeneity was assessed using the I² test. Data were pooled using a fixed-effects model when studies displayed low heterogeneity (I² < 25%) and a random-effects model for high heterogeneity (I² > 75%). No studies displayed intermediate heterogeneity. Relative risk (RR) results were calculated using the Mantel-Haenszel test with confidence interval (CI) set at 95%. P values < .05 were considered as statistically significant. Forest plots were constructed for each variable.

A total of 44 relevant articles published from November 1979 to October 2011 were identified. After titles and abstracts were screened, 8 articles were selected for further analyses. Two of these were removed due to duplication of data from other studies, resulting in 6 total articles. One article was a randomized prospective control trial, whereas the remaining 5 were observational studies. The full texts of 2 observational studies were in a foreign language (1 in French, 1 in Croatian). The full text for the Croatian article was available for translation. Data for the French study were obtained from the English abstract. Study sample sizes varied from 75 to 1287, including both ureteroureterostomy and ureteroneocystostomy. The number of ureteroureterostomies performed varied from 16 to 311. In all studies, 4756 patients were included, of which 1277 underwent ureteroureterostomy and 3479 underwent ureteroneocystostomy. In the single prospective study, randomization was performed through alternation. Characteristics of the studies are shown in Table 1.^1,5-9

The overall complication rate was not significantly different between the 2 groups (RR of 1.22; 95% CI, 0.90-1.65; P = .20) (Figure 1). Homogeneity among the studies was excellent (I² = 0%). Five papers discussed stricture, obstruction, and stone complications (Figure 2). Four studies identified an increased incidence of stenosis after ureteroureterostomy, with only 1 of these reaching statistical significance. The overall result was an increased risk of stricture, obstruction, and stone formation after ureteroureterostomy (RR of 0.63; P = .006), with low heterogeneity (I² = 8%).

Hematuria was evaluated in 1 study, which found no significant difference (Figure 3). All 6 studies compared rates of urine leak and fistula formation (Figure 4), with only 1 of these evaluating ureteroureterostomy. One study displayed a higher incidence of leak and fistula after ureteroureterostomy (RR of 0.13; 95% CI, 0.07-0.24). None of the other 4 studies found a significant difference between the 2 procedures, leading to an overall difference that was not significant between the 2 procedures (RR of 0.77; P = .75, I² = 92%). Vesicoureteral reflux was evaluated in 4 studies (Figure 5), with all 4 finding VUR to be more common after ureteroneocystostomy (RR of 6.82; P = .007). A similar result was found for the 3 studies that evaluated urinary tract infection (Figure 6) (RR of 2.29; P = .004). Both VUR and urinary tract infection showed low heterogeneity.

Transplant surgeons have plenty of options to choose from when performing ureteral reimplant procedures. Most transplant surgeons prefer to use ureterneocystostomy for primary anastomosis, of which extravesical techniques have gained considerable favor. Although no consensus exists, recent studies have found that extravesical techniques, in conjunction with stenting, may reduce duration of the surgical procedure, complications, and patient morbidities compared with intravesical techniques.^10-12

Few centers use ureteroureterostomy as a primary anastomosis, with most reserving this procedure as a secondary option for repeat anastomosis for correction.^5,6,9 However, 2 of the retrospective studies reviewed here performed terminoterminal ureteroureteral anastomoses at the beginning of the study period only as a secondary option but then switched this procedure to the primary option. Both studies cited ease of performance and a perceived decrease in complication rates as the rationale.^5,9 Only the randomized control trial performed primary ureteroureterostomy for the complete study period.⁶

Overall urologic complication rates after extravesical reimplant procedures have been reported to range from 2% to 12%.^10,13-16 Ureteroureterostomy or pyeloureterostomy have displayed similar rates, ranging from 1.9% to 12.6%.^6,17,18 The results reviewed here fall within this range. Interestingly, none of the studies included in this meta-analysis found a statistically significant difference when comparing overall complication rates of ureteroureterostomy versus ureteroneocystostomy. This is consistent with the compiled results, which demonstrated no significant difference in overall complication rates. These 2 approaches have distinct advantages and disadvantages that are reflected when complications are divided into categories.

Ureteroureterostomy has the potential to cause ureteral stenosis at the anastomotic line, hematoma formation, and intraluminal blood clots, all leading to ureteral obstruction. Rates of obstruction for extravesical anastomoses range from 1% to 8%, which is similar to those found in the reviewed studies (0%-5.4%).^10,11 Only 1 of the reviewed studies demonstrated a significantly higher rate of obstruction with ureteroureterostomy than with ureteroneocystostomy.⁹ The most common cause of early obstruction was hematoma, whereas ureteral stenosis and stone formation were most often the causes of late obstruction. Overall, the incidence of obstruction was higher after ureteroureteral anastomosis. However, this technique may proffer the advantage of maintaining a natural ureteral orifice for future endourologic treatment. In treatment of stenosis, Nie and associates were able to perform balloon dilation in all patients who received a ureteroureterostomy, whereas nearly 70% of ureteroneocystostomy patients required a repeat anastomosis for correction.⁹ Faenza and associates found that significantly more patients required surgery for treatment of early stenosis (P = .008).⁵However, a primary ureteroneocystostomy reserves the possibility of using a ureteroureterostomy for repeat anastomosis.

Only 1 study reported hematuria as a complication and found no difference between the 2 procedures. Both anastomotic techniques fell within the range reported in the literature (0.3%-14%).^10,11 Analyses of urine leak and fistula formation complications displayed conflicting results. Nie and associates⁹ found that leakage was more common after ureteroneocystostomy, whereas Maricic and associates⁸ demonstrated a decrease after switching from ureteroureterostomy to ureteroneocystostomy. Ureteral necrosis secondary to ischemia is the most frequent cause of urine leakage, which can result from technical issues during both procurement and transplant. Recipient age, site of arterial anastomosis, and preexisting bladder conditions also have been cited as factors that predispose patients to urine leak.¹⁹ Nie and associates⁹ reported an association between urine leak and deceased donors and patients with diabetes. However, Maricic and associates⁸ did not differentiate complication by donor type or comorbidity. Overall, this meta-analysis found no association between procedure and the incidence of urine leak, with significant heterogeneity between studies.

The maintenance of a natural ureteral orifice is a distinct advantage in ureteroureteral anastomosis. As would be expected, the rate of VUR was more common after ureteroneocystostomy. In fact, none of the 3 studies reported VUR as a complication after ureteroureterostomy. In conjunction with this, ureteroureterostomy was associated with a lower incidence of urinary tract infection. The lower incidences of VUR, in conjunction with early catheter removal reported by several studies after ureteroureterostomy, were likely contributing factors.^5,6 This technique also maintains an orthotopic access point for future endourologic procedures, as mentioned above.

Our present study has several limitations. Only 1 of the 6 studies was a randomized control trial. Of the 5 retrospective reviews, 2 dictated the procedure by surgeon preference and/or patient selection, whereas the other 3 actually switched from 1 procedure to the other, which may have introduced selection bias and surgeon skill as confounding factors. There were also differences in how complications were reported. One study assessed VUR in all patients, whereas the other 2 only reported VUR in symptomatic patients, thus potentially underestimating the incidence. However, the complication rates found in these studies were consistent with published literature. In addition, although all studies placed stents after ureteroureterostomy, the use of stents after ureteroneocystostomy was variable. Despite this, it is not clear whether the variability in use of stents affected the primary endpoints. In the current climate of routine stent use after all ureteral anastomotic procedures, we would recommend that a future prospective trial with standardized stenting procedures be used to compare patients undergoing ureteroureterostomy with a homogeneous ureteroneocystostomy cohort.

Although none of the studies that we reviewed displayed a significant difference in overall complication rates, individual complications varied by procedure. Maintenance of an orthotopic ureteral orifice eliminated VUR as a complication after ureteroureterostomy, leading to a decrease in urinary tract infections. Urine leak, obstruction, and hematuria were found to be similar between the 2 procedures. Both ureteroureterostomy and ureteroneocystostomy appear to be good options for ureteral reimplant procedures during kidney transplant. With the differing individual complication rates and the scarcity of data comparing the 2 methods, no specific recommendation regarding which technique should be used can currently be elucidated. We believe that further prospective studies comparing ureteroneocystostomy and primary ureteroureterostomy may reveal which technique is superior regarding complication rates.

Volume: 14 Issue: 3 June 2016

FULL TEXT