- Email: [email protected]
Lessons from 151 ureteral reimplantations for postcystectomy ureteroenteric strictures: A single-center experience over a decade
Urologic Oncology: Seminars and Original Investigations ] (2016) ∎∎∎–∎∎∎
Original article
Lessons from 151 ureteral reimplantations for postcystectomy ureteroenteric strictures: A single-center experience over a decade Vignesh T. Packiam, M.D.*, Vijay A. Agrawal, M.D., Andrew J. Cohen, M.D., Joseph J. Pariser, M.D., Scott C. Johnson, M.D., Gregory T. Bales, M.D., Norm D. Smith, M.D., Gary D. Steinberg, M.D. Section of Urology, University of Chicago, Chicago, IL Received 6 June 2016; received in revised form 27 September 2016; accepted 5 October 2016
Abstract Objectives: Ureteroenteric anastomotic strictures are common after cystectomy with urinary diversion. Endoscopic treatments have poor long-term success, although ureteral reimplantation is associated with morbidity. Predictors of successful open repair are poorly defined. Our objective was to characterize outcomes of ureteral reimplantation after cystectomy and identify risk factors for stricture recurrence. Patients and methods: We performed a retrospective review of 124 consecutive patients with a total of 151 open ureteral reimplantations for postcystectomy ureteroenteric strictures between January 2006 and December 2015. Baseline clinicopathologic characteristics and perioperative outcomes were examined. Predictors for stricture recurrence were assessed by univariable testing and univariate Cox proportional hazards regression. Results: Most patients underwent preoperative drainage by percutaneous nephrostomy (PCN; 43%) or percutaneous nephroureterostomy (PCNU; 44%). Major iatrogenic injuries included enterotomies requiring bowel anastomosis (3.2%) and major vascular injuries (2.4%). Overall, 60 (48%) patients suffered 90-day complications, of which 15 (12%) patients had high-grade complications. Median length of stay was 6 days [interquartile range: 5, 8] and median follow-up was 21 months [interquartile range: 5, 43]. The overall success rate per ureter was 93.4%. On univariate analysis, the only significant predictor of stricture recurrence was preoperative PCNU placement compared with PCN placement or no drainage (success rates: 85.5% vs. 98.9%, respectively, P ¼ 0.002). Cox proportional hazards regression demonstrated that preoperative PCNU placement yielded a hazard ratio of 10.2 (95% CI: 1.27–82.6) for stricture recurrence (P o 0.005). Stricture recurrence was independent of previous endoscopic interventions (P ¼ 0.42). Stricture length was unable to be assessed. Conclusions: Postcystectomy ureteral reimplantation was associated with relatively low rates of major iatrogenic injuries and high-grade complications. Preoperative PCN placement rather than PCNU may yield better results. r 2016 Elsevier Inc. All rights reserved.
Keywords: Cystectomy; Percutaneous nephrostomy; Stricture; Ureteroenteric stricture; Ureteral reimplantation
1. Introduction Radical cystectomy with urinary diversion is the gold standard treatment for muscle-invasive bladder cancer [1]. Benign ureteroenteric anastomotic strictures persist as common long-term complications despite refinements in surgical technique, consistently afflicting 3% to 10% of This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. * Corresponding author. Tel.: þ1-773-702-9757; fax: þ1-773-702-1001. E-mail address: [email protected] (V.T. Packiam). http://dx.doi.org/10.1016/j.urolonc.2016.10.005 1078-1439/r 2016 Elsevier Inc. All rights reserved.
patients in large cystectomy series [2–5]. The etiology of benign ureteroenteric strictures is most commonly attributed to ischemia, suspected from suboptimal distal ureteral neovascularization from enteric segments [6]. Other potential risk factors relate to poor healing and inflammation are as follows: smoking, radiation, urinary tract infection (UTI), and urine leak [4–6]. Endoscopic treatments are often attempted as first-line management despite low success rates [7]. These treatments yield even poorer outcomes for strictures that are left sided, 41 cm in length, and with poor ipsilateral renal function [8–10]. Open ureteral reimplantation yields
2
V.T. Packiam et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
excellent success (490%), although some surgeons avoid this surgery as first-line management owing to concerns of both iatrogenic and postoperative morbidity from reoperative laparotomy [9,11]. A clear understanding of the incidence of the most worrisome complications, including iatrogenic bowel and vascular injuries, is precluded by the small size of existing series. Predictors for stricture recurrence after ureteral reimplantation are also poorly understood. Although etiologies may be similar to those originally responsible for initial stricture formation, intervening percutaneous, and endoscopic treatments before ureteral reimplantation pose potentially avoidable risk factors. Performance of these procedures in addition to ureteral stenting can be associated with infection, inflammation, and fibrosis, compromising long-term outcomes in the relatively poorly vascularized ureteroenteric anastomosis [12,13]. We sought to rigorously characterize outcomes from our large series of 151 ureteral reimplantations and subsequently identify risk factors for stricture recurrence following ureteral reimplantation.
2. Patients and methods 2.1. Study population Institutional review board approval was obtained for this study. We retrospectively identified 124 consecutive patients who underwent open ureteral reimplantation between January 2006 and December 2015 for ureteroenteric strictures after radical cystectomy. We excluded reimplantations performed with concomitant conversion to different urinary diversion. Strictures were diagnosed by ultrasound or computed tomography imaging on routine follow-up or for pain, elevated creatinine, or UTI. When diagnosis of obstruction was equivocal, renal scintigraphy was obtained. No percutaneous decompression was performed for patients with stable creatinine and immediately scheduled reimplantation. Incidentally, visualized strictures in reimplanted contralateral ureters during open exploration were also not decompressed. Percutaneous nephrostomy (PCN) decompression was performed for all other strictures. Attempts at stent internalization and percutaneous nephroureterostomy (PCNU) placement were routinely made if anterograde nephrostogram showed passage of contrast through the stricture. Given their poor long-term success [8,11], percutaneous endoscopic treatments including stricture dilation and endoureterotomy were only attempted for patients with poor overall clinical status or for select rightsided strictures. After endoscopic treatment, nephroureteral stents were removed to assess for treatment success, and ureteral reimplantation was performed if obstruction recurred in patients with good overall clinical status. Biopsies were performed if recurrent malignancy was suspected from clinical history or imaging or both.
Our surgical technique for ureteral reimplantation in this setting has previously been published [8]. Briefly, a midline abdominal incision is made and the strictured ureteroenteric segment is identified and excised until healthy tissue is evident. Neobladders without chimney modifications involve incision of the neobladder with reimplantation of ureters into the posterior wall. Hautmann neobladders with chimney modifications afford easier repairs; the chimney is readily dissected and brought proximally, sometimes even mobilized under the sigmoid mesentery for long left strictures. For ileal conduit and Indiana pouch strictures, a midline laparotomy incision is made and the diversion is followed from the fascia to the level of the ureteroenteric stricture. These strictures are often associated with adhesions to areas of prior lymphadectomy and the iliac vessels. After excision of the stricture, the ureter is spatulated and reimplanted in an end-to-side Bricker fashion for ileal conduits, Indiana pouches, and the chimney of Hautmann neobladders. Running vs. interrupted anastomotic suturing technique was performed based on surgeon preference. Iatrogenic deserosalizations were oversewn and enterotomies were managed with excision of affected bowel and stapled anastomosis. A 6 Fr ureteral stent was left for 6 weeks postoperatively. A routine perioperative pathway was instituted and did not change over the course of the study. All patients underwent annual surveillance laboratories and imaging either locally or at our institution. 2.2. Assessment of data Operative reports and patient charts were reviewed to create this database immediately before this study. Specifically, baseline characteristics, perioperative outcomes, and stricture recurrence were assessed. Select major postcystectomy complications were examined based on factors that have been implicated with stricture formation. Preoperative urinary drainage categorization (none, PCN, and PCNU) was based on the type of drainage present in the operating room. Patients with prior PCNU converted to PCN were considered PCN, and vice versa. Attempted conversions of PCN to PCNU, replacements of PCNU with PCN, and endoscopic dilations, incisions, and biopsies were recorded. Complications were assessed rigorously as recommended [14] with additional stratifications based on timing (o30 days, 30–90 d), and Clavien-Dindo classification (low [1–2] and high [3–5]) [15]. Treatment failure or stricture recurrence, was defined as radiographic or clinical signs or both of recurrent obstruction requiring renal drainage or repeat ureteral reimplantation or both or nephrectomy. 2.3. Assessment of risk factors for stricture recurrence Owing to potentially differing conditions for each stricture in bilateral reimplantations (i.e., left PCN and right PCNU), success rates of repair were calculated per ureter during assessment of risk factors for stricture recurrence.
V.T. Packiam et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
All clinicopathologic and perioperative risk factors were examined using Fisher exact test for association with stricture recurrence. To identify clinically identifiable cutoffs as predictors for success, continuous variables were converted into categorical variables using the 50th percentile as a dichotomous cutoff. 2.4. Statistical analyses Statistical analysis was performed using Stata, version 13.0 (College Station, TX). Continuous variables were presented as median and interquartile range (IQR) and compared using the Mann-Whitney U test. Categorical variables were presented as whole numbers and percentages and compared using the Fisher exact test. Kaplan-Meier analysis and log-rank test were used to compare likelihood of treatment success by risk factors that significantly affected success on univariate analysis. Univariate Cox proportional hazards regression and additional univariate analysis was performed for significant variables. Two-sided P values were reported, and a P o 0.05 was considered statistically significant. 3. Results 3.1. Baseline characteristics of patients There were a total of 151 ureteroenteric strictures in 124 patients, as 27 patients had bilateral strictures. Overall, 12 (10%) patients had benign or pT0 pathology on original cystectomy, and most patients (69/124; 56%) had an orthotopic neobladder for urinary diversion (Table 1). The median interval from cystectomy to initial stricture diagnosis was 7 months [IQR: 3.5–21]. Most patients had strictures identified on imaging (74/124; 61%) and ultimately underwent drainage by PCN (53/124; 43%) or PCNU (55/124; 44%). Moreover, 7 patients (5.6%) underwent endoscopic stricture biopsy and 14 (11%) underwent endoscopic stricture dilation or endoureterotomy with Holmium laser. The median interval from initial stricture diagnosis to subsequent reimplantation was 52 days [IQR 30–121]. 3.2. Perioperative characteristics of patients Most intraoperative complications were deserosalizations (10/124; 8.1%), followed by enterotomies requiring bowel anastomosis (4/124; 3.2%), and significant vascular injuries (3/124; 2.4%) (Table 2). Median length of stay was 6 days [5, 8]. A total of 60 (48%) patients experienced postoperative complications within 90 days; 15 (12.1%) suffered a high-grade complication. Overall, 10 (8.1%) complications occurred between 30 and 90 days; 5 highgrade complications included percutaneous drains placed for abdominal abscess (2/124; 1.6%), direct visual internal
3
Table 1 Clinicopathologic characteristics of 124 patients with open ureteroenteric anastomotic revision Variable Age, y (median, IQR) BMI, kg/m2 (median, IQR) Sex, male (%)
65 [60–73] 28.8 [25.3–34.0] 96 (77%)
Race (%) White Black Other
92 (74%) 21 (17%) 11 (9%)
CCI (median, IQR) GFR, ml/min (median, IQR) Creatinine, mg/dl (median, IQR) Smoking history (%) History of radiation (%) Neoadjuvant chemotherapy (%) Robotic (%)
3 [3–4] 51 [37–61] 1.4 [1.1–1.7] 95 (77%) 6 (4.8%) 7 (5.6%) 7 (5.6%)
Diversion, n (%) Ileal conduit Indiana pouch Orthotopic neobladder
39 (31%) 16 (13%) 69 (56%)
Pathologic T stage (%) oT2 4T2
77 (62%) 47 (38%)
Variant histology (%) Positive nodes (%) Adjuvant chemotherapy (%)
16 (13%) 10 (8.1%) 23 (19%)
Select postcystectomy complications (%) Urine leak Fistula
15 (12%) 9 (7.3%)
Interval to stricture, mo (median, IQR) Early (o6 mo) Late (46 mo)
7 [3.5–21] 51 (41%) 73 (59%)
Presentation (%) Imaging Pain Renal failure UTI/Pyelonephritis
74 (60.7%) 5 (4.1%) 19 (15.6%) 24 (19.7%)
Preoperative urinary drainage (%) None PCN PCNUa
16 (12.9%) 53 (42.7%) 55 (44.4%)
Laterality (%) Left Right Bilateral
62 (50%) 35 (28%) 27 (22%)
BMI ¼ body mass index; CCI ¼ Charlson comorbidity index; GFR ¼ Glomerular filtration rate. a Includes 5 patients with bilateral strictures with PCNU on one side and PCN on the other.
urethrotomy for bladder neck contracture (2/124; 1.6%), and bilateral PCN for urine leak (1/124; 0.8%). In patients with small bowel obstruction, 13% (16/124) required nasogastric tube placement, whereas 7.3% (9/124) received total parenteral nutrition. No thromboembolic complications
V.T. Packiam et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
4
Table 2 Perioperative characteristics of 124 patients undergoing open ureteroenteric anastomotic revision Variable Anastomotic suturing technique Running Interrupted
Overall
112 (90.3%) 12 (9.7%)
Concomitant procedures (%) DVIU Bowel anastomosisa Hernia repair
7 (5.6%) 14 (11.3%) 14 (11.3%)
Intraoperative complications (%) Deserosalizationb Enterotomyb Vascular injuryc
10 (8.1%) 4 (3.2%) 3 (2.4%)
Operating room time, min (median, IQR) Estimated blood loss (median, IQR) Malignant pathology (%) Length of stay, d (median, IQR) Timing of complications (%)d o30 d 30–90 d
Low (1–2) High Clavien Clavien
3.3. Risk factors for stricture recurrence
162 [134–200] 187 [100–200] 4 (3.2%) 6 [5–8] 50 (40.3%) 10 (8.1%)
40 5
10 5
Individual 30-d complications (%)e Infectious 30 (24.2%) UTI/pyelonephritis 10 (8.1%) Clostridium difficile 8 (6.5%) Wound 12 (9.7%) Bowel 29 (23.4%) Ileus 9 (7.3%) Small bowel obstructionf 17 (13.7%) Bowel leak 3 (2.4%) Bleeding 3 (2.4%) Fascial dehiscence 3 (2.4%) Urine leak 4 (3.2%) Cardiac 3 (2.4%) Pulmonary 4 (3.2%)
29 10 8 11 25 9 16 0 3 0 1 3 1
1 0 0 1 4 0 1 3 0 3 3 0 3
30-d reoperation rate (%) 30-d readmission rate (%) 30-d mortality rate (%) Overall success rate (%) Left Right Bilateral Median time to failure, mo (median, IQR) Median follow-up, mo (median, IQR)
6 (4.8%) 12 (9.7%) 1 (0.8%) 116 59/62 32/35 25/27
(93.5%) (95.2%) (91.4%) (92.6%)
9 [3–24] 21 [5–43]
DVIU ¼ direct visual internal urethrotomy. A total of 7 entero-diversion fistula repairs, 3 entero-cutaneous fistula repairs, and 4 iatrogenic bowel injuries. b All deserosalizations were oversewn, all enterotomies were excised and reanastomosed. c All vascular injuries occurred in patients with neobladders. d Timing per patient rather than per complication. e Data presented for each individual complication. f In all, 16 required nasogastric tube placements and 9 required total parenteral nutrition. a
occurred. Further, 6 (4.8%) patients underwent reoperation within 30 days for bowel leak (3/124; 2.4%) or fascial (3/124; 2.4%) dehiscence and 12 (9.7%) patients were readmitted within 30 days. There was single 30-day mortality for a patient with multiple medical comorbidities including coronary artery disease, hypertension, and stage 4 chronic kidney disease. He was admitted on postoperative day 1 to the intensive care unit for hypotension, sepsis, and respiratory failure. He passed away on postoperative day 2; the underlying cause of death was ultimately unknown.
3.3.1. Univariate analysis of risk factors for stricture recurrence We assessed risk factors for stricture recurrence on a total of 151 reimplanted ureters. Table 3 shows both significant and the most plausible risk factors. There were a total of 10 (6.6%) recurrent strictures. The overall success rate per ureter was 93.4% (141/151). On Kaplan-Meier analysis, the estimated 5-year failure-free survival rate was 83% (95% CI: 68%–92%). Median follow-up was 21 months [5, 43]. Overall, 10 reimplantations (6.6%) were performed for recurrent strictures after open repair; none had 2 prior repairs. Median time to stricture recurrence per ureter was 9.5 months [IQR: 1–23]. There were no significant differences in success rates based on laterality, timing of presentation, or diversion type. Interestingly, UTI/ pyelonephritis (P ¼ 0.34), prior chemotherapy receipt (P ¼ 0.44), poor renal function (P ¼ 0.37), diabetes (P ¼ 0.64), symptomatic presentation (P ¼ 0.58) were not significant predictors for failure. Smoking history, history of prior radiation, and urine leak after ureteral reimplantation had weak association with success (all P ¼ 0.11). Success rates were significantly higher in ureters with preoperative PCN (62/62; 100%) or no drainage (26/27; 96.3%) compared with those with preoperative PCNU (53/62; 85.5%) (P ¼ 0.002). When ureters that underwent no prior drainage and PCN placement were grouped together and compared with those who underwent PCNU, difference in success rates remained significant (P ¼ 0.002). 3.3.2. Analysis of endoscopic procedures and specific percutaneous drainage factors in relation to stricture recurrence Stricture recurrence was independent of previous endoscopic stricture dilation or endoureterotomy (P ¼ 0.42). The only patient with recurrent stricture and history of endoscopic procedure had previously undergone 2 endoscopic dilations and subsequent PCNU placement. To further understand the effects of PCNU, we explored whether temporary ureteral stenting was also associated with stricture recurrence. There were no recurrent strictures in 8 ureters with prior PCNU exchanged for PCN before surgery. There were also no recurrent strictures in 28 ureters with previous unsuccessful PCNU placement attempts. Interestingly,
V.T. Packiam et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
5
Table 3 Success rates by risk factor for 151 strictures after open ureteroenteric anastomotic revision; univariable analysis Risk factor
No. success/total no. (% success)
Overall Laterality Left Right
141/151 (93.4%)
P value
84/89 (94.4%) 57/62 (91.9%)
0.74
Interval to stricture Early (o6 mo) Late (46 mo)
56/62 (90.3%) 85/89 (95.5%)
0.32
Diversion type Ileal conduit Indiana pouch Orthotopic neobladder
43/48 (89.6%) 20/20 (100%) 78/83 (94.0%)
0.37
34/34 (100%) 107/117 (91.5%)
0.11
Smoking history No Yes History of radiation No Yes
134/142 (94.4%) 7/9 (77.8%)
Preoperative urinary drainage None 26/27 (96.3%) PCN 62/62 (100%) PCNU 53/62 (85.5%)
0.11
0.002*
Anastomotic suturing technique Running 127/136 (93.4%) Interrupted 14/15 (93.3%)
1.00
Postreimplantation urine leak No 134/142 (94.4%) Yes 7/9 (77.8%)
0.11
Fig. Kaplan-Meier analysis of success rates per stricture by prior urinary drainage procedure (no drainage/PCN vs. PCNU). The number of patients at risk at 1, 3, and 5 years was 56, 32, and 10, respectively, for the no drainage/PCN group and 39, 22, and 14, respectively, for PCNU group. (Color version of figure is available online.)
between ureters with no drainage or PCN placement vs. those with PCNU (Fig; P ¼ 0.007). The estimated 5-year failure-free survival rate was 99% (95% CI: 92%–100%) for ureters with no drainage or PCN placement, compared with 83% (95% CI: 68%–92%) for ureters with PCNU placement. Cox proportional hazards regression revealed PCNU placement had a 10.2 hazard ratio for stricture recurrence (95% CI: 1.27–82.6, P o 0.005), compared with no drainage or PCN placement.
4. Discussion
Bolded values are significant. * P ¼ 0.002 when comparing PCNU vs. combined no drainage/PCN groups; P ¼ 0.003 when only comparing PCNU vs. PCN groups.
for the single patient with bilateral ureteral reimplantations and recurrent unilateral stricture, the recurrence developed on the right side that had a PCNU, whereas the left side with prior PCN remained patent. Finally, in ureters with PCNU, there was no association with stricture recurrence and median time interval between PCNU placement and ureteral reimplantation (103 [46, 216] days for no stricture recurrence vs. 103 [71, 138] days for stricture recurrence, P ¼ 0.77). We performed univariate analysis comparing baseline characteristics of ureters with and without preoperative PCNU (Supplementary Table). There were no significant differences between groups. Multivariate analysis was precluded by the limited number of stricture recurrences. 3.3.3. Kaplan-Meier analysis and univariate Cox proportional hazards regression assessing PCNU as risk factor for stricture recurrence Kaplan-Meier analysis and log-rank test confirmed a significant difference in stricture recurrence-free survival
To our knowledge, we assessed the largest series of ureteral reimplantation for postcystectomy ureteroenteric strictures. Our study has several key findings. First, overall success rate was high (93.4%). Although there is considerable perioperative morbidity, the incidence of serious iatrogenic injuries and high-grade postoperative complications was relatively low. Preoperative PCNU placement was associated with stricture recurrence compared with no drainage or PCN placement. Finally, prior endoscopic intervention did not affect long-term stricture recurrence rates after ureteral reimplantation. Management of ureteroenteric anastomotic strictures after radical cystectomy is challenging. Given the lack of guidelines, most specific recommendations are from expert opinion [9,11,16,17]. The poor efficacy of endoscopic procedures is well-documented [6–8], and there are concerns of the implications of failed endoscopic procedures, with single series showing subsequent increased blood loss and operative time during subsequent ureteral reimplantation [16]. Despite this, ureteral reimplantation causes many surgeons to hesitate when considering the morbidity of repeat laparotomy. Laparoscopic and robotic ureteral reimplantations have been reported, although these have been limited to only highly experienced centers to date [18].
V.T. Packiam et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
6
Table 4 Summary of series reporting on open revision of ureteroenteric anastomosis Study
No. of patients
Age (y)
Laterality (left/ right/ bilateral, %)
Ileal conduit diversion (%)
Cystectomy interval (mo)
LOS (d)
Overall major complicationsa (%)
Success rate, % (f/u mo)
Msezane et al. [19] Schöndorf et al. [9]
41 35
64 NR
54%/19%/27% 63%/37%/–
15% 26%
NR 10
NR NR
88% (20) 91% (29)
Nassar et al. [20] DiMarco et al. [11]
32 24
57 50
56%/10%/34% 46%/54%/–
22% 100%
13 66% o6 mo 34% 4 6 mo 7 100% 4 6 mo
NR 15
NR 8%
Laven et al. [16] Milhoua et al. [21] Tal et al [22] Shah et al. [17]
15 15c 14 12
61 66 NR NR
54%/31%/15% 43%/43%/14% NR NR
7% 43% NR NR
NR 6 7 NR
6 NR NR NR
27% 20% 7% NR
84% (47) 94 % (12) 94% (24) 86% (36)b 80 % (35) 80 % (18) 93 % (55) NR
LOS ¼ length of stay; NR ¼ not reported. Includes intraoperative and postoperative complications. b Rates for initial repair. c All results (except success rate) are reported for subset of 7 patients with initial primary open revision. a
All of these considerations increase the importance of clearly understanding the morbidity and complications arising from ureteral reimplantation. We captured all 90-day events per recent recommendations [14] and found that 48% of patients experienced complications in this timeframe. However, only 12% of patients suffered major (Clavien 3–5) complications. This expands on data from previous series that published nonspecific complication rates ranging from 7% to 27% [9,11,16,17,19–22]. The possibility of major iatrogenic injuries can deter surgeons from open anastomotic revision. In our cohort, we showed that rates of iatrogenic bowel injuries requiring anastomosis and major vascular injuries are low (3.2% and 2.4%, respectively), in experienced hands. Deserosalizations did not significantly increase overall or major complications in our series. Our success rate of 93.4% mirrored prior studies (Table 4) [9,11,19]. Prior studies have identified significant risk factors for initial stricture formation after cystectomy: patient (history of smoking or radiation), operative (running anastomotic sutures or not stenting postoperatively), and postoperative (UTI or urine leak) [4–6,12]. However, predictors of recurrent stricture after ureteral reimplantation are poorly characterized; the only previously identified predictor was stricture length 41 cm [11]. Stricture length was not reported on final pathology at our institution, and thus could not be evaluated. We found weak association with stricture recurrence and smoking history, prior radiation, and urine leak. Interestingly, although left-sided strictures have been shown to be at higher risk for stricture after initial urinary diversion and confer poorer outcomes after endoscopic interventions, laterality had minimal effect on success rates after open ureteral reimplantation. The dramatic influence of PCNU placement on stricture recurrence was a novel finding. Some oncologists and reconstructive surgeons have suspected that ureteral stenting
results in inflammation, infection, devascularization, and fibrosis that can compromise ureteral reimplantation [6]. This may be accentuated by the especially poor blood supply that is unique to the ureteroenteric anastomosis [12,13]. Anecdotally, we have found that identification of the exact transition point of diseased and healthy ureter is significantly more difficult in the setting of preoperative stenting as it reduces hydronephrosis and creates an inflammatory response along the ureter. In a small subset of patients, replacement of PCNU by PCN was not associated with any recurrent strictures. There was also no difference in duration of preoperative PCNU in patients with or without recurrent strictures. Although these findings may be secondary to small sample size, they have potential management implications. First, PCN placement should be preferred for patients with planned open repairs in the near future. Secondly, as the presence of PCNU at time of ureteral reimplantation seems to be a poor risk factor independent of duration, and replacement by PCN mitigates these risks, then replacement of PCNU with PCN a few weeks before surgery may be an effective strategy to balance patient comfort with durability of reimplantation. Endoscopic treatments are often initially selected to manage anastomotic strictures. Our institution has favored ureteral reimplantation based on poor endoscopic results shown by Laven et al. [16]. Although the overall rate of prior percutaneous endoscopic stricture incision or dilation procedures in our series was low (11%), we found minimal effect of failed endoscopic procedures on long-term success after ureteral reimplantation. Thus, endoscopic stricture incision or dilation appears reasonable for select patients with poor clinical status who do not have the risk factors for endoscopic failure: left-sided, 41 cm, poor ipsilateral renal function [16]. If these procedures fail, our results suggest these patients can still undergo effective surgical repair. In our experience, most patients were satisfied with ureteral
V.T. Packiam et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
reimplantation and willing to accept its associated potential morbidity owing to its high success rates and freedom from percutaneous drainage or ureteral stenting. There are several limitations to this study. It is retrospective and from a large tertiary referral center, which may limit its generalizability. Although it is the largest series of its kind, some analyses were still limited by sample size when assessing certain risk factors for stricture recurrence. Stricture length was unable to be assessed. Some cases of stricture recurrence may have been missed by surveillance imaging, as we avoided more invasive, but definitive assessments of obstruction such as anterograde pyelography [18]. The very low number of recurrent strictures after reimplantation precluded multivariate analysis for PCNU placement, which could have attempted to control for confounding factors.
[6]
[7]
[8]
[9]
[10]
[11]
5. Conclusions [12]
Ureteral reimplantation for postcystectomy strictures yielded a high success rate (93.4%) that was not influenced by prior endoscopic interventions. The rate of major iatrogenic injuries and high-grade complications were low. Stricture recurrence was significantly more likely in patients with preoperative PCNU placement. Therefore, our results suggest that endoscopic intervention is acceptable as initial management for select patients who are poor operative candidates, and PCNU should be avoided for patients with planned open repairs. Additional studies are needed to confirm these findings.
[13]
[14]
[15]
Appendix A. Supporting information
[16]
Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j. urolonc.2016.10.005.
[17]
References
[18]
[1] Stenzl A, Cowan NC, De Santis M, Kuczyk MA, Merseburger AS, Ribal MJ, et al. Treatment of muscle-invasive and metastatic bladder cancer: update of the EAU guidelines. Eur Urol 2011;59:1009–18, http://dx.doi.org/10.1016/j.eururo.2011.03.023. [2] Sullivan JW, Grabstald H, Whitmore WF. Complications of ureteroileal conduit with radical cystectomy: review of 336 cases. J Urol 1980;124:797–801. [3] Kouba E, Sands M, Lentz A, Wallen E, Pruthi RS. A comparison of the Bricker versus Wallace ureteroileal anastomosis in patients undergoing urinary diversion for bladder cancer. J Urol 2007; 178:945–8, http://dx.doi.org/10.1016/j.juro.2007.05.030. [4] Large MC, Cohn JA, Kiriluk KJ, Dangle P, Richards KA, Smith ND, et al. The impact of running versus interrupted anastomosis on ureterointestinal stricture rate after radical cystectomy. J Urol 2013;190:923–7, http://dx.doi.org/10.1016/j.juro.2013.02.091. [5] Anderson CB, Morgan TM, Kappa S, Moore D, Clark PE, Davis R, et al. Ureteroenteric anastomotic strictures after radical cystectomy
[19]
[20]
[21]
[22]
7
—does operative approach matter? J Urol 2013;189:541–7, http: //dx.doi.org/10.1016/j.juro.2012.09.034. Tyritzis SI, Wiklund NP. Ureteral strictures revisited…trying to see the light at the end of the tunnel: a comprehensive review. J Endourol 2015;29:124–36, http://dx.doi.org/10.1089/end.2014.0522. Kramolowsky EV, Clayman RV, Weyman PJ. Management of ureterointestinal anastomotic strictures: comparison of open surgical and endourological repair. J Urol 1988;139:1195–8. Laven BA, O’Connor RC, Steinberg GD, Gerber GS. Long-term results of antegrade endoureterotomy using the holmium laser in patients with ureterointestinal strictures. Urology 2001;58: 924–9. Schöndorf D, Meierhans-Ruf S, Kiss B, Giannarini G, Thalmann GN, Studer UE, et al. Ureteroileal strictures after urinary diversion with an ileal segment—is there a place for endourological treatment at all? J Urol 2013;190:585–90, http://dx.doi.org/10.1016/j.juro.2013.02.039. Wolf JS, Elashry OM, Clayman RV. Long-term results of endoureterotomy for benign ureteral and ureteroenteric strictures. J Urol 1997;158:759–64. DiMarco DS, LeRoy AJ, Thieling S, Bergstralh EJ, Segura JW. Longterm results of treatment for ureteroenteric strictures. Urology 2001;58:909–13. Mattei A, Birkhaeuser FD, Baermann C, Warncke SH, Studer UE. To stent or not to stent perioperatively the ureteroileal anastomosis of ileal orthotopic bladder substitutes and ileal conduits? results of a prospective randomized trial. J Urol 2008;179:582–6, http://dx.doi. org/10.1016/j.juro.2007.09.066. Keane PF, Bonner MC, Johnston SR, Zafar A, Gorman SP. Characterization of biofilm and encrustation on ureteric stents in vivo. Br J Urol 1994;73:687–91. Shabsigh A, Korets R, Vora KC, Brooks CM, Cronin AM, Savage C, et al. Defining early morbidity of radical cystectomy for patients with bladder cancer using a standardized reporting methodology. Eur Urol 2009;55:164–76, http://dx.doi.org/10.1016/j.eururo.2008.07.031. Dindo D, Demartines N, Clavien P-A. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240:205–13. Laven BA, O’Connor RC, Gerber GS, Steinberg GD. Long-term results of endoureterotomy and open surgical revision for the management of ureteroenteric strictures after urinary diversion. J Urol 2003;170: 1226–30, http://dx.doi.org/10.1097/01.ju.0000086701.68756.8f. Shah SH, Movassaghi K, Skinner D, Dalag L, Miranda G, Cai J, et al. Ureteroenteric strictures after open radical cystectomy and urinary diversion: The University of Southern California experience. Urology 2015;86:87–91, http://dx.doi.org/10.1016/j.urology.2015. 03.014. Rosales A, Emiliani E, Salvador JT, Peña JA, Gaya JM, Palou J, et al. Laparoscopic management of ureteroileal anastomosis strictures: initial experience. Eur Urol 2016http://dx.doi.org/10.1016/j.eururo. 2016.02.045. Msezane L, Reynolds WS, Mhapsekar R, Gerber G, Steinberg G. Open surgical repair of ureteral strictures and fistulas following radical cystectomy and urinary diversion. J Urol 2008;179:1428–31, http://dx.doi.org/10.1016/j.juro.2007.11.083. Nassar OAH, Alsafa MES. Experience with ureteroenteric strictures after radical cystectomy and diversion: open surgical revision. Urology 2011;78:459–65, http://dx.doi.org/10.1016/j.urology.2011.01.040. Milhoua PM, Miller NL, Cookson MS, Chang SS, Smith JA, Herrell SD. Primary endoscopic management versus open revision of ureteroenteric anastomotic strictures after urinary diversion—single institution contemporary series. J Endourol 2009;23:551–5. Tal R, Sivan B, Kedar D, Baniel J. Management of benign ureteral strictures following radical cystectomy and urinary diversion for bladder cancer. J Urol 2007;178:538–42, http://dx.doi.org/10.1016/j. juro.2007.03.142.
Original article
Lessons from 151 ureteral reimplantations for postcystectomy ureteroenteric strictures: A single-center experience over a decade Vignesh T. Packiam, M.D.*, Vijay A. Agrawal, M.D., Andrew J. Cohen, M.D., Joseph J. Pariser, M.D., Scott C. Johnson, M.D., Gregory T. Bales, M.D., Norm D. Smith, M.D., Gary D. Steinberg, M.D. Section of Urology, University of Chicago, Chicago, IL Received 6 June 2016; received in revised form 27 September 2016; accepted 5 October 2016
Abstract Objectives: Ureteroenteric anastomotic strictures are common after cystectomy with urinary diversion. Endoscopic treatments have poor long-term success, although ureteral reimplantation is associated with morbidity. Predictors of successful open repair are poorly defined. Our objective was to characterize outcomes of ureteral reimplantation after cystectomy and identify risk factors for stricture recurrence. Patients and methods: We performed a retrospective review of 124 consecutive patients with a total of 151 open ureteral reimplantations for postcystectomy ureteroenteric strictures between January 2006 and December 2015. Baseline clinicopathologic characteristics and perioperative outcomes were examined. Predictors for stricture recurrence were assessed by univariable testing and univariate Cox proportional hazards regression. Results: Most patients underwent preoperative drainage by percutaneous nephrostomy (PCN; 43%) or percutaneous nephroureterostomy (PCNU; 44%). Major iatrogenic injuries included enterotomies requiring bowel anastomosis (3.2%) and major vascular injuries (2.4%). Overall, 60 (48%) patients suffered 90-day complications, of which 15 (12%) patients had high-grade complications. Median length of stay was 6 days [interquartile range: 5, 8] and median follow-up was 21 months [interquartile range: 5, 43]. The overall success rate per ureter was 93.4%. On univariate analysis, the only significant predictor of stricture recurrence was preoperative PCNU placement compared with PCN placement or no drainage (success rates: 85.5% vs. 98.9%, respectively, P ¼ 0.002). Cox proportional hazards regression demonstrated that preoperative PCNU placement yielded a hazard ratio of 10.2 (95% CI: 1.27–82.6) for stricture recurrence (P o 0.005). Stricture recurrence was independent of previous endoscopic interventions (P ¼ 0.42). Stricture length was unable to be assessed. Conclusions: Postcystectomy ureteral reimplantation was associated with relatively low rates of major iatrogenic injuries and high-grade complications. Preoperative PCN placement rather than PCNU may yield better results. r 2016 Elsevier Inc. All rights reserved.
Keywords: Cystectomy; Percutaneous nephrostomy; Stricture; Ureteroenteric stricture; Ureteral reimplantation
1. Introduction Radical cystectomy with urinary diversion is the gold standard treatment for muscle-invasive bladder cancer [1]. Benign ureteroenteric anastomotic strictures persist as common long-term complications despite refinements in surgical technique, consistently afflicting 3% to 10% of This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. * Corresponding author. Tel.: þ1-773-702-9757; fax: þ1-773-702-1001. E-mail address: [email protected] (V.T. Packiam). http://dx.doi.org/10.1016/j.urolonc.2016.10.005 1078-1439/r 2016 Elsevier Inc. All rights reserved.
patients in large cystectomy series [2–5]. The etiology of benign ureteroenteric strictures is most commonly attributed to ischemia, suspected from suboptimal distal ureteral neovascularization from enteric segments [6]. Other potential risk factors relate to poor healing and inflammation are as follows: smoking, radiation, urinary tract infection (UTI), and urine leak [4–6]. Endoscopic treatments are often attempted as first-line management despite low success rates [7]. These treatments yield even poorer outcomes for strictures that are left sided, 41 cm in length, and with poor ipsilateral renal function [8–10]. Open ureteral reimplantation yields
2
V.T. Packiam et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
excellent success (490%), although some surgeons avoid this surgery as first-line management owing to concerns of both iatrogenic and postoperative morbidity from reoperative laparotomy [9,11]. A clear understanding of the incidence of the most worrisome complications, including iatrogenic bowel and vascular injuries, is precluded by the small size of existing series. Predictors for stricture recurrence after ureteral reimplantation are also poorly understood. Although etiologies may be similar to those originally responsible for initial stricture formation, intervening percutaneous, and endoscopic treatments before ureteral reimplantation pose potentially avoidable risk factors. Performance of these procedures in addition to ureteral stenting can be associated with infection, inflammation, and fibrosis, compromising long-term outcomes in the relatively poorly vascularized ureteroenteric anastomosis [12,13]. We sought to rigorously characterize outcomes from our large series of 151 ureteral reimplantations and subsequently identify risk factors for stricture recurrence following ureteral reimplantation.
2. Patients and methods 2.1. Study population Institutional review board approval was obtained for this study. We retrospectively identified 124 consecutive patients who underwent open ureteral reimplantation between January 2006 and December 2015 for ureteroenteric strictures after radical cystectomy. We excluded reimplantations performed with concomitant conversion to different urinary diversion. Strictures were diagnosed by ultrasound or computed tomography imaging on routine follow-up or for pain, elevated creatinine, or UTI. When diagnosis of obstruction was equivocal, renal scintigraphy was obtained. No percutaneous decompression was performed for patients with stable creatinine and immediately scheduled reimplantation. Incidentally, visualized strictures in reimplanted contralateral ureters during open exploration were also not decompressed. Percutaneous nephrostomy (PCN) decompression was performed for all other strictures. Attempts at stent internalization and percutaneous nephroureterostomy (PCNU) placement were routinely made if anterograde nephrostogram showed passage of contrast through the stricture. Given their poor long-term success [8,11], percutaneous endoscopic treatments including stricture dilation and endoureterotomy were only attempted for patients with poor overall clinical status or for select rightsided strictures. After endoscopic treatment, nephroureteral stents were removed to assess for treatment success, and ureteral reimplantation was performed if obstruction recurred in patients with good overall clinical status. Biopsies were performed if recurrent malignancy was suspected from clinical history or imaging or both.
Our surgical technique for ureteral reimplantation in this setting has previously been published [8]. Briefly, a midline abdominal incision is made and the strictured ureteroenteric segment is identified and excised until healthy tissue is evident. Neobladders without chimney modifications involve incision of the neobladder with reimplantation of ureters into the posterior wall. Hautmann neobladders with chimney modifications afford easier repairs; the chimney is readily dissected and brought proximally, sometimes even mobilized under the sigmoid mesentery for long left strictures. For ileal conduit and Indiana pouch strictures, a midline laparotomy incision is made and the diversion is followed from the fascia to the level of the ureteroenteric stricture. These strictures are often associated with adhesions to areas of prior lymphadectomy and the iliac vessels. After excision of the stricture, the ureter is spatulated and reimplanted in an end-to-side Bricker fashion for ileal conduits, Indiana pouches, and the chimney of Hautmann neobladders. Running vs. interrupted anastomotic suturing technique was performed based on surgeon preference. Iatrogenic deserosalizations were oversewn and enterotomies were managed with excision of affected bowel and stapled anastomosis. A 6 Fr ureteral stent was left for 6 weeks postoperatively. A routine perioperative pathway was instituted and did not change over the course of the study. All patients underwent annual surveillance laboratories and imaging either locally or at our institution. 2.2. Assessment of data Operative reports and patient charts were reviewed to create this database immediately before this study. Specifically, baseline characteristics, perioperative outcomes, and stricture recurrence were assessed. Select major postcystectomy complications were examined based on factors that have been implicated with stricture formation. Preoperative urinary drainage categorization (none, PCN, and PCNU) was based on the type of drainage present in the operating room. Patients with prior PCNU converted to PCN were considered PCN, and vice versa. Attempted conversions of PCN to PCNU, replacements of PCNU with PCN, and endoscopic dilations, incisions, and biopsies were recorded. Complications were assessed rigorously as recommended [14] with additional stratifications based on timing (o30 days, 30–90 d), and Clavien-Dindo classification (low [1–2] and high [3–5]) [15]. Treatment failure or stricture recurrence, was defined as radiographic or clinical signs or both of recurrent obstruction requiring renal drainage or repeat ureteral reimplantation or both or nephrectomy. 2.3. Assessment of risk factors for stricture recurrence Owing to potentially differing conditions for each stricture in bilateral reimplantations (i.e., left PCN and right PCNU), success rates of repair were calculated per ureter during assessment of risk factors for stricture recurrence.
V.T. Packiam et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
All clinicopathologic and perioperative risk factors were examined using Fisher exact test for association with stricture recurrence. To identify clinically identifiable cutoffs as predictors for success, continuous variables were converted into categorical variables using the 50th percentile as a dichotomous cutoff. 2.4. Statistical analyses Statistical analysis was performed using Stata, version 13.0 (College Station, TX). Continuous variables were presented as median and interquartile range (IQR) and compared using the Mann-Whitney U test. Categorical variables were presented as whole numbers and percentages and compared using the Fisher exact test. Kaplan-Meier analysis and log-rank test were used to compare likelihood of treatment success by risk factors that significantly affected success on univariate analysis. Univariate Cox proportional hazards regression and additional univariate analysis was performed for significant variables. Two-sided P values were reported, and a P o 0.05 was considered statistically significant. 3. Results 3.1. Baseline characteristics of patients There were a total of 151 ureteroenteric strictures in 124 patients, as 27 patients had bilateral strictures. Overall, 12 (10%) patients had benign or pT0 pathology on original cystectomy, and most patients (69/124; 56%) had an orthotopic neobladder for urinary diversion (Table 1). The median interval from cystectomy to initial stricture diagnosis was 7 months [IQR: 3.5–21]. Most patients had strictures identified on imaging (74/124; 61%) and ultimately underwent drainage by PCN (53/124; 43%) or PCNU (55/124; 44%). Moreover, 7 patients (5.6%) underwent endoscopic stricture biopsy and 14 (11%) underwent endoscopic stricture dilation or endoureterotomy with Holmium laser. The median interval from initial stricture diagnosis to subsequent reimplantation was 52 days [IQR 30–121]. 3.2. Perioperative characteristics of patients Most intraoperative complications were deserosalizations (10/124; 8.1%), followed by enterotomies requiring bowel anastomosis (4/124; 3.2%), and significant vascular injuries (3/124; 2.4%) (Table 2). Median length of stay was 6 days [5, 8]. A total of 60 (48%) patients experienced postoperative complications within 90 days; 15 (12.1%) suffered a high-grade complication. Overall, 10 (8.1%) complications occurred between 30 and 90 days; 5 highgrade complications included percutaneous drains placed for abdominal abscess (2/124; 1.6%), direct visual internal
3
Table 1 Clinicopathologic characteristics of 124 patients with open ureteroenteric anastomotic revision Variable Age, y (median, IQR) BMI, kg/m2 (median, IQR) Sex, male (%)
65 [60–73] 28.8 [25.3–34.0] 96 (77%)
Race (%) White Black Other
92 (74%) 21 (17%) 11 (9%)
CCI (median, IQR) GFR, ml/min (median, IQR) Creatinine, mg/dl (median, IQR) Smoking history (%) History of radiation (%) Neoadjuvant chemotherapy (%) Robotic (%)
3 [3–4] 51 [37–61] 1.4 [1.1–1.7] 95 (77%) 6 (4.8%) 7 (5.6%) 7 (5.6%)
Diversion, n (%) Ileal conduit Indiana pouch Orthotopic neobladder
39 (31%) 16 (13%) 69 (56%)
Pathologic T stage (%) oT2 4T2
77 (62%) 47 (38%)
Variant histology (%) Positive nodes (%) Adjuvant chemotherapy (%)
16 (13%) 10 (8.1%) 23 (19%)
Select postcystectomy complications (%) Urine leak Fistula
15 (12%) 9 (7.3%)
Interval to stricture, mo (median, IQR) Early (o6 mo) Late (46 mo)
7 [3.5–21] 51 (41%) 73 (59%)
Presentation (%) Imaging Pain Renal failure UTI/Pyelonephritis
74 (60.7%) 5 (4.1%) 19 (15.6%) 24 (19.7%)
Preoperative urinary drainage (%) None PCN PCNUa
16 (12.9%) 53 (42.7%) 55 (44.4%)
Laterality (%) Left Right Bilateral
62 (50%) 35 (28%) 27 (22%)
BMI ¼ body mass index; CCI ¼ Charlson comorbidity index; GFR ¼ Glomerular filtration rate. a Includes 5 patients with bilateral strictures with PCNU on one side and PCN on the other.
urethrotomy for bladder neck contracture (2/124; 1.6%), and bilateral PCN for urine leak (1/124; 0.8%). In patients with small bowel obstruction, 13% (16/124) required nasogastric tube placement, whereas 7.3% (9/124) received total parenteral nutrition. No thromboembolic complications
V.T. Packiam et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
4
Table 2 Perioperative characteristics of 124 patients undergoing open ureteroenteric anastomotic revision Variable Anastomotic suturing technique Running Interrupted
Overall
112 (90.3%) 12 (9.7%)
Concomitant procedures (%) DVIU Bowel anastomosisa Hernia repair
7 (5.6%) 14 (11.3%) 14 (11.3%)
Intraoperative complications (%) Deserosalizationb Enterotomyb Vascular injuryc
10 (8.1%) 4 (3.2%) 3 (2.4%)
Operating room time, min (median, IQR) Estimated blood loss (median, IQR) Malignant pathology (%) Length of stay, d (median, IQR) Timing of complications (%)d o30 d 30–90 d
Low (1–2) High Clavien Clavien
3.3. Risk factors for stricture recurrence
162 [134–200] 187 [100–200] 4 (3.2%) 6 [5–8] 50 (40.3%) 10 (8.1%)
40 5
10 5
Individual 30-d complications (%)e Infectious 30 (24.2%) UTI/pyelonephritis 10 (8.1%) Clostridium difficile 8 (6.5%) Wound 12 (9.7%) Bowel 29 (23.4%) Ileus 9 (7.3%) Small bowel obstructionf 17 (13.7%) Bowel leak 3 (2.4%) Bleeding 3 (2.4%) Fascial dehiscence 3 (2.4%) Urine leak 4 (3.2%) Cardiac 3 (2.4%) Pulmonary 4 (3.2%)
29 10 8 11 25 9 16 0 3 0 1 3 1
1 0 0 1 4 0 1 3 0 3 3 0 3
30-d reoperation rate (%) 30-d readmission rate (%) 30-d mortality rate (%) Overall success rate (%) Left Right Bilateral Median time to failure, mo (median, IQR) Median follow-up, mo (median, IQR)
6 (4.8%) 12 (9.7%) 1 (0.8%) 116 59/62 32/35 25/27
(93.5%) (95.2%) (91.4%) (92.6%)
9 [3–24] 21 [5–43]
DVIU ¼ direct visual internal urethrotomy. A total of 7 entero-diversion fistula repairs, 3 entero-cutaneous fistula repairs, and 4 iatrogenic bowel injuries. b All deserosalizations were oversewn, all enterotomies were excised and reanastomosed. c All vascular injuries occurred in patients with neobladders. d Timing per patient rather than per complication. e Data presented for each individual complication. f In all, 16 required nasogastric tube placements and 9 required total parenteral nutrition. a
occurred. Further, 6 (4.8%) patients underwent reoperation within 30 days for bowel leak (3/124; 2.4%) or fascial (3/124; 2.4%) dehiscence and 12 (9.7%) patients were readmitted within 30 days. There was single 30-day mortality for a patient with multiple medical comorbidities including coronary artery disease, hypertension, and stage 4 chronic kidney disease. He was admitted on postoperative day 1 to the intensive care unit for hypotension, sepsis, and respiratory failure. He passed away on postoperative day 2; the underlying cause of death was ultimately unknown.
3.3.1. Univariate analysis of risk factors for stricture recurrence We assessed risk factors for stricture recurrence on a total of 151 reimplanted ureters. Table 3 shows both significant and the most plausible risk factors. There were a total of 10 (6.6%) recurrent strictures. The overall success rate per ureter was 93.4% (141/151). On Kaplan-Meier analysis, the estimated 5-year failure-free survival rate was 83% (95% CI: 68%–92%). Median follow-up was 21 months [5, 43]. Overall, 10 reimplantations (6.6%) were performed for recurrent strictures after open repair; none had 2 prior repairs. Median time to stricture recurrence per ureter was 9.5 months [IQR: 1–23]. There were no significant differences in success rates based on laterality, timing of presentation, or diversion type. Interestingly, UTI/ pyelonephritis (P ¼ 0.34), prior chemotherapy receipt (P ¼ 0.44), poor renal function (P ¼ 0.37), diabetes (P ¼ 0.64), symptomatic presentation (P ¼ 0.58) were not significant predictors for failure. Smoking history, history of prior radiation, and urine leak after ureteral reimplantation had weak association with success (all P ¼ 0.11). Success rates were significantly higher in ureters with preoperative PCN (62/62; 100%) or no drainage (26/27; 96.3%) compared with those with preoperative PCNU (53/62; 85.5%) (P ¼ 0.002). When ureters that underwent no prior drainage and PCN placement were grouped together and compared with those who underwent PCNU, difference in success rates remained significant (P ¼ 0.002). 3.3.2. Analysis of endoscopic procedures and specific percutaneous drainage factors in relation to stricture recurrence Stricture recurrence was independent of previous endoscopic stricture dilation or endoureterotomy (P ¼ 0.42). The only patient with recurrent stricture and history of endoscopic procedure had previously undergone 2 endoscopic dilations and subsequent PCNU placement. To further understand the effects of PCNU, we explored whether temporary ureteral stenting was also associated with stricture recurrence. There were no recurrent strictures in 8 ureters with prior PCNU exchanged for PCN before surgery. There were also no recurrent strictures in 28 ureters with previous unsuccessful PCNU placement attempts. Interestingly,
V.T. Packiam et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
5
Table 3 Success rates by risk factor for 151 strictures after open ureteroenteric anastomotic revision; univariable analysis Risk factor
No. success/total no. (% success)
Overall Laterality Left Right
141/151 (93.4%)
P value
84/89 (94.4%) 57/62 (91.9%)
0.74
Interval to stricture Early (o6 mo) Late (46 mo)
56/62 (90.3%) 85/89 (95.5%)
0.32
Diversion type Ileal conduit Indiana pouch Orthotopic neobladder
43/48 (89.6%) 20/20 (100%) 78/83 (94.0%)
0.37
34/34 (100%) 107/117 (91.5%)
0.11
Smoking history No Yes History of radiation No Yes
134/142 (94.4%) 7/9 (77.8%)
Preoperative urinary drainage None 26/27 (96.3%) PCN 62/62 (100%) PCNU 53/62 (85.5%)
0.11
0.002*
Anastomotic suturing technique Running 127/136 (93.4%) Interrupted 14/15 (93.3%)
1.00
Postreimplantation urine leak No 134/142 (94.4%) Yes 7/9 (77.8%)
0.11
Fig. Kaplan-Meier analysis of success rates per stricture by prior urinary drainage procedure (no drainage/PCN vs. PCNU). The number of patients at risk at 1, 3, and 5 years was 56, 32, and 10, respectively, for the no drainage/PCN group and 39, 22, and 14, respectively, for PCNU group. (Color version of figure is available online.)
between ureters with no drainage or PCN placement vs. those with PCNU (Fig; P ¼ 0.007). The estimated 5-year failure-free survival rate was 99% (95% CI: 92%–100%) for ureters with no drainage or PCN placement, compared with 83% (95% CI: 68%–92%) for ureters with PCNU placement. Cox proportional hazards regression revealed PCNU placement had a 10.2 hazard ratio for stricture recurrence (95% CI: 1.27–82.6, P o 0.005), compared with no drainage or PCN placement.
4. Discussion
Bolded values are significant. * P ¼ 0.002 when comparing PCNU vs. combined no drainage/PCN groups; P ¼ 0.003 when only comparing PCNU vs. PCN groups.
for the single patient with bilateral ureteral reimplantations and recurrent unilateral stricture, the recurrence developed on the right side that had a PCNU, whereas the left side with prior PCN remained patent. Finally, in ureters with PCNU, there was no association with stricture recurrence and median time interval between PCNU placement and ureteral reimplantation (103 [46, 216] days for no stricture recurrence vs. 103 [71, 138] days for stricture recurrence, P ¼ 0.77). We performed univariate analysis comparing baseline characteristics of ureters with and without preoperative PCNU (Supplementary Table). There were no significant differences between groups. Multivariate analysis was precluded by the limited number of stricture recurrences. 3.3.3. Kaplan-Meier analysis and univariate Cox proportional hazards regression assessing PCNU as risk factor for stricture recurrence Kaplan-Meier analysis and log-rank test confirmed a significant difference in stricture recurrence-free survival
To our knowledge, we assessed the largest series of ureteral reimplantation for postcystectomy ureteroenteric strictures. Our study has several key findings. First, overall success rate was high (93.4%). Although there is considerable perioperative morbidity, the incidence of serious iatrogenic injuries and high-grade postoperative complications was relatively low. Preoperative PCNU placement was associated with stricture recurrence compared with no drainage or PCN placement. Finally, prior endoscopic intervention did not affect long-term stricture recurrence rates after ureteral reimplantation. Management of ureteroenteric anastomotic strictures after radical cystectomy is challenging. Given the lack of guidelines, most specific recommendations are from expert opinion [9,11,16,17]. The poor efficacy of endoscopic procedures is well-documented [6–8], and there are concerns of the implications of failed endoscopic procedures, with single series showing subsequent increased blood loss and operative time during subsequent ureteral reimplantation [16]. Despite this, ureteral reimplantation causes many surgeons to hesitate when considering the morbidity of repeat laparotomy. Laparoscopic and robotic ureteral reimplantations have been reported, although these have been limited to only highly experienced centers to date [18].
V.T. Packiam et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
6
Table 4 Summary of series reporting on open revision of ureteroenteric anastomosis Study
No. of patients
Age (y)
Laterality (left/ right/ bilateral, %)
Ileal conduit diversion (%)
Cystectomy interval (mo)
LOS (d)
Overall major complicationsa (%)
Success rate, % (f/u mo)
Msezane et al. [19] Schöndorf et al. [9]
41 35
64 NR
54%/19%/27% 63%/37%/–
15% 26%
NR 10
NR NR
88% (20) 91% (29)
Nassar et al. [20] DiMarco et al. [11]
32 24
57 50
56%/10%/34% 46%/54%/–
22% 100%
13 66% o6 mo 34% 4 6 mo 7 100% 4 6 mo
NR 15
NR 8%
Laven et al. [16] Milhoua et al. [21] Tal et al [22] Shah et al. [17]
15 15c 14 12
61 66 NR NR
54%/31%/15% 43%/43%/14% NR NR
7% 43% NR NR
NR 6 7 NR
6 NR NR NR
27% 20% 7% NR
84% (47) 94 % (12) 94% (24) 86% (36)b 80 % (35) 80 % (18) 93 % (55) NR
LOS ¼ length of stay; NR ¼ not reported. Includes intraoperative and postoperative complications. b Rates for initial repair. c All results (except success rate) are reported for subset of 7 patients with initial primary open revision. a
All of these considerations increase the importance of clearly understanding the morbidity and complications arising from ureteral reimplantation. We captured all 90-day events per recent recommendations [14] and found that 48% of patients experienced complications in this timeframe. However, only 12% of patients suffered major (Clavien 3–5) complications. This expands on data from previous series that published nonspecific complication rates ranging from 7% to 27% [9,11,16,17,19–22]. The possibility of major iatrogenic injuries can deter surgeons from open anastomotic revision. In our cohort, we showed that rates of iatrogenic bowel injuries requiring anastomosis and major vascular injuries are low (3.2% and 2.4%, respectively), in experienced hands. Deserosalizations did not significantly increase overall or major complications in our series. Our success rate of 93.4% mirrored prior studies (Table 4) [9,11,19]. Prior studies have identified significant risk factors for initial stricture formation after cystectomy: patient (history of smoking or radiation), operative (running anastomotic sutures or not stenting postoperatively), and postoperative (UTI or urine leak) [4–6,12]. However, predictors of recurrent stricture after ureteral reimplantation are poorly characterized; the only previously identified predictor was stricture length 41 cm [11]. Stricture length was not reported on final pathology at our institution, and thus could not be evaluated. We found weak association with stricture recurrence and smoking history, prior radiation, and urine leak. Interestingly, although left-sided strictures have been shown to be at higher risk for stricture after initial urinary diversion and confer poorer outcomes after endoscopic interventions, laterality had minimal effect on success rates after open ureteral reimplantation. The dramatic influence of PCNU placement on stricture recurrence was a novel finding. Some oncologists and reconstructive surgeons have suspected that ureteral stenting
results in inflammation, infection, devascularization, and fibrosis that can compromise ureteral reimplantation [6]. This may be accentuated by the especially poor blood supply that is unique to the ureteroenteric anastomosis [12,13]. Anecdotally, we have found that identification of the exact transition point of diseased and healthy ureter is significantly more difficult in the setting of preoperative stenting as it reduces hydronephrosis and creates an inflammatory response along the ureter. In a small subset of patients, replacement of PCNU by PCN was not associated with any recurrent strictures. There was also no difference in duration of preoperative PCNU in patients with or without recurrent strictures. Although these findings may be secondary to small sample size, they have potential management implications. First, PCN placement should be preferred for patients with planned open repairs in the near future. Secondly, as the presence of PCNU at time of ureteral reimplantation seems to be a poor risk factor independent of duration, and replacement by PCN mitigates these risks, then replacement of PCNU with PCN a few weeks before surgery may be an effective strategy to balance patient comfort with durability of reimplantation. Endoscopic treatments are often initially selected to manage anastomotic strictures. Our institution has favored ureteral reimplantation based on poor endoscopic results shown by Laven et al. [16]. Although the overall rate of prior percutaneous endoscopic stricture incision or dilation procedures in our series was low (11%), we found minimal effect of failed endoscopic procedures on long-term success after ureteral reimplantation. Thus, endoscopic stricture incision or dilation appears reasonable for select patients with poor clinical status who do not have the risk factors for endoscopic failure: left-sided, 41 cm, poor ipsilateral renal function [16]. If these procedures fail, our results suggest these patients can still undergo effective surgical repair. In our experience, most patients were satisfied with ureteral
V.T. Packiam et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
reimplantation and willing to accept its associated potential morbidity owing to its high success rates and freedom from percutaneous drainage or ureteral stenting. There are several limitations to this study. It is retrospective and from a large tertiary referral center, which may limit its generalizability. Although it is the largest series of its kind, some analyses were still limited by sample size when assessing certain risk factors for stricture recurrence. Stricture length was unable to be assessed. Some cases of stricture recurrence may have been missed by surveillance imaging, as we avoided more invasive, but definitive assessments of obstruction such as anterograde pyelography [18]. The very low number of recurrent strictures after reimplantation precluded multivariate analysis for PCNU placement, which could have attempted to control for confounding factors.
[6]
[7]
[8]
[9]
[10]
[11]
5. Conclusions [12]
Ureteral reimplantation for postcystectomy strictures yielded a high success rate (93.4%) that was not influenced by prior endoscopic interventions. The rate of major iatrogenic injuries and high-grade complications were low. Stricture recurrence was significantly more likely in patients with preoperative PCNU placement. Therefore, our results suggest that endoscopic intervention is acceptable as initial management for select patients who are poor operative candidates, and PCNU should be avoided for patients with planned open repairs. Additional studies are needed to confirm these findings.
[13]
[14]
[15]
Appendix A. Supporting information
[16]
Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j. urolonc.2016.10.005.
[17]
References
[18]
[1] Stenzl A, Cowan NC, De Santis M, Kuczyk MA, Merseburger AS, Ribal MJ, et al. Treatment of muscle-invasive and metastatic bladder cancer: update of the EAU guidelines. Eur Urol 2011;59:1009–18, http://dx.doi.org/10.1016/j.eururo.2011.03.023. [2] Sullivan JW, Grabstald H, Whitmore WF. Complications of ureteroileal conduit with radical cystectomy: review of 336 cases. J Urol 1980;124:797–801. [3] Kouba E, Sands M, Lentz A, Wallen E, Pruthi RS. A comparison of the Bricker versus Wallace ureteroileal anastomosis in patients undergoing urinary diversion for bladder cancer. J Urol 2007; 178:945–8, http://dx.doi.org/10.1016/j.juro.2007.05.030. [4] Large MC, Cohn JA, Kiriluk KJ, Dangle P, Richards KA, Smith ND, et al. The impact of running versus interrupted anastomosis on ureterointestinal stricture rate after radical cystectomy. J Urol 2013;190:923–7, http://dx.doi.org/10.1016/j.juro.2013.02.091. [5] Anderson CB, Morgan TM, Kappa S, Moore D, Clark PE, Davis R, et al. Ureteroenteric anastomotic strictures after radical cystectomy
[19]
[20]
[21]
[22]
7
—does operative approach matter? J Urol 2013;189:541–7, http: //dx.doi.org/10.1016/j.juro.2012.09.034. Tyritzis SI, Wiklund NP. Ureteral strictures revisited…trying to see the light at the end of the tunnel: a comprehensive review. J Endourol 2015;29:124–36, http://dx.doi.org/10.1089/end.2014.0522. Kramolowsky EV, Clayman RV, Weyman PJ. Management of ureterointestinal anastomotic strictures: comparison of open surgical and endourological repair. J Urol 1988;139:1195–8. Laven BA, O’Connor RC, Steinberg GD, Gerber GS. Long-term results of antegrade endoureterotomy using the holmium laser in patients with ureterointestinal strictures. Urology 2001;58: 924–9. Schöndorf D, Meierhans-Ruf S, Kiss B, Giannarini G, Thalmann GN, Studer UE, et al. Ureteroileal strictures after urinary diversion with an ileal segment—is there a place for endourological treatment at all? J Urol 2013;190:585–90, http://dx.doi.org/10.1016/j.juro.2013.02.039. Wolf JS, Elashry OM, Clayman RV. Long-term results of endoureterotomy for benign ureteral and ureteroenteric strictures. J Urol 1997;158:759–64. DiMarco DS, LeRoy AJ, Thieling S, Bergstralh EJ, Segura JW. Longterm results of treatment for ureteroenteric strictures. Urology 2001;58:909–13. Mattei A, Birkhaeuser FD, Baermann C, Warncke SH, Studer UE. To stent or not to stent perioperatively the ureteroileal anastomosis of ileal orthotopic bladder substitutes and ileal conduits? results of a prospective randomized trial. J Urol 2008;179:582–6, http://dx.doi. org/10.1016/j.juro.2007.09.066. Keane PF, Bonner MC, Johnston SR, Zafar A, Gorman SP. Characterization of biofilm and encrustation on ureteric stents in vivo. Br J Urol 1994;73:687–91. Shabsigh A, Korets R, Vora KC, Brooks CM, Cronin AM, Savage C, et al. Defining early morbidity of radical cystectomy for patients with bladder cancer using a standardized reporting methodology. Eur Urol 2009;55:164–76, http://dx.doi.org/10.1016/j.eururo.2008.07.031. Dindo D, Demartines N, Clavien P-A. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240:205–13. Laven BA, O’Connor RC, Gerber GS, Steinberg GD. Long-term results of endoureterotomy and open surgical revision for the management of ureteroenteric strictures after urinary diversion. J Urol 2003;170: 1226–30, http://dx.doi.org/10.1097/01.ju.0000086701.68756.8f. Shah SH, Movassaghi K, Skinner D, Dalag L, Miranda G, Cai J, et al. Ureteroenteric strictures after open radical cystectomy and urinary diversion: The University of Southern California experience. Urology 2015;86:87–91, http://dx.doi.org/10.1016/j.urology.2015. 03.014. Rosales A, Emiliani E, Salvador JT, Peña JA, Gaya JM, Palou J, et al. Laparoscopic management of ureteroileal anastomosis strictures: initial experience. Eur Urol 2016http://dx.doi.org/10.1016/j.eururo. 2016.02.045. Msezane L, Reynolds WS, Mhapsekar R, Gerber G, Steinberg G. Open surgical repair of ureteral strictures and fistulas following radical cystectomy and urinary diversion. J Urol 2008;179:1428–31, http://dx.doi.org/10.1016/j.juro.2007.11.083. Nassar OAH, Alsafa MES. Experience with ureteroenteric strictures after radical cystectomy and diversion: open surgical revision. Urology 2011;78:459–65, http://dx.doi.org/10.1016/j.urology.2011.01.040. Milhoua PM, Miller NL, Cookson MS, Chang SS, Smith JA, Herrell SD. Primary endoscopic management versus open revision of ureteroenteric anastomotic strictures after urinary diversion—single institution contemporary series. J Endourol 2009;23:551–5. Tal R, Sivan B, Kedar D, Baniel J. Management of benign ureteral strictures following radical cystectomy and urinary diversion for bladder cancer. J Urol 2007;178:538–42, http://dx.doi.org/10.1016/j. juro.2007.03.142.