Long-Term Complications of Conduit Urinary Diversion Mark S. Shimko,* Matthew K. Tollefson, Eric C. Umbreit, Sara A. Farmer, Michael L. Blute and Igor Frank From the Department of Urology (MSS, MKT, ECU, MLB, IF) and Health Sciences Research (SAF), Mayo Medical School and Mayo Clinic, Rochester, Minnesota Abbreviations and Acronyms ECOG Eastern Cooperative Oncology Group Submitted for publication June 11, 2010. Study received institutional review board approval. Nothing to disclose. Supplementary material for this article can be obtained at http://mayoresearch.mayo.edu/ diversion-complications/. * Correspondence: 200 First St. SW, Rochester, Minnesota 55905 (telephone: 507-284-3982; FAX: 507-284-4951; e-mail: shimko.mark@mayo. edu). Editor s Note: This article is the fifth of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions on pages 758 and 759. Purpose: We evaluated long-term surgical complications and clinical outcomes in a large group of patients treated with conduit urinary diversion. Materials and Methods: We identified 1,057 patients who underwent radical cystectomy with conduit urinary diversion using ileum or colon at our institution from 1980 to 1998 with complete followup information. Patients were followed for long-term clinical outcomes and analyzed for the incidence of diversion specific complications. Results: A total of 844 patients died at a median of 4.1 years (range 0.1 to 28.1) following cystectomy. Median followup of the surviving 213 patients was 15.5 years (range 0.3 to 29.1). There were 643 (60.8%) patients with 1,453 complications directly attributable to the urinary diversion performed with a mean of 2.3 complications per patient. Bowel complications were the most common, occurring in 215 patients (20.3%), followed by renal complications in 213 (20.2%), infectious complications in 174 (16.5%), stomal complications in 163 (15.4%) and urolithiasis in 162 (15.3%). The least common were metabolic abnormalities, which occurred in 135 patients (12.8%), and structural complications, which occurred in 122 (11.5%). Increasing age at cystectomy (HR 1.21, p 0.001), increasing Eastern Cooperative Oncology Group performance status (HR 1.23, p 0.02) and recent era of surgery (HR 1.68, p 0.001) were significantly associated with a higher incidence of complications. Conclusions: Conduit urinary diversion is associated with a high overall complication rate but a low reoperation rate. Long-term followup of these patients is necessary to closely monitor for potential complications from the urinary diversion that can occur decades later. Key Words: cystectomy, urinary diversion, postoperative complications BLADDER cancer is the 5th most common malignancy in the United States with an estimated 70,980 new diagnoses in 2009 and an estimated 14,300 deaths. 1 Up to a quarter of these patients have muscle invasion at diagnosis. 2 Although chemotherapy and radiation have begun to have a larger role in the treatment of this disease process, 3 5 the gold standard therapy for muscle invasive disease remains radical cystectomy. Removal of the bladder necessitates urinary reconstruction to preserve renal function. The 3 common types of urinary reconstruction are orthotopic bladder substitution, continent cutaneous diversion and conduit diversion. Orthotopic reconstruction and continent cutaneous diversion were introduced in an attempt to improve quality of life. However, the conduit urinary diversion described by Bricker 6 continues to be the most 562 www.jurology.com 0022-5347/11/1852-0562/0 Vol. 185, 562-567, February 2011 THE JOURNAL OF UROLOGY Printed in U.S.A. 2011 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC. DOI:10.1016/j.juro.2010.09.096
LONG-TERM COMPLICATIONS OF CONDUIT URINARY DIVERSION 563 common form of urinary reconstruction performed following cystectomy and occurs in up to 80% of radical cystectomy procedures in the United States. 7 The current literature has focused primarily on early complications after cystectomy due to high early mortality from the primary disease and from patient medical comorbidities. However, renewed attention has been directed toward long-term complications of the diversion considering that many patients undergoing cystectomy may have a high likelihood of durable survival. All forms of urinary reconstruction have been shown to have high complication rates and impact the postoperative quality of life. 7 16 Due to the high early mortality of bladder cancer and significant medical comorbidities found in this patient population, long-term outcomes of the procedure have been studied with relatively small patient cohorts. For example, the largest recently published series had only 18 patients with more than 15 years of followup. 8 Therefore, we evaluated the long-term complications specifically associated with the conduit urinary diversion, including bowel, renal, stomal, infectious, structural and metabolic complications, as well as urolithiasis that occurred a minimum of 30 days after surgery performed between 1980 and 1998. MATERIALS AND METHODS Our institutional cystectomy registry is comprised of pertinent clinical and pathological information for all patients who underwent a radical cystectomy at our institution starting in 1980. It is a composite of retrospective and prospective data collection that is populated with data directly from patient medical records and patient correspondence. After appropriate institutional review board approval this registry was queried for all patients who underwent cystectomy with conduit urinary reconstruction between 1980 and 1998. Patients with less than 30 days of followup were excluded, as were complications occurring within 30 days of surgery. Surgical Technique During the study period urinary reconstruction was performed by numerous surgeons, but a common technique was used for all conduits. A 20 cm segment of ileum (or rarely, colon) was isolated approximately 20 cm proximal to the ileocecal junction. The bowel was re-anastomosed primarily in a side-to-side fashion. The ureteroileal anastomosis was performed as originally described by Bricker in an end-to-side fashion. 6 Ureteral stents were typically placed for the more recent years of the study and were left in place for 5 to 14 days until the patient tolerated a general diet. The conduit segment was brought through the belly of the rectus muscle to the skin and a nipple stoma created. Followup Followup surveillance in patients after radical cystectomy during the period of this study generally consisted of clinical visits every 3 months for 2 years, every 6 months for 2 years and annually thereafter. Serum electrolytes were monitored during each visit and upper tracts were imaged every 3 to 6 months initially and then on an annual basis. Urine cultures were obtained when clinically relevant for systemic signs of infection such as febrile illness or malaise. For patients followed elsewhere pertinent information was obtained starting 14 months after the most recent followup by sequentially sending 2 written questionnaires to the patient and a letter to their local physician at 2-month intervals followed by telephone interviews to obtain the pertinent information. Statistical Analysis Complication rates at 5, 10, 15 and 20 years following cystectomy were estimated using the Kaplan-Meier method. Hazard ratios for increasing age at cystectomy, gender, era of surgery, ECOG performance status and perioperative chemotherapy were estimated using the Cox proportional hazards model. Statistical analyses were done using the SAS software package. RESULTS We identified 1,057 patients who underwent radical cystectomy with conduit urinary diversion between 1980 and 1998 at our institution. A segment of ileum was used for creation of the conduit in 1,045 patients (98.9%) and colon was used for the remaining 12 patients (1.1%). The majority of patients (842 of 1,057, 79.7%) were male and median age at surgery was 69 years old (range 31 to 92). Median followup after surgery for the entire cohort was 6.3 years (range 0.1 to 29.1). For those alive at last evaluation followup was 15.5 years (range 0.3 to 29.1). There were 97 patients who had at least 20 years of followup (fig. 1). Additional characteristics of the patient cohort are shown in table 1. Pathological findings at cystectomy for the cohort are shown in table 2. A total of 1,453 conduit related complications occurred in 643 patients (61%) with a mean of 2.3 Patients (number) 1200 1000 800 600 400 200 0 1057 574 409 215 0 5 10 15 20 Time from cystectomy (years) Figure 1. Study population and followup 97
564 LONG-TERM COMPLICATIONS OF CONDUIT URINARY DIVERSION Table 1. Patient demographics No. pos family history (%) 36 (3.4) No. tobacco use: Never 203 Current 329 Historical 525 Mean kg/m 2 body mass index 27.1 No. ECOG performance status (%):* 0 773 (74) 1 215 (20) 2 53 (5) 3 8 (1) 4 1 (0.1) 5 0 (0) * In 1,050 patients. complications per patient. These occurred at a median of 1.1 years (range 0.1 to 25.7) from surgery. The breakdown of complications according to age is shown in table 3. At a median of 1.7 years (range 0.1 to 15.8) 61 patients (5.8%) required reoperation for complications. By 20 years following surgery the incidence of any complication was 79.9% (fig. 2). Of 1,057 patients 527 (49.9%) experienced a complication within 5 years of surgery. Of the 276 patients who survived 5 years complication-free, 116 (42.0%) eventually experienced a complication. Increasing age at cystectomy was associated with a higher incidence of complications (HR 1.21, p 0.001), as was a higher ECOG performance status (HR 1.23, p 0.02) and the most recent era of cystectomy (HR 1.68, p 0.001). Gender and perioperative chemotherapy were not significantly associated with a higher rate of complications (table 4). Bowel Bowel related complications were reported in 215 patients (20.3%) at a median of 1.5 years (range 0.1 to 17.3). Bowel obstruction was the most common Table 3. Distribution of complications with age at cystectomy Pt Age No. With Complications (%) Younger than 50 24 (71) 50 59 98 (67) 60 69 235 (61) 70 79 242 (68) 80 44 (66) In 988 patients. complication occurring in 169 patients (16.0%) at a median of 1.7 years (range 0.1 to 17.3), of whom 12 (7.1%) required reoperation. An abscess related to the bowel anastomosis occurred in 38 patients (3.6%) at a median of 0.9 years (range 0.1 to 21.6), of whom 3 (7.9%) required open drainage with the remainder treated percutaneously. Enteric fistulas occurred in 29 patients (2.7%) at a median of 1.9 years (range 0.1 to 21.1), none of whom required reoperation. Renal Renal complications were reported in 213 patients (20.2%) at a median of 2.2 years (range 0.1 to 29.6). From the overall cohort 6.9% (73 of 1,057) of patients had preexisting renal failure defined as a creatinine greater than 2.0 mg/dl and new onset chronic renal failure developed in 19.0% (201 of 1,057) at a median of 2.3 years (range 0.1 to 29.6). There were 26 patients (2.5%) with progression to renal replacement therapy at a median of 8.4 years (range 0.9 to 23.5), and 22 (2.1%) had loss of a functional renal unit at a median of 2.4 years (range 0.2 to 23.5). Infectious Infectious complications occurred in 174 patients (16.5%) at a median of 1.8 years (range 0.1 to 25.7). Pyelonephritis occurred in 127 patients (12.0%) at a Table 2. 2002 American Joint Committee on cancer pathological staging of cohort No. Pts (%) Tumor:* Ta 31 (3) TIS 222 (21) T1 185 (18) T2 230 (22) T3 239 (23) T4 72 (7) Node: Nx 129 (12) N0 802 (76) N1 71 (7) N2/3 55 (5) Metastasis: Mx/M0 1,038 (98.2) M1 19 (2) * In 1,047 patients. Figure 2. Cumulative events curve of complications for each category and for all patients by year from surgery.
LONG-TERM COMPLICATIONS OF CONDUIT URINARY DIVERSION 565 Table 4. Impact of various factors on the incidence of complications Hazard Ratio (95% CI) p Value 10-yr increase in age at cystectomy 1.21 (1.10, 1.32) 0.001 Gender: F 1.0 (reference) M 0.92 (0.76, 1.11) 0.38 ECOG performance status:* 0 1.0 (reference) Greater than 0 1.23 (1.03, 1.47) 0.02 Yr of cystectomy: 1980 1984 1.0 (reference) 1985 1989 1.04 (0.84, 1.30) 0.71 1990 1994 1.24 (1.00, 1.52) 0.05 1995 1998 1.68 (1.35, 2.09) 0.001 Neoadjuvant/primary chemotherapy 1.09 (0.93, 1.28) 0.31 Adjuvant/salvage chemotherapy 0.81 (0.57, 1.16) 0.26 * In 985 patients. median of 2.3 years (range 0.1 to 25.7). Recurrent urinary tract infections occurred in 73 patients (6.9%) at a median of 2.1 years (range 0.1 to 21.6). Stomal Stoma related complications occurred in 163 patients (15.4%) at a median of 2.3 years (range 0.2 to 23.4). The most common stomal complication was a peristomal hernia which occurred in 147 patients (13.9%) at a median of 2.4 years (range 0.2 to 18.3). Of these patients 12 (8.2%) required repair or resiting of their stoma due to complications of the hernia. Stomal stenosis occurred in 22 patients (2.1%) at a median of 9.2 years (range 0.2 to 23.4). Two patients (9.1%) required revision of their stoma to correct the stenosis and the remainder were treated with intermittent catheterization or dilation. Stone Stone related complications occurred in 162 patients (15.3%) at a median of 2.5 years (range 0.1 to 24.9). Upper tract urolithiasis occurred in 141 patients (13.3%) at a median of 2.5 years (range 0.1 to 24.9). Conduit stones were found in 48 patients (4.5%) at a median of 3.0 years (range 0.2 to 22.9), of whom 27 (56.3%) had associated upper tract stones and 21 (43.8%) had conduit stones as the sole stone burden. Of the patients with stones 32 (19.8%) required conduit or upper tract endoscopy and lithotripsy. Metabolic A total of 135 patients (12.8%) experienced metabolic complications at a median of 1.9 years (range 0.1 to 25.9). Metabolic acidosis, defined as a serum bicarbonate level less than 20 mg/dl or requiring treatment with an alkalinizing agent at a median of 1.0 years (range 0.1 to 24.2), was found in 108 patients (10.2%). In addition, 32 (3.0%) patients had low vitamin B12 levels, which were normal before cystectomy, at a median of 9.1 years (range 0.4 to 25.8). Structural A total of 122 patients (11.5%) had structural complications at a median of 1.5 years (range 0.1 to 25.0), which were defined as structural complications of the conduit not including the stoma. Of these patients 106 (10.0%) had anastomotic strictures at a median of 1.1 years (range 0.1 to 25.0). Most of these patients were treated with antegrade stent placement but 15 (14.2%) required open ureteroileal anastomosis revision. A conduit stricture developed in 25 patients (2.4%) at a median of 9.4 years (range 0.2 to 24.1). DISCUSSION To our knowledge this is the largest series with long-term followup of complications following conduit urinary diversion. Other retrospective series have shown mixed results with regard to the superiority of 1 type of urinary reconstruction compared to another, 11 16 and the conduit diversion continues to be the most popular form of reconstruction due to multiple factors, including reduced operative time and familiarity to most urologists. Frequently a lower complication rate is given as justification to perform a conduit diversion but reports have failed to confirm a decreased complication rate with this procedure. 8 Our current analysis is unique because of the large number of patients with more than 2 decades of followup from an incontinent urinary diversion. The overall morbidity in our series was high (61%) and consistent with previously published reports. 7,8,12 16 At 20 years of followup the incidence of any complication was 79.9% (fig. 2). A significant portion of these complications was minor, and although there was a high overall complication rate (80% at 20 years), there was a low reoperation rate (6%). These results should be considered when discussing the various reconstructive options with patients before cystectomy. Every complication type had a wide range of time to first occurrence, indicating that these patients require close long-term surveillance even decades after cystectomy. Increasing age at cystectomy, increasing ECOG performance status and era of surgery were all associated with a significant increase in complication rates (table 4). The increase in complications in the most recent era may be attributable to more aggressive treatment of urothelial carcinoma as well as improved availability of health records of recent patients. Not all complications are significant in the decision of the type of urinary reconstruction following cystectomy. For example, bowel complications were the most common, occurring in 20% of the population. It is expected that this would be similar be-
566 LONG-TERM COMPLICATIONS OF CONDUIT URINARY DIVERSION tween various reconstructive options as they all use bowel for the reconstruction. Compared to series on orthotopic reconstruction and continent catheterizable pouches, patient selection has a large role. 7 For example, prolonged exposure of the bowel mucosa to urine has long been known to be toxic to those with preexisting renal or hepatic dysfunction. 17 In contrast, it has been suggested that an ileal neobladder may provide some renal protective function. 10,18 These considerations are important variables when counseling patients on the appropriate urinary reconstruction. In addition, patients with multiple medical comorbidities are often recommended for a conduit diversion as this is associated with decreased operative time. This makes unbiased comparisons between these reconstructions difficult and should be considered whenever comparing complications of different reconstructive options. 14,15,19 The high complication rate further suggests that the conduit is not the low risk, low complication procedure that it once appeared to be and potential complications do not necessarily support 1 type of urinary reconstruction over another. Many complications occurring decades after surgery are likely attributable to other causes but the conduit diversion was believed to be a contributing factor. For example, the development of metabolic acidosis 24 years after surgery may be secondary to other disease processes, but the impaired acid secreting ability of those with conduits due to reabsorption of urine electrolytes may contribute to the acidosis. In our study the complications occurring years later included vitamin B12 deficiency and strictures of the conduit. Vitamin B12 deficiency occurred at a median interval of greater than 9 years from surgery. This fits with the known physiology of B12 metabolism, as the human body stores should last approximately 5 years. 17 Earlier occurrences are likely due to preexisting nutritional deficiency. Conduit strictures, commonly related to adhesions, occurred at a median of 9 years from surgery as well. These 2 complications in particular occur late and underscore the need for vigorous long-term monitoring of these patients from a physiological and functional standpoint. Although the need for surveillance after radical cystectomy remains hotly debated, our study underscores that many of these complications occur long after a patient is considered cured of malignancy. For example, all complication types occurred decades after surgery. The clinical impact of these complications is uncertain as many of the reported complications were minor or asymptomatic and discovered during routine followup. It remains unclear whether routine screening for these complications is necessary or whether evaluation should be reserved for patients with symptoms. This is in addition to any oncologic monitoring that is necessary, although the same studies should monitor for both. Although this is the largest single institution study on long-term conduit complications following radical cystectomy to date to our knowledge, there are certain inherent limitations. This study is retrospective in nature and consequently, there are selection biases and variability in care. Inconsistent followup, changes in imaging availability and techniques during the study, and the lack of multicenter outcomes all limit the applicability of this study. The overall morbidity of the conduit is likely to be understated due to the high early mortality of this patient population due to the nature of urothelial carcinoma as well as the multiple comorbid conditions that are frequently present. The early median time to occurrence is more likely related to early mortality than a true timeline of when to expect these complications. Some patients lost to followup may have been followed locally without complication which led to an overestimation of the true complication rate, whereas some complications may have been handled locally and not reported. These potential biases may have led to an overestimation and underestimation, respectively, of the true complication rate. CONCLUSIONS Our long-term experience demonstrates that conduit urinary diversion is associated with high overall complication rates but a low reoperation rate. Long-term followup of these patients is necessary not only from an oncologic standpoint but also to closely monitor for potential complications from their urinary diversion, which can occur decades later. REFERENCES 1. Jemal A, Siegel R, Ward E et al: Cancer statistics, 2009. CA Cancer J Clin 2009; 59: 225. 2. Messing EM, Young TB, Hunt VB et al: Comparison of bladder cancer outcome in men undergoing hematuria home screening versus those with standard clinical presentations. Urology 1995; 45: 387. 3. Mak RH, Zietman AL, Heney NM et al: Bladder preservation: optimizing radiotherapy and integrated treatment strategies. BJU Int 2008; 102: 1345. 4. Milosevic M, Gospodarowicz M, Zietman A et al: Radiotherapy for bladder cancer. Urology 2007; 69: 80.
LONG-TERM COMPLICATIONS OF CONDUIT URINARY DIVERSION 567 5. Shipley WU, Kaufman DS, Zehr E et al: Selective bladder preservation by combined modality protocol treatment: long-term outcomes of 190 patients with invasive bladder cancer. Urology 2002; 60: 62. 6. Bricker EM: Bladder substitution after pelvic evisceration. Surg Clin North Am 1950; 30: 1511. 7. Gore JL, Saigal CS, Hanley JM et al: Variations in reconstruction after radical cystectomy. Cancer 2006; 107: 729. 8. Madersbacher S, Schmidt J, Eberle JM et al: Long-term outcome of ileal conduit diversion. J Urol 2003; 169: 985. 9. Tanaka T, Kitamura H, Takahashi A et al: Longterm functional outcome and late complications of Studer s ileal neobladder. Jpn J Clin Oncol 2005; 35: 391. 10. Studer UE, Burkhard FC, Schumacher M et al: Twenty years experience with an ileal orthotopic low pressure bladder substitute lessons to be learned. J Urol 2006; 176: 161. 11. Autorino R, Quarto G, Di Lorenzo G et al: Health related quality of life after radical cystectomy: comparison of ileal conduit to continent orthotopic neobladder. Eur J Surg Oncol 2009; 35: 858. 12. Frich PS, Kvestad CA and Angelsen A: Outcome and quality of life in patients operated on with radical cystectomy and three different urinary diversion techniques. Scand J Urol Nephrol 2009; 43: 37. 13. Gburek BM, Lieber MM and Blute ML: Comparison of Studer ileal neobladder and ileal conduit urinary diversion with respect to perioperative outcome and late complications. J Urol 1998; 160: 721. 14. Gore JL and Litwin MS: Quality of care in bladder cancer: trends in urinary diversion following radical cystectomy. World J Urol 2009; 27: 45. 15. Gore JL, Yu HY, Setodji C et al: Urinary diversion and morbidity after radical cystectomy for bladder cancer. Cancer 2010; 116: 331. 16. Somani BK, Gimlin D, Fayers P et al: Quality of life and body image for bladder cancer patients undergoing radical cystectomy and urinary diversion a prospective cohort study with a systematic review of literature. Urology 2009; 74: 1138. 17. McDougal WS: Metabolic complications of urinary intestinal diversion. J Urol 1992; 147: 1199. 18. Thoeny HC, Sonnenschein MJ, Madersbacher S et al: Is ileal orthotopic bladder substitution with an afferent tubular segment detrimental to the upper urinary tract in the long term? J Urol 2002; 168: 2030. 19. Froehner M, Brausi MA, Herr HW et al: Complications following radical cystectomy for bladder cancer in the elderly. Eur Urol 2009; 56: 443.