- Email: [email protected]
Short-term outcomes after incontinent conduit for gynecologic cancer: Comparison of ileal, sigmoid, and transverse colon
YGYNO-975457; No. of pages: 5; 4C: Gynecologic Oncology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Short-term outcomes after incontinent conduit for gynecologic cancer: Comparison of ileal, sigmoid, and transverse colon Zaid M. Tabbaa a,1, Jo Marie T. Janco a,1, Andrea Mariani a, Sean C. Dowdy a, Michaela E. McGree b, Amy L. Weaver b, William A. Cliby a,⁎ a b
Division of Gynecologic Surgery, Mayo Clinic, 200 1st St. SW, Rochester, MN 55905, USA Division of Biomedical Statistics and Informatics, Mayo Clinic, Rochester, MN, USA
H I G H L I G H T S • The 30-day rate of significant conduit-related complications after incontinent conduit formation was 15.1%. • By 90 days, the Kaplan–Meier estimated rate of significant conduit-related complication was 21.8%. • There were lower rates of significant conduit-related complications in the transverse colon conduit group.
a r t i c l e
i n f o
Article history: Received 12 February 2014 Accepted 26 March 2014 Available online xxxx Keywords: Urinary conduit Incontinent Outcomes Gynecologic malignancy
a b s t r a c t Objective. The aim of this study is to estimate the overall rates of significant incontinent conduit-related complications and compare rates between conduit types. Methods. This was a retrospective review of 166 patients who underwent incontinent urinary diversion from April 1993 through April 2013. Patients were categorized by conduit type—ileal, sigmoid colon, and transverse colon. Significant conduit-related complications were assessed at 30 and 90 days after surgery. Significant conduit-related complication was defined as any of the following: ureteral stricture, conduit leak, conduit obstruction, conduit ischemia, ureteral anastomotic leak, stent obstruction requiring intervention via interventional radiology procedure or reoperation, and renal failure. Results. A total of 166 patients underwent formation of an incontinent urinary conduit, most commonly during exenteration for gynecologic malignancy. There were 129 ileal, 11 transverse colon, and 26 sigmoid conduits. The overall significant conduit-related complication rate within 30 days was 15.1%. Complication rates for ileal, transverse and sigmoid conduits were 14.7%, 0%, and 23.1%, respectively (Fisher's exact test, p = 0.24). By 90 days, the Kaplan–Meier estimated rates of significant complications were 21.8% overall, and 22.3%, 0%, and 28.9%, respectively, by conduit type (log-rank test, p = 0.19). The most common significant conduit-related complications were conduit or ureteral anastomotic leaks and conduit obstructions. By 1 and 2 years following surgery, the Kaplan–Meier estimated overall rate of significant conduit-related complication increased to 26.5% and 30.1%, respectively. Conclusions. Our study suggests that there are multiple appropriate tissue sites for use in incontinent conduit formation, and surgical approach should be individualized. Most significant conduit-related complications occur within 90 days after surgery. © 2014 Elsevier Inc. All rights reserved.
Introduction Formation of a urinary conduit is a critical aspect of reconstruction after pelvic exenteration for gynecologic malignancy. While there are several options, incontinent conduits continue to be a good choice for ⁎ Corresponding author. E-mail address: [email protected] (W.A. Cliby). 1 Co-authors.
simple diversion and preferable for many patients. These may be formed using any portion of the intestine, most commonly the distal ileum or sigmoid or transverse colon. There has been debate in the surgical literature regarding the optimal tissue and type of urinary conduit; while a comparison of colonic versus ileal conduits was published in 2012 [1] the largest recent series was published in 1996 [2] and was limited to outcomes for transverse colonic conduits. The last direct comparison of all conduit types appeared in 1986 [3]. In several studies examining outcomes for patients undergoing pelvic exenteration, use
http://dx.doi.org/10.1016/j.ygyno.2014.03.568 0090-8258/© 2014 Elsevier Inc. All rights reserved.
Please cite this article as: Tabbaa ZM, et al, Short-term outcomes after incontinent conduit for gynecologic cancer: Comparison of ileal, sigmoid, and transverse colon, Gynecol Oncol (2014), http://dx.doi.org/10.1016/j.ygyno.2014.03.568
2
Z.M. Tabbaa et al. / Gynecologic Oncology xxx (2014) xxx–xxx
of transverse colon conduits was associated with fewer serious surgical complications [4,5], with a possible advantage due to use of nonradiated bowel. However, other series have found no significant advantages to the use of one location of bowel harvest for the conduit versus another [1,2]. Recent studies have focused on continent versus incontinent urinary conduits, with similar complication rates reported aside from an increased risk of stone formation in continent conduits [6–9]. However, most studies lack long-term follow-up or patient satisfaction data, which are important in evaluating continent conduit outcomes. In one study 54% of patients who had a continent conduit reported that they would have chosen an incontinent conduit if given the option again, due to issues with self-catheterization [1]. In a German study comparing incontinent ileal conduit at the time of pelvic exenteration to formation of a continent ileal pouch, the mode of urinary diversion did not significantly impact survivor quality of life, but surgeries for stomal complications and treatment of stones were more common in continent diversions [10]. For many patients an incontinent conduit will remain a preferred option. As there is little recent literature comparing outcomes with use of various types of incontinent conduits, we analyzed our institutional experience using three types of conduits, with detailed analysis of type and timing of significant conduit-related complications and focused review of the precedent literature. These findings should be of value in counseling patients and making a reasoned surgical choice. Methods This was a retrospective review of patients undergoing incontinent urinary diversion by a gynecologic oncologist at Mayo Clinic from April 1993 through April 2013, either alone or at the time of total or anterior pelvic exenteration. Institutional Review Board approval was obtained for this study. Patients were excluded if a gynecologic oncologist did not perform surgery or if the patient did not consent to the use of their medical records for research purposes. The method of ileal conduit formation used was that described by Bricker [11]. Sigmoid conduits were created in the setting of end colostomies with no distal anastomosis. Urinary diversion using a transverse colon conduit was performed as described by Schmidt [12], with modification for use of modern gastrointestinal staplers, and is described briefly. Sufficient conduit length is obtained to allow formation of an everting stoma, and is usually based upon adequate blood supply from the middle colic artery. The greater omentum is dissected off the surface of the transverse colon, which is then divided, with bowel continuity reestablished by end-to-end anastomosis. The mesocolic defect is loosely closed. The end of the conduit chosen to be proximal is closed in two layers and is fixed to the posterior parietal peritoneum at or near the midline. The ureters are then dissected proximally, spatulated, and sutured into the conduit, with deployment of ureteric stents. The distal end of the conduit is then brought to the stoma site and matured. Medical records were reviewed and the following information was abstracted: age, body mass index (BMI), smoking status, presence of comorbidities (e.g. diabetes, hypertension), prior urinary tract issue, prior abdominal surgery, prior radiation exposure, preoperative creatinine, surgical indication, type of conduit, type of procedure, and concomitant pelvic floor or vaginal reconstruction. Follow-up was limited to clinical notes and correspondence contained in the medical record. All complications were recorded along with the date noted, as well as necessary interventions. We defined significant conduit-related complications as the following: ureteral stricture, conduit leak, conduit obstruction, conduit ischemia, ureteral anastomosis leak, stent obstruction requiring intervention via interventional radiology procedure or reoperation, and renal failure. All of the mentioned complications were recorded if noted in the chart, whether found on routine imaging or by symptomatology. The percentage of significant conduit-related complications requiring intervention was also recorded. Renal failure was defined as renal dysfunction requiring dialysis or noted as a prolonged clinical
problem in the chart. The date of the last relevant clinical follow-up, vital status and date of last follow-up or death were recorded. Results were summarized using standard descriptive statistics. Baseline patient characteristics were compared between the three conduit groups using the Wilcoxon rank-sum test for continuous variables and the Fisher's exact test for categorical variables. The complication rates within the first 30 days were compared between the three conduit groups using Fisher's exact test. The complication rates by 90 days, 1 year and 2 years, respectively, were estimated using the Kaplan– Meier method to take into account the varying duration of follow-up. The Kaplan–Meier methodology was chosen to estimate the 90-day complication rates due to censorship of patient deaths within the first 90 days and those who did not have 90 days of relevant clinical follow-up. All calculated p-values were two-sided and p-values less than 0.05 were considered statistically significant. Analyses were performed using the SAS version 9.2 software package (SAS Institute, Inc.; Cary, NC). Results During the 20-year period, a total of 166 patients underwent formation of an incontinent urinary conduit. The ileum was used for 129 conduits, the transverse colon for 11 and the sigmoid colon for 26. Patient characteristics were similar across conduit type, with a mean age of 60.2 years for the cohort as a whole. Body mass index was also similar across the groups with a mean of 28.0 kg/m2. The majority of patients were not current smokers, had a normal preoperative creatinine, a history of prior abdominal surgery, and had received prior radiation, with pelvic external beam therapy being the most common modality of treatment (Table 1). The indication for surgery is listed in Table 2; most were done as part of exenterative surgery though nearly 10% were due to fistula related to prior/current malignancy. In this cohort, only 7.2% (N = 12) underwent formation of a conduit only; the remainder underwent conduit formation with either total (44.0%, N = 73) or anterior (48.8%, N = 81) pelvic exenteration. Nearly three-fourths (70.5%, N = 117) had pelvic floor reconstruction with an omental flap, muscular flap, or combination. The primary outcome for this study was the rate of significant conduit-related complications, as defined in the methods, within 30 days and 90 days of surgery (Table 3). Of the 166 patients, 25 experienced a significant conduit-related complication within 30 days postoperatively, for an overall rate of 15.1%. Of the 141 patients without conduit-related complications, all but 10 had at least 30 days of relevant clinical follow-up (1 died post-operatively from a pulmonary embolism and 9 cases had between 6 and 24 days of follow-up). The 30-day conduit-related complication rate was 14.7% (19/129) for the ileal conduit group, 0.0% (0/11) for transverse colon, and 23.1% (6/26) for sigmoid colon. There was no statistically significant difference between the groups for 30-day significant conduit-related complications (p = 0.24, Fisher's exact test) though power was limited due to low numbers and lack of complications in the transverse colon group. There were 3 non-conduit anastomotic leaks diagnosed within the first 30 days after surgery, all in the ileal conduit group, for an ileal conduit rate of 2.3% and an overall rate of 1.8%. Radiation was not associated with complication rates when examining separate conduit groups. Within the first 90 days postoperatively, 34 patients (crude rate = 20.5%, Kaplan–Meier rate by 90 days = 21.8%) developed a significant conduit-related complication. Specifically for the ileal, transverse, and sigmoid conduit groups, Kaplan–Meier rates for significant 90-day conduit-related complication rates were 22.3% (N = 27), 0.0% (N = 0), and 28.9% (N = 7), respectively (log-rank p = 0.19). Smoking status, diabetes, prior urinary tract issue, and prior radiation were not associated with either of the two outcome measures (Table 4). We evaluated the management of significant conduit-related complications. Twenty-five patients had a 30-day conduit-related complication, of which 14 required an interventional radiology (IR) procedure
Please cite this article as: Tabbaa ZM, et al, Short-term outcomes after incontinent conduit for gynecologic cancer: Comparison of ileal, sigmoid, and transverse colon, Gynecol Oncol (2014), http://dx.doi.org/10.1016/j.ygyno.2014.03.568
Z.M. Tabbaa et al. / Gynecologic Oncology xxx (2014) xxx–xxx
3
Table 1 Patient characteristics. Characteristic
Age (years), mean (SD) BMI (kg/m2), mean (SD) Preoperative creatinine (mg/dL), mean (SD) Current smoker, N (%) Diabetes, N (%) Hypertension, N (%) Prior urinary tract issue, N (%) Prior abdominal surgery, N (%) Prior radiation, N (%) Vaginal brachytherapy Pelvic external beam Tandem and ovoids Interstitial implants Intraoperative Unknown Procedure, N (%) Conduit only Conduit and total exenteration Conduit and anterior exenteration Pelvic floor reconstruction, N (%) Vaginal reconstruction, N (%)
Type of conduit
Pa
Total (N = 166)
Ileal (N = 129)
Transverse colon (N = 11)
Sigmoid (N = 26)
59.5 (13.3) 28.1 (6.8) 1.0 (0.4) 21/120 (18) 20 (16) 48 (37) 53 (41) 105 (81) 111 (86) 46 91 5 4 31 11
66.0 (7.9) 27.6 (6.3) 0.8 (0.2) 0/11 (0) 1 (9) 5 (45) 6 (55) 10 (91) 11 (100) 3 8 0 0 5 2
60.9 (12.9) 27.6 (6.5) 1.0 (0.2) 3/25 (12) 6 (23) 11 (42) 6 (23) 21 (81) 24 (92) 7 17 0 1 2 5
60.2 (13.0) 28.0 (6.7) 1.0 (0.3) 24/156 (15) 27 (16) 64 (39) 65 (39) 136 (82) 146 (88) 56 116 5 5 38 18
12 (9) 43 (33) 74 (57) 84 (65) 37 (29)
0 (0) 5 (45) 6 (55) 10 (91) 1 (9)
0 (0) 25 (96) 1 (4) 23 (88) 10 (38)
12 (7) 73 (44) 81 (49) 117 (70) 48 (29)
0.18 0.97 0.06 0.37 0.62 0.77 0.14 0.87 0.46
b0.001
0.02 0.20
Abbreviations: BMI, body mass index; SD, standard deviation. a Wilcoxon rank-sum p-value presented for continuous variables and Fisher's exact p-value presented for categorical variables.
(10 of 19 in the ileal group and 4 of 6 in the sigmoid colon group). Five of the 25 patients with complications required a reoperation for a 30-day significant conduit-related complication (4 of 19 in the ileal group and 1 of 6 in the sigmoid colon group). Of the 34 patients with a 90-day significant conduit-related complication, 16 required an IR procedure (11 of 27 in the ileal group and 5 of 7 in the sigmoid group). Eight of these 34 patients required a reoperation for a 90-day significant conduit-related complication (7 of 27 in the ileal group and 1 of 7 in the sigmoid colon group). The significant conduit-related complications requiring intervention within 90 days in the sigmoid conduit group were as follows—4 patients developed a conduit leak, 1 developed conduit ischemia, and 1 developed obstruction and renal failure. Of note, despite historic descriptions of hyperchloremic acidosis attributed to use of the ileum, we did not see cases in any of the groups. The median duration of clinical follow-up after surgery was approximately 11, 15, and 9 months, respectively, in the ileal, transverse and sigmoid colon groups. By 1 and 2 years following surgery, the Kaplan–
Meier estimated rate of significant conduit-related complications increased to 26.5% and 30.1%, respectively. Long-term significant conduit-related complications (first after 90 days) occurred in 11 of the 166 patients (5 within 1 year and 6 between 1.6 and 6.1 years following surgery). Nine of those were in the ileal group; 4 ureteral strictures, 2 renal failures, 2 ureteral obstructions and 1 conduit leak and obstruction. In the transverse colon group there was 1 ureteral obstruction and 1 renal failure. No long-term significant conduit-related complications were reported in the sigmoid group. Discussion Urinary diversion via an incontinent conduit continues to be an essential tool for the gynecologic oncologist, notwithstanding debate regarding the ideal type. We did not observe a statistically significant difference between ileal, transverse colon, or sigmoid colon conduits in the rates of significant conduit-related complications. We did observe
Table 2 Indication for surgery. Indication
Recurrent cervical cancer Recurrent vaginal cancer Recurrent endometrial cancer Recurrent vulvar cancer Fistulaa Primary vaginal cancer Primary cervical cancer Recurrent ovarian cancer Genital tract melanoma Primary leiomyosarcoma Primary vulvar cancer Incontinence (history of leiomyosarcoma) Uterine cancer Neurogenic bladder Recurrent anal cancer Recurrent rectal cancer Primary anal cancer a
Type of conduit
Total (N = 166)
Ileal (N = 129)
Transverse colon (N = 11)
Sigmoid (N = 26)
N (%)
N (%)
N (%)
N (%)
43 (33) 20 (16) 20 (16) 13 (10) 11 (9) 4 (4) 5 (3) 3 (2) 2 (1) 2 (2) 0 1 (0.8) 1 (0.8) 1 (0.8) 1 (0.8) 1 (0.8) 1 (0.8)
2 (18) 2 (18) 1 (9) 3 (27) 3 (27) 0 0 0 0 0 0 0 0 0 0 0 0
11 (42) 3 (12) 4 (15) 5 (19) 1 (4) 1 (4) 0 0 0 0 1 (4) 0 0 0 0 0 0
56 (34) 25 (15) 25 (15) 21 (13) 15 (9) 5 (3) 5 (3) 3 (2) 2 (1) 2 (1) 1 (0) 1 (0.6) 1 (0.6) 1 (0.6) 1 (0.6) 1 (0.6) 1 (0.6)
Vesicovaginal, rectovaginal, urethrovaginal, ureterovaginal.
Please cite this article as: Tabbaa ZM, et al, Short-term outcomes after incontinent conduit for gynecologic cancer: Comparison of ileal, sigmoid, and transverse colon, Gynecol Oncol (2014), http://dx.doi.org/10.1016/j.ygyno.2014.03.568
4
Z.M. Tabbaa et al. / Gynecologic Oncology xxx (2014) xxx–xxx
Table 3 Significant conduit-related complications. Type of significant conduit-related complicationa First within 30 days N (crude rate, %) 95% CI Ureteral stricture, n Conduit leak, n Conduit obstruction, n Conduit ischemia, n Ureteral anastomotic leak, n Stent obstruction, nb Renal failure, n First within 90 days N (Kaplan–Meier rate, %) 95% CI Ureteral stricture, n Conduit leak, n Conduit obstruction, n Conduit ischemia, n Ureteral anastomotic leak, n Stent obstruction, nb Renal failure, n a b
Ileal (N = 129)
Transverse colon (N = 11)
Sigmoid (N = 26)
Total (N = 166)
19 (14.7) (9.1, 22.0) 1 8 4 0 5 2 1
0 (-) (0, 28.5) 0 0 0 0 0 0 0
6 (23.1) (9.0, 43.7) 0 4 1 1 0 0 1
25 (15.1) (10.0, 21.4) 1 12 5 1 5 2 2
27 (22.3) (14.5, 29.4) 4 13 4 0 5 2 1
0 (-) (-) 0 0 0 0 0 0 0
7 (28.9) (8.3, 44.9) 0 5 1 1 0 0 1
34 (21.8) (15.0, 28.0) 4 18 5 1 5 2 2
Some patients had more than one complication on the same date. Requiring intervention via interventional radiology procedure or additional surgery.
that a) sigmoid conduits more commonly resulted in significant conduit-related complications that required additional intervention for repair; however, the increased rate of intervention and repair did not reach statistical significance; and b) there were lower rates of significant conduit-related complications in the transverse colon conduit group, though small numbers limited statistical power. Similar to other institutions, our surgical practice has evolved away from use of the sigmoid colon for conduit formation. Theoretical benefits include use of tissue already being used for end colostomy and avoiding the necessity of a small or large bowel reanastomosis. However, in practice the use of tissue from within the primarily radiated field may lead to increased complications despite the avoidance of additional bowel manipulation. Indeed, in our series we observed an increase in significant conduit-related complications that required intervention and additional procedures in the sigmoid conduit group. The observed significant conduit-related complications (leak, ischemia, obstruction) are likely attributable to use of previously irradiated tissue. A 1986 study from MD Anderson examined the complications associated with formation of an ileal conduit and reported a trend toward higher incidence of complications with use of the sigmoid colon, and a Table 4 Evaluation of risk factors. Risk factor
Current smokerc No (N = 132) Yes (N = 24) Diabetes No (N = 139) Yes (N = 27) Prior urinary tract issue No (N = 101) Yes (N = 65) Prior radiation No (N = 20) Yes (N = 146) a b c
No. (%) with a significant conduit-related complication within 30 days
Pa
Kaplan–Meier rate for incidence of significant conduit-related complication by 90 days
0.99 21 (16) 3 (13)
Pb
0.79 21.9% 25.0%
0.25 19 6 (22)
0.35 20.6% 28.1%
0.83 16 (16) 9 (14)
0.83 20.8% 23.5%
0.51 4 (20) 21 (14)
Fisher's exact test p-value. Log-rank test p-value. Smoking status available on 156 of the 166 patients.
0.95 20.0% 22.1%
low incidence of conduit leaks in the transverse colon group, albeit with a lack of statistical significance due to fewer patients in the transverse colon group [3]. We observed a similar trend and similar lack of power due to numbers of transverse colon conduits. Others also reported a higher rate of complications with use of the sigmoid colon and ileum, and advocated the use of the transverse colon for conduit formation, particularly because of its lack of radiation exposure in a population of patients with usually heavily irradiated pelvises [4,5]. Other advantages are that the transverse colon may be more easily mobilized to optimize stomal placement, allows for higher ureteral anastomosis if working with shorter segments of viable ureter, and does not necessitate a small bowel reanastomosis. However, in cases requiring lower colon resection with proximal ligation of the inferior mesenteric artery, further disruption to the colonic blood supply required by transverse colon resection may be unwise. Tailored intraoperative decisions should take these factors into account given the data on both conduit types. In contrast, several recent studies have not demonstrated such a difference, albeit in series either small or limited to transverse colon conduits alone. Segreti et al. examined transverse colon urinary conduits performed in 86 patients, with the following complications—6% loss of renal function, 22% partial ureteral obstruction requiring nephrostomy or stenting, and 8% conduit revision. Complications tended to be relatively remote from surgery, with median time to conduit revision of 26 months. They concluded that the rate of complications was comparable to other tissue sites used, with no distinct advantage with use of the transverse colon [2]. This was also largely the conclusion of a series comparing 47 patients with ileal conduits to 27 with colonic conduits. Overall, 16% developed a conduit or conduit-related anastomotic leak, of which the majority (9 of 12) occurred within the first month. There was no difference between groups at either 30 days or 90 days postoperatively, although the numbers were lower in the transverse colon group (17% vs. 4% at 30 days). An interesting observation was that the transverse colon group did have a statistically significant lower rate of general significant complications, defined as any leaks, fistulae, septic events, or percutaneously drained collections (40% in the ileal group vs. 19% in the transverse colon group, p b 0.001) [1]. Most significant conduit-related complications occur within the first 30 postoperative days, although additional serious complications may occur within 90 postoperative days and beyond. The most common significant conduit-related complications are leak and obstruction, with associated pelvic/abdominal abscess in some cases. Our rate of significant conduit-related complications was similar to those reported, with
Please cite this article as: Tabbaa ZM, et al, Short-term outcomes after incontinent conduit for gynecologic cancer: Comparison of ileal, sigmoid, and transverse colon, Gynecol Oncol (2014), http://dx.doi.org/10.1016/j.ygyno.2014.03.568
Z.M. Tabbaa et al. / Gynecologic Oncology xxx (2014) xxx–xxx
an overall rate of significant conduit-related complications of 15.1% within 30 postoperative days. Although the number of patients in the transverse colon conduit group was small (n = 11), we observed no significant conduit-related complications in that group. Within 90 postoperative days, the rate of significant conduit-related complications for those who received a transverse colon conduit was 0.0% versus 22.3% and 28.9% in the ileal and sigmoid colon conduit groups respectively, as calculated using the Kaplan–Meier method. Overall, differences did not reach statistical significance. Coupled with existing literature comparing rates of complications related to transverse colon conduits the data suggests a probable benefit to using non-irradiated tissues for conduit creation without a concomitant drawback. Limitations of the current study include limited number of transverse colon conduits and the nature of retrospective review of data. Nonetheless these data contribute to the literature a detailed assessment of both timing and nature of significant conduit-related complications after incontinent conduit formation. For this population of heavily radiated patients, consideration should be given to transverse colon as an attractive alternative to ileal conduits for non-continent diversions. Factors to consider should include the extent and field of radiation, the types of colonic resections planned, the site of stoma placement, available ureteral length and the patient's comorbidities. Extended postoperative follow-up is imperative for these patients for whom significant morbidity is a distinct probability. Conflict of interest statement
5
References [1] Teixeira SC, Ferenschild FT, Solomon MJ, Rodwell L, Harrison JD, Young JM, et al. Urological leaks after pelvic exenterations comparing formation of colonic and ileal conduits. Eur J Surg Oncol 2012;38:361–6. [2] Segreti EM, Morris M, Levenback C, Lucas KR, Gershenson DM, Burke TW. Transverse colon urinary diversion in gynecologic oncology. Gynecol Oncol 1996;63:66–70. [3] Hancock KC, Copeland LJ, Gershenson DM, Saul PB, Wharton JT, Rutledge FN. Urinary conduits in gynecologic oncology. Obstet Gynecol 1986;67:680–4. [4] Soper JT, Berchuck A, Creasman WT, Clarke-Pearson DL. Pelvic exenteration: factors associated with major surgical morbidity. Gynecol Oncol 1989;35:93–8. [5] Orr Jr JW, Shingleton HM, Hatch KD, Taylor PT, Austin Jr JM, Partridge EE, et al. Urinary diversion in patients undergoing pelvic exenteration. Am J Obstet Gynecol 1982;142:883–9. [6] Urh A, Soliman PT, Schmeler KM, Westin S, Frumovitz M, Nick AM, et al. Postoperative outcomes after continent versus incontinent urinary diversion at the time of pelvic exenteration for gynecologic malignancies. Gynecol Oncol 2013;129:580–5. [7] Karsenty G, Moutardier V, Lelong B, Guiramand J, Houvenaeghel G, Delpero JR, et al. Long-term follow-up of continent urinary diversion after pelvic exenteration for gynecologic malignancies. Gynecol Oncol 2005;97:524–8. [8] Spahn M, Weiss C, Bader P, Frohneberg D, Studer UE, Burkhard FC. The role of exenterative surgery and urinary diversion in persistent or locally recurrent gynecological malignancy: complications and survival. Urol Int 2010;85:16–22. [9] Goldberg GL, Sukumvanich P, Einstein MH, Smith HO, Anderson PS, Fields AL. Total pelvic exenteration: the Albert Einstein College of Medicine/Montefiore Medical Center Experience (1987 to 2003). Gynecol Oncol 2006;101:261–8. [10] Forner DM, Lampe B. Ileal conduit and continent ileocecal pouch for patients undergoing pelvic exenteration: comparison of complications and quality of life. Int J Gynecol Cancer 2011;21:403–8. [11] Bricker EM. Bladder substitution after pelvic evisceration. Surg Clin North Am 1950;30:1511–21. [12] Schmidt JD, Hawtrey CE, Buchsbaum HJ. Transverse colon conduit: a preferred method of urinary diversion for radiation-treated pelvic malignancies. J Urol 1975;113:308–13.
The authors declare that there are no conflicts of interest.
Please cite this article as: Tabbaa ZM, et al, Short-term outcomes after incontinent conduit for gynecologic cancer: Comparison of ileal, sigmoid, and transverse colon, Gynecol Oncol (2014), http://dx.doi.org/10.1016/j.ygyno.2014.03.568
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Short-term outcomes after incontinent conduit for gynecologic cancer: Comparison of ileal, sigmoid, and transverse colon Zaid M. Tabbaa a,1, Jo Marie T. Janco a,1, Andrea Mariani a, Sean C. Dowdy a, Michaela E. McGree b, Amy L. Weaver b, William A. Cliby a,⁎ a b
Division of Gynecologic Surgery, Mayo Clinic, 200 1st St. SW, Rochester, MN 55905, USA Division of Biomedical Statistics and Informatics, Mayo Clinic, Rochester, MN, USA
H I G H L I G H T S • The 30-day rate of significant conduit-related complications after incontinent conduit formation was 15.1%. • By 90 days, the Kaplan–Meier estimated rate of significant conduit-related complication was 21.8%. • There were lower rates of significant conduit-related complications in the transverse colon conduit group.
a r t i c l e
i n f o
Article history: Received 12 February 2014 Accepted 26 March 2014 Available online xxxx Keywords: Urinary conduit Incontinent Outcomes Gynecologic malignancy
a b s t r a c t Objective. The aim of this study is to estimate the overall rates of significant incontinent conduit-related complications and compare rates between conduit types. Methods. This was a retrospective review of 166 patients who underwent incontinent urinary diversion from April 1993 through April 2013. Patients were categorized by conduit type—ileal, sigmoid colon, and transverse colon. Significant conduit-related complications were assessed at 30 and 90 days after surgery. Significant conduit-related complication was defined as any of the following: ureteral stricture, conduit leak, conduit obstruction, conduit ischemia, ureteral anastomotic leak, stent obstruction requiring intervention via interventional radiology procedure or reoperation, and renal failure. Results. A total of 166 patients underwent formation of an incontinent urinary conduit, most commonly during exenteration for gynecologic malignancy. There were 129 ileal, 11 transverse colon, and 26 sigmoid conduits. The overall significant conduit-related complication rate within 30 days was 15.1%. Complication rates for ileal, transverse and sigmoid conduits were 14.7%, 0%, and 23.1%, respectively (Fisher's exact test, p = 0.24). By 90 days, the Kaplan–Meier estimated rates of significant complications were 21.8% overall, and 22.3%, 0%, and 28.9%, respectively, by conduit type (log-rank test, p = 0.19). The most common significant conduit-related complications were conduit or ureteral anastomotic leaks and conduit obstructions. By 1 and 2 years following surgery, the Kaplan–Meier estimated overall rate of significant conduit-related complication increased to 26.5% and 30.1%, respectively. Conclusions. Our study suggests that there are multiple appropriate tissue sites for use in incontinent conduit formation, and surgical approach should be individualized. Most significant conduit-related complications occur within 90 days after surgery. © 2014 Elsevier Inc. All rights reserved.
Introduction Formation of a urinary conduit is a critical aspect of reconstruction after pelvic exenteration for gynecologic malignancy. While there are several options, incontinent conduits continue to be a good choice for ⁎ Corresponding author. E-mail address: [email protected] (W.A. Cliby). 1 Co-authors.
simple diversion and preferable for many patients. These may be formed using any portion of the intestine, most commonly the distal ileum or sigmoid or transverse colon. There has been debate in the surgical literature regarding the optimal tissue and type of urinary conduit; while a comparison of colonic versus ileal conduits was published in 2012 [1] the largest recent series was published in 1996 [2] and was limited to outcomes for transverse colonic conduits. The last direct comparison of all conduit types appeared in 1986 [3]. In several studies examining outcomes for patients undergoing pelvic exenteration, use
http://dx.doi.org/10.1016/j.ygyno.2014.03.568 0090-8258/© 2014 Elsevier Inc. All rights reserved.
Please cite this article as: Tabbaa ZM, et al, Short-term outcomes after incontinent conduit for gynecologic cancer: Comparison of ileal, sigmoid, and transverse colon, Gynecol Oncol (2014), http://dx.doi.org/10.1016/j.ygyno.2014.03.568
2
Z.M. Tabbaa et al. / Gynecologic Oncology xxx (2014) xxx–xxx
of transverse colon conduits was associated with fewer serious surgical complications [4,5], with a possible advantage due to use of nonradiated bowel. However, other series have found no significant advantages to the use of one location of bowel harvest for the conduit versus another [1,2]. Recent studies have focused on continent versus incontinent urinary conduits, with similar complication rates reported aside from an increased risk of stone formation in continent conduits [6–9]. However, most studies lack long-term follow-up or patient satisfaction data, which are important in evaluating continent conduit outcomes. In one study 54% of patients who had a continent conduit reported that they would have chosen an incontinent conduit if given the option again, due to issues with self-catheterization [1]. In a German study comparing incontinent ileal conduit at the time of pelvic exenteration to formation of a continent ileal pouch, the mode of urinary diversion did not significantly impact survivor quality of life, but surgeries for stomal complications and treatment of stones were more common in continent diversions [10]. For many patients an incontinent conduit will remain a preferred option. As there is little recent literature comparing outcomes with use of various types of incontinent conduits, we analyzed our institutional experience using three types of conduits, with detailed analysis of type and timing of significant conduit-related complications and focused review of the precedent literature. These findings should be of value in counseling patients and making a reasoned surgical choice. Methods This was a retrospective review of patients undergoing incontinent urinary diversion by a gynecologic oncologist at Mayo Clinic from April 1993 through April 2013, either alone or at the time of total or anterior pelvic exenteration. Institutional Review Board approval was obtained for this study. Patients were excluded if a gynecologic oncologist did not perform surgery or if the patient did not consent to the use of their medical records for research purposes. The method of ileal conduit formation used was that described by Bricker [11]. Sigmoid conduits were created in the setting of end colostomies with no distal anastomosis. Urinary diversion using a transverse colon conduit was performed as described by Schmidt [12], with modification for use of modern gastrointestinal staplers, and is described briefly. Sufficient conduit length is obtained to allow formation of an everting stoma, and is usually based upon adequate blood supply from the middle colic artery. The greater omentum is dissected off the surface of the transverse colon, which is then divided, with bowel continuity reestablished by end-to-end anastomosis. The mesocolic defect is loosely closed. The end of the conduit chosen to be proximal is closed in two layers and is fixed to the posterior parietal peritoneum at or near the midline. The ureters are then dissected proximally, spatulated, and sutured into the conduit, with deployment of ureteric stents. The distal end of the conduit is then brought to the stoma site and matured. Medical records were reviewed and the following information was abstracted: age, body mass index (BMI), smoking status, presence of comorbidities (e.g. diabetes, hypertension), prior urinary tract issue, prior abdominal surgery, prior radiation exposure, preoperative creatinine, surgical indication, type of conduit, type of procedure, and concomitant pelvic floor or vaginal reconstruction. Follow-up was limited to clinical notes and correspondence contained in the medical record. All complications were recorded along with the date noted, as well as necessary interventions. We defined significant conduit-related complications as the following: ureteral stricture, conduit leak, conduit obstruction, conduit ischemia, ureteral anastomosis leak, stent obstruction requiring intervention via interventional radiology procedure or reoperation, and renal failure. All of the mentioned complications were recorded if noted in the chart, whether found on routine imaging or by symptomatology. The percentage of significant conduit-related complications requiring intervention was also recorded. Renal failure was defined as renal dysfunction requiring dialysis or noted as a prolonged clinical
problem in the chart. The date of the last relevant clinical follow-up, vital status and date of last follow-up or death were recorded. Results were summarized using standard descriptive statistics. Baseline patient characteristics were compared between the three conduit groups using the Wilcoxon rank-sum test for continuous variables and the Fisher's exact test for categorical variables. The complication rates within the first 30 days were compared between the three conduit groups using Fisher's exact test. The complication rates by 90 days, 1 year and 2 years, respectively, were estimated using the Kaplan– Meier method to take into account the varying duration of follow-up. The Kaplan–Meier methodology was chosen to estimate the 90-day complication rates due to censorship of patient deaths within the first 90 days and those who did not have 90 days of relevant clinical follow-up. All calculated p-values were two-sided and p-values less than 0.05 were considered statistically significant. Analyses were performed using the SAS version 9.2 software package (SAS Institute, Inc.; Cary, NC). Results During the 20-year period, a total of 166 patients underwent formation of an incontinent urinary conduit. The ileum was used for 129 conduits, the transverse colon for 11 and the sigmoid colon for 26. Patient characteristics were similar across conduit type, with a mean age of 60.2 years for the cohort as a whole. Body mass index was also similar across the groups with a mean of 28.0 kg/m2. The majority of patients were not current smokers, had a normal preoperative creatinine, a history of prior abdominal surgery, and had received prior radiation, with pelvic external beam therapy being the most common modality of treatment (Table 1). The indication for surgery is listed in Table 2; most were done as part of exenterative surgery though nearly 10% were due to fistula related to prior/current malignancy. In this cohort, only 7.2% (N = 12) underwent formation of a conduit only; the remainder underwent conduit formation with either total (44.0%, N = 73) or anterior (48.8%, N = 81) pelvic exenteration. Nearly three-fourths (70.5%, N = 117) had pelvic floor reconstruction with an omental flap, muscular flap, or combination. The primary outcome for this study was the rate of significant conduit-related complications, as defined in the methods, within 30 days and 90 days of surgery (Table 3). Of the 166 patients, 25 experienced a significant conduit-related complication within 30 days postoperatively, for an overall rate of 15.1%. Of the 141 patients without conduit-related complications, all but 10 had at least 30 days of relevant clinical follow-up (1 died post-operatively from a pulmonary embolism and 9 cases had between 6 and 24 days of follow-up). The 30-day conduit-related complication rate was 14.7% (19/129) for the ileal conduit group, 0.0% (0/11) for transverse colon, and 23.1% (6/26) for sigmoid colon. There was no statistically significant difference between the groups for 30-day significant conduit-related complications (p = 0.24, Fisher's exact test) though power was limited due to low numbers and lack of complications in the transverse colon group. There were 3 non-conduit anastomotic leaks diagnosed within the first 30 days after surgery, all in the ileal conduit group, for an ileal conduit rate of 2.3% and an overall rate of 1.8%. Radiation was not associated with complication rates when examining separate conduit groups. Within the first 90 days postoperatively, 34 patients (crude rate = 20.5%, Kaplan–Meier rate by 90 days = 21.8%) developed a significant conduit-related complication. Specifically for the ileal, transverse, and sigmoid conduit groups, Kaplan–Meier rates for significant 90-day conduit-related complication rates were 22.3% (N = 27), 0.0% (N = 0), and 28.9% (N = 7), respectively (log-rank p = 0.19). Smoking status, diabetes, prior urinary tract issue, and prior radiation were not associated with either of the two outcome measures (Table 4). We evaluated the management of significant conduit-related complications. Twenty-five patients had a 30-day conduit-related complication, of which 14 required an interventional radiology (IR) procedure
Please cite this article as: Tabbaa ZM, et al, Short-term outcomes after incontinent conduit for gynecologic cancer: Comparison of ileal, sigmoid, and transverse colon, Gynecol Oncol (2014), http://dx.doi.org/10.1016/j.ygyno.2014.03.568
Z.M. Tabbaa et al. / Gynecologic Oncology xxx (2014) xxx–xxx
3
Table 1 Patient characteristics. Characteristic
Age (years), mean (SD) BMI (kg/m2), mean (SD) Preoperative creatinine (mg/dL), mean (SD) Current smoker, N (%) Diabetes, N (%) Hypertension, N (%) Prior urinary tract issue, N (%) Prior abdominal surgery, N (%) Prior radiation, N (%) Vaginal brachytherapy Pelvic external beam Tandem and ovoids Interstitial implants Intraoperative Unknown Procedure, N (%) Conduit only Conduit and total exenteration Conduit and anterior exenteration Pelvic floor reconstruction, N (%) Vaginal reconstruction, N (%)
Type of conduit
Pa
Total (N = 166)
Ileal (N = 129)
Transverse colon (N = 11)
Sigmoid (N = 26)
59.5 (13.3) 28.1 (6.8) 1.0 (0.4) 21/120 (18) 20 (16) 48 (37) 53 (41) 105 (81) 111 (86) 46 91 5 4 31 11
66.0 (7.9) 27.6 (6.3) 0.8 (0.2) 0/11 (0) 1 (9) 5 (45) 6 (55) 10 (91) 11 (100) 3 8 0 0 5 2
60.9 (12.9) 27.6 (6.5) 1.0 (0.2) 3/25 (12) 6 (23) 11 (42) 6 (23) 21 (81) 24 (92) 7 17 0 1 2 5
60.2 (13.0) 28.0 (6.7) 1.0 (0.3) 24/156 (15) 27 (16) 64 (39) 65 (39) 136 (82) 146 (88) 56 116 5 5 38 18
12 (9) 43 (33) 74 (57) 84 (65) 37 (29)
0 (0) 5 (45) 6 (55) 10 (91) 1 (9)
0 (0) 25 (96) 1 (4) 23 (88) 10 (38)
12 (7) 73 (44) 81 (49) 117 (70) 48 (29)
0.18 0.97 0.06 0.37 0.62 0.77 0.14 0.87 0.46
b0.001
0.02 0.20
Abbreviations: BMI, body mass index; SD, standard deviation. a Wilcoxon rank-sum p-value presented for continuous variables and Fisher's exact p-value presented for categorical variables.
(10 of 19 in the ileal group and 4 of 6 in the sigmoid colon group). Five of the 25 patients with complications required a reoperation for a 30-day significant conduit-related complication (4 of 19 in the ileal group and 1 of 6 in the sigmoid colon group). Of the 34 patients with a 90-day significant conduit-related complication, 16 required an IR procedure (11 of 27 in the ileal group and 5 of 7 in the sigmoid group). Eight of these 34 patients required a reoperation for a 90-day significant conduit-related complication (7 of 27 in the ileal group and 1 of 7 in the sigmoid colon group). The significant conduit-related complications requiring intervention within 90 days in the sigmoid conduit group were as follows—4 patients developed a conduit leak, 1 developed conduit ischemia, and 1 developed obstruction and renal failure. Of note, despite historic descriptions of hyperchloremic acidosis attributed to use of the ileum, we did not see cases in any of the groups. The median duration of clinical follow-up after surgery was approximately 11, 15, and 9 months, respectively, in the ileal, transverse and sigmoid colon groups. By 1 and 2 years following surgery, the Kaplan–
Meier estimated rate of significant conduit-related complications increased to 26.5% and 30.1%, respectively. Long-term significant conduit-related complications (first after 90 days) occurred in 11 of the 166 patients (5 within 1 year and 6 between 1.6 and 6.1 years following surgery). Nine of those were in the ileal group; 4 ureteral strictures, 2 renal failures, 2 ureteral obstructions and 1 conduit leak and obstruction. In the transverse colon group there was 1 ureteral obstruction and 1 renal failure. No long-term significant conduit-related complications were reported in the sigmoid group. Discussion Urinary diversion via an incontinent conduit continues to be an essential tool for the gynecologic oncologist, notwithstanding debate regarding the ideal type. We did not observe a statistically significant difference between ileal, transverse colon, or sigmoid colon conduits in the rates of significant conduit-related complications. We did observe
Table 2 Indication for surgery. Indication
Recurrent cervical cancer Recurrent vaginal cancer Recurrent endometrial cancer Recurrent vulvar cancer Fistulaa Primary vaginal cancer Primary cervical cancer Recurrent ovarian cancer Genital tract melanoma Primary leiomyosarcoma Primary vulvar cancer Incontinence (history of leiomyosarcoma) Uterine cancer Neurogenic bladder Recurrent anal cancer Recurrent rectal cancer Primary anal cancer a
Type of conduit
Total (N = 166)
Ileal (N = 129)
Transverse colon (N = 11)
Sigmoid (N = 26)
N (%)
N (%)
N (%)
N (%)
43 (33) 20 (16) 20 (16) 13 (10) 11 (9) 4 (4) 5 (3) 3 (2) 2 (1) 2 (2) 0 1 (0.8) 1 (0.8) 1 (0.8) 1 (0.8) 1 (0.8) 1 (0.8)
2 (18) 2 (18) 1 (9) 3 (27) 3 (27) 0 0 0 0 0 0 0 0 0 0 0 0
11 (42) 3 (12) 4 (15) 5 (19) 1 (4) 1 (4) 0 0 0 0 1 (4) 0 0 0 0 0 0
56 (34) 25 (15) 25 (15) 21 (13) 15 (9) 5 (3) 5 (3) 3 (2) 2 (1) 2 (1) 1 (0) 1 (0.6) 1 (0.6) 1 (0.6) 1 (0.6) 1 (0.6) 1 (0.6)
Vesicovaginal, rectovaginal, urethrovaginal, ureterovaginal.
Please cite this article as: Tabbaa ZM, et al, Short-term outcomes after incontinent conduit for gynecologic cancer: Comparison of ileal, sigmoid, and transverse colon, Gynecol Oncol (2014), http://dx.doi.org/10.1016/j.ygyno.2014.03.568
4
Z.M. Tabbaa et al. / Gynecologic Oncology xxx (2014) xxx–xxx
Table 3 Significant conduit-related complications. Type of significant conduit-related complicationa First within 30 days N (crude rate, %) 95% CI Ureteral stricture, n Conduit leak, n Conduit obstruction, n Conduit ischemia, n Ureteral anastomotic leak, n Stent obstruction, nb Renal failure, n First within 90 days N (Kaplan–Meier rate, %) 95% CI Ureteral stricture, n Conduit leak, n Conduit obstruction, n Conduit ischemia, n Ureteral anastomotic leak, n Stent obstruction, nb Renal failure, n a b
Ileal (N = 129)
Transverse colon (N = 11)
Sigmoid (N = 26)
Total (N = 166)
19 (14.7) (9.1, 22.0) 1 8 4 0 5 2 1
0 (-) (0, 28.5) 0 0 0 0 0 0 0
6 (23.1) (9.0, 43.7) 0 4 1 1 0 0 1
25 (15.1) (10.0, 21.4) 1 12 5 1 5 2 2
27 (22.3) (14.5, 29.4) 4 13 4 0 5 2 1
0 (-) (-) 0 0 0 0 0 0 0
7 (28.9) (8.3, 44.9) 0 5 1 1 0 0 1
34 (21.8) (15.0, 28.0) 4 18 5 1 5 2 2
Some patients had more than one complication on the same date. Requiring intervention via interventional radiology procedure or additional surgery.
that a) sigmoid conduits more commonly resulted in significant conduit-related complications that required additional intervention for repair; however, the increased rate of intervention and repair did not reach statistical significance; and b) there were lower rates of significant conduit-related complications in the transverse colon conduit group, though small numbers limited statistical power. Similar to other institutions, our surgical practice has evolved away from use of the sigmoid colon for conduit formation. Theoretical benefits include use of tissue already being used for end colostomy and avoiding the necessity of a small or large bowel reanastomosis. However, in practice the use of tissue from within the primarily radiated field may lead to increased complications despite the avoidance of additional bowel manipulation. Indeed, in our series we observed an increase in significant conduit-related complications that required intervention and additional procedures in the sigmoid conduit group. The observed significant conduit-related complications (leak, ischemia, obstruction) are likely attributable to use of previously irradiated tissue. A 1986 study from MD Anderson examined the complications associated with formation of an ileal conduit and reported a trend toward higher incidence of complications with use of the sigmoid colon, and a Table 4 Evaluation of risk factors. Risk factor
Current smokerc No (N = 132) Yes (N = 24) Diabetes No (N = 139) Yes (N = 27) Prior urinary tract issue No (N = 101) Yes (N = 65) Prior radiation No (N = 20) Yes (N = 146) a b c
No. (%) with a significant conduit-related complication within 30 days
Pa
Kaplan–Meier rate for incidence of significant conduit-related complication by 90 days
0.99 21 (16) 3 (13)
Pb
0.79 21.9% 25.0%
0.25 19 6 (22)
0.35 20.6% 28.1%
0.83 16 (16) 9 (14)
0.83 20.8% 23.5%
0.51 4 (20) 21 (14)
Fisher's exact test p-value. Log-rank test p-value. Smoking status available on 156 of the 166 patients.
0.95 20.0% 22.1%
low incidence of conduit leaks in the transverse colon group, albeit with a lack of statistical significance due to fewer patients in the transverse colon group [3]. We observed a similar trend and similar lack of power due to numbers of transverse colon conduits. Others also reported a higher rate of complications with use of the sigmoid colon and ileum, and advocated the use of the transverse colon for conduit formation, particularly because of its lack of radiation exposure in a population of patients with usually heavily irradiated pelvises [4,5]. Other advantages are that the transverse colon may be more easily mobilized to optimize stomal placement, allows for higher ureteral anastomosis if working with shorter segments of viable ureter, and does not necessitate a small bowel reanastomosis. However, in cases requiring lower colon resection with proximal ligation of the inferior mesenteric artery, further disruption to the colonic blood supply required by transverse colon resection may be unwise. Tailored intraoperative decisions should take these factors into account given the data on both conduit types. In contrast, several recent studies have not demonstrated such a difference, albeit in series either small or limited to transverse colon conduits alone. Segreti et al. examined transverse colon urinary conduits performed in 86 patients, with the following complications—6% loss of renal function, 22% partial ureteral obstruction requiring nephrostomy or stenting, and 8% conduit revision. Complications tended to be relatively remote from surgery, with median time to conduit revision of 26 months. They concluded that the rate of complications was comparable to other tissue sites used, with no distinct advantage with use of the transverse colon [2]. This was also largely the conclusion of a series comparing 47 patients with ileal conduits to 27 with colonic conduits. Overall, 16% developed a conduit or conduit-related anastomotic leak, of which the majority (9 of 12) occurred within the first month. There was no difference between groups at either 30 days or 90 days postoperatively, although the numbers were lower in the transverse colon group (17% vs. 4% at 30 days). An interesting observation was that the transverse colon group did have a statistically significant lower rate of general significant complications, defined as any leaks, fistulae, septic events, or percutaneously drained collections (40% in the ileal group vs. 19% in the transverse colon group, p b 0.001) [1]. Most significant conduit-related complications occur within the first 30 postoperative days, although additional serious complications may occur within 90 postoperative days and beyond. The most common significant conduit-related complications are leak and obstruction, with associated pelvic/abdominal abscess in some cases. Our rate of significant conduit-related complications was similar to those reported, with
Please cite this article as: Tabbaa ZM, et al, Short-term outcomes after incontinent conduit for gynecologic cancer: Comparison of ileal, sigmoid, and transverse colon, Gynecol Oncol (2014), http://dx.doi.org/10.1016/j.ygyno.2014.03.568
Z.M. Tabbaa et al. / Gynecologic Oncology xxx (2014) xxx–xxx
an overall rate of significant conduit-related complications of 15.1% within 30 postoperative days. Although the number of patients in the transverse colon conduit group was small (n = 11), we observed no significant conduit-related complications in that group. Within 90 postoperative days, the rate of significant conduit-related complications for those who received a transverse colon conduit was 0.0% versus 22.3% and 28.9% in the ileal and sigmoid colon conduit groups respectively, as calculated using the Kaplan–Meier method. Overall, differences did not reach statistical significance. Coupled with existing literature comparing rates of complications related to transverse colon conduits the data suggests a probable benefit to using non-irradiated tissues for conduit creation without a concomitant drawback. Limitations of the current study include limited number of transverse colon conduits and the nature of retrospective review of data. Nonetheless these data contribute to the literature a detailed assessment of both timing and nature of significant conduit-related complications after incontinent conduit formation. For this population of heavily radiated patients, consideration should be given to transverse colon as an attractive alternative to ileal conduits for non-continent diversions. Factors to consider should include the extent and field of radiation, the types of colonic resections planned, the site of stoma placement, available ureteral length and the patient's comorbidities. Extended postoperative follow-up is imperative for these patients for whom significant morbidity is a distinct probability. Conflict of interest statement
5
References [1] Teixeira SC, Ferenschild FT, Solomon MJ, Rodwell L, Harrison JD, Young JM, et al. Urological leaks after pelvic exenterations comparing formation of colonic and ileal conduits. Eur J Surg Oncol 2012;38:361–6. [2] Segreti EM, Morris M, Levenback C, Lucas KR, Gershenson DM, Burke TW. Transverse colon urinary diversion in gynecologic oncology. Gynecol Oncol 1996;63:66–70. [3] Hancock KC, Copeland LJ, Gershenson DM, Saul PB, Wharton JT, Rutledge FN. Urinary conduits in gynecologic oncology. Obstet Gynecol 1986;67:680–4. [4] Soper JT, Berchuck A, Creasman WT, Clarke-Pearson DL. Pelvic exenteration: factors associated with major surgical morbidity. Gynecol Oncol 1989;35:93–8. [5] Orr Jr JW, Shingleton HM, Hatch KD, Taylor PT, Austin Jr JM, Partridge EE, et al. Urinary diversion in patients undergoing pelvic exenteration. Am J Obstet Gynecol 1982;142:883–9. [6] Urh A, Soliman PT, Schmeler KM, Westin S, Frumovitz M, Nick AM, et al. Postoperative outcomes after continent versus incontinent urinary diversion at the time of pelvic exenteration for gynecologic malignancies. Gynecol Oncol 2013;129:580–5. [7] Karsenty G, Moutardier V, Lelong B, Guiramand J, Houvenaeghel G, Delpero JR, et al. Long-term follow-up of continent urinary diversion after pelvic exenteration for gynecologic malignancies. Gynecol Oncol 2005;97:524–8. [8] Spahn M, Weiss C, Bader P, Frohneberg D, Studer UE, Burkhard FC. The role of exenterative surgery and urinary diversion in persistent or locally recurrent gynecological malignancy: complications and survival. Urol Int 2010;85:16–22. [9] Goldberg GL, Sukumvanich P, Einstein MH, Smith HO, Anderson PS, Fields AL. Total pelvic exenteration: the Albert Einstein College of Medicine/Montefiore Medical Center Experience (1987 to 2003). Gynecol Oncol 2006;101:261–8. [10] Forner DM, Lampe B. Ileal conduit and continent ileocecal pouch for patients undergoing pelvic exenteration: comparison of complications and quality of life. Int J Gynecol Cancer 2011;21:403–8. [11] Bricker EM. Bladder substitution after pelvic evisceration. Surg Clin North Am 1950;30:1511–21. [12] Schmidt JD, Hawtrey CE, Buchsbaum HJ. Transverse colon conduit: a preferred method of urinary diversion for radiation-treated pelvic malignancies. J Urol 1975;113:308–13.
The authors declare that there are no conflicts of interest.
Please cite this article as: Tabbaa ZM, et al, Short-term outcomes after incontinent conduit for gynecologic cancer: Comparison of ileal, sigmoid, and transverse colon, Gynecol Oncol (2014), http://dx.doi.org/10.1016/j.ygyno.2014.03.568