Michael L. Blute, Department of Urology, 200 First Street SW, Rochester, MN 55905, USA. e-mail: firstname.lastname@example.org
To review our experience with orthotopic neobladder reconstruction (ONR) in women, highlighting functional and oncological outcomes, as ONR has been used increasingly for urinary diversion in women after radical cystectomy (RC), largely due to a better understanding of the natural history of urothelial cell carcinoma (UCC) and of the anatomy underlying the female continence mechanism, but defining the safety and long-term efficacy of ONR remain important to expanding its use.
PATIENTS AND METHODS
We identified 59 women who had RC with ONR between 1995 and 2006 at the Mayo Clinic. Their records were reviewed for clinicopathological demographics, and functional and oncological outcomes.
The median (range) age at surgery of the women was 62 (20–82) years, and the median follow-up was 29.2 (1–141) months. Fifty-three women had RC for malignant disease, including UCC in 47. Five women (8.5%) required resection of the anterior vaginal wall during surgery, while 39 (66%) had concurrent hysterectomy. One patient had a positive surgical margin, at the left distal ureter, on final pathology. Thirteen (22%) patients had perioperative complications, including two (3%) who required reoperation. In addition, three patients (5%) developed a neobladder-vaginal fistula after RC with ONR, requiring surgical repair. At the last follow-up, daytime urinary continence (defined as needing no pads) was reported by 44/49 (90%) patients, while 28/49 (57%) had achieved continence at night. Seventeen (31%) patients required intermittent self-catheterization to facilitate emptying of the neobladder. Fifteen (28%) women had recurrence of disease, at a median of 8 (2–36) months after ONR, including seven (13%) who developed a pelvic recurrence. Twelve (20%) patients died during the follow-up, with four (8%) dying from UCC at a median of 14 (11–65) months after surgery.
ONR provides a safe and effective option for urinary diversion in women, with oncological and functional outcomes similar to those for men. Thus, the technique remains a preferred option for urinary diversion in appropriately selected women treated with RC.
Radical cystectomy (RC) remains the reference standard treatment for patients with muscle-invasive urothelial cell carcinoma (UCC) of the bladder, and for patients with high-risk noninvasive UCC . In women, this procedure has classically involved en bloc cystectomy, hysterectomy, bilateral salpingo-oophorectomy, anterior vaginal wall resection, and total urethrectomy , followed by urinary diversion with ileal conduit or continent cutaneous reservoir. Over the past two decades, the use of orthotopic neobladder reconstruction (ONR), which obviates the need for an abdominal wall stoma or cutaneous appliance, and offers patients the ability to maintain a more natural volitional voiding pattern, has become an increasingly popular choice for urinary diversion after RC in men. Indeed, a previous study showed, using validated questionnaires, that patients’ quality of life was better after ONR than after an ileal conduit diversion, with preserved self-esteem, less leakage leading to wet clothing, and a shorter period of physical rehabilitation after surgery . Moreover, 97% of patients from that study who had ONR reported that they would recommend that diversion to a friend, while only 36% of patients treated with ileal conduit diversion would recommend a conduit .
Women were initially excluded from ONR due to concerns about the increased risks of urethral recurrence and postoperative urinary incontinence. However, an improved understanding of the anatomy of the female rhabdosphincter , and studies detailing the natural history of UCC in women [5–10], have led to the increased use of ONR for women after RC, with reports to date showing oncological outcomes similar to those reported in men after ONR [11,12] or ileal conduit diversion .
Importantly, several technical modifications to RC have been developed for women undergoing planned ONR, in an effort to improve the functional outcomes. For example, preserving the anterior vaginal wall has been described as a method to maintain pelvic support for the neobladder and thereby improve urinary continence . Also, neurovascular preservation at the time of RC and ONR was used to preserve sexual function .
We review our institution’s experience with RC and ONR in women, to detail the perioperative outcomes and the functional and oncological results.
PATIENTS AND METHODS
After review from the Institutional Review Board, we identified all women who had RC with ONR between 1995 and 2006 at the Mayo Clinic. Patients were considered candidates for ONR if they had a preoperative serum creatinine level of <2.0 mg/dL, normal preoperative bowel function, and were physically able and willing to use intermittent self-catheterization. Patients with an intraoperative frozen-section analysis showing tumour at the distal urethral margin, and those in whom the rhabdosphincter was felt to be compromised after RC, were excluded from ONR during RC. Our technique for RC in women with planned ONR was described previously . The vaginal wall was spared in all patients with no palpable tumour at the bladder base. The ON was constructed using a segment of terminal ileum according to the technique described by Studer and Zingg .
Patient records were reviewed for clinicopathological demographics and functional and oncological outcomes. All patients were followed at our institution, the follow-up comprising voided urine cytology, upper tract imaging, electrolyte panel, and serum creatinine levels at 3-month intervals for the first 2 years after surgery, at 6-month intervals for the next 3 years, and annually thereafter. Perioperative complications were defined as those occurring within 30 days of surgery. Recurrence was defined by radiographic or pathological confirmation of cancer after RC. Day- and night-time urinary continence was assessed by patient interview during the follow-up visits and through an examination of the medical record. Only those patients who required no use of pads or medication for urinary leakage were considered continent. The recurrence-free, cancer-specific, and overall survival were estimated using the Kaplan-Meier method, with patients who had RC and ONR for benign disease excluded from this analysis.
We identified 59 women who had RC with ONR at our institution, with a median (range) age at surgery was 62 (20–82) years; other patient demographics are listed in Table 1. The indications for RC are also shown in Table 1; UCC was the main indication, in 47/59 (80%) patients. At the time of surgery, 15/59 (25%) patients had a history of smoking, while 20/59 (33%) were current smokers.
Table 1. The pathology at RC in women undergoing ONR
Squamous cell carcinoma
Small cell carcinoma
Non-cancer indication for cystectomy
Tumour stage (53 women)
Tumour grade (35 women)
Lymph node status
At the time of RC, 39/59 (66%) women had a concurrent hysterectomy, with the remainder having had a previous hysterectomy. In addition, five of 59 (8%) patients had en bloc vaginal wall resection with RC, while the anterior vaginal wall was left intact in 54 (92%). One patient was found to have a positive margin on final pathology, at the distal left ureter. Moreover, 13 (22%) patients had a perioperative complication (Table 2), that required reoperation in two (3%), i.e. one who required endoscopic cauterization of a bleeding duodenal ulcer, and a second who had a laparotomy with bowel resection for a postoperative small-bowel obstruction. In addition, three (5%) patients developed a neobladder-vaginal fistula after ONR, which occurred at 1.5, 2 and 7 months, respectively, after surgery, and required operative repair in all cases. Moreover, de novo chronic renal insufficiency (creatinine level >2 mg/dL) developed in three (5%) women after ONR, and was managed medically in two and by conversion to an ileal conduit in the third patient. Indeed, four (7%) patients have subsequently required conversion of the neobladder to an ileal conduit during the follow-up, one for de novo chronic renal insufficiency, one for recurrent neobladder-vaginal fistulae, and two because of a pelvic recurrence of UCC.
Table 2. Perioperative complications after ONR in women
Acute renal failure
Deep venous thrombosis
The median (range) follow-up after ONR was 29.2 (1–141) months; 15 (28%) of the 53 patients who had RC for cancer have had recurrence of disease, at a median of 8 (2–36) months after surgery, including seven (13%) who had a recurrence in the pelvis. Two of the pelvic recurrences occurred at the vaginal cuff, two were at the pelvic side-wall, and one patient each recurred in the left ureter, the urethra, and the neobladder lumen. The primary tumour pathology, management and outcome of these seven patients are listed in Table 3. In all, 12 of 53 (23%) women who had RC with ONR for malignancy had died at the last follow-up, including four (8%) who died of metastatic UCC at a median of 14 (11–65) months after ONR. The 5-year pelvic recurrence-free (Fig. 1a), cancer-specific (Fig. 1b) and overall survival (Fig. 1c) rates for women after RC with ONR for cancer were 85.1%, 87.7% and 82.9%, respectively.
Table 3. Pelvic recurrences in women after RC with ONR
Pathology at cystectomy
Time to pelvic recurrence, months
Location of pelvic recurrence
Treatment of pelvic recurrence
AWOSR, alive without subsequent recurrence; SCC, small cell carcinoma.
Excision of neobladder, ileal conduit formation
Excision of neobladder, ileal conduit formation
Chemotherapy + radiation
Chemotherapy + resection of vaginal cuff mass
Left distal ureter
Excision of neobladder, ileal conduit formation
Died from metastatic SCC
Excluding the four patients who had conversion to an ileal conduit, information on the urinary functional status after surgery was available in 49 of 55 (89%) of the remaining women. At the last follow-up, complete daytime urinary continence (no pad requirement) was reported by 44/49 (90%) of these patients, while 28 (57%) remained continent at night. Two of the women with daytime incontinence had a periurethral injection with a bulking agent and thereafter reported an improvement in function. Meanwhile, intermittent self-catheterization was required by 17 (35%) women to facilitate emptying of the neobladder.
The use of ONR after RC in women has developed after clinical and pathological studies helped to establish the natural history of TCC in female patients, and further elucidate the neurovascular and fascial relationships of the female rhabdosphincter. Nevertheless, concerns remain about the oncological and functional outcomes of this relatively new approach to urinary diversion. Here, we report the safety and efficacy of ONR in a large series of women undergoing RC. Specifically, we report a perioperative complication rate of 22% among 59 consecutive women who had RC with ONR at our institution over a 10-year period. We also report an 88% 5-year cancer-specific survival among patients undergoing ONR for malignant disease. At a median follow-up of nearly 3 years after ONR, 90% of women maintained complete urinary continence during the daytime and 57% remained continent at night.
The results of the present series are consistent with several previous series that evaluated the outcomes of ONR in women. For example, despite early concerns of an increased risk of urethral recurrence with an intact female urethra, previous studies reported a urethral recurrence rate in women after RC of 1.4–12%[7–9], and in the present study, only one patient (2%) developed a urethral recurrence after ONR. These results are also comparable with the reported incidence of urethral recurrence in men after RC without urethrectomy in a long-term follow-up . However, the most accurate means of predicting urethral recurrence, and thereby selecting appropriate patients for ONR, continues to be debated. Although Stein et al. and Coloby et al. independently reported that carcinoma at the bladder neck is an important risk factor for urethral tumour involvement, as 53% and 100% of patients from each series, respectively, with urethral involvement had concomitant bladder neck involvement, nevertheless about half of patients with bladder neck involvement have no coexistent urethral disease . The ability of intraoperative frozen-section analysis to correctly identify the status of the proximal urethra was also evaluated. Indeed, when compared with the final pathology report, frozen sections were found in one study to correctly identify the disease status in all 29 specimens assessed . Thus, we currently require a negative distal urethral margin on intraoperative frozen-section analysis before proceeding with ONR in women (and men) undergoing RC.
In addition to providing durable oncological efficacy, preservation of urinary function is an important goal of ONR, and as such, documenting patients’ ability to maintain continence after surgery is a key measure of functional outcome. Historically, ONR was withheld as a choice of urinary diversion in women because it was thought that preserving the bladder neck and an adequate length of urethra were required for urinary continence. However, Colleselli et al. reassessed the anatomy of the female rhabdosphincter through cadaver dissection, computerized three-dimensional reconstructions of histological sections, and CT analyses of fetal specimens. These investigators found that the female rhabdosphincter lies in the caudal two-thirds of the female urethra, and that the branches from the pudendal nerve that supply the sphincter course deep to the endopelvic fascia and enter the urethra laterally . Therefore, appropriately modifying the dissection during RC to protect the sphincter and its nerve supply, e.g. by avoiding dissection distal to the bladder neck and by preserving the endopelvic fascia and supporting structures, might facilitate the preservation of urinary continence. We noted that 90% of women were completely continent after ONR during the day, which is consistent with the findings from previous series of ONR in women (82–100%) [5,6,19,20]. The present night-time continence rate of 57% was somewhat lower than the rates reported in separate studies, of 67–86%[5,6,19,20]. However, continence rates depend on the definition of urinary continence used, and the manner in which continence was determined, both of which have been highly variable across studies to date. For example, several previous investigators classified patients who use up to one pad per day as continent [6,20], whereas our definition of continence required patients to be completely dry with no use of pads.
Difficulty with neobladder emptying, also referred to as hypercontinence, can also be occur after ONR, and might require patients to use intermittent self-catheterization to prevent the sequelae of urinary retention, such as upper tract deterioration and infection. Indeed, hypercontinence has been noted more commonly among women than men after orthotopic diversion, for while Simon et al. reported that 11.5% of 655 male patients had at least one episode of neobladder emptying failure requiring therapy , various studies reported a need for intermittent catheterization in 9.5–40% of females after orthotopic urinary diversion [13,18,22–24]. The aetiology for the increased incidence of retention in women after ONR is probably multifactorial, and includes possible kinking of the urethrovesical angle in women due to descent of the neobladder within the pelvis after RC with concurrent hysterectomy. The present rate of self-catheterization among women after ONR (35%) is slightly higher than the published results to date (10–27%) [6,20,25], although we continue to investigate refinements in surgical technique, such as using a culdoplasty to provide additional support to the neobladder, and thus maintain the anatomical relationship between the neobladder and urethra, which we think will further improve functional outcomes. Nevertheless, we continue to provide preoperative counselling about the potential need for intermittent self-catheterization for all patients scheduled for ONR.
Our results also confirm the safety of ONR in women, as the perioperative complication rate of 22% is consistent with contemporary series, with reported complications in 16–61%[12,25] of patients after ONR, and is not significantly different from the reported incidence of complications after an ileal conduit [1,25]. Indeed, Nieuwenhuizen et al. recently compared the complication rates associated with ileal conduit, Indiana pouch, orthotopic diversion, and nerve-sparing cystectomy with ONR. Although patients who had a diversion with an ileal conduit had fewer late complications, most of the late complications in non-conduit patients were minor, and consisted primarily of UTIs . Thus, ONR does not appear to place patients at a greater risk of major complications than other forms of urinary diversion.
Indeed, additional technical modifications to the procedure were implemented in an attempt to further decrease the incidence of long-term complications. For example, whenever possible, we preserve the anterior vaginal wall during RC with planned ONR, as this technique has been described as a means of providing additional pelvic support to improve postoperative continence, and, by minimizing the number of overlapping suture lines, of potentially decreasing the rate of fistula between the neobladder and vagina [13,24]. Chang et al. reported only one fistula among 21 patients who had ONR with vaginal wall-sparing, in a patient who had an inadvertent vaginal wall injury that was repaired during surgery. Similarly, we preserved the anterior vaginal wall in 54/59 (92%) of the present patients, and there was a 4% (two/54) rate of neobladder-vaginal fistulae in these patients, compared to on of five who had resection of the anterior vaginal wall.
We recognize that our study is limited by its retrospective, single-institution design. In addition, we acknowledge that functional status was obtained by patients’ response to a direct physician inquiry, and not using a validated questionnaire. Nevertheless, as ONR has become an increasingly popular method of urinary diversion in patients undergoing RC, due to the quality-of-life advantages afforded by orthotopic diversion, we think that, with the safety and long-term efficacy of the procedure shown here, ONR should remain the preferred option for reconstruction after RC in women with a urethral margin free of tumour on intraoperative frozen-section analysis.