Radical cystectomy (RC) is the first-line surgical treatment for nonmetastatic, muscle-invasive bladder cancer and select cases of nonmuscle-invasive bladder cancer.^1-4 Urinary diversion (UD) is required after RC and selecting the best diversion involves extensive patient counseling. Options for UD include continent cutaneous diversions, orthotopic neobladders, bowel conduits, and cutaneous ureterostomy (CU). After being popularized by Bricker in the 1950s, the ileal conduit (IC) has since become the most performed UD in multiple countries and comprises >80% of UD after RC in the United States.^5,6 The CU has offered a simple alternative that circumvents the bowel-related complication of diversion with IC. However, widespread adoption of CU has been limited by the ureteral stenosis rates that have been reported at upwards of 70%.⁷ In our study we sought to present our center’s experience of patients with bladder cancer undergoing RC+CU compared to RC+IC to assess for differences in perioperative outcomes.

We conducted a retrospective, cohort study on our patients who underwent open or robotic RC with CU or IC, as performed by a single surgeon (M.T.) from 2020 to 2022. The included patients required a minimum of 3 months follow-up. Our primary outcomes were operative time and length-of-stay. Secondary outcomes were 30- and 90-day postoperative Clavien-Dindo (C.D.) complications, 30- and 90-day readmissions, and ureteral stent outcomes. These quantitative and categorical outcomes were compared with Mann-Whitney U or Fischer’s exact tests, respectively. The likelihood of any urinary drainage tube replacement, including stent, was represented by “replacement-free survival,” and assessed with Kaplan-Meier survival analyses with log-rank test. Single-stoma CU was performed using a modified Ariyoshi technique illustrated in Figure 1 to create a ureteral nipple.⁸ IC was performed using standard techniques with a Wallace-type ureteroenteric anastomosis.^8,9 Flexible, 7F, single-J ureteral stents were placed into each renal pelvis per the CU or IC stoma. All patients were evaluated for stent removal at approximately 2 weeks and then 3 months, postoperatively. No postoperative prophylactic antibiotics were given while stents were in place, except for the day of removal.

We identified 53 patients who received RC, of which UD was performed with CU or IC in 53% and 47% of cases, respectively (Table 1). The groups were similar regarding median age (CU, 73 years and IC 70 years), and were predominantly male sex (CU, 75% and IC, 80%). There were more patients with BMI >30 kg/m² undergoing IC over CU (42% and 14%, respectively, P = .03) and similar proportions of those with BMI <30 kg/m² in both groups. The median operative times for open RC+UD were 1.7 hours shorter in those receiving CU over IC (3.6 and 5.3 hours, respectively, P < .001), because of the avoidance of bowel manipulation necessary in IC. Postoperatively, the median length of stay was 1 day shorter for CU compared to IC patients (3 days vs 4 days, respectively, P = .03). Both groups had similar 30- and 90-day major complication rates (C.D. 3 to 5) and readmissions, as shown in Table 2. Ureteral stent(s) were removed sooner in the IC group compared to CU (22 days and 34 days, respectively, P = .002) with similar tube-free rates at last follow-up (84% and 83%, respectively, P > .99). As outlined in Table 3, only 11% of CU patients developed ureteral strictures that necessitated stent replacement. Lastly, illustrated in Figure 2, replacement-free survival was similar among the groups at 6 months postoperative (CU 78.6% and IC 77.1%, P = .36) and 12 months postoperative (CU 78.6% and IC 70.1%, P = .36).

Table 1. Preoperative Demographics and Intraoperative Details of Patients Undergoing Radical Cystectomy + Urinary Diversion

Table 2. Postoperative Outcomes of Patients Undergoing Radical Cystectomy + Urinary Diversion

Table 3. Ureteral Stent Outcomes After Radical Cystectomy With Cutaneous Ureterostomy or Ileal Conduit

Our study suggests that RC with single-stoma CU is a safe and feasible UD that compares favorably to IC at intermediate-term follow-up. CU has the advantages of decreased operating room time and length of stay that may expand the indications for RC to include more comorbid patients. The findings of decreased length of stay may be expected in the absence of bowel-related reconstruction, which also contributed to faster operative times. Interestingly, rates of 90-day bowel-related complications including paralytic ileus requiring nasogastric tube were similar between the CU and IC groups (16% and 26%, respectively, P = .50), indicating that intraoperative bowel manipulation rather than bowel excision and reconstruction may be responsible for these complications. Lastly, we reported a low 11% ureteral stenosis rate in our CU patients, drastically lower than reported in previous literature, and this compared similarly to the 12% ureteroenteric stricture rate in the IC patients.⁷ As our RC data mature, we look forward to comparisons with larger cohorts with long-term follow-up to better understand long-term complication rates, changes in renal function, and drainage tube dependence in patients undergoing CU or IC. In conclusion, we recommend that bladder cancer patients pursuing RC should be counseled on single-stoma CU as a viable UD option alongside IC.