OPEN: Open Urethroplasty versus Endoscopic Urethrotomy for Recurrent Bulbar Urethral Stricture
OPEN was an open-label, patient-randomized superiority trial of open urethroplasty versus endoscopic urethrotomy in 222 men with recurrent bulbar urethral stricture at 38 UK NHS hospitals, with the 24-month voiding symptom score as its primary endpoint. Symptom scores did not differ (adjusted mean difference −0.36, 95% CI −1.74 to 1.02), while 15 men allocated to urethroplasty and 29 allocated to urethrotomy needed a reintervention (hazard ratio 0.52, 95% CI 0.31–0.89).[1]
- Design and population
- Multicenter, open-label, patient-randomized superiority trial. 222 men with bulbar urethral stricture and at least one previous stricture operation randomized at 38 UK NHS hospitals; two ineligible men were excluded after randomization, leaving 108 and 112 by allocation.
- Arms
- Open urethroplasty versus endoscopic urethrotomy
- Report and follow-up
- 2020 · Symptom score profile over 24 months after randomization; time to reintervention to database closure (at least 24 and up to 48 months)
| Endpoint | Results | Effect | Finding |
|---|---|---|---|
| Area-under-the-curve profile of the 0-24 patient-reported voiding symptom score over 24 months (lower is better); adjusted mean difference, urethroplasty minus urethrotomy | 7.4 (SD 3.8), n=69 vs 7.8 (SD 4.2), n=90 | difference -0.36; CI -1.74–1.02; P 0.6 | Not met |
| Time to first reintervention (hazard ratio, urethroplasty versus urethrotomy); men with at least one reintervention | 15 men (of 93 with data available; 108 allocated) vs 29 men (of 104 with data available; 112 allocated) | HR 0.52; CI 0.31–0.89; P 0.017 | Met |
Harms: Any adverse event (by procedure received): 50/82 (61.0%) vs 30/115 (26.1%); Perioperative mouth pain: 12/82 (14.6%) vs 2/115 (1.7%); Perioperative wound infection: 4/82 (4.9%) vs 0/115
ISRCTN98009168 · Report DOI
Design
Eligible men were adults with bulbar urethral stricture who had already had at least one surgical intervention for it. Current perineal sepsis and a urethrocutaneous fistula were exclusions. Men were identified at 53 NHS providers (38 randomized at least one man); 1,262 were identified and 222 were randomized between February 27, 2013 and December 23, 2015. Two men found ineligible after randomization were excluded, leaving 108 allocated to urethroplasty and 112 to urethrotomy.[1]
At baseline, mean stricture length was 2.0 cm (SD 1.4) in the urethroplasty group and 1.7 cm (SD 1.1) in the urethrotomy group, but length was recorded in only 67 and 63 men. Men had had a mean of 1.9 and 1.8 previous interventions of any type, mostly urethrotomy (1.6 and 1.4); prior urethroplasty averaged 0.1 in each group. The last procedure was 12 months or more earlier in 67% and 68%, the cause was unknown in 70% and 72%, and 21% and 13% were currently using intermittent self-dilatation. Mean baseline maximum flow rate was 10.0 and 9.7 mL/s and mean baseline voiding score 13.5 and 13.2.[1]
Allocation was 1:1 through a central automated system with minimization on recruiting site and time since the last procedure (under or over 12 months). Participants and surgeons knew the allocation; central trial staff were masked. The primary outcome was the area under the curve of the patient-reported voiding symptom score, which sums six items (delay before starting, poor stream strength, straining, intermittent stream, incomplete emptying, post-micturition dribbling) each scored 0 to 4, for a total of 0 to 24. Questionnaires were sent at baseline, before the operation, at 3, 6, 9, 12 and 24 months after it, and at 18 and 24 months after randomization. The primary analysis required a baseline score, an early score (up to 12 months after the operation) and a later score (18 or 24 months after randomization), and analyzed men as randomized.[1]
The investigators set the minimum clinically important difference at more than 10% of the symptom profile, with 90% power. After recruitment proved slow, the required sample was revised to 170 men, and the target of 210 allowed for 19% attrition. The trial was also powered (104 men, 90% power) to detect a 30% relative reduction in reintervention. A reintervention was any procedure after the allocated one, excluding self-dilatation. Recurrence was a reintervention performed or scheduled, or deterioration of the maximum flow rate or voiding score to its pre-intervention value. Improvement in maximum flow rate was an increase of at least 10 mL/s.[1]
Results
The primary analysis included 69 of 108 men allocated to urethroplasty and 90 of 112 allocated to urethrotomy. The mean voiding score profile over 24 months was 7.4 (SD 3.8) after urethroplasty and 7.8 (SD 4.2) after urethrotomy; p = 0.6.[1] Multiple-imputation intention-to-treat analysis gave a difference of −0.33 (95% CI −1.74 to 1.09) and a modified intention-to-treat analysis −0.52 (95% CI −2.0 to 0.96).[1] Subgroups by time since last procedure, age, stricture length and number of previous interventions showed no heterogeneity. The full monograph reports mean scores falling from 13.4 to 6 with urethroplasty and from 13.2 to 6.4 with urethrotomy between baseline and 24 months.[1][2]
Forty-four men had at least one reintervention, and there were 52 reinterventions overall. Among men who had one, the median interval from surgery was 474 days (IQR 399–577) after urethroplasty and 308 days (IQR 211–448) after urethrotomy. Counting multiple events per man gave a hazard ratio of 0.49 (95% CI 0.30–0.82; p = 0.006), and a per-protocol analysis of men who received the allocated procedure gave 0.28 (95% CI 0.15–0.55; p < 0.001). Among men who had surgery (93 in the urethroplasty arm and 104 in the urethrotomy arm), recurrence occurred in 19 and 39, respectively (hazard ratio 0.46, 95% CI 0.29–0.72; p = 0.001).[1][2]
An improvement in maximum flow rate of at least 10 mL/s at 12 or 24 months was reported in 18 of 93 men (19%) in the urethroplasty arm and 13 of 104 (13%) in the urethrotomy arm; among the 44 and 63 men with flow-rate data, respectively, the adjusted odds ratio was 2.64 (95% CI 1.14–6.15; p = 0.024). At 3 months the odds ratio was 2.1 (95% CI 1.05–4.12).[1][2] Impact of symptoms on daily activity (difference 0.06, 95% CI −0.19 to 0.30) and satisfaction with sexual function (difference 0.35, 95% CI −0.06 to 0.75; p = 0.090) did not differ.[1]
In the base-case cost-effectiveness analysis of 46 men per group, urethroplasty cost £2,148 more per patient over 24 months (95% CI £689–£3,606) with a QALY difference of −0.01 (95% CI −0.17 to 0.14). In a 10-year Markov model the base case cost £8,026 and 7.61 QALYs for urethroplasty versus £6,553 and 7.60 QALYs for urethrotomy, an incremental cost of £301,073 per QALY.[3]
Surgery and treatment delivery
Treatment received departed from allocation. In the urethroplasty group, 71 of 108 men (66%) had a urethroplasty, 22 (20%) had a urethrotomy or dilatation instead and 15 (14%) had no procedure. In the urethrotomy group, 93 of 112 (83%) had urethrotomy or dilatation, 11 (9.8%) had a urethroplasty and 8 (7.1%) had no procedure.[1]
Of the 71 urethroplasties in the urethroplasty group, 45 were dorsal graft, 9 anastomotic without transection, 9 anastomotic with transection, 7 ventral graft and 1 was a perineal urethrostomy. The 11 crossovers to urethroplasty comprised 4 dorsal graft, 4 anastomotic without transection, 2 ventral graft and 1 anastomotic with transection. Endoscopic treatment was optical cold-knife urethrotomy in 105 of the 115 men who received it (15 in the urethroplasty group, 90 in the urethrotomy group), hot-knife urethrotomy in 1 and dilatation in 9 (7 and 2).[1]
Intermittent self-dilatation was started after allocation in 29 of 112 men (26%) in the urethrotomy group and 9 of 108 (8.3%) in the urethroplasty group; all 9 had crossed over to urethrotomy.[1]
Adverse events
Adverse events are reported by procedure received (82 men had urethroplasty and 115 urethrotomy). At least one adverse event occurred in 50 of 82 men (61.0%) after urethroplasty and 30 of 115 (26.1%) after urethrotomy. Perioperative events after urethroplasty were mouth pain in 12 (14.6%, graft harvest site), wound infection in 4 (4.9%) and urinary infection in 3 (3.7%); after urethrotomy mouth pain occurred in 2 (1.7%), with no wound or urinary infection. During follow-up, erectile dysfunction was recorded in 4 of 82 (4.9%) and 3 of 115 (2.6%), wound pain in 5 (6.1%) and 1 (0.9%), and urinary tract infection in 5 (6.1%) and 6 (5.2%).[1]
At least one serious adverse event occurred in 7 of 82 men (8.5%) after urethroplasty and 10 of 115 (8.7%) after urethrotomy; two of the 22 serious events were related to the intervention.[1] Two urethrotomy patients were readmitted (four readmissions in total), and one urethrotomy patient died of deep vein thrombosis and pulmonary embolism, reported as unrelated to the intervention. The monograph reports two postoperative complications requiring reintervention in the urethroplasty group and five in the urethrotomy group.[1][2]
Limitations
The symptom analysis excluded 39 of 108 men allocated to urethroplasty and 22 of 112 allocated to urethrotomy, and the reasons for the unequal loss are unknown; the investigators suggest that more men received their allocated treatment in the urethrotomy group, or the shorter wait for urethrotomy. Of those excluded, 15 and 8 had no surgery at all. Sensitivity analyses with multiple imputation and pattern-mixture models were robust except under extreme assumptions. The 159 men analyzed were fewer than the 170 required after the sample-size revision.[1]
Participants and surgeons were unblinded, 37 of 108 men allocated to urethroplasty (34%) and 19 of 112 allocated to urethrotomy (17%) received a different procedure or none, and stricture length was missing in 90 of 220 men.[1] Symptoms and costs were assessed over 24 months, whereas reinterventions were tracked to database closure, at least 24 and up to 48 months. The cost-effectiveness analysis did not include the loss of quality of life during recovery from later procedures.[1][3]
In conversation
- Parallel to ROBUST III (2021). Both trials used endoscopic treatment as the comparator in recurrent strictures; ROBUST III tested a paclitaxel-coated balloon, OPEN tested urethroplasty.
- Parallel to ROBUST III (2021). The AUA panel read both randomized endoscopic-management trials together; OPEN used a symptom outcome and compared urethroplasty with urethrotomy.
ROBUST III, which the AUA guideline discusses together with OPEN, tested a paclitaxel-coated balloon with dilation or urethrotomy against dilation or urethrotomy alone, and used urethral patency and freedom from retreatment as endpoints. OPEN's experimental arm was urethroplasty and its primary endpoint was patient-reported symptoms.[4]
Guidelines and approvals
- AUA, Urethral Stricture Disease guideline (published 2016, amended 2023). Surgeons should offer urethroplasty instead of repeated endoscopic management for recurrent anterior urethral strictures after failed dilation or direct visual internal urethrotomy . The supporting discussion cites OPEN's reintervention and flow-rate results and describes the trial as a pragmatic comparison whose symptom endpoints did not differ.[4]
- EAU, Urethral Strictures limited update (2026). The section on direct vision internal urethrotomy for recurrent strictures reviews OPEN. The guideline recommends not performing repetitive (more than two) direct vision internal urethrotomies or dilatations if urethroplasty is a viable option .[5]
Report history
| Published | Report | Cutoff | Headline results | Citation |
|---|---|---|---|---|
| 2020-07-04 | Primary report · primary | — | Area-under-the-curve profile of the 0-24 patient-reported voiding symptom score over 24 months (lower is better); adjusted mean difference, urethroplasty minus urethrotomy: 7.4 (SD 3.8), n=69 vs 7.8 (SD 4.2), n=90; difference -0.36; CI -1.74–1.02; P 0.6 Not met Time to first reintervention (hazard ratio, urethroplasty versus urethrotomy); men with at least one reintervention: 15 men (of 93 with data available; 108 allocated) vs 29 men (of 104 with data available; 112 allocated); HR 0.52; CI 0.31–0.89; P 0.017 Met Multiple reintervention events per participant (Andersen-Gill hazard ratio): 44 men had at least one reintervention and there were 52 reinterventions overall.; HR 0.49; CI 0.3–0.82; P 0.006 Time to first reintervention, per-protocol analysis (men who received the allocated procedure): Secondary analysis restricted to men who underwent the allocated procedure.; HR 0.28; CI 0.15–0.55; P <0.001 Time to recurrence (hazard ratio, urethroplasty versus urethrotomy); men with recurrence: 19 men vs 39 men; HR 0.46; CI 0.29–0.72; P 0.001 Improvement in Qmax of at least 10 mL/s from baseline at 12 or 24 months (odds ratio, urethroplasty versus urethrotomy): 18/44 with data (data missing for 49 of 93) vs 13/63 with data (data missing for 41 of 104); OR 2.64; CI 1.14–6.15; P 0.024 Profile score over 24 months (0 none to 3 a lot); adjusted mean difference, urethroplasty minus urethrotomy: 1.1 (SD 0.8), n=69 vs 1.0 (SD 0.7), n=90; difference 0.06; CI -0.19–0.3; P 0.6 Profile score over 24 months; adjusted mean difference, urethroplasty minus urethrotomy: 2.9 (SD 1.2), n=63 vs 2.5 (SD 1.2), n=87; difference 0.35; CI -0.06–0.75; P 0.090 | Goulao B, Carnell S, Shen J, et al. "Surgical Treatment for Recurrent Bulbar Urethral Stricture: A Randomised Open-label Superiority Trial of Open Urethroplasty Versus Endoscopic Urethrotomy (the OPEN Trial)." *Eur Urol.* 2020;78(4):572-580. doi:10.1016/j.eururo.2020.06.003 DOI |
| 2020 | NIHR HTA monograph · secondary | — | Mean difference in area under the curve of repeated 0-24 voiding symptom scores: The monograph abstract prints -1.78 as the lower confidence limit; its results text and tables and the journal report print -1.74. Primary analysis included 69 (63%) of 109 and 90 (80%) of 113 men allocated.; difference -0.36; CI -1.78–1.02; P 0.6 Mean (SD) 0-24 voiding symptom score at baseline and 24 months after randomization: 13.4 (4.5) to 6 (5.5) vs 13.2 (4.7) to 6.4 (5.3); — Descriptive Time to reintervention (hazard ratio, urethroplasty versus urethrotomy): —; HR 0.52; CI 0.31–0.89; P 0.02 | Pickard R, Goulao B, Carnell S, et al. "Open urethroplasty versus endoscopic urethrotomy for recurrent urethral stricture in men: the OPEN RCT." *Health Technol Assess.* 2020;24(61):1-110. doi:10.3310/hta24610 DOI |
| 2021-05-03 | Cost-effectiveness analysis · secondary | — | Base-case mean total cost in pounds sterling (2017 prices) at 24 months; difference, urethroplasty minus urethrotomy: 4869 vs 2721; difference 2148; CI 689–3606 Base-case mean QALYs at 24 months after randomization (EQ-5D-5L); difference, urethroplasty minus urethrotomy: 1.74 vs 1.75; difference -0.01; CI -0.17–0.14 10-year Markov model base case: total cost in pounds sterling and QALYs per patient: 8026; 7.61 QALYs vs 6553; 7.60 QALYs; — | Shen J, Vale L, Goulao B, et al. "Endoscopic urethrotomy versus open urethroplasty for men with bulbar urethral stricture: the OPEN randomised trial cost-effectiveness analysis." *BMC Urol.* 2021;21(1):76. doi:10.1186/s12894-021-00836-1 DOI |
References
1. Goulao B, Carnell S, Shen J, et al. "Surgical Treatment for Recurrent Bulbar Urethral Stricture: A Randomised Open-label Superiority Trial of Open Urethroplasty Versus Endoscopic Urethrotomy (the OPEN Trial)." Eur Urol. 2020;78(4):572-580. doi:10.1016/j.eururo.2020.06.003
2. Pickard R, Goulao B, Carnell S, et al. "Open urethroplasty versus endoscopic urethrotomy for recurrent urethral stricture in men: the OPEN RCT." Health Technol Assess. 2020;24(61):1-110. doi:10.3310/hta24610
3. Shen J, Vale L, Goulao B, et al. "Endoscopic urethrotomy versus open urethroplasty for men with bulbar urethral stricture: the OPEN randomised trial cost-effectiveness analysis." BMC Urol. 2021;21(1):76. doi:10.1186/s12894-021-00836-1
4. Wessells H, Morey A, Souter L, Rahimi L, Vanni A. "Urethral Stricture Disease Guideline Amendment." J Urol. 2023;210(1):64-71. doi:10.1097/JU.0000000000003482
5. Lumen N, Campos-Juanatey F, Dimitropoulos K, et al. EAU Guidelines on Urethral Strictures. European Association of Urology; 2026.