After radiotherapy strictures, anastomotic repair had a third of the recurrence risk

Design

two-centre retrospective cohort with Cox regression, 2001 to 2024

Population

212 men with radiation-induced bulbomembranous urethral stenosis; 163 anastomotic urethroplasty, 49 buccal mucosa graft

Primary outcome

stricture recurrence, defined as stricture under 16Fr confirmed on cystoscopy

Effect

buccal graft HR 3.43 (95% CI 1.03 to 11.47, P = .046); 10-year recurrence 20% vs 6.8% (log-rank P = .01)

Two centres reviewed 212 men undergoing posterior urethroplasty for bulbomembranous urethral stenosis after prostate cancer radiotherapy between 2001 and 2024 — 163 anastomotic urethroplasty, 49 buccal mucosa graft onlay. Median stenosis length was 2.0 cm; aetiology was brachytherapy in 46 per cent, external beam in 45 per cent and combination in 9 per cent. Median follow-up in men without recurrence was 97 months.

On Cox regression, buccal graft urethroplasty was independently associated with recurrence (HR 3.43, 95% CI 1.03 to 11.47, P = .046). Age, stenosis length, diabetes, smoking, obesity and prior endoscopic treatment were not. Cumulative recurrence at 1, 2 and 10 years was 17, 18 and 20 per cent after buccal graft against 3.3, 5.0 and 6.8 per cent after anastomotic repair (log-rank P = .01). The techniques did not differ in 90-day complications, patient satisfaction, de novo erectile dysfunction or incontinence.

The recurrence curves tell you something useful beyond the comparison: almost all the buccal graft failures happened in the first year, then the curve flattened. That is the behaviour of a graft failing to take in irradiated tissue, which is the mechanistic reason to prefer excision and anastomosis where the stenosis is short enough to allow it.

• Prefer anastomotic urethroplasty for radiation-induced bulbomembranous stenosis where technically feasible
• Note that incontinence was 18 per cent with either technique — counsel for it regardless
• Watch buccal graft repairs closely in the first year, where almost all failures occurred
• Do not use stenosis length alone to justify a graft; length was not an independent predictor of recurrence
• Discuss erectile dysfunction risk as similar between techniques, not as a reason to choose one

Why it matters

It gives a technique preference in a reconstruction where the choice has been down to surgeon habit.

Don't overread it

Retrospective and non-randomised — technique was selected by feasibility, so the comparison carries confounding by indication.

The statistics, in plain English

A hazard ratio of 3.43 with a confidence interval from 1.03 to 11.47 only just excludes 1, and P = .046 sits at the edge of conventional significance — with 49 patients in the graft arm, the size of the difference is uncertain even though its direction is supported by the separated survival curves. This was not randomised: surgeons chose anastomosis where anatomy allowed it, so the graft group may have had harder strictures.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.