HBOT Is the Only Randomised Option for Radiation-Damaged Bladders. The Trial Drew a Line First.

A dark grid of small frosted tiles representing 69 studies, exactly two of them lit warm amber against deep navy, one circled inside a soft light pool

Late urinary damage is the part of pelvic radiotherapy nobody puts on a survivorship leaflet. It arrives months or years after treatment ends, ranges from a nuisance to a haemorrhage, and lands hardest on prostate and gynaecological cancer survivors who were told the hard part was behind them. A scoping review published in Clinical and Translational Radiation Oncology has now mapped how the field treats it, across 69 studies of previously irradiated patients, and the map contains one finding that will interest anyone who follows hyperbaric medicine: in a literature almost entirely built on retrospective case series, the single intervention with randomised evidence and five-year follow-up is hyperbaric oxygen therapy.

That is the headline, and it is a real one. It is also, on a careful reading of the same review, a statement about a narrow slice of a wide problem.

What sixty-nine studies actually contain

The authors, L. Mullaney and S. Bennani, searched PubMed and Embase for primary studies published between January 2015 and April 2026 and included 69 studies reported across 70 publications. They sorted the field into seven urinary syndromes — radiation cystitis and haematuria, irritative lower urinary tract symptoms, stress incontinence, urethral stricture, ureteral stricture, fistula with pubic osteomyelitis, and end-stage urinary injury — and three management pathways: bladder-preserving, reconstructive, and salvage.

The evidence behind that taxonomy is thin in a quotable way. Of the 69 studies, 58 are retrospective and nine are prospective cohorts. Exactly two are randomised controlled trials. One is a trial of intravesical sodium hyaluronate with nine-month follow-up. The other is RICH-ART, the Nordic hyperbaric oxygen trial, which the review describes as having the strongest evidence of any intervention for radiation cystitis, “supported by randomised data and long-term follow-up.”

Reporting is patchy: radiation modality appeared in 56 per cent of the included publications, total dose in 33 per cent, and the interval from radiotherapy to toxicity in 70 per cent.

Against that background, the alternatives to hyperbaric oxygen look less like competitors than like improvisations. GreenLight laser photocoagulation prevented recurrent haematuria requiring further surgery in 82.7 per cent of patients, but disabling overactive bladder symptoms developed in 31 per cent and two patients went on to cystectomy. Fibrin glue and formalin rest on single-arm series of 20 and 8 patients; a cystoscopically applied self-assembling peptide gel reduced haematuria in 93 per cent of 15 men at six weeks, while 27 per cent had a Clavien grade 3 or higher complication within 90 days.

The trial, in the numbers

RICH-ART ran at five Nordic university hospitals and randomised 87 patients between May 2012 and December 2017. Eligibility required pelvic radiotherapy completed at least six months earlier, a urinary domain score below 80 on the Expanded Prostate Cancer Index Composite, and referral to a participating hyperbaric clinic for symptoms of late radiation cystitis.

Eight patients withdrew consent immediately after randomisation, seven of them in the standard-care arm, leaving 79 in the intention-to-treat analysis. Patients received 30 to 40 daily sessions of 100 per cent oxygen at 240 to 250 kPa for 80 to 90 minutes, or standard care with no restrictions on other treatment. Nobody was masked.

The primary outcome was the change in the EPIC urinary total score at six to eight months. It improved by 17.8 points with hyperbaric oxygen and 7.7 points with standard care, a between-group difference of 10.1 points (95 per cent CI 2.2 to 18.1; p=0.013). The scoping review, summarising the trial’s secondary findings, notes that the bowel score improved by 13.2 versus 4.9 points and that 64 per cent of treated patients had reduced cystoscopic injury on the RTOG/EORTC late morbidity scale against 18 per cent of controls (p=0.0012). Seventeen of the 41 treated patients, or 41 per cent, had transient grade 1 to 2 adverse events during therapy, related to sight and hearing.

A 10-point improvement on a symptom questionnaire is not a cure, and the trial’s authors did not claim one. They claimed relief of symptoms, tolerability and safety, where the previous standard was a catheter, a transfusion or an operating theatre.

Who the trial could not answer for

The exclusion criteria are where this story turns, because they name the patients who are hardest to help. RICH-ART excluded anyone with ongoing bleeding requiring more than 500 mL of transfusion in the previous four weeks, anyone with a permanent urinary catheter, anyone whose bladder capacity had fallen below 100 mL, and anyone with a fistula in the bladder. Prior hyperbaric treatment for late radiation injury was also excluded, as were the usual contraindications.

Those four exclusions describe, almost item by item, the salvage end of the review’s own map. Irradiated vesicovaginal fistulae closed in only 25 per cent of cases in the largest series the review identified, with 70 per cent requiring diversion. Refractory bleeding and a devastated outlet are the indications for cystectomy or urinary diversion, operations with substantial perioperative morbidity that nonetheless produce some of the strongest patient-reported gains in the field. Contraction below 100 mL is exactly the bladder that has stopped being a bladder.

So the randomised evidence for hyperbaric oxygen sits at the moderate end of a spectrum, in patients with chronic symptoms and a bladder still worth preserving. It is a legitimate place to sit, but not the place where a clinician facing a bleeding, fistulised, fibrotic pelvis can point to a trial and say the question is settled.

The review makes the adjacent point about timing without dressing it up. Earlier treatment after a first haematuria episode was associated with better response and less recurrence in the observational cohorts, and patients who achieved a complete response had significantly longer haematuria-free survival than partial responders, 71.4 versus 36.4 months (p<0.001). The trial itself gives no answer on when to start, because it did not test timing.

Five years out, with no control group left

The long-term report, published in eClinicalMedicine in 2025, followed the 70 patients who received hyperbaric oxygen — including those randomised to standard care who were offered the treatment once the primary endpoint was reached. Mean EPIC urinary total rose from 46.6 to 64.6 at six months, an improvement of 18.0 points (95 per cent CI 14.2 to 21.8), and held at 19.1 points (95 per cent CI 13.3 to 24.9) at five years. Responders, defined as an improvement of at least nine points, were 48 of the 70, or 68.6 per cent, and maintained a mean gain of 22.9 points. The 22 non-responders, 31.4 per cent, showed no early benefit and no late one. Nine patients, 12.8 per cent, needed further hyperbaric courses for recurrent symptoms. Follow-up was terminated six months early in May 2022, the authors state, mainly for lack of funding.

Durability at five years in a condition that otherwise tends to drift downward is a genuinely good result. It is also, unavoidably, a single-group result: after crossover there is no randomised comparison left, so the five-year number describes the trajectory of treated patients, not the size of an effect.

The pressure question is still retrospective

Two more pieces of the puzzle come from cohort data, and both cut in usefully awkward directions. A Spanish series of 105 patients treated at 2.4 ATA for 80 minutes, median 40 sessions, reported haematuria control in 92.4 per cent and recurrence in 24.7 per cent over a median 63 months, with no serious complications and better responses when treatment came sooner after the first bleeding episode. Patient-reported follow-up of 54 patients found symptom severity and anxiety improved and transfusions fell, while 39 per cent still had intermittent haematuria and 26 per cent needed further treatment.

Then there is dose. A 2026 two-site retrospective comparison of 93 patients, published in Neurourology and Urodynamics, found fewer patients treated at 2.5 ATA had gross haematuria within a year of therapy than those treated at 2.0 ATA. Time to recurrence, however, was indistinguishable at 10.2 versus 9.6 months, no other urinary symptom differed, and adverse events were higher at 2.5 ATA in a mixed-effects model. The authors are explicit that this supports an association, not a causal claim.

The review adds a further clinical modifier that has nothing to do with the chamber. Three of the included cohorts found anticoagulant or antiplatelet therapy independently associated with recurrence or repeat intervention, with hazard ratios from 1.8 to 3.3.

Where the regulators and the payers stopped

Both of the documents that govern American practice cover this territory without ever naming the bladder. The Undersea and Hyperbaric Medical Society’s 15th-edition indications list carries entry 11, “Delayed Radiation Injury (Soft Tissue and Bony Necrosis),” among its 15 indications. The word “cystitis” does not appear anywhere on that page; nor does it appear in CMS National Coverage Determination 20.29, which covers “soft tissue radionecrosis as an adjunct to conventional treatment” and “osteoradionecrosis as an adjunct to conventional treatment” without listing a single organ syndrome. Coverage runs through a tissue category; clinicians translate from there.

The society’s own definitional language is stricter than the wellness market assumes. It describes hyperbaric oxygen as physician-prescribed medical-grade oxygen at not less than 2.0 ATA in a chamber built to ASME PVHO-1 and NFPA 99, typically for 90 to 120 minutes per treatment. Its section on unproven hyperbaric treatment — the low-pressure chambers sold in wellness centres and spas — warns that such facilities often operate without fire-safety and chamber-construction standards, “putting those exposed at risk for serious injury and even death.”

The Food and Drug Administration’s contribution is a different kind of caution. Its August 2025 letter to health care providers classifies hyperbaric chambers as Class II devices cleared through the 510(k) pathway under product code CBF, and states plainly that the agency is aware of “fires that occurred with HBOT devices that resulted in serious injuries and deaths,” and that “the root cause of these events is not known.” The letter also says serious adverse events are believed to be rare. It contains no discussion of indications at all. The regulatory boundary for radiation injury is set by clearance and by coverage, and the safety boundary by the fire record.

What the cautious reading looks like

The 2023 Cochrane review of hyperbaric oxygen for late radiation tissue injury is the more disciplined counterweight. Across 18 studies and 1,071 participants published between 1985 and 2022, complete resolution or significant improvement of late radiation injury reached a risk ratio of 1.39 (95 per cent CI 1.02 to 1.89) on five trials and 468 participants, with substantial heterogeneity and low certainty. Death at one year was unaffected (RR 0.93, 0.47 to 1.83, low certainty). On harms, the review is unusually precise: reduced visual acuity had a risk ratio of 4.03 (1.65 to 9.84) on high-certainty evidence, and ear barotrauma a risk ratio of 9.08 when no sham pressurisation was used, falling to 1.07 when it was.

Read together, the two syntheses are not in contradiction. One says that within late pelvic urinary injury, hyperbaric oxygen is the best-evidenced option available. The other says that the best-evidenced option in this field still rests on small trials of low certainty, with real and measurable harms, and that nobody knows the right dose, the right duration or the right patient. Both can be true of a therapy that works modestly, safely and without a decent denominator.

For patients, the practical content sits in the review’s closing argument rather than in its evidence table. Management should follow the syndrome, the tissue quality and the cumulative burden of treatment rather than a fixed ladder of escalation, and repeated low-yield intervention is not obviously kinder than a definitive operation. The review found no study comparing early with delayed surgery, so it does not claim to know the right moment for a referral. It simply notes that the question has not been asked — which is also true, RICH-ART’s exclusion list suggests, of hyperbaric oxygen’s hardest cases.

HBOToday Editorial reports on peer-reviewed research and regulatory positions. This article is not medical advice; treatment decisions belong with qualified clinicians. Hyperbaric oxygen therapy is a prescription therapy with cleared indications, meaningful risks and a documented fire-safety record, and most wellness and longevity uses are not FDA-cleared.

Sources

← Back to blog