Gastroenterology · Medical Literature
Hyperbaric oxygen therapy alters bowel perfusion and improves outcomes in patients with treatment-refractory ulcerative colitis: a prospective pilot trial.
Abstract
Hyperbaric oxygen therapy (HBOT) delivers 100% oxygen in a pressurized chamber and enhances tissue oxygenation and neovascularization. While effective in radiation proctitis, acute severe ulcerative colitis, and perianal fistulizing Crohn's disease, its role in refractory ulcerative colitis (UC) remains underexplored. PARADOX was a prospective, open-label, phase 2a pilot trial in biologic-experienced patients with moderate-to-severe refractory UC (total Mayo > 5, Mayo endoscopic subscore (MES) ≥ 2, failure of ≥ 2 advanced therapies). Patients received 10 or 20 daily HBOT sessions (2.4 atmospheres absolute, 120 minutes/day) while continuing stable background therapy. The primary endpoint was composite clinical and endoscopic response at week 12 (≥ 3-point and 30% Mayo score reduction, rectal bleeding score = 0, and ≥ 1-point decrease in MES). Secondary outcomes included symptomatic, biochemical, and transmural response by intestinal ultrasound (IUS), including perfusion metrics via contrast-enhanced ultrasound (CEUS). In total, 16 patients (8 per group) were included in this pilot study. At week 12, composite response was achieved in 2/8 (10-session) and 4/8 (20-session) patients. Improvements in clinical, endoscopic, and IUS parameters were observed in both groups. Clinical responders showed increased CEUS perfusion at week 12 (peak enhancement Δ12.1 dB), while non-responders declined (Δ -5.4 dB). This pattern was consistent across wash-in/wash-out metrics, supporting a volume perfusion-based response. HBOT was well tolerated, with no serious adverse events or treatment discontinuations. HBOT is a well-tolerated adjunctive therapy in refractory UC, which improved clinical outcomes and perfusion kinetics. These findings support further evaluation in a randomized, dose-optimized trial.Euclinicaltrials.eu, EU CT number 2024-515278-28-00.
Clinical relevance
In biologic-experienced patients with moderate-to-severe refractory ulcerative colitis who have failed at least two advanced therapies, does adjunctive hyperbaric oxygen therapy (HBOT) produce clinically meaningful composite clinical and endoscopic response at 12 weeks, and does it alter bowel perfusion as measured by contrast-enhanced ultrasound (CEUS)? __SNAPSHOT__ {"studyDesign":"Prospective, open-label, non-randomized, dose-ranging phase 2a pilot trial (PARADOX) with sequential cohort enrollment","setting":"Single center: Amsterdam UMC, tertiary IBD referral center, Amsterdam, The Netherlands","sampleSize":"n=16 (10-session cohort: n=8; 20-session cohort: n=8); enrolled January 2023–December 2024","population":"Patients aged ≥16 years with confirmed UC of ≥4 months duration, moderate-to-severe disease activity (total Mayo score ≥5, Mayo endoscopic subscore [MES] ≥2), and prior failure of ≥2 classes of advanced therapies (biologics or small molecules). Background medications were required to be stable and at optimized doses for ≥8 weeks prior to enrollment.","intervention":"Hyperbaric oxygen therapy (HBOT): 10 or 20 daily sessions at 2.4 atmospheres absolute (ATA), 120 minutes/session with two 5-minute air breaks, administered in supervised multiplace hyperbaric chambers, as adjunct to ongoing stable background therapy","comparator":"No control or placebo arm; sequential cohort dose-ranging design (not applicable for between-arm comparison in the traditional sense)","primaryOutcome":"Composite clinical and endoscopic response at week 12, defined as ≥3-point and ≥30% reduction in total Mayo score, rectal bleeding subscore=0, and ≥1-point decrease in MES","followUp":"26 weeks (assessments at end of HBOT course, week 12, and week 26)","fundingSource":"Investigator-initiated; specific funding source not reported in the available abstract and conference data"}
Key findings
Among 16 enrolled patients, the composite primary outcome (clinical and endoscopic response at week 12) was achieved in 6/16 (38%) overall: 2/8 (25%) in the 10-session cohort and 4/8 (50%) in the 20-session cohort. No p-value for between-cohort comparison was reported. Improvements in clinical, endoscopic, symptomatic, biochemical, and histological parameters were observed in both groups, with a trend toward greater benefit in the 20-session group. No NNT is calculable given the absence of a control arm. Mechanistically, CEUS demonstrated a significant divergence by week 12: clinical responders showed a mean increase in peak enhancement of +12.1 dB (95% CI: 0.8 to 34.3) from baseline, while non-responders showed a numerical decrease of –5.4 dB (95% CI: –20.9 to 4.0); the 95% CI for non-responders crosses zero, indicating statistical non-significance for this subgroup. B-mode intestinal ultrasound (IUS) showed preserved bowel wall thickness in responders and declining thickness in non-responders. HBOT was well tolerated through week 26; the most common treatment-emergent adverse events were fatigue, worsening UC symptoms, and nasopharyngitis. No serious adverse events or treatment-related discontinuations were reported.
Practice implications
The PARADOX trial provides preliminary, hypothesis-generating evidence that adjunctive HBOT can achieve composite clinical and endoscopic response in approximately one-third to one-half of highly refractory UC patients who have exhausted multiple biologic and small molecule therapies. The mechanistic signal from CEUS—showing increased bowel perfusion specifically among responders—is novel and supports a vascular/regenerative mode of action distinct from conventional anti-inflammatory pathways. However, multiple critical limitations preclude any change in practice at this time: the trial was open-label and uncontrolled (no sham HBOT group), making placebo effect and observer bias unquantifiable; the sequential cohort design (rather than true randomization) introduces substantial risk of confounding, including baseline imbalances between dose groups (median age 48 vs. 64 years; median Mayo score 10.5 vs. 9); and the sample size of 8 patients per cohort is insufficient to establish dose-response relationships or estimate effect sizes with precision. The study did not establish superiority over any comparator, durability of response beyond week 26, or the optimal HBOT regimen. Patients with severe or acutely deteriorating disease, pediatric populations (age 16–18), and those not at a center with multiplace hyperbaric chamber access were not adequately represented. Findings are based on the published abstract and conference data only; the full peer-reviewed manuscript was published in the Journal of Crohn's and Colitis (Volume 20, Issue 5, 2026) but was not fully accessible for this summary, which should be noted as a limitation. __LIMITATIONS__ No control arm or sham HBOT comparator: precludes attribution of observed improvements to HBOT itself versus background therapy, spontaneous fluctuation, or placebo effect,Open-label design: unblinded patients and assessors introduce performance and detection bias; endoscopy and IUS were centrally adjudicated (a partial mitigation),Sequential (non-randomized) cohort design: meaningful baseline differences between dose groups (age, Mayo score) may confound apparent dose-response signals,Very small sample size (n=8 per cohort): severely underpowered to detect between-group differences or estimate response rates with clinical precision; all effect estimates carry wide uncertainty,Single-center study at a specialized tertiary referral center: limits generalizability to community or non-academic settings and to centers without multiplace hyperbaric chambers and CEUS expertise,CEUS and IUS are exploratory mechanistic endpoints; divergence between responders and non-responders is biologically interesting but subject to selection bias inherent in post-hoc responder/non-responder stratification,No formal statistical comparison between 10- and 20-session cohorts was reported; the trend toward greater response with 20 sessions cannot be interpreted as a confirmed dose-response relationship,Absence of a sham or placebo control means patient-reported and symptomatic outcomes are particularly vulnerable to bias,Colectomy-free survival and long-term durability beyond 26 weeks were not assessed,Specialized infrastructure required (multiplace hyperbaric chambers, technical CEUS expertise) limits practical replicability across most IBD programs,Full peer-reviewed manuscript data were not fully accessible for this summary; additional detail on statistical analyses, histologic outcomes, and subgroup data may be present in the complete paper __BOTTOM_LINE_CATEGORY__ promising_but_preliminary __BOTTOM_LINE__ The PARADOX trial demonstrates that adjunctive HBOT is feasible and well tolerated in highly biologic-refractory UC, with a composite clinical-endoscopic response rate of 38% (6/16) overall at week 12 and a mechanistically coherent CEUS signal of increased bowel perfusion in responders. These findings are hypothesis-generating and insufficient to change practice, but they provide a compelling rationale for a randomized, sham-controlled, adequately powered trial. Clinicians should not offer HBOT as a standard-of-care adjunct for refractory UC outside of a clinical trial context.