This article originally appeared on WoundReference.


It is time for the third installment of diseases that are considered “off-label” for hyperbaric oxygen ... yet, these diseases have some (sometimes a lot of) evidence supporting effectivity and plausible pathophysiology for HBOT use. This blog installment will cover inflammatory bowel disease (IBD).

We need to make a couple of distinctions and disclaimers before starting on this journey. First, IBD is not irritable bowel syndrome. While there is diarrhea involved, that’s where the similarity stops. Irritable bowel should never be mistaken for IBD.

And, there is no indication for irritable bowel syndrome to be any better due to hyperbaric oxygen. Second, I recognize that there is a distinct difference between ulcerative colitis (UC) and Crohn’s disease (CD), but the mechanism by which HBOT is thought to work will not be much different in either case. So, where the literature is specific about recruiting only CD or only UC, I will so state in this review.

Like our cardiac pre-conditioning blog, there will be a number of papers to consider. The literature for HBOT and IBD begins in the 1980s with the most recent studies published in 2019! We are in luck because there are several “review articles” that have taken the literature to that point and reviewed or attempted to classify the research by effectivity.

I will include the literature review papers for your use rather than an exhaustive review of the literature. That said, I still plan to hit the highlights of research and evidence for/against HBOT in IBD as we go along.

Ready? Let’s get this show on the road.

The early literature

One of the first reported case presentations using HBOT for CD complications occurred in 1989. This was a woman with severe perineal CD that defied medical therapy for eight years. The senior hyperbaric physician author just happened to be one of the clinical “fathers” of hyperbaric medicine, Jefferson C. Davis, MD.

Dr. Davis was a career United States Air Force physician and one of the founders of International ATMO in San Antonio. The patient is a 48-year-old woman who had CD for 17 years, having taken all known forms of therapy at that time without success. She had multiple surgeries for debilitating fistulas of lower abdominal, perianal, buttock, and vulvar lesions.

Because this was a chronic non-healing wound, Dr. Davis decided that the patient might be a good candidate for hyperbaric oxygen therapy (HBOT). No medical or surgical therapy (including a proctocolectomy) had any effect on the wounds. He was able to show with perineal transcutaneous oximetry that the wounded tissue was hypoxic (15-18 mmHg).

In the hyperbaric chamber, at 2.4 ATA, tissue oxygen levels were greater than 800 mmHg and sometimes approached 1200 mmHg during oxygen breathing. Over 2.5 months, the patient had resolution of all perineal wounds except one area near her buttocks. She received 67 treatments during this time. This therapy was declared a success with evidence of complete healing.

Dr. Davis attributed the healing to chronic non-healing hypoxic wound and HBOT. Wrong! Right treatment: wrong mechanism of action. (I had the privilege of meeting Dr. Davis ... I can't criticize his clinical acumen.) Read on, it will take years for the research laboratory to catch up to clinical medicine ... again ... but it will.

Over the next several years, she had some episodes of recurrence, all of which were noted to be less severe than her original wounds. Each of these outbreaks was treated with 20 treatments of HBOT and saw resolution of the wounds. For our purposes, this case is simply a record of a 'hypoxic, chronic problem wound.' At that time, little was known about the etiology of CD and therapy was all over the map. Clearly, the thinking at the time was that CD contains an inflammatory as well as an autoimmune component based on treating the patients with corticosteroids and azathioprine (an anti-rejection drug).

HBOT was deemed successful because it reversed the chronic tissue hypoxia. While hypoxia is a key component of the resulting wounds, we now have the benefit of gene-level testing as to cellular components that are up-regulated or down-regulated by HBOT. Keep reading ... more hints ahead.

Before we discuss the next reference to HBOT for refractory CD perineal lesions, I should interject my opinion about HBOT and CD. This is another disease in which there is plenty of evidence for adjunctive HBOT, so...

...if I had a severe CD patient with refractory perineal disease, I would definitely recommend a course of 40 HBOT sessions in order to lessen the need for complicated surgery that has a 50% or greater likelihood of failure. These radical surgeries frequently do not stop the perineal disease course. To treat or not to treat, that is the question! For me, the answer is to treat.

A case report nearly identical to the one above was published in 1990. To save details, the HBOT regimen was 2.0 ATA for 62 treatments, divided into 3 stages in the surgical/reconstructive process. This paper focused on the immediate resolution of a purulent drainage to serosanguinous.

Hence, the discussion section focuses on HBOT and enhancing the effects of antibiotics in hypoxic wounded tissues. This is good logic for the day, but it has no bearing on the methods of action known today. Right treatment for the wrong reason ... again ...

The HBOT experience from 1989 to 1997

HBOT for Crohn's Disease — what is the mechanism of action?

How can I summarize the HBOT experience from 1989-97? There are a number of case reports of successful resolution of perineal wounds with addition of adjunctive HBOT. However, the mechanism of action remains unclear, and each case report theorizes a different mechanism of action.

What happens in 1997? For the first time, a cogent mechanism of action emerges! An immunology paper is released that states (in Vivo and in Vitro) Tumor Necrosis Factor-alpha (TNF), Interleukin-1, and Interleukin-6 are down-regulated by HBOT in patients with perineal CD. Aha! And, here we go. From our wound care knowledge, TNF, IL-1 and IL-6 are elevated when macrophages and monocytes are activated in the inflammatory phase of wound healing. Hence, wounds will not heal.

Down-regulating these factors tends to improve wound healing. This particular paper measures these cytokines before, during, and after a series of 20 HBO treatments. Interesting in that they found persistent, chronic perineal wounds healed, but after HBOT finished, the cytokine levels increased toward pre-treatment levels. Hmmmm ... This paper raises more questions than answers, but it is the first to recognize that HBOT ameliorates cytokine precursors of inflammatory cascades.

What else inhibits TNF and Interleukins?

Interesting question! There are a number of 'biologic' medications that have shown some efficacy in treating IBD. I'm a pretty simple person, so this explanation will try to simplify a complicated subject. When you look at the generic name for IBD drugs, we see a number of them terminating in “-mab.”

The “mab” ending signifies that the drug is a monoclonal antibody. Some of the drugs are fully human mab origin and some are not. Some of these drugs attack/decrease production of TNF, integrins and others attack the Interleukins. Beyond this explanation, further research is for the reader.

The point (in relation to HBOT) is that HBOT mimics some of the effects of the mab drugs, inhibiting TNF and Interleukins without setting up a potentially allergic reaction to further use of the biologic. Most of the patients reported for the HBOT intervention have already tried and failed biologic medications.

1997 to 2012: more case reports on HBOT for Crohn's disease

For the sake of brevity, I will summarize several more case reports and case series between 1997 and 2012. At this point, the only evidence of effectiveness is the resolution of severe perineal wounds in patients with CD in response to adjunctive HBOT. Unfortunately, it appears that the series (20, 40, 60+ treatments) isn't permanently curative.

There are a variety of responses to wound recidivism. Some authors have chosen to give 20 additional treatments when the CD perineal wounds recur. Uniformly, the reports state that wound recurrence is less severe than the original wounds and that the wound responds quickly to additional HBOT. A 2012 review of HBOT treatments for IBD reiterates some of the same points that we have made above.

The recent HBOT experience

HBOT for ulcerative colitis

Until this point, all of the papers have concerned advanced CD, and none have combined UC cases in the mix. Pagoldh et al. designed a prospective, randomized, open-label clinical trial for HBOT intervention and UC, published in 2013.

The primary objective of the study was to assess effects of HBOT on Mayo Score, laboratory tests, and fecal weight. The short version of the results show ... ummmm ... no improvement vs the control group in any of the categories. It appears that UC responds quite differently to HBOT than CD. Correction, HBOT appears to have no place in UC based on this study, even though the two disease states share similar TNF activity and several papers showing severe UC exacerbations are ameliorated with HBOT.

Speaking for myself, Pagoldh's conclusion doesn't make sense. The mechanism of action for HBOT should be the same for either UC or CD. How would HBOT work in one disease and not the other? I'm not attributing value to the following statement, just stating fact: The Pagoldh study is the only “negative” HBOT study in the IBD literature.

There are problems with the Pagoldh study. In fact, I would be reticent to accept the conclusions of this study as it has been performed. First, the power of the study is a problem. They set n=10 for each group. Only eight patients were enrolled in the 'control' arm of the study, and five completed while three did not. On the intervention side, there were 10 HBOT patients, of which only four completed the HBOT trial and six withdrew. In my mind, there is a significant amount of statistical bias, and one might easily interpret the data to rule out a process that is actually good (a Type II error).

In addition, while the study was randomized, it was not blinded, nor was there a sham control arm. This raises some potential methodological errors that may have tilted the conclusions of the study. Overall, I wish that the study contained more patients in both arms and was free of potential bias by using a sham control arm.

Stay with me now ... more to come. There is yet another systematic review of HBOT in IBD. Coming up shortly.

Patients with refractory inflammatory bowel disease 'heal' after pelvic exenteration and adjunct HBOT

Many of the patients with IBD have proctocolectomy procedures. You would logically think that the disease is within the colon, and that when you remove the affected bowel, the disease severity ceases. Not so simple. Patients with persistent perineal draining sinus cavities fail to heal from the proctocolectomy. Up to 33% of these patients can have continued draining sinuses for a year or more following muscle free flap procedures to close the perineal cavity left by the surgery. (I remember taking care of these patients when I was on surgery rotations in medical school. Definitely a difficult management!)

Chan and colleagues (2013) studied four patients who had medically refractory IBD and were undergoing a pelvic exenteration procedure and either a vertical or transverse rectus flap for closure. These four patients (2 with CD and 2 with UC) had HBOT (25 - 30 treatments preoperatively) consisting of 2.4 ATA for 90 minutes of oxygen breathing. When logistically possible, 10 treatments postoperatively were delivered. All four patients were declared 'healed' by 3 months postoperatively. None of the four patients had recurring sinus formation for 35 months follow up period.

Reviews conclude that adjunct HBOT is potentially efficacious for refractory inflammatory bowel disease

Dulai and colleagues from the Inflammatory Bowel Disease Center at Dartmouth-Hitchkock performed a systematic review of hyperbaric oxygen and inflammatory bowel disease in 2014. Their review of the literature was much more thorough than we've gone through to this point. They found 17 studies, involving more than 600 patients (286 with CD and 327 with UC). Overall response rate was 86% (the same whether UC or CD). With CD perineal sinus, 18/40 completely healed and 17/40 partially healed.

They conclude that HBOT is relatively safe with minor complications and potentially efficacious for IBD patients who are refractory to medical management (worst of the worst). Still, there have been no blinded, randomized, sham controlled studies. Ahhh, but Dulai and colleagues will come back with exactly that study in 2017 published in the American Journal of Gastroenterology.

A retrospective review of 32 consecutive patients with refractory UC were reviewed. (Remember the Pagoldh study? This one refutes its findings.) Over the period from 1994-2011, these patients had the standard care as well as 40 HBOT treatments. This research group reported a number of clinical and laboratory parameters (worth the read, if you are interested). Pre-HBOT stool frequency was 7, whereas post-HBOT was 1. Bloody stools went from 10 to 0. Endoscopy severity score decreased significantly. Biopsies with analysis for CD44 stem cells were greatly improved post-HBOT.

A randomized controlled study shows that HBOT is effective for ulcerative colitis

Dulai returns with a phase 2A pilot multi-center, randomized, double-blind, sham-controlled study in 2017. This still does NOT answer the questions, nor is it definitive. It is a pilot study and proof of concept. This study looked at patients who were hospitalized due to a UC moderate-severe flare. Even with low recruitment, we see the following trends. Clinical remission at day 5 and day 10 were 50% for the HBOT group to 0% for the sham-control.

HBOT patients did not require progression to secondary medications (10% vs 63% in the sham group). HBOT patients did not require urgent colectomy (0% vs 38% in the sham group). While this study needs to be done again with a much larger number of patients, the design looks very good. Sham is provided by taking patients to 1.34 ATA and breathing air the entire time. As an aside, Dr. Lin Weaver and staff from Salt Lake City have shown that patients compressed to 1.2 and 1.3 ATM are required to clear ears, feel heat of compression, and have no idea what final pressure they have been exposed to. Hence, an excellent sham design.

And there you have it ... from 1989 to 2019!

So, what is the conclusion on adjunctive HBOT for inflammatory bowel disease?

Of all the papers reviewed, only 1 (Pagoldh) was a negative trial. There are statistical reasons that we should be reticent about accepting that study. In fact, there were a number of case studies and case series with UC that were positive, including the Dulai pilot study. Dulai suffers the same statistical lack of power as Pagoldh. Based on the above, I would discount the Pagoldh paper, because Dulai follows the general trend of all the UC case series papers. Yes, I guess I'm biased, but we all are to some extent. So:

For severe IBD, medically refractory, I would recommend using adjunctive HBOT off-label for these patients. For the HBOT regimen, I would start with 40 treatments and allow patient response/or non-response to guide further treatments.

Related Topics

In case you missed it, see the introduction to this blog series.

This blog series focuses on three conditions that are off-label and have plausible literature evidence for improvement after HBOT: