A new randomized controlled trial (RCT) on the use of hyperbaric oxygen therapy (HBOT) for diabetic foot ulcers (DFUs) has been published in the high-quality diabetes journal, Diabetes Care. The RCT found that HBOT did not improve healing or amputation rates in patients with ischaemic DFUs compared to usual care. But, these findings have created some controversy; so what is the full story?
What do we know about HBOT in DFU?
The use of HBOT has long been thought to be beneficial in healing DFUs. The theory behind the benefit is that it increases the amount of oxygen to the tissue at the site of the DFU which in turn heals the DFU quicker. But in practice we don’t categorically know if this is the case, as the studies in this area have varied considerable in patients studied, methodology used and findings.
For instance the recent Cochrane Review concluded, “in people with foot ulcers due to diabetes, HBOT significantly improved the ulcers healed in the short term, but not the long term, and the trials had various flaws in design and/or reporting that means we are not confident in the results.” And the International Diabetic Foot wound healing guidelines recommend to “consider the use of systemic hyperbaric oxygen therapy, even though further blinded and randomised trials are required to confirm its cost-effectiveness, as well as to identify the population most likely to benefit from its use.”
However, after considering the theory and studies most experts think HBOT should be most beneficial in people with ischaemic DFUs. Which is exactly what this new RCT from the Netherlands hoped to look at.
What did this new study do then?
Well this was a large multi-centre RCT investigating HBOT in patients with ischaemic DFUs only. The study recruited patients from 24 hospitals in the Netherlands and one in Belgium. Patients needed to have diabetes, an ulcer on their lower leg, and a toe pressure <50mmHg. They then collected a whole range of baseline characteristics, including; demographic, diabetes history, comorbidity history, medication, PAD and DFU characteristics.
Patients were then randomized to receive a HBOT intervention of up to 40 x 90 minute sessions of HBOT (5 days per week until healed or 40 sessions) or no HBOT (control group). Otherwise according to the authors all patients in both groups received revascularization if required and standard care decided on by their treating clinicians, such as debridement, offloading and antibiotics if needed.
Patients were then followed up at 3, 6 and 12 months. The main outcomes they were looking for at each follow up visit were if the ulcer had healed, they had an amputation or died. They also measured quality of life and costs but the authors say they will report those findings in a future article.
So what did they find?
First, they recruited 120 eligible patients (60 in the HBOT intervention; 60 control). Both groups had the same baseline characteristics, except the HBOT group was younger and had higher HbA1c.
Second, after 12 months they found statistically similar results in both groups: for DFUs that had completely healed (50% HBOT; 47% control), average time to healing (202 days HBOT; 217 control), amputations (37% HBOT; 48% control), deaths (8% HBOT; 15% control), and those that were still alive and had not had a major amputation called ‘amputation free survival’ (82% HBOT; 68% control). But, they also found that only 39 (65%) in the HBOT group completed most (>75%) of their HBOT treatments.
So lastly, they compared the 39 patients who completed most of their planned HBOT treatments to all the other 81 patients. They still found no statistical differences for those completely healed (57% HBOT; 44%), but did find more amputation free survival in the HBOT group (92% HBOT; 67% control).
What was good or bad about this study?
Like last week we decided to do a quick assessment of this study’s quality using the new recommended diabetic foot study quality tool and we quickly rated this study 14/21 (67%) for quality.
The main strengths of this study were: i) it was a large multi-centre study of patients with ischemic DFUs, we would have rated them down for including lower leg ulcers but since they only reported 1 “DFU” above the ankle in each group we let them off the hook; ii) they reported comprehensive baseline characteristics, including thorough PAD histories and non-invasive vascular tests; iii) they thoroughly followed patients for 12 months and reported robust statistical analyses.
The limitations though were: i) they didn’t report what other DFU treatments were provided in each group, such as how they debrided, offloaded, dressed DFUs; ii) they didn’t recruit the numbers of patients they originally calculated they needed; iii) they didn’t ‘blind’ the patients, treating clinicians or the outcome raters, so everyone knew who got what treatment which could have biased findings; and iii) unlike the controls a large proportion of the HBOT group (35%) didn’t finish their planned treatments.
So what does that all mean?
Well it seems as the authors conclude, “HBOT did not significantly improve complete wound healing or limb salvage in patients with diabetes and lower-limb ischemia”. So what was the controversy?
Well many experts wrote into Diabetes Care to complain about some of the study’s limitations, i.e. the study didn’t report other DFU treatment, didn’t recruit the numbers they set out too and many patients dropped out. All in all they suggest the study was ‘under-powered’ (meaning there would have been a difference if they had recruited more patients) to detect a difference between the groups, and when they did look at the subgroup of patients receiving most HBOT treatments there was a difference.
However, as the authors counter in response, “it remains open for discussion whether our conclusions would have been different or even clinically more relevant if we would have been able to include the initially planned number of patients”. And they also importantly noted that “many patients with ischemic ulcers and diabetes are unable to complete a full HBOT regimen, mostly due to their overall bad health condition, and may therefore not benefit from HBOT (anyway)”.
So does that answer the question of is HBOT beneficial for healing DFUs? Well certainly not entirely, but it does seem to bring us a step or two closer. From these findings it seems HBOT might help in certain types of ischaemic DFU patients, but not all ischaemic DFU patients. However, in those certain types of ischaemic DFU patients, patients would need to be assessed to see if they are medically fit enough to complete all the HBOT treatments necessary. And what are those certain types of ischaemic DFU patients? Well ….. we don’t know and more RCTs are still needed; the search continues J.