Hyperbaric Oxygen Therapy Facilitates Healing of Chronic Foot Ulcers in Patients With Diabetes

This study was initiated, designed, and performed by the authors. A steering committee was in charge of organization, data handling, and general conduct of the study. A blinded clinical event committee evaluated and classified all reported events as well as clinical outcome. The protocol was approved by the ethics committee at the Lund University.

The HODFU study was a randomized, single-center, double-blinded, placebo-controlled clinical trial that evaluated the effect of HBOT on ulcer healing in patients with diabetes and chronic foot ulcers. The outcomes for the group receiving HBOT were compared with those of the group receiving treatment with hyperbaric air. The study design and rationale were reported in detail previously (18).

All patients had diabetes and at least one full-thickness wound below the ankle for >3 months. They were previously treated at a diabetes foot clinic for a period of no <2 months. All patients were assessed by a vascular surgeon at the time of inclusion and only patients with adequate distal perfusion or nonreconstructable peripheral vascular disease were included in the study. Patients having an acute foot infection were included when the acute phase was resolved. Oral or local antibiotic treatment did not exclude patients from study participation. Exclusion criteria for study participation were contraindications for hyperbaric treatment (severe obstructive pulmonary disease, malignancy, and untreated thyrotoxicosis), current drug or alcohol misuse, vascular surgery in the lower limbs within the last two months, participation in another study, or suspected poor compliance. All participants provided written informed consent.

Patients were stratified based upon arterial toe blood pressure (≤35 mmHg vs. >35 mmHg) before being randomly assigned to either of the treatment arms. Randomization was done in blocks of 10 using sealed envelopes.

The study was performed in an ambulatory setting. Treatment sessions were given in a multi-place hyperbaric chamber five days per week for eight weeks (40 treatment sessions). The treatment period could be extended to 10 weeks, but the number of treatments was not allowed to exceed 40. An HBOT session included a period of compression in air for 5 min, followed by a treatment period at 2.5 atmospheres absolute (ATA) for 85 min, and then a decompression period of 5 min. Patients from both groups could be treated in the same session as study gases were administered by masks and air or 100% oxygen entered the chamber through separate double-blinded pipes (19). Study treatment was given as an adjunct to regular treatment at the multidisciplinary diabetes foot clinic, which included treatment of infection, revascularization, debridement, off-loading, and metabolic control according to high international standards (18). Investigators did not intervene in the daily routine clinical management of the patients. Outcomes were measured every first week during the treatment period (first 8–10 weeks) and then at three-month intervals. Ulcers were graded using the Wagner classification system and ulcer areas were measured using Visitrak Digital (Smith & Nephew, Hull, England) (20).

The primary end point was healing of the index ulcer. The index ulcer was defined as the ulcer with the largest area and a duration of at least three months at the time of inclusion. An ulcer was considered healed when it was completely covered by epithelial regeneration and remained so until the next visit in the study. Wagner grade 4 ulcers were considered healed when the gangrene had separated and the ulcer below was completely covered by epithelial regeneration. If a patient died during the follow-up period, he/she was censored at the time of the death. If a major amputation (above ankle amputation) was required, the ulcer was considered not to have healed. Secondary end points reported in this paper are major amputations and death.

The treatment code was not broken until the last patient had completed the 1-year follow-up visit.

Statistical evaluation was initially performed as an intention-to-treat analysis. An analysis was also performed on those patients who received >35 treatments (per protocol analysis). Categorical variables were analyzed using contingency tables (Fisher exact test) and continuous variables using Mann-Whitney U test. A two-sided P value <0.05 was considered statistically significant. Statistical analysis was performed using Statistica software, version 8.0 (Statsoft, Tulsa, OK).

During the inclusion period, a total of 164 eligible participants were registered: 23 (14%) did not consent to participation, 47 (29%) were excluded according to the study protocol, and 94 (57%) underwent randomization (Fig. 1).

The two groups had similar characteristics at baseline (Table 1). The median ulcer duration was 10 months and the median ulcer area was 3 cm². A majority of index ulcers were classified as Wagner grade 3 and 4 (20). Vascular surgery was previously performed in the affected lower limb in 55% of the participants.

Fifty-four (57%) patients completed all 40 treatment sessions, and 75 (80%) completed >35 sessions. Nine patients completed <10 treatment sessions, and the remaining 10 patients completed between 14 and 28 sessions. Early treatment termination was due to claustrophobia in two patients and worsening medical condition in nine patients (two deaths, two major amputations, and five hospitalizations). Eight patients decided to withdraw from the study.

Open vascular surgery in the affected lower limb was not performed on any study participant during the 1st year after inclusion, whereas percutaneous transluminal angioplasty (PTA) was carried out in 10 patients (Table 2).

In the intention-to-treat analysis, complete healing of the index ulcer was achieved in 37 patients at 1-year of follow-up in 25/48 (52%) in the HBOT group and 12/42 (29%) in the placebo group (P = 0.03). In the per-protocol analysis, complete healing of the index ulcer occurred in 23/38 (61%) in the HBOT group and 10/37 (27%) in the placebo group (P = 0.009) (Fig. 2). The largest differences in healing rates between groups were seen at the nine-month follow-up. Numbers needed to treat to avert nonhealing of a chronic foot ulcer were 4.2 and 3.1 according to the intention-to-treat and per-protocol analysis, respectively. During the 1st-year of follow-up, new ulcers developed in nine patients in the HBOT group and eight patients in the control group.

Four participants—one in the HBOT group and three in the placebo group—died during their 1st year in the study. The patient in the HBOT group received eight treatments and died of multiple organ failure 20 days after inclusion. She was 87-years-old and had a medical history including severe peripheral artery disease, neuropathy, myocardial infarction, and heart failure. In the placebo group, the causes of death were myocardial infarction in two patients (at 162 days after 39 treatments and 218 days after 22 treatments) and sepsis derived from an infected foot ulcer in one patient (at 144 days after 40 treatments).

Three major amputations were performed in the HBOT group as compared with one in the placebo group within the 1st year. In the HBOT group, two of the three amputations were done within 2 months after inclusion and the third was done at 191 days. The amputation in the placebo group was performed 98 days after inclusion. All four patients had an arterial toe blood pressure ≤15 mmHg in the affected lower limb. Two of these ulcers were classified as Wagner grade 4 and two were classified as grade 3 ulcers (20). Four minor amputations were performed in each group during the 1st year of follow-up.

All together 3,225 treatment sessions were given in the HODFU study. As discussed above, one fatal outcome was seen in the HBOT group during the treatment period, and a relation between HBOT and this fatal outcome cannot be excluded. In the placebo group, one patient was hospitalized for 24 h after temporarily losing consciousness after a treatment session.

Hypoglycemia (symptoms and blood-glucose <3.0 mmol/l) within 6 h after treatment occurred in two and four patients in the HBOT and placebo groups, respectively (n.s.). One of the patients in the HBOT group was hospitalized due to the event.

One patient in the HBOT group endured barotraumatic otitis. Another four patients, two in each group, required myringotomy with tube placement due to pain caused by the inability to equilibrate air pressure through the eustachian tube.

In the HBOT group, treatment-related dizziness was seen in one patient and the worsening of cataracts in another. One case of minor head injury occurred in the placebo group after a fall inside the hyperbaric chamber. Oxygen toxicity, seizures, or pneumothorax were not seen.

The present study supports the concept that adjunctive treatment with HBOT enhances foot ulcer healing in selected patients with diabetes. Accordingly, in our patients with long-standing chronic ulcers at 1-year follow-up, HBOT doubled the number of healed ulcers as compared with adjunctive treatment with hyperbaric air used as placebo.

In this study, arterial toe blood pressure was not a discriminating factor for beneficial effect of HBOT (data not shown). Patients with arterial toe blood pressure as low as 5 mmHg did heal in the HBOT group, and we could not identify a lower level of arterial toe blood pressure for nonhealing. Ulcer healing rate in the placebo group might be lower than expected, probably mainly reflecting a long prestudy period of failure to heal.

Our findings are in agreement with those reported in previous randomized studies focused on ulcer healing by Abidia et al. (17), Duzgun et al. (21), Kessler et al.(22), and Kalani et al. (23). Abidia et al. evaluated the effect of HBOT compared with hyperbaric air in a double-blind study of ischemic Wagner grade 1 and 2 ulcers. In that study of 18 patients, a nonsignificant improvement of healing rate following HBOT was seen after six weeks, which reached statistical significance at 1-year follow-up. In the unblinded, randomized study by Duzgun et al., the effect of HBOT was compared with standard therapy in 100 patients with a foot ulcer duration of at least four weeks. During a mean follow-up period of 92 weeks, primary healing was achieved in 66% of patients receiving HBOT compared with 0% following standard therapy (21). In the randomized trial by Kessler et al., the effect of two daily 90-min sessions of HBOT five days a week for two weeks was compared with regular treatment in 28 hospitalized patients with neuropathic Wagner grade 1 to 3 ulcers (22). After two weeks of treatment, the reduction in ulcer area was doubled in the HBOT group (P = 0.037). However, this improvement disappeared during the next two weeks of follow-up. The study by Kalani et al. included 38 patients with ischemic ulcers without full-thickness gangrene. After three years, 76% of the 17 patients receiving HBOT had healed their ulcers to intact skin compared with 48% of those given conventional treatment. However, these data should be interpreted with caution as the randomization procedure was not sustained through the study (23).

In contrast to our findings, some studies have shown the beneficial effect of HBOT in preventing amputations (8,21,23). Our study had insufficient power to analyze the rate of major amputations, which were only performed due to life-threatening infection and refractory pain. In the studies by Faglia et al. (8), Kalani et al. (23), and Duzgun et al. (21), amputation rates were substantially higher than in our study, suggesting these studies used other indications for major amputation (i.e., long-term unhealed ulcers) (8,21,23). This might indirectly be mirrored in toe blood pressure levels that were ≤15 mmHg in our amputated patients compared with 42 ± 30 mmHg in the study by Kalani et al. (23). Also, a possible bias effect cannot be excluded in open studies where the investigator may be involved in the decision to amputate.

Although >50% of our patients had had at least one previous vascular intervention at the time of randomization and the need for—or possibility of—vascular surgical intervention had been excluded at time of inclusion, PTA was performed in 10 patients (11%) within one year of randomization. This partly reflects a general problem in evaluating the effects of different treatment methods for diabetic foot ulcers and is not unique for our study. For instance in the study by Faglia et al. (8), vascular surgical intervention was performed in 38% of the patients during the follow-up period. The implications of these interventions on study outcome have been debated (24). In our study, however, PTA intervention cannot explain the higher ulcer healing rate in the HBOT group as only three ulcers, one of which was in the placebo group, healed following PTA.

The frequency (6%) of middle ear barotrauma or the need for myringotomy with tube placement was not higher in our study population of relatively old patients with severe cardiovascular disease and neuropathy compared with previously reported data in younger patients (25). In our treatment schedule, an HBOT session included a period of compression in air for 5 min, followed by a treatment period at 2.5 ATA for 85 min, and then a decompression period of 5 min. This schedule has been shown to be safe for scuba divers and our study showed it to be safe in our diabetic foot ulcer population where 3,225 treatments were given without decompression sickness, seizures, or pneumothorax.

In conclusion, the HODFU study, a double-blinded, randomized, placebo-controlled trial, showed that adjunctive treatment with HBOT facilitates healing of chronic foot ulcers in selected patients with diabetes.

This study was supported by unrestricted grants from Thelma Zoégas Foundation, Region Skåne Foundation, and the medical faculty of Lund University.

No potential conflicts of interest relevant to this article were reported.

This study was published in abstract form at the 16th International Congress on Hyperbaric Medicine, Beijing, China, 26–29 October 2008; the European Committee for Hyperbaric Medicine's Oxygen & Infection Conference, Stockholm, Sweden, 7–9 May 2009; the 19th European Wound Management Association Conference, Helsinki, Finland, 20–22 May 2009; and the 7th Diabetic Foot Study Group Conference, Bled, Slovenia, 25–28 September 2009.

The authors thank Dr. John Jeffery for scientific advice and help with the English text; Dr. Thomas Kjellström and Dr. Jan Apelqvist for helpful support; and registered nurse Carina Johansson and assistant nurses Gisela Torildsson and Magnus Andersson at the Hyperbaric Unit and research nurses Anneli Svensson and Lisbeth Lundell at the Research Unit at Helsingborg Hospital for collaboration and invaluable help.