Dear Readers:

To honor T. K. Hunt, MD, Emeritus Professor of Surgery at the University of California, San Francisco, we devote this Evidence Corner to hyperbaric oxygen (HBO). Professor Hunt has been a mentor to many and a key contributor to the science of wound healing. Among his distinguished works, he and his colleagues discovered how chronic tissue hypoxia stimulates capillary proliferation.[1] As a result of this work, some hypothesized that increasing oxygen gradients by increasing the oxygen supply to tissue surrounding chronic wounds would improve healing, particularly for ischemic wounds. However, the evidence supporting this hypothesis is sparse and variable. Confusion arises in this literature from variability in techniques of applying HBO as well as from inconsistency in measuring oxygen tension in wounds or surrounding tissue associated with HBO therapy. The currently accepted definition for HBO is systemic intermittently inhaled 100-percent oxygen in chambers pressurized above one atmosphere absolute or ATA. One ATA is defined as the atmospheric pressure at sea level, approximately 14.7 pounds per square inch or 101.3 kiloPascals. The evidence described here includes a systematic review of HBO effects in the wound care literature on ischemic wounds and a randomized controlled trial (RCT) exploring HBO effects on nonischemic diabetic foot ulcers.

Laura L. Bolton, PhD
Department Editor, WOUNDS

Hyperbaric Oxygen Effects on Wound Healing: A Systematic Review

Reference: Wang C, Schwaitzberg S, Berliner E, et al. Hyperbaric oxygen for treating wounds: A systematic review of the literature. Arch Surg 2003;138:272–9.

Rationale: On request from the Centers for Medicare and Medicaid Services, the Agency for Healthcare Research and Quality assessed the safety and efficacy of HBO in treating hypoxic wounds. The resulting systematic review of the literature served as the basis for this publication.

Objective: Determine the conditions and indications for which HBO is an effective adjunct treatment for hypoxic wounds and any contraindications or adverse effects of HBO.

Methods: A systematic literature review of the MEDLINE database was conducted from January 1998 to August 2001 to supplement and include references from prior systematic reviews plus articles suggested by reviewers. Only published articles reporting original data on clinical outcomes for at least five human subjects were included. Randomized, controlled trials (RCTs), nonrandomized comparison studies, and case series reporting amputation, wound healing, duration of hospitalization, and/or infection control were reviewed. The authors extracted data on patient demographics, conditions, diagnostic criteria, wound duration, measurements of periwound transcutaneous partial pressure of oxygen (TcPO2), study design, adverse effects of treatment, and major clinical outcomes.

Results: Of the 57 studies on 2070 patients, 7 were RCTs on 430 patients. Two RCTs on chronic diabetic ulcers reported reduced amputation risk in the HBO (n=50) vs. control (n=50) arms. One RCT on crush injuries and suturing of severed limbs reported 94 percent of the 17 HBO wounds healed in a mean of 50 days versus 55 percent of the 18 controls healed in a mean of 56 days (p<0.01 for percent healed with no significant difference in healing times). Two RCTs on compromised skin grafts reported less wound dehiscence (p=0.001) and less delayed wound healing (p=0.001) in 80 HBO versus 80 control wounds and improved survival of skin grafts (p<0.01). The remaining two RCTs reported less osteoradionecrosis (p=0.01; total n=74) and healing improvement based on clinical signs and symptoms and x-ray interpretations (no significance reported, n=12). Dosage and treatment frequency and duration varied, usually ranging from 2 to 3 ATA for up to 44 sessions of up to 90 minutes each in either mono- or multiple-patient HBO chambers. While the case series generally concluded that HBO was beneficial, they were inadequately controlled, so it was not clear whether HBO added benefit beyond standard wound care including wound debridement and antibiotics. Adverse reactions reportedly related to HBO therapy were reported in at least 42 (2.02%) of the 2070 patients. These included oxygen-related seizures, earaches requiring tympanostomy tubes in some patients, transient vision changes, and one seizure-related death two hours after decompression.

Conclusions: The overall methodological quality of the studies was deemed poor, with only seven RCTs. Serious adverse events included seizures and barotraumatic otitis in seven studies. High quality RCTs evaluating the short- and long-term benefits and risks of HBO therapy are needed to improve the basis for clinical decisions about HBO use to improve recovery and healing.

HBO Accelerates Healing of Nonischemic Diabetic Foot Ulcers

Reference: Kessler L, Bilbault P, Ortega F, et al. Hyperbaric oxygenation accelerates the healing rate of nonischemic chronic diabetic foot ulcers. Diabetes Care 2003;26(8):2378–82.

Rationale: In-vivo wound healing studies in the rat have reported acceleration of angiogenesis, but a case series suggests that this result may depend on peripheral arterial disease in nondiabetic patients.

Objective: This prospective RCT compared healing efficacy of HBO therapy on nonischemic diabetic foot ulcers.

Methods: Twenty-eight patients with type 1 or type 2 diabetes without clinical signs of arteriopathy, gangrene, sepsis, emphysema, proliferating retinopathy, or claustrophobia, having Wagner grades 1 to 3 neuropathic foot ulcers that had not improved in three months were randomly assigned to receive either HBO or full standard treatment. Absence of arteriopathy was confirmed by palpation of arterial pulses in the lower extremity as well as normal lower limb Doppler scans and transcutaneous oxygen tensions >30mmHg measured at the dorsum of the foot. All patients were hospitalized for two weeks of conventional treatment during which those randomized to the hyperbaric group received additional treatment with HBO at 2.5 ATA in an intermittent protocol totaling 90 minutes twice daily five days a week. Both groups were provided Barouk shoes for offloading pressure over the foot ulcers and blood glucose control with 2 to 3 daily injections of insulin for most patients as well as appropriate antibiotics to address microbiologically identified infections. Healing was measured as ulcer area at Day 0 before HBO and Days 15 and 30 after initiating HBO therapy.
Results: The two groups did not differ on any of the clinical measures at baseline. Wound surface area decreased 41.8 percent during the hospitalized two weeks of HBO therapy (n=14) or 21.7 percent for the hospitalized control group (n=13). After release from the hospital and cessation of HBO therapy healing rates were comparable with 0- to 4-week–percent reduction in ulcers size of 61.9 percent for the HBO group, with 2 healed versus 55.1 percent for the control group and none healed. One HBO patient (7.1%) experienced barotraumatic otitis resulting in withdrawal from the study.

Conclusions: The authors conclude, “HBO doubles the mean healing rate of nonischemic foot ulcers in selected diabetic patients.”

Clinical Perspective

Hypoxic or ischemic wounds. The evidence for using HBO on hypoxic wounds is generally positive but less than compelling and describes significant risk of adverse events associated with HBO therapy. Appropriately powered, well-designed prospective RCTs are needed to identify the timing and procedures for HBO administration and criteria for identifying wounds that may benefit from it.

Nonischemic ulcers. Comparable four-week healing rates in the HBO and control group suggest limited enduring clinical value of two weeks of HBO on nonischemic, diabetic, neuropathic ulcers when appropriate offloading and glucose control are applied. If the doubled healing rate had been maintained during four weeks of HBO, the clinical advantage of a potential 83-percent reduction in ulcer size versus 43 percent for control patients may have outweighed the risks of adverse events associated with HBO, but the study was not designed to test this hypothesis. There remains insufficient evidence for using HBO on nonischemic ulcers.

General conclusions. While periwound TcPO2 over 200mmHg for patients in-chamber at 2.5 ATA predicts tissue survival and repair,[2] oxygen alone cannot do the job. In fact, more than 80 percent of neuropathic diabetic foot ulcers heal in 10 weeks with total contact casting plus a hydrocolloid dressing[3] or within three months of surgery.[4] Nature shares a vital message with us through this research. Once the cause of tissue damage is diagnosed and alleviated and any necrotic tissue is removed, the environment for healing requires not just one molecule, but all the essential moisture, blood gases, electrolytes, nutrients, cytokines, etc. in the right proportions at the right times to foster healing. To test the value of any therapeutic modality, all of these steps and essential factors should be addressed equally in the experimental and control group with only the experimental modality varied, preferably using blinded procedure.