Fracture Healing and Roles of Hyperbaric Oxygen

  • Michael B Strauss, MD, Anna M. Tan, DPM, and Lientra Q. Lu, BS
  • Volume 07 - Issue 4

Hyperbaric oxygen (HBO) treatments for fracture healing is not an approved indication according to the guidelines formulated in the 13th edition of the Undersea and Hyperbaric Medical Society Hyperbaric Oxygen Therapy Indications.¹ Use of HBO for fracture healing is, therefore, considered an “off-label” therapy for this modality. Consequently, the Centers for Medicare and Medicaid Services, as well as most other third-party payers, do not reimburse for using HBO as an adjunct to fracture healing. Since it is estimated that 300,000 (4.7%) of the 6.3 million fractures that occur in the USA yearly are slow to heal or do not heal, other modalities need to be considered to improve these statistics.2,3 Hyperbaric oxygen is one of them.

Bone is a remarkable tissue. It has multiple functions, including support and protection of the soft tissues of the body, a framework for muscles to act through joints, a fundamental component of calcium homeostasis, a site for generation of hematopoietic elements of the bloodstream and a reservoir for red blood cells. These functions of bone continue to occur while the fracture is healing or even in the presence of delayed or nonhealing of a fracture. Bone can also heal without leaving a scar, a trait it shares with the liver as the only body tissues that have the ability to do such.

Three requirements are essential for fractures to heal: adequate perfusion, appropriate stabilization and an environment for progenitor cells to be induced to form bone. When one or more of these elements is deficient, delayed or nonhealing of the fracture is likely to occur, as in the 4.7% of fractures in the USA that become nonunions. Hyperbaric oxygen has mechanisms to mitigate two of the three reasons fractures become nonunions. These are augmentation of perfusion- oxygenation and enhancement of the environment for fracture healing. This chapter describes the physiology of fracture healing, reports on the laboratory studies that support the use of HBO for bone healing and reviews the limited clinical experiences where HBO was used as an adjunct for fracture healing. From this information, a “new look” is warranted for using HBO in a select complement of fractures.

Physiology of Bone Growth and Fracture Healing

In the developing individual, two types of bone growth occur: enchondral and intramembranous. Enchondral bone formation occurs in the growth plates of long bones. It starts with a cartilaginous anlage. As the cartilage cells enlarge, they calcify and then outstrip their blood supply (Figure 1). Invasion of blood vessels bring in bone formation cell precursors, which line the calcified cartilage and lay down new bone along the cartilage anlage. The dead cartilage cells are reabsorbed by osteoclasts/chondroclasts. Intramembranous bone formation occurs under the periosteum of flat bones such as the clavicle. Bone formation occurs in situ under the periosteum and is incorporated into the outer surface of the cortical bone.

FIGURE 1. Bone formation at the growth plate


Legend: Oxygen availability is highly regulated in the growth plate. Cartilage multiplies in an oxygenated environment adjacent to epiphyseal blood vessels, but enlarges to lengthen the physis and eventually exceeds the diffusion distance of O2. The cartilage cells die from hypoxia; this initiates blood vessel invasion between the columns of cartilage. Osteoblasts generate new bone along the blood vessels.

Fracture healing incorporates features of both types of bone formation with highly regulated oxygen tensions in the healing environment. Intramembranous bone formation is visualized on x-ray as periosteal new bone. Fracture healing has many analogies to soft tissue wound healing (Figure 2). The stages include injury, hematoma formation, inflammatory response, formation of granulation tissue, mitogenesis to soft callus, maturation to hard callus/new bone formation and remodeling. Two factors are essential for each stage to progress to the next. First, there must be adequate perfusion- oxygenation to the fracture site so that the cells in each stage of fracture healing can express themselves with their oxygen-dependent functions. Second, the environment of the bone surrounding the fracture must act as an inducer for the pluripotential cells of the granulation tissue phase to transform to osteogenesis precursors rather than continue to form collagen that would heal the soft tissue wound.

Three cell types are associated with bones: osteocytes, osteoblasts and osteoclasts. Osteocytes are the cells embedded in the bone matrix. They are anything but “resting” bone cells, being very metabolically active in directing osteoblastic and osteoclastic activity as well as in the calcium phosphorus regulation in the bone tissue. Osteoclasts are multinucleate giant cells from the macrophage linage that are induced to form alkaline and acid phosphatases to remove calcified tissues. They are very metabolically active and have 100 times the oxygen requirements as the osteoblast. Osteoblasts are the bone- building cells. Their ability to build bone is about one one- hundredth that of the osteoclast’s ability to remove bone. Consequently, stabilization (as well as alignment during the remodeling stage) as stated above, is an important component of fracture healing. Without it, bone is subject to deformities and, subsequently, pathological fractures, abnormal mechanics leading to arthritis, and wound development at the apex of the deformity.

FIGURE 2. Fracture healing, O2 tensions and roles of hyperbaric oxygen

figure2Legend: Oxygen availability and regulation, as in the growth plate, is required throughout the fracture healing process. This occurs without problems in the 95.3% of fractures that heal normally. In the other 300,000 fractures that go onto nonunion in the USA each year, identifiable factors and especially ischemia—hypoxia are the cause. Hyperbaric oxygen can mitigate the ischemia—hypoxia cause

Key: * = Growth factors, ** = Skeletal muscle—compartment syndrome


Dr. Bassett in the late 1960s demonstrated fracture cell precursors when compressed in a hypoxic environment formed cartilage and in the hyperoxic environment formed bone (Figure 3).4 This model conforms to what is observed in nonunions where avascular cartilage and fibrous tissue bridge the bone ends.
Another consideration is that healthy, “raw” bone edges at the fracture site contribute to the induction effect. Death of bone cells immediately adjacent to the fracture is postulated to occur with all but the simplest fracture patterns. This dead bone is removed by the osteoclast, a very oxygen-dependent cell.


Bones are in a continual state of remodeling. Structural unsound bone is removed by the osteoclast, while osteoblasts form new bone along lines of stress. When osteoclast function is inhibited by bisphosphonates, an anti-osteoporosis medication, bone remodeling is impaired. Although the bone may appear structurally sound (i.e., thick cortical margins) on x-ray, the failure to remove and rebuild bone along the lines of stress make it subject to pathological fractures, as seen in femurs of longtime bisphosphate drug users.

Laboratory Studies with Oxygen and Fracture Healing

Many articles have been published about the effects of different oxygen tensions on fracture healing in animal models. Online libraries such as PubMed, Rubicon, Elsevier, Google Scholar, and VA Medical Library have been searched to obtain the most relevant studies on this subject; most of them are prospective, case-control experiments and are reviewed in this section.

In 1932, Ham et al. observed that osteogenic cells have a “dull potentiality” — being able to form bone or cartilage in response to the degree of perfusion-oxygenation in the area in which it differentiates.6 If oxygenation was compromised, fracture healing was disrupted and cartilage formed as is so often observed at a fracture nonunion site. Several earlier fracture-healing animal studies demonstrated that increased vascularity at the fracture site was needed before fracture healing occurred (Table 1).7-12  Degree of perfusion and rate of angiogenesis at the fracture site correlated highly with the speed of callus resorption and bone union. Cavadias and Trueta reported that periosteal and endosteal vessels invaded the undifferentiated mesenchyme at the fracture site.12  This confirmed that perfusion and oxygen delivery is an integral part of bone formation. Kolodny’s findings also supported the perfusion-oxygenation hypothesis when he created fractures on canine tibia and disrupted the nutrient artery sites to the bone. Nonunions uniformly occurred. Gothman observed that for a fracture to heal in a nonunion fracture model, angiogenesis/revascularization must occur within 14 days after the injury and the blood supply to the site needed to increase by a four- to fivefold factor; otherwise, nonunion would continue.9  From their study on the vascular pattern of canine’s undisplaced closed fractures, Rhinelander and Baragry found that increased blood flow associated with fracture healing stimulated long bone overgrowth.11

TABLE 1. Animal studies where vascularity was investigated in fractures
Reference Animal Treatment Results
Kolodny Dog radius Fracture & disruption of nutrient artery separating bone & soft tissue Nonunion resulted; decreased to total absence of callus formation vs. control limb
Trueta Rabbit Kuntscher (clover---leaf) type of intramedullary nailing of fracture model At 7---10 days, new bone layer appears along the diaphysis→→ gradually becomes denser. Periosteal callus appears earlier and more robust in rodded side & reflected periosteo perfusion
Gothman Monkey tibia Intramedullary nailing vs. plating vs. external fixation In nailed fractures, there was initially an 85.6% reduction in blood flow, but after 90 days, they had significantly greater blood flow than externally fixed and plate fixed fractures
Wray Immature rat Tibial fracture without immobilization éé Fractured femur length (i.e. overgrowth secondary to increased perfusion,; éé vascular bed observed at fracture site
Rhinelander Dog Undisplaced closed fractures vascular pacern vs. normal Vasodilatation of existing arterial tree, éé medullary (predominantly) and periosteal (auxiliary) circulations
Cavadias Rabbit Permeable barriers implanted in fractures Periosteal & endosteal blood vessels invade the implant bed


Reports in animal models demonstrate the harmful effects of hypoxia on bone precursors in fractures (Table 2). When Bassett subjected undifferentiated bone precursors of chick embryonic tissue to a hypoxic environment and pressure, cartilage formed (Figure 3).4 With induced hypoxia of varying degrees in rats, Vaes and Nichols noted that more lactate accumulated in bones with low oxygen tensions and there was a decrease in synthesis of new bone matrix.14 In canine tibial fractures exposed to hypoxia, Heppenstall reported that there was a decreased rate of bone union.24

FIGURE 3. Effects of pressure and oxygen tensions on undifferentiated mesenchyme

figure3Legend: Dr. Bassec's findings (Reference 4) confirmed that undifferentiated pluripotential cells form bone when subjected to compressive forces in an enriched oxygen environment and form cartilage in a hypoxic environment. Cartilage is typically found in the nonunion site.


With 85% oxygen and 5% carbon-dioxide partial pressures, Sledge and Dingle observed increased rate of cartilage degradation and bone formation in the dog model.15 With rats subjected to hyperbaric oxygen exposures several hours a day for 14 to 20 days, both Coulson et al. and Wray and Rogers noted an increase in callus formation and decrease in break strength.16,17  This latter finding was consistent with HBO augmentation of osteoclast activity. This resulted in temporary weakening of bone during the bone remodeling process. Yablon and Cruess demonstrated increased healing of fractures with twice-a-day HBO exposures in rats.18 Laurnen and Kelly using radioisotopes (strontium-85 washout) showed a sixfold increase in blood flow in dog tibial fractures, while oxygen consumption, carbon- dioxide production and blood calcium and pH changes remained the same as in the opposite control limb.19 These findings confirm that increased metabolism occurs with the concomitant need for oxygen in fracture healing.

Niinikoski et al. and Penttinen et al. studied biochemical changes in rat tibial fractures.20,21  They found that increased oxygen tensions stimulated precursors  of fracture healing and mineral productions such as Ca, Mg, Zn, etc. This led to increased callus formation at the fracture sites. According to Brighton and Krebs’s study on the rat epiphyseal growth plate utilizing mega vitamin A doses and high oxygen tensions, the pretreatment low oxygen tension promoted calcium release from the blood stream to the growth plate.22  Subsequently, the high oxygen tensions resulted in hydroxylation of proline, an essential amino acid bone formation precursor generated by the undifferentiated fibroblast. The study also revealed a progressive decrease in oxygen tensions from the epiphysis to the metaphysis of the growth plate (Figure 1). In an experiment on murine calvaria fractures exposed to two atmospheres absolute pressures continuously for 24 hours, Gray et al. found an increase in collagen deposition and a decrease in bone formation and resorption.23

TABLE 2. Animal studies demonstrating the effects of varying oxygen tensions on fracture healing
Reference Animal Exposure Results
Bassex Chick embryo Compression plus high (35% O2) and low O2 tensions Bone formation with high O2 tensions; cartilage formation with low O2 tensions
Vaes Rat Hypoxia of varying degrees If bone O2 tensions remained low, lactate accumulated,; êê synthesis of new matrix
Sledge Chick embryo 85% O2, 5% CO2 Degradation of cartilage septa, conversion of cartilage to bone
Coulson Rat 3 ATA, 2 hr. daily; 14 days éé Breaking strength of fractures
Wray Rat 2 ATA, 6 hr. daily; 20 days éé callus, decreased strength
Yablon Rat 3 ATA, 1 hr. BID; 4---40 days Improved healing with BID HBO
Laurnen Dog Strontium---85 clearance after 10 minutes in tibiae 12 weeks after fracture No difference in fx side & control side pO2, pCO2, pH. O2 consumption & CO2 production increased, reflecting blood flow has stabilized.
Niinikoski Rat 2.5 ATA, 2 hr. BID; 5---21 days éé callus; éé precursors
PenZnen Rat 2.5 ATA, 2 hr; BID Callus 19---42% éé nitrogen 15---60% éé minerals (e.g. Ca, Mg, Zn, etc.) 20---63% éé strengths unchanged
Brighton Rat Low O2 initially in the cartilage anlage of the growth plate

Low O2 →→ Calcium release; 2 requirement for proline →→ OH-­‐proline

Gray Mice calvaria 2 ATA, 24 hr. continuous exposure éé collagen; éé bone formation & bone resorption
Heppenstall Dog Hypoxia (50%) Delay fracture healing

As the animal review shows, in the 1960s and 1970s there was much interest in studying the roles of oxygen and blood flow in general and HBO in particular on fracture healing in animal studies. Unfortunately, this interest waned after this time probably because of improved orthopaedic fracture fixation techniques, application of callotaxis principles (Ilizarov), new and improved bone grafting alternatives, pulsed electrical stimulation of bone and use of inducers such as bone morphogenetic protein. Consistent findings associated with HBO exposures included increased and speedier callus formation and faster bony union. The fracture strength, however, was initially weaker than compared to the control limbs. These observations substantiate the mechanisms that HBO exhibit in fracture healing. In fracture healing, the oxygenated environment is essential for the pluripotential cells in the precallus to generate bone. In contrast, the physis (growth plate) accounts for the longitudinal growth of long bones in children. In this situation, the cells in the cartilage columns enlarge to the point that they exceed the oxygen diffusion gradient from the epiphyseal arteries and die (Figure 1). This stimulates angiogenesis from the metaphyseal blood vessels, which, in turn, provides a sufficiently oxygenated environment for the osteoblast to lay down new bone along the dead cartilage cell anlage.

In summary, the increased oxygen tensions in the fracture- healing environment accelerate callus formation and osteogenesis. In addition, the activity of the highly oxygen dependent osteoclast is also stimulated. This results in faster bone resorption for the remodeling and ultimate strengthening process of the fracture. In the interim, because of the speeded-up remodeling process, the bone exposed to HBO is temporarily weaker. By the ends of the observation periods, bone strengths were usually equal in both the HBO and control limbs.

These observations are consistent with the first authors’ experiences that in about 20% of the cases where HBO was used for fracture healing and/or arrest of refractory osteomyelitis, delayed onset stress fractures occurred after there was x-ray evidence of bone union.

A final observation from the animal studies is that “pulses” of HBO, that is intermittent exposures, are sufficient to achieve the bone healing effect. In contrast, the study of a continuous HBO exposure was detrimental to bone healing, which is consistent with oxygen toxicity (probable generation of free radicals) from the single, prolonged HBO exposure.22

Clinical Review of Hyperbaric Oxygen in Fracture Healing

While there have been numerous articles exploring the effects of HBO and fracture healing in animal models, little comparative clinical evidence of effectiveness exists. Extensive literature search via Pub Med, Ovid, Cochrane Library, and Google Scholar revealed a dearth of information, with search results consisting primarily of case reports and prospective series. These studies and most recent literature will be summarized in the following section.

Lindstrom et al. reported on 20 subjects requiring intramedullary nailing for closed tibial fractures who received HBO as an adjunct therapy. They measured transcutaneous oxygen tensions in the lower leg, limb temperature and distal arterial flow via Doppler. The trial, however, did not report any clinical outcomes in relation to fracture healing.25 Porcellini et al. reviewed patients with serious vascular injures in addition to fractures. Hyperbaric oxygen therapy was utilized in 7 out of 34 patients to control bacterial contamination and improve wound healing. Outcomes regarding HBO and its effect on fracture healing, however, were not explicitly reviewed.26

Braune et al. detailed the case of a patient with 17-year-old posttraumatic pseudarthrosis of the dens axis following conservative treatment of unstable Anderson’s type II odontoid fracture. Following surgical intervention and revisions, x-rays revealed an unstable posttraumatic Blauth’s type III odontoid nonunion in association with wound dehiscence, exposed autograft and internal fixation hardware. Hyperbaric oxygen treatments were used as a therapeutic option in conjunction with a surgical salvage procedure. Complete wound healing was observed after 25 days. In addition, radiographs showed bone fusion with incorporation of the autologous bone graft and solid atlantoaxial fusion.27

In a 2005 Cochrane review study, the authors were unable to find any articles to meet its inclusion criteria for justifying HBO for fracture healing.29 Thus, the conclusion was that there was insufficient evidence to support or refute HBO in use for delayed bony healing and nonunion of fractures. There is one randomized double blinded control trial, however, in which the authors studied fracture healing in crush injuries.28  Bouachour et al. reported complete healing in 94% of the HBO group versus 33% in the control limb, while need for repeat surgery (after initial debridement and stabilization) was 6% in the HBO group compared to 33% in the patients who did not receive HBO. Both observations were statistically significant (p <0.05). All patients had crush injuries (Gustilo grade 2-B open fractures) where vascular compromise and threatened flaps were an integral part of the pathophysiology of the injury.

The most recent studies regarding HBO and fracture healing are from China. Chen observed the effect of intramedullary nail fixation combined with HBO in the treatment of long bone fractures in children. The authors concluded that those in the HBO group had lower levels of pain and swelling with earlier ambulation and shorter healing times.30 In another study from China, Yu found that HBO significantly improved the formation of callus in 44 patients with delayed union of tibial fractures.31 In his randomized controlled study, the 22 patients in the test group who received HBO had a “total effective rate” of 95%. The author concluded that HBO promoted the healing of fractures and reduced the incidence of delayed union and nonunion.

In summary, only meager clinical evidence from the literature supports the use of HBO for fracture healing and especially nonunions. Further studies including prospective series and randomized control trials are needed to advance HBO as an adjunct to fracture healing, prevention of delayed union, and healing of nonunion fractures. Nonetheless, the current knowledge of fracture healing mechanisms and the animal studies reported above are consistent with specialized roles for HBO in fracture management.

Personal Experiences with Hyperbaric Oxygen in Fracture Healing

Through the years, the first author (MBS) has had a number of patients for which hyperbaric oxygen has influenced healing  of fractures. This group needs to be differentiated from the multitude of cases where HBO has been used as an adjunct to manage open fractures and crush injuries as described in the Bouachour study previously cited as well as in cases of chronic refractory osteomyelitis.28,32  The subset of patients in questions had delayed or nonunions of their lower-extremity fractures, or fracture healing was likely to be impaired due to the severity of the injury. The first experiences occurred with three active-duty US Navy personnel who had nonunions of their tibia fractures after motorcycle accidents. After nonhealing of the fracture was ascertained, each underwent a segmental (about 5 cm in length) fibula ostectomy and Sarmiento patellar tendon cast bracing. The rationale for the fibular osteotomy was to allow compressive forces to increase across the fracture site with weight bearing. Without subsequent evidence of healing, arrangements were made for the patients to receive three US Navy Treatment Table 5 HBO exposures a week for four weeks at the Naval Special Warfare recompression chamber at the US Naval Amphibious Base in Coronado, California. All three fractures became stable, allowing the discontinuation of cast bracing. X-rays indicated the stability was attributed to periosteal new bone formation.

Subsequently, the first author used HBO for a small number of “last-resort” fracture situations referred to him at Long Beach Memorial Medical Center in Long Beach, California. The following are five anecdotal cases in which HBO appeared to change favorably the patients’ outcomes:

One of the earliest cases was a middle-aged male who had bilateral nonunions of his distal femurs after Rush rodding and bone grafting. During the course of 20 hyperbaric treatments, the rods were removed, and then the femoral fractures stabilized with external fixation. The fractures healed after another 10 HBO treatments and subsequent removal of the external fixators 12 weeks after applications.

A second case involved a nonunion of the proximal femoral shaft in a male in his 20s. After cast immobilization failed, an external fixator was applied, but there was hardware failure with breakage of the fixator. With HBO treatments while remaining nonweight bearing on the extremity, the fracture healed.

A third case was that of male smoker in his 40s who after a motorcycle accident sustained a severe proximal tibial fracture that was managed with medial and lateral platings. The proximal 20% of the skin over the anterior aspect of the tibia sloughed, exposing both plates and the necrotic anterior tibial fragment (Figure 4). The patient was referred for HBO treatments as alternative to an above-knee amputation. With staged surgeries while receiving ongoing HBO treatments, the wound was debrided, hardware removed and the fracture stabilized with external fixation spanning from the distal femur to the middle tibia. After granulation tissue covered the cavity, autologous bone grafting was used to fill the defect, and a rotational gastrocnemius muscle flap used to provide soft- tissue coverage. Twenty-five years later, the patient continues to be a community ambulator with moderate loss of knee flexion.

FIGURE 4. Major skin slough and devitalized proximal tibia bone

figure-4Legend: Severe proximal tibial injury after motorcycle accident, referred for HBO treatments at adjunct to orthopaedic and plastic surgery intervention to prevent an above knee amputation


The fourth case was that of a male in his 20s who developed an infected tibial nonunion after plating of a closed shaft fracture and more than 15 subsequent surgeries. Eighteen months after the injury without evidence of healing or control of the infection, he had made the decision to undergo a below-knee amputation (Figure 5). His wife, a nurse at our medical center, convinced him to try HBO as an alternative to an amputation. After two weeks of HBO treatments, the nonunion site was debrided and the fracture site stabilized with an external fixator. A week later, when the nonunion site was covered with healthy granulation tissue, autologous bone grafting (Papineau technique) was done. Hyperbaric oxygen treatments were continued for two additional weeks. Blood vessels grew into the open bone graft, and gradually the graft epithelialized over a three-month period. At that time, the fixator was removed, and bony union was confirmed by clinical exam and x-rays.

FIGURE 5. Infected nonunion with multiple surgical attempts to resolve problem

figure-5Legend: Infected tibial nonunion after motorcycle accident and over 20 previous surgeries. HBO used at an adjunct to debridement and external fixation followed by staged bone grafting.


A fifth case concerned a male also in his 20s who sustained a severely comminuted open distal tibia and fibula fractures with a 5-cm gap due to loss of bone after the initial debridement. A spanning external fixator with single half pins in the calcaneus and proximal to the fracture were used to temporarily  maintain alignment until plastic surgeons could perform a microvascular ipsilateral fibular graft to span the defect. Unfortunately, the fixator did not maintain the length, and at the time of the fibular grafting, the gap had shortened 4 cm. The plastic surgeons where stymied as to what to do: Accept the shortening and use 1 cm of the 5-cm fibular graft to span the remaining gap, discard the fibula graft, or seek an immediate second orthopaedic surgeon’s opinion. The latter option was chosen. The vascularized fibular graft was morcellized into fine fragments and packed into the gap. A rigid external fixator was applied with transfixing pins placed in the foot and above the fracture. Hyperbaric oxygen treatments were started to manage swelling and the threatened flaps after the four-and-a-half-hour surgery. Two weeks later, the patient was returned to the operating room (OR) where the morcellized fibular graft was stretched out by readjusting the fixator. This maneuver nearly regained the full tibia length. At four weeks, the axial and rotatory alignments were “fine- tuned” in the OR to achieve an essentially normal-appearing extremity, and HBO treatments (40 in total) were stopped at casted for another six weeks, at which time x-rays showed consolidation and full incorporation of the bone graft. The patient resumed walking with a normal gait.

A final case was of a male in his late teens whose leg got caught in a rotating drum water extractor for drying rags while working at a car wash.  An open fracture with 20 or more comminuted bone fragments resulted (Figure 6). The patient refused an amputation and was transferred to our medical center for HBO treatments. Stabilization was achieved with an external fixator. At three months, the fractures had developed callus and the fixator replaced with a walking cast. At six months, bony union was achieved. Several conclusions can be drawn from these favorable outcomes using HBO as an adjunct for fracture healing. First, HBO was used in conjunction with “state-of-the-art” orthopaedic management. It should always be used as an adjunct to orthopaedic management and not a substitute for it. Hyperbaric oxygen treatments alone would not  have achieved the same results. Second, a course of HBO treatments before definitive orthopaedic interventions appeared to help the fracture site respond to the final, definitive orthopaedic intervention. Staged surgeries using HBO before surgery to generate a favorable environment at the nonunion site for bone grafting and after surgery to promote bony union is logical, based on the mechanisms of HBO.33 Third, HBO favored bone formation of the undifferentiated mesenchyme that invaded the fracture site in the fracture milieu. In this capacity, it may have acted as an inducer of osteoblastic and osteoclastic activity as well as providing an oxygen environment sufficient for these cells to function. Fourth, HBO supported angiogenesis (i.e., vascular ingrowth of the avascular bone graft scaffold) in those cases were bone grafting was used in the management. Fifth, early utilization of HBO in the course of healing of fractures that are highly predictable of complications (e.g., the first author’s cases 3, 5, and 6 described above) is supported by the above observations. Finally, HBO promotes bone remodeling by the osteoclast, which is the most metabolically active of the three bone cell types. This was apparent in cases 5 and 6 above where extensive remodeling occurred. It is also seen as a minor complication with delayed-onset stress fractures after resuming ambulation. This has been observed in about one- fifth of our patients treated for crush injuries, osteomyelitis and nonunions of long bones.

FIGURE 6. Limb threatening open fracture with recommendation for a below-knee amputation


Legend: Extremely comminuted, open mid-shaft tibial fracture secondary to twitching injury. Early wound infection. Patient referred for HBO treatments when he refused a leg amputation. Remarkable remodeling (osteoclast effect), but delayed stress fracture shown healed at 18 months.


There is much supporting information that perfusion- oxygenation is a key factor in bone formation and growth. This is observed in limb length overgrowth of children’s lower-extremity fractures and is attributed to the increased flow associated with fracture healing. For similar reasons, bone overgrowth and hypertrophy are associated with arteriovenous fistulas. The propensity for excessive bone formation is seen after muscle injury (i.e., heterotopic ossification) and after brain and spinal cord injuries.

Increased perfusion associates with both the inflammatory reaction in myositis and disruption of sympathetic nervous system vasoconstriction with the neurological conditions. Although acromegaly is attributed to pituitary tumors and increased elaboration of growth hormone, there appear to be concomitant changes in blood flow from the hormone.34 This may be the reason that increased bone mass is associated with this condition. Finally, the formation of bone spurs is possibly due to an inflammatory response increasing blood flow to bone from repetitive trauma to the site either from traction effects (spurs pointing away from the metaphyses) or over underlying bony deformities.

Additional information showing the importance of blood flow and oxygen availability for bone viability is observed in those situations in which blood supply is lost. Examples include: 1) osteonecrosis of the femoral head associated with fractures, steroids, alcoholism, and dysbarism; 2) osteoradionecrosis of the mandible due to sclerosis of blood vessels from radiation; 3) jaw-bone necrosis from bisphosphonates with failure of bone remodeling (from the osteoclast) to maintain channels for blood vessels to perfuse the mandible; 4) osteonecrosis in conjunction with osteopetrosis for similar mechanisms attributed to jaw-bone necrosis; and 5) bone death in purpura fulminans and frostbite from sludging and obstruction. Bone viability can be confirmed with the technetium nuclear medicine scan.

The approximately one in 25 fractures (4.7 %) in the USA in which healing is delayed is usually predictable. Soft tissue damage, energy exchange, comminution, displacement, contamination, interruption of vascular supply and inadequate management are obvious causes. A multicenter review of more than 300,000 fractures reported that the highest nonunion rates were in the metatarsals (15.5%), scaphoids (15.3%), tibia and fibula (14%) and femur (9%).3 For all open fractures, the nonunion rate was 10.9% and ranged from 4.4% in those patients with a single fracture to 24% in patients with 7 fractures. Odds ratios were the highest for patients on a combination of nonsteroidal anti-inflammatory drugs and opioids (1.84), management requiring surgery (1.78), open fracture (1.66), concomitant anticoagulation (1.49), high-energy injury (1.38), osteoporosis (1.24), insulin requirement (1.21) and smoking (1.20). Other factors had odds ratios less than 1.20 and included such items as use of antibiotics, use of bisphosphonates for osteoporosis, vitamin D deficiency, use of diuretics and renal insufficiency, in that order. Surprisingly age in 10-year increments from 18 to 63 did not show significant differences in rates of nonunions (range from 4.5 to 5.5%). A surprising omission of the study was absence of information on perfusion to the fracture site. Nonetheless, the five risk factors of multiple fractures, NSAIDs plus opioid use, open fractures, anticoagulation and osteoarthritis plus rheumatoid arthritis increased the risk of nonunion by 50%.3

Can information of this sort provide justification for using HBO as an adjunct for prevention of nonunions and achieve fracture healing? The answer is a qualified yes if tools currently available are utilized to justify the decision to use HBO. The first step, of course, is the clinical evaluation and assessment of perfusion directly and oxygenation indirectly. Palpation and Doppler detection of pulses provide a direct assessment of perfusion. Skin (and wound base) coloration, skin temperature and capillary refill provide indirect evidence of perfusion and oxygenation. Finally, juxta-wound transcutaneous oxygen measurements provide objective information as to tissue oxygenation.35

Two user-friendly clinical tools can supplement the fracture evaluation information and help justify decision-making regarding salvage or amputation of the lower limb nonunion. These tools are the Wellness and Goal Scores that are also found in our osteomyelitis foot wound chapter in the fourth edition of Hyperbaric Medicine Practice (Tables 3 and 4). Each is a 0-to-10 score based on five assessments graded with objective findings from 2 points (best) to 0 points (worst). Scores on these two tools when each is 5 points or greater support the decision to do everything possible to heal the nonunion. If ischemia-hypoxia is associated with the nonunion, and the factors having the highest odds ratios for nonunion (i.e., NSAIDS plus opioid use, operating room treatment, open fracture, anticoagulation and high-energy injury) are present, we feel there is substantial justification for using HBO as an adjunct to other orthopaedic interventions. Even if orthopaedic interventions were used before and failed to resolve the nonunion, repeat bone grafting, callotaxis, use of bone morphogenetic protein, electrical stimulation and/or microvascular bone transplants should be utilized in conjunction with HBO treatments when the indications from the Wellness and Goal Score tools support the decision to heal the nonunion.

Unfortunately, there is minimal clinical information to support the use of HBO for fracture healing (Boauchour et al., Chen ,Yu and the first author’s anecdotal observations).28,30,31  Mitigation of hypoxia was the design for most of the animal studies dealing with fracture healing.

Impairment of oxygenation on a perfusion basis in tibial nonunions has been documented with transcutaneous oxygen measurements.36 Information such as this supports the use of HBO since hyperoxygenation of plasma and tissue fluids is a primary mechanism of HBO.33 When a score we devised titled Rational Indications for Using HBO is applied to fracture healing, only modest justification (i.e., 4 points out of 10 possible points) is summated (Table 5). Obviously, new animal studies using perturbations from the clinical literature associated with nonunions and randomized controlled clinical trials are needed before HBO can be unequivocally recommended for this problem.


While hyperbaric oxygen is not an approved use for nonunions, its mechanisms support its use when ischemia- hypoxia is contributing to the failure of bone union. Identifiable clinical factors, juxta-wound oxygenation measurements and imaging studies can confirm whether or not perfusion-oxygenation is adequate to the bone at the nonunion site. User-friendly tools such as the Wellness and Goal Scores support decisions to salvage or amputate the nonunion. Although there is only sparse clinical information to support the use of HBO as an adjunct to fracture healing, the laboratory studies consistently validate the need for oxygen in fracture healing in general and the nonunion in particular. Even though only one in 25 fractures (300,000) of the 6.3 million fractures that occur in the USA each year go onto nonunions, they present major treatment and financial challenges. When “tried-and-true” orthopaedic interventions such as bone grafting, optimal stabilization, bone morphogenetic protein, callotaxis and pulsatile electrical on sonographic bone stimulation have not resolved the nonunion and there is justification to avoid a lower limb amputation, then HBO should be utilized in the management as an adjunct to one or more of the above orthopaedic techniques. There is no information, study, or observation where bone perfusion- oxygen is not a primary consideration in fracture healing.


  1. Weaver LK. Undersea and Hyperbaric Medical Society Hyperbaric Oxygen Therapy Indications, 13th ed. Best Publishing Company, North Palm Beach, FL; 2014.
  2. Physical Fields-OrthoInfo - AAOS. Accessed August 26, 2016 from
  3. Zura R, Ziong Z, Einhorn T, et al. Epidemiology of fracture nonunion in 18 human bones. JAMA Surg. Published online September 07, 2016. Doi: 10.1001/jamasurg.2016.2775.
  4. Bassett CA, Herrmann I. Influence of oxygen concentration and mechanical factors on differentiation of connective tissues in vitro. Nature. 1961; 190:460-461.
  5. Johnson LC. Kinetics of Osteoarthritis. Lab Invest. 1959; 8:1223-1241.
  6. Ham AW. Cartilage and Bone. Special Cytology. 2nd ed. New York, NY; 1932.
  7. Kolodny AW. Cartilage and Bone. Special Cytology. 2nd ed. New York, NY; 1932.
  8. Truta J, Cavadias AX. Vascular changes caused by the Kuntscher type of nailing: an experimental study in the rabbit. J Bone Joint Surg Am. 1955; 37-B(3):492-505.
  9. Gothman L. Arterial changes in experimental fractures of the monkey’s tibia treated with intramedullary nailing. A microangiographic study. Acta Chir Scand. 1961; 121:56-66.
  10. Wray JB, Goodman, HO. Post-fracture vascular phenomena and long- bone overgrowth in the immature skeleton of the rat. J Bone Joint Surg Am. 1961; 43-A:1047-1055.
  11. Rhinelander FW, Baragry R. Microangiography in bone healing. I. Undisplaced closed fractures. J Bone Joint Surg Am. 1962; 44-A:1273- 1298.
  12. Cavadias AX, Trueta J. An experimental study of the vascular contribution to the callus of fracture. Surg Gynecol Obstet. 1965; 120:731-747.
  13. Gothman L. Vascular reactions in experimental fractures. Microangiographic and radioisotope studies. Acta Chir Scand Suppl. 1961; Suppl 284:1-34.
  14. Vaes GM, Nichols G Jr. Oxygen tension and the control of bone cell metabolism. Nature. 1962; 193:379-380.
  15. Sledge CB, Dingle JT. Oxygen-induced resorption of cartilage in organ culture. Nature. 1965; 205:140-141.
  16. Coulson DB, Ferguson AB Jr, Diehl RC Jr. Effect of hyperbaric oxygen on the healing femur of the rat. Surg Forum. 1966: 17:449-450.
  17. Wray JB, Rogers LS. Effect of hyperbaric oxygenation upon fracture healing in the rat. J Surg Res. 1968: 8(8):373-378.
  18. Yablon IG, Cruess RL. The effect of hyperbaric oxygen on fracture healing in rats. J Trauma. 1968; 8(2):186-202.
  19. Laurnen EL, Kelly PJ. Blood flow, oxygen consumption, carbon- dioxide production, and blood-calcium and pH changes in tibial fractures in dogs. J Bone Joint Surg Am. 1969; 51(2):298-308.
  20. Niinikoski J, Penttinen R, Kulonen E. Effect of hyperbaric oxygenation on fracture healing in the rat: a biochemical study. Calcif Tissue Res. 1970; Suppl:115-116.
  21. Penttinen R, Niinikoski J, Kulonen E. Hyperbaric oxygenation and fracture healing. A biochemical with rats. Acta Chir Scand. 1972; 138(1):39-44.
  22. Brighton CT, Krebs AG. Oxygen tension of healing fractures in the rabbit. J Bone Joint Surg Am. 1972; 54(2):323-332.
  23. Gray DH, Hamblen DL. The effects of hyperoxia upon bone in organ culture. Clin Orthop Relat Res. 1976; (119):225-230.
  24. Heppenstall RB, Goodwin CW, Brighton CT. Fracture healing in the presence of chronic hypoxia. J Bone Joint Surg Am. 1976; 58(8):1153-1156.
  25. Lindstrom T, Gullichsen E, Lertola K, Niinikoski J. Effects of hyperbaric oxygen therapy on perfusion parameters and transcutaneous oxygen measurements in patients with intramedullary nailed tibial shaft fractures. Undersea & Hyperbaric Medicine. 1998; 25(2):87-9.
  26. Porcellini M, Bernardo B, Capasso R, Bauleo A, Baldassarre M. Combined vascular injuries and limb fractures. Minerva Cardioangiologica. 1997;45(4):131-8.
  27. Braune C, Hamm J, Bohmer D, Scale D, Zichner L. Hyperbaric oxygenation as a successful therapeutic approach in oral wound dehiscence after operative stabilization of an unstable post- traumatic odontoid non-union. Archives of Orthopaedic and Trauma Surgery. 2002; 122(2):115-9.
  28. Bouachour G, Cronier P, Gouello JP, et al. Hyperbaric oxygen therapy in the management of crush injuries: A randomized double-blind placebo-controlled clinical trial. J Trauma. 1996; 41:333-339.
  29. Bennett MH, Stanford R, Turner R. Hyperbaric oxygen therapy for promoting fracture healing and treating fracture non-union (Cochrane Review). In: The Cochrane Library, Issue 1, 2006. Oxford: Update Software.
  30. Chen Guo-ping. Effect of elastic stable intramedullary nail combined with hyperbaric oxygen on children with long bone fractures. Chinese Journal of General Practice. 2012.
  31. Yu Wenzhong. The effects of hyperbaric oxygen therapy on the treatment of adult patients with delayed union of tibial fracture. China Continuing Medical Education. 2014; 6(8):61-63.
  32. Strauss MB, Lu LQ, Miller SS, Hart BB. Hyperbaric Oxygen for the Management of Chronic Refractory Osteomyelitis. In: Hyperbaric Medicine Practice, 4th edition, Kindwall (Ed.). In Press.
  33. Strauss MB, Hart GB, Lu LQ. Role of Oxygen and Hyperbaric Oxygen Mechanisms. In: Hyperbaric Medicine Practice, 4th edition, Kindwall (Ed.). In Press.
  34. Kapitola J, Marek J, Jaboda I, et al. Decreased 133Xe clearance in the proximal femur in acromegaly. Exp Clin Endocrinol. 1986; 87(1):112- 115.
  35. Moon H, Strauss MB, La S, Miller SS. The validity of transcutaneous oxygen measurements in predicting healing of diabetic foot ulcers. Undersea Hyperb Med. 2016; 43(6):671-678.
  36. Kwong LM, Johanson PH, Klein SR. Postocclusive transcutaneous oximetry in followup assessment of tibial nonunion and healed tibia fractures. J Trauma. 1988; 28(7):947-954.


About the Author


MICHAEL STRAUSS, M.D., has been a regular contributor to WCHM with almost 30 articles and citations to date. He is appreciative of Dr. Whelan’s (the editor of the 4th edition of Hyperbaric Medicine Practice) approval to share with the readers of WCHM at this time the fracture-healing chapter from this soon-to-be-published text. As an orthopaedic surgeon, Dr. Strauss is understandably well versed in fracture healing. Even during his orthopaedic residency, Dr. Strauss was able to use his background in hyperbaric and undersea medicine to arrange treatments for several tibial fracture nonunions (see text). His integration of hyperbaric oxygen therapy to the subject of fracture healing is a singular contribution to the hyperbaric medicine literature.



ANNA M. TAN, DPM, is the chief resident at the Long Beach Memorial Medical Center podiatric medicine and surgery program. She has been collaborating with Dr. Strauss on the five-part WCHM series “The Prevention of New and Recurrent Foot Wounds.”Dr. Tan will be the first author in the upcoming fourth part of the series with her article titled “Protective Footwear.” In addition to contributing to this fracture-healing chapter for the 4th edition of Hyperbaric Medicine Practice, she has also coauthored with Dr. Strauss the chapter “Hyperbaric Oxygen in the Evaluation and Management of Diabetic Foot and Other Wounds.” She has a special interest working with patients with challenging foot and ankle problems, and in her spare time enjoys cooking, traveling and Bikram yoga.



LIENTRA LU is a research coordinator at the VA Medical Center in Long Beach, California, under the guidance of Dr. Ian Gordon, a vascular surgeon, and Dr. Michael Strauss, the first author of this paper. She is also working with Dr. Strauss and Dr. Tan on the series of five articles, “The Prevention of New and Recurrent Foot Wounds.” Miss Lu is the first author of the third paper, titled “Evaluation and Management of Foot Skin and Toenails,” also in this publication. Additionally, she has coauthored with Dr. Strauss five chapters for the 4th edition of Hyperbaric Medicine Practice, including mechanisms of hyperbaric oxygen therapy, chronic refractory osteomyelitis, diabetic foot ulcers, and crush injuries in addition to the nonhealing fracture chapter, which they were given permission to use in this edition of WCHM. Miss Lu is helping with diabetic foot and venous leg ulcer studies at the VA Medical Center while also serving as an assistant in patient care at the PAVE (Prevention-Amputation, Veterans Everywhere) Clinic there. In addition, Miss Lu is helping with Dr. Strauss’s Long Beach Wound Score Research Project. She also works with the American Red Cross in disaster preparedness.















Contact Us

Best Publishing Company
631 US Highway 1, Suite 307
North Palm Beach, FL 33408

This email address is being protected from spambots. You need JavaScript enabled to view it.



Copyright © 2018 Best Publishing Company, a company of WCHMedia Group, Inc | All rights reserved
Find more information at