Review

Wound healing and its impairment in the diabetic foot

Vincent Falanga
Lancet 2005; 366: 173643 Departments of Dermatology and Biochemistry, Boston University, Boston, MA, USA (Prof V Falanga MD) Department of Dermatology and Skin Surgery, Roger Williams Medical Center, Elmhurst Building, 50 Maude Street, Providence, RI 02908, USA (Prof V Falanga MD) vfalanga@bu.edu

Optimum healing of a cutaneous wound requires a well-orchestrated integration of the complex biological and molecular events of cell migration and proliferation, and of extracellular matrix deposition and remodelling. Cellular responses to inammatory mediators, growth factors, and cytokines, and to mechanical forces, must be appropriate and precise. However, this orderly progression of the healing process is impaired in chronic wounds, including those due to diabetes. Several pathogenic abnormalities, ranging from disease-specic intrinsic aws in blood supply, angiogenesis, and matrix turnover to extrinsic factors due to infection and continued trauma, contribute to failure to heal. Yet, despite these obstacles, there is increasing cause for optimism in the treatment of diabetic and other chronic wounds. Enhanced understanding and correction of pathogenic factors, combined with stricter adherence to standards of care and with technological breakthroughs in biological agents, is giving new hope to the problem of impaired healing. The healing of a wound requires a well orchestrated integration of the complex biological and molecular events of cell migration, cell proliferation, and extracellular matrix (ECM) deposition. Cellular responses to inammatory mediators, to growth factors and cytokines, and to mechanical forces must be appropriate and precise. These fundamental processes are similar to those guiding embryogenesis, tissue and organ regeneration, and even neoplasia.14 However, denite differences exist between adult wounds and these other systems. In cutaneous injuries that heal readily and do not have an underlying pathophysiological defect (acute wounds), the main evolutionary force may have been to achieve repair quickly and with the least amount of energy. Hence, such wounds heal with a scar and no regeneration. In wounds with preexisting pathophysiological abnormalities (chronic wounds, such as diabetic ulcers), evolutionary adaptations have probably not occurred; impaired healing is the result. However, there is much cause for optimism for the treatment of chronic wounds, because of tremendous strides in our scientic understanding of the repair process and how that knowledge can be used to develop new approaches to treatment. time after injury. Wounds that are restricted to the supercial layer of the dermis (partial-thickness wounds) still have a reservoir of keratinocytes in the hair follicles and other skin appendages left in the wound bed, and thus can heal both from the edges and from within the wound. Conversely, full-thickness wounds can only heal from the edges, and contraction plays an important mechanism for wound closure in these deeper wounds.5,6

Events and phases of wound healing

The fundamental biological and molecular events after cutaneous injury, with information mainly derived from experimental wounds in animals, cannot be separated and categorised in a clear-cut way. However, it has been useful to divide the repair process into four overlapping phases of coagulation, inammation, migration-proliferation (including matrix deposition), and remodelling. These phases are shown in gure 1, which also highlights the main events during each phase and the key types of cells implicated. Whereas acute wounds go through the linear progression of overlapping biological and molecular events illustrated in gure 1, chronic non-healing wounds do not. Some areas of chronic wounds are in different phases at the same time and, presumably, progression to the next phase does not occur in synchrony. These overall differences between acute and chronic wounds are not restricted to lack of progression alone. Certain events occur abnormally in the healing-impaired wound, highlighting the need to be cautious in extrapolating lessons learned from acute wounds to the situation in chronic wounds. Studies in animals have shown that isolated abnormalities can markedly modulate the healing process.7 Impaired healing is found in mice with combined deciency of molecules that have a critical role in inammation (E-selectins and P-selectins), and in mice without plasminogen, urokinase plasminogen activator, and tissue plasminogen activator (double knockout), broblast growth factor-2 (basic broblast growth factor), or inducible nitric oxide.1,2,7 Conversely, decreased healing occurs in transgenic mice overexpressing some tissue metalloproteinases (eg, matrix metalloproteinase [MMP]1) and antisense to CD44, the receptor for hyaluronic acid.7 Unexpectedly, some mutations lead to accelerated healing,
www.thelancet.com Vol 366 November 12, 2005

Basic aspects of normal wound healing

The type, size, and depth of cutaneous injury have important implications for events at the cellular and molecular level. Scalpel injury (ie, after surgical procedures) causes less overall and diffuse tissue damage than burns or radiation, can be primarily closed (by suture), and generally results in less scarring. Small and supercial cutaneous defects can resurface mainly by epidermal migration, and do not have to rely on actual keratinocyte proliferation and its more substantial lag
Search strategy and selection criteria I searched PubMed by matching wound healing and wounds with the search terms keratinocytes, diabetes, hemidesmosomes, integrins, MMPs, contraction, neuropathic ulcers, gene therapy, stem cell therapy, growth factors, tissue engineering. I mainly selected publications from the past 6 years. Relevant articles and book chapters were also included. No restriction was applied on language of publication.

1736

Review

as reported with Smad-3 or skn-1a knockout mice.8 These ndings offer the promise of improving healing in human beings, by manipulating growth factors, ECM, and signalling pathways.

Time

Phases Coagulation Fibrin plug formation, release of growth factors, cytokines, hypoxia

Main cell types Platelets

Specific events Platelet aggregation and release of fibrinogen fragments and other proinflammatory mediators

Phases 1 and 2: coagulation and inammation

The different phases of wound healing not only overlap, but also have ramications beyond their more obvious immediate purposes. Soon after injury, a brin plug forms and inammatory cells are quickly recruited to the wound. Coagulation is needed for haemostasis and wound protection. The brin plug consists of platelets embedded in a meshwork of mainly polymerised brinogen (brin), bronectin, vitronectin, and thrombospondin; it is an immediate way to ward off bacteria and provide temporary wound coverage,3,7 but also has other roles. During their incorporation within the plug, platelets aggregate and release a wide range of growth factors, including platelet-derived growth factor (PDGF) and transforming growth factor (TGF) 1.1,3 These and other growth factors, the activation of which also depends on pH and other parameters within the injured tissue, have an early role in cell recruitment and a later one in ECM formation.7 As another example of multiple effects, thrombin polymerisation of brinogen to brin yields fragments, such as brinopeptides A and B, which can recruit inammatory cells to the wound.3 Then, through the endothelial expression of selectins, leucocytes are slowed down in the bloodstream enough that stronger forces generated by binding to integrins will help their movement through endothelial gaps and into the extracellular space (diapedesis).2 Again, these inammatory cells recruited to the wound have several purposes. Neutrophils and macrophages, whose function is impaired in diabetes, aid in wound debridement. However, both cell types produce several key growth factors and mediators that keep fuelling the repair process: for example, connective tissue growth factor was rst identied in neutrophils.3 Immediately after injury, the wound is hypoxic because of damage to the blood vessels. This seemingly deleterious situation has some benecial effects, and might help prepare for the next phase of healing. Hypoxia increases keratinocyte migration, early angiogenesis, proliferation and clonal expansion of broblasts, and the transcription and synthesis of crucial growth factors and cytokines, including PDGF, vascular endothelial growth factor, and TGF 1.9 Later, within the next 23 days, inammatory and dermal cells recruited to the injury site produce a powerful armamentarium of growth factors and cytokines.7 Circulating monocytes take up residence at the injury site as tissue macrophages, and so do broblasts and endothelial cells as they form the early granulation tissue that begins the process of contraction.
Hours

Neutrophils, monocytes Inflammation Cell recruitment and chemotaxis, wound debridement Selectins slow down blood cells and binding to integrinsdiapedesis Macrophages Hemidesmosome breakdownkeratinocyte migration

Days Migration/proliferation Epidermal resurfacing, fibroplasia, angiogenesis, ECM deposition, contraction Keratinocytes, fibroblasts, endothelial cells

Myofibroblasts Weeks to months Remodelling Scar formation and revision, ECM degradation, further contraction and tensile strength

Cross-talk between MMPs, integrins, cells, cytokinescell migration, ECM production

Phenotypic switch to myofibroblasts from fibroblasts

Figure 1: Phases of wound healing, major types of cells involved in each phase, and selected specic events

wound closure needs to be addressed. Formation of ECM proteins, angiogenesis, contraction, and keratinocyte migration are essential components of these phases. Matrix proteins, including collagens, bronectin, and vitronectin, provide substrates for cell movement, vehicles for changing cell behaviour, and structures that return function and integrity to the tissue.10 Angiogenesis makes possible the re-supply of oxygen and other nutrients. Contraction, aided by the formation of ECM, granulation tissue, and the emergence of myobroblasts, is a rapid and efcient way of achieving wound closure. The balance between contraction and keratinocyte-dependent closure has much to do with the depth and location of the wound and the presence of complications due to infection, and seems to be impaired in diabetic wounds. Another critical balance is the deposition, persistence, and dynamic remodelling of ECM proteins. Excessive deposition of some matrix proteins, such as collagens and bronectin, has been reported in diabetic wounds.9

The role of integrins

As shown in gure 1, cell movement is critical for broplasia, angiogenesis, and keratinocyte-dependent wound closure. Keratinocytes need to migrate through or below the brin meshwork, and broblasts and endothelial cells are recruited to the nascent granulation tissue. At this time, MMPs and other enzymes (tissue plasminogen activator and urokinase plasminogen activator) are needed to free cells and structures from their more stable surroundings.7,11 Integrins are also essential, because they represent the language by which cells communicate with the matrix and with each other.
1737

Phases 3 and 4: migration-proliferation and remodelling

As the inammatory phase of wound healing is toned down (gure 1), wound contraction begins, but stable
www.thelancet.com Vol 366 November 12, 2005

Review

Integrins are and transmembrane cell-surface receptors that bind the ECM to cytoskeletal structures. There are at least 24 heterodimers, which are formed from a pool of 18 different and eight subunits. For cells to move, they have to be freed from their stable conguration and location in the tissue.11 Dermal broblasts, which up to 34 days after wounding are in a resting state with dermal collagen, undergo a switch from 2 to 3 and 5 integrin subunits, the latter being more efcient in negotiating the migration of broblasts through the early brin-rich matrix. Basic broblast growth factor is crucial to angiogenesis and can induce vascular endothelial growth factor, the t-1 receptors of which are upregulated on endothelial cells and are decreased in wounds with impaired healing. Again, several interactions are implicated. For example, endothelial cells are incapable of responding to angiogenic stimuli without the expression of v 5 integrin. During migration, endothelial cells exhibit high-afnity forms of v 5.7,9,11

ultimately lead to induction of MMP1 (collagenase 1 or interstitial collagenase). Acting like molecular scissors, the MMPs help regulate matrix degradation and cellular movement. Type 4 and type 7 collagen, essential components of the basement membrane and anchoring brils, are cut by MMP9. Other non-collagenous matrix components are degraded by MMP10 (stromelysin-2).3,11,16

Keratinocyte proliferation
Keratinocytes start migrating to ll the wound defect within a few hours after injury. A keratinocyte proliferative burst then occurs, which is especially important for larger wounds where migration of cells alone is insufcient to close the defect.2 Keratinocyte proliferation also involves regulation of TP53, CKAP4, and TP73 tumour suppression genes, epidermal growth factor, and TGF , among other signals. Fibroblast and keratinocyte proliferation rely in part on the TGF -related activins.17

Wound contraction and ECM remodelling

Although keratinocyte migration is very important in wound closure, events leading to angiogenesis and wound contraction play a major part in both acute and chronic wounds. Failure of timely and rapid contraction seems to be a major problem in diabetic ulcers, for example. Within a week from injury, several events have occurred, including matrix deposition, aided by such growth factors as PDGF, TGF , broblast growth factors, and vascular endothelial growth factor, phenotypic changes in broblasts to myobroblasts (TGF -induced), and early remodelling.3 Thus, overall, ECM is formed, providing initial support and a conduit for cell migration, and begins to be degraded, as a result of serine proteases and MMPs. Sequential deposition of collagens, rst type 3 and then type 1, and their hydroxylation peak at about 3 weeks. The wound continues to contract, with maximum tensile strength being 60% of the previously unwounded skin.3,6

Keratinocyte migration
The disassembly of hemidesmosomes, which provide anchorage of the basal keratinocytes to the underlying basement membrane, is a good example of a highly organised structure being torn down for the purpose of cell migration. This disassembly and keratinocyte migration require cross-talk between growth factors, MMPs, integrins, and structural proteins. Among critical factors affecting hemidesmosome disassembly are the unravelling of laminin-5 binding to 6 4 integrin, receptor clustering, interactions of integrins with ECM components, formation of lamellipodia needed for cell movement,12,13 the molecular switches GTPases (Rho, Rac, Cdc42),13,14 and the state of phosphorylation of the integrin subunits. For example, a shift from the stable and resting assembly of laminin-5 with 6 4 is caused by phosphorylation of this integrin heterodimer, causing binding to the unlocked 3 1 integrin and facilitation of lamellipodia formation and keratinocyte movement.13 In addition to lamellipodia extension, basal keratinocytes leapfrog over the basal cells near the wound. Interestingly, in the embryo and probably in the adult cornea, a pursestring mechanism for wound closure is operative, so that an actin cable forms within minutes, followed by keratinocytes being pulled together.1 Kinases, such as mitogen-activated protein kinase, are activated in basal and suprabasal keratinocytes by further action of integrins or the release of interleukin 1 . Calcium concentrations and entry into the cells also have a central role in migration, proliferation, and differentiation.15 MMPs and other enzymes are important components of the wound that facilitate cell movement and the eventual remodelling of ECM. To negotiate the brin clot, keratinocytes need to upregulate tissue plasminogen activator and urokinase plasminogen activator. The interactions between 3 1, keratinocytes, and collagen
1738

Impaired healing: the diabetic ulcer

The linear progression paradigm for normal wound healing shown in gure 1 has been highly valuable in understanding the basic biology of tissue repair. However, one should not oversimplify. Even during the normal process of wound healing complications can occur, including infection, thrombosis, and ischaemia. Also, lessons learned from experimental models, on which gure 1 and the previous discussion are based, cannot be completely extrapolated to the situation encountered in diabetic wounds. There, intrinsic pathobiological abnormalities and extrinsic factors contribute to an even more complex wound microenvironment. Unfortunately, a valid model of chronic wounds in animals has not yet been developed. Clinical and experimental evidence suggests that diabetic ulcers and other types of chronic wounds do not follow an orderly and reliable progression of wound healing. Parts of the chronic wound may be stuck in different phases, having lost the ideal synchrony
www.thelancet.com Vol 366 November 12, 2005

Review

of events that leads to rapid healing.18,19 In the case of diabetic ulcers, healing impairment is caused by several intrinsic factors (neuropathy, vascular problems, other complicating systemic effects due to diabetes) and extrinsic factors (wound infection, callus formation, and excessive pressure to the site). Traditionally, this set of predisposing abnormalities in diabetes has been referred to as the pathogenic triad of neuropathy, ischaemia, and trauma. However, this too is an oversimplication. One pathogenic abnormality can lead to another, and vicious cycles of pathogenicity develop in the diabetic foot. Moreover, the triad does not specically mention infection, which plays a major role in healing impairment, hospitalisation, and high incidence of limb loss.20,21 Diabetic ulcers are also quite heterogeneous, depending on the underlying predominant abnormality. In a sense, no diabetic ulcer is completely pure from a pathogenic standpoint. An ulcer can be mainly attributed to vascular occlusion or neuropathy. However, neuroischaemic ulcers are common, if not the rule.22 Moreover, infection, the location of the ulcer, and foot deformity and callus have to be factored in. For this reason, developing an adequate and universally agreed classication of diabetic ulcers from a pathogenic standpoint has been a challenge. An extremely useful series of publications and updates have come from consensus statements by an international working group on diabetes, which highlights these important considerations and the need to properly dene abnormalities and classications.2325

denervation and autonomic neuropathy, lead to the maldistribution of bloodow. Poly (ADP-ribose) polymerase (PARP), a nuclear enzyme responsive to oxidative DNA damage, can also lead to cell necrosis and changes in microcirculatory reactivity. In diabetic neuropathy, the neurovascular response, dependent on the C-nociceptive nerve bres and adjacent C bres, is impaired, leading to defects in the secretion of substance P, calcitonin generelated peptide, and histamine. Hence, vasodilatation is impaired, particularly in situations of stress from trauma and pressure.30,31 The assessment of bloodow in the diabetic foot is complicated by the presence of medial calcication, which renders simple measurement of ankle-brachial pressure index unreliable.32 Thus, non-invasive assessment for vascular disease requires other tests, including absolute ankle pressure, toe pressure measurements, and colour duplex ultrasonography.22 Measurements of transcutaneous oxygen, especially around the wound, might also be helpful from a prognostic standpoint.33

Neuropathy
Some neuropathological problems have already been mentioned, as they are tied to microcirculatory defects. Motor, sensory, and autonomic bres are all affected. The consequences are predictable. Because of sensory decits, the diabetic patient does not have protective symptoms guarding against pressure and heat. Thus, trauma can initiate the development of an ulcer. Absence of pain, probably combined with abnormal vasodilatory autoregulation, contributes to the pathogenesis of Charcot foot, which further impairs the ability to sustain pressure. Similarly, the addition of motor bre abnormalities leads to undue physical stress on the insensate foot, the development of further anatomical deformities (arched foot, clawing of toes), and might play a part in the development of infection, since bacterial growth is enhanced in tissues with high compressive forces.20,22,3436 Although intuitively correct, the link between glucose control and the development or stabilisation of neuropathic abnormalities is not absolutely proven. This type of evidence would necessitate randomised prospective trials that would clearly not be ethical. However, longterm trials aimed at determining the relation and correlations in a large cohort of patients might be possible. Measurements with the Semmes-Weinstein 10 g nylon monolament can assess protective sensation. Vibratory sensation can be measured with a biothesiometer.34,35

Vasculopathy and endothelial cell abnormalities

Patients with diabetes, particularly those with type 1, have more macrovascular disease than non-diabetic people, with more distal distribution from the supercial femoral artery to the pedal arch and involvement of the metatarsal artery.26 Microcirculatory deciencies occur early in diabetes. These abnormalities include a reduction of capillary size, thickening of the basement membrane, and arteriolar hyalinosis. The thickening of the basement membrane interferes with physiological exchanges, and leads to altered migration of leucocytes (contributing to infection), decreased maximal hyperaemia, and abnormal autoregulatory capacity.27,28 Impaired endothelial function might involve a reduction of nitric oxide synthetase. Importantly, the lumen of microvessels is not decreased in diabetes.29 The long-standing myth of small-vessel disease accounted for the unfortunate and incorrect notion that revascularisation would not help diabetic patients. Nevertheless, although true luminal occlusion of small blood vessels does not occur, bloodow is maldistributed. Abnormal bloodow might also explain the development of Charcot foot, which results in dramatic changes in bone alignment and great susceptibility to pressure forces in the insensate foot. Clear links exist between vasculopathy and neuropathy in the diabetic foot. Shunts in the microcirculation, together with the presence of sympathetic nerve
www.thelancet.com Vol 366 November 12, 2005

Infection
Infection is not a stated component of the pathogenic triad for development of diabetic foot ulcers, but is an extremely important cause of morbidity and hospitalisation, amputation, and impaired healing. Whether it has a role in the initial development of the ulcer, especially when combined with trauma, is unclear. There are several reasons for the increased incidence of infection in the
1739

Review

diabetic foot compared with other types of chronic wounds. The role of stress and compressive forces favouring overgrowth of bacteria has been mentioned,36 as has decreased function of macrophages and neutrophils.37,38 However, it is the combination of factors, including vascular abnormalities, that has the essential role in this major complication. Infection can spread rapidly in diabetic ulcers. Limb-threatening cellulitis, abscesses, and osteomyelitis need immediate attention. High bacterial burden without the classic signs of infection is also detrimental to healing.39 The presence of bacterial biolms in diabetic wounds is still speculative. As discussed earlier, temporary hypoxia after injury may be benecial in stimulating cell movement, angiogenesis, and production of growth factors. However, prolonged hypoxia is detrimental, in part by exaggerating these early physiological events and by causing reperfusion injury and the formation of oxygen radicals. Together with hyperglycaemia and other metabolic effects of diabetes, hypoxia adversely affects neutrophil and macrophage function.40

Debridement: multiple benecial effects

Proper debridement involves removal of the necrotic wound bed and callus, as the latter can contribute to increased pressure on the insensate foot.41,42 In a retrospective analysis, debridement increased the therapeutic effect of topically applied PDGF.42 However, we have now begun to realise that debridement, by removing diseased tissue, actually corrects several cellular and molecular abnormalities. One hypothesis is that debridement resets the stage for proceeding towards the normal wound healing sequence.9,43 Several observations and mechanistic reports lend support to this still unproven view. Diabetic ulcers appear to be stuck in the proliferative phase, with an excess of matrix proteins, including bronectin.19 Thus, remodelling or turnover of matrix might be inadequate, which ultimately affects cell migration and probably the stability of the healed wound. These abnormalities could have consequences for growth factors, which can become trapped and unavailable for the healing process.44 Directly or indirectly, hyperglycaemia alters the balance of MMP concentrations and proteolytic activity. Diabetes is associated with decreased concentrations of urokinase plasminogen activator and increased tissue plasminogen activator inhibitor, a situation that might result in decreased brinolysis and impaired matrix deposition.45

replicative senescence, but are perhaps caused by more complex interactions between the resident cells and the chronic wound.49 Chronic wound broblasts do show decreased expression of type 2 TGF receptors, with impaired phosphorylation of transduction signals, including Smad2, Smad3, and mitogen-activated protein kinase.50 Although much more work is needed to clearly dene the phenotypical abnormalities in diabetic wound cells, these ndings have clear-cut implications for therapeutic intervention. For example, growth factors delivered in a simplistic topical approach might not nd a regularly suitable and responsive target cell population. Other cellular abnormalities exist in diabetes. Macrophages in diabetes show a decrease in release of cytokines, including tumour necrosis factor , interleukin 1 , and vascular endothelial growth factor.38 Excessive activation of some MMPs, such as MMP9, can impair cell migration and lead to breakdown of some necessary matrix proteins and growth factors.51 Although there is no direct evidence that the proliferative activity of keratinocytes is affected in diabetes, migration may well be impaired and studies with cells from diabetic wounds are needed.52 Therefore, going back to the original premise mentioned earlier, it is possible that proper debridement of diabetic ulcers corrects many more subtle abnormalities, at least partly, by removal of altered resident cells and matrix material.

Correcting impaired healing

Is the diabetic foot ulcer truly a chronic wound, and is impaired healing simply the result of failure to provide timely treatment for the ulcer or due to poor patient compliance? This point of view might apply to purely neuropathic ulcers, in which ofoading alone can lead to rapid healing. However, as stated earlier, diabetic ulcers are heterogeneous. The treatment and the outcome depend very much on the presence or extent of arterial insufciency, the degree of neuropathy, the ulcer location, presence of Charcot deformity, and the persistent propensity to infection. Figure 2 provides a guideline for the approach to diabetic ulcers. Thorough assessment of the patient and the wound is crucial, as is the immediate need to control glucose concentrations, treat infection, and correct perfusion abnormalities. Ofoading is crucial, and has been the subject of much discussion. Areas of controversy exist. For example, a recent systematic review53 favours the use of hyperbaric oxygen in the treatment of diabetic foot ulcers. However, even that review admits to methodological problems and the need for further studies. Similarly, although moist wound healing is widely practised in the management of other types of chronic wounds, the answer in diabetic ulcers is more difcult; a more delicate balance may be needed to avoid maceration of tissues while promoting conditions that prevent eschar formation and facilitate cell migration within the wound.54 Control of oedema and removal of exudate are important. There is
www.thelancet.com Vol 366 November 12, 2005

Wound cell abnormalities

Very importantly, some of the resident cells in diabetic ulcers become phenotypically altered. Fibroblasts isolated from diabetic foot ulcers are probably senescent and show a decreased proliferative response to growth factors.18 Similar studies in other types of chronic wounds are in agreement with these ndings, having shown decreased broblast response to TGF 1,46 platelet-derived growth factor,47 and other cytokines.48 Evidence suggests that phenotypic changes in wound cells are not due only to
1740

Review

now some evidence that negative wound pressure may lead to faster healing.55 The diabetic wound needs to be assessed both from the pathogenic standpoint and for its extent. The Wagner classication can be used to assess extent.56 Surgical intervention, including vascular reconstruction and debridement may be required immediately. Ways to achieve the optimum wound bed have been discussed.43 Figure 2 addresses optimisation of the wound bed as good clinical practice (a term adopted by regulatory agencies overseeing clinical trials that covers both evidence-based approaches and widely accepted ones that have yet to be proven conclusively). In previous work I have referred to some of these steps as wound bed preparation.43 The remainder of the paradigm shown in gure 2 gives suggestions for when to use more advanced therapies. Even after complete wound closure, constant vigilance is required, in terms of glucose control, daily attention to any breaks in the skin, and ofoading. Preventing wound recurrence is of critical importance.

controls.62 Notably, patients in the active group had decreased incidence of osteomyelitis and amputation, possibly because of faster healing. However, the study was not initially powered to study these complications. In another 12-week randomised study with living foreskin broblasts in a vicryl mesh, incidence of complete wound closure of neuropathic foot ulcers was 30% in the active group and 18% in the control group.63 Widely differing incidences of wound closure have been reported in the control groups of these trials with PDGF and bioengineered skin. The reasons for this discrepancy are unclear. The results might reect different extents of disease and variable adherence to ofoading in different protocols and groups of investigators. The extent of debridement might also vary between different study sites. Another criticism of these trials is that optimum ofoading, by total contact casting or other variations, can presumably achieve similar or even better results.64 However, types of ofoading methods have not been
Assessment of patient Treat systemic conditions and poor nutritional status, achieve glucose control

Technological advances
Recent technological advances have led to very promising breakthroughs in the treatment of diabetic and other types of chronic wounds.57 The realisation of the crucial role of growth factors in normal wound healing has already led to the development and regulatory approval of topically applied growth factors, particularly PDGF-BB.41,58 Four placebo-controlled trials of PDGF-BB in neuropathic ulcers have been done, with the best result being a 15% increased incidence of wound closure at 20 weeks (50% healing in the growth factor-treated group).59 There is a need to improve these results with growth factors. Greater efciency of delivery of growth factors, by gene therapy or by cell therapy, is now possible and being tested.60,61 Better understanding of the phenotypic changes in resident cells, which may be unresponsive to growth factors and which were discussed earlier, may further improve the therapeutic outcome of growth factor therapy. In addition to the use of growth factors, there has also been considerable interest in the application of ECM proteins to accelerate healing of diabetic foot ulcers, including collagen and hyaluronic acid. In the future, we will probably see combination therapies of ECM with growth factors, provided we can overcome the regulatory hurdles. Growth-factor therapy requires knowledge about the dose of peptide to be used and, from a regulatory standpoint, is a challenge if multiple cytokines and growth factors are to be tested in clinical trials.9,57 Partly for these reasons, cell therapy with bioengineered skin has had recent success in both testing and results. Two main types of living bioengineered skin have been tested and proven to be effective in diabetic neuropathic foot ulcers. In a randomised 12-week trial of 208 patients with neuropathic ulcers, a bilayered construct comprising living broblasts and keratinocytes from neonatal foreskin led to complete wound closure in 56% of patients, compared with 38% in
www.thelancet.com Vol 366 November 12, 2005

Assessment of wound

Immediate considerations

Preliminary diagnosis Further assessmentwound biopsy, vascular studies, blood tests

1) Perfusion/oxygenation 2) Treat infection, abscess 3) Surgical debridement 4) Surgical evaluation

Diagnosis and further management

Good clinical practice

Debridement, improve oxygenation

Control of moisture/ exudate/oedema

Treat infection, decrease bacterial colonisation

Specific therapies off loading, surgery

Follow-up evaluation Non-healing wound Healing wound

Reassessment and other treatments

Slow or intermittent healing

Satisfactory and progressive healing

Prevention of recurrence

Healed wound

Continue present treatment

Figure 2: A management strategy for treatment of diabetic foot wounds, taking into account systemic and wound-related pathophysiological abnormalities

1741

Review

compared directly with advanced biological treatments (or even a combined approach). Still, from a therapeutic and a purely scientic standpoint, these results are important since they have shown a benecial effect of biological agents in the treatment of chronic wounds. The mechanisms of action of bioengineered skin might involve new matrix deposition, increased availability of growth factors, and perhaps recruitment of stem and progenitor cells to the wound site.65 However, strong evidence exists that true engraftment or prolonged persistence of cells from these allogeneic constructs does not occur.66,67 Ultimately, for chronic wounds to heal in a timely fashion and even be resistant to the pathophysiological forces driving their recurrence, structures may need to be reconstituted and the wound site repopulated with healthier cells. There is great interest in delivery of stem or progenitor cells, either applied topically or recruited from the circulation.68 Some preliminary work suggests that topically applied autologous bone-marrow cultured cells can heal human chronic wounds that are recalcitrant to other treatments, including growth factors and bioengineered skin.69 Recruitment of CD34 cells from the circulation have shown promise in ichaemic limbs.70 Also, CD34 cells from diabetic animals produce fewer endothelial cells than in control animals.71 However, delivery of stem cells to the wound may need to be followed by other interventions, such as split-thickness skin grafting or application of growth factors or bioengineered skin (Falanga V, unpublished). One hypothesis is that stem cells need to be directed for favourable differentiation that benets the wound. Clinical decisions about when to use advanced or more experimental therapies can be based on healing rates. Studies in venous and diabetic ulcers suggest that advancement of more than 07 mm per week is 80% sensitive and specic for eventual wound closure.72 As stated earlier, advances in the treatment of chronic wounds, particularly diabetic wounds, are promising. However, the intrinsic pathophysiological abnormalities that lead to ulceration in the rst place cannot be ignored. At the moment, no known therapy will be effective without concomitant correction of ischaemia, treatment of infection, and adequate ofoading.20,35 To address these issues, a multidisciplinary team is often needed. Still, the stage is set for incorporating in the approach to impaired wound healing the great advances in the basic sciences, embryology, regeneration, tissue engineering, and stemcell biology. These developments will also take advantage of the encouraging breakthroughs in correcting the metabolic abnormalities of diabetes. There is good reason to believe that the near future will be marked by therapeutic approaches increasingly rooted in scientic advances from the laboratory and the bedside.
Conict of interest statement In the past 3 years, I have received honoraria and research grant support from Smith & Nephew, Novartis, Organogenesis, and Johnson & Johnson.

Acknowledgments This work was funded by US National Institutes of Health grants AR42936, AR46557, and DK067836. References 1 Martin P, Parkhurst SM. Parallels between tissue repair and embryo morphogenesis. Development 2004; 131: 302134. 2 Martin P. Wound healingaiming for perfect skin regeneration. Science 1997; 276: 7581. 3 Singer AJ, Clark RA. Cutaneous wound healing. N Engl J Med 1999; 341: 73846. 4 Vascotto SG, Beug S, Liversage RA, Tsildis C. Identication of cDNAs associated with late dedifferentiation in adult newt forelimb regeneration. Dev Dyn 2005; 233: 34755. 5 Thomas DW, Harding KG. Wound healing. Br J Surg 2002; 89: 120305. 6 Ramasastry SS. Acute wounds. Clin Plast Surg 2005; 32: 195208. 7 Mehendale FMP, Martin P. The cellular and molecular events of wound healing. In: Falanga V, ed. Cutaneous wound healing. London: Martin Dunitz, 2001: 1537. 8 Ashcroft GS, Yang X, Glick AB, et al. Mice lacking Smad3 show accelerated wound healing and an impaired local inammatory response. Nat Cell Biol 1999; 1: 26066. 9 Falanga V. The chronic wound: impaired healing and solutions in the context of wound bed preparation. Blood Cells Mol Dis 2004; 32: 8894. 10 Falanga V. Physiology and pathophysiology of wound healing. In: Veves A, Giurini JM, Lo Gerfo FW, eds. The diabetic foot: medical and surgical management. Totowa, NJ: Human Press, 2002: 5973. 11 Santoro MM, Gaudino G. Cellular and molecular facets of keratinocyte reepithelization during wound healing. Exp Cell Res 2005; 304: 27486. 12 Iyer V, Pumiglia K, DiPersio CM. Alpha3beta1 integrin regulates MMP-9 mRNA stability in immortalized keratinocytes: a novel mechanism of integrin-mediated MMP gene expression. J Cell Sci 2005; 118: 118595. 13 Choma DP, Pumiglia K, DiPersio CM. Integrin alpha3beta1 directs the stabilization of a polarized lamellipodium in epithelial cells through activation of Rac1. J Cell Sci 2004; 117: 394759. 14 Danen EH, van Rheenen J, Franken W, et al. Integrins control motile strategy through a Rhocolin pathway. J Cell Biol 2005; 169: 51526. 15 Nahm WK, Philpot BD, Adams MM, et al. Signicance of N-methylD-aspartate (NMDA) receptor-mediated signaling in human keratinocytes. J Cell Physiol 2004; 200: 30917. 16 Toy LW. Matrix metalloproteinases: their function in tissue repair. J Wound Care 2005; 14: 2022. 17 Bamberger C, Scharer A, Antsiferova M, et al. Activin controls skin morphogenesis and wound repair predominantly via stromal cells and in a concentration-dependent manner via keratinocytes. Am J Pathol 2005; 167: 73347. 18 Loots MA, Kenter SB, Au FL, et al. Fibroblasts derived from chronic diabetic ulcers differ in their response to stimulation with EGF, IGFI, bFGF and PDGF-AB compared to controls. Eur J Cell Biol 2002; 81: 15360. 19 Loots MA, Lamme EN, Zeegelaar J, Mekkes JR, Bos JD, Middelkoop E. Differences in cellular inltrate and extracellular matrix of chronic diabetic and venous ulcers versus acute wounds. J Invest Dermatol 1998; 111: 85057. 20 Jeffcoate WJ, Harding KG. Diabetic foot ulcers. Lancet 2003; 361: 154551. 21 Jeffcoate WJ, Price P, Harding KG. Wound healing and treatments for people with diabetic foot ulcers. Diabetes Metab Res Rev 2004; 20 (suppl 1): S7889. 22 Vowden P VK. The management of diabetic foot ulcers. London: Martin Dunitz, 2001. 23 Apelqvist J, Bakker K, van Houtum WH, Nabuurs-Franssen MH, Schaper NC, on behalf of the International Working Group on the Diabetic Foot. International consensus and practical guidelines on the management and the prevention of the diabetic foot. Diabetes Metab Res Rev 2000; 16 (suppl 1): S8492. 24 Bakker K, Schaper NC. [New developments in the treatment of diabetic foot ulcers]. Ned Tijdschr Geneeskd 2000; 144: 40912. 25 Schaper NC, Apelqvist J, Bakker K. The international consensus and practical guidelines on the management and prevention of the diabetic foot. Curr Diab Rep 2003; 3: 47579.

1742

www.thelancet.com Vol 366 November 12, 2005

Review

26

27 28

29

30

31

32

33

34 35 36

37

38

39

40

41

42

43 44 45

46

47

48

49

Ferrier TM. Comparative study of arterial disease in amputated lower limbs from diabetics and nondiabetics (with special reference to feet arteries). Med J Aust 1967; 1: 511. Dinh T, Veves A. Microcirculation of the diabetic foot. Curr Pharm Des 2005; 11: 230109. Dinh TL, Veves A. A review of the mechanisms implicated in the pathogenesis of the diabetic foot. Int J Low Extrem Wounds 2005; 4: 15459. LoGerfo FW, Coffman JD. Current concepts. Vascular and microvascular disease of the foot in diabetes. Implications for foot care. N Engl J Med 1984; 311: 161519. Veves A, Akbari CM, Primavera J, et al. Endothelial dysfunction and the expression of endothelial nitric oxide synthetase in diabetic neuropathy, vascular disease, and foot ulceration. Diabetes 1998; 47: 45763. Decker P, Muller S. Modulating poly (ADPribose) polymerase activity: potential for the prevention and therapy of pathogenic situations involving DNA damage and oxidative stress. Curr Pharm Biotechnol 2002; 3: 27583. Mackaay AJ, Beks PJ, Dur AH, et al. The distribution of peripheral vascular disease in a Dutch Caucasian population: comparison of type II diabetic and non-diabetic subjects. Eur J Vasc Endovasc Surg 1995; 9: 17075. Pecoraro RE, Ahroni JH, Boyko EJ, Stensel VL. Chronology and determinants of tissue repair in diabetic lower-extremity ulcers. Diabetes 1991; 40: 130513. Boulton AJ, Kirsner RS, Vileikyte L. Clinical practice. Neuropathic diabetic foot ulcers. N Engl J Med 2004; 351: 4855. Brem H, Sheehan P, Boulton AJ. Protocol for treatment of diabetic foot ulcers. Am J Surg 2004; 187: 1S10S. Ctercteko GC, Dhanendran M, Hutton WC, Le Quesne LP. Vertical forces acting on the feet of diabetic patients with neuropathic ulceration. Br J Surg 1981; 68: 60814. Naghibi M, Smith RP, Baltch AL, et al. The effect of diabetes mellitus on chemotactic and bactericidal activity of human polymorphonuclear leukocytes. Diabetes Res Clin Pract 1987; 4: 2735. Zykova SN, Jenssen TG, Berdal M, Olsen R, Myklebust R, Seljelid R. Altered cytokine and nitric oxide secretion in vitro by macrophages from diabetic type II-like db/db mice. Diabetes 2000; 49: 145158. Robson MC, Steed DL, Franz MG. Wound healing: biologic features and approaches to maximize healing trajectories. Curr Probl Surg 2001; 38: 72140. Patel V, Chivukala I, Roy S, et al. Oxygen: from the benets of inducing VEGF expression to managing the risk of hyperbaric stress. Antioxid Redox Signal 2005; 7: 137787. Steed DL, for the Diabetic Ulcer Study Group. Clinical evaluation of recombinant human platelet-derived growth factor for the treatment of lower extremity diabetic ulcers. J Vasc Surg 1995; 21: 7181. Steed DL, Donohoe D, Webster MW, Lindsley L, for the Diabetic Ulcer Study Group. Effect of extensive debridement and treatment on the healing of diabetic foot ulcers. J Am Coll Surg 1996; 183: 6164. Falanga V. Classications for wound bed preparation and stimulation of chronic wounds. Wound Repair Regen 2000; 8: 34752. Falanga V, Eaglstein WH. The trap hypothesis of venous ulceration. Lancet 1993; 341: 100608. Marutsuka K, Woodcock-Mitchell J, Sakamoto T, Sobel BE, Fujii S. Pathogenetic implications of hyaluronan-induced modication of vascular smooth muscle cell brinolysis in diabetes. Coron Artery Dis 1998; 9: 17784. Hasan A, Murata H, Falabella A, et al. Dermal broblasts from venous ulcers are unresponsive to the action of transforming growth factor-beta 1. J Dermatol Sci 1997; 16: 5966. Agren MS, Steenfos HH, Dabelsteen S, Hansen JB, Dabelsteen E. Proliferation and mitogenic response to PDGF-BB of broblasts isolated from chronic venous leg ulcers is ulcer-age dependent. J Invest Dermatol 1999; 112: 46369. Stanley AC, Fernandez NN, Lounsbury KM, et al. Pressure-induced cellular senescence: a mechanism linking venous hypertension to venous ulcers. J Surg Res 2005; 124: 11217. Stephens P, Cook H, Hilton J, et al. An analysis of replicative senescence in dermal broblasts derived from chronic leg wounds predicts that telomerase therapy would fail to reverse their diseasespecic cellular and proteolytic phenotype. Exp Cell Res 2003; 283: 2235.

50

51

52

53

54

55

56 57

58

59

60

61

62

63

64

65

66 67

68 69 70

71

72

Kim BC, Kim HT, Park SH, et al. Fibroblasts from chronic wounds show altered TGF-beta-signaling and decreased TGF-beta Type II receptor expression. J Cell Physiol 2003; 195: 33136. Signorelli SS, Malaponte G, Libra M, et al. Plasma levels and zymographic activities of matrix metalloproteinases 2 and 9 in type II diabetics with peripheral arterial disease. Vasc Med 2005; 10: 16. Acikgoz G, Devrim I, Ozdamar S. Comparison of keratinocyte proliferation in diabetic and non-diabetic inamed gingiva. J Periodontol 2004; 75: 98994. Kranke P, Bennett M, Roeckl-Wiedmann I, Debus S. Hyperbaric oxygen therapy for chronic wounds. Cochrane Database Syst Rev 2004: CD004123. Hilton JR, Williams DT, Beuker B, Miller DR, Harding KG. Wound dressings in diabetic foot disease. Clin Infect Dis 2004; 39 (suppl 2): S10003. Armstrong DG, Lavery AL, for the Diabetic Foot Study Consortium. Negative pressure wound therapy after partial diabetic foot amputation: a multicentre, randomised controlled trial. Lancet 2005; 366: 170410. Frykberg RG. Diabetic foot ulcers: pathogenesis and management. Am Fam Physician 2002; 66: 165562. Bennett SP, Grifths GD, Schor AM, Leese GP, Schor SL. Growth factors in the treatment of diabetic foot ulcers. Br J Surg 2003; 90: 13346. Smiell JM, Wieman TJ, Steed DL, Perry BH, Sampson AR, Schwab BH. Efcacy and safety of becaplermin (recombinant human platelet-derived growth factor-BB) in patients with nonhealing, lower extremity diabetic ulcers: a combined analysis of four randomized studies. Wound Repair Regen 1999; 7: 33546. Wieman TJ, Smiell JM, Su Y. Efcacy and safety of a topical gel formulation of recombinant human platelet-derived growth factor-BB (becaplermin) in patients with chronic neuropathic diabetic ulcers. A phase III randomized placebo-controlled double-blind study. Diabetes Care 1998; 21: 82227. Eming SA, Krieg T, Davidson JM. Gene transfer in tissue repair: status, challenges and future directions. Expert Opin Biol Ther 2004; 4: 137386. Tinsley RB, Faijerson J, Eriksson PS. Efcient non-viral transfection of adult neural stem/progenitor cells, without affecting viability, proliferation or differentiation. J Gene Med 2005; published early online Aug 12. DOI:0.1002/jgm.823. Veves A, Falanga V, Armstrong DG, Sabolinski ML. Graftskin, a human skin equivalent, is effective in the management of noninfected neuropathic diabetic foot ulcers: a prospective randomized multicenter clinical trial. Diabetes Care 2001; 24: 29095. Marston WA, Hanft J, Norwood P, Pollak R. The efcacy and safety of Dermagraft in improving the healing of chronic diabetic foot ulcers: results of a prospective randomized trial. Diabetes Care 2003; 26: 170105. Armstrong DG, Nguyen HC, Lavery LA, van Schie CH, Boulton AJ, Harkless LB. Off-loading the diabetic foot wound: a randomized clinical trial. Diabetes Care 2001; 24: 101922. Falanga V, Isaacs C, Paquette D, et al. Wounding of bioengineered skin: cellular and molecular aspects after injury. J Invest Dermatol 2002; 119: 65360. Grifths M, Ojeh N, Livingstone R, Price R, Navsaria H. Survival of Apligraf in acute human wounds. Tissue Eng 2004; 10:118095. Phillips TJ, Manzoor J, Rojas A, et al. The longevity of a bilayered skin substitute after application to venous ulcers. Arch Dermatol 2002; 138: 107981. Conrad C, Huss R. Adult stem cell lines in regenerative medicine and reconstructive surgery. J Surg Res 2005; 124: 20108. Badiavas EV, Falanga V. Treatment of chronic wounds with bone marrow-derived cells. Arch Dermatol 2003; 139: 51016. Kudo FA, Nishibe T, Nishibe M, Yasuda K. Autologous transplantation of peripheral blood endothelial progenitor cells (CD34 ) for therapeutic angiogenesis in patients with critical limb ischemia. Int Angiol 2003; 22: 34448. Sivan-Loukianova E, Awad OA, Stepanovic V, Bickenbach J, Schatteman GC. CD34 blood cells accelerate vascularization and healing of diabetic mouse skin wounds. J Vasc Res 2003; 40: 36877. Falanga V, Sabolinski ML. Prognostic factors for healing of venous and diabetic ulcers. Wounds 2000; 12: 42A46A.

www.thelancet.com Vol 366 November 12, 2005

1743