Preparing the Wound Bed 2003: Focus on Infection and Inflammation
The Preparing the Wound Bed Concept
Wound bed preparation was first described in 2000 by Sibbald et al1 and Falanga.2 This approach to wound management stresses that successful diagnosis and treatment of patients with chronic wounds requires holistic care and a team approach. The whole patient, the underlying cause, and patient-centered concerns must be considered before looking at the wound itself (see Figure 1).
This article updates the wound bed preparation model and examines the evidence base and expert opinion regarding infection. Three components of wound bed preparation are relevant in this respect: Tissue debridement, infection/inflammation, and moisture balance. Even when these factors have been corrected, some wounds do not heal and have an abrupt or steep epidermal edge, with no migration of epidermal cells across the wound surface. These four components of wound bed preparation can easily be remembered using the mnemonic TIME - tissue, infection/inflammation, moisture, and edge effect.
To date, expert opinion has provided recommendations for preparing the wound bed. It remains the best guide because the evidence base is still weak (see Appendix 1). A template for wound bed preparation, revised since 2000, is presented in Table 1.
Scope of Review
This review examines the management of infection and inflammation in all types of chronic wounds (pressure ulcers, diabetic foot ulcers, venous and arterial leg ulcers, and inflammatory ulcers), and does not include burns, surgical, and other acute wounds. Wounds that have a suspected diagnosis of pyoderma gangrenosum or vasculitis are also included.
A number of systematic reviews have been conducted regarding the management of infection in chronic wounds. Although little of the work published in this area passes the rigorous assessment criteria set by the reviewers to direct clinicians to the most appropriate interventions, readers may be interested in noting the study conclusions. The sources below were searched for reviews on wound management within the scope of this review. All the systematic reviews listed are based on searches of the major databases, individual inquiries to non-indexed journals, searches of meetings abstracts, and personal interviews with researchers in order to identify unpublished material.
* DARE (the Database of Abstracts of Reviews of Effects). Available at www.nhscrd.york.ac.uk. A database of systematic reviews, assembled by staff at the University of York, including reviews in the Cochrane Library and elsewhere
* The Cochrane Library, free to UK users through www.nelh.nhs.uk
* TRIP (Turning Research Into Practice) through www.tripdatabase.com. It includes a wide range of UK and US clinical effectiveness resources and evidence-based guidelines
* Health Technology Assessments (HTA reviews).
In the absence of evidence that matches the gold standard of the double-blinded, randomized, controlled trial, clinicians have to rely on expert opinion and evidence available from controlled and uncontrolled studies. This review presents a consensus from two expert groups: the International Wound Bed Preparation Advisory Board and the Canadian Chronic Wound Advisory Board.
Acute and Chronic Wounds
Acute wounds are caused by external trauma and usually heal within a predictable time frame by progressing through a natural series of phases including inflammation and granulation to re-epithelialization and re-modeling. During acute wound healing, the inflammatory response is initiated by the release of cytokines and growth factors that induce vasodilatation and an increase in blood flow to the site of injury. Increased vascular permeability and an influx of phagocytic cells also occur. Antibodies trap and remove micro-organisms, foreign debris, and bacterial toxins and enzymes. The symptoms and signs of the inflammatory response are pain, erythema, swelling, and increased temperature.
By contrast, chronic wounds do not heal in a predictable fashion and are usually the result of endogenous mechanisms associated with underlying conditions. In a chronic wound infected with a persistent microbial population, the inflammatory response produces a chronic influx of neutrophils that release cytolytic enzymes, free oxygen radicals, and inflammatory mediators that injure host tissue. Localized thrombosis and vasoconstriction lead to tissue hypoxia, a condition that promotes bacterial proliferation, especially of anaerobic organisms.
Chronic wounds frequently appear to be "stuck" in the inflammatory stage of healing. Inflammation is a part of the normal healing process but prolonged inflammation and contact with chronic wound fluid may be harmful.3 Persistent inflammation or systemic diseases that cause inflammatory local wound beds (eg, vasculitis and pyoderma gangrenosum) can be confused with infection.
The inflammatory response consists of a cellular response and a humoral (fluid or antibody) response. The humoral response involves B-cells, which develop into plasma cells that secrete antibodies. The cellular response involves T-cells (named for their maturation in the thymus), macrophages (activated monocytes), and granulocytes (neutrophils, eosinophils, and basophils). Two main subpopulations of T-cells include the T-helper cells (TH) and cytotoxic T-cells (TC). Cytokines are small polypeptides that regulate both the cellular and humoral immune responses. They exert powerful effects on chemotaxis, proliferation, and differentiation of inflammatory cells and also have important actions on non-inflammatory wound cells such as fibroblasts, epithelial cells, and vascular endothelial cells. Persistent inflammation occurs in a chronic wound because the levels of key pro-inflammatory cytokines, such as TNF-beta and IL-1, are elevated, probably due to the presence of bacteria, fungi, and viruses in the open wound (see Figure 2).
Tissue Perfusion and Bacterial Colonization
Adequate tissue perfusion is a critical factor in getting wounds to heal rapidly. Good blood perfusion allows oxygen, nutrients, and cells to be delivered to the wound and limits the opportunity for micro-organisms to colonize. A tissue oxygen tension of greater than 40 mm Hg is conducive to healing but healing is unlikely to occur if levels are less than 20 mm Hg.4 Acute wounds in otherwise healthy individuals usually have oxygen tensions of 60 mm Hg to 90 mm Hg; whereas, chronic non-healing wounds are frequently hypoxic due to poor blood perfusion and oxygen tensions can be as low as 5 mm Hg to 20 mm Hg.
The initial assessment of a chronic wound must include an evaluation of the vascular supply to ensure blood flow is adequate. A palpable pulse indicates a blood supply of 80 mm Hg in the foot, 70 mm Hg in the hand, and 60 mm Hg in the neck. If a regional pulse is not palpable, Doppler or other vascular assessments (toe pressures or transcutaneous oxygen saturation) are necessary to determine healability.5
Hypoxic conditions cause cell death and tissue necrosis, which create ideal growing conditions for micro-organism contaminants and lead to colonization. Wound colonization is the presence of replicating micro-organisms within a wound that are not causing injury to the host. Anaerobes are likely to proliferate in low oxygen tension conditions and to continue to proliferate as the remaining oxygen is consumed by facultative bacteria. Arterial or venous insufficiency, trauma, blood loss, and edema all interfere with tissue perfusion and increase the likelihood of microbial proliferation compared to a healthy host. Critical colonization occurs when the levels of pro-inflammatory cytokines produced by the micro-organisms and the host neutralize host resistance and prevent or delay wound healing.
Local and Systemic Factors That Delay Healing
Neurotropic foot ulcers are usually caused by local pressure and loss of protective sensation with poor blood supply. Venous leg ulcers occur in people with venous insufficiency and local edema. Pressure ulcers are directly caused by sustained external pressure but are exacerbated by the patient's general health and comorbidities. Other factors such as advanced age, obesity, smoking, poor nutrition, and a compromised immune system all delay wound healing.
Inflammation and Infection
The diagnosis and management of infection and inflammation in chronic wounds is beset with difficulties for the wound care practitioner. A clinically infected wound can have serious consequences for the patient and can add to the overall cost of care. Treatment of infected surgical wounds can add up to 10 days of hospital care to the length of treatment.6,7
However, the role of micro-organisms in chronic wounds, the definition of infection, the role of quantitative and qualitative culture of wounds, the sampling methods that should be used, and the most appropriate course of treatment for infected wounds and for non-infected wounds that fail to heal are subjects for debate and disagreement. Ideally, systematic reviews that form the basis for many official treatment guidelines should help resolve some of these controversies, but those currently available offer little guidance regarding the diagnosis and management of infected wounds (see Appendix 1). Clinicians must rely on expert opinion and evidence secured from controlled and uncontrolled studies.
Interaction of Wounds with Bacteria
All wounds contain bacteria at levels ranging from contamination, through colonization, critical colonization (also known as increased bacterial burden or occult infection) to infection. The increased bacterial burden may be confined to the superficial wound bed or may be present in the deep compartment and surrounding tissue of the wound margin.
Several systemic and local factors increase the risk of infection (see Table 2). Great emphasis is frequently placed on bacterial burden; whereas, host resistance is often the critical factor in determining whether infection will take place. Host resistance is lowered by poor tissue perfusion as outlined above, poor nutritional status, local edema, and other behavioral factors such as smoking and drug and alcohol abuse. Other systemic factors that impair healing include comorbidities and medication that the patient may be taking for other conditions. Local characteristics of the wound also affect healing and the risk of infection. The presence of foreign material such as necrotic debris, retained packing materials, or small fragments of gauze dressing will significantly decrease host resistance and decrease the number or virulence of bacteria necessary to cause bacterial infection.
Contamination and colonization. Contamination is the presence of non-replicating organisms in a wound. Once a wound has been created, whether through surgery, trauma, or endogenous mechanisms, the probability for contamination is 100%. Contaminating micro-organisms arise from the external environment, surrounding skin, and endogenous sources such as the gastrointestinal (GI) and oral tracts. The normal flora of the oral and GI mucosae are diverse and present in large numbers. They will have little impact on a minor wound that is healing rapidly but may establish large colonies on slowly healing chronic wounds. Most chronic wounds become contaminated from endogenous secretions, healthcare providers, or the environment.8
It is widely believed that aerobic or facultative pathogens such as Staphylococcus aureus, Pseudomonas aeruginosa and the beta-hemolytic streptococci are primarily responsible for delayed healing and infection in all types of wounds, but this has largely been based on studies in which the culture and isolation of anaerobic bacteria was minimal or omitted.8 When wound colonization is investigated using microbiological techniques appropriate to anaerobic species, anaerobes form a significant proportion of the colonizers in acute and chronic wounds. Bowler et al8 reviewed a number of studies involving wounds of varying etiologies and concluded that anaerobes constitute, on average, 38% of the total number of microbial isolates even in clinically non-infected wounds.
The under-reporting of anaerobes in wounds may be due to the fact that culturing, isolating, and identifying anaerobes are more demanding than practices required for aerobic species. Also, some clinicians hold the view that anaerobes are not detrimental to wound healing.9-11 However, anaerobes can be highly virulent and may be the cause of postoperative infections when routine culture fails to yield bacterial growth.12
Critical colonization, increased bacterial burden, or covert infection. While the signs of frank infection are generally easy to identify, making a judgment about wounds that display persistent inflammation can be difficult. Wounds may display signs of covert infection where the host is harmed enough to impede healing but not enough to cause typical inflammatory symptoms. Covert infection is difficult to diagnose, as many of the signs listed above may be absent, but the most obvious sign is failure of the wound to heal or stalling after making signs of progress. Atrophy or deterioration of previously healthy granulation tissue, discoloration of granulation tissue to pale gray or deep red, and increased friability are less serious signs, along with excess exudate that is watery and serous rather than purulent.
Another confounding factor in diagnosis is the presence of biofilms, which may give the wound a healthy pink appearance while harboring large colonies of bacteria.13 Proliferating bacteria attach to the wound bed and secrete a glycocalyx coating that helps protect the micro-organisms from antimicrobial agents. These protected colonies can undergo genetic mutation to alter their sensitivity to antimicrobials. Biofilms intermittently release single viable bacterial cells that lead to local infections or weakening of the collagen matrix in recently healed wounds, causing breakdown and re-ulceration.14-16
The ability to make a diagnosis can be further complicated by the presence of certain non-infectious conditions that produce an inflammatory response, such as pyoderma gangrenosum and vasculitis. Swelling, redness, and increased temperature are usually taken to indicate infection. Persistent inflammation also can produce these signs, but they are usually not accompanied by wound breakdown unless associated with a secondary infection or a systemic inflammatory response that recruits more cellular infiltrates and inflammatory mediators. It is important to differentiate between inflammatory and infective ulcers (see Figure 3).
Infection. Infection is the presence of replicating micro-organisms in a wound with associated host injury. No fixed point determines when a wound moves from being colonized to being infected. Infection occurs when the pathogenic activities of the micro-organisms present in a wound overcome the natural defenses of the host's immune system. Invasion of the micro-organisms into host tissue brings about various local and systemic responses as described above. The likelihood of a wound becoming infected is clearly related to microbial load, the type of micro-organism, and the ability of the host to resist infection:
Infection = (number x virulence)/Host resistance
Although numerical values cannot be inserted, this formula expresses the relationship between increasing numbers of organisms and virulence that, together, can overcome the ability of the host to contain infection.
Infected Wounds: Causative Species
Infected wounds less than 1 month in duration usually have a high percentage of Gram-positive organisms. In infected wounds of longer than 1-month's duration, the wound is likely to acquire multiple organisms including Gram-negatives and anaerobes in addition to the Gram-positive bacterial flora.17
In their review, Bowler et al8 summarize the studies published in this area and conclude that the overall average percent frequency of anaerobic bacteria in infected wounds is 48% (compared with 38% in non-infected wounds). They conclude that a clear correlation exists between the incidence of anaerobic bacteria and wound infection. Studies involving chronic wounds within the scope of this review are summarized in Table 3.
Infected diabetic foot ulcers. People with diabetes have compromised immunity that leads to a reduced resistance to infection. This is further exacerbated if blood sugars are poorly controlled: patients with a Hb A1C (glycosylated hemoglobin) of greater than 0.12 will have impaired neutrophil chemotaxis. Diabetes may suppress the classical inflammatory signs of infection.
S. aureus is a common aerobic isolate in diabetic foot ulcers along with Staphylococcus epidermidis, Streptococcus spp, P. aeruginosa, Enterococcus spp, and coliform bacteria.20,22,23 When appropriate microbiological techniques are used, anaerobes are also isolated from up to 95% of diabetic wounds,24 the most common isolates being Peptostreptococcus, Bacteroides and Prevotella species.18,20,23-25
Infected leg ulcers and pressure ulcers (decubitus). The flora in chronic venous leg ulcers is often polymicrobial, with anaerobes constituting approximately 30% of the total number of isolates in non-infected wounds. S. aureus is the most common pathogen found in leg ulcers21,26,27 but the incidence of anaerobes increases in clinically infected leg ulcers to approximately 49%.21 The flora of infected pressure ulcers is also polymicrobial and is often similar to that seen in some acute necrotizing soft tissue infections.28
Is the causative organism relevant? As is the case in many areas of wound management, assumptions about chronic wounds are often based on observations of acute wounds. For example, S. aureus is considered to be responsible for delayed wound healing in traumatic, surgical, and burn wounds largely because of its high incidence in these wounds.12,29-32 Other organisms that cause concern include P. aeruginosa and beta-hemolytic streptococci.
However, although polymicrobial wounds are also frequently colonized with these species, it has yet to be demonstrated that they are the main species responsible for wound infection. Some studies have identified specific micro-organisms responsible for delayed wound healing or wound infection, but Bowler and colleagues8 note that, in some of these studies, it is unclear whether the identified species were the only ones isolated, and in other cases selective culture media could have biased the results.
Other studies have failed to correlate particular species with infection.9,10,27,33-35 In a study of chronic leg ulcers, Trengove36 reported that no single micro-organism or group of bacteria was more detrimental to healing than any other and the probability for healing was significantly lower if four or more bacterial groups were present in the ulcer. Bowler and Davies21 also reported that more species were isolated in infected than in non-infected leg ulcers.
Based on the collective evidence, the role of specific micro-organisms is still debatable; it may be that the presence of a number of different types of organism is the key factor. Virulence is also important. Beta-hemolytic streptococci produce a number of exotoxins and spreading factors that enable it to cause infection at lower concentrations than other organisms.17
Most chronic wounds contain more than three species of micro-organisms, which increases the risk of infection because they may develop synergies with each other.37 The combined effects of aerobes and anaerobes in wounds may be synergistic, producing effects that are not seen with just one type of micro-organism. Oxygen consumption by aerobic bacteria brings about tissue hypoxia, which favors the growth of anaerobic bacteria; one bacterium may produce specific nutrients that are required by other micro-organisms; and some anaerobes are able to impair the host immune cell functions, providing a competitive advantage to themselves and other micro-organisms.8
In wounds that are infected with a number of species, distinguishing which is the causative organism is impossible. Bacterial swabs or wound cultures do not diagnose infection, but they can be used as guidance for antimicrobial therapy. The diagnosis of infection is based on clinical symptoms and signs.
Microbial Load and Infection
A number of studies have been conducted in an attempt to assess the impact of microbial load on wound healing. In 1964, Bendy et al38 reported that healing in decubitus ulcers was inhibited if the bacterial load was greater than 106 colony forming units (CFU)/mL of wound fluid. Superficial wound swabs were used in this study, but other studies using the gold standard (tissue biopsy specimens) reported similar results in pressure ulcers and surgical wounds.39-41 Quantitative microbiology has a role to play in predicting the risk of infection; many studies have shown that bacterial load correlates with risk.8 However, it cannot be taken in isolation as an indicator of infection because other factors (eg, reduced host resistance or the presence of foreign objects in the wound) can significantly reduce the bacterial load required to trigger infection.
Diagnosis is primarily a clinical skill, and microbiological data should be used to supplement the clinical diagnosis, not the other way around. The progress of a wound along the continuum to an infected state cannot be predicted by the presence of a specific type of bacterium. Neither bacterial load nor type can be used in isolation to provide a definitive diagnosis of infection because the immune response of the host is critical in determining whether a wound will become infected. A healthy host will be able to tolerate a higher bacterial load and to resist infection from otherwise highly virulent pathogens better than a compromised host.
The classical signs of infection in acute wounds include pain, erythema, edema, purulent discharge, and increased heat. These are related to the inflammatory process occurring in the wound. Increased blood flow produces the rise in temperature, and fluid leaking from intravascular spaces accumulates in the tissue, causing visible swelling. Vasoactive mediators such as histamine produce the characteristic erythema, and pain is caused through activation of plasma-derived mediators near unmyelinated nerve fiber endings.
For chronic wounds, it has been suggested that other signs should be added: delayed healing, increased exudate, bright red discoloration of granulation tissue, friable and exuberant granulation, new areas of slough or breakdown on the wound surface, undermining, foul odor, and new areas of wound breakdown.42 Purulent exudate, usually white and creamy, is common in infections produced by bacteria such as S. aureus. By contrast, serous exudate is thin and clear in color. Serous exudate may be increased in a chronic wound with increased bacterial burden before purulence is noted.
It has been suggested that chronic wounds should show some evidence of healing within 4 weeks to progress to healing within 12 weeks. If this time limit is exceeded, increased bacterial burden or infection should be suspected as one of the causes of delayed healing.43
Discolored granulation tissue arises from excessive angiogenetic responses caused by pathogens, while friable granulation tissue bleeds easily with light pressure. Healthy granulation tissue is pink-red and firm with a moist translucent appearance. When infected, it will appear dull and may have patches of greenish or yellow discoloration. Certain anaerobic species such as Bacteroides fragilis and streptococci can produce a dullish, dark-red hue, while Pseudomonas will produce green or blue patches which may fluoresce with a black or Woods light.
Undermining is probably caused by a lack of granulation tissue that has been inhibited or digested by bacteria along some areas of the wound bed. Foul odor is usually caused by Gram-negative bacilli, Pseudomonas species, or anaerobic bacteria.35 Faulty collagen formation arises from increased bacterial burden and results in over-vascularized friable loose granulation tissue that usually leads to wound breakdown.
Deep infection will often cause erythema and warmth extending 2 cm or more beyond the wound margin.44 This increased inflammatory response is painful and will cause the wound to increase in size or lead to satellite areas of tissue breakdown that cause adjacent ulceration. Deep infections, especially in ulcers of long duration, can often lead to osteomyelitis.45 Probing to bone is a simple clinical test that carries a high probability of osteomyelitis, especially in people with diabetic foot ulcers (see Table 4).46
Gardiner et al47,48 examined the reliability and validity of clinical signs of infection. These studies identified various symptoms and signs of infection and compared diagnoses made using these signs to the results of quantitative cultures from tissue biopsies. Increasing pain, friable granulation tissue, foul odor, and wound breakdown all demonstrated validity based on sensitivity, specificity, discriminatory power, and positive predictive values.
A checklist subsequently was constructed to test the ability of different observers to distinguish these signs (reliability). A high total percent agreement was noted between observers, but it is also important to assess the discriminatory power of the sign or symptom when infection is present compared to when it is absent. A kappa test was performed to quantify the usefulness of each criterion. High kappa values (above 0.7) were present with: increasing pain (1.0), edema (0.93), wound breakdown (0.89), delayed healing (0.87), friable granulation (0.8), purulent exudate (0.78), and serous exudate (0.74). The sample size was small (N = 31) and five different types of wounds were included. Although these observations must be considered preliminary, these characteristics will assist clinicians to more accurately identify infection in chronic wounds.
The Role of Microbiology in Diagnosis
Culturing a chronic wound that is healing at an expected rate and does not display any signs or symptoms of infection is unnecessary. Because all wounds are contaminated and colonized, a culture simply confirms the presence of micro-organisms without providing any information as to whether they are having a detrimental effect on the host.
However, bacterial swabs can provide information on the predominant flora within a non-progressing, deteriorating, or heavily exudating wound. Microbiological tests also can screen for multi-resistant bacteria such as methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant enterococcus (VRE). The degree of inflammatory response is measured by the presence and quantity of neutrophils per high power field in the Gram stain of the swab contents before inoculating the specimen on growth media.
If a wound has progressed beyond the stage of increased bacterial burden or covert infection, systemic antimicrobial therapy would be required to supplement local treatment, and the microbial analysis would assist in targeting therapy. However, quantifying the bacterial load through invasive tissue biopsy is not always necessary in clinical practice. A number of alternative non-invasive techniques have been assessed and provide similar information with far less trauma to the patient.
Which culture technique should be used?
Quantitative sampling (tissue biopsy) has merits, and a strong association exists between the number of organisms in a wound and the ability of the wound to heal. Once bacterial load reaches 106 CFU/g of tissue, wound healing is usually impaired.49 However, these findings need to be viewed in perspective. At least 20% of wounds colonized with more than 105 CFU/g of tissue will still heal,50 and normal skin flora present in high quantities appear to enhance wound healing.51 On the other hand, some micro-organisms (eg, beta-hemolytic Streptococci, Mycobacterium tuberculosis, Treponema pallidum, Corynebacteria diphtheriae, Bacillus anthracis, Francisella spp and Brucella spp) can be detrimental in small numbers. Thus, quantitative microbiology does not necessarily provide an unambiguous diagnosis of infection.
Quantitative biopsy also may have poor sensitivity and reliability. Woolfrey et al52 showed a 25% chance of missing an organism using biopsy, probably due to uneven distribution of organisms within the wound bed and the techniques used to clean the specimen. Results can vary by 2 logs in 27% of paired isolates.
Qualitative aspects are at least as important as overall bacterial load, and evidence is growing that microbiology obtained by a swab may adequately correlate with qualitative findings obtained through tissue biopsy. Growth in the fourth quadrant (4+ or heavy growth) or the most dilute streaking of the culture plate by the swab corresponds approximately to a growth of 105 CFU/g of tissue as determined by quantitative biopsy. Some clinicians argue that wound fluid may not yield micro-organisms that would be detected in biopsied tissue. The counter-argument is that in most cases the colonizing bacteria come from exogenous sources and would be present in the superficial compartment before reaching the deep tissues.
In a study on diabetic foot infections, Wheat et al18 showed that the results obtained with swabs are similar to those obtained with tissue biopsy. A fairly high rate of false positive and negative results occurred using the swab, but most of the false positives were commensals that did not require antimicrobial therapy, and those that were missed would not have resulted in a different treatment had they been recorded. The authors concluded that 92% of antibiotic therapy would have been adequate based on the swab alone.
Sapico et al35 found similar results in a study on chronic pressure ulcers, demonstrating a 75% concordance between swab and biopsy results. Ehrenkranz et al53 demonstrated that an irrigation-aspiration technique could produce similar results to qualitative biopsy.
Procedure for Taking a Swab
In most cases, wounds should not be cultured if no evidence of infection or impaired healing is noted unless screening is being performed for colonization of multi-resistant organisms. The wound bed must first be cleaned with saline and superficially debrided so the cultures from the superficial wound compartment more closely resemble those in the deep wound compartment. Some clinicians have recommended alginate or rayon-tipped swabs in the belief that the fatty acids contained in cotton swabs might inhibit growth in certain bacteria. However, the organisms commonly encountered in infection are likely to withstand the environment of a cotton swab.
Pre-moistening a swab in the transport media is useful if the surface of the wound is dry (it can improve the yield) but is not necessary if the wound is already moist. The swab should be taken from the granulation tissue surface of the wound, avoiding debris and frank collections of pus. The tip of the swab should be rolled on its side for one full rotation over the part of the wound granulation tissue with the most obvious signs of infection, avoiding slough and surface purulent discharge. A zigzag pattern can be used for wounds larger than 5cm2. This technique is likely to increase the yield of non-significant colonizers; a preferred alternative is to take more than one regional swab from the upper and lower areas of the wound. If pus or discrete abscesses are collected locally, the fluid should be aspirated into a syringe using a needle. The fluid is an ideal specimen for culture. Likewise, deep curettings from the debridement process should be sent for analysis; these closely correlate to biopsy samples.
Transport to the laboratory should be carried out promptly. If infection is diagnosed, a properly obtained swab will provide the information that is needed for most clinical situations.
Changes in microbial flora over time. The microbial flora in a chronic wound changes over time in a predictable fashion.17 Combining this information with clinical signs may allow identification of the invading pathogens while waiting for culture results (see Table 5).
Managing Inflamed and Infected Wounds
Managing non-healing chronic wounds should proceed in a stepwise and logical fashion:
1. Treat the cause. Identify and correct the underlying factors that produced the wound or are impeding its healing.
2. Healable wound (a wound that has an adequate blood supply and favorable host factors that will promote healing if the wound bed is adequately prepared): If the wound still fails to heal and evidence of increased superficial bacterial burden or delayed healing with no evidence of deep infection are observed, use local antimicrobials with debridement and moisture balance. With evidence of deep infection or if the wound fails to heal within 2 weeks with topical antimicrobials, systemic antibiotics should be considered.
3. Non-healable wound (inadequate blood supply or decreased host resistance): Antiseptics may be used to limit bacterial proliferation.
Holistic patient assessment. The first step in managing possible infection is to assess the patient for underlying causes that could be impeding wound healing. This is the basis for wound bed preparation (see Figure 1). Local wound management measures are unlikely to succeed if the patient is not receiving adequate treatment for conditions that are known to impair healing, such as heart failure, uncontrolled diabetes, inadequate compression therapy, poor nutrition. Corrective measures such as pressure relief, revascularization of ischemic tissue, control of edema, and all other relevant measures should be carried out before local wound care treatments are applied.
This philosophy is encapsulated in the TIME approach to the management of chronic wounds (see Table 6), where the objective is to assess the evidence and provide recommendations for best care in each of four areas:
1. T for tissue
2. I for infection and inflammation
3. M for moisture balance
4. E for epidermal advancement.
Vasculitis is an inflammatory process involving the blood vessel walls.54-57 Depending upon the extent of local injury, the vessel walls become leaky, leading to fixed urticarial lesions and purpura if red blood cells are extravasated. When inflammatory infiltrate penetrates the vessel wall and surrounds the periphery of the vessel, the skin surface lesion becomes palpable. A lesion of palpable purpura on the lower extremities is the hallmark of cutaneous vasculitis. These areas can develop local necrosis with further blood vessel damage and the loss of nutrients to surrounding tissue. This results in a necrotic eschar on the surface of the skin with crust formation and subsequent ulceration. The lower legs are a common site for cutaneous vasculitis because relative stasis, cooling, re-duplication of the basement membrane, and gravitational forces all assist circulating immune complexes of systemic vasculitic diseases to be trapped.
Vasculitis can be divided into large, medium, and small vessel vasculitis. Large vessel vasculitis manifests as giant cell (temporal) arteritis or Takayasu’s arteritis, neither of which generally result in chronic skin ulcers. Giant cell arteritis often involves the temporal artery in patients older than 50 years and is associated with polymyalgia rheumatica. Takayasu’s arteritis usually occurs in individuals younger than 50 years and is predominantly located in the aorta and larger arterial branches.
Medium-sized vessel vasculitis can be divided into polyarteritis nodosa and Kawasaki’s Disease (usually seen in children and associated with a mucocutaneous lymph node syndrome with involvement of the coronary arteries). Polyarteritis nodosa may present with leg ulcers although a localized variant of polyarteritis nodosa that involves the skin and adjacent structures only, often on the lower legs, may be noted. The systemic form may have acute renal involvement. In addition, infectious variants associated with acute bacterial endocarditis, IV drug abuse, and various infections have been seen. When lesions occur on the lower legs, they often start with painful nodules that can develop necrotic crusts and form deep painful ulcers. These can be extremely difficult to control; treatment of the underlying cause, along with local wound care, is usually necessary. A differential diagnosis also can include a primary inflammation of the fat or panniculitis.
Small vessel vasculitis includes conditions associated with anti-neutrophil cytoplasmic auto-antibodies (ANCA): Wegener’s Granulomatosis, Churg-Strauss Syndrome, and microscopic polyangitis. Churg-Strauss Syndrome has eosinophilic rich lung involvement, while Wegener’s Granulomatosis and microscopic polyangiitis involve the kidneys and the lungs. Cutaneous lesions can occur and present with palpable purpura and necrosis. Local wound care measures will not be successful until patients with these conditions are treated systemically, usually with prednisone and cyclophosphamide. These diseases are life-threatening and treatment has significant side effects.
The most common small vessel vasculitis diseases associated with lesions and chronic ulcers are cutaneous leukocytoclastic vasculitis, Henoch-Schönlein purpura, and essential cryoglobulinemic vasculitis. Cutaneous leukocytoclastic vasculitis is common on the lower legs, may have associated arthralgias, but often no other systemic organ involvement. However, checking for joint, liver, or kidney involvement is important. Lung, heart, GI, and central nervous system changes are much less common. Henoch-Schönlein purpura is associated with IgA deposits in the vessel walls of the affected skin, the kidneys are frequently affected, and GI tract with hemorrhage and arthralgias may occur. Essential cryoglobulinemic vasculitis is associated with cryoglobulins in the skin, and glomeruli are often involved.
As indicated in the algorithm (see Figure 4), these patients require a skin biopsy of a fresh palpable purpuric lesion or a deep wedge biopsy of a nodule for histological diagnosis. Diagnosis is made by skin biopsy, demonstrating the changes of vasculitis and the pattern of other organ involvement along with laboratory studies. Depending on the diagnosis, systemic and local treatment can be initiated.
Pyoderma gangrenosum often starts with a small pustule that quickly evolves into painful, fluid-filled blisters that enlarge rapidly, forming hemorrhagic blisters or bullae that become necrotic. Ulceration that forms in the center is associated with a translucent rolled border containing the inflammatory cellular infiltrate that may be many centimeters in diameter. Skin biopsy reveals a massive infiltrate of neutrophils into the skin; although vasculitis may accompany these lesions, they display pathergy — that is, sharp debridement can trigger an extension of the inflammatory process. Until the acute inflammatory process has eased or the patient has been treated systemically, aggressive local care is contraindicated.58-60
Pyoderma gangrenosum may be idiopathic without any apparent underlying cause or association. In about half of all cases, pyoderma gangrenosum may be associated with rheumatoid arthritis, inflammatory bowel disease (ulcerative colitis or Crohn’s Disease), and myeloproliferative disorders (plasma cell dyscrasias, leukemias, and lymphomas). The approach to investigation is outlined in Figure 5.
Treating Infected and Inflamed Ulcers
Local wound care. If all systemic factors have been identified and corrected and the wound still fails to heal, the prime objective of treatment will be to reduce bacterial burden using debridement, moisture balance, and topical antimicrobial agents. If inflammation due to active vasculitis or pyoderma gangrenosum is suspected, sharp debridement is not recommended, although other aspects of management are similar to those for infected wounds (see Figure 6 and Figure 7).
Role of moisture (see Table 7). Moist interactive dressings provide fluid balance in a wound bed, preventing excess fluid from macerating the wound margins or promoting bacterial growth. The choice of dressing is based on form and function, wear time, caregiver skill, cost, and patient preference.
Antiseptics (see Table 8). Antiseptics can be toxic both to microbial cells and host cells, although this appears to be concentration-dependent and can be modified by altering the form in which the agent is delivered. Concern over the use of antiseptics, especially hydrogen peroxide and aniline dyes (crystal violet, neutral red, mercurochrome), has largely centered on their potential for cytotoxicity toward important components of wound healing such as fibroblasts, keratinocytes, and leukocytes. However, the cytotoxic effects were observed in vitro where concentrations were high and have not been observed in lower in vivo concentrations where they may retain their antimicrobial activity without damaging host cells.60-62
These agents are generally reserved for those patients with non-healable wounds (those with inadequate blood supply or lowered host resistance) and may be used for short periods in patients with increased bacterial burden in the superficial compartment if the bacteria are of more concern than potential cellular toxicity. This may be the case if the wound surface shows significant necrotic debris that has not been debrided. A deep infection or sinus beneath a moist wound surface will facilitate bacterial proliferation and hinders host response. Preferred local antiseptic agents that have a combination of acceptable cellular toxicity, broad-spectrum antibacterial coverage, residual effect, and low sensitization potential include povidone iodine and chlorhexidine.
Topical antimicrobials (see Table 9). Topical antimicrobials are used to reduce bioburden; therefore, the choice has to be related to the identity of the causative organisms, assessed either through bacteriological culture or clinical judgment. The choice of topical agent should also include an awareness of its potential to induce sensitization. Neomycin is a well-known allergen, but lanolin and perfumes contained in the delivery vehicles of other agents are also common sensitizers and should ideally be avoided.
Iodine. Iodine is a potent broad-spectrum antiseptic agent. Its role in wound management is controversial because some iodine formulations (povidone iodine) were shown in vitro to impair the functioning of cells involved in wound healing. However, this was not observed in vivo where concentrations used were lower than 1%.63 Although the microbial load was significantly reduced using povidone iodine, healing was not accelerated in this study.
Improved formulations are now available that release low levels of iodine over longer periods (cadexomer iodine). These have been shown to be effective against wound pathogens without impairing wound healing.64 It is believed that the sustained release of iodine overcomes the neutralizing effect of organic material in the wound. A review of the literature concludes that cadexomer iodine is safe, effective, and economical in the treatment of many chronic wounds. In addition to its antimicrobial effects, this formulation of iodine into an absorbent dressing also acts as a debriding agent that removes pus and debris from chronic wounds.65
Cadexomer iodine also has been shown to inhibit proliferation of methicillin-resistant S. aureus in experimental wounds.66 Expert opinion now supports the use of this form of iodine in healable chronic wounds that have an increased bacterial burden in the superficial wound compartment.67
Topical povidone iodine should be considered for non-healing wounds, with or without clinical signs of infection.
Silver. The antimicrobial properties of silver have been recognized and exploited for thousands of years, even though the mechanism of action was not initially understood. The first documented silver preparation to be used in medicine was a 1% silver nitrate solution used to prevent neonatal ocular infections. In 1887, Von Behring68 showed that 0.25% and 0.01% silver nitrate solutions were effective against typhoid and anthrax bacilli, respectively. In the early 1900s, hammered foil and colloidal silver were used to treat non-healing wounds, and it was noted that they effected a decrease in erythema (rubor). In the 1920s, the US Food and Drug Administration acknowledged that colloidal silver was an effective wound treatment.69 Research into antibiotics in the 1940s shifted the emphasis away from silver, and it was 30 years before Fox70 introduced 1% silver sulfadiazine cream for the treatment of burn wounds.
The effects of silver sulfadiazine cream in chronic venous leg ulcers were demonstrated by Bishop et al70 in an early prospective, randomized trial that reported statistically significant reduction in ulcer size compared to copper complexes or placebo.71 More recently, Mi et al72 showed that silver sulfadiazine incorporated into a bilayer chitosan wound dressing can provide long-term inhibition of P. aeruginosa and S. aureus.
Silver was first incorporated into modern dressings adsorbed onto charcoal. The silver kills bacterial organisms that are adsorbed into the dressing, and the charcoal provides a wound deodorizer. At that time, film dressings were the backbone for a calcium sodium phosphate polymer matrix that releases most silver over the first few hours with some delayed release over the next few days, but this dressing has limited fluid handling capabilities. Several newer delayed release vehicles for silver have been developed that incorporate longer dressing wear time with moisture balance. In some products, autolytic debridement also may be available.
Clinicians know that silver is effective against a broad range of aerobic, anaerobic, Gram-negative, and Gram-positive bacteria, as well as yeast, fungi, and viruses.73-75 Silver has an effect on bacterial DNA, enzymes, and membranes, requiring several bacterial mutations for resistant organisms to appear. It has a very low mammalian cell toxicity, low sensitization potential, and is not used systemically, making it an ideal agent for treatment of increased bacterial growth in the superficial compartment.
In vitro concentrations of silver as low as 10 µg/L can control bacteria, but higher concentrations are delivered in some topical wound dressings. The minimum inhibitory concentration (MIC) in wounds in one study was estimated to be between 20 and 40 µg/L.76 A study on common wound pathogens using a complex organic growth medium found that the MIC in vitro ranged between 5 and 12.5 µg/mL.73
In a study of 29 chronic wound patients not healing at the expected rate, Sibbald et al77 applied nanocrystalline silver dressings after baseline superficial bacterial swabs and quantitative biopsies. Better healing was related to improvement noted in the semi-quantitative surface swabs and the deep bacterial biopsies. The majority of the chronic wounds showed improvement in the surface swabs. Quantitative bacterial biopsy results did not demonstrate any improvement in the deep compartment quantitative bacterial count. If the deep compartment was out of bacterial balance and this was delaying healing, topical silver dressings did not reverse the impaired healing response or the increased bacteria in the deep compartment.
Uncommon silver allergic sensitization has been reported, but no other significant adverse effects have been noted despite the large amounts of silver used in burn wound treatment. On the other hand, the nitrate molecule in silver nitrate may be pro-inflammatory, while the cream base in silver sulfadiazine reacts with serous exudate to produce a pseudo-eschar that must be removed before reapplication.78 In both of these preparations, a large excess of silver has to be supplied to the wound to compensate for inactivation; therefore, new technologies have been developed to improve the controlled release of silver ions. The silver ions can provide antibacterial and anti-inflammatory properties topically as well as providing moisture balance with absorptive dressing cores.
Nanocrystalline silver. Nanocrystalline silver is composed of small crystals of less than 20 nm that appear to possess crystalline characteristics.79 In amorphous matter, atoms and molecules interact only with their nearest neighbor; whereas, in crystals each component interacts with immediate and distant neighbors through the crystal lattice. During dissolution, the silver reaches a steady state where the concentration in solution is between 70 and 100 µg/mL74 and antimicrobial levels can be maintained in the dressing for at least 7 days. Nanocrystalline silver is effective in vitro against a broad range of bacteria, including MRSA and VRE.
For more than century, the anti-inflammatory effects of silver have been observed and documented, but the mechanism by which silver exerted this effect was not understood. The anti-inflammatory effect was largely masked or even countered by the silver preparations available in the 20th century (silver nitrate, silver sulfadiazine); whereas, the development of nanocrystalline silver may modulate the effects of matrix metalloproteinases (MMPs). This was suggested by Wright et al80 who studied MMP, cell apoptosis, and healing in a porcine wound model in which wounds were dressed with nanocrystalline silver, silver nitrate, and saline soaks. Their observations suggest that the nanocrystalline silver may modulate the actions of the MMPs. Another pilot study examined the wound fluid of 10 patients whose wounds were managed either with a nanocrystalline dressing or a control. Patients using the nanocrystalline dressing had lower MMP-9 and TNF-ß levels relative to the controls.80
Systemic treatment. Antibiotic therapy is useful for the treatment of infected wounds in the deep compartment and helps prevent infection from spreading into soft tissues beyond the wound,81 but repeated use can lead to the development of bacterial resistance. Antibiotics should be reserved for increased bacterial burden of the superficial compartment that does not respond to topical treatment.
If infection extends beyond the ulcer margin (edema, warmth, tenderness, increasing ulcer size, or the presence of new satellite areas) or if it is possible to probe to bone, systemic antibiotics are indicated.18 A bacterial swab is not needed to diagnose infection, but it can be used to guide antimicrobial therapy and to identify the presence of resistant organisms such as MRSA or VRE.
Patient response to antibiotics can be monitored through clinical assessment of the symptoms and signs of infection with special attention to pain and ulcer size. In patients with less obvious infection clinically, an eosinophil sedimentation rate higher than 40 without another cause or an elevated C-reactive protein, it may be useful to monitor resolution of deep infections. A negative X-ray does not rule out osteomyelitis, and probing to bone is a useful clinical sign.
Preparing the wound bed is a holistic approach to the diagnosis and treatment of chronic wounds that integrates expert knowledge into best clinical practices. This article explores the problem of chronic wounds with infection or prolonged inflammation (see Figures 8-13). Inflammatory conditions such as pyoderma gangrenosum and cutaneous vasculitis can mimic infection.
Bacteria and wounds exist in a continuum ranging from contamination to colonization to critical colonization to infection. The most important factor limiting infection is host resistance. The type of organism (with a few exceptions) is less important than the number. Infection is diagnosed clinically with microbiological tests used to identify organisms and their antibiotic susceptibility. Topical antimicrobial treatment has been improved with new ionized silver dressings.
By introducing the concept of TIME as an approach to local wound care, the clinician is reminded of the need for tissue debridement, the control of inflammation or infection, and moisture balance. For non-responsive wounds, advanced therapeutic modalities are available that treat the epidermal edge effect or non-advancing border.
Expanding the information on the aspects of infection and inflammation paves the way for future articles that assess tissue debridement, moisture balance, and the epidermal edge effect with a similar approach. In the end, this paradigm will help clinicians improve the quality of life and healing for patients with chronic wounds.
The authors wish to thank Smith & Nephew for supporting the work of the International Advisory Board on Wound Bed Preparation. Members of the International Advisory Board on Wound Bed Preparation include: Gregory G. Schultz, Elizabeth A. Ayello, Caroline Dowsett, Vincent Falanga, Keith Harding, Marco Romanelli, R. Gary Sibbald, Michael C. Stacey, Luc Teot, are Wolfgang Vanscheidt. Members of the Canadian Wound Care Advisory Board (R. Gary Sibbald, Chair) are: Alain Brassard, Keith Bowering, Cathy Burrows, Karen Campbell, Richard Cloutier, Pat Coutts,Lincoln D'Sousa, Laurent Delorme, Ken Dolynchuk, Louise Forest-Lalonde, Wayne Gulliver, Pam Houghton, Tim Kalla, David Keast, Brian Kunimoto, Marie-France Magie, Heather Orsted, Diane St-Cyr, and Laura Teague
A special thank you to Valerie Cocker, for assistance with inflammation/infection algorithms and to Jude Douglass for assistance in research and preparation of the manuscript.