Critical Colonization — The Concept under Scrutiny
The term critical colonization has attracted increasing attention over the past 5 years. Accepted by some and viewed with derision by others, critical colonization has been regarded by a number of publications as synonymous with local infection1-4 through its association with the criteria for wound infection developed by Cutting and Harding,5 a benchmark for diagnosis that appears to have gained relatively broad acceptance. Others have dismissed the concept of critical colonization as a myth, expressing the view that a wound is either infected or not, with no prodromal phase of infection.6 In order to advance understanding of delayed healing in the absence of an obvious clinical cause, the basic concept of critical colonization deserves consideration.
If alternative explanations for delayed healing can be identified, patient morbidity potential can be reduced; therefore, delayed healing must be placed accurately into context to help avoid making or perpetuating inappropriate assumptions. This paper reviews the emergence of the concept of critical colonization from an historical perspective, discusses assumptions that have been made, and presents scientific evidence collated from the literature. This approach draws some parallels with the criteria for wound infection developed by Cutting and Harding5 where a review of the literature led to the collation of traditional and additional diagnostic features of wound infection and the development of an entirely new approach to identifying clinical wound infection.
Development of the Concept
The term critical colonization was first coined in 1996 by Davis7 in a poster presented to a joint meeting of the Wound Healing Society and the European Tissue Repair Society. Using case studies, Davis demonstrated how delayed healing in wounds could be reversed through appropriate use of topical antiseptics. She also defined the condition of the wound in relation to bacterial presence. Using a modified model for infection first published by Ayton,8 Davis introduced the notion of critical colonization within the infection spectrum (from sterile to contaminated to colonized to critically colonized to infection) and defined it as “multiplication of organisms without invasion but interfering with wound healing.”
Davis also stated that “the classic signs of infection must be reassessed to include the ‘critically colonized’ wound,”7 offering the first association with local infection. In support of her treatise, Davis cited Danielson9 and a clinical study with microbiological screening by Trengove10 espousing the notion that the presence of pathogens, with or without host reaction, could interfere with healing. Currently, the absence of a host response is viewed as a fundamental link to understanding the concept of critical colonization.
Davis continued to champion the term critical colonization but little if any notice was taken until Kingsley11 renamed her model of wound infection “the Wound Infection Continuum.” This model would appear to bear the most relevance to acute wounds because even though the first stage — sterility — is not a feature or even a therapeutic goal in chronic wounds, the wound infection continuum is most closely associated with chronic wounds.
While the term critical colonization may sound novel, the underpinning concept possibly has been part of the wound healing lexicon under guises related to delayed healing. A review of the literature in relation to delayed healing reveals the use of a number of possibly synonymous phrases, including silent infection, covert infection,12occult infection,13refractory wound,14subclinical infection,15indolent wound,16stunned wound,17subacute infection, and recalcitrant wound.18
Placing Critical Colonization into Clinical Context
Role in the wound infection continuum. With heightened focus on wound microbiology and infection in recent years, the wound infection continuum has been proposed as a model to account for an increasing microbial load (bioburden) and related pathology.11,16 Although the concept of critical colonization is not universally accepted, clinicians and researchers generally agree that the term needs definitive characterization in order to validate its consideration in infection management.19,20
Wound infection development depends on microbial and host factors. An in vivo study21 has shown that the number of bacterial species and the number of organisms are important factors in the development of infection. However, these findings have yet to be clinically validated. In a theoretical hypothesis, Heinzelmann et al22 submit that the host response, or immune status of the individual, is a key factor in the development of infection; the triggering of a host response has been used as a diagnosis of infection for 2,000 years.
For those who accept it as a distinct entity, critical colonization is a stage where wound healing is delayed without the overt signs and symptoms of infection23; it occurs despite optimum treatment.16 It would appear to be a contradiction that a microbiologically induced delay in healing could occur without eliciting a host response. Could such a situation arise without the host response playing a part? A number of authors have reported delayed ulcer healing influenced by micro-organisms: Lookingbill et al24 in a quantitative bacteriology study on 13 leg ulcers; Daltrey et al25 in a bacteriological study of 74 pressure ulcers in 53 patients; Halbert et al26 in a bacteriological study of 82 patients with 100 ulcerated limbs; and Hansson et al27 studying the qualitative and quantitative bacteriology of 58 patients with leg ulcers. In a retrospective review of patients with various inflammatory wounds such as necrobiosis lipoidica, Drosou28 provides an additional perspective, stating it is likely that subclinical damage to tissue as a result of bacterial contamination exists and cites Hermanns15 in support of the premise.
Clinically, host response to wound infection is recognized by the classic signs and symptoms of inflammation — ie, redness, swelling, warmth, and pain. Spreading erythema around the wound is usually indicative of infection such as erysipelas or cellulitis.27 However, not all erythematous reactions are immunologically generated. Recent findings from a series of clinical cases have shown that Morganella species (notably M. morganii) commonly found in wounds express histamine in physiologically significant amounts30; therefore, periwound erythema could be attributable to M. morganii colonization. This Gram-negative bacillus inhabits the gastro-intestinal tract and is a part of the normal fecal flora. It has been reported in a single case report on chronic leg ulceration31 and in Chiclero’s ulcer in a microbiological study involving 26 patients32 but is not routinely considered in bacterial samples acquired from wounds. Hansson et al27 found M. morganii (identified as Proteus morganii) in 23% of venous leg ulcers (n = 58) studied. Conversely, Bowler and Davies,33 in reviewing data from a prospective clinical study where swabs from 44 infected leg ulcers were compared with 30 from non-infected ulcers, found this bacterium in infected but not non-infected leg ulcers. In this study, the diagnosis of ulcer infection was determined on the basis of clinical signs including erythema, cellulitis, edema, increased pain, increased exudate, and warmth.
Microbial factors. In literature reviews of the cell biology of chronic wounds, delayed healing has been intimately linked with uncontrolled inflammation34 or immunopathology.35 This is not visually evident in many chronic wounds because it is not always accompanied by the classical signs of inflammation. However, delayed healing is histologically evident.36 The acolytes of critical colonization believe delayed healing often can be attributed to microbial factors37 and that frequently diagnosis is confirmed only retrospectively once antimicrobial measures have been taken and found to be effective.
How, then, can delayed healing be associated with microbial factors and not elicit an obvious host response? Three or more potential bacterial modes of action, described in the literature, can delay wound healing without any apparent inflammatory or immunological response: the expression of immuno-evasion,38 biofilm formation,39 and suppression of cellular wound healing responses.40 These modes have been identified following in vitro work and can occur when the wound is colonized by certain specific bacteria.
Pseudomonas aeruginosa. An organism commonly found in chronic wounds33 and associated with chronic infection,41Pseudomonas aeruginosa is known to form biofilms42 and secrete immuno-evasive factors43 active against polymorpho nucleocytes (PMNs). To this effect, activation of the type III secretion system, a recently identified virulence determinant of P. aeruginosa, has been reported from in vitro study using clinical isolates.44 It has been postulated from in vitro study that P. aeruginosa is likely to be of far greater significance to wound chronicity, tissue invasion, and infection than previously recognized.38-40,45,46
As indicated in a summary of clinical and microbiology findings by Hamilton and Danielsen,47 experienced wound clinicians have long noticed occasional green coloration in chronic wounds and have associated it with the presence of P. aeruginosa and delayed healing. The green pigment, pyocyanin, is a phenazine, a highly diffusible exotoxic metabolite described in an in vitro study by Denning et al.48 In a review of published clinical and in vitro data by Lau et al,41 pyocyanin has been shown to inhibit many cell functions and impair host defenses through apoptosis. In vitro laboratory research on clinical samples49 has shown that many pathogens, including P. aeruginosa, induce inappropriate or premature apoptosis of immune cells such as macrophages and neutrophils and that this can be pro-inflammatory.50
P. aeruginosa has evolved immuno-evasive strategies by which it affects host immunity.51In vitro studies have shown pyocyanin and other similar phenazines to have pro-apoptotic action on human neutrophils.43 This is postulated to be a clinically important mechanism of persistence of P. aeruginosa in human tissue.43 What mediates the change in the infective potential of this organism? P. aeruginosa has been shown in a literature review to be a phenotypically unstable pathogen, particularly in chronic infection.52 The virulence of P. aeruginosa is controlled by an N-acyl homoserine lactone (AHL)-dependent quorum sensing system. The organism has been shown in vitro to have the capability to modulate its own quorum-sensing dependent pathogenic potential through an AHL-acylase enzyme.53 This may in part explain how under certain circumstances P. aeruginosa may be a delayer of wound healing and under other circumstances an infecting organism.
Other aerobes and anaerobes also have been recognized for down-regulating the immune response. In an in vitro microbiology study, Bowler et al54 summarized the role of succinate (a short-chain fatty acid) produced by aerobes and anaerobes; in vitro studies by Rotstein55,56 demonstrate how succinate may increase the risk of infection by impairing host cell function.
Staphylococcus aureus. Staphylococcus aureus also is an important human wound pathogen that interferes with host-cell functions. According to in vitro studies, impaired healing often is observed in S. aureus-infected wounds where the extracellular adherence protein (EAP) has been implicated.57 Extracellular adherence protein has been shown in in vitro studies to be a potent anti-inflammatory58 and anti-angiogenic agent, preventing recruitment of inflammatory cells to the wound site as well as inhibiting neovascularization.59
Odor-producing micro-organisms. Another phenomenon is that of wound malodor, a common characteristic of chronic wounds linked in in vitro studies to short-chain fatty acids (SCFAs).54 These volatile compounds are the metabolic by-products of anaerobic bacterial metabolism.60 Malodor is associated with organisms known to generate SCFAs such as Bacteroides spp and anaerobic cocci.59 In an in vitro study, Stephens et al40 demonstrated that Peptostreptococci-generated SCFA inhibited the growth of key cells responsible for wound healing — eg, keratinocytes, fibroblasts, and endothelial cells. If translated to the in vivo situation, this could result in delayed healing from uncomplicated colonization (ie, no perceived clinical or cellular effects) without the bioburden necessarily reaching a theoretical infection threshold. Hansson et al’s in vivo study27 found Peptostreptococcus species (identified as P. magnus, P. asaccharolyticus, and P. prevotii) in 30% of venous leg ulcers. This is a clinically significant level of species-specific colonization and indicates the importance of anaerobic involvement in chronic wound bacteriology.
Short-chain fatty acids studies in vitro have been shown to play a part in impairing neutrophil chemotaxis and phagocytosis.40 The low pH of all chronic wounds facilitates succinate activity and provides a milieu that down-regulates neutrophil function.55,56
Differentiating Critical Colonization as a Distinct Stage of Infection
From the theses presented, it can be observed that a chronic wound colonized but not infected with one or more of certain bacteria (among them Morganella spp, P. aeruginosa, and Peptostreptococcus spp.) may exhibit erythema and delayed healing without a traditional or otherwise evident host response. Scenarios involving these organisms and possibly others yet to be identified have been used to postulate the concept of critical colonization.
The critical nature of colonization takes on a far greater significance when viewed in this light. A low level of colonization may be all that is required to delay healing and is far removed from that required for local infection to be diagnosed in terms of the level of bioburden and the demonstration of signs of infection (host response). The clinical indicators of infection developed by Cutting and Harding5 (and validated by Cutting61 a few years later regarding the decisions made by nurses on the infection state of a variety of wounds) should remain firmly aligned within the domain of local infection. Simply renaming local infection to be called critical colonization has no value.62
The fundamental message is that a number of possible mechanisms may allow micro-organisms to contribute to delayed healing without overt signs of infection. This is not to be confused with the subtle signs of infection as proposed by Cutting and Harding.5 Critical colonization is currently better explained from a microbiological perspective than from a clinical perspective. This should encourage clinicians to pay closer attention to delayed healing and its assessment. Currently, it has yet to be determined how frequently delayed healing can be attributed to a microbiological cause or to other factors. In chronic wounds, the fact that colonization is the norm should precipitate the conclusion that delayed healing is more likely than not to be microbiological in origin.
When encountered clinically, delayed healing may be perceived as an idiosyncratic event that defies rational explanation. In the absence of firm evidence to explain delayed healing — eg, malnutrition, smoking, comorbidities, less-than-optimal care — critical colonization should be considered not as a confounding feature but as a clinical probability based on the rationale presented. To put this into context, many clinicians have seen an indolent wound improve following topical antimicrobial treatment, retrospectively confirming the diagnosis of critical colonization.
Clearly, these are areas for research before mechanisms can be clearly defined. The thought processes and concepts outlined in this paper may offer a suitable starting point.
Wound microbiology, particularly in the so-called chronic wound, has justifiably achieved a high profile. While wound infection is a cause of morbidity and subsequent increased patient management costs, the state of delayed healing also presents cause for concern. Healing delays adversely affect patient quality of life and are used as justification for expensive modern wound treatments. The term critical colonization describes the situation of delayed healing with a microbial cause. It is likely that this state will vary between individuals and over time. It should be viewed microbiologically not purely quantitatively but also qualitatively, where its manifestation is dependent on the species present and thereafter by the expression of virulence determinants by those species. The goal in such situations is to consider treatment such as topical antiseptics that control the bioburden so healing may proceed. This being the case, recognizing that critical colonization is a distinct, clinically important stage in the wound infection continuum is important; not acknowledging that critical colonization is a cause of delayed healing (even without a traditional host response) impedes early diagnosis. Additional studies need to ascertain the point at which a wound is critically colonized as well as appropriate treatment to provide at that juncture to avoid additional morbidity and care costs.