Diagnosis and 2020 CONTRIBUTING AUTHORS Management of ANDREW J.M. BOULTON, MD, DSC (HON), FACP, FRCP Professor of Medicine, University of Manchester, Manchester, UK, and Visiting Professor of Medicine, Diabetic Foot University of Miami Miller School of Medicine, Miami, FL DAVID G. ARMSTRONG, DPM, MD, PHD Professor of Surgery, Keck School of Medicine of Infections the University of Southern California and Director, Southwestern Academic Limb Salvage Alliance (SALSA), Los Angeles, CA MATTHEW J. HARDMAN, PHD Professor of Wound Healing and Director of Research, Hull York Medical School, Hull, UK MATTHEW MALONE, PHD, FFPM RCPS (GLASG) Director of Research, South Western Sydney Local Health District, Limb Preservation and Wound Research Academic Unit, Ingham Institute of Applied Medical Research, Liverpool, Sydney, NSW, Australia, and Conjoint Senior Lecturer, Western Sydney University, School of Medicine, Infectious Diseases and Microbiology, Campbelltown Campus, Sydney, Australia JOHN M. EMBIL, MD, FACP, FRCPC Professor, Departments of Medicine (Section of Infectious Diseases) and Medical Microbiology, Max Rady College of Medicine, University of Manitoba, Winnipeg, Manitoba, Canada CHRISTOPHER E. ATTINGER, MD Chief, Division of Wound Healing, MedStar Georgetown University Hospital, Washington, DC BENJAMIN A. LIPSKY, MD, FACP, FIDSA, FRCP (LONDON), FFPM RCPS (GLASG) Emeritus Professor of Medicine, University of Washington School of Medicine, Seattle, WA, and Green Templeton College, University of Oxford, UK JAVIER ARAGÓN-SÁNCHEZ, MD PHD Chief, Department of Surgery and Diabetic Foot Unit, La Paloma Hospital, Las Palmas de Gran Canaria, Spain HO KWONG LI, MBBS, MRCP (UK), DTM&H Clinical Research Training Fellow, Medical Research Council—Centre for Molecular Bacteriology & Infection and Department of Infectious Disease, Imperial College London, London, UK GREGORY SCHULTZ, PHD Professor of Obstetrics and Gynaecology and Director of the Institute for Wound Research, University of Florida, Gainesville, FL ROBERT S. KIRSNER, MD, PHD Chairman and Harvey Blank Professor, Dr. Phillip Frost Department of Dermatology and Cutaneous This publication has been supported by unrestricted educational grants to the Surgery, University of Miami Miller School of American Diabetes Association from Healogics, Inc., and Organogenesis, Inc. Medicine, Miami, FL Diagnosis and Management of Diabetic Foot Infections is published by the American Diabetes Association, 2451 Crystal Drive, Arlington, VA 22202. Contact: 1-800-DIABETES, professional.diabetes.org.

The opinions expressed are those of the authors and do not necessarily reflect those of Healogics, Inc., Organogenesis, Inc., or the American Diabetes Association. The content was developed by the authors and does not represent the policy or position of the American Diabetes Association, any of its boards or committees, or any of its journals or their editors or editorial boards.

©2020 by American Diabetes Association. All rights reserved. None of the contents may be reproduced without the written permission of the American Diabetes Association.

To request permission to reuse or reproduce any portion of this publication, please contact [email protected].

Front cover image Credit: Mehau Kulyk/Science Source; Description: Foot bones. Colored X-ray of the bones in a healthy human foot, seen in top view.

Back cover image Credit: Du Cane Medical Imaging Ltd/Science Source; Description: Normal foot. Colored X-ray of a healthy human foot seen from the side (lateral projection). Dear ADA Journal Subscriber: About half of all hospital admissions for amputations involve people with diabetes, and diabetic foot infections (DFIs) are a major contributing cause. Fortunately, the past decade has brought numerous new treatments for conditions of the diabetic foot, and research is pointing the way toward more effective management of DFIs. This monograph,Diagnosis and Management of Diabetic Foot Infections, is published by the American Diabetes Association to inform primary care providers and other health care professionals of the latest developments in diabetic foot care. Supported by unrestricted educational grants from Healogics, Inc., and Organogenesis, Inc., this publication is a follow-up to a 2018 ADA compendium on foot complications (available at professional.diabetes.org/monographs and PMID 30958663). Whereas the first publication offered a broad overview of diabetic foot conditions, this second volume presents a detailed discussion of DFIs specifically. Led by Andrew J.M. Boulton, MD, DSc (Hon), FACP, FRCP, and David G. Armstrong, DPM, MD, PhD, members of the international expert writing team contribute important insights into the identification and successful treatment of DFIs from the fields of medicine and clinical care, basic scientific research, wound healing, infection control, microbiology, and surgery. The monograph begins from the viewpoint of basic science, describing the impact of infection on the healing process of experimental wounds. Next comes an explanation of biofilm development in chronic diabetic foot ulcers, the microbiology of DFIs, and the crucial role of debridement in ensuring positive outcomes of treatment. The authors provide a practical guide to the diagnosis and clinical management of DFIs and address current controversies in the treatment of diabetic osteomyelitis. The monograph ends with a review of new topical treatments and the role of emerging technologies in infection control. This essential resource brings together everything you need to know about recent major strides made in the management of DFIs. We hope it aids you in your efforts to keep acute foot injuries from progressing to chronic infected foot ulcers in your patients with diabetes and thereby to improve their long- term health and well-being. Sincerely,

Kenneth P. Moritsugu, MD, MPH, FACPM, FAADE(H) Rear Admiral, USPHS (Ret) Interim Chief Science, Medical, and Missions Officer Team of Teams Leader American Diabetes Association Andrew J.M. Boulton, MD, DSc (Hon), FACP, FRCP1,2 David G. Armstrong, DPM, MD, Diagnosis and PhD3,4 Matthew J. Hardman, PhD5 Matthew Malone, PhD, FFPM RCPS Management of (Glasg)6,7 John M. Embil, MD, FACP, FRCPC8 Diabetic Foot Christopher E. Attinger, MD9 Benjamin A. Lipsky, MD, FACP, FIDSA, FRCP (London), FFPM RCPS (Glasg)10,11 Infections Javier Aragón-Sánchez, MD, PhD12 Ho Kwong Li, MBBS, MRCP (UK), DTM&H13,14 Gregory Schultz, PhD15 ABSTRACT | This compendium is a follow-up to the 2018 American Dia- 2 betes Association compendium Diagnosis and Management of Diabetic Robert S. Kirsner, MD, PhD Foot Complications. Whereas the first compendium offered a broad gen- eral overview of diabetic foot conditions, this second volume presents 1 a detailed discussion of the prevention and treatment of diabetic foot University of Manchester, Manchester, UK 2University of Miami Miller School of infections (DFIs), a major contributor to high amputation rates among Medicine, Miami, FL people with diabetes. The treatise begins from the viewpoint of basic 3Keck School of Medicine of the University science, describing the impact of infection on the healing process of of Southern California, Los Angeles, CA experimental wounds. There follow overviews of biofilm development 4Southwestern Academic Limb Salvage Alliance (SALSA), Los Angeles, CA in chronic diabetic foot ulcers (DFUs), the microbiology of DFIs, and the 5Hull York Medical School, Hull, UK crucial role of debridement in ensuring positive outcomes of DFI treat- 6South Western Sydney Local Health ment. Next, the authors provide a practical guide to the diagnosis and District, Limb Preservation and Wound clinical management of DFIs. Current controversies regarding the treat- Research Academic Unit, Ingham Institute of Applied Medical Research, Liverpool, ment of osteomyelitis are addressed, including the relative value of anti- Sydney, NSW, Australia biotics versus surgery and the use of intravenous versus oral antibiotics. 7Western Sydney University, School The compendium closes with a look at new topical treatments and the of Medicine, Infectious Diseases and role of emerging technologies in infection control. Microbiology, Campbelltown Campus, Sydney, Australia 8Max Rady College of Medicine, University of Manitoba, Winnipeg, Manitoba, Canada fter the success and positive reception of the American Diabetes 9MedStar Georgetown University Hospital, A Association’s 2018 compendium Diagnosis and Management of Washington, DC Diabetic Foot Complications (1) the association asked us to proceed 10University of Washington School of with a second volume. The first publication offered a broad general Medicine, Seattle, WA 11Green Templeton College, University of overview of diabetic foot issues, encompassing the etiopathogenesis Oxford, UK of complications, screening, and wound classification; management 12Department of Surgery and Diabetic Foot Unit, La Paloma Hospital, Las Palmas de of diabetic foot ulcers (DFUs) and diabetic foot infections (DFIs); Gran Canaria, Spain recognition and treatment of peripheral artery disease (PAD) and 13Medical Research Council—Centre Charcot neuroarthropathy; off-loading, wound management, and for Molecular Bacteriology & Infection, London, UK adjunctive therapies; and maintenance of the foot in remission. 14Department of Infectious Disease, In the past few years, there has been a renaissance in diabetic foot Imperial College London, London, UK care with respect to evidence-based treatments (2). Examples include 15University of Florida, Gainesville, FL the LeucoPatch system (3), topical oxygen delivery (4,5), and, for neu- Address correspondence to roischemic ulcers, sodium octasulfate dressings (6). There has also Andrew J.M. Boulton, been progress in the management of DFUs with infection. For exam- [email protected], ple, a randomized controlled trial (RCT) confirmed that treatment with and David G. Armstrong, antibiotics is noninferior to local surgery for localized diabetic foot os- [email protected]. teomyelitis (DFO) (7). (See related discussion on p. 12.) Most recently, the OVIVA (Oral Versus Intravenous Antibiotics) trial confirmed that, ©2019 by the American Diabetes for complex bone and joint infections, oral antibiotic therapy had sim- Association, Inc. ilar outcomes to intravenous (IV) the Infectious Disease Society of DFUs are characterized by accu- therapy (8). (See related discus- America will soon be updating its mulation of senescent cells and sion on p. 13.) Because foot infec- 2012 guidelines (11) on the treat- lack responsiveness to the normal tions are a major contributor to ment of infected DFUs. cues that drive timely repair. At amputation, we decided to make the molecular level, there are still infected DFUs the focus of this major gaps in our knowledge of second foot care compendium. DFU ontogenesis, underscored by The first sections herein cover How Infection a distinct lack of biological thera- the impact of infection on healing Impairs Wound pies for wound care. of experimental wounds (p. 2), the Healing How does this profound state of importance of biofilms (p. 4), and dysregulation occur, and perhaps a general overview of the microbi- DFUs are a complex, multifac- more importantly, why do some ology of DFIs (p. 6). Although de- torial clinical problem that, de- DFUs resolve quickly, where- bridement of DFUs was covered in spite decades of research, remain as others persist for months or the first compendium, we deemed poorly understood at the basic even years? It seems increasingly it important enough to include science level. Over millennia, likely that the answer lies in here as well, given its pivotal role in our bodies have evolved a high- the concept of wound infection, the management of DFIs. Where- ly orchestrated healing response or more specifically, in every as molecular markers might in- to skin wounds through which wound’s unique bacterial signa- form the extent of wound debride- inflammation, cell proliferation, ture, or microbiome. ment in the future (9), at present, migration, angiogenesis, matrix Infection is defined clinically clinical assessment is used to en- deposition, and remodeling are by the cardinal signs and symp- sure that adequate debridement sequentially activated to repair toms of redness, heat, swelling, is achieved, as reviewed starting injured tissue. We know that, in and pain, with nearly half of on p. 7. Subsequent sections cover DFUs, repair becomes dysregu- all DFUs classified as infected the management of infected DFUs lated with widespread tissue ne- (12). Diagnosis of DFU infection (p. 9) and discussions of antibiot- crosis, hyperproliferative wound strongly correlates with subse- ics versus surgery for osteomy- callus, and excessive inflamma- quent amputation and increased elitis (p. 12) and the OVIVA trial tion. This drives a matrix metallo- mortality. The current dogma is (p. 13). There is little doubt that proteinase (MMP)–rich proteo- that open wounds become con- the OVIVA trial will challenge the lytic wound bed, which impedes taminated with pathogenic mi- current management of osteomy- granulation, delays vasculariza- croorganisms, which then col- elitis, in which IV antibiotics are tion, and results in a hypoxic local onize the tissue (Figure 1). This still commonly used. The remain- environment. At the cellular level, transition from contamination ing sections cover potential topi- cal treatments for DFIs (p. 15) and the role of modern technology in infection control (p. 17). In addition to this compendium and the previous one (1), we want to direct readers’ attention to two forthcoming publications that will surely be of interest. First, in the wake of the 8th International Sym- posium on the Diabetic Foot held in May 2019 in The Netherlands, the International Working Group on the Diabetic Foot (IWGDF) has recently published its revised and updated guidelines on the man- FIGURE 1 Gram stain allows visualization of bacteria within a DFU tissue sample. Gram-positive agement of DFIs (10). In addition, bacteria stain dark blue, gram-negative bacteria stain pink, and DFU tissue stains light blue.

2 | DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS to colonization is aided by the complexity and diversity of the terial virulence mechanisms have unique DFU microenvironment: chronic wound microbial ecosys- been studied in detail. For exam- hyperglycemic, ischemic tissue tem. Bacteria, fungi, and other ple, Pseudomonas aeruginosa is with dry, cracked neuropathic microorganisms do not exist in optimized for adherence to the epidermis. The transition to isolation; they use an array of sig- surface of the skin epithelium but wound infection is further sup- naling molecules to interact with is also able to invade into the deep- ported by functional deficiencies each other and with host tissue. er wound bed tissue (16). Pseudo- in diabetic immune cells, which Established microbial communi- monas can even exist inside host at this early stage of infection are ties exist as structurally complex cells (keratinocytes), where it unable to efficiently phagocytose biofilms, altering their metabolic effectively evades the immune bacteria or release antimicro- state, gene expression, and envi- system. Cells exposed to live bial factors to manage the in- ronment (15). Biofilm bacteria P. aeruginosa or Pseudomonas- vading pathogens. There is good are surrounded by a protective derived lipopolysaccharide in evidence that the tipping point matrix that confers both pro- vitro demonstrate impaired mi- to local infection is marked by tection from the host immune gration, increased proliferation, a shift in wound bacterial com- system and recalcitrance to ex- and increased cell death. These position, specifically toward in- ogenous antimicrobials. In many findings are supported by in vivo creased local anaerobic bacteria ways, the number of microorgan- models demonstrating impaired (13). Interestingly, wound de- isms present is irrelevant; it is wound healing after the adminis- bridement has been shown to be the environment in which those tration of live bacteria or bacterial particularly effective at remov- organisms exist that directs phe- components directly on experi- ing these unwanted anaerobes. notype and virulence, and these mental wounds. If left untreated, local infection factors ultimately trigger the Current research, underpinned can quickly transit to widespread host response that manifests as by the powerful new technologies overt infection with involvement clinical signs of infection. of shotgun metagenomics and of other tissues such as bone. Wound microorganisms direct- long-read sequencing, is opening The concept that wound bac- ly modulate wound cell behav- our eyes to the bewildering com- teria influence healing (both ior and modify healing capacity. plexity of the skin and wound positively and negatively) in the Compared to the relatively new microbiota. Historically, we have absence of infection may seem area of DFU microbial composi- relied on culture-based methods counterintuitive. Yet, recent culture- tion profiling, far more is known to identify wound bacteria. In- independent studies have re- about how bacteria interact with deed, bacterial culture remains vealed that DFU microbiomes human skin cells and tissues, the front-line diagnostic tech- in the absence of infection are an area for which the relevant nique for wound infection. The dynamic, heterogeneous, and laboratory techniques have been problem with these culture-based incredibly complex. Wound bac- readily available for years. Wound methods is that they identify only terial signatures strongly link to cells sense the bacterial compo- a small fraction of constituent healing outcome in noninfected sition of a wound via an array of microbes. Bacterial culture has DFUs (14), suggesting a host- pattern recognition receptors promoted the idea that Staphylo- microbe relationship that is more (PRRs). These PRRs recognize coccus aureus, Streptococcus, and subtle than previously thought. specific bacterial components P. aeruginosa are the main or- Indeed, it is likely that DFU bac- known as pathogen-associated ganisms in DFUs. However, in a teria directly modify the wound molecular pattern molecules, game-changing shotgun sequenc- milieu, altering the local micro- which in turn upregulate host an- ing study of 100 DFUs, Kalan et climate (e.g., pH) and producing timicrobial peptides to actively al. (17) reported detailed genus-, metabolites that directly affect modulate the skin/wound micro- species-, and even strain-lev- cellular healing. bial composition. el characterization of the DFU So, why do some wounds pres- Defects in a number of PRRs microbiome. Intriguingly, these ent with high bioburden but no have been linked to diabetes, offer- authors showed that a strain of signs of infection and vice versa? ing a mechanism for the observed S. aureus (SA10757) linked by The answer appears to lie in the bacterial dysbiosis. Specific bac- metagenomics to clinical healing

DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS | 3 outcome also directly delayed demonstrated that most acute healing in a mouse model. A key Biofilms in the wounds are not complicated by theme in emerging microbiome Context of DFUs biofilms. It is also pertinent to studies is the strong links among note that, whereas biofilms may reduced diversity, increased tem- form in wounds such as DFUs, poral stability, and poor healing. WHAT ARE BIOFILMS? not all biofilms may act pathogen- The individual nature of the There is no current consensus ically or be detrimental to wound microbiome poses an ongoing among researchers regarding healing. the definition of biofilm, but col- challenge. When the microbi- lectively from within the litera- ome signature differs substan- ture, definitions often describe BIOFILMS AND THEIR tially among individuals, how medically related biofilms as OCCURRENCE IN DIABETIC does one look for common “heal- “aggregates of microorganisms FOOT WOUNDS ing” or “nonhealing” signatures? embedded in a matrix of extra- Biofilms are not the prima- The answer is to employ care- cellular polymeric substances.” ry mechanism behind the de- fully designed (but expensive Biofilms can attach to host tissue velopment of foot ulceration; and time consuming) longitudi- or in-dwelling medical devices or rather, ulceration occurs as a nal profiling studies looking for exist in fluids adjacent to those result of precipitating factors common changes in composi- surfaces. In contrast to planktonic that include peripheral neurop- tion or diversity. microorganisms, biofilms demon- athy (loss of protective sensa- So, what does the future hold strate reduced growth rates and tion), altered foot architecture, for the treatment of infect- altered gene expression. These trauma, and PAD. These fac- ed DFUs? It is widely accept- changes may help to explain why tors contribute to breaks in the protective barrier of the skin. ed that the number of infected biofilms show an enhanced tol- However, once pathogenic bio- DFUs requiring treatment will erance to antimicrobials and the films become established in continue to increase, driven host immune response. DFUs, they may contribute as a by the expanding number of cause of chronic and persistent diabetic and elderly patients. BIOFILMS IN HUMAN infections, which may delay ul- In the short term, emerging HEALTH AND DISEASE cer healing. In vitro and animal technologies to detect bacteria Biofilms can be found in medi- model research has demonstrat- (e.g., MolecuLight i:X; Smith cal, industrial, and natural envi- ed that biofilms can delay wound ronments and may affect human & Nephew, Hertfordshire, UK) healing. However, the transla- health both positively and nega- will be essential to guide treat- tional evidence from human tively. When biofilms are impli- ment. In the next 10 years or clinical studies demonstrating cated in human disease, they are so, there will likely be wide- biofilms as causal mechanisms widely acknowledged as a cause spread implementation of point- for delayed ulcer healing or as of chronic and persistent infec- of-care microbiota profiling drivers of chronic infections in tions (18). It is generally believed methodologies in clinics. With the feet of people with diabetes that biofilms are not the cause of these methods, bacterial com- is scant and requires further ex- acute infections; these are caused munity–level genetic resistance ploration (17,19,20). by planktonic microorganisms. profiling will revolutionize the When breaches in the skin en- ways in which we diagnose, velope occur in the feet of people DIAGNOSIS OF BIOFILMS manage, and treat infection. Ul- with diabetes, most open wounds Contrary to general belief, bio- timately, we will reach a level are colonized by microorganisms; films are not visible to naked-eye of understanding of the host- however, this does not mean the observation. Wound bed material, microbe continuum that per- microorganisms will act patho- such as slough or fibrin, or obser- mits a personalized treatment genically and incite a host re- vations of a shiny translucent lay- approach, augmenting the en- sponse (i.e., clinical signs of in- er may be mistaken for biofilms. dogenous host defense while fection). Likewise, not all wounds There is no scientific validation supporting skin commensals. will form biofilms. Evidence has to suggest that these clinical fea-

4 | DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS tures are biofilms or proponents confirm their presence. Further- within the preexisting diabet- of biofilm. Additionally, biofilms more, microbiological specimens ic foot literature. Most DFUs do not form homogenously on sent for conventional culture do contain polymicrobial biofilms. the surface of wounds. They are not necessarily identify microor- Aerobic gram-positive cocci heterogeneously distributed in ganisms in biofilms (because of (staphylococci and streptococ- aggregates and may form on both their reduced growth), and most ci) are predominant. In addition the wound surface and deeper clinical microbiology laborato- to aerobic species, other bacteria structures, making diagnosis and ries will not have the capability commonly identified in the same treatment more difficult. to perform biofilm-specific cul- foot ulcers include fastidious an- In keeping with current expert ture approaches. aerobes (namely, those belonging guidelines (10), the optimal sam- To circumvent the lack of di- to Clostridiales Family XI), Cory- pling technique for determining agnostics, the clinical character- nebacterium sp., and gram-nega- pathogens from infected DFUs istics that best define a chronic tive rods (namely, Klebsiella spp., is to obtain an appropriate tissue infection may aid clinicians in Acinetobacter spp., Enterobacter specimen. Obtaining tissue speci- identifying potential biofilm in- spp., P. aeruginosa, and Esche- mens is also the only direct way to fections in DFUs (Figure 2). richia coli). visualize the presence of biofilms using microscopy (18). Biofilm vi- MICROBIOLOGY OF MANAGEMENT OF BIOFILMS sualization is best performed us- Across the spectrum of human BIOFILMS IN DFUS ing microscopy approaches such diseases caused by biofilms (i.e., as scanning electron, confocal, Most studies exploring the pres- ence of biofilms in DFUs have periodontal, in-dwelling medical transmission, or light microscopy. device, and human tissue or bone employed molecular (DNA) The major limitation is that these infections), the standard treat- sequencing technologies and techniques are typically confined ment approach is to physically thus report an extended view of to bench research and are not remove the biofilms. Removing the diabetic foot microbiome. clinically practical. biofilms from infected tissue or Nonetheless, the predominant Adding to the complexity of bone via debridement is one of the microorganisms identified from diagnosing biofilm infections, most important treatment strat- DFUs with biofilm formations there are currently no routine egies (15). However, it may not diagnostic tests or biomarkers to are those commonly reported be possible to completely eradi- cate biofilms from tissue or bone Failure of antibiotic treatment Prolonged classic signs and symptoms of because treating clinicians are and recurrence of the infection, inflammation and infection. Inflammation is unable to see biofilms and there- particularly if evidence is typically reported as being low-grade. Clinicians provided that the same organism should remain aware that acute-on-chronic fore may not debride all infected is responsible on multiple time presentations may also occur. It is hypothesized tissue and because biofilms are points and no known antibiotic that biofilm dispersal may increase the number tolerant to antimicrobials and the resistance exists. of planktonic microorganisms. host immune response. Systemic antibiotics may have little effect Evidence of multiple against chronic biofilm infections Medical history of or recurrent clinical in DFUs; therefore, antimicrobial biofilm-predisposing infections in the same stewardship must be considered condition (e.g., ulcer that resolve with implanted medical antibiotic therapy only with controlled use to help man- device). to recur after therapy age planktonic bacteria (acute in- has ceased. fection) and to prevent associated systemic infections. In the context of local wound Culture negative results despite Secondary or covert optimal sampling technique signs of infection such as care for DFUs, there is also a lack and a high suspicion of friable granulation tissue of high-quality evidence from infection by the clinician. or wound undermining. human in vivo clinical trials per- taining to the effectiveness of top- FIGURE 2 Clinical signs and symptoms that may indicate chronic biofilm infection of a DFU. ical antimicrobial agents against

DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS | 5 biofilms. There are currently infections that require treatment are two or more clinical findings no agents with Level 1 evidence are characterized by bacterial of infection (Table 1) (10). (i.e., based on RCTs, systematic invasion of skin, other soft tis- Antimicrobial therapy is re- reviews, or meta-analyses). In the sues, or bone. The clinical man- quired for the treatment of DFIs absence of RCTs or case-control ifestations of DFIs include local (11,21) but is not required for un- studies, clinicians should con- warmth, erythema, induration, infected wounds (10,11,21). Initial sider the cost burden of using pain or tenderness, and puru- treatment of DFIs typically in- topical agents for which there are lence. The presence of these clin- cludes empiric antibiotic therapy, limited data. ical signs and symptoms should because specimen culture results alert clinicians to consider pro- usually are not yet available. For ceeding with further evaluation infected wounds, specimen cul- and treatment (11,21). ture results help to identify the Evolving There are several classification causative pathogens, allowing Microbiology of DFIs systems to define the presence for any necessary modification and severity of infection in the to optimize antimicrobial thera- In evaluating and treating DFUs, feet of people with diabetes. The py. Specimen culture results for it is important to distinguish be- IWGDF categorizes feet as un- clinically uninfected DFUs do not tween microbial colonization and infected if there are no local or meaningfully benefit decisions on infection. Although ulcers may be systemic manifestations of infec- therapy (10,11,21). colonized with microorganisms, tion and as infected when there DFUs typically are colonized

TABLE 1 IWGDF Classification for Defining the Presence and Severity of DFIs

IWGDF CLASSIFICATION DEFINITION UNINFECTED 1 – Uninfected No local or systemic symptoms or signs of infection INFECTED* ⊲ At least two of these features are present: • Local swelling or induration • Erythema >0.5 cm around the wound • Local tenderness or pain • Local increased warmth • Purulent discharge ⊲ No other cause of skin inflammation (e.g., trauma, gout, Charcot arthropathy, fracture, thrombosis, or venous stasis disease) 2 – Mild infection ⊲ Infection with no systemic manifestations (see below) involving: • Only skin or subcutaneous tissue (not any deeper tissues), and • Any erythema present does not extend >2 cm3 around the wound† 3 – Moderate infection ⊲ Infection with no systemic manifestations, and involving: • Erythema extending ≥2 cm from the wound margin, and/or • Tissue deeper than skin and subcutaneous tissues (e.g., tendon, muscle, joint, bone; see below) 4 – Severe infection ⊲ Any foot infection with two or more associated systemic manifestations (per systemic inflamma- tory response syndrome [SIRS] criteria): • Temperature >38° C or <36° C • Heart rate >90 bpm • Respiratory rate >20 breaths/min or partial pressure of carbon dioxide <4.3 kPa (32 mmHg) • White blood cell count >12,000/mm3, <4,000/mm3, or >10% immature (band) forms 3(O) or 4(O) – Moderate ⊲ Moderate or severe infection that also involves bone (osteomyelitis)‡ or severe infection with associated osteomyelitis Adapted from ref. 10. *Infection refers to involvement of any part of the foot, not just a wound or an ulcer. †In any direction from the rim of the wound. The presence of major foot ischemia makes diagnosis and treatment of infection more difficult. ‡If osteomyelitis is present in the absence of two or more signs or symptoms of local or systemic inflammation, the foot is classified as having grade 3(O) (if <2 SIRS criteria) or grade 4(O) (if ≥2 SIRS criteria).

6 | DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS initially by gram-positive bacteria resistant pathogen causing DFI, ment, possibly contributing to the such as S. aureus (methicillin- accounting for one-third of in- predominance of gram-negative susceptible S. aureus or methicil- fections (22). In a retrospective pathogens (30). lin-resistant S. aureus [MRSA]) case-control study from China, In Africa, some studies of DFIs and the β-hemolytic streptococci MRSA infections, both hospi- have shown a predominance of such as group A Streptococcus tal-acquired and especially com- gram-positive pathogens such as (Streptococcus pyogenes), group munity-associated, were prev- S. aureus (38,39), but one study B Streptococcus (Streptococcus alent (29). Another study from showed that gram-negative ba- agalactiae), or group C or G Strep- China showed that MRSA ac- cilli were more common than tococcus. As an ulcer becomes counted for 24.5% of all strains of gram-positive bacteria, with chronic, it may contain devital- S. aureus (30). antimicrobial resistance noted ized, necrotic tissue and become In India, DFIs commonly are in the gram-negative but not colonized with gram-negative caused by gram-negative bacil- the gram-positive bacteria (40). bacteria such as P. aeruginosa; li (31). Furthermore, pathogens In Morocco and Brazil, gram- the Enterobacteriaceae, including from India (32,33) and other de- negative pathogens may be more E. coli, Proteus spp., and Klebsi- veloping nations, such as Egypt prevalent than gram-positive ella spp.; and anaerobic bacte- (24) and China (34), have be- bacteria (41,42), and the study ria such as Bacteroides spp. and come highly resistant to antimi- from Brazil showed MRSA in 22% Clostridium spp. (11,21). With crobials. DFIs in regions with of wounds, including one-third hospitalization and treatment warmer climates, including In- of MRSA strains that also were with broad-spectrum antibiotics, dia, the Middle East, and Africa, resistant to vancomycin (41). ulcer flora may change to include most commonly are caused by In summary, DFI are caused antimicrobial-resistant pathogens P. aeruginosa (35), attributed in primarily by gram-positive bac- such as MRSA, vancomycin- part to the warm, humid envi- teria such as S. aureus in North resistant Enterococcus, and gram- ronment causing foot sweating, America and Europe. Although S. negative bacteria that produce self-treatment with antimicrobi- aureus remains a frequent patho- extended spectrum β-lactamase or als, and self-contamination from gen in developing countries (42), carbapenemase enzymes (22–24). suboptimal perineal and hand gram-negative pathogens may Most current knowledge about hygiene (36,37). be predominant strains in India, the microbiology of DFIs has been A large multicenter study from Pakistan, the Middle East, Africa, derived from studies conducted in Beijing, China, showed that, of China, and Brazil (28–33,39–42). North America and Europe, where the microorganisms recovered the most prevalent pathogens are from DFIs, gram-negative ba- gram-positive bacteria such as S. cilli accounted for 57.5%, and aureus (10,11,21). In these West- gram-positive bacilli accounted Debridement: ern countries, MRSA has become for 39.6%. The most prevalent The First Step in a frequently occurring DFI patho- microorganisms recovered were Controlling DFIs gen (25,26). Infection with MRSA Enterobacteriaceae (41%) and also may occur in patients who do Staphylococcus spp. (25.4%), of Establishing a healthy wound not have any typical risk factors which S. aureus comprised 17.1%, bed through adequate debride- for acquisition of MRSA (e.g., hos- and MRSA comprised 24.5% of ment of infected, senescent, and pitalization or previous antibiot- all S. aureus isolates (30). Many devitalized tissue is central to ic therapy) because of increased patients in this study from Bei- the progression of normal wound prevalence of MRSA in the com- jing who had DFUs, especially healing in diabetic ulcers. In this munity (27). MRSA infections of older adults, were receptive to section, we review the minor and DFUs are an increasing problem receiving traditional medicines, major excisional debridement in India, as are polymicrobial and and >50% of outpatients were techniques that represent cur- gram-negative bacilli infections prescribed antibiotics. The high rent medical practice (43). (28). In Detroit, MI, in the United incidence of self-treatment with Debridement is derived from States, MRSA has been reported antibiotics may have caused de- the French débridement, which as the most common multidrug- lays in seeking medical treat- means to remove a constraint.

DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS | 7 The clinical definition of debride- ings and dry gauze to facilitate selectively dissolves necrotic tis- ment has subsequently evolved to removal of infected and nonvia- sue and biofilms into a nutrient- include the removal of nonviable or ble tissue. Wet-to-dry dressing rich food source, while sparing contaminated tissue that impedes involves applying moist gauze to a healthy tissue. This process re- normal tissue growth. Debride- wound, then removing it once dry duces the bacterial burden that ment enables the wound and sur- and adherent to underlying tissue. often plagues gangrenous, recalci- rounding tissue to promote nor- Both wet-to-dry dressings and trant wounds. mal healing by removing infection, dry gauze erratically tear necrotic Surgical Debridement biofilms, and senescent cells. tissue from the underlying wound Excisional debridement can be Ultimately, durable restoration and are often painful; they are accomplished as a minor proce- of soft-tissue coverage is de- usually insufficient for adequate dure, in the clinic or at patients’ pendent on the satisfaction of a wound bed preparation due to bedside, or as a major procedure number of requisite objectives, fluid loss, surfacing cooling, vaso- under regional block or general including eradication of infection constriction, impaired immune anesthesia in the operating room. or reduction of bioburden, im- response, and local tissue hypoxia. When performed as a minor provement of local blood flow, re- Biological Debridement procedure, excisional debride- vitalization of the wound bed, and Autolytic dressings (i.e., hydro- ment is limited by pain (mitigable correction of biomechanical ab- gels, hydrocolloids, and polymeric with local block or topical lido- normalities. Debridement, when membrane formulations) are in- caine) and the risk of having to performed correctly, optimizes dicated for wounds with necrotic deal with major bleeding. Clinic or diabetic wound healing by meet- tissue or fibrin coats and act to bedside debridement can be an ef- ing these objectives. soften fibrotic wound margins as fective means of wound tempori- they stimulate release of endoge- zation or a definitive treatment in DEBRIDEMENT nous proteolytic enzymes. Using some cases and can be performed TECHNIQUES these dressings is relatively pain- easily with local anesthetic in pa- Multiple techniques are used to less, which represents a major tients with retained sensibility. debride DFUs, and these can be advantage for patients, particu- Such minor procedures involve categorized as mechanical, bio- larly those who are sensate. These the use of a curette to scrape the logical, and surgical methods. We dressings most often benefit pa- coagulum, which contains both have found the surgical approach tients who have minimal necrotic MMPs and biofilms, off of the to be the most effective. However, loads and cannot tolerate more wound surface. Although the un- determining the most appropriate aggressive forms of debridement. derlying tissue may look clean, , which a curette only removes 0.5 loga- technique mandates the consider- Enzymatic ointments rely on directly hydrolyzing pep- rithm of bacteria. Using a knife or ation of host-specific factors (e.g., tide bonds, are recommended for scissors and a pickup, a surgeon comorbidities, compliance, and moist or fibrotic wounds, partic- can actually cut all necrotic tissue social support) and wound-related ularly in patients who are poor within the wound. By staying at factors (e.g., infection/contamina- surgical candidates. Enzymes se- the interface between the necrot- tion status, perfusion, and viabili- lectively digest devitalized tissue. ic and viable tissue, the surgeon ty), as well as the resources avail- This process causes less trauma can minimize the risk of bleeding. able at the treatment facility. The to healthy tissue than surgical Again, remember that staying at European Wound Management debridement, but it debrides at a the wound surface minimally de- Association’s guidelines for de- very slow rate. creases the bacterial count. Any bridement provide specific infor- Maggot debridement therapy, bleeding can usually be handled mation regarding the indications, using the radiated larvae of the with direct pressure or topical sil- contraindications, and potential blowfly Phaenicia sericata, is a ver nitrate. adverse effects associated with proven, cost-effective alternative As a major procedure, exci- each technique (44). for treating drug-resistant, chron- sional debridement involves the Mechanical Debridement ically infected wounds in patients direct excision of all infected, Mechanical debridement includes who are poor surgical candidates. necrotic, and inflamed tissue of the use of both wet-to-dry dress- Maggots secrete an enzyme that a wound using a combination

8 | DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS of knife (steel or hydrosurgical), surgeon with a color reference to scissors, curettes, rongeurs, pow- help in removing the entire base Managing Infected er burrs, and sagittal saws. Surgi- of the wound. In addition, the sur- cal debridement is the preferred geon must be familiar with nor- DFUs method used for situations in mal tissue colors (i.e., red [muscle Among patients with diabetes which urgent or emergent wound and blood], white [tendon, fascia, presenting with a foot wound, decompression is required, deep- nerve, and bone], and yellow [fat]). about half have clinical evidence er structures (e.g., bone, joints, or Using these colors as a guide pro- of infection (10). DFIs typically tendons) are involved, and major vides an endpoint to debridement begin in a break in the protec- bleeding is anticipated. and helps to prevent removal of tive skin envelope, which allows Major surgical excisional de- healthy tissue. Finally, excision of organisms that are either intro- bridement procedures involve 3–5 mm of marginal tissue during duced by trauma or colonizing the tangential excision of all gross- initial debridement removes se- surrounding skin to gain entrance ly contaminated and devitalized nescent cells and residual bio- to subcutaneous tissues. Unless tissue until only normal tissue films from chronic wound edges checked by host defenses or med- is present. Removing the indu- and permits healthy underlying ical interventions, infections can rated and inflamed soft tissue at cells to progress through the stag- spread contiguously to deeper soft all borders of the wound ensures es of normal wound healing. Pre- tissues, including tendons, liga- the removal of deeply buried bio- and post-debridement cultures ments, joints, and bone. The de- films that can easily recolonize should be obtained to guide future velopment of a foot infection in a the wound. Thin serial slicing therapy. Surgical wounds may re- person with diabetes is a sentinel minimizes the amount of viable quire debridement every 2–3 days event; it is the most common dia- tissue sacrificed, while ensuring until negative post-debridement betes-related reason for hospital- that only healthy tissue remains. cultures are obtained. izations and in most countries is Gentle tissue handling, sharp Debridement is an essential now the principal cause of lower- dissection, and pinpoint or bi- component in the management extremity amputations. The good polar cauterization of bleeding of both acute and chronic wounds news, however, is that recent stud- vessels serve to minimize trauma because it removes infected tis- ies have demonstrated that rapid and promote tissue viability. The sue and contaminants, biofilms, recognition and appropriate man- surgeon should avoid harmful and senescent cells that impede agement of DFIs can usually avert maneuvers such as crushing skin the normal progression of wound these adverse outcomes (10). edges with forceps or clamps, healing. The mechanical, biolog- burning tissues with electro- ical, and surgical techniques de- cautery, or suture-ligating healthy scribed above each have a role, DEFINING INFECTION perivascular tissues. depending on the nature of the Because all open wounds will be Three technical adjuncts can specific wound and patient char- colonized by microorganisms, we be used in combination to ensure acteristics. The appropriate use cannot define a DFI merely by the adequate debridement of the en- of technical adjuncts such as growth of microorganisms (even tire wound: 1) topical staining topical tissue staining, tangen- potentially virulent pathogens) on of the wound surface with dye tial excision, and color-guided culture of the wound. Rather, in- (skin marker or methylene blue), debridement can enhance the fection is defined by the response 2) use of a color-guided approach efficiency of wound bed prepara- of the host (i.e., the presence of to debridement (down to healthy tion and expedite time to closure. at least two of the classic signs red, yellow, and white tissue), and Although debridement methods and symptoms of inflammation). 3) tangential excision of indurat- have improved over time, there is However, these findings may be ed or senescent wound margins. potential for further refinement. altered in patients with peripher- Before debridement, methylene Future prospective RCTs will aid al neuropathy or PAD, which are blue with a cotton applicator or in the establishment of evidence- comorbidities in most patients the pulled-out tip of a skin mark- based guidelines to standardize with a diabetic foot wound. Thus, er should be liberally applied to debridement practices for com- some clinicians accept the pres- the wound base. This provides the plex wound management. ence of secondary signs such as

DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS | 9 friable granulation tissue, wound resonance imaging or radiolabeled DFI TREATMENT undermining, and foul odor as ev- scintigraphy, may be appropriate DFIs can progress rapidly. Thus, idence of infection. for some patients in whom the while awaiting the results of cul- Once a DFI is diagnosed, clas- presence or absence of osteomyeli- tures (and any additional diag- sifying its severity using stan- tis is uncertain, or when planning nostic studies), clinicians should dardized criteria helps to define a surgical intervention (46). Defin- initiate empiric antibiotic thera- both the approach to treatment itively diagnosing bone infection py for most DFIs. Base the choice and the prognosis. Classifica- requires aseptically collecting a of a regimen on the clinical char- tion requires careful clinical ex- bone specimen either during sur- acteristics and severity of the amination of the wound and re- gery or percutaneously. Findings infection, any clues to the likely view of results of laboratory and of a positive culture or histologi- pathogens or recent culture re- imaging tests to determine the cal evidence of inflammation and sults, any history of recent antibi- depth and extent of infection and necrosis (preferably both) are the otic therapy, and knowledge of lo- whether there is bone involve- criterion standard for diagnosing cal antibiotic resistance patterns. ment or evidence of systemic in- osteomyelitis. The microbiology of DFIs is fection. Clinicians should probe discussed in more detail else- such wounds to delineate their CULTURES where in this compendium (p. depth and seek palpable bone, A clinically uninfected diabetic 6). Briefly, in Western countries, which is highly suggestive of os- foot wound should not be cul- the most common pathogens are teomyelitis, or foreign bodies. tured because it does not re- aerobic gram-positive cocci, es- Infection involving an area of quire antimicrobial therapy. All pecially S. aureus. For nonsevere <2 cm of skin and only superfi- appropriately diagnosed DFIs infections, in the absence of risk cial tissues is classified as mild, should be cultured to define the factors for gram-negative patho- whereas those with ≥2 cm of cel- causative pathogens and their gens (e.g., previous antibiotic lulitis or involving subcutaneous antibiotic susceptibilities. Tis- therapy or hospitalization) or ob- tissues are deemed moderate. sue specimens collected by cu- ligate anaerobes (e.g., ischemia In the forthcoming 2020 update rettage or biopsy provide more or gangrene), relatively narrow- to the IWGDF’s infection clas- specific and sensitive culture spectrum therapy (active against sification system, the presence results than wound swabs. For staphylococci and streptococci) of bone infection (osteomyeli- osteomyelitis, cultures of bone often suffices. For severe infec- tis) is designated separately (as more accurately reveal (and tions, it is safer to initially pre- “O”) and is not part of the clas- generally demonstrate fewer) scribe a broader-spectrum regi- sification of moderate or severe pathogens than those of even men (48). For a clinically stable infection (10). The presence of deep soft tissue. Blood cultures patient with a chronic infection findings of systemic inflammato- are only needed for patients or at risk for unusual or resistant ry response syndrome, especially with evidence of sepsis syn- pathogens (e.g., due to recent an- fever or leukocytosis, defines a drome. More recent studies us- tibiotic treatment), discontinuing severe infection. ing molecular microbiological or withholding antibiotic therapy For all but the mildest DFIs, (genotypic) techniques have for a few days may reduce the risk clinicians should obtain a com- demonstrated that, compared of false-negative cultures. plete blood count, as well as plain to standard (phenotypic) mi- Topical antimicrobial therapy X-rays to look for foreign bodies, crobiology, there are consider- is discussed elsewhere in this tissue gas, or bone abnormalities. ably more microorganisms of compendium (p. 15). Few data Levels of inflammatory markers, many more species (especially support its effectiveness for especially serum C-reactive pro- obligate anaerobes) (47). What mild infections when used alone, tein, erythrocyte sedimentation remains unclear, however, is and most DFIs require systemic rate, and perhaps procalcitonin, whether it is clinically beneficial antibiotic therapy (48). For se- may help in defining the severity to direct antimicrobial therapy vere infections, initial parenteral and monitoring the progress of the against all of these identified or- therapy (usually for a few days, infection (45). Advanced imaging ganisms, many of which are not followed by a switch to oral thera- techniques, especially magnetic classic pathogens. py) is often safest; otherwise, oral

10 | DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS antibiotic agents with good bio- In addition to antimicrobi- OUTCOMES OF DFI availability are sufficient. Issues al therapy, most patients with a TREATMENT related to selecting predomi- DFI require some type of surgical In addition to the involvement nantly oral versus IV antibiotic procedure; these range from bed- of bone in a DFI, factors that ap- therapy are discussed on p. 13. side sharp debridement to more pear to decrease the likelihood of Clinicians should review pa- extensive operative soft-tissue successful treatment include in- tients’ clinical responses to em- and bone resection. Emergent fection with antibiotic-resistant piric therapy and their culture surgery is required for DFI pa- pathogens (especially MRSA, and sensitivity results to deter- tients with complications such P. aeruginosa, and gram-negative mine whether the selected em- as compartment syndrome, nec- bacilli with extended-spectrum piric treatment regimen requires rotizing fasciitis, or gas gangrene, β-lactamases) and the presence adjustment. Several factors help but other surgical procedures of severe PAD or end-stage re- in determining the most appro- are mostly considered urgent nal disease. Patients with these priate definitive antibiotic reg- or elective. Operating surgeons risk factors require especially imen, including the safety, cost, must have a thorough under- careful follow-up to ensure that and availability of various agents. standing of how to drain infec- the infection is responding. For It is best to follow the princi- tions that may involve several of patients whose infection fails ples of antimicrobial steward- the compartments in the foot. In to respond, consider imaging to ship: treat with the narrowest- general, it is best to perform sur- detect previously undisclosed deep-tissue involvement that re- spectrum regimen appropriate, gical drainage of deep soft-tissue quires drainage or resection, and for the shortest duration neces- infection, especially abscesses, sary (49). A key point is that anti- obtaining optimally collected as soon as practical, rather than biotics treat infections, but there specimens for repeat culture. waiting for the infectious pro- is no good evidence that they help Despite the difficulties in diag- cess to “cool off” with medical heal wounds or prevent DFIs. nosing and treating DFIs, with therapy. Thus, although a foot wound proper management, clinicians Because most cases of DFO may take months to heal, anti- can expect to achieve resolution are chronic and accompanied by biotic treatment of 10–14 days of >90% of mild and moderate necrotic bone, surgical resection (until the signs and symptoms soft-tissue infections. Appropriate is usually the preferred treat- of infection resolve) is sufficient treatment can also resolve infec- for most soft-tissue infections. ment approach. Recent studies tions in >75% of DFO cases (often The required therapy duration have demonstrated that, in about with minor bone resection) and for bone infections is less clear, one-third of cases in which the severe infections (usually with but treatment for 4–6 weeks (or surgeon biopsies the presumed surgical debridement) (10). Elim- shorter if all infected bone is re- uninfected bone at the resection inating the clinical manifestations sected) is usually adequate. margin, cultures are positive. In of infection is a key first step in There is no evidence to support these cases, patients probably managing DFIs, but patients with recommending any proposed ad- require further anti-infective these infections also need appro- junctive treatments (e.g., hyper- treatment. Starting on p. 12, we priate wound care, including pres- baric oxygen therapy or negative discuss the issues involved in sure off-loading, wound cleansing pressure wound therapy) spe- deciding when to consider non- and debridement, revasculariza- cifically for treating DFIs. Pro- surgical (antibiotic) treatment tion of ischemic limbs, and opti- duction of biofilms by causative for DFO. Because bone infec- mized glycemic control. The best pathogens appears to contribute tion recurs in about one-third of predictor of the development of to the difficulty in eradicating -in patients, often months after ap- a foot infection in patients with fections and healing wounds, but parently successful treatment, diabetes is a history of previous it is not clear whether any of the clinicians should consider osteo- DFI, so clinicians should carefully currently available antibiofilm myelitis as being only in remis- follow patients who have had such agents are clinically effective, as sion until 1 year after treatment, infections and also teach them and discussed in the sections starting after which the infection can be their caregivers optimal preven- on p. 4 and p. 17. considered fully cured. tion techniques.

DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS | 11 percutaneous bone biopsy, the re- infection from the bone. However, Surgery Versus mission rate was 64%, and bone although removing bone deformi- culture–based antibiotic therapy ties seems to play a role in mini- Antibiotics in DFO was the only variable found to be mizing the risk of recurrence, this The traditional approach to pa- associated with remission in mul- has not yet been demonstrated in tients with diabetes and foot os- tivariate analyses (53). Another a prospective study. teomyelitis has been a surgical study reported an 81% remission Over the past few years, a new one. However, this approach may rate in patients in whom per- surgical concept called “conser- be associated with biomechani- ulcer bone samples were obtained vative surgery” has been estab- cal changes in the foot, significant after bone debridement. The kind lished; this refers to removal of the loss of quality of life, and early of bone debridement used in this necrotic (and much of the infect- mortality. For these reasons, cli- series was quite similar to real ed) bone and soft tissue without nicians now favor a more conser- surgical debridement, but it was amputating any part of the foot. vative approach to DFO. carried out on an outpatient basis This approach has an acceptably The most conservative ap- (52). Therefore, it is not possible low rate of infection recurrence proach is the use of antibiotics to to know the definitive role of anti- (4.6%) (54). Conservative surgery achieve remission. Treating pa- biotics in this series. A remission aims to preserve the soft-tissue tients with DFO exclusively with rate of 82.3% was achieved in a envelope and more distal tissues antibiotics offers the potential to series in which the authors ex- and is successful in treating al- avoid hospitalization and the ex- clusively used empiric antibiotic most half of patients admitted for pense and risk involved with sur- treatment (51). Although these se- DFO. Aesthetically, this approach gical procedures. Furthermore, it ries demonstrated that remission could be an appropriate alterna- may help to avoid the biomechan- might be achieved in about two- tive to amputation for patients. ical disturbances associated with thirds of cases, their designs pre- Theoretically, conservative sur- surgical resection of all or part of cluded them from demonstrating gery could also be a safe alterna- the foot. The available evidence the potential advantages of this tive to amputation that minimiz- regarding treating DFO exclu- approach (i.e., reduced cost and es the risk of ulcer recurrences sively with antibiotics is based on recurrence of ulcers) compared to because it removes small pieces several retrospective series, one surgical treatment. Furthermore, of bone. However, this has not yet prospective noncontrolled series, there is no information available been demonstrated. Any removal and one RCT (50). These series about the timing of treatment or of pieces of bone leads to biome- are highly heterogeneous, with the clinical and radiological signs chanical disturbances. Indeed, a variation in several factors, in- indicating that antibiotic treat- prospective series dealing with cluding the use of completely dif- ment should be stopped and sur- the outcomes of conservative ferent criteria to define remission gery carried out. surgery reported that re-ulcer- or cure of bone infection and the ations at a new site were asso- method of obtaining microbiolog- CONSERVATIVE SURGERY ciated with a plantar location of ical samples; exclusively empiric Surgery for DFO is required in the ulcer during the first episode treatment was used in one series cases involving spreading soft- and with Charcot deformity (54). (51), whereas “bone debridement” tissue infection, the presence of Recurrences after the removal as a part of the medical treatment severe infection, or where antibi- of a metatarsal head could be as was used in others (52). otic therapy alone is likely to be high as 41% after 13 months of Bone culture provides the most ineffective. Surgical treatment follow-up (55). However, nonsur- accurate microbiological infor- could theoretically have some gical treatment of DFO could also mation, and surgical or percuta- advantages. It removes necrot- be associated with a high rate of neous bone biopsy is the optimal ic bone, bacteria, and biofilms. re-ulceration. Recurrences may specimen collection method for Furthermore, for cases in which be as high as 40% within 1 year obtaining a noncontaminated the ulcer is associated with bone after healing of a DFU, 60% at 3 bone sample (50). In a series in deformities or bony prominenc- years, and up to 65% at 5 years which 52% of patients without es, surgery could correct these (56). Recurrences after nonoper- ischemia or gangrene underwent problems while also removing the ative treatment frequently occur

12 | DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS at the same site as the previous offer some conclusions. Surgery sociated with IV catheters, the ulceration, even despite the use of should be primarily elected in evidence supporting the superi- an orthosis, whereas recurrences cases of: ority of IV administration and the after surgery occur at different ⊲ Severe infection, gangrene, belief that antibiotics given this sites due to pressure transfer. spreading soft-tissue infection, way are somehow “stronger” for To clarify the definitive role of or destruction of the soft-tissue a number of infection syndromes surgery and exclusively antibiotic envelope; is limited and under scrutiny (58). treatment when treating DFO, a ⊲ Infections associated with sub- The OVIVA trial (8) was a group of authors designed a ran- stantial bone necrosis or pro- pragmatic, multicenter, parallel- domized comparative trial (7). gressive bone destruction de- group, randomized, open-label The authors compared primari- spite antibiotic treatment; trial to assess the noninferiority ly antibiotic treatment based on ⊲ Osteomyelitis associated with of oral (PO) antibiotics compared deep soft-tissue microbiological deformities or biomechanical to injectable antibiotics in bone samples (90 days) with primarily disturbances that are surgically and joint infections, including conservative surgical treatment correctable; and those of the diabetic foot. It was (plus 10 days of antibiotics). The ⊲ Infecting pathogens that are re- the largest clinical trial to date exclusion criteria for this study sistant to available antibiotics. to assess orthopedic infection were strict: patients with severe Antibiotics should be primarily outcomes based on different an- infections, spreading soft-tissue elected in cases involving: tibiotic administration strategies infections, limb ischemia, ne- ⊲ Noncomplicated forefoot os- (59). A condensed overview of the crosis, or Charcot changes. No teomyelitis and when bone bi- trial and its key findings has been patients with mid- or rearfoot os- opsy techniques are available; published (8). teomyelitis were included in the ⊲ Noncomplicated osteomyelitis Researchers recruited adult pa- study. No differences were found for which a per-ulcer bone bi- tients presenting with orthopedic between the two groups regarding opsy can be easily collected as infections from 26 UK centers healing rates (considered the re- an outpatient procedure; and randomly assigned them to mission definition), healing time, ⊲ A high likelihood of poor post- receive either IV or PO antibiotic and complications requiring a operative biomechanics of the therapy for 6 weeks, with or with- subsequent surgery. foot; out prior surgical debridement. No difference in minor am- ⊲ Patients who are too medically Adult patients were eligible if putations was found between unstable for surgery; and they had a bone or joint infection the two groups (P = 0.336). No ⊲ Development of osteomyelitis that would normally warrant a recurrences were found in the and critical limb ischemia while prolonged course of antibiotics two groups during the follow-up waiting for revascularization. such as native bone and joint os- period. Two re-ulcerations were teomyelitis, including DFO, ver- detected in the antibiotic group tebral osteomyelitis, and pros- (9.5%), and four were found in the thetic joint infections. Notable surgical group (21%) during the IV Versus Oral exclusions were patients who did 12 weeks of follow-up after heal- Antibiotics for not have oral antibiotic options ing (P = 0.670). This study clari- because of bacterial antibiotic fied some previous questions, but Osteomyelitis: resistance and those who had a its small sample size and short Lessons from the concurrent syndrome mandating follow-up period were important OVIVA Trial prolonged IV therapy (e.g., sepsis limitations. It could be concluded or endocarditis). Follow-up was that, in mild and moderate cas- The conventional management for 1 year, with clinical assess- es of neuropathic nonischemic of complex bone and joint infec- ments made at 6 weeks, 3 months, forefoot ulcers complicated by tions usually comprises surgical and 1 year. osteomyelitis, antibiotic treat- debridement, followed by a pro- The choice of antibiotics was ment is as safe as surgery. longed course of IV antibiotics made by an accredited infection Based on the current medical lasting at least 4–6 weeks (57). specialist taking into account literature and expert opinion, we Alongside the inherent risk as- local epidemiology, antibiotic

DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS | 13 TABLE 2 Baseline Surgical Procedures Performed in Patients in the OVIVA Trial (Total N = 1,054) sites (81.1%), with the foot ac- counting for 16.6%. Diabetes was IV Group PO Group Total the most common comorbidity Chronic native osteomyelitis, debrided 153/527 169/527 322 (30.6%) (19.5%). The primary outcome of definite Chronic native osteomyelitis, not debrided 25/527 29/527 54 (5.1%) treatment failure (imputed lost to Debridement with implant retention 124/527 123/527 247 (23.4%) follow-up) occurred in 74 of 506 Removal of infected metalware 89/527 78/527 167 (15.8%) patients (14.6%) in the IV group Prosthetic joint implant removed 68/527 67/527 135 (12.8%) and 67 of 509 patients (13.2%) in Single-stage joint revision 47/527 43/527 90 (8.5%) the PO group, giving an absolute Vertebral osteomyelitis, debrided 8/527 5/527 13 (1.2%) risk difference of –1.5% (95% CI Vertebral osteomyelitis, not debrided 13/527 13/527 26 (2.5%) –5.7 to 2.8), which satisfied the pre- specified noninferiority margin of Adapted from ref. 8. 7.5%. Further analyses including a resistance, drug bioavailability, of randomized therapy, and re- worst-case sensitivity analysis in allergies, drug interactions, and source utilization. The full meth- which missing endpoint data were contraindications. Adjunctive odological details are available in assumed to be treatment failure in oral rifampin as an anti-biofilm the published protocol (60). the PO group and treatment suc- agent and antibiotic-impregnated Of 1,054 randomized patients, cess in the IV group also supported bone cement were each permitted 39 (3.7%) were lost to follow-up noninferiority (Figure 3). in both groups at the discretion of with no available endpoint data. Secondary outcomes for which the infection specialist. Patients in both groups were well the difference between the two The primary outcome was the matched in baseline characteris- groups was statistically signif- rate of definite treatment failure tics such as demographics (me- icant included: 1) early discon- (based on operative findings and dian age 60 years, 64.3% male), tinuation of randomized therapy microbiological and histopatho- surgical procedure performed (18.9% in the IV group vs. 12.8% in logical criteria) as assessed by (Table 2), histology (infected the PO group, P = 0.006); 2) com- an endpoint committee blinded to 51.5%, equivocal 2.8%, uninfect- plications from the IV catheter the randomization strategy. Sec- ed 6%, not performed 39.7%), and (9.4% in the IV group vs. 1.0% in ondary outcomes included possi- microbiology results (Staphylo- the PO group, P <0.001); and, 3) ble and probable treatment failure, coccus spp. 64.8%, polymicrobial median hospital length of stay (14 serious adverse events, IV catheter 18%, culture negative 15.5%). In- days in the IV group vs. 11 days in complications, Clostridium diffi- fections of the lower limb com- the PO group, P <0.001). Second- cile infections, early termination prised the majority of affected ary outcomes for which the dif-

SUBGROUP PO GROUP IV GROUP RISK DIFFERENCE (95% CI) -1.4 NI Margin Intention-to-treat group 70/527 77.3/527 -1.5 Modified intention-to-treat group 67/509 74/506 -2.5 Per-protocol group 61/466 69/443 -2.1 Worst-case sensitivity analysis 85/527 74/527

-7.5 -5.0 -2.5 0.02.5 5.07.5 Favors PO Favors IV

FIGURE 3 Risk difference in failure rates in the OVIVA trial, according to analysis performed. Failure rates are expressed as number of patients with treatment failure/total number of patients. NI, noninferiority. Adapted from ref. 8.

14 | DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS ference between the two groups in the IV group vs. 71 days in the of chronic diabetic foot wounds. was not statistically significant PO group). These conditions include poor included: 1) probable or possible Notable points in this study are arterial perfusion, impaired im- treatment failure (1.2% in the IV that it is representative of real- mune cell functions, and neu- group vs. 2.0% in the PO group); world circumstances and prag- ropathies. They all result in part 2) occurrence of at least one seri- matic but necessarily open-label, from chronically elevated blood ous adverse event (27.7% in the IV given the logistics and risks of glucose levels that lead to non- group vs. 26.2% in the PO group); administering matched placebos. enzymatic glycation of multiple and, 3) C. difficile infection (1.7% The vast majority of patients in proteins and activation of ad- in the IV group vs. 1.0% in the PO this study had preceding surgery vanced glycation end products/ group). in the form of prosthesis removal receptor for advanced glycation Predefined and post-hoc sub- or debridement, highlighting the end products pathways that reg- group analyses did not demon- basis of effective osteomyelitis ulate functions of inflammatory strate an advantage toward: 1) the management. The antibiotics se- cells, neuronal cells, and wound type of surgery performed (i.e., lected were specifically tailored cells (fibroblasts, epithelial cells, osteomyelitis debrided or not de- to be the most appropriate for and vascular endothelial cells) brided) (P = 0.26), retention of each patient, but the resultant (61). Obviously, good clinical metalware (P = 0.13), or antibiotic heterogeneity hindered subgroup management of these comorbidi- cement used (P = 0.98); 2) infect- analysis. ties is essential to reduce the risk ing pathogen isolated (P = 0.30); The OVIVA trial provided evi- of an acute wound converting into or, 3) presence of peripheral vas- dence that challenges the widely a stalled, chronic wound. cular disease (P = 0.47), which held belief that the treatment of High bioburden, which is a may directly affect the delivery osteomyelitis requires IV antibi- rather poorly defined term (i.e., and concentration of antibiotic at otics. If oral regimens are appro- ≥105 cfu/g of tissue), is associated distal sites of infection. priately selected, they can be as ef- with the failure of an acute The most frequently used IV fective, more convenient, and less wound to heal in patients with or antibiotics were glycopep- costly (with a conservative non- without diabetes and is, in most tides (41.1%) and cephalosporins surgical treatment per-patient wounds, a combination of both (33.2%), reflecting their con- savings estimated at £2,740 GBP). planktonic and biofilm bacteria, venient once-daily dosing and More importantly, the use of oral as well as fungal species. How- coverage of the predominant antibiotics negates the signifi- ever, in the past 10 years, animal staphylococcal infections. The cantly increased IV catheter- model studies of wound healing quinolones (36.5%) and combi- related adverse events observed have demonstrated that forma- nation therapy (16.6%) compris- in this study. Certainly, oral anti- tion of bacterial biofilms signifi- ing ciprofloxacin and either clin- biotics are not necessarily suit- cantly delays healing (62). This damycin or doxycycline were the able for all cases of osteomyelitis; evidence has led to an increased most common oral antibiotics but without a doubt, not all cases focus on the role of bacterial bio- prescribed, reflecting their high of osteomyelitis mandate the use films in impairing skin wound oral bioavailability and bone pen- of IV antibiotics. healing, including in patients etration profile. Adjunctive oral with diabetes. rifampin for at least 6 weeks was Clinical studies have now es- administered in the PO group tablished that bacteria in the more frequently than in the IV Topical Treatments biofilm phenotype are present group (22.9 vs. 31.4%) but did for DFIs in a high percentage (probably not significantly affect outcome. >80%) of chronic skin wounds Oral follow-on therapy beyond As elucidated throughout this (63,64). The ability of bacteria the 6-week study period was compendium, multiple comor- in biofilm communities to sur- observed in the vast majority of bidities can contribute to the fail- vive under conditions that nor- patients (76.7%), but the median ure of acute wounds in the legs mally kill planktonic bacteria total duration of antibiotics did or feet of people with diabetes to very effectively is explained by not differ significantly (78 days heal, leading to the development a combination of several factors

DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS | 15 (65). These include the difficul- aggressive debridement of bio- evidence to support the use of ty of phagocytic inflammatory films and topical treatments that topical antimicrobials to con- cells (neutrophils and macro- have been shown in laboratory trol (either eradicate or prevent) phages) to engulf and kill large or clinical studies to be effective DFIs. However, several new top- masses of biofilm bacteria that at killing residual biofilm bac- ical treatments do appear to sig- are tightly attached to extra- teria. As the bioburden level of nificantly reduce infection by cellular matrix, bone cortex, or biofilm bacteria is reduced, the both planktonic and biofilm bac- innate surfaces such as metal- level of inflammation (neutro- teria based on laboratory and an- lic orthopedic implants. Also, phils and macrophages) and the imal wound-healing studies, and the dense exopolymeric matrix elevated levels of proteases and a few pilot clinical studies have of many bacterial biofilms has a reactive oxygen species will also reported some improved healing. high negative charge density be- be reduced, which will allow the Unfortunately, there are no pub- cause of acidic polysaccharides chronic wound to move out of a lished, randomized, appropriate- (polyalginic acid in P. aerugino- chronic inflammatory phase into ly controlled, multicenter clin- sa biofilms) and free bacterial an active healing (repair) phase. ical studies that provide Level 1 DNA that can limit diffusion of Topical treatments can then evidence that any of these new positively charged antimicrobi- “step down” to less frequent and topical treatments significantly als such as silver ions. In addi- aggressive debridement com- reduce wound bacterial biobur- tion, individual bacteria locat- bined with standard antimicro- den, including mature, toler- ed deep inside mature biofilms bial dressings that can effectively ant biofilm bacteria, or improve frequently become metabolically kill planktonic bacteria and pre- healing of chronic wounds. dormant, which provides toler- vent reformation of biofilm com- Localizing Bacterial Biofilms ance to antibiotics that typically munities in the wound bed. Fi- and Assessing Successful only kill metabolically active bac- nally, when the DFU wound bed Debridement teria by interfering with essen- has been adequately prepared, Two of the biggest challenges tial bacterial enzyme and protein topical treatments can “step up” clinicians face in implementing systems. This combination of to advanced wound treatments BBWC are knowing where bio- factors contributes to bacterial such as amnion/chorion dress- films are located on a chronic biofilms being a major common ings, growth factors, and skin wound bed and assessing wheth- cause of persistent infections in grafts that will effectively stimu- er the biofilms have been re- skin wounds and multiple other late healing because the proteins moved by debridement. A recent clinical conditions (66,67). that comprise these advanced initial clinical study reported Building on this base of lab- wound treatments and their re- that bacterial biofilms were de- oratory and clinical data about ceptors on wound cells will sur- tected and localized on chronic biofilms and chronic skin vive and function normally to wound beds using a simple and wounds, an international panel of promote healing. rapid membrane blotting tech- wound care clinicians and basic nique followed by brief staining scientists produced consensus NEW APPROACHES TO of the membrane with a cation- guidelines for identification and TOPICAL TREATMENT ic colored dye that bound to the treatment of biofilms in chronic Building on the principles and negatively charged components nonhealing wounds, including concepts of BBWC to control in- of the biofilm exopolymeric ma- DFUs (15). A key take-home fection and inflammation, it is trix (69). Importantly, if no bio- message from the guidelines is important to know whether new film staining was detected after that treatment of chronic DFUs topical approaches can help con- debridement, there was a signif- should be based on the princi- trol infection and inflammation icant reduction in generation of ples of biofilm-based wound care and what evidence may support wound slough and a reduction (BBWC) that emphasize a “step- their use. A 2019 IWGDF sys- in wound area in the following down-then-step-up” approach. tematic review (49) and a recent week compared to wounds that This approach involves starting Cochrane systematic review (68) had residual biofilm staining on treatment with a combination of found no compelling published the membrane.

16 | DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS Another new technology Concentrated Nonionic (63). Additionally, acute wounds (MolecuLight i:X) that has been Surfactant Wound Gel made experimentally, inoculated reported to localize bacteria in Bacterial biofilms tend to be with biofilm-forming bacteria, wounds and on surrounding tightly attached (sessile) to ex- and made to induce biofilms have skin uses blue light to stimulate tracellular matrix components at demonstrated delayed healing. fluorescence of fluorochrome the surface of wound beds and in Biofilms develop quickly, and molecules synthesized by some the superficial layers under the wound-isolated P. aeruginosa strains of common wound patho- surface of the wound bed, which in vitro displays characteristics gens. This technology does not is one reason that simply wiping of mature biofilm within 10 appear to distinguish between with a gauze pad does not effec- hours (77). As noted previously planktonic and biofilm pheno- tively remove all biofilm bacteria (p. 4–5), biofilms cannot be de- types of bacteria but may be (74). A recent laboratory study tected using routine clinical useful in localizing bacteria in showed that daily application of techniques. Chronic wounds lack wounds to guide and assess de- a concentrated nonionic surfac- overt clinical signs of infection, bridement of bacteria (70). tant gel containing a preserva- making lack of clinical suspicion tive (Plurogel) with gauze wiping a confounding problem in wound Negative Pressure Wound eliminated mature biofilms of biofilm identification. Therapy Plus Instillation of P. aeruginosa and S. aureus grown Antimicrobial Solutions on pig skin explants (75). TREATMENT MODALITIES Negative pressure wound therapy In summary, new topical treat- High-quality studies are lacking (NPWT) has become a standard ments have shown positive re- and thus represent a needed op- treatment for many types of acute sults in significantly reducing portunity with regard to elucidat- and chronic wounds. This ther- levels of planktonic and biofilm ing the beneficial role of biofilm apy primarily stimulates forma- bacteria using laboratory models eradication and its relationship to tion of new granulation tissue by of infected skin wounds. Results healing in patients with chronic exerting a combination of macro- from clinical studies are needed wounds. Studies to date have pri- and microdistortion forces on to more completely understand marily focused on approaches that wound cells that alter patterns the effects of these treatments in treat or prevent biofilms using of gene expression and by induc- actual clinical use. in vitro or in vivo models or have ing localized regions of ischemia involved patients using purport- in the wound bed. In general, ed anti-biofilm approaches (i.e., NPWT alone does not dramati- Role of Modern BBWC) that result in improved cally reduce planktonic or biofilm Technology in the healing. However, these latter bacteria in infected wounds (71). studies often have not: 1) studied However, laboratory studies have Prevention and a single chronic wound type; 2) shown dramatic reductions (>4 Management of demonstrated the presence of bio- logarithms) in planktonic and Biofilms and DFIs films before treatment; 3) demon- biofilm bacteria levels on pig skin strated the eradication of biofilms explants with mature biofilms Although biofilms are more com- originally present; 4) confirmed when NPWT was combined with mon in chronic wounds than in eventual healing status; or, 5) in- instillation of various antimi- acute wounds, one of the first as- cluded randomized, controlled crobial solutions (72), and these sociations of biofilms and wounds groups. Despite these significant findings were confirmed in a -pi followed the electron microscop- limitations, reducing, removing, lot clinical study (73). Well con- ic examination of sutures and or preventing biofilms remains a trolled clinical studies are need- staples removed from healed sur- logical approach to help clinicians ed to rigorously assess whether gical wounds (76). However, the heal chronic wounds. Table 3 lists NPWT plus instillation reduces fact that chronic wounds often commercially available products levels of planktonic and biofilm harbor biofilms is part of the ra- that have demonstrated anti- bacteria in DFUs and improves tionale linking their presence to biofilm activity in vitro or with progression toward healing. delayed healing of such wounds lower levels of evidence.

DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS | 17 TABLE 3 Commercially Available Products with Some Level of Evidence Supporting Anti-Biofilm Activity

PRODUCT/COMPONENTS MECHANISM OF ACTION Polyhexamethylene biguanide (PHMB), ⊲ The antimicrobial activity of PHMB owes itself to its molecular shape, charge, and including: delivery system. It causes damage to cytoplasm membrane and prevents the efflux • Prontosan (B Braun) pump, so microbes are unable to reduce the levels of antimicrobial present. • PHMB and the surfactant betaine • Suprasorb X + PHMB (Lohmann & Rauscher) • PuraPly AM (Organogenesis) • HydroClean (Hartmann) + hydro- responsive wound dressing + PHMB Plurogel (Medline) ⊲ This concentrated surfactant gel creates a moist healing environment that promotes autolytic debridement. It includes hydrophilic and hydrophobic components, the mi- celles of which link to form a micelle matrix. The hydrophilic surface of the micelles softens and loosens wound debris, which is then trapped by the hydrophobic inner core BlastX (NextScience) (benzalkonium ⊲ BlastX dissolves the extrapolysaccharide matrix, thereby exposing 80–90% of other- chloride 0.13%, polyethylene glycol, wise encapsulated bacteria and allowing their removal. sodium citrate, citric acid, and water) Nanoparticles (NPs) ⊲ Silver NPs capitalize on the studied antimicrobial and anti-biofilm properties of silver. • Silver (Ag) ⊲ The antimicrobial and anti-biofilm activity of ZnO NPs could be caused by release of • Zinc oxide (ZnO) reactive oxygen species. • Polyethylenimine/diazeniumdiolate- ⊲ Nitric oxide (NO) has shown promise in dispersing biofilm; however, it needs to be doped poly(lactic-co-glycolic acid) supplied to the biofilm in a sustainable manner. PLGA-PEI/NO NPs aim to achieve this (PLGA-PEI/NO) by supplying NO to biofilm over an extended time period. Lasers/photodynamic therapy ⊲ Blue laser light inhibits biofilm formation in vitro and in vivo by inducing oxidative stress ⊲ A “smart” material was developed that incorporated gold nanorods and an adsorbed protease (protease-conjugated gold nanorods [PGs]). When illuminated with near-in- frared light, PGs physically damage bacteria, prevent biofilm and exotoxin produc- tion, eliminate preexisting biofilm and exotoxin, and inhibit bacterial quorum-sensing systems. Hypochlorous acid ⊲ Use of this acid results in chemical inactivation of various cellular processes via oxidation of sulfhydryl enzymes and amino acids; chlorination of amino acids; and inhibition of protein synthesis. Monofilament fiber debridement ⊲ Millions of polyester fibers loosen and bind debris. This product is available in pad technology (MFDT; Debrisoft, Lohmann and lolly forms. & Rauscher) Dispersin B wound spray (Kane ⊲ Enzyme isolated from Aggregatibacter actinomycetemcomitans breaks down po- Biotech) ly-N-acetylglucoasmine, a major polysaccharide constituent of biofilms. Microlyte Ag Ga (Imbed Biosciences) ⊲ This technology interferes with iron metabolism. Gallium ions compete with iron ions, resulting in inhibition of various bacterial cellular processes, including growth and biofilm production. This interaction also results in increased sensitivity to silver ions. Cadexomer iodine (IODOSORB; Smith ⊲ This spherical starch bead lattice containing 0.9% iodine as a high absorptive & Nephew) capacity. Iodine is slowly released as it encounters wound fluid. Iodine has a direct bactericidal effect. IODOSORB destroys biofilms, collapses glycocalyces, and traps bacteria within beads.

18 | DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS APPROACHES TO BIOFILM film removal in vivo (82). Other anti-biofilm agents include TREATMENT AND Antibiotics and Antiseptics ethylenediaminetetraacetic acid, PREVENTION Biofilms evade the host’s immune gallium, acetyl salicylic acid, and Debridement system and may be up to 1,500 others. Physical removal of biofilm bacte- times more resistant to antibiot- Novel anti-biofilm agents cur- rently under development in- ria is the standard means of man- ics than planktonic cells. Systemic clude anti-adhesion molecules, aging biofilms, and debridement antibiotics are not part of routine quorum-sensing inhibitors, and is considered essential for effec- standard care for clinically nonin- selectively targeted antimicrobial tive biofilm control, most com- fected DFUs, and biofilm bacteria peptides (STAMPs) (86). Given monly with sharp debridement are more resistant to topical anti- that adhesion is the first stage in to remove and suppress biofilm biotics than planktonic bacteria. biofilm formation, anti-adhesion (78). However, biofilms are often There are multiple mechanisms molecules may keep bacteria in deep in wound tissue, and remov- for biofilm antibiotic resistance: the planktonic state, making them al by more conservative surgical biofilms may contain a subpopu- more susceptible to the host im- debridement techniques such as lation of specialized survivor cells, mune system and antibiotics. Sor- by curetting may be incomplete. the drug target may be modified or tase, from gram-positive bacteria, Other debridement techniques unexpressed, biofilms are less sus- is shared by most of the surface such as plasma-mediated bipo- ceptible because they contain few- proteins and is an anti-adhesion lar radiofrequency ablation was er growing bacteria, or the antimi- candidate because it covalently found to be better at least in one crobial agent may not adequately anchors surface proteins to pep- model (79). Even with removal, penetrate the biofilm. tidoglycan. Quorum sensing is biofilms rapidly recover from Anti-Biofilm Agents crucial to the maturation of bio- mechanical disruption to reform With increased knowledge of bio- films; thus, using quorum-sensing within 24 hours, implying that film formation, anti-biofilm agents inhibitors theoretically can main- debridement may best be used that disrupt biofilm community tain bacteria in a planktonic state in combination with other treat- functions and defenses may allow by affecting cell-to-cell commu- ments (80). other concomitant therapies and nication, making bacteria more Laser and Ultrasound natural host mechanisms to work susceptible to the immune sys- Laser treatment, particularly low- more effectively to promote heal- tem and antibiotics. Two leading level laser therapy (LLLT), is ing. As an example, lactoferrin, a candidates for quorum-sensing thought to accelerate wound heal- component of tears, mucus, and inhibitors include the furanones ing by increasing the proliferation human milk, prevents biofilm for- and ribonucleic acid III inhibiting of cells involved in wound heal- mation by altering bacterial mo- peptide. Species-specific control ing and the synthesis of collagen, tility and decreasing bacterial sur- of biofilms may be achieved with while also decreasing the inflam- face attachment. Lactoferrin use STAMPs, which are an essential matory response. Laser therapy in the disruption of Pseudomonas part of the innate immune sys- has also been shown to cause dis- biofilms has been described and tem that contain species-specific aggregation of microorganisms shown to act synergistically with binding domains. (81). Results from LLLT studies xylitol, which impairs matrix de- Biofilm Extracellular Enzymes have shown significant bacteri- velopment (83). Studies regarding Other novel strategies include cidal potential without damaging BBWC for wound treatment in pa- matrix-degrading enzymes, which tissue. LLLT can reduce both cell tients with critical limb ischemia degrade the protective polysac- viability and biofilm growth. have demonstrated a statistically charide layer and make cells more In addition to lasers, ultrasound significant increase in wound heal- susceptible to antibiotics (78). therapy has been used to investi- ing with BBWC (84,85). In these The final stage of the biofilm life gate oral biofilms, as well as cu- examples, BBWC included stan- cycle, dispersion, allows bacteria taneous wound healing. Noncon- dard care plus lactoferrin and xyli- to release from the biofilm and tact ultrasonic waves can reduce tol compounded in a methylcellu- colonize other areas. However, biofilm in vitro and assist in bio- lose gel as the anti-biofilm agents. dispersion results in an increased

DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS | 19 number of easier-to-eradicate charide and infect biofilm cells, is an important goal. Last year, planktonic bacteria and thus is when experimentally tracing the the U.S. National Institute of a therapeutic topic of interest. interaction of bacteriophage with Diabetes and Digestive and Kid- To facilitate dispersion, bacteria bacterial biofilms using fluores- ney Diseases highlighted this produce extracellular enzymes cent and chromogenic probes. need by creating a consortium such as glycosidases, proteas- dedicated to advancing this na- es, and deoxyribonucleases that scent field (87). We believe it is degrade the adhesive compo- highly likely that numerous fu- nents in the biofilm matrix. Two Conclusion ture therapeutics will come bun- common wound pathogens, P. The worldwide burden of diabet- dled with a companion diagnostic aeruginosa and S. aureus, pro- ic foot complications—particu- to help target when or whether it duce enzymes that can degrade larly DFUs—has been growing is likely to be most effective. the biofilm matrix. As an exam- (1,56). In this compendium, we Measuring what we manage ple, P. aeruginosa-produced al- have reviewed a number of prom- is of course crucially important. ginate lyase increases biofilm ising strategies to heal wounds, However, globally accepted defi- cell detachment and antibiotic in addition to some essentials of nitions of risk are arguably even effectiveness (86). The enzyme care such as high-quality surgi- more central to high-quality care. dispersin B is able to degrade the cal wound debridement and off- In the past year, the IWGDF and polysaccharide matrix of staphy- loading of pressure (56). We have the Global Vascular Guidelines lococci and is being developed as also included treatments that have Writing Group both published a wound care gel. Other matrix- received increased support in the important efforts to define risk degrading enzymes include literature such as cultured tissue and better guide care worldwide deoxyribonuclease, lysostaphin, products, topical oxygen thera- (10,88). Both groups have made proteinase K, and trypsin. py, sucrose octasulfate dressings, efforts to refine the terminology Vaccination and autologous platelet-rich fibrin for PAD. Instead of the previous New treatment strategies are dressings (2–6). term “critical limb ischemia,” or being explored, including the Importantly, major strides CLI, the preferred term is now potential of vaccination against have occurred in the manage- “chronic limb-threatening isch- pathogens. In the area of oral ment of wound infection during emia,” or CLTI (88). This new biofilms, numerous studies have the past decade, and these are term better describes the con- documented effective vaccina- also summarized herein. Where- dition and marries well with tion against oral pathogens (78). as clinically noninfected wounds the concept of chronic DFUs. In Often, biofilms are polymicro- do not require antibiotic therapy, addition, a more universally ac- bial, and some bacteria, such we have explained that infected cepted and validated definition of as Fusobacterium nucleatum in wounds should be cultured, tak- limb threats (“wound, ischemia, the mouth, bridge other species ing tissue from the wound base and foot infection,” or WIfI) is within the biofilm community. after debridement, and that anti- becoming the new norm in inter- Finally, there is great interest biotic therapy should be targeted disciplinary teams (88,89). WIfI in developing passive immuno- appropriately. Undoubtedly, the classification allows clinicians to therapy or vaccines for bacterial results of the OVIVA trial will measure and communicate the infections such as staphylococci. engender much debate in the multifactorial and dynamic na- Vaccine targeting can disrupt bio- near future about the need for IV ture of diabetic foot disease with film attachment and growth. Bac- antibiotics in osteomyelitis. an easily understood scoring sys- teriophages (viruses that affect Although we have made ad- tem. This is similar in nature to bacteria) can be used effectively vances in approaches to tissue the use of “tumor, node, metasta- in managing infection, mainly repair and wound healing, our sis,” or TNM, terminology in the because of their bactericidal ac- knowledge of the constantly field of oncology. tivity. Studies involving the in- changing wound milieu remains Although healing wounds re- teraction of bacteriophages and elusive. Indeed, the development mains a key goal, one could argue biofilms have shown that phages of companion diagnostics and that it is a bit too narrow a fo- can degrade biofilm exopolysac- biomarkers to help direct therapy cus. Indeed, after a wound heals,

20 | DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS 40% of people with diabetes will the Prevention and Management of chronic diabetic foot ulcers: the TWO2 re-ulcerate within 1 year, and Biofilms and DFIs.” A.J.M.B. and D.G.A. study. Diabetes Care. Epub ahead of nearly two-thirds will do so by 3 are the guarantors of this work. print on 16 October 2019 (doi: 10.2337/ years (56). Thus, recurrence is dc19-0476) DUALITIES OF INTEREST not only common, it is likely. The A.J.M.B., D.G.A., M.J.H., J.M.E., C.E.A., 6. Edmonds M, Lázaro-Martínez foot in diabetes, once healed, is in B.A.L., J.A.-S., G.S., and R.S.K. have JL, Alfayate-García JM, et al. remission. no relevant dualities of interest to Sucrose octasulfate dressing Our goals, as mentioned in the disclose. M.M. has received research versus control dressing in patients first compendium (1), must focus grants and consultant fees for with neuroischaemic diabetic foot on innovation in methods to max- educational activities from Smith ulcers (Explorer): an international, imize ulcer-free, hospitalization- & Nephew in the area of wound multicentre, double-blind, free, and activity-rich days. As we biofilm. The work of H.K.L. was randomised, controlled trial. Lancet have shown, much progress has funded by the National Institute for Diabetes Endocrinol 2018;6:186–196 been made toward this end, but Health Research—Health Training 7. Lázaro-Martínez JL, Aragón- there is much work yet to do. We Assessment. Sánchez J, García-Morales E. hope this volume will help readers Antibiotics versus conservative with the early detection, identifi- REFERENCES surgery for treating diabetic foot cation, and treatment of infection 1. Boulton AJM, Armstrong DG, Kirsner osteomyelitis: a randomized in diabetic feet, for it is infection RS, et al. Diagnosis and Management comparative trial. Diabetes Care in neuroischemic ulcers that often of Diabetic Foot Complications. 2014;37:789–795 leads to amputation. American Diabetes Association, 8. Li HK, Rombach I, Zambellas R, et ACKNOWLEDGMENTS Arlington, VA, 2018 al.; OVIVA Trial Collaborators. Oral The authors thank Dr. Elly Trepman for 2. Edmonds M. A renaissance in versus intravenous antibiotics for review and helpful suggestions on the diabetic foot care: new evidence- bone and joint infection. N Engl J Med section concerning the microbiology based treatments. Lancet Diabetes 2019;380:425–436 of DFIs. Endocrinol 2018;6:837–838 9. Brem H, Stojadinovic O, Diegelmann Editorial and project management 3. Game F, Jeffcoate W, Tarnow RF, et al. Molecular markers in services were provided by Debbie L, et al.; LeucoPatch II Trial patients with chronic wounds to Kendall of Kendall Editorial in Team. LeucoPatch system for the guide surgical debridement. Mol Med Richmond, VA. management of hard-to-heal diabetic 2007;13:30–39 foot ulcers in the UK, Denmark, 10. Lipsky BA, Senneville E, Abbas AUTHOR CONTRIBUTIONS and Sweden: an observer-masked, A.J.M.B. and D.G.A. served as co- ZG, et al. IWGDF guideline on the randomised controlled trial. Lancet editors and, as such, co-wrote the diagnosis and treatment of foot Diabetes Endocrinol 2018;6:870–878 infection in persons with diabetes. introduction and conclusion and Diabet Metab Res Rev. In press. Also reviewed and edited the entire 4. Niederauer MQ, Michalek JE, Liu available from iwgdfguidelines.org/ manuscript. M.J.H. wrote “How Q, Papas KK, Lavery LA, Armstrong wp-content/uploads/2019/05/05- Infection Impairs Wound Healing.” DG. Continuous diffusion of oxygen IWGDF-infection-guideline-2019.pdf. M.M. wrote “Biofilms in the Context improves diabetic foot ulcer healing Accessed 5 November 2019 of DFUs.” J.M.E. wrote “Evolving when compared with a placebo Microbiology of DFIs.” C.E.A. wrote control: a randomised, double-blind, 11. Lipsky BA, Berendt AR, Cornia PB, “Debridement: The First Step in multicentre study. J Wound Care et al.; Infectious Diseases Society of Controlling DFIs.” B.A.L. wrote 2018;27(Suppl. 9):S30–S45 America. 2012 Infectious Diseases “Managing Infected DFUs.” J. A.-S. 5. Frykberg RG, Franks PJ, Edmonds Society of America clinical practice wrote “Surgery Versus Antibiotics in M, et al., on behalf of the TWO2 Study guideline for the diagnosis and DFO.” H.K.L. wrote “IV Versus Oral Group. A multinational, multicenter, treatment of diabetic foot infections. Antibiotics for Osteomyelitis: Lessons randomized, double-blinded, placebo- Clin Infect Dis 2012;54:e132–e173 from the OVIVA Trial.” G.S. wrote controlled trial to evaluate the efficacy 12. Prompers L, Huijberts M, Apelqvist “Topical Treatments for DFIs.” R.S.K. of cyclical topical wound oxygen J, et al. High prevalence of ischaemia, wrote “Role of Modern Technology in therapy (TWO2) in the treatment of infection and serious comorbidity in

DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS | 21 patients with diabetic foot disease from molecular and microscopic Staphylococcus aureus in diabetic in Europe: baseline results from approaches. Clin Microbiol Infect foot infection in India: a growing the Eurodiale study. Diabetologia 2019;25:332–339 menace. Int J Low Extrem Wounds 2007;50:18–25 21. Lipsky BA, Berendt AR, Deery HG, 2019;18:236–246 13. Kalan LR, Brennan MB. The role et al.; Infectious Diseases Society of 29. Wang SH, Sun ZL, Guo YJ, et al. of the microbiome in nonhealing America. Diagnosis and treatment of Meticillin-resistant Staphylococcus diabetic wounds. Ann N Y Acad Sci diabetic foot infections. Clin Infect Dis aureus isolated from foot ulcers in 2019;1435:79–92 2004;39:885–910 diabetic patients in a Chinese care 14. Williams H, Campbell L, Crompton 22. Henig O, Pogue JM, Cha R, et al. hospital: risk factors for infection RA, et al. Microbial host interactions Epidemiology of diabetic foot infection and prevalence. J Med Microbiol and impaired wound healing in mice in the metro-Detroit area with a 2010;59:1219–1224 and humans: defining a role for focus on independent predictors for 30. Li X, Qi X, Yuan G, et al. BD14 and NOD2. J Invest Dermatol pathogens resistant to recommended Microbiological profile and clinical 2018;138:2264–2274 empiric antimicrobial therapy. Open characteristics of diabetic foot 15. Schultz G, Bjarnsholt T, James Forum Infect Dis 2018;5:ofy245 infection in northern China: a GA, et al.; Global Wound Biofilm 23. Tascini C, Lipsky BA, Iacopi E, et al. retrospective multicentre survey in Expert Panel. Consensus guidelines KPC-producing Klebsiella pneumoniae the Beijing area. J Med Microbiol for the identification and treatment rectal colonization is a risk factor for 2018;67:160–168 of biofilms in chronic nonhealing mortality in patients with diabetic 31. Ramakant P, Verma AK, Misra R, et wounds. Wound Repair Regen foot infections. Clin Microbiol Infect al. Changing microbiological profile 2017;25:744–757 2015;21:790.e1–e3 of pathogenic bacteria in diabetic 16. Ruffin M, Brochiero E. Repair 24. Hassan MA, Tamer TM, Rageh foot infections: time for a rethink on process impairment by Pseudomonas AA, Abou-Zeid AM, Abd El-Zaher EHF, which empirical therapy to choose? aeruginosa in epithelial tissues: major Kenawy ER. Insight into multidrug- Diabetologia 2011;54:58–64 features and potential therapeutic resistant microorganisms from microbial 32. Zubair M, Ahmad J. Potential risk avenues. Front Cell Infect Microbiol infected diabetic foot ulcers. Diabetes factors and outcomes of infection with 2019;9:182 Metab Syndr 2019;13:1261–1270 multidrug resistance among diabetic 17. Kalan LR, Meisel JS, Loesche MA, et 25. Tentolouris N, Jude EB, Smirnof I, patients having ulcers: 7 years study. al.; Strain- and species-level variation Knowles EA, Boulton AJ. Methicillin- Diabetes Metab Syndr 2019;13:414– in the microbiome of diabetic wounds resistant Staphylococcus aureus: an 418 is associated with clinical outcomes increasing problem in a diabetic foot 33. Miyan Z, Fawwad A, Sabir R, Basit and therapeutic efficacy. Cell Host clinic. Diabet Med 1999;16:767–771 A. Microbiological pattern of diabetic Microbe 2019;25:641–655.e5 26. Dang CN, Prasad YD, Boulton foot infections at a tertiary care center 18. Høiby N, Bjarnsholt T, Moser C, AJ, Jude EB. Methicillin-resistant in a developing country. J Pak Med et al.; ESCMID Study Group for Biofilms Staphylococcus aureus in the diabetic Assoc 2017;67:665–669 and Consulting External Expert Werner foot clinic: a worsening problem. 34. Wu M, Pan H, Leng W, Lei X, Zimmerli. ESCMID guideline for the Diabet Med 2003;20:159–161 Chen L, Liang Z. Distribution of diagnosis and treatment of biofilm 27. King MD, Humphrey BJ, Wang YF, microbes and drug susceptibility in infections 2014. Clin Microbiol Infect Kourbatova EV, Ray SM, Blumberg HM. patients with diabetic foot infections 2015;21(Suppl. 1):S1–S25 Emergence of community-acquired in Southwest China. J Diabetes Res 19. Johani K, Malone M, Jensen S, et methicillin-resistant Staphylococcus 2018;2018:9817308 al. Microscopy visualisation confirms aureus USA 300 clone as the 35. Spichler A, Hurwitz BL, Armstrong multi-species biofilms are ubiquitous predominant cause of skin and soft- DG, Lipsky BA. Microbiology of in diabetic foot ulcers. Int Wound J tissue infections. Ann Intern Med diabetic foot infections: from Louis 2017;14:1160–1169 2006;144:309–317 Pasteur to ‘crime scene investigation.’ 20. Johani K, Fritz BG, Bjarnsholt T, et 28. Viswanathan V, Pendsey S, BMC Med 2015;13:2 al. Understanding the microbiome of Radhakrishnan C, Rege TD, Ahdal 36. Leung HB, Ho YC, Wong WC, diabetic foot osteomyelitis: insights J, Jain R. Methicillin-resistant Guerin J. Seasonal variations in

22 | DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS non-traumatic major lower limb in the foot in diabetes: a systematic et al. Does osteomyelitis in the feet amputation in Hong Kong Chinese review. Diabetes Metab Res Rev 2019. of patients with diabetes really recur diabetic patients. Hong Kong Med J In press after surgical treatment? Natural 2007;13:379–381 46. Lauri C, Tamminga M, Glaudemans history of a surgical series. Diabet 37. Uçkay I, Aragón-Sánchez J, Lew AWJM, et al. Detection of osteomyelitis Med 2012;29:813–818 D, Lipsky BA. Diabetic foot infections: in the diabetic foot by imaging 55. Molines-Barroso RJ, Lázaro- what have we learned in the last 30 techniques: a systematic review and Martínez JL, Aragón-Sánchez J, years? Int J Infect Dis 2015;40:81–91 meta-analysis comparing MRI, white García-Morales E, Beneit-Montesinos 38. Uçkay I, Gariani K, Pataky Z, blood cell scintography, and FDG-PET. JV, Álvaro-Afonso FJ. Analysis of Lipsky BA. Diabetic foot infections: Diabetes Care 2017;40:1111–1120 transfer lesions in patients who state-of-the-art. Diabetes Obes Metab 47. Pereira SG, Moura J, Carvalho E, underwent surgery for diabetic foot 2014;16:305–316 Empadinhas N. Microbiota of chronic ulcers located on the plantar aspect diabetic wounds: ecology, impact, of the metatarsal heads. Diabet Med 39. Mutonga DM, Mureithi MW, Ngugi 2013;30:973–976 NN, Otieno FCF. Bacterial isolation and potential for innovative treatment and antibiotic susceptibility from strategies. Front Microbiol 2017;8:1791 56. Armstrong DG, Boulton AJM, Bus diabetic foot ulcers in Kenya using 48. Peters EJG, Senneville E, Abbas ZG, SA. Diabetic foot ulcers and their microbiological tests and comparison et al. Interventions in the management recurrence. N Engl J Med 2017;376: with RT-PCR in detection of S. of infection in the foot in diabetes: a 2367–2375 aureus and MRSA. BMC Res Notes systematic review (update). Diabetes 57. Lew DP, Waldvogel FA. 2019;12:244 Metab Res Rev. In press Osteomyelitis. Lancet 2004;364:3 40. Belefquih B, Frikh M, Benlahlou 49. Uçkay I, Berli M, Sendi P, 69–379 Y, et al. Diabetic foot infection in Lipsky BA. Principles and practice 58. Li HK, Agweyu A, English M, Bejon Morocco: microbiological profile. of antibiotic stewardship in the P. An unsupported preference for Wounds 2016;28:89–98 management of diabetic foot intravenous antibiotics. PLoS Med 41. Perim MC, Borges Jda C, Celeste infections. Curr Opin Infect Dis 2015;12:e1001825 2019;32:95–101 SR, et al. Aerobic bacterial profile and 59. Conterno LO, Turchi MD. antibiotic resistance in patients with 50. Aragón-Sánchez J, Lipsky BA. Antibiotics for treating chronic diabetic foot infections. Rev Soc Bras Modern management of diabetic foot osteomyelitis in adults. Cochrane Med Trop 2015;48:546–554 osteomyelitis: the when, how and why Database Syst Rev 2013;9:CD004439 of conservative approaches. Expert 42. Jaju K, Pichare A, Davane M, 60. Li HK, Scarborough M, Zambellas Rev Anti Infect Ther 2018;16:35–50 Nagoba B. Profile and antibiotic R, et al. Oral versus intravenous susceptibility of bacterial pathogens 51. Game FL, Jeffcoate WJ. Primarily antibiotic treatment for bone and joint associated with diabetic foot non-surgical management of infections (OVIVA): study protocol for ulcers from a rural area. Wounds osteomyelitis of the foot in diabetes. a randomised controlled trial. Trials 2019;31:158–162 Diabetologia 2008;51:962–967 2015;16:583 43. Anghel EL, DeFazio MV, Barker 52. Lesens O, Desbiez F, Vidal M, 61. Schultz G, Acosta JB, Cowan JC, Janis JE, Attinger CE. Current et al. Culture of per-wound bone L, Stechmiller J. Linking the concepts in debridement: science specimens: a simplified approach for advanced glycation endproducts/ and strategies. Plast Reconstr Surg the medical management of diabetic receptor for advanced glycation 2016;138(Suppl. 3):82S–93S foot osteomyelitis. Clin Microbiol endproducts pathway in diabetics 44. Strohal R, Dissemond J, Jordan Infect 2011;17:285–291 with with inflammation and topical O’Brien J, et al. EWMA document: 53. Senneville E, Lombart A, Beltrand antiinflammatory treatments of debridement: an updated overview E, et al. Outcome of diabetic foot chronic wounds. Adv Wound Care and clarification of the principle osteomyelitis treated nonsurgically: a (New Rochelle) 2010;1:248–253 role of debridement. J Wound Care retrospective cohort study. Diabetes 62. Schierle CF, De la Garza M, Mustoe 2013;22:5 Care 2008;31:637–642 TA, Galiano RD. Staphylococcal 45. Senneville E, Abbas ZG, Aragón- 54. Aragón-Sánchez J, Lázaro- biofilms impair wound healing by Sánchez J, et al. Diagnosis of infection Martínez JL, Hernández-Herrero C, delaying reepithelialization in a murine

DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS | 23 cutaneous wound model. Wound The effect of negative pressure wound 81. Basso FG, Oliveira CF, Fontana Repair Regen 2009;17:354–359 therapy with periodic instillation A, et al. In vitro effect of low-level 63. James GA, Swogger E, Wolcott using antimicrobial solutions on laser therapy on typical oral microbial R, et al. Biofilms in chronic wounds. Pseudomonas aeruginosa biofilm on biofilms. Braz Dent J 2011;22:502–510 Wound Repair Regen 2008;16:37–44 porcine skin explants. Int Wound J 82. Nishikawa T, Yoshida A, Khanal 2013;10(Suppl. 1):48–55 64. Malone M, Bjarnsholt T, McBain A, et al. A study of the efficacy of AJ, et al. The prevalence of biofilms in 73. Yang C, Goss SG, Alcantara ultrasonic waves in removing biofilms. chronic wounds: a systematic review S, Schultz G, Lantis Ii JC. Effect of Gerodontology 2010;27:199–206 and meta-analysis of published data. J negative pressure wound therapy 83. Snyder RJ, Bohn G, Hanft J, et al. Wound Care 2017;26:20–25 with instillation on bioburden in Wound biofilm: current perspectives chronically infected wounds. Wounds and strategies on biofilm disruption and 65. Fux CA, Costerton JW, Stewart 2017;29:240–246 treatments. Wounds 2017;29:S1–S17 PS, Stoodley P. Survival strategies of infectious biofilms. Trends Microbiol 74. Fazli M, Bjarnsholt T, Kirketerp- 84. Black CE, Costerton JW. Current 2005;13:34–40 Møller K, et al. Nonrandom distribution concepts regarding the effect of of Pseudomonas aeruginosa wound microbial ecology and biofilms 66. Costerton JW, Stewart PS, and Staphylococcus aureus in on wound healing. Surg Clin North Am Greenberg EP. Bacterial biofilms: a chronic wounds. J Clin Microbiol 2010;90:1147–1160 common cause of persistent infections. 2009;47:4084–4089 Science 1999;284:1318–1322 85. Wolcott RD, Rhoads DD. A study of 75. Yang Q, Larose C, Della Porta AC, biofilm-based wound management in 67. del Pozo JL, Patel R. The challenge subjects with critical limb ischaemia. of treating biofilm-associated bacterial Schultz GS, Gibson DJ. A surfactant- J Wound Care 2008;17:145–148, infections. Clin Pharmacol Ther based wound dressing can reduce 150–152, 154–155 2007;82:204–209 bacterial biofilms in a porcine skin explant model. Int Wound J 86. Alkawash MA, Soothill JS, Schiller 68. Dumville JC, Lipsky BA, Hoey C, 2017;14:408–413 NL. Alginate lyase enhances antibiotic Cruciani M, Fiscon M, Xia J. Topical killing of mucoid Pseudomonas antimicrobial agents for treating 76. Gristina AG, Price JL, Hobgood aeruginosa in biofilms. APMIS foot ulcers in people with diabetes. CD, Webb LX, Costerton JW. Bacterial 2006;114:131–138 Cochrane Database Syst Rev colonization of percutaneous sutures. 2017;6:CD011038 Surgery 1985;98:12–19 87. National Institute of Diabetes and Digestive and Kidney Diseases. Diabetic 69. Nakagami G, Schultz G, Gibson 77. Harrison-Balestra C, Cazzaniga foot consortium—biomarker meeting. DJ, et al. Biofilm detection by AL, Davis SC, Mertz PM. A wound- Available from www.niddk.nih.gov/ wound blotting can predict slough isolated Pseudomonas aeruginosa news/meetings-workshops/2018/ development in pressure ulcers: a grows a biofilm in vitro within 10 hours diabetic-foot-consortium-biomarkers. prospective observational study. and is visualized by light microscopy. Accessed 28 October 2019 Wound Repair Regen 2017;25:131–138 Dermatol Surg 2003;29:631–635 88. Conte MS, Bradbury AW, Kolh P, 70. Rennie MY, Lindvere-Teene L, 78. Manci KA, Kirsner RS, Ajdic D. et al. Global vascular guidelines on Tapang K, Linden R. Point-of-care Wound biofilms: lessons learned from the management of chronic limb- oral biofilms. Wound Repair Regen fluorescence imaging predicts the threatening ischemia. J Vasc Surg 2013;21:352–362 presence of pathogenic bacteria in 2019;69(Suppl. 6):3S–125S.e40 wounds: a clinical study. J Wound 79. Nusbaum AG, Gil J, Rippy MK, 89. Mills JL Sr, Conte MS, Armstrong Care 2017;26:452–460 et al. Effective method to remove DG, et al.; Society for Vascular wound bacteria: comparison of 71. Weed T, Ratliff C, Drake DB. Surgery Lower Extremity Guidelines various debridement modalities in Quantifying bacterial bioburden Committee. The Society for Vascular an in vivo porcine model. J Surg Res during negative pressure wound Surgery Lower Extremity Threatened therapy: does the wound VAC 2012;176:701–707 Limb Classification System: risk enhance bacterial clearance? Ann 80. Cooper R, Okhiria O. Biofilms, stratification based on wound, Plast Surg 2004;52:276–279 wound infection and the issue of ischemia, and foot infection (WIfI). J 72. Phillips PL, Yang Q, Schultz GS. control. Wounds UK 2006;2:48–57 Vasc Surg 2014;59:220–234.e1–2

24 | DIAGNOSIS AND MANAGEMENT OF DIABETIC FOOT INFECTIONS Suggested citation: Boulton AJM, Armstrong DG, Hardman MJ, et al. Diagnosis and Management of Diabetic Foot Infections. Arlington, Va., American Diabetes Association, 2020 (https://doi.org/10.2337/db2020-01)