Diabetic Foot Infections

Updated July, 2011

Authors: Paul B. Cornia, M.D., Benjamin A. Lipsky, M.D.

Foot infections are common, costly, potentially limb or even life-threatening complications of diabetes mellitus. Diabetic foot infection may be defined most simply as any acute or chronic inflammatory response to a microbial invasion in the infra-malleolar area in a person with diabetes. Because of the comorbidities associated with diabetes, these infections may begin as a seemingly minor problem but often progress, sometimes rapidly, if not managed appropriately. Proper treatment often requires appropriate wound care (usually including debridement) as well as antimicrobial therapy. Most often antibiotic therapy must be initiated empirically in persons with diabetic foot infection while awaiting the results of wound cultures. Because of the complexity of diabetic foot infection, an evidence-based, well coordinated, multi-disciplinary team approach improves outcomes (1, 7, 27).


Most diabetic foot infections begin with a break in the protective cutaneous barrier, typically in the form of a neuropathic ulcer. The lifetime risk of foot ulceration in persons with diabetes is about 15-25% and about 60% of ulcers are clinically infected at presentation (4). A recent prospective study found that despite patient education and provider follow-up, 9% of diabetic patients developed a foot infection during a two year observation period (4); nearly all of these were precipitated by a foot wound. In developed countries, 85% of lower extremity amputations in diabetic patients are preceded by a foot ulcer. A complex, incompletely understood interplay of risk factors may conspire to cause foot ulceration and foot infection in persons with diabetes (Table 1 ).

back to top


Peripheral sensory neuropathy plays the central role in the development of foot ulcers, primarily by causing a loss of protective sensation. Importantly, most patients do not recognize the loss of protective sensation, underscoring the need for preventive efforts including serial surveillance screening and daily foot inspection. Skin ulceration usually results from unperceived repetitive sheer-type trauma due to altered weight bearing (foot deformities and excess plantar pressure), ill fitting shoes, or less commonly, skin-penetrating trauma. Peripheral motor neuropathy can cause abnormal foot biomechanics and lead to distorted foot anatomy. Dry, thickened, and cracked skin related to peripheral autonomic neuropathy increases the risk of skin breaks, offering a portal of entry for bacteria. Most foot ulcers develop on the plantar surface of the foot, especially at the metatarsal heads and to a lesser degree on the toes and calcaneum, the sites of highest pressure with standing and ambulation. Once a foot ulcer develops, several poorly characterized immunological and metabolic perturbations may impair healing and allow the infection to progress. Another risk factor for developing an infection, and for worsening its severity, is limb ischemia. Peripheral arterial disease typically affecting the major arteries between the knee and ankle, is twice as common in persons with diabetes as in non-diabetics (14). Diminished tissue perfusion inhibits wound healing and impairs delivery of antibiotics to infected tissue. Diabetes-related immune dysfunction, caused by impaired neutrophil and macrophage chemotaxis and phagocytosis, also predisposes to more frequent, rapidly spreading and severe infections (13).

back to top


Selecting an appropriate empiric antibiotic regimen requires knowing the usual etiologic organisms (Table 2). Numerous studies have assessed the microbiology of diabetic foot infection, but because of the heterogeneity of the patient populations and culture methods used the reported results vary substantially (24). Isolation of multiple species of bacteria from a wound specimen is common with many studies reporting a mean of 2-5 isolates per case (24). Isolation of bacteria from a wound specimen does not define pathogenicity and distinguishing pathogens from colonizers may be difficult. Nonetheless, in almost all cases collecting appropriately obtained specimens for culture is helpful. The results of these cultures may allow tailoring of the antibiotic regimen – either to a narrower spectrum or one targeted to specific bacteria not covered by the initial empiric regimen. Almost all studies, however, show that aerobic gram-positive cocci, particularly Staphylococcus aureus, are responsible for most acute infections, especially in patients who have not recently received antibiotic therapy. Methicillin-resistantStaphylococcus aureus(MRSA) is an increasingly frequent pathogen in diabetic foot infections. Recent studies have reported that an increasing frequency of MRSA isolates in diabetic foot infections. A recent review of studies conducted between 1993 and 2007 found a prevalence of MRSA in diabetic foot infection ranging from 5-30%; the majority reported rates of 10-20% (9). Coagulase-negative staphylococci are also frequently isolated, and may be mistakenly dismissed as contaminants. Because persons with diabetes are immunologically compromised, these are often true pathogens. Beta-hemolytic streptococci (usually group B) are also relatively frequent pathogens especially in patients with cellulitis. Enterococci are among the more common isolates in many studies, but their clinical significance is uncertain. Because cephalosporins are active against many common pathogens in diabetic foot infection, but not enterococci, treatment with this class predisposes to infection withEnterococcusspp.

Aerobic gram-negative bacilli, usually including Enterobacteriaceae (Escherichia coli, Proteus, Klebsiella, and Enterobacter) are also frequently isolated from diabetic foot infection, especially chronic or previously treated infections. Gram-negative rods are rarely the sole, or even predominant, pathogen. Many broad-spectrum antibiotic agents will cover Enterobacteriaceae, but notP Pseudomonas aeroginosa. Pseudomonas deserves specific mention, as it is a relatively frequent isolate and usually requires specifically targeted therapy. This water-borne organism often colonizes, and sometimes infects, wounds that have been soaked or subjected to hydrotherapy. It is also reported more frequently from countries with warmer climates (11, 17, 26); this may be a consequence of excessive sweating into shoes or wearing open or no footwear with exposure to the soil. Gram-negative organisms that elaborate extended spectrum beta-lactamases (ESBL) are also a growing problem in diabetic foot infection. As withPseudomonas, these generally require specifically targeted regimens.

Obligately anaerobic organisms may also cause diabetic foot infection. Recognizing these isolates requires obtaining proper (usually tissue) specimens and then quickly and appropriately processing them. The most frequent anaerobic isolates are peptococci and peptostreptococci, and less often Bacteroides species. These may be important pathogens, but almost exclusively occur as part of a polymicrobial infection with an ischemic or necrotic wound. The clinician must also be aware that the causative organism may change over time, especially if the patient’s infection fails to clinically respond to simple regimens.

(Printable Version of Causes of Diabetic Foot Infections)

back to top

Initial Assessment

Not all ulcers are infected. Since all wounds are colonized by microorganisms, infection must be diagnosed clinically rather than microbiologically. Various authoritative committees (Infectious Disease Society of America [IDSA], International Working Group on the Diabetic Foot [IWGDF], and American Diabetes Association) have defined infection in the diabetic foot as the presence of purulent secretions or at least two symptoms or signs of infection (erythema, warmth, tenderness, pain, or induration). Importantly, local and systemic inflammatory responses to infection may be diminished in those with peripheral neuropathy or arterial insufficiency – additional signs of infection that may be useful include purulent secretions, friable or discolored granulation tissue, undermining of the wound edges or a foul odor. In a patient with limb ischemia, infection may reach a limb-threatening state before the patient or clinician recognizes the problem. Eradicating infection in a wound will certainly facilitate healing, but it will usually take additional time for the wound to completely heal. As long as the wound remains, it is at continued risk of re-infection. Thus, curing infection is a separate, albeit related, issue to wound healing.

Because of the complex nature of diabetic foot infection and the potential for rapid worsening (sometimes within hours), the clinician must assess the patient promptly, methodically and repeatedly (Figure 1 and Figure 2, IDSA guidelines). Evaluate for systemic evidence of infection (e.g., fever, chills, leukocytosis), examine the affected limb (i.e., for foot deformities, altered biomechanics, neuropathy, and arterial insufficiency) and finally the wound (size see photo, depth, tissues involved see photo , necrosis/gangrene see photos, foreign objects). There are several classification schemes for diabetic foot ulcers see photos and the lack of consensus on wound definitions and infection classification makes comparison of published studies difficult and is confusing to clinicians. Most, however agree that the critical factors in evaluating a diabetic foot wound are its depth and the limb’s vascular status. The recently published guidelines from the IWGDF (15) and IDSA (5) (update in progress, publication planned 2011) are similar (Table 3). The IDSA scheme has been validated and predicts clinical outcome (6). Assessing the component features should influence decisions regarding site of therapy (inpatient vs. outpatient); the spectrum, route of administration and duration of antibiotic therapy; the urgency of any necessary surgical intervention; and likely, the outcome. Identifying causative pathogens using proper wound culturing techniques guides antibiotic therapy, especially for chronic infections and persons recently treated with antibiotics. A Gram-stained smear of a wound specimen can provide real-time information on the likely causative organisms. When selecting an initial antibiotic regimen, it is most helpful for deciding whether or not to add coverage for gram-negative rods in a patient with mild infection.

back to top

Determining the Severity of Infection

Systemic Evidence of Infection:

Systemic symptoms and signs of infection include fevers, chills, diaphoresis, anorexia, hemodynamic instability (tachycardia, hypotension), metabolic derangements (e.g., acidosis, dysglycemia, volume depletion, renal failure), leukocytosis and inflammatory markers. Surprisingly to many clinicians, these are uncommon in patients with a diabetic foot infection. When systemic signs or symptoms are present they generally signify severe infection with extensive tissue involvement or more virulent pathogens. But, elevated temperature, white blood cell count, or sedimentation rate are absent in up to 50% of severe diabetic foot infection (10).

Extent of Tissue Involvement:

A key factor in determining the outcome of a diabetic foot infection is to assess the wound depth and which tissues are involved. This requires first debriding see photos any necrotic material or callus, then gently probing to any abscesses, sinus tracts, foreign bodies or bone or joint involvement see photos . Occasionally, defining the extent of infection requires an imaging study (usually MRI) or surgical exploration. If there is any concern for necrotizing deep space infection see photos , an experienced surgeon should evaluate the patient. Deeper and more extensive infections may respond more slowly to appropriate antibiotic therapy. Palpating bone see photos in a diabetic foot ulcer using a steel probe (a positive “probe-to-bone” test) is a simple and useful bedside test to aid in the diagnosis of osteomyelitis (6, 8). The positive predictive value approaches 90% when the pre-test probability of osteomyelitis is high (12), but is closer to 55% when the prevalence (usually ~20%) is lower (21). Visibly exposed bone probably provides similar information as probing bone.

back to top

Assessment of Peripheral Arterial Perfusion:

The presence of PAD is an independent risk factor for developing a diabetic foot infection (20) and is present in up to 40% of cases see photos . The presence of significant arterial insufficiency in an infected limb adversely affects host immunological responses and wound healing and impairs delivery of systemic antibiotics to the infected tissues. The absence of pedal pulses suggests peripheral artery disease, but this method of assessment of arterial perfusion is often not reliable, especially in persons with diabetes. Determining the ratio of ankle to brachial artery systolic blood pressure (ankle/brachial index [ABI]) is a simple, reliable, non-invasive, bedside procedure to assess for peripheral artery disease (30) and should be performed in most patients with diabetic foot infection especially if pedal pulses are absent or diminished. An ABI <0.90 is abnormal, and <0.40 signifies severe ischemia (Table 4), but arterial calcification of vessels may falsely elevate the ankle/brachial index in patients with diabetes. Assessing the transcutaneous partial pressure of oxygen (TcpO2) in the skin of the foot is another non-invasive method of assessing peripheral arterial perfusion. TcpO2 values of less than 30mmHg signify critical limb ischemia and predict poor wound healing. If significant peripheral artery disease is suspected based on history, physical examination, or non-invasive testing, vascular surgery consultation is appropriate. Limb revascularization may be necessary to cure infection and promote healing. Having access to an active lower extremity revascularization program can decrease amputation rates and increase the incidence of foot sparing surgeries that have a more favorable long-term outcome.

back to top

Laboratory Diagnosis

Obtaining Cultures in Diabetic Foot Infections

Properly obtained wound cultures (Table 5) are useful for guiding antibiotic therapy in diabetic foot infections, particularly in patients with chronic infections or who have recently been treated with antibiotics. Culture specimens should be obtained after the wound has been cleansed and debrided. A sample obtained by curettage, the aseptic scraping of tissue at an ulcer base using a scalpel blade or dermal curette, more accurately identifies pathogens than a wound swab see photo. Swabs are often contaminated with normal skin flora or colonizers (24) and are less likely to grow anaerobic, and some fastidious aerobic organisms. Specimens must be promptly transported to the laboratory, in an appropriate sterile transport system, where they should be processed for aerobic and anaerobic cultures and a Gram-stained smear. Other acceptable methods of culturing wounds include aspiration of cellulitic tissue or purulent secretions, and tissue biopsy obtained either at the bedside or at surgery. A bone biopsy, obtained surgically or percutaneously see photo, processed for culture (and histological assessment, if possible) is the criterion standard for diagnosing osteomyelitis. The results of wound or sinus tract cultures do not accurately reflect those of bone culture. Blood cultures are not frequently positive in these infections but should be obtain in patients with systemic symptoms and signs of infection. In the minority of cases with bacteremia,S. aureusis the most frequently isolated pathogen.

back to top

Imaging Studies

Imaging studies may be useful in a patient with diabetic foot infection to assess for any foreign material, soft tissue abscessessee photos, or bony abnormalities. Plain radiographs are usually the appropriate first study but have limited diagnostic utility in assessing for osteomyelitis see photos . They lack sensitivity early in infection because abnormalities on plain film may not be evident until 50% of the bone is resorbed which typically requires 2-3 weeks. They also lack specificity because neuroarthropathy (Charcot foot) may have a similar radiographic appearance. If suspicion for osteomyelitis remains despite an initial negative radiograph, repeating plain films in a few weeks can either exclude the diagnosis (if still negative) or suggest that it has developed (if there is cortical erosion, periosteal elevation or other suggestive changes in one underlying the affected soft tissue).

Radionucleotide bone scans (using bisphosphonate-linked technetium or other radionuclides) are more sensitive than plain radiographs for diagnosing osteomyelitis, but uptake occurs with any type of inflammation see photos, resulting in poor specificity (~50%). Labeled (e.g., with Indium111) white cell or immunoglobulin scans have better specificity (~75%) than bone scans see photos . Among imaging modalities, magnetic resonance imaging see photo has the best overall sensitivity (>90%) and specificity (>80%) for detecting osteomyelitis and higher resolution for soft tissue abnormalities. It is now considered the imaging procedure of choice (17, 21), but it is still relatively expensive and often not readily available.<

(Printable Version of Assessment of Diabetic Foot Infections)

back to top


Antibiotic Therapy

Diabetic foot infection may be limb- or even life-threatening and may progress rapidly. Too narrow an empiric antibiotic regimen may therefore lead to a poor clinical outcome. Furthermore, diabetic foot infection are commonly polymicrobial. Thus, clinicians should generally opt for broad-spectrum empiric therapy while awaiting culture and sensitivity results. This is not, however, always necessary. Unnecessarily broad-spectrum therapy also has potential adverse consequences, such as increasing antibiotic resistance (a potential problem for both the individual patient and society as a whole), increased financial cost, and potentially increased risk of drug toxicity. Properly obtained specimens for microbial evaluation are crucial to optimally direct antibiotic therapy and to potentially allow the empiric antibiotic regimen to be narrowed or targeted.

Deciding upon an empiric antibiotic regimen for diabetic foot infection is challenging, but understanding some basic principles can guide clinicians (22). An important place to start is to realize that about half of diabetic foot wounds have no clinical evidence of infection (20). Some believe that the presence of a high density of organisms, usually defined as >105 per gram of tissue, represents “critical colonization” that may require antibiotic therapy (5), but the few published trials of antibiotic therapy for uninfected lesions do not support this practice. Thus, clinically uninfected wounds generally require neither a culture nor antibiotic therapy.

For clinically infected lesions, mild, and most moderate, infections (Table 3) can be treated with relatively narrow spectrum agents predominantly directed at staphylococci and streptococci. An exception to this would be a patient who has recently been treated with an antibiotic. This may predispose to infection with more unusual and antibiotic resistant organisms. More severe infections generally mandate a broader-spectrum regimen. All empiric regimens should cover aerobic gram-positive cocci, especially Staphylococcus aureus, since they are the most frequently isolated organism from a diabetic foot infection (acute and chronic; mild, moderate and severe).

The decision as to whether or not to provide empiric coverage for MRSA is based on the presence of known risk factors and the local prevalence MRSA. Previous history of MRSA infection or colonization may be the best predictor. Other risk factors include recent long-term antibiotic use, recent hospitalization, prolonged duration of the wound, and presence of osteomyelitis, but these have not been consistently reported in all studies.

Aerobic gram-negative bacilli are also commonly isolated as part of a polymicrobial infection see photo, especially in a chronic wound in a patient who has received antibiotic therapy. Because Pseudomonas aeruginosa is a hydrophilic organism, hydrotherapy (e.g., soaking the affected foot in water or debridement with water lavage) is the main risk factor. Even when isolated in diabetic foot infection, Pseudomonas is usually part of a mixed infection and rarely the predominant pathogen and patients often improve without anti-pseudomonal antibiotic therapy. Selecting a regimen aimed at this frequently antibiotic resistant organism should thus be reserved for patients with one of the noted risk factors or who have green-blue colored wound drainage (Table 2) and a more severe infection see photo. As previously mentioned, recent reports from developing countries with a tropical or arid climate have shown more frequent isolation of aerobic gram-negative bacilli, including pseudomonas, rather than gram-positive cocci. Clinicians in such regions may consider broader spectrum (i.e., covering both aerobic gram-negative bacilli and gram-positive cocci) empirical therapy.

Obligate anaerobic bacteria are isolated from a substantial minority of cases, but almost always as part of a polymicrobial infection and usually associated with a chronic wound. Anaerobes are more often isolated from necrotic or gangrenous wounds with limb ischemia. A foul, feculent odor (the so called “fetid foot”) is also a clue to infection with anaerobes. Since oxygen is a potent anti-anaerobic agent, adequate treatment may require only debridement of necrotic tissue and exposing the remaining organisms to air. An empiric antibiotic regimen directed at anaerobic organisms may be appropriate when clinical suspicion is high and the infection is moderate to severe. The anaerobes in diabetic foot infection are mainly gram-positives (peptococci or peptostreptococci) rather than Bacteroides spp. Especially with anaerobic infections, antibiotic therapy is no substitute for adequate debridement and drainage.

One final consideration is fungal infection. While tinea infections of the webspaces between the toes and onychomycosis are frequent in diabetic patients, pathogenic fungal infections are not. Occasionally, however, a patient who has received multiple antibiotics, who has severe hyperglycemia, or who is receiving corticosteroid therapy will develop an invasive soft tissue fungal infection.

Surprisingly few studies have been performed to assess the efficacy of various antibiotics to treat diabetic foot infection. Furthermore, the available studies are not well standardized, making comparisons difficult. In virtually all comparative studies the regimens produced equivalent results. Thus, while several agents have been used successfully to treat diabetic foot infection, no specific antibiotic regimens have emerged as being preferred for a particular type of infection. Based on a review of available studies, the IDSA formulated suggested antibiotic regimens for the treatment of soft-tissue diabetic foot infection, as shown in Table 6. The general approach is to select as narrow spectrum, safe, and convenient a regimen as possible (IDSA guidelines). In about two-thirds of patients this will need to be empiric. When culture (or Gram-stained smear) results are available, the clinician should of course use this information. After 2-3 days, the clinician should reassess the patient and the infection and consider altering the definitive regimen based on the culture and sensitivity results and the patient’s clinical response to the selected regimen.

Mild infections see photo may be treated with relatively narrow spectrum oral therapy directed at these aerobic gram-positive cocci, such as penicillinase-resistant penicillins (e.g., dicloxacillin), first-generation cephalosporins (e.g., cephalexin), or clindamycin. Some mildly infected wounds can be treated with topical antimicrobial therapy, either antiseptics (e.g., silver or iodine based) or antibiotic (e.g., mupirocin or bacitracin) but there are currently few published studies to support this approach. New topical products are now undergoing testing and may emerge as useful treatments for selected cases.

Moderate see photos to severe infections see photos often necessitate empirical regimens with activity against commonly isolated gram-negative bacilli, MRSA (if there are risk factors as previously discussed) and perhaps Enterococcus species. Enterococci are relatively frequent isolates, especially in patients previously treated with cephalosporins, but they are often colonizers and rarely primary pathogens. Broadening empirical antibiotic coverage based on previously mentioned clinical findings (Table 1 ) is often appropriate. As a general rule, we err on the side of overly broad initial coverage, put our effort into obtaining high quality specimens for culture, ensuring proper wound care and thinking more about the definitive antibiotic regimen to complete the course when culture results are available. We think most errors in treatment consist of failing to alter (and curtail) antibiotic therapy rather than in making a poor initial antibiotic choice.

Additional considerations include the site of treatment and route of drug administration. Intravenous antibiotic therapy and hospitalization are not only costly but also potentially expose patients to iatrogenic complications and nosocomial infection. The severity of infection is usually the most important determinant of the need for hospitalization. Other factors may also influence the decision, such as the clinician’s desire to closely monitor the wound and response to treatment, a need for expedited diagnostic testing or consultant evaluation, and adverse social circumstances (e.g., inability of the patient to adequately care for the wound, off-load the foot, or uncertainty regarding compliance with antibiotics or follow-up care). Most patients with mild to moderate infections can be successfully treated as outpatients with oral antibiotic therapy ( 23). Many oral antibiotics have high bioavailability allowing therapeutic serum levels. An antibiotic’s pharmacokinetic and pharmacodynamic properties determine its serum, and therefore tissue, levels. When parenteral therapy is needed, it can often be offered by outpatient intravenous therapy with a once daily agent when this service is available. The arterial supply to the foot must also be adequate to ensure delivery of therapeutic antibiotic levels to the infected tissue.

After wound culture and antibiotic susceptibility results become available, the clinician should consider narrowing the antibiotic coverage to the isolated pathogen(s). Clinical assessment of the patient and wound, including the response to the empirical regimen, are paramount in tailoring the antibiotic regimen. If the wound is clinically responding, it is reasonable to continue an empirical regimen, even if it does not have activity against all the isolated organisms. Conversely, if a wound is not improving, broaden antibiotic coverage to cover all isolated organisms, even those whose pathogenicity is uncertain. The optimal duration of antibiotic therapy has not been studied, but we recommend one to two weeks for mild to moderate soft tissue infections and at least two weeks (occasionally longer) for deeper or more complex soft tissue infections (Table 7). It is not usually necessary to continue antibiotic therapy for soft tissue infections beyond the point when all clinical signs if infection have resolved see photos .

back to top

Surgical Intervention

Surgical procedures designed to control infection and preserve as much the affected limb as possible are frequently required in the treatment of diabetic foot infection (29). Early, thorough debridement of infected and necrotic tissue helps define the extent of infection and may reduce the likelihood of future amputation. Severe infections may be immediately limb-threatening and require urgent surgical consultation. Unexplained, persistent foot pain, particularly in a patient known to have advanced neuropathy (i.e., a previously insensate foot) or distortion of the superficial anatomy (e.g., bulging of the plantar surface see photo or dorsal induration see photo in a patient with a plantar ulcer) may indicate deep space infection that warrants prompt evaluation by a surgeon. When amputation is necessary, the surgeon should make every effort to perform a minor (below the ankle) rather than a major amputation to preserve functionality. If severe peripheral arterial disease is present, a vascular surgeon should be consulted promptly. Non-urgent amputation should rarely be done without a vascular surgery consultation. Early revascularization (i.e., within days) is generally preferred to delaying the procedure for a prolonged course of potentially ineffective antibiotic therapy. Endovascular or bypass procedures can be limb salvaging in some patients with a diabetic foot infection.

back to top

Wound Care

After the initial assessment and treatment, the patient requires ongoing wound care. Regular follow-up evaluations and repeated debridement of any residual dead tissue can help promote continued healing. Many dressings and wound-care products are marketed but none are favored based on evidence. A moist dressing that allows daily wound inspection is optimal. The roles of adjunctive treatments such as hyperbaric oxygen and granulocyte colony stimulating factors have not yet been adequately defined. Meta-analyses of these treatments suggest that an as yet undefined subset of patients may benefit from these treatments. Considering their cost they should rarely be used as initial therapy but rather reserved for patients who are not responding to what should otherwise be appropriate treatment.

Many clinicians and patients do not recognize the importance of off-loading pressure from the affected foot ( 2, 4). A variety of footwear devices are designed to redistribute pressure off the affected part of the foot. The total contact cast has been shown to be the best device, but the clinician and patient cannot easily view the infected wound in this device. In some patients healing sandals, half shoes, and removable cast walkers are adequate. Complete offloading of the affected limb may also be achieved with crutches, a walker or a wheelchair; however, these devices may increase the risk of ulceration of the contralateral limb and confinement to a wheelchair may not be practical and leads to deconditioning. With the exception of the total contact cast (which the patient cannot remove), patient compliance is often suboptimal. Select a dressing and off-loading device with the fact in mind that infected wounds must generally be examined and dressed daily.

back to top


Osteomyelitis frequently complicates diabetic foot infection but its evaluation and management remain controversial (16). Except for direct inoculation by a puncture wound, bone infection spreads from soft tissue to cortex to marrow. Thus, nearly all cases in the setting of a diabetic foot infection are chronic osteomyelitis. Traditionally, surgical resection of infected and necrotic bone has been viewed as the best approach to curing these infections. Recent retrospective reports have demonstrated clinical success in ~65-80% of selected cases using prolonged (3-6 months) antibiotic (usually fluoroquinolone) therapy with little or no surgery. Prospective trials are needed to define in which clinical circumstances osteomyelitis may be appropriately treated without surgery. Using the results of bone culture is the preferred means to guide antibiotic therapy, especially if it is to be prolonged. When all the necrotic bone has been resected, 4-6 weeks of antibiotic therapy for the residual infected bone is usually sufficient. If nonsurgical treatment is selected, initial parenteral therapy (for perhaps 1-2 weeks) is recommended. Oral antibiotic(s) with good bioavailability may then be used to complete a course of therapy. If all infected and necrotic bone has been resected, a brief course of antibiotic therapy (1-2 weeks) for any residual soft tissue infection is sufficient (3). While much discussed, the data on the accuracy of measuring and importance of achieving adequate bone levels of antibiotics to treat osteomyelitis are weak. Some studies support the use of clindamycin; clinical experience suggests that beta-lactams and selected fluoroquinolones work well and some recent case series support using linezolid (for the shortest necessary duration and with careful monitoring for hematological and neurological toxicity).

back to top


Having a diabetic foot infection increases the risk of future recurrence – emphasize this to the patient at this teachable moment to underscore the importance of preventive strategies. Identifying patients with peripheral neuropathy, foot deformity or peripheral artery disase, and educating them about the risk of foot ulcer and diabetic foot infection is a simple, but critical and often overlooked, way to prevent diabetic foot infection (28). Primary and secondary prevention measures include optimizing metabolic (glycemic, lipid) control, wearing appropriate footwear, avoiding mechanical or thermal foot trauma, and performing a daily foot inspection. Fungal infections of the foot are more common in people with diabetes. Patients with tinea pedis should be treated with a topical anti-fungal, while those with severe onychomycosis may benefit from oral anti-fungal therapy. Persons with major foot complications should be referred to a specialist (e.g., podiatric, orthopedic or vascular surgeons).

back to top


Diabetic foot infection is a common, complex complication of diabetes. An early and thorough clinical evaluation, with attention to the presence of systemic symptoms and signs infection, the extent of local tissue involvement, and an assessment of peripheral arterial perfusion is essential. Obtaining proper culture specimens (e.g., biopsy or curettage of the cleansed, debrided ulcer base or aspiration of purulent secretions) will help guide antibiotic selection. Most patients should have plain radiographs of the foot. If osteomyelitis is suspected, this is best diagnosed by MRI and bone biopsy for microbiological and histological evaluation. Initial antibiotic therapy is usually empirical. In the absence of culture results, this should be directed at the most commonly encountered pathogens, i.e., aerobic gram-positive cocci. Certain clinical clues can suggest the presence of gram-negative rods or anaerobes. The spectrum of pathogens targeted by the empirical regimen, as well as the route of administration, duration of therapy, and need for hospitalization are primarily determined by the severity of infection. The antibiotic regimen should be re-assessed when culture and sensitivity tests become available, but any changes should be guided by the patient’s clinical response to the empirical agents. Deep space and extensive infections typically require surgical interventions. Consider early revascularization for a severely ischemic limb. Amputation may be required, but much of the foot can often be spared. Since infection often recurs, patients must be educated to examine their feet routinely and counseled to seek prompt medical attention at the first appearance of symptoms or signs of infection.

(Printable Version of Management of Diabetic Foot Infections)

back to top


1. Armstrong DG, Harless LB. Outcomes of preventative care in a diabetic foot clinic. J Foot Ankle Surg. 1998; 37: 460-6. [PubMed]

2. Armstrong DG, Lavery LA, Nixon BP, Boulton AM. It’s not what you put on, but what you take off: techniques for debriding and off-loading the diabetic foot wound. Clin Infect Dis. 2004; 39 Suppl 2: S92-9. [PubMed]

3. Berendt AR, Lipsky BA, Jeffcoate WJ, et al. Progress report on the diagnosis and treatment of diabetic foot osteomyelitis. International Working Group on the Diabetic Foot, interactive DVD available at www.idf.org/bookshop.

4. Boulton AJ. Pressure and the diabetic foot: clinical science and off-loading techniques. Am J Surg 2004; 187: 17S-24S. [PubMed]

5. Bowler PG, Duerden BI, Armstrong DG. Wound microbiology and associated approaches to wound management. Clin Microbiol Rev. 2001; 14: 244-69. [PubMed]

6. Butalia S, Palda VA, Sargeant RJ, Detsky AS, Mourad O. Does this patient with diabetes have osteomyelitis of the lower extremity? JAMA. 2008; 299: 806-13. [PubMed]

7. Crane M, Werber B. Critical pathway approach to diabetic pedal infections in a multidisciplinary setting. J Foot Ankle Surg. 1999; 38: 30-3. [PubMed]

8. Dinh MY, Abad CL, Safdar N. Diagnostic accuracy of the physical examination and imaging tests for osteomyelitis underlying diabetic foot ulcers: meta-analysis. Clin Infect Dis. 2008; 47: 519-27. [PubMed]

9. Eleftheriadou I, Tentolouris N, Argiana V, Jude E, Boulton AJ. Methicillin-resistant Staphylococcus aureus in diabetic foot infections. Drugs 2010; 70(14): 1785-97. [PubMed]

10. Eneroth M, Apelqvist J, Stenstrom A. Clinical characteristics and outcome in 223 diabetic patients with deep foot infections. Foot Ankle Int 1997; 18: 716-22. [PubMed]

11. Gadepalli R, Dhawan B, Sreenivas V, Kapil A, Ammini AC, Chaudhry R. A clinico-microbiological study of diabetic foot ulcers in an Indian tertiary care. Diabetes Care. 2006; 29: 1727-32. [PubMed]

12. Grayson ML, Gibbons GW, Balogh K, Levin E, Karchmer AW. Probing to bone in infected pedal ulcers. A clinical sign of underlying osteomyelitis in diabetic patients. JAMA 1995; 273: 721-3. [PubMed]

13. Geerlings SE, Hoepelman AI. Immune dysfunction in patients with diabetes mellitus. FEMS Immuneol Med Microbiol. 1999; 26: 259-65. [PubMed]

14. Gregg EW, Solie P, Paulose-Ram R, et al. Prevalence of lower-extremity disease in the US adult population ≥40 years or age with and without diabetes: 1999-2000 National Health and Nutritional Examination Survey. Diabetes Care. 2004; 27: 1591-97. [PubMed]

15. International Working Group on the Diabetic Foot. International consensus on the diabetic foot [CD-ROM]. Brussels: International Diabetes Foundation, May 2003. < [PubMed]

16. Jeffcoate WJ, Lipsky BA. Controversies in diagnosing and managing osteomyelitis of the foot in diabetes. Clin Infect Dis. 2004;39: S115-22. [PubMed]

17. Kandemir O, Akbay E, Sahin E, Milcan A, Gen R. Risk factors for infection of the diabetic foot with multi-drug resistant organisms. J Infect; 2007;54(5): 439-45. [PubMed]

18. Kapoor A, Page S, LaValley M, Gale DR, Felson DT. Magnetic resonance imaging for diagnosing foot osteomyelitis: a meta-analysis. Arch Intern Med. 2007; 167: 125-32. [PubMed]

19. Lavery LA, Armstrong DG, Murdoch DP, Peters EJ, Lipsky BA. Validation of the Infectious Disease Society of America’s diabetic foot classification system. Clin Infect Dis. 2007; 44: 562-5. [PubMed]

20. Lavery LA, Armstrong DG, Wunderlich RP, Mohler MJ, Wendel CS, Lipsky BA. Risk factors for foot infection in patients with diabetes mellitus. Diabetes Care. 2006; 29: 1288-93. [PubMed]

21. Lavery LA, Armstrong DG, Peters EJ, Lipsky BA. Probe-to-bone test for diagnosing diabetic foot osteomyelitis: reliable or relic? Diabetes care. 2007; 30: 270-4. [PubMed]

2 2. Lipsky BA. Empirical therapy for diabetic foot infections: are there clinical clues to guide antibiotic selection? Clin Microbiol Infect. 2007; 13: 351-3. [PubMed]

23. Lipsky BA, Pecoraro RE, Larson SA, Hanley ME, Ahroni JH. Outpatient management of uncomplicated lower-extremity infections in diabetic patients. Arch Intern Med. 1990; 150: 790-7. [PubMed]

24. Lipsky BA, Pecoraro RE, Wheat LJ. The diabetic foot: soft tissue and bone infection. Infect Dis North Am. 1990; 4: 409-32. [PubMed]

25. Lipsky BA, Berendt AR, Deery HG, et al. Diagnosis and treatment of diabetic foot infections. Clin Infect Dis. 2004;39: 885-910. [PubMed]

26. Martínez-Gómez DA, Ramírez-Almagro C, Campillo-Soto A, Morales-Cuenca G, Pagán-Ortiz J, Aguayo-Albasini JL. [Diabetic foot infections. Prevalence and antibiotic sensitivity of the causative microorganisms]. Enferm Infecc Microbiol Clin. 2009 Jun;27(6): 317-21.

27. Prompers L, Huijberts M, Apelqvist J, et al. Delivery of care to diabetic patients with foot ulcers in daily practice: results from the Eurodiale Study, a prospective cohort study. Diabet Med 2008;25: 700-7.

28. Singh N, Armstrong DG, Lipsky BA. Preventing foot ulcers in patients with diabetes. JAMA. 2005; 293: 217-28. [PubMed]

29. van Baal JG. Surgical treatment of the infected diabetic foot. Clin Infect Dis. 2004; 39 Suppl 2: S123-8. [PubMed]

30. Wilson JF, Laine C, Goldmann D. In the clinic. Peripheral arterial disease. Ann Intern Med. 2007; 146: ITC 3-1 – ITC 3-16. [ PubMed]

Table 1. Risk Factors for Foot Ulceration and Infection

Risk factor

Mechanism leading to ulceration, impaired wound healing or infection

Peripheral sensory neuropathy

Loss of protective sensation (e.g., repetitive shear-type stress leading to ulceration)

Peripheral motor neuropathy

Abnormal foot anatomy and biomechanics resulting in excess pressure

Peripheral autonomic neuropathy

Impaired sweating leading to dry, cracked skin

Arterial insufficiency

Diminished delivery of nutrient, oxygen, neutrophils, etc. leading to impaired wound healing and clearance of infection


Immune system (e.g., neutrophil) dysfunction and excess collagen cross-linking

Patient disability or non-adherence

Reduced vision (unable to inspect feet), prior amputation, lack of regular follow-up with medical care, poor hygiene, inappropriate footwear


Table 2. Pathogens Associated with Diabetic Foot Infection Syndromes

Diabetic foot infection syndrome


Cellulitis without ulceration

Beta-hemolytic streptococci (especially group B) and Staphylococcus aureus

Ulcer or wound, recently developed and no prior antibiotic treatment

S. aureus and beta-hemolytic streptococci

Ulcer or wound, chronic or recent antibiotic treatment

Usually polymicrobial – S. aureus and beta-hemolytic streptococci plus Enterobacteriaceae. Enterococci if previous cephalosporin therapy.

Ulcer or wound, prior hydrotherapy or green-blue colored drainage

Pseudomonas aeroginosa (often in combination with other organisms)

Extensive necrosis or gangrene, ischemic limb, feculent odor (“fetid foot”)

Polymicrobial – mixed aerobic gram-positive cocci (including enterococci), Enterobacteriaceae, nonfermentative gram-negative rods, and obligate anaerobes


MRSA; ESBL-producing gram-negative rods

MRSA = Methicillin-resistant Staphylococcus aureus

ESBL = Extended spectrum beta-lactamase


Table 3. Clinical Classification of Diabetic Foot Infections

Infection* severity

Clinical manifestations of infection


Wound lacking purulence or any manifestations of inflammation


Infection localized to the skin and subcutaneous tissue (cellulitis/erythema extends ≤2 cm around an ulcer) without evidence of systemic illness


More extensive local infection (i.e., local spread ≥2cm beyond an ulcer, lymphangitic streaking, abscess, gangrene, or involvement of deep soft tissue, muscle, fascia, tendon, joint or bone) without evidence systemic illness or severe metabolic derangements


Infection with systemic toxicity or severe metabolic derangements

* Infection defined as the presence of purulent secretions (pus) or ≥2 signs or symptoms of inflammation (erythema, warmth, tenderness, induration, pain)


Table 4. Interpretation of Ankle-Brachial Index Results

Ankle-brachial index (ABI)*



Poorly compressible vessels, arterial calcification




Mild obstruction


Moderate obstruction


Severe obstruction

* Obtained by measuring the systolic blood pressure in the ankle divided by that in the brachial artery


Table 5. Recommendations for Collection of Specimens for Culture from Diabetic Foot Wounds


Cleanse and debride wound before obtaining specimen(s) for culture

Obtain tissue specimen for culture by scraping with a sterile scalpel or dermal curette (curettage) or biopsy from the base of a debrided ulcer

Aspirate any purulent secretions using sterile needle/syringe

Promptly send specimens for culture in sterile container or appropriate transport media for aerobic and anaerobic culture

Do Not

Culture clinically uninfected lesions, unless for epidemiological studies

Obtain specimen for culture without first cleansing or debriding the wound

Obtain specimen for culture by swabbing the wound or wound drainage


Table 6. Suggested Antibiotic Regimens for the Treatment of Soft-Tissue Diabetic Foot Infections

Severity of infection

Route of administration

Recommended agents (choose on or more)*

Alternative agents*




(500 mg q.i.d.)



(250 mg q.i.d.)



(300 mg t.i.d.)



(875/125 mg b.i.d.)


(750 mg q.d.) ±


(300 mg t.i.d.)



(2 double-strength b.i.d.)


Intravenous until stable, then transition to an oral equivalent (or tailor based on culture results)


(3.0 gm q.i.d.)


Clindamycin (450 mg q.i.d.) + ciprofloxacin (750 mg b.i.d.)


(3.3 gm q.i.d.)


Clindamycin (600 mg q.i.d.) + ceftazidime (2 gm t.i.d.)


Ertapenem (Invanz®)

(1 gm q.d.)


Prolonged intravenous


(500 mg q.i.d.)


Clindamycin (900 mg q.i.d.) + tobramycin (5.1 mg/kg/d) + ampicillin (50 mg/kg q.i.d.)

Vancomycin (15 mg/kg b.i.d.) + aztreonam (2.0 gm t.i.d.) + metronidazole (7.5 mg/kg q.i.d.)

*Based on published evidence of efficacy for complicated skin infection or diabetic foot infections. Choice should be based on any available culture results, clinical factors (allergies, co-morbidities, etc.), local antibiotic susceptibility data, availability and cost.


Table 7. Suggested Route, Setting, and Duration of Antibiotic Therapy by Clinical Syndrome

Site and severity of infection

Route of antibiotic administration

Usual setting for therapy

Duration of therapy*

Soft tissue only





1-2 weeks, may extend up to 4 weeks if infection slow to resolve


Oral (or initial parenteral)

Outpatient or inpatient

2-4 weeks


Initial parenteral, switch to oral when clinically stable

Initial inpatient, discharge to outpatient when clinically stable

2-4 weeks

Bone or joint**


Post-amputation, no residual infected tissue

Parenteral or oral

Outpatient, unless patient ill or requires parenteral therapy (and no available OPAT)

2-5 days

Post-amputation, residual infected soft tissue but not infected bone

Parenteral or oral


2-4 weeks

Post-amputation, residual infected but viable bone

Initial parenteral, then consider switch to oral


4-6 weeks

Post-amputation, residual non-viable bone or no surgery

Initial parenteral, then consider switch to oral


>3 months

OPAT = Outpatient antibiotic therapy

* Treat until resolution of all (or most) of the original symptoms and signs of infection

** With or without concurrent soft tissue infection


Figure 1. Approach to treating a diabetic patient with a foot wound


Figure 2. Approach to treating a diabetic patient with a foot infection.


Consider hospitalization if any of the following criteria are present: systemic toxicity (e.g., fever and leukocytosis); metabolic instability (e.g., severe hypoglycemia or acidosis); rapidly progressive or deep tissue infection, substantial necrosis or gangrene, or presence of critical ischemia; requirement of urgent diagnostic or therapeutic interventions; and inability to care for self or inadequate home support.


Diabetic Foot Infections