You are on page 1of 5

Medical therapy of diabetic foot infections

Warren S. Joseph, DPM, FIDSA,a and Benjamin A. Lipsky, MD, FACP, FIDSA, FRCP,b Philadelphia, Pa;
and Seattle, Wash

Diabetic foot infections are a common and often serious problem, accounting for a greater number of hospital bed days
than any other complication of diabetes. Despite advances in both antibiotic therapy and surgical management, these
infections continue to be a major risk factor for amputations of the lower extremity. Although a number of wound size
and depth classification systems have been adapted for use in codifying diabetic foot ulcerations, none are specific for
infection. In 2003, the International Working Group on the Diabetic Foot developed guidelines for managing diabetic
foot infections, including the first severity scale specific for these infections. The following year, the Infectious Diseases
Society of America (IDSA) published their Diabetic Foot Infection Guidelines. In this article, we review some of the
critical points from the Executive Summary of the IDSA document and provide a commentary following each issue to
update the reader on any pertinent changes that have occurred since the publication of the original document in 2004.
The importance of a multidisciplinary limb salvage team, apropos this special joint issue of the American Podiatric
Medical Association and the Society for Vascular Surgery, cannot be overstated. ( J Vasc Surg 2010;52:67S-71S.)

Until 20 years ago, what little literature there was represent the most comprehensive multinational attempt to
available on treating infections of the foot in persons with codify a treatment approach to a condition that contributes to
diabetes suggested that: almost all were polymicrobial; 71,000 lower extremity amputations per year in the United
gram-negative bacilli and obligate anaerobes were frequent States alone (www.diabetes.org). Although an updated set of
pathogens; virtually all such patients needed to be hospital- DFI Guidelines is slated to be published by the IDSA Com-
ized; and they should be treated with parenteral, broad- mittee sometime in 2010, and several new studies specifically
spectrum antibiotic therapy. Studies published since 1990 investigating antibiotic treatment of DFIs have been pub-
have demonstrated that most of these assertions were at lished in the past 6 years,4 many of the original points pre-
least exaggerations, if not myths. With the accumulating sented in the Executive Summary of the original document are
case series and investigations on treatment of diabetic still valid and deserving of review. Furthermore, the validated
foot infections (DFIs), especially randomized clinical severity classification scale is now widely accepted and used to
trials and carefully conducted microbiological studies, by classify patients enrolled in DFI studies and will remain un-
the early 2000s, it was time for convening experts in the changed in the updated document.
field to develop evidence-based guidelines on diagnosing
and treating DFIs. SEVERITY SCALE
In 2003, the International Working Group on the
Diabetic Foot (IWGDF) appointed a committee that Many classification systems have been developed to
drafted DFI guidelines1 that were approved by members categorize diabetic foot wounds (especially ulcerations),
from over 60 participating countries and subsequently pub- some of which include the presence or absence of infection.
lished and distributed in multiple languages on DVD (www. None of these, however, have specifically classified the
iwgdf.org). The following year, the Infectious Diseases Soci- severity of infection. Probably the most recognized, and
ety of America (IDSA) empanelled an expert committee to most venerable, system was that proposed by Wagner.5 In
draft diabetic foot infection clinical guidelines that were pub- this classification with four grades, more useful as a descrip-
lished in 2004.2 These included an infection severity classifi- tor of size and depth of a wound, the presence of an
cation scale that was subsequently validated and shown to infection is not considered until a Grade III ulceration,
predict clinical outcomes.3 These documents, taken together, which is described as a deep ulcer with abscess, osteomyeli-
tis, or joint sepsis. It is reasonable to ask, why can’t a less
From the Roxborough Memorial Hospital, Philadelphia;a University of extensive Grade 0, I, or II ulcer become infected? A newer,
Washington and Primary Care Clinic, Antibiotic Research, VA Puget
Sound HCS, Seattle.b
validated system first proposed by investigators at the
Competition of interest: Dr Joseph is on the Speaker’s Bureau and is a University of Texas diabetic foot group is a derivation of
consultant for Pfizer & Merck. the Wagner scheme.6 This system includes infection as a
This article is being co-published in the Journal of Vascular Surgery® and the comorbidity modifier that can be added to any wound
Journal of the American Podiatric Medical Association.
depth category, but it does not specify the severity of the
Reprint requests: Dr Warren S. Joseph, 1109 Old Ford Rd., Huntingdon
Valley, PA 19006 (e-mail: wsjoseph@comcast.net). infection.
The editors and reviewers of this article have no relevant financial relationships In order to address the lack of an adequate severity
to disclose per the JVS policy that requires reviewers to decline review of any score for DFI, the IWGDF and the IDSA have each devel-
manuscript for which they may have a competition of interest. oped a mostly interchangeable, four level grading system
0741-5214/$36.00
Copyright © 2010 by the Society for Vascular Surgery and the American
specifically for infection (Table I). The presence of infec-
Podiatric Medical Association. tion in a foot wound is defined clinically, not by a positive
doi:10.1016/j.jvs.2010.06.010 culture. Ulcerations are classified as either uninfected (lack-
67S
JOURNAL OF VASCULAR SURGERY
68S Joseph and Lipsky September Supplement 2010

Table I. Clinical classification schemes (Infectious Diseases Society of America [IDSA] and International Working
Group on the Diabetic Foot [IWGDF]) for a diabetic foot infectiona

IDSA infection IWGDF PEDIS


Clinical manifestations of infection severity gradeb

Wound lacking purulence or any manifestations of inflammation Uninfected 1


Presence of purulence or ⱖ2 manifestations of inflammation (pain, tenderness, erythema, Mild 2
warmth, or induration), but any cellulitis/erythema extends ⱕ2 cm around the ulcer,
and infection is limited to the skin or superficial subcutaneous tissues; no other local
complications or systemic illness.
Infection (as defined above) in a patient who is systemically well and metabolically stable, Moderate 3
but which is associated ⱖ1 of the following characteristics: cellulitis extending ⬎2 cm,
lymphangitic streaking, spread beneath the superficial fascia, deep-tissue abscess,
gangrene, and involvement of muscle, tendon, joint, or bone.
Infection in a patient with signs or symptoms of systemic toxicity (eg, fever, chills, Severe 4
leukocytosis) or metabolic instability (eg, tachycardia, hypotension, confusion,
vomiting, acidosis, severe hyperglycemia, or azotemia)
a
The presence of critical ischemia of the affected limb often increases the severity of the infection.
b
International Consensus on the Diabetic Foot PEDIS system: perfusion, extent/size, depth/tissue loss, infection, and sensation.

Table II. Suggested antibiotic regimens for diabetic foot infections, based on clinical severity

Agent(s) Mild Moderate Severe

Advised route Oral for most Oral or parenteral, based on clinical Intravenous, at least
situation and agent(s) selected initially
Dicloxacillin 公
Clindamycin 公
Cephalexin 公
Trimethoprim-sulfamethoxazole 公
Amoxicillin/clavulanate 公 公
Levofloxacin 公 公
Cefoxitin 公
Ceftriaxone 公
Ampicillin/sulbactam 公
Linezolida ⫾ aztreonam 公
Ertapenem 公
Cefuroxime ⫾ metronidazole 公
Ticarcillin/clavulanate 公
Piperacillin/tazobactam 公 公
Levofloxacin or ciprofloxain ⫹ clindamycin 公 公
Imipenem-cilastatin 公
Vancomycina ⫹ ceftazidime ⫾ metronidazole 公
Similar agents of the same class may be substituted. Some of these regimens may not have FDA approval for complicated skin and skin structure infections, and
only linezolid is currently specifically approved for diabetic foot infections.
Reprinted from Lipsky et al.2
a
Where methicillin-resistant Staphylococcus aureus infection is proven or likely.

ing clinical evidence of purulence or inflammation) or of a wound, with infection severity (eg, “the patient pre-
infected, with a severity of mild, moderate, or severe. The sents with a University of Texas 2B wound with an IDSA
differentiation between these categories is based on clinical category moderate infection”). One of the benefits of the
signs and symptoms, thus corroborating the importance of infection severity scheme is that it can be used to help select
making a clinical diagnosis of infection, as spelled out in an appropriate antibiotic regimen for DFIs, as shown in
point #5 of the Executive Summary described below. This Table II.
scale has been validated in a prospective observational study
published by Lavery et al.3 When applied to 1666 patients EXECUTIVE SUMMARY REVIEW
with a foot ulceration, there was a significant increase in The IDSA DFI guidelines contained a 19-point Exec-
rates of hospitalization and lower extremity amputation utive Summary with the key recommendations for diagnos-
with increased severity of the infection. These classification ing and treating these infections. The recommendations
systems have now been widely accepted and are being still remain valid and are critical to a successful approach to
employed in academic and clinical circles. Some clinicians managing DFIs. We will address a few of the most impor-
have taken to combining systems that rate the size/depth tant recommendations numbering them as they were in the
JOURNAL OF VASCULAR SURGERY
Volume 52, Number 12S Joseph and Lipsky 69S

2004 guidelines, to make for easier reference to the original diagnosed infection in order to aid in antibiotic selec-
document. The original text is in bold italicized print, tion. A recent IWGDF consensus statement on dia-
followed by our commentary. betic foot osteomyelitis suggests that culture-based
diagnosis and treatment are likely the best predictors of
2. Diabetic foot infections require attention to local a positive outcome.7 Even in cases of osteomyelitis, the
(foot) and systemic (metabolic) issues and coordinated role of diagnostic imaging studies has been recently
management, preferably by a multidisciplinary foot- challenged. Currently, magnetic resonance imaging is
care team. The team managing these infections should the best imaging study for defining both soft tissue and
include, or have ready access to, an infectious diseases bone infections.8 Other studies have been investi-
specialist or a medical microbiologist. Apropos this gated, however, with positron emission tomography
special issue jointly published by the American Podiat- (FDG-PET) combined with computed tomography
ric Medical Association and the Society for Vascular scanning being the most promising.9
Surgery, the concept of a multidisciplinary team ap- 9. Available evidence does not support treating clini-
proach cannot be overstated. The team requires mem- cally uninfected ulcers with antibiotic therapy. Anti-
bers trained in the skill set necessary for limb salvage biotic therapy is necessary for virtually all infected
and includes those concentrating on not only the wounds, but it is often insufficient without appropri-
anatomic disturbances of the lower extremity but also ate wound care. Inappropriate use of antibiotics leads
the perfusion to the limb. This approach (ie, care for to bacterial resistance. This observation has gained
the diabetic foot wound by at least an irreducible team greater importance with the development of resistance
of a foot specialist and a vascular surgeon) is aptly of the common gram-positive pathogens, especially
named the “toe and flow” concept (Armstrong, DA MRSA.10,11 An increasing threat of treating soft tissue
personal communication), an issue recognized in #14 infections, including DFI, is the varying levels of tol-
below. erance of formerly susceptible microorganisms to van-
4. Aerobic gram-positive cocci (especially Staphylococcus comycin.12 This can take the form of just increasing
aureus) are the predominant pathogens in diabetic minimum inhibitory concentrations (“MIC creep”) or
foot infections. Patients who have chronic wounds, or frank resistance (vancomycin-resistant S. aureus, or
who have recently received antibiotic therapy, may VRSA). In fact, the first two reported cases of VRSA13
also be infected with gram-negative rods, and those back in 2002 were isolated from diabetic foot wounds.
with foot ischemia or gangrene may have obligate anaer- With advances in the agents available for antibiotic
obic pathogens. It has become increasingly clear that treatment of these resistant gram-positive organisms
Staphylococcus aureus, including methicillin-resistant has come a newer threat (eg, multi-drug resistant
strains (MRSA), and Streptococcus spp, most often Group (MDR) gram-negative bacteria). Organisms, such as
B, are the primary pathogens in most DFI. A number of species of Klebsiella and Proteus, once relatively suscep-
studies have shown that antibiotic therapy directed at just tible to a wide range of antibiotics, now may produce
these organisms can be successful, even when culture extended-spectrum beta-lactamases (ESBL) or carbap-
results demonstrate various other gram-negative bacteria enemases, rendering them resistant to most commonly
or obligate anaerobes. This has led to the promulgation used drugs.14 A major reason for the emergence of
of the “Head of the Snake” concept—the aerobic gram- these resistant organisms is inappropriate, typically un-
positive bacteria constitute the crucial head of the snake, necessary, and overlong, antibiotic treatment.
while other isolated organisms are part of the body. If 10. Select an empirical antibiotic regimen on the basis of
antibiotic therapy is directed at the primary organisms, the severity of the infection and the likely etiologic
thereby “cutting off the head” of the snake, the residual agent(s). Therapy aimed solely at aerobic gram-
“body” would die. Thus, antibiotic therapy can often be positive cocci may be sufficient for mild-to-moderate in-
narrowly targeted at only aerobic gram-positive cocci. fections in patients who have not recently received
5. Wound infections must be diagnosed clinically on the antibiotic therapy. Broad-spectrum empirical therapy
basis of local (and occasionally systemic) signs and is not routinely required but is indicated for severe
symptoms of inflammation. Laboratory (including infections, pending culture results and antibiotic sus-
microbiological) investigations are of limited use for ceptibility data. Take into consideration any recent
diagnosing infection, except in cases of osteomyelitis. antibiotic therapy and local antibiotic susceptibility
This point represents a cultural shift away from oft- data, especially the prevalence of methicillin-resistant
practiced technology- (mostly imaging) based ap- S. aureus (MRSA) or other resistant organisms. De-
proach to diagnosis. In order to diagnose infection in finitive therapy should be based on both the culture
a diabetic foot wound, the patient should present with results and susceptibility data and the clinical re-
at least two of the clinical signs and symptoms of sponse to the empirical regimen. The previously dis-
inflammation including; erythema, edema, warmth, cussed “head of the snake” paradigm addresses initiat-
tenderness, induration, or pain. Cultures should be ing antibiotic therapy aimed principally (if not solely)
obtained, not to diagnose the infection, but rather to at aerobic gram-positive organisms for many DFI.
determine which organism is causing the clinically Even the 2004 Guidelines acknowledged the increas-
JOURNAL OF VASCULAR SURGERY
70S Joseph and Lipsky September Supplement 2010

ing prevalence of MRSA, but the problem has clearly limb salvage. A recent consensus statement from the
worsened in most parts of the world. At the time the IWGDF7 classified DFI surgery as being “urgent” only
Guidelines were prepared, clinical trials specifically in cases of necrotizing or deep soft tissue infections,
looking at antimicrobial management of DFI were major abscess, or gangrene. “Nonurgent” surgery may
sparse, and rates of MRSA isolation were 10% or less. be needed when there is risk of significant compromise
More recent work has shown that these rates are in- of the soft tissue envelope, loss of mechanical function,
creasing.15 In some institutions, MRSA has supplanted or if the degree of bone involvement is life- or limb-
methicillin-susceptible S. aureus (MSSA) as the pri- threatening. However, none of this “toe” surgery can
mary pathogen isolated in DFI. This raises the impor- be accomplished without adequate “flow” to the limb.
tant question: When should empiric MRSA coverage Likewise, antibiotic therapy of the “toe” infection
be instituted? As stated in the summary point, the needs adequate “flow” to achieve therapeutic levels (of
decision to cover MRSA should be based on knowl- antibiotic and leukocytes) in the target infected tissues.
edge of local prevalence, but at what “breakpoint” is
Thus, a full circle is completed from the first executive
the decision made? Unfortunately, there are no hard
summary review point to the last. A team approach is
(or easy) and fast rules. The severity of the infection
required, with specialists who are interested in foot
along with the risk tolerance of the clinician should
infections and who possess the necessary skill set to
each probably play a role. For mild infections, clini-
succeed at diabetic limb salvage. Optimal outcomes for
cians may tolerate a higher local prevalence or risk of
patients presenting with DFIs have repeatedly been
MRSA infection before initiating empiric MRSA ther-
shown to depend on care by a team of specialists, includ-
apy, since the patient may do well with local wound
ing those with training in podiatric and vascular surgery
care regardless of whether or not the organism is
along with infectious diseases, diabetes, wound care, and
covered. However, for moderate to severe infections,
internal medicine. To this end, the IDSA DFI Commit-
the threshold for initiating anti-MRSA therapy should
tee has been expanded to include stakeholders from
be lower (eg, an expected prevalence of ⬎10%) given
specialties across the spectrum of diabetic foot care,
the potential for serious sequelae if aggressive antibi-
including podiatric surgery and vascular surgery. All of
otic therapy is not initiated. This has given rise to the
these voices will be included in the 2010 update.
concept of “escalation” or “de-escalation” therapies
for empiric MRSA coverage. If the organism is not
empirically covered, the clinician can always “escalate” REFERENCES
(broaden the spectrum) the choice of therapeutic 1. Lipsky BA. A report from the international consensus on diagnosing and
agents to include MRSA once initial culture results are treating the infected diabetic foot. Diabetes Metab Res Rev 2004;20
received. Likewise, if there is a greater concern that it (Suppl 1):S68-77.
2. Lipsky BA, Berendt AR, Deery HG, Embil J, Joseph WS, Karchmer
may be present, broader spectrum empiric coverage is
AW, et al. Infectious Disease Society of America. Diagnosis and treat-
begun with the option of “de-escalation” (narrowing ment of diabetic foot infections. Clin Infect Dis 2004;39:885-910.
the spectrum) if cultures reveal only susceptible organ- 3. Lavery LA, Armstrong DG, Murdoch DP, Peters EJ, Lipsky BA. Vali-
isms. The availability of several new anti-MRSA anti- dation of the Infectious Diseases Society of America’s diabetic foot
biotic agents has made covering this organism easier infection classification system. Clin Infect Dis 2007;44:562-5.
4. Rao N, Lipsky BA. Optimising antimicrobial therapy in diabetic foot
than it was a decade ago.16 infections. Drugs 2007;67:195-214.
14. Seek surgical consultation and, when needed, inter- 5. Van Acker K, De Block C, Abrams P, et al. The choice of diabetic foot
vention for infections accompanied by a deep abscess, ulcer classification in relation to the final outcome. Wounds 2002;14:
extensive bone or joint involvement, crepitus, substan- 16-25.
6. Armstrong DG, Lavery LA, Harkless LB. Validation of a diabetic
tial necrosis or gangrene, or necrotizing fasciitis.
wound classification system. The contribution of depth, infection, and
Evaluating the limb’s arterial supply and revascular- ischemia to risk of amputation. Diabetes Care 1998;21:855-9.
izing when indicated are particularly important. 7. Berendt AR, Peters EJ, Bakker K, Embil JM, Eneroth M, Hinchliffe RJ,
Surgeons with experience and interest in the field et al. Diabetic foot osteomyelitis: a progress report on diagnosis and a
should be recruited by the foot-care team, if possible. systematic review of treatment. Diabetes Metab Res Rev 2008;24
(Suppl 1):S145-61.
This final point is at the heart of the entire “toe and 8. Teh J, Berendt T, Lipsky BA. Rational imaging. Investigating suspected
flow” concept. By the definition used by the US Food bone infection in the diabetic foot. BMJ 2009;339:b4690.
and Drug Administration, a complicated skin and skin 9. Nawaz A, Torigian DA, Siegelman ES, Basu S, Chryssikos T, Alavi A.
structure infection (cSSSI) is one in which the patient Diagnostic performance of FDG-PET, MRI, and plain film radiography
(PFR) for the diagnosis of osteomyelitis in the diabetic foot. Mol
either has a significant underlying medical comorbidity
Imaging Biol 2010;12:335-42.
or an immunocompromising disease; in these cases, 10. Bowling FL, Jude EB, Boulton AJ. MRSA and diabetic foot wounds:
the potential for greater severity of the infection may contaminating or infecting organisms? Curr Diab Rep 2009;9:440-4.
require more aggressive surgical approaches (eg, inci- 11. Dang CN, Prasad YD, Boulton AJ, Jude EB. Methicillin-resistant
sion and drainage or resection of infected and necrotic Staphylococcus aureus in the diabetic foot clinic: a worsening problem.
Diabet Med 2003;20:159-61.
tissue). Although appropriate antibiotic therapy can be 12. Gould IM. Clinical relevance of increasing glycopeptide MICs
critical in achieving optimal outcomes, for many mod- against Staphylococcus aureus. Int J Antimicrob Agents 2008;31
erate to severe DFI, surgical intervention can be key to (Suppl 2):1-9.
JOURNAL OF VASCULAR SURGERY
Volume 52, Number 12S Joseph and Lipsky 71S

13. Centers for Disease Control and Prevention (CDC). Vancomycin- 15. Lipsky BA, Tabak YP, Johannes RS, Vo L, Hyde L, Weigelt JA. Skin and
resistant Staphylococcus aureus–Pennsylvania, 2002. MMWR Morb soft tissue infections in hospitalised patients with diabetes: culture
Mortal Wkly Rep 2002;51:902. isolates and risk factors associated with mortality, length of stay and
14. Varaiya A, Dogra J, Kulkarni M, Bhalekar P. Extended spectrum beta cost. Diabetologia 2010;53:914-23.
lactamase (ESBL) producing Escherichia coli and klebsiella pneu- 16. Nicolau DP, Stein GE. Therapeutic options for diabetic foot infections:
moniae in diabetic foot infection. Indian J Med Microbiol 2008;26: a review with an emphasis on tissue penetration characteristics. J Am
281-2. Podiatr Med Assoc 2010;100:52-63.