European Review for Medical and Pharmacological Sciences 2019; 23(2 Suppl.

): 26-37

Diabetic foot infections: a comprehensive

overview
D. PITOCCO1, T. SPANU2, M. DI LEO1, R. VITIELLO3, A. RIZZI1,
L. TARTAGLIONE1, B. FIORI2, S. CAPUTO1, G. TINELLI4, F. ZACCARDI1,
A. FLEX5, M. GALLI3, A. PONTECORVI6, M. SANGUINETTI2
1
Diabetes Care Unit, Endocrinology, University Hospital “A. Gemelli”, Catholic University of the
Sacred Heart, Rome, Italy
2
Institute of Microbiology, University Hospital “A. Gemelli”, Catholic University of the Sacred Heart,
Rome, Italy
3
Institute of Orthopedic Surgery, Internal Medicine, University Hospital “A. Gemelli”, Catholic
University of the Sacred Heart, Rome, Italy
4
Institute of Vascular Surgery, University Hospital “A. Gemelli”, Catholic University of the Sacred
Heart, Rome, Italy
5
Institute of Internal Medicine, University Hospital “A. Gemelli”, Catholic University of the Sacred
Heart, Rome, Italy
6
Institute of Endocrinology, University Hospital “A. Gemelli”, Catholic University of the Sacred
Heart, Rome, Italy

Abstract. – Diabetic foot ulcers (DFUs), a mi- ity of life, and they give rise to a need for visits
cro-vascular complication, are associated with by health-care providers, wound care, antimicro-
a substantial increase in morbidity and mortal- bial therapy, and often surgical procedures/de-
ity. DFUs are a complicated mixture of neurop-
athy, peripheral arterial diseases, foot deformi-
bridements. As such, these infections comprise the
ties, and infections. Foot infections are frequent most frequent grounds for both diabetes-associat-
and potentially devastating complications. In- ed hospitalization and lower extremity losses,. In
fection prospers in more than half of all foot ul- an acute presentation with diabetic foot infection
cers and is the factor that most often leads to (DFI), there is frequently a delay in the recognition
lower extremity amputation. The complications of the causative organism, which may compel the
of microbial flora span the spectrum from su-
use of empirical antibiotics,. According to Peters,
perficial cellulitis to chronic osteomyelitis and
gangrenous extremity lower limb amputations. the incidence of foot infections in people with dia-
Wounds without confirmed soft tissue or bone betes ranges from an overall lifetime risk of 4% to
infections do not require antibiotic therapy. Mild a yearly risk of 7%. If infection advances to deeper
and moderate infections need empiric therapy structures, including the underlying bone, diabet-
covering Gram-positive cocci, while severe in- ic foot osteomyelitis (DFO) develops. DFIs are the
fections caused by drug-resistant organisms most frequent diabetes-related complication re-
require broad-spectrum anti-microbials target-
ing aggressive Gram-negative aerobes and obli-
quiring hospitalization, and DFO is present in 44-
gate anaerobes. 68% of patients with DFIs admitted to the hospital.
Infections in foot lesions should be clinically
Key Words: defined by the presence of inflammation or pu-
Diabetes mellitus, Diabetic foot ulcer, Infection, Am- rulence, and then classified by severity. This
putation, Antibiotics, Mmicrobiology. approach helps clinicians make decisions about
which patients to hospitalize, send for imaging
procedures or recommend for surgical interven-
tions. Many organisms, alone or in combination,
Introduction can cause DFIs, but Gram-positive cocci (GPC),
especially staphylococci, are the most common.
Infections in ulcerated feet in patients with dia- To achieve more successful outcomes and ulti-
betes are a primary cause of morbidity, including mately avoid amputations, a systematic approach
discomfort, and reduced physical and mental qual- to the management of DFIs must be adopted. If

26 Corresponding Author: Dario Pitocco, MD; e-mail: [email protected]

 Diabetic foot infections: a comprehensive overview

not treated promptly and appropriately, DFI can amputation is needed, a high-level (i.e., transtib-
become incurable or even lead to septic gangrene. ial) procedure is often indicated, more because
At least 60% of non-traumatic lower limb ampu- of irreversible ischaemia than because of uncon-
tations occur among people with diabetes. trolled infection. However, most amputations
The existence of osteomyelitis further raises reflect the multimodal foot problems related to
the costs of hospitalization because of the need diabetes, which highlights the need for a multidis-
for additional diagnostic studies, prolonged med- ciplinary approach (Figure 1). All clinicians reg-
ical treatment and surgeries. Notably, the use of ularly seeing persons with diabetes should have
antibiotics in such cases is at least doubled. When an understanding of how to prevent, diagnose

Figure 1. Multimodal foot problems related to diabetes and the need for a multidisciplinary approach.

27
 D. Pitocco, T. Spanu, M. Di Leo, R. Vitiello, A. Rizzi, et al

and treat DFIs. Given the increasing amount of is also connected to a damage of collateral cir-
research in this area, this review aims to make cli- culation. This is considered to be an atheroscle-
nicians aware of recent developments in this field. rotic obstructive disease of large vessels, which
leads to peripheral arterial disease of the lower
Pathogenesis of Diabetic Foot Infection extremities. Little is known about the biology of
The foot is the crossroad for many pathological peripheral arterial disease (PAD) in individuals
processes in diabetics and it is an area in which with diabetes, but it is believed that the vascular
almost all gears of the lower limb are involved: changes observed with other manifestations of
skin, subcutaneous tissue, muscles, bones, joints, atherosclerotic disease are also applicable to pa-
nerves, and blood vessels. DFI is more often the tients with both peripheral arterial disease and di-
consequence than the cause of diabetic foot ul- abetes. Peters et al suggest that previous amputa-
cers. These infections usually begin with a split tion, peripheral vascular disease, and neuropathy
in the cutaneous envelope, typically in a site of are significant risk factors for DFI. Microangiop-
trauma (mechanical/ thermal) or ulceration. In- athy results in capillary basal membrane thick-
fection is best defined as an invasion by micro-or- ening, altered nutrient exchange, tissue hypoxia
ganisms and their multiplication in host tissues and microcirculation ischemia. There is evidence
induces inflammatory responses. This is followed of diabetic foot from as early as ancient Egyptian
by tissue destruction. DFI is defined by infection times, as mummies with prosthetic toes have been
in soft tissue or bone anywhere below the malleo- discovered. Pryce reported a case of a foot ulcer
li in a diabetic person. Several factors predispose associated with diabetes in 1887.
diabetic patients to developing a DFI, including A lack of attention to foot hygiene and the use
neuropathy, vasculopathy, immunopathy, and foot of poorly fitting footwear are the major factors that
biomechanics. are preventable in the development of infection.
Sensory loss due to peripheral neuropathy in Diabetic foot infection may range from fungal in-
the diabetic foot is always considered to be the fections of the nail to severe necrotising limb- or
earliest developed and most prominent threat, and life-threatening infections. Early diagnosis and
features in the development of ulcers. About 60% prompt definitive treatment may be delayed due to
of diabetic patients with foot ulcers have neurop- a lack of foot sensation, the patient’s poor eyesight,
athy. Nerve dysfunction in diabetic patients may and poor judgement by the physician. Abrasions,
be described as sensory, motor, or autonomic. The rashes and loss of skin integrity can be the initiat-
lack of balances in the musculature of the foot due ing factors in the development of diabetic foot in-
to motor neuropathy result in atrophy with mus- fection. Approximately 60% of foot infections start
cle wastage, dislocation of fat pads and associ- in webbed spaces and 30% in nails, while 10% are
ated foot deformities, such as foot drop, clawed secondary to punctures. In a diabetic, the clinical
and hammerhead toes, and equinus deformities, presentation ranges from acute cellulitis to life
creating areas susceptible to trauma. As a conse- threatening necrotising fasciitis.
quence of the reduced sensation, insults to lower Debridement must be meticulous and repeated
extremities often go unnoticed, which progres- debridement is often necessary. Treatment prior-
sively worsens the state, as the affected lesion is ities are: 1) aggressive treatment of infections, 2)
subjected to repetitive plantar pressure and shear diagnosis of ischemia and evaluation for possi-
forces from ambulation and weight bearing that ble revascularization, 3) relief of pressure on the
damage the sensory nerves of extremities. With wound and 4) improvement of the wound envi-
the loss of sweat and oil gland functions, the di- ronment with debridement dressing and advanced
abetic foot becomes dry and keratinized. Auto- care treatments. Treatment of a complicated dia-
nomic neuropathy leads to sudomotor function betic foot ulcer can involve numerous pathways.
and abnormal blood flow to the soles of the feet. After a complete ulcer evaluation, including mea-
With displacement in functions of the foot’s sweat surements, x-rays and fundoscopy, an arterial
and sebaceous glands, the skin becomes dry and Doppler ultrasound can be utilized.
keratinizes, so that it more easily cracks, generat- Immunopathy has been implicated in the di-
ing a portal for infection,. abetic patient’s inherent susceptibility to infec-
Diabetic angiopathy is the most frequent cause tion as well as the potential to mount a normal
of morbidity and mortality in diabetic patients. inflammatory response. Impaired host defences
Macroangiopathy reveals as diffuse multi-seg- secondary to hyperglycaemia include defects in
mental involvement of the lower limb vessels. It leukocyte function and morphologic changes to

28
 Diabetic foot infections: a comprehensive overview

macrophages. Bagdade et al demonstrated that luses to fully visualize the wound. The diag-
leukocyte phagocytosis was significantly reduced nosis of infection is based on the presence of
in patients with poorly controlled diabetes, and purulence, or at least two classic symptoms or
that an improvement in microbiocidal rates was signs of inflammation (e.g., erythema, edema,
directly correlated with the correction of hyper- warmth, tenderness, pain or induration). How-
glycaemia. Decreased chemotaxis of growth ever, in some cases, patients with diabetes may
factors and cytokines, coupled with an excess of have a dull neuro-inflammatory response, such
metalloproteinases, impede normal wound heal- that they do not manifest typical signs of infec-
ing by creating a prolonged inflammatory state. tion. Secondary signs in cases of neuropathic
Fasting hyperglycaemia and the presence of an foot include friable or discoloured granulation
open wound create a catabolic state. A negative tissue, a foul odour, non-purulent discharges,
nitrogen balance ensues secondary to insulin and delayed wound healing. A proper emphasis
deprivation, caused by gluconeogenesis from should be placed on evaluating the risk factors of
protein breakdown. This metabolic dysfunction DFI, including positive probe-to-bone (PTB), the
impairs the synthesis of proteins, fibroblasts and presence of an ulceration for more than 30 days,
collagen, and further systemic deficiencies are a history of recurrent foot ulcers, a traumatic eti-
propagated, which lead to nutritional compro- ology, the presence of peripheral arterial disease
mise. A research indicates impairment of the im- in the involved limb, a history of lower extremity
mune system at serum glucose levels ≥150 ml/dl. amputation, a lack of protective sensations, renal
Patients with diabetes tolerate infections poorly insufficiency or a history of walking barefoot.
and infections adversely affect diabetic control. An adequate description of ulcer characteristics,
This repetitive cycle leads to uncontrolled hy- such as size, depth, base, margins, appearance,
perglycaemia, which further affects the host’s re- and location, is necessary for mapping progress
sponse to infection. during treatment. A thorough assessment of the
presence of granulation tissue or slough should
Foot Architecture be made in the floor of the ulcer to determine
The distinctive framework of the foot, which subsequent management. Patients with a diabetic
has several interconnected bony compartments, foot infection should also be properly evaluated
favours the spread of infection via proximal cal- for arterial insufficiency and neuropathic condi-
caneal convergence or direct perforation of septae. tions on a structured schedule based on defined
In addition, the soft tissues of the foot, like plantar risk factors. The presence of fever, tachycardia or
tendons, aponeurosis, muscles sheaths, and fascia, tachypnea may indicate an infected wound. The
cannot resist infections. Infection of the bone is an vascular status should be documented by palpat-
outcome of the contiguous spreading of infection ing all peripheral pulses, or by using a hand-held
to cortex (osteis) or bone marrow (osteomyelitis). Doppler for non-palpable or faint pedal pulses.
A sterile metal probe is inserted into the ulcer. If ABI is a common non-invasive tool used in diag-
it penetrates to the bone, it nearly always indicates nosing PAD, but false elevations due to calcified
the presence of infected bone. Ulcers >60 mm2 arteries warrant more vascular studies. Neuro-
in size, chronic discharge from the sinus tract, an logical examinations are also needed to clinical-
erythrocyte sedimentation rate >70 mm/hour, or ly manage diabetic foot ulcers and healing.
the presence of sausage toe, suggest the presence
of underlying osteomyelitis. Plain radiographs are Microbiology of Diabetic Foot Ulcers
a cost-effective method for confirming osteomyeli- The management of diabetic foot disease pri-
tis. Other sensitive techniques include CT scans, marily focuses on avoiding amputation of the
MRI, and radioisotope scans providing high-reso- lower extremities. The basic principles of wound
lution images of bone and soft tissues,. Some inde- healing apply equally to DFU patients and pa-
pendent risk factors for DFI include wounds that tients with wounds at any other site. The heal-
penetrate bones, recurrent lesions, and a history of ing of DFUs will occur in the presence of three
amputations, neuropathy, wounds with traumatic conditions: adequate arterial inflow, appropriate
etiologies, and the presence of PAD. control of infection, and the offloading of the
wound site and the immediate surrounding area
Assessment of Infection infection is defined as invasion and colonization
Clinical assessment requires appropriate de- by pathogenic microbes in a foot wound, which
bridement to remove necrotic sections and cal- causes local tissue damage favoured by hyper-

29
 D. Pitocco, T. Spanu, M. Di Leo, R. Vitiello, A. Rizzi, et al

glycaemia-mediated deranged host defences. Staphylococcus aureus among the aerobes and
Infections begin as minor problems and later Peptostreptococcus spp among the anaerobes.
progress to involve deep tissues, joints, or bones, Infection in previously untreated DFUs is
especially if unmanaged. caused by Gram-positive cocci, mostly in a mo-
Exploring the microbial etiology is an import- no-microbial state, whereas chronic or severely
ant aspect of DFI management. Foot wounds in infected lesions harbour polymicrobial strata with
diabetics lack many of the protective barriers and a mix of Gram-negative aerobes and anaerobes.
mechanisms associated with intact skin, there- Breen et al demonstrated that Staphylococcus au-
by providing a portal for invasive microorgan- reus is the most important pathogen in DFIs and a
isms. The presence of non-replicating pathogens component of polymicrobial etiology. Among the
is termed “contamination”, while wounds with Gram-negative group, Escherichia coli, Klebsiel-
fast-dividing microbes are “colonized”. Criti- la pneumoniae, and Proteus species are the most
cal colonization or a state of transition between common pathogens, followed by Pseudomonas
colonization and invasion delays wound healing, aeruginosa. Chronic or previously treated wounds
and altered microbial-host interactions increase harbour Gram-negative bacilli, especially from
virulence. Diffused immune responses accelerate the Enterobacteriace family. Wounds involving
the process. When a colonized wound progresses deep tissues or ischemic necrosis are invaded by
to an infected wound, a microbiological analysis obligate anaerobes. Pseudomonas infections are
should be undertaken to evaluate the underlying common in wounds soaked in wet dressings and
pathogen(s). Management of a clinically overt di- frequently seen in warmer regions. In foot ulcers,
abetic foot infection requires apposite systemic the methicillin-resistant staphylococcus aureus
antibiotic therapy, which is best guided by iden- (MRSA) was high but MSRA was eradicated by
tifying the causative pathogens. Proper specimen regular debridement and topical treatments.
collection, such as deep-tissue samples, can reveal Our preliminary study showed that the
the true flora and are preferred over wound swabs, Gram-positive cocci are the most common patho-
as the latter may reveal colonizing agents and pro- gens, and we observed that the percentage of
vide false results. A curettage or tissue scraping MRSA was extremely high at 59.4% (Table I). We
from the base of the ulcer provides a more accu- found that the six most common microbial species
rate result if promptly sent for aerobic and anaer- constitute almost 70% of diabetic foot infections
obic analysis. Swab or tissue specimens should be registered at our clinic (Figures 2 and 3). These
evaluated for phenotypic testing in line with CLSI microbial species showed the following rates of
guidelines. This can be attained through a culture resistance. For S. aureus, we detected a rate of
of a specimen using selective or standard growth resistance to oxacillin of 59.4% (MRSA), 65.9%
media along with antimicrobial sensitivity testing. to amoxicillin/clavulanic, 64.1% to ciprofloxacin,
Traditional microscopy and staining techniques, 0.6% to teicoplanin (n=2) and 0.4% to vancomy-
such as the Gram-stained smear, can provide ad- cin (n=1). We found no resistance to linezolid. For
ditional organism characterization. Disadvantag-
es of these techniques include the fact that they
take at least a couple of days to process, they miss Table I. A total 765 episodes of DFI in 482 adult patients were
some facultative organisms, and they are less use- identified.
ful in patients undergoing antibiotic therapy. Expected pathogens Frequency Percentage
Longstanding DFUs with severe infections are
usually polymicrobial. In a clinico-microbiolog- S. aureus 335 26.8
ical study of 80 diabetic foot patients by Gade- Ent. faecalis D 166 13.3
P. aeruginosa 164 13.1
palli et al32, 82.5% demonstrated polymicrobial E. coli 85 6.8
flora with an average of 2.3 species per patient P. mirabilis 65 5.2
and an aerobic to anaerobic ratio of 5.5. The B. fragilis 49 3.9
most commonly isolated pathogens were Staph- A. baumannii 44 3.5
ylococcus aureus, Proteus spp, and Escherich- K. pneumoniae 41 3.3
Ent. faecium D 34 2.7
ia coli. Among anaerobes, Peptostreptococcus S. agalactiae B 25 2.0
spp, Veilonella species, and Bacteroides species E. cloacae 21 1.7
were predominant. In another study, Zubair et al S. haemolyticus 21 1.7
reported polymicrobial etiology in 65% cases of Morganella morganii 19 1.5
DFI with a predominance of Escherichia coli and S. maltophilia 12 1.0

30
 Diabetic foot infections: a comprehensive overview

Figure 2. Overall antibiotic resistance.

P. aeruginosa, no antibiotics with a sensitivity fotazidime, and 1.2% to amikacin. For B. fragilis,
of 100% were observed. We found the following we observed resistance of 21.4% to piperacillin,
rates of resistance to other antibiotics: 58.5% to but no resistance to amoxicillin/clavulanate, imi-
ciprofloxacin, 44.5% to gentamicin, 34.1% to ce- penem or meropenem.
fotazidime, 39.1% to imipenem, 28.2% to mero- The prevention of ulcers infected by multi-
penem, and 18.3% to amikacin. For P. mirabilis, drug resistant organisms (MDRO) should be in
no resistance to meropenem was observed, while focus and resistance patterns should be careful-
we detected the following rates of resistance to ly monitored. A mild or moderate DFI can be
other antibiotics: 79.7% to ampicillin, 60.9% to treated with oral antimicrobials, while chronic
ciprofloxacin, 45.2% to cefotaxime, 42.6% to infection requires inpatient antimicrobial therapy
amoxicillin/clavulanate, 37.5% to cefotazidime, or surgical treatment, as well as controlled met-
and 14.1% to amikacin. For E. faecalis D, no re- abolic derangements. Patients with DFI should
sistance to linezolid was observed. We detected initially be treated with an empirical regime cov-
the following rates of resistance to other antibi- ering Gram-positive cocci. The spectrum can be
otics: 4.3% to vancomycin, 4.2% to teicoplanin, broadened to cover Gram-negative aerobes in
and 3% to ampicillin. For E. coli, no resistance chronic infections. Wounds that are necrotic, foul
to imipenem and meropenem was observed, but smelling or gangrenous require anti-anaerobic
we detected the following rates of resistance to microbials. A definite diagnosis of bone infection
other antibiotics: 80% to ampicillin, 69.4% to cip- usually requires histological findings consistent
rofloxacin, 25% to amoxicillin/clavulanate, 37.6% with bone infection along with microbiological
to cefotaxime, 35.3% to gentamicin, 31.8% to ce- examinations of an aseptically obtained bone

31
 D. Pitocco, T. Spanu, M. Di Leo, R. Vitiello, A. Rizzi, et al

Figure 3. Prevalence (%) of antibiotic resistance, by year.

sample. However, this is typically only necessary it was devised for dysvascular foot, it has been
when the diagnosis is in doubt or determining the used for lesion classification for the past 25 years.
causative pathogen’s antibiotic susceptibility is This six-grade classification system takes into
crucial. The primary treatment for bone infection consideration the depth of the ulcer, the presence
should be parenteral and can be prolonged up to of gangrene, and the extent of tissue necrosis.
six weeks. Chronic osteomyelitis requires surgi- Even though Wagner-Meggit’s grading is one
cal intervention (i.e., removal of bone). of the most widely used classification systems,
it does not take into account important clinical
Classification of Diabetic Foot Infections parameters, such as ischemia, infection, or other
An adequate description of ulcer characteris- co-morbid factors (Table II).
tics, such as size, depth, appearance, and location, The University of Texas system primarily grades
allows for mapping of progress during treatment. ulcers based on depth, and then, it stages which di-
The evaluation should determine the etiology of vide patients who have clean ulcers and those who
the ulcer and ascertain whether the lesion is neu- are infected. More specifically, grade 0 in the Texas
ropathic, ischemic or neuro-ischemic. Various System classification (Table III) represents a pre-
classification systems have been used to evaluate or postulcerative site. Grade 1 ulcers are superficial
the severity of diabetic foot lesions. These sys- wounds through either the epidermis or the epider-
tems attempt to encompass different characteris- mis and dermis, but that do not penetrate to ten-
tics of the ulcer, including size, depth, ischemia, don, capsule, or bone. Grade 2 wounds penetrate to
infection, and neuropathy. tendon or capsule, but the bone and joints are not
One of the most commonly used classification complicated. Grade 3 wounds infiltrate bone or in-
systems is the Wagner-Meggit system. Although to a joint. Each wound grade consisted of 4 stages:

32
 Diabetic foot infections: a comprehensive overview

Table II. Wagner classification system. ischemic, or neuroischemic, determined by how

0 Pre-ulcerative, with no open lesion or cellulitis complications such as peripheral neuropathy and ar-
1 Superficial ulcer terial disease affect the ulcer’s etiology.
2 Deep ulcer upto tendons and joint tissue
3 Deep ulcer with abscess, osteomyelitis, and joint sepsis Choosing an Appropriate Antibiotic
4 Localized gangrene of forefoot or heel
5 Gangrene of entire foot/global gangrene
As soon as a diagnosis of DFI is established,
antibiotic treatment should be initiated. NICE
(2016) guidelines state that all primary care set-
tings should have care pathways in place for man-
clean wounds (A), nonischemic infected wounds aging DFIs with specific antibiotic regimens that
(B), ischemic wounds (C), and infected ischemic take local resistance issues into account. The anti-
wounds (D). The S(AD) SAD classification (Table biotic choice should be based on the likely proven
IV) grades 5 ulcer features (size, depth, sepsis, ar- causative pathogens, the severity of the infection
teriopathy, and denervation) on a 4-point scale (0- and evidence of efficacy for DFIs while being
3). Similarly, the International Working Group on cognisant of cost. The NICE guidelines also rec-
the Diabetic Foot has suggested the PEDIS clas- ommend that the choice of antibiotic treatment
sification (Perfusion (ischaemia), Extent (area), should be influenced by the care setting, patient
Depth, Infection, Sensation (neuropathy)), which preferences, the clinical situation, and the pa-
grades the wound on a 5-feature basis: perfusion tient’s medical history.
(arterial supply), extent (area), depth, infection, and The IWGDF and NICE make specific recom-
sensation. Finally, according to the Infectious Dis- mendations concerning antimicrobial therapy for
eases Society of America guidelines, the infected DFIs depending on the severity:
diabetic foot is subclassified into the categories of   - For mild infections, initially offer oral antibiot-
uninfected, mild (restricted involvement of only ics effective against Gram-positive organisms.
skin and subcutaneous tissues), moderate (more   - A one- to two-week course of antibiotic ther-
extensive or affecting deeper tissues), and severe apy is usually sufficient for mild infections.
(accompanied by systemic signs of infection or   - For moderate and severe infections, admin-
metabolic instability). ister antibiotics effective against Gram-posi-
In an uncomplited clinical classification approach, tive and Gram-negative organisms, including
diabetic foot ulcers can be described as neuropathic, anaerobic bacteria.

Table III. University of Texas Classification System.

0 1 2 3

A No open lesion Superficial wound Affected tendons/capsules Affected bone/joint

B With infection With infection With infection With infection
C Ischemic Ischemic Ischemic Ischemic
D Infection/Ischemia Infection/Ischemia Infection/Ischemia Infection/Ischemia

Table IV. Diabetic foot infection classification schemes: Infectious Diseases Society of America (IDSA).
Clinical description IDSA IWGDF

Wound without purulence or any manifestations of inflammation Uninfected 1

≥ 2 Manifestations of inflammation (purulence or erythema, pain, tenderness,
warmth, or induration); any cellulitis or erythema extends 52 cm Mild 2
around ulcer, and infection is limited to skin or superficial subcutaneous tissues;
no local complications or systemic illness
Infection in a patient who is systemically well and metabolically stable but has ≥ 2 cm; Moderate 3
lymphangitis; spread beneath fascia; deep tissue abscess; gangrene;
muscle, tendon, joint, or bone involvement
Infection in a patient with systemic toxicity or metabolic instability Severe 4
(e.g., fever, chills, tachycardia, hypotension, confusion, vomiting, leukocytosis,
acidosis, hyperglycemia, or azotemia)

33
 D. Pitocco, T. Spanu, M. Di Leo, R. Vitiello, A. Rizzi, et al

  - For moderate infections, offer oral or initial quency, duration of study antibiotic treatment, or
parental administration depending on the length of hospitalization. In another randomized
clinical situation and choice of antibiotic. study, double-blinded, multicentre trial in diabet-
 - For severe infections, administer parental ic adults (n=586) with a foot infection classified as
therapy with a switch to oral therapy based moderate-to-severe it was found that ertapenem
on the clinical situation and the response to were equivalent to those for patients treated with
treatment. piperacillin/tazobactam. The once-daily ertape-
  - For DFO, offer six weeks of antibiotic ther- nem is advantageous in the DFI setting, in spite
apy according to local protocols for patients of the fact that ertapenem does not cover most
who do not undergo surgical resection of the enterococci or Pseudomonas aeruginosa. Usual-
infected bone. ly, moderate and severe DFI are typically treated
 - For those who have undergone surgical inter- with intravenous antibiotic therapy for two to four
vention and all infected bone are resected, offer weeks, with four to six weeks of therapy for os-
no more than one week of antibiotic therapy37. teomyelitis.
Lipsky et al43 do not recommend prophylactic Operating intervention of moderate to severe
treatment of clinically uninfected wounds with DFI is often essential, and includes aggressive
antimicrobial therapy, and they advise against incision, drainage and debridement of non-viable
the use of any specific type of dressing for DFI soft tissue and bone. With an increasing infection
with the aim of preventing an infection or im- severity, there was a statistically significant trend
proving its outcome. toward an increased risk for amputation, an in-
creased anatomic level of amputation. An increas-
Treatment of Diabetic Foot Infections ing infection severity was associated to a signifi-
Founded on the results of the available studies, cant trend toward increasing risk for experiencing
no single drug or combination of agents seems to other diabetic foot-related complications, such as
be superior to any others. We have several antibi- neuropathy, vascular disease, and history of am-
otic agents for treating DFIs, both by the oral and putation. Foot infections can extend proximally
parenteral routes. It is important to bear in mind into the leg through the tarsal tunnel, resulting in
that while antibiotics are necessary for treating a rapidly ascending limb- and life-threatening infec-
DFI, they are not usually sufficient. All patients tions. Early surgical treatment of DFI may reduce
will need appropriate wound care (debridement, the need for major amputations. An aggressive
dressings, and pressure off-loading) and most will surgical approach against foot infection in diabet-
need some surgical interventions. An internation- ic patients may reduce the need for above-ankle
al survey found that developing a stewardship pro- amputation. Treatment of diabetic foot infection
gramme was associated with reductions in 96% requires the combination of early surgical treat-
of hospitals for inappropriate prescribing, 86% for ment and antimicrobial therapy. Many surgeons
broad-spectrum antibiotic use, 80% for antibiotic still advocate a transtibial amputation (TTA) as the
expenditure, 71% for healthcare-acquired infec- primary surgical option for non-healing foot ulcer-
tions, 65% for length of stay or mortality, and 58% ations. A bioabsorbable calcium sulphate antibiotic
for bacterial resistance. MRSA infections protract beads into the surgical wound can increase the ef-
wound healing times and hospitalization stays, in- fects of TMA for diabetic ulcerations of the fore-
crease the need for surgical procedures, and re- foot. This method could have a significant impact
sult in treatment failure. The antibiotic regimen on the management of diabetic forefoot ulcerations
should take account of an agent active against by preventing additional hospital stays for opera-
Gram-positive cocci having a care for MRSA in tive revisions, and thereby improving the patient’s
high-risk patients. Treatment for previously treat- quality of life.
ed or severe DFI should include extended cover- Adjunctive therapies include the use of antibi-
age for Gram-negative bacilli and enterococcus otic-impregnated beads, the application of nega-
species. Gangrenous and foul-smelling wounds tive-pressure wound therapy. Adjunctive HBOT
may necessitate anti-anaerobic therapy. In a ran- has a positive effect on wound healing in diabetic
domized study of ampicillin/sulbactam versus foot with infection. Conservative treatment, in-
imipenem/cilastatin for the treatment of mod- cluding prolonged, culture-guided parenteral and
erately severe DFI in 90 patients, there were no oral antibiotics, is efficacious without amputation
significant differences between the treatments in in a large percentage of diabetic patients admitted
terms of clinical success rate, adverse-event fre- for a foot skin ulcer or suspected osteomyelitis55.

34
 Diabetic foot infections: a comprehensive overview

Predictors of Treatment Failure Conflict of Interest

in Diabetic Foot Infections The authors have not received any funding or benefits from
Clinical failure rates were 46% for patients the industry to conduct this study.
with risk factors (elevated white blood cell count,
C-reactive protein or erythrocyte sedimentation
rate; high wound severity score; inpatient treat- Reference
ment; low serum albumin; male sex; and skin
temperature of affected foot >10°C above that of  1) Singh N, A rmstrong DG, L ipsky BA. Preventing foot
unaffected foot) compared with 10% for patients ulcers in patients with diabetes. JAMA 2005; 293:
with no risk factors and 16-17% for patients with 217-228.
one risk factor56. Increased WBC and severe UT  2) National diabetes fact sheet, 2011. US Department
wounds (grades 2 and 3) were significant inde- of Health and Human Services. Center for Dis-
ease Control and Prevention, Atlanta, GA (2011).
pendent risk factors for clinical failure in patients
 3) L ipsky BA. Empirical therapy for diabetic foot infec-
treated for DFI in the SIDESTEP study46. Clinical tions: are there clinical clues to guide antibiotic
failure was noted in 23% of the patients with UT selection? Clin Microbiol Infect 2007; 13: 351‐353.
wounds 2B,D and 3B,D at baseline, which can  4) Richard JL, Sotto A, L avigne JP. New insights in
be compared to 11% for a wound stage of 0 or diabetic foot infection. World J Diabetes 2011; 2:
1. The mean WBC was 9,777 cells/mm3 for those 24‐32.
patients who failed treatment, compared to 7,933  5) Peters EJ. Pitfalls in diagnosing diabetic foot infec-
cells/mm3 for those with a favourable response. tions. Diabetes Metab Res Rev 2016; 32 (Supp 1):
CRP and ESR values greater than 9.1 and 54.4, 254-260.
respectively, were associated with treatment fail-  6) Lipsky BA, Aragon-Sanchez J, Diggle M, Embil J, Kono S,
Lavery L, Senneville É, Urbančič-Rovan V, Van Asten S;
ure. A meta-analysis of data57 from randomized International Working Group on the Diabetic Foot, Pe-
controlled trials on DFIs showed a treatment fail- ters EJ. IWGDF guidance on the diagnosis and man-
ure rate of 22.7% in 18 studies. The isolation of agement of foot infections in persons with diabetes.
MRSA was found to be a significant factor associ- Diabetes Metab Res Rev 2016; 32 (Suppl 1): 45-74.
ated with treatment failure, although the presence  7) A rmstrong DG, L ipsky BA. Diabetic foot infections:
or absence of OM did not affect the outcome. In stepwise medical and surgical management. Int
a retrospective cohort study46 of the outcomes of Wound J 2004; 1: 123-132.
conservatively treated DFIs, fever, increased se-  8) Centers for Disease Control and Prevention, US De-
partment of Health and Human Services. National
rum creatinine, prior hospitalization for DFI and diabetes fact sheet: general information and na-
gangrenous lesions were independent factors as- tional estimates on diabetes in the United States,
sociated with treatment failure. 2003. Atlanta: US Centers for Disease Control
and Epidemiology, 2003.
 9) Mutluoglu M, Sivrioglu AK, Eroglu M, Uzun G, Turhan
Conclusions V, Ay H, Lipsky BA. The implications of the presence
of osteomyelitis on outcomes of infected diabetic foot
wounds. Scand J Infect Dis 2013; 45: 497-503.
The prevalence of MRSA is high and the in-
10) Faglia E, Clerici G, C aminiti M, Curci V, Somalvico F.
correct use of antibacterials, hospital environ- Influence of osteomyelitis location in the foot of
ment, osteomyelitis, and nasal carriage of MRSA diabetic patients with transtibial amputation. Foot
give to infection with MRSA. Understanding the Ankle Int 2013; 34: 222-227.
pathophysiology and promptly identifying risk 11) Darbellay P, Uçkay I, Dominguez D, Mugnai D, Filtri
factors for DFI are essential. A thorough evalua- L, L ew D, A ssal M. Diabetic foot infection: a multi-
tion of DFI that utilises a multidisciplinary team disciplinary approach. Rev Med Suisse 2011; 7:
is recommended to achieve optimal outcomes. It 894-897.
is important to classify accurately DFI in order 12) A rmstrong DG, L avery LA. Diabetic foot ulcers:
to guide treatment regimens, facilitate consistent prevention, diagnosis and classification. Am Fam
Phys 1998; 57: 1325-1332.
communication among health-care providers,
13) Reiber GE, Vileikyte L, Boyko EJ, del Aguila M, Smith
and predict patient outcomes. The IDSA and UT DG, L avery LA. Causal pathways for incident low-
classifications provide relatively simplistic and er-extremity ulcers in patients with diabetes from
objective methods for classifying DFI. Prompt two settings. Diabetes Care 1999; 22: 157-162.
recognition and treatment of DFI are mandatory 14) Noor S, Zubair M, Ahmad J. Diabetic foot ulcer–a re-
for ensuring maximal limb salvage. view on pathophysiology, classification and microbial
etiology. Diabetes Metab Syndr 2015; 9: 192-199.

35
 D. Pitocco, T. Spanu, M. Di Leo, R. Vitiello, A. Rizzi, et al

15) A hmad J. The diabetic foot. Diabetes Metab Syndr 29) Edwards R, Harding KG. Bacteria and wound heal-
Clin Res Rev 2016; 10: 48-60. ing. Curr Opin Infect Dis 2004; 17: 91-96.
16) Joseph WS, L eFrock JL. The pathogenesis of diabet- 30) A bbas ZG, Lutale JK, Ilondo MM, A rchibald LK. The
ic foot infections immunopathy, angiopathy. J Foot utility of Gram stains and culture in the manage-
Surg 1987; 26: 7-11. ment of limb ulcers in persons with diabetes. Int
Wound J 2012; 9: 677-682.
17) Peters EJ, L avery LA, A rmstrong DG. Diabetic lower
extremity infection: influence of physical, psycho- 31) G adepalli R, Dhawan B, Sreenivas V, K apil A, A mmini
logical and social factors. J Diabetes Complica- AC, Chaudhry R. A clinicomicrobiological study of
tions 2005; 19: 107-112. diabetic foot ulcers in an Indian tertiary care hos-
pital. Diabetes Care 2006; 29: 1727-1732.
18) Pryce TD. A case of perforating ulcers of both feet
associated with diabetes and ataxic symptoms. 32) Zubair M, A bida M, Jamal A. Incidence, risk factors
Lancet 1887; 130: 11-12. for amputation among patients with diabetic foot
ulcer in a North Indian tertiary hospital. Foot 2011;
19) L ipsky BA, Pecoraro RE, L arson SA, Hanley ME, 22: 24-30.
A hroni JH. Outpatient management of uncompli-
cated lower extremity infections in diabetic pa- 33) Breen JD, K archmer AW. Staphylococcus aureus in-
fections in diabetic patients. Infect Dis Clin North
tients. Arch Intern Med 1990; 150: 790-797.
Am 1995; 9: 11-24.
20) Bagdade JD, Root RK, Bulger RJ. Impaired leuko-
34) Dang CN, Prasad YD, Boulton AJ, Jude EB. Methi-
cyte function in patients with poorly controlled di-
cillin-resistant Staphylococcus aureus in the dia-
abetes. Diabetes 1974; 23: 9-15.
betic foot clinic: a worsening problem. Diabet Med
21) Inzucchi SE. Clinical practice. Management of hy- 2003; 20: 159-161.
perglycemia in the hospital setting. N Engl J Med 35) Pitocco D, Spanu T, Di L eo M, Vitiello R, Rizzi A,
2006; 355: 1903-1911. Tartaglione L, C aputo S, Tinelli G, Z accardi F, Flex A,
22) Hostetter MK. Perspectives in diabetes. Handicaps G alli M, Pontecorvi A, Sanguinetti M. Spectrum of
to host defense. Effects of hyperglycemia on C3 and germs and their antibiotic pattern in patients with
Candida albicans. Diabetes 1990; 39: 271-275. infected diabetic foot ulcers: an observational
23) Sentochnik DE, Eliopoulos GM. Infection and diabe- study of five years. Data not published.
tes. In: Kahn CR, Weir GC, editors. Joslin’s diabe- 36) K andemir O, A kbay E, Sahin E, Milcan A, Gen R.
tes mellitus. 13th ed. Philadelphia: Lea & Febiger, Risk factors for infection of the diabetic foot with
1994; pp. 867-868. multi-antibiotic resistant microorganisms. J Infect
24) K eidar Z, Militianu D, Melamed E, Bar -Shalom R, Is- 2007; 54: 439-445.
rael O. The diabetic foot: initial experience with 37) Wagner FW Jr. The dysvascular foot: a system
18F-FDG PET/CT. J Nucl Med 2005; 46: 444-449. for diagnosis and treatment. Foot Ankle 1981; 2:
64-122.
25) B asu S, Chryssikos T, Houseni M, Scot M alay D,
Shah J, Zhuang H, A lavi A. Potential role of FDG 38) L avery LA, A rmstrong DG, Harkless LB. Classifica-
PET in the setting of diabetic neuro-osteoar- tion of diabetic foot wounds. J Foot Ankle Surg
thropathy: can it differentiate uncomplicated 1996; 35: 528-531.
Charcot’s neuroarthropathy from osteomyelitis 39) M acfarlane RM, Jeffcoate WJ. Classification of di-
and soft tissue infection. Nucl Med Commun abetic foot ulcers: the S(AD) SAD system. Diabet
2007; 28: 465-472. Foot 1999; 2: 123-131.
26) L ipsky BA, Peters EJ, Berendt AR, Senneville E, Bak- 40) Schaper NC. Diabetic foot ulcer classification sys-
ker K, Embil JM, L avery LA, Urban či č-Rovan V, Jeff- tem for research purposes: a progress report on
coate WJ; International Working Group on Diabetic criteria for including patients in research stud-
Foot. Specific guidelines for the treatment of di- ies. Diabetes Metab Res Rev 2004; 20 Suppl 1:
abetic foot infections 2011. Diabetes Metab Res S90-S95.
Rev 2012; 28 (Suppl. 1): 234-235. 41) L avery LA, A rmstrong DG, Murdoch DP, Peters EJG,
27) A merican Diabetes A ssociation. Peripheral arterial L ipsky BA. Validation of the infectious diseases So-
disease in people with diabetes. Diabetes Care ciety of America’s diabetic foot infection classifi-
2003; 26: 3333-3341. cation system. Clin Infect Dis 2007; 44: 562-565.
28) B oulton AJ, A rmstrong DG, A lbert SF, Frykberg 42) NICE (2016) Diabetes in Adults NICE Quality
RG, Hellman R, K irkman MS, L avery LA, L emaster Standards (QS6) (online) Available at: https://
JW, M ills JL Sr, M ueller MJ, Sheehan P, Wukich www.nice.org.uk/guidance/ QS6/chapter/Intro-
DK; American Diabetes Association; American duction (accessed 20.08.2016)
Association of Clinical Endocrinologists. Com- 43) L ipsky BA, A ragon -Sanchez J, Diggle M, Embil J, Kono
prehensive foot examination and risk assess- S, L avery L, Senneville É, Urbančič-Rovan V, Van A s-
ment: a report of the task force of the foot care ten S; International Working Group on the Diabetic
interest group of the American Diabetes Asso- Foot, Peters EJ. IWGDF guidance on the diagno-
ciation, with endorsement by the American As- sis and management of foot infections in persons
sociation of Clinical Endocrinologists. Diabetes with diabetes. Diabetes Metab Res Rev 2016; 32
Care 2008; 31: 1679-1685. (Suppl 1): 45-74.

36
 Diabetic foot infections: a comprehensive overview

44) Vardakas KZ, Horianopoulou M, Falagas ME. Fac- above-ankle amputation? Clin Infect Dis 1996;
tors associated with treatment failure in patients 23: 286-291.
with diabetic foot infections: An analysis of data 51) Sanders LJ, Dunlap G. Transmetatarsal amputa-
from randomized controlled trials. Diabetes Res tion: a successful approach to limb salvage. J Am
Clin Pract 2008; 80: 344-351. Pod Med Assoc 1992; 82: 129-130.
45) Barriere SL. Clinical, economic and societal impact 52) K rause FG, deVries G, Meakin C, K alla TP, Younger
of antibiotic resistance. Expert Opin Pharmacoth- AS. Outcome of transmetatarsal amputations in
er 2015; 16: 151-153. diabetics using antibiotic beads. Foot Ankle Int
46) Eleftheriadou I, Tentolouris N, A rgiana V, Jude E, 2009; 30: 486-493.
Boulton AJ. Methicillin-resistant Staphylococcus 53) Chen CE, Ko JY, Fong CY, Juhn RJ. Treatment of di-
aureus in diabetic foot infections. Drugs 2010; 70: abetic foot infection with hyperbaric oxygen ther-
1785-1797. apy. Foot Ankle Surg 2010; 16: 91-95.
47) McK innon PS, Paladino JA, Grayson ML, Gibbons 54) Pittet D, Wyssa B, Herter -Clavel C, Kursteiner K,
GW, K archmer AW. Cost-effectiveness of ampicil- Vaucher J, L ew PD. Outcome of diabetic foot in-
lin/sulbactam versus imipenem/cilastatin in the fections treated conservatively: a retrospective
treatment of limb-threatening foot infections in di- cohort study with long-term follow-up. Arch Intern
abetic patients. Clin Infect Dis 1997; 24: 57-63. Med 1999; 159: 851-856.
48) Lipsky BA, Armstrong DG, Citron DM, Tice AD, Morgen- 55) Chen Z, Cheng L, Feng G. Bone inflammation and
stern DE, Abramson MA. Ertapenem versus piperacillin/ chronic recurrent multifocal osteomyelitis. Eur
tazobactam for diabetic foot infections (SIDESTEP): Rev Med Pharmacol Sci 2018; 22: 1380-1386.
prospective, randomised, controlled, double-blinded, 56) L ipsky BA, Sheehan P, A rmstrong DG, Tice AD, Polis
multicentre trial. Lancet 2005; 366: 1695-1703. AB, A bramson MA. Clinical predictors of treatment
49) L avery LA, A rmstrong DG, Murdoch DP, Peters EJ, failure for diabetic foot infections: data from a pro-
Lipsky BA. Validation of the Infectious Diseases spective trial. Int Wound J 2007; 4: 30-38.
Society of America’s diabetic foot infection classi- 57) Vardakas KZ, Horianopoulou M, Falagas ME. Fac-
fication system. Clin Infect Dis 2007; 44: 562-565. tors associated with treatment failure in patients
50) Tan JS, Friedman NM, H azelton -M iller C, Flana - with diabetic foot infections: an analysis of data
gan JP, File TM. Jr. Can aggressive treatment from randomized controlled trials. Diabetes Res
of diabetic foot infections reduce the need for Clin Pract 2008; 80: 344-351.

37