You are on page 1of 10

|

Received: 20 May 2017    Accepted: 19 August 2017

DOI: 10.1111/ijcp.13006

CONSENSUS

Diabetic foot infection: Antibiotic therapy and good practice


recommendations

Nicholas D. Barwell1  | Marion C. Devers2 | Brian Kennon3 | Helen E. Hopkinson3 | 


Claire McDougall4 | Matthew J. Young5 | Hannah M. A. Robertson6 | Duncan Stang4 | 
Stephanie J. Dancer4 | Andrew Seaton3 | Graham P. Leese7 | on behalf of the Scottish
Diabetes Foot Action Group

1
Forth Valley Royal Hospital, Larbert, UK
2
Abstract
Monklands Hospital, Airdrie, UK
3 Background: Healthcare events related to diabetic foot disease carry a burden of morbid-
Queen Elizabeth University Hospital,
Glasgow, UK ity, mortality and economic cost. Prompt identification of clinical infection with appropri-
4
Hairmyres Hospital, East Kilbride, UK ate tissue sampling limits use of broad spectrum empirical antibiotics and improves
5
Edinburgh Royal Infirmary, Edinburgh, UK antibiotic stewardship. Staphylococcus aureus remains the commonest infecting organism
6
Aberdeen Royal Infirmary, Aberdeen, UK
and high-­dose flucloxacillin remains the empirical antibiotic of choice for antibiotic naïve
7
Ninewells Hospital, Dundee, UK
patients. Barriers to microbe-­specific treatment include: adequate tissue sampling, delays
Correspondence in culture results, drug allergies and the emergence of multidrug-­resistant organisms
Nicholas D. Barwell, Consultant Physician,
Department of Diabetes and Endocrinology,
which can complicate the choice of targeted antibiotics. Even appropriate antibiotic treat-
Forth Valley Royal Hospital, Larbert, UK. ment carries a risk of adverse events including the selection of resistant organisms.
Email: nicholas.barwell@nhs.net
Aims: Multidisciplinary clinical assessment of a diabetic foot infection is supported by
the use of appropriate imaging modalities and deep tissue sampling, both of which are
encouraged to enhance sampling accuracy. Narrow-­spectrum, high dose, short dura-
tion antimicrobial therapy is ideal. Further clarity in these areas would be of benefit to
clinicians involved in management of diabetic foot infections.
Methods: A combination of literature review with expert discussion was used to gen-
erate consensus on management of diabetic foot infection, with a specific focus on
empirical antimicrobial therapy.
Results: Gram positive organisms represent the commonest pathogens in diabetic foot
infection. However there are developing challenges in antimicrobial resistance and
antibiotic availability.
Discussion: Recommendations for empirical therapy, including the choice of alterna-
tive oral agents and use of outpatient antibiotics would be of benefit to those involved
in diabetic foot care.
Conclusion: This paper provides advice on empirical antibiotic therapy that may be
used as a framework for local guideline development to support clinicians in the man-
agement of diabetic foot infection.

This update on antimicrobial recommendations for diabetic foot ulcer treatment is a consensus statement based on clinical trial evidence, review of international guidelines and expert opinion.
In the context of individual treatment decisions, local microbiology results and advice should be paramount in informing responsible clinicians. However, the spectrum of infecting pathogens
causing foot infection is consistent and supports the consideration of empirical antibiotic advice. These recommendations are suggested to support clinicians and in conjunction with regional
pathogen variations and antibiotic susceptibilities, provide a basis for local guideline development.

Int J Clin Pract. 2017;e13006. wileyonlinelibrary.com/journal/ijcp © 2017 John Wiley & Sons Ltd  |  1 of 10
https://doi.org/10.1111/ijcp.13006
|
2 of 10       BARWELL et al.

1 |  INTRODUCTION T A B L E   1   IDSA guideline for infection grading, incorporating


infection component of IWGDF PEDIS classification23,25

Foot ulceration and infection continue to represent an important Severity of Infection Description
1,2
source of morbidity in people with diabetes mellitus. In an acute pre- 1. No infection
sentation with diabetic foot infection (DFI), there is frequently a delay 2. Mild (Grade 2)
in the identification of the causative organism, which may compel use No systemic illness with either:
of empirical antibiotic(s).3,4 Whilst this allows the prompt initiation of
Two or more features of inflammation:
antimicrobial therapy, it carries a risk of treatment failure, the selection
Swelling/Induration, Erythema, Pain, Warmth, Pus
of resistant organisms and complications related to drug toxicity and
Or
alteration of the host microbiome.5
Cellulitis < 2 cm. Confined to skin/subcutaneous tissue
The lifetime risk of foot complication within the diabetic popu-
3. Moderate (Grade 3)
lation is considered to be around 25%,6 with a point prevalence of
around 2%7 and the development of a diabetes-­related foot ulcer is Above with either:

associated with an increased risk of lower extremity amputation and lymphatic streaking, deep tissue infection

death.7-10
In addition to these outcomes, diabetic foot ulceration is (subcutaneous fascia, tendon, bone), abscess
associated with hospital admission, prolonged inpatient length of stay Or
and physical and psychological morbidity.7,8,11 Cellulitis > 2 cm
Sharp debridement of ulcers, combined with pressure offloading, 4. Severe (Grade 4)
management of infection and the involvement of a multidisciplinary Any infection with systemic toxicity (fever, shock, vomiting,
team to optimise glycaemia, cardiovascular risk, assess potential re- confusion, metabolic instability). Presence of critical
vascularisation and surgical intervention are considered to be the basis
of ulcer care.12-15 In order to promote wound healing and reduce risk
of amputation, involvement of the multidisciplinary team is recom- bone in an infected ulcer increases the likelihood that osteomyelitis
mended early in the natural history of the ulcer. 14-19 is present, but is not pathognomonic of osteomyelitis.14,31 It is, how-
ever, an independent predictor of lower extremity amputation (LEA).28
The combination of both radiography and positive probe-­to-­bone has
2 |  IDENTIFICATION OF DFI high sensitivity for the diagnosis of osteomyelitis.32,33 The removal of
bone fragments during debridement is consistent with osteomyelitis
Establishing presence of infection is an important component of and these fragments should be sent for culture in a sterile container.34
ulcer care. Not all ulcers are infected and given increasing antibiotic Clinical findings such as a globally inflamed digit which has lost its
resistance and risk of antibiotic-­related adverse events including di- usual contours and skin markings (“sausage” digit) also suggest under-
arrhoea and Clostridium difficile infection, the goal for antibiotic use lying osteomyelitis but may also be observed in fracture or Charcot
in foot ulceration is to treat an active infective process, aiming to neuroarthropathy.35,36 Differentiation of infection vs Charcot neuro-
tailor therapy to the appropriate pathogen(s).20 The presence of mi- arthropathy is often clinically challenging, however detailed discussion
crobes within clinically non-­infected ulcers does not influence rates of the exploration of these two diagnoses is beyond the scope of this
of healing.21 In the absence of infection, antibiotic therapy represents article.
a risk for the selection of drug-­resistant pathogens and therapeutic Further specific assessment of the infective burden associated
21-23
complications. with a foot ulcer can be categorised using the Infectious Disease
The identification of an infective process within a diabetic foot Society of America (IDSA) grading tool (Table 1). Based on current ev-
ulcer is predicated on clinical assessment combined with supportive idence review, the authors would not suggest any update or addition
investigations.23-25 Inflammatory change within the ulcer bed and the to the previous consensus statement from 200923 and would reiterate
presence of increasing exudate or pus is suggestive of infection.24-26 the use of high dose, narrow-­spectrum antibiotics for relatively short
The validity of these clinical factors has been challenged by a study duration. In the context of empirical therapy, previous positive mi-
which has shown poor correlation between microbial load and the crobiology results should always be considered and cover for specific
presence of clinical signs.22 However, the IDSA criteria have been pathogens might be indicated in relation to the clinical context (eg,
27,28
validated as a useful tool for grading foot infections. Coexisting methicillin-­resistant S. aureus in people in long-­term care).37,38
skin and/or soft tissue infection (SSTI) in association with the ulcer Assessment of ulcer characteristics including infection can be stan-
border also suggests an active infective process, usually seen as red, dardised using validated classifications schemes such as the University
warm, swollen and inflamed skin.3,24-26 This may be accompanied by of Texas rating,39 SINBAD 40
or IWGDF PEDIS.41 The current itera-
a change in odour from the ulcer or the development of pain (unusual tion of the Scottish National Diabetes Database (SCI-­Diabetes) utilises
in a neuropathic foot).22,28 The diagnosis of deep infection such as assessment tools based on the University of Texas system. An ulcer
osteomyelitis is, despite the development of new imaging modalities, grading system helps to standardise assessment, indicates prognosis
often made using serial plain radiography.29,30 The ability to probe to and forms a useful basis for audit and research.42
BARWELL et al. |
      3 of 10

and infection. The presence of inflammatory infiltrates, bone loss


and marrow oedema is observed in both conditions, but predomi-
nance of osteoclasts over osteoblasts is more indicative of Charcot
neuroarthropathy.35 Despite these limitations, the authors support
the use of bone biopsy for pathogen detection or to resolve diag-
nostic uncertainty between infection and Charcot neuroarthopathy.
The development of local protocols which consider available imag-
ing and surgical resource is recommended, based on bone sampling
principles outlined in Figure 1.48-50
For forefoot ulcers complicated by osteomyelitis, use of MRI scan-
ning in a small cohort was shown to support clinical decision-­making,
specifically duration of antibiotic therapy and requirement for surgical
intervention.51 Three-­monthly MRI scanning was used to determine if
an osteomyelitic bone signal had changed. Although the relapse rate
(initial healing, followed by recurrence of a lesion) at 31 months was
F I G U R E   1   Example guidance on use of deep tissue or bone
low (13%) compared with other studies,52 this approach did promote
biopsy by percutaneous route in suspected deep seated foot
infection in patients with diabetes48–50 prolonged antibiotic therapy (median 6 months, range 3-­12). Other
groups have reported resolution of osteomyelitis with much shorter
3 | IMAGING IN THE DIAGNOSIS OF DFI courses of antibiotics53,54 and we would advise caution with the length
of treatment as described.
The sensitivity of plain X-­ray in detecting osteomyelitis has been
recorded at 54%,29 but is much greater with serial plain X-­ray.43
The extent of deep tissue involvement including osteomyelitis may 4 | DIAGNOSTIC MICROBIOLOGY IN DFI
not be clinically apparent, even with the use of sterile metal probes
and serial plain X-­ray imaging.30 The presence of a non-­healing ulcer Wound bioburden in the absence of clinical infection should not result
may raise clinical suspicion of deep tissue infection, although this in the initiation of antimicrobial therapy.21,25 However, it is accepted
was not a contributory factor in a recent study of chronic ulcera- that culture results may be limited by culturing/sampling techniques
tion.44 Both isotope white cell scans and triple phase bone scans and previous antibiotic exposure.21,55 The role of polymicrobial infec-
are highly sensitive in detecting osteomyelitis and deep tissue in- tion and the degree to which microbe-­produced biofilm influences
fection but lack specificity (being unable to distinguish between culture results are controversial. It is likely that adherent biofilm pro-
inflammatory changes related to Charcot neuroarthropathy), ana- duced by microbes such as S. aureus contributes to the antimicrobial
tomical resolution and may continue to show positive results for resistance of the organism in acute infection.56,57 Biofilm may also
45,46
long periods after clinical resolution. MRI scanning is between shield other pathogens from detection,58 as the combination of mo-
90% and 100% sensitive for osteomyelitis but alone, has a speci- lecular techniques and deep tissue sampling often reveals a polymi-
ficity of <80%. Imaging protocols which combine Single Positron crobial presence.59,60
Emission Computed Tomography (SPECT) and MRI may improve A French group has utilised molecular techniques to determine the
the sensitivity and specificity of detection of foot infection.47 An pathogenicity of microbes detected in foot ulcers. Oligonucleotide mi-
alternative diagnostic combination using SPECT–CT scanning with croarrays determining the expression of virulence factors from S. au-
67
the radioisotope Ga and aseptic bone puncture results in even reus combined with detection of S. aureus clones considered to be
greater specificity (>93%) than hybrid SPECT/MRI.48 Benefits in- “actively infecting” show promise in establishing which ulcers contain
clude direct bone sampling to enhance pathogen detection, with a bioburden likely, or unlikely, to cause harm and therefore may help
potential reduction in false positive microbiology because of com- choose antibiotic therapy.61
mensal contamination. In participants with a positive SPECT-­CT Further studies in small cohorts of people with diabetic foot ul-
but negative bone puncture, none had evidence of persisting ul- cers have been performed by the same group.62 These studies not
ceration or clinical suspicion of osteomyelitis at 1-­year follow up.48 only identify a much more diverse bacterial flora within the ulcer, but
This technique represents a potential advance in supporting clinical appear to be able to distinguish actively pathogenic organisms, from
decision-­making with respect to the diagnosis of osteomyelitis and non-­
pathogenic or colonising organisms with pathogenic potential
may provide a greater degree of confidence in pathogen detection and have greater sensitivity in detecting pathogenic species in deep
67
and treatment. However, SPECT-­CT with Ga imaging is not a tissue samples. The presence of ‘colonising’ bacteria (with pathogenic
widely available imaging modality (Personal communication Colville potential) may be a predictor of relapsing foot infection or chronic
D, 2015) and the bone puncture process is invasive and requires persistent ulceration.63,64 The presence of additional-­site methicillin-­
stringent asepsis and preparation. Histological analysis of bone may resistant S. aureus carriage (eg, nasal/perineal) may also predispose to
be of benefit in differentiation between Charcot neuroarthropathy re-­infection.65,66 These issues highlight known limitations in current
|
4 of 10      

T A B L E   2   Empirical antibiotic guideline for diabetic foot infection

  Mild Moderate Severe Osteomyelitis

Antibiotic naive Flucloxacillin 1q QDS (oral) Flucloxacillin 1 g QDS (oral) Flucloxacillin 2 g QDS (IV) Acute (IDSA/IWGDF-PEDIS Grade 4) Flucloxacillin 2 g QDS
Or 2g QDS (IV) + Clindamycin 600 mg QDS (IV) (IV)
+/− Metronidazole 400 mg TDS +/− Gentamicin (IV) (Max 4 days) or + Clindamycin 600 mg TDS (oral)
(oral) Aztreonam 2 g TDS (IV) +/− Gentamicin (IV) (Max 4 days) or
+/− Metronidazole 400 mg TDS (oral) or Aztreonam 2 g TDS (IV) or Ciprofloxacin 750 mg BD (oral)
500 mg TDS (IV) Non-Acute (IDSA/IWGDF-PEDIS Grade 2-3)
Clindamycin 600 mg TDS (oral)
+ Ciprofloxacin 750 mg BD (oral)
Non-Antibiotic Doxycycline 100 mg BD (oral) Clindamycin 600 mg TDS (oral) Piperacillin-­Tazobactam 4.5 g TDS (IV) Acute
naive Or or 600 mg QDS (IV) + Clindamycin 600 mg QDS (IV) Vancomycin (IV)
Clindamycin 450 mg TDS Or + Clindamycin 600 mg TDS (oral)
(oral) Co-­amoxiclav 625 mg TDS +/− Gentamicin (IV) (Max 4 days) or
(oral) or 1.2 g Aztreonam 2 g TDS (IV) or Ciprofloxacin
TDS (IV) 750 mg BD (oral)
Penicillin allergy   Clindamycin 600 mg TDS (oral) Vancomycin (IV) Non-Acute
or + Clindamycin 600 mg QDS (IV) Clindamycin 600 mg TDS (oral) +
600 mg QDS (IV) +/− Gentamicin (IV) (Max 4 days) or Ciprofloxacin 750 mg BD (oral)
Aztreonam 2 g TDS (IV)
+/− Metronidazole 400 mg TDS (oral) or
500 mg TDS (IV)
Duration/ Notes 7 days 7 days (including IV to oral 7-­10 days (including IV to oral switch) 6-­12 weeks therapy. Usually at least 2 weeks of IV therapy in
switch) 14 days IV therapy if S. aureus bacteraemia acute setting but oral therapy may be suitable in non-­acute
Review need for Gram-­negative cover and setting
rationalise therapy depending on microbiology Review need for Gram-­negative cover and rationalise therapy
depending on microbiology
MRSA known Doxycycline 100 mg BD (oral) Vancomycin (IV) Vancomycin (IV) Acute
carrier or Or Or + Clindamycin 600 mg QDS (IV) Vancomycin (IV)
proven Clindamycin 600 mg TDS Clindamycin 600 mg QDS (IV) Consider additional Gram-­negative and + Rifampicin 450 mg BD (oral) or Sodium fusidate 500 mg
infection (oral) Or anaerobic cover TDS (oral)
Or Co-­trimoxazole 960 mg BD (IV) Consider additional Gram-­negative and anaerobic cover
Co-­trimoxazole 960 mg BD Or if Resistant to above: Non-Acute
(oral) Linezolid 600 mg BD Usually combination therapy depending on sensitivities and
Or if Resistant to above: infection specialist advice
Linezolid 600 mg BD (oral)

(Continues)
BARWELL et al.
BARWELL et al. |
      5 of 10

methods of sampling, culture and organism detection. The combina-

available. Dosing of vancomycin, gentamicin and teicoplanin should be as per local guidance. Be aware of potential drug interactions with rifampicin, doxycycline, ciprofloxacin, metronidazole and linezolid.
This is for guidance only and local practice/ guidance may vary. Infection specialist advice should be sought if any uncertainty. Targeted therapy based on good microbiological sampling is always preferred when
Review need for Gram-­negative cover and rationalise therapy
tion of biofilm,59,67 suboptimal sampling techniques 68
and a more

setting but oral therapy may be suitable in non-­acute setting


+ Clindamycin 600 mg TDS (oral) or Rifampicin 450 mg BD

4-­6 weeks therapy (may require >6 weeks in non-­surgically


diverse ulcer microbiome may contribute to reduced efficacy of anti-

treated). Usually at least 2 weeks of IV therapy in acute


microbial therapy.63,64,67
The use of molecular techniques shows potential to deliver more
rapidly available results along with more comprehensive evaluation of

(oral) or Sodium fusidate 500 mg TDS (oral)


the ulcer microbiome.61,62 However, the availability of advanced mo-
lecular diagnostics is limited in most NHS laboratories. There is also a
paucity of outcome data available to illustrate whether a more com-
plete description of the wound microbiome can direct more effective
Daptomycin 6-­8 mg/kg (IV)

depending on microbiology
antimicrobial therapy and improve outcomes.
Ceftriaxone 2g (IV)

Teicoplanin (IV)

5 | ANTIBIOTIC THERAPY: RISKS,


Osteomyelitis

BENEFITS AND COMPLICATIONS


Or

Or

Antibiotic therapy choice is influenced by efficacy, organism sus-


ceptibility and side effect profile. Any antibiotic may be associated
rationalise therapy depending on microbiology
Not applicable as requires initial hospitalisation

with subsequent C. difficile infection (CDI), especially cephalosporins,


14 days IV therapy if S. aureus bacteraemia
Review need for Gram-­negative cover and

clindamycin, co-­
amoxiclav, quinolones, piperacillin-­
tazobactam and
7-­10 days (including IV to oral switch)

carbapenems. All β-­lactam antibiotics, quinolones and macrolides are


associated with increased selection of methicillin-­resistant S. aureus
(MRSA). In Scotland, MRSA generally does not show greater virulence
than MSSA; however, the efficacy of antimicrobial therapy is reduced.
In addition, carbapenem therapy leads to Candidal overgrowth69
and promotion of resistant coliforms.70,71 Risk of CDI is reportedly
increased in people over 65 years, immunosuppressed, hospital in-
patients,72,73 those on proton pump inhibitor agents and those who
Severe

have previously experienced CDI.74-76 Recent changes in antibiotic


prescribing patterns in Scotland have been associated with changing
Dosing assumes normal renal and hepatic function. Please refer to BNF for dose adjustments.

epidemiology of CDI with significantly fewer cases overall but propor-


Community-­associated MRSA

tionally more observed with community-­onset CDI.72,73,77


Linezolid 600 mg BD (oral)
7 days (including IV to oral
Daptomycin 6 mg/kg (IV)

sensitive to Clindamycin,

In all cases, treatment of an infected diabetic foot lesion should be


Ceftriaxone 2 g OD (IV)

focused to a narrow spectrum of pathogen cover, ideally directed by


culture results.24,25 Empirical antibiotic therapy is the first-­line step for
doxycycline and
Teicoplanin (IV)

Co-­trimoxazole

patients with more severe infection with sepsis or associated skin and
is usually

soft tissue infection (Table 1). Wherever possible, appropriate samples


Moderate

switch)

should always be sent for microbiological analysis before initiation


Or

Or

Or

of antibiotic therapy and this group would encourage use of histor-


ical microbiology results to influence choice of empirical therapy. In
Clindamycin, doxycycline and
7 days Community associated
MRSA is usually sensitive to

poorly healing or recurrent ulcers, the authors suggest that clinicians


may also consider the use of screening cultures (eg, nose, perineum,
throat swabs) to provide greater knowledge of the patient’s own skin/
Co-­trimoxazole

mucosal microbiome. These results have the potential to direct paral-


Not applicable

lel eradication therapies (eg, MRSA) and allow an appreciation of re-­


infection risk.65,66 The decision to utilise antibiotic treatment may be
Mild
T A B L E   2   (Continued)

supported using the algorithm in Figure 3.


The International Working Group on the Diabetic Foot (IWGDF) has
published comprehensive guidance on diabetic foot infections.24 Whilst
Duration/Notes

these guidelines represent the foundation of practice for treatment of


DFI, national and regional variation in pathogens and antimicrobial sus-
OPAT

ceptibility exist in the literature.78,79 The recommendations outlined in


Table 2 reflect consensus on regional/national pathogens, antimicrobial
 
|
6 of 10       BARWELL et al.

susceptibility and antibiotic availability (Table 2). For this reason, these pertinent in the current climate of admission avoidance and outpa-
recommendations may be of more practical use to UK-­based clinicians. tient management. The decision to utilise empirical therapy coupled
The duration of antibiotic therapy remains a contentious subject. with potentially suboptimal tissue sampling creates an opportunity
Superficial SSTI can be assessed clinically and 1-­2 weeks of ther- for more prolonged use of broad-­spectrum agents. These cannot
apy is usually adequate.24,25,80,81 However, deep-­
seated infection necessarily be adequately rationalised before inducing antibiotic re-
or osteomyelitis may require prolonged antibiotic therapy and fur- sistance in the infecting pathogen. They may even encourage com-
ther assessment may benefit from re-­imaging (eg, MRI).51 Traditional mensal proliferation which could become problematic in chronic
advice favouring 6-­12 weeks of treatment with antibiotics for non-­ wound situations.
surgically treated osteomyelitis is supported by heterogeneous The diabetic foot ulcer prevalence of MRSA of between 20% and
25
data from retrospective studies. There is inadequate evidence to 30% in some countries88 has lead to a return to use of non-­β-­lactam
support treatment beyond 6 weeks from randomised controlled tri- antimicrobial agents, such as rifampicin, fusidic acid, trimethoprim
als.80,81 A recent French study suggested that resolution of osteomy- and sulphamethoxazole. These agents can prove effective for both
elitis occurred in around two-­thirds of patients treated for 6 weeks. susceptible S. aureus and MRSA in various combinations but would
In this study, not all patients received intravenous antibiotics and rapidly encourage antibiotic resistance if used in isolation.89 The ox-
many received narrow-­spectrum therapy because of the use of bone azolidinone agent linezolid is also active against MRSA but as an es-
sampling and avoidance of empirical therapy where clinically appro- sentially bacteriostatic agent, may also induce resistance.90 Linezolid
priate.53 Both the 2012 IDSA and the 2016 IWGDF guidelines con- has well documented risks of peripheral and optic neuropathy, lactic
sider that a minimum of 4 weeks of antibiotic therapy is required in acidosis, bone marrow toxicity (requiring full blood count monitoring),
the presence of infected or necrotic bone.20,24,25 In the absence of is subject to potential drug interactions including serotonin syndrome
surgical intervention, 6 weeks of antibiotic therapy is suggested24 al- when co-­prescribed with SSRIs and reduced concentration when co-­
though patients who do not undergo any surgical resection may need prescribed with rifampicin. Linezolid may also affect insulin sensitivity
24,25
a longer duration. to alter risk of hypoglycaemia.91 Despite these limitations, linezolid
Retrospective analysis of clinical osteomyelitis treatment in a gen- can permit outpatient management, thus saving costs and with careful
eral diabetic foot ulcer population has shown resolution with medical monitoring may be an appropriate alternative to outpatient parenteral
treatment alone in 66.9% of cases.82 Ulcers were considered healed antibiotic therapy (OPAT).92 The duration of Linezolid therapy for the
with sustained skin closure at 3 months after antibiotic completion majority of patients is usually limited to 2 weeks, although a small pro-
(Personal communication Rajbhandari SM, 2015). Around 50% of these portion may require 4 weeks of treatment.92 The novel cephalosporin
82
patients required multiple antibiotic courses. A randomised controlled agent ceftaroline fosamil has shown efficacy in intravenous treatment
trial in a single specialist centre achieved similar (75%) healing rates in of Gram-­positive infections, including MRSA, and has been effective
patients with forefoot osteomyelitis; these patients received medical in treatment of diabetic foot infections (DFI).93,94 Ceftaroline may be
treatment and showed comparable healing rates to those treated with considered for use in DFI for inpatients for whom the usual glycopep-
a primarily surgical approach (86.3%, P = .33). Patients with peripheral tide therapy has proved unsuccessful or is contraindicated. This agent
arterial disease, exposed bone and poorly controlled diabetes (HbA1c may be considered for use in circumstances where either linezolid or
−1 83
>10%/85.8 mmol.mol ) were excluded. These are all features com- daptomycin would be an alternative option.
mon to “real-­world” diabetes foot clinics. However, there is developing In addition to a return to use of older antibiotics for MRSA, use
evidence that many diabetic foot ulcers with osteomyelitis can be man- of other agents with Gram-­positive activity such as the tetracycline
aged by conservative, non-­surgical treatment. group and co-­trimoxazole, has been included in local guideline devel-
Studies are underway to assess whether the strategy of using oral opment. The prevalence of MRSA infections in Scotland has declined
therapy (usually utilising oral bio-­available agents with appropriate in recent years, partly as a result of both universal and targeted screen-
spectrum of activity and good bone penetration) is non-­inferior to ing programmes and better antimicrobial stewardship.95-97 Although
traditional IV therapy in bone and joint infection (including diabetes-­ there are concerns regarding CDI risk with quinolone antibiotics, cip-
84
associated osteomyelitis). A published Cochrane review of a small rofloxacin represents a suitable oral option for Gram-­negative cover,
number of appropriate studies has suggested equipoise,85 except in and in combination (eg, rifampicin) for treatment of S. aureus, demon-
patients with septicaemia. strates good tissue penetration in skin and soft tissue as well as bone.
Attempts to assess the efficacy of specific antibiotic regimens for Levofloxacin has similar oral bioavailability and bone penetration
diabetic foot ulcers are complicated by the heterogeneous nature of properties to ciprofloxacin and may be considered in some susceptible
study design. In many studies, it is unclear the extent to which “usual Gram-­positive bone infections, usually in combination. Fluoroquinolone
care” includes core factors such as sharp debridement and offloading. antibiotics should be used cautiously in people with risk factors98 for
In addition, those with osteomyelitis and peripheral arterial disease QTc prolongation (including the co-­prescription of some antidepres-
are frequently excluded from studies and there is a lack of consistency sants) and pretreatment and follow-­up ECGs are advisable in these
with respect to reporting of severity criteria and end-­point data.86,87 circumstances.99 Consideration should also be given to the potential
The paucity of novel antibiotic therapies is an important factor ecological impact of the fluoroquinolone, including the risk of multi-
in the ongoing clinical decision-­making process. This is particularly drug resistance in commensal Gram-­negative organisms,100 methicillin
BARWELL et al. |
      7 of 10

as gentamicin require close monitoring. In addition, side effects such


as hyperkalaemia occur more frequently in people with reduced renal
function.105 Consideration of renal function is therefore encouraged in
both empirical antibiotic choice and dosing in those with chronic kid-
ney disease and in all patients the authors recommend a maximum of
4 days of empirical gentamicin106 (Table 2). An alternative strategy to
limit nephrotoxicity is to consider synergistic dosing of gentamicin in
combination with other agents. Specialist antimicrobial advice should
be sought in this case.

6 | OUTPATIENT INTRAVENOUS
ANTIBIOTIC THERAPY (OPAT)
F I G U R E   2   Key points for treatment of DFI
Inevitably, a minority of patients with deep-­seated diabetic foot infec-
tions will require prolonged IV therapy because of organism suscepti-
bility or intolerance/failure of an otherwise appropriate oral regimen.
In these circumstances and presuming suitability, OPAT can be con-
sidered. The resultant reduction in inpatient stay is associated with
reduced healthcare-­associated costs and complications as well as im-
proved patient satisfaction.107 A number of antibiotics lend themselves
to ease of administration in the OPAT setting including the once daily
agents ceftriaxone, ertapenem, daptomycin and teicoplanin. Depending
on the OPAT service, IV therapy may be self (or carer) administered or
via specialised infusion devices. UK good practice recommendations
for safe OPAT practice have been published.108 In the context of se-
vere infection (Tables 1 and 2), OPAT should only be considered as a
“step-­down” option once the individual is clinically stable enough for
outpatient management and assuming no appropriate oral therapy.

7 | CONCLUSIONS

Foot ulceration in people with diabetes continues to be challenging


for both those affected and the healthcare professionals who look
after them. Identification and treatment of infection is an important
step in management although despite useful progress in both imag-
ing of deep tissues and detection of pathogens, clinical assessment
using parameters such as “sausage” digits, probe-­to-­bone testing and
IDSA infection grading criteria (Table 1)25,109 remain the most useful
and appropriate guidelines. Key points with respect to treating DFI
F I G U R E   3   Stepwise approach to antibiotic choice are summarised in Figure 2. Whilst narrow spectrum, tailored antibi-
otic therapy is recommended, it is accepted that empirical treatment
is often required, ideally after appropriate tissue sampling (Figure 3).
resistance in S. aureus101 and host C. difficile.102 Whilst oral antibiotic In the absence of novel antibiotics, the attached guideline should
combination therapy is frequently recommended to reduce the risk of prove to be a useful tool in supporting clinical decision-­making with
resistance development in target organisms, combinations may be lim- respect to antimicrobial therapy (Table 2). Antibiotic choice should
ited by drug-­drug interactions. Particular attention is required when then be reviewed once culture results become available.
considering combining rifampicin with tetracyclines, sodium fusidate
or Linezolid because of potential reduction in efficacy.103
AC KNOW L ED G EM ENTS
Chronic kidney disease is a common comorbidity in people with
diabetes which also increases the risk of foot ulceration.104 Direct The authors would like to thank Drs S Jamdar and B Cooke,
nephrotoxic effects from agents with a narrow therapeutic index such Department of Clinical Microbiology, Forth Valley Royal Hospital and
|
8 of 10       BARWELL et al.

K Hill, Antimicrobial Pharmacist, NHS Tayside and the members of 16. Wang C, Mai L, Yang C, et al. Reducing major lower extremity ampu-
the Scottish Antimicrobial Prescribing Group for reviewing the article. tations after the introduction of a multidisciplinary team in patient
with diabetes foot ulcer. BMC Endocr Disord. 2016;16:38.
17. Trautner C, Haastert B, Mauckner P, Gatcke LM, Giani G. Reduced
AUTHOR CONTRI B UTI O N S incidence of lower-­limb amputations in the diabetic population of
a German city, 1990-­2005: results of the Leverkusen Amputation
All authors contributed to the planning, writing and editing of the Reduction Study (LARS). Diabetes Care. 2007;30:2633‐2637.
18. Krishnan S, Nash F, Baker N, Fowler D, Rayman G. Reduction in dia-
manuscript.
betic amputations over 11 years in a defined U.K. population: bene-
fits of multidisciplinary team work and continuous prospective audit.
Diabetes Care. 2008;31:99‐101.
D ISCLOSURE
19. Gottrup F, Holstein P, Jorgensen B, Lohmann M, Karlsmar T. A new
None. concept of a multidisciplinary wound healing center and a national
expert function of wound healing. Arch Surg. 2001;136:765‐772.
20. Abbas M, Uckay I, Lipsky BA. In diabetic foot infections antibiotics
O RCI D are to treat infection, not to heal wounds. Expert Opin Pharmacother.
2015;16:821‐832.
Nicholas D. Barwell  http://orcid.org/0000-0001-8870-8123 21. Gardner SE, Haleem A, Jao YL, et al. Cultures of diabetic foot ulcers
without clinical signs of infection do not predict outcomes. Diabetes
Care. 2014;37:2693‐2701.
22. Gardner SE, Hillis SL, Frantz RA. Clinical signs of infection in diabetic
REFERENCES
foot ulcers with high microbial load. Biol Res Nurs. 2009;11:119‐128.
23. Leese GP, Nathwani D, Young MJ, et al. Good practice guidance for
1. Skrepnek GH, Mills JL Sr, Armstrong DG. A Diabetic Emergency One
the use of antibiotics in patients with diabetic foot ulcers. Diabetes
Million Feet Long: disparities and Burdens of Illness among Diabetic
Foot J. 2009;12:62‐78.
Foot Ulcer Cases within Emergency Departments in the United
24. Lipsky BA, Aragon-Sanchez J, Diggle M, et  al. IWGDF guidance
States, 2006-­2010. PLoS ONE. 2015;10:e0134914.
on the diagnosis and management of foot infections in persons
2. Apelqvist J, Larsson J, Agardh CD. Long-­term prognosis for diabetic
with diabetes. Diabetes Metab Res Rev. 2016;32(Suppl 1):45‐74.
patients with foot ulcers. J Intern Med. 1993;233:485‐491.
**Detailed review of most recent evidence in management of diabetic
3. Lipsky BA. Empirical therapy for diabetic foot infections: are there
foot infections
clinical clues to guide antibiotic selection? Clin Microbiol Infect.
25. Lipsky BA, Berendt AR, Cornia PB, et  al. 2012 Infectious Diseases
2007;13:351‐353.
Society of America clinical practice guideline for the diagno-
4. Richard JL, Sotto A, Lavigne JP. New insights in diabetic foot infec-
sis and treatment of diabetic foot infections. Clin Infect Dis.
tion. World J Diabetes. 2011;2:24‐32.
2012;54:e132‐e173. **Represents current best practice in understand-
5. Selva Olid A, Sola I, Barajas-Nava LA, Gianneo OD, Bonfill Cosp X,
ing when to utilise antimicrobial therapy in diabetic foot infection.
Lipsky BA. Systemic antibiotics for treating diabetic foot infections.
26. Lipsky BA, Berendt AR, Deery HG, et al. Diagnosis and treatment of
Cochrane Database Syst Rev. 2015;(9):Cd009061.
diabetic foot infections. Clin Infect Dis. 2004;39:885‐910.
6. Hartemann-Heurtier A, Senneville E. Diabetic foot osteomyelitis.
27. Lavery LA, Armstrong DG, Murdoch DP, Peters EJ, Lipsky BA.
Diabetes Metab. 2008;34:87‐95.
Validation of the Infectious Diseases Society of America’s diabetic
7. Foot Care for People with Diabetes: The Economic Case for Change
foot infection classification system. Clin Infect Dis. 2007;44:562‐565.
[press release]. UK: NHS Diabetes, 2011.
28. Pickwell K, Siersma V, Kars M, et al. Predictors of lower-­extremity
8. Boulton AJ, Vileikyte L, Ragnarson-Tennvall G, Apelqvist J. The global
amputation in patients with an infected diabetic foot ulcer. Diabetes
burden of diabetic foot disease. Lancet. 2005;366:1719‐1724.
Care. 2015;38:852‐857.
9. Young MJ, McCardle JE, Randall LE, Barclay JI. Improved sur-
29. Dinh MT, Abad CL, Safdar N. Diagnostic accuracy of the physical
vival of diabetic foot ulcer patients 1995-­2008: possible impact
examination and imaging tests for osteomyelitis underlying diabetic
of aggressive cardiovascular risk management. Diabetes Care.
foot ulcers: meta-­analysis. Clin Infect Dis. 2008;47:519‐527.
2008;31:2143‐2147.
30. Peters EJ. Pitfalls in diagnosing diabetic foot infections. Diabetes
10. Kennon B, Leese GP, Cochrane L, et al. Reduced incidence of lower-­
Metab Res Rev. 2016;32(Suppl 1):254‐260.
extremity amputations in people with diabetes in Scotland: a nation-
31. Lam K, van Asten SA, Nguyen T, La Fontaine J, Lavery LA. Diagnostic
wide study. Diabetes Care. 2012;35:2588‐2590.
accuracy of probe to bone to detect osteomyelitis in the diabetic
11. Habacher W, Rakovac I, Gorzer E, et al. A model to analyse costs and
foot: a systematic review. Clin Infect Dis. 2016;63:944‐948.
benefit of intensified diabetic foot care in Austria. J Eval Clin Pract.
32. Aragon-Sanchez J, Lipsky BA, Lazaro-Martinez JL. Diagnosing dia-
2007;13:906‐912.
betic foot osteomyelitis: is the combination of probe-­to-­bone test
12. Apelqvist J, Ragnarson-Tennvall G, Persson U, Larsson J. Diabetic foot
and plain radiography sufficient for high-­risk inpatients? Diabet Med.
ulcers in a multidisciplinary setting. An economic analysis of primary
2011;28:191‐194.
healing and healing with amputation. J Intern Med. 1994;235:463‐471.
33. Alvaro-Afonso FJ, Lazaro-Martinez JL, Aragon-Sanchez J, Garcia-
13. Apelqvist J, Larsson J. What is the most effective way to reduce in-
Morales E, Garcia-Alvarez Y, Molines-Barroso RJ. Inter-­observer re-
cidence of amputation in the diabetic foot? Diabetes Metab Res Rev.
producibility of diagnosis of diabetic foot osteomyelitis based on a
2000;16(Suppl 1):S75‐S83.
combination of probe-­to-­bone test and simple radiography. Diabetes
14. Hingorani A, LaMuraglia GM, Henke P, et al. The management of diabetic
Res Clin Pract. 2014;105:e3‐e5.
foot: a clinical practice guideline by the Society for Vascular Surgery in
34. Lesens O, Desbiez F, Vidal M, et al. Culture of per-­wound bone spec-
collaboration with the American Podiatric Medical Association and the
imens: a simplified approach for the medical management of diabetic
Society for Vascular Medicine. J Vasc Surg. 2016;63(2 Suppl):3s‐21s.
foot osteomyelitis. Clin Microbiol Infect. 2011;17:285‐291.
15. Bakker K, Apelqvist J, Schaper NC. Practical guidelines on the man-
35. Ertugrul BM, Lipsky BA, Savk O. Osteomyelitis or Charcot neuro-­
agement and prevention of the diabetic foot 2011. Diabetes Metab
osteoarthropathy? Differentiating these disorders in diabetic
Res Rev. 2012;28(Suppl 1):225‐231.
BARWELL et al. |
      9 of 10

patients with a foot problem. Diabet Foot Ankle. 2013, doi: 10.3402/ 54. Zeun P, Gooday C, Nunney I, Dhatariya K. Predictors of outcomes
dfa.v4i0.21855. in diabetic foot osteomyelitis treated initially with conservative
36. Rajbhandari SM, Sutton M, Davies C, Tesfaye S, Ward JD. ‘Sausage toe’: (nonsurgical) medical management: a retrospective study. Int J Low
a reliable sign of underlying osteomyelitis. Diabet Med. 2000;17:74‐77. Extrem Wounds. 2016;15:19‐25.
37. Epstein L, Mu Y, Belflower R, et  al. Risk factors for invasive 55. Dancer SJ. The effect of antibiotics on methicillin-­ resistant
methicillin-­resistant staphylococcus aureus infection after recent Staphylococcus aureus. J Antimicrob Chemother. 2008;61:246‐253.
discharge from an acute-­care hospitalization, 2011-­2013. Clin Infect 56. Dancer SJ. How antibiotics can make us sick: the less obvious ad-
Dis. 2016;62:45‐52. verse effects of antimicrobial chemotherapy. Lancet Infect Dis.
38. Marwick C, Santiago VH, McCowan C, Broomhall J, Davey P. 2004;4:611‐619.
Community acquired infections in older patients admitted to hospi- 57. McCarthy H, Rudkin JK, Black NS, Gallagher L, O’Neill E, O’Gara JP.
tal from care homes versus the community: cohort study of microbi- Methicillin resistance and the biofilm phenotype in Staphylococcus
ology and outcomes. BMC Geriatr. 2013;13:12. aureus. Front Cell Infect Microbiol. 2015;5:1.
39. Monteiro-Soares M, Martins-Mendes D, Vaz-Carneiro A, Sampaio 58. Bhinu VS. Insight into biofilm-­ associated microbial life. J Mol
S, Dinis-Ribeiro M. Classification systems for lower extremity am- Microbiol Biotechnol. 2005;10:15‐21.
putation prediction in subjects with active diabetic foot ulcer: 59. Dowd SE, Wolcott RD, Sun Y, McKeehan T, Smith E, Rhoads D.
a systematic review and meta-­ analysis. Diabetes Metab Res Rev. Polymicrobial nature of chronic diabetic foot ulcer biofilm infections
2014;30:610‐622. determined using bacterial tag encoded FLX amplicon pyrosequenc-
40. Ince P, Abbas ZG, Lutale JK, et al. Use of the SINBAD classification ing (bTEFAP). PLoS ONE. 2008;3:e3326.
system and score in comparing outcome of foot ulcer management 60. Bjarnsholt T. The role of bacterial biofilms in chronic infections.
on three continents. Diabetes Care. 2008;31:964‐967. APMIS Suppl. 2013;136:1‐51.
41. Chuan F, Tang K, Jiang P, Zhou B, He X. Reliability and validity of the 61. Sotto A, Richard JL, Messad N, et al. Distinguishing colonization from
perfusion, extent, depth, infection and sensation (PEDIS) classifica- infection with Staphylococcus aureus in diabetic foot ulcers with min-
tion system and score in patients with diabetic foot ulcer. PLoS ONE. iaturized oligonucleotide arrays: a French multicenter study. Diabetes
2015;10:e0124739. Care. 2012;35:617‐623.
42. O’Loughlin A, McIntosh C, Dinneen SF, O’Brien T. Review paper: 62. Dunyach-Remy C, Cadiere A, Richard JL, et  al. Polymerase chain
basic concepts to novel therapies: a review of the diabetic foot. Int J reaction-­ denaturing gradient gel electrophoresis (PCR-­ DGGE): a
Low Extrem Wounds. 2010;9:90‐102. promising tool to diagnose bacterial infections in diabetic foot ul-
43. Markanday A. Diagnosing diabetic foot osteomyelitis: narrative re- cers. Diabetes Metab. 2014;40:476‐480.
view and a suggested 2-­step score-­based diagnostic pathway for 63. Smith K, Collier A, Townsend EM, et  al. One step closer to under-
clinicians. Open Forum Infect Dis. 2014;1:ofu060. standing the role of bacteria in diabetic foot ulcers: characterising
44. Musa HG, Ahmed ME. Associated risk factors and management of the microbiome of ulcers. BMC Microbiol. 2016;16:54.
chronic diabetic foot ulcers exceeding 6 months’ duration. Diabet 64. van Asten SA, La Fontaine J, Peters EJ, Bhavan K, Kim PJ, Lavery LA.
Foot Ankle. 2012, doi: 10.3402/dfa.v3i0.18980. The microbiome of diabetic foot osteomyelitis. Eur J Clin Microbiol
45. Palestro CJ, Love C. Radionuclide imaging of musculoskeletal infection: Infect Dis. 2016;35:293‐298.
conventional agents. Semin Musculoskelet Radiol. 2007;11:335‐352. 65. Lavery LA, Fontaine JL, Bhavan K, Kim PJ, Williams JR, Hunt NA. Risk
46. Palestro CJ, Love C. Nuclear medicine and diabetic foot infections. factors for methicillin-­resistant Staphylococcus aureus in diabetic foot
Semin Nucl Med. 2009;39:52‐65. infections. Diabet Foot Ankle. 2014, doi: 10.3402/dfa.v5.23575.
47. Zavadovskaya VD, Zorkal’tsev MA, Udodov VD, et al. [Possibilities of 66. Stanaway S, Johnson D, Moulik P, Gill G. Methicillin-­ resistant
a software-­based hybrid single photon emission computed tomog- Staphylococcus aureus (MRSA) isolation from diabetic foot ul-
raphy/magnetic resonance imaging in the diagnosis of complicated cers correlates with nasal MRSA carriage. Diabetes Res Clin Pract.
diabetic foot syndrome]. Vestn Rentgenol Radiol. 2015;(6):24‐29. 2007;75:47‐50.
48. Aslangul E, M’Bemba J, Caillat-Vigneron N, et al. Diagnosing diabetic 67. Mottola C, Mendes JJ, Cristino JM, Cavaco-Silva P, Tavares L,
foot osteomyelitis in patients without signs of soft tissue infection Oliveira M. Polymicrobial biofilms by diabetic foot clinical isolates.
by coupling hybrid 67 Ga SPECT/CT with bedside percutaneous Folia Microbiol. 2016;61:35‐43.
bone puncture. Diabetes Care. 2013;36:2203‐2210. **Recommended 68. Uckay I, Aragon-Sanchez J, Lew D, Lipsky BA. Diabetic foot infec-
to encourage development of local/regional guidelines combining en- tions: what have we learned in the last 30 years? Int J Infect Dis.
hanced imaging and bone biopsy. 2015;40:81‐91.
49. Senneville E, Melliez H, Beltrand E, et  al. Culture of percutaneous 69. Hesstvedt L, Arendrup MC, Poikonen E, Klingpor L, Friman V, Nordoy
bone biopsy specimens for diagnosis of diabetic foot osteomyelitis: I. Differences in epidemiology of candidaemia in the Nordic coun-
concordance with ulcer swab cultures. Clin Infect Dis. 2006;42:57‐62. tries -­what is to blame? Mycoses. 2017;60:11‐19.
50. Senneville E, Morant H, Descamps D, et al. Needle puncture and 70. Hsueh PR, Chen WH, Luh KT. Relationships between antimicrobial
transcutaneous bone biopsy cultures are inconsistent in patients use and antimicrobial resistance in Gram-­negative bacteria causing
with diabetes and suspected osteomyelitis of the foot. Clin Infect Dis. nosocomial infections from 1991-­2003 at a university hospital in
2009;48:888‐893. Taiwan. Int J Antimicrob Agents. 2005;26:463‐472.
51. Valabhji J, Oliver N, Samarasinghe D, Mali T, Gibbs RG, Gedroyc WM. 71. Lee CM, Lai CC, Wang YY, Lee MC, Hsueh PR. Impact of susceptibil-
Conservative management of diabetic forefoot ulceration compli- ity profiles of Gram-­negative bacteria before and after the introduc-
cated by underlying osteomyelitis: the benefits of magnetic reso- tion of ertapenem at a medical center in northern Taiwan from 2004
nance imaging. Diabet Med. 2009;26:1127‐1134. to 2010. Diagn Microbiol Infect Dis. 2013;75:94‐100.
52. Game FL, Jeffcoate WJ. Primarily non-­surgical management of os- 72. Taori SK, Wroe A, Hardie A, Gibb AP, Poxton IR. A prospective study
teomyelitis of the foot in diabetes. Diabetologia. 2008;51:962‐967. of community-­associated Clostridium difficile infections: the role of
53. Tone A, Nguyen S, Devemy F, et  al. Six-­week versus twelve-­week antibiotics and co-­infections. J Infect. 2014;69:134‐144.
antibiotic therapy for nonsurgically treated diabetic foot osteomyeli- 73. Reilly J, Cairns S, Fleming S, et al. Results from the second
tis: a multicenter open-­label controlled randomized study. Diabetes Scottish national prevalence survey: the changing epidemiol-
Care. 2015;38:302‐307. **Supports consideration of shorter courses of ogy of healthcare-­associated infection in Scotland. J Hosp Infect.
antimicrobial therapy for osteomyelitis 2012;82:170‐174.
|
10 of 10       BARWELL et al.

74. Bloomfield LE, Riley TV. Epidemiology and risk factors for community-­ 95. Lawes T, Lopez-Lozano JM, Nebot CA, et  al. Effects of na-
associated Clostridium difficile infection: a narrative review. Infect Dis tional antibiotic stewardship and infection control strategies on
Ther. 2016;5:231‐251. hospital-­associated and community-­associated meticillin-­resistant
75. Fellmeth G, Yarlagadda S, Iyer S. Epidemiology of community-­onset Staphylococcus aureus infections across a region of Scotland: a non-­
Clostridium difficile infection in a community in the South of England. linear time-­series study. Lancet Infect Dis. 2015;15:1438‐1449.
J Infect Public Health. 2010;3:118‐123. 96. Cairns S, Packer S, Reilly J, Leanord A. Targeted MRSA screening can
76. Bignardi GE. Risk factors for Clostridium difficile infection. J Hosp be as effective as universal screening. BMJ. 2014;349:g5075.
Infect. 1998;40:1‐15. 97. Lawes T, Edwards B, Lopez-Lozano JM, Gould I. Trends in
77. Collier A, McLaren J, Godwin J, Bal A. Is Clostridium difficile associ- Staphylococcus aureus bacteraemia and impacts of infection control
ated with the ‘4C’ antibiotics? A retrospective observational study in practices including universal MRSA admission screening in a hospital
diabetic foot ulcer patients. Int J Clin Pract. 2014;68:628‐632. in Scotland, 2006-­2010: retrospective cohort study and time-­series
78. Noor S, Ahmad J, Parwez I, Ozair M. Culture-­based screening of intervention analysis. BMJ Open 2012;2:e000797. doi: 10.1136/
aerobic microbiome in diabetic foot subjects and developing non-­ bmjopen-2011-000797
healing ulcers. Front Microbiol. 2016;7:1792. 98. Tsai LH, Weng YM, Lin CC, Kuo CW, Chen JC. Risk screening for long
79. Hatipoglu M, Mutluoglu M, Turhan V, et al. Causative pathogens and QT prior to prescribing levofloxacin. Am J Emerg Med. 2014;32:1153.
antibiotic resistance in diabetic foot infections: a prospective multi-­ e1-3.
center study. J Diabetes Complications. 2016;30:910‐916. 99. Panicker GK, Karnad DR, Kadam P, Badilini F, Damle A, Kothari S.
80. Grigoropoulou P, Eleftheriadou I, Jude EB, Tentolouris N. Diabetic Detecting moxifloxacin-­induced QTc prolongation in thorough QT
foot infections: an update in diagnosis and management. Curr and early clinical phase studies using a highly automated ECG analy-
DiabRep. 2017;17:3. sis approach. Br J Pharmacol. 2016;173:1373‐1380.
81. Peters EJ, Lipsky BA, Berendt AR, et al. A systematic review of the 100. Yanat B, Rodriguez-Martinez JM, Touati A. Plasmid-­mediated quino-
effectiveness of interventions in the management of infection in the lone resistance in Enterobacteriaceae: a systematic review with
diabetic foot. Diabetes Metab Res Rev. 2012;28(Suppl 1):142‐162. a focus on Mediterranean countries. Eur J Clin Microbiol Infect Dis.
82. Acharya S, Soliman M, Egun A, Rajbhandari SM. Conservative man- 2017;36:421‐435.
agement of diabetic foot osteomyelitis. Diabetes Res Clin Pract. 101. Weber SG, Gold HS, Hooper DC, Karchmer AW, Carmeli Y.
2013;101:e18‐e20. Fluoroquinolones and the risk for methicillin-­resistant Staphylococcus
83. Lazaro-Martinez JL, Aragon-Sanchez J, Garcia-Morales E. Antibiotics aureus in hospitalized patients. Emerg Infect Dis. 2003;9:1415‐1422.
versus conservative surgery for treating diabetic foot osteomyelitis: 102. Brown KA, Khanafer N, Daneman N, Fisman DN. Meta-­analysis of
a randomized comparative trial. Diabetes Care. 2014;37:789‐795. antibiotics and the risk of community-­associated Clostridium difficile
84. Li HK, Scarborough M, Zambellas R, et  al. Oral versus intravenous infection. Antimicrob Agents Chemother. 2013;57:2326‐2332.
antibiotic treatment for bone and joint infections (OVIVA): study 103. Kim BN, Kim ES, Oh MD. Oral antibiotic treatment of staphylo-
protocol for a randomised controlled trial. Trials. 2015;16:583. coccal bone and joint infections in adults. J Antimicrob Chemother.
85. Conterno LO, Turchi MD. Antibiotics for treating chronic osteomyeli- 2014;69:309‐322.
tis in adults. Cochrane Database Syst Rev. 2013;9:Cd004439. 104. Al-Rubeaan K, Al Derwish M, Ouizi S, et al. Diabetic foot complica-
86. Crouzet J, Lavigne JP, Richard JL, Sotto A. Diabetic foot infection: a tions and their risk factors from a large retrospective cohort study.
critical review of recent randomized clinical trials on antibiotic ther- PLoS ONE. 2015;10:e0124446.
apy. Int J Infect Dis. 2011;15:e601‐e610. 105. Calderon-Ortiz R, Colton-Verge P, Muniz-Ortega M, Lespier L,
87. Jeffcoate WJ, Bus SA, Game FL, Hinchliffe RJ, Price PE, Schaper NC. Cordova H. Life threatening hyperkalemia chronic kidney diseases
Reporting standards of studies and papers on the prevention and patients treated with trimethoprim-­sulfamethoxazole: a case series.
management of foot ulcers in diabetes: required details and markers Bol Asoc Med P R. 2011;103:15‐17.
of good quality. Lancet Diabetes Endocrinol. 2016;4:781‐788. 106. Tamma PD, Cosgrove SE, Maragakis LL. Combination therapy for
88. Tascini C, Piaggesi A, Tagliaferri E, et al. Microbiology at first visit of treatment of infections with gram-­negative bacteria. Clin Microbiol
moderate-­to-­severe diabetic foot infection with antimicrobial activ- Rev. 2012;25:450‐470.
ity and a survey of quinolone monotherapy. Diabetes Res Clin Pract. 107. Chapman AL, Dixon S, Andrews D, Lillie PJ, Bazaz R, Patchett JD.
2011;94:133‐139. Clinical efficacy and cost-­effectiveness of outpatient parenteral an-
89. Wu WS, Chen CC, Chuang YC, et  al. Efficacy of combination oral tibiotic therapy (OPAT): a UK perspective. J Antimicrob Chemother.
antimicrobial agents against biofilm-­embedded methicillin-­resistant 2009;64:1316‐1324.
Staphylococcus aureus. J Microbiol Immunol Infect. 2013;46:89‐95. 108. Chapman AL, Seaton RA, Cooper MA, et  al. Good practice recom-
90. Iguchi S, Mizutani T, Hiramatsu K, Kikuchi K. Rapid Acquisition of mendations for outpatient parenteral antimicrobial therapy (OPAT)
Linezolid Resistance in Methicillin-­Resistant Staphylococcus aureus: in adults in the UK: a consensus statement. J Antimicrob Chemother.
Role of Hypermutation and Homologous Recombination. PLoS ONE. 2012;67:1053‐1062. **Useful evidence for local clinicians/groups to
2016;11:e0155512. support the development of OPAT service
91. Walker KTR, Sandeep TC, Devers M, Bain D, Barwell ND. Linezolid-­ 109. Schaper NC. Diabetic foot ulcer classification system for research
induced hypoglycaemia in a patient with Type 1 Diabetes. Br J purposes: a progress report on criteria for including patients in re-
Diabetes Vasc Dis. 2012;12:186‐188. search studies. Diabetes Metab Res Rev. 2004;20(Suppl 1):S90‐S95.
92. Young MJ, Hodges G, McCardle JE. Cost avoidance using linezolid for
methicillin-­resistant Staphylococcus aureus infections in a specialist
diabetes foot clinic. J Antimicrob Chemother. 2012;67:2974‐2975. How to cite this article: Barwell ND, Devers MC, Kennon B,
93. Evans JD, Udeani G, Cole P, Friedland HD. Ceftaroline fosamil for et al; on behalf of the Scottish Diabetes Foot Action Group.
the treatment of acute bacterial skin and skin structure infections in
Diabetic foot infection: Antibiotic therapy and good practice
obese patients. Postgrad Med. 2014;126:128‐134.
94. Lipsky BA, Cannon CM, Ramani A, et al. Ceftaroline fosamil for treat- recommendations. Int J Clin Pract. 2017;e13006. https://doi.
ment of diabetic foot infections: the CAPTURE study experience. org/10.1111/ijcp.13006
Diabetes Metab Res Rev. 2015;31:395‐401.