Cellulitis - Management of Infection Guidance for Primary Care

Publication: 30/09/2010  --
Last review: 04/09/2012  
Next review: 01/04/2018  
Clinical Guideline
ID: 2256 
Approved By: Trust Clinical Guidelines Group 
Copyright© Leeds Teaching Hospitals NHS Trust 2012  


This Clinical Guideline is intended for use by healthcare professionals within Leeds unless otherwise stated.
For healthcare professionals in other trusts, please ensure that you consult relevant local and national guidance.


Illness Comments Preferred option Alternative

If patient afebrile and healthy other than cellulitis, use oral flucloxacillin alone 1,2C

If febrile and ill, admit for IV treatment 1 (C)

For patients with recurrent cellulitis, look for and manage any underlying condition (e.g. lymphoedema). Wound swabs in such patients may be useful to target antibiotic therapy.

Stop Clindamycin and seek alternative if diarrhoea occurs.


Flucloxacillin electronic Medicines Compendium information on Flucloxacillin
500mg – 1g QDS







If exposure to seawater/coral injury:

Doxycycline electronic Medicines Compendium information on Doxycycline
200mg OD

If history of MRSA colonisation or MRSA grown from wound swab:

Doxycycline electronic Medicines Compendium information on Doxycycline
200mg OD


7 days (may be extended up to 14 days depending on clinical response).1,2,3C

Doxycycline electronic Medicines Compendium information on Doxycycline
200mg OD

Over 65 years:
Clarithromycin electronic Medicines Compendium information on Clarithromycin
250-500mg BD

Under 65 years:
Clindamycin electronic Medicines Compendium information on Clindamycin
300–450 mg QDS










7 days (may be extended up to 14 days depending on clinical response).1,2,3C

Clinical assessment
Assess the skin and soft tissue for evidence of cellulitis. If there is a clinical diagnosis of cellulitis consider empirical therapy.

Microbiological assessment
A microbiological diagnosis of cellulitis is often not possible. If there is no wound, pus or discharge to swab, a skin swab is unlikely to identify the microbiological cause of the cellulitis. If there is a wound, pus or discharge and the infection is of a moderate or greater severity send a swab for microbiological culture and sensitivity.

Microbiological interpretation
A number of bacteria may be found on a skin swab. The common bacteria are listed below along with their significance. They are group into pathogens, bacteria which are likely to be causing cellulitis when this has been clinically diagnosed, and colonisers, these are unlikely to be causing cellulitis. Pathogens can colonise the skin so their identification must be correlated with a clinical diagnosis of infection before treating the patient. Colonisers can also cause infection in some circumstances e.g. immunosuppression, vascular disease. Treating of colonisers should not be completed unless there is a risk factor for infection with these bacteria and a patient has failed empirical therapy against the pathogen group.

Staphylococcus aureus: is a common cause of cellulitis. It can though colonise the skin so treatment should not be initiated because it is identified in a microbiology sample unless there is clinical evidence of infection.

Group A streptococcus (S. pyogenes): is the most common cause of cellulitis. It can though colonise the skin so treatment should not be initiated because it is identified in a microbiology sample unless there is clinical evidence of infection. 

Skin flora: These include coagulase negative staphylococci, diphtheroids and propionibacteria. These are skin commensals and are not causes of cellulitis.

Enteric flora: These include E. coli and Klebsiella spp. It is common for wounds to become colonised with these bacteria. They are rarely a cause of cellulitis. If a patient has not responded to empirical antimicrobial therapy and these bacteria are cultured as the predominant organism, i.e. reported as the predominant organism and not as mixed enteric flora, please discuss with a microbiologist if you are considering treatment.

Pseudomonas: P. aeruginosa is a common coloniser of the skin, especially moist wounds. It is an uncommon cause of cellulitis except with predisposing conditions e.g. immunosuppression/vascular disease. Treatment for P. aeruginosa is not routinely indicated. If a patient has failed to respond to empirical therapy for cellulitis and has a risk factor for infection and it has been isolated as a predominant organism in culture please contact microbiology to discuss treatment.

Anaerobes: Anaerobes are a common coloniser of wounds. This colonisation can result in an unpleasant wound smell which can be treated with topical metronidazole. They are not common causes of cellulitis in isolation but if felt to be causing soft tissue infection can be treated with oral metronidazole in addition to standard therapy for cellulitis.

Sensitivities available on request
Some bacteria are uncommon causes of skin and soft tissue infection. We do not routinely release sensitivities for these bacteria to prevent overuse of antibiotics. If you would like to treat a patient whose sample has grown one of these uncommon causes of skin and soft tissue infection please contact microbiology on (0113) 392 3962.

Evidence level=C 

Principles of Treatment

  1. This guidance is based on the best available evidence but its application must be modified by professional judgement.
  2. A dose and duration of treatment is suggested. In severe or recurrent cases consider a larger dose or longer course
  3. Choices are given as Preferred option or Alternative for patients intolerant of the preferred option and 2nd Line for when an alternative is required because of treatment failure
  4. Only send microbiology specimens if there is a clinical suspicion of infection. Inappropriate specimens (e.g. routine ulcer swab or routine catheter specimen of urine) lead to inappropriate antibiotic prescribing.
  5. Prescribe an antibiotic only when there is likely to be a clear clinical benefit.
  6. Limit prescribing over the telephone to exceptional cases.
  7. Use simple generic antibiotics if possible. Avoid broad spectrum antibiotics (e.g. co-amoxiclav, quinolones and cephalosporins) when narrow spectrum antibiotics remain effective, as they increase risk of Clostridium difficile, MRSA and resistant UTI's.
  8. Avoid widespread use of topical antibiotics (especially those agents also available as systemic preparations).
  9. In pregnancy AVOID tetracyclines, aminoglycosides, quinolones, and high dose metronidazole. Short-term use of trimethoprim (unless low folate status or taking another folate antagonist such as antiepileptic or proguanil) or nitrofurantoin (at term, theoretical risk of neonatal haemolysis) is unlikely to cause problems to the foetus.
  10. We recommend clarithromycin as it has less side-effects than erythromycin, greater compliance as twice rather than four times daily & generic tablets are similar cost. In children erythromycin may be preferable as clarithromycin syrup is twice the cost.
  11. Where a ‘best guess’ therapy has failed or special circumstances exist, microbiological advice can be obtained from LTHT Microbiology (Mon-Fri 9am-5pm and Sat and Sun 9am-1pm: (0113) 39 23962/28580; Otherwise via LTHT switchboard - ask for the On call Microbiology Registrar)


Note: Doses are oral and for adults unless otherwise stated. Please refer to BNF for further information.
Letters indicate strength of evidence:
A+ = systematic review: D = informal opinion

Back to top


Record: 2256
  • to provide a simple, empirical approach to the treatment of common infections
  • to promote the safe, effective and economic use of antibiotics
  • to minimise the emergence of bacterial resistance and reduce the incidence of Healthcare Associated Infections in the community
Clinical condition:


Target patient group:
Target professional group(s): Primary Care Doctors
Adapted from:

This guidance was initially developed in 1999 by practitioners in South Devon, as part of the S&W Devon Joint Formulary Initiative, and Cheltenham & Tewkesbury Prescribing Group and modified by the PHLS South West Antibiotic Guidelines Project Team, PHLS Primary Care Co-ordinators and members of the Clinical Prescribing Sub-group of the Standing Medical Advisory Committee on Antibiotic Resistance. It was further modified following comments from Internet users. The guidance has been updated annually as significant research papers, systematic reviews and guidance have been published. The Health Protection Agency works closely with the authors of the Clinical Knowledge Summaries.

Evidence base

Grading of guidance recommendations
The strength of each recommendation is qualified by a letter in parenthesis.

Study design

Recommendation grade

Good recent systematic review of studies


One or more rigorous studies, not combined


One or more prospective studies


One or more retrospective studies


Formal combination of expert opinion


Informal opinion, other information


Clinical Knowledge Summaries web. BNF (No 55), SMAC report - The path of least resistance (1998), SDHCT Medical Directorate guidelines + GU medicine guidelines, Plymouth Management of Infection Guidelines project LRTI and URTI.


  1. CREST Guidelines on the management of cellulitis in adults. Clinical Resource Efficiency Support Team. 2005. Expert consensus is that people who have no signs of systemic toxicity and no uncontrolled co-morbidities can usually be managed on an outpatient basis with oral antibiotics. Flucloxacillin 500mg QDS (or clarithromycin 500mg BD for those with penicillin allergy) are suitable oral antibiotics because they cover staphylococci and streptococci, the most commonly implicated pathogens. Clindamycin 300mg QDS is also recommended as a further alternative for people with penicillin allergy. Most cases of uncomplicated cellulitis can be treated successfully with 1-2 weeks of treatment.
  2. Jones, G.R. Principles and practice of antibiotic therapy for cellulitis. CPD Journal Acute Medicine. 2002; 1(2): 44-49. Oral agents will be as effective as intravenous agents for cellulitis if they can maintain the free antibiotic level above the MIC of the pathogen for more than 40% of the dose interval. Flucloxacillin 500 mg, clarithromycin 500 mg and clindamycin 300 mg are suitable oral doses.
  3. Morris AD. Cellulitis and erysipelas. Clinical Evidence. 2007. London. BMJ Publishing Group. This systematic review found no RCTs of antibiotics compared with placebo of sufficient quality for inclusion. Although 11 RCTs were identified that compared antibiotic treatments, these studies were small and only powered to demonstrate equivalence, not superiority, between antibiotics. Two RCTs using intravenous flucloxacillin were found, but none using oral flucloxacillin. Oral azithromycin was compared with erythromycin, cloxacillin, and cefalexin in three RCTs. Oral co-amoxiclav was compared with fleroxacin in one sub-group analysis.
  4. Fischer RG and Benjamin DK Jr. Facial cellulitis in childhood: a changing spectrum. Southern Medical Journal. 2002; 95: 672-674. Buccal cellultis is commonly due to Haemophilus influenzae infection, although rates are decreasing following the Hib immunization programme. The Health Protection Agency and the Association of Medical Microbiologists recommends co-amoxiclav for empirical treatment of facial cellulitis because it is broader spectrum than flucloxacillin and also covers H. influenzae.

Back to top

Approved By

Trust Clinical Guidelines Group

Document history

LHP version 1.0

Related information

Not supplied