Published on 19/03/2015 by admin
Filed under Dermatology
Last modified 22/04/2025
This article have been viewed 4161 times
Adrian H.M. Heagerty
Evidence Levels: A Double-blind study B Clinical trial ≥ 20 subjects C Clinical trial < 20 subjects D Series ≥ 5 subjects E Anecdotal case reports
Cellulitis is strictly an acute, subacute, or chronic infection of the subcutaneous tissues, whereas erysipelas is an infection of the dermis and superficial subcutis. Infection of the more superficial layers gives rise to superficial edema and inflammation, with the consequent development of a palpable, often advancing edge. The causative organism is usually regarded as Streptococcus, though many organisms have been isolated, including Haemophilus influenzae, and more rarely staphylococci, Aeromonas hydrophilia, and Pseudomonas aeruginosa, as well as fungi and Gram-negative rods. Fulminating and necrotic cellulitis and fasciitis may occur rarely, usually in relation to immunosuppression or atypical organisms. These are rare, but necrotizing fasciitis may have a mortality of up to 50%.
The management of cellulitis and erysipelas should initially be directed to trying to identify the organism responsible for the infection, and then directing appropriate antimicrobial therapy. Any underlying and predisposing condition should be identified and treated to prevent subsequent recurrence. Perhaps the commonest condition that is not identified and treated is toe web tinea pedis, which provides a portal of entry for infection.
Uncomplicated cellulitis and erysipelas may be managed without admission if the patient does not exhibit signs of systemic toxicity. In such cases oral broad-spectrum antibiotics, chosen to cover group A streptococci and staphylococci, may be sufficient, supplemented with a single parenteral loading dose or long-acting preparation. The drug of choice is oral penicillin V (phenoxymethylpenicillin) with or without flucloxacillin, or erythromycin, if the patient has a known penicillin allergy. Newer macrolides, such as clarithromycin, may be more acceptable with fewer side effects. Some authorities have recommended the use of clindamycin rather than a macrolide because of apparent increased tissue penetration, but this may be associated with an increased incidence of Clostridium difficile superinfection. Further, although most group A β-hemolytic streptococci are sensitive to this, increasing numbers of MRSA are displaying resistance to clindamycin, and widespread use may exacerbate this resistance.
Immunocompromised patients, those with signs of systemic toxicity, and otherwise debilitated patients should be treated as inpatients with intravenous antimicrobials (e.g., penicillin G – benzylpenicillin) or one of the newer antibiotics (e.g., ciprofloxacin, ticarcillin, teicoplanin, or imipenem/cilastatin). If there is evidence of head and neck disease or sinus infection, amoxicillin combined with clavulanic acid should be considered to cover H. influenzae infection.
Sites of entry for infection should be sought, such as excoriations in eczema or following trauma, and these should be treated.
Blood cultures
ASO titer/anti-DNase B
Cultures of aspirates and lesions
Skin scrapings for mycology
Blood cultures may be positive and significant in only about 25% of cases, but should always be taken if there is any evidence of systemic toxicity. Swabs of wounds and broken skin may be helpful, but surface swabs of unbroken skin provide little or no useful information. If available, aspirate of bullae may yield positive cultures. Slightly better rates for isolation than those of needle aspirates have been achieved with punch skin biopsies.
Rising titers of streptococcal antibodies (ASO titer and anti-DNase B) may be helpful, but are more commonly of value retrospectively.
In the case of cellulitis or erysipelas of the lower leg, skin scrapings from toe webs should be taken for mycological examination. Facial erysipelas should warrant sinus radiographs to exclude underlying sinusitis. Crepitus should prompt the clinician to the presence of either clostridia or non-spore-forming anaerobes, either alone or mixed with other bacteria such as Pseudomonas, Escherichia coli, or Klebsiella spp.
Jorup-Ronstrum C, Britton S, Gavlevik A, Gunnarsson K, Redman AC. Infection 1984; 12: 390–4.
In this study of 60 patients there appeared to be no appreciable benefit from intravenous rather than oral therapy with penicillin for erysipelas, and so oral therapy is recommended if there are no associated complications with the infection.
Cox NH, Colver GB, Paterson WD. J R Soc Med 1998; 91: 634–7.
A case note review of 92 patients admitted for inpatient care for ascending cellulitis of the leg revealed a portal of entry, most commonly minor injury. The mean hospital stay was 10 days. Bacteriology was seldom helpful, but group G streptococci were the most frequently identified pathogens. Benzylpenicillin was used in 43 cases (46%). The authors emphasize the need for benzylpenicillin, treatment of tinea pedis, and retrospective diagnosis of streptococcal infection by serology.
Aly AA, Roberts NM, Seipol KS, MacLellan DG. Med J Aust 1996; 165: 553–6.
This retrospective survey examined the management of 118 patients with lower limb cellulitis in a tertiary teaching hospital. In 79% of cases there was underlying disease, but only 20% were investigated. Blood cultures were taken from 55%, all with negative results. A combination of flucloxacillin and penicillin was given intravenously for a mean of 6 days to 76% of patients, and where documented 94% responded within 5 days. However, 40% of patients had intravenous therapy for longer than this, and 10% for 10 days or more. The length of inpatient stay averaged 13 days, prolonged stay being associated with surgical intervention or intercurrent problems, but 15% of patients had no clear indication for an extended stay. The authors concluded that excessive microbiological investigations, inadequate investigation, and treatment of underlying disease with prolonged use of intravenous antibiotics and questionable use of combinations of antibiotic therapy were common.
Sjoblom AC, Bruchfeld J, Eriksson B, Jorup-Rönström C, Karkkonen K, Malmborg AS, et al. Infection 1992; 20: 30–3.
Tissue and serum blood levels were measured in 45 patients with erysipelas after oral penicillin (phenoxymethylpenicillin); the minimal inhibitory concentrations were exceeded for streptococci isolated, supporting the role of oral therapy.
Vinen J, Hudson B, Chan B, Vinen J, Hudson B, Chan B, Fernandes C. Clin Drug Invest 1996; 12: 221–5.
A randomized comparative study in 58 patients with cellulitis; intravenous ceftriaxone cured 92%, but intravenous flucloxacillin cured only 64% after 4 to 6 days.
Bernard P, Plantin P, Roger H, Sassolas B, Villaret E, Legrain V, et al. Br J Dermatol 1992; 127: 155–9.
This prospective randomized multicenter trial compared oral roxithromycin with intravenous benzylpenicillin. Overall efficacy was similar.
Fleischer GR, Wilmott CM, Capos JM. Antimicrob Agents Chemother 1983; 24: 679–81.
Amoxicillin with clavulanic acid was compared with cefaclor in children with impetigo and cellulitis due to staphylococci, streptococci, and Haemophilus spp. There was a 100% response to therapy with the combination, compared to 90% with the cephalosporin; the incidence of relapse and re-infection and side effects was small, but greater with the combination therapy.
Seaton RA, Bell E, Gourlay Y, Semple L. J Antimicrob Chemother 2005; 55: 764–7.
The safety and efficacy of a nurse-led outpatient parenteral antibiotic therapy service for cellulitis were examined in 114 patients and 230 historical controls. No alteration in outcomes, complications, or re-admission rates was seen compared to the earlier physician-supervised outpatient treatment. Treatment duration was reduced from 4 to 3 days.
Gouin S, Chevalier I, Gautier M, Lamarre V. Paediatr Child Health 2008; 44: 214–18.
The clinical outcomes of 92 children receiving outpatient treatment in a day treatment center were examined prospectively, after a mean of 2.5 days of intravenous therapy 73 patients (79.3%) were switched to oral agents and discharged from the day treatment center. There were no relapses in this group.
There is a trend towards increasing use of parenteral antibiotics given out of hospital for the treatment of cellulitis, either at home or in a day treatment center.
Wood MJ, Logan MN. J Antimicrob Chemother 1986; 18: 159–64.
Twenty-one patients with cellulitis or other soft tissue infection were treated with oral ciprofloxacin; 19 were clinically cured or improved, and one was withdrawn from the study because of nausea and vomiting. Nine of the original 18 bacterial isolates were eradicated, but the majority of failures were due to staphylococci and streptococci.
Turpin PJ, Taylor GP, Logan MN, Wood MJ. J Antimicrob Chemother 1988; 21: 117–22.
Twenty-four patients with cellulitis or other soft tissue infection were treated with once-daily teicoplanin, resulting in clinical cure or improvement without severe adverse reactions, but with a rise in the plasma platelet count.
Sexton DJ, Wlodaver CG, Tobey LE, Yangco BG, Graziani AL, MacGregor RR. Chemotherapy 1991; 37: 26–30.
Of 102 patients enrolled in this study with mild to moderately severe skin and soft tissue infections, 74 were evaluable, with 20 having cellulitis, 23 wound infections, and 31 abscesses. Imipenem/cilastatin was given intramuscularly using doses of 500 or 750 mg 12-hourly. In this study there was no assessment by type of infection, but 82% were cured and 16% improved. Eight patients reported minor side effects.
Stevens DL, Smith LG, Bruss JB, McConnell-Martin MA, Duvall SE, Todd WM, Hafkin B. Antimicrob Agents Chemother 2000: 12: 3408–13.
Eight hundred and twenty-six hospitalized adult patients were randomized to receive linezolid (600 mg intravenously) every 12 hours or oxacillin (2 g intravenously) every 6 hours; following sufficient clinical improvement, patients were switched to the respective oral agents (linezolid 600 mg orally every 12 hours or dicloxacillin 500 mg orally every 6 hours). The clinical cure rates were 88.6% and 85.8%, respectively.
Bergkvist PI, Sjobeck K. Scand J Infect Dis 1997; 29: 377–82
Although prednisolone may predispose to infection, its use in combination with intravenous antibiotics reduced the median time to cure by 1 day (5 vs 6 days); at the 90th centile healing time was 10 days vs 14.6 days, and median hospital stay was reduced from 6 to 5 days. The relapse rate within 3 weeks was approximately the same in both groups.
Gough A, Clapperton M, Rolando N, Foster A, Philpott-Howard J, Edmonds ME. Lancet 1997; 350: 855–9.
This randomized controlled trial compared the ability of G-CSF to improve clinical outcome in the treatment of cellulitis in diabetes mellitus, using resolution as the end-point. G-CSF stimulates the neutrophil response, which is impaired in diabetes mellitus, but is important for defence against infection. The risk is principally that of high white cell counts, which may predispose to coronary and cerebral vascular events.
Mathur MN, Patrick WG, Unsworth IP, Bennett FM. Aust NZ J Surg 1995; 65: 367–9.
This case report of Aeromonas hydrophilia cellulitis, unresponsive to antibiotics and surgical debridement, responded to hyperbaric oxygen therapy. Although there are few objective reports of similar treatment in streptococcal necrotizing fasciitis, it has been suggested that in all types of necrotizing fasciitis hyperbaric oxygen reduces mortality.
Duvanel T, Merot Y, Saurat JH. Lancet 1985; 1: 1401.
Sixteen patients who received weekly intramuscular penicillin as prophylaxis were followed and assessed at 2 years. On cessation of prophylaxis the risk of recurrence rapidly returned to the non-treatment/no prophylaxis level.
Morris A. Clin Evid 2004; 12: 2268–74.
Although there is a consensus that successful treatment of predisposing factors such as leg edema, tinea pedis, and traumatic wounds reduces the risk of developing cellulitis, there are no randomized controlled trials or observational studies to support these.
UK Dermatology Clinical Trials Network’s PATCH Trial Team. Br J Dermatol 2012; 166: 169–78.
A double-blind, randomized controlled trial examining whether prophylactic antibiotics prescribed after an episode of cellulitis of the leg can prevent further episodes. A total of 123 patients were recruited and randomized between low dose oral phenoxymethyl penicillin and placebo for 6 months. Recurrence rates were 20% on penicillin vs 33% on placebo (p = 0.08).
Treatment of Skin Disease Comprehensive Therapeutic Strategies 4e
WhatsApp us
Penicillin G
Penicillin G with flucloxacillin
Penicillin V
Amoxicillin with clavulanic acid
Ceftriaxone
Roxithromycin
Ciprofloxacin
Teicoplanin
Imipenem/cilastatin
Linezolid
Oxacillin/dicloxacillin
Prednisolone as an adjunct to antibiotics
Granulocyte colony-stimulating factor
Hyperbaric oxygen
Intramuscular weekly penicillin
Treatment of predisposing factors
