Are You Confident of the Diagnosis?

Cellulitis is a diffuse acute inflammatory infectious process that involves the dermal and subcutaneous layers of the skin. Erysipelas is a specific varient of cellulitis with characteristic features that is generally superficial and caused by beta-hemolytic streptococci. The spectrum of cellulitis also includes cellulitis with gangrene or necrosis, and skin infections characterized predominantly by abscess formation; these are discussed elsewhere. Other related conditions include impetigo and lymphangitis.

What you should be alert for in the history

The key components of the history, in addition to the presenting symptoms, include exposures or preceding events such as recent major or minor trauma, surgery or other breaks in the skin (Figure 1), and potential exposure to unusual pathogens, for example via animal scratches or bites, other environmental exposures, or contact with individuals with streptococcal or staphylococcal infections.

Figure 1.

Cellulitis of the lower leg after a punch biopsy of the skin. (Courtesy of Bryan Anderson, MD)

Continue Reading

The other key historical components are those related to underlying host predispositions for infection This includes local anatomic factors that can increase risk for cellulitis, such as history of chronic skin disorders, chronic edema or lymphedema, chronic open wounds, or peripheral vascular disease, and also systemic processes that increase the risk of development of more severe infection.

Characteristic findings on physical examination

Findings on physical examination include induration and swelling, erythema, warmth and tenderness and may be accompanied by fever and varying degrees of systemic toxicity. The typical appearance of erysipelas includes bright red, thickened induration of the skin with well-demarcated, often raised, borders (Figure 2) .

Figure 2.

Erysipelis. Red dermal plaque with a well-defined border. (Courtesy of Bryan Anderson, MD)

In other forms of cellulitis the margins may not be raised and may be more indistinct, and there may be patchy areas of involvement. Local findings may also include underlying furuncles or small abscesses, blisters or ulcerations from where the cellulitis may have originated. Tender regional lymph nodes and even lymphangitic streaking can also be seen. In more severe infections, bullae and skin necrosis may develop.

Expected results of laboratory studies

Laboratory findings in uncomplicated cellulitis include leukocytosis and elevated CRP. Specific serologic tests, such as antistreptolysin O (ASO) titers, are not very useful for the acute diagnosis of infection. Where cellulitis is not accompanied by ulcers or an abscess, aspiration of the leading edge of the area of involvement has good specificity and has provided information on the microbiology of this infection, but the sensitivity in nonimmunocompromised patients is low and thus this is not routinely recommended, though yield may be higher in aspiration of fluctuant areas.

Cultures of areas of ulceration and abrasion can identify responsible pathogens, but may also recover surface colonizing organisms; yield and specificity are higher for culture of draining furuncles or abscesses, especially for staphylococcal infections. Yield of blood cultures in nonimmunocompromised patients with uncomplicated cellulitis is low. Imaging findings in cellulitis are generally nonspecific and are more important in providing evidence for deeper, more serious necrotizing soft tissue infections or underlying osteomyelitis.

Diagnosis confirmation

The differential diagnosis of cellulitis includes a wide variety of infectious and noninfectious conditions. Noninfectious conditions seen with relative frequency that may mimic cellulitis include superficial and deep venous thromboses, dermatologic conditions such as contact dermatitis, eczema, urticaria and angioedema, insect bites and stings, foreign body reactions, cutaneous drug reactions, systemic processes such as Sweet’s syndrome, rheumatologic inflammatory processes such as gout, lupus and polychondritis, and a variety of panniculitis syndromes. Another very common process misdiagnosed as cellulitis is the erythema of chronic lymphadema, especially lymphadema with concurrent stasis dermatitis (often presenting as lipodermatosclerosis)

In one recent study, 28% of patients admitted from an emergency department with presumed cellulitis had alternate diagoses, most commonly stasis dermatitis. Infectious processes that can mimic cellulitis include the erythema migrans rash of Lyme Disease and tinea (ringworm) infection; the presentation of severe cellulitis may overlap with that of necrotizing fasciitis and other more severe skin and soft tissue infections. Ultrasound may be helpful to differentiate cellulitis in an extremity from the erythema, warmth and swelling of a venous thrombosis.

Who is at Risk for Developing this Disease?

Cellulitiis is an extremely common infection in the US. Recent studies utilizing large ambulatory and hospital databases give estimates of from 6.3 million to as high as 12 million office visits for skin and soft tissue infections, predominantly cellulitis, in 2005. Some studies suggest that these rates have increased over the past decade, coincident with emergence of community-associated methicillin-resistant Staphylococcus aureus (MRSA).

In one population-based survey of a more carefully defined cohort with lower extremity cellulitis, the incidence of cellulitis increased with increasing age but there were no differences in rates between sexes. In this study and other similar studies, approximately 20% of these patients required hospitalization and 20% had recurrent infections within 2 years.

An individual’s risk for developing cellulitis include those risks from exposure to particular cellulitis pathogens; exogenous and endogenous factors that disrupt the integrity of the normal local mucosal barriers, thus enhancing the ability of microorganisms to invade the subcutaneous tissue and host local and systemic factors that impair ability to controlevolving infections.

The more virulent bacterial pathogens of cellulitis, especially group A streptococcus and S aureus, can be transmitted among household members and other close contacts, and outbreaks of both streptococcal erysipelas and staphylococcal skin infections have occurred. Individuals who are chronically colonized by S aureus in the anterior nares or on the skin are also at increased risk for subsequent development of staphylococcal skin infections from their own colonizing strains following even minor disruption of host skin barriers.

Other potential exposures to less common virulent cellulitis pathogens would include outdoor activities, trauma, contact with fresh, brackish or salt water, contact with soil, and animal, insect or human bites,

Exogenous disruption of mucosal barriers can occur from traumatic injuries including penetrating trauma or thermal burns, injection drug use or minor cuts and scrapes. Even trivial trauma such as small razor cuts can provide a portal of entry for cellulitis pathogens. Examples of endogenous factors that lead to disruption of skin barriers can include chronic skin conditions, for example eczema and psoriasis and other chronic dermatitis, as well as acute dermatlologic processes such as poison ivy or antecedent varicella or herpes zoster infections.

Examples of other endogenous factors contributing to local development of celluliits include chronic edema and lymphedema, chronic stasis ulcers, prior lymphatic disruption, and web space intertrigo. Systemic immune modifying processes that may presdispose to cellulitis and erysipelas include diabetes mellitus, chronic renal and liver disease, neutropenia, other immunosuppressive conditions and medications, and HIV infection.

In one population based study of risk factors for lower extremity erysipelis, disruption of cutaneous barriers by a vareity of processes including ulcers, wounds, fissuring of the toe webs, pressure ulcers or leg dermatosis markedly increased risk of erysipelis with an odds ratio of 23.8. Other important risk factors in this study were lymphedema, edema and obesity. Local factors were much more significant in this study than systemic illnesses such as diabetes and alcohol abuse.

What is the Cause of the Disease?

Erysipelas and cellulitis are infections that are caused when infectious organisms gain access and spread through the dermal and subcutaneous tissue layers, causing tissue damage as well as provoking a vigorous host inflammatory response. The processes involved in the pathophysiology of erysipelas and cellulitis are similar, although erysipelas generally refers to infections that involve the more superficial layers of the subcutaneous tissue and is generally caused by beta hemolytic streptococci.


Factors important in the pathogenesis of this process are factors related to the exposure to and colonization with the primary pathogens of cellulitis, most commonly beta-hemolytic streptococci and S aureus, factors that provide disruption of the integrity of the normal dermal epithelial barriers, and factors that that facilitate the multiplication of microrganisms and the spread of infection through the subcutaneous tissue. These include bacterial virulence factors and a wide variety of host immune deficits.

More virulent organisms may require only minor disruption of host barriers and can cause extensive disease in otherwise healthy individuals; less virulent organisms are more opportunistic and cause disease predominantly in the setting of more disruption of host barriers or greater degrees of immune compromise.

Most commonly, cellulitis results from bacteria directly gaining access into the deep dermal layers and subcutaneous tissues through minor or major breaks in the skin. Once established in these tissues, invading microorganisms stimulate potent host immune and inflammatory response in an attempt to control the infection, but specific virulence factors expressed by the most successful cellulitis pathogens enable them to succesfully evade these host defenses.

Streptococcal infections generally present as diffuse tissue infections without discrete abscess formation. For staphylococcal infections, the focus of infection often originates in the hair follicles and with an intense neutrophilic response. The resulting infection can be primarily that of an evolving small abscess, of cellulitis without apparent abscess or purulent drainage, or a mixed picture of both abscess formation and intense surrounding cellulitis.

Rarely, cellulitis occurs as a result of the hematogenous spread of invasive organisms from other sites. This occurs predominantly those with significant immune defects. More often, the bacteremia that occurs in patients with cellulitis is a consequence rather than a cause of their cellulitis.

Cellulitis can also manifest from the contiguous spread of infection from deeper focal processes such as osteomyelitis, septic arthritis, dental abscesses or enteric abscesses and fistulas. Management of the cellulitis complicating a deeper infection requires addressing the primary focus of infection in addition to treating the cellulitis.

Processes involved in the pathogenesis of staphylococcal skin infections include colonization with S aureus, followed subsequently by infection with endogenous strains. Less commonly S aureus can cause infection by direct inoculation into a wound or other skin break by transmission from another individual, as may occur in the health care setting. The major reservoir for S aureus is the anterior nares. Approximately 20% of individuals are chronically colonized with S aureus in the anterior nares and another 30% may be transient carriers; those chronically colonized are at much higher risk for subsequent clinical staphylococcal infections.

Other common sites of colonization include the groin and axilla, as well as any open wound or area of chronic skin inflammation such as areas of eczema or psoriasis. Group A beta-hemolytic streptococci are the major cause of bacterial pharyngitis, and pharyngeal streptococcal colonization is common. However, the association between pharyngeal carriage and skin colonization and infection is less well understood, and there may be differences in virulence factors and potential between pharyngeal strains and those causing severe cellulitis.

Bacterial virulence factors include preformed bacterial toxins, secreted enzymes with tissue digesting activities including proteases, lipases, hyaluronidases, phospholipases, nucleases and many others, and other specific cell products. Some factors suggested to be important in the pathogenesis of the epidemic of CAMRSA skin infections include phenol soluble modulins, protein A, and the Panton-valentine leukocidin, though the roles of some of these factors remains incompletely understood.

Important streptococcal virulence factors include a variety of structural cell surface components such as M protein, hyaluronic acid capsules, peptidoglycan, lipoteichoic acid and many others that are involved in colonization and evasion of host neturophil responses, in addition to a broad array of tissue-destructive enzymes and additional secreted exotoxins. Some of these secreted toxins, such as the toxic shock syndrome toxins, may act as superantigens and are repsonsible for the severe systemic manifestations complicating some episodes of streptococcal and staphylococcal cellulitis. Unusual late complications of group A streptococcal infections, especially in children and young adults, may include rheumatic fever and other post streptococcal inflammatory syndromes and post streptococcal glomerulonephritis.

The consequence of the spread of infectious organisms, release of toxins, ongoing tissue breakdown and activation of host macrophage and neutrophil responses is the stimulation of a strong host inflammatory response. This inflammatory response is a major contributor to the erythema, edema, pain and other clinical manifestations of the infection and is the rationale for studies of anti-inflammatory therapies as adjuncts to antimicrobials for the treatment of cellulitis Infection may spread through and destroy the lymphatic system, resulting in worsening edema, clinical manifestations of lymphangitis, and producing pain and swelling in regional lymph nodes.

Systemic Implications and Complications

Complications of cellulitis include progression of infection, secondary bacteremias and other metastatic spread of infection, and systemic sepsis syndromes from both local and metastatic infection. The most common late complication of cellulitis is the development of recurrent episodes of this infection.

A common sequelae of cellulitis is evolution of discrete absesses, which will require drainage in addition to antimicrobial therapy. Cellulitis may also progress to more severe, deeper processes such as necrotizing fasciitis and other necrotizing soft tissue infections; early on these processes may be difficult to distinguish from severe cellulitis.

Careful clinical monitoring, imaging studies including CAT scan or MRI looking for evidence of gas in the soft tissue, and use of scoring systems such as the Laboratory Risk Indicator for Necrotizing Fasciitis may help distinguish uncomplicated cellulitis from these more severe, surgically managed processes. Severe local inflammation and edema from cellulitis can also result in compartment syndrome, which would require fasciotomy.

Other local complications may include extension to bone and development of osteomyeliits, especially when cellulitis complicates a chronic ulcer. Recognition of underlying osteomyelitis has implications for determining duration of antibiotic therapy as well as need for surgical debridement. Magnetic resonance imaging (MRI) is the most sensitive study for identification of underlying osteomyeilitis. Plain radiographs lack sensitivity, especially early in the course.

Nuclear medicine studies such as bone scan and tagged white blood cell studies may also help distinguish uncomplicated celluitis from cellulitis associated with osteomyelitis. On physical examination, probe of an open ulcer to bone also has high specificity for the diagnosis of osteomyelitis.

Cellulitis will often spread through lymphatic channels to involve the regional lymph nodes, causing lymphangitis and lymphadenitis. The inflammatory process may also involve the venous circulation. Venous thrombosis, in addition to being a mimic of cellulitis and a predisposition to subsequent episodes of cellulitis, can occasionally also be a complication of acute cellulitis. Doppler studies are the most useful studies for identification of superficial and deep venous thromobosis.

Blood cultures are positive in less than 5% of episodes of erisyipelis and cellulitis but are more common in the subset of patients who are hospitalized. In one study, 18.5% of hospitalized pateints with limb cellulitis had positive blood cultures, in this study the most common blood culture isolates were with gram-negative organisms and non-group A beta-hemolytic streptococci. Risk factors for positive blood cultures included not receiving prior antibiotics, proximal limb infections, short duration of illness, and having 2 or more associated co-morbidities.

Infections with group A streptococci and S. aureus can also produce toxin-mediated disease syndromes, including staphylococcal and streptococcal toxic shock syndrome, that can occur even in the absence of bacteremia. In addition to sepsis syndromes and septic shock, other complications of bacteremia may include development pf metastatic foci of infections such as endocarditis, septic arthritis and osteomyelitis. Cellulitis is also a risk factor for infection of prosthetic joints and other orthopedic devices, even in the absence of documented bacteremia.

The major long term-potential complication of an episode of cellulitis is the risk for recurrent episodes of infection. Each episode of cellulitis will result in some degree of lymphatic inflammation and lymphatic destruction, leading to worsening of chronic edema and lymphedema, and increasing the propensity for further episodes of cellulitis in the affected limb or region. In one study of lower extremity cellulitis, 16.7% of patients recurred within 2 years. Infection in the tibial region, a history of malignancy, or dermatitis in the limb were independent predictors of recurrence.

Several clinical staging systems have been proposed for management of patients with cellulitis to assist with decisions on need for hospitalization and parenteral antibiotics, including the widely used Eron classification system based on expert panel recommendations (2000), the Clinical Resource Efficiency Support Team (CREST) Guidelines on the Management of Cellulitis in Adults from Scotland (2003), and the recent “Dundee” Classification system from a retrospective review of another cohort of hospitalized cellulitis patients in Scotland (2011). None of these has been prospectively validated.

However, regardless of the details of the staging system, patients with higher stage disease as determined by having findings of sepsis or septic shock in addition to other physiologic or clinical parameters, had very high in-hospital and 30 day mortality. In the recent study by Marwick using the “Dundee” staging system, overall mortality in a cohort of 205 hospitalized patients was 9% but was 33% in the highest stage cohort.

Treatment Options

Treatment options are summarized in Table I.

Table I.
Systemic Antimicrobial Therapy Adjuvant Pharmacologic Therapy Local Care Measures Surgical Interventions
Oral Antimicrobial Therapy Corticosteroids Elevation of the affected limb Drainage of Abscesses
Parenteral antimicrobial therapy in the Hospital Non-Steroidal Anti-inflammatory Agents Treatment of edema with compressive measures and/or diuresis Surgical exploration and debridement of necrotic tissue
Outpatient Parenteral antimicrobial therapy Treatment of web space and other chronic tinea infections Fasciotomy for compartment syndromes
Chronic prophylactic oral antimicrobial therapy Treatment of chronic dermatitis with emollients and other measures

Empiric Antimicrobial Therapeutic Choices for Cellulitis and Erisypelis

The primary modality of treatment of cellulitis and erysipelas is antimicrobial therapy. Antimicrobial therapy is targeted at the most common skin pathogens, which are generally beta-hemolytic streptococci and S aureus. Topical antibiotics are not sufficient for the treatment of erysipelas or cellulitis and systemic therapy is required (Table II).

Table II.
Antibiotic Route of Administration Standard Dosing Regimen Comments
penicllin oral 500mg four times a day No staphylococcal activity
intravenous 3 Million units every 4hours IM dosing also available
amoxicillin oral 500mg three times a day No staphylococcal activity
amoxicillin-clavulanic acid oral 875mg twice a day or 500mg three times a day No MRSA activity
cephalexin oral 500mg four times a day No MRSA activity
dicloxacillin oral 500mg four times a day No MRSA activity
clindamycin oral 300-450mg three to four times a day Active against many CA-MRSA strains
intravenous 600mg three times a day
linezolid oral or intravenous 600mg twice a day Active against MRSA. and most gram positive pathogens
Monitor CBC
doxycycline oral or intravenous 100mg twice a day Active against most MRA but less reliable streptococcal activity
Minocycline has similar properties
Trimethoprim-sulfamethoxazole oral one to two double strength tablets twice a day Active against most MRSA
Poor streptococcal activity. Intravenous formulation also available
Cefazolin intravenous 1g every 8 hours No MRSA activity
Ceftriaxone intravenous 1-2g daily No MRSA activity
Nafcillin intravenous 2g every 4 hours No MRSA activity
vancomycin intravenous 15-20mg/kg every 12 hours Active against MRSA and most gram positive pathogens. Dosing based on levels
daptomycin intravenous 4mg/kg once a day Active against MRSA and most gram positive pathogens
Monitor CPK
Higher doses may be necessary

Choice of agents and route of administration (oral or parenteral) may be modified by disease severity and host immune status.In sicker or more immunosuppresed patients, the margin for error in the selection of empiric antmicrobial therapy is smaller. Therapy may also be modified on the basis of specific risk factors for infection with unusual organisms. The major recent change in approach to empiric antimicrobial therapy for cellulitis is the need to consider risk for infection with MRSA and community associated MRSA (CA-MRSA) in a broader range of patients.

First-line antimicrobial therapy for erysipelas, which is primarily caused by beta-hemolytic streptococci, is penicillin. Penicillin can be administered orally to nonsystemically ill patients without major comorbidities; in other circumstances, intravenous or intramuscular dosingcan be used. Oral amoxicillin provides similar antistreptococcal activity to oral penicillin, good bioavailability and the advantage of less frequent dosing.

For patients with minor penicillin allergy such as rash who can tolerate cephalosporins, an oral first=generation cephalosporin may an acceptable alternative. For highly penicillin allergic patients, other oral alternatives would include a macrolide or clindamycin. However, rates of resistance of beta-hemolytic streptococci to macrolides and clindamycin are increasing.

For management of uncomplicated cellulitis, antimicrobial regimens are generally targeted against beta-hemolytic streptococci and S aureus. In many instances this will now require agents with activity against MRSA, especially in areas where rates of CA-MRSA in the community are high. The need for staphylococcal activity is dictated by the type of infection. Streptococci are the most common pathogen in uncomplicated nonpurulent cellulitis without associated abscess, ulcerations, recent surgery or trauma.

Recently published guidelines from the Infectious Diseases Society of America do not recommend empiric coverage for CA-MRSA for uncomplicated, nonpurulent cellulitis.These guidelines do recommend using agents active against MRSA for cellulitis with purulent drainage or exudates or for patients failing to respond to regimen without MRSA activity.Several of the inexpensive and well tolerated oral agents with MRSA activity, including trimethoprim-sulfamethoxazole and the tetracyclines, do not have reliable streptococcal activity, thus, the clinical characterization of the type of cellulitis has become even more important.

Historically, oral options for uncomplicated cellultis that have treated both streptococci and methicillin-susceptible S aureus have included a semisynthetic penicillin such as cloxacillin or dicloxacillin, an oral first-generation cephalosporin such as cephalexin, or amoxicillin-clavulanic acid. Staphylococcal activity of macrolides is now insufficient as to consider these as reliable alternatives in this setting.

For highly penicillin-allergic patients, clindamycin is a recommended alternative, although rates of clindamycin resistance among S aureus strains has increased. Linezolid is also active against the vast majority of streptococcal and staphylococcal strains and is a good alternative in highly beta-lactam allergic patients, but is generally reserved for more severe infections.

For more severe infections or those in higher-risk patients, parenteral options offering good activity for non-urulent cellulitis where MRSA is not a significant concern include an intravenous first generation cephalosporin such as cefazolin, a third-generation cephalosporin such as ceftriaxone, or an intravenous semisynthetic penicillin (nafcillin or oxacillin). For penicillin-allergic patients, vancomycin, clindamycin and linezolld are parenteral options.

In patients with clinical findings of purulent cellulitis or specific risk factors for MRSA, such as a history of MRSA, recent hospitalization or stay in long term-care facility, chronic ongoing medical care, chronic wounds, or recent antimicrobial therapy, acceptable cellulitis regimens should include agents active against MRSA. This is especially important in more severely ill, hospitalized patients. Conversely, in those patients who are not systemically ill and where the primary process is a cutaneous MRSA abscess with relatively minimal component of surrounding cellulitis, incision and drainage even without antimicrobial therapy is an acceptable treatment alternative.

Oral agents active against most MRSA include clindamycin, linezolid, trimethoprim-sulfamethoxazole (TMS) and tetracyclines. As TMS and the tetracyclines are not reliably active against beta-hemolytic streptococci, oral regimens where both staphyloccci and streptococci are concerns may need to include a streptococcal agents such as a penicillin in addition to TMS or a tetracycline.

Clindamycin is active against many but not all CA-MRSA strains but is not recommended for staphylococci with inducible erythromycin resistance. Linezolid continues to have excellent MRSA and streptococcal activity and is both an oral and parenteral option. Quinolones are not recommended for treatment of MRSA infections.

Parenteral options for more severely ill patients requiring an MRSA active regimen would include vancomycin and daptomycin in addition to linezolid, other new agents parenteral agents with good MRSA activity that have been studied for complicated skin and soft tissue infections in clincal trials include tigecycline, telavancin, and ceftaroline.

Antibiotic regimens need to be modified in situations where cellulitis caused by unusual pathogens other than staphylococci and streptococci is suspected. Some examples of situations and specific pathogens that need to be considered would include:

-Dog and cat bites: Pasteurella multocida, Capnocytophaga

-Human bites: Haemophilus, Eikenella, Bacteroides and other anaerobes

-Surgical postoperative infections: Enteric gram negative rods, enterococci, other organisms, depending on the location of the surgical wound

-Salt water: Vibrios

-Fresh water: Aeromonas

-Neutropenic and other severely immunocompromised: Enteric gram negative rods and Pseudomonas

Optimal Therapeutic Approach for this Disease


Oral antimicrobial therapy

Most patients with uncomplicated erysipelas and cellulitis can be safely managed with oral antimicrobial therapy. This includes those without evidence of systemic toxicity or significant immunocompromise or major comorbidities as well as many patients with a single comorbility such as peripheral vascular disease or venous insufficiency or morbid obesity. Oral antibiotics are targeted at beta-hemolytic streptococci and/or S. aureus including MRSA depending on clinical features.

Outpatient parenteral antimicrobial therapy

Can be utilized in nonsystemically ill patients who do not otherwise require hospital observation but may benefit from broader antimicrobial regimens or use of agents such as vancomycin that can only be given parenterally. More often this option is utliized in improving patients after a period of inpatient parenteral therapy.

Inpatient parenteral antimicrobial therapy

Indicated in more severely ill patients with more systemic toxicity, underlying immune compromise or multiple co-morbitidies. Approximately 20% of patients will require hospitalization and parenteral antibiotic therapy. Initial antibiotic regimens in these patients are often broader in spectrum than those used in less severely ill or immunocompromised patients and usually include streptococcal and staphylococcal activity.

Typical total durations of antimicrobial therapy are 7 to 114 days but durations are often determined by clinical response and resolution of findings. In many instances, the clinical findings of inflammation can persist beyond eradication of infecting bacteria resulting in significant overtreatment. One recent study demonstrated equivalence of 5 or 10 days of oral therapy in a cohort of patients with uncomplicated cellulitis

Chronic prophylactic antibiotics should be reserved only for those patients with frequent and severe recurrences


Limb elevation

Decreases local edema and swelling and may hasten clearance of inflammatory mediators of infection and is the most simple, useful and often neglected measure. Elevation can be more rigorously enforced in the hospital environment.

Other measures for edema and lymphadema

Use of compressive stockings and dressings and diuresis may decrease edema. These interventions must be used with caution in the setting of active cellulitis and in hemodynamically unstable patients. These measures may be more important in preventing recurrences.

Treatment of tinea web space infections

Web space infections are a common portal of entry for cellulitis pathogens.

Treatment of chronic dermatitis with topical emolliments


The mainstay of therapy of abscesses associated with cellulitis is drainage in addition to antimicrobial therapy. For minor staphylococcal infections, where the predominant problem is the abscess with minimal surrounding cellulitis, drainage without antibiotics may be adequate therapy.

Surgical debridement may be necessary in severe cellultis that progress to tissue necrosis, in infections associated with wounds or those associated with chronic osteomyelitis and devitalized bone.

Fasciotomy may be necessary with cellulitis and marked tissue edema that results in compartment syndrome.


Use of corticosteroids

There is a single randomized clincal trial of prednisolone with antibiotics in the treatment of erysipelas in which steroids decreased time to healing and duration of antibiotic days without increased risk of relapse or increased toxicity. Given the limited data, use of corticosteroids is not routinely recommended.

Use of nonsteroidal anti-inflammatory drugs (NSAIDS)

One small randomized trial of ibuprofen plus antibiotics or antibiotics alone showed possible increased rates of resolution in those treated with adjunctive ibuprofen. However, there is also conflicting data on NSAID agent use and risk of severe group A streptococcal cellulitis in children. Routine use of non-steroidal agents is not recommended

Patient Management

The monitoring of patients being treated for cellulitis includes monitoring for findings of clinical progression that would indicate a potentially more severe infection, such as necrotizing fasciitis that might require surgical intervention, or would suggest potential failure of the initial empiric antimicrobial regimen and need to revise or broaden the regimen.

For sick, hospitalized patients, clinical monitoring includes monitoring of fever curve, white blood cell count (WBC) and clinical examination as well as consideration for imaging studies such as MRI or CAT scan that might not have been initially performed. The initial clinical decision process involves identification of those patients who are ill enough to require hospitalization for either intravenous therapy or other monitoring and those patients who can be safely managed with outpatient antimicrobial therapy.

Those patients treated with oral agents as outpatients who fail to improve may require escalation of care including hospitalization and initiation of parenteral antibiotics. Other benefits of hospitalization include more rigorous institution of local care measures such as bed rest and limb elevation.

Antibiotic regimens may need to be modifed based on microbiologic data, if available, and patients need to be evaluated for any evidence of toxicity to antimicrobial agents. Some agents such as vancomycin require monitoring of drug levels; others may require obtaining of other specific laboratory values such as monitoring of complete blood counts (CBCs) and platelet count in patients on linezolid or creatine phosphokinase levels in patients on daptomycin.

Hospitalizated patients who improve may be considered for transition to oral therapy or home parenteral therapy when clinically improving. The duration of antibiotic therapy has never been carefully evaluated in controlled clinical trials. Treatment is generally continued until marked improvement of the clinical signs and symptoms of infection, but this may result in overtreatment. Treatment durations are typically 5 to 10 days, but may be longer with associated bacteremia, osteomyeltis or significant local wounds and underlying immune compromise.

Patients with episodes of cellulitis remain at high risk of recurrance; 20% of those with limb cellulitis will have another episode within 2 years. Prevention of recurrences includes measures targeted at correction of risk factors such as edema and lymphedema and chronic tinea and webspace infections and chronic dermatitis. For diabetic patients, longterm control of blood sugars is also important.

Patients with frequent recurrences may benefit from self-directed early administration of antimicrobial therapy at first onset of symptoms. Chronic antibiotic prophylaxis targed at beta-hemolytic streptococci has also been employed in patients with frequent episodes and significant morbidity from their episodes. The most commonly used regimens have been Intramuscular benzathine penicillin 1.2 to 2.4 million units every 3-4 weeks, or low dose oral penicillin or erythromycin twice a day.

Unusual Clinical Scenarios to Consider in Patient Management

Physicians must be attuned to evidence of unusual precipitating events or exposures or other host predispositions that would raise concern for unusual pathogens that might not be treated withthe “usual” empiric antibiotic regimens. Concern for possible atypical pathogens would be an indication to obtain punch biopsies and cutures, including mycobacterial and fungal cultures and possibly even viral cultures if a possibe underlying herpetic infection is suspected.

Some of the unusual exposures and bacterial pathogens include Vibrios in salt water injuries and shellfish ingestion, Aeromonas from fresh water injuries, Pasturella and Capnocytophaga from cat and dog bites, Erysipeiothrix in butchers and fishermen, as well as many others. Atypcial organisms such as rapidly growing mycobacteria, nocardia and fungal organisms also need to be considered in particular circumstances, especially for exposure related infections that fail to improve after courses of standard antimicrobial therapy, and infections in patients with underlying immunodeficiencies.

What is the Evidence?

Stevens , DL, Bisno , AL, Chambers , HF, Everett , ED, Dellinger , P, Goldstein, EJC. “Practice guidelines for the diagnosis and management of skin and soft tissue infections”. Clin Infect Dis . vol. 41. 2005. pp. 1373-1406. (The most recent Infectious Diseases Society of America practice guidelines for management of skin and soft tissue infections that includes focused discussions of the severe skin and soft tissue infection syndromes and current recommendations for management. An update of these guidelines is currently in progress with anticipated publication in 2012.)

Falagas ME and Vergidis , PI. ” Narrative review: Diseases that masquerade as infectious cellulitis”. Ann Intern Med. vol. 42. 2005. pp. 47-55. (Good overview and discussion of the many non-infectious processes that can be mistaken for infectious cellulitis.)

Dupuy, , A, Benchikhi , H, Roujeau , JC, Bernard , P, Vaillant , L, Chosidow , O. “Risk factors for erisypelis of the leg (cellulitis): case control study”. BMJ . vol. 318. 1999. pp. 1591-4. (One of several well done case-control studies demonstrating the importance of chronic edema/ lymphedema and other anatomic factors as risks for erysipelas and cellulitis.)

Liu , C, Bayer , A, Cosgrove , SE, Daum , RS, Fridkin , SK, Gorwitz , RJ. “Clinical practice guidelines by the Infectious Diseases Society of America for the treatment of methicillin-resistant Staphylococcus aureus infections in adults and children”. Clin Infect Dis . vol. 52. 2011. pp. 1-38. (New guidelines that discuss oral and parenteral options for management of MRSA infections including skin and soft tissue infections, includes discussion of distinction in treatment regimens for purulent and non-purulent cellulitis.)

Marwick , C, Broomhall , J, McCowan , C, Phillips , G, Gonzalez-McQuire , S, Akhras , K. ” Severity assessment of skin and soft tissue infections: cohort study of management and outcomes for hospitalized patients”. J Antimicrob Chemother . vol. 66. 2011. pp. 387-97. (The most recent of several proposed staging systems for severity of skin and soft tissue infections that correlates severity with mortality and addresses the overtreatment of lower stage infections and undertreatment of higher stage infections )

Stevens , DL, Eron , LJ. ” Cellulitis and soft tissue infections”. Ann Intern Med. vol. 150. 2009. pp. ITC-11. (Excellent clinical overview from the American College of Physicians addressing many of the clinical aspects of management of patients with cellulitis.)

Jenkins , TC, Knepper , BC, Sabel AL Sarcone , EE, Long , JA, Haukoos , JS. “Decreased antibiotic utilization after Implementation of a guideline for inpatient cellulitis and cutaneous abscess”. Arch Intern Med. vol. 171. 2011. pp. 1072-79. (An important study deomonstrating the use of a guideline for antibiotic selection in treating skin and soft tissue infections that decreased antibiotic exposure without changing outcome of treatment.)

Hepburn , MJ, Dooley , DP, Skidmore , PJ, Ellis , MW, Starnes , WF, Hasewinkle , WC. “Comparison of short course (5 days) and standard (10 days) of treatment for uncomplicated cellulitis”. Arch Int Med . vol. 16. 2004. pp. 1669-74. (A well-done trial comparing 5 days of antibiotics plus 5 days of placebo to 10 days of antibiotics in uncomplicated cellulitis demonstrating equivalent outcomes in both groups.)

Bergkvist , PI, Sjöbeck , K. “Antibiotic and prednisolone therapy of erysipelas: a randomized, double blind, placebo controlled study”. Scand J Infect Dis . vol. 29. 1997. pp. 377-92. (The single published trial showing some benefit and no adverse consequence for use of corticosteroids in acute erysipelas.)

Leclerc , S, Teixeira , A, Mahe , E, Descamps , V, Crickx , B, Chosidow , O. “Recurrent erysipelas; 47 cases”. Dermatology . vol. 214. 2007. pp. 52-7. (Describes experience and outcomes with the use of chronic oral or parenteral antibiotic prophylaxis in a group of patients with recurrent cellulitis.)