Official reprint from UpToDate®

www.uptodate.com ©2019 UpToDate, Inc. and/or its affiliates. All Rights Reserved.

Cellulitis and skin abscess in adults: Treatment

Authors: Denis Spelman, MBBS, FRACP, FRCPA, MPH, Larry M Baddour, MD, FIDSA, FAHA
Section Editor: Franklin D Lowy, MD
Deputy Editor: Meg Sullivan, MD

All topics are updated as new evidence becomes available and our peer review process is complete.

Literature review current through: Oct 2019. | This topic last updated: Apr 01, 2019.

INTRODUCTION

Patients with skin and soft tissue infection may present with cellulitis, abscess, or both [1-3].

Treatment of cellulitis and skin abscess are reviewed here.

Issues related to clinical manifestations and diagnosis of cellulitis and abscess are discussed
separately. (See "Cellulitis and skin abscess: Clinical manifestations and diagnosis".)

Issues related to skin and soft tissue infections associated with specific epidemiologic factors (such
as diabetes, animal bites, and water exposure) are discussed separately. (See "Clinical
manifestations, diagnosis, and management of diabetic infections of the lower extremities" and "Soft
tissue infections following water exposure" and "Animal bites (dogs, cats, and other animals):
Evaluation and management" and "Human bites: Evaluation and management".)

Issues related to infection involving the gluteal area and perineum are discussed separately. (See
"Perianal and perirectal abscess" and "Pilonidal disease".)

GENERAL PRINCIPLES

Overview — The approach to management of skin and soft tissue infection depends on the clinical
presentation:

● Patients with nonpurulent infection (ie, cellulitis or erysipelas in the absence of abscess or
purulent drainage) should be managed with empiric antibiotic therapy (algorithm 1). (See
'Nonpurulent infection' below.)
 ● Patients with drainable abscess should undergo incision and drainage; the technique is
discussed separately (see "Technique of incision and drainage for skin abscess"). In addition,
antibiotic therapy is warranted if clinical criteria are met, as discussed below (algorithm 2). (See
'Drainable abscess present' below.)

● Patients with purulent cellulitis (ie, cellulitis associated with purulent drainage in the absence of
drainable abscess) should be managed with antibiotic therapy (algorithm 2). (See 'Purulent
cellulitis (no drainable abscess)' below.)

Issues related to choosing between oral and parenteral therapy are discussed below. (See 'Oral
versus parenteral therapy' below.)

Attention to antibiotic dosing is important, particularly in obese individuals; underdosing (particularly

in those with morbid obesity and lymphedema) may result in higher rates of treatment failure [4].

The approach to empiric antimicrobial therapy should be modified as indicated in the setting of
known pathogens, underlying conditions (such as diabetes), and special circumstances (such as
animal bites and water exposure). Management of patients in these settings is discussed in detail
separately. (See "Clinical manifestations, diagnosis, and management of diabetic infections of the
lower extremities" and "Animal bites (dogs, cats, and other animals): Evaluation and management"
and "Soft tissue infections following water exposure" and "Human bites: Evaluation and
management".)

Other components of management include elevation of the affected area and treatment of underlying
conditions (such as edema or underlying cutaneous disorders) if present [2]. Elevation facilitates
gravity drainage of edema and inflammatory substances. The skin should be sufficiently hydrated to
avoid dryness and cracking without interdigital maceration.

Oral versus parenteral therapy — Patients with mild infection may be treated with oral antibiotics.
Treatment with parenteral antibiotics is warranted in the following circumstances:

● Systemic signs of toxicity (eg, fever >100.5°F/38°C, hypotension, or sustained tachycardia)

● Rapid progression of erythema
● Progression of clinical findings after 48 hours of oral antibiotic therapy
● Inability to tolerate oral therapy
● Proximity of the lesion to an indwelling medical device (eg, prosthetic joint or vascular graft)

The decision to initiate parenteral therapy should be based on individual clinical circumstances such
as severity of clinical presentation and patient comorbidities. As an example, the presence of an
immunocompromising condition (such as neutropenia, recent organ transplant, advanced HIV
infection, B cell or T cell deficiency, or use of immunosuppressive agents) should lower the threshold
for parenteral therapy.

CLINICAL APPROACH

Nonpurulent infection — Forms of nonpurulent skin and soft tissue infection include cellulitis and
erysipelas. (See "Cellulitis and skin abscess: Clinical manifestations and diagnosis", section on
'Cellulitis and erysipelas'.)

Management of cellulitis and erysipelas should include elevation of the affected area and treatment
of underlying conditions. (See 'Overview' above.)

Many patients with cellulitis have underlying conditions that predispose them to developing recurrent
cellulitis (these include tinea pedis, lymphedema, and chronic venous insufficiency). In such patients,
treatment should be directed at both the infection and the predisposing condition if modifiable. As an
example, patients with edema may benefit from treatment with compressive stockings and diuretic
therapy.

Cellulitis — Patients with nonpurulent cellulitis (eg, cellulitis with no purulent drainage or exudate
and no associated abscess) should be managed with empiric therapy for infection due to beta-
hemolytic streptococci and methicillin-susceptible Staphylococcus aureus (MSSA) [1-3]. Common
options are cefazolin for intravenous therapy and cephalexin for oral therapy; options and doses are
summarized in the algorithm (algorithm 1). Issues related to choosing between oral and parenteral
therapy are discussed above. (See 'Oral versus parenteral therapy' above.)

This approach is supported by the following studies:

● A randomized trial including 496 patients with nonpurulent cellulitis (in the modified intention-to-
treat analysis) noted similar clinical cure rates among those treated with cephalexin plus placebo
(for empiric treatment of beta-hemolytic streptococci and MSSA; 69 percent) and those treated
with cephalexin plus trimethoprim-sulfamethoxazole (TMP-SMX; for empiric treatment of beta-
hemolytic streptococci and methicillin-resistant S. aureus [MRSA]; 76 percent; difference 7.3
percent; 95% CI -1.0 to 15.5 percent; p=0.09) [5]. While these findings raise the possibility that
addition of TMP-SMX may be somewhat superior to cephalexin alone, the results were likely
skewed by a relatively large number of patients who did not complete the full course of therapy.

● A randomized trial including 153 patients with cellulitis without abscess noted comparable cure
rates among those treated with cephalexin (for empiric treatment of beta-hemolytic streptococci
and MSSA; 82 percent) and those treated with cephalexin and TMP-SMX (for empiric MRSA
coverage; 85 percent) [6].
Additional empiric coverage for MRSA is warranted in the following circumstances [2,7]:

● Systemic signs of toxicity (eg, fever >100.5°F/38°C, hypotension, or sustained tachycardia)

● Prior episode of MRSA infection or known MRSA colonization
● Lack of clinical response to antibiotic regimen that does not include activity against MRSA
● Presence of risk factor(s) for MRSA infection (including recent hospitalization, residence in a
long-term care facility, recent surgery, hemodialysis, and HIV infection)
● Proximity of the lesion to an indwelling medical device (eg, prosthetic joint or vascular graft)

Issues related to treatment of MRSA infection are discuss further below. (See 'Approach to antibiotic
therapy' below.)

Deepening of erythema may be observed following initiation of antimicrobial therapy. This may be
due to destruction of pathogens that release enzymes increasing local inflammation and should not
be mistaken for therapeutic failure.

Patients with cellulitis typically have symptomatic improvement within 24 to 48 hours of beginning
antimicrobial therapy, although visible improvement of clinical manifestations in more severe cases
may take up to 72 hours. Persistence of erythema and/or systemic symptoms after this period of
time should prompt consideration of resistant pathogens or alternative diagnoses. In such cases,
culture data should be reviewed carefully, pursuit of radiographic evaluation for deeper infection is
appropriate, and broadening antibiotic therapy to include coverage for gram-negative bacilli pending
further diagnostic data is reasonable. (See "Cellulitis and skin abscess: Clinical manifestations and
diagnosis", section on 'Diagnosis'.)

The duration of therapy should be individualized depending on clinical response. In general, five days
of therapy is appropriate for patients with uncomplicated cellulitis whose infection has improved
within this time period [2,8]. Extension of antibiotic therapy (up to 14 days) may be warranted in the
setting of severe infection, slow response to therapy, or immunosuppression.

In one study including 216 patients hospitalized with nonpurulent cellulitis, 90 percent of patients had
improvement in clinical findings and serum C-reactive protein concentration 3 days after initiation of
antimicrobial therapy [9,10]. More than half of patients had residual inflammation at the end of
therapy (median 11 to 15 days), but relapse occurred in only 16 percent of these cases.

Erysipelas — Patients with erysipelas should be managed with empiric therapy for infection due to
beta-hemolytic streptococci.

Patients with systemic manifestations (such as fever and chills) should be treated with parenteral
therapy. Appropriate choices include cefazolin, ceftriaxone, or flucloxacillin (algorithm 1). Cefazolin
has activity against streptococci as well as MSSA, which is useful in settings where erysipelas cannot
be reliably distinguished from cellulitis. Ceftriaxone has activity against streptococci (and may be
used for activity against MSSA in some circumstances), and its once-daily dosing allows for
convenient outpatient administration.

Patients with mild infection or those who have improved following initial treatment with parenteral
antibiotic therapy may be treated with oral penicillin or amoxicillin (algorithm 1). In the setting of beta-
lactam allergy, cephalexin (if the patient can tolerate cephalosporins), clindamycin, trimethoprim-
sulfamethoxazole, or linezolid may be used; data regarding use of trimethoprim-sulfamethoxazole are
limited [2,11]. Macrolides (such as erythromycin) may not be adequate in areas with relatively high
resistance rates among beta-hemolytic streptococci [2,12].

The duration of therapy should be individualized depending on clinical response; 5 to 14 days is

usually appropriate.

Recurrent infection

First recurrence — Recurrent cellulitis is common; 22 to 49 percent of patients with cellulitis

report at least one prior episode [3]. Recurrences occur in approximately 14 percent of cellulitis cases
within one year and 45 percent of cases within three years, usually in the same location [3]. In one
review of more than 447,000 index admissions for cellulitis, the readmission rates was 9.8 percent;
most readmissions are due to skin and soft tissue infections [13].

Underlying conditions that predispose to recurrent cellulitis include [14-22]:

● Edema due to impaired lymphatic drainage

● Venous insufficiency
● Obesity
● Immunosuppression
● Fissuring or maceration of the interdigital toe spaces
● Tinea pedis

Patients with suspected recurrent infection should be evaluated carefully to rule out alternative
diagnoses. (See "Cellulitis and skin abscess: Clinical manifestations and diagnosis", section on
'Differential diagnosis'.)

The approach to treatment of a recurrent episode is the same as the approach for an initial episode.
In addition, predisposing condition(s) should be identified and alleviated if possible [3]. As an
example, patients with edema may benefit from treatment with compressive stockings and diuretic
therapy. (See "Clinical staging and conservative management of peripheral lymphedema" and
"Medical management of lower extremity chronic venous disease" and "Dermatophyte (tinea)
infections".)
For patients with recurrent infection due to S. aureus, attempting decolonization is reasonable; this is
discussed further separately. (See "Methicillin-resistant Staphylococcus aureus (MRSA) in adults:
Prevention and control", section on 'Targeted decolonization' and "Methicillin-resistant
Staphylococcus aureus (MRSA) in children: Prevention and control".)

Subsequent recurrences — For patients with three to four episodes of cellulitis per year in the
setting of predisposing factors that cannot be alleviated, suppressive antibiotic therapy may be
warranted (for as long as the predisposing factors persist) [2].

Serologic testing for beta-hemolytic streptococci may be a useful diagnostic tool to help guide the
choice of suppressive antibiotic therapy. (See "Cellulitis and skin abscess: Clinical manifestations and
diagnosis", section on 'Diagnosis'.)

Antibiotic options for suppressive therapy include:

● For patients with known or presumed beta-hemolytic streptococcal infection [23,24]:

• Penicillin V (250 to 500 mg orally twice daily)

• Erythromycin (250 mg orally twice daily)
• Penicillin G benzathine intramuscular injections (1.2 million units for patients who weigh >27
kg; 600,000 units for patients who weigh ≤27 kg) administered every two to four weeks

● For patients with known or presumed staphylococcal infection [25]:

• Cefadroxil (500 mg orally twice daily; for treatment of MSSA infection)

• Clindamycin (150 mg orally once daily)
• Trimethoprim-sulfamethoxazole (one double-strength tablet orally twice daily)

Suppressive therapy may be continued for several months with interval assessments for efficacy and
tolerance. If recurrent cellulitis occurs, the patient should be reevaluated promptly; patients may be
given instructions to self-initiate antibiotic therapy at onset of symptoms prior to seeking immediate
medical attention.

Support for suppressive antibiotic therapy for prevention of recurrent infection comes from the
following studies:

● In a study of 209 cases of cellulitis, recurrences were observed in 17 percent of patients; among
143 patients with erysipelas, 29 percent had recurrent infection [14,26]. Early episodes of
cellulitis cause lymphatic inflammation, and repeated infection can lead to lymphedema.
Supportive care with elevation of the affected area and treatment of underlying predisposing
conditions are paramount. (See "Clinical staging and conservative management of peripheral
lymphedema".)
 ● In a systematic review and meta-analysis of five trials with a total of over 500 patients with at
least one prior episode of cellulitis, prophylactic antibiotic use reduced the risk of subsequent
cellulitis (relative risk [RR] 0.46, 95% CI 0.26-0.79) [27]. Findings from two of the included studies
also demonstrated that antibiotic prophylaxis is cost-effective [28].

● In a subsequent randomized trial that included 274 patients with two or more episodes of lower
extremity cellulitis, penicillin (250 mg orally twice daily) nearly halved the risk of recurrence
during 12 months of prophylaxis (hazard ratio 0.55; 95% CI 0.35 to 0.86; p = 0.01), but the
protective effect diminished rapidly after the prophylaxis period ended [23]. A lower likelihood of
response was observed among patients with a body mass index ≥33, multiple previous episodes
of cellulitis, or lymphedema of the leg. These findings warrant further investigation since patients
in these categories are most likely to receive long-term prophylaxis.

Purulent infection — Purulent infection refers to presence of a drainable abscess or cellulitis

associated with purulent drainage (in the absence of drainable abscess) (algorithm 2). An infection
involving purulence (whether the process began as an abscess [with secondary cellulitis] or as a
cellulitis [with secondary purulence]) is potentially attributable to S. aureus, which should be reflected
in the choice of empiric antimicrobial therapy.

Drainable abscess present — Patients with drainable abscess should undergo incision and
drainage (algorithm 2) [29,30]. The technique for incision and drainage and the role of culture are
discussed separately. (See "Technique of incision and drainage for skin abscess".)

Individuals at risk for endocarditis warrant empiric antibiotic therapy prior to incision and drainage; an
oral antibiotic with activity against MRSA and beta-hemolytic Streptococcus should be administered
one hour prior to procedure (algorithm 2) [31,32]. (See "Antimicrobial prophylaxis for the prevention of
bacterial endocarditis".)

Role of antibiotic therapy — For patients undergoing incision and drainage of a skin abscess,
we suggest antibiotic treatment. In particular, we favor antibiotic treatment for patients with any of
the following:

● Single abscess ≥2 cm [33-35]

● Multiple lesions
● Extensive surrounding cellulitis
● Associated immunosuppression or other comorbidities
● Systemic signs of toxicity (eg, fever >100.5°F/38°C, hypotension, or sustained tachycardia)
● Inadequate clinical response to incision and drainage alone
● Presence of an indwelling medical device (such as prosthetic joint, vascular graft, or pacemaker)
 ● High risk for adverse outcomes with endocarditis (these include a history of infective
endocarditis, presence of prosthetic valve or prosthetic perivalvular material, unrepaired
congenital heart defect, or valvular dysfunction in a transplanted heart)
● High risk for transmission of S. aureus to others (such as in athletes or military personnel)

However, because many abscesses can be treated successfully with incision and drainage alone,
expert opinion varies, and it is reasonable to forgo antibiotic therapy in otherwise healthy patients
who have small abscesses (eg, <2 cm) and none of the above factors [36-38]. An antibiotic-sparing
approach may be particularly compelling in patients who have multiple antibiotic allergies or
intolerances.

The 2014 Infectious Diseases Society of America (IDSA) guidelines on the management of skin and
soft tissue infections did not recommend routine antibiotic therapy for patients with mild skin
abscesses in the absence of systemic infection, immunocompromising conditions, extremes of age,
or multiple abscess, based on earlier data that suggested similar cure rates with incision and
drainage alone [2]. However, subsequent large trials have indicated a benefit to antibiotic therapy,
even in patients with small abscesses:

● In a randomized trial including 1220 patients >12 years of age (median 35 years) with abscess 2
to 5 cm in diameter who underwent incision and drainage, treatment with TMP-SMX (320
mg/1600 mg twice daily) resulted in higher cure rates 7 to 14 days after treatment than placebo
(80.5 versus 73.6 percent) [33]. Wound cultures were positive for MRSA in 45 percent of cases. A
subsequent subgroup analysis of this cohort also demonstrated improved outcomes among the
TMP-SMX recipients [35].

● In another randomized trial including more than 780 patients with skin abscess ≤5 cm (45
percent were ≤2 cm) who underwent incision and drainage, treatment with TMP-SMX or
clindamycin each resulted in higher cure rates at 10 days than placebo (82 or 83 percent versus
69 percent) [34]. MRSA was isolated in 49 percent of cases.

The above randomized trials were included in a systematic review and meta-analysis of more than
2400 patients with drained abscess treated with antibiotics or placebo [39]. The rate of treatment
failure was lower among patients who received antibiotics (7 versus 16 percent); the odds ratio for
clinical cure was 2.3 (95% CI 1.7-3.1). A separate systematic review and meta-analysis including more
than 400 patients concluded treatment with TMP-SMX or clindamycin reduced the risk of treatment
failure compared with placebo, albeit with some risk of drug-adverse (primarily gastrointestinal)
events; cephalosporins failed to reduce treatment failure rates [40].

Antimicrobial therapy may also decrease the risk of recurrent skin abscess. In one randomized trial,
new infections at one month of follow-up were less common among those who received clindamycin
than those who received TMP-SMX or placebo [34]. In another randomized trial, the likelihood of
recurrent abscess formation was lower in patients who received TMP-SMX than in patients who
received placebo [38].

The approach to empiric antibiotic selection and duration of therapy is as described below. (See
'Approach to antibiotic therapy' below.)

Purulent cellulitis (no drainable abscess) — Patients with cellulitis associated with purulent
drainage (in the absence of drainable abscess) should be managed with antibiotic therapy (algorithm
2). The approach to empiric antibiotic selection and duration of therapy is as described below. (See
'Approach to antibiotic therapy' below.)

Approach to antibiotic therapy — The approach to empiric antibiotic selection is the same for all
patients with purulent infection, including:

● Patients with drainable abscess and relevant clinical criteria (see 'Role of antibiotic therapy'
above)

● Patients with cellulitis and associated purulent drainage, in the absence of a drainable abscess
(see 'Purulent cellulitis (no drainable abscess)' above)

It is useful to document the baseline appearance of the physical findings at the start of antibiotic
therapy. A baseline digital image should be taken to monitor progress.

Pathogens to cover — Patients with purulent infection (as defined in the preceding section)
should be managed with empiric therapy for infection due to MRSA, pending culture results
(algorithm 2) [1,2]. Empiric therapy for infection due to beta-hemolytic streptococci is usually not
necessary. Empiric therapy selection should be tailored to culture and susceptibility results when
available.

This approach is supported by findings of a study including 422 patients with purulent soft tissue
infection in which MRSA was the dominant organism (isolated from 59 percent of patients), followed
by MSSA (isolated from 17 percent of patients); beta-hemolytic streptococci accounted for a much
smaller proportion of these infections (2.6 percent) [41].

Initial management of patients with purulent infection in association with a pressure ulcer, a perioral
or peri-rectal site of infection, or prominent skin necrosis consists of antimicrobial therapy that
includes empiric coverage for MRSA as well as gram-negative and anaerobic organisms (pending
culture and susceptibility results) (algorithm 2).

Choosing an antibiotic agent — Issues related to choosing between oral and parenteral therapy
are discussed above. (See 'Oral versus parenteral therapy' above.)
Options for empiric oral therapy of purulent infection (ie, with activity against MRSA) include
clindamycin, TMP-SMX, or tetracyclines (doxycycline or minocycline) (table 1 and algorithm 2). The
efficacy of clindamycin and TMP-SMX for treatment of uncomplicated skin infection may be
considered comparable; this was illustrated in a randomized trial that included 524 patients with
uncomplicated skin infections, including both cellulitis and abscesses (cure rates for clindamycin and
TMP-SMX were 80 and 78 percent, respectively) [42]. For oral anti-MRSA coverage, however, we
generally favor TMP-SMX, doxycycline, or minocycline; clindamycin is less frequently chosen due to
its greater risk of Clostridioides (formerly Clostridium) difficile infection [43]. In addition, in vitro
susceptibility of MRSA to clindamycin must be confirmed [44,45].

Oxazolidinones (linezolid or tedizolid), delafloxacin, and omadacycline are additional agents with
activity against MRSA; they should be reserved for circumstances in which none of the other
regimens listed can be used. (See "Methicillin-resistant Staphylococcus aureus (MRSA) in adults:
Treatment of skin and soft tissue infections".)

For patients who are at high risk of an adverse outcome if infective endocarditis occurs (eg,
individuals with prior infective endocarditis, prosthetic heart material, unrepaired congenital heart
defect, and valvular dysfunction in a transplanted heart), we favor covering beta-hemolytic
streptococci in addition to MRSA pending culture data because of the possibility (albeit small) of
streptococcal involvement. In such patients who are undergoing incision and drainage, antibiotics
should be administered 60 minutes prior to the incision.

Antimicrobial therapy with activity against MRSA and beta-hemolytic streptococci may be achieved
with clindamycin or TMP-SMX monotherapy; there may be regional differences in MRSA susceptibility
to these agents [11,46]. In one systematic review that included 10 randomized trials evaluating the
utility of TMP-SMX for treatment of skin and soft tissue infection due to beta-hemolytic streptococci
or S. aureus, eight studies demonstrated efficacy of TMP-SMX for these conditions, although the
number of nonpurulent cellulitis cases in these trials due to beta-hemolytic streptococci was limited
[11]. In vitro susceptibility data support the notion that TMP-SMX is active against Streptococcus
pyogenes [46]. Because of uncertain streptococcal coverage with doxycycline or minocycline, we add
amoxicillin to each of these agents to ensure adequate antistreptococcal activity.

Options for empiric parenteral therapy of MRSA include vancomycin and daptomycin (algorithm 2);
alternative parenteral agents with activity against MRSA are summarized in the table (table 2) and are
discussed in detail separately. These agents also have activity against beta-hemolytic Streptococcus.
(See "Methicillin-resistant Staphylococcus aureus (MRSA) in adults: Treatment of skin and soft tissue
infections", section on 'Parenteral antibiotic therapy'.)
 Duration of therapy — The appropriate duration of therapy for treatment of skin and soft tissue
infection depends on the nature of the clinical presentation, and the clinical response should guide
duration of therapy.

Patients with mild infection who warrant outpatient management with oral antibiotic therapy should
have repeat evaluation after 24 to 48 hours to verify clinical response [2]. Patients with MRSA
responsive to oral therapy are typically treated for 5 days; extension of the duration (up to 14 days)
may be warranted in the setting of severe infection, slow response to therapy, or immunosuppression.
Lack of response may be due to infection with resistant organism(s), inadequate adherence, or
presence of a deeper, more serious infection than previously realized.

Patients with infection warranting parenteral therapy (in the absence of bacteremia or involvement
beyond soft tissue) are typically treated for a total duration of 5 to 14 days. Once there are signs of
clinical improvement with no evidence of systemic toxicity, antibiotics may be transitioned from
parenteral to oral therapy.

For patients with abscess that was detected radiographically, follow-up imaging may be useful for
assessing response to therapy. (See "Cellulitis and skin abscess: Clinical manifestations and
diagnosis", section on 'Diagnosis'.)

Recurrent infection — The approach to management of recurrent purulent cellulitis is the same as

the approach to management of the initial episode.

A recurrent abscess at a site of previous infection should prompt consideration of additional causes
such as pilonidal cyst, hidradenitis suppurativa, or presence of foreign material [2]. Surgical
exploration and debridement may be warranted, and suppressive antibiotics may be reasonable if no
drainable collection or treatable underlying condition is found. In patients with recurrent abscess
beginning in early childhood, evaluation for a neutrophil disorder is warranted. (See "Primary
disorders of phagocytic function: An overview".)

For patients with recurrent infection due to S. aureus, attempting decolonization is reasonable; this is
discussed separately. (See "Methicillin-resistant Staphylococcus aureus (MRSA) in adults: Prevention
and control", section on 'Targeted decolonization' and "Methicillin-resistant Staphylococcus aureus
(MRSA) in children: Prevention and control".)

SOCIETY GUIDELINE LINKS

Links to society and government-sponsored guidelines from selected countries and regions around
the world are provided separately. (See "Society guideline links: Skin and soft tissue infections".)
SUMMARY AND RECOMMENDATIONS

● Patients with skin and soft tissue infection may present with cellulitis, abscess, or both. The
approach to management depends on the nature of the clinical presentation. (See 'General
principles' above.)

● In general, patients with mild infection may be treated with an oral antibiotic regimen. The
decision to initiate parenteral antibiotic therapy should be based on individual clinical
circumstances such as severity of clinical presentation and patient comorbidities. We
recommend that patients with signs of systemic toxicity or rapid progression of erythema be
treated initially with parenteral antibiotics (Grade 1B). Parenteral therapy is also appropriate for
patients with persistence or progression of symptoms despite 48 hours of oral antibiotic therapy,
inability to tolerate oral therapy, or proximity of the lesion to an indwelling medical device (eg,
prosthetic joint or vascular graft). (See 'Oral versus parenteral therapy' above.)

● The management of patients with nonpurulent infection (ie, cellulitis or erysipelas in the absence
of abscess or purulent drainage) consists of antibiotic therapy (algorithm 1). (See 'Nonpurulent
infection' above.)

• Patients with nonpurulent cellulitis (eg, cellulitis with no purulent drainage or exudate and no
associated abscess) should be managed with empiric therapy for infection due to beta-
hemolytic streptococci and methicillin-susceptible Staphylococcus aureus (MSSA);
additional empiric coverage for methicillin-resistant S. aureus (MRSA) is warranted in the
circumstances summarized above (algorithm 1). (See 'Cellulitis' above.)

• Patients with erysipelas should be managed with empiric therapy for infection due to beta-
hemolytic streptococci. (See 'Erysipelas' above.)

• The duration of antibiotic therapy for treatment of nonpurulent infection should be

individualized depending on clinical response. In general, 5 days of therapy is appropriate for
patients with uncomplicated infection who have improved within this time period. Extension
of the duration (up to 14 days) may be warranted in the setting of severe infection and/or
slow response to therapy.

● For patients with recurrent nonpurulent cellulitis, we suggest administration of suppressive

antibiotic therapy (Grade 2B). Predisposing factors that can be alleviated should be addressed
whenever possible. (See 'Recurrent infection' above.)

● The management of patients with purulent infection depends on whether a drainable abscess is
present (see 'Purulent infection' above):
 • Patients with drainable abscess should undergo incision and drainage (algorithm 2). (See
'Drainable abscess present' above and "Technique of incision and drainage for skin
abscess".)

• The role of antimicrobial therapy in the setting of drainable abscess depends on individual
clinical circumstances.

- In general, we recommend antibiotic therapy for patients with multiple lesions, extensive
surrounding cellulitis, associated comorbidities or immunosuppression, signs of
systemic infection, or inadequate clinical response to incision and drainage alone
(Grade 1B), and we suggest antibiotic therapy for patients with skin abscess ≥2 cm
(Grade 2A), an indwelling device, or high risk for transmission of S. aureus to others
(Grade 2B). (See 'Role of antibiotic therapy' above.)

- For otherwise healthy patients with no risk factors, we suggest administering

antimicrobial therapy (Grade 2C). However, because many abscesses can be treated
successfully with incision and drainage alone, expert opinion varies, and it is reasonable
to forgo antibiotic therapy in otherwise healthy patients who have small (eg, <2 cm)
abscesses and no other comorbidities. (See 'Role of antibiotic therapy' above.)

• Patients with purulent cellulitis (ie, cellulitis associated with purulent drainage in the
absence of drainable abscess) should be managed with antibiotic therapy (algorithm 2).
(See 'Purulent cellulitis (no drainable abscess)' above.)

● Initial treatment for patients with purulent infection who warrant antibiotic therapy should consist
of treatment for MRSA (pending culture and susceptibility results) (algorithm 2). In addition,
initial treatment of patients with purulent infection in association with a pressure ulcer, a perioral
or peri-rectal site of infection, or prominent skin necrosis should consist of antimicrobial therapy
that includes empiric coverage for gram-negative and anaerobic organisms (pending culture and
susceptibility results). The duration of therapy should be individualized depending on clinical
response. (See 'Approach to antibiotic therapy' above.)

● The duration of antibiotic therapy for treatment of purulent infection should be individualized
depending on clinical response. Patients with MRSA responsive to oral therapy are typically
treated for 5 days; extension of the duration (up to 14 days) may be warranted in the setting of
severe infection, slow response to therapy, or immunosuppression. Patients with infection
warranting parenteral therapy (in the absence of bacteremia or involvement beyond soft tissue)
are typically treated for a total duration of 5 to 14 days. Once there are signs of clinical
improvement with no evidence of systemic toxicity, antibiotics may be transitioned from
parenteral to oral therapy. (See 'Duration of therapy' above.)
 Use of UpToDate is subject to the Subscription and License Agreement.

Topic 110530 Version 22.0

GRAPHICS

Clinical approach to management of nonpurulent cellulitis in adults

This algorithm outlines our approach to initial empiric antibiotic therapy of patients with cellulitis or erysipelas without an abscess or purulen
patient allergies or medication intolerance, expected toxicities or drug interactions, physician familiarity with the different antibiotics, expecte
MRSA: methicillin-resistant Staphylococcus aureus; MSSA: methicillin-susceptible S. aureus; IV: intravenously.
* The decision to administer antimicrobial therapy parenterally should be individualized based on clinical presentation and patient characteristics. Pare
infection, such as the presence of an immunocompromising condition (eg, neutropenia, recent organ transplant, advanced HIV infection, B cell or T cel
¶ Five days of antibiotic therapy is generally sufficient; extension up to 14 days may be warranted for slow response to therapy.
Δ Intravenous antibiotic dosing as follows (if 2 doses are listed for a given agent, the higher one is for patients with higher weights [eg, >120 kg] or mor
Cefazolin 1 to 2 g IV every 8 hours.
Ceftriaxone 1 to 2 g IV every 24 hours.
Ciprofloxacin 400 mg IV every 12 hours.
Clindamycin 600 to 900 mg IV every 8 hours.
Daptomycin 4 to 6 mg/kg IV every 24 hours.
Flucloxacillin 2 g IV every 6 hours.
Nafcillin 2 g IV every 4 hours.
Oxacillin 2 g IV every 4 hours.
Vancomycin 15 to 20 mg/kg/dose IV every 8 to 12 hours (not to exceed 2 g per dose).
A one-time 25 mg/kg loading dose is appropriate for severely ill patients.
A target trough level of 15 to 20 mcg/mL is appropriate in patients with severe and deep-seated infections.
A target trough of 10 to 15 mcg/mL is appropriate for nonsevere infections.
◊ Oral antibiotic dosing as follows (if 2 doses are listed for a given agent, the higher one is for patients with higher weights [eg, >120 kg] or more severe
Amoxicillin 875 mg orally twice daily.
Cefadroxil 1 g orally daily.
Cephalexin 500 mg orally 4 times daily.
Clindamycin 450 mg orally 3 times daily.
Dicloxacillin 500 mg orally 4 times daily.
Doxycycline 100 mg orally twice daily.
Flucloxacillin 500 mg orally 4 times daily.
Minocycline 200 mg orally once, then 100 mg orally every 12 hours.
Penicillin V potassium 500 mg orally every 6 hours.
Trimethoprim-sulfamethoxazole 1 to 2 double-strength tablets orally every 12 hours.
§ Penicillin and amoxicillin are preferred for erysipelas. For patients with a penicillin allergy, cephalexin (depending on the allergy), clindamycin, and trim
¥ Risk factors for MRSA include:
Recent (eg, within the prior 1 to 2 months) hospitalization or surgery.
Residence in a long-term care facility.
Hemodialysis.
HIV infection.
‡ For oral treatment of beta-hemolytic Streptococcus and MRSA, we generally favor trimethoprim-sulfamethoxazole, doxycycline, or minocycline becau
antistreptococcal activity and so are administered with amoxicillin. Other active options include linezolid, tedizolid, and delafloxacin, but these should b
† Vancomycin is the preferred parenteral anti-MRSA option. Daptomycin is an acceptable alternative agent for patients who have prior infection or colo
vancomycin. Additional alternative anti-MRSA agents include ceftaroline, linezolid, tedizolid, delafloxacin, omadacycline, telavancin, dalbavancin, and o
MRSA for further discussion of the approach to antibiotic selection.

Graphic 111732 Version 8.0

Approach to management of drainable abscess or skin infection with purulent drainage in adults
This algorithm outlines our approach to initial empiric antibiotic therapy and abscess management of patients with skin abscesses or skin inf
further depends on factors such as patient allergies or medication intolerance, expected toxicities or drug interactions, physician familiarity w

MRSA: methicillin-resistant S. aureus; IV: intravenously; MIC: minimum inhibitory concentration.

* This includes a history of infective endocarditis, prosthetic valve or perivalvular material, and unrepaired congenital heart disease. Refer to other UpTo
¶ We suggest antibiotic therapy for all patients with skin abscesses. However, because many abscesses can be successfully treated with incision and
the first box. An antibiotic-sparing approach may be particularly compelling in patients who have multiple antibiotic allergies or intolerances.
Δ The decision to administer antimicrobial therapy parenterally should be individualized based on clinical presentation and patient characteristics. Pare
infection, such as the presence of an immunocompromising condition (eg, neutropenia, recent organ transplant, advanced HIV infection, B cell or T cel
◊ Wound cultures are not necessary in healthy patients who do not receive antibiotics. For patients receiving antibiotics for a purulent skin infection, we
Severe local infection (eg, extensive cellulitis).
Systemic signs of infection (eg, fever).
History of recurrent or multiple abscesses.
Failure of initial antibiotic therapy.
Extremes of age (young infants or older adults).
Immunocompromising condition.
Indications for prophylaxis against infective endocarditis.
The community patterns of S. aureus susceptibility are unknown or rapidly changing.
§ Five days of antibiotic therapy is generally sufficient; extension up to 14 days may be warranted for slow response to therapy. In cases of slow improv
¥ Oral antibiotic dosing as follows (if 2 doses are listed for a given agent, the higher one is for patients with higher weights [eg, >120 kg] or more severe
Amoxicillin 875 mg orally twice daily.
Amoxicillin-clavulanate 875 mg orally twice daily.
Clindamycin 450 mg orally 3 times daily.
Ciprofloxacin 500 mg orally twice daily.
Doxycycline 100 mg orally twice daily.
Levofloxacin 750 mg orally once daily.
Metronidazole 500 mg orally 3 times daily.
Minocycline 200 mg orally once, then 100 mg orally twice daily.
Trimethoprim-sulfamethoxazole 1 to 2 double-strength tablets orally twice daily.
‡ For oral anti-MRSA coverage, we generally favor trimethoprim-sulfamethoxazole, doxycycline, or minocycline because of the greater associated risk o
but these should be reserved for circumstances in which the other options cannot be used.
† Intravenous antibiotic dosing as follows (if 2 doses are listed for a given agent, the higher one is for patients with higher weights [eg, >120 kg] or mor
Ampicillin-sulbactam 3 g IV every 6 hours.
Ceftriaxone 1 to 2 g IV every 24 hours.
Ciprofloxacin 400 mg IV every 12 hours.
Daptomycin 4 to 6 mg/kg IV every 24 hours.
Levofloxacin 750 mg IV once daily.
Metronidazole 500 mg IV every 8 hours.
Piperacillin-tazobactam 3.375 or 4.5 g IV every 6 hours.
Ticarcillin-clavulanate 3.1 g IV every 4 hours.
Vancomycin 15 to 20 mg/kg/dose IV every 8 to 12 hours (not to exceed 2 g per dose).
A one-time 25 mg/kg loading dose is appropriate for severely ill patients.
A target trough level of 15 to 20 mcg/mL is appropriate in patients with severe and deep-seated infections.
A target trough of 10 to 15 mcg/mL is appropriate for nonsevere infections.
** Vancomycin is the preferred parenteral anti-MRSA option. Daptomycin is an acceptable alternative agent for patients who have prior infection or colo
Additional alternative anti-MRSA agents include ceftaroline, linezolid, tedizolid, delafloxacin, omadacycline, telavancin, dalbavancin, and oritavancin; us
to MRSA for further discussion of the approach to antibiotic selection.
¶¶ Because of uncertain streptococcal coverage with doxycycline or minocycline, we add amoxicillin to each of these agents to ensure coverage, altho

Graphic 114919 Version 9.0

Oral antimicrobial therapy for treatment of skin and soft tissue infections due to methicillin-resistant
Staphylococcus aureus (MRSA) in adults

Treatment Adult dose

Clindamycin 450 mg orally 3 times daily

Trimethoprim-sulfamethoxazole (cotrimoxazole) 1 or 2 DS tablets twice daily

Doxycycline 100 mg orally twice daily

Minocycline 200 mg orally once, then 100 mg orally twice daily

Linezolid 600 mg orally twice daily

Tedizolid 200 mg orally once daily

Delafloxacin 450 mg orally twice daily

Omadacycline 300 mg orally once daily

The doses recommended above are intended for patients with normal renal function; the doses of some of these agents must be
adjusted in patients with renal insufficiency.

DS: double strength (ie, 160 mg trimethoprim with 800 mg sulfamethoxazole per tablet).

Data from:
1. Stevens DL, Bisno AL, Chambers HF, et al. Practice guidelines for the diagnosis and management of skin and soft tissue infections: 2014
update by the Infectious Diseases Society of America. Clin Infect Dis 2014; 59:e10.
2. Liu C, Bayer A, Cosgrove SE, et al. 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 2011; 52:e18 (note: for TMP-SMX dose in
osteomyelitis, refer to p e38).

Graphic 64178 Version 26.0

Parenteral antimicrobial therapy for treatment of skin and soft tissue infections due to methicillin-
resistant Staphylococcus aureus (MRSA) in adults

Drug Adult dose

Antibiotics of choice*

Vancomycin 15 to 20 mg/kg/dose every 8 to 12 hours, not to exceed 2 g per dose

Daptomycin ¶ 4 to 6 mg/kg IV once daily

Alternative agents

Short-acting agents with parenteral or oral dosing

Linezolid 600 mg IV (or orally) twice daily

Tedizolid 200 mg IV (or orally) once daily

Delafloxacin 300 mg IV twice daily (or 450 mg orally twice daily)

Omadacycline Δ 100 mg IV once daily (or 300 mg orally once daily)

Short-acting agent with parenteral dosing ◊

Ceftaroline 600 mg IV every 12 hours

Long-acting agents with parenteral dosing

Dalbavancin Single-dose regimen: 1500 mg once

Two-dose regimen: Initial dose 1000 mg, followed by 500 mg dose one week later

Oritavancin 1200 mg IV as a single dose

Telavancin 10 mg/kg once daily

IV: intravenously.
* In areas outside the United States where teicoplanin is available, some use it as the drug of choice for initial therapy of gram-positive
pathogens, while others favor its use for patients with intolerance to vancomycin.
¶ Dosing of daptomycin for treatment of skin and soft tissue infection is 4 to 6 mg/kg IV once daily; dosing for treatment of bacteremia is at
least 6 mg/kg IV once daily. (Refer to the UpToDate topic on treatment of MRSA bacteremia for further discussion.)
Δ A loading dose may be used: 200 mg IV over 60 minutes OR 100 mg IV over 30 minutes twice.
◊ Ceftobiprole has broad-spectrum activity against gram-positive and gram-negative organisms and is available in some countries outside the
United States.

Data from: Lui C, Bayer A, Cosgrove SE, et al. 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 2011; 52:285.

Graphic 110072 Version 8.0