American Journal of Clinical Dermatology

https://doi.org/10.1007/s40257-021-00659-8

REVIEW ARTICLE

Cellulitis: A Review of Current Practice Guidelines and Differentiation

from Pseudocellulitis
Michelle A. Boettler1 · Benjamin H. Kaffenberger2 · Catherine G. Chung2,3

Accepted: 14 November 2021

© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2021

Abstract
Cellulitis, an infection involving the deep dermis and subcutaneous tissue, is the most common reason for skin-related hos-
pitalization and is seen by clinicians across various disciplines in the inpatient, outpatient, and emergency room settings,
but it can present as a diagnostic and therapeutic challenge. Cellulitis is a clinical diagnosis based on the history of present
illness and physical examination and lacks a gold standard for diagnosis. Clinical presentation with acute onset of redness,
warmth, swelling, and tenderness and pain is typical. However, cellulitis can be difficult to diagnose due to a number of
infectious and non-infectious clinical mimickers such as venous stasis dermatitis, contact dermatitis, eczema, lymphedema,
and erythema migrans. Microbiological diagnosis is often unobtainable due to poor sensitivity of culture specimens. The
majority of non-purulent, uncomplicated cases of cellulitis are caused by β-hemolytic streptococci or methicillin-sensitive
Staphylococcus aureus, and appropriate targeted coverage of this pathogen with oral antibiotics such as penicillin, amoxicil-
lin, and cephalexin is sufficient. Even with rising rates of community-acquired methicillin-resistant Staphylococcus aureus,
coverage for non-purulent cellulitis is generally not recommended.

1 Introduction
Key Points
Cellulitis is an infection involving the deep dermis and sub-
Cellulitis is a common dermatologic condition that cutaneous tissue. While this is a common condition, it can
presents across multiple disciplines in the inpatient, present a diagnostic and therapeutic challenge. Over 14 mil-
outpatient, and emergency room settings, with a large lion patients in the US are treated for cellulitis annually,
impact on healthcare costs in the US. accounting for over 1% of all US hospitalizations and a cost
Currently, cellulitis is a clinical diagnosis with no gold of over $7 billion [1–3]. Additionally, it is estimated the
standard for diagnosis. Differentiation from pseudocel- misdiagnosis of cellulitis results in $195–$515 million in
lulitis and other clinical mimickers may be difficult, and avoidable health care spending annually in the US [4]. In
early consultation with a dermatologist may decrease this review, practice guidelines on cellulitis directed towards
misdiagnosis and improve outcomes. the practicing dermatologist will be discussed, including the
clinical spectrum, differential diagnoses, diagnostic proce-
Empiric treatment of cellulitis should be informed by dures, and treatment options.
clinical severity, risk factors, and likely etiological
organisms.
2 Epidemiology
* Catherine G. Chung
[email protected] In one observational retrospective cohort study utilizing
administrative claims data, the incidence of skin and soft
1
The Ohio State University College of Medicine, Columbus, tissue infection (SSTI) was 45.73 episodes per person-years
OH, USA
(PY) in the ambulatory setting and 2.19 episodes per PY
2
Division of Dermatology, Department of Internal Medicine, in the inpatient setting [5]. Cellulitis accounts for 10.1%
The Ohio State University Wexner Medical Center,
Columbus, OH, USA
of infectious disease hospitalizations, with an age-adjusted
3
rate of 156.2 hospitalizations per 100,000 persons and a
Department of Pathology, The Ohio State University Wexner
Medical Center, Columbus, OH, USA
52.2% rate increase from 1998 to 2006 [6]. In one study,

Vol.:(0123456789)
 M. A. Boettler et al.

the total number and incidence of ambulatory SSTIs sig- Systemic symptoms and signs, such as fevers, chills,
nificantly increased from 1997 to 2005, with abscess/cellu- fatigue, and leukocytosis, indicate a more severe infection.
litis accounting for 95% of this increase; annual visits more Depending on the clinical setting and inclusion criteria, the
than doubled from 4.6 million to 9.6 million and visit rates percentage of patients with fever ranges from 22.5 to 71%
increased from 17.3 to 32.5 visits per 1000 population [7]. [10–12, 23, 24]. The most common site of infection is the
The mean age in hospitalized patients ranges from 56 lower extremities; however, any area of skin or soft tissue
to 66.5 years [4, 8–11], with no sex predilection in some can be affected, but is rarely bilateral [12, 19, 25]. Orbital,
studies [4, 8, 11] but with sex predilection in some studies buccal, and perianal cellulitis are variants of cellulitis dif-
with small sample sizes [9, 10, 12]. Risk factors associated ferentiated by anatomic location. Cellulitis can be compli-
with cellulitis include previous history of cellulitis, pres- cated by the development of sepsis, lymphedema, recurrent
ence of Staphylococcus aureus and/or β-hemolytic strep- cellulitis, or abscess formation [26–28].
tococci (BHS) in the toe webs, tinea pedis, obesity, older
age, lymphedema/chronic leg edema, venous insufficiency,
prior saphenectomy, psoriasis, and presence of a wound [9, 4 Evaluation
13–16]. Risk factors may vary by geographic location, as,
among Japanese patients with different patient backgrounds Cellulitis is a clinical diagnosis based on the history of
such as lower body mass index (BMI), hypoalbuminemia, present illness and physical examination and lacks a gold
lymphedema, hypertension, and hyperlipidemia were sig- standard for diagnosis. The severity of symptoms, history of
nificantly associated with cellulitis hospitalizations, whereas recurrent infections, risk factors for specific microorganisms,
classic risk factors of chronic venous insufficiency, periph- and presence or absence of purulence direct antimicrobial
eral circulatory disturbance, and deep vein thrombosis were selection, route of administration, and whether outpatient
not [17, 18]. versus inpatient treatment is warranted.
Sex differences do exist and vary among different coun-
tries [18]. When comparing sex differences between Japa-
nese hospitalized cellulitis patients, males tend to be young,
obese with diabetes, and have skin barrier dysfunction,
whereas females tend to be cancer-bearing with low body-
weight [18]. Of hospitalized Spanish patients with cellulitis,
females are significantly more likely to be older, have prior
cellulitis history, have edema/lymphedema, and location
of cellulitis other than in the lower extremities; males are
significantly more likely to have positive pus cultures [19].

3 Clinical Presentation

Cellulitis typically presents with acute onset of redness

(rubor), warmth (calor), swelling (tumor), and tenderness
and pain (dolor) (Fig. 1). The spreading borders are often
poorly demarcated yet smooth. Erysipelas is a subtype of
cellulitis that is more superficial and with a sharply demar-
cated, often raised border [20]. Erysipeloid, a clinical variant
caused by Erysipelothrix rhusiopathiae, is most often local-
ized to the back of one hand and/or fingers with erythema
and well-defined, raised borders [21]. Streptococcal celluli-
tis may present as blistering distal dactylitis, involving only
the fat pad of the distal finger or, more rarely, the toe [22].
Clinical manifestations range from mild (e.g. localized ery-
thema) to severe, with cutaneous abscesses, purulence, and
life-threatening spread, including bacteremia. Lymphangitis,
tender regional lymph nodes, bullae, petechiae, and ecchy-
moses may develop. Peau d’orange or ‘orange peel’ skin may Fig. 1  Acute bacterial cellulitis presenting with characteristic bright
appear secondary to edema causing perifollicular dimpling. red erythema and edema of the lower leg
Cellulitis: A Review of Current Practice Guidelines and Differentiation from Pseudocellulitis

Superficial swabs, biopsy, and blood or needle aspirate or more targeted antimicrobial therapies. Additionally,
cultures have low yield and are not routinely recommended administrative data have demonstrated that hospitals that
for non-purulent cellulitis by the 2014 Infectious Diseases lack access to dermatology consultations have higher rates
Society of America guidelines; however, in patients with of cellulitis and lower rates of alternative skin disease
malignancy on chemotherapy, neutropenia, severe cell-medi- diagnosis-related groups per bed size [44]. While special-
ated immunodeficiency (SCID), immersion injuries, and ani- ists may not be available in every hospital setting, there is
mal bites, blood cultures are recommended and cultures and moderate agreement among dermatologists in differentiat-
microscopic examination of needle aspirates, tissue biopsy, ing and diagnosing cellulitis and pseudocellulitis, accord-
or swabs should be considered [29]. Skin swabs and super- ing to a recent study [45].
ficial wound cultures are often polymicrobial that cannot
distinguish naturally occurring transient bacteria and colo-
nizers from pathogenic microorganisms, which can compli-
cate the overall picture. Immunocompromised patients are 5 Imaging
more likely to have positive blood culture results and have a
wider differential diagnosis with higher risk for atypical bac- Imaging studies are often unnecessary in the work-up of
terial, viral, fungal, and parasitic agents [29, 30]. Of those cellulitis. However, given that clinical and laboratory evalu-
patients in the emergency department (ED) or inpatient set- ation can be non-specific and lack sensitivity, in the appro-
ting, the percentage of patients with a positive blood cul- priate setting radiographic imaging can provide diagnostic
ture ranges from 1 to 11% [23, 26, 31–34]. A 2020 scoping clues, evaluate disease extension, and aid in treatment plan-
review examining pretest probability of bacteremia in adult ning. Imaging is also helpful in the setting of rapid progres-
non-neutropenic inpatients with cellulitis demonstrated the sion or severe systemic manifestations to detect underlying
risk is low (< 10%) and low–moderate (10% to < 20%) in deep tissue involvement. Computed tomography (CT) and
patients with severe comorbidities such as liver cirrhosis and magnetic resonance imaging (MRI) have high spatial and
diabetes mellitus [26, 33, 35, 36]. The percentages are likely contrast resolution, providing detailed anatomic information
lower in the ambulatory setting. [46]. Although CT provides higher sensitivity for the detec-
Laboratory evaluation is often unnecessary in cases of tion of soft-tissue gas, MRI is the mainstay imaging tool
uncomplicated cellulitis or patients without comorbidi- for the diagnosis of soft-tissue infections [46]. Non-specific
ties. Findings tend to be non-specific: in patients hospi- soft-tissue swelling can be seen on radiograph [46]. Deep
talized or presenting to the ED with cellulitis, leukocytosis venous ultrasonography (U/S) can be useful to exclude non-
ranges from 34 to 51% [11, 23, 37, 38], while elevated infectious causes of soft-tissue swelling such as DVT [46,
inflammatory markers, such as erythrocyte sedimentation 47]. However, in the ED setting, the use of U/S for clinical
rate (ESR) and C-reactive protein (CRP), range from 59 cellulitis changed physician management in approximately
to 97% [11, 23, 38]. Although these laboratory tests are half of all cases [47]. U/S imaging of cellulitis is non-spe-
not specific to cellulitis, some may be helpful to indicate cific, demonstrating subcutaneous edema [46].
disease severity. In one study comparing cellulitis patients
with and without sepsis, higher leukocyte and neutrophil
counts, serum creatinine, and CRP were seen in patients 6 Microbiology
with cellulitis who developed sepsis [26]. Anti-DNase-
B (ADB) and anti-streptolysin-O (ASO) antibodies can Cellulitis is often separated into purulent and non-purulent
be used to detect recent BHS infection [39]. Of patients cellulitis. Microbiological diagnosis is often unobtainable
with diffuse, non-culturable cellulitis, 73% had at least due to poor sensitivity of culture specimens. The clinical
one positive titer, with 50% mounting both antibodies, isolation rate of a pathogen is < 20% in non-purulent cellu-
29% mounting ADB, and 21% mounting ASO [39]. ASO litis, and the relative frequencies of pathogens are therefore
and ADB antibodies are however non-specific and can be difficult to establish [29]. A systematic review of bacteremia
elevated in rheumatic fever, glomerulonephritis, pharyngi- in the setting of cellulitis demonstrated BHS as the most
tis, tonsillitis, and scarlet fever; thus, they may be helpful common pathogen (57%), followed by Gram-negative bacte-
in supporting the diagnosis but should be interpreted with ria (28%) and Staphylococcus aureus (14%) [20, 25, 39, 48].
caution [39]. In one study, patients diagnosed with BHS by serology or
Data are available supporting the involvement of spe- culture had a 97% response to B-lactam antibiotic treatment,
cialists, with multiple studies showing benefits with der- while those who did not have BHS by serology or culture
matology and infectious disease consultations [40–43]. had a 91% response, supporting BHS as the most common
Having specialists available may lead to a discussion of etiology of cellulitis even in the absence of pathogen isola-
alternative etiologies, deeper microbial analysis, and/ tion despite the emergence and uptick in methicillin-resistant
 M. A. Boettler et al.

Staphylococcus aureus (MRSA) infections [39]. Staphylo- Table 1  Risk factors and associated pathogens in cellulitis
coccus aureus is the most common isolate of acute puru-
Risk factor Pathogen
lent SSTI [49]. There is an increasing concern over anti-
biotic resistance and rising rates of community-associated Animal and human bites
MRSA. However, in a randomized clinical trial comparing Human bite [51] Peptostreptococcus
anti-methicillin-sensitive Staphylococcus aureus (MSSA) Fusobacterium
and anti-MRSA antibiotics in the outpatient treatment of Veillonella
Clostridium
cellulitis without abscess, the addition of trimethoprim-sul- α-hemolytic Streptococci
famethoxazole to cephalexin did not improve outcomes [50]. β-hemolytic Streptococci
The most common pathogen of cellulitis is bacteria; how- Staphylococcus aureus
ever, immunosuppression creates a unique environment for Staphylococcus epidermidis
Corynebacterium sp.
opportunistic infection with fungi and atypical bacteria such Eikenella corrodens
as Mycobacterium species (Fig. 2). Bacteroidesfragilis
Pathogens related to specific risk factors are charted Prevotella
in Table 1. Inoculation via animal or human bite, aquatic Porphyromonas
exposure, immunosuppression, and chronic liver and kidney Cat bite [51] Pasteurella multocida
Bacteroides fragilis
disease can increase the likelihood of an atypical pathogen, Prevotella
necessitating a different treatment choice. Porphyromonas
Peptostreptococcus
Fusobacterium sp.
Veillonella pawula
7 Histopathology Dog bite [51] Pasteurella multocida
Staphylococcus sp.
Currently, tissue biopsy is considered low yield and is not Streptococcus sp.
recommended unless the patient is immunocompromised Corynebacterium sp.
(e.g. history of transplant, in the setting of systemic ster- Bacteroides fragilis
Prevotella
oids or other immunosuppressive treatment, or in individu- Porphyromonas
als with HIV/AIDS), febrile with neutropenia, or at risk of Peptostreptococcus
specific pathogens (water or animal bites exposure) [1, 29]. Fusobacterium sp.
If there is no improvement with antibiotic initiation and/or Veillonella pawula
Capnocytophaga canimorsus
there is suspicion of non-infectious pseudocellulitis, punch
Marine water exposure [52] Vibrio spp.
biopsy should be obtained for histopathologic analysis. His- Aeromonas spp.
topathology of cellulitis includes dermal edema, lymphatic Shewanella spp.
and/or vascular dilation, and prominent diffuse neutrophilic Erysipelothrix rhusiopathiae
Mycobacterium marinum
Streptococcus iniae
Clostridium
Pseudomonas
Plesiomonas
Immunosuppression Mycobacterium hemophilum [53,
54]
Cryptococcus neoformans [53]
Mycobacterium tuberculosis [54]
Vibrio cholerae [55]
Helicobacter [56, 57]
Fusarium [58]
Chryseobacterium meningosepti-
cum [59]
Cedecea sp. [60]
Serratia marcescens [61]
Stenotrophomonas maltophilia
[62]
Streptococcus pneumoniae [63]
Escherichia coli [64, 65]
Campylobacter fetus [66]
Shewanella putrefaciens [67]
Fig. 2  Cryoptococcal cellulitis in a patient with chronic liver disease Hemophilus influenzae [68]
presenting with well-circumscribed bright red erythema and edema of
the upper extremity
Cellulitis: A Review of Current Practice Guidelines and Differentiation from Pseudocellulitis

Table 1  (continued) metals such as nickel is important to determine an etiology

Risk factor Pathogen [77].
Lymphedema is a chronic progressive localized form
Chronic liver disease [61] Vibrio cholerae [55] of tissue swelling due to excessive retention of lymphatic
Group A Streptococci
Staphylococcus aureus
fluid in the interstitial compartment caused by impaired
Escherichia coli lymphatic drainage [28]. Lymphedema often causes over-
Klebsiella spp. lying cutaneous changes (Fig. 4) and is commonly compli-
Pseudomonas aeruginosa cated by secondary skin infections, with one study demon-
Aeromonas spp.
strating 29% of patients with lymphedema were admitted
Chronic kidney disease Cedecea sp. [60]
Helicobacter cinaedi [57]
for intravenous antibiotics over the course of 12 months
Escherichia coli [65] [78]. Information on family history, history of penetrating
trauma, infection, cancer, heart failure, hypothyroidism,
hypoalbuminemia, sepsis, venous or lymphatic obstruc-
infiltration that involves the dermis and may extend to the tion, inguinal/axillary radiation, lymphadenectomy, and
soft tissues. In cases where cellulitis is accompanied by travel to areas endemic for filariasis should be elicited
tissue necrosis, secondary necrotizing vasculitis may also [28]. Lymphoscintigraphy is the standard imaging tool to
be present. Identification of microorganisms ranges from confirm the diagnosis of lymphedema [28]. Acute inflam-
numerous (especially in the setting of necrosis) to scant. matory edema is characterized by bilateral, erythematous,
Lymphocytes, histiocytes, and granulation tissue may and edematous plaques involving areas of dependence and
develop in established, more chronic cases [69–72]. sparing areas of pressure, most commonly the thighs and
abdomen of critically ill patients with acute fluid overload
[79]. This type of pseudocellulitis can mimic cellulitis
8 Differential Diagnosis clinically and histopathologically, given the presence of
neutrophils [79].
Cellulitis can be difficult to diagnose due to a number of Erythema migrans, or primary Lyme disease, is an ery-
infectious and non-infectious clinical mimickers (Table 2) thematous macule or papule that expands centrifugally to
that may present with the characteristic warmth, erythema, form a patch of erythema that may develop a targetoid, ‘bull-
edema, pain, fever, and/or leukocytosis seen in cellulitis. In seye’ appearance caused by Ixodes tick transmitting Borrelia
fact, 30 [43, 73] to 74% [74] of cellulitis cases are eventually burgdorferi. The expanding nature of erythema can clini-
diagnosed as a pseudocellulitis. Of the causes of pseudocel- cally mimic cellulitis, but pertinent history regarding travel
lulitis, venous stasis dermatitis, contact dermatitis, eczema, or residence in an endemic, history of recent tick bite within
lymphedema, and erythema migrans are among the top con- 2 weeks, late spring/summer, and locations atypical for bac-
tributors to misdiagnosis [43, 73, 74]. A thorough history terial cellulitis such as the axilla, popliteal fossa, groin, and
and clinical examination and specialist consultation with waist can provide clues to Lyme disease [80].
dermatology when available can help differentiate cellulitis DVT, necrotizing soft tissue infection (NSTI), septic
from pseudocellulitis [44]. arthritis, and gout are less common causes of pseudocel-
Venous stasis dermatitis is caused by venous hyperten- lulitis but can cause significant morbidity and mortality if
sion resulting from retrograde flow, which subsequently misdiagnosed and are therefore essential to differentiate.
creates an inflammatory process and further tissue changes DVT is caused by a blood clot obstruction of the venous
[75]. Stasis dermatitis presents as poorly demarcated ery- system, which can migrate to the pulmonary system or
thematous plaques of the lower legs bilaterally, classically brain, causing pulmonary embolism or stroke, respec-
involving the medial malleolus, whereas cellulitis of the tively. DVT often presents similarly to cellulitis, with uni-
lower extremities is typically unilateral and infrequently, if lateral limb swelling, erythema, and pain. In a systematic
ever, bilateral. Duplex ultrasound can be utilized to demon- review and meta-analysis on the risk of DVT in patients
strate venous reflux. with cellulitis and erysipelas, 3.1% of 1054 patients with
Contact dermatitis is due to an irritant or allergen causing cellulitis or erysipelas had any type of DVT via compres-
a delayed type IV hypersensitivity reaction (Fig. 3), with sion ultrasound [81]. The overall prevalence of DVT in
pruritus as the classic symptom, differentiating it from cel- patients with cellulitis or erysipelas appears to be low;
lulitis. The use of medical adhesives over a wound can cause however, if there is concern for DVT, patients should be
a hypersensitivity reaction that must be distinguished from sent to an urgent care center or ED for further evaluation.
cellulitis [76]. Investigating triggers of contact dermatitis Symptoms of gout and septic arthritis can cause a rapid
such as the use of topical antibiotics, hair dye, cosmetic onset of pain, erythema, swelling, and warmth over an over-
products, fragrances, rubber, latex, plants, acrylates, and lying joint. Patients with underlying joint pain require joint
 M. A. Boettler et al.

Table 2  Differential diagnosis of cellulitis

Characteristics Evaluation

Infectious
Necrotizing soft tissue infection Rapidly progressive erythema or purpura, pain out of proportion, fever, Evaluation by general surgery
systemic toxicity, swelling, hemorrhagic or bluish bullae, skin necrosis or if clinically suspicious
extensive ecchymosis, gas, or crepitus
Cutaneous abscess Nodule with fluctuance and drainage in addition to features of cellulitis
Herpes simplex Grouped vesicles on an erythematous base. Most often appear in the trigemi- Polymerase chain reaction
nal or sacral ganglia distribution but may occur at other body locations is the most sensitive and
preferred diagnostic test
if clinical presentation is
unclear [89, 90]
Herpes zoster Painful dermatomal erythematous patch or plaque with grouped vesicles Polymerase chain reaction
without crossing the midline unless disseminated is the most sensitive and
preferred diagnostic test
if clinical presentation is
unclear [90]
Erythema migrans Well-demarcated circular erythematous macule, patch, or plaque that History of a tick bite or recent
expands and may develop central clearing, forming a targetoid bullseye travel to an endemic area
appearance
Septic arthritis Painful, swollen, warm, erythematous skin overlying a joint. It can be Joint aspiration for diagnosis
accompanied by fever and leukocytosis
Non-infectious inflammatory
Sweet syndrome Abrupt painful, edematous, erythematous papules, plaques, or nodules that Both major criteria and two
coalesce. Fever and leukocytosis frequently accompany the cutaneous of four minor criteria are
lesions required for diagnosis [91]
Contact dermatitis Acute presentation of erythema or vesicles. Scaling and fissures are seen
chronically. Well-demarcated geometric or non-organic pattern and distri-
bution. Pruritus is the most common symptom
Erythema nodosum Erythematous, tender, immobile nodules and plaques most commonly on the
bilateral shins [92]
Gout Painful, swollen, warm, erythematous skin overlying a joint. It can be History of gout. Joint aspira-
accompanied by fever and leukocytosis. Subcutaneous gout nodules (tophi) tion for diagnosis
appear as white or yellow
Acute bursitis Painful, swollen, warm, erythematous skin overlying a bursa. Often due to
trauma or infection
Pyoderma gangrenosum Papule, pustule, vesicle, or nodule that rapidly expands and forms an erosion
or ulcer. Pathergy often seen
Familial mediterranean fever Erysipelas-like reaction with a tender, erythematous, raised lesion over the Recurrent nature and positive
lower extremities during an attack family history
Vascular
Stasis dermatitis Bilateral, chronic history that improves with leg elevation, compression
therapy, and topical corticosteroids. Often hyperpigmentation, varicose
veins, ulcerations
Deep vein thrombosis Unilateral swelling, erythema, warmth, or tenderness. Risk factors include Ultrasound is a sensitive and
prolonged immobilization, prior history of deep vein thrombosis, active relatively cheap tool for
malignancy, recent surgery, and pregnancy diagnosis [93]
Erythromelalgia Bilateral, warm, erythematous extremities with burning paresthesia that is
out of proportion to clinical examination and improves with cooling
Lymphatics
Lymphedema Feeling of heaviness and discomfort commonly accompanies swelling. Pit- Lymphoscintigraphy is the
ting is variable in patients with lymphedema and can be absent if chronic. standard imaging tool to
Cutaneous and subcutaneous thickening if severe confirm the diagnosis [28]
Acute inflammatory edema Bilateral, erythematous, edematous plaques most frequently on the abdomen
and thighs. Often in critically ill, fluid overloaded patients with high body
mass index
Cellulitis: A Review of Current Practice Guidelines and Differentiation from Pseudocellulitis

Table 2  (continued)
Characteristics Evaluation

Neoplasm
Carcinoma erysipeloides Cutaneous metastasis presenting as a fixed erythematous, well-demarcated Most often associated with
patch or plaque often overlying a primary cancer breast carcinoma, which
presents on the chest [94]
Miscellaneous
Fixed drug eruption Usually non-tender eruption soon after drug ingestion. Most often on the
face, lips, trunk, genitalia, hands, and feet [95]
Hypersensitivity reaction Pruritic reaction with variable presentation depending on the type of hyper-
sensitivity

hemodynamic collapse, and organ failure, with delays in sur-

gical debridement, antibiotic use, and resuscitation impact-
ing morbidity and mortality. The absence of fever or early
cutaneous manifestations, non-specific radiographic tests,
recent history of non-penetrating trauma, surgery, or child-
birth, and chronic underlying diseases can complicate diagno-
sis [82]. Classic clinical findings of NSTI include soft tissue
edema (75%), erythema (72%), severe pain (72%), tenderness
(68%), fever (60%), and skin blebs, bullae, or necrosis (38%)
[83]. While certain ‘hard’ clinical signs are associated with
necrotizing fasciitis, such as hemodynamic instability, crepi-
tance, skin necrosis, bullae, and gas on x-ray, a majority of
Fig. 3  Allergic contact dermatitis to neomycin
patients unfortunately present with none of these [84]. The
Laboratory Risk Indicator for Necrotizing Fasciitis (LRINEC)
is a clinical tool to screen for necrotizing fasciitis based on
CRP, total white blood cell count, hemoglobin, sodium, cre-
atinine, and glucose [85]. An LRINEC score of six or greater
confers a higher risk of necrotizing fasciitis [85]. The LRINEC
score is most specific for severe disease [86]. Newer scoring
systems such as the SIARI score (Site other than the lower
limb, Immunosuppression, Age ≤ 60 years, Renal impairment
[creatinine > 141 µmol/L], and Inflammatory markers [CRP >
150, white blood cell count > 25]) developed in New Zealand
showed greater diagnostic ability compared with the LRINEC
score [87] (Table 3). A modified LRINEC score has also been
created with alteration of laboratory parameters (including
fibrinogen levels and erythrocyte counts) and the addition of
clinical symptoms, leading to higher positive predictive value
without losing specificity [88].
If clinical suspicion of necrotizing fasciitis is high,
Fig. 4  Chronic lymphedema: bilateral lower extremity edema with immediate surgical evaluation and antibiotic treatment are
overlying skin changes, including thickening, hyperpigmentation, and required.
scale

fluid analysis via arthrocentesis for diagnosis and treatment 9 Treatment

to prevent joint destruction and permanent damage.
NSTI, such as necrotizing cellulitis and necrotizing A systematic review and meta-analysis showed a high
fasciitis, is a severe, life-threatening infection character- overall treatment failure rate of nearly 20%, likely due to
ized by widespread tissue destruction, systemic toxicity, a significant proportion of cellulitis misdiagnoses [96]. As
 M. A. Boettler et al.

Table 3  The laboratory risk indicator for necrotizing fasciitis [85] rates of relapse or recurrence, or patients’ pain scores and
satisfaction between intravenous versus oral routes of anti-
Parameter Range Score
biotic administration [103]. In the hospital setting, the use of
Hb (g/dL) > 13.5 0 antimicrobials with broad aerobic gram-negative, anti-pseu-
11–13.5 1 domonal, or broad anaerobic activity are frequently used
< 11 2 without a corresponding decrease in clinical failure [104]. A
White blood cells (­ 109/L) < 15 0 multicenter clinical trial on the use of single, renally admin-
15–25 1 istered, long-acting intravenous dalbavancin for stable ED
> 25 2 patients with advanced SSTI showed a significant reduction
Sodium (mmol/L) < 135 2 in the hospitalization rate [105]. In severely compromised
Creatinine (μmol/L) > 141 2 patients with malignancy on chemotherapy, neutropenia,
Glucose > 10 1 SCID, immersion injuries, and animal bites, broad spectrum
C-reactive protein > 150 4 coverage such as vancomycin plus piperacillin/tazobactam
or imipenem/meropenem may be considered [29]. Purulent
Score ≤ 5 = < 50% risk (low); 6–7 = intermediate risk; ≥ 8 = > 75%
risk (high)
cellulitis should be managed with incision and drainage,
with the addition of antibiotics for those who are immuno-
compromised and display systemic signs of infection, signs
mentioned previously, the causative microbe in cellulitis is of deeper infection, or antibiotic failure [29].
often not recovered, and treatment is most often initiated Antibiotic options and standard dosing based on the
empirically. The majority of non-purulent, uncomplicated route of administration are shown in Table 4. Poor out-
cases of cellulitis are caused by BHS or MSSA and appropri- comes are significantly associated with age, previous epi-
ate targeted coverage of this pathogen is sufficient. Patients sodes of cellulitis, prior wounds and skin lesions, venous
can often be managed with oral antibiotics. Suitable options insufficiency, lymphedema, immunosuppression, and
include penicillin, amoxicillin, and cephalexin. Cephalexin, involvement of the lower limbs [8].
dicloxacillin, and amoxicillin-clavulanate cover for MSSA A number of patients are susceptible to recurrent cel-
in addition to BHS. Patients with a penicillin allergy can lulitis infections. A previous episode of cellulitis has a
be treated with clindamycin or trimethoprim-sulfamethox- recurrence rate ranging from 8 to 20% annually, especially
azole, which demonstrate near equivalent cure rates [97]. if occurring on the legs [29]. In a prospective case-control
The recommended duration of treatment is 5 days unless study, 49% of hospitalized cellulitis patients had a positive
infection has not improved during this time span, and treat- history of at least one cellulitis episode before entering
ment should be extended [29, 98]. Even with rising rates the study [106]. Obesity (BMI ≥ 30) and previous ipsilat-
of community-acquired MRSA, β-lactam and non-β-lactam eral surgical procedure of the site of cellulitis were sig-
antibiotic clinical failure rates for uncomplicated cellulitis nificantly more common in these patients [106]. Episodes
do not differ [99]. Coverage of MRSA for non-purulent cel- of cellulitis cause post-inflammatory lymphatic damage,
lulitis is generally not recommended unless patients have leading to a vicious cycle since lymphedema predisposes
failed initial antibiotic treatment, markedly impaired host to cellulitis [107]. A Cochrane systematic review of inter-
defenses, systemic inflammatory response syndrome and ventions for recurrent cellulitis demonstrated antibiotic
hypotension, cellulitis associated with penetrating trauma, prophylaxis, compared with no treatment or placebo,
evidence of MRSA infection elsewhere, nasal colonization decreased the risk of cellulitis recurrence after treatment
with MRSA, community-associated MRSA-prevalent set- by 69%, and reduced the incidence rate of cellulitis by
ting, or injection drug use [29, 100, 101]. Oral antibiotics 56% [108]. However, the protective effects of antibiotic
with MRSA coverage such as trimethoprim-sulfamethoxa- prophylaxis did not persist after treatment cessation [108].
zole, clindamycin, doxycycline, and linezolid may be chosen Proposed regimens include oral penicillin 250 mg twice
by providers. The rates of MRSA vary considerably depend- daily, oral penicillin 2–4 g daily depending on body weight
ing on geographic location, hospitalization, and residence, (1 g twice daily if < 90 kg; 1 g + 2 g daily if 90–120 kg;
and should be taken into consideration [102]. Systemic signs 2 g twice daily if > 120 kg), intramuscular penicillin 1.2
of infection (e.g. fever, tachycardia, leukocytosis) warrant million units every 15 days, and oral erythromycin 250
systemic antibiotics [29]. However, the benefit of the intra- mg twice daily [108]. It has been suggested that those who
venous versus oral route of antibiotic administration for experience three to four episodes of cellulitis per year may
cellulitis of similar severity has not been proven to be sig- benefit from prophylactic antibiotics, whereas the system-
nificant [98, 103]. Various randomized control trials have atic review demonstrated the effects are most relevant for
shown no evidence of difference in clinical response rates, people after just two episodes of leg cellulitis occurring
mean days until no advancement of the area of cellulitis, within a period of up to 3 years [29, 108]. Adverse effects
Cellulitis: A Review of Current Practice Guidelines and Differentiation from Pseudocellulitis

Table 4  Antimicrobial treatment options for cellulitis by route Treatment for long-term success of cellulitis includes
targeting risk factors. It is estimated 10–30% of people
Antibiotic Standard dosing
who have one episode of cellulitis will experience repeated
Oral attacks across different time intervals [108]. Diabetes mel-
Penicillin 500 mg PO qid litus, history of previous episodes of cellulitis, edema of
­Amoxicillina 500 mg PO tid the lower extremities, peripheral vascular disease, obesity,
Amoxicillin-clavulanatea 875–125 mg PO bid immunosuppression, alcoholism, and dermatomycosis are
Dicloxacillin 500 mg PO qid among the most common underlying conditions associated
­Cephalexina 500 mg PO qid with cellulitis [12, 108]. The treatment or management of
­Clindamycinb 300 mg PO tid underlying predisposing conditions may prevent recurrence.
Alternatives Various methods to reduce limb lymphedema and edema,
Trimethoprim- 1–2 double-strength tablets PO venous insufficiency, local skin care and hygiene maintain
sulfamethoxazolea,b q12h the integrity of skin and decrease the likelihood of creation
­Doxycyclineb 100 mg PO qid of a nidus for infection. Combined decongestive therapy is
­Minocyclineb 200 mg PO once, then 100 mg a multimodal approach to lymphedema that includes com-
PO q12h
pression therapy, manual lymphatic drainage, exercise, and
Flucloxacillin 500 mg PO qid
skincare and is regarded as the standard of treatment [28].
­Delafloxacinb 450 mg PO bid
The acquisition of acute pneumatic compression devices for
­Linezolidb 600 mg PO bid
lymphedema is associated with significant reductions in epi-
­Tedizolidb 200 mg PO bid
sodes of cellulitis and patient care costs [109]. Weight loss
Intravenous
for obese patients is strongly recommended and leg elevation
­Penicillina 2–4 million units IV, q4h–q6h
in early-stage disease can reduce fluid accumulation [28].
­Cefazolina 1–2 g IV q8h
Diuretics do not play a role in lymphedema and surgical
Nafcillin 1–2 g IV q4–6h
intervention is considered when conservative management
Ceftriaxone 1–2 g IV q24h
fails [28]. Lymphaticovenular anastomosis is a potential
­Clindamycinb 600–900 mg IV q12h
surgical intervention for lymphedema that has shown a sta-
Alternatives
tistically significant reduction in the rate of cellulitis [110].
­Vancomycina,b 15–20 mg/kg IV q8–12h
Interdigital toe space examination and treatment for fis-
­Daptomycina,b 4–6 mg/kg IV q24h
suring, scaling, or maceration in lower extremity cellulitis
­Ceftarolinea,b 600 mg IV q12h
may eradicate pathogen colonization and reduce the inci-
­Linezolidb 600 mg IV bid
dence of recurrent infections [29]. In a randomized control
­Tedizolidb 200 mg IV bid
trial of participants with chronic leg edema and recurrent
­Dalbavancina,b 1500 mg IV single dose
cellulitis, compression therapy caused a lower incidence of
­Ciprofloxacina 400 mg IV q12h
recurrent cellulitis than conservative treatment and the trial
Flucloxacillin 2 g IV q6h
was stopped for efficacy [111]. Antibiotic prophylaxis fail-
Oxacillin 1–2 g IV q4h
ure with penicillin is associated with a BMI ≥ 33, three or
This is a standard guide of possible treatment regimens and is not more previous episodes of cellulitis, and presence of edema,
meant to be all-inclusive. Clinicians should refer to their institution’s which a large proportion of patients who benefit from com-
antimicrobial stewardship guidelines and other resources for prescrib- pression therapy have [111, 112]. However, compression
ing
therapy should not be used during an active infection.
PO orally, bid twice daily, TID three times daily, QID four times
daily, IV intravenous, qxh every x hours, MRSA methicillin-resistant
Staphylococcus aureus
a
May need adjustment for renal function 10 Patient Education
b
Provides MRSA coverage
Patients should be counseled on cellulitis and the pre-
scribed antibiotics. Completion of the full course of anti-
are rare and most commonly consist of gastrointestinal biotics should be encouraged despite improvement before
symptoms, mainly nausea and diarrhea; rash, with no the course ends; improvement should be seen within
cases of severe cutaneous adverse reactions reported; and 24–48 h of initiating antibiotics. Providers may find util-
thrush [108]. Importantly, antimicrobial resistance was not ity to mark the margins of the infection to indicate to
assessed. patients whether it is improving or worsening. Patients
 M. A. Boettler et al.

should inform their provider if the margin is spreading 4. Weng QY, Raff AB, Cohen JM, Gunasekera N, Okhovat
or not improving, the pain worsens, or persistent fever JP, Vedak P, et al. Costs and consequences associated with
misdiagnosed lower extremity cellulitis. JAMA Dermatol.
develops. 2017;153(2):141–6.
Patients should maintain appropriate hand hygiene when 5. Miller LG, Eisenberg DF, Liu H, Chang CL, Wang Y, Luthra R,
attending to the site of infection and keep the area clean and et al. Incidence of skin and soft tissue infections in ambulatory
dry. As feasible, patients should elevate the site to reduce and inpatient settings, 2005–2010. BMC Infect Dis. 2015;15:362.
6. Christensen KL, Holman RC, Steiner CA, Sejvar JJ, Stoll BJ,
edema [29]. Schonberger LB. Infectious disease hospitalizations in the United
States. Clin Infect Dis. 2009;49(7):1025–35.
7. Hersh AL, Chambers HF, Maselli JH, Gonzales R. National
11 Summary trends in ambulatory visits and antibiotic prescribing for skin and
soft-tissue infections. Arch Intern Med. 2008;168(14):1585–91.
8. Collazos J, de la Fuente B, García A, Gómez H, Menéndez C,
Cellulitis is a common condition that causes significant Enríquez H, et al. Cellulitis in adult patients: a large, multicenter,
health care costs, can be difficult to distinguish clinically, observational, prospective study of 606 episodes and analysis
and lacks a gold standard for diagnosis. A thorough his- of the factors related to the response to treatment. PLoS ONE.
2018;13(9): e0204036.
tory and clinical examination can assist in the diagnosis, 9. Björnsdóttir S, Gottfredsson M, Thórisdóttir AS, Gunnarsson
and evaluation with laboratory and imaging investigation GB, Ríkardsdóttir H, Kristjánsson M, et al. Risk factors for acute
is usually not necessary. The clinical isolation rate of a cellulitis of the lower limb: a prospective case-control study. Clin
pathogen is low and treatment is most often begun empiri- Infect Dis. 2005;41(10):1416–22.
10. Chartier C, Grosshans E. Erysipelas. Int J Dermatol.
cally. The most significant risk factor for the development 1990;29(7):459–67.
of cellulitis is a prior episode of cellulitis, and treatment 11. Lazzarini L, Conti E, Tositti G, de Lalla F. Erysipelas and cel-
should also address underlying conditions and patient lulitis: clinical and microbiological spectrum in an Italian tertiary
education. care hospital. J Infect. 2005;51(5):383–9.
12. Koutkia P, Mylonakis E, Boyce J. Cellulitis: evaluation of possi-
ble predisposing factors in hospitalized patients. Diagn Microbiol
Declarations Infect Dis. 1999;34(4):325–7.
13. Quirke M, Ayoub F, McCabe A, Boland F, Smith B, O’Sullivan
Funding No sources of funding were used to assist in the preparation R, et al. Risk factors for nonpurulent leg cellulitis: a systematic
of this article. review and meta-analysis. Br J Dermatol. 2017;177(2):382–94.
14. McNamara DR, Tleyjeh IM, Berbari EF, Lahr BD, Martinez
JW, Mirzoyev SA, et al. Incidence of lower-extremity cellulitis:
Conflicts of Interest Michelle Boettler, Benjamin Kaffenberger and a population-based study in Olmsted county, Minnesota. Mayo
Catherine Chung state no conflicts of interest. Clin Proc. 2007;82(7):817–21.
15. Dupuy A, Benchikhi H, Roujeau JC, Bernard P, Vaillant L,
Ethics Approval Not applicable. Chosidow O, et al. Risk factors for erysipelas of the leg (cel-
lulitis): case-control study. BMJ. 1999;318(7198):1591–4.
Informed Consent Not applicable. 16. Wakkee M, de Vries E, van den Haak P, Nijsten T. Increased
risk of infectious disease requiring hospitalization among
Consent for publication Not applicable. patients with psoriasis: a population-based cohort. J Am Acad
Dermatol. 2011;65(6):1135–44.
Data Availability Data sharing is not applicable to this article as no 17. Norimatsu Y, Ohno Y. Predictors for readmission due to cel-
datasets were generated or analyzed during the current study. lulitis among Japanese patients. J Dermatol. 2021;48(5):681–4.
18. Norimatsu Y, Ohno Y. Sex-based differences in Japanese
Code availability Not applicable. patients with cellulitis. J Dermatol. 2021;48(11):1797–8.
19. Collazos J, de la Fuente B, de la Fuente J, García A, Gómez
Author Contributions All authors contributed to the review conception H, Rivas-Carmenado M, et al. Sex differences in hospitalized
and design. The first draft of the manuscript was written by Michelle adult patients with cellulitis: a prospective, multicenter study.
Boettler and all authors commented on previous versions of the manu- Int J Infect Dis. 2021;104:584–91.
script. All authors read and approved the final manuscript. 20. Eriksson B, Jorup-Rönström C, Karkkonen K, Sjöblom AC,
Holm SE. Erysipelas: clinical and bacteriologic spectrum and
serological aspects. Clin Infect Dis. 1996;23(5):1091–8.
References 21. Veraldi S, Girgenti V, Dassoni F, Gianotti R. Erysipeloid: a
review. Clin Exp Dermatol. 2009;34(8):859–62.
22. Olivieri I, Scarano E, Padula A, Giasi V, Priolo F. Dactyli-
1. Raff AB, Kroshinsky D. Cellulitis: a review. JAMA.
tis, a term for different digit diseases. Scand J Rheumatol.
2016;316(3):325–37.
2006;35(5):333–40.
2. Zhang M, Markova A, Harp J, Dusza S, Rosenbach M, Kaf-
23. Hook EW, Hooton TM, Horton CA, Coyle MB, Ramsey PG,
fenberger BH. Dermatology-specific and all-cause 30-day and
Turck M. Microbiologic evaluation of cutaneous cellulitis in
calendar-year readmissions and costs for dermatologic diseases
adults. Arch Intern Med. 1986;146(2):295–7.
from 2010 to 2014. J Am Acad Dermatol. 2019;81(3):740–8.
24. Musher DM, Fainstein V, Young EJ. Treatment of cel-
3. Peterson RA, Polgreen LA, Cavanaugh JE, Polgreen PM. Increas-
lulitis with ceforanide. Antimicrob Agents Chemother.
ing incidence, cost, and seasonality in patients hospitalized for
1980;17(2):254–7.
cellulitis. Open Forum Infect Dis. 2017;4(1):ofx008.
Cellulitis: A Review of Current Practice Guidelines and Differentiation from Pseudocellulitis

25. Bruun T, Oppegaard O, Kittang BR, Mylvaganam H, Lange- the emergency department: a double cohort study. Diagn Micro-
land N, Skrede S. Etiology of cellulitis and clinical prediction of biol Infect Dis. 2017;87(4):371–5.
streptococcal disease: a prospective study. Open Forum Infect 43. Ko LN, Garza-Mayers AC, St John J, Strazzula L, Vedak P,
Dis. 2016;3(1):ofv181. Shah R, et al. Effect of dermatology consultation on outcomes
26. Collazos J, de la Fuente B, de la Fuente J, García A, Gómez H, for patients with presumed cellulitis: a randomized clinical trial.
Menéndez C, et al. Factors associated with sepsis development JAMA Dermatol. 2018;154(5):529–36.
in 606 Spanish adult patients with cellulitis. BMC Infect Dis. 44. Ly N, Goldenberg M, Korman AM, Spaccarelli N, Wang H,
2020;20(1):211. Chung C, et al. A retrospective study of cellulitis outcomes in
27. Picard D, Klein A, Grigioni S, Joly P. Risk factors for abscess Ohio hospitals with or without access to dermatology residency
formation in patients with superficial cellulitis (erysipelas) of the programs. Int J Dermatol. 2021. https://​doi.​org/​10.​1111/​ijd.​
leg. Br J Dermatol. 2013;168(4):859–63. 15611.
28. Grada AA, Phillips TJ. Lymphedema: diagnostic workup and 45. Korman AM, Kroshinsky D, Raff AB, Mostaghimi A, Micheletti
management. J Am Acad Dermatol. 2017;77(6):995–1006. RG, Rosenbach M, et al. A survey-based study of diagnostic
29. Stevens DL, Bisno AL, Chambers HF, Dellinger EP, Goldstein and treatment concordance in standardized cases of cellulitis
EJ, Gorbach SL, et al. Practice guidelines for the diagnosis and and pseudocellulitis via teledermatology. J Am Acad Dermatol.
management of skin and soft tissue infections: 2014 update by 2020;82(5):1221–3.
the Infectious Diseases Society of America. Clin Infect Dis. 46. Hayeri MR, Ziai P, Shehata ML, Teytelboym OM, Huang BK.
2014;59(2):e10-52. Soft-tissue infections and their imaging mimics: from cellulitis
30. Carey CF, Dall L. Diagnosis of cellulitis in the immunocompro- to necrotizing fasciitis. Radiographics. 2016;36(6):1888–910.
mised host. Can J Infect Dis. 1990;1(4):133–5. 47. Tayal VS, Hasan N, Norton HJ, Tomaszewski CA. The effect
31. Ko LN, Garza-Mayers AC, St John J, Strazzula L, Vedak P, Dobry of soft-tissue ultrasound on the management of cellulitis in the
AS, et al. Clinical usefulness of imaging and blood cultures in emergency department. Acad Emerg Med. 2006;13(4):384–8.
cellulitis evaluation. JAMA Intern Med. 2018;178(7):994–6. 48. Gunderson CG, Martinello RA. A systematic review of bactere-
32. Perl B, Gottehrer NP, Raveh D, Schlesinger Y, Rudensky B, Yin- mias in cellulitis and erysipelas. J Infect. 2012;64(2):148–55.
non AM. Cost-effectiveness of blood cultures for adult patients 49. Moran GJ, Krishnadasan A, Gorwitz RJ, Fosheim GE, McDougal
with cellulitis. Clin Infect Dis. 1999;29(6):1483–8. LK, Carey RB, et al. Methicillin-resistant S. aureus infections
33. Lee CY, Kunin CM, Chang C, Lee SS, Chen YS, Tsai HC. Devel- among patients in the emergency department. N Engl J Med.
opment of a prediction model for bacteremia in hospitalized 2006;355(7):666–74.
adults with cellulitis to aid in the efficient use of blood cultures: 50. Pallin DJ, Binder WD, Allen MB, Lederman M, Parmar S, Filbin
a retrospective cohort study. BMC Infect Dis. 2016;16(1):581. MR, et al. Clinical trial: comparative effectiveness of cephalexin
34. Paolo WF, Poreda AR, Grant W, Scordino D, Wojcik S. Blood plus trimethoprim-sulfamethoxazole versus cephalexin alone for
culture results do not affect treatment in complicated cellulitis. J treatment of uncomplicated cellulitis: a randomized controlled
Emerg Med. 2013;45(2):163–7. trial. Clin Infect Dis. 2013;56(12):1754–62.
35. Fabre V, Sharara SL, Salinas AB, Carroll KC, Desai S, Cosgrove 51. Griego RD, Rosen T, Orengo IF, Wolf JE. Dog, cat, and human
SE. Does this patient need blood cultures? A scoping review of bites: a review. J Am Acad Dermatol. 1995;33(6):1019–29.
indications for blood cultures in adult nonneutropenic inpatients. 52. Finkelstein R, Oren I. Soft tissue infections caused by marine
Clin Infect Dis. 2020;71(5):1339–47. bacterial pathogens: epidemiology, diagnosis, and management.
36. van Daalen FV, Kallen MC, van den Bosch CMA, Hulscher Curr Infect Dis Rep. 2011;13(5):470–7.
MEJL, Geerlings SE, Prins JM. Clinical condition and comor- 53. Sayabovorn N, Chongtrakool P, Chayakulkeeree M. Cryptococ-
bidity as determinants for blood culture positivity in patients with cal fungemia and Mycobacterium haemophilum cellulitis in a
skin and soft-tissue infections. Eur J Clin Microbiol Infect Dis. patient receiving ruxolitinib: a case report and literature review.
2017;36(10):1853–8. BMC Infect Dis. 2021;21(1):27.
37. Raff AB, Weng QY, Cohen JM, Gunasekera N, Okhovat JP, 54. Franco-Paredes C, Marcos LA, Henao-Martínez AF, Rod-
Vedak P, et al. A predictive model for diagnosis of lower extrem- ríguez-Morales AJ, Villamil-Gómez WE, Gotuzzo E, et al.
ity cellulitis: a cross-sectional study. J Am Acad Dermatol. Cutaneous mycobacterial infections. Clin Microbiol Rev.
2017;76(4):618-25.e2. 2018;32(1):e00069-e118.
38. Krasagakis K, Valachis A, Maniatakis P, Krüger-Krasagakis 55. Maraki S, Christidou A, Anastasaki M, Scoulica E. Non-O1, non-
S, Samonis G, Tosca AD. Analysis of epidemiology, clini- O139 Vibrio cholerae bacteremic skin and soft tissue infections.
cal features and management of erysipelas. Int J Dermatol. Infect Dis (Lond). 2016;48(3):171–6.
2010;49(9):1012–7. 56. Flahou B, Haesebrouck F, Smet A, Yonezawa H, Osaki T,
39. Jeng A, Beheshti M, Li J, Nathan R. The role of beta-hemo- Kamiya S. Gastric and enterohepatic non-Helicobacter pylori
lytic streptococci in causing diffuse, nonculturable cel- Helicobacters. Helicobacter. 2013;18(Suppl 1):66–72.
lulitis: a prospective investigation. Medicine (Baltimore). 57. Shimizu S, Shimizu H. Cutaneous manifestations of Helicobacter
2010;89(4):217–26. cinaedi: a review. Br J Dermatol. 2016;175(1):62–8.
40. Wells A, Gupta P, Tian F, Adkins E, Kaffenberger B. The effect 58. Dignani MC, Anaissie E. Human fusariosis. Clin Microbiol
of implementing teledermatology in patients presenting with Infect. 2004;10(Suppl 1):67–75.
cellulitis versus pseudocellulitis in an academic emergency 59. Bloch KC, Nadarajah R, Jacobs R. Chryseobacterium meningo-
department setting: a pilot study. J Clin Aesthet Dermatol. septicum: an emerging pathogen among immunocompromised
2020;13(4):43–4. adults. Report of 6 cases and literature review. Medicine (Balti-
41. Gupta P, Tolliver S, Zhang M, Schumacher E, Kaffenberger BH. more). 1997;76(1):30–41.
Impact of dermatology and teledermatology consultations for 60. Thompson DK, Sharkady SM. Expanding spectrum of oppor-
patients admitted with cellulitis: a pilot study. J Am Acad Der- tunistic Cedecea infections: current clinical status and multidrug
matol. 2020;82(2):513–5. resistance. Int J Infect Dis. 2020;100:461–9.
42. Jain SR, Hosseini-Moghaddam SM, Dwek P, Gupta K, Elsayed S, 61. Ekpanyapong S, Reddy KR. Infections in cirrhosis. Curr Treat
Thompson GW, et al. Infectious diseases specialist management Options Gastroenterol. 2019;17(2):254–70.
improves outcomes for outpatients diagnosed with cellulitis in
 M. A. Boettler et al.

62. Falagas ME, Kastoris AC, Vouloumanou EK, Dimopoulos G. 83. McHenry CR, Piotrowski JJ, Petrinic D, Malangoni MA. Deter-
Community-acquired Stenotrophomonas maltophilia infec- minants of mortality for necrotizing soft-tissue infections. Ann
tions: a systematic review. Eur J Clin Microbiol Infect Dis. Surg. 1995;221(5):558–63 (Discussion 63-5).
2009;28(7):719–30. 84. Wall DB, Klein SR, Black S, de Virgilio C. A simple model to
63. Khan T, Martin DH. Streptococcus pneumoniae soft tissue help distinguish necrotizing fasciitis from nonnecrotizing soft
infections in human immunodeficiency virus. Am J Med Sci. tissue infection. J Am Coll Surg. 2000;191(3):227–31.
2011;342(3):235–8. 85. Bechar J, Sepehripour S, Hardwicke J, Filobbos G. Laboratory
64. Paterson DL, Gruttadauria S, Lauro A, Scott V, Marino IR. Spon- risk indicator for necrotising fasciitis (LRINEC) score for the
taneous gram-negative cellulitis in a liver transplant recipient. assessment of early necrotising fasciitis: a systematic review
Infection. 2001;29(6):345–7. of the literature. Ann R Coll Surg Engl. 2017;99(5):341–6.
65. Yoon TY, Jung SK, Chang SH. Cellulitis due to Escherichia 86. Stevens DL, Bryant AE. Necrotizing soft-tissue infections. N
coli in three immunocompromised subjects. Br J Dermatol. Engl J Med. 2017;377(23):2253–65.
1998;139(5):885–8. 87. Cribb BI, Wang MTM, Kulasegaran S, Gamble GD, Mac-
66. Ichiyama S, Hirai S, Minami T, Nishiyama Y, Shimizu S, Cormick AD. The SIARI Score: a novel decision support tool
Shimokata K, et al. Campylobacter fetus subspecies fetus cel- outperforms LRINEC Score in necrotizing fasciitis. World J
lulitis associated with bacteremia in debilitated hosts. Clin Infect Surg. 2019;43(10):2393–400.
Dis. 1998;27(2):252–5. 88. Borschitz T, Schlicht S, Siegel E, Hanke E, von Stebut E.
67. Chen YS, Liu YC, Yen MY, Wang JH, Wann SR, Cheng DL. Improvement of a clinical score for necrotizing fasciitis: ‘Pain
Skin and soft-tissue manifestations of Shewanella putrefaciens Out of Proportion’ and high CRP levels aid the diagnosis.
infection. Clin Infect Dis. 1997;25(2):225–9. PLoS ONE. 2015;10(7): e0132775.
68. Crowe HM, Levitz RE. Invasive Haemophilus influenzae disease 89. Ramchandani M, Kong M, Tronstein E, Selke S, Mikhaylova
in adults. Arch Intern Med. 1987;147(2):241–4. A, Magaret A, et al. Herpes simplex virus type 1 shedding in
69. Beckman EN, Leonard GL, Castillo LE, Genre CF, Pankey GA. tears and nasal and oral mucosa of healthy adults. Sex Transm
Histopathology of marine vibrio wound infections. Am J Clin Dis. 2016;43(12):756–60.
Pathol. 1981;76(6):765–72. 90. Schmutzhard J, Riedel HM, Wirgart BZ, Grillner L. Detec-
70. Hurwitz RM, Leaming RD, Horine RK. Necrotic celluli- tion of herpes simplex virus type 1, herpes simplex virus type
tis. A localized form of septic vasculitis. Arch Dermatol. 2 and varicella-zoster virus in skin lesions. Comparison of
1984;120(1):87–92. real-time PCR, nested PCR and virus isolation. J Clin Virol.
71. Musher DM. Cutaneous and soft-tissue manifestations of 2004;29(2):120–6.
sepsis due to Gram-negative enteric bacilli. Rev Infect Dis. 91. von den Driesch P. Sweet’s syndrome (acute febrile neutro-
1980;2(6):854–66. philic dermatosis). J Am Acad Dermatol. 1994;31(4):535–56
72. Kolivras A, Provost P, Thompson CT. Erysipelas-like erythema (Quiz 57-60).
of familial Mediterranean fever syndrome: a case report with 92. StatPearls. Treasure Island, FL; StatPearls; 2021.
emphasis on histopathologic diagnostic clues. J Cutan Pathol. 93. Kruger PC, Eikelboom JW, Douketis JD, Hankey GJ. Deep vein
2013;40(6):585–90. thrombosis: update on diagnosis and management. Med J Aust.
73. Levell NJ, Wingfield CG, Garioch JJ. Severe lower limb cel- 2019;210(11):516–24.
lulitis is best diagnosed by dermatologists and managed with 94. Cox SE, Cruz PD. A spectrum of inflammatory metastasis to
shared care between primary and secondary care. Br J Derma- skin via lymphatics: three cases of carcinoma erysipeloides. J
tol. 2011;164(6):1326–8. Am Acad Dermatol. 1994;30(2 Pt 2):304–7.
74. Strazzula L, Cotliar J, Fox LP, Hughey L, Shinkai K, Gee SN, 95. Korkij W, Soltani K. Fixed drug eruption. A brief review. Arch
et al. Inpatient dermatology consultation aids diagnosis of cel- Dermatol. 1984;120(4):520–4.
lulitis among hospitalized patients: a multi-institutional analy- 96. Obaitan I, Dwyer R, Lipworth AD, Kupper TS, Camargo
sis. J Am Acad Dermatol. 2015;73(1):70–5. CA, Hooper DC, et al. Failure of antibiotics in cellulitis tri-
75. Sundaresan S, Migden MR, Silapunt S. Stasis dermatitis: als: a systematic review and meta-analysis. Am J Emerg Med.
pathophysiology, evaluation, and management. Am J Clin 2016;34(8):1645–52.
Dermatol. 2017;18(3):383–90. 97. Miller LG, Daum RS, Creech CB, Young D, Downing MD,
76. Thornton NJ, Gibson BR, Ferry AM. Contact dermatitis and Eells SJ, et al. Clindamycin versus trimethoprim-sulfameth-
medical adhesives: a review. Cureus. 2021;13(3): e14090. oxazole for uncomplicated skin infections. N Engl J Med.
77. Johansen JD, Werfel T. Highlights in allergic contact der- 2015;372(12):1093–103.
matitis 2018/2019. Curr Opin Allergy Clin Immunol. 98. Williams OM, Brindle R. Antibiotic route and duration of
2019;19(4):334–40. therapy for cellulitis: data extracted from a multi-center clini-
78. Moffatt CJ, Franks PJ, Doherty DC, Williams AF, Badger C, cal trial. Int J Antimicrob Agents. 2020;56(3): 106076.
Jeffs E, et al. Lymphoedema: an underestimated health prob- 99. Madaras-Kelly KJ, Remington RE, Oliphant CM, Sloan KL,
lem. QJM. 2003;96(10):731–8. Bearden DT. Efficacy of oral beta-lactam versus non-beta-
79. Marchionne EM, McCalmont TH, Pincus LB, LeBoit PE, Fox lactam treatment of uncomplicated cellulitis. Am J Med.
LP. Acute inflammatory edema: a mimicker of cellulitis in 2008;121(5):419–25.
critically ill patients. J Am Acad Dermatol. 2019;81(4):931–6. 100. Khawcharoenporn T, Tice A. Empiric outpatient therapy with
80. Nadelman RB, Wormser GP. Erythema migrans and early trimethoprim-sulfamethoxazole, cephalexin, or clindamycin for
Lyme disease. Am J Med. 1995;98(4A):15S-23S (Discussion cellulitis. Am J Med. 2010;123(10):942–50.
23S-24S). 101. Cenizal MJ, Skiest D, Luber S, Bedimo R, Davis P, Fox P,
81. Gunderson CG, Chang JJ. Risk of deep vein thrombosis in et al. Prospective randomized trial of empiric therapy with
patients with cellulitis and erysipelas: a systematic review and trimethoprim-sulfamethoxazole or doxycycline for outpatient
meta-analysis. Thromb Res. 2013;132(3):336–40. skin and soft tissue infections in an area of high prevalence
82. Stevens DL, Bryant AE, Goldstein EJ. Necrotizing soft tissue of methicillin-resistant Staphylococcus aureus. Antimicrob
infections. Infect Dis Clin N Am. 2021;35(1):135–55. Agents Chemother. 2007;51(7):2628–30.
Cellulitis: A Review of Current Practice Guidelines and Differentiation from Pseudocellulitis

102. Livermore DM, Pearson A. Antibiotic resistance: location, 107. Soo JK, Bicanic TA, Heenan S, Mortimer PS. Lymphatic
location, location. Clin Microbiol Infect. 2007;13(Suppl abnormalities demonstrated by lymphoscintigraphy after lower
2):7–16. limb cellulitis. Br J Dermatol. 2008;158(6):1350–3.
103. Cross ELA, Jordan H, Godfrey R, Onakpoya IJ, Shears A, 108. Dalal A, Eskin-Schwartz M, Mimouni D, Ray S, Days W,
Fidler K, et al. Route and duration of antibiotic therapy in Hodak E, et al. Interventions for the prevention of recur-
acute cellulitis: a systematic review and meta-analysis of rent erysipelas and cellulitis. Cochrane Database Syst Rev.
the effectiveness and harms of antibiotic treatment. J Infect. 2017;6(6): CD009758.
2020;81(4):521–31. 109. Karaca-Mandic P, Hirsch AT, Rockson SG, Ridner SH. The
104. Jenkins TC, Knepper BC, Sabel AL, Sarcone EE, Long JA, cutaneous, net clinical, and health economic benefits of
Haukoos JS, et al. Decreased antibiotic utilization after imple- advanced pneumatic compression devices in patients with
mentation of a guideline for inpatient cellulitis and cutaneous lymphedema. JAMA Dermatol. 2015;151(11):1187–93.
abscess. Arch Intern Med. 2011;171(12):1072–9. 110. Mihara M, Hara H, Furniss D, Narushima M, Iida T, Kikuchi K,
105. Talan DA, Mower WR, Lovecchio FA, Rothman RE, Steele et al. Lymphaticovenular anastomosis to prevent cellulitis asso-
MT, Keyloun K, et al. Pathway with single-dose long-acting ciated with lymphoedema. Br J Surg. 2014;101(11):1391–6.
intravenous antibiotic reduces emergency department hospi- 111. Webb E, Neeman T, Bowden FJ, Gaida J, Mumford V, Bissett B.
talizations of patients with skin infections. Acad Emerg Med. Compression therapy to prevent recurrent cellulitis of the leg. N
2021;28(10):1108–17. Engl J Med. 2020;383(7):630–9.
106. Karppelin M, Siljander T, Vuopio-Varkila J, Kere J, Huhtala 112. Thomas KS, Crook AM, Nunn AJ, Foster KA, Mason JM, Chal-
H, Vuento R, et al. Factors predisposing to acute and recurrent mers JR, et al. Penicillin to prevent recurrent leg cellulitis. N
bacterial non-necrotizing cellulitis in hospitalized patients: Engl J Med. 2013;368(18):1695–703.
a prospective case-control study. Clin Microbiol Infect.
2010;16(6):729–34.