Cellulitis Mimics – ED Considerations

Author: Garrett K. Blumberg, MD (EM Resident Physician, Parkland Memorial Hospital / UT Southwestern) // Editors: Alex Koyfman, MD (@EMHighAK, EM Attending Physician, UT Southwestern Medical Center / Parkland Memorial Hospital) and Brit Long, MD (@long_brit)

In your bustling ED, EMS brings in a disheveled, homeless 34-year-old female complaining of a rash on her left elbow. Upon walking into the room, you see an uncomfortable appearing woman with a red, swollen, left elbow with erythema and fluctuance spanning her lateral forearm and upper arm. She holds her arm in flexion. The patient reports approximately one week ago the erythema was much smaller and located only in her anterior cubital fossa. She noticed the erythema became fluctuant, and her friend attempted to drain the wound with a needle. Over the past 24 hours, however, she reports worsening severe pain, and she is very hesitant to move her left elbow. The nurse provides you with vital signs showing she is tachycardic, normotensive, and febrile. While in the patient’s room you begin to smell a foul odor and note that the patient’s wound has come to a head and is draining what appears to be dish-water fluid.


Cellulitis is a common pathology seen in the emergency department (ED) with 2.3 million emergency department (ED) visits annually in the US[1] and is defined as superficial spreading skin infection without an underlying collection of pus[2]. Patients who present with cellulitis often have erythema (rubor), edema (tumor), warmth (calor), and pain (dolor) that develops over days[2]. Systemic signs of infection may or may not be present. If so, systemic symptoms may present before cutaneous inflammation.

WBC, ESR, CRP are commonly elevated, but normal levels do not rule out the diagnosis. 35-50% of patients will have a leukocytosis, 60-92% will have an elevated ESR, and 75-95% will have an elevated CRP. These laboratory values may also be elevated in cellulitis mimics and thus are a nothing to hang your hat on. Blood cultures are often not helpful, except for certain patient populations including immunocompromised, neutropenic, and seriously ill or toxic patient.

Cellulitis located in the lower extremity is often caused by streptococcus which colonizes the interdigital toe spaces. Other common infectious causes include: staphylococcus aureus, anaerobes, and polymicrobial infection. When breaks in the skin occur or venous/lymphatic drainage is obstructed, bacteria are able to spread through the superficial tissues creating a localized area of infection. Predisposing risk factors include: blockage in drainage return (venous insufficiency, saphenous veinectomy in coronary artery bypass), breaks in the skin (trauma, ulceration, edema), inflammatory disease of the skin (allergic contact dermatitis, atopic dermatitis, venous eczema), older age, and diabetes.

Why do these risk factors predispose patients? Blockage of drainage of lymphatic fluid or venous return prevents the lymphatic system to clear microbes that have breached the skin. In the case of edema, it is believed that microtears serve as entry ports for bacteria. Obesity is thought to be a risk factor for cellulitis due to the compression of the lymphatic flow by excess adipose tissue. In the lower extremities, abnormal interdigital spaces is a predisposing factor. Interdigital intertrigo is associated with lower limb cellulitis with an odds ratio from 3.2-5.4 [3].

Of the 2.3 million cases, approximately 14-17% are admitted. In a cross-sectional study published in JAMA Dermatology evaluating patients who were initially diagnosed with a lower extremity cellulitis and admitted to the hospital, 30.5% were incorrectly diagnosed [1]. This high percentage is consistent with prior studies showing a high rate of misdiagnosis [4, 5]. Fortunately, common mimickers of cellulitis are relatively benign: venous stasis, lymphedema, gout, contact dermatitis. However this post will discuss those diagnoses emergency physicians must not miss.

These diagnoses include septic bursitis, septic arthritis, deep vein thrombosis/phlegmasia cerulea dolens, necrotizing fasciitis, flexor tenosynovitis, fight bite, orbital cellulitis, and toxic shock syndrome. Less emergent diagnoses include erysipelas, abscess, felon, paronychia, and gouty arthritis.

These diagnosis often mimic cellulitis. Nearly all have a similar constellation of symptoms: rubor, tumor, calor, and dolor due to the underlying pathophysiology. An infection or a nidus of inflammation stimulates an immune reaction, leading to vasodilation and IL-1 and TNF production, which recruit neutrophils to the site of inflammation. At first glance many of these diseases may appear the same and only begin to crystalize upon a thorough history and physical exam. Many of these mimics can easily be missed if not appreciated or sought out.


Septic Bursitis

Septic bursitis is inflammation and infection of a bursa commonly due to direct inoculation from trauma, spread from nearby soft tissue (cellulitis), or hematogenous spread. Patients present with swelling, warmth, erythema, and tenderness over a bursa. The most common locations of septic bursitis are at the olecranon and prepatellar bursae.

Cellulitis may precede or accompany septic bursitis. Only 40-44% patients will develop a fever [6]. Thus distinguishing cellulitis and septic bursitis is difficult. However, if palpation of the site of infection reveals fluctuance, it is most likely septic bursitis or an abscess. Knowledge of the locations of bursae and ultrasound imaging are helpful in differentiating an abscess from a bursae.

It is often difficult to distinguish septic bursitis from aseptic bursitis. Some findings can help differentiate the two such as warmth over the bursal skin, presence of erythema over the bursa, aspirate WBC, and glucose of aspirate/serum ratio.

Signs and Measurements Septic Bursitis Aseptic Bursitis
Fever >38/100.4 20-58% 0%
Warmth 100% 50%
Pre-bursal cellulitis 63-100% 25-50%
Pain Severe. Present in >90% Mild. Present in 45%
Erythrocyte Sedimentation Rate >20 in 80% of cases Elevated
C-reactive protein Elevated in 100% cases Elevated
Aspirate WBC >10,000

(>1,000 in prepatellar)

Cell predominance Neutrophils Monocytes
Serum/aspirate glucose Aspirate glucose is <50% of the serum glucose Glucose 70-80% in 100% cases

Adapted from [7].

Septic Prepatellar Bursitis

Septic Prepatellar Bursitis usually is the result of trauma to the knee resulting in direct inoculation. The classic presentation of a septic prepatellar bursitis is a swollen, painful, erythematous, and warm prepatellar bursae. Differentiating aseptic from septic bursitis is difficult but can be assisted by aspirating bursal fluid. Send the fluid for cells, gram stain, cultures, and crystals. The aspirate is considered infected in prepatellar bursitis if the WBC is >1,000/uL[8]. The treatment of prepatellar septic bursitis is controversial, and no good guidelines exist at this time. Treatment can range anywhere from oral antibiotics alone to surgery [8]. Most patients will respond to nonsurgical treatment and will only require antibiotics such as dicloxacillin or clindamycin. Aspiration can be performed for diagnostic as well as therapeutic measures. It is important to note that prepatellar bursitis appears to have a high risk of outpatient failure and may benefit from parenteral antibiotics as this resolves the infection faster. However, consideration both the patient’s reliability, history, and physical exam will help guide the decision to discharge or admit the patient[8].

Septic Olecranon Bursitis

Approximately 20% of all bursitis cases of the olecranon are septic in nature[8]. Most often septic olecranon bursa have overlying redness or cellulitis features and are painful. However these factors are often unreliable. Skin warmth and aspiration of bursal fluid are needed to diagnose septic bursitis in the olecranon. Gram stain shows no growth 50% of the time. Cell count for aseptic bursitis will have WBC <1,000/mm3 whereas septic bursitis will have WBC >10,000/mm3. If the WBC count is between these two numbers, the clinical history and exam and cell differential should be used to differentiate the two. PMNs favor the diagnosis of septic bursitis, whereas monocytes favor aseptic bursitis. Serum to aspirate glucose levels can also help distinguish septic versus aseptic[8].

Treatment of olecranon bursitis should be based on history and physical exam. If aseptic bursitis is suspected, patients can often be treated with NSAIDs, ice, compression, and rest, without drainage. If there is concern for septic bursitis such as warmth, overlaying cellulitis, and moderate pain, the bursa should be drained, and the patient must be given systemic antibiotics[8]. Patients can be treated with oral antibiotics; however, studies have shown that parenteral antibiotics require fewer days to sterilize the bursal fluid[8]. Antibiotics must cover staph aureus, which is the most common etiology. If concerned for MSSA, consider giving nafcillin or oxacillin or cefazolin. If concern for MRSA is present, vancomycin should be administered.

Septic joint

A septic joint is infection of the joint space, usually due to bacteremia, trauma with direct inoculation, or very uncommonly from spread of osteomyelitis. Time is cartilage. If this diagnosis is missed, the infection in the joint space will eat away at the cartilage leading to long term sequela. Additionally, in hospital mortality rates of patients with septic arthritis is as high as 15% [9]. Infections of the synovial fluid commonly result from staphylococcus aureus and streptococcus species. If traumatic inoculation occurs from an external wound, gram negative bacilli may be present. Approximately 50% of cases of septic joint involve the knee [9]. There is often a history of osteoarthritis, rheumatoid arthritis, pseudogout, gout, or Charcot arthropathy that predisposes patients to infection of the affected joint. Physical exam usually reveals one joint that is red, swollen, and exquisitely painful to passive and active movement.

Arthrocentesis is required to evaluate for septic joint. The risk of iatrogenic infection from arthrocentesis is 0.1% in immunocompetent patients and 0.5% in immunosuppressed patients. For patients on warfarin, the estimated risk of clinically significant hemorrhage is below 10%. The fluid should be sent for cell count, gram stain, culture, and crystals. Synovial lactate should be obtained if available, as this demonstrates the greatest test characteristics for diagnosis.Screen Shot 2017-03-28 at 7.18.59 PM

If clinical gestalt and arthrocentesis fluid point towards septic arthritis, parenteral vancomycin and a third-generation cephalosporin should be administered. It is important to note that not all septic joints will have bacteria on gram stain and thus gram stain cannot not rule out the diagnosis of septic joint. Additionally, not all septic joints with have >50,000 WBC/mm3. In one retrospective study evaluating arthrocentesis cell count, 19 patients of 49 (39%) with culture-positive synovial fluid had a synovial fluid count <50,000WBC/mm3. Thus, the sensitivity of the 50,000 WBC/mm3 cut off was 61% (95% CI, 48-75%)[11].  Additional studies have found that most MRSA infections result in synovial fluid with WBC count <20,000[12]. Given these new confounding studies, the decision to admit and begin empiric antibiotics while awaiting culture results weighs in the balance of the clinical picture. Take into consideration the entire picture: clinical presentation, age, risk factors (history of arthritis, prior septic arthritis, history of joint surgery, ESRD, diabetes, cancer, immunosuppression, IV drug use[11]), presence of fever, synovial fluid white blood cell count, crystals, and gram stain[12]. Orthopedic surgery should be consulted for septic arthritis [13].  For more on septic arthritis, please see http://www.emdocs.net/septic-joint-reminders-updates-and-pitfalls/.


Deep vein thrombosis is defined as a blood clot in the deep venous system. The deep venous system includes the anterior tibial, posterior tibial, peroneal, popliteal, and femoral veins. 25% of all DVT’s will present with tenderness and erythema in the swollen extremity which are similar findings seen in cellulitis. The site of this erythema and warmth may be distal to the location of a more proximal DVT. Calf pain elicited by passive dorsiflexion of the foot, otherwise known as Homan’s sign, is neither sensitive nor specific for DVT and should not be used as a predictive indicator. A 2005 meta-analysis found there are limited findings on physical exam that strongly point towards or away from the diagnosis of DVT[14]. However, a difference in the calf diameter was of potential value for ruling in DVT with a likelihood ratio of 1.8 (CI 1.5-2.2). The converse also holds true, as absence of calf swelling and no difference between calf sizes provide evidence against DVT[14]. Diagnostic test include ultrasound, which is 95% sensitive and specific for DVT. In this same low risk group, a negative D-Dimer is sensitive enough to rule out DVT. However if the patient is moderate to high risk per Wells Criteria (or strong clinical suspicion) and the ultrasound is negative, a repeat should be performed 5-7 days. If positive, treatment should begin immediately with anticoagulation or inferior vena cava filter if anticoagulation is contraindicated or if the patient develops a DVT while on therapeutic warfarin.

Phlegmasia cerulea dolens

Phlegmasia cerulea dolens (PCD) literally stands for “edema, blue, painful” and has a classic triad of leg swelling, acute ischemic pain, and discoloration [15]. In this disease process, a large obstructing thrombosis within the deep venous system results in a low venous flow state of the lower extremity. Clot has been reported to be located anywhere from the femoral vein tracking into the IVC. Analogous to the low flow state seen in some forms of priapism, ischemia and tissue necrosis are the ultimate result with grave consequences. Mortality rates for PCD range from 25-40%, and amputation rates are approximately 20-50% [15]. Death occurs from massive fluid sequestration leading to circulatory collapse and shock.

Similar to DVT, there is a high association of PCD with hypercoagulable states such as malignancy and oral contraception. Malignancy is the most common risk factor. As would be expected, pulmonary embolism is commonly associated with PCD. The most reliable and fastest technique to diagnosis PCD is by duplex ultrasound imaging. However, ultrasound comes with some limitations such as difficulty visualizing clot in the pelvis and abdomen. Additionally, ultrasound only allows for diagnosis. Imaging studies such as CT or MR venography allows for both diagnosis and evaluation of clot extension, as well as treatment by catheter-directed thrombolysis.

Initial general treatment of PCD includes absolute bed rest, leg elevation, fluids, and IV heparin. The initial dose is 10,000 to 15,000 units followed by continual infusion [15]. Follow the aPTT for a goal of 2 times normal lab value. Monitor the platelet count to ensure the patient does not develop heparin induced thrombocytopenia. If the patient does not improve after 4-6 with anticoagulation, invasive treatment is required. This is where treatment options become controversial without solid guidelines. However, one of the more common treatment options that has risen in favor over the past decade is catheter-directed thrombolysis [15, 16]. This method of treatment can be used 4-6 hours after a patient has failed to respond to heparin anticoagulation, or alternatively as the first line therapy if a patient is late in the PCD disease process[16]. This technique is not perfect and has been associated with subsequent PE due to fragments of thrombus embolizing. Consultation with interventional radiology or vascular surgery is required to assist in management for this rare but fatal disease.

Flexor tenosynovitis

Flexor tenosynovitis (FTS) is a surgical emergency. It is an infection of the tendon and synovial sheath of the flexor surface of the hand. Misdiagnosis or failure to diagnose this disease may results in adhesions, tendon vascular compromise, and necrosis with loss of finger function and potential loss of hand function as the infection spreads. Thus it is critical to suspect flexor tenosynovitis when cellulitis presents on the hand.

Infection of the tendon sheath most commonly occurs following penetrating trauma to the finger, followed next by blunt trauma. Frequently, however, patients will not report prior trauma. Risk factors significantly associated with increased risk of amputation of the affected digit include age >45 years, diabetes mellitus, peripheral vascular disease, renal failure, subcutaneous purulence, ischemic changes, and polymicrobial infection[17].

Often Kanavel’s four cardinal signs are not all present, and unfortunately the sensitivity and likelihood ratios of these findings have not been thoroughly studied or validated. In 2007, Phang et al. performed a retrospective study of 75 confirmed suppurative flexor tenosynovitis by operative report over 5 years. They found that fusiform swelling was the most common finding occurring in 97% (73/75 patients) of cases. Next most common was pain with passive extension 72% (54/75 patients), followed by digit held in flexion 69% (54/75 patients), and lastly tenderness to the flexor tendon sheath 64% (48/75 patients) [17]. It has been suggested that tenderness to the flexor tendon sheath is a late sign of disease and thus should not be used to rule out disease [18]. Overall, Kanavel’s signs are not to be trusted without critical appraisal of the patient’s history and risk factors. These signs may assist with examination, but it is important to acknowledge these signs have not been validated.

Staphylococcus is the most common isolated organism. Other commonly found bacteria include streptococcal species, gram negatives in the setting of trauma, and anaerobes. Initial antibiotic treatment consists of vancomycin with clindamycin or ceftriaxone. FTS can also be caused by disseminated Neisseria gonorrhoeae in at risk populations. Treatment includes immobilization, elevation, and consultation with hand surgeon. If the infection is caught early, you may have saved your patient from surgery as some early cases are managed nonoperatively with IV antibiotics. 

Fight bite

Close fist injury results from injury to the hand after fight in which the patient has punched his or her opponent in the mouth. The injury usually occurs over the 3rd, 4th, or 5th metacarpophalangeal Screen Shot 2017-03-28 at 7.19.24 PMjoint. Due to the great force of the punch and the sharp edges of the incisors, patients are at risk of inoculating the skin, soft tissue, tendon sheath, joint space, bone, or deep spaces of the hand to bacteria. Infection is usually polymicrobial: streptococcal species, staphylococcus, Eikenella corrodens, fusobacterium, peptostreptococcus, and candida. Do you suspect deep palmar space, tendon, joint involvement? If so, place the patient on broad spectrum antibiotics and consult hand surgeon for open debridement and irrigation. Radiograph should be obtained to rule out fracture and foreign body such as tooth fragments.

Necrotizing soft tissue infection

Necrotizing soft tissues infections are limb and life threatening, with a mortality rate of 70-100% if left untreated or missed. Infection can involve any layer: skin, subcutaneous tissue, superficial or deep fascia, and muscle. There is a great deal of confusion in nomenclature. Regardless of the different names such as, “necrotizing fasciitis, gas gangrene, phagedena, or non-clostridial crepitant cellulitis” the pathogenesis and treatment are the same.

If cellulitis is suspected, necrotizing fasciitis should always be considered. Necrotizing soft tissue infections are usually distinguishable from cellulitis by clinical examination. Findings consistent with necrotizing soft tissue infection include pain out of proportion, edema or tenderness extending beyond the erythematous boarder of infection, vesicles or bullae (sometimes hemorrhagic, which can also be seen in cellulitis), bleeding into the skin (petechiae or ecchymosis) with a central area of grey/white concerning for severe ischemia, crepitus due to gas formation from anaerobic bacteria, cutaneous anesthesia (la belle indifference later in disease due to destruction of cutaneous nerve fibers), fluctuance, woody induration, rapid expansion of margins of inflammation despite antibiotics, drainage of necrotic area with murky, and thin purulent liquid that resembles dishwater and may be foul-smelling due to the presence of anaerobic bacteria. The table below demonstrates likelihood ratios for these findings.

Sign/Symptom Likelihood Ratio
Pain out of proportion 4.5
Redness extending beyond the margins 3.1
Fluctuance 5
Diarrhea 6
Hypotension 8
Recent surgery 7
Bullae 8
Necrotic skin 30
Altered mental status 3.3

Absence of all 9 likely rules out necrotizing fasciitis. If 3 or more are present, NF should be very high on the differential diagnosis [19].

The LRINEC score takes into consideration the patient’s WBC, hemoglobin, sodium, creatinine, glucose, and CRP, and relies solely on laboratory data. Initially, it was believed to have a sensitivity of 89.9% (CI 81.9-94.6%)[20]. However, multiple validating studies have shown much lower sensitivities. One validation study in the UK which found a sensitivity as low as 43.2% (CI 32.5-54.6%) [21]. Thus the LRINEC score is not a safe tool to rule out the diagnosis of necrotizing fasciitis.

Fortunately, ultrasound and CT can assist in the diagnosis of necrotizing fasciitis. On ultrasound, cellulitis often reveals cobble stoning. Whereas in necrotizing fasciitis, subcutaneous thickening, air, and fascial fluid (STAFF) can be seen[22]. This will appear as hypoechoic fluid tracking along the deep fascial layer with air present in the soft tissue on ultrasound [23]. Plain films are helpful to see subcutaneous tissue but have poor sensitivity. CT is the most sensitive modality to evaluate for soft-tissue gas. Additionally, CT can show the degree of underlying involvement of infection and help determine if osseous structures are involved[24]. The ability to rapidly obtain a CT scan versus the “in process” time of an emergent MRI makes the CT scanner more advantageous.

Cellulitis vs. Necrotizing Fasciitis

Screen Shot 2017-03-28 at 7.19.43 PM

If diagnosis remains uncertain, a “finger test” may be performed. This consists of a 2-cm incision in the affected area and inserting a sterile gloved finger to dissect the subcutaneous tissue from the deep fascia with minimal resistance. A positive test indicates a necrotizing infection.

Initial measures require resuscitation, surgical consultation, and broad-spectrum antimicrobials. Definitive treatment is emergent surgical consultation for irrigation and debridement. Antibiotics recommended by the IDSA include vancomycin, plus piperacillin-tazobactam or a carbapenem, plus clindamycin [26]. Alterative treatment can be vancomycin plus ceftriaxone and metronidazole [26]. These antimicrobials cover polymicrobial (mixed aerobic-anaerobic microbes) or monomicrobial such as group A streptococcus or CA-MRSA infections. For more on necrotizing fasciitis, please see: http://www.emdocs.net/necrotizing-fasciitis-pearls-pitfalls/.

Orbital cellulitis

Orbital cellulitis may mimic cellulitis with erythema and swelling of the periorbital soft tissue. However orbital cellulitis harbors infection deep to the orbital septum. History from patients will point towards pain with extraocular movement. Physical exam may demonstrate proptosis, chemosis, and poor visual acuity. The most common organisms are staphylococcus aureus, staphylococcus epidermidis, streptococcus species, and anaerobes. CT orbits should be obtained immediately, and ophthalmology should be consulted. Antibiotics should be initiated. Lastly, check intraocular pressure. Lateral canthotomy may be indicated if intraocular pressures are >40mmHg[13]{Tintinalli, 2016 #25}. Definite treatment occurs with antibiotics and operative drainage. It is very important to consider this diagnosis when evaluating cellulitis of the face. Complications from orbital cellulitis include vision loss, meningitis, brain abscess, cavernous sinus thrombosis, frontal bone osteomyelitis.

Toxic Shock Syndrome

Toxic shock syndrome is a deadly disease that often occurs in young, otherwise healthy patients. It is commonly misdiagnosed as viral illness or simple soft tissue infection. Two bacterial causes are staphylococcus and streptococcus. Both secrete toxin which acts as superantigen to stimulate a large quantity of T cell, resulting in a cytokine storm. Young people do not have the antibodies to target the toxin, and therefore TSS affects younger people who are otherwise healthy.

Staphylococcal TSS that is menstrually related (tampons) accounts for 50%, with non-menstrual related (soft tissue infections, sinusitis, sinus packing, pneumonia) accounting for the other 50%. Staphylococcal TSS clinical presentation includes symptoms that mimic gastroenteritis (vomiting, diarrhea) or viral/influenza illness (fever, myalgias, headache) in a young patient and is thus often misdiagnosed. Niduses of infection such as tampons, soft tissue infection, etc. are often overlooked because there is not significant inflammation/swelling at the site due to staphylococcal toxin which suppresses neutrophil activity. Diffuse erythema of the skin and strawberry tongue is often transient or very subtle. Eventually these patients who initially appear to have viral illness go on to develop renal failure, hypotension, and shock.

Streptococcal TSS is due to soft tissue infection (cellulitis, necrotizing fasciitis, surgical wound, myositis). Patients manifest with local signs of infection or a mixed picture of local and toxigenic infection. In the toxigenic variety, patients will appear present similarly to staphylococcal TSS with gastroenteritis or viral/influenza-like symptoms. In contrast, the local infection variety will present with invasive local streptococcal infection such as cellulitis, necrotizing fasciitis, and even pneumonia or pharyngitis. There may or may not be a prior history of gastrointestinal symptoms or viral prodrome. Often these patients will progress to septic shock that is refractory to standard treatments (IV antibiotics, 30mL/kg fluids resuscitation, norepinephrine drip, vasopressin drip, hydrocortisone). Failure of these treatments is an indication you may be dealing with toxic shock.

Diagnosis is difficult, and unfortunately there are no clinically focused guidelines [27]. Thus, the most important factor is to simply suspect TSS and discuss with infectious disease.


Anti-toxin producing antibiotics such as clindamycin [28] and linezolid [29] should be provided. Both antibiotics inhibit production of toxin production by halting bacterial ribosomal synthesis.

Source control

Involve the surgeon early even if the potential source (abscess, surgical wound, necrotizing fasciitis) appears benign. Remember these bacteria produce toxins which prevent neutrophil migration into the site of infection, thus tricking the physician into overlooking the nidus of infection.


In conjunction with infectious disease specialist and critical care, IVIG may be administered. Patients with TSS often do not have antibodies that target the toxin produced in this disease. IVIG theoretically targets the toxin for destruction by the host immune system. Studies have hinted at the potential benefit of IVIG, but no study has statistically proven a decrease in mortality [30]. Part of the difficulty of researching IVIG is the high dollar cost.

In conclusion, consider the likelihood of TSS and treat according to your suspicion with anti-toxin antibiotics such as clindamycin and linezolid, with source control. IVIG is more likely to be given in the ICU setting rather than in the emergency department.



An abscess is a soft tissue infection that has developed into a collection of pus within the dermis and deeper skin tissue. Given this wall of protection, abscesses require drainage. Patients will present with painful, tender, and fluctuant red nodule surmounted by a pustule with a rim of erythema and swelling. During incision and drainage, it is important to break up loculations by probing the wound. Drainage of an abscess should be purulent. If an abscess drains “dish-water” appearing fluid, be concerned for necrotizing fasciitis (bacteria that result in necrotizing fasciitis have enzymes which destroy neutrophils seen in pus, resulting in clear drainage).

If the diagnosis of cellulitis versus abscess cannot be made on physical exam, ultrasound is can confirm abscess. Ultrasound for cellulitis reveals cobble stoning, whereas an abscess will appear as a well-defined pocket of fluid.

Screen Shot 2017-03-28 at 7.20.20 PM

Per the 2014 IDSA guidelines for soft tissue infections, all abscesses should be drained. The decision to administer antibiotics against MRSA as an adjunct to I&D should be based on the presence of SIRS criteria (temperature >38C or <36C, respiratory rate >24, heart rate >90, white blood cell count >12,000 or <4,000) [26]. Antibiotics may assist preventing abscess recurrence. If a patient has markedly impaired host defenses, significant cellulitis, or meets SIRS criteria with hypotension, antibiotics such as vancomycin, linezolid, clindamycin, doxycycline, or trimethoprim-sulfamethoxazole must be administered to cover for MRSA [26].


Erysipelas is a form of cellulitis that presents with a well demarcated boarder. Unlike cellulitis, erysipelas affects the superficial dermis and superficial lymphatics, whereas cellulitis involves the dermis and subcutaneous fat. Given its proximity to the epidermis, the spread of erythema is more clearly visible and linear. Providers often have false notion that erysipelas refers only to facial cellulitis, however erysipelas can be located anywhere on the body (most commonly the face and legs).  This infection is usually due to staphylococcus aureus and streptococcus species. Diagnosis is largely clinical and can be identified by the defined boarder of erythema. Treatment includes anti-MRSA antibiotics if found on the face. If found on the lower extremity, treatment with a penicillin or cephalosporin is usually sufficient.


A felon is an infection of the pulp of the fingertip. The mechanism of bacteria infiltrating the pulp of the finger defines what bacterial should be suspected. Infections are often due to MRSA, streptococcus pyogenes, anaerobes, or polymicrobial bacteria. Like abscesses and paronychia, if there is no overlaying inflammation, then drainage alone may be sufficient. If the infection is associated with overlaying inflammation, however, antibiotics for MRSA and streptococcus should be administered. Alternatively, consider clindamycin or amoxicillin-clavulanate if anaerobic bacteria are suspected such as is the case with traumatic injury. It is important to drain the felon by the lateral approach to protect the neurovascular bundle [13].

Screen Shot 2017-03-28 at 7.20.41 PM


A paronychia is an infection of the lateral nail fold, often present with minor trauma to the nail. Pain, tenderness, and swelling of one of the lateral folds of the nail are often present. Often there is no pus drainage. If this is the case, treat with warm soaks, elevation, and Screen Shot 2017-03-28 at 7.20.56 PMantibiotics. Drainage of a small paronychia can be performed by lifting the nail fold with either a needle or a number 11 blade, allowing drainage. If pus is seen below the nail, then a portion of the nail may have to be removed, being careful to avoid injury to the nailbed [13].

Gouty arthritis

Gout is due to crystal formation in the joint space. Precipitation of crystals in the joint with neutrophil migration leads to inflammation due to vasodilation in the surrounding area. Redness, swelling, pain, and edema develop and can spread past the joint space into surrounding tissue.  Patients may also be febrile during acute attacks, further complicating the picture. In contrast to the more slowly developing cellulitis (development over days), the peak intensity of a gout attack is acute, ranging 6-12 hours. Pain can be isolated to the joint despite erythema that has spread past the joint. This is in stark contrast to cellulitis where pain is elicited over all areas of erythema. The erythema seen in gouty attacks is often described more commonly as “dusky” or more violaceous than that of cellulitis.

A history of gout may also help differentiate gout from cellulitis. However as mentioned before in the septic arthritis section, gouty joints are at high risk of septic arthritis and should be tapped if concerned for infection. Serum laboratory test such as serum uric acid is not helpful during an attack since 40-50% of patients have normal levels with an acute episode [2, 31]. Treatment includes NSAIDs and Colchicine.


  • Cellulitis and many of its mimics are characterized by rubor, tumor, calor, and dolor.
  • Location of infection and mechanism of injury often provides clues as to the diagnosis and what bacteria are implicated.
  • Cellulitis mimics include: septic bursitis, septic joint, deep vein thrombosis, phlegmasia cerulea dolens, necrotizing fasciitis, flexor tenosynovitis, fight bite, orbital cellulitis, and toxic shock syndrome. Less emergent cellulitis mimics include: erysipelas, abscess, felon, paronychia, and gouty arthritis.
  • Remember to involve your surgeon early if considering necrotizing fasciitis, septic arthritis, phlegmasia cerulea dolens, flexor tenosynovitis, fight bite, or orbital cellulitis.


References / Further Reading

  1. Weng, Q.Y., et al., Costs and Consequences Associated With Misdiagnosed Lower Extremity Cellulitis. JAMA Dermatol, 2016.
  2. Hirschmann, J.V. and G.J. Raugi, Lower limb cellulitis and its mimics: part I. Lower limb cellulitis. J Am Acad Dermatol, 2012. 67(2): p. 163 e1-12; quiz 175-6.
  3. David, C.V., et al., Diagnostic accuracy in patients admitted to hospitals with cellulitis. Dermatol Online J, 2011. 17(3): p. 1.
  4. Davis, J.P., et al., Tri-state toxic-shock syndrome study. II. Clinical and laboratory findings. J Infect Dis, 1982. 145(4): p. 441-8.
  5. Levell, N.J., C.G. Wingfield, and J.J. Garioch, Severe lower limb cellulitis is best diagnosed by dermatologists and managed with shared care between primary and secondary care. Br J Dermatol, 2011. 164(6): p. 1326-8.
  6. Ho, G., Jr., A.D. Tice, and S.R. Kaplan, Septic bursitis in the prepatellar and olecranon bursae: an analysis of 25 cases. Ann Intern Med, 1978. 89(1): p. 21-7.
  7. Wasserman, A.R., L.D. Melville, and R.H. Birkhahn, Septic Bursitis: A Case Report and Primer for the Emergency Clinician. The Journal of Emergency Medicine, 2009. 37(3): p. 269-272.
  8. Aaron, D.L., et al., Four common types of bursitis: diagnosis and management. J Am Acad Orthop Surg, 2011. 19(6): p. 359-67.
  9. Carpenter, C.R., et al., Evidence-based diagnostics: adult septic arthritis. Acad Emerg Med, 2011. 18(8): p. 781-96.
  10. Esterhai, J.L., Jr. and I. Gelb, Adult septic arthritis. Orthop Clin North Am, 1991. 22(3): p. 503-14.
  11. McGillicuddy, D.C., et al., How sensitive is the synovial fluid white blood cell count in diagnosing septic arthritis? Am J Emerg Med, 2007. 25(7): p. 749-52.
  12. Frazee, B.W., C. Fee, and L. Lambert, How common is MRSA in adult septic arthritis? Ann Emerg Med, 2009. 54(5): p. 695-700.
  13. Tintinalli, J.E., et al., Tintinalli’s emergency medicine a comprehensive study guide. 2016, New York; Chicago; San Francisco: McGraw-Hill Education.
  14. Goodacre, S., A.J. Sutton, and F.C. Sampson, Meta-analysis: The value of clinical assessment in the diagnosis of deep venous thrombosis. Ann Intern Med, 2005. 143(2): p. 129-39.
  15. Chinsakchai, K., et al., Trends in management of phlegmasia cerulea dolens. Vasc Endovascular Surg, 2011. 45(1): p. 5-14.
  16. Geroulakos, G., J. Hossain, and T. Tran, Economy-class Syndrome Presenting as Phlegmasia caerulea dolens. European Journal of Vascular and Endovascular Surgery, 2000. 20(1): p. 102-104.
  17. Pang, H.N., et al., Factors affecting the prognosis of pyogenic flexor tenosynovitis. J Bone Joint Surg Am, 2007. 89(8): p. 1742-8.
  18. Kennedy, C.D., J.I. Huang, and D.P. Hanel, In Brief: Kanavel’s Signs and Pyogenic Flexor Tenosynovitis. Clin Orthop Relat Res, 2016. 474(1): p. 280-4.
  19. Alayed, K.A., C. Tan, and N. Daneman, Red Flags for Necrotizing Fasciitis: A Case Control Study. Int J Infect Dis, 2015. 36: p. 15-20.
  20. Wong, C.-H., et al., The LRINEC (Laboratory Risk Indicator for Necrotizing Fasciitis) score: A tool for distinguishing necrotizing fasciitis from other soft tissue infections*. Critical Care Medicine, 2004. 32(7): p. 1535-1541.
  21. Al-Hindawi, A., et al., Validation of the laboratory risk indicator for necrotising fasciitis scoring system (LRINEC) in a Northern European population. Journal of Plastic Reconstructive and Aesthetic Surgery, 2017. 70(1): p. 141-143.
  22. Castleberg, E., N. Jenson, and V.A. Dinh, Diagnosis of necrotizing faciitis with bedside ultrasound: the STAFF Exam. West J Emerg Med, 2014. 15(1): p. 111-3.
  23. Thom, C. and M. Warlaumont, A Necrotizing Fasciitis Fake Out on Point-of-Care Ultrasound-Watch the Shadow. J Emerg Med, 2016.
  24. Chaudhry, A.A., et al., Necrotizing fasciitis and its mimics: what radiologists need to know. AJR Am J Roentgenol, 2015. 204(1): p. 128-39.
  25. Kehrl, T., Point-of-care ultrasound diagnosis of necrotizing fasciitis missed by computed tomography and magnetic resonance imaging. J Emerg Med, 2014. 47(2): p. 172-5.
  26. Stevens, D.L., 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(2): p. e10-52.
  27. Case definitions for infectious conditions under public health surveillance. Centers for Disease Control and Prevention. MMWR Recomm Rep, 1997. 46(Rr-10): p. 1-55.
  28. Linner, A., et al., Clinical efficacy of polyspecific intravenous immunoglobulin therapy in patients with streptococcal toxic shock syndrome: a comparative observational study. Clin Infect Dis, 2014. 59(6): p. 851-7.
  29. Coyle, E.A., R. Cha, and M.J. Rybak, Influences of linezolid, penicillin, and clindamycin, alone and in combination, on streptococcal pyrogenic exotoxin a release. Antimicrob Agents Chemother, 2003. 47(5): p. 1752-5.
  30. Darenberg, J., et al., Intravenous immunoglobulin G therapy in streptococcal toxic shock syndrome: a European randomized, double-blind, placebo-controlled trial. Clin Infect Dis, 2003. 37(3): p. 333-40.
  31. Hirschmann, J.V. and G.J. Raugi, Lower limb cellulitis and its mimics: part II. Conditions that simulate lower limb cellulitis. J Am Acad Dermatol, 2012. 67(2): p. 177 e1-9; quiz 185-6.

3 thoughts on “Cellulitis Mimics – ED Considerations”

Leave a Reply

Your email address will not be published. Required fields are marked *