Infections of skin, muscle, and necrotizing soft tissue infections

Diagnosis

Cellulitis is primarily a clinical diagnosis. Inflammation of the reticular dermis and hypodermis results in the typical findings: heat, pain, swelling, and redness in the affected region. The lower extremities are most commonly affected: 70%–80% of cases. Additional clinical markers include leukocytosis and elevated erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP), though the absence of inflammatory markers does not exclude a diagnosis of cellulitis. Cellulitis can be associated with abscesses. When possible, purulent material should be cultured to guide antibiotic therapy, but initiation of antimicrobials targeted at common gram-positive organisms or broad-spectrum antibiotics if the patient is in a high-risk group (see later) need not be delayed for culture results. Imaging is not necessary for all patients with cellulitis. However, ultrasonography, computed tomography (CT), and/or magnetic resonance imaging (MRI) may identify organized collections that require surgical intervention. In 606 episodes of cellulitis, just under half had imaging (46%), and of those 60 (22%) had purulent collection that was not detected on physical examination. Blood cultures are not recommended for all patients with cellulitis, but are indicated for those with signs of systemic inflammation, sepsis, and for neutropenic and immunocompromised patients.

Microbiology

The most common organisms responsible for cellulitis include Streptococcus pyogenes and Staphylococcus aureus, which cause approximately three-quarters of the infections in which a causative organism is identified. Different causative organisms can be seen, however, in different patient populations ( Table 115.1 ). Purulent cellulitis is most commonly associated with staphylococcal infections, with high rates of rates of methicillin-resistant S. aureus (MRSA), even in community-acquired infections in many areas.

Treatment

The cornerstone to cellulitis treatment includes early and appropriate initiation of antibiotic therapy and identification and surgical drainage of any purulent collections. Although some studies suggest that simple cellulitis is cured in over 90% of patients without MRSA coverage, even in areas with high MRSA prevalence, patients with severe cellulitis with associated signs of systemic infection should have empiric antibiotic coverage for MRSA. This is especially true for patients with evidence of prior MRSA colonization or positive nasal swab, purulent drainage, a history of injection drug use, or recent penetrating trauma. If the patient is in a high-risk group, as outlined in Table 115.1 , broad-spectrum antibiotic therapy is appropriate while awaiting culture results. Local community and hospital resistance profiles are also important in guiding initial therapy selection. The addition of empiric gram-negative and/or anaerobic therapy is appropriate in certain patient populations, including patients with severe infections, neutropenia, on chemotherapy, cell-mediated immunodeficiency, immersion injuries and/or animal bites, diabetes, and pelvic lymph node dissections, patients who are immunocompromised, or in children. ^,

The duration of antibiotics depends on patient factors and the severity of systemic infection. For the patients treated by intensivists, intravenous antibiotic therapy for at least 5 days is appropriate; however, patients whose symptoms fail to resolve should receive a longer duration, with up to 14 days of treatment reported in several studies. ^, Failure of initial antibiotic therapy or relapse of cellulitis occurs in up to 8%–20% of lower extremity infections.

Surgical incision and drainage of purulent collections and/or aspiration and placement of drains into large fluid collections is necessary for purulent cellulitis. Although the majority of collections are noted on examination (65%), some deeper collections may require imaging. If a patient does not respond to initial therapy, the intensivist should consider the presence of a deeper abscess and proceed with imaging. Aspiration and/or serial evaluation may be appropriate for abscesses that are small (<1.5 cm) or associated with high morbidity with open surgical drainage. However, in a study of cutaneous abscesses without surrounding cellulitis, aspiration alone only resolved 25% of a patient’s symptoms, regardless of size. ^,

Outcome

Mortality and outcomes after cellulitis vary greatly and are driven by the severity of infection and the patient’s comorbidities. In a large series of patients admitted to Spanish hospitals with cellulitis, 3% of patients died, but only 28% of the deaths were related to cellulitis. A recent meta-analysis of hospitalized patients with cellulitis estimated the overall in-hospital mortality rate to be 1%. In patients with severe cellulitis without a necrotizing component, however, mortality rates approach 20%–40%. Recurrence of infections occurs in up to one in five patients with lower extremity cellulitis.

Diagnosis

Early diagnosis of NSTI is essential, as initial débridement is required within hours of seeking care to decrease the risk of mortality. ^{, ,} Distinguishing severe cellulitis from NSTI is challenging. Misdiagnoses occur in up to 75% of NSTIs, and in some large series diagnoses were delayed in 100% of patients. Patients presenting early in their disease process may have skin findings that are consistent with simple cellulitis because the cutaneous manifestations of the underlying fascial necrosis have not yet manifested or may be very subtle ( Fig. 115.1 ). ^, In patients with chronic wounds, distinguishing the progression of NSTI is even more challenging. ^, Although NSTI usually presents in one anatomic location, there have been reports of multifocal NSTI. ^,

Given the challenge in diagnosis, several studies have tried to determine which clinical features can distinguish severe cellulitis from NSTI. Crepitus from gas-producing organisms is a concerning finding for NSTI, but it will be present in only 10% of NSTI patients. The most common signs of NSTI in a 2014 systematic review of 1463 patients were swelling (81%) and pain (79%). In other studies, however, swelling was only 70% sensitive and 5% specific to distinguish NSTI from cellulitis. Additional skin findings that should raise concern for NSTI include woody induration of the skin, skin sloughing or blistering, and cutaneous gangrene ( Fig. 115.2 ). Pain out of proportion to examination should also be a very concerning sign for necrotizing infection. ^{, ,} Laboratory values have also been evaluated to distinguish patients with a necrotizing component. An ICU-based sepsis study found no difference in admission CRP, white blood cell (WBC) count, creatinine, or lactate between 23 patients with severe cellulitis and 31 NSTI patients, though the NSTI group had higher rates of organ dysfunction and sepsis. Other groups have found erythrocyte counts, fibrinogen levels, pain on examination, and clinical evidence of renal failure helped to distinguish between cellulitis and NSTI.

Several authors have developed scoring systems to distinguish cellulitis from an NSTI by combining different symptoms, signs, and laboratory values. Wong and colleagues’ Laboratory Risk Indicator for Necrotizing Fasciitis (LRINEC) score based on WBC, creatinine, sodium, hemoglobin, glucose and CRP, has been the most studied, with the predictive values of scores of ≥6 (high risk) or ≥8 (very high risk) that range in sensitivity (42%–92%) and specificity (63%–78%). ^, Given this range of performance, other scores have been developed. Cribb and colleagues developed the SIARI score. This score combines six data derived variables: (1) s ite other than lower limb (2) i mmunosuppression, (3) a ge ≤ 60 years, (4) r enal impairment and (5/6) i nflammatory markers (i.e. WBC, CRP). In their recent 300-patient cohort, their SIARI score outperformed the LRINEC score in both sensitivity (81% vs. 59%) and specificity (73% vs. 64%). Alayed and colleagues considered several factors, including clinical findings (hemorrhagic bullae, pain out of proportion, necrotic skin, progressive erythema, and fluctuance), markers of sepsis (hypotension and altered sensorium), history of surgery in the prior 90 days, and recent diarrhea. They found that patients with at least three of these findings were virtually guaranteed to have an NSTI. Although scoring systems may be helpful to raise the index of suspicion, there should be a low threshold for surgical evaluation of underlying tissue if NSTI is considered possible.

Imaging

The role of imaging (e.g., ultrasound, MRI, CT, and plain x-ray) in diagnosing NSTIs has been evaluated in many studies. ^, MRI is likely the most precise, and MRI findings combined with LRINEC have improved the ability to predict an NSTI. However, when there is a concern for NSTI, MRI and other imaging modalities should not delay early surgical exploration and débridement, especially for patients with symptoms of sepsis. ^{, , ,}

Imaging showing gas in the soft tissues (see Fig. 115.1 ), much like crepitus on examination, conveys a high likelihood of NSTI, but its absence does not rule out NSTI. ^, Other CT scan findings associated with NSTI are lack of contrast enhancement of tissue, multiple fluid collections, and subfascial inflammation. Fluid collections deeper than 2–4 mm have had an increased association with NSTI. Other ultrasound findings can include thickening of the fascia and emphysema in the tissues. Imaging may also be considered in truncal NSTI to evaluate for an underlying source of intraabdominal infection presenting as an NSTI.

Operative exploration

Those evaluating severe SSTIs should maintain a high index of suspicion for NSTI, especially when the severity of illness is not consistent with the physical examination findings, and ensure early surgical consultation. Early surgical consultation facilitates wound exploration to confirm the diagnosis and to debride affected tissue as quickly as possible. ^{, , ,} At the time of exploration, clinical findings that are associated with an NSTI include thin, foul-smelling dishwasher fluid; fascial sliding; purulence; thrombosed vessels; failure of the muscle to respond to electric stimulation; and necrotic tissue at any layer, including skin, subcutaneous, fascia, or muscle. ^{, , ,} Tissue cultures and samples of any purulent collections should be sent at the time of initial débridement to guide subsequent narrowing of antimicrobial therapy, but sampling error can lead to false negatives on pathologic evaluation and should not be the only diagnostic consideration.