Acute Bacterial Skin and Skin Structure Infections (ABSSSI)

Practice Guidelines for Management and Care Transitions in the Emergency Department and Hospital

Charles V. Pollack Jr., MA, MD; Alpesh Amin, MD, MBA; William T. Ford Jr., MD, SFHM; Richard Finley, MD; Keith S. Kaye, MD, MPH; Hien H. Nguyen, MD, MS; Michael J. Rybak, PHARMD, MPH; David Talan, MD


J Emerg Med. 2015;48(4):508-519. 

In This Article


Management of ABSSSI: Time for a New Focus

Skin and skin structure infections are common causes for presentation to the ED. The majority of these patients can be treated effectively as outpatients with oral antimicrobial agents, with or without minor surgical intervention. For patients with the more serious ABSSSI, who are deemed to require parenteral therapy, effective communication and care transition between the ED and the hospital medicine service are particularly important. A clearly defined methodology for evaluating the need for inpatient care and subsequent outpatient and inpatient management is currently absent. The objective of such a methodology should be to provide highest quality care. Secondarily, the goal is to give appropriate and convenient care at the lowest possible cost, while minimizing complications, readmissions, and inappropriate antibiotic use. We will first discuss why ABSSSIs are a high-priority infection that deserve immediate attention, and then describe why a consistent, quality-driven program to manage these patients is necessary.

Why ABSSSI and MRSA are High-priority Infections

  • MRSA is prevalent in ABSSSI.

  • Rates of MRSA infection and hospitalizations are dramatically increasing.

  • MRSA and ASSSI have high rates of morbidity and other associated medical conditions.

  • MRSA infection is a risk factor for subsequent hospitalization and death.

  • Inadequate treatment of MRSA ABSSSI due to antibiotic resistance is likely a factor in relapse.

  • Urgent need exists to reduce hospitalization through the use of more effective outpatient treatment strategies.

  • Effective outpatient management can reduce cost and improve patient outcomes and satisfaction.

Profile of ABSSSI Today: The Importance of MRSA

Patients with ABSSSI present to EDs with a broad range of disease severity, ranging from cellulitis to serious, life-threatening, necrotizing infections. These infections may arise as the result of minor injuries that break the skin, from animal bites, to gunshot and knife wounds; or with no clear precipitating event.

As suggested by the FDA guidance above, staphylococci, predominantly MRSA, and to a lesser extent, methicillin-susceptible S. aureus, cause most skin infections. Some infections, such as those associated with a diabetic foot ulcer, are additionally caused by aerobic Gram-negative and anaerobic bacteria, and others, such as dog and cat bites, are due to specific zoonotic pathogens, such as Pasteurella spp. The etiology of cellulitis without a wound or drainage is unclear due to lack of culturable material, but is thought to be predominantly due to Streptococcal species such as S. pyogenes. In most instances, ED clinicians begin empiric antibiotic therapy for infections prior to knowing the results of culture and susceptibility testing. Given that in many cases of ABSSSI there are no positive wound or blood cultures giving subsequent definitive guidance, it is all the more important that empiric coverage be guided by recognition of the most common pathogens, which are Gram-positive cocci. Gram-negative coverage is not empirically indicated in most cases of ABSSSI.[2,3]

The choice of empiric antibiotics has not always been optimal. For example, in one study of ED practice in 2004, before MRSA was recognized as a frequent cause of ABSSSI, 59% of patients were found to have infection caused by MRSA. Only 57% of these patients were prescribed an empiric antibiotic to which the isolate demonstrated in vitro resistance.[4]

The wide spectrum of illness caused by MRSA includes not only ABSSSIs, but also bacteremia and endocarditis, pneumonia, bone and joint infections, necrotizing fasciitis/myositis, central nervous system infections, and toxic shock and sepsis syndromes. The emergence of MRSA is, however, a truly recent phenomenon. Prior to 2000, MRSA was more commonly encountered in nosocomial infections than in community-onset infections. A study of cutaneous abscesses associated with intravenous drug use in the early 1990s showed no methicillin resistance among the S. aureus isolates.[5] Then, between 2001 and 2005, the prevalence of MRSA infections among ABSSSI cases increased from 29% to 64% in a single Los Angeles ED.[6] In a geographically diverse network of EDs in the United States collecting isolates in the early 2000s, MRSA was identified as the most common cause of ABSSSI.[4] Similarly, data from the SENTRY Antimicrobial Surveillance Program that evaluated causes and types of SSTI from 1998 to 2004 revealed that S. aureus was present in 44.6% of isolates in North America, of which 35.9% were methicillin resistant.[7] By 2008, empiric coverage for MRSA in the ED management of ABSSSI was common.[8]

Compared with hospital-associated MRSA (HA-MRSA), community-acquired MRSA (CA-MRSA) tends to be more virulent and may carry genes that encode the Panton-Valentine leukocidin, as well as many other exotoxins that are associated with tissue necrosis and greater severity of disease.[9–13] Although many strains have emerged globally, in the United States, USA300 is the most common CA-MRSA strain.[6,8,9,14,15]

Scope of the Problem: the ED Perspective

According to the National Hospital Ambulatory Medical Care Survey, the number of ED visits for skin infections almost tripled from the late 1990s until 2005, and continues to increase at an alarming rate (Figure 1).[1]

Figure 1.

Annual visits to United States emergency departments for selected acute bacterial skin and skin structure (16). MRSA = methicillin-resistant Staphylococcus aureus.

Data published in 2009 showed that ABSSSI (though discussed using different terminology at that time) accounted for almost 870,000 hospital admissions in the United States in 2004. This represents an increase of almost 30% in the incidence of this diagnosis over a 4-year period.[16] Other studies have reported a 50% increase in outpatient visits for skin and skin structure infections over an 8-year period.[17] These increases in disease incidence have been attributed primarily to the epidemic spread of CA-MRSA.[16–18] The fact that CA-MRSA is now common, even though colonization in the general population is low, suggests that CA-MRSA may colonize other body sites or have virulence characteristics that favor de novo colonization and infection in susceptible hosts.[19] In addition, during much of the first decade of this century, insufficient attention was paid by empiric prescribers to the role of MRSA in ABSSSI, and it is likely that prescribing practices played a role in the substantially increased incidence of MRSA infections.

Morbidity and Mortality

Although HA-MRSA has been increasingly implicated in life-threatening and fatal nosocomial infections, the impact of CA-MRSA infections may be underestimated because hospital-based bacteriologic surveillance studies often do not capture data on infected patients who are not admitted. In light of this information gap, Delaney and colleagues conducted a large cohort study comprised of 1439 patients in general practices diagnosed with CA-MRSA in the UK between 2001 and 2004, and 14,090 matching noninfected patients from the General Practice Research Database, to assess mortality associated with community-diagnosed CA-MRSA infections.[20] Their principal finding was that patients who develop a CA-MRSA infection are at significantly greater risk of mortality during the year after diagnosis, compared with patients who did not contract it: 21.8% vs 5.0%, respectively, as shown in Figure 2.[20] The patients with CA-MRSA commonly had comorbidities, such as diabetes, that are also associated with poor outcomes, so a truly causative link between MRSA and 1-year mortality may not exist; nonetheless, CA-MRSA may serve as a marker for "sicker" patients.

Figure 2.

Kaplan-Meier plot of the cumulative probability of death up to 1 year in patients after diagnosis with methicillin-resistant Staphylococcus aureus (MRSA) (20).

The Delaney study also revealed some important clinical associations between CA-MRSA infections and comorbid conditions among patients in the general practice population. Figure 3 shows the prevalence of various comorbidities in CA-MRSA patients, as compared to the prevalence of those diseases in patients who did not have MRSA infections.[20] Note that diabetes, cardiovascular disease and stroke, peripheral vascular disease, chronic obstructive pulmonary disease, renal failure, and cancer were all more common in patients with MRSA infections than in MRSA noninfected patients. In the same study, patients with a diagnosis of MRSA were more likely to die (events/100 patient-years) than those without MRSA (unadjusted hazard ratio [HR] 4.85; 95% confidence interval [CI] 4.25–5.54), as shown in Figure 4.[20]

Figure 3.

Comorbidities in patients with community-acquired methicillin-resistant Staphylococcus aureus (MRSA) (20). CVD = cardiovascular disease; PVD = peripheral vascular disease; COPD = chronic obstructive pulmonary disease.

Figure 4.

All-cause mortality and hospitalization within 1 year (20). MRSA = methicillin-resistant Staphylococcus aureus.

The HR for death from any cause in patients with a diagnosis of CA-MRSA was 4.08 (95% CI 3.54–4.69) as compared to patients without CA-MRSA, after adjustment for age, gender, and the comorbid conditions. Patients with CA-MRSA were also at greater risk of hospitalization or death within 1 year after diagnosis, with an adjusted HR of 3.31 (95% CI 2.97–3.69), compared to patients without MRSA. Antibiotic treatment for any cause in the year prior to study entry exerted only a small effect on mortality after adjustment for other factors, with an adjusted HR of 1.30 (95% CI 1.14–1.49). Among CA-MRSA patients without comorbidities, the risk of mortality was comparable to that for all patients, with an adjusted HR of 4.86 (95% CI 4.10–5.77).[20] The results of this study suggest that CA-MRSA is at least a marker of poor outcomes and, particularly in patients with significant comorbidities, should prompt particular attention to hospital care and follow-up. A recent analysis of US Healthcare Cost and Utilization Project (HCUP) data by Talan found that the mortality rate of patients hospitalized for ABSSSI was 0.5%.[21]

Important Comorbidities with MRSA ABSSSI

Diabetes. Data suggest that patients with types 1 and 2 diabetes are at 1.6 and 1.3 times greater risk of developing ABSSSIs, respectively, compared with those without diabetes.[22] Given the increasing incidence of type 2 diabetes, this association may be in part responsible for increasing infection rates among the general population.

Suaya and colleagues investigated the relationship among diabetes, ABSSSIs, and complications of ABSSSIs in International Classification of Diseases, 9th Revision-stratified records of 140,652 patients with and 1,539,692 patients without diabetes, who were between the ages of birth and 64 years and had at least 1 episode of ABSSSI.[22] The overall sample was comprised of 2,227,402 distinct episodes of ABSSSI, of which 10% occurred in patients with diabetes.[22] Among patients with ABSSSIs whose initial diagnoses occurred in outpatient settings, the most commonly observed complication was osteomyelitis, which occurred in 3.3% of patients with, and 0.4% of patients without, diabetes (p < 0.01). Bacteremia, endocarditis, septicemia, and sepsis were the most commonly reported inpatient complications associated with ABSSSI, and occurred in 25% of patients with, and 16% of patients without, diabetes (p < 0.01). Rates of hospital admission were significantly higher among patients with, compared to those without, diabetes for all sites and types of ABSSSI, with the exception of folliculitis.[22] These results suggest the need for more cautious evaluation and management of diabetic patients with ABSSSI, and closer follow-up, due to the potential for complications.

Advanced Age. Data from HCUP reveal that the highest rate of MRSA hospitalization is among the elderly—360.8 MRSA stays per 100,000 patients over 65 years of age.[23] This is more than three times higher than for any other age group: 114.7 stays for infants, 19.2 for 1- to 17-year-olds, 58.1 for 18- to 44-year-olds, and 111.5 for 45- to 64-year-olds per 100,000 (Figure 5).[23] This may reflect higher hospital exposure, a higher burden of comorbidities, diminished immune function, or some combination of these and other factors.

Figure 5.

Methicillin-resistant Staphylococcus aureus infections by age group (23).

ED as the "Portal of Entry"

Inpatients with MRSA ABSSSI infections are more likely to be first cared for in the ED than to be transfers from another hospital, or to be transfers from long-term care settings (Figure 6).[23] Although it may not be surprising that the largest number of patients with MRSA infections were admitted from the ED, this finding underscores the importance of having a well-defined process for identifying, risk-stratifying, and empirically managing patients with ABSSSI both in the ED and all the way through their hospital course.

Figure 6.

Source of admission for discharges with methicillin-resistant Staphylococcus aureus (MRSA) infections (23).

Cost of Treatment

Cost containment and cost-efficient patient management are top priorities today for hospitals, health systems, and Accountable Care Organizations. The estimated mean cost of an ABSSSI hospitalization in the United States is $8023 with a 4.9-day length of stay and associated risks. Lee and colleagues developed an economic simulation model to quantify CA-MRSA-associated (specifically; these figures may not be generalizable to all ABSSSI) costs from societal and third-party payer perspectives.[24] They noted that hospitalization rates and mortality are important cost drivers and that CA-MRSA infections result in a substantial economic burden on third-party payers and society. Major contributors to the total societal economic burden include CA-MRSA-attributable productivity losses. Their findings were as follows:[24]

  • Cost of a single CA-MRSA case to a third-party payer: $2277–$3200

  • Cost to society: $7070–$20,489, depending on patient age

  • Annual burden to U.S. third-party payers: $478 million to $2.2 billion

  • Annual burden to society (U.S.): $1.4–13.8 billion on society, depending on the CA-MRSA definitions and incidences

Results of an observational study that was conducted in a primary care setting built on the above findings showed that 1 in 5 patients with CA-MRSA who presents to a primary care setting is likely to require additional interventions, each of which is associated with a cost of almost $2000 per incident.[25]

Risk Stratification in the ED

A major component of clinical decision-making in the ED involves risk stratification.[26–29] This is certainly true of ABSSSI, for which life- or limb-threatening infection must be distinguished from less severe infections. We describe and recommend a step-wise approach to ABSSSI risk stratification.

Assess for Hemodynamic Instability. For patients with skin and skin structure infections who are hemodynamically unstable, resuscitation should begin immediately, accompanied by appropriate antimicrobial coverage. Necrotizing infections should be considered (see next). These patients are typically admitted to the hospital, often to intensive care settings. Hemodynamically stable but acutely ill patients with ABSSSI should be screened for sepsis as per institutional protocols/pathways.[30]

Assess for Possible Necrotizing Fasciitis. This is largely a clinical assessment, taking into account signs of severe sepsis, disproportionate pain, rapidity of advancement, and evidence of soft tissue gas, compartment syndrome, or muscle necrosis (e.g., elevated creatine phosphokinase). This evaluation can be enhanced by use of the Laboratory Risk Index for Necrotizing Fasciitis (LRINEC). This score was developed about 10 years ago on the basis of a study of 145 patients who were admitted with the diagnosis of necrotizing fasciitis, compared to a control population of 309 patients hospitalized for cellulitis or abscess.[31] Abnormalities of the total white blood cell count (15–25,000/mm3 = 1 point; > 25,000 = 2 points), hemoglobin (11–13.5 g/dL = 1 point, < 11 g/dL = 2 points), sodium (<135 mEq/L = 2 points), glucose (>180 mg/dL = 1 point), serum creatinine (>1.6 mg/dL = 2 points), and C-reactive protein (>150 mg/L = 4 points) were used to construct a score by converting the regression coefficients of independently predictive factors for diagnosing necrotizing fasciitis. The cutoff value for the LRINEC score was ≥ 6 points, with a positive predictive value of 92.0% and negative predictive value of 96.0%. The score is one tool that can be used along with clinical judgment and serial observations (e.g., observing a drawn border of the lesion for progression) to determine if necrotizing ABSSSI is likely.

Necrotizing fasciitis is a clinical diagnosis for which emergency surgical exploration and treatment, if the diagnosis is confirmed, should not be delayed. Expedited computed tomography (CT) and magnetic resonance imaging (MRI), by showing better than plain radiography the extent and detail of deep tissue involvement, and gas when present, may have a role in stable patients in whom the diagnosis is uncertain. These advanced imaging findings, when present, however, are often nonspecific and thus, may further delay potentially critical surgery. It is important to note that findings of necrotizing fasciitis—subcutaneous thickening, air, and fascial fluid—have also been appreciated on bedside ultrasound, and use of this modality in the ED may save time.[32,33]

Evaluate for Unstable Comorbidities. Comorbidities may have a significant impact on outcomes and mortality in ABSSSI and may independently require inpatient management. If the patient has diabetic ketoacidosis or has an acute exacerbation of any other disease that could jeopardize the success of outpatient treatment, hospital admission is indicated.

Assess for High-risk Locations and Lesions That Require Extensive Surgery. Infections involving certain areas may require more careful evaluation and monitoring in the hospital, such as orbital cellulitis and hand infections. In larger and more indurated lesions in which there is a possibility of a deep abscess that cannot be discerned by clinical examination, soft tissue ultrasound may provide valuable information. Patients with large and deep abscesses, and infected areas that require extensive debridement, may require hospitalization for operative treatment and intensive physical therapy. Inadequate drainage is often a reason for apparent failure of an initial outpatient antibiotic regimen.[34]

Identify Social and Personal Factors That Would Interfere With Successful Outpatient Care. The patient's home and emotional condition should be evaluated. Those with no social support and for whom such support cannot be found or arranged, or who are deemed to be psychologically unstable or unreliable with regard to following treatment instructions and meeting follow-up needs, may need inpatient treatment. Ongoing intravenous drug use may also be an important consideration, both as a portal of entry for ABSSSI and a concern for reliability regarding compliance.

Many patients who do not have these indications for hospitalization can be managed as outpatients with close follow-up and treated with various outpatient strategies. These strategies include oral antibiotics, a once-a-day parenteral antibiotic given in the ED with the peripheral catheter left for next-day follow-up and additional dosing, insertion of a peripheral intravenous central catheter (PICC) and outpatient parenteral antibiotic therapy (OPAT), and new extended half-life parenteral lipoglycopeptides for single-dose or weekly dose treatment.[35–37]

Neither the presence nor magnitude of fever precludes outpatient management. For example, two randomized, double-blind, double-dummy trials comparing dalbavancin, two injections one week apart, to intravenous vancomycin for at least 3 days followed by oral linezolid, found similar response rates and no septic deaths among 1315 admitted ABSSSI patients with areas of erythema > 300 cm2 and frequent fever (84% had temperature > 38.0°C). Overall, approximately 25% of the patients in these two trials received all treatment for their infections on an outpatient basis.[34,36] The availability of oritavancin has the potential to increase the use of that option even further, as a second dose is not required.[38]

Choice of Antibiotics

Considerations in selection of an empiric antibiotic for the treatment of ABSSSI in the ED include likely bacteriologic etiology, susceptibility, spectrum, host status (e.g., neutropenia, immunocompromise, diabetes), compliance, allergies, and location. Guidelines from the Infectious Diseases Society of America published in 2014 avoid recommending specific agents over one another.[39] This reflects an evidence basis of clinical trials that were designed to be noninferiority comparisons. We therefore follow that approach, recognizing that emergency physicians and hospitalists frequently must make empiric treatment decisions that are complex for any individual patient.

Current recommendations for the treatment of skin infections when MRSA is suspected or identified are shown below.[39] For each of the two disease categories, the recommended antibiotic and adult dose (in nonpregnant patients with normal hepatic and renal function) are provided:

  • Purulent cellulitis (defined as cellulitis associated with purulent drainage or exudate in the absence of a drainable abscess)

    • Clindamycin, 300–450 mg orally (p.o.) 3 times daily

    • TMP-SMX (trimethoprim-sulfamethoxazole), 1–2 double strength tab p.o. twice daily

    • Doxycycline, 100 mg p.o. twice daily

    • Minocycline, 200 mg × 1, then 100 mg p.o. twice daily

    • Linezolid, 600 mg p.o. twice daily

  • Complicated SSTI (now called ABSSSI)

    • Vancomycin, 15–20 mg/kg/dose intravenously (i.v.) every 8–12 hours (q8–12h)

    • Linezolid, 600 mg p.o./i.v. q12 h

    • Daptomycin, 4 mg/kg/dose i.v. q24 h

    • Telavancin, 10 mg/kg/dose i.v. q24 h

    • Clindamycin, 600 mg p.o./i.v. q8h

    • Tedizolid, 200 mg p.o./i.v. q24 h

    • Dalbavancin, 1000 mg i.v. on day 1, then 500 mg i.v. on day 8

    • Oritavancin, 1200 mg i.v. single dose

Some of the common considerations pertinent to emergency physicians and hospitalists with regard to the choice of FDA-approved parenteral antibiotics with MRSA activity for more serious ABSSSI are summarized in Table 1. These represent the opinions and clinical experience of the authors and are not official guidelines or recommendations. They are not meant to represent a complete list of all important considerations. They also serve as a reminder that understanding the perspectives of all members of the treatment team can help to create a smoother transition from the ED to the hospital and better coordination of care. In general, hospital-based providers should be attuned to identifying opportunities for improving antimicrobial stewardship, choosing drugs and doses that are safe and effective for the acutely ill patient, accounting for comorbidities such as impaired renal function that may impact dosing and monitoring regimens, and following culture and sensitivity results once available. "Antimicrobial stewardship" refers to coordinated interventions designed to improve and measure the appropriate use of antimicrobials by promoting the selection of the optimal antimicrobial drug regimen, dose, duration of therapy, and route of administration.[40]

Current Payment and Regulatory Forces Impacting the Management of ABSSSI

Health care demands have forced hospitals and health systems to reevaluate care provided for many patient populations. There are a number of factors that have led to the development of programs focused on different diseases. Table 2 outlines many of the factors that may lead emergency physicians and hospitalists to streamline and standardize management of ABSSSI.[40–43]

As health care becomes increasingly regulated, medical practice becomes increasingly scrutinized. Not only are improved outcomes without complications expected, but in addition, care must be delivered in a manner that is streamlined, efficient, and without waste. It may be difficult for frontline physicians to keep up with the litany of mandates.

Evaluating ABSSSI as an example in this new and ever-changing care and reimbursement paradigm suggests how a systems-based and standardized management plan may be helpful. Programs should evaluate and limit the use of relatively low-yield diagnostic tests, such as erythrocyte sedimentation rate, for which there is no defined role in ABSSSI unless osteomyelitis is suspected. Use of ED ultrasound instead of CT scans in ABSSSI to evaluate for deep abscesses can save time and resources (Choosing Wisely).[43] Use of MRI should be reserved for situations in which the result will importantly impact management.[2] Standard antimicrobial regimens for empiric therapy (based on local resistance patterns), standard approaches to bacteriologic testing, and uniform patient follow-up procedures increase consistency and quality of care (Affordable Care Act [ACA]). Finally, emergency physicians and hospitalists must remain current with the availability of new antimicrobial options (such as long-acting agents with activity against MRSA) and with local capabilities for home infusion services and close follow-up.

Multidisciplinary ABSSSI evaluation and management programs can potentially contribute to better outcomes and coordination of care from the ED to the hospital (in patients who require admission) and from the ED/hospital to the outpatient setting (ACA, Two-Midnight Rule). As the majority of ABSSSI patients have a low mortality rate, hospital avoidance should be a consideration for ABSSSI programs, particularly for patients who have been assessed for the potential of complications and bad outcomes in a standardized fashion. The use of OPAT programs and new extended half-life parenteral agents promote cost-effective care and improve ED throughput by allowing ED discharge (ACA, ED throughput). By avoiding admission, naturally, readmissions would be minimized (ACA). Additionally, not only do OPAT programs improve care transitions out of the ED, they may also improve transitions out of the hospital, ensuring patients who may not have an available primary care physician close follow-up for their infection (ACA).

Another consideration specific to ABSSSI care is a different perspective on drug acquisition cost, based on the settings in which the drugs are used. The inclusion of "cost" is relevant for all branded agents (all have higher acquisition cost compared to generics), but drug acquisition cost is only one cost component in overall total cost per episode of care. With longer-acting antimicrobial agents, there is a potential opportunity for broader "cost avoidance" through shortening or wholly averting inpatient admission, reducing ancillary care requirements, and reducing the use of PICC lines. Shifting to an outpatient or ED-only site of care can also trigger a different reimbursement formula, such as under Medicare, where drugs are separately reimbursed. This is in stark contrast to the inpatient, fixed payment by diagnosis-related group.[44]

ABSSSI programs developed on epidemiologic evidence, incorporating evidenced-based recommendations, and considering total, overall costs have the potential to improve patient outcomes, satisfaction, value, and will address the changing health care paradigm.