Incision and Drainage

Abscess Etiology and Pathogenesis

Localized pyogenic infections may develop in any region of the body. Abscesses generally begin as a localized superficial cellulitis. Some organisms cause necrosis and liquefaction, as well as the accumulation of leukocytes and cellular debris. This is followed by loculation and subsequent walling off of these products, all of which result in the formation of one or more abscesses. Any process or event that causes a breach in the skin's defensive epithelial barrier increases risk for the development of an abscess. This includes primary dermatologic conditions as well as trauma. The lymphatic tissues may be involved in this form of lymphangitis or “streaking.”

An isolated abscess is often associated with local symptoms only, lacking fever, systemic complaints, or abnormal blood tests. Systemic signs of toxicity or fever suggest deeper tissue involvement, bacteremia, or both. As the process progresses, the area of liquefaction increases until it “points” and eventually ruptures into the area of least resistance. This may be toward the skin or the mucous membrane, into the surrounding tissues, or into a body cavity. If the abscess is particularly deep, spontaneous drainage may not occur. In some cases, a fistulous tract can arise and lead to the formation of a chronic draining sinus. This development—or the recurrence of an abscess that was previously drained—should broaden the etiologic differential. For example, recurrent abscesses in the perineal or lower abdominal area should raise suspicion for inflammatory bowel disease as the trigger, and recurring abscesses in the axilla or groin should raise the possibility of hidradenitis suppurativa (HS). Chronic abscesses may be associated with an immunocompromised state or intravenous drug use (IVDU), and recurrent abscesses may suggest the possibility of a retained foreign body, underlying osteomyelitis, or the presence of an atypical or drug-resistant organism ( Fig. 37.2 ).

Various organisms that colonize normal skin can cause necrosis and liquefaction with subsequent accumulation of leukocytes and cellular debris. Loculation and subsequent walling off of these products leads to abscess formation. The cause of an abscess depends on its anatomic location and flora indigenous to that area. Different organisms cause disease based on environmental exposure. For example, direct inoculation of extraneous organisms may occur during a mammalian bite (e.g., Eikenella , Pasteurella ), exposure to saltwater (e.g., various Vibrio strains) or freshwater (e.g., Aeromonas ), or meat or fish exposure (e.g., Erysipelothrix rhusiopathiae ). Pseudomonas folliculitis has been associated with the use of a hot-tub, as this organism thrives in a warm, wet environment.

Staphylococcal strains, which are normally found on the skin, produce rapid necrosis, early suppuration, and localized infections with large amounts of creamy yellow pus, the typical manifestation of an abscess. Conversely, group A β-hemolytic streptococcal infections tend to spread through tissues and cause a more generalized infection characterized by erythema, edema, a serous exudate, and little or no necrosis, typical manifestations of cellulitis. Anaerobic bacteria, which proliferate in the oral and perineal regions, produce necrosis with profuse brownish, malodorous pus and may cause both abscesses and cellulitis.

Normal skin is extremely resistant to bacterial invasion, and few organisms are capable of penetrating intact epidermis. In a normal healthy host with intact skin, the topical application of even very high concentrations of pathogenic bacteria does not result in infection. The requirements for infection usually include a high concentration of pathogenic organisms, such as in hair follicles in the adnexa; occlusion of glands or other structures that prevent desquamation and normal drainage; a moist environment; adequate nutrients; and trauma to the corneal layer, which allows organisms to penetrate into deeper tissues. Tissue perfusion may also play a role in the ability to prevent infection. Trauma may be the result of abrasions, shaving, insect bites, hematoma, injection of chemical irritants, incision, or occlusive dressings that macerate the skin. The presence of a foreign body can potentiate skin infections by enabling a lower number of bacteria to establish an infection. For example, abscesses occasionally develop at suture sites in otherwise clean wounds. In addition, abscesses can develop at any site used for body piercing. Ear piercings through the cartilage of the pinna seem to be at particular risk for infection because of the avascularity of auricular cartilage.

When favorable factors are present, the normal flora that colonize cutaneous areas flourish and infect the skin and deeper structures. In persons performing manual labor, the arms and the hands are infected most frequently. In women, the axilla and submammary regions are frequently infected because of minor trauma from shaving, contact with undergarments, a moist environment, and an abundance of bacteria in these areas. Infections may develop anywhere on the body in intravenous (IV) drug users, although the upper extremities are most commonly affected. Deep soft tissue abscesses can be caused by an addict's attempts to access deep venous structures when peripheral venous access sites are exhausted. In addition, areas with compromised blood supply are more prone to infection because normal host defenses, including cell-mediated immunity, are less available.

Special Considerations

Parenteral drug users, insulin-dependent diabetics, hemodialysis patients, cancer patients, transplant recipients, and individuals with acute leukemia have an increased frequency of abscess formation when compared with the general population. At initial evaluation, the patient may emphasize an exacerbation of the underlying disease process or an unexplained fever, with symptoms of an abscess being a secondary complaint. In these situations, abscesses tend to have exotic or uncommon bacteriologic or fungal causes and typically respond poorly to therapy. Patients with diabetes-induced ketoacidosis (DKA) should be evaluated extensively for an infectious process; a rectal examination should be included with the physical examination to rule out a perirectal abscess as the infectious trigger of DKA. This is also true for patients who are immunocompromised. There are several reasons why patients with diabetes and parenteral drug users are at increased risk for abscess formation: intrinsic immune deficiency, an increased incidence of staphylococcal carriage, potentially compromised tissue perfusion, and frequent needle punctures, which allow a mode of entry for pathogenic bacteria.

Patients who use IV drugs frequently use veins in the neck and in the femoral areas, which can produce abscesses and other infectious complications at these sites. Any abscess near a vein of the antecubital fossa or dorsum of the hand should alert the clinician to possible IV drug use; however, substance users may also inject directly into the skin (“skin popping”), which can cause cutaneous abscesses distant from veins ( Fig. 37.3 ).

A foreign body may serve as a nidus for abscess formation. Patients with a history of IVDU frequently break needles in skin that has been toughened by multiple injections, so the clinician should maintain a high index of suspicion for retained needle fragments. If an abscess is recurrent or if the patient is a known or suspected IV drug user, consider radiographs or other techniques to search for foreign bodies, an underlying septic joint, or osteomyelitis. Ultrasound is also a useful adjunct to evaluate for foreign bodies that are not radiopaque.

MRSA

First acknowledged in the 1960s as a cause of infection in patients in health care settings, MRSA has now become the most common identifiable cause of community-acquired skin and soft tissue infections in many metropolitan areas in the United States. The spread of this organism is considered an epidemic and it is very virulent and aggressive. One study found an MRSA prevalence as high as 75% to 80% in some parts of the country.

Virulent community-acquired MRSA (CA-MRSA) causes rapid and destructive soft tissue infection because of the presence of two bacterial toxins elaborated by the omnipresent USA-300 and USA-400 strains. Panton-Valentine leukocidin enhances tissue necrosis, and phenol-soluble modulin is toxic to neutrophils. Methicillin resistance is mediated by PBP-2a, a penicillin-binding protein encoded by the mecA gene, which permits the organism to grow and divide in the presence of methicillin and other β-lactam antibiotics. S. aureus acquires methicillin resistance through a mobile staphylococcal cassette chromosome (SCC) that contains the mecA gene complex (SCC mec ). MRSA probably arose as a result of antibiotic selective pressure.

A single clone probably accounted for most MRSA isolates discovered during the 1960s; by 2004, six major MRSA clones had emerged. The spread of resistance is thought to be mediated by horizontal transfer of the mecA gene and related regulatory sequences thereon.

In 1980, the spread of MRSA from hospitals into communities became evident. More recently, community-acquired infections have occurred more frequently, even in people without known risk factors. A small pustule can become a large abscess in 24 to 48 hours ( Fig. 37.4 ). Such lesions are often mistaken for a spider bite or drug use because of their rapid progression and seemingly spontaneous onset in an otherwise healthy person. These observations have led to the identification of some risk factors for CA-MRSA, including skin trauma (e.g., lacerations, tattoos, IV and intradermal drug use, shaving), incarceration, shared razors or towels, and close contact with others colonized or infected with MRSA. Animals can also carry MRSA and can function as a source of transmission. Importantly, many patients with CA-MRSA have no identifiable risk factors for acquisition of the disease. CA-MRSA tends to be more virulent than health care–associated MRSA (HA-MRSA) and is associated with more frequent serious complications such as osteomyelitis, joint infections, sepsis, and death. However, these organisms fortunately tend to be susceptible to a broader array of antibiotics.

The prevalence of MRSA has increased in both health care and community settings. For example, the prevalence of methicillin resistance among S. aureus isolates in intensive care units in the United States has been reported at 60%, and more than 90,000 invasive infections by MRSA occurred in the United States in 2005.

HA-MRSA and CA-MRSA differ with respect to their clinical epidemiology and molecular structures. HA-MRSA is defined as MRSA infection that occurs following hospitalization (hospital onset, formerly “nosocomial”) or MRSA infection that occurs outside the hospital within 12 months of exposure to a health care setting (e.g., history of surgery, hospitalization, dialysis, or residence in a long-term care facility—community onset instead of community acquired).

HA-MRSA is usually associated with severe, invasive disease, including skin and soft tissue infection, bloodstream infection, and pneumonia. In fact, S. aureus continues to be a significant cause of surgical site infections. HA-MRSA strains tend to be resistant to multiple drugs. MRSA is one of the few pathogens routinely implicated in nearly every type of hospital-acquired infection. This is probably related in part to the organism's capacity for biofilm formation on indwelling lines and tubes in hospital settings. Biofilm facilitates survival and multiplication of MRSA on these surfaces, thereby prolonging the duration of exposure of the organism to antibiotics, as well as promoting the potential for the development of genetic resistance.

CA-MRSA is defined as MRSA infection that occurs in the absence of health care exposure. It is often associated with skin and soft tissue infections in young, otherwise healthy individuals. Most CA-MRSA strains are sensitive to non–β-lactam antibiotics, although a multidrug-resistant isolate has been described in men who have sex with men. This strain contains the pUSA03 plasmid and carries resistance genes for β-lactams, fluoroquinolones, tetracycline, macrolides, clindamycin, and mupirocin.

The CA-MRSA and HA-MRSA classifications are no longer distinct as MRSA colonization can develop in one realm and manifestations of infection in another. In the mid-2000s in San Francisco, the annual incidence of CA-MRSA surpassed that of HA-MRSA.

Furthermore, community-onset HA-MRSA infections have been observed with increasing frequency. This was illustrated in a study of 209 patients discharged from hospitalized care; within 18 months following hospital discharge, 49% of new MRSA infections began outside the hospital. In another series of 102 patients with CA-MRSA infections, 29% had molecular typing consistent with HA-MRSA.

CA-MRSA was initially reported in injecting drug users in the early 1980s and has since become the most frequent cause of skin and soft tissue infections seen in US EDs and ambulatory clinics. In an assessment of the prevalence of MRSA across the United States, Moran and colleagues compiled data from adults who sought treatment of acute skin and soft tissue infections in EDs in 11 American cities in August 2004. S. aureus was isolated from three fourths of the 422 patients who met the study criteria. Seventy-eight percent of the S. aureus isolates were resistant to methicillin. MRSA was isolated from 59% of patients in the study. The prevalence of MRSA ranged from 15% to 74% in the participating EDs. MRSA was the most common identifiable cause of skin and soft tissue infections in all but one of the EDs.

Frazee and associates, reporting from an ED in northern California, found that half of the 137 patients in their study were either infected with or colonized by MRSA. Three fourths of all S. aureus isolates were MRSA. In addition, 76% of cases met a strict clinical definition of CA-MRSA. The incidence of CA-MRSA, genetically unrelated to nosocomial isolates, increased steadily from 1990 to 2001 and then dramatically in 2002 and each year thereafter.

MRSA has also emerged as a potential sexually transmitted disease. Roberts and colleagues described their treatment of two patients who came to their urban ED with abscesses, probably transmitted by heterosexual oral-genital contact. Both tested positive for MRSA. A 2010 case report drew a similar conclusion. It reported orogenital transmission of MRSA to an immunocompetent 22-year-old man who tested orally positive for MRSA and group B (genital) Streptococcus after oral contact with a female partner in whom MRSA-positive gluteal lesions had previously been diagnosed. MRSA abscesses have been described following skin-to-skin contact during a lap dance.

In a retrospective chart review, Roberts and colleagues found that 18% of the 524 subcutaneous abscesses treated in their urban ED in 2006 were confined to the genital area. Almost three fourths of the 272 outpatient wound cultures performed on that year's patient population were positive for MRSA.

Manifestations of Cutaneous Abscesses

The diagnosis of cutaneous abscess formation is usually straightforward. The presence of a fluctuant mass in an area of induration, erythema, and tenderness is clinical evidence that an abscess exists (see Fig. 37.1 ). An abscess may appear initially as a definite, tender, soft tissue mass, but in some cases a distinct abscess may not be readily evident. If the abscess is deep, as is true of many perirectal, pilonidal, and breast abscesses, the clinician may be misled by the presence of a firm, tender, indurated area without a definite mass. If the findings on physical examination are equivocal, needle aspiration or ultrasound examination may be performed to assist in the diagnosis. This approach may also identify a mycotic aneurysm or an inflamed lymph node simulating an abscess. A specific entity commonly mistaken for a discrete abscess is the sublingual cellulitis of Ludwig's angina (see Chapter 65 ).

Parenteral injection of illicit drugs can produce simple cutaneous abscesses that unpredictably advance to extensive necrotizing soft tissue infections. The emergency clinician must maintain a high index of suspicion to avoid missing this potentially life-threatening condition. Cellulitis and abscess formation can lead to bacteremia and sepsis, especially in immunocompromised patients.

The pain of an abscess often brings the patient to the hospital before it spontaneously ruptures, or the patient can have a draining abscess that appears to have undergone spontaneous rupture and is self-resolving. The patient may have even punctured the abscess in an attempt to drain it. In most cases, a formal I&D procedure will be necessary to effectively manage the condition, even though copious drainage may not be encountered. Although no formal drainage may be required after the spontaneous rupture of a simple cutaneous abscess, conditions such as a perirectal abscess, Bartholin gland abscess, and breast abscess are usually best managed with further appropriate drainage and packing.

Imaging

Ultrasound Box 37.1: Cellulitis and Abscesses

Ultrasound offers a distinct advantage when evaluating a patient with suspected soft tissue infection and may change management. A recent study from Academic Emergency Medicine by Tayal and colleagues evaluated the effect of soft tissue ultrasound on the management of cellulitis in the emergency department. The authors found that in patients with a low suspicion for abscess, ultrasound changed management in 56% of cases. Peritonsillar abscesses are difficult to diagnose from the physical examination alone, and some clinicians may feel hesitant to attempt blind drainage. Ultrasound of suspected peritonsillar abscesses has been found to be reliable in making the diagnosis. The overall size of the abscess, as well as its proximity to the carotid artery, can be evaluated with ultrasound, which will perhaps improve the confidence of the clinician in attempting drainage.

General Considerations

Typically, a high-frequency (7.5 to 10 MHz) transducer should be used to evaluate the superficial soft tissues. The higher frequency will allow the clinician sufficient resolution to identify changes consistent with soft tissue infection. The entire area should be scanned in detail, in multiple planes, to identify fluid pockets. Surrounding structures in the area should also be evaluated, especially when incision and drainage are planned. When evaluating the posterior pharynx for a potential peritonsillar abscess, an intracavitary transducer should be used.

Normal Soft Tissue

Normal soft tissue is characterized by well-defined layers, with clear demarcation between these layers ( Fig. 37.US1 ). The top of the screen corresponds to the most superficial soft tissue, including the epidermis and dermis. It should appear hyperechoic (light gray to white), thin, and clearly separate from the underlying layers. Subcutaneous tissue is found beneath the dermis and is of varying thickness. However, as with the most superficial layers, this layer should appear thin and well demarcated from the surrounding layers. Underneath the subcutaneous tissue, muscle will typically be seen as layers of striated tissue separated by bright layers of fascia.

Cellulitis

Cellulitis is recognized on ultrasound by thickening of the skin and subcutaneous layers ( Fig. 37.US2 ). The tissue may also appear more hyperechoic than normal soft tissue. When a significant amount of edema is present within the tissue, bands of hypoechoic (dark gray) or anechoic (black) fluid may be seen within the area of thickened tissue. This is known as “cobblestoning” ( Fig. 37.US3 ). Cobblestoning appears as thin bands of fluid throughout the tissue and can be distinguished from an abscess by the lack of a discrete fluid collection.

Abscess

An abscess is seen as a focal, discrete fluid collection within an area of cellulitis ( Fig. 37.US4 ). The presence of surrounding cellulitis is the key to distinguishing an abscess from other fluid collections such as cysts. The character of the fluid may be variable, depending on the content of the abscess. Collections that are completely fluid will appear as anechoic (black) areas, whereas areas with more solid components will appear to have “internal echoes” within the collections ( Fig. 37.US5 ). Once a focal fluid collection has been located, it can be evaluated in detail to determine the overall size and depth from the surface. Peritonsillar abscesses appear as rounded hypoechoic (dark gray) to anechoic (black) collections of variable size. In addition to confirming the presence of an abscess, the location and depth of the carotid artery can also be judged before an attempt at aspiration.

References:

• 1. Tayal VS, Hasan N, Norton HJ, et. al.: The effect of soft-tissue ultrasound on the management of cellulitis in the emergency department,. Acad Emerg Med 2006; 13: pp. 384-388.
• 2. Blaivas M, Theodoro D, Duggal S: Ultrasound-guided drainage of peritonsillar abscess by the emergency physician. Am J Emerg Med 2003; 21: pp. 155-158.
• 3. Lyon M, Blaivas M: Intraoral ultrasound in the diagnosis and treatment of suspected peritonsillar abscess in the emergency department. Acad Emerg Med 2005; 12: pp. 85-88.

Laboratory Findings

A complete blood count (CBC), blood cultures, and Gram stain are not standard or required for the treatment of straightforward cutaneous abscesses in the ED. Recommendations for culturing abscesses encountered in the ED are confusing and clinical practice varies. Firm recommendations for the emergency clinician are difficult to standardize, partly because of insufficient data but also because the recommendations promulgated are not confined to ED abscess treatment. In addition, “complicated” and “uncomplicated” criteria are somewhat arbitrary. Traditionally, culturing the contents of a readily drainable cutaneous abscess was not indicated, nor standard. It simply provided no additional useful information to the clinician under most circumstances. Many clinicians still forgo routine culturing, even in the CA-MRSA milieu. Currently, there is no agreement concerning routine culturing, and reasonable arguments can be made for a culture or no-culture approach to most abscesses treated in the ED. The authors support selective, not routine culturing but acknowledge that some now consider cultures to be indicated for all abscesses drained in the ED. A culture will potentially identify an unusual or resistant organism, especially if I&D is not curative. Culture will also permit identification of antibiotic susceptibility and assist in customization of antibiotic therapy. Cohort results also provide a framework for local epidemiology and resistance patterns. Culturing the abscess contents will distinguish between MRSA and nonresistant abscesses and will provide useful sensitivity information when managing complicated cases.

Culturing should be performed for recurrent, unusual, or atypical abscesses. This information could be useful if the patient responds poorly to initial surgical drainage, if secondary spread of the infection occurs, or if bacteremia develops. It also appears prudent to obtain cultures from abscesses and other purulent skin and soft tissue infections in patients already taking antibiotics, in immunosuppressed patients, in those with signs of systemic illness, in patients who have not responded adequately to initial treatment, if there is concern for a cluster or outbreak of infection, or in patients with severe local infection. Severe local infection can be defined as an abscess larger than 5 cm in diameter, multiple lesions, or extensive surrounding cellulitis. However, the degree of surrounding cellulitis qualifying as “extensive” is ill defined.

When obtaining a specimen for culture, the most accurate and complete culture results will be obtained if one aspirates pus with a needle and syringe before I&D. The material should be cultured for aerobic and anaerobic bacteria. Most clinicians, however, still culture free-flowing purulent material obtained with a cotton swab during I&D. For uncomplicated ED abscesses, this culture technique is standard and adequate to isolate aerobic organisms, including MRSA. A “sterile” culture from a specimen collected with a standard cotton swab after incision is frequently the result of improper anaerobic culture technique. In selected patients, such as immunocompromised hosts or IV drug users, isolation of possible anaerobic organisms by needle aspiration with a syringe through appropriately cleaned (e.g., with chlorhexidine) skin before I&D will enhance the results of culture and can add information that may be clinically relevant to subsequent therapy. There is a general misconception that foul-smelling pus is a result of E. coli . This foul odor is actually caused by the presence of anaerobes; the pus associated with E. coli is odorless.

The discovery of solid or suspicious material in an abscess should prompt histologic evaluation because a malignancy may mimic cutaneous abscesses (see Fig. 37.2 A and B ).

The majority of patients with an uncomplicated cutaneous abscess will have a normal CBC and will not experience fever, chills, or malaise. Therefore, in the absence of extenuating circumstances, it is not standard to analyze blood from patients with typical cutaneous abscesses because laboratory test results do not lead to a specific therapeutic path. An abscess may produce leukocytosis, depending on the severity and duration of the purulent process; however, the presence or absence of leukocytosis has virtually no diagnostic or therapeutic implications. Bacteremia may occasionally be manifested as a peripheral abscess resulting from septic emboli, and it usually produces clinical characteristics dissimilar to those associated with simple cutaneous abscesses. A cutaneous abscess itself rarely produces bacteremia.

Gram stain is neither indicated nor standard in the care of uncomplicated simple abscesses. However, patients who appear “toxic” or immunocompromised and those who require prophylactic antibiotics (see the section on Prophylactic Antibiotics later in this chapter) may benefit from Gram stain in addition to cultures. Gram stain results have been shown to correlate well with subsequent culture results, so in compromised hosts the test can be used to direct the choice of antibiotic therapy. Anaerobic infections should be suspected when multiple organisms are noted on Gram stain, when a foul odor is associated with the purulence, when free air is noted on radiographs of the soft tissue, and when no growth is reported on cultures.

Indications for and Contraindications to I&D

Surgical I&D is the definitive treatment of a soft tissue abscess ; antibiotics alone are often inadequate. Drainage of a suppurative focus generally results in marked resolution of the symptoms in most uncomplicated cases. In the initial stages, only induration and inflammation may be found in an area destined to produce an abscess. Premature incision, before localization of pus, will not be curative and may theoretically be deleterious because extension of the infectious process and, rarely, bacteremia can result from manipulation. In some cases, the application of heat to an area of inflammation may ease the pain, speed resolution of the cellulitis, and facilitate the localization and accumulation of pus. Nonsurgical methods are not a substitute for surgical drainage and should not be continued for more than 24 to 36 hours before the patient is reevaluated.

Prophylactic and Therapeutic Antibiotic Therapy

The utility of antibiotics remains unproven for prophylaxis against and for treatment of uncomplicated and adequately drained cutaneous abscesses in immunocompetent hosts. For simple abscesses, I&D alone is likely to be quite adequate and curative, even if the causative organism is MRSA. Routine administration of antibiotics after I&D is not currently standard, although the topic is subject to ongoing investigations. Simply stated, drainage alone for uncomplicated abscesses is usually curative. Furthermore, the use of antibiotics may in fact be harmful. Antibiotic misuse has also been shown to complicate resistance patterns, both in general and in specific patients.

Even though no randomized controlled trial data definitively demonstrate the need for antibiotic therapy in conjunction with I&D of uncomplicated cutaneous abscesses in healthy, immunocompetent patients (without the specific types of valvular heart disease discussed later), there is strong consensus for antibiotic treatment of abscesses associated with the following conditions: severe or extensive disease (e.g., involving multiple sites of infection) or rapid progression in the presence of associated cellulitis, signs and symptoms of systemic illness, associated comorbid conditions, immunosuppression, extremes of age, abscess in an area difficult to drain (e.g., face, hand, and genitalia), associated septic phlebitis, and lesions that are unresponsive to I&D alone.

Antibiotic use is indicated for abscesses with associated severe cellulitis (this term is not well defined and generally is a clinical judgement) and those with purulent cellulitis . Purulent cellulitis is defined as cellulitis associated with purulent drainage or exudate in the absence of a drainable abscess. Purulent cellulitis is usually caused by MRSA. Nonpurulent cellulitis, defined as cellulitis with no purulent drainage or exudate and no associated abscess, is usually due to β-hemolytic streptococci. Empirical therapy for MRSA, pending culture results, is recommended for patients who have failed non-MRSA treatment, for those with a previous history of or risk factors for MRSA, and for those with severe infection or systemic signs and symptoms. Empirical therapy for infection with β-hemolytic streptococci is likely to be unnecessary under these circumstances.

The duration of therapy for skin and soft tissue infections has not been well defined, although no differences in outcome were observed in adult patients with uncomplicated cellulitis receiving 5 versus 10 days of therapy in a randomized, controlled trial. In the Food and Drug Administration licensing trials for complicated skin and soft tissue infections, patients were typically treated for 7 to 14 days. However, in the outpatient setting of uncomplicated infections, 3 to 5 days of antibiotic therapy is reasonable but should be individualized on the basis of the patient's clinical condition and response to treatment. Accordingly, a return wound inspection or primary care follow-up is an important component of the care plan.

Patients with a history of IVDU who have an abscess and fever require parenteral antibiotic therapy after blood has been drawn for culture, until bacterial endocarditis can be ruled out. Additionally, patients who have extensive cellulitis or are clinically septic require immediate IV antibiotics, as well as aggressive surgical drainage of any significant abscess. By administering IV ampicillin/sulbactam (2 g/1 g) every 6 hours, Talan and colleagues achieved 100% eradication of pathogens from major abscesses in hospitalized IV drug users and non–drug users.