The increase in infections caused by microorganisms resistant to conventional antimicrobials represents a major therapeutic limitation in the hospital setting.

Soon after the introduction of penicillin in 1941 the first strains of penicillin-resistant Staphylococcus aureus were isolated. Hospital strains of methicillin (oxacillin)-resistant S aureus appeared 2 decades later, and subsequently became one of the main bacterial causes of nosocomial infections. However, since the 1980s S aureus has also been detected in non-hospital-acquired infections,1 and its incidence has grown exponentially in many countries, including the United States, where it is the leading cause of community-acquired skin and soft tissue infections in several states.2

Community-acquired methicillin-resistant S aureus (CA-MRSA) differs from its nosocomial counterpart in its clinical, epidemiological, and microbiological characteristics (Table 1),3–7 and thus requires radically different therapeutic management in most cases.

Given that CA-MRSA predominantly affects skin and soft tissue, we reviewed 11 cases diagnosed in our hospital in order to describe the main clinical features of the disease and the correlation with available demographic and microbiological data. Based on our findings, we propose several specific clinical criteria to improve the initial empirical treatment of skin and soft tissue infections in outpatients.

This is a descriptive, observational, retrospective study of adult patients with a microbiological diagnosis of skin and soft tissue infection caused by CA-MRSA. The data were collected between June 2009 and December 2011 from the skin infection database of the dermatology department at the Hospital General Universitario Gregorio Marañón in Madrid, Spain.

All cases had been microbiologically confirmed by culture followed by antibiogram and molecular analysis.

The clinical variables studied were age, sex, clinical presentation (lesion type, presence of pus, necrosis, and location), the course of the lesions, treatments received, and the response to treatment. We analyzed the origin of the patients, the presence of recurrent skin infections in the patient's family, risk practices (contact sports or activities, shared use of personal hygiene items, membership of recruiting groups or institutions), contact with the health sector (employment, previous hospital admissions), and contact with domestic or farm animals.

Data for the patients studied are summarized in Table 2.

Of the patients studied, 91% (10/11) were younger than 40 years, and the male to female ratio was 1:1.2. The individuals were healthy, immunocompetent patients with no prior hospital admissions and no contact with the healthcare sector through either employment or family.

Eight (73%) of the 11 patients presented with abscesses (Figs. 1-4); half of these cases (4/8) involved perilesional cellulitis, and 63% (5/8) presented abscesses on exposed areas of the limbs. Surgical drainage removed small amounts of purulent material in 37% (3/8) of cases, and none in the remainder (63%; 5/8). Tissue necrosis, forming a cavity within the subcutaneous tissue, was the main feature of all lesions.

Figure 1.

Abscess and perilesional cellulitis.

(0.13MB).

Figure 2.

Characteristic appearance of abscess after drainage: necrosis and absence or scarcity of purulent exudate.

(0.19MB).

Figure 3.

Erythematous-violaceous border and fibrinous center after spontaneous drainage of an abscess on the leg.

(0.17MB).

Figure 4.

Abscess next to a tattoo on a young woman.

(0.15MB).

Primary follicular involvement (staphylococcal folliculitis and anthrax) was observed in 2 patients, and recurrent paronychia in another. Due to the nature of the lesions, it was not possible to clearly establish the amount of purulent exudate in these 3 patients.

Thirty-six percent (4/11) of the patients were Spanish, 45% (5/11) were Latin American, and 18% (2/11) were Polish. Possible risk factors associated with the development of infection included sports (2 patients, 18%), dance (2 patients, 18%), and noninvasive cosmetic procedures (1 patient, 9%).

Two patients had a history of recurrent, uncomplicated skin infections.

All cases had been confirmed by culture followed by an antibiogram and molecular analysis. The bacterial strains isolated in 64% (7/11) of the cases studied were resistant to erythromycin. Levofloxacin resistance was also detected in 1 case (1/7) and inducible clindamycin resistance in another (1/7), both of which were confirmed by D-test. Genes encoding Panton-Valentine leukocidin (PVL) were expressed in all staphylococcal isolates analyzed (except 1 case in which no analysis was performed as the patient came from another health care center).

Of the patients analyzed, 55% (6/11) had been previously treated with amoxicillin-clavulanate with no clinical response. Of these, 67% (4/6) experienced treatment failure after a second round of antibiotic therapy with β-lactam (cefuroxime/cloxacillin) or erythromycin.

All abscesses (8/8) were surgically drained and antibiotics administered in 88% of cases. Ciprofloxacin and cotrimoxazole were the drugs of choice in 73% of patients (8/11). In 1 case treatment with oral clindamycin was discontinued due to gastrointestinal intolerance (diarrhea). Two patients with uncomplicated abscesses were treated with intravenous vancomycin as an alternative to β-lactam antibiotics. None of the patients presented systemic complications or myofascial involvement.

CA-MRSA is emerging as a causative agent of skin and soft tissue infections. Its incidence remains low in Spain, in contrast to other industrialized countries, where it constitutes one of the main community-acquired bacterial pathogens.8,9

The epidemiological data presented here are similar to those reported in the scientific literature: those affected are young, healthy patients who engage in contact sports/activities or come from endemic areas such as Central or South America.9

Often, patients present single abscesses, located in areas of skin that are exposed or subject to contact during sports/activity.10 Tissue necrosis is the main characteristic of these abscesses and very little pus is produced. A study evaluating the presence of eschar and perilesional cellulitis as clinical predictors in abscesses caused by CA-MRSA reported a sensitivity of 92% and a positive predictive value of 94%.11 These findings may be explained by the cytotoxic effects that PVL is thought to exert on neutrophils and epidermal cells, and by the release of proinflammatory factors secondary to granulocyte lysis.12 However, further studies are needed to confirm the role of this toxin in the pathogenesis of infections caused by CA-MRSA.

The differential diagnosis of these lesions should include pyoderma gangrenosum, spider bites,13 loxoscelism,14 and cutaneous leishmaniasis.

Other common cutaneous manifestations have been described, including folliculitis, carbuncles (recurrent infections), and impetigo (especially in children).15

CA-MRSA strains have microbiological characteristics that differentiate them from nosocomial strains; they are usually resistant only to β-lactam antibiotics and carry the genes that encode the PVL subunits LukF-PV and LukS-PV. However, depending on the source of the strain CA-MRSA may be also resistant to other antibiotics, as is the case with the predominant clone in the USA, USA300 (ST8-MRSA-IV), which is highly resistant to macrolides.7 Therefore, where possible, samples should be taken in all cases for microbiological analysis of the lesions.

In this series 91% of patients received antimicrobial therapy, combined with surgical abscess drainage. However, some authors claim that lesion drainage and daily local treatment is sufficient in cases of extrafacial abscesses of less than 5cm in diameter in nondiabetic, immunocompetent patients without myofascial involvement.16–18Table 3 summarizes the various relevant therapeutic modalities for the treatment of infections caused by CA-MRSA.

Since this bacterium mainly causes uncomplicated skin infections, treatment does not usually require hospitalization. However, cases of septic arthritis, pyomyositis, fascitis,20 and necrotizing pneumonia21 have been described. Necrotizing pneumonia21 should be considered in patients with uncomplicated skin infections and past or simultaneous symptoms of respiratory infection.22 These cases always require antimicrobial treatment and close clinical observation.

The determination of carrier status by nasal swab is not recommended as a regular procedure.16 Nasal colonization is often low in patients with lesions that are not associated with community-acquired MRSA outbreaks (the majority of cases in Spain), and the bacteria may be detected in other locations such as the inguinal/perineal area, the axillary folds, the rectum, or the pharynx.23 There is no consensus on the measures necessary to achieve decolonization of this bacteria,24 although some authors have proposed showering with antiseptic soap solutions (chlorhexidine or povidone iodine, although use of the latter is limited by its staining of the skin and the risk of local sensitization or irritation25), or using topical intranasal (mupirocin) and systemic (cotrimoxazole) antibiotics,26,27 with varying results.

Uncomplicated infections of the skin and soft tissue are one of the main reasons for visits to dermatological practices and emergency health services. The epidemiology of these infections has changed in recent years, with the appearance of pathogens that were hitherto considered exclusively nosocomial. CA-MRSA represents a paradigm in dermatological practice, as its main pathogenic targets are the skin and skin adnexa. The collection of samples for culture and subsequent antibiogram remains the gold standard in diagnosis. However, while awaiting the results of microbiological tests, we believe it is essential to individualize the initial empirical antibiotic treatment and to include this pathogen in the bacterial spectrum covered.

Although the small number of cases analyzed and the retrospective nature of our study do not allow us to generalize based on the results obtained, we propose that the use of β-lactam antibiotics as an initial therapeutic measure should be avoided in young, healthy patients with cutaneous abscesses with a significant necrotic component and scant purulent exudate, with or without cellulitis. Similarly, in cases of recurrent skin infections, family outbreaks of abscesses, folliculitis, and carbuncles, or treatment failure after 1 or more cycles of β-lactam antibiotics, CA-MRSA should be considered as a potential causative pathogen.

The authors declare that they have no conflicts of interest.