The epidemic of community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA) has spread rapidly both in the community [1] and in nosocomial settings [2]. Preliminary studies suggest that CA-MRSA may have colonization dynamics [3, 4] different from those of traditional MRSA strains, including having a disproportionately large effect on certain populations [5, 6].

Colonization with MRSA is felt to precede infection [7] with the anterior nares being the primary site of colonization [8]. However, colonization outside the nares—at the axilla, inguinal region, oropharynx, wounds, and gastrointestinal tract—occurs [9, 10]. Decolonization with mupirocin [11] or with mupirocin and chlorhexidine [12] is less successful and requires more attempts if >1 body site is colonized. Because colonized individuals are at increased risk of MRSA infection and can serve as a source of MRSA cross-transmission, the reduced effectiveness of decolonization among persons with higher MRSA colonization burden (ie, higher proportion of anatomic sites colonized per patient) has implications at both the patient and community level.

A 2003–2004 survey of the US population [13] observed an MRSA nasal colonization prevalence in the community of 1.5%; 19.7% of isolates were genotyped as USA300, the predominant CA-MRSA strain in the United States [14]. The low proportion of nares colonization with USA300 strains, despite the widespread CA-MRSA epidemic, raises the question of extranasal colonization and whether the true burden of USA300 MRSA colonization among the general population was underestimated. Outbreak investigations of USA300 MRSA infections often have not been able to detect nasal colonization at the time of infection and nasal culture results have not necessarily predicted who would develop CA-MRSA infection [15, 16]. This suggests either that colonization outside the nares plays a role in infection development or that the infections occur without, or with only transient, colonization—the “hit and run” theory [17].

CA-MRSA has had a disproportionately greater impact on certain patient populations. For example, in our experience patients infected with human immunodeficiency virus (HIV) have had >6-fold higher risk of CA-MRSA skin and soft-tissue infections (SSTIs) than have HIV-negative patients [5]. In addition, prevalence of nasal colonization with CA-MRSA is higher in HIV-infected patients [18] and HIV infection appears to be associated with persistent colonization [19]. However, risk for colonization, even among HIV-infected individuals, appears to be unevenly distributed [20].

A hypothesis to explain why certain community populations have increased risk of CA-MRSA infection is that persons within these populations have a higher colonization burden with USA300 MRSA, thereby facilitating spread of strains within the population. The objective of our study was to evaluate the epidemiology, prevalence, and colonization burden of CA-MRSA strains among HIV-infected and HIV-negative patients who were recently admitted to an acute care hospital. Given preliminary data demonstrating that current and former detainees are at increased risk of CA-MRSA, we included this population in our study.

Abstract

Background. The epidemic of community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA) has had a disproportionate impact on patients with human immunodeficiency virus (HIV).

Methods. We evaluated CA-MRSA colonization burden (number of colonized sites per total number sampled) among HIV-infected and HIV-negative inpatients within 72 hours of hospitalization. From March 2011 through April 2012, we obtained cultures from nasal and extranasal sites (throat, axilla, inguinal, perirectal, and chronic wound if present) and collected risk factor data.

Results. Of 745 patients (374 HIV-infected, 371 HIV-negative), 15.7% were colonized with CA-MRSA at any site: 20% of HIV and 11% of HIV-negative patients (relative prevalence = 1.8, P = .002). HIV-infected patients had a higher prevalence of nasal, extranasal, and exclusive extranasal colonization as well as higher colonization burden. Perirectal and inguinal areas were the extranasal sites most frequently colonized, and 38.5% of colonized patients had exclusive extranasal colonization. Seventy-three percent of isolates were identified as USA300. Among HIV-infected patients, male sex, younger age, and recent incarceration were positively associated whereas Hispanic ethnicity was negatively associated with higher colonization burden. Among HIV-negative patients, temporary housing (homeless, shelter, or substance abuse center) was the only factor associated with higher colonization burden. Predictors of USA300 included HIV, younger age, illicit drug use, and male sex; all but 1 colonized individual with current or recent incarceration carried USA300.