The prevalence of S. aureus colonization in our HIV-infected population is consistent with rates reported for the general population [17]. However, only 3 individuals (corresponding to 2/190 nasal samples and 2/110 pharyngeal samples) were colonized by MRSA and none of the isolates belonged to a typical CA-MRSA lineage. Our data contrast with most reports from the USA, where a significant association between HIV infection and MRSA colonization has been well documented [3–7]. A recent meta-analysis assessing MRSA colonization prevalence and risk factors in HIV-infected individuals reported that 6.9 % of the population studied were MRSA carriers and that this rate rose to 8.8 % when only studies from North America were included [5]. Although data on MRSA carriage in HIV-infected individuals in Europe are scarce, lower prevalence rates have been reported (0 % to 2.8 %) [5, 9–12], and the findings in our series are consistent with reports from other European studies.
Several US studies have described an association between CA-MRSA colonization and risk factors such as sexual behavior, substance abuse, incarceration, and area of residence [4, 6, 7]. These risk groups were represented in our sample (Table 1).
We recently reported a series of MRSA infection in HIV-infected patients in Spain in which we detected a higher risk for MRSA infection among individuals with poorly controlled HIV infection and immigrants (mostly South Americans) [18]. Nevertheless, the overall prevalence was still lower than that reported for the USA. The absence of CA-MRSA colonization in HIV-infected individuals in our area could partly explain the low rate of CA-MRSA infection in this population.
Our study is the first to assess the prevalence of MRSA colonization in HIV-infected patients in Spain. However, it has some limitations. Although our sample is representative of our cohort of HIV-infected patients, the number of individuals studied is lower than in other series. Additionally, while MRSA colonization of extranasal areas such as the buttocks and the perianal, inguinal and axillary regions has been well documented [4, 7, 19, 20], the only extranasal location studied in our case was the pharynx, and paired nasal and pharyngeal samples were only available for a subset of 110 patients. Thus, the rate of MRSA colonization in our cohort might have been underestimated. The importance of extranasal colonization has been demonstrated for both hospital- and community-acquired MRSA [4, 7, 19, 20]. Recent studies in the community setting have observed that nasal-only screening could miss up to 51 % of MRSA colonized individuals [4, 19]. In the subset of patients with both nasal and pharyngeal samples in our series, MRSA was detected in both samples in one patient while the other one was an exclusive pharyngeal carrier. None of the patients had exclusive nasal colonization in this subgroup. Although we do not discount the importance of extranasal colonization (the colonization rate would have been 50 % lower if our study had been limited to nasal samples), the rate of MRSA colonization in our cohort was low.
We were unable to genetically characterize one MRSA isolate. However, PVL expression, which is characteristic in CA-MRSA strains in our area, was not detected, and the antibiotic susceptibility pattern, while unspecific, was indistinguishable from the dominant hospital-acquired MRSA lineage. Thus, the molecular characterization results for the available isolates, the absence of PVL production, and the antibiotic resistance patterns detected suggested healthcare-related acquisition in all cases.
