Prevalence of community-associated methicillin-resistant Staphylococcus aureus colonization in men who have sex with men

INTRODUCTION

Infections attributable to strains of community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA) are being increasingly described among men who have sex with men (MSM). ^1,2 Although much of the published literature thus far has focused on MSM who are HIV-positive, recent data confirm that the spread of CA-MRSA among MSM can occur independently of HIV infection and may be sexually transmitted in this population. ³ Risk factors for infection with CA-MRSA have been analysed in various studies of HIV-positive MSM, and have included high-risk sexual behaviours, use of crystal methamphetamine and recent history of sexually transmitted disease. ^1,2,4

However, there are few data describing the potential role of asymptomatic colonization in propagating the continued spread of CA-MRSA among MSM. Such information is important to capture, as the possibility of widespread rectal or nasal carriage of CA-MRSA among MSM would prompt and inform the development of relevant prevention and decolonization protocols. In addition, as most observed infections with CA-MRSA in MSM involve the buttocks or perianal region, the possibility of widespread rectal carriage contributing to these infections is not understood and requires clarification. The purpose of this cross-sectional study was to determine the prevalence of CA-MRSA colonization among a large cohort of MSM followed at two inner city family practice clinics. Outbreaks of infection with CA-MRSA in MSM had occurred in the year preceding the onset of the study at both clinics, and new infections were continuing to be observed at both sites at the time of study initiation.

METHODS

RESULTS

A total of 500 subjects were enrolled in the study, of whom 298 (59.6%) were HIV-positive. Demographic characteristics of the cohort are summarized in Table 1. The median age of the entire cohort was 43 years (interquartile range [IQR] 37, 49). Seventy-one (14.2%) participants had received antibiotic therapy within three months preceding the study. The median viral load and CD4+ cell count of the HIV-positive participants were 1.69 log₁₀copies/mL (IQR 1.69, 2.83) and 420 cells/mm³ (IQR 280, 650), respectively. A total of 181 (60.7%) HIV-positive participants had a viral load of <50 copies/mL, and 36 (12.1%) were receiving daily TMP-SMX for Pneumocystis jirovecii disease prophylaxis.

Of the 500 screened patients, 8 (1.6% [95% CI: 0.6–2.6%]) had MRSA cultured from either the rectum (n = 2), the nares (n = 3) or both sites (n = 3). All isolates met the predefined criteria for being community-associated. Four of the eight patients from whom CA-MRSA was cultured were HIV-positive. The median number of sexual partners in the three months preceding the study was significantly higher among participants colonized with CA-MRSA relative to those who were not (7 [IQR 5, 12] vs. 2 [IQR 1, 4]; P = 0.0091). There were no significant differences observed between the groups with respect to other variables examined, including recent (i.e. within three months) treatment for a sexually transmitted disease (P = 0.09), antibiotic use (P = 0.09) and engagement in high-risk sexual behaviours (P = 0.15).

Isolates were uniformly sensitive to vancomycin, rifampin, mupirocin, fusidic acid and tetracycline. Resistance to TMP-SMX was observed in one HIV-positive participant despite a lack of recent exposure to this agent. Clindamycin and erythromycin resistance were observed in two and 10 of the 11 isolates, respectively.

Based on PFGE patterns, nine of the 11 isolates were identified as strains of USA300 (CMRSA-10), and a further isolate was identified as USA400 (CMRSA-7). The remaining isolate had a PFGE fingerprint pattern that was not described in the national MRSA database. Ten of the 11 isolates were PVL-positive, and all were found to carry the SCCmec type IVa gene.

DISCUSSION

The prevalence of CA-MRSA nasal or rectal carriage in our cohort of MSM was 1.6%, implying that colonization with this organism has not occurred in this population to a significant extent. These findings are similar to those attained in surveillance studies of other community dwelling populations and are mirrored by the pooled CA-MRSA colonization prevalence of 1.3% attained by meta-analysis of studies conducted among healthy community members. ^11,12 Widespread colonization with CA-MRSA is therefore unlikely contributing to the ongoing nature of the outbreaks of purulent skin and soft tissue infections being observed in the MSM population in the clinic catchment area. It is therefore more likely that new infections with CA-MRSA in our MSM clients are occurring secondary to direct skin-to-skin contact with an infected person or by indirect transmission mediated by contaminated clothing, equipment or other fomites. The colonizing strains observed in our cohort were consistent with those most frequently cited as being the aetiologic agents of CA-MRSA-related infections. ^1–3

Our study adds to the existing small body of literature examining CA-MRSA colonization in MSM. In the only other study looking at a cohort comprised exclusively of MSM, seven of 158 participants had a nasal culture that was positive for CA-MRSA. ¹⁴ Rectal carriage was not assessed by these investigators. Given that five of 11 isolates of CA-MRSA in our study were cultured from rectal swabs, future studies addressing the question of CA-MRSA prevalence among MSM should include collection and culture of rectal swabs, as the sole reliance on nasal swabs may underestimate the prevalence of colonization in this population. In a separate study performed at an urban outpatient clinic, 15 of 279 (5.4%) patients had MRSA cultured from the nares, 13 of whom were MSM. However, although the majority of patients enrolled in the study were MSM, this was not a study performed exclusively in this population. In addition, it is unclear what criteria were in place to ascertain whether strains were community-derived or health-care associated. ¹⁵ Finally, an additional published study examined the prevalence of nasal MRSA colonization in 162 HIV-positive outpatients, 144 of whom were men. ¹⁶ A colonization rate of 6% was observed in this cohort, although it is unclear if these isolates met the definition for being community-derived, and whether the men were MSM.

Our study has several strengths, including the large sample of participants enrolled from two community clinics, use of predefined criteria for determining whether isolates were community-associated, molecular typing of strains and assessment of both nasal and rectal carriage. However, there are several limitations to our study. As both clinics had experienced outbreaks of CA-MRSA infections among MSM in the year preceding the initiation of this study, infection control measures and educational initiatives directed towards the community may have resulted in a decreased rate of colonization relative to a period preceding the identification of CA-MRSA as a significant public health concern among this cohort. In addition, we were statistically limited in our ability to perform any multivariable regression analysis to assess the role of healthcare-associated or demographic variables as risk factors for CA-MRSA colonization, given the small number of patients with a positive culture. Similarly, the finding of significantly more sexual contacts among participants who were colonized may be spurious given the small number of outcomes observed and large number of comparisons drawn. Still, this finding cannot be dismissed entirely, given prior evidence identifying sex with multiple partners as a risk factor for CA-MRSA skin infections in MSM. In addition, the possibility of non-response bias cannot be excluded as refusal rates and reasons for refusal were not recorded in the study. Finally, it is possible that self-collection of rectal swabs may have affected the results, although patients were instructed on the proper collection technique and a study coordinator was available if problems with this procedure arose. No participant was required to repeat the rectal swab due to incorrect specimen collection.

In summary, we observed a low prevalence of colonization with CA-MRSA among a large cohort of MSM seen at two inner city family practice clinics. These data suggest that new infections are not occurring as a result of widespread asymptomatic nasal or rectal colonization in this community. In addition, our data do not support routine screening of MSM for asymptomatic carriage of CA-MRSA. Instead, strategies aimed at disrupting the continual spread of this organism within the MSM community would likely be most successful if emphasis was placed on basic infection control measures in venues such as bath-houses, gyms or other centres where contact with potentially contaminated surfaces can occur.