Incidence and Persistence of Staphylococcus aureus Nasal Colonization in a Community Sample of HIV-Infected and -Uninfected Drug Users
- Maureen Miller1,
- Christian Cespedes2,
- Meera Bhat2,
- Peter Vavagiakis4,
- Robert S. Klein5, and
- Franklin D. Lowy2,3
- 1Department of Epidemiology and Biostatistics, School of Public Health, New York Medical College, Valhalla
- 2Division of Infectious Diseases, Department of Medicine, New York, New York
- 3Department of Pathology, Columbia University, College of Physicians <mp; Surgeons, New York, New York
- 4Panna Technologies, Montefiore Medical Center and the Albert Einstein College of Medicine, New York, New York
- 5Division of Infectious Diseases, Department of Medicine and the Department of Epidemiology and Population Health, Montefiore Medical Center and the Albert Einstein College of Medicine, New York, New York
- Reprints or correspondence: Dr. Maureen Miller, Dept. of Epidemiology and Biostatistics, School of Public Health, New York Medical College, Valhalla, NY 10595 (maureen_miller{at}nymc.edu).
Abstract
A longitudinal study of 282 community-based drug users was conducted from February 1999 through September 2000. Both the incidence (15.0 cases per 100 person-years at risk; 95% confidence interval, 10.2–20.7 cases per 100 person-years at risk) and persistence of Staphylococcus aureus carriage were increased among human immunodeficiency virus (HIV)–seropositive individuals.
Although little is known about the correlates of Staphylococcus aureus colonization, even less is known about the predictors of incident S. aureus colonization in the community. Recently documented increases in community-associated methicillin-resistant S. aureus (MRSA) infection suggest that there is already a reservoir of methicillin-resistant strains [1,2–3]. The use of antimicrobials as both prophylactic and therapeutic agents for opportunistic infections in HIV-infected individuals and for cutaneous infections in injection drug users may increase the likelihood of nasal carriage of antimicrobial-resistant S. aureus [4] and suggests that these groups may be “hidden” contributors to a community-based reservoir.
We prospectively observed a subsample of participants from a large, well-characterized, community-based sample of current and former drug users [5]. In addition to monitoring the incidence, persistence, and DNA profiles of S. aureus colonizing strains, we examined potential predictors of incident and persistent colonization.
Methods. From February 1999 through March 2000, 500 respondents were recruited; 282 persons (56%) participated in at least 1 follow-up interview, with a mean duration of follow-up (±SD) of 11 ± 4 months. After providing informed consent, respondents completed a standardized interview that assessed sociodemographic characteristics, drug use, and medical history. Limited physical examinations were performed. HIV serostatus and CD4 lymphocyte counts were determined, and anterior nasal culture specimens were obtained (Becton Dickinson Culturette Systems).
Reevaluations were scheduled every 6 months. Incident S. aureus colonization is defined as the first positive swab result after a negative swab result at a previous visit. The study was approved by the institutional review boards of Columbia University and Montefiore Medical Center (New York, NY).
The swab specimens obtained from the subjects' anterior nares were first cultured on Mannitol salt agar (Becton Dickinson). After a 48-h incubation period at 37°C, positive colonies from each sample were isolated onto 5% sheep blood agar plates (Becton Dickinson), and a single colony was selected for additional analysis. S. aureus species identification was confirmed by coagulase and protein A detection kit (Murex StaphAurex). Persistent colonization was defined as a minimum of 2 specimens testing positive for the presence of S. aureus. Isolates were classified as consistently negative, transiently colonized (defined as 1 specimen testing positive for S. aureus), or persistently colonized (defined as 67%–100% of all specimens testing positive for S. aureus).
All positive isolates were compared by PFGE [6, 7] and analyzed using BioNumerics software, version 4.00 (Applied Maths), and by staphylococcal protein A (spa) typing [8]. Isolates identified as methicillin resistant by PCR [9] were further analyzed for their specific staphylococcal chromosomal cassette (SCC) mec type, using a multiplex PCR assay [10], and were validated using primers for the pls region of the SCC mec complex (F-CAACAGGAGCAGATGGCTG, R-GGTGAGCCTGGTTCTAGAG), as well as the ccr gene complex primers [11]. H. De Lencastre provided control strains containing the different SCC mec types.
For univariate analyses, we used χ2 and Student's t tests to examine differences between S. aureus variables and sociodemographic characteristics and other potential risk factors. Additional analyses were conducted for the subgroup of HIV-infected individuals. Incidence rates were calculated using person-years of follow-up from baseline to the midpoint between visits in which the last negative result occurred and the first positive result occurred and, for persons who remained negative for S. aureus, until the final interview date. The rates of S. aureus colonization incidence are reported as cases per 100 person-years at risk (PYAR). Risk factor data for incident colonization were derived from the first visit with a positive test result; risk factor data for the participants who had persistently negative test results were derived from the most recent follow-up visit. Cox proportional hazards ratios and 95% CIs summarize the relationship between incidence rates by statistically significant predictors. Risk factor data for persistent S. aureus colonization were based on baseline data or on the interview associated with the first positive test result for participants with incident and persistent colonization. Multivariable logistic regression analyses to control for biologically plausible factors associated with S. aureus were conducted. Two-tailed P values were statistically significant at P <t; .05; Fisher's exact test was used to determine P values when expected cell counts were <t;5. SAS, version 9.1.3 (SAS Institute), was used for all statistical analyses.
Results. Sixty-five (23%) of 282 respondents were colonized with S. aureus at the baseline interview, 50 (18%) of 282 at the first follow-up interview, 21 (21%) of 98 at the second interview, and 5 (56%) of 9 at the third interview. The mean age (±SD) of the cohort was 41 ± 7 years (range, 23–68 years), 59% were male, and approximately two-thirds were Hispanic. table 1 presents data that examine potential risk factors for S. aureus colonization. Only HIV seropositive status was associated with S. aureus colonization (P = .03). Multivariable analyses confirmed HIV infection as the unique correlate of S. aureus colonization (data not shown).
Table 1
The prevalence of Staphylococcus aureus colonization and potential risk indicators among 282 current or former drug users at the baseline interview.
Thirty-one (14%) of 217 initially S. aureus-negative respondents became colonized with S. aureus. The incidence rate of S. aureus colonization was 15.0 cases per 100 PYAR (95% CI, 10.2–20.7 cases per 100 PYAR). HIV-seropositive individuals were twice as likely as HIV-seronegative individuals to become colonized with S. aureus (hazard ratio, 2.2; 95% CI, 1.1–4.6); the incidence rates were 24.4 cases per 100 PYAR (95% CI, 10.2–20.7 cases per 100 PYAR) and 8.8 cases per 100 PYAR (95% CI, 4.4–14.8 cases per 100 PYAR) for HIV-seropositive and HIV-seronegative respondents, respectively.
Thirty-six (38%) of the 96 respondents colonized with S. aureus were persistently colonized, and most (67%) of these respondents were colonized with the same strain over time. The only predictor of persistent colonization was HIV-seropositive status (P = .002).
MRSA was detected in 9 individuals during the first 3 interview periods. Seven of these 9 individuals were infected with HIV. Two HIV-infected individuals had incident colonization with MRSA.
One hundred thirty-one subjects (46%) were infected with HIV. No incident HIV infections were identified. Approximately one-third of HIV-infected individuals received HIV-related therapy. Neither antiretroviral therapy nor antimicrobial use were associated with prevalent, incident, or persistent S. aureus colonization. Among HIV-infected individuals, the median CD4 cell count was 252 cells/µL; CD4 cell count was not associated with prevalent, incident, or persistent S. aureus colonization.
Analysis of the PFGE profiles for the isolates failed to reveal a predominant clonal type in the cohort. of the 93 S. aureus-positive specimens available for typing, 14 (15%) were spa type 7. Twelve (86%) of 14 spa type 7 specimens were from HIV-infected individuals (P = .003). No individual was colonized with spa type 7 in >1 specimen. SCC mec typing revealed that 4 of the 8 individuals for whom specimens were available were colonized with SCC mec type IV; 3 of these 4 individuals were colonized with persistent MRSA.
Discussion. These data are, to our knowledge, the first to provide evidence of high rates of incident S. aureus colonization in a high-risk, community-based sample. Moreover, HIV serostatus was significantly and uniquely associated with S. aureus incidence and persistence and with MRSA colonization. No therapeutic or clinical factors of HIV infection were found to increase the risk of colonization. This suggests that, although HIV infection increases vulnerability through an unidentified biological or epidemiological mechanism, risk may be occurring at a level other than the individual one.
In terms of possible underlying mechanisms of S. aureus transmission, colonization risk may be environmental, such as a high background prevalence of S. aureus (e.g., high levels of colonization among the health care providers of HIV-infected individuals [4, 9] and unhygienic drug-use environments [12, 13]). Alternatively, transmission risk could be related to the social networks of HIV-infected individuals (e.g., HIV-infected individuals may socialize in the community, which could provide greater opportunity for exposure to disease [14]).
Although there was no predominant clonal type that suggested unique colonizing advantage, there was a replication of the association between HIV status and colonization with S. aureus spa type 7 multilocus sequence type 8 [4, 15,16–17]. The basis for this unique association is unclear but suggests a biologic rather than solely an epidemiologic explanation. Sequence type 8 is of interest, because it has been shown to harbor all 4 SCC mec types. It is, therefore, especially receptive to the SCC mec element. The basis for the unique association of sequence type 8 with HIV-infected patients might include the presence of particular adhesins that can facilitate adherence to epitopes expressed or made more accessible on host cells by HIV infection. Demonstration of these isolates in the community illustrates the potential for their persistence and spread in an otherwise-hidden reservoir.
Much of the data are based on self-reported information; therefore, associations should be interpreted with caution. Self-reported data may underreport individual practices, particularly those related to illicit drug use, because this recruited sample of current and former drug users has valid reasons for not reporting the full extent of their drug use [5]. In addition, recent drug use was defined as having occurred in the past 6 months and may not reflect current practices.
This study demonstrates strong associations between HIV infection and incident S. aureus colonization; the association with HIV infection likely explains the high incidence of S. aureus infection in the HIV-infected population. The basis for this association, whether it relates to environmental factors or to underlying biologic factors associated with HIV infection, remains to be established.
Acknowledgments
Financial support. National Institutes of Health (DA15018 and DA11868) and Center for AIDS Research (5P30AI051519).
Potential conflicts of interest. All authors: no conflicts.
- Received January 5, 2007.
- Accepted March 21, 2007.
- © 2007 of the Infectious Diseases Society of America
References
This Article
- AbstractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 8.186
5-Yr impact factor: 7.898
For The Media
For Authors
- Services for authors
- Self-archiving Policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
