Abstract
Background
Persons living with Human Immunodeficiency Virus (PLWH) are disproportionately burdened with methicillin resistant Staphylococcus aureus (MRSA). Our objective was to evaluate prevalence and risks for MRSA colonization in PLWH.
Methods
Adults were recruited from Johns Hopkins University AIDS Service in Baltimore, MD, USA. A risk questionnaire and specimen collection from anatomical sites with culture susceptibility and genotyping were completed. Generalized estimating equation modeling identified MRSA colonization risk factors.
Results
Of 500 participants, most were black (69%), on antiretroviral therapy (ART) (87%), with undetectable viral loads (73.4%). Median CD4 count was 487 cells/mm3 (IQR: 316 – 676.5 cells/mm3). MRSA prevalence was 15.4%, predominantly from the nares (59.7%). Forty percent were nares negative yet colonized elsewhere. Lower odds for colonization were associated with recent sexual activity (AOR: 0.84, P<0.001) and ART (AOR: 0.85, P=0.011). Increased odds were associated with lower income (<$25,000 vs. >$75,000 AOR: 2.68, P<0.001), recent hospitalization (AOR: 1.54, P<0.001), incarceration (AOR: 1.55, P<0.001), use of street drugs or (AOR: 1.43, P<0.001 and skin abscess (AOR: 1.19, P<0.001).
Conclusions
Even with high MRSA prevalence, the proportion identified through nares surveillance alone was low, indicating the importance of screening multiple anatomical sites. Associations were not found with same-sex coupling or black race. MRSA prevention might be a benefit of ART in PLWH.
Keywords: MRSA, HIV, prevalence, risk, anatomical site
INTRODUCTION
Methicillin resistant Staphylococcus aureus (MRSA) continues to cause excess morbidity and mortality among persons living with Human Immunodeficiency Virus (PLWH). In the U.S., PLWH have substantially higher incidence of MRSA infections than the general population (12.3/1000 person years compared to 1 to 2/1000 person years)1 and MRSA remains a substantial reason for hospital admission.2 Metropolitan areas throughout the country have documented a substantial increase in MRSA infections,3-5 peaking in 2008, with an incidence five times greater in PLWH compared to HIV-uninfected persons within a large healthcare system.6 Despite recent reported declines in skin and soft tissue infection, PLWH continue to shoulder a disproportionate burden of disease.7
Key risk factors for MRSA colonization and/or infection have been identified as a result of these data and include substance abuse;8,9 high risk sexual practices in persons with greater numbers of sex partners, regardless of sexual orientation;10 and having a sexual partner with a known skin infection.10 Additional risks for MRSA infection among PLWH include male sex;7 incarceration history;7 lower CD4 counts;4,5,10 high viral load;4,11 recent hospital admission;12 β-lactam antibiotic use3; lack of cotrimoxazole prophylaxis;5,11 and known MRSA infection in the last 12 months.3
Despite recent attention to this issue, many questions related to the HIV/MRSA interface remain unanswered. To further understand risk and enable the improvement of this population's health and well being as well as limit the spread of MRSA within and among partners, this study assessed the overall prevalence of MRSA colonization at multiple body sites among PLWH and their primary partners at multiple body sites. In addition, we determined the risk factors that are associated with MRSA colonization.
METHODS
Study Setting and Participant Selection
To assess colonization prevalence among PLWH, we conducted a cross-sectional epidemiologic evaluation of MRSA among persons within the Johns Hopkins University AIDS Service (JHUAS). The JHUAS is a hospital-based practice that provides specialty care at the Moore Clinic on the Johns Hopkins Hospital campus in downtown Baltimore, Maryland, and at Green Spring Station (GSS) in Baltimore County. Urban Baltimore has a high incidence of HIV infection as well as MRSA colonization within outpatient populations.13,14 Greater than 50% of our clients reside in East Baltimore and > 75% within the city limits. The Moore Clinic follows an average of 2,000 clients annually including the uninsured, with a majority residing in Baltimore City. Most are African-American (77%) with major HIV transmission risks of intravenous drug use and heterosexual sex. Co-located services include viral hepatitis clinic, counseling, case management, social work, lab services, wound care and an outpatient pharmacy. The GSS medical office serves a smaller patient cohort (650) that is primarily white (64%), with a greater proportion residing in Baltimore's surrounding counties. The major HIV transmission risk at GSS is men who have sex with men (MSM). The GSS clinic also provides services to persons without HIV infection and does not accept the uninsured. There is an onsite laboratory and outpatient pharmacy. Many providers work at both the Green Spring Station and Moore Clinic locations.
Sample size calculations were informed by previous work showing a strong association between MRSA and previous or current abscess15 along with community prevalence data for Baltimore City. Based on the projected probability of MRSA among persons with abscess (0.30) and the probability of abscess in the entire cohort (0.18), a sample size of 500 was needed to obtain 90% power at alpha = 0.05 for a two-tailed test. This sample size would allow us to detect associations between having a current abscess and MRSA colonization.
Subjects were recruited from March 1, 2010 to June 30, 2010, with microbiologic analysis completed in April 2011. Eligible primary subjects were adult men and women ages 18 years and older, able to read and/or understand spoken English who receive care within the JHUAS. Potential participants were approached and screened consecutively during the recruitment period-- every 10th patient entering the Moore Clinic and all patients at Green Spring Station. We gathered specimens from sexual partners to examine similarities and differences between partners and MRSA colonization. Sexual partners were eligible for inclusion if referred by the index HIV positive subject, regardless of the partners HIV status. Participants were offered a $25 gift card for participation. The study was approved by the Johns Hopkins Medicine Institutional Review Board.
Collection of Clinical Specimens
A total of 6 swabs for men and 7 for women were collected, plus an additional swab for anyone with a wound. Anterior nares (N), axillary (A), throat (T), groin (G), perineum (P), rectal (R) swabs were obtained using BactiSwab II dual-headed culturettes (Remel, Lenexa, KS). The perineum is the area between the anus and scrotum or vulva. Vaginal swabs were collected on both women and post-operative transgendered women. Wound specimens were obtained from any patient presenting with an open or draining wound. Two trained registered nurses collected all patient swabs. All swabs were collected during the clinic visit at study enrollment transported at room temperature and stored at 5°C until processing within 24 hours of collection.
Risk Factor Evaluation
A survey instrument was designed with a set of questions used in a questionnaire previously implemented within the same population to assess factors related to a person's hospital and community risks for MRSA acquisition. Another 13 questions were added to assess behavioral risk factors, leading to a final 64-item questionnaire. The current study did not evaluate all 64 items. A study team member administered the questionnaire face-to-face during the clinic visit at study enrollment. Interviewers were trained in sexual history taking and pilot tested the survey instrument. Interviews were conducted in a private setting and confidentiality of responses was ensured to address the tendency for subjects to provide socially desirable responses. Both clinic populations are routinely screened for sexually history and sexually transmitted infections (STI) at each clinic visit. The time frame for any sexual activity or drug use was the 12 months prior to the interview. Medical records were reviewed by research nurses for clarity of HIV-related or sexual history, comorbidities, admission data and other self-reported information from the questionnaire. Information from medical records was used if discrepancies occurred with self-reported information.
Microbiologic Evaluation
Using standard culture methods, a swab was streaked onto CHROMagar-MRSA (BD Diagnostics, Inc.) [CHROM-MRSA], a selective medium, and then placed in Trypticase soy broth (TSB) with 6.5% NaCl. Gram-positive cocci that were catalase positive and latex agglutination positive were identified presumptively as S. aureus. The presumptive S. aureus isolates recovered from the enrichment broth were subbed and sent to the BD Phoenix™ Automated Microbiology System (Phoenix) (BD Diagnostics, Sparks, MD)16 for identification and susceptibility testing when the matching CHROM-MRSA agar plates were negative. All isolates growing on CHROM-MRSA were considered MRSA.
Staphylococcal genotyping was performed using the Ibis Staphylococcus Typing and Characterization kit (Ibis, Inc. Carlsbad, CA). This multiplex, broad based, molecular assay, performed in a microtiter plate format, uses primer sets that are capable of identifying S. aureus (tufB and nuc) and distinguishing MSSA from MRSA (mecA). It also incorporates primers that determine the presence of Panton Valentine Leukocidin genes (LukS & LukD, Toxic Shock Toxin 1 (TSST1), and high (mupA) and low level (ileS) mupirocin resistance. Strain characterization is performed by multi-locus sequence typing using a unique set of sequences from seven common housekeeping genes for S. aureus. The assay is based upon the principle of polymerase chain reaction-electrospray ionization mass spectrometry (PCR ESI-MS), and is performed on the Ibis T5000 instrument.17,18 The assay was performed according to the manufacturer's instructions and interpreted according to the guidelines in the reference by Wolk, et al.19
Statistical Methods
Descriptive statistics were computed on all study variables. Frequency distributions were used to summarize categorical variables and measures of central tendency and dispersion to summarize continuous variables. Statistical models were constructed to model MRSA status as a function of demographic, behavioral, and clinical characteristics. Because between-site differences in this multi-site study were substantial, we accounted for them with the implementation of generalized linear models with an unknown within-site correlation structure. Model parameters were estimated using generalized estimating equations (GEE). Bivariate associations between all variables and the outcome of MRSA status were examined in unadjusted GEE models. A multivariable adjusted model was chosen based on clinical relevance, statistical significance, and quasi-likelihood under the independence model criterion (QIC). Odds ratios and 95% confidence intervals were reported. The data analysis for this paper was generated using SAS software, Version 9.3 of the SAS System for Windows.
RESULTS
Study Population Characteristics
Table 1 lists the characteristics of the participants in the study. All were HIV positive. Of those participating, a majority (87%) was on HIV medications. Participants’ median CD4 count was 487 cells/mm3 (IQR: 316 – 676.5 cells/mm3), with 11.6% (n=58) enrolled with a CD4 count less than 200 cells/mm3; the majority, 73.4% (n=367), had an undetectable viral load at enrollment.
Table 1.
Summary Statistics of Study Measures
| GSS (n=150) | Moore (n=350) | ||||
|---|---|---|---|---|---|
| Overall sample (n=500) n (%) | MRSA-neg (n=136) n (%) | MRSA-pos (n=14) n (%) | MRSA-neg (n=287) n (%) | MRSA-pos (n=63) n (%) | |
| Sex | |||||
| Male | 330 (66.0) | 113 (83.1) | 13 (92.9) | 173 (60.3) | 31 (49.2) |
| Female | 170 (34.0) | 23 (16.9) | 1 (7.1) | 114 (39.7) | 32 (50.8) |
| Race | |||||
| Black | 345 (69.0) | 36 (26.5) | 3 (21.4) | 254 (88.5) | 52 (82.5) |
| Other | 13 (2.6) | 7 (5.2) | 1 (7.1) | 4 (1.4) | 1 (1.6) |
| White | 142 (28.4) | 93 (68.4) | 10 (71.4) | 29 (10.1) | 10 (15.9) |
| Education | |||||
| No high school/no GED | 131 (26.2) | 0 (0) | 0 (0) | 103 (35.9) | 28 (44.4) |
| High school/GED | 143 (28.6) | 14 (10.3) | 2 (14.3) | 109 (38.0) | 18 (28.6) |
| Some college/ vocational | 111 (22.2) | 42 (30.9) | 5 (35.7) | 52 (18.1) | 12 (19.1) |
| College grad or more | 115 (23.0) | 80 (58.8) | 7 (50.0) | 23 (8.0) | 5 (7.9) |
| Yearly Income | |||||
| <25000 | 351 (70.5) | 24 (17.9) | 3 (21.4) | 264 (92.0) | 60 (95.2) |
| 25,001-50,000 | 62 (12.5) | 37 (27.6) | 6 (42.9) | 17 (5.9) | 2 (3.2) |
| 50,001-75,000 | 46 (9.2) | 36 (26.9) | 3 (21.4) | 6 (2.1) | 1 (1.6) |
| >75,000 | 39 (7.8) | 37 (27.6) | 2 (14.3) | 0 (0) | 0 (0) |
| Been arrested | |||||
| Yes | 40 (8.0) | 4 (2.9) | 0 (0) | 26 (9.1) | 10 (15.9) |
| No | 460 (92.0) | 132 (97.1) | 14 (100) | 261 (90.9) | 53 (84.1) |
| Hands on customer contact job | |||||
| Yes | 94 (18.8) | 49 (36.0) | 4 (28.6) | 39 (13.6) | 2 (3.2) |
| No | 406 (81.2) | 87 (64.0) | 10 (71.4) | 248 (86.4) | 61 (96.8) |
| Sexual orientation | |||||
| Different sex | 315 (63.0) | 95 (69.9) | 8 (57.1) | 69 (24.0) | 13 (20.6) |
| Same sex/both sexes | 185 (37.0) | 41 (30.1) | 6 (42.9) | 218 (76.0) | 50 (79.4) |
| Substance abuse (‘yes’ to street drugs)1 | |||||
| Yes | 253 (50.6) | 31 (22.8) | 4 (28.6) | 173 (60.3) | 45 (71.4) |
| No | 247 (49.4) | 105 (77.2) | 10 (71.4) | 114 (39.7) | 18 (28.6) |
| Sexually active1 | |||||
| Yes | 336 (67.2) | 111 (81.6) | 11 (78.6) | 177 (61.7) | 37 (58.7) |
| No | 164 (32.8) | 25 (18.4) | 3 (21.4) | 110 (38.3) | 26 (41.3) |
| # Sex Partners in last 30 days | |||||
| 0 | 220 (44.3) | 50 (36.8) | 4 (28.6) | 133 (46.3) | 33 (52.4) |
| 1 | 240 (48.3) | 70 (51.5) | 7 (50.0) | 135 (47.0) | 28 (44.4) |
| 2 or more | 37 (7.4) | 16 (11.8) | 3 (21.4) | 16 (5.7) | 2 (3.2) |
| STI2 | |||||
| Yes | 16 (3.2) | 8 (5.9) | 1 (7.1) | 6 (2.1) | 1 (1.6) |
| No | 484 (96.8) | 128 (94.1) | 13 (92.9) | 281 (97.9) | 62 (98.4) |
| Current abscess | |||||
| Yes | 39 (7.8) | 9 (6.6) | 2 (14.3) | 22 (7.7) | 6 (9.5) |
| No | 461 (92.2) | 127 (93.4) | 12 (85.7) | 265 (92.3) | 57 (90.5) |
| Prior Abscess1 | |||||
| Yes | 67 (13.4) | 16 (11.8) | 2 (14.3) | 38 (13.2) | 11 (17.5) |
| No | 433 (86.6) | 120 (88.2) | 12 (85.7) | 249 (86.8) | 52 (82.5) |
| Hospitalized1 | |||||
| Yes | 142 (28.5) | 25 (18.4) | 2 (14.3) | 85 (29.7) | 30 (47.6) |
| No | 357 (71.5) | 111 (81.6) | 12 (85.7) | 201 (70.3) | 33 (52.4) |
| On isolation3 | |||||
| Yes | 86 (17.3) | 17 (12.8) | 2 (14.3) | 55 (19.2) | 12 (19.0) |
| No | 410 (82.7) | 116 (87.2) | 12 (85.7) | 231 (80.8) | 51 (81.0) |
| HIV Medication | |||||
| Yes | 430 (86.7) | 125 (92.6) | 14 (100) | 242 (84.9) | 49 (79.0) |
| No | 66 (13.3) | 10 (7.4) | 0 (0) | 43 (15.1) | 13 (21.0) |
| On prophylaxis4 | |||||
| Yes | 86 (17.3) | 11 (8.1) | 0 (0) | 59 (20.8) | 16 (25.4) |
| No | 411 (82.7) | 125 (91.9) | 14 (100) | 225 (79.2) | 47 (74.6) |
| Viral Load | |||||
| Undetectable | 367 (73.4) | 115 (84.6) | 14 (100.0) | 196 (68.3) | 42 (66.7) |
| Detectable | 133 (26.6) | 21 (15.4) | 0 (0) | 91 (31.7) | 21 (33.3) |
| CD4 | |||||
| <200 | 58 (11.6) | 5 (3.7) | 0 (0) | 42 (14.6) | 11 (17.5) |
| 201-350 | 92 (18.4) | 20 (14.7) | 0 (0) | 55 (19.2) | 17 (27.0) |
| 351-500 | 113 (22.6) | 27 (19.9) | 6 (42.9) | 69 (24.0) | 11 (17.4) |
| >500 | 237 (47.4) | 84 (61.8) | 8 (57.1) | 121 (42.2) | 24 (38.1) |
Missing: Hospitalized n=1; Isolation n=4; Prophylaxis n=3; Income n=2; HIV Meds n=4; number of sex partners in 30 days n=3
Within previous 12 months.
Within previous 6 months.
On any type of isolation during any previous hospitalization.
Prophylaxis for HIV-associated opportunistic infections includes any use of: trimethoprim/sulfamethoxezole; dapsone; azithromycin; fluconazole.
Of the study's 500 participants, 66% were male and 69% were Black. Although the majority of participants had received at least a high school education (74%), the majority (71%) also had an annual income of less than $25,000. Over half of the participants reported sexual activity within the previous 12 months (67%). Almost two-thirds identified heterosexual (63%).
Isolate Characteristics
A total of 217 MRSA isolates from 77 individuals were identified and characterized with PCR ESI-MS technology. Of those, 86.2% were identified as a USA 300 strain type. The remaining 13.8% of the isolates included strain types USA 100, 700, 900, 1000 and 200/1100. The rates for Panton Valentine Leukocidin genes (LukS & LukD), Toxic Shock Toxin 1 (TSST1), and high level (mupA) mupirocin resistance, were 77.4%, 5.1% and 2.8% respectively. No isolates exhibited low level (IleS) mupirocin resistance. All isolates positive for both mupirocin resistance as well as PVL were identified as USA 300 strain type, while the TSST positive isolates were a mix of both USA 300 and USA 100 strains. Of all 217 MRSA isolates from the 77 individuals, the proportions were highest for the nares (21.2%) and throat (20.7%), followed by the rectum (16.1%), groin (15.2%), perineum (15.2%), vagina (5.1%), axilla (4.1%), and wound (2.3%).
MRSA Colonization by Body Site
Among 77 individuals with MRSA colonization (15.4% prevalence), the most common anatomical sites were nares of 46 individuals (59.7%) and throat of 45 individuals (58.4%) (Table 2). Forty-eight (62.3%) of those colonized were culture positive at more than one site, and only one individual was colonized with multiple strain types. Thirty-one individuals (40.3%) were culture negative from the nares, but found to be colonized with MRSA at another anatomical site. Among those with MRSA colonization, the number colonized at only one site was 29 (37.7%), 2 to 3 sites was 20 (26.0%), 4 to 5 sites was 22 (28.6%) and 6 individuals were colonized at 6 ore more sites (7.8%). Two-sided Fisher's exact tests were performed to examine associations between MRSA colonization sites and sexual activity in the prior 12 months and found not to be associated with the throat (p=0.318), perineum (p= 0.563), groin (p= 0.702), vagina (p=0.533) or rectum (p=0.850). The prevalence of USA strains was high across all body sites, ranging from 81.8% for vaginal and up to 94.1% (n=32) for rectal sites. All 5 wounds colonized with MRSA were USA strains.
Table 2.
MRSA Prevalence Among Body Sites by Nares Result.
| Positive body site | Nares Negative (n=454) N (%) | Nares Positive (n=46) N (%) | Total Positive n=77 N (%) | USA Positive1 n=66 N (%) |
|---|---|---|---|---|
| Nares | ---- | ----- | 46 (59.7) | 39 (84.8) |
| Throat | 15 (3.3) | 30 (65.2) | 45 (58.4) | 37 (82.2) |
| Rectal | 11 (2.4) | 23 (50.0) | 34 (44.2) | 32 (94.1) |
| Groin | 9 (2.0) | 24 (52.2) | 33 (41.6) | 28 (84.8) |
| Perineum | 6 (1.3) | 26 (56.5) | 32 (42.9) | 29 (90.6) |
| Axilla | 1 (0.2) | 8 (17.4) | 9 (11.7) | 8 (88.9) |
| Wound2 | 1 (3.2) | 4 (66.7) | 5 (13.5) | 5 (100) |
| Vaginal3 | 4 (2.6) | 7 (36.8) | 11 (6.5) | 9 (81.8) |
Percent USA out of total positive for each body site.
Total 37 individuals with a wound; 31 nares negative and 6 nares positive.
Total 170 women; 152 nares negative and 19 nares positive.
Sex Partners Screening and MRSA Status
We obtained data on 34 dyads (n=68) to examine similarities and differences between partners and MRSA colonization. Ten couples (29%) had at least one partner who was colonized with MRSA in at least one body site. Four of the 10 couples were both partners positive for MRSA. All four of these subject/partner pairs (100%) shared the same strain type and all were identified as USA 300 by PCR ESI-MS technology. Only one of the four MRSA-positive concordant couples was positive for colonization in several body sites; the female was positive for vaginal, rectal, perineum and groin while the male partner was positive for the axilla and throat. The remaining three couples had the female partner positive at one body site (nares in two and vaginal in one), while the male partner was positive at multiple sites.
Risk Factors for MRSA Colonization
Unadjusted and multivariable models are presented in Table 3. In the unadjusted model, persons with the lowest annual income (<$25,000) had the greatest odds of colonization compared to those with the highest income (>$75,000) (OR, 3.96, 95% CI 3.61, 4.32). In the adjusted model, these relationships remained, but somewhat attenuated. Compared with the highest income (>$75,000), persons earning less than $25,000 had 2.68 odds of colonization (95% CI 2.33, 3.08). Including race and income in the GEE model resulted in multicollinearity between the two variables. Further, the 1.13 odds of colonization by race as black relative to was not significant (95% CI 0.67, 1.90), therefore, race was removed from the final model. Similarly, while there were 1.73 (95% CI 1.16, 2.58) odds of colonization with heterosexual orientation, this relationship did not hold when adjusting for other factors and sexual orientation was not included in the final model.
Table 3.
GEE Model Results
| Unadjusted Models | Adjusted Model | |||
|---|---|---|---|---|
| OR (95% CI) | p-value | OR (95% CI) | p-value | |
| Sex | ||||
| Male | 0.61 (0.52, 0.71) | <0.0001 | 0.75 (0.53, 1.05) | 0.0888 |
| Female | Reference | Reference | ||
| Education | ||||
| No high school/no GED | 2.34 (1.40, 3.93) | 0.0013 | ||
| High School/GED | 1.33 (0.80, 2.21) | 0.2690 | ||
| Some college/vocational | 1.57 (1.38, 1.80) | <0.0001 | ||
| College grad or more | Reference | |||
| Yearly income | ||||
| <25,000 | 3.96 (3.61, 4.35) | <0.0001 | 2.68 (2.33, 3.08) | <0.0001 |
| 25,001-50,000 | 2.82 (2.37, 3.37) | <0.0001 | 2.50 (1.87, 3.32) | <0.0001 |
| 50,001-75,000 | 1.76 (1.32, 2.34) | 0.0001 | 1.67 (1.35, 2.06) | <0.0001 |
| 75,001 + | Reference | |||
| Been arrested | ||||
| Yes | 2.07 (1.92, 2.23) | <0.0001 | 1.55 (1.34, 1.80) | <0.0001 |
| No | Reference | Reference | ||
| Hands on customer contact job | ||||
| Yes | 0.33 (0.17, 0.66) | 0.0017 | ||
| No | Reference | |||
| Sexual orientation | ||||
| Different sex | 1.73 (1.16, 2.58) | 0.0077 | ||
| Same sex/both sexes | Reference | |||
| Substance abuse (‘yes’ to street drugs)1 | ||||
| Yes | 1.69 (1.38, 2.07) | <0.0001 | 1.43 (1.38, 1.48) | <0.0001 |
| No | Reference | Reference | ||
| Sexually Active1 | ||||
| Yes | 0.74 (0.60, 092) | 0.0027 | 0.84 (0.83, 0.85) | <0.0001 |
| No | Reference | Reference | ||
| # Sex Partners in last 30 days | ||||
| 0 | 1.22 (1.15, 1.28) | <0.0001 | ||
| 1 | Reference | |||
| 2 or more | 0.95 (0.47, 1.90) | 0.8819 | ||
| Current or prior abscess2 | ||||
| Yes | 1.44 (1.09, 1.91) | 0.0109 | 1.19 (1.17, 1.20) | <0.0001 |
| No | Reference | Reference | ||
| Hospitalized1 | ||||
| Yes | 1.91 (1.51, 2.43) | <0.0001 | 1.54 (1.22, 1.94) | 0.0003 |
| No | Reference | Reference | ||
| HIV Medication | ||||
| Yes | 0.66 (0.54, 0.80) | <0.0001 | 0.85 (0.75, 0.96) | <0.0108 |
| No | Reference | Reference | ||
| On prophylaxis3 | ||||
| Yes | 1.20 (1.08, 1.33) | 0.0009 | ||
| No | Reference | |||
| CD4 | ||||
| <200 | 1.36 (1.22, 1.52) | <0.0001 | ||
| 201-350 | 1.46 (1.03, 2.06) | 0.0329 | ||
| 351-500 | 1.16 (0.61, 2.19) | 0.6556 | ||
| >500 | Reference | |||
OR=Odds Ratio; CI=Confidence Interval
Missing: Hospitalized n=1; Prophylaxis n=3; Income n=2; HIV Meds n=4.
Non-significant univariate results: race; on any type of isolation during hospitalization; undetectable viral load.
Non-significant multivariate analysis: sexual orientation; race; CD4 count; prophylaxis; number of sex partners; hands-on job; education.
Previous 12 months.
Abscess groups combined to be current or history of abscess (yes/no). Reference group was “no”.
Prophylaxis for HIV-associated opportunistic infections includes any use of: trimethoprim/sulfamethoxezole; dapsone; azithromycin; fluconazole.
Lower adjusted odds of colonization were associated sexual activity in the previous 12 months and use of HAART. Participants who reported being sexually active within the prior 12 months had lower odds of MRSA colonization (AOR 0.84, 95% CI 0.83, 0.85). Taking HIV medication was associated with lower odds of MRSA colonization (AOR 0.85, 95% CI 0.75, 0.96). In the unadjusted analysis, men were less likely to be colonized with MRSA than women (AOR 0.61, 95% CI 0.52, 0.71); however this relationship did not remain in the final adjusted model (AOR 0.75, 95% CI 0.53, 1.05).
Additional risks for MRSA were conferred for several attributes. Those who had been arrested were 55% more likely to have MRSA colonization (AOR 1.55, 95% CI 1.34, 1.80). Using street drugs in the previous 12 months conferred independent additional risk for MRSA infection (AOR 1.343, 95% CI 1.38, 1.48). Participants with a history of or with a current abscess (AOR 1.19, 95% CI 1.17, 1.20), or having been hospitalized (AOR 1.54, 95% CI 1.22, 1.94) were also significantly more likely to be colonized with MRSA.
DISCUSSION
MRSA is a preventable infection that can lead to significant morbidity and mortality among PLWH. The findings from the present study provide new information about risk factors for MRSA colonization, and are particularly relevant to the ongoing debate regarding body site colonization and sexual acquisition among this vulnerable population. Colonization of the groin in particular has been associated with skin and soft tissue infections,20 suggesting the importance of the anatomical site of colonization in setting surveillance and treatment priorities. In the present study, we found that 40.3% of all MRSA colonized individuals were culture negative in the nares suggests that surveillance of the nares alone can miss substantial MRSA colonization among PLWH and may contribute to both transmission in hospital and community settings.
Being sexually active in the previous 12 months was found to lower the odds of MRSA colonization, suggesting that other means of acquisition may play a greater role in transmission in our study population. Comparing those reporting sexual activity with sexually inactive individuals, we found significant differences (chi2 p=0.041) in the proportion with CD4+ T cell counts over 500 (50.3 vs. 41.5) and below 200 (10.1 vs. 14.6). Consequently, sexually active individuals may be in better overall health, thus providing one possible explanation of the protective effect of this characteristic. Four of the 10 sexual partner pairs were found to carry the same strain, indicating likely household transmission.
Although several recent studies report increased MRSA transmission among MSM populations,3,4,6 we did not find an association with same-sex coupling when adjusting for other risk factors. The loss of a significant association may have been a consequence of confounding by other factors. Two-sided Fisher's exact tests were performed to examine associations with sexual orientation. Compared to heterosexual orientation, those reporting same or both sex orientation were more likely to be sexually active (76.8% vs. 61.1%, p=0.001) and report annual income over $75,000 (16.9% vs. 8%, p<0.001), both of which were protective in adjusted analysis. Compared to same or both sex orientation, heterosexual orientation was positively associated with substance abuse (29.2% vs. 11.9%, p<0.001), and being hospitalized (32.2% vs. 22.2%, p<0.010) or arrested (10.1% vs. 4.3%, p<0.025) in the previous 12 months, which were risks for colonization in adjusted analysis. Lack of associations in multivariate analysis for multiple sex partners may be influenced by low prevalence (n=37, 7.4%), and under-reporting related to social desirability bias. Neither age nor race was associated with increased MRSA colonization.
Use of HIV antiretroviral therapy (ART) lowered the odds of MRSA colonization. While several studies have looked at use of ART as a risk factor for MRSA,21 three of the four studies that found any association did so only in univariate analyses,20,22,23 and only one study showed significance in a multivariate analysis, with a reduced odds of MRSA infection when taking ART.11 Our findings suggest that receiving ART may lower the odds of colonization as well.
Clinical indicators of immune status, viral load, and CD4 count, were not found to be significantly associated MRSA colonization. It may be that being on ART was a surrogate indicator of greater engagement in outpatient primary care and fewer encounters with hospitals and other acute healthcare settings where contact with and transmission of MRSA can occur. Greater engagement in primary care, which enhances primary prevention, may be protective of MRSA colonization regardless of the stage of HIV disease.
Studies of MRSA surveillance within general populations have found a greater proportion identified with the nares alone compared to our study in PLWH, and confirmed that sampling multiple anatomical sites greatly increases screening sensitivity. Within general populations on hospital or ICU admission, nares cultures identified from only about 66%24 to 81.5%25 of all MRSA colonized patients. Sensitivity analysis has shown that sampling multiple sites can increase yield with nares and perineum to 89.6%26 and with nares, throat and groin screening to 98%.27 Screening nares only when MRSA prevalence is low (<6%) yields 68% of all colonized patients; with high prevalence 73%; and in the ICU 75%.28 Even with a very high (15.4%) prevalence in our study of PLWH, the proportion identified through nares alone (58.9%) was much lower than proportions identified in the general population. It is clear that among PLWH, screening multiple body sites is of even greater importance.
MRSA colonization remains common among PLWH and predictors of colonization are highly variable depending on the clinical context and the study sample characteristics. We found clear differences in a number of socioeconomic and health indicators between the two clinic sites as seen in Table 1. The GSS serves a primarily, white, more educated and male population with higher incomes and greater CD4+ T cell counts compared to the Moore Clinic's primarily inner-city, African-American, less educated clients with lower incomes and lower CD4+ T cell counts. However, the choice of the GEE method for modeling associations can help to account for intra-clinic correlations. In addition, both clinics are part of the same medical service and share the same clinicians, which likely limits differences in service delivery as a factor in the associations we found.
In this cross-sectional analysis, we have identified the importance of multi-body site evaluation in this patient population and uncovered additional risk associated with colonization in this group. It is time for hospital epidemiologist and infection preventionist to reconsider surveillance approaches for populations known to have a high colonization prevalence for MRSA.
Limitations
Although robust estimation of standard errors was used in fit of the GEE models, the small number of clinic sites likely results in an underestimation of the standard errors. Using a cross sectional method, this study examined prevalence of MRSA at a single point in time and does not account for changing relationships among risk factors and MRSA acquisition over time. We attempted to reduce the likelihood of providing socially desirable responses, particularly regarding drug use and sexual behaviors, by using previously implemented survey questions, providing a private interview location and confirming confidentiality of responses. To control for this potential bias, respondent answers were verified by medical record review and abstraction. Sample size was determined based on an estimated 18% prevalence of current abscess; however prevalence in the study sample was 7.8% overall. Consequently, our ability to detect smaller effect sizes and the likelihood that statistically significant associations reflect true effects may be limited.
Highlights.
Risks for MRSA colonization were modeled for a clinical cohort of persons living with HIV
40% were nares negative yet colonized at another anatomical site
Lower odds for colonization were associated with recent sexual activity and use of antiretroviral therapy
Increased odds were associated with lower income, recent hospitalization, incarceration, use of street drugs and skin abscess
Screening for MRSA at multiple anatomical sites is indicated for persons living with HIV
ACKNOWLEDGEMENTS
Financial support. This publication was made possible by the Johns Hopkins Institute for Clinical and Translational Research (ICTR); the National Center for Advancing Translational Sciences (NCATS), a component of the National Institutes of Health (NIH); and NIH Roadmap for Medical Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of the Johns Hopkins ICTR, NCATS or NIH.
The authors value the support of the Johns Hopkins University AIDS Service staff and are particularly grateful to their patients at the Moore Clinic and Greenspring Station offices. Marguerite B. Lucea, PhD, MPH, RN, with the Department of Community and Public Health of the Johns Hopkins University School of Nursing provided assistance in the early study period. We also acknowledge the funding and support of the Johns Hopkins Institute for Clinical and Translational Research and the National Center for Advancing Translational Sciences of the National Institutes of Health.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Potential conflicts of interest. All authors report no conflicts of interest relevant to this article.
Contributor Information
Jason E. Farley, Department of Community and Public Health Johns Hopkins University School of Nursing 525 N. Wolfe Street, Room 525 Baltimore, MD 21205.
Matthew J. Hayat, School of Public Health, Georgia State University PO Box 3984 Atlanta, GA 30302.
Paul L. Sacamano, Department of Epidemiology Johns Hopkins University Bloomberg School of Public Health 615 N. Wolfe Street Baltimore, MD 21205.
Tracy Ross, Department of Medical Microbiology Johns Hopkins University School of Medicine 600 North Wolfe Street Baltimore, MD 21287.
Karen Carroll, Department of Medical Microbiology Johns Hopkins University School of Medicine 600 North Wolfe Street Baltimore, MD 21287.
REFERENCES
- 1.Crum-Cianflone N, Weekes J, Bavaro M. Recurrent community-associated methicillin-resistant Staphylococcus aureus infections among HIV-infected persons: incidence and risk factors. AIDS Patient.Care STDS. 2009;23(7):499–502. doi: 10.1089/apc.2008.0240. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Crum-Cianflone NF, Grandits G, Echols S, et al. Trends and causes of hospitalizations among HIV-infected persons during the late HAART era: what is the impact of CD4 counts and HAART use? J.Acquir.Immune.Defic.Syndr. 2010;54(3):248–257. doi: 10.1097/qai.0b013e3181c8ef22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Diep BA, Chambers HF, Graber CJ, et al. Emergence of multidrug-resistant, community-associated, methicillin-resistant Staphylococcus aureus clone USA300 in men who have sex with men. Annals of internal medicine. 2008;148(4):249–257. doi: 10.7326/0003-4819-148-4-200802190-00204. [DOI] [PubMed] [Google Scholar]
- 4.Crum-Cianflone NF, Burgi AA, Hale BR. Increasing rates of community-acquired methicillin-resistant Staphylococcus aureus infections among HIV-infected persons. International journal of STD & AIDS. 2007;18(8):521–526. doi: 10.1258/095646207781439702. [DOI] [PubMed] [Google Scholar]
- 5.Skiest DJ, Brown K, Cooper TW, Hoffman-Roberts H, Mussa HR, Elliott AC. Prospective comparison of methicillin-susceptible and methicillin-resistant community-associated Staphylococcus aureus infections in hospitalized patients. The Journal of infection. 2007;54(5):427–434. doi: 10.1016/j.jinf.2006.09.012. [DOI] [PubMed] [Google Scholar]
- 6.Delorenze GN, Horberg MA, Silverberg MJ, Tsai A, Quesenberry CP, Baxter R. Trends in annual incidence of methicillin-resistant Staphylococcus aureus (MRSA) infection in HIV-infected and HIV-uninfected patients. Epidemiology and infection. 2013 Nov;141(11):2392–2402. doi: 10.1017/S0950268813000228. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Popovich KJ, Hota B, Aroutcheva A, et al. Community-associated methicillin-resistant Staphylococcus aureus colonization burden in HIV-infected patients. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2013 Apr;56(8):1067–1074. doi: 10.1093/cid/cit010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Al Rawahi GN, Schreader AG, Porter SD, Roscoe DL, Gustafson R, Bryce EA. Methicillin-resistant Staphylococcus aureus nasal carriage among injection drug users: six years later. Journal of clinical microbiology. 2008;46(2):477–479. doi: 10.1128/JCM.01596-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Huang H, Cohen SH, King JH, Monchaud C, Nguyen H, Flynn NM. Injecting drug use and community-associated methicillin-resistant Staphylococcus aureus infection. Diagnostic microbiology and infectious disease. 2008;60(4):347–350. doi: 10.1016/j.diagmicrobio.2007.11.001. [DOI] [PubMed] [Google Scholar]
- 10.Lee NE, Taylor MM, Bancroft E, et al. Risk factors for community-associated methicillin-resistant Staphylococcus aureus skin infections among HIV-positive men who have sex with men. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2005;40(10):1529–1534. doi: 10.1086/429827. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Mathews WC, Caperna JC, Barber RE, et al. Incidence of and risk factors for clinically significant methicillin-resistant Staphylococcus aureus infection in a cohort of HIV-infected adults. Journal of acquired immune deficiency syndromes (1999) 2005;40(2):155–160. doi: 10.1097/01.qai.0000179464.40948.b9. [DOI] [PubMed] [Google Scholar]
- 12.Drapeau CM, Angeletti C, Festa A, Petrosillo N. Role of previous hospitalization in clinically-significant MRSA infection among HIV-infected inpatients: results of a case-control study. BMC infectious diseases. 2007;7:36. doi: 10.1186/1471-2334-7-36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Farley JE, Ross T, Stamper P, Baucom S, Larson E, Carroll KC. Prevalence, risk factors, and molecular epidemiology of methicillin-resistant Staphylococcus aureus among newly arrested men in Baltimore, Maryland. Am.J.Infect.Control. 2008;36(9):644–650. doi: 10.1016/j.ajic.2008.05.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Farley JE, Stamper PD, Ross T, Cai M, Speser S, Carroll KC. Comparison of the BD GeneOhm methicillin-resistant Staphylococcus aureus (MRSA) PCR assay to culture by use of BBL CHROMagar MRSA for detection of MRSA in nasal surveillance cultures from an at-risk community population. Journal of clinical microbiology. 2008;46(2):743–746. doi: 10.1128/JCM.02071-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Graham PL, III, Lin SX, Larson EL. A U.S. population-based survey of Staphylococcus aureus colonization. Annals of internal medicine. 2006;144(5):318–325. doi: 10.7326/0003-4819-144-5-200603070-00006. [DOI] [PubMed] [Google Scholar]
- 16.Carroll KC, Borek AP, Burger C, et al. Evaluation of the BD Phoenix automated microbiology system for identification and antimicrobial susceptibility testing of staphylococci and enterococci. Journal of clinical microbiology. 2006;44(6):2072–2077. doi: 10.1128/JCM.02636-05. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ecker DJ, Sampath R, Massire C, et al. Ibis T5000: a universal biosensor approach for microbiology. Nat.Rev.Microbiol. 2008;6(7):553–558. doi: 10.1038/nrmicro1918. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wolk DM, Blyn LB, Hall TA, et al. Pathogen profiling: rapid molecular characterization of Staphylococcus aureus by PCR/electrospray ionization-mass spectrometry and correlation with phenotype. Journal of clinical microbiology. 2009;47(10):3129–3137. doi: 10.1128/JCM.00709-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Hall TA, Sampath R, Blyn LB, et al. Rapid molecular genotyping and clonal complex assignment of Staphylococcus aureus isolates by PCR coupled to electrospray ionization-mass spectrometry. J Clin Microbiol. 2009 Jun;47(6):1733–1741. doi: 10.1128/JCM.02175-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Peters PJ, Brooks JT, McAllister SK, et al. Methicillin-resistant Staphylococcus aureus colonization of the groin and risk for clinical infection among HIV-infected adults. Emerg Infect Dis. 2013 Apr;19(4):623–629. doi: 10.3201/eid1904.121353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Shadyab A, Crum-Cianflone N. Methicillin-resistant Staphylococcus aureus (MRSA) infections among HIV-infected persons in the era of highly active antiretroviral therapy: a review of the literature. HIV.Med. 2012 doi: 10.1111/j.1468-1293.2011.00978.x. [DOI] [PubMed] [Google Scholar]
- 22.Hidron AI, Moanna A, Rimland D. The rise and fall of methicillin-resistant Staphylococcus aureus infections in HIV patients. Aids. 2011 Apr 24;25(7):1001–1003. doi: 10.1097/QAD.0b013e328343c595. [DOI] [PubMed] [Google Scholar]
- 23.Ramsetty SK, Stuart LL, Blake RT, Parsons CH, Salgado CD. Risks for methicillin-resistant Staphylococcus aureus colonization or infection among patients with HIV infection. HIV Med. 2010 Jul 1;11(6):389–394. doi: 10.1111/j.1468-1293.2009.00802.x. [DOI] [PubMed] [Google Scholar]
- 24.Matheson A, Christie P, Stari T, et al. Nasal swab screening for methicillin-resistant Staphylococcus aureus--how well does it perform? A cross-sectional study. Infect Control Hosp Epidemiol. 2012 Aug;33(8):803–808. doi: 10.1086/666639. [DOI] [PubMed] [Google Scholar]
- 25.Lauderdale TL, Wang JT, Lee WS, et al. Carriage rates of methicillin-resistant Staphylococcus aureus (MRSA) depend on anatomic location, the number of sites cultured, culture methods, and the distribution of clonotypes. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology. 2010 Dec;29(12):1553–1559. doi: 10.1007/s10096-010-1042-8. [DOI] [PubMed] [Google Scholar]
- 26.Bitterman Y, Laor A, Itzhaki S, Weber G. Characterization of the best anatomical sites in screening for methicillin-resistant Staphylococcus aureus colonization. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology. 2010 Apr;29(4):391–397. doi: 10.1007/s10096-009-0869-3. [DOI] [PubMed] [Google Scholar]
- 27.Lautenbach E, Nachamkin I, Hu B, et al. Surveillance cultures for detection of methicillin-resistant Staphylococcus aureus: diagnostic yield of anatomic sites and comparison of provider- and patient-collected samples. Infect Control Hosp Epidemiol. 2009 Apr;30(4):380–382. doi: 10.1086/596045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.McKinnell JA, Huang SS, Eells SJ, Cui E, Miller LG. Quantifying the impact of extranasal testing of body sites for methicillin-resistant Staphylococcus aureus colonization at the time of hospital or intensive care unit admission. Infect Control Hosp Epidemiol. 2013 Feb;34(2):161–170. doi: 10.1086/669095. [DOI] [PMC free article] [PubMed] [Google Scholar]