Abstract
Methicillin-resistant Staphylococcus aureus (MRSA) is an emerging problem in horses; however, the epidemiology of infection and colonization is poorly understood. This study evaluated factors associated with MRSA colonization at the time of admission to a veterinary teaching hospital. A case-control study evaluating historical factors was performed. Previous colonization of the horse, previous identification of colonized horses on the farm, antimicrobial administration within 30 days, admission to the neonatal intensive care unit, and admission to a service other than the surgical service were risk factors for community-associated colonization. A better understanding of risk factors for MRSA colonization is important to elucidate the epidemiology of this emerging veterinary and zoonotic pathogen, and to design evidence-based infection control programs.
Résumé
Facteurs de risques reliés à la colonisation par le Staphylococcus aureus résistant à la méthicilline chez des chevaux admis dans un hôpital vétérinaire d’enseignement. Le Staphilococcus aureus résistant à la méthicilline (SARM) est un problème émergeant chez les chevaux alors que l’épidémiologie de l’infection et de la colonisation est encore mal connue. Cette étude a pour but d’évaluer les facteurs associés à la colonisation par le SARM au moment de l’admission dans un hôpital vétérinaire d’enseignement. Une étude cas-témoins évaluant les facteurs historiques a été réalisée. La colonisation du cheval, l’identification des chevaux colonisés sur la ferme, l’administration d’antimicrobiens en deçà de 30 jours, l’admission dans une unité de soins intensifs néonataux et l’admission dans un service autre que le service de chirurgie constituaient des facteurs de risques pour la colonisation associée au groupe. Une meilleure compréhension des facteurs de risque associés à la colonisation par le SARM est importante pour élucider l’épidémiologie de ce pathogène vétérinaire et zoonotique émergeant et pour mettre au point des programmes de contrôle basés sur des données scientifiques.
(Traduit par Docteur André Blouin)
Introduction
Methicillin-resistant Staphylococcus aureus (MRSA) appears to be an emerging pathogen in horses. Methicillin-resistant Staphylococcus aureus infections have been reported in horses in multiple countries, and transmission of MRSA between horses and humans has been identified, both in veterinary hospitals and in the community (1–6). There is concern that MRSA colonization may be widespread in the horse population, at least in certain regions. One study of horses on farms in Ontario, and in New York State, USA, reported a prevalence of colonization of 4.7% (4), and a study of colonization of horses at the time of admission to a veterinary teaching hospital reported a prevalence of 27/1000 admissions (7). In contrast, no colonized horses were identified during screening of 200 horses in the community in the Netherlands (8) or 40 horses in the United Kingdom (1), yet the authors of the latter study suggested that MRSA was present in the general horse population, based on the diversity of MRSA isolates found in hospitalized horses.
Methicillin-resistant Staphylococcus aureus colonization in horses is of concern for multiple reasons. Colonization at the time of admission to a veterinary hospital has been identified as a risk factor for development of a clinical MRSA infection (7). Colonized horses may also transmit MRSA to other horses and humans, and zoonotic MRSA infections from horses have occurred (3,6).
Risk factors for community-associated MRSA (CA-MRSA) infection and colonization in horses have not been adequately evaluated. Identification of risk factors would allow for a better understanding of the epidemiology of MRSA in horses. It would also facilitate the design of appropriate infection control programs that might include active screening and, potentially, isolation of horses deemed at higher risk for colonization, similar to screening methods that are used in some human healthcare facilities (9). The objective of this study was to identify risk factors for CA-MRSA colonization in horses admitted to a veterinary teaching hospital.
Materials and methods
Study group
As part of ongoing MRSA control measures, nasal swabs are collected for MRSA culture from all horses presented to the Ontario Veterinary College–Veterinary Teaching Hospital (OVC-VTH). Enrichment culture is performed, as has been described elsewhere (7). Screening results from horses admitted to OVC-VTH between October 14, 2002, and April 20, 2005, were evaluated for this study. Horses from which MRSA was isolated from a nasal swab at the time of admission were classified as having CA-MRSA. Four controls, consisting of horses that were screened and found negative for MRSA at the time of admission, were randomly selected for each case.
Data collection
Data were gathered on 19 categorical, potential risk factors or confounders, which are listed in Table 1. Animals were classified as having received antimicrobials, if they had been administered oral or parenteral antimicrobials within 30 d of admission. Information on the specific antimicrobials received within that time period was also obtained. Previous admission to the OVC-VTH and coming from a farm with a history of MRSA infection or colonization were considered as putative risk factors because of increased opportunities for prior exposure to MRSA. Admission to the Foal Watch program, a 24-hour neonatal intensive care unit program, was included, based on previous experiences with MRSA colonization in foals admitted to this program, including zoonotic transmission of MRSA to hospital personnel (6).
Table 1.
Univariate analysis of putative, categorical risk factors considered for an association with MRSA colonization in horses admitted to the veterinary teaching hospital of the Ontario Veterinary College between October 14, 2002, and April 20, 2005 (n = 340)
| Variable | Value | Number of horses | OR | 95% CI | P |
|---|---|---|---|---|---|
| Age | < 1 mo | 38 | 1.0 | Ref | Ref |
| 1 to 6 mo | 33 | 0.23 | 0.08–0.65 | 0.005a | |
| 6 mo to 1 y | 15 | 0.26 | 0.07–0.98 | 0.047a | |
| 1 to 3 y | 52 | 0.21 | 0.09–0.54 | 0.001a | |
| 3 to 10 y | 132 | 0.07 | 0.02–0.12 | < 0.001a | |
| 11 to 20 y | 65 | 0.21 | 0.09–0.52 | 0.001a | |
| > 20 y | 5 | b | |||
| Breed | Mixed | 24 | 1.0 | Ref | Ref |
| Thoroughbred | 144 | 10.45 | 1.37–79.83 | 0.024a | |
| Standardbred | 91 | 4.18 | 0.52–33.52 | 0.178 | |
| Warmblood | 38 | 1.28 | 0.11–14.91 | 0.845 | |
| Other | 43 | 4.47 | 0.52–38.76 | 0.174 | |
| Gender | Male, intact | 79 | 1.0 | Ref | Ref |
| Female | 169 | 0.73 | 0.40–1.34 | 0.307 | |
| Male, castrated | 92 | 0.20 | 0.08–0.50 | 0.001a | |
| Season | Winter | 69 | 1.0 | Ref | Ref |
| Spring | 110 | 1.28 | 0.66–2.48 | 0.464 | |
| Summer | 82 | 0.17 | 0.05–0.49 | 0.001a | |
| Fall | 79 | 0.34 | 0.14–0.81 | 0.015a | |
| Reason | Elective | 167 | 1.0 | Ref | Ref |
| Emergency | 173 | 1.54 | 0.87–2.74 | 0.112 | |
| Admitting service | Medicine | 172 | 1.0 | 0.25–0.77 | 0.004a |
| Surgery | 156 | 0.44 | |||
| Ophthalmology | 2 | b | |||
| Theriogenology | 10 | b | |||
| Foal watch | No | 326 | 1.0 | 4.24–96.34 | < 0.001a |
| Yes | 14 | 16.94 | |||
| Purpose of horse | Pleasure | 82 | 1.0 | 1.0 | < 0.001a |
| Show | 28 | b | 2.14–10.03 | 0.701 | |
| Breeding | 143 | 4.63 | 0.30–2.24 | ||
| Racing | 87 | 0.82 | |||
| Nasogastric tube placed | No | 281 | 1.0 | Ref | 0.482 |
| Yes | 59 | 0.77 | 0.33–1.65 | ||
| Previous admission to OVC | No | 271 | 1.0 | Ref | Ref |
| Yes | 69 | 1.24 | 0.61–2.42 | 0.503 | |
| Previously positive for MRSA | No | 335 | 1.0 | 1 | < 0.001a |
| Yes | 5 | 16.61 | 1.59–821.03 | ||
| Farm positive for MRSA | No | 285 | 1.0 | Ref | Ref |
| Yes | 55 | 9.33 | 4.69–18.54 | < 0.001a | |
| Received antibiotics | No | 263 | 1.0 | Ref | Ref |
| Yes | 77 | 4.34 | 2.35–7.98 | < 0.001a | |
| Received penicillin | No | 307 | 1.0 | Ref | Ref |
| Yes | 33 | 3.90 | 1.71–8.75 | < 0.001a | |
| Received aminoglycosides | No | 312 | 1.0 | Ref | Ref |
| Yes | 28 | 3.96 | 1.63–9.44 | < 0.001a | |
| Received cephalosporins | No | 329 | 1.0 | Ref | Ref |
| Yes | 11 | 7.54 | 1.83–35.92 | < 0.001a | |
| Received trimethoprim sulfa | No | 308 | 1.0 | Ref | Ref |
| Yes | 32 | 2.64 | 1.11–6.05 | 0.011* | |
| Received fluoroquinolones | No | 338 | 1.0 | Ref | Ref |
| Yes | 2 | 3.97 | 0.04–312.79 | 0.295 | |
| Discharge status | Alive | 295 | 1.0 | Ref | Ref |
| Dead | 45 | 1.32 | 0.57–2.87 | 0.457 |
OR: Odds ratio, CI: confidence interval, Ref: reference value
Statistically significant at P < 0.05
None of the horses in this category tested positive for MRSA
Data collected on admitting service was originally categorized as “medical,” “surgical,” “ophthalmologic,” and “theriogenologic.” However, since none of the latter 2 categories included colonized horses, these observations were combined with those of the horses admitted to the medical service to increase the number of records usable by the model.
Model building
Risk factors were screened for inclusion in the model by testing each independently for a significant association with MRSA status, using univariate logistic regression (Intercooled Stata 9.1; Stata Corp LP, College Station, Texas, USA). Fisher’s exact tests were used to test associations between dichotomous variables when 1 or more of the cell values were small (< 20). Those variables that achieved a liberal significance level of P ≤ 0.20 in the univariate analyses were considered for inclusion in the multivariate model. Significant risk factors were further screened by testing for pair-wise correlations between sets. Any pairs of variables found to be highly correlated (r ≥ 0.8) or strongly associated [Odds ratio (OR) < 0.3 or > 8] were evaluated further to eliminate those variables with the weaker strength of association with the outcome, as determined by odds ratios.
A backward elimination approach to logistic regression was used initially to perform a descriptive analysis without accounting for any clustering of cases by farm of origin. Variables were considered to be significantly associated with the outcome if P ≤ 0.05. The presence of confounding factors was evaluated by removing each insignificant variable, 1 at a time, and noting the effect of that elimination on the coefficients for the remaining variables. If at least 1 of the coefficients for the remaining variables changed by more than 25%, the eliminated variable was deemed to be a confounder and restored to the model. The model achieved through this manner was deemed the “main effects model.” Biologically plausible interactions were assessed by creating interaction terms and adding them to the main effects model, 1 by 1, to test for their significance. The potential impact of clustering by farm of origin was subsequently accounted for by adding a random effect to create a mixed logistic regression model. Analyses were performed both by using a combination of all horses that were administered antimicrobials and by separate analysis of individual antimicrobial classes.
Age as a continuous variable was assessed for a linear relationship with the outcome by plotting a smoothed scatterplot of the log odds of the outcome against each predictor.
Because the data involved solely binary values and a limited range of covariate patterns, goodness-of-fit of the final model was assessed primarily by using the Pearson and deviance χ2 tests. When a model’s fit was questionable, the Hosmer-Lemeshow goodness-of-fit test was also employed. To identify outlier observations, both Pearson and deviance residuals were calculated. The data were also examined for influential and unusual (leverage) observations, using delta-betas and hat matrix calculations, respectively (10).
Results
Community-associated MRSA colonization was identified in 69/3372 (2.0%) horses from which an admission nasal swab was collected. Two hundred and seventy-six control horses were enrolled; however, 5 were subsequently removed, because it was determined that the initial nasal swab had been collected more than 24 h after admission.
In the univariate analyses, many variables were found to be significantly associated with MRSA colonization (Table 1); however, when controlling for other variables through multivariate analysis, some factors lost that association. These non-significant factors were subsequently confirmed to be related to other, stronger, predictors and consequently they were dropped from the multivariate logistic regression model. For example, most categories of “age” became insignificant when controlling for “Foal Watch” and “antimicrobials,” which were both significant factors in the multivariate model. Further analysis showed that “age” in its continuous form was associated with both “Foal Watch” (OR = 0.002, P = 0.033) and “antimicrobials” (OR = 0.88, P < 0.001). Because of this fact, and because the graphical association between the log odds of the outcome (MRSA status) and “age” in its continuous form was determined to be a nonlinear association that could not be resolved by transforming “age” to a quadratic term, we concluded that “Foal Watch” and “antimicrobials” better captured the true association, so “age” was removed from further analyses.
Two final main effects models were produced. Only 5 factors were deemed significant for retention in the 1st final model, where the type of antimicrobial was not specified, as follows: antimicrobial use (“antimicrobials”), testing positive for MRSA previously, coming from a test-positive farm, being admitted by the surgery service, and being on Foal Watch (Table 2). When the variable “antimicrobials” was replaced by specific classes of antimicrobials, only “penicillin” and “trimethoprim-sulfa (TMS)” were deemed significant (Table 3), in addition to the other 4 aforementioned variables. No significant interaction terms involving any variables in the main effects models were identified.
Table 2.
Summary of risk factors for methicillin-resistant Staphylococcus aureus colonization in horses admitted to the Ontario Veterinary College between October 14, 2002, and April 20, 2005, that were colonized with methicillin-resistant Staphylococcus aureus, using a mixed logistic regression model to account for clustering at the level of farm of origin and antibiotic use as a main effect
| Coefficient | OR | 95% CI for OR | P | |
|---|---|---|---|---|
| Antibiotics | 1.59 | 4.93 | 2.14–11.34 | < 0.001a |
| Previously positive | 4.32 | 75.88 | 3.52–1637.51 | 0.006a |
| Positive farm | 2.26 | 9.57 | 3.70–24.78 | < 0.001a |
| Foal watch | 2.55 | 12.78 | 1.98–82.34 | 0.007a |
| Surgery | −0.94 | 0.39 | 0.17–0.89 | 0.025a |
| Intercept | −2.64 | N/A | N/A | < 0.001a |
OR: Odds ratio, CI: confidence interval, N/A: not applicable
Statistically significant at P < 0.05
Table 3.
Summary of risk factors for methicillin-resistant Staphylococcus aureus colonization in horses admitted to the Ontario Veterinary College between October 14, 2002, and April 20, 2005, using a mixed logistic regression model to account for clustering at the level of farm of origin and penicillin and trimethoprim-sulfa (TMS) use as main effects
| Coefficient | OR | 95% CI for OR | P | |
|---|---|---|---|---|
| TMS | 1.24 | 3.46 | 1.16–10.34 | 0.026a |
| Penicillin | 1.67 | 5.29 | 1.82–15.36 | 0.002 |
| Previously positive | 4.39 | 80.36 | 3.77–1714.29 | 0.005a |
| Positive farm | 2.35 | 10.48 | 4.05–27.12 | < 0.001a |
| Foal watch | 2.44 | 11.47 | 1.81–72.85 | 0.010a |
| Surgery | −1.01 | 0.36 | 0.15–0.83 | 0.016a |
| Intercept | −2.56 | N/A | N/A | < 0.001a |
OR: Odds ratio, CI: confidence interval, N/A: not applicable
Statistically significant at P < 0.05
Pearson and deviance χ2 tests showed that the final model using “antimicrobials” fit the observed data reasonably well (P = 0.655 and 0.590, respectively). None of the observations’ residuals exceeded ± 1.5. No significant unusual observations were detected. Covariate patterns that were found to be influential were justifiably so (due to large numbers of observations within those patterns). Similar observations were made regarding a model where trimethoprim-sulfa (“TMS”) replaced “antimicrobials,” with the Pearson and deviance χ2 tests confirming a reasonable fit (P = 0.810 and 0.830, respectively).
However, the model that included both “TMS” and “penicillin” was found to have inadequate fit (P = 0.048 and 0.130 for the Pearson and deviance χ2 tests). Consequently, a Hosmer-Lemeshow goodness-of-fit test was run, demonstrating adequate fit under its criteria (P = 0.465). Four particular covariate patterns out of a total of 22 were associated with standardized residuals in excess of 2, but since the data related to these observations were not erroneous, no justification could be found for removing them from the analysis. The covariate pattern identified as having high leverage consisted of only 2 observations: 2 MRSA-positive horses that did not possess any of the risk factors included in the model. A different pattern, identified as being highly influential, was comprised of 20 MRSA-positive horses that possessed only 2 risk factors (they had tested positive previously, and they were admitted for reasons other than surgery). Removing this group from the analysis led to improved fit (P = 0.293 and 0.280 for the Pearson and deviance χ2 tests). In doing so, the significance of being admitted to the surgical service versus other services decreased to P = 0.25. Ultimately, we chose to leave the 20 influential observations in the model.
In both mixed-effects models, using “farm of origin” as a random effect, the intra-class correlation was found to be 0.31, suggesting a significant amount of clustering at this level (P = 0.002 to 0.004).
Discussion
This study identified a variety of factors that were associated with CA-MRSA colonization in horses, information that is useful for furthering an understanding of MRSA in this species and for the development of MRSA control programs. It was not surprising that previous identification of colonization in the individual horse was an important risk factor. While duration of colonization has not been adequately explored and persistent colonization has not been identified, colonization of up to 5 mo duration has been reported (11). Consequently, at our hospital, all horses that were infected or colonized at the time of their last visit are considered colonized until screening results are reported. These horses are isolated and handled with contact precautions until screening results are available, and this protocol seems to be a reasonable recommendation, based on the results of this study.
It was similarly unsurprising that residence on a farm where 1 or more horses had previously been diagnosed with MRSA infection or colonization was also a risk factor, as identification of multiple colonized horses on a farm is not uncommon (3,4). A previous point prevalence study of MRSA colonization of horses on farms reported isolation of MRSA from 46/391 (12%) of horses from farms with a history of MRSA infection or colonization, including 1 farm where 45% of horses were colonized (4). The highly social nature of horse behavior leads to frequent contact between horses, particularly contact with the nares. Shared facial-contact items, such as water and feed containers, could play a role in transmission, as MRSA contamination of items such as these has been reported in an equine hospital (12). Further, the nature of human-horse interaction also frequently involves direct contact with the nares during petting of the face and nose, creating opportunities for interspecies transmission. Because of the high prevalence of colonization identified on some farms, our hospital has a policy whereby any horse from a designated ‘high-risk’ farm is considered as colonized until a negative screening swab is obtained. It is important to note, however, that fewer than 50% of colonized horses were from a farm with previously reported MRSA cases, so while a history of previous MRSA on the farm may be a useful indicator of higher risk horses, it should not be the sole criterion for determination of whether screening is required.
It was not particularly surprising that, in light of a recent report that administration of cephalosporins or aminoglycosides during hospitalization was associated with hospital-associated MRSA (HA-MRSA) colonization in horses, antimicrobial therapy was associated with CA-MRSA colonization in this study (13). Previous antimicrobial therapy has also been reported as a risk factor in humans (14,15), yet there is increasing recognition of CA-MRSA infection in humans without risk factors, such as antimicrobial use (16,17). These latter reports have tended to focus on the USA300 [Canadian epidemic MRSA (CMRSA)-10] clone of MRSA, which has emerged as a tremendously important cause of CA-MRSA infection in humans in North America. In contrast, USA500 (CMRSA-5) has accounted for the vast majority of MRSA infections in horses in North America (3,4,6,7) and USA300 (CMRSA-10) has yet to be reported in horses. While the reason for the predominance of USA500 (CMRSA-5) is unclear, it may impact on the epidemiology of MRSA in horses and this difference in predominant MRSA clones may hamper direct comparison with humans. Regardless, the association of antimicrobials with MRSA colonization highlights the need for prudent antimicrobial use in veterinary medicine, as has been discussed elsewhere (18).
Interestingly, administration of cephalosporins and aminoglycosides, while found to be significantly associated with CA-MRSA status in the univariate analyses in this study and while reported to be associated with HA-MRSA colonization using multivariate analysis in another study (7), failed to achieve significance when controlling for other factors in the present study. Rather, penicillin and TMS were the only antimicrobials that maintained their strong association. This finding emphasizes the need to consider putative risk factors in concert with others when investigating relationships with any outcome.
Based on previous reports of MRSA infection and colonization in critically ill neonatal foals (3,6), it was not particularly surprising that admission to the Foal Watch program was a risk factor. Reasons for this are unclear and may relate to the intensiveness of management of foals on farms prior to referral. It may also relate to farm characteristics that were not evaluated in the model, particularly when one considers that the high cost of neonatal foal care may create some bias in this referred population, as many horse breeders are unwilling or unable to afford this level of veterinary care. One interesting factor to consider is that most foals admitted to the Foal Watch program are less than 24 h of age, which highlights how quickly after birth MRSA colonization can develop.
The finding that admission to the surgery service was a protective factor was somewhat surprising and is without a ready explanation. It is possible that this relates to disease severity and the amount of medical care provided prior to admission.
Interestingly, previous hospitalization was not a risk factor for horses in this study. In humans, contact with healthcare facilities is a commonly reported risk factor for CA-MRSA infection or colonization (19,20); however, colonization and infection in individuals without recognized risk factors are increasingly being reported (21–23). Care must be taken in comparing human and equine data, because the nature of the equine population and equine healthcare is quite different from that of humans. In fact, some major subsets of the equine population, such as breeding farms, may mimic the human healthcare situation more closely than the human community situation because of the frequent contact of horses with veterinary personnel, frequent antimicrobial use, regular inter-farm movement of horses, and care by a constant group of farm personnel that may concurrently be managing sick horses.
This study used timing of isolation of MRSA as the criterion for designating CA-MRSA colonization. Currently, the epidemiology of MRSA infection and colonization appears to be changing, and there is considerable debate about the optimal definition of CA-MRSA in humans (24,25,17). The use of time of sampling as the sole criterion for differentiating CA- from HA-colonization does risk including some horses with HA- MRSA colonization, as some horses that were identified as colonized at admission could have actually been infected during previous hospitalization. This does hamper some of the epidemiological conclusions that can be made; however, the results of this study are nonetheless important as they indicate factors that are associated with horses being colonized upon admission to a veterinary hospital, something that is important to know for the design and implementation of optimal infection control programs. Further studies should be undertaken to design optimal farm- and hospital-based screening programs to lessen the dissemination of this potentially important veterinary and zoonotic pathogen. CVJ
References
- 1.Baptiste KE, Williams K, Williams NJ, et al. Methicillin-resistant staphylococci in companion animals. Emerg Infect Dis. 2005;11:1942–1944. doi: 10.3201/eid1112.050241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.O’Mahony R, Abbott Y, Leonard FC, et al. Methicillin-resistant Staphylococcus aureus (MRSA) isolated from animals and veterinary personnel in Ireland. Vet Microbiol. 2005;109:285–296. doi: 10.1016/j.vetmic.2005.06.003. [DOI] [PubMed] [Google Scholar]
- 3.Weese JS, Archambault M, Willey BM, et al. Methicillin-resistant Staphylococcus aureus in horses and horse personnel, 2000–2002. Emerg Infect Dis. 2005;11:430–435. doi: 10.3201/eid1103.040481. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Weese JS, Rousseau J, Traub-Dargatz JL, Willey BM, McGeer A, Low DE. Community-associated methicillin-resistant Staphylococcus aureus in horses and humans who work with horses. J Am Vet Med Assoc. 2005;226:580–583. doi: 10.2460/javma.2005.226.580. [DOI] [PubMed] [Google Scholar]
- 5.Cuny C, Kuemmerle J, Stanek C, Willey B, Strommenger B, Witte W. Emergence of MRSA infections in horses in a veterinary hospital: Strain characterisation and comparison with MRSA from humans. Euro Surveill. 2006;11:44–47. [PubMed] [Google Scholar]
- 6.Weese JS, Caldwell F, Willey BM, et al. An outbreak of methicillin- resistant Staphylococcus aureus skin infections resulting from horse to human transmission in a veterinary hospital. Vet Microbiol. 2006;114:160–164. doi: 10.1016/j.vetmic.2005.11.054. [DOI] [PubMed] [Google Scholar]
- 7.Weese JS, Rousseau J, Willey BM, Archambault M, McGeer A, Low DE. Methicillin-resistant Staphylococcus aureus in horses at a veterinary teaching hospital: Frequency, characterization, and association with clinical disease. J Vet Intern Med. 2006;20:182–186. doi: 10.1892/0891-6640(2006)20[182:msaiha]2.0.co;2. [DOI] [PubMed] [Google Scholar]
- 8.Busscher JF, van Duijkeren E, Sloet van Oldruitenborgh-Oosterbaan MM. The prevalence of methicillin-resistant staphylococci in healthy horses in the Netherlands. Vet Microbiol. 2005;113:131–136. doi: 10.1016/j.vetmic.2005.10.028. [DOI] [PubMed] [Google Scholar]
- 9.West TE, Guerry C, Hiott M, Morrow N, Ward K, Salgado CD. Effect of targeted surveillance for control of methicillin-resistant Staphylococcus aureus in a community hospital system. Infect Control Hosp Epidemiol. 2006;27:233–238. doi: 10.1086/500372. [DOI] [PubMed] [Google Scholar]
- 10.Dohoo I, Martin W, Stryn H. Veterinary Epidemiologic Research. Charlottetown, PEI: AVC Inc; 2003. [Google Scholar]
- 11.Weese JS, Rousseau J. Attempted eradication of methicillin-resistant Staphylococcus aureus colonisation in horses on two farms. Equine Vet J. 2005;37:510–514. doi: 10.2746/042516405775314835. [DOI] [PubMed] [Google Scholar]
- 12.Weese JS, Goth K, Ethier M, Boehnke K. Isolation of methicillin- resistant Staphylococcus aureus from the environment in a veterinary teaching hospital. J Vet Int Med. 2004;18:468–470. doi: 10.1892/0891-6640(2004)18<468:iomsaf>2.0.co;2. [DOI] [PubMed] [Google Scholar]
- 13.Weese JS, Archambault M, Rousseau J, Willey BM, McGeer A, Low DE. Methicillin-resistant Staphylococcus aureus (MRSA) surveillance in horses at a veterinary teaching hospital. 11th Int Symp Staphylococci Staphylococcal Infect; Charleston, South Carolina. 2004. [Google Scholar]
- 14.Hidron AI, Kourbatova EV, Halvosa JS, et al. Risk factors for colonization with methicillin-resistant Staphylococcus aureus (MRSA) in patients admitted to an urban hospital: Emergence of community-associated MRSA nasal carriage. Clin Infect Dis. 2005;41:159–166. doi: 10.1086/430910. [DOI] [PubMed] [Google Scholar]
- 15.Karas JA, Enoch DA, Emery MM. Community-onset healthcare- associated MRSA bacteraemia in a district general hospital. J Hosp Infect. 2006;62:480–486. doi: 10.1016/j.jhin.2005.10.001. [DOI] [PubMed] [Google Scholar]
- 16.Frazee BW, Salz TO, Lambert L, Perdreau-Remington F. Fatal community- associated methicillin-resistant Staphylococcus aureus pneumonia in an immunocompetent young adult. Ann Emerg Med. 2005;46:401–404. doi: 10.1016/j.annemergmed.2005.05.023. [DOI] [PubMed] [Google Scholar]
- 17.King MD, Humphrey BJ, Wang YF, Kourbatova EV, Ray SM, Blumberg HM. Emergence of community-acquired methicillin- resistant Staphylococcus aureus USA 300 clone as the predominant cause of skin and soft-tissue infections. Ann Intern Med. 2006;144:309–317. doi: 10.7326/0003-4819-144-5-200603070-00005. [DOI] [PubMed] [Google Scholar]
- 18.Morley PS, Apley MD, Besser TE, et al. Antimicrobial drug use in veterinary medicine. J Vet Intern Med. 2005;19:617–629. doi: 10.1892/0891-6640(2005)19[617:aduivm]2.0.co;2. [DOI] [PubMed] [Google Scholar]
- 19.Baggett HC, Hennessy TW, Rudolph K, et al. Community-onset methicillin-resistant Staphylococcus aureus associated with antibiotic use and the cytotoxin Panton-Valentine leukocidin during a furunculosis outbreak in Rural Alaska. J Infect Dis. 2004;189:1565–1573. doi: 10.1086/383247. [DOI] [PubMed] [Google Scholar]
- 20.Lu PL, Chin LC, Peng CF, et al. Risk factors and molecular analysis of community methicillin-resistant Staphylococcus aureus carriage. J Clin Microbiol. 2005;43:132–139. doi: 10.1128/JCM.43.1.132-139.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Centers for Disease Control and Prevention. Four pediatric deaths from community-acquired methicillin-resistant Staphylococcus aureus. Minnesota and North Dakota. MMWR. 1999;48:707–710. [PubMed] [Google Scholar]
- 22.Huang YC, Su LH, Chen CJ, Lin TY. Nasal carriage of methicillin-resistant Staphylococcus aureus in school children without identifiable risk factors in northern Taiwan. Pediatr Infect Dis J. 2005;24:276–278. doi: 10.1097/01.inf.0000154333.46032.0f. [DOI] [PubMed] [Google Scholar]
- 23.Kazakova SV, Hageman JC, Matava M, et al. A clone of methicillin-resistant Staphylococcus aureus among professional football players. N Engl J Med. 2005;352:468–475. doi: 10.1056/NEJMoa042859. [DOI] [PubMed] [Google Scholar]
- 24.Tacconelli E, Venkataraman L, De Girolami PC, EM DA. Methicillin-resistant Staphylococcus aureus bacteraemia diagnosed at hospital admission: Distinguishing between community-acquired versus healthcare-associated strains. J Antimicrob Chemother. 2004;53:474–479. doi: 10.1093/jac/dkh107. [DOI] [PubMed] [Google Scholar]
- 25.Lesens O, Hansmann Y, Brannigan E, et al. Healthcare-associated Staphylococcus aureus bacteremia and the risk for methicillin resistance: Is the Centers for Disease Control and Prevention definition for community-acquired bacteremia still appropriate? Infect Control Hosp Epidemiol. 2005;26:204–209. doi: 10.1086/502527. [DOI] [PubMed] [Google Scholar]