Cross-border comparison of the admission prevalence and clonal structure of meticillin-resistant Staphylococcus aureus

Journal of Hospital Infection (2009) 71, 320e326

Available online at www.sciencedirect.com

www.elsevierhealth.com/journals/jhin

Cross-border comparison of the admission prevalence and clonal structure of meticillinresistant Staphylococcus aureus ¨ck a, L. Brakensiek a, A. Mellmann a, F. Kipp b, M. Henderikx c, R. Ko D. Harmsen d, I. Daniels-Haardt e, C. von Eiff b, K. Becker b, M.G.R. Hendrix c, A.W. Friedrich a,* a

Institute of Hygiene, University Hospital Mu¨nster, Mu¨nster, Germany Institute of Medical Microbiology, University Hospital Mu¨nster, Mu¨nster, Germany c Laboratory for Microbiology, Enschede, The Netherlands d Department of Periodontology, University Hospital Mu¨nster, Mu¨nster, Germany e Institute for Health and Work Northrhine-Westphalia (LIGA), Mu¨nster, Germany b

Received 21 August 2008; accepted 1 December 2008 Available online 6 February 2009

KEYWORDS Admission; Cross-border; EUREGIO; Germany; MRSA; Netherlands; Prevalence

Summary Since patient exchange between hospitals sharing a common catchment area might favour regional spread of meticillin-resistant Staphylococcus aureus (MRSA), the reliable detection of patients colonised at admission is crucial. Thus, hospitals in the DutcheGerman border area EUREGIO MRSA-net aim at synchronising their local MRSA standards in order to prevent unidentified inter-hospital as well as cross-border spread. This assumes enhanced knowledge of MRSA prevalence and risk factors associated with MRSA carriage at admission. We conducted nasal MRSA screening of all inpatients admitted to 39 German hospitals (in the period 1 November to 30 November 2006) and to one Dutch hospital (in the period 1 July to 30 September 2007) in the EUREGIO MRSA-net. A total of 390 MRSA cases were detected among 25 540 patients screened. The admission prevalence was 1.6 MRSA/100 patients (6.5% of all S. aureus) in the German and 0.5 MRSA/100 patients (1.4% of all S. aureus) in the Dutch part of the border region. Overall, the predominating S. aureus protein A gene (spa) sequence types were t003, t032 and t011. One isolate (t044) carried Panton-Valentine leukocidin (PVL) encoding genes. Altogether, 79% and 67% of all MRSA

* Corresponding author. Address: Institute of Hygiene, University Hospital Mu ¨nster, Robert-Koch-Str. 41, 48149 Mu ¨nster, Germany. Tel.: þ49-251-83-55366; fax: þ49-251-83-55688. E-mail address: [email protected] 0195-6701/$ - see front matter ª 2008 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhin.2008.12.001

MRSA admission screening

321 patients in the German and Dutch regions respectively, were identifiable by the classical nosocomial risk factors assessed. In patients lacking all risk factors assessed, spa types t011 and t034 were predominant (P < 0.001). ª 2008 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.

Introduction The prevalence of meticillin-resistant Staphylococcus aureus (MRSA) isolated from blood cultures is about 20 times lower in The Netherlands than in Germany.1 The reason for this has been attributed to the stringent Dutch ‘search and destroy’ policy and to optimised antibiotic stewardship.2 Recently, mathematical simulations revealed how even regions with medium or high MRSA prevalence, as hypothesised for Germany, can control and reduce MRSA by adoption of strategies for screening and precautionary isolation.3 Therefore, German healthcare institutions might learn from the Dutch facilities how to prevent the intra- and interhospital spread of MRSA. Hence, we created the EUREGIO MRSA-net Twente/Mu ¨nsterland (www.mrsa-net.eu) as a crossborder network of hospitals and other healthcare institutions in the central part of the DutcheGerman border area and addressed the following two major problems in an area-wide prospective multicentre study. (1) Since data on the prevalence of patients admitted with MRSA are lacking in the German project area and indicate increasing prevalence in the Dutch area, we characterised the actual prevalence of MRSA among patients admitted to hospitals on the Dutch and German sides of the border.4 (2) Since MRSA are detected with increasing frequency

Table I Category

I II III IV Total

in patients lacking classical nosocomial risk factors, it becomes questionable whether ‘search’ policies based on screening of defined risk groups are sufficient for the reliable identification of MRSA patients.5 We therefore asked whether carriers of MRSA are still detectable by screening for risk factors and we sought to determine the major risk factors associated with MRSA.

Methods Setting and screening The EUREGIO MRSA-net geographically comprises six German districts (codes DEA33, DEA34, DEA35, DEA37, DEA38 and DE94B, level 3, according to the Nomenclature of Territorial Units for Statistics) and two Dutch regions (codes NL213, NL222) which are inhabited by 1.7 million people on the German side and 990 000 on the Dutch side of the border. A nasal screening programme for MRSA was carried out in the German study area from 1 to 30 November 2006 in 33 acute care hospitals representing 100% of the regional beds (N ¼ 10139; 6/1000 inhabitants) for acute inpatient care and six further specialised hospitals (Table I). In the EUREGIO Dutch area, the screening was performed between 1 July and 30 September

Results of MRSA screening in hospitals within the German part of the EUREGIO MRSA-net a

No. of hospitalsa 20 12 1 6 39

No. of patients screened (% of all admissions) 8883 11 628 2376 679 23 566

(89) (80) (76) (97) (83)

No. of S. aureus

No. of MRSA

2241 2819 695 124 5879

156 180 27 18 381

MRSA/100 patients (range)b 1.8 1.6 1.1 2.7 1.6

(0.3e4.2) (0.6e3.3) (0.0e7.7) (0.0e7.7)

Average no. of risk points per patient (95% CI, range)c 0.8 (0.6e0.9, 0.2e1.7) 0.5 (0.4e0.6, 0.4e0.7) ed 1.7 (1.3e2.2, 1.3e2.4) 0.8 (0.7e1.0, 0.4e2.4)

a Categories: I (<300 beds); II (300e799 beds); III (>800 beds); IV (specialised in geriatric psychiatry and rehabilitative medicine). b Range between different hospitals. c Risk points of all patients were added up and divided by the number of patients screened; 95% confidence interval (CI); range between different hospitals. d Not available.

322 2007 in one of four acute care hospitals with 13% of all regional patient beds (N ¼ 2435, 2.5/1000 inhabitants). During the study periods, all inpatients provided a nasal swab at admission (cotton swabs in transport medium) using a standardised sampling technique. All ambulatory patients and newborns were excluded. Results of the screening were communicated directly to the hospitals enabling the prompt implementation of infection control measures.

Microbiological diagnostics and typing Microbiological identification of S. aureus was performed on Columbia blood agar and an MRSA-selective medium (bioMe ´rieux), using direct cultures and a 24 h enrichment culture in dextrose broth. Species diagnostic and susceptibility testing were performed by VITEK 2 automated systems (bioMe ´rieux, Marcy l’Etoile, France). Meticillin resistance was confirmed by mecA polymerase chain reaction or penicillin-binding protein 2a (PbP2a) latex agglutination test.6 All MRSA were characterised using Staphylococcus protein A gene (spa) typing.7 Cluster formation of spa types (spa clonal complexes, spa-CC) was done using the Based Upon Repeat Pattern (BURP) algorithm of the Ridom StaphType software (Ridom GmbH, Wu ¨rzburg, Germany).8,9 Differences in the distribution of spa-CCs were analysed statistically using the c2-test for independence (P < 0.05). Multilocus sequence typing was carried out on every MRSA isolated in the Dutch area and on samples for every spa type within spa-CC011.10 Panton-Valentine leukocidin (PVL) encoding genes (lukS-PV, lukF-PV) were detected according to Lina et al.11

Risk factor assessment The screening on admission was done in accordance with the Ethical Committee of the University Hospital Mu ¨nster (2006-268-f-S). Healthcare professionals obtained the following potential risk factors for MRSA carriage at admission: previous carriage or infection with MRSA, contact with another person with MRSA carriage or infection, receipt of any antibiotics during the previous six months, hospitalisation for >24 h within the previous six months (including data on which country), direct transfer of the patient from other hospitals and from long-term care facilities to the admitting hospital (including data on which country), need for long-term care, presence of indwelling devices, skin lesions and burns, and need for haemodialysis.

R. Ko ¨ck et al. All data were collected using a standardised retrieval questionnaire. Because one hospital (>800 beds) carries out a general admission screening of all patients according to its routine local standard, it contributed to the study by taking swabs, but conducted a risk factor assessment only partially. The collected risk data were not included in this study. Statistical analysis of risk factors was done using Epi Info Version 3.3.2 (CDC, Atlanta, GA, USA). c2Test or Fisher’s exact test, where appropriate, was used for assessing risk factors for MRSA carriage. Every risk factor associated with MRSA at P < 0.2 in a univariate analysis was included in an initial logistic regression model. Subsequently, variables for which the c2-statistic of the likelihood ratio test indicated no statistical significance were eliminated from the model. P < 0.05 was considered significant. In order to analyse variations in the risk profile of patients admitted to different hospitals, risk points were assigned to the independent risk factors according to their b-coefficients in the regression model and the average number of risk points per patient was calculated for each hospital.

Results EUREGIO German area In total, 28 378 patients were admitted to the hospitals in the German area of the EUREGIO. Of these, 23 566 patients (83%) provided nasal swabs (Table I). The rate of MRSA/100 patients (95% confidence interval) was 1.4 (1.0e1.4) in general and traumatic surgery (N ¼ 4710 patients), 1.2 (0.7e1.7) in other surgical departments (N ¼ 5496 patients), 2.1 (1.6e2.6) in internal medicine (N ¼ 7592 patients), 1.7 (0.8e2.6) in other conservative departments (N ¼ 1363 patients), 0.8 (0.5e1.1) in paediatrics (N ¼ 1375 patients), 2.6 (0.6e4.6) in intensive care (N ¼ 508 patients), 2.6 (1.0e4.2) in geriatric/rehabilitative wards (N ¼ 757 patients) and 1.2 (0e3.6) in patients for whom no department was given (N ¼ 1765). Geographically, the rate of MRSA/100 patients was 1.4 in the districts DEA33 and DEA34, 1.5 in DE94B and DEA38, 1.8 in DEA35 and 2.2 in DEA37. A total of 375 MRSA (98.4% of all MRSA) were available for spa typing (Table II). All spa types within spa-CC011 exhibited MLST ST398. One isolate associated with spa type t044 carried PVL genes (0.3% of all MRSA).

MRSA admission screening

323

Table II Distribution of spa-CCs and spa types among isolates from the German and the Dutch parts of the EUREGIO, Twente/Mu ¨nsterland EUREGIO MRSA-net area German

spa-CCa 003

004 008/305 011 012 015 346 379 Excludedb Singleton

Dutch

a b c d

Not typedc Total 011 003 total

spa types

Outstanding spa type (% of all isolates within spa-CC)

Total (%)

t001, t002, t003, t010, t014, t035, t045, t067, t105, t151, t422, t481, t603, t1227, t1686, t2020 t004, t065, t2009 t008, t305 t011, t034, t108, t567, t1451, t2011 t012, t018, t318 t015, t069, t073 t085, t346, t2058 t005, t020, t022, t032, t379, t910, t1147, t1499, t1559, t1770, t1865 t132, t463, t535, t1344, t2006 t044, t161, t240, t247, t684, t786, t1687, t2013, t2019

t003 (80%)

169 (44.4)

t011, t108, t899c,d t3271

t004 t008 t011 t018 t015 t085 t032

(66%) (71%) (57%) (50%) (75%) (50%) (81%)

18 7 65 4 8 4 84

(4.7) (1.8) (17.1) (1.1) (2.1) (1.1) (22.1)

5 (1.3) 11 (2.9)

t011 (75%)

6 381 8 1 9

(1.6) (100.0) (89) (11) (100.0)

spa clonal complex (spa-CC) as determined by the Based Upon Repeat Pattern (BURP) algorithm. spa types containing <5 repeats were excluded from BURP clustering. Isolates were not forwarded to spa typing. spa type t899 was grouped as a singleton by BURP, but is multilocus sequence type ST398.

In all, 13 565 of 21 190 patients (64%) interviewed (in hospitals of categories I, II and IV) concerning potential risk factors lacked all the factors stated in the questionnaire. The distribution of risk factors and the results of logistic regression are shown in Table III. Of the patients with MRSA, 35.6% were associated with one, 24% with two and 19.8% with three and more risk factors, respectively. A total of 73 MRSA patients (20.6%) lacked all risk factors assessed. The distribution of spa-CCs in the isolates from these patients differed from those found in MRSA patients with at least one risk factor for spa-CC003 (P < 0.001) and spa-CC011 (P < 0.001). Spa-CC003 and spa-CC011 accounted for 24.7% and 37% of the non-risk-factor-associated isolates versus 49.3% and 12.7% of the risk-factor-associated isolates, respectively. Risk points were assigned to each risk factor (Table III) and the average risk points per patient admitted to each hospital were calculated (Table I).

EUREGIO Dutch area During three months, 1974 nasal swabs were obtained (95% of 2077 patients admitted). In all, 624 S. aureus (31.6% of all screenings) were isolated, nine of which were MRSA (1.4% of all

S. aureus). The MRSA rate was 0.5/100 patients. All isolates were PVL negative and were spa-typed t011 (N ¼ 6, 67%, MLST ST398), t108 (N ¼ 1, 11.1%, MLST ST398), t899 (N ¼ 1, 11.1%, MLST ST398) and t3271 (N ¼ 1, 11.1%, MLST ST225). A total of 1364 patients (69%) lacked all the risk factors assessed. The average patient admitted had 0.6 risk points. Of the MRSA patients, N ¼ 3 lacked all the risk factors assessed (33%), N ¼ 5 had been previously hospitalised in a Dutch hospital, N ¼ 4 had received antibiotics, N ¼ 1 lived in a long-term care facility and N ¼ 1 was fitted with a urinary catheter. None of the MRSA patients (and N ¼ 9 of the patients without MRSA) had been hospitalised in a foreign country. In univariate analysis, only previous hospitalisation in a Dutch hospital was significantly associated with MRSA (P ¼ 0.03).

Discussion In this study, which may be one of the largest MRSA screening studies ever conducted, the MRSA prevalence rate on admission was 1.6 MRSA/100 patients (1.6%) in the German EUREGIO area. This is low compared to previous studies from Switzerland (3.3%), the USA (3.4%), France (6.6%) and the UK (6.7%).12e15 The prevalence in the Dutch EUREGIO

324

R. Ko ¨ck et al.

Table III Multivariate analysis of risk factors associated with MRSA carriage at admission (German part of the EUREGIO MRSA-net) Risk factor Known MRSA carrier Contact to MRSA carriers Hospitalisation for >24 h in the last 6 months Direct transfer from other hospitals Long-term care facilities Antibiotics Haemodialysish Burnsh Skin lesions Catheters Need for long-term care a b c d e f g h

MRSA (N ¼ 354)a

No MRSA (N ¼ 20 836)a

b-Coefficient

ORb

95% CIb

27 9 173c

232 177 4806d

4.20 1.64 0.91

66.8 5.2 2.5

25.5 1.7 1.9

174.9 15.9 3.2

32e

651f

1.02

2.8

1.5

5.1

58g 89 3 0 48 49 55

558g 2742 179 39 415 541 662

3.64 0.29 eh eh 4.0 2.09 1.22

38.2 1.3 eh eh 54.5 8.1 3.4

19.6 1.0 eh eh 26.2 4.1 1.8

74.2 1.8 eh eh 113.6 15.9 6.4

Pb

Assigned risk points

<0.001 0.0044 <0.001

4 2 1

<0.001

1

<0.001 0.0776 eh eh <0.001 <0.001 <0.001

4 0 eh eh 4 2 1

Number of patients for whom risk factors were available. Odds ratio (OR), 95% confidence interval (CI) and P-value as obtained in logistic regression. N ¼ 173 in Germany. N ¼ 4797 in Germany, N ¼ 2 in Denmark, N ¼ 1 in Poland, N ¼ 1 in Turkey, N ¼ 1 in Spain, N ¼ 4 in unknown foreign country. N ¼ 32 from Germany. N ¼ 647 from Germany, N ¼ 2 from Spain, N ¼ 2 from Angola. All in Germany. Risk factors showing no statistical significance in univariate analysis were not included in logistic regression.

was 3-fold lower (0.5%) than in the German area, but was nonetheless higher than expected, because a study from 2004 reported the prevalence on admission in The Netherlands to be 0.03%.2 In this context, the discordant screening periods and the fact that data in our study were derived from one of four regional Dutch hospitals (13% of regional patient beds) may be limitations and might reflect the local situation only. Examination of MRSA rates within the German study area showed variations between individual hospitals (0.0 to 7.7 MRSA/100 patients) and the various departments (0.8e2.6 MRSA/100 patients) which might be attributable to differences in the implementation of hygiene measures, to local variations in the prescribing of antimicrobial agents and to different case-mix indices.12,13,16 This is supported by our findings that the average risk points per patient at admission ranged from 0.2 to 2.4 (Table I), indicating an unequal distribution of risk patients. Furthermore, these variations show that there is a risk of MRSA dissemination by interinstitutional transfer of patients. Since we observed that almost 5% of patients directly admitted from another regional hospital were colonised with MRSA and >50% of all regional German MRSA exhibited only two classical nosocomial spa types (t003 and t032), the need for consistent screening and synchronised guidelines for MRSA is

highlighted. If preventive measures are carried out only in single hospitals, non-cooperators will compromise the preventive achievements, nullifying the efforts of those who are compliant by facilitating a regional to-and-fro exchange of unidentified carriers.17,18 It was shown previously that estimation of MRSA rates from clinical cultures fails to detect 85% of patients colonised.19 Furthermore, it was reported that it is worth investing resources in MRSA screening because it can be cost-effective in preventing nosocomial MRSA infections.20,21 In this study, we confirmed previously reported risk factors for the carriage of MRSA.12,15 However, recent reports emphasised that MRSA colonisation is increasingly reported in patients lacking any preceding contact to a healthcare setting.5 These community-associated MRSA cases might interfere with MRSA ‘search’ policies which focus on risk patients. In our study, 20.6% of MRSA patients in the German and 33% of MRSA patients in the Dutch project area showed none of the risk factors assessed. This is in accord with data from Denmark reporting that 24% of MRSA patients in Copenhagen carry MRSA with a community onset, but without healthcare-associated risk factors.22 In contrast to previous reports from other regions showing that patients with community-associated MRSA frequently carry PVL-encoding genes, we identified only one PVL-MRSA exhibiting the most frequent

MRSA admission screening European PVL-associated spa type t044 in our study.22e24 In the USA, PVL-MRSA, apart from their pathogenic impact for community-acquired infections, have been introduced into the hospitals and have emerged as a cause of healthcare-associated infections; our findings underline recent observations indicating a relatively low prevalence of PVL-MRSA (1e2% of all MRSA) in Germany.25,26 Focusing on those isolates that lacked all risk factors assessed, spa typing revealed that among these MRSA spa-CC011 (MLST ST398) was significantly overrepresented (37% of all isolates compared to 12.7% of the isolates found among patients associated with at least one risk factor). MRSA ST398 strains have recently been associated with livestock and farming in The Netherlands and their potential pathogenicity has been reported.27e29 In the Dutch EUREGIO 89% of all MRSA exhibited ST398 and in the German EUREGIO the prevalence of ST398 was high compared with national data.30 This might corroborate recent reports that the occurrence of MRSA ST398 is linked to regions with a high density of pig and cattle farming, which is true for the Dutch and the German EUREGIO, and that screening of farmers in Dutch hospitals leads to a three-fold increase in the local incidence of MRSA.4,31 Further investigation on the prevalence of MRSA among regional livestock is warranted in order to determine whether contact with animals can explain the high prevalence of MRSA ST398 in the EUREGIO. In summary, we showed that the prevalence of MRSA on admission is 1.6/100 patients on the German and 0.5/100 patients on the Dutch side of the border. The majority of MRSA patients in regional German hospitals are identifiable by screening for classical nosocomial risk factors. Nevertheless, molecular typing demonstrated that the MRSA lineage ST398 predominates among Dutch MRSA and is emerging among the German isolates, suggesting a potential livestock-related reservoir for community-associated MRSA. Furthermore, spa typing indicated the frequent spread of MRSA between regional hospitals in the German EUREGIO.

Acknowledgements We thank all Dutch and German hospitals participating in the EUREGIO MRSA-net network for the screening; also the regional laboratories for the microbiological diagnostics. The participants in the project are named at www.mrsa-net.eu. We are grateful to J. van der Palen for his help with statistical analysis.

325 Conflict of interest statement D. Harmsen partially owns the company Ridom GmbH, Wu ¨rzburg which distributes the Ridom StaphType software mentioned in the article. No other authors have commercial or potential dual interests. Funding sources This study was performed within the framework of the INTERREG-IIIa project ‘EUREGIO MRSAnet Twente/Mu ¨nsterland’ (reference no.: 2-EUR-V-1 ¼ 96).

References 1. Tiemersma EW, Bronzwaer SLAM, Lyytikainen O, et al. Methicillin-resistant Staphylococcus aureus in Europe, 1999e2002. Emerg Infect Dis 2004;10:1627e1634. 2. Wertheim HFL, Vos MC, Boelens HAM, et al. Low prevalence of methicillin-resistant Staphylococcus aureus (MRSA) at hospital admission in the Netherlands: the value of search and destroy and restrictive antibiotic use. J Hosp Infect 2004;56:321e325. 3. Bootsma MCJ, Diekmann O, Bonten MJM. Controlling methicillin-resistant Staphylococcus aureus: quantifying the effects of interventions and rapid diagnostic testing. Proc Natl Acad Sci USA 2006;103:5620e5625. 4. van Rijen MML, Van Keulen PH, Kluytmans JA. Increase in a Dutch hospital of methicillin-resistant Staphylococcus aureus related to animal farming. Clin Infect Dis 2008;46: 261e263. 5. Klevens RM, Morrison MA, Nadle J, et al. Invasive methicillin-resistant Staphylococcus aureus infections in the United States. J Am Med Assoc 2007;298:1763e1771. 6. Becker K, Pagnier I, Schuhen B, et al. Does nasal cocolonization by methicillin-resistant coagulase-negative staphylococci and methicillin-susceptible Staphylococcus aureus strains occur frequently enough to represent a risk of false-positive methicillin-resistant S. aureus determinations by molecular methods. J Clin Microbiol 2006;44: 229e231. 7. Mellmann A, Friedrich AW, Rosenkotter N, et al. Automated DNA sequence-based early warning system for the detection of methicillin-resistant Staphylococcus aureus outbreaks. PLoS Med 2006;3. e33. 8. Mellmann A, Weniger T, Berssenbrugge C, et al. Based Upon Repeat Pattern (BURP): an algorithm to characterize the long-term evolution of Staphylococcus aureus populations based on spa polymorphisms. BMC Microbiol 2007;7:98. 9. Mellmann A, Weniger T, Berssenbrugge C, et al. Characterization of clonal relatedness among the natural population of Staphylococcus aureus strains by using spa sequence typing and the BURP (based upon repeat patterns) algorithm. J Clin Microbiol 2008;46:2805e2808. 10. Enright MC, Day NP, Davies CE, Peacock SJ, Spratt BG. Multilocus sequence typing for characterization of methicillin-resistant and methicillin-susceptible clones of Staphylococcus aureus. J Clin Microbiol 2000;38:1008e1015. 11. Lina G, Piemont Y, Godail-Gamot F, et al. Involvement of Panton-Valentine leukocidin-producing Staphylococcus aureus in primary skin infections and pneumonia. Clin Infect Dis 1999;29:1128e1132.

326 12. Harbarth S, Sax H, Fankhauser-Rodriguez C, Schrenzel J, Agostinho A, Pittet D. Evaluating the probability of previously unknown carriage of MRSA at hospital admission. Am J Med 2006;119:275. e15e23. 13. Davis KA, Stewart JJ, Crouch HK, Florez CE, Hospenthal DR. Methicillin-resistant Staphylococcus aureus (MRSA) nares colonization at hospital admission and its effect on subsequent MRSA infection. Clin Infect Dis 2004;39:776e782. 14. Rioux C, Armand-Lefevre L, Guerinot W, Andremont A, Lucet J. Acquisition of methicillin-resistant Staphylococcus aureus in the acute care setting: incidence and risk factors. Infect Control Hosp Epidemiol 2007;28:733e773. 15. Gopal Rao G, Michalczyk P, Nayeem N, Walker G, Wigmore L. Prevalence and risk factors for meticillin-resistant Staphylococcus aureus in adult emergency admissions e a case for screening all patients? J Hosp Infect 2007;66: 15e21. 16. Livermore DM, Pearson A. Antibiotic resistance: location, location, location. Clin Microbiol Infect 2007;13(Suppl 2): 7e16. 17. Strommenger B, Kettlitz C, Weniger T, Harmsen D, Friedrich AW, Witte W. Assignment of Staphylococcus isolates to groups by spa typing, SmaI macrorestriction analysis, and multilocus sequence typing. J Clin Microbiol 2006; 44:2533e2540. 18. Smith DL, Levin SA, Laxminarayan R. Strategic interactions in multi-institutional epidemics of antibiotic resistance. Proc Natl Acad Sci USA 2005;102:3153e3158. 19. Salgado CD, Farr BM. What proportion of hospital patients colonized with methicillin-resistant Staphylococcus aureus are identified by clinical microbiological cultures? Infect Control Hosp Epidemiol 2006;27:116e121. 20. Diller R, Sonntag AK, Mellmann A, et al. Evidence for cost reduction based on pre-admission MRSA screening in general surgery. Int J Hyg Environ Health 2008;211: 205e212. 21. Nulens E, Broex E, Ament A, et al. Cost of the meticillinresistant Staphylococcus aureus search and destroy policy

R. Ko ¨ck et al.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

in a Dutch university hospital. J Hosp Infect 2008;68: 301e307. Bartels M, Boye K, Larsen A, Skov R, Westh H. Rapid increase of genetically diverse methicillin-resistant Staphylococcus aureus, Copenhagen, Denmark. Emerg Infect Dis 2007;13:1533e1540. Kluytmans-Vandenbergh MFQ, Kluytmans JAJW. Communityacquired methicillin-resistant Staphylococcus aureus: current perspectives. Clin Microbiol Infect 2006;12(Suppl. 1): 9e15. Tristan A, Bes M, Meugnier H, et al. Global distribution of Panton-Valentine leukocidin-positive methicillin-resistant Staphylococcus aureus, 2006. Emerg Infect Dis 2007;13: 594e600. Seybold U, Kourbatova EV, Johnson JG, et al. Emergence of community-associated methicillin-resistant Staphylococcus aureus USA300 genotype as a major cause of health careassociated blood stream infections. Clin Infect Dis 2006; 42:647e656. von Eiff C, Friedrich AW, Peters G, Becker K. Prevalence of genes encoding for members of the staphylococcal leukotoxin family among clinical isolates of Staphylococcus aureus. Diagn Microbiol Infect Dis 2004;49:157e162. De Neeling AJ, van den Broek MJ, Spalburg EC, et al. High prevalence of methicillin resistant Staphylococcus aureus in pigs. Vet Microbiol 2007;122:366e372. Voss A, Loeffen F, Bakker J, Klaassen C, Wulf M. Methicillinresistant Staphylococcus aureus in pig farming. Emerg Infect Dis 2005;11:1965e1966. van Belkum A, Melles DC, Peeters JK, et al. Methicillin-resistant and -susceptible Staphylococcus aureus sequence type 398 in pigs and humans. Emerg Infect Dis 2008;14:479e483. Witte W, Strommenger B, Stanek C, Cuny C. Methicillin-resistant Staphylococcus aureus ST398 in humans and animals, Central Europe. Emerg Infect Dis 2007;13:255e258. van Loo I, Huijsdens X, Tiemersma E, et al. Emergence of methicillin-resistant Staphylococcus aureus of animal origin in humans. Emerg Infect Dis 2007;13:1834e1839.