Comparison of an enrichment broth-enhanced commercial PCR ...

Eur J Clin Microbiol Infect Dis (2006) 26:155–160 DOI 10.1007/s10096-007-0269-5

ARTICLE

Comparison of an enrichment broth-enhanced commercial PCR procedure versus bacteriological culture for separating non-colonized from suspected or colonized MRSA individuals J. H. T. Wagenvoort & M. F. H. A. van de Cruijs & C. T. M. Meuwissen & J. M. H. Gronenschild & E. I. G. B. De Brauwer

Published online: 22 February 2007 # Springer-Verlag 2007

Abstract The aim of the study presented here was to evaluate an enrichment broth-enhanced commercial PCR procedure for excluding the presence of meticillin-resistant Staphylococcus aureus (MRSA) in patient samples in less than 36 h. In The Netherlands to date, all MRSA epidemics have been successfully controlled with the Dutch searchand-destroy policy. However, PCR facilitates more rapid screening for MRSA than traditional culture. One commercial PCR option is the hyplex StaphyloResist® PCR assay (Biologische Analysensystem GmbH, Lich, Germany), which detects Staphylococcus aureus and the mecA gene in MRSA as well as in coagulase-negative staphylococci (CoNS). This assay was used to test a total of 939 specimens obtained from 346 individuals. Following resolution of all discrepancies, the prevalence, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) for all separate specimens were 9.0, 97.6, 83.7, 37.4 and 99.7%, respectively, and for specimens grouped according to daily episode submitted per individual, they were 7.5, 97.4, 77.2, 26.2 and 99.7%, respectively. These results led to the introduction of this PCR into the hospital laboratory’s routine for the purpose outlined above.

J. H. T. Wagenvoort (*) : M. F. H. A. van de Cruijs : C. T. M. Meuwissen : J. M. H. Gronenschild : E. I. G. B. De Brauwer Department of Medical Microbiology, Atrium Medical Centre Parkstad, P.O. Box 4446, 6401 CX Heerlen, The Netherlands e-mail: [email protected]

Introduction In The Netherlands, any epidemic with meticillin-resistant Staphylococcus aureus (MRSA) occurring to date has been successfully controlled with the Dutch search-anddestroy policy [1, 2]. Since 2001, 12 epidemics have occurred at the Atrium Medical Center, Parkstad, which is an 800-bed hospital serving the Limburg Parkstad Rhine– Meuse Euregion; each epidemic resulted in closure of the affected department for more than 1 month. This severely interfered with the hospital’s normal patient-care routine and led to hospital costs of circa EUR 250,000 per epidemic [3]. Since no direct relation to stay at a hospital outside of The Netherlands was found for the patients involved, it seems that MRSA not only spreads during hospital epidemics, it also spreads in the communities within our region [4]. Intra-family transmission has also been documented, particularly with multi-locus sequence type (MLST) 5 [3]. Consequently, there is an obvious need for a quick and sensitive method of identifying MRSA or excluding it as a possible diagnosis. PCR assays screen for MRSA more rapidly than traditional culture. The processing time with a PCR assay varies from 5 h using directly sampled swabs to 1.5 days when an overnight enrichment broth is employed. In comparison, culture averages 5 days for MRSA detection. This rapid processing time of PCR has been demonstrated by several in-house assays, and sensitivity and specificity values of up to and exceeding 90% have been reported [5, 6]. However, use of these tests is generally limited to local settings. The worldwide prevalence of MRSA has highlighted the need for rapid tests and new commercial PCR

156

assays are forthcoming; the prospects for comparably favorable results are promising [7–10]. One of the newly developed commercial PCR options is the hyplex StaphyloResist® PCR assay, manufactured by Biologische Analysensystem GmbH (BAG, Lich, Germany). This multiplex-PCR assay detects both a species-specific genetic element of S. aureus and the mecA gene responsible for meticillin resistance. The PCR assay result is regarded as positive if both aspects are positive. Its structure also allows it to show false-positive results due to the simultaneous detection of meticillinsusceptible S. aureus and the mecA gene of meticillinresistant coagulase-negative staphylococci (CoNS) present in the same patient specimen. Our aim was primarily to evaluate this rapid PCR for obtaining MRSA-negative results as soon as possible. The specimens that elicit a positive PCR assay signal can then be suspected to contain MRSA and should be evaluated further using traditional bacteriological culture. Our approach was to apply an enrichment-broth step to the PCR assay as well as to culture. In a previous study, application of an additional enrichment-broth step was favorable for achieving remarkably more (circa 25%) MRSA-positive results for clinical specimens [11]. We expected that the higher yield already demonstrated for culture would be reflected by a better start for the PCR. This approach was previously confirmed using another PCR assay developed in-house [6]. Described here is an approach that is capable of excluding the presence of MRSA in patient samples in less than 36 h, which is superior to most traditional plating methods currently used in many laboratories.

Materials and methods In this study a total of 939 swab specimens obtained from 346 individuals during initial and follow-up investigations conducted between October 2005 and January 2006 were evaluated for the presence of MRSA using overnight enrichment followed by testing with the hyplex StaphyloResist® PCR assay and traditional culture. The specimens were obtained from the following groups: (a) patients, (b) nursing home and rehabilitation centre inhabitants who were actually or suspected to be colonized with MRSA, (c) contacts of these inhabitants and patients (including other patients, healthcare workers and family members of these) during an outbreak, and (d) patients returning from a hospital stay outside of The Netherlands. Specimens were taken from the nose, throat and perineum and, if present, wounds or other infected anatomic locations. All samples were stored at 4°C between the time of collection and testing in this study.

Eur J Clin Microbiol Infect Dis (2006) 26:155–160

Bacterial strains While 41 different PFGE types of MRSA have circulated in the Limburg Parkcity Rhine–Meuse Euregion in various frequencies during the past 5 years, only eight of these 41 PFGE types were present during the time period investigated. There was a notable predominance (circa 50%) of MRSA strains identified as pediatric New York clone MLST 5 in the various epidemics from which the isolates were obtained. To evaluate whether the commercial PCR could detect all 41 of the PFGE types of MRSA found circulating in this region at various times to date, we additionally tested the stock collection of MRSA strains. These strains were tested in pure culture, with the S. aureus PCR and with the mecA PCR. A second recently developed variant of the PCR mecA, which was not available during the investigation period, was additionally used in the follow-up of one PCR-negative stock strain of MRSA. Also included in our panel, as strain no. 42, was the heterogeneous “Berlin” MRSA (MLST 45) from the Rijnmond outbreak. Around the year 2002, this strain was responsible for one-third of all new MRSA strains detected in The Netherlands [5], and according to the German National Reference Centre of the Robert–Koch Institute, it belongs to the top five strains isolated in German hospitals [12]. Given the proximity of Germany to the Rhine–Meuse Euregion, investigation of this particular strain is also important. Bacterial culture for MRSA The specimens were initially pooled in one non-selective peptone enrichment prior to subculture on agar media [3, 10]. All swabs from each screened individual were quenched in the peptone broth. If an individual tested positive for MRSA, follow-up specimens obtained from different anatomic locations were processed separately. The enrichment broth was incubated for 18 h at 37°C. The broth was subinoculated onto a blood agar plate made selective with aztreonam (25 mg/l) and an oxacillin resistance screening agar base (Oxoid, Basingstoke, UK). To gain an indication of the possible presence of MRSA, a ceftizoxime (30 μg) disc was placed onto the first inoculation streak on the blood agar plate. Suspected colonies were screened using the Denka Seiken® latex agglutination PBP-2 screen test and evaluated further using a previously described procedure [11] and various standard procedures [10]. In accordance with routine practice, all positive MRSA isolates were sent to the National Institute of Health and the Environment (RIVM) for confirmation and registration in the Dutch MRSA surveillance system; there, they were also typed by PFGE and assigned cluster numbers.


PCR Since the mix of culture-positive versus-negative results for all kinds of specimens submitted to our laboratory for MRSA screening during the past 2 years was circa 1:10, the PCR assay was expected to achieve the same ratio. The same enrichment broth used for culture was also employed for the PCR. All non-selective peptone broth media used for enrichment were stored at 4°C until a PCR run of 49 specimens was performed. These runs were carried out blindly up to a maximum of 8 weeks later and the technicians had no prior knowledge of the culture results. The sample preparation, multiplex performance and reverse hybridization of the PCR assay were performed as follows: Sample preparation After microbiological processing, the same enrichment broth was used for the PCR assay. Culture broth (100 μl) was added to 1 ml saline EDTA buffer provided by the manufacturer and mixed well. All specimens were centrifuged for 2 min at 10,000×g and the supernatant was discarded by decanting. All pellets were resuspended in 200 μl hyplex lysis buffer (BAG, Lich, Germany) and mixed well. These solutions were heated to 99°C on a heating block for 10 min and centrifuged for 2 min at 10,000 × g. Five microliters of the sample supernatant was used for the PCR assay. Multiplex PCR Amplification was performed in a thermocycler with a heated lid (Mastercycler personal, Eppendorf, Hamburg, Germany). Amplifications were carried out in 0.2ml tubes (Eppendorf, Hamburg, Germany). Two microliters of the primer mix and 1 μl of the nucleotide mix of the hyplex StaphyloResist® test system (BAG, Lich, Germany) were added for each reaction. Tth polymerase (BAG, Lich, Germany) was used. Finally, 5 μl of the sample supernatant was added to the PCR mix. The cycling conditions were set according to the manufacturer’s instructions: initial denaturation step at 94°C for 5 min, 35 cycles of denaturation at 94°C for 25 s, annealing at 52°C for 25 s, and elongation at 72°C for 20 s increasing by 1 s per cycle. After a final elongation step of 3 min at 72°C, the amplification products were used in reverse hybridization. Reverse hybridization Amplification products were heated at 95°C for 5 min, cooled immediately using an ice block, and 15 μl was then mixed with 150 μl of cool hybridization buffer. Fifty microliters of this mixture was inserted into the color-coded microwell plate cavities containing immobilized oligonucleotide capture probes specific for the mecA gene and S. aureus. Hybridization of amplification products with immobilized oligonucleotide capture probes took place during

157

incubation at 50°C for 30 min. To remove unbound or loosely bound PCR products, three stringent washing steps with a prewarmed (50°C) wash solution were required. After an additional washing step with a second wash buffer, incubation with a peroxidase (POD) conjugate for 30 min at room temperature, three washing steps, and addition of tetramethylbenzidin (TMB) substrate as a developer, the results were read. A positive result is indicated when a blue color develops during incubation for 15 min at room temperature. Stopping the color reaction by adding a stop solution changes the blue color to yellow. The results were measured using a microwell-plate photometer at 450 nm. As a reference, the optical density (OD) was also measured at 630 nm and the test result was obtained by subtracting A630 from A450. OD values of ≥0.150 that were also threetimes higher than the negative control (NC) were defined as positive. For each run, a StaphyloResist-positive control and a negative (sterile water) control were processed like the original samples. The reverse hybridization step was performed automatically using a Personal Lab ELISAprocessor (Adaltis, Freiburg, Germany), which is able to handle two microwell plates in parallel. The parallel detection of the two most common CoNS carrying mecA (S. epidermidis and S. haemolyticus) is possible with the hyplex StaphyloResist® PCR assay, but we did not use this feature in this study. There were four possible outcomes for the specimens tested in this study: (a) a PCR-positive and (MRSA) culture-positive result (PCR+, culture+); (b) a PCR-negative and (MRSA) culture-negative result (PCR−, culture−), for which no further action was necessary; (c) a PCRpositive and (MRSA) culture-negative result (PCR+, culture−), whereby samples were positive and borderline positive for mecA and S. aureus according to PCR but negative by culture and thus regarded as “suspicious for MRSA” (these were re-examined by repeat culture of the enrichment broth, not only for MRSA, but also for the presence of S. aureus and CoNS); and (d) a PCR-negative and (MRSA) culture-positive result (PCR−, culture+), which led to retesting with PCR and/or culture for elucidation of the reason. The prevalence, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were calculated for all separate specimens as well as for specimens grouped per individual per day. Culture discrepancies were evaluated separately.

Results The prevalence of MRSA-positive specimens was 9%. In total, 346 individuals were screened and 20 (6%) of them

158


Table 1 Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of an enrichment broth-enhanced MRSA PCR assay versus enrichment broth-enhanced culture for detecting MRSA in 939 specimens evaluated PCR

Culture+

Culture−

Performance (%)

PCR+

83

139

PCR−

2 85

715

Sensitivitya 97.6 PPVb 37.4 Specificityc 83.7 NPVd 99.7

+ positive, − negative Data are no. of both culture- and PCR-positive results/no. of culturepositive results b Data are no. of both culture- and PCR-positive results/no. of PCRpositive results c Data are no. of both culture- and PCR-negative results/no. of culturenegative results d Data are no. of both culture- and PCR-negative results/no. of PCRnegative results a

tested positive. Among a total of 939 separate specimens, 715 were PCR−/culture− and 61 were PCR+/culture+. The PCR+/culture+ and PCR−/culture− results were clearcut and no further testing was required. The OD value for the PCR-positive results was a median of the mecA PCR of 2,413 (range 156–3,502) and the S. aureus PCR of 2,459 (range 166–3,600), respectively. Initially, the discrepancies were as follows: 139 PCR+/ culture− and 24 PCR−/culture+. For the PCR−/culture+ group of specimens, it was necessary to make some technical modifications to the laboratory procedure, as commented upon in the discussion section; afterwards, 16 of 24 specimens became PCR+ for MRSA. Finally, by replacing the enrichment broth that had been stored at 4°C with a fresh one and reincubating it, six of eight remaining discrepancies were resolved. This left just two of 85 PCR−/ culture+ specimens remaining. Table 1 shows the results obtained after as many PCR discrepancies as possible were resolved and includes the values for sensitivity, specificity, PPV and NPV. These results were also listed as specimens grouped according to daily episode submitted per individual (Table 2). The sensitivity, specificity, PPV and NPV percentages were comparable to those in Table 1. Of a total of 495 screening episodes, one (0.02%) MRSA-infected

individual with a PCR−/culture+ result was missed. On an individual level, 326 of the 346 (circa 95%) individuals screened tested negative. The 139 PCR+/culture− specimens listed in Table 1 were cultured again. During this process, six of the specimens were found to be culture-positive for MRSA, which reduced the sensitivity of this method to 92%. While five of these six specimens were from individuals already known to be MRSA carriers, one belonged to a new MRSA patient whose status was previously missed; this led to a new round of investigations on the ward where the patient was still being treated, in accordance with the Dutch search-and-destroy policy. During this screening no MRSA was found, including from the patient whose MRSApositive status was detected just 3 weeks earlier. Three other PCR+/culture− specimens were from the same individuals who tested PCR+/culture+ within a range of 5 days, which suggests MRSA was present and detectable by PCR but not by culture; this brought the number of potentially false-negative culture results to nine, which may indicate “additional” decreased sensitivity of culture. Methicillin-sensitive Staphylococcus aureus (MSSA) and CoNS were present in 89 of the 139 PCR+/culture− MRSA specimens. During the study, eight PFGE types circulating in our region during the past 5 years were detected by the PCR. With the PCR assay, 40 of 41 different MRSA PFGE types detected in pure culture were positively identified in both PCR arms. The PCR+ samples detected were the MLST 5 MRSA strain and the heterogeneous (“Berlin”) MLST 45 MRSA strain. The 41st PFGE type was only detected later with an updated version of the hyplex StaphyloResist® PCR assay, which was released by the manufacturer in January 2006 and consequently not used in our comparative study.

Discussion The high sensitivity and NPV of the commercial PCR tested indicate it is useful for screening MRSA-negative individuals. The lower specificity and PPV do not affect the

Table 2 Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of an enrichment broth-enhanced PCR assay versus enrichment broth-enhanced culture for detecting MRSA, with specimens grouped according to daily episode submitted per individual Group

MRSA group result

No. (%) of individual daily episodesa

Performance (%)

PCR−/culture− PCR+/culture+ PCR+/culture− PCR−/culture+

Negative Positive Negative Missed MRSA

353 (71.3) 37 (7.4) 104 (21) 1 (0.2)

Sensitivity 97.4 Specificity 77.2 PPV 26.2 NPV 99.7

+ positive, − negative Total no. of specimens grouped according to daily episode submitted per individual (n=495).

a


method’s utility for screening, since all positive PCR findings are cultured before any individual is definitively determined to be an MRSA carrier. During the test period it was necessary to modify the laboratory environment for successful implementation of this molecular biological test. It is advisable that the PCR (initially) be performed by reliable and experienced technicians who then train others. To improve the performance of the PCR, the enrichment broth can be made selective with antibiotics, and after overnight incubation at 37°C, the overgrown enrichment broth has to be homogenized by vortexing. Additionally, an optimal secure test procedure can be promoted using the following measures: decanting and removing remaining drops with a paper towel, ensuring optimal contact between the heating block and the specimens, and using a stable quality of demineralised water. Finally, we noticed the yield of PCR-positive results in retested discrepant samples was better after subinoculation was performed in a fresh enrichment broth; this indicates some degradation occurred during storage of the original broth at 4°C. For clinical utility, we suggest the results of Table 1 are better judged if the specimens are grouped according to daily episodes per individual (Table 2), since this includes specimens obtained from different anatomic locations of the same patient. Although these specimens are often not all positive for MRSA in culture, when combined, they give a higher yield per individual [11]. One positive MRSA result is sufficient for deciding that an individual is colonized with MRSA. During the test period, the number of different PFGE types of MRSA strains in circulation was limited to eight and included the aforementioned pediatric New York MRSA clone of MLST 5. However, the favorable test results obtained using stocked MRSA strains of other PFGE types indicate the PCR is able to detect them as well. For more than one-third of the PCR+/culture− specimens detected in this study, the expected S. aureus and mecA CoNS could not be cultured despite the extra yield of MRSA facilitated by enrichment. This could be due to the development of conditions toxic for certain staphylococci during the 8 weeks of storage at 4°C, but the reasons were not elucidated. A PCR-positive result can reflect mixed cultures of S. aureus and mecA-positive CoNS and not indicate MRSA. Technically speaking, these specimens are not really falsely positive, they simply reflect the correct technical performance of the PCR assay, which was developed to detect the S. aureus and/or mecA markers. Following repeat culture of the broth for PCR+/culture− specimens, 8% more MRSA-positive results were detected by bacteriology. This shows the PCR is able to detect lowlevel colonization, even below the detection level of conventional routine culture. This shakes the belief in

159

culture as the gold standard for MRSA detection. Although PCR is not completely sensitive (97%), neither is culture (92%). It is important to realize that MRSA-positive individuals will be missed with all applied assays; in our study that amounted to one MRSA patient who was missed by PCR as well as culture. In a previous report, Becker et al. [13] considered a low PPV of circa 40% unacceptable for a molecular test in a low-prevalence setting. However, such an opinion is dependent on the reason for applying the PCR. In our view, the high NPV (99.7%) of the PCR we tested is of particular importance in a setting where positive findings are relatively rare, as is currently the case in Dutch hospitals. Our approach is to screen patients and HCWs in order to weed out the majority of MRSA-negative individuals from “the rest,” and for this purpose the PCR is clearly of value in our low-prevalence (0.25%) setting. The “negative” approach was also evaluated favorably in a study of an alternative PCR assay [6]. By separating noncolonized patients from those suspected to be colonized with MRSA, further transmission on a ward can be stopped at an early stage, a concept that ideally fits the Dutch search-and-destroy policy, which aims to reduce the presence of MRSA in the hospital to zero as soon as possible. At the time of writing, the number of peer-reviewed publications of results obtained with the hyplex StaphyloResist® MRSA PCR-Elisa assay was limited to one [14]. That study found values for sensitivity, specificity, PPV and NPV of 92, 90, 31 and 99%, respectively. The sensitivity and NPV values we found were high and are in agreement with both these findings and those obtained with an alternative PCR assay [8]; however, both of those evaluations were performed under different conditions (i.e., no enrichment broth was used) and in countries with markedly higher prevalences of MRSA. The value of using an enrichment broth has been clearly demonstrated with the culture method, resulting in a 25% greater yield of positive results [11]. Although we have not evaluated this step separately, we aimed to devise an optimal detection approach with regard to the PCR and thus applied the PCR assay on the enrichment broth. This procedure has the additional advantage of reducing the possibility of falsenegative PCR results since the use of a pre-PCR enrichment broth also stimulates the multiplication of small quantities of MRSA. Although we did not specifically investigate this point, we believe the 1-day delay incurred by applying the enrichment-broth step may clearly pay off in the form of a higher yield, and the PCR results are still obtained 2 days faster than with traditional culture, which often takes up to 5 days. The emergence of community-acquired MRSA has highlighted the need for hospital-based active surveillance programs that are rapid and can respond to the changing

160

epidemiology of MRSA colonization and infection. In this regard, it must be remembered that the risk factors for MRSA colonization are shifting and are different for distinct geographic regions and patient populations. One particular advantage of the PCR assay we evaluated is its ability to test 92 specimens simultaneously; this makes it possible to screen an entire department of circa 40 patients and attending healthcare workers where MRSA is suspected to circulate. As each run is completed in 2 h, it is easy to complete multiple runs within 1 work day (8 h). The costs associated with the test include the following: purchase of a microcentrifuge, thermocycler and Elisa en pipetting robot apparatus, amounting to circa EUR 75,000; the per-test price of EUR 12 per run; and the cost of one technician for the days of work involved. Such costs must be weighed against those caused by an outbreak, which amount to approximately EUR 250,000 in our area. The outcome of this investigation convinced the administration board of our hospital to implement this PCR in the hospital laboratory’s routine for the purpose outlined above. Acknowledgment The authors acknowledge the assistance of their hospital’s infection control practitioners in collecting specimens and background data, Mrs. K. Hofman and Mrs. S. Costongs for secretarial assistance, and the medical microbiology technicians for performing bacteriological assays.

References 1. Wagenvoort JHT (2000) Dutch measures to control MRSA and the expanding European Union. Eurosurveillance 5:26–28 2. Dutch Working Party on Infection Prevention (2005) Policy for methicillin-resistant Staphylococcus aureus (in Dutch). http:// www.wip.nl 3. Wagenvoort JHT, De Brauwer EIGB, Sijstermans MLH, Toenbreker HMJ (2005) Risk of re-introduction of methicillinresistant Staphylococcus aureus into the hospital by intrafamilial spread from and to healthcare workers. J Hosp Infect 59:67–76 4. Wagenvoort JHT, De Brauwer E, Toenbreker H, Heck M, Wannet W (2002) Endemic MRSA shakes the Dutch “search and destroy”


5.

6.

7.

8.

9.

10.

11.

12. 13.

14.

strategy. Brit Med J 324:240b, http://bmj.com/cgi/eletters/324/ 7331/240/b Hagen RM, Seegmuller I, Navai J, Kappstein I, Lehn N, Miethke T (2005) Development of real-time PCR assay for rapid identification of methicillin-resistant Staphylococcus aureus from clinical samples. Int J Med Microbiol 295:77–86 Nilsson P, Alexandersson H, Ripa T (2005) Use of broth enrichment and real-time PCR to exclude the presence of methicillin-resistant Staphylococcus aureus in clinical samples: a sensitive screening approach. Clin Microbiol Infect 11:1027–1034 Warren DK, Liao RS, Merz LR, Eveland M, Dunne WM Jr (2004) Detection of methicillin-resistant Staphylococcus aureus directly from nasal swab specimens by a real-time PCR assay. J Clin Microbiol 42:5578–5581 Huletsky A, Lebel P, Picard FJ, Bernier M, Gagnon M, Bocher N, Bergeron MG (2005) Identification of methicillin-resistant Staphylococcus aureus carriage in less than 1 hour during a hospital surveillance program. Clin Infect Dis 40:976–981 Kaiser P, Reischl U, Miethke T, Lehn N, Schneider-Brachert W (2005) Comparison of two different real-time PCR-assays for rapid identification of MRSA directly from clinical swabs. In: Program and abstracts of the 45th ICCAC™ Washington. Abstract no. D-1714 Brown DF, Edwards DI, Hawkey PM, Morrison D, Ridgway GL, Towner KJ, Wren MW (2005) Guidelines for the laboratory diagnosis and susceptibility testing of methicillin-resistant Staphylococcus aureus (MRSA). On behalf of the joint working party of the British Society for Antimicrobial Chemotherapy, Hospital Infection Society and Infection Control Nurses Association. J Antimicrob Chemother 56:1000–1018 Wagenvoort JHT, Werink TJ, Gronenschild JMH, Davies BI (1996) Optimization of detection of methicillin-resistant Staphylococcus aureus phage type III-29. Infect Control Hosp Epidemiol 17:208–209 Robert Koch Institute (2004) Zur MRSA-Situationen in Deutschland im Jahr 2004. Epidemiol Bull 4:376–380 Becker K, Pagnier I, Schuhen B, Wenzelburger F, Friedrich AW, Kipp F, Peters G, von Eiff C (2005) Does nasal colonization by methicillin-resistant coaugulase-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 44:229–231 Daeschlein G, Assadian O, Daxboeck F, Kramer A (2006) Multiplex PCR-Elisa for direct detection of MRSA in nasal swabs advantageous for rapid identification of non-MRSA carriers. Eur J Clin Microbiol Infect Dis 25:328–330