Successful Control of Widespread Methicillin‐Resistant Staphylococcus aureus Colonization and Infection in a Large Teaching Hospital in The Netherlands
Objective. The low prevalence of infection and colonization with methicillin‐resistant Staphylococcus aureus (MRSA) in The Netherlands is ascribed to a national “search‐and‐destroy” policy. We describe the measures that were implemented to control widespread MRSA colonization and infection in a Dutch hospital.
Design. Descriptive intervention study.
Setting. Teaching medical center with a capacity of 679 beds, including 16 intensive care beds.
Interventions. MRSA colonization and infection were identified using conventional culture with a selective broth. Isolates were typed using pulsed‐field gel electrophoresis. Measures to control the epidemic included screening of contacts (patients and hospital staff), screening of patients at readmission or discharge, strict isolation of MRSA‐positive patients, decolonization of colonized staff and patients, the development of an electronic signal identifying MRSA‐positive patients, and the development of a culture information–system for hospital personnel.
Results. Awareness of uncontrolled dissemination of MRSA began in November 2001. Because the clone involved had a low minimum inhibitory concentration for oxacillin, at first it was not recognized as MRSA. In February 2002, when major screening efforts started, it appeared that MRSA had spread all over the hospital and that many staff members were colonized. By the end of December 2005, a total of 600 patients and 135 staff members were found to be newly colonized. The yearly incidence of cases of MRSA colonization and infection decreased from 351 in 2002 to 56 in 2005. Typing of the isolates showed that 3 MRSA clones were predominant. Outbreaks of colonization involving these clones did not occur after 2003.
Conclusion. Our observations show that strict application of “search‐and‐destroy” measures can effectively control a huge epidemic of MRSA colonization and infection.
Received November 27, 2006; accepted March 5, 2007; electronically published June 19, 2007.
Some believe that high rates of methicillin‐resistant Staphylococcus aureus (MRSA) infection and colonization are an inescapable consequence of modern health care worldwide, but some individual hospitals, and even whole countries, have maintained control over MRSA.1,2 In The Netherlands, the prevalence of MRSA infection and colonization in the community and in hospitals is among the lowest in the world. In 2000, the prevalence in the community was estimated at 0.03%.3 In hospitals, 0.3% of invasive S. aureus isolates were methicillin resistant.1,3 Still, The Netherlands and most Nordic countries maintain levels below or around 1%, whereas in 2004, 44% of the clinical isolates of S. aureus from the United Kingdom were methicillin resistant, and in 2003, 64% of nosocomial S. aureus infections in the Centers for Disease Control and Prevention National Nosocomial Infections Surveillance System hospitals were methicillin resistant.1,4,5
The low prevalence of MRSA infection and colonization in The Netherlands likely results from a national “search‐and‐destroy” policy, formally implemented in 1988. The restricted use of antibiotics may also help to control MRSA.6 The policy includes screening patients at risk for MRSA colonization, as well as hygienic measures, including strict isolation of colonized or infected patients, decolonization of patients, and identification and decolonization of MRSA‐positive hospital staff.7,8 MRSA contamination of the inanimate environment is addressed as well.
The required extent of “search‐and‐destroy” measures is debated because the approach is labor‐intensive, costly, and has important effects on patients and hospital staff. Less stringent alternative policies have been tried. Recently, Cepeda et al.9 studied the effect of single‐room isolation for MRSA‐colonized patients in intensive care units (ICUs). They concluded that isolation measures made no difference in the spread of MRSA. However, in the Cepeda et al. study, the screening of patients at ICU admission was incomplete and hospital staff were not screened at all. Furthermore, the use of barrier precautions was the same in the 2 groups studied, and hand hygiene compliance was only 21%.9,10
We hypothesize that MRSA can only effectively be suppressed to very low levels by a combination of measures. This was indeed the conclusion of a recent analysis that tested the effect of various measures by modelling.11 In this article, we describe a hospitalwide outbreak of MRSA colonization and infection in a large teaching hospital, involving 3 strains. Several measures to overcome the outbreak were taken simultaneously, and this resulted in the control of MRSA in the hospital within 2 years.
Methods
Setting
The study period extended from November 2001 through December 2005. The setting was the Medical Center Rijnmond‐Zuid. This teaching medical center, which includes 2 locations, provides care for approximately 260,000 inhabitants of Rotterdam, the second‐largest city in The Netherlands. Its 679‐bed capacity includes 16 intensive care beds.
MRSA Sampling Procedures
Patients were screened for MRSA colonization by collecting swab samples from the nares, throat and perineum. If applicable, other sites were swabbed as well, such as wounds or the insertion sites of vascular catheters. If MRSA grew in cultures of samples from any of these sites , the patient was classified as MRSA‐positive. Hospital staff were screened for MRSA colonization by collection of swab samples from the nares and throat, and, if applicable, from wounds or other skin defects. We preferred to obtain swab samples at the start of a staff member's shift.
Microbiological Methods
Swab samples were cultured according to standard methods, ie, they were inoculated in a selective phenol‐red mannitol broth, as described elsewhere.12 After 48 hours of incubation at 35°C, a 5‐μL loop of broth was subcultured onto a blood agar plate. After 24 and after 48 hours of incubation, suspected colonies were tested with an agglutination test (StaphaurexPlus; Remel). In case of doubt, an S. aureus–specific DNA hybridization test was used (Accuprobe; Gen‐Probe). Oxacillin (later cefoxitin) susceptibility was tested by disk diffusion, in accordance with guidelines of the Clinical and Laboratory Standards Institute (formerly the National Committee on Clinical and Laboratory Standards [NCCLS]).13 The identity of isolates that had a minimum inhibitory concentration of oxacillin and/or cefoxitin compatible with resistance or intermediate susceptibility was confirmed with a DNA hybridization test. The presence of penicillin‐binding protein 2 (PBP2′) was determined with the PBP2′–latex agglutination test (Oxoid). From March 2002 onward, a mecA gene polymerase chain reaction was also available. The isolate recovered from each newly identified MRSA‐positive individual was sent to the national reference laboratory at the National Institute for Public Health and the Environment (RIVM), Bilthoven, The Netherlands, where it was typed by pulsed‐field gel electrophoresis (PFGE) and assigned an MRSA‐cluster number.14 Persons who had sequential positive test results for the same MRSA strain were counted once in analyses, at the date of first MRSA isolation. Persons from whom 2 (or more) PFGE types were recovered were counted twice (or more).
Measures to Control the Epidemic
Before the outbreak was recognized, no surveillance cultures were performed. Screening for MRSA infection or colonization was confined to patients who had been admitted from facilities outside The Netherlands. Hospital guidelines were used to restrict the prescription of antibiotics. Before the outbreak, from January through October 2001, 14 new cases of infection or colonization with MRSA had been identified.
Until March 2002, the hospital employed only 2 (part‐time) infection control practitioners and 1 clinical microbiologist. As this workforce did not meet the hospital's infection control needs, assistance was provided by the neighboring Erasmus Medical Center, University Medical Center Rotterdam. For 3 months, an infection control practitioner and a clinical microbiologist from this center were added to the staff. All measures subsequently introduced to control the epidemic are listed in the Table, and some are clarified below.
Dedicated space and laboratory resources. Because of a shortage of negative pressure–ventilation isolation rooms, an entire ward was designated as an MRSA ward in February 2002. Later, an MRSA dialysis unit and an MRSA outpatient clinic were added, both located outside the hospital building to minimize opportunities for cross‐infection with MRSA. It was necessary to reopen an old laboratory in April 2002 to accommodate the greatly increased demand for screening cultures, ranging from 54,000 to 86,000 cultures annually (in 2002 and 2003, respectively), an extra workload that required up to 5 technicians working 7 days a week. (Figure 1).
Figure 1. Time line of the outbreak of methicillin‐resistant Staphylococcus aureus (MRSA) colonization and infection. Both patients and staff were tested; positivity rates are for newly identified cases.
Control of transmission and MRSA eradication. In the hospital information system, a signal system to identify MRSA‐positive patients was developed. Before admission and outpatient contacts, healthcare workers could thus verify the MRSA status of a patient and set up isolation, if necessary.
If a new patient was identified as MRSA positive while on a ward, the patient was isolated, and everyone in the ward, including staff, was screened for MRSA infection or colonization immediately. If transmission of MRSA was detected, involving either a patient or staff member, the ward was closed to new admissions until all individuals who were sources of MRSA were identified and either isolated or discharged from the ward. The ward was only reopened after thorough environmental disinfection. Colonized staff members were considered a source of MRSA. Initially, some staff members were reluctant to provide samples for culture. Noncompliance was addressed in staff meetings, where it was stressed that screening serves a communal interest and only works if everyone complies. We thus managed to screen nearly 100% of the staff.
Whenever MRSA colonization was identified in a staff member, he or she was sent home after a new set of swab samples were obtained. This new set included samples from the perineum, to verify carrier status and extent of colonization. At this point, eradication therapy was started, consisting of 5 days of treatment with mupirocin nasal ointment and a triclosan‐containing soap for washing the body and hair. If the new set of swab samples obtained before the start of treatment were negative for MRSA on culture, the staff member could resume work. If not, swab samples were obtained 10, 15, and 20 days after the start of treatment. If any of these swab samples were MRSA positive on culture, a 7‐day course of therapy with 2 oral antibiotics effective against the particular isolate (generally rifampicin and trimethoprim) was added to the treatment. Multiple antibiotics were used to prevent the development of antibiotic resistance and to improve the treatment success rate (which was close to 100%). If the swab samples obtained on day 10 of treatment were MRSA negative on culture, the worker was allowed to resume work. Two additional sets of negative culture results were required to confirm the success of the eradication treatment. Similar MRSA eradication therapy was instituted for patients who did not have contraindications to therapy (ie, patients with no skin defects or catheters present). Eradication therapy was administered to patients to reduce the opportunity for MRSA transmission in the hospital, as well as in the community.
MRSA screening. From the beginning of the epidemic, contacts of affected patients and staff had been screened. Beginning in April 2002, several rounds of screening cultures were performed for all hospital staff with patient contact, to exclude the possibility of undetected MRSA sources among hospital personnel. Critical care wards with a relatively high incidence of MRSA colonization (eg, the ICU and hemodialysis units) were temporarily screened weekly. Because surrounding hospitals worried about MRSA infection and colonization in patients transferred from the Medical Center Rijnmond‐Zuid, all patients from previously affected wards were screened at discharge. This also served as surveillance tool.
After laboratory capacity was increased, former patients who had been admitted during the first months of the epidemic were approached through the mail for MRSA screening. About 800 former patients responded to this request for self‐administered nose and throat swab samples (response rate, 70%), which resulted in 13 additional diagnoses of MRSA colonization.
Results
Onset and Course of the Epidemic
In November 2001, MRSA was isolated from a patient who was transferred from the Medical Center Rijnmond‐Zuid to another hospital. This finding was reported to the microbiologist at the Medical Center Rijnmond‐Zuid. Subsequently, limited contact screening was started, which was gradually broadened when it became clear that additional patients were colonized with MRSA.
A complicating factor was the fact that the strain could not easily be identified as MRSA because tests for antibiotic resistance produced inhibition zones around oxacillin disks that were compatible with methicillin susceptibility. In addition, testing with the Vitek 1 system (bioMérieux) showed a low oxacillin minimal inhibitory concentration of 1‐8 μg/mL. Thus, the strain had the phenotype of methicillin susceptible S. aureus, even though later analysis showed that it contained the mecA gene. Only after this became clear was the strain found to be present across the hospital. By systematic screening of contacts, starting in February 2002, many new cases of MRSA colonization were identified among patients and also among hospital staff (Figure 2). The strain was identified as PFGE type 16 by the RIVM reference lab. This PFGE type is similar to the Berlin epidemic clone or the USA600 strain.15
Figure 2. Monthly incidence of newly identified methicillin‐resistant Staphylococcus aureus (MRSA)‐positivity during the outbreak
The exact time of introduction of this MRSA clone remains unknown, but the February 2002 finding of a substantial number of colonized hospital staff, from various departments, suggests that the clone had been around for some time. During the first quarter of 2002, the incidence of new cases of MRSA colonization increased, mainly as a result of improved case finding and a dramatic increase in the number of persons screened (Figures 1 and 2), from 101 (of which 21 [21%] were MRSA‐ positive) during November through December 2001 to a maximum of 3,085 (of which 17 [0.55%] were newly positive) screened in August 2002 .
Molecular typing of strains showed that isolates of cluster 16 were initially predominant, but subsequently 2 other PFGE types, 37 and 38, were found to have spread. These types were confirmed to differ from each other and from type 16 by additional typing methods performed by the reference laboratory. Thus, 3 epidemic strains were present simultaneously, accounting for 94% of all MRSA isolates until April 2002. Most MRSA isolates from the epidemic strains were susceptible to rifampicin, tetracycline, trimethoprim‐sulfamethoxazole, and fusidic acid. All cluster 16 isolates were susceptible to aminoglycosides, and some to quinolones, and all cluster 37 and 38 isolates were resistant to these antibiotic classes.
With increased screening, eradication of MRSA from colonized hospital staff, and other measures, the number of newly colonized patients and staff members gradually decreased until September 2002, when transmission among patients recurred (Figure 2). After identifying several MRSA‐positive patients in the ICU, equipment and materials were screened and found to be MRSA‐positive (data not shown). Thus, environmental contamination was clearly an issue that needed to be addressed.
The percentage of new cases of MRSA colonization and infection involving the epidemic types (16, 37, and 38) gradually decreased, from 89% in 2002 to 34% in 2004 and to 14% in 2005. Other types were isolated mainly from admittance screening cultures performed for patients who had previously been hospitalized abroad. The incidence of MRSA colonization and infection per 1,000 admissions varied from fewer than 0.5 cases in some departments (such as pediatrics, obstetrics, and gynecology) to more than 5 in others (eg, 6.4 cases in internal medicine, and 10.7 cases in surgery).
Discussion
Our results show that, through a combination of measures, a large outbreak of MRSA colonization and infection can be controlled. A first prerequisite is adherence to basic principles of infection control. Secondly, to get results quickly, maximum efforts should be made to prevent further transmission. Therefore, institutions should have or create isolation facilities, and high‐risk patients should be isolated and screened for MRSA. The use of barrier precautions in a standard single room could be helpful when there are insufficient negative‐pressure isolation rooms available.
Communication lines between the wards and the microbiology laboratory should be short, and results should be readily available. As patients colonized with MRSA often need readmission, we found it imperative to have a signal identifying MRSA carriage in the patient information system. In addition, it proved helpful to screen patients before admission or at discharge. The burden of patient isolation can be further reduced if culture results are made available quickly.
In addition to patients, hospital staff can serve as a source of MRSA transmission; in our study, we found that many were colonized early in the outbreak (Figure 2).16,17 Because the identification of MRSA was initially incomplete, the epidemic strain continued to circulate in the hospital, thereby allowing it to colonize some of the staff. Therefore, to control an outbreak of colonization, identification and treatment of staff who are carrying MRSA should be complete. The entire staff should feel responsible for achieving this goal—not only infection control practitioners. Stigmatization of and any other harm to staff members who are MRSA‐positive should be prevented. Fortunately, decolonization of otherwise healthy persons without skin lesions is usually successful. Recolonization that occurs soon after treatment may well come from an MRSA‐positive spouse.
In 2005, a total of 56 newly MRSA‐positive individuals were detected, which was higher than the number detected before the epidemic started. We nevertheless think that the epidemic was over for the following reasons: (1) the epidemic strains had almost disappeared; (2) most new cases of colonization involved unrelated strains; (3) transmission became a rare finding; (4) only a few healthcare workers were found to be MRSA positive, resulting from a low infectious pressure in the hospital; (5) far more persons were screened at this late stage than before the epidemic started, and we are convinced that both the diagnostic accuracy rate for MRSA colonization and the MRSA recovery rate dramatically improved during the epidemic; and (6) former patients who did not receive a diagnosis of MRSA colonization (but were nonetheless colonized) and patients infected by former patients in our catchment area may have raised the prevalence.
Initially, the laboratory capacity did not permit extensive screenings. Therefore, only the persons at highest risk were screened, and case finding remained incomplete, as it did among former patients, who often need readmission. This category cannot be ignored, which is illustrated by the finding of 13 new cases of MRSA colonization among 800 former patients, who had been hospitalized up to 6 months earlier. The hospital board decided to continue screening at discharge, a measure in addition to those specified by the Dutch guidelines. Though this practice was initially started to diagnose transmission unnoticed during hospitalization, we realize that most of the case patients found through these screenings were probably already MRSA positive on admission, because these findings usually concern single cases.
We do not know how much each of the measures taken contributed to the control of MRSA, but we are convinced they have to be taken together and have to be implemented completely to get quick and lasting results. Several studies show the beneficial effect of a single measure (eg, active detection and isolation of colonized patients) that reduced the prevalence of MRSA infection and/or colonization.18,19 But our aim was more ambitious, as we wanted to eliminate MRSA transmission in the hospital. Controlling MRSA was expensive, at an estimated cost of €3 million, including loss of productivity.20 Though the costs of “search‐and‐destroy” strategies are considerable, this strict regimen is considered cost effective in The Netherlands.21
Could our methods work in countries where MRSA is highly endemic? We cannot fairly compare our setting with hospitals in those countries. Control of high endemicity requires greater capacity for isolation of patients and screening of all patients at, or preferably before, admission. And, though this may require a paradigm shift, screening and treatment of personnel is important, particularly in situations involving epidemic clones and situations in which nosocomial transmission is observed.
Acknowledgments
We thank Prof. Dr. H. A. Verbrugh and Dr. M. C. Vos for critically reading the manuscript.
Potential conflicts of interest. All authors report no conflicts of interest relevant to this article.
References
- 1. European Antimicrobial Resistance Surveillance System (EARSS) Annual Report. EARSS‐2003. Bilthoven, The Netherlands: the National Institute for Public Health and the Environment (RIVM); 2004.
- 2. Boyce JM, Cookson B, Christiansen K, et al. Methicillin‐resistant Staphylococcus aureus. Lancet Infect Dis 2005; 5:653‐663.
- 3. Wertheim HF, Vos MC, Boelens HA, 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; 56:321‐325.
- 4. National Nosocomial Infections Surveillance (NNIS) System Report, data summary from January 1992 through June 2003, issued August 2003. Am J Infect Control 2003; 31:481‐498.
- 5. Klevens RM, Edwards JR, Tenover FC, McDonald LC, Horan T, Gaynes R. Changes in the epidemiology of methicillin‐resistant Staphylococcus aureus in intensive care units in US hospitals, 1992‐2003. Clin Infect Dis 2006; 42:389‐391.
- 6. Goossens H, Ferech M, Vander SR, Elseviers M. Outpatient antibiotic use in Europe and association with resistance: a cross‐national database study. Lancet 2005; 365:579‐587.
- 7. Dutch Working Party on Infection Prevention (STWIP). Policy for methicillin‐resistant Staphylococcus aureus. Guideline 35A. 1994. Available at: http://www.wip.nl. Accessed November 23, 2006.
- 8. Verhoef J, Beaujean D, Blok H, et al. A Dutch approach to methicillin‐resistant Staphylococcus aureus. Eur J Clin Microbiol Infect Dis 1999; 18:461‐466.
- 9. Cepeda JA, Whitehouse T, Cooper B, et al. Isolation of patients in single rooms or cohorts to reduce spread of MRSA in intensive‐care units: prospective two‐centre study. Lancet 2005; 365:295‐304.
- 10. Farr BM. What to think if the results of the National Institutes of Health randomized trial of methicillin‐resistant Staphylococcus aureus and vancomycin‐resistant Enterococcus control measures are negative (and other advice to young epidemiologists): a review and an au revoir. Infect Control Hosp Epidemiol 2006; 27:1096‐1106.
- 11. Bootsma MC, Diekmann O, Bonten MJ. Controlling methicillin‐resistant Staphylococcus aureus: quantifying the effects of interventions and rapid diagnostic testing. Proc Natl Acad Sci U S A 2006; 103:5620‐5625.
- 12. Wertheim H, Verbrugh HA, van Pelt C, de Man P, van Belkum A, Vos MC. Improved detection of methicillin‐resistant Staphylococcus aureus using phenyl mannitol broth containing aztreonam and ceftizoxime. J Clin Microbiol 2001; 39:2660‐2662.
- 13. National Committee on Clinical and Laboratory Standards (NCCLS). Performance Standards for Antimicrobial Susceptibility Testing: 12th Informational Supplement. Wayne, PA: NCCLS; 2002:M100‐S12.
- 14. Tenover FC, Arbeit R, Archer G, et al. Comparison of traditional and molecular methods of typing isolates of Staphylococcus aureus. J Clin Microbiol 1994; 32:407‐415.
- 15. Wannet WJ, Spalburg E, Heck ME, Pluister GN, Willems RJ, de Neeling AJ. Widespread dissemination in The Netherlands of the epidemic Berlin methicillin‐resistant Staphylococcus aureus clone with low‐level resistance to oxacillin. J Clin Microbiol 2004; 42:3077‐3082.
- 16. Lessing MP, Jordens JZ, Bowler IC. Molecular epidemiology of a multiple strain outbreak of methicillin‐resistant Staphylococcus aureus amongst patients and staff. J Hosp Infect 1995; 31:253‐260.
- 17. Saiman L, Cronquist A, Wu F, et al. An outbreak of methicillin‐resistant Staphylococcus aureus in a neonatal intensive care unit. Infect Control Hosp Epidemiol 2003; 24:317‐321.
- 18. Huang SS, Yokoe DS, Hinrichsen VL, et al. Impact of routine intensive care unit surveillance cultures and resultant barrier precautions on hospital‐wide methicillin‐resistant Staphylococcus aureus bacteremia. Clin Infect Dis 2006; 43:971‐978.
- 19. Thompson RL, Cabezudo I, Wenzel RP. Epidemiology of nosocomial infections caused by methicillin‐resistant Staphylococcus aureus. Ann Intern Med 1982; 97:309‐317.
- 20. de Jong B, Verbrugh HA. Ramp in het ziekenhuis. Med Contact 2003;15.
- 21. Vriens M, Blok H, Fluit A, Troelstra A, Van Der WC, Verhoef J. Costs associated with a strict policy to eradicate methicillin‐resistant Staphylococcus aureus in a Dutch University Medical Center: a 10‐year survey. Eur J Clin Microbiol Infect Dis 2002; 21:782‐786.