mrsa - the hospital infection society

Journal of Hospital Infection (2006) 63S, S1eS44

www.elsevierhealth.com/journals/jhin

Guidelines for the control and prevention ofmeticillin-resistant Staphylococcus aureus(MRSA) in healthcare facilities*

J.E. Coia a, G.J. Duckworth b, D.I. Edwards c, M. Farrington d,C. Fry e, H. Humphreys f,*, C. Mallaghan g, D.R. Tucker h,for the Joint Working Party of the British Society of AntimicrobialChemotherapy, the Hospital Infection Society, and the InfectionControl Nurses Association

a Department of Bacteriology, Glasgow Royal Infirmary, Glasgow, UKb Health Protection Agency, London, UKc 39 Wallenger Avenue, Gidea Park, Essex, UKd Clinical Microbiology and Public Health Laboratory, Health Protection Agency,Addenbrooke’s Hospital, Cambridge, UKe Department of Health, London, UKf Department of Microbiology, RCSI Education and Research Centre, Beaumont Hospital,Dublin, Irelandg East Midlands (South) Health Protection Agency, Enderby, Leicester, UKh Department of Infection Control, St Thomas’ Hospital, London, UK

Available online 3 April 2006

KEYWORDSStaphylococcusaureus; Methicillinresistance; Meticillin;Cross infection;Infection control;Handwashing;Decontamination;Populationsurveillance; Diseasereservoirs; Vancomycinresistance; Microbialdrug resistance;

Summary Meticillin-resistant Staphylococcus aureus (MRSA) remains en-demic in many UK hospitals. Specific guidelines for control and preventionare justified because MRSA causes serious illness and results in significantadditional healthcare costs. Guidelines were drafted by a multi-disciplinarygroup and these have been finalised following extensive consultation. Therecommendations have been graded according to the strength of evidence.Surveillance of MRSA should be undertaken in a systematic way and shouldbe fed back routinely to healthcare staff. The inappropriate or unneces-sary use of antibiotics should be avoided, and this will also reduce the like-lihood of the emergence and spread of strains with reduced susceptibilityto glycopeptides, i.e. vancomycin-intermediate S. aureus/glycopeptide-intermediate S. aureus (VISA/GISA) and vancomycin-resistant S. aureus

* In this document, ‘meticillin’ has been used in place of the established ‘methicillin’ in accordance with the new InternationalPharmacopoeia guidelines.* Corresponding author. Tel.: þ3531 8093708/3710; fax: þ3531 8093709.

E-mail address: [email protected]

0195-6701/$ - see front matter ª 2006 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.doi:10.1016/j.jhin.2006.01.001

mailto:[email protected]

http://www.elsevierhealth.com/journals/jhin

S2 J.E. Coia et al.

Workload; Antisepsis;Colonisation;Antibiotic stewardship;Patient isolation

(VRSA). Screening for MRSA carriage in selected patients and clinical areasshould be performed according to locally agreed criteria based upon as-sessment of the risks and consequences of transmission and infection.Nasal and skin decolonization should be considered in certain categoriesof patients. The general principles of infection control should be adoptedfor patients with MRSA, including patient isolation and the appropriatecleaning and decontamination of clinical areas. Inadequate staffing, espe-cially amongst nurses, contributes to the increased prevalence of MRSA.Laboratories should notify the relevant national authorities if VISA/GISAor VRSA isolates are identified.ª 2006 The Hospital Infection Society. Published by Elsevier Ltd. All rightsreserved.

1. Preamble

Guidelines for the control of meticillin-resistantStaphylococcus aureus (MRSA) infections in hospi-tals in the UK have been published previously bya Joint Working Party of the British Society for Anti-microbial Chemotherapy and the Hospital InfectionSociety in 19861 and 1990,2 and together with the In-fection Control Nurses Association in 1998.3 Withthe increased media and public interest, the adventof glycopeptide-resistant S. aureus and new drugs,including linezolid and teicoplanin, the Departmentof Health’s Specialist Advisory Committee on Anti-microbial Resistance (SACAR) asked the three socie-ties to revise the guidelines. SACAR also requestedan enhanced focus on: (i) prophylaxis and therapyof MRSA infections; (ii) the laboratory diagnosisand susceptibility testing of MRSA; and (iii) the pre-vention and control of MRSA infections in the UK.The last is the subject of this report. The first twowill be published elsewhere. These guidelines ex-clude evidence and recommendations for MRSA inpaediatric, neonatal and dental patients for whominsufficient evidence exists. Where possible, rec-ommendations have been given based on the evi-dence available, even though the evidence basemay be poor. Most of the evidence reviewed con-cerns acute care settings. Nonetheless, many ofthe recommendations and principles in the guide-lines will apply in other healthcare settings.

A systematic review was conducted covering theliterature from the beginning of 1996 to the end ofJune 2004, thus focusing on the period since thepreparation of the last guidelines. This review hasalso been published in this supplement. Data sour-ces included MEDLINE, EMBASE, the CumulativeIndex of Nursing and Allied Health Literature, theCochrane Clinical Trials Register, the NationalHealth Service (NHS) Centre for Reviews and Dis-semination Database of Reviews of Effectiveness,

and the Health Management Information Consor-tium Database. The focus of the review centred onthe following questions:

e To what extent does the screening of patientsbefore or on admission to hospital reduce theincidence of MRSA transmission and what arethe costs?

e To what extent does the use of MRSA surveil-lance data reduce the incidence of MRSA trans-mission and what are the costs?

e To what extent does the isolation or cohortingof patients prevent the spread of MRSA andwhat are the costs?

e To what extent does environmental cleaningwith detergent or detergent plus disinfectantcontribute to the control of MRSA infectionand what are the costs?

2. Grades of evidence andrecommendations

Each recommendation, as graded by the US Cen-ters for Disease Control and Prevention (CDC), iscategorized on the basis of existing scientific data,theoretical rationale, applicability and economicimpact. These grades were chosen in preferenceto those published by the Scottish IntercollegiateGuidelines Network or the National Institute forClinical Excellence as they include scientific evi-dence and are not exclusively clinical. The CDC/Hospital Infection Control Practices Advisory Com-mittee (HICPAC) system for categorizing recom-mendations is as follows.

e Category 1a. Strongly recommended forimplementation and strongly supported bywell-designed experimental, clinical or epide-miological studies.

Guidelines for MRSA in healthcare facilities S3

e Category 1b. Strongly recommended for imple-mentation and strongly supported by certainexperimental, clinical or epidemiological stud-ies and a strong theoretical rationale.

e Category 1c. Required for implementation, asmandated by federal or state regulation orstandard. The UK equivalent is to operatewithin European Union or UK Health & SafetyLegislation.

e Category 2. Suggested for implementation andsupported by suggestive clinical or epidemio-logical studies or a theoretical rationale.

e No recommendation. Unresolved issue. Practicesfor which insufficient evidence exists or for whichthere is no consensus regarding efficacy.

2.1. Surveillance

Surveillance must be undertaken routinely as part ofthe hospital’s infection control programme and mustbe a recognized element of the clinical governanceprocess. As such, there should be clear arrangementsidentifying those responsible for acting on the resultsin individual hospital directorates (Category 1b).

For benchmarking purposes, surveillance datashould be collected and reported in a consistentway, to agreed case definitions and using agreedspecialty activity denominators, with stratificationaccording to case mix (Category 1b).

Surveillance data should be fed back to hospitalstaff routinely, readily intelligible to most hospitalstaff, considered regularly at hospital senior man-agement committees, and used in local infectioncontrol training.

MRSA surveillance should include:

e any mandatory requirements (Category 1c);e results of microbiological investigations for

clinical purposes (Category 1b); ande results of microbiological investigations under-

taken for screening purposes (Category 1b).

The dataset should include:

e patient, laboratory, unit/ward and hospitalidentifiers;

e patient demographics (address, age, sex);e date of admission;e date of onset of infection (if appropriate);e site of the primary infection, if appropriate (if

bacteraemia, source of the bacteraemia);e date specimen taken;e site of specimen (blood culture, wound, etc.);e where the MRSA was acquired (hospital, com-

munity, specialty, etc.);

e whether part of an outbreak; ande antimicrobial susceptibilities.

Other desirable items include the primary di-agnosis, an assessment of severity of underlyingillnesses, prior antimicrobial therapy and possiblerisk factors for infection (Category 2).

2.2. Antibiotic stewardship

e Avoidance of inappropriate or excessive antibi-otic therapy and prophylaxis in all healthcaresettings (Category 1a).

e Ensuring that antibiotics are given at the cor-rect dosage and for an appropriate duration(Category 1b).

e Limiting the use of glycopeptide antibiotics tosituations where their use has been shown tobe appropriate. If possible, prolonged coursesof glycopeptide therapy should be avoided(Category 1a).

e Reducing the use of broad-spectrum antibi-otics, particularly third-generation cephalo-sporins and fluoroquinolones, to what isclinically appropriate (Category 1b).

e Instituting antibiotic stewardship programmesin healthcare facilities, key components ofwhich include the identification of key person-nel who are responsible for this, surveillance ofantibiotic resistance and antibiotic consump-tion, and prescriber education (Category 1c).

2.3. Screening

Active screening of patients for MRSA carriageshould be performed and the results should belinked to a targeted approach to the use ofisolation and cohorting facilities (Category 2).

Certain high-risk patients should be screenedroutinely, and certain high-risk units should bescreened at least intermittently in all hospitals.The fine detail regarding which patients arescreened should be determined locally by the in-fection control team and must be discussed with theappropriate clinical teams and endorsed by therelevant hospital management structure. They willbe influenced by the local prevalence of MRSA in thehospital and unit concerned, the reason for admis-sion of the patient, the risk status of the unit towhich they are admitted, and the likelihood that thepatient is carrying MRSA. Patients at high risk ofcarriage of MRSA include those who are:

e known to have been infected or colonized withMRSA in the past (Category 1b);

S4 J.E. Coia et al.

e frequent re-admissions to any healthcare facil-ity (Category 1b);

e direct interhospital transfers (Category 1b);e recent inpatients at hospitals abroad or hospi-

tals in the UK which are known or likely tohave a high prevalence of MRSA (Category1b); and

e residents of residential care facilities wherethere is a known or likely high prevalence ofMRSA carriage (Category 1b).

Other risk groups may be defined by localexperience, based on screening initiatives or out-break epidemiology. Published examples have in-cluded injecting drug users, patients infected withhuman immunodeficiency virus, and members ofprofessional contact sport teams (Category 2).

Units caring for patients at high risk for sufferingserious MRSA infections or with a high proportionof MRSA infections among colonized patients in-clude: intensive care, neonatal intensive care,burns, transplantation, cardiothoracic, orthopae-dic, trauma, vascular surgery, renal, regional,national and international referral centres (allCategory 1b), and other specialist units as de-termined by the infection control team and asagreed with the senior clinical staff of the unitsand relevant hospital management structure.

Patients on elective surgical units (e.g. ortho-paedic, vascular), usually with short inpatientstays, are at lower risk of MRSA acquisition thanpatients on trauma and emergency units, or mixedunits. Due account of these differences should betaken when local screening policies are beingestablished (Category 2).

All patients who are at high risk for carriage ofMRSA should be screened at the time of admissionunless they are being admitted directly to isolationfacilities and it is not planned to attempt to clearthem of MRSA carriage (Category 2).

Regular (e.g. weekly or monthly, according tolocal prevalence) screening of all patients onhigh-risk units should be performed routinely(Category 2).

In addition, screening all patients (regardlessof their risk-group status) should be consideredon admission to high-risk units (Category 2). Thedecision about whether or not to perform routineadmission screening should be made explicitly bythe infection control team in consultation withthe senior clinical staff of the units, and shouldbe agreed with the relevant hospital managementstructure. Such ‘blanket’ screening may be usedintermittently, and may be especially worthwhileif the local prevalence of MRSA carriage in suchpatients is higher than usual for the UK, if there

are sufficient local isolation/cohorting resourcesto manage carriers effectively, and if local poli-cies for clearance of carriage and/or use ofsurgical prophylaxis with glycopeptides are inplace.

The following sites should be sampled forpatients (Category 1b): anterior nares, skin lesionsand wounds and sites of catheters, catheter urine,groin/perineum, tracheostomy and other skinbreaks in all patients, and sputum from patientswith a productive cough. The umbilicus should besampled in all neonates. One should also considersampling the throat.

The decision whether to perform screening ofpatients on admission to other wards or regularscreening of inpatients on other wards should bedecided by the local infection control team inconsultation with the senior clinical staff of theunits, and as agreed with the relevant hospitalmanagement structure (Category 2). In principle,hospitals with significant problems with MRSAtransmission or a high prevalence of MRSA car-riage or infection should consider performingmore widespread and regular screening than unitswith a low prevalence. However, this approachhas resource implications and should first be usedin areas where the clinical impact of high MRSAprevalence is highest (i.e. in the ‘high-risk’clinical areas). The aim is to identify all positivepatients within the hospital to allow targeting ofisolation and cohorting facilities in order tominimize the risk of onward transmission to otherpatients.

When possible, patients awaiting elective ad-mission who satisfy local requirements for screen-ing should be screened before admission by theirgeneral practitioners or in pre-admission clinics(Category 2). Patients who are at high continuingrisk of acquiring MRSA between the time of pre-admission screening and that of admission (e.g.they reside in a residential care facility which isknown to have a high prevalence of MRSA) must berescreened on admission, and should be isolated orcohorted according to policies in place on theadmitting unit until both sets of screening resultsare known.

Action to be taken if screeningresults are positiveIn general, detection of patients colonized orinfected with MRSA on a ward should be anindication for increased screening (Category 2).Little evidence exists to guide the details of anappropriate response, but this should be influ-enced by the risk group of the affected unit, bythe number of newly detected MRSA-positive


patients, by the adequacy of nurse numbers tostaff the ward, and by the availability of isolationand cohorting facilities. There is always a delaybetween MRSA acquisition by a patient and itspresence being detectable by screening samples,so it is recommended that at least three screens atweekly intervals should be performed beforea patient can be considered to be at low risk ofhaving acquired MRSA if they have been nursed inproximity to unknown and unisolated MRSA-positive patients or by the same staff (Category 2).The screening for MRSA in each unit within ahospital should be the subject of regular audit,with the results reviewed by the hospital’s in-fection control committee. The results should alsobe made available to management.

Screening of staff is not recommended routinely,but if new MRSA carriers are found among thepatients on a ward, staff should be asked about skinlesions. Staff with such lesions should be referredfor screening and for consideration of dermatolog-ical treatment by the relevant occupational healthdepartment (Category 1b). Staff with persistentcarriage at sites other than the nose should beconsidered for referral for appropriate specialistmanagement (e.g. ear, nose and throat; dermatol-ogy) who should arrange follow-up screeningaccording to local protocols (Category 1b).

Staff screening is indicated if transmissioncontinues on a unit despite active control mea-sures, if epidemiological aspects of an outbreakare unusual, or if they suggest persistent MRSAcarriage by staff (Category 2).

Care is needed to distinguish between transientcarriage (i.e. nasal carriage which is lost withina day or so of removal from contact with MRSA-positive patients and carries little risk of onwardtransmission) and prolonged carriage (especiallyassociated with skin lesions) (Category 1b). This isusually best achieved by screening staff as theycome on duty at the beginning of their shift andnot as they leave at the end of their shift.

Nurses, doctors, physiotherapists, other alliedhealth professionals and non-clinical support staff(e.g. porters) should be considered for screening,and the implications for onward spread by staffworking on other wards should also be considered(Category 2).

The special difficulties and risks posed byagency and locum staff should be considered(Category 1b).

Appropriate sampling sites for staff screeninginclude anterior nares, throat and any areas ofabnormal or broken skin (Category 1b). As aguide to use of eradication measures, oneshould consider screening the hairline and groin/

perineum of staff members found to be MRSApositive.

It is recommended that a minimum of threescreens at weekly intervals, while not receivingantimicrobial therapy, should be performed beforea previously positive staff member can be consid-ered to be clear of MRSA (Category 2). Localpolicies should be developed to guide postclear-ance sampling of staff (Category 2), and due noteshould be taken of the individual’s risk of trans-mission to patients when agreeing their continua-tion or return to work. In principle, only staffmembers with colonized or infected hand lesionsshould be off work while receiving courses ofclearance therapy.

No recommendation is made about performanceof ‘discharge screening’.

Performance of active screening for MRSA ineach unit within a hospital must be the subject ofregular audit, with the results reviewed andminuted by the hospital’s infection control com-mittee and made available to the appropriatehospital management structure (Category 1b).

Units with highly prevalent, endemic MRSAshould consider focusing screening, control mea-sures and other resources on high-risk units atfirst, with the intention of rolling them out tolower-risk areas after the position has improved(Category 2). Screening should not be seen as anend in itself, but rather it should be linked tospecific, locally determined packages of controlmeasures.

Geographically adjacent healthcare facilities,and those exchanging large numbers of patientsbecause of clinical links, should liaise to agreecommon and efficient screening measures thatshould be linked to common and efficient controlmeasures (Category 2). Such links should capitalizeon any developing networking relationships amongclinical and laboratory units, such as those en-couraged through the Pathology Modernizationinitiative.

Results of screening cultures should be madeavailable promptly to the clinical and infectioncontrol teams of other healthcare facilities towhom a patient is to be, or has recently been,transferred (Category 1b). Refusal to accept trans-fer of a patient is not justifiable on the basis of therisk posed to other patients by an individual’scarriage of or infection with MRSA. All units shouldhave procedures in place and adequate facilitiesfor containment of MRSA.

Trusts should develop local protocols for inform-ing patients, carers, relatives and staff membersof their MRSA status with due regard for confiden-tiality (Category 2).

S6 J.E. Coia et al.

2.4. Decolonization

Nasal decolonizationPatients receiving prophylaxis for an operativeprocedure and in an outbreak situation under theadvice of the infection control team should un-dergo nasal decolonization. This should beachieved by applying mupirocin 2% in a paraffinbase to the inner surface of each nostril (anteriornares) three times daily for five days. The patientshould be able to taste mupirocin at the back ofthe throat after application (Category 1b).

Mupirocin should not be used for prolongedperiods or used repeatedly (i.e. for more thantwo courses for five days) as resistance may beencouraged (Category 1a).

Nasal decolonization using topical nasal mupir-ocin should be used with other forms of interventionsuch as skin decolonization with 4% chlorhexidinegluconate aqueous solution (Category 2).

Throat decolonizationSystemic treatment should only be prescribed onthe advice of the consultant microbiologist in thehospital, with appropriate monitoring [e.g. regularliver function tests (LFTs) to monitor effects of thedrugs on the liver]. If treatment is required, thisshould be restricted to one course of treatment,the course should not be repeated and the possibleside-effects should be explained to the patient(Category 1b).

Systemic treatment should be given in conjunc-tion with nasal mupirocin and skin decolonization(Category 1b).

Local treatment for throat carriage such asantiseptic gargles or sprays may be used to reducethe organism load (no recommendation).

Skin decolonizationSkin decolonization using 4% chlorhexidine body-wash/shampoo, 7.5% povidone iodine or 2% triclo-san is useful in eradicating or suppressing skincolonization for short times, particularly pre-operatively to reduce the risk of surgical siteinfections (Category 1a).

Patients should bathe daily for five days withthe chosen antiseptic detergent. The skin shouldbe moistened and the antiseptic detergent shouldbe applied thoroughly to all areas before rinsing inthe bath or shower. Special attention should bepaid to known carriage sites such as the axilla,groin and perineal area. The antiseptic should alsobe used for all other washing procedures and forbed bathing. Hair should be washed with anantiseptic detergent (Category 1a).

After satisfactory completion of a course oftreatment, i.e. each bath and hairwash, cleanclothing, bedding and towels should be provided(Category 2).

For patients with eczema, dermatitis or otherskin conditions, attempts should be made to treatthe underlying skin condition. Advice on suitableeradication protocols for these individuals shouldbe sought from a consultant dermatologist. Oil-atum bath additive or Oilatum plus (with addedbenzalkonium chloride 6% and triclosan 2%) may beused with these patients; these should onlybe prescribed on the advice of a dermatologist(Category 2).

2.5. Patient management

General principlesThe general principles of infection control shouldbe adopted for the management of patients withMRSA. Good infection control practice should beplaced at the centre of clinical practice, andrequires the explicit support of the organizationalexecutive to ensure that it is seen as having anappropriate position within the organization andcan be enforced as a matter of clinical governance(Category 1b).

Standard infection control principlesA standard approach to isolation precautionsshould be adopted in accordance with the generalprinciples of infection control, rather than intro-ducing specific guidance for the management ofMRSA that may lead to differing standards (Cate-gory 1b).

Management of MRSA-infectedor -colonized patientsPatients should be managed in accordance withthe type of facility in which they receive care, theresources available, and the level of risk that isposed to them and to others. Patients (and thefacilities that may house them) classified as beingat high risk of contracting MRSA or for whom theconsequence of infection may have a high impactwill require a rigorous approach to screening,placement and treatment. Patients identifiedwith MRSA infection or colonization should beinformed of their condition, and local arrange-ments should be made to ensure ease of identifi-cation if re-admission to the facility occurs(Category 1b).


Patient isolationPatient isolation for those infected or colonizedwith MRSA will be dependent on the facilitiesavailable and the associated level of risk. Wherenew buildings or refurbishment are planned,published guidelines should be adopted to providethe most appropriate facilities for patient care.Isolation should be in a designated closed areathat should be clearly defined; in most facilities,this will be either single-room accommodation orcohort areas/bays with clinical handwashing fa-cilities. Consideration should be given to theprovision of isolation wards to contain MRSAspread. The procedures for isolation should beclearly stated, and where necessary explained, tostaff, patients and visitors. Hospital staff enteringisolation facilities should be required to adopt theprescribed isolation precautions rigorously andthese should be audited regularly. Non-staffvisitors should be requested to adopt the neces-sary level of precautions to minimize the risk ofspread of MRSA to other areas of the facility(Category 1b).

Cleaning and decontaminationManagement of the environment and equipmentshould be considered as central to decrease thespread of MRSA. Cleaning regimens for isolationfacilities should focus on the minimization of dustand the removal of fomites from contact areas. Thisshould be a two-fold approach; firstly, the manage-ment of the occupied facility, and then the termi-nal clean of the facility after discharge of thepatient. Cleaning regimens and products should bein accordance with local policy, but should includethe removal of organic material with a general-purpose detergent. Cleaning regimens and theirperformance should be audited regularly.

Patient equipment, e.g. wheelchairs, hoists,slings, sphygmomanometer cuffs, etc., shouldeither be capable of being decontaminated andbe decontaminated before use with other pa-tients, or should be single-patient use and dis-carded as clinical waste at the end of a period ofusage (Category 1b).

Patient movementThe movement of patients with MRSA withina facility should be kept to a minimum to reducethe risk of cross-infection and any potential em-barrassment for the patient. Where patients needto attend departments for essential investigations,the receiving area should be notified of thepatient’s MRSA status in advance of the transfer,and arrangements should be put in place tominimize their contact with other patients, i.e.

to be called forward when the department is readyfor them and to ensure that they are not held incommunal waiting areas. Staff should adopt iso-lation precautions whilst in contact with thepatient.

Arrangements for transfer to other healthcarefacilities, e.g. hospitals, residential care homes,etc., should include notification of the individual’sMRSA status, as appropriate (Category 1b).

Surgical/invasive proceduresPrior to any planned invasive procedure, effortsshould be made to minimize the level of risk ofinfection through topical and systemic decoloniza-tion, and prophylactic antimicrobial therapy, asappropriate.

It may be considered desirable to place theindividual at the end of a procedure list. However,in mechanically filtered environments such asoperating theatre suites, the number of air ex-changes should render this unnecessary. Goodinfection control practices, which should be inplace between all patients, should reduce the riskof cross-infection (Category 1b).

TransportationThe risk of cross-infection from an MRSA-colonizedor -infected patient to other patients in anambulance is minimal. Good infection controlpractices and routine cleaning should suffice toprevent cross-infection (Category 2).

DischargeGenerally, there is no requirement for patientscolonized with MRSA to continue with extendederadication protocols after discharge. This may bevaried in the event of anticipated re-admission toa hospital, especially for a planned invasive pro-cedure. It is appropriate that individuals/groupsinvolved in further care are informed of theindividual’s known MRSA status at discharge.

Patients and their appropriate contacts shouldbe fully briefed and given relevant information onMRSA, its implications and significance prior todischarge in order to reduce unnecessary anxietyand concern when returning to the home environ-ment (Category 2).

2.6. Nursing staff workload and MRSAtransmission

The Working Party emphasizes that inadequatenurse staffing is incompatible with effective in-fection control. Infection control teams and hos-pital managements should bear nursing workloadin mind (including staff numbers, grades and levels

S8 J.E. Coia et al.

of experience, and patient acuity) when planninglocal responses to MRSA and when reacting tooutbreaks, and adequate staffing resources mustbe given a high priority for all patient care areas(Category 1a).

Improving nurse staffing levels on an affectedward may allow improved adherence to localinfection control policies (Category 2), and shouldbe considered as a component of a package ofmeasures to control local outbreaks (Category 2).

2.7. Control of vancomycin-intermediateand -resistant S. aureus (VISA and VRSA)

In the absence of randomized controlled trial dataand on the basis of the descriptive studies outlinedabove and a strong theoretical rationale, recom-mendations for the control of these organismsremain the province of existing best practice andprofessional opinion. These measures can be con-veniently considered under the headings of pre-vention, surveillance and precautions.

PreventionAntibiotic resistance flourishes when antimicrobialdrugs are abused, misused and dispensed at levelslower than treatment guidelines dictate. Virtuallyall strains of S. aureus with reduced susceptibilityto glycopeptide antibiotics described to date arethought to have arisen from pre-existing reservoirsof MRSA, usually in patients with chronic underly-ing disease who have received multiple and/orprolonged courses of glycopeptide treatment. Itseems logical, therefore, to ensure that measuresoutlined elsewhere in this document for controlof MRSA are implemented within the healthcare in-stitution, and that careful antibiotic stewardship isemployed to minimize the inappropriate use ofglycopeptide agents (Category 1b).

Where the use of such agents is deemedappropriate, clinicians should ensure that ade-quate dosages are given to ensure that therapeu-tic levels are obtained at the site of infection andthat duration of therapy is not unnecessarilyprolonged. These measures will help to reducethe likelihood of resistant strains arising de novo(Category 1b).

SurveillanceIt is vital that clinicians and microbiologists remainaware of the potential for emergence of strains ofS. aureus with reduced susceptibility to glyco-peptide antibiotics, and that this awareness isreflected in ongoing laboratory-based surveillance

programmes. The detection of intermediate-levelresistance is challenging for laboratories. This is es-pecially true for strains that are heterogeneous intheir expression of glycopeptide resistance. A highlevel of suspicion must be maintained, particularlyin patients who have received multiple and/or pro-longed courses of glycopeptide antibiotics or whoare known to be colonized/infected with MRSAand vancomycin-resistant enterococci (VRE). De-tailed recommendations and levels of evidence forthe laboratory detection of these strains are givenin the Guidelines for the Laboratory Diagnosis andSusceptibility Testing of MRSA 3a.

The laboratory must notify the relevant clini-cian and infection control personnel as soon aspossible after the isolation of a presumptive S. au-reus isolate with reduced glycopeptide sensitivityin order that control measures can be imple-mented with minimum delay. It is also importantthat the relevant national surveillance network isnotified to ensure that accurate information aboutthe epidemiology and spread of these organisms isgathered (Category 1b).

Control precautions (all Category 1b)Action to be taken on identificationof a case of VISA/glycopeptide-intermediateS. aureus (GISA) or VRSA

e The laboratory should immediately notify the rel-evant clinician and infection control personnel.

e The infection control team should immediatelyidentify where the patient is and where the pa-tient has been during all of the current admis-sion, including transfers from other healthcarefacilities.

e The relevant national surveillance organiza-tion, e.g. Health Protection Scotland, HealthProtection Agency in England and Wales, andthe Health Protection Agency (CommunicableDisease Surveillance Centre) in Northern Ire-land, should be notified.

If the patient is still an inpatient

e The number of healthcare workers caring forthe patient should be reduced. This will causeproblems for those who are allocated to carefor the patient. These healthcare workers willneed support.

e Healthcare workers with chronic skin condi-tions, e.g. eczema or psoriasis, should not beinvolved in direct care of the patient.

e All staff caring for the patient should be madeaware of how the organism is transmitted andthe precautions necessary to prevent this.


e The patient should be cared for in a single roomwith toilet facilities and a wash hand basin.

e The patient and visitors must understand theneed for isolation.

e Fans should not be used to control the patient’stemperature.

e Appropriate infection control proceduresshould be implemented:1. Standard precautions should be used.

Gowns/disposable aprons and disposablegloves should be worn by all those enteringthe patient’s room. Clean, non-sterilegloves and gowns/aprons are adequate.Consideration should be given to use of the-atre-style greens in addition to protectiveclothing to ensure that healthcare workersdo not take uniforms home to launder.

2. Disposable masks and eye protection shouldbe worn by carers for procedures likely togenerate aerosols/splashing. Use of closedsuction systems will help to reduceaerosols.

3. Hand hygiene should be performed withan antibacterial preparation before andafter patient contact. Visibly soiled handsshould be washed with soap prior todisinfection.

4. Non-disposable items that cannot be easilycleaned or disinfected (e.g. sphygmoma-nometer cuffs) should be dedicated for useonly by the infected/colonized patient.

5. Patient charts and records should be keptoutside the isolation room.

6. Linen should be treated as infected. It mustbe discarded into alginate bags within thepatient’s room and a secondary bag outsidethe room.

7. All waste should be discarded into a clinicalwaste bag inside the room, and bags shouldsubsequently be disposed of according tohospital policy.

8. Transfers of colonized/infected patientswithin and between institutions should beavoided unless essential, and the receivinginstitution should be made aware of the pa-tient’s colonization/infection status prior totransfer.

9. After discharge, the room in which the pa-tient was cared for should be cleaned ac-cording to local disinfection policy, withspecial attention given to horizontal sur-faces and dust-collecting areas. Hot waterand detergent are usually satisfactory. Cur-tains should be changed.

10. Compliance with infection control proce-dures should be monitored.

Screening (all Category 1b)Patients

e Nose, axillae, perineum, skin lesions and ma-nipulated sites of the index case and all otherpatients in the unit should be screened for car-riage of VISA/GISA or VRSA.

e The infection control team should review theadmission history of the patient and determineif screening needs to be extended to otherareas and other units alerted.

Staff

e Agreement with staff on the need for screeningshould be sought.

e Nose, axillae and perineum of healthcareworkers and others with close physical contactwith the case should be screened for carriageof VISA/GISA or VRSA.

e Healthcare workers who maintain contactwith the patient will require weekly screen-ing. This may require significant support forthese staff.

e Feedback of results and maintenance of confi-dentiality should be considered.

Eradication (all Category 1b)

e Eradication of colonization/carriage of pa-tients and healthcare workers should be at-tempted (see section on eradication of MRSAcarriage).

e Colonized staff should be excluded from workuntil eradication of carriage is achieved.

3. Background

MRSA was first reported in 1961;4 it has since beenregarded both as a rare condition and of doubtfulclinical significance,5 and as a major pathogen inmany countries.6 Control is necessary because ofthe recent emergence of VISA and VRSA.7,8 Insome countries, such as The Netherlands, the pro-portion of S. aureus bloodstream infections thatare meticillin resistant is small9 (under 1%) com-pared with Germany (19%), Belgium (28%), France(33%), the USA (50%) and the UK (40%).9,10 Thelow rates in some northern European countriesmay reflect a more vigorous ‘search and destroy’policy combined with lower bed occupancy rates,or may reflect exposure to different strains ofMRSA with less propensity for spread.

S10 J.E. Coia et al.

3.1. Why is control and preventionstill important?

MRSA remains common in the UK.11 Nonetheless,up to the early 1990s, MRSA accounted for lessthan 5% of S. aureus blood culture isolates. How-ever, there has been a dramatic change in thelast 10 years. The prevalence of meticillin resis-tance amongst strains of S. aureus causing blood-stream infection in the UK between 1990 and theearly 2000s increased from 2% to >40%, and themean overall rates of MRSA bacteraemia per 1000occupied beds ranged from 0.10 to 0.19.12 In anall-island prospective study of MRSA in Ireland,the prevalence rate per 100 000 population washigher in the south (14.0) compared with the north(11.4), and the incidence of invasive infectionranged from 5% to 10%.13 Throughout Europe,there is considerable variation in the prevalenceof MRSA, varying from low in the Scandinaviancountries to high in the UK, Ireland, Spain andItaly, with the proportion of MRSA of S. aureusisolates amongst blood cultures increasing signifi-cantly between 1999 and 2002 in both the UK andIreland.14

The reasons for continuing efforts to controlMRSA, i.e. to prevent its occurrence in clinicalareas that are MRSA free and minimize the prev-alence and clinical impact (see below) where MRSAis not uncommon or even endemic, remain valid inthe opinion of the Working Party. Nevertheless,justification for not implementing specific mea-sures has been argued by others.15e17 Amongst thereasons offered for relative inactivity include theview that these bacteria do not spread easily, arenot virulent, specific measures advocated to con-trol MRSA are counter-productive and, further-more, they divert energies from other importantareas of infection prevention. Finally, it is arguedthat the clinical impact of MRSA is no greaterthan that of meticillin-sensitive S. aureus (MSSA).Others acknowledge the clinical impact of MRSAbut have been obliged, due to other consider-ations, to relax control measures and have docu-mented the consequences.18

It is mistaken to believe that specific measuresto control MRSA are at the expense of measuresto control and prevent infection with otherpathogens such as Gram-negative bacteria, assuggested in one study from a burns unit.19 Theexperience of some countries such as Finland,where two successive MRSA outbreaks in the early1990s were managed successfully and where MRSAis largely confined to long-term facilities ratherthan acute hospitals, suggests that it is possible

in the non-endemic situation to control thespread of MRSA and also to eradicate it.20,21

Whether it is possible to eradicate MRSA in hospi-tals where MRSA is endemic is debatable, but it ispossible to control spread and minimize the clin-ical impact.

MRSA control measures have additional advan-tages to those of controlling MRSA alone as theyaccentuate the awareness of the importance ofhealthcare-associated infection and assist in thecontainment of other multi-antibiotic-resistantbacteria.22 One of the reasons for the relativelack of success in the control of MRSA may be inad-equate resources and the failure of healthcareprofessionals to comply with good infection con-trol practice. A recent report, which incorporateda literature review and surveillance cultures ina 500-bed hospital in North America, confirmedthat, amongst other things, poor adherence to iso-lation precautions and handwashing accounted forthe apparent ineffectiveness of control mea-sures.23 Furthermore, an investigation of contacttransmission of MRSA in Australia showed that17% of contacts between a healthcare workerand an MRSA-colonized patient resulted in trans-mission of MRSA from the patient to the health-care worker’s gloves.24 However, compliancerates with glove use in the same study were 75%amongst the healthcare workers surveyed butonly 27% amongst doctors.24 In a study of risk fac-tors for MRSA transmission in an adult intensivecare unit (ICU), staff shortages were the only sig-nificant variable associated with clusters of cases,but a mean of only 59% of patient contacts werefollowed by recommended hand disinfection pro-cedures.25 Furthermore, the authors calculatedthat an increase of 12% in hand hygiene compli-ance would have decreased the potential forMRSA transmission significantly.

Recent North American guidelines for the controland prevention of both MRSA and multi-drug-resistant enterococci emphasize the importanceof good infection control practice such as handhygiene protocols, but also recommend specificmeasures to control MRSA such as decolonization,active surveillance cultures and barrier precau-tions.26 The arrival of clinically significant strainsof vancomycin-heteroresistant S. aureus and strainsthat are fully resistant to vancomycin may meanthat there will be fewer effective therapeutic op-tions available to treat S. aureus.27,28 Consequently,specific measures to control MRSA as part of an over-all strategy of hospital infection prevention will helpto reduce the number of patients likely to acquireboth MRSA and strains resistant to vancomycin.


3.2. What is the true impact of MRSA?

ClinicalAlthough the majority of patients who acquireMRSA are merely colonized, not ill and do notrequire antibiotic therapy, a proportion (aboutone-third, depending on the patient population)of patients develop infection, including invasiveinfection, which may result in death. The numberof patients in whom infection with MRSA has beenassociated with death as recorded on death cer-tificates increased from 8% in 1993 to 44% in 1998in England and Wales.29 In a retrospective compar-ison of 504 bacteraemia patients with either MSSAor MRSA bacteraemia, mortality was greater in theMRSA group (14% vs 8%, P< 0.05).30 Many historicalor retrospective studies are difficult to assess be-cause of deficiencies in data capture and becausedue allowance has not been made for inadequateinitial antibiotic therapy. In a prospective studycarried out over a four-year period, 84 patientswith MRSA bacteraemia were compared with 100patients with MSSA bacteraemia.31 Multi-variateanalysis revealed that overall mortality was high-est in the MRSA group and that meticillin resis-tance was independently associated with death.A meta-analysis of nine suitable studies revealedthat all but one found an increased risk of deathfrom MRSA bacteraemia, the relative risk com-pared with MSSA bacteraemia arising from all ofthese studies being 2.12.32 A more recent publica-tion that assessed studies published between 1980and 2000 found no studies that showed a lowermortality in patients with MSSA bacteraemia com-pared with MRSA bacteraemia, seven studies thatshowed a higher mortality in patients with MRSAbacteraemia, and 24 studies where there was nodifference in mortality.33 However, when the stud-ies were combined in a meta-analysis, the odds ra-tio for increased mortality from MRSA bacteraemiawas statistically significant.33

There is also significant morbidity and mortalityassociated with other invasive MRSA infections. Ina prospective study of patients with ventilator-associated pneumonia caused by MRSA or MSSA,the presence of bacteraemia and septic shock wasmore frequent in the MRSA group, and mortalitydirectly due to pneumonia was significantly higheramongst patients with MRSA infection.34 In a pro-spective study of patients with MRSA and MSSA sur-gical site infections, patients with MRSA hada longer mean duration of hospital stay witha higher mortality.35 Meticillin resistance remainedan independent factor influencing mortality onmulti-variate analysis in this study.

Inadequate or inappropriate infection controlmeasures, including those directed at controllingMRSA, may have an adverse impact on hospital-acquired infection. There was a significant in-crease in the overall rate of hospital-acquiredinfection in a US hospital from 4.5% to 5.9% ata time when MRSA spread in that particularinstitution.36

FinancialA variety of attempts have been made to docu-ment the increased costs associated with MRSA,but separating the true cost of MRSA infectionscompared with the cost of MSSA, and the cost ofthe actual interventions to control and preventMRSA from the consequences of colonization andinfection, are very difficult. In a prospective case-control study, the median hospital stay attribut-able to primary nosocomial MSSA bacteraemia wasfour days compared with 12 days for MRSA,and the overall costs were $9661 and $27 083,respectively.37

It is difficult to extrapolate from local data tonational data when assessing the true costs ofMRSA in the healthcare sector and in societygenerally, because the incidence and prevalencevaries from hospital to hospital, and it is difficultto standardize costs between hospitals. Nonethe-less, in a Canadian hospital in which 20 patientswith MRSA infections were compared with 79colonized patients between 1996 and 1998, thecost of isolation and management of colonizedpatients was 1363 Canadian dollars per admission;extrapolating that throughout Canada, the authorsconcluded that the annual costs associated withMRSA infection in Canadian hospitals were be-tween 42 and 59 million Canadian dollars.38 InThe Netherlands, where MRSA is relatively uncom-mon, it has been calculated that the cost of keep-ing one medical centre in Utrecht free of MRSAover a 10-year period (1991e2000) by implement-ing a ‘search and destroy’ policy, i.e. vigorousscreening of possible MRSA cases, isolation, decon-tamination with topical agents and effectivefollow-up, was V2.8m.39 The implementation ofthis policy was associated with 2265 lost hospitalbed-days and wards being closed on 48 occasions.39

The financial consequences of MRSA, if it hadspread and caused infection requiring treatmentover the 10-year period, were not calculated butwould probably have been well in excess of this.

MRSA isolated from superficial sites and in long-stay patients in the community may have littleclinical or financial impact. In contrast, MRSA inthe ICU often results in bloodstream infection,


ventilator-associated pneumonia, intravascular-device-associated infections and urinary tract in-fections, with significant financial implications. Acarrier in the ICU may also act as a reservoir forMRSA acquisition by many very ill patients at risk ofinvasive infection over many days or even weeks.

In a case-control study of patients in a medicalICU in France between 1993 and 1997 with a prev-alence of MRSA carriage of 4%, the mean attribut-able cost associated with MRSA infection wascalculated as $9275, and the total cost of anMRSA control programme ranged from $340 to$1480 per patient.40 The authors also made an ef-fort to calculate the impact of control measures,depending upon the cost of those control measuresand their success in reducing incidence. A study oftwo tertiary neonatal units where efforts to con-trol spread and prevent infections were differentrevealed interesting findings. In the first hospital,where there were 18 colonized patients and fourinfections over a 10-month period, the costsranged from $48 617 to $68 637. In the second hos-pital, where efforts at control were less success-ful, 75 bacteraemias and 14 deaths over 31months were recorded, with costs totalling $1.3million.41 There are, of course, the additionalcosts to patients and their families (e.g. loss of in-come), and to society (e.g. absence from theworkforce) that also need to be considered.

3.3. Do control measures work and arethey worthwhile?

The objective of control measures should be toprevent the acquisition of MRSA and eradicate itwhen it does arise in centres where it is not currentlyprevalent. In hospitals where MRSA is endemic, theobjective is to minimize spread and, in particular,avoid as far as possible the clinical impact in high-risk patients such as those in the ICU or in other keyclinical areas. Harbarth et al. argued that the num-ber of patients with MRSA bacteraemia correlateswith the hospital-wide prevalence of MRSA, andthat even where control measures are delayed, con-trol measures have a substantial impact on both thereservoir of MRSA patients and the attack rate ofMRSA bacteraemia.42 This is also supported by themodelling studies in the Health Technology Assess-ment systematic review of isolation policies.43 Fur-thermore, some recent studies have suggested thatMRSA control measures as part of an infection con-trol programme can also reduce the impact of othermulti-resistant bacteria in ill patients.44 After theinstitution of good infection control practice, theincidence of MRSA colonization decreased (from7.7% to 2.6%) and the percentage of patients with

antibiotic-resistant Klebsiella pneumoniae also de-clined (1.7% to 0%).44 In one of the largest resource-limited hospitals in the world, targeted interventionprogrammes, in which staff and patients werescreened for MRSA carriage, patient carriers wereisolated, and mupirocin and chlorhexidine were ad-ministered to carriers, resulted in the percentage ofpatients with MRSA bacteraemia in the ICU decliningfrom 1% to 0.5%; however, this increased one yearafter the study when the intervention measureswere withdrawn.45 Feedback of MRSA rates is alsoimportant; when this was undertaken for Clostrid-ium difficile infections, there was a decrease in in-cidence but this increased again when the feedbackwas discontinued.46

In a French study assessing the efficacy ofa control programme during the mid 1990s, therate of MRSA infection decreased from 5.9 to 0.8/1000 patient-days as did the prevalence of MRSAcarriage and the ratio of MRSA to all S. aureus.47 Ina Spanish study, three time periods were studied,i.e. pre-outbreak, during an outbreak of MRSA andwhen a control programme was instituted. The num-ber of cases per 1000 patient-days was 3.2, 8.2 and2.0 during the respective periods in the ICU.48 Theauthors estimated that the programme prevented76% of expected MRSA cases and 85% of expected fa-talities due to MRSA in the ICU. Another study in anICU on the effect of application of mupirocin oint-ment to the nose with whole-body washing usingchlorhexidine in patients colonized with MRSA toprevent pneumonia showed that there was a signifi-cant reduction in infection.49

In contrast, in a recent UK study conducted in twoICUs to assess the effectiveness of patient isolationduring two periods, one of which involved notmoving positive patients to an isolation room, theauthors found no difference in the MRSA acquisitionor transmission rates, and concluded that isolationpolicies should be re-evaluated.50 An accompanyingcommentary argued that their conclusions werepremature because admission cultures were ob-tained in only 80e87% of patients (possibly insuffi-cient to prevent dissemination) and becausecompliance with hand hygiene was only 21%.51

Nonetheless, this study demonstrates the need forwell-designed studies to be carried out on specificinterventions, addressing confounding factors.

When assessing the cost-effectiveness of MRSAcontrol programmes, a number of variables have tobe considered. These include the cost of theintervention and the cost of MRSA infection.Laboratory costs for MRSA screening are quitelow; in one Canadian hospital, these were foundto be $8.34 per specimen with a total cost of$30 632 during 1996 for a 980-bed hospital.52


A model has been proposed to explore the impactof MRSA acquisition and different MRSA screeningtests; it was concluded that taking a sample fromthe nose alone and inoculating directly on to a ci-profloxacin Baird-Parker agar, without broth incu-bation, was the most cost-effective approach.53

However, other aspects have to be costed andthese include hospital-wide programmes of ade-quate cleaning and environmental decontamina-tion. Interventions to improve hospital hygienedecreased the percentage of environmental sitespositive for MRSA from 32% to 0.47% in one re-port.54 Enhanced environmental decontaminationis likely to assist in controlling the spread ofMRSA as well as other bacteria such as C. difficile.It is also likely to improve the aesthetic appear-ance of the hospital, resulting in other beneficialhealth and psychological effects and demonstrat-ing the values of the organization.

3.4. Conclusions

The data available to date strongly implicate MRSAas a significant hospital-acquired infection result-ing in additional morbidity and mortality as well ascontributing to healthcare costs. This applies topatients at particular risk, e.g. patients requiringintensive care and patients following major sur-gery, and the elderly, which comprise an increas-ing proportion of patients in acute hospitals and inother healthcare institutions. Furthermore, pa-tients and the public are increasingly seeingMRSA and rates of MRSA infections as indicatorsof the quality of patient care. They require re-assurance that all healthcare professionals aretaking reasonable and sensible precautions tominimize spread. Although it is very difficult tocarry out double-blind randomized controlled tri-als on specific aspects of recommended controlprogrammes, because MRSA colonization and in-fection rates vary considerably from time to timeand from centre to centre, control measures havebeen shown to be effective, resulting in reducedmortality as well as helping to contain healthcarecosts. Consequently, the Working Party is of thestrong opinion that an active MRSA control andprevention programme, as part of an overall in-fection control strategy within a hospital, con-tinues to be the recommended approach.

4. Surveillance

‘Epidemiologic surveillance is the ongoing system-atic collection, analysis, and interpretation of

health data essential to the planning, implemen-tation, and evaluation of public health practice,closely integrated with the timely dissemination ofthese data to those who need to know. The finallink in the surveillance chain is the application ofthese data to prevention and control.’

Thacker SB, Berkelman RL. Public health surveil-lance in the United States. Epidemiol Rev 1998;10:164e190.

4.1. Background

Most reviews of interventions to prevent andcontrol MRSA document the difficulty of establish-ing the effect of particular infection control in-terventions, as multiple interventions are usuallyused together. Surveillance, however, is a criticalpart of any infection control programme. It mustnot be an end in itself, but should be undertaken toimprove the quality of care. It is the instrument forearly recognition of changes in patterns of infec-tion, identifying the size of the problem, monitor-ing trends and comparing rates, evaluating theeffectiveness of interventions, identifying areasfor further investigation or research, re-inforcinggood practice, and influencing key hospital staffand decision makers. Robust surveillance cannot beundertaken on a ‘shoestring’ budget, but requiresresourcing for the collection, collation, analysisand interpretation of data. Typically, this requiresinput from staff with information technology skillsto the infection control team.

Seminal work in this area has demonstrated thathospitals with infection surveillance and controlprogrammes reporting wound infection rates backto surgeons reduced rates of hospital-acquiredinfection by 20%, highlighting the importance offeeding back data to inform decision makers.These reductions were further augmented incre-mentally when surveillance had been in operationfor at least one year and there were dedicatedinfection control practitioners (this study alsodemonstrated the cost-effectiveness of such pro-grammes).55 More recent work has also empha-sized the importance of feedback of surveillanceinformation. For instance, Stone et al. demon-strated that feedback of C. difficile rates withinvolvement of clinicians was associated with re-ductions in the incidence of C. difficile diarrhoea.When feedback was relaxed, rates rose, as didthose for MRSA.46 In addition, a descriptive studyby Curran et al. investigated the hospital-widefeedback of MRSA acquisition data monthly to


clinical staff using statistical process control chartsfor each unit and its effect on rates of MRSA.56

Twenty-five months of retrospective data and 21months of prospective data were collected, andthere were monthly reductions in MRSA acquisitionfrom two months into the feedback process. Duringthe prospective period, acquisition rates fell byapproximately 50%, suggesting that regular feed-back of surveillance data to healthcare staff hada positive effect on staff behaviour. In other stud-ies, the surveillance usually formed part of a wideraggressive control programme.57,58

4.2. Sources of surveillance information

The most commonly utilized sources of surveil-lance information are microbiological reports,usually collected as part of ‘alert organism’ sur-veillance. However, most hospital control plans forMRSA also include interventions that are a sourceof surveillance data. Thus, many aggressive con-trol strategies are based on screening patients toidentify those colonized or infected with MRSA inorder to segregate them from unaffected pa-tients.59,60 Many hospitals employ admissionscreening to keep units at high risk for seriousMRSA infections free of MRSA, and routine patientscreening on a regular basis may also be under-taken. Occasionally, discharge screening may beused to identify acquisitions during the admission.Staff may be screened in certain circumstances(see below). Prevalence surveys may be under-taken periodically across part of or the whole hos-pital to identify its infection status. These actionsall result in information that may be useful for sur-veillance purposes.

4.3. Surveillance: information for action

The key steps in surveillance are the systematiccollection of data, and its analysis, interpretationand dissemination for action. As noted earlier,feeding back the surveillance data to the staff inthe hospital units surveyed is, in itself, an impor-tant infection control intervention. Moreover, forthe benefits of surveillance to be maximized,surveillance needs to be tied in actively to systemimprovements. This is the basis of the qualityimprovement model, where the infection controlpractitioner must be able to identify the differ-ence between natural and unnatural variation inreal time, as well as all the possible causes and themost important causes, so that actions to improvethe system can be recognized and implemented.This requires knowledge of the important elementsin the system that are not performed or are not

performed sufficiently well, e.g. hand hygiene orinformation about isolation precautions. There arevarious methods for displaying root causes ofparticular problems, such as the Fishbone cause-and-effect chart (Figure 1). This can be useful inidentifying and addressing all factors contributingto acquisition and transmission of infection in a hi-erarchical way.61 The numbers of MRSA acquisi-tions can be tracked clearly using a statisticalprocess control (SPC) chart (Figure 2). This isa chronological chart of past events used to pre-dict, with some degree of certainty, where futureresults should fall if the system remains stable andin control. SPCs distinguish between natural varia-tion (part of the system) and unnatural variation(outwith the system). They are becoming increas-ingly used to present the data and to identifywhen numbers of cases are exceeding normal ex-pectations for that unit. They can also identify sig-nificant reductions in variation that would occur ifsystem improvements were implemented and sus-tained. An SPC chart must be updated regularly,preferably at monthly intervals. The initial objec-tive is for results to stay within control limits. How-ever, in the longer term, the objective is to reducevariation through sustained system improvementsin infection control. An SPC chart will identify ifthe system has changed but not identify the causeof the change. CuSums and other statisticalmethods of depicting data may also be used.

Consequently, the information from the SPCchart or other statistical method being used toidentify changes away from expected levels ofacquisition needs to be linked to the causes of newacquisitions, with a focus on those that have thebiggest impact. This approach is designed to reduceMRSA, but does not indicate the level at which itshould be applied. National and organizational dataare essential for trend analysis, but action to reduceMRSA acquisition should be taken at the ward or unitlevel as individual units vary in the patient risks,procedures, environment, facilities and staff. Ac-tions must be tailored to each clinical unit’s needs.

For surveillance to operate fully as a tool forquality improvement, it is important that it isrecognized as being an inherent part of thehospital’s clinical governance process, with cleararrangements in place that identify those respon-sible for taking action on the results in individualhospital departments. The results should be con-sidered by the relevant senior management com-mittee on a regular basis. They should also be usedto inform local infection control training. Thus, thedata should be disseminated in a timely mannerand should be readily intelligible to a wide varietyof hospital practitioners, from cleaners to

ain branches represent the main factors. AHPs,ource: E. Curran.

Guid

elin

es

for

MRSA

inhealth

care

facilitie

sS15

Figure 1 Fishbone cause and effect diagram displaying factors that impact upon the transmission of MRSA. The mallied health professionals; PPE, personal protective equipment; HCWs, healthcare workers; IC, infection control. S


Figure 2 Statistical process control chart showing new meticillin-resistant Staphylococcus aureus acquisitions ona ward by month, before and after the introduction of system improvements. The centre line (CL) shows the meannumber of acquisitions, whilst the upper (UCL) and lower (LCL) control limits show the range in which the data areexpected if infection control systems in the ward remain stable. If new results fall within this range, the system isstable or ‘in control’. If new acquisitions rise above the UCL, the system is out of control and infection control staffshould identify likely cause(s) and intervene to bring the situation back to the ‘in control’ range. Apart from remaining‘in control’, it is also important to put in training improvements to reduce the mean and control limits over time.Sources: E. Curran and Health Protection Agency.

clinicians, so that they can be utilized in trainingor update sessions.

Many different methods may be used for col-lecting and collating surveillance information andfeeding it back to clinical teams.62,63 Infectioncontrol software packages are available that canassist with the collation and analysis of surveil-lance information [Health Protection Agency Infec-tion Control IT Implementation and EvaluationProject. Report prepared for the Department ofHealth, August 2005. Available at: http://hpa.org.uk/infections/topics_az/hai/ICITIAE_report_Dec_2005.pdf. Last accessed 26 January 2006].

Although the primary focus of surveillance shouldbe as a tool for use at local level, surveillance atregional and national level is important to identifydiffering trends and for benchmarking. When com-paring units within a hospital, it is often sufficient tomeasure numbers of infections. However, for datato be comparable between hospitals, specialtyactivity denominators become necessary and activecase finding in a consistent fashion to agreed casedefinitions is paramount. National surveillance datamay be used to drive action, as has been the casewith the mandatory S. aureus bacteraemia surveil-lance system established in England in 2001(Figure 3).64e66 This, allied to the setting of targetsand performance indicators, has been used by theUK Government to highlight unacceptably high rates

of serious MRSA infections and to drive hospitalmanagement to take action.67

4.4. Recommendations

Surveillance must be undertaken routinely as partof the hospital’s infection control programme, andshould be a recognized element of the clinicalgovernance process. As such, there should be cleararrangements identifying those responsible foracting on the results in individual hospital direc-torates (Category 1b).

For benchmarking purposes, surveillance datashould be collected and reported in a consistentway, to agreed case definitions and using agreedspecialty activity denominators, with stratificationaccording to case mix (Category 1b).

Surveillance data should be fed back to hospitalstaff routinely, readily intelligible to most hospitalstaff, considered regularly by hospital senior man-agement committees, and used in local infectioncontrol training.

MRSA surveillance should include:

e any mandatory requirements (Category 1c);e results of microbiological investigations for

clinical purposes (Category 1b); ande results of microbiological investigations under-

taken for screening purposes (Category 1b).

http://hpa.org.uk/infections/topics_az/hai/ICITIAE_report_Dec_2005.pdf




Figure 3 Staphylococcus aureus bacteraemias in England, 1992e2005. Mandatory surveillance of S. aureus bacter-aemias started in April 2001. The data prior to this are from the routine laboratory reporting system (voluntary) to thePublic Health Laboratory Service and, latterly, the Health Protection Agency. Numbers of reported bacteraemias in-creased following the start of mandatory surveillance, reflecting improved laboratory reporting of bacteraemias.MSSA, meticillin-sensitive S. aureus; MRSA, meticillin-resistant S. aureus.

The dataset should include:

e patient, laboratory, unit/ward and hospitalidentifiers;

e patient demographics (address, age, sex);e date of admission;e date of onset of infection (if appropriate);e site of the primary infection, if appropriate (if

bacteraemia, source of the bacteraemia);e date specimen taken;e site of specimen (blood culture, wound, etc.);e where the MRSA was acquired (hospital, com-

munity, speciality, etc.);e whether part of an outbreak; ande antimicrobial susceptibilities.

Other desirable items include the primary di-agnosis, an assessment of the severity of underly-ing illnesses, prior antimicrobial therapy andpossible risk factors for infection (Category 2).

5. Prevention and control of MRSAinfections

5.1. Antibiotic stewardship

Inappropriate antibiotic use promotes the emer-gence and spread of antibiotic resistance. The

emergence of meticillin resistance in previouslysensitive strains of S. aureus appears to be rela-tively rare. Excessive use of antibiotics, however,promotes the spread of existing strains of MRSAthrough reduction in colonization resistance inpatients and by giving resistant strains a survivaladvantage in the hospital environment.68

Antibiotic use and compliance with local guide-lines needs to be audited. Inappropriate antibioticuse, e.g. underdosing, multiple or excessive durationof courses, and the use of broad-spectrum agents aremajor factors in the spread of antibiotic resistance inhealthcare settings. Numerous antibiotic classeshave been associated with MRSA colonization andinfection in different studies.69e73 Exposure tobroad-spectrum antibiotics, particularly third-generation cephalosporins and fluoroquinolones, isan independent risk factor for MRSA colonizationand infection in numerous studies.72,74e77 Further-more, antibiotic stewardship programmes havebeen shown to result in significant reductions inMRSA colonization and infection rates.78,79

Colonization or infection with glycopeptide-resistant and intermediately-resistant S. aureus isstrongly associated with prolonged exposure toglycopeptides and prior colonization or infectionwith MRSA.27,80,81 Promotion of prudent glycopep-tide use has been shown to reduce the prevalenceof VRE in ICUs,82 and it follows that prudent


glycopeptide use should also be promoted to pre-vent glycopeptide resistance in staphylococci.83

In addition to general (e.g. compliance withhand hygiene measures) and specific (e.g. screen-ing for carriage) measures to control MRSA, atten-tion must be given to the appropriate use ofantibiotics. This includes the following.

e Avoidance of inappropriate or excessive antibi-otic therapy and prophylaxis in all healthcaresettings (Category 1a).

e Ensuring that antibiotics are given at the cor-rect dosage and for an appropriate duration(Category 1b).

e Limiting the use of glycopeptide antibiotics tosituations where their use has been shown tobe appropriate. If possible, prolonged coursesof glycopeptide therapy should be avoided(Category 1a).

e Reducing the use of broad-spectrum antibi-otics, particularly third-generation cephalo-sporins and floroquinolones, to what isclinically appropriate (Category 1b).

e Instituting antibiotic stewardship programmesin healthcare facilities, key components ofwhich include the identification of key person-nel who are responsible for this, surveillance ofantibiotic resistance and antibiotic consump-tion, and prescriber education (Category 1c).

5.2. Screening for MRSA

IntroductionScreening for MRSA should be directed at thecommon sites of carriage and infection. In S. aureuscarriers of both meticillin-susceptible and -resis-tant strains, the anterior nares are persistently orintermittently colonized, whereas carriage atother normal body sites is generally less frequentand persistent.84 Carriage is commonly persistentat sites of damaged or diseased skin (e.g. wounds,eczema) and at sites of insertion of foreign bodiessuch as intravenous catheters. Colonization of thethroat may be a marker of persistent carriage inotherwise healthy staff members, and oropharyn-geal carriage may persist in those with poor dentalcare, inadequately cleaned dentures or unhealthytonsils. All studies of the value of screening pa-tients and staff may inevitably be criticized meth-odologically. Screening is not a control measure initself, and the clinical efficacy of any screeningprogramme can only be measured by its ability todirect interventions such as isolation, cohortingand decolonization of a subset of patients.

Implementation of these interventions varieswidely among reported studies. Additional influen-tial variables, such as the use of epidemiologicalfactors to guide isolation of newly admitted pa-tients before screening results are available, alsodiffer among published reports. The authors ofthe recent Health Technical Assessment system-atic review endorsed screening as a componentof control policies, but did not find sufficientdata to assess its individual contribution.43

Nevertheless, the Working Group agree with theconclusions of other MRSA guideline developmentgroups who, while acknowledging the imprecisionof the published evidence, were convinced by itsconsistency, which strongly suggests a causal re-lationship between programmes incorporating ac-tive surveillance cultures and isolation precautionsand successful control of MRSA.26,85 This approachis scientifically and managerially logical becausecolonized and infected patients are the primaryreservoir of MRSA infection for others, and theiridentification by active screening allows focusingof effective but limited infection control resourceson positive patients.86 Fifteen full publications be-tween 1982 and 2002 were adduced by the Societyof Healthcare Epidemiology of America (SHEA)Working Party 26,40,42,87e98 in support of the valueof active surveillance.85e91 Several of these re-ports provided particularly significant additionalinformation. Jernigan et al. demonstrated a 15.6-fold reduced rate of transmission (95% confidenceinterval 5.3e45.6, P< 0.0001) from patients whowere known to be MRSA carriers by surveillancecultures compared with those whose screening re-sults were not available.87 Later re-analysis of thedata showed that the effect was similar if patientswho were only colonized with MRSA were studied(relative risk 11.9, 95% confidence interval 3.25e47.5, P¼ 0.00014).99 Patients known to be positivewere isolated with additional droplet or infectioncontrol precautions, while those not known to bepositive were managed with standard or universalprecautions. In a further study by the same group,these additional precautions were shown to becost-effective.41 Secondly, on a Dutch ICU, Vrienset al. found a 38-fold greater rate of transmissionfrom unisolated unknown positive patients nursedwith universal precautions compared with identi-fied isolated positive patients cared for withgown, mask and gloves.100

Forceville et al. introduced active surveillancewith other measures in a stepwise fashion to anICU, and demonstrated a reduction in MRSA acqui-sition on the unit from 8.6% to 0% (P< 0.001) overa six-year period.101 The methodology of this studymay be criticized because the baseline rate was


only established over a one-month period, and thelowest rate was only seen in the final year, whichmay weaken the directness of the link between in-tervention and outcome. Surveillance screeningdata were used to produce control charts for regu-lar feedback on MRSA local acquisition to wardclinical staff.56 Twenty-five months of retrospec-tive data and 21 months of prospective datawere collected and, during the prospective inter-vention period, acquisition rates fell by about 50%.

A recent study in an 850-bedded communityhospital in Italy with endemic MRSA (>50% of theS. aureus infections being meticillin resistant)reported the effects on MRSA bloodstream infec-tions of introducing patient screening, targetedenhanced contact precautions, feedback of MRSArates to ward staff, and mupirocin clearance ofMRSA carriage.102 A sustained fall in the incidenceof MRSA bacteraemia was observed from 0.64 to0.30 per 1000 admissions. Historical controlswere used, but the baseline was established over18 months, screening and interventions were pro-gressively introduced over 30 months, and thiswas followed by a 24-month observation period.This reduction occurred despite rising usage ofcentral venous catheters. Interestingly, a rise inthe rate of MSSA central venous catheter infectionwas noted (0.81 to 1.59 per 1000 admissions, rela-tive risk 1.96, 95% confidence interval 1.32e2.93,P¼ 0.001), which suggests that the change inMRSA rate was independent of any change in cath-eter management.

There is no good-quality, comprehensive guid-ance from the literature on which patients andbody sites should be screened, and a variety ofstrategies have been adopted in published studiesand are currently in use in UK healthcare facilities.Recommendations have been made based on con-sensus within the Working Party relating to screen-ing of patients on admission to hospital and ofinpatients on particular wards and units, and alsorelating to screening of staff members.

On the grounds of lack of evidence of clinicaland cost-effectiveness, routine screening of alladmissions to hospital is not advocated. However,local conditions may justify such a policy, perhapstemporarily, and such decisions should be madelocally by the infection control team and beagreed with the relevant hospital managers anddirectors. This remains an unresolved issue. Thedecision to perform surveillance cultures before orat admission to a unit must be considered inrelation to whether isolation or cohorting ofpatients on admission is to be used; whetherdetails of patient management are to be influ-enced, such as attempts at decolonization or

prophylaxis with glycopeptides for surgical pro-cedures; and the availability of rapid methods fordetection of MRSA carriage methods3a.

‘Discharge screening’ as advocated in the 1998guidelines may be useful as an epidemiologicalmeasure of the prevalence of MRSA colonizationand acquisition on a unit, but it cannot guidethe targeting of contact and isolation precautionsat individual patients. There is no other validreason for screening patients before discharge tothe community, and carriage of MRSA is nota valid reason for exclusion from residentialcare homes.

There is evidently wide variation in practiceamong UK hospitals over screening patients whowere known MRSA carriers in the past and mayhave received clearance therapy, and patients whohave been exposed to MRSA. Some units specifytwo and others specify three sets of screens,between 48 h and seven days apart, and a varietyof screening sites and screening methodologieshave been used. Very few publications have ad-dressed this issue. Bannister recommended fivescreens (with direct inoculation of a solid selectivemedium) for screening of previously positive pa-tients before they were transferred to ‘suscepti-ble’ hospital units.103 The Working Party are notaware of any high-quality evidence to guide thechoice among these alternatives, and it is recom-mended that this should be the subject of futurestudy. On the grounds of practicability, the Work-ing Party recommend that three screens shouldnormally be performed one week apart, beginningat least 48 h after antibiotic and antiseptic therapyhas stopped. Nevertheless, a total of five screensmay be prudent in high-risk situations, and bothrelapse and re-infection may occur.

To guide the definition of risk groups, there islittle good-quality published evidence on theprevalence and duration of MRSA carriage afterattempted clearance of carriage and discharge ofpatients from hospital. A recent study from TheNetherlands reported 95% clearance rates andprolonged clearance in patients with negativepost-treatment screens and without conventionalrisk factors for MRSA persistence (e.g. abnormalrespiratory tract, skin lesions or foreign bodiessuch as percutaneous gastroenterostomytubes).104 The authors recommended that pa-tients could only be considered as risk free forsubsequent MRSA carriage after remaining screennegative and free from risk factors for 12 months.This is a stringent requirement and the WorkingParty doubt that its application would be usefulor practicable in the current UK healthcaresystem.


RecommendationsActive screening of patients for MRSA carriageshould be performed and the results should belinked to a targeted approach to use of isolationand cohorting facilities (Category 2). The WorkingParty advocate that certain high-risk patientsshould be screened routinely, and certain high-risk units should be screened at least intermit-tently, in all hospitals. The fine details of whichpatients are screened should be determined lo-cally by the infection control team and must bediscussed with the appropriate clinical teams andendorsed by the relevant hospital managers anddirectors. These details will be influenced by thelocal prevalence of MRSA in the hospital and unitconcerned, the reason for admission of the pa-tient, the risk status of the unit to which they areadmitted, and the likelihood that the patient iscarrying MRSA. Patients regularly move among riskareas, and consideration should be given to ex-tending screening measures to low-risk units ifnumerous patients are transferred from them tohigh-risk areas. Patients at high risk of carriage ofMRSA include those who are:

e known to have been infected or colonized withMRSA in the past (Category 1b);

e frequent re-admissions to any healthcare facil-ities (Category 1b);

e direct interhospital transfers (Category 1b);e recent inpatients at hospitals abroad or hospi-

tals in the UK which are known or likely tohave a high prevalence of MRSA (Category1b); and

e residents of residential care facilities wherethere is a known or likely high prevalence ofMRSA carriage (Category 1b).

Other risk groups may be defined by localexperience, based on screening initiatives or out-break epidemiology. Published examples have in-cluded injecting drug users, patients infected withhuman immunodeficiency virus, individuals witheczema, dermatitis and psoriasis, and members ofprofessional contact sport teams (Category 2).

Units caring for patients at high risk for sufferingserious MRSA infections or with a high proportionof MRSA infections among colonized patients in-clude: intensive care, neonatal intensive care,burns, transplantation, cardiothoracic, orthopae-dic, trauma, vascular surgery, renal, internationalreferral centres (all Category 1b), and other spe-cialist units (e.g. bone marrow transplant units) asdetermined by the infection control team and asagreed with the senior clinical staff of the units andrelevant hospital management structure.

Patients on elective surgical units (e.g. ortho-paedic, vascular), usually with short inpatientstays, are at lower risk of MRSA acquisition thanpatients on trauma and emergency units, or mixedunits, and due account of these differences shouldbe taken when local screening policies are beingestablished (Category 2).

All patients who are at high risk for carriage ofMRSA should be screened at the time of admissionunless they are being admitted directly to isolationfacilities and it is not planned to attempt to clearthem of MRSA carriage (Category 2).

In addition, all inpatients on high-risk units shouldbe screened regularly (e.g. weekly or monthly,according to local prevalence) (Category 2).

In addition, screening all patients (regardless oftheir risk-group status) should be considered onadmission to high-risk units and to those units,especially elective surgical wards, designated asbeing ‘MRSA free’ (Category 2). The decision aboutwhether or not to perform routine admissionscreening should be made explicitly by the in-fection control team in consultation with thesenior clinical staff of the units, and should beagreed with the relevant hospital managementstructure. Such ‘blanket’ screening may be usedintermittently, and may be especially worthwhileif the local prevalence of MRSA carriage in suchpatients is higher than usual for the UK, if thereare sufficient local isolation/cohorting resourcesto manage carriers effectively, and if local policiesfor clearance of carriage and/or use of surgicalprophylaxis with glycopeptides are in place.

For patients, the following sites should besampled (Category 1b): anterior nares, skin lesionsand wounds and sites of catheters, catheter urine,groin/perineum (specimens should be taken withsensitivity), tracheostomy and other skin breaks inall patients, and sputum from patients with a pro-ductive cough. The umbilicus should be sampled inall neonates. One should also consider samplingthe throat as a marker of persistent colonizationthat may require more intensive clearancetherapy.

The decision to perform screening of patients onadmission to other wards or to screen inpatients onother wards regularly should be made by the localinfection control team in consultation with thesenior clinical staff of the units and as agreed withthe relevant hospital managers and directors(Category 2). In principle, hospitals with significantproblems with MRSA transmission or a high preva-lence of MRSA carriage or infection should considerperforming more widespread and regular screeningthan units with a low prevalence. However, thisapproach has resource implications and should first


be used in areas where the clinical impact of highMRSA prevalence is highest (i.e. in the ‘high-risk’clinical areas). The aim is to identify all positivepatients within the hospital to allow targeting ofisolation and cohorting facilities to minimize therisk of onward transmission to other patients.

When possible, patients awaiting elective ad-mission who satisfy local requirements for screen-ing should be screened before admission by theirgeneral practitioners or in pre-admission clinics(Category 2). Patients who are at high continuingrisk of acquiring MRSA between the time of pre-admission screening and that of admission (e.g.they reside in a residential care facility which isknown to have a high prevalence of MRSA) must berescreened on admission and should be isolated orcohorted according to policies in place on theadmitting unit until both sets of screening resultsare known.

Action to be taken if screening results are positiveIn general, detection of patients colonized orinfected with MRSA on a ward should be anindication for increased screening (Category 2).Little evidence exists to guide the details of anappropriate response, but this should be influ-enced by the risk group of the affected unit, bythe number of newly detected MRSA-positivepatients, by the adequacy of nurse numbers tostaff the ward, and by the availability of isolationand cohorting facilities. There is always a delaybetween MRSA acquisition by a patient and itspresence being detectable by screening samples,and so it is recommended that at least threescreens should be performed at weekly intervalsbefore a patient can be considered to be at lowrisk of having acquired MRSA if they have beennursed in proximity to unknown and unisolatedMRSA-positive patients or by the same staff (Cat-egory 2). The screening for MRSA in each unitwithin a hospital should be the subject of regularaudit, with the results reviewed by the hospital’sinfection control committee. The results shouldalso be made available to management.

Screening of staff is not recommended rou-tinely, but the Working Party considers it to bevaluable under certain circumstances. It is a con-troversial area and therefore guidance is provided.If new MRSA carriers are found among the patientson a ward, staff should be asked about skin lesions.Staff with such lesions should be referred forscreening and for consideration of dermatologicaltreatment by the relevant occupational healthdepartment (Category 1b). Staff with persistentcarriage at sites other than the nose should beconsidered for referral for appropriate specialist

management (e.g. ear, nose and throat; derma-tology) who should arrange follow-up screeningaccording to local protocols (Category 1b).

e Staff screening is indicated if transmission con-tinues on a unit despite active control mea-sures, if epidemiological aspects of anoutbreak are unusual, or if they suggest persis-tent MRSA carriage by staff (Category 2).

e Care is needed to distinguish between tran-sient carriage (i.e. nasal carriage which is lostwithin a day or so of removal from contactwith MRSA-positive patients and carries littlerisk of onward transmission) and prolonged car-riage (especially associated with skin lesionsand throat colonization) (Category 1b). Thisdistinction is usually best achieved by screen-ing staff as they come on duty at the beginningof their shift and not as they leave at the end oftheir shift.

e Nurses, doctors, physiotherapists, other alliedhealth professionals and non-clinical supportstaff (e.g. porters) should be considered forscreening, and the implications for onwardspread by staff working on other wards shouldalso be considered (Category 2).

e The special difficulties and risks posed byagency and locum staff should be considered.It may be appropriate to consider the MRSAprevalence in units where staff of this typehave recently worked when allocating workareas and tasks among agency and locum staff;this task is easier if agency and locum staffworking on high-risk units can be drawn froma known pool of staff (Category 1b).

e Appropriate sampling sites for staff screeninginclude anterior nares and any areas of abnor-mal or broken skin (Category 1b). As a guideto use of eradication measures and as markersof staff who may be at high risk of sheddingMRSA to the environment, one should also con-sider screening throat, hairline and groin/peri-neum (specimens should be taken withsensitivity) of staff members found positiveon initial screens.

e It is recommended that a minimum of threescreens at weekly intervals (while not receivingantimicrobial therapy) should be performedbefore a previously positive staff member canbe considered to be clear of MRSA (Category2). Local policies should be developed to guidepostclearance sampling of staff (Category 2),and due note should be taken of the individu-al’s risk of transmission to patients whenagreeing their continuation or return to work.For example, a staff member colonized with


MRSA working in an ICU or neonatal unit repre-sents a greater potential risk to patients thana staff member with MRSA working in an outpa-tients department. In principle, it is recom-mended that only staff members withcolonized or infected hand lesions should beoff work while receiving courses of clearancetherapy, but this decision should be based onlocal risk assessments. To aid staffing re-sources, it may be possible to re-allocate staffcarriers temporarily to low-risk tasks or units.

The Working Party makes no recommendationabout performance of ‘discharge screening’.

Performance of active screening for MRSA ineach unit within a hospital must be the subject ofregular audit, with the results reviewed andminuted by the hospital’s infection control com-mittee and made available to the appropriatehospital managers and directors, and reportedvia the appropriate clinical governance structures(Category 1b).

Hospitals with highly prevalent, endemic MRSAshould consider focusing screening, control mea-sures and other resources on high-risk units ini-tially, with the intention of rolling them out tolower-risk areas after the position has improved(Category 2). Screening should not be seen as anend in itself, but rather it should be linked tospecific, locally determined packages of controlmeasures.

Geographically adjacent healthcare facilitiesand those exchanging large numbers of patientsbecause of clinical links should liaise to agreecommon and efficient screening measures thatshould be linked to common and efficient controlmeasures (Category 2). Such links should capitalizeon any developing networking relationships amongclinical and laboratory units, such as those en-couraged through the Pathology Modernizationinitiative.

Results of screening cultures should be madeavailable promptly to the clinical and infectioncontrol teams of other healthcare facilities towhom a patient is to be, or has recently been,transferred (Category 1b). Refusal to accept trans-fer of a patient is not justifiable on the basis of therisk posed to other patients by an individual’scarriage of or infection with MRSA. All units shouldhave procedures in place and adequate facilitiesfor containment of MRSA.

5.3. Decolonization

MRSA decolonization refers mainly to the useof topical agents such as nasal ointment and

bodywash/shampoo to eradicate/reduce nasaland skin carriage.

Systemic antibiotics may be used to clearpersistent carriage (e.g. persistent throat car-riage). Complete eradication is not always possiblebut a decrease of carriage can reduce the risk oftransmission in healthcare settings. Decolonizationwill also reduce the risk of inoculation to thepatient’s own surgical wound during the operation.

The efficacy of any decolonization regimen willdepend on the presence of wounds, skin lesionsand foreign bodies such as urinary catheters,nasogastric tubes, haemodialysis lines, etc. Anydecolonization regimen should be carried out un-der the advice and supervision of the hospitalinfection control team.

The following section provides evidence forrecommendations concerning the use of decoloni-zation strategies in patients and staff to reducethe spread and incidence of MRSA. Suppression ofcarriage, eradication of carriage or both have beenused at times to help control the spread of MRSA.Previous guidance suggests treatment of carriers ofMRSA in certain circumstances, with evidencesuggesting that patients and staff carriers areimportant sources of MRSA for subsequent spread.3

The studies that have addressed the use of topicaldecolonization strategies in the eradication ofMRSA include a recent Cochrane review of random-ized controlled trials.105 The remaining studies areobservational, plus one controlled trial and a con-trolled before and after study. In addition to thesestudies, the most recent guidelines for the preven-tion of MRSA published by SHEA were appraised us-ing the Appraisal of Guidelines for Research andEvaluation tool.26 In these guidelines, it was shownthat treating colonized or infected healthcareworkers who were epidemiologically implicatedin outbreaks has helped to control outbreaks.106

For healthy healthcare workers, topical treatmentwith intranasal mupirocin ointment twice daily forfive days was associated with a 91% reduction inthe prevalence of S. aureus carriage, but recoloni-zation was noted in 26% of decolonized healthcareworkers within four weeks, with relapse or re-infection being the possible causes.107 Higher suc-cess rates were reported for eradicating MRSAfrom hospitalized patients using a protocol that in-cluded mupirocin intranasal ointment, daily chlo-rhexidine baths, systemic therapy, removal andreplacement of foreign bodies (e.g. catheters)and routine cleaning of the environment. Anystudies using a combination of interventionstogether gave better success rates of decoloniza-tion, but only for a restricted time before recoloni-zation. Therefore, it is difficult to evaluate the


effectiveness of one intervention alone, suchas nasal mupirocin or chlorhexidine 4% forskinwashing.108

Nasal carriageA Cochrane systematic review of randomizedcontrolled trials involving patients colonized withMRSA tested the efficacy of intranasal mupirocinointment compared with placebo in reducingcolonization and preventing infection among per-sistent carriers of S. aureus. Reviewers concludedthat there is insufficient evidence to support thewidespread use of topical or systemic antimicro-bial therapy for eradicating nasal or extranasalMRSA. Nevertheless, selective, short-term use ofmupirocin for specific patient groups may be use-ful, e.g. patients about to undergo majorsurgery.105

The effect of a five-day course of mupirocin inreducing nasal colonization in both mupirocin-susceptible (MS) and mupirocin-resistant (MR) pa-tients with MRSA (colonized, infected or both) inone hospital demonstrated that mupirocin waseffective in eradicating MS MRSA, but strains ofMR MRSA persisted after treatment suggestingtreatment failure rather than recolonization.109

Another study assessed the effectiveness of a pro-gramme of intranasal mupirocin throughout a neo-natal unit during an outbreak of MRSA. Theprogramme was effective in removing MRSA fromthe unit for three months and it therefore hasa role when conventional methods have failed.110

A prospective study considered the incidence ofMRSA carriage at the time of admission and therate of acquisition during hospitalization in pa-tients with liver disease.111 All nasal carriers ofMRSA were treated with mupirocin for five daysand underwent daily bodywashing using 4% chlo-rhexidine alternating with liquid soap. Nasalscreening of all patients continued weekly. Al-though 98.8% were culture negative after oneweek of treatment, subsequent recolonizationwas common, particularly in long-term patients,and subsequent treatment failure was due to MRstrains. As mupirocin did not reduce the risk ofMRSA and high-level MR strains emerged, they rec-ommended that mupirocin should only be usedonce.

In another study, patients were screened onadmission and staff were screened for nasalcarriage of MRSA on an ICU. Those positive forMRSA were treated with a five-day course ofmupirocin.112 Patients were placed in cohort isola-tion until discharge and a staff education pro-gramme of handwashing was introduced withmonitoring for compliance. Although MRSA rates

fell from 1.23 per 1000 to 0.53, follow-up wasonly for six months and it is not clear which ofthe interventions (i.e. isolation, mupirocin, hand-washing education programme or monitoring of ad-herence) was the most effective in reducing MRSA.

Throat carriageTreatment of throat carriage should only beconsidered in exceptional circumstances, e.g.when there is evidence that there is transmissionfrom a throat carrier in a continuing outbreak orwhen the patient carrying MRSA in the throat hasexperienced episodes of invasive infection. Adviceon choice of treatment and duration should besought from the consultant microbiologist. Throatcarriage may be associated with the presence offoreign bodies such as dentures and nasogastrictubes. Underlying conditions such as chronic ton-sillitis may also contribute to persistent throatcarriage.

In a study to assess the efficacy and safety ofa short course of oral vancomycin combined withintranasal mupirocin in the eradication of MRSA inpatients and staff, 69% (N¼ 24) were MRSA free af-ter one course.113 Eleven required further treat-ment and 80% (N¼ 28) reported side-effects.Consequently, great caution should be exercisedwhen considering this option. While the combina-tion of oral vancomycin and mupirocin is effectivein the elimination of MRSA colonization, the safetyof vancomycin requires further study.

Skin carriageIn a controlled study to eradicate skin carriage aspart of peri-operative prophylaxis to reduce MRSAsurgical site infections in patients undergoingorthopaedic prosthetic surgery, intranasal mupir-ocin was commenced one day pre-operatively andcontinued for five days.114 In addition, patientshad a pre-operative shower or bath using 2% triclo-san, and patients were followed-up for 12 months.The study investigators concluded that this regi-men could reduce the incidence of MRSA surgicalsite infection after orthopaedic surgery, probablyby reducing nasal MRSA carriage in the endemicsetting without selecting for mupirocin resistance.

A randomized controlled trial in two long-term-care facilities involved patients colonized withMRSA and compared single or combination, topicalor systemic antimicrobials with placebo.115 Investi-gators concluded that mupirocin was effective indecolonizing MRSA in persistent carriers, but theefficacy had decreased by day 90 after study entry.They suggested that pulse therapy with 14-daytreatment every two to three months may beuseful.


It has never been proven that, once colonized,any group of patients remain permanently MRSAfree after a decolonization regimen. There is alsothe possibility of confounding by re-acquisition ofthe same strain, especially for patients housed inhospital and in contact with other MRSA carriers.The evidence from the above study shows that anindividual can remain decolonized for up to 90days in a community healthcare facility.

RecommendationsNasal decolonizationPatients receiving prophylaxis for an operativeprocedure and in an outbreak situation under theadvice of the infection control team should un-dergo nasal decolonization. This should be un-dertaken by applying mupirocin 2% in a paraffinbase to the inner surface of each nostril (anteriornares) three times daily for five days. The patientshould be able to taste mupirocin at the back ofthe throat after application (Category 1b).

Mupirocin should not be used for prolongedperiods or used repeatedly (i.e. for more thantwo courses for five days), as resistance may beencouraged (Category 1a). The presence of a for-eign body such as nasogastric tube may reduce theefficacy of treatment with nasal mupirocin (Cate-gory 2).

Nasal decolonization using topical nasal mupir-ocin should be used with other forms of interventionsuch as skin decolonization with 4% chlorhexidinegluconate aqueous solution (Category 2).

Throat decolonizationSystemic treatment should only be prescribed onthe advice of the consultant microbiologist in thehospital, with appropriate monitoring (e.g. LFTs).If treatment is required, this should be restrictedto one course of treatment. The course should notbe repeated and the possible side-effects shouldbe explained to the patient (Category 1b).

Systemic treatment should be given in conjunc-tion with nasal mupirocin and skin decolonization(Category 1b).

The value of local treatment for throat carriagesuch as antiseptic gargles or sprays is uncertain,but may reduce the organism load (norecommendation).

Skin decolonizationSkin decolonization using 4% chlorhexidine body-wash/shampoo, 7.5% povidone iodine or 2% triclo-san is useful in eradicating or suppressing skincolonization for short times, particularly pre-operatively to reduce the risk of surgical siteinfections (Category 1a).

Patients should bathe daily for five days withthe chosen antiseptic detergent. The skin shouldbe moistened and the antiseptic detergent shouldbe applied thoroughly to all areas before rinsing inthe bath or shower. Special attention should bepaid to known carriage sites such as the axilla,groin and perineal area. The antiseptic should alsobe used for all other washing procedures and forbed bathing. Hair should be washed with anantiseptic detergent (Category 1a).

After satisfactory completion of a course oftreatment, clean clothing, bedding and towelsshould be provided after each bath and hairwash(Category 2).

For patients with eczema, dermatitis or otherskin conditions, attempts should be made to treatthe underlying skin condition. Advice on suitableeradication protocols for these individuals shouldbesought from a consultant dermatologist. Oilatumbath additive or Oilatum plus (with added benzal-konium chloride 6% and triclosan 2%) may be usedwith these patients; these should only be prescribedon the advice of a dermatologist (Category 2).

Careful consideration should be given in neo-nates regarding the appropriate use of agents usedfor decolonization. This should be discussed withthe infection control team and the paediatrician/neonatologist (Category 2).

5.4. Patient management

IntroductionThe proper management and placement of pa-tients with infectious conditions is one of thefundamentals of infection control in minimizingthe impact and potential transmission of anyinfectious condition, including MRSA. Evidence tosupport specific interventions is lacking due tomany studies employing multiple interventions inthe management of these patient groups, withoutdemonstrating the value of individual measures.However, there is good evidence that these in-terventions can substantially reduce the incidenceof MRSA, even in settings with a high level ofendemic MRSA.43 Accordingly, this section focuseson the principles of infection control and their ap-plication to the management of MRSA patients.Standard principles for isolation are detailed inthe appendices. This section is based on a reviewof the 1998 guidelines, encompassing current rec-ommendations as appropriate.3

General principlesThe primary objectives of infection control are theprevention of acquisition and spread of infectionby patients and staff. The priorities for targeted


control procedures are high-risk areas such asICUs, elective surgery (e.g. orthopaedic, cardiac,etc.) and patients who are identifiable as partic-ularly susceptible to infection.116

Infection control is the responsibility of all staffassociated with patient care. A high standard ofinfection control is required in all wards and units,although the level of risk may vary, and is animportant part of total patient care.117 Contractorsof health services should add their weight to theserequirements by specifically including infectioncontrol in contract specifications,118 and requiringthe active participation of senior clinical and man-agement staff in the control of hospital infection.117

Provision for good infection control will be an expec-tation of agencies, e.g. Healthcare Commission,Clinical Negligence Scheme for Trusts.

It is essential that infection control is seen as anorganizational responsibility and priority, that ade-quate facilities and resources are provided, and thatappropriate infection control staff and supportservices are available.116 The infection controlteam should agree a long-term strategy for controlwith hospital senior management, which shouldinclude provision for adequate isolation and hand-washing facilities in new or upgraded units orhospitals.118,119

To ensure a high standard of practice, it isessential that a strong emphasis is placed oninfection control being included as a core elementof induction programmes for all new staff. Itshould be included within undergraduate and post-graduate training,117 and in continuing educationprogrammes for all staff groups.120

The general principles of infection control applyto all healthcare facilities, and are applicable tothe control of MRSA. Infection control practice andfacilities should be monitored and auditedregularly by infection control and/or clinical staff,as appropriate, against published standards.121

Current evidence-based written policies andguidelines should be available in all wards andunits. The establishment of a link nurse/practi-tioner scheme should be considered to supportinfection control teams.116

It is the view of the Working Party that there isa lack of robust studies on the effectiveness of theuse of isolation wards for containing the spread ofMRSA. However, a series of studies appear todemonstrate that the employment of isolationwards is of benefit but that the numerical burdenof MRSA cases may overwhelm the facilities avail-able, and the individual’s underlying conditionmay be compromised if patients are removedfrom specialist care.122 The majority of caseswill be isolated in single rooms or isolation

wards. Where multiple cases are present, it maybe considered prudent to cohort patients inclosed bays to maximize resources, includingstaffing, within the speciality of their underlyingcondition.

Management of MRSA-infectedor -colonized patientsGiven that MRSA is considered to be widespreadwithin the NHS, the course of action taken for themanagement of patients colonized and/or in-fected with MRSA will depend on a variety offactors that will be influenced locally by the levelof MRSA within the organization.43 Factors deter-mining isolation practices include:

e type of hospital or facility (tertiary referral,district general, single speciality, intermediatecare);

e type of ward (non-acute, acute, admission, in-tensive care or other high-risk unit);

e facilities available for patient isolation;e ward design (availability of single-room accom-

modation; ‘Nightingale’, i.e. open-plan with nobays; wards with bays or cubicles);

e whether affected patients are likely to beheavy shedders of MRSA (e.g. those with burnsor infected eczema);

e resistance pattern, virulence and potentialtransmissibility of the organism (e.g. E-MRSA16, VISA, VRSA, etc.).

Risk of cross-infection can be categorized ac-cording to the risk or vulnerability of other patientson the ward or unit. The introduction of differentcategories of infection risk and infection controlresponses depends on the susceptibility of patientsand available resources. Categories of risk and theappropriate control procedures are outlined belowbut these may overlap. It is not possible to beprescriptive for all circumstances as decisionsneed to be based on the local situation. Local riskassessment together with these guidelines shouldindicate the appropriate course of action of theinfection control team. The trust’s infection controlpolicy should identify which clinical areas are in-cluded in each clinical risk category (see below),i.e. local priorities need to be specifically discussed,agreed and minuted as part of a specific policy.Depending on local clinical practice and ward casemix, it may be more practical for some hospitals tomerge clinical risk categories.

Patients who are colonized or infected withMRSA should be informed of their condition and itsimplications, and should be clearly identified forinfection control purposes.3


Recommended methods for improving identifi-cation of MRSA patients include:

e labelling medical records, in a manner agreedlocally to preserve patient confidentiality, sothat they are recognized immediately onadmission;

e electronic labelling of patient database;e ready access to a database of known affected

patients in the admissions and accident andemergency departments.

Patients may also be given cards indicating thatthey have been infected or colonized with MRSA inthe past (this has the benefit of informing staff ifthe patient is admitted to a different hospital orhealthcare centre for treatment other than thatwhere the original diagnosis was made).

General principles for MRSA managementActive and timely intervention can be effective inreducing the overall numbers of colonized andinfected patients.122 The infection control teamshould continue to assess the MRSA incidence, andwhether most cases are new acquisitions withinthe hospital, or admissions and transfers of alreadyinfected/colonized patients. Where an increasedincidence occurs, the general principles should bereviewed and re-inforced, with emphasis on moni-toring compliance with infection control policies;cleaning in affected wards should be increased,including a schedule for thorough cleaning of allwards in rotation; antibiotic policies should be re-viewed, particularly antibiotics used for prophylaxisand empirical therapy; and movement of patientsbetween wards should be reduced.26 Units contain-ing vulnerable patients at high risk of developinginvasive infection (e.g. patients with multipletrauma, patients following major life-saving sur-gery, patients on immunosuppressive therapy)should be given priority for maximal precautions.

Approach in high-risk areas of a hospitalHigh-risk areas, including specialist wards or units(as defined below), are those where the conse-quence of uncontrolled MRSA is serious because ofthe risk of invasive infection and difficulties intreatment.3 These include: intensive care, neona-tal intensive care, burns, transplantation, cardio-thoracic, orthopaedic, trauma, vascular surgery,renal, regional, national or international referralcentres, and other specialist areas as determinedby the infection control team, and as agreedwith the senior clinical staff of the units and therelevant trust management structure.

Approach in medium-risk areasThe level of risk within other areas of a hospital,e.g. admission wards, general surgical, urological,paediatric, general medical, elderly medicine,etc., and the measures that need to be appliedwill be determined locally by risk assessment andin accordance with the MRSA burden and thefacilities available.

Approach in low-risk areasThe level of intervention where there is a low riskof invasive infection, but a high risk of colonizationin low-risk areas, e.g. psychiatric, psycho-geriatricand long-term-care facility, including residentialhomes, should be determined by risk assessment,e.g. if it is a regular feeder to high-risk areas.

Category of isolationStandard source isolation procedures should beinstituted for affected patients. The patient’smedical and psychological welfare should not becompromised by unnecessarily restrictive infectioncontrol practices. The infection control team shouldbe contacted in case of doubt.

Individual single rooms or isolation wards (withen-suite facilities, if available) should be employedas the preferred standard of accommodation.26

Although the value of cohorting patients in areassuch as ICUs is disputed, nursing MRSA patients ina defined area of a ward and using designatedstaff may be a necessary alternative to using singlerooms where there are a number of affectedcases.50 Consideration will need to be given tothe need to protect the privacy and dignity of theindividual. Such areas should be capable of physicalseparation from other ward areas.122

The implications of MRSA colonization, infectionand treatment should be explained to the patientand close relatives at the time of diagnosis andideally prior to transfer to a side-room, isolationunit or designated area.123 Information leafletsshould be available giving general information onMRSA in the language appropriate to the recipient.Patients and visitors should be encouraged to playtheir part by complying with local recommenda-tions for hand hygiene. Leaflets on MRSA may beobtained from either local resources, the HealthProtection Agency (www.hpa.org.uk), the Associa-tion of Medical Microbiologists (www.amm.co.uk/newamm/html/public.htm), the Infection ControlNurses’ Association (www.icna.co.uk), or nationalresources such as the National Resource for Infec-tion Control (www.nric.org.uk) or the Departmentof Health’s simple guide to MRSA (www.dh.gov.uk/PolicyAndGuidance/HealthAndSocialCareTopics/HealthcareAcquiredInfection/HealthcareAcquired

http://www.hpa.org.uk

http://www.amm.co.uk/newamm/html/public.htm

http://www.amm.co.uk/newamm/html/public.htm

http://www.icna.co.uk

http://www.nric.org.uk

http://www.dh.gov.uk/PolicyAndGuidance/HealthAndSocialCareTopics/HealthcareAcquiredInfection/HealthcareAcquiredGeneralInformation/HealthcareAcquiredGeneralArticle/fs/en&percnt;63CONTENT_ID&percnt;614093113&percnt;38chk&percnt;617/XgcQ




GeneralInformation/HealthcareAcquiredGeneralArticle/fs/en?CONTENT_ID¼4093113&chk¼7/XgcQ).

Isolation proceduresHigh standards of hand decontamination are re-quired to minimize the risk of cross-infection.Hands should be adequately decontaminated be-fore and after patient contact, and on leaving anisolation facility. Hands should be decontaminatedby thorough washing and/or the application ofa 70% alcohol hand rub preparation.124

Disposable aprons or gowns should be worn byall staff handling the patient or having contactwith their immediate environment. This also ap-plies to visitors who assist with the patient’s bodilycare. Visitors who only have social contact withthe patient, such as shaking hands, do not need towear protective clothing but do need to decon-taminate their hands after leaving the room.

Gloves do not obviate the need for hand de-contamination and should be worn when there iscontact with body fluids and handling of contam-inated dressings or linen.

Masks are occasionally necessary for healthcareworkers such as when procedures may generatestaphylococcal aerosols, e.g. during sputum suc-tion or chest physiotherapy.

The door should be kept closed to minimizespread to adjacent areas. If this is likely to com-promise patient care, for instance in elderly con-fused patients, a risk assessment should be maderegarding whether the door may be kept open. Suchpatients often benefit from being nursed togetherin a cohort with other MRSA patients. The side-roomdoor must be kept shut during procedures that maygenerate staphylococcal aerosols, such as chestphysiotherapy or bed-making, etc.

Visitors to the cubicle or ward and staff fromother wards and departments, e.g. physiothera-pists, radiographers, other medical teams, stu-dents, etc., should only enter after permissionand instruction from the nurse in charge.

A card or information sheet detailing isolationprecautions should be displayed prominently.

Cleaning and decontaminationThere is little evidence of the role that theenvironment and equipment play in the trans-mission of MRSA in a facility,26,125,126 although inone extended outbreak of MRSA, a strong associa-tion was demonstrated when enhanced cleaningwas introduced and the outbreak ended.54 Ac-cordingly, general principles should be adoptedto minimize the bacterial burden within a facility.The ability of MRSA to survive in dust demon-strates the need for dust minimization, the

removal of fomites from contact surfaces, and theappropriate disposal of contaminated waste andlinen.127 Local policies for environmental cleaningand equipment decontamination, and waste andlinen management should state the necessary stan-dards, and should be applied rigorously.128 Thisguidance should be based on agreed national stan-dards which should be subjected to monitoring.129

Adequate handwashing facilities and antibacte-rial hand rub/gel should be available for staff andvisitor hand decontamination before and aftercontact with the patient or their immediateenvironment.

Instruments or equipment should preferably besingle-patient use. Multiple-patient-use itemsshould be decontaminated appropriately beforeuse on another patient in accordance with localpolicy or the manufacturer’s instructions.

There is some evidence that standard cleaningregimens are ineffective in eliminating MRSA con-tamination.126 There is a requirement for protocolsto be agreed for enhanced levels of cleaning, to in-clude additional time to enable the removal of allreservoirs of dust, e.g. ventilation ducts, radia-tors, equipment, etc. MRSA-contaminated patientareas should be cleaned after the patient’s dis-charge with hot water and detergent in accor-dance with local environmental decontaminationprotocol. Curtains should be removed and laun-dered if not single-use disposable curtains. Pillowsand mattress covers should be checked for dam-age. Therapy beds may need specialist cleaningin accordance with the manufacturer’s/hirer’sinstructions.

Additional general measures that may reducespread include installing antibacterial hand rubdispensers at the ward entrance and the individualbedsides, and using individual staff dispensers. Allstaff, patients and visitors entering and leaving theward should be required to decontaminate theirhands with the antibacterial hand rub.124

Finally, there should be planned, periodic andthorough cleaning of the whole ward, includingbedding and curtains.126,129

Clinical wasteAll waste should be categorized as clinical waste,and disposed according to local policy.130

LinenAll linen from patients infected or colonized withMRSA should be considered to be contaminated/infected, including bedding and adjacent curtains,and should be managed in accordance with localpolicy and national guidance.131






Ward closureWard closure to new admissions may need to beconsidered in certain circumstances on the basis ofrisk assessment. Factors influencing considerationof ward closure to admissions include:

e risk status of patients to be admitted, e.g.elective orthopaedics, coronary artery bypasssurgery;

e number of cases;e MRSA strain, e.g. virulence, resistance, etc.;e availability of alternative facilities;e staffing levels.

Before re-opening to new admissions, effectiveenvironmental decontamination is required.

Transfer and discharge of MRSA-colonizedor -infected patientsWithin the hospital

e Transfer of MRSA-affected patients to otherwards should be minimized to reduce therisk of spread, but this should not compro-mise other aspects of care, such asrehabilitation.

e Transport of the infected/colonized patientshould be supervised carefully;

e Lesions should be occluded whenever possiblewith an impermeable dressing;

e Attendants who may be in contact with the pa-tient should wear disposable plastic aprons toprotect their clothing whilst in contact withthe patient. Aprons should be removed whencontact with the patient has finished and dis-posed of as clinical waste;

e Gloves need only be worn if staff transportingthe patient have skin abrasions;

e The trolley or chair should be decontaminatedin accordance with local policy after use by thepatient and before being used for anotherpatient. All linen should be dealt with as in-fected, in accordance with local policy;

e Staff should decontaminate their hands thor-oughly after dealing with the patient andcleaning the trolley or chair.

Visits to outpatients and specialist departmentsVisits by MRSA patients to other departmentsshould be kept to a minimum. If this is necessary,either for investigation or treatment, prior ar-rangements should be made with staff of thereceiving department, so that control of infectionmeasures for that department can be imple-mented. These should include:

e dealing with these patients at the end of theworking session if at all possible;

e the patient should spend the minimum time inthe department, being sent for when the de-partment is ready and not left in a waitingarea with other patients;

e staff coming into close contact with the patientshould wear disposable aprons. Staff shouldavoid direct contact with other patients whilstdealing with an MRSA-positive patient;

e equipment and the number of staff attendingshould be kept to a minimum;

e surfaces with which the patient has had directcontact should be decontaminated with hotwater and detergent;

e linen should be treated as infected;e staff should decontaminate their hands after

contact with the patient.

Surgical interventionEvery effort should be made to eliminate orsuppress colonization or infection with MRSAbefore surgery.132 Decolonization of patients isaddressed in Section 5.3. If not possible, thefollowing should be performed to minimize thenumbers of bacteria colonizing the patient andreduce the risk of introduction of the organisminto an open wound. As part of pre-operativepreparation:

e bathe/shower the patient with an antiseptic de-tergent, applied direct to the skin as a wash,and rinsed off;

e cover affected lesions with an impermeabledressing;

e clean the area adjacent to the lesion with alco-holic chlorhexidine;

e apply mupirocin to the nose before the opera-tion if the patient is a nasal carrier;

e prophylactic antibiotic cover for surgical proce-dures in colonized or infected patients, followingdiscussion with a medical microbiologist (seeCG Gemmell, DI Edwards, AP Fraise et al. Guide-lines for the prophylaxis and treatmentof methicillin-resistant Staphylococcus aureus(MRSA) infections in the UK. J Antimicrob Che-mother. Advanced access February 27, 2006);

e consideration may be given to placing patientsat the end of the operating session. However,with effective theatre ventilation systems,there should be an adequate number of air ex-changes to provide a safe environment within15 min of removal of the MRSA patient fromthe operating theatre;133,136

e theatre surfaces in close contact or nearthe patient, such as the operating table or


instrument trolley, should be decontaminatedwith hot water and detergent followed by alco-hol before being used for the next patient;

e patients may be allowed to recover after surgeryin the operating theatre or an area not occupiedby other patients to avoid possible contamina-tion of the usual recovery area. If this is not pos-sible, the patients should be segregated as far aspossible within the recovery area, and nursed bystaff dedicated to their care, employing stan-dard isolation precautions.

Ambulance transportation

e Most MRSA carriers may be transported withother patients in the same ambulance withoutany special precautions, other than changingthe bedding used by the carrier. High-risk cate-gories of susceptible patients should not betransported in the same ambulance as a knownMRSA-positive patient. Lesions should becovered;

e The ambulance service should be notified inadvance by the responsible ward staff, if con-sidered necessary by the infection controlteam;

e There is no evidence that ambulance staff ortheir families are put at risk by transporting pa-tients with MRSA;

e To minimize the risk of cross-infection with anyinfectious agent, ambulance staff should usean antibacterial hand rub after contact withall patients as part of good, basic infectioncontrol practice;

e If further measures are required in special cir-cumstances, the infection control team shouldinform the ambulance service;

e No additional cleaning of the ambulance is re-quired after transporting an MRSA-positive pa-tient. Further information can be found in‘National Guidance and Procedures for Infec-tion Prevention and Control’ by the AmbulanceAssociation (June 2004).

Transfer to another hospitalMRSA infection or colonization should not be abarrier to good clinical care. Consequently, inter-hospital transfers should not be prevented ordelayed, although unnecessary movement shouldbe avoided.

Identification of infected or colonized patients isthe responsibility of the transferring hospital. Be-fore transfer, a member of the clinical team for the

patient, or a member of the infection control team,at the transferring hospital should inform either thereceiving ward staff or the infection control team atthe receiving hospital of the patient’s status.

Discharge of patientsMRSA patients should be discharged promptly fromhospital when their clinical condition allows. Thegeneral practitioner and others involved in thepatient’s care should be informed.

MRSA carriers will not normally require specialtreatment after discharge from hospital. If a treat-ment course needs to be completed in particularcircumstances, the infection control team shouldadvise the general practitioner about this.

If the patient is discharged to a residential carefacility, the medical and/or nursing staff should beinformed in advance. Colonization or infectionwith MRSA is not a contra-indication to the transferof a patient to a nursing or convalescent home.

It is essential that patients, their relatives andcarers should be fully briefed on MRSA, informedthat there is no risk of infection to healthyrelatives and contacts outside the hospital, andthat normal social interaction should not becompromised.

Patients should be advised that if they arehospitalized in future, they should advise admit-ting staff that they have been identified as carriersof MRSA in the past to ensure that they aremanaged appropriately.

There is no indication for routine screeningbefore discharge to the community, includingdischarge to residential or care homes.

Deceased patientsThe infection control precautions for handlingdeceased patients are the same as those used inlife. Any lesions should be covered with imperme-able dressings. Plastic body bags are not neces-sary, but may be employed as part of generalpractice in accordance with standard precautionsfor all patients. There is negligible risk to mortuarystaff or undertakers provided that standard in-fection control precautions are employed.

RecommendationsGeneral principlesThe general principles of infection control should beadopted for the management of patients with MRSA.Good infection control practice should be placed atthe centre of clinical practice, and requires theexplicit support of the organizational executive toensure that it is seen as having an appropriateposition within the organization and can be enforcedas a matter of clinical governance (Category 1b).


Standard infection control principlesA standard approach to isolation precautions shouldbe adopted in accordance with the general princi-ples of infection control, rather than introducingspecific guidance for the management of MRSA thatmay lead to differing standards (Category 1b).

Management of MRSA-infectedor -colonized patientsPatients should be managed in accordance with thetype of facility in which they are cared for, theresources available, and the level of risk that isposed to them and to others. Patients (and thefacilities that may house them) classified as being athigh risk of either contracting MRSA or for whom theconsequence of infection may have a high impactwill require a rigorous approach to screening,placement and treatment. Patients identified withMRSA infection or colonization should be informedof their condition, and local arrangements shouldbe made to ensure ease of identification ifre-admission to the facility occurs (Category 1b).

Patient isolationPatient isolation for those infected or colonizedwith MRSA will be dependent on the facilitiesavailable and the associated level of risk. Wherenew builds or refurbishment are planned, pub-lished guidelines should be adopted to provide themost appropriate facilities for patient care. Iso-lation should be in a designated closed area, whichshould be clearly defined; in most facilities, thiswill be either single-room accommodation or co-hort areas/bays with clinical handwashing facili-ties. Consideration should be given to the provisionof isolation wards to contain MRSA spread. Theprocedures for isolation should be stated clearly,and where necessary explained, to staff, patientsand visitors. Hospital staff entering isolation facil-ities should be required to adopt rigorously theprescribed isolation precautions, which should beaudited regularly. Non-staff visitors should berequested to adopt the necessary level of pre-cautions to minimize the risk of spread of MRSA toother areas of the facility (Category 1b).

Cleaning and decontaminationManagement of the environment and equipment isimportant in reducing the spread of MRSA. Cleaningregimens for isolation facilities should focus on theminimizing of dust and the removal of fomites fromcontact areas. This should be a two-fold approach;firstly, the management of the occupied facility,and then the terminal clean of the facility afterdischarge of the patient. Cleaning regimens and

products should be in accordance with local policy,but should include the removal of organic materialwith a general-purpose detergent. Cleaning regi-mens should be audited regularly

Patient equipment, e.g. wheelchairs, hoists,slings, sphygmomanometer cuffs, etc., shouldeither be capable of being decontaminated (andbe decontaminated before use with other pa-tients), or should be single-patient use and dis-carded as clinical waste at the end of a period ofuse (Category 1b).

Patient movementThe movement of patients with MRSA withina facility should be minimized to reduce the riskof cross-infection and any potential embarrass-ment for the patient. Where patients need toattend departments for essential investigations,the receiving area should be notified of thepatient’s MRSA status in advance of the transfer,and arrangements should be put in place tominimize their contact with other patients, i.e.to be called forward when the department is readyfor them and to ensure that they are not held incommunal waiting areas. Staff should adopt stan-dard precautions whilst in contact with thepatient.

Arrangements for transfer to other healthcarefacilities, e.g. hospitals, residential care homes,etc., should include notification of the individual’sMRSA status, as appropriate (Category 1b).

Surgical/invasive proceduresPrior to any planned invasive procedure, effortsshould be made to minimize the risk of infectionthrough topical and systemic decolonization, andprophylactic antimicrobial therapy, asappropriate.

It may be considered desirable to place theindividual at the end of a procedure list, althoughin environments with mechanically filtered venti-lation, such as operating theatre suites, thenumber of air exchanges should render this un-necessary. Good infection control practices, whichshould be in place between all patients, shouldreduce the risk of cross-infection (Category 1b).

TransportationThe risk of cross-infection from an MRSA-colonizedor -infected patient to other patients in anambulance environment is minimal. Good infec-tion control practices and routine cleaning(see also Ambulance Service Association guide-lines) should suffice to prevent cross-infection(Category 2).


DischargeGenerally, there is no requirement for patientscolonized with MRSA to continue with extendederadication protocols after discharge. This may bevaried in the event of anticipated re-admission toa hospital, especially for a planned invasive pro-cedure. It is appropriate that individuals/groupsinvolved in further care are informed of theindividual’s known MRSA status at discharge.

Patients and their appropriate contacts should befullybriefedandgiven relevant informationonMRSA,its implications and significance prior to discharge toreduce unnecessary anxiety and concern when re-turning to the home environment (Category 2).

5.5. Nursing staff workload and MRSAtransmission

Many studies have identified peaks in ward nursingstaff workload with increased rates of nosocomialinfection,18,25,90,134e153 and some are prospectivein design.25,70,90,147 This effect is consistent in allreports, and has most commonly been reportedin connection with staphylococcal cross-infection(including MRSA) and central-venous-catheter-associated bloodstream infection. This associationhas been found in a wide range of individual hospi-tal units, including adult and neonatal ICUs andburns units, and also at the whole-hospital and na-tional level. Various methodologies for assessingnursing workload have been used in these studies,ranging from simple measures of staff and patientnumbers, to more complex assessments of nursingcapability and of patient care requirements. Thewhole issue of nursing staff working conditionsand the resulting impact on nosocomial infectionhas been reviewed recently.145

There are obvious conflicts between the work-load demands made on nursing and other staff andtheir ability to perform recommended hand hygieneprocedures.139,140 This was highlighted by Voss andWidmer, who emphasized that clinical workloadpressures on nurses reduce the time available forperforming routine infection control proceduresand raise the likelihood that micro-organisms willbe transferred among patients.141 Similarly, ina large prospective study, Pittet et al. found thatmembers of nursing staff were least likely to per-form hand hygiene when they were most busy (andconsequently when they had the highest frequencyof times per hour when they should perform it).142

The details of control measures themselves intro-duced in response to MRSA outbreaks also affectward nursing staff workload (e.g. whether cohortingor closure of wards to new admissions is used143),

and it is logical to infer that extra infection controlprocedures are unlikely to be fully implemented byoverstretched staff. Finally, the inter-relationshipsamong workload, performance of hand hygiene andMRSA transmission on an ICU were modelled byGrundmann et al. They estimated that a 12% im-provement in performance of hand hygiene wouldhave compensated for the increased transmissionotherwise seen during periods of maximal pressureon staff.25 One state in the USA has introduced min-imum staffing ratios, while others have prohibitedmandatory overtime for nurses and held hospitalsaccountable for implementing approved staffingplans.146

RecommendationsThe Working Party emphasizes that inadequatenurse staffing is incompatible with effective in-fection control. Infection control teams and hos-pital managements should bear nursing workloadin mind (including staff numbers, grades and levelsof experience, and patient acuity) when planninglocal responses to MRSA and when reacting tooutbreaks, and adequate staffing resources mustbe given a high priority for all patient care areas(Category 1a).

Improving nurse staffing levels on an affectedward may allow improved adherence to localinfection control policies (Category 2), and shouldbe considered as a component of a package ofmeasures to control local outbreaks (Category 2).

6. S. aureus with reduced susceptibilityto vancomycin

Susceptibility to vancomycin in S. aureus is definedas a minimum inhibitory concentration (MIC)�4 mg/L.154,155 Until relatively recently, strainswith an MIC nearing that breakpoint were unheardof. However, since the first description of VRE inthe mid 1980s156 and the subsequent demonstrationin vitro that such resistance mediated by vanA geneswas transmissible to S. aureus,157 there was consid-erable anxiety and speculation amongst clinical mi-crobiologists about the potential for the emergenceof clinical isolates of S. aureus with reduced suscep-tibility to glycopeptide antibiotics. These fears wererealized when the first case of clinical infection withS. aureus with reduced susceptibility to vancomycinwas reported in Japan in 1997.158

Since that time, a small but increasing number ofclinical isolates of S. aureus with reduced glycopep-tide susceptibility have been reported.27 Becausethese cases share common features (patients have


almost invariably had previous MRSA colonization/infection and have received long and repeatedcourses of glycopeptide therapy159) it is nowapparent that this is not a homogeneous group andstrains differ in their phenotypic expression andunderlying genotypic mechanism of resistance.Further details of the nomenclature, mechanismsof resistance, characterization and laboratorydetection of these strains is provided in recentrecommendations.3a The following two paragraphsdefining the terminology of VISA, VRSA, GISA, etc.are taken from the aforementioned document.

There is some confusion regarding the definitionand nomenclature of reduced susceptibility to gly-copeptides in S. aureus, which is compounded by dif-ferences in breakpoints. The first S. aureus withreduced susceptibility to vancomycin (MIC 8 mg/L)was designated ‘VRSA’ by the British Society of Anti-microbialChemotherapybreakpoints.ThefirstUS re-port referred to these isolates as ‘VISA’ according toNational Committee of Clinical Laboratory Standardsbreakpoints. The broader generic term ‘GISA’ was in-troduced as these isolates also had reduced suscepti-bility to teicoplanin, although some have reducedsusceptibility to teicoplaninalone (TISA). Inaddition,two resistance phenotypes, heterogeneous and ho-mogeneous,havebeen recognized. Isolateswithhet-erogeneous intermediate resistance (hGISA orheteroGISA) appear to be susceptible to glycopep-tides (vancomycin MIC� 4 mg/L) but contain sub-populations of cells at frequencies of >10�6 thatexhibit reduced susceptibility (vancomycin MIC 8e16 mg/L). The term ‘S. aureus with reduced vanco-mycin susceptibility’ has been suggested for isolateswith vancomycin MIC of �4 mg/L, which may moreclosely resemble GISA than glycopeptide-susceptibleS. aureus (GSSA).

These definitions are largely phenotypic. TheWorking Party recommends that the terms ‘VISA’,‘TISA’ and ‘GISA’ should be used for isolates withhomogeneous low-level resistance, hVISA, hTISAand hGISA should be used for isolates with hetero-geneous low-level resistance, and VRSA should beused for isolates with higher levels of resistance tovancomycin (MIC� 32 mg/L). It should be appreci-ated, however, that infections caused by strainswith homogeneous low-level resistance are unlikelyto respond to therapy with glycopeptides in seriousinfections, and this may also be the case with someheterogeneous low-level-resistant isolates.

6.1. Recent epidemiology

VISA/GISA and VRSA infections are still relativelyrare. Only a handful of cases of infection caused by

VISA27,160e165 and three cases of infection caused byVRSA28,166,167 have been reported in the USA. Al-though the first VISA isolate was described in Japanin 1997,158 a recent nationwide survey suggestedthat such strains are not widely disseminated in Jap-anese hospitals.168 France,169 Germany,170 theUK,171 Spain,172 Hong Kong,173 Greece,174 Italy,175

Australia,176 Brazil,177 Korea178 and Poland179 haveall described isolation of S. aureus strains with re-duced susceptibility to vancomycin. While some ofthese isolates have been associated with failure oftreatment of individual patients, most (especiallyhVRSA) have only been identified as part of retro-spective testing surveys. Tenover argued that sinceMRSA may be highly transmissible in healthcare set-tings, it is prudent to assume that VISA/GISA (and byinference, VRSA) may be no less transmissible.180

Although VISAs have been described in many partsof the world, these are comparatively rare and donot generally appear to be a problem.

As indicated above, in virtually all cases infec-tions with these organisms have arisen in patientswith significant previous morbidity, previous MRSAcolonization or infection, and prior glycopeptidetherapy. Fridkin et al. performed prospective sur-veillance and a nested case-control study of US pa-tients infected with S. aureus with reducedvancomycin susceptibility.159 This confirmed ante-cedent vancomycin use and prior MRSA infection inthe two to three months prior to the current infec-tion as independent risk factors.

Further elucidation of the epidemiology ofthese organisms will be dependent on clinicallaboratory vigilance. It is particularly importantthat sufficiently robust methods are used to detectlow levels of glycopeptide resistance in S. aureusisolates.181

6.2. General prevention and controlmeasures

As a result of the relatively limited experience ofVISA/GISA infections worldwide, and the evensmaller number of documented VRSA infections,there is no randomized clinical trial data on which tobase control measures for these organisms. None-theless, guidelines for the management of patientsinfected and/or colonized with isolates of S. aureuswith reduced susceptibility to glycopeptides havebeen published and are broadly based upon existingguidance for the control of VRE infections.182,183

This guidance has been implemented in those casesof VISA/GISA/VRSA infection described in the litera-ture, and some of these reports provide limitedevidence for the effectiveness of these measures,which is discussed in more detail below.


6.3. Effectiveness of control measuresfor VISA/GISA and VRSA

There have been no randomized controlled trials toassess the effectiveness of control measures forVISA/GISA and VRSA. The following suggested mea-sures are based upon the interim guidelines of theCDC for prevention and control of staphylococcalinfection associated with reduced susceptibility tovancomycin.182,183 The evidence to support theeffectiveness of this guidance is largely basedupon observational studies that have looked eitherretrospectively or prospectively for evidence oftransmission to known contacts after precautionshave been implemented.

In the USA, all infection control personnel dealingwith confirmed VISA/GISA infections to date haveadopted practices based to a large extent upon theHICPAC guidelines. Approximately 400 culture spec-imens obtained from contacts (healthcare workersand family members) failed to identify any addi-tional individuals colonized or infected with VISA/GISA.160 In most of these patients, contact precau-tions were already in place as a result of pre-existingMRSA and/or VRE colonization. Retrospective anal-ysis of chronic haemodialysis patients in the USA col-onized with VISA/GISA also failed to find evidence oftransmission to other patients or healthcareworkers after implementation of the HICPAC recom-mendations.184 Similar findings have been reportedfor a VISA-colonized patient in the home healthcaresetting.164 Similarly, secondary spread of VRSA hasnot been documented following implementationof this guidance.28

It has been demonstrated that these strains dohave the potential for nosocomial transmission, asevidenced by dissemination within Japanese hospi-tals.185 In a recent report from Brazil, the introduc-tion of control measures based upon the HICPACguidance was associated with cessation of ongoingtransmission of VISA/GISA strains.156 In anothersmall study in a French ICU, eradication of VISA/GISA colonization in three staff members (one ther-apist and two nurses) was associated with a reduc-tion of VISA/GISA colonization and/or infectionrates of ICU patients from 1.5% to 0.2%.186

6.4. Recommendations for controlof VISA and VRSA

In the absence of randomized controlled trial dataand on the basis of the descriptive studies outlinedabove and a strong theoretical rationale, recom-mendations for the control of these organismsremain the province of existing best practice and

professional opinion. These measures can be con-veniently considered under the headings of pre-vention, surveillance and precautions.

PreventionAntibiotic resistance flourishes when antimicrobialdrugs are ‘abused, misused and dispensed at levelslower than treatment guidelines dictate’.187 Virtu-ally all strains of S. aureus with reduced susceptibil-ity to glycopeptide antibiotics described to date arethought to have arisen from pre-existing reservoirsof MRSA, usually in patients with chronic underlyingdisease who have received multiple and/or pro-longed courses of glycopeptide treatment.159,188 Itseems logical, therefore, to ensure that measuresoutlined elsewhere in this document for control ofMRSA are implemented within the healthcare insti-tution, and that careful antibiotic stewardship isemployed to minimize the inappropriate use of gly-copeptide agents (Category 1b).83

Where the use of such agents is deemed appropri-ate, clinicians should ensure that adequate dosagesare given to ensure that that therapeutic levels areobtainedat the siteof infectionand that thedurationof therapy is not unnecessarily prolonged. Thesemeasures will help to reduce the likelihood of re-sistant strains arising de novo (Category 1b).

SurveillanceIt is vital that clinicians and microbiologists remainaware of the potential for emergence of strains ofS. aureus with reduced susceptibility to glycopep-tide antibiotics, and that this awareness is re-flected in ongoing laboratory-based surveillanceprogrammes.189 The detection of intermediate-level resistance is challenging for laborato-ries.190,191 This is especially true for strains thatare heterogeneous in their expression of glycopep-tide resistance.28,192e194 A high level of suspicionmust be maintained, particularly in patients whohave received multiple and/or prolonged coursesof glycopeptide antibiotics or who are known tobe colonized/infected with MRSA and VRE. De-tailed recommendations and levels of evidencefor the laboratory detection of these strains aregiven in Ref. 3a.

The laboratory must notify the relevant clini-cian and infection control personnel as soon aspossible after the isolation of a presumptive S. au-reus isolate with reduced glycopeptide sensitivityin order that control measures can be imple-mented with minimum delay. The isolate shouldalso be forwarded to a reference laboratory. It isalso important that the relevant national surveil-lance network is notified to ensure that accurate


information about the epidemiology and spread ofthese organisms is gathered (Category 1b).

Control precautions (all Category 1b)Action to be taken on identificationof a case of VISA/GISA or VRSA

e The laboratory should immediatelynotify the rel-evant clinician and infection control personnel;

e The infection control team should immediatelyidentify where the patient is and where the pa-tient has been during all of the current admis-sion, including transfers from other healthcarefacilities;

e The relevant national surveillance organiza-tion, e.g. the Health Protection in Scotland,the Health Protection Agency in England andWales, and the Health Protection Agency (Com-municable Disease Surveillance Centre) inNorthern Ireland, should be notified.

If the patient is still an inpatient

e The number of healthcare workers caring forthe patient should be reduced. This will causeproblems for those who are allocated to carefor the patient;

e Healthcare workers with chronic skin condi-tions, e.g. eczema or psoriasis, should not beinvolved in direct care of the patient;

e All staff caring for the patient must be aware ofhow the organism is transmitted and the pre-cautions necessary to prevent this;

e The patient should be cared for in a single roomwith toilet facilities and a wash hand basin;

e The patient and visitors need to understand theneed for isolation;

e Fans should not be used to control the patient’stemperature;

e Appropriate infection control proceduresshould be implemented.1. Use standard precautions. Gowns/dispos-

able aprons and disposable gloves shouldbe worn by all those entering the patient’sroom. Clean, non-sterile gloves and gowns/aprons are adequate. Staff should usetheatre-style scrub suits in addition to pro-tective clothing to ensure that healthcareworkers do not take uniforms home tolaunder.

2. Disposable masks and eye protection shouldbe worn by carers for procedures likely togenerate aerosols/splashing. Use of closedsuction systems will help to reduce aerosols.

3. Hand hygiene should be performed withan antibacterial preparation, before andafter patient contact. Visibly soiled handsshould be washed with soap prior todisinfection.

4. Non-disposable items that cannot be easilycleaned or disinfected (e.g. sphygmoma-nometer cuffs) should be dedicated for usesolely by the infected/colonized patient.

5. Patient charts and records should be keptoutside the isolation room.

6. Linen should be treated as infected. It mustbe discarded into alginate bags within thepatient’s room and a secondary bag outsidethe room.

7. All waste should be discarded into a clinicalwaste bag inside the room, and bags shouldsubsequently be disposed of as per hospitalpolicy.

8. Transfers of colonized/infected patientswithin and between institutions should beavoided unless essential, and the receivinginstitution should be made aware of the pa-tient’s colonization/infection status prior totransfer.

9. After discharge, the room in which the pa-tient has been cared for should be cleaned,with special attention given to horizontalsurfaces and dust-collecting areas. Hotwater and detergent are usually satisfac-tory. Curtains should be changed.

10. Compliance with infection control proce-dures should be monitored.

Screening (all Category 1b)Patients

e Nose, axillae, perineum, skin lesions and ma-nipulated sites of the index case and all otherpatients in the unit should be screened for car-riage of VISA/GISA or VRSA;

e The infection control team should review theadmission history of the patient and determineif screening needs to be extended to otherareas.

Staff

e Agreement with staff on the need for screeningshould be sought;

e Nose, axillae and perineum of healthcareworkers and others with close physical contactwith the case should be screened for carriageof VISA/GISA or VRSA;

e Healthcare workers who maintain contactwith the patient will require weekly screening.


This may require significant support for thesestaff;

e Feedback of results and maintenance of confi-dentiality should be considered.

Eradication (all Category 1b)

e Eradication of colonization/carriage of pa-tients and healthcare workers should be at-tempted (see section on eradication of MRSAcarriage);

e Colonized staff should be excluded from workuntil eradication of carriage is achieved.

7. Recommendations for futureresearch

e The Working Party recommend a study of theclinical and cost-effectiveness of rapid screen-ing methods (such as polymerase chain reac-tion) for MRSA, linked to their ability to directefficient use of physical isolation facilitiesand procedures, decolonization proceduresand glycopeptide surgical prophylaxis;

e The Working Party recommend that studiesshould be carried out to determine the sensitivityand clinical effectiveness of different screeningstrategies for considering patients free fromMRSA carriage (or as not having acquired carriageafter exposure); e.g.whether two or three setsofswabs should be taken between 48 h and oneweek apart, and of what sites;

e The Working Party recommend performance ofstudies investigating the impact of staff work-load on MRSA infection and control, on thevalue of setting limits on staff workload, andon the impact of infection control measureson staff workload. The Working Party endorsethe recommendations in this area made bythe EPIC guideline project;144

e The Working Party advocate an assessment ofthe effectiveness of current and emerging ap-proaches to environmental decontaminationand their impact on MRSA colonization and in-fection rates;

e The Working Party recommend that researchshould be carried out in comparing the effec-tiveness of different cleaning regimens andtechnologies together with an assessment oftheir cost-effectiveness. For example, a com-parison of conventional cleaning techniqueswith steam cleaning or cleaning with hydrogenperoxide vapour in their ability to eradicateS. aureus and other important pathogens, e.g.C. difficile;

e The Working Party recommend that researchshouldbeconducted into theeffectivenessofdif-ferent skin decolonization/disinfection regimensin eradicating MRSA and the cost-effectiveness ofthe regimens. Suitable agents may include triclo-san, hexachlorophene, chlorhexidine or povi-done iodine alone and in combination;

e The Working Party recommend that researchshould be carried out into the effectivenessof using local treatment for throat carriageof MRSA; e.g. a comparison of vancomycin ortyrothricin throat lozenges and the use of mu-pirocin, alone or in combination, togetherwith a cost/benefit analysis;

e The Working Party recommend that further re-search should be done to produce guidelinesfor the management of MRSA in the commu-nity, e.g. care homes and primary care. Itwould be useful to attempt to link the strainof community-acquired MRSA with the fre-quency and nature of hospital attendance.This will need to be linked to an assessmentof different case definitions for hospital- andhealthcare-associated MRSA, as well as com-munity-acquired MRSA. Agreement on appro-priate denominators and what constitutesa ‘new case’ of MRSA is also required.

Acknowledgements

This exercise was initiated by the Specialist Advi-sory Committee on Antimicrobial Resistance, anindependent advisory committee set up to provideexpert scientific advice on resistance issues arisingfrom medical, veterinary and agricultural use ofantimicrobials. Established in 2001 followingrecommendations in the House of Lords SelectCommittee on Science and Technology’s originalreport ‘Resistance to Antibiotics and other Antimi-crobial Agents’, the Committee advises the UKGovernment on its strategy to minimize illness anddeath due to antimicrobial-resistant infection andto maintain the effectiveness of antimicrobialagents in their medical, veterinary and agriculturaluse.

The Working Party thanks Potenza Atiogbe, Cen-tral Library, Health Protection Agency, Colindale,London, UK for the initial electronic literaturesearch, Heather Loveday, CM Pellowe, SRLJ Jonesand Robert Pratt, Richard Wells Research Centre,Thames Valley University, London, UK for thesystematic review, and Evonne Curran for contribu-tions on the quality improvement model and figures.The Working Party is also grateful to all those whoresponded during the consultation period.


Comment on the editorial process

This Working Party Report was put out forconsultation on 30 June 2005 (the consultationperiod closed on 1 August 2005) and was amendedin light of the comments received prior to itssubmission to the Journal of Hospital Infection.This national consultation exercise amongst majorstakeholders and other interested parties replacedthe Journal’s peer review process.

Appendices

Standard principles for preventinghospital-acquired infectionsy,195

Hospital environmental hygiene

e The hospital environment must be visiblyclean, free from dust and soilage, and accept-able to patients, their visitors and staff.

e Where a piece of equipment is used for morethan one patient (e.g. commode, bath hoist), itmust be cleaned following every episode of use.

e Statutory requirements must be met in relationto the safe disposal of clinical waste, laundryarrangements for used and infected linen,food hygiene and pest control.

e All staff involved in hospital hygiene activitiesmust be included in education and training re-lated to the prevention of hospital-acquiredinfection.

Hand hygiene

e Hands must be decontaminated immediatelybefore each and every episode of direct pa-tient contact/care and after any activity orcontact that potentially results in hands be-coming contaminated.

e Hands that are visibly soiled or potentiallygrossly contaminated with dirt or organic mate-rial must be washed with liquid soap and water.

e Application of an alcohol-based hand rub orhandwashing with liquid soap and water mustbe performed to decontaminate hands betweencaring for different patients, or between differ-ent caring activities for the same patient.

e All wrist and (ideally) hand jewellery must beremoved at the beginning of each clinical shift

y These guidelines on the standard principles for preventinghospital-acquired infections are being reviewed and will bepublished as a supplement to the Journal of Hospital Infectionduring 2006.

before regular hand decontamination begins.Cuts and abrasions must be covered with wa-terproof dressings.

e Effective handwashing involves three stages:preparation, washing and rinsing, and drying.Preparation requires wetting hands under tepidrunning water before applying liquid soapor an antimicrobial preparation. The handwashsolution must come into contact with all thesurfaces of the hand. The hands must berubbed together vigorously for a minimum of10e15 s, paying particular attention to thetips of the fingers, the thumbs and the areasbetween the fingers. Hands should be rinsedthoroughly prior to drying with good-qualitypaper towels.

e When decontaminating hands with an alcoholhand rub, hands should be free of dirt and or-ganic material. The hand rub solution mustcome into contact with all surfaces of thehand. The hands must be rubbed together vig-orously, paying particular attention to the tipsof the fingers, the thumbs and the areas be-tween the fingers, until the solution has evapo-rated and the hands are dry.

e An emollient hand cream should be appliedregularly to protect skin from the drying ef-fects of regular hand decontamination. If a par-ticular soap, antimicrobial handwash or alcoholproduct causes skin irritation, occupationalhealth advice should be sought.

The use of personal protective equipment

e Protective equipment should be selected onthe basis of an assessment of the risk of trans-mission of micro-organisms to the patient, andthe risk of contamination of healthcare practi-tioners’ clothing and skin by patients’ blood,body fluids, secretions and excretions.

e Gloves must be worn for invasive procedures,contact with sterile sites, non-intact skin andmucous membranes, all activities that havebeen assessed as carrying a risk of exposureto blood, body fluids, secretions and excre-tions, and when handling sharp or contami-nated instruments.

e Gloves should be worn as single-use items.Gloves should be put on immediately before anepisode of patient contact or treatment andshould be removed as soon as the activity iscompleted. Gloves should be changed betweencaring for different patients, or between differ-ent care/treatment activities for the samepatient.


e Gloves must be disposed of as clinical wasteand hands should be decontaminated followingthe removal of gloves.

e Gloves conforming to European Communitystandards and of an acceptable quality mustbe available in all clinical areas.

e Alternatives to natural rubber latex (NRL)gloves must be available for use by practi-tioners and patients with NRL sensitivity.

e Powdered and polythene gloves should not beused in healthcare activities.

e Disposable plastic aprons should be wornwhen there is a risk that clothing or uniformmay become exposed to blood, body fluids,secretions and excretions, with the exceptionof sweat.

e Full-body fluid-repellent gowns should be wornwhere there is a risk of extensive splashing ofblood, body fluids, secretions and excretions,with the exception of sweat, on to the skin ofhealthcare practitioners.

e Plastic aprons should be worn as single-useitems for one procedure or episode of patientcare, and then discarded and disposed of asclinical waste.

e Face masks and eye protection should be wornwhere there is a risk of blood, body fluids, se-cretions and excretions splashing into theface and eyes.

e Respiratory protective equipment should beused when clinically indicated.

The safe use and disposal of sharps

e Sharps must not be passed directly from handto hand, and handling should be kept toa minimum.

e Needles must not be bent or broken prior to useor disposal.

e Needles and syringes must not be disassembledby hand prior to disposal.

e Needles should not be recapped.e Used sharps must be discarded into a sharps

container (conforming to UN3291 and BS 7320standards) at the point of use. These mustnot be filled above the mark indicating thatthey are full. Containers in public areas mustnot be placed on the floor and should be lo-cated in a safe position.

e The use of needlestick-prevention devicesshould be considered where there are clear in-dications that they will provide safe systems ofworking for healthcare practitioners.

e A rigorous evaluation of needlestick-preventiondevices should be conducted to determinetheir effectiveness, acceptability to

practitioners, impact on patient care and costbenefit prior to widespread introduction.

References

1. Report of a Combined Working Party of the Hospital Infec-tion Society and British Society for Antimicrobial Chemo-therapy. Guidelines for the control of meticillin-resistantStaphylococcus aureus. J Hosp Infect 1986;7:193e201.

2. Working Party Report. Guidelines for the control of epi-demic methicillin-resistant Staphylococcus aureus.J Hosp Infect 1990;16:351e377.

3. Duckworth G, Cookson B, Humphreys H, Heathcock R. Re-vised methicillin-resistant Staphylococcus aureus infectioncontrol guidelines for hospitals. Report of a Working Partyof the British Society for Antimicrobial Chemotherapy, theHospital Infection Society and the Infection Control NursesAssociation. J Hosp Infect 1998;39:253e290.

3a. Brown DFJ, Edwards DI, Hawkey PM, et al. Guidelinesfor the laboratory diagnosis and susceptibility testingof methicillin-resistant Staphylococcus aureus (MRSA).J Antimicrob Chemother 2005;56:1000e1018.

4. Jevons MP. Celbenin-resistant staphylococci. BMJ 1961;1:124e125.

5. Barber M. Methicillin-resistant staphylococcus. J ClinPathol 1961;14:385e393.

6. Harbarth S, Pittet D. Methicillin-resistant Staphylococcusaureus. Lancet Infect Dis 2005;5:653e663.

7. Hiramatsu K. Reduced susceptibility of Staphylococcus au-reus to vancomycin e Japan, 1996. MMWR Morb MortalWkly Rep 1997;46:624e626.

8. Centers for Disease Control and Prevention. Staphylococ-cus aureus resistant to vancomycin e United States. JAMA2002;288:824e825.

9. European Antimicrobial Resistance Surveillance System(EARSS). Annual report EARSS e 2003. Bilthoven, TheNetherlands: RIVM; 2004.

10. Salgado CD, Farr BM, Calfee DP. Community-acquired meth-icillin-resistant Staphylococcus aureus: a meta-analysisof prevalence and risk factors. Clin Infect Dis 2003;36:131e139.

11. Communicable Disease Surveillance Centre, Public HealthLaboratory Service. Antimicrobial resistance in 2000.England and Wales. London: Public Health LaboratoryService; 2002.

12. Johnson AP, Pearson A, Duckworth G. Surveillance andepidemiology of MRSA bacteraemia in the UK. J AntimicrobChemother 2005;56:455e462.

13. McDonald P, Mitchell E, Johnson H, et al. Epidemiologyof MRSA: the north/south study of MRSA in Ireland 1999.J Hosp Infect 2003;54:130e134.

14. Tiemersma EW, Bronzwaer SLAM, Lyytikainen O, et al.Methicillin-resistant Staphylococcus aureus in Europe,1999e2002. Emerg Infect Dis 2004;10:1627e1634.

15. Lacey RW. Multi-resistant Staphylococcus aureus e a suit-able case for inactivity? J Hosp Infect 1987;9:103e105.

16. Barrett SP, Mummery RV, Chattopadhyay B. Trying tocontrol MRSA causes more problems than it solves.J Hosp Infect 1998;39:85e93.

17. Teare EL, Barret SP. Stop the ritual of tracing colonisedpeople. BMJ 1997;314:665e666.

18. Farrington M, Redpath C, Trundle C, Coomber S,Brown NM. Winning the battle but losing thewar: methicillin-resistant Staphylococcus aureus (MRSA)infection at a teaching hospital. Q J Med 1998;91:539e548.


19. Shannon T, Edgar P, Villarreal C, Herndon DN, Phillips LG,Heggers JP. Much ado about nothing: methicillin-resistantStaphylococcus aureus. J Burn Care Rehabil 1997;18:326e331.

20. Kotilainen P, Routamaa M, Peltonen R, et al. Elimination ofepidemic methicillin-resistant Staphylococcus aureus froma university hospital and district institutions, Finland.Emerg Infect Dis 2003;9:169e175.

21. Kerttula A-M, Lyytikainen O, Salmenlinna S, Vuopio-Varkila J. Changing epidemiology of methicillin-resistantStaphylococcus aureus in Finland. J Hosp Infect 2004;58:109e114.

22. Rubinovitch B, Pittet D. Screening for methicillin-resistantStaphylococcus aureus in the endemic hospital: whathave we learned? J Hosp Infect 2001;47:9e18.

23. Boyce JM, Havill NL, Kohan C, Dumigan DG, Ligi CE. Do in-fection control measures work for methicillin-resistantStaphylococcus aureus? Infect Control Hosp Epidemiol2004;25:395e401.

24. McBryde ES, Bradley LC, Whitby M, McElwain DLS. Aninvestigation of contact transmission of methicillin-resistant Staphylococcus aureus. J Hosp Infect 2004;58:104e108.

25. Grundmann H, Hori S, Winter B, Tami A, Austin DJ. Riskfactors for the transmission of methicillin-resistant Staph-ylococcus aureus in an adult intensive care unit: fittinga model to the data. J Infect Dis 2002;185:481e488.

26. Muto CA, Jernigan JA, Ostrowsky BE, et al. SHEA guidelinesfor preventing nosocomial transmission of multi-drug-resistant strains of Staphylococcus aureus and Enterococ-cus. Infect Control Hosp Epidemiol 2003;24:362e386.

27. Srinivasan A, Dick JD, Perl TM. Vancomycin resistance instaphylococci. Clin Microbiol Rev 2002;15:430e438.

28. Staphylococcus aureus resistant to vancomycin e UnitedStates,2002.MMWRMorbMortalWklyRep2002;51:565e567.

29. Anon. Report: Deaths involving MRSA: England and Wales,1999e2003. Health Stat Q 2005;25:60e65.

30. Selvey LA, Whitby M, Johnson B. Nosocomial methicillin-resistant Staphylococcus aureus bacteremia: is it anyworse than nosocomial methicillin-sensitive Staphylococ-cus aureus bacteremia? Infect Control Hosp Epidemiol2000;21:645e648.

31. Romero-Vivas J, Rubio M, Fernandez C, Picazo JJ. Mortal-ity associated with nosocomial bacteraemia due to methi-cillin-resistant Staphylococcus aureus. Clin Infect Dis1995;21:1417e1423.

32. Whitby M, McLaws M-L, Berry G. Risk of death from meth-icillin-resistant Staphylococcus aureus bacteraemia:a meta-analysis. Med J Aust 2001;175:264e267.

33. Cosgrove SE, Sakoulas G, Perencevich EN, Schwaber MJ,Karchmer AW, Carmeli Y. Comparison of mortality associ-ated with methicillin-resistant and methicillin-susceptibleStaphylococcus aureus bacteremia: a meta-analysis. ClinInfect Dis 2003;36:53e59.

34. Rello J, Torres A, Ricart M, et al. Ventilator-associatedpneumonia by Staphylococcus aureus: comparison ofmethicillin-resistant and methicillin-sensitive episodes.Am J Respir Crit Care Med 1994;150:1545e1549.

35. Engeman JJ, Carmeli Y, Cosgrove SE, et al. Adverseclinical and economic outcomes attributable to methicil-lin resistance amongst patients with Staphylococcusaureus surgical site infection. Clin Infect Dis 2003;36:592e598.

36. Stamm AM, Long MN, Belcher B. Higher overall nosocomialinfection rate because of increased attack rate of methi-cillin-resistant Staphylococcus aureus. Am J Infect Control1993;21:70e74.

37. Abramson M, Sexton DJ. Nosocomial methicillin-resistantand methicillin-susceptible Staphylococcus aureus primarybacteraemia: at what costs? Infect Control Hosp Epidemiol1999;20:408e411.

38. Kim T, Oh PI, Simor AE. The economic impact of methicil-lin-resistant Staphylococcus aureus in Canadian hospitals.Infect Control Hosp Epidemiol 2001;22:99e104.

39. Vriens M, Blok H, Fluit A, Troelstra A, van der Werken C,Verhoef J. Costs associated with a strict policy to eradi-cate methicillin-resistant Staphylococcus aureus in a Dutchuniversity medical centre: a 10 year survey. Eur J Clin Mi-crobiol Infect Dis 2002;21:782e786.

40. Chaix C, Durand-Zaleski I, Alberti C, Brun-Buisson C. Con-trol of endemic methicillin-resistant Staphylococcusaureus. A costebenefit analysis in an intensive care unit.JAMA 1999;282:1745e1751.

41. Karchmer TB, Durbin LJ, Simonton BM, Farr BM. Cost-ef-fectiveness of active surveillance cultures andcontact/droplet precautions for control of methicillin-re-sistant Staphylococcus aureus. J Hosp Infect 2002;51:126e132.

42. Harbarth S, Martin Y, Rohner P, Henry N, Auckenthaler R,Pittet D. Effect of delayed infection control measures ona hospital outbreak of methicillin-resistant Staphylococcusaureus. J Hosp Infect 2000;46:43e49.

43. Cooper BS, Stone SP, Kibbler CC, et al. Isolation measuresin the hospital management of methicillin resistant Staph-ylococcus aureus (MRSA): systematic review of the litera-ture. BMJ 2004;329:1e8.

44. Souweine B, Traore O, Aublet-Cuvelier B, et al. Role of in-fection control measures in limiting morbidity associatedwith multi-resistant organisms in critically ill patients.J Hosp Infect 2000;45:107e116.

45. Blumberg LH, Klugman KP. Control of methicillin-resistantStaphylococcus aureus bacteraemia in high-risk areas. EurJ Clin Microbiol Infect Dis 1994;13:82e85.

46. Stone S, Kibbler C, How A, Balestrini A. Feedback isnecessary in strategies to reduce hospital acquiredinfection. BMJ 2000;321:302.

47. Cosseron-Zerbib M, Roque Afonson AM, Naas T, et al. Acontrol programme for MRSA (methicillin-resistantStaphylococcus aureus) containment in a paediatricintensive care unit: evaluation and impact on infectionscaused by other micro-organisms. J Hosp Infect 1998;40:225e235.

48. Valls V, Gomez-Herruz P, Gonzalez-Palacios R, et al. Long-term efficacy of a program to control methicillin-resistantStaphylococcus aureus. Eur J Clin Microbiol Infect Dis1994;13:90e95.

49. Rumbak MJ, Cancio MR. Significant reduction in methicillin-resistant Staphylococcus aureus ventilator-associatedpneumonia associated with the institution of a preventionprotocol. Crit Care Med 1995;23:1200e1203.

50. Cepeda JA, Whitehouse T, Cooper B, et al. Isolation of pa-tients in single rooms or cohorts to reduce spread of MRSAin intensive-care units: prospective two-centre study. Lan-cet 2005;365:295e304.

51. Huskins WC, Goldmann DA. Controlling methicillin-resistant Staphylococcus aureus, aka ‘Superbug’. Lancet2005;365:273e275.

52. Papia G, Louie M, Tralla A, et al. Screening high-risk pa-tients for methicillin-resistant Staphylococcus aureus onadmission to the hospital: is it cost effective? Infect Con-trol Hosp Epidemiol 1999;20:473e477.

53. Kunori T, Cookson B, Roberts JA, Stone S, Kibbler CC. Cost-effectiveness of different MRSA screening methods. J HospInfect 2002;51:189e200.


54. Rampling A, Wiseman S, Davis L, et al. Evidencethat hospital hygiene is important in the control ofmethicillin-resistant Staphylococcus aureus. J HospInfect 2001;49:109e116.

55. Haley RW, Culver DH, White JW, et al. The efficacy of in-fection surveillance and control programs in preventingnosocomial infections in US hospitals. Am J Epidemiol1984;121:182.

56. Curran ET, Benneyan JC, Hood J. Controlling methicillin-resistant Staphylococcus aureus: a feedback approachusing annotated statistical process control charts. InfectControl Hosp Epidemiol 2002;23:13e18.

57. Coello R, Fernandez C, Arroyo P, et al. Success and failure inthe control of methicillin-resistant Staphylococcus aureus.In: Coello R, Casewell MW, et al, editors. Methicillin-resistant Staphylococcus aureus. Royal Tunbridge Wells,Kent: Wells Medical Limited; 1993. p. 95e98.

58. Hartstein AI, Denny MA, Morthland VH, LeMonte AM. Con-trol of methicillin-resistant Staphylococcus aureus in a hos-pital and an intensive care unit. Infect Control HospEpidemiol 1995;16:405e411.

59. Spicer JW. Three strategies in the control of staphylococciincluding methicillin-resistant Staphylococcus aureus.J Hosp Infect 1995;5(Suppl. A):45e49.

60. Vandenbroucke-Grauls CM. Control of methicillin-resistant Staphylococcus aureus in the Netherlands. In:Coello R, Casewell MW, editors. Methicillin-resistantStaphylococcus aureus. Royal Tunbridge Wells, Kent: WellsMedical Limited; 1993. p. 75e78.

61. Curran E. MRSA: monitoring quality. Br J Infect Control2000;2:20e23.

62. Gaynes RP. Surveillance of nosocomial infections. In:Bennett JV, Brachman PS, editors. Hospital infections.Boston: Lippincott-Raven; 1998. p. 65e84.

63. Giesecke J. Routine surveillance of infectious diseases. In:Giesecke J, editor. Modern infectious disease epidemiol-ogy. London: Arnold; 2002. p. 148e159.

64. Public Health Laboratory Service. The first year of theDepartment of Health’s mandatory MRSA bacteraemiasurveillance scheme in acute NHS Trusts in England:April 2001eMarch 2002. CDR Wkly 2002;12(25):1e17,http://www.hpa.org.uk/cdr/PDFfiles/2002/cdr2502.pdf.Last accessed 26 January 2006.

65. Department of Health Press Release. New MRSA figurespublished. Available on Department of Health (UK) websiteat: http://www.dh.gov.uk/PublicationsAndStatistics/PressReleases/PressReleases/Notices/fs/en?CONTENT_ID=4085856&chk=WsJjaB; http://www.dh.gov.uk/assetRoot/04/08/58/94/04085894.pdf2004. Last accessed 26 January2006.

66. Health Protection Agency Communicable Disease Surveil-lance Centre for the Department of Health 2005. MRSAsurveillance system e results. Available on Depart-ment of Health’s website at: http://www.dh.gov.uk/PublicationsAndStatistics/Publications/PublicationsStatistics/PublicationsStatisticsArticle/fs/en?CONTENT_ID¼4085951&chk¼HBt2QD; 2002. Last accessed 26 January 2006.

67. Department of Health. Bloodborne MRSA infection ratesto be halved by 2008. London: Department of Health.Available from: http://www.dh.gov.uk/PublicationsAndStatistics/PressReleases/PressReleasesNotices/fs/en?CONTENT_IDþ4093533&chkþMY%BkD/; 2004. Last accessed 26 Janu-ary 2006.

68. Enright MC, Robinson DA, Randle G, Feil EJ, Grundmann H,Spratt BG. The evolutionary history of methicillin-resistantStaphylococcus aureus (MRSA). Proc Natl Acad Sci USA2002;99:7687e7692.

69. Thomas JC, Bridge J, Waterman S, Vogt J, Kilman L,Hancock G. Transmission and control of methicillin-resistant Staphylococcus aureus in a skilled nursing facility.Infect Control Hosp Epidemiol 1989;10:106e110.

70. Shimada M, Kamakura T, Itasaka H, Matsumata T,Hashizume M, Sugimachi K. The significance of methicillin-resistant Staphylococcus aureus infection in generalsurgery: a multivariate analysis of risk factors andpreventive approaches. Surg Today 1993;23:880e884.

71. Onorato M, Borucki MJ, Baillargeon G, et al. Risk factorsfor colonization or infection due to methicillin-resistantStaphylococcus aureus in HIV-positive patients: a retro-spective caseecontrol study. Infect Control Hosp Epide-miol 1999;20:26e30.

72. Washio M, Mizoue T, Kajioka T, et al. Risk factors formethicillin-resistant Staphylococcus aureus (MRSA) infec-tion in a Japanese geriatric hospital. Public Health 1997;111:187e190.

73. Safdar N, Maki DG. The commonality of risk factors fornosocomial colonization and infection with antimicrobial-resistant Staphylococcus aureus, enterococcus, Gram-negative bacilli, Clostridium difficile, and Candida. AnnIntern Med 2002;136:834e844.

74. Dziekan G, Hahn A, Thune K, et al. Methicillin-resistantStaphylococcus aureus in a teaching hospital: investigationof nosocomial transmission using a matched caseecontrolstudy. J Hosp Infect 2000;46:263e270.

75. Hori S, Sunley R, Tami A, Grundmann H. The NottinghamStaphylococcus aureus population study: prevalence ofMRSA among the elderly in a university hospital. J HospInfect 2002;50:25e29.

76. Crowcroft NS, Ronveaux O, Monnet DL, Mertens R. Methi-cillin-resistant Staphylococcus aureus and antimicrobialuse in Belgian hospitals. Infect Control Hosp Epidemiol1999;20:31e36.

77. Campillo B, Dupeyron C, Richardet JP. Epidemiology ofhospital-acquired infections in cirrhotic patients: effectof carriage of methicillin-resistant Staphylococcus aureusand influence of previous antibiotic therapy and norfloxa-cin prophylaxis. Epidemiol Infect 2001;127:443e450.

78. Frank MO, Batteiger BE, Sorensen SJ, et al. Decrease inexpenditures and selected nosocomial infections followingimplementation of an antimicrobial-prescribing improve-ment program. Clin Perform Qual Health Care 1997;5:180e188.

79. Fukatsu K, Saito H, Matsuda T, Ikeda S, Furukawa S,Muto T. Influences of type and duration of antimicrobialprophylaxis on an outbreak of methicillin-resistant Staph-ylococcus aureus and on the incidence of wound infection.Arch Surg 1997;132:1320e1325.

80. Stelfox HT, Bates DW, Redelmeier DA. Safety of patientsisolated for infection control. JAMA 2003;290:1899e1905.

81. Paterson DL. Reduced susceptibility of Staphylococcus au-reus to vancomycin e a review of current knowledge.Commun Dis Intell 1999;23:69e73.

82. Fridkin SK, Lawton R, Edwards JR, Tenover FC,McGowan Jr JE, Gaynes RP. Monitoring antimicrobial useand resistance: comparison with a national benchmarkon reducing vancomycin use and vancomycin-resistant en-terococci. Emerg Infect Dis 2002;8:702e707.

83. Centers for Disease Control and Prevention. Recommenda-tions for preventing the spread of vancomycin resistance:recommendations of the Hospital Infection Control Advi-sory Committee (HICPAC). MMWR Morb Mortal Wkly Rep1995;44:1e33.

84. Sanford MD, Widmer AF, Bale MJ, Jones RN, Wenzel RP. Ef-ficient detection and long-term persistence of the carriage

http://www.hpa.org.uk/cdr/PDFfiles/2002/cdr2502.pdf

http://www.dh.gov.uk/PublicationsAndStatistics/PressReleases/PressReleasesNotices/fs/en&percnt;63CONTENT_ID+4093533&chk+MY&percnt;BkD/



http://http://www.dh.gov.uk/assetRoot/04/08/58/94/04085894.pdf2004

http://http://www.dh.gov.uk/assetRoot/04/08/58/94/04085894.pdf2004

http://www.dh.gov.uk/PublicationsAndStatistics/Publications/PublicationsPolicyAndGuidance/PublicationsPolicyAndGuidanceArticle/fs/en&percnt;63CONTENT_ID&percnt;614086665&percnt;38chk&percnt;61jXDWU6












of methicillin-resistant Staphylococcus aureus. Clin InfectDis 1994;19:1123e1128.

85. New Zealand Ministry of Health. Guidelines for the controlof methicillin-resistant Staphylococcus aureus in NewZealand. Wellington: New Zealand Ministry of Health,http://www.moh.govt.nz/cd/mrsa; 2002. Accessed 9January 2005.

86. Farr BM, Jarvis WR. Would active surveillance cultures helpcontrol healthcare-related methicillin-resistant Staphylo-coccus aureus infection? Infect Control Hosp Epidemiol2002;23:65e68.

87. Jernigan JA, Titus MG, Groschel DRM, Getchell-White SL,Farr BM. Effectiveness of contact isolation duringa hospital outbreak of methicillin-resistant Staphylococcusaureus. Am J Epidemiol 1996;142:496e504.

88. Salmenlinna S, Lyytikainen O, Kotilainen P, Scotford R,Siren E, Vuopio-Varkila J. Molecular epidemiology ofmethicillin-resistant Staphylococcus aureus in Finland.Eur J Clin Microbiol Infect Dis 2000;19:101e107.

89. Thompson RL, Cabezudo I, Wenzel RP. Epidemiology ofnosocomial infections caused by methicillin-resistantStaphylococcus aureus. Ann Intern Med 1982;97:309e317.

90. Haley RW, Cushion NB, Tenover FC, et al. Eradication ofendemic methicillin-resistant Staphylococcus aureus in-fections from a neonatal intensive care unit. J Infect Dis1995;171:614e624.

91. Jernigan JA, Clemence MA, Stott GA, et al. Control ofmethicillin resistant Staphylococcus aureus at a universityhospital: one decade later. Infect Control Hosp Epidemiol1995;16:686e696.

92. Jans B, Suetens C, Struelens M. Decreasing MRSA rates inBelgian hospitals: results from the national surveillancenetwork after introduction of national guidelines. InfectControl Hosp Epidemiol 2000;21:419.

93. Back NA, Linnemann Jr CC, Staneck JL, Kotagal UR. Con-trol of methicillin-resistant Staphylococcus aureus in a neo-natal intensive-care unit: use of intensive microbiologicsurveillance and mupirocin. Infect Control Hosp Epidemiol1996;17:227e231.

94. Calfee DP, Farr BM. Infection control and cost control inthe era of managed care. Infect Control Hosp Epidemiol2002;23:407e410.

95. Law MR, Gill ON. Hospital-acquired infection withmethicillin-resistant and methicillin-sensitive staphylo-cocci. Epidemiol Infect 1988;101:623e629.

96. Murray-Leisure KA, Geib S, Graceley D, et al. Control ofepidemic methicillin resistant Staphylococcus aureus. In-fect Control Hosp Epidemiol 1990;11:343e350.

97. Nicolle LE, Dyck B, Thompson G, et al. Regional dissemina-tion and control of epidemic methicillin-resistant Staphy-lococcus aureus. Infect Control Hosp Epidemiol 1999;20:202e205.

98. Kotilainen P, Routamaa M, Peltonen R, et al. Eradication ofmethicillin-resistant Staphylococcus aureus from a healthcenter ward and associated nursing home. Arch InternMed 2001;161:859e863.

99. Geffers C, Farr BM. Risk of transmission of nosocomialmethicillin-resistant Staphylococcus aureus (MRSA) frompatients colonized with MRSA. Infect Control HospEpidemiol 2005;26:114e115.

100. Vriens MR, Fluit AC, Troelstra A, Verhoef J, Van DerWerken C. Are MRSA more contagious than MSSA in a surgi-cal intensive care unit? Infect Control Hosp Epidemiol2002;23:491e494.

101. Forceville X, Faibis F, Lahilaire P, et al. Decrease ofinfection rate of methicillin resistant Staphylococcus

aureus acquired in a French intensive care unit, under re-inforcement of specific isolation. Med Mal Infect 2002;32:346e358.

102. Pan A, Carnevale G, Catenazzi, et al. Trends in methicillin-resistant Staphylococcus aureus (MRSA) bloodstream infec-tions: effect of the MRSA ‘search and isolate’ strategy ina hospital in Italy with hyperendemic MRSA. Infect ControlHosp Epidemiol 2005;26:127e133.

103. Bannister BA. Management of patients with epidemicmethicillin-resistant Staphylococcus aureus: experience atan infectious diseases unit. J Hosp Infect 1987;9:126e131.

104. Vriens MR, Blok HEM, Gigengack ACM, et al. Methicillin-resistant Staphylococcus aureus carriage among patientsafter hospital discharge. Infect Control Hosp Epidemiol2005;26:629e633.

105. Loeb M, Main C, Walker-Dilks C, Eady A. Antimicrobialdrugs for treating meticillin resistant Staphylococcus au-reus colonisation [Cochrane Review]. In: The CochraneLibrary, Issue 3. Chichester: John Wiley & Sons, Ltd;2003.

106. The AGREE Collaboration. Appraisal of guidelines forresearch and evaluation (AGREE) instrument. London:St. Georges Hospital Medical School. Available from:www.agreecollaboration.org; 2001.

107. Doebbeling BN, Breneman DL, Neu HC, et al. Elimina-tion of Staphylococcus aureus nasal carriage in healthcare workers: analysis of six clinical trials with mupir-ocin calcium ointment. The Mupirocin CollaborativeStudy Group. Clin Infect Dis 1993;17:466e474.

108. Karchmer TB, Jernigan JA, Simonton BM, Farr BM. Eradica-tion of meticillin-resistant S. aureus (MRSA) colonisationwith different regimens. Presented at the Ninth AnnualMeeting of the Society for Healthcare Epidemiology ofAmerica, 18e20 April 1999, San Francisco, CA, USA.Abstract 65.

109. Walker ES, Vasquez JE, Dula R, Bullock H, Sarubbi FA. Mu-pirocin-resistant meticillin-resistant Staphylococcus au-reus: does mupirocin remain effective? Infect ControlHosp Epidemiol 2003;24:342e346.

110. Hitomi S, Kubota M, Mori N, et al. Control of meticillin-resistant Staphylococcus aureus outbreak in a neonatalintensive care unit by selective use of nasal mupirocinointment. J Hosp Infect 2000;46:123e129.

111. Dupeyron C, Campillo B, Bordes M, Faubert E,Richardet JP, Mangeney N. A clinical trial of mupirocin inthe eradication of meticillin-resistant Staphylococcusaureus nasal carriage in a digestive diseases unit. J HospInfect 2002;52:281e287.

112. Wang JT, Lin SF, Chiu L, et al. Molecular epidemiology andcontrol of nosocomial meticillin-resistant Staphylococcusaureus infection in a teaching hospital. J Formos MedAssoc 2004;103:32e36.

113. Maraha B, van Halteren J, Verzijl JM, Wintermans RG,Buiting AG. Decolonisation of meticillin-resistant Staphy-lococcus aureus using oral vancomycin and topicalmupirocin. Clin Microbiol Infect 2002;8:671e675.

114. Wilcox MH, Hall J, Pike H, et al. Use of peri-operativemupirocin to prevent meticillin-resistant Staphylococcusaureus (MRSA) orthopaedic surgical site infection. J HospInfect 2003;54:196e201.

115. Mody L, Kauffman CA, McNeil SA, Galecki AT, Bradley SF.Mupirocin-based decolonisation of Staphylococcus aureuscarriers in residents of 2 long term care facilities: a rando-mised, double-blind, placebo-controlled trial. Clin InfectDis 2003;37:1467e1474.

116. Department of Health. Hospital infection control: guid-ance on the control of infection in hospitals. Prepared by

http://www.moh.govt.nz/cd/mrsa

http://www.agreecollaboration.org


the Hospital Infection Working Group of the Department ofHealth and PHLS. London: Department of Health; 1995.

117. Department of Health. Winning ways: working together toreduce healthcare associated infection in England. Reportfrom the Chief Medical Officer. London: Department ofHealth; 2003.

118. Department of Health. Infection control in the built envi-ronment. HBN 30. London: NHS Estates; 2002.

119. Department of Health. In patient accommodation: optionsfor choice. Supplement 1. Isolation facilities in acute set-tings. HBN 4. London: NHS Estates; 1997.

120. Report by the Controller and Auditor General. Improvingpatient care by reducing the risk of hospital acquiredinfection: a progress report. HC 876 Session 2003e2004;London: HMSO 2004.

121. Department of Health. Standards for better health.London: Department of Health, http://www.dh.gov.uk/PublicationsAndStatistics/Publications/PublicationsPolicyAndGuidance/PublicationsPolicyAndGuidanceArticle/fs/en?CONTENT_ID¼4086665&chk¼jXDWU6; 2004. Last accessed30 October 2005.

122. Cooper BS, Stone SO, Kibbler CC, et al. Systematic reviewof isolation policies in the hospital management ofmethicillin-resistant Staphylococcus aureus: a review ofthe literature with epidemiological and economic model-ling. Health Technol Assess 2003;7:1e194.

123. Department of Health. The patient’s charter: privacy anddignity and the provision of single sex accommodation.London: Department of Health, http://www.dh.gov.uk/assetRoot/04/01/14/24/04011424.pdf; 1997. Lastaccessed 30 October 2005.

124. Boyce JM, Pittet D. Guideline for hand hygiene inhealth-care settings: recommendations of the HealthcareInfection Control Practices Advisory Committee and theHICPAC/SHEA/APIC/IDSA Hand Hygiene Task Force. InfectControl Hosp Epidemiol 2002;23(Suppl.):S3eS41.

125. Ayliffe GAJ, Babb JR, Taylor LJ. Hospital-acquiredinfection. Principles and prevention. 3rd ed. Oxford: But-terworth-Heinemann; 1999.

126. French GL, Otter JA, Shannon KP, et al. Tackling contam-ination of the hospital environment by methicillin-resistant Staphylococcus aureus (MRSA): a comparisonbetween conventional terminal cleaning and hydrogenperoxide vapour decontamination. J Hosp Infect 2004;57:31e37.

127. Duckworth GL, Jordens JZ. Adherence and survivalproperties of an ‘epidemic’ methicillin-resistant strainof Staphylococcus aureus compared with those ofmethicillin-sensitive strains. J Med Microbiol 1990;32:195e200.

128. Wiseman S. Implementing effective infection control inthe hospital. Nurs Times 2004;100:36.

129. NHS Estates. Standards for cleanliness in the NHS: a frame-work in which to measure performance outcomes. Leeds:NHS Estates, http://www.patientexperience.nhsestates.gov.uk/clean_hospitals/ch_content/home/home.asp; 2003.Accessed 28 May 2005.

130. Health Care Standards Unit. Standards for better health.Signpost C4e (Waste). London: HMSO, http://www.hcsu.org.uk/index.phf?option¼com_content&task¼view&id¼47&Itemid¼0; 2003. Accessed 28 May 2005.

131. NHS Executive. Hospital laundry arrangements for usedand infected linen. HSG(95)18. Heywood: Health Publica-tions Unit; 1995.

132. Ayliffe GAJ, Fraise AP, Geddes AM, Mitchell K. Control ofhospital infection. A practical handbook. 4th ed. London:Arnold; 2000.

133. Hoffman PN, Williams J, Stacey A, et al. Report of a Work-ing Party of the Hospital Infection Society. Microbiologicalcommissioning and monitoring of operating theatre suites.J Hosp Infect 2002;52:1e28.

134. Haley RW, Bregman DA. The role of understaffing andovercrowding in recent outbreaks of staphylocccal infec-tion in a neonatal special-care unit. J Infect Dis 1982;145:875e885.

135. Arnow PM, Allyn PA, Nichols EM, Hill DL, Pezzio M,Bartlett RH. Control of methicillin resistant Staphylococ-cus aureus in a burn unit: role of nurse staffing. J Trauma1982;22:954e959.

136. Fridkin SK, Pear SM, Williamson TH, Galgiani JN, Jarvis WR.The role of understaffing in central venous catheter-associated bloodstream infections. Infect Control HospEpidemiol 1996;17:150e158.

137. Vicca AF. Nursing staff workload as a determinantof methicillin-resistant Staphylococcus aureus spread inan adult intensive care unit. J Hosp Infect 1999;43:109e113.

138. Harbarth S, Sudre P, Dharan S, Cadenas M, Pittet D. Out-break of Enterobacter cloacae related to understaffing,overcrowding, and poor hygiene practices. Infect ControlHosp Epidemiol 1999;20:598e603.

139. Pittet D. Improving compliance with hand hygiene in hospi-tals. Infect Control Hosp Epidemiol 2000;21:381e386.

140. HICPAC. Guideline for hand hygiene in health-care settings.MMWR Morb Mortal Wkly Rep 2002;51:RR-16, http://www.cdc.gov/mmwr/PDF/rr/rr5116.pdf. Last accessed23 January 2006.

141. Voss A, Widmer AF. No time for handwashing!?Handwashing versus alcoholic rub: can we afford 100%compliance? Infect Control Hosp Epidemiol 1997;18:205e208.

142. Pittet D, Mourouga P, Perneger TV. Compliance with handwashing in a teaching hospital. Ann Int Med 1999;130:126e130.

143. Farrington M, Trundle C, Redpath C, Anderson L. Effectson nursing workload of different MRSA control strategies.J Hosp Infect 2000;46:118e122.

144. Epic Guideline Project. The epic project: developing na-tional evidence-based guidelines for preventing healthcareassociation infections phase 1: guidelines for preventinghospital-acquired infections. London: HMSO, http://www.dh.gov.uk/assetRoot/04/07/73/68/04077368.pdf; 2001.Last accessed 27 January 2006.

145. Stone PW, Clark SP, Cimiotti J, Corres-de-Araujo R. Nurses’working conditions: implications for infectious disease.Emerg Infect Dis 2004;10:1984e1989.

146. Kovner C, Jones C, Zhan C, Gergen PJ, Basu J. Nursestaffing and postsurgical adverse events: an analysis ofadministrative data from a sample of U.S. hospitals,1990e1996. Health Serv Res 2002;37:611e629.

147. Alonso-Echanove J, Edwards JR, Richards MJ, et al. Effectof nurse staffing and antimicrobial-impregnated centralvenous catheters on the risk for bloodstream infectionsin intensive care units. Infect Control Hosp Epidemiol2003;2:916e925.

148. Anderson BM, Lindemann R, Bergh R, et al. Spread ofmethicillin-resistant Staphylococcus aureus in a neonatalintensive care unit associated with understaffing, over-crowding and mixing of patients. J Hosp Infect 2002;50:18e24.

149. Stenenga J, Bell E, Matlow A. The role of nurse understaff-ing in nosocomial viral gastrointestinal infections on a gen-eral pediatrics ward. Infect Control Hosp Epidemiol 2002;23:133e136.

http://www.dh.gov.uk/PublicationsAndStatistics/Publications/PublicationsPolicyAndGuidance/PublicationsPolicyAndGuidanceArticle/fs/en?CONTENT_ID-4086665&chk-XDWU6






http://www.dh.gov.uk/assetRoot/04/07/73/68/04077368.pdf


http://www.patientexperience.nhsestates.gov.uk/clean_hospitals/ch_content/home/home.asp

http://www.patientexperience.nhsestates.gov.uk/clean_hospitals/ch_content/home/home.asp

http://www.hcsu.org.uk/index.phf?option=com_content&38task=view&id=47&Itemid=0







http://www.cdc.gov/mmwr/PDF/rr/rr5116.pdf

http://www.cdc.gov/mmwr/PDF/rr/rr5116.pdf




150. Amravadi RK, Jacobson BC, Solomon DH, Fischer MA. ICUnurse-to-patient ratio is associated with complicationsand resource use after esophagectomy. Inten Care Med2000;26:1857e1862.

151. Robert J, Fridkin SK, Blumberg HM, et al. The influence ofthe composition of the nursing staff on primary blood-stream infection rates in a surgical intensive care unit.Infect Control Hosp Epidemiol 2000;21:12e17.

152. Kovner C, Gergen PJ. Nurse staffing levels and adverseevents following surgery in U.S. hospitals. J Nurse Schol-arsh 1998;30:315e321.

153. Archibald LK, Manning ML, Bell LM, Banerjee S, Jarvis WR.Patient density, nurse-to-patient ratio and nosocomialinfection risk in a pediatric intensive care unit. PediatrInfect Dis J 1997;16:1045e1048.

154. Working Party of the British Society for Antimicrobial Che-motherapy. Supplementary report. J Antimicrob Chemo-ther 1996;38:1103e1105.

155. National Committee for Clinical Laboratory Standards.Methods for dilution. Antimicrobial susceptibility testsfor bacteria that grow aerobically. Approved standard,M7-A4. 4th edn. Vilanova: NCCLS; 1997.

156. Kaplan AH, Gilligan PH, Facklam RR. Recovery of resistantenterococci during vancomycin prophylaxis. J Clin Micro-biol 1988;26:1216e1218.

157. Noble WC, Virani Z, Cree RG. Co-transfer of vancomycinand other resistance genes from Enterococcus faecalisNCTC 12201 to Staphylococcus aureus. FEMS Microbiol Lett1992;72:195e198.

158. Hiramatsu K, Hanaki H, Ino T, Yabuta K, Oguri T,Tenover FC. Methicillin-resistant Staphylococcus aureusclinical strain with reduced vancomycin susceptibility.J Antimicrob Chemother 1997;40:135e136.

159. Fridkin SK, Hageman J, McDougal LK, et al. Vancomycin-intermediate Staphylococcus aureus Epidemiology StudyGroup. Epidemiological and microbiological characteriza-tion of infections caused by Staphylococcus aureus withreduced susceptibility to vancomycin, United States,1997e2001. Clin Infect Dis 2003;36:429e439.

160. Smith TL, Pearson ML, Wilcox KR, et al. Emergence ofvancomycin resistance in Staphylococcus aureus. Glyco-peptide-intermediate Staphylococcus aureus WorkingGroup. N Engl J Med 1999;340:493e501.

161. Rotun SS, McMath V, Schoonmaker DJ, et al. Staphylococ-cus aureus with reduced susceptibility to vancomycin iso-lated from a patient with fatal bacteremia. Emerg InfectDis 1999;5:147e149.

162. Khurshid MA, Chou T, Carey R, et al. Staphylococcus au-reus with reduced susceptibility to vancomycin e Illinois,1999. MMWR Morb Mortal Wkly Rep 2000;48:1165e1167.

163. Naimi TS, Anderson D, O’Boyle C, et al. Vancomycin-intermediate Staphylococcus aureus with phenotypicsusceptibility to methicillin in a patient with recurrentbacteremia. Clin Infect Dis 2003;36:1609e1612.

164. Hageman JC, Pegues DA, Jepson C, et al. Vancomycin-intermediate Staphylococcus aureus in a home health-carepatient. Emerg Infect Dis 2001;7:1023e1025.

165. Walsh TR, Howe RA. The prevalence and mechanisms ofvancomycin resistance in Staphylococcus aureus. Ann RevMicrobiol 2002;56:657e675.

166. Weigel LM, Clewell DB, Gill SR, et al. Genetic analysis ofa high-level vancomycin-resistant isolate of Staphylococ-cus aureus. Science 2003;302:1569e1571.

167. Tenover FC, Weigel LM, Appelbaum PC, et al. Vancomycin-resistant Staphylococcus aureus isolate from a patient inPennsylvania. Antimicrob Agents Chemother 2004;48:275e280.

168. Ike Y, Arakawa Y, Ma X, et al. Nationwide surveyshows that methicillin-resistant Staphylococcus aureusstrains heterogeneously and intermediately resistant tovancomycin are not disseminated throughout Japanesehospitals. J Clin Microbiol 2001;39:4445e4451.

169. Ploy MC, Grelaud C, Martin C, de Lumley L, Denis F.First clinical isolate of vancomycin-intermediate Staphylo-coccus aureus in a French hospital. Lancet 1998;351:9110e9112.

170. Bierbaum G, Fuchs K, Lenz W, Szekat C, Sahl HG. Presenceof Staphylococcus aureus with reduced susceptibility tovancomycin in Germany. Eur J Clin Microbiol Infect Dis1999;18:691e696.

171. Howe RA, Bowker KE, Walsh TR, Feest TG, MacGowan AP.Vancomycin-resistant Staphylococcus aureus. Lancet 1998;351:602.

172. Ariza J, Pujol M, Cabo J, et al. Vancomycin in surgical in-fections due to methicillin-resistant Staphylococcus au-reus with heterogeneous resistance to vancomycin.Lancet 1999;353:1587e1588.

173. Wong SS, Ho PL, Woo PC, Yuen KY. Bacteremia caused bystaphylococci with inducible vancomycin heteroresist-ance. Clin Infect Dis 1999;29:760e767.

174. Kantzanou M, Tassios PT, Tseleni-Kotsovili A, Legakis NJ,Vatopoulos AC. Reduced susceptibility to vancomycinof nosocomial isolates of methicillin-resistant Staphylo-coccus aureus. J Antimicrob Chemother 1999;43:729e731.

175. Marchese A, Balistreri G, Tonoli E, Debbia EA, Schito GC.Heterogeneous vancomycin resistance in methicillin-resistant Staphylococcus aureus strains isolated in a largeItalian hospital. J Clin Microbiol 2000;38:866e869.

176. Ward PB, Johnson PD, Grabsch EA, Mayall BC, Grayson ML.Treatment failure due to methicillin-resistant Staphylo-coccus aureus (MRSA) with reduced susceptibility to vanco-mycin. Med J Aust 2001;175:480e483.

177. Oliveira GA, Dell’Aquila AM, Masiero RL, et al. Isolation inBrazil of nosocomial Staphylococcus aureus with reducedsusceptibility to vancomycin. Infect Control Hosp Epide-miol 2001;22:443e448.

178. Kim MN, Hwang SH, Pyo YJ, Mun HM, Pai CH. Clonal spreadof Staphylococcus aureus heterogeneously resistant tovancomycin in a university hospital in Korea. J Clin Micro-biol 2002;40:1376e1380.

179. Krzyszton-Russjan J, Gniadkowski M, Polowniak-Pracka H, Hagmajer E, Hryniewicz W. The first Staphylo-coccus aureus isolates with reduced susceptibility tovancomycin in Poland. J Antimicrobial Chemother2002;50:1065e1069.

180. Tenover FC. Implications of vancomycin-resistant Staphy-lococcus aureus. J Hosp Infect 1999;43(Suppl):S3eS7.

181. Tenover FC, Lancaster MV, Hill BC, et al. Characterizationof staphylococci with reduced susceptibilities to vancomy-cin and other glycopeptides. J Clin Microbiol 1998;36:1020e1027.

182. Edmond MB, Wenzel RP, Pasculle AW. Vancomycin-resistant Staphylococcus aureus: perspectives on mea-sures needed for control. Ann Intern Med 1996;124:329e334.

183. Hospital Infection Control Practises Advisory Committee,CDC, Atlanta, USA Interim guidelines for prevention andcontrol of staphylococcal infection associated with re-duced susceptibility to vancomycin. MMWR Morb MortalWkly Rep 1997;46:626e628.

184. D’Agata EM. Antimicrobial-resistant, Gram-positive bacte-ria among patients undergoing chronic hemodialysis. ClinInfect Dis 2002;35:1212e1218.


185. Hiramatsu K, Aritaka N, Hanaki H, et al. Dissemination inJapanese hospitals of strains of Staphylococcus aureusheterogeneously resistant to vancomycin. Lancet 1997;350:1670e1673.

186. Ploy MC, Francois B, Mounier M, Vignon P, Denis F. Nasalcarriage of vancomycin-intermediate Staphylococcus au-reus among intensive care unit staff. Clin Infect Dis2001;33:1951.

187. World Health Organization. Report on infectious diseases.Geneva: World Health Organization. Available from:www.who.int/infectious-disease-report/2000/index.html;2000. Last accessed 25 January 2006.

188. Wenzel RP, Edmond MB. Vancomycin-resistant Staphylo-coccus aureus: infection control considerations. ClinInfect Dis 1998;27:245e249.

189. Kralovic SM, Danko LH, Roselle GA. Laboratory reporting ofStaphylococcus aureus with reduced susceptibility to van-comycin in United States Department of Veterans AffairsFacilities. Emerg Infect Dis 2002;8:402e407.

190. Tenover FC, Biddle JW, Lancaster MV. Increasing resis-tance to vancomycin and other glycopeptides in Staphylo-coccus aureus. Emerg Infect Dis 2001;7:327e332.

191. Hussain FM, Boyle-Vavra S, Shete PB, Daum RS. Evidencefor a continuum of decreased vancomycin susceptibilityin unselected Staphylococcus aureus clinical isolates.J Infect Dis 2002;186:661e667.

192. Midolo PD, Korman TM, Kotsanas D, Russo P, Kerr TG.Laboratory detection and investigation of reducedsusceptibility to vancomycin in oxacillin-resistant Staph-ylococcus aureus. Eur J Clin Microbiol Infect Dis 2003;22:199e201.

193. Wootton M, Howe RA, Hillman R, Walsh TR, Bennett PM,MacGowan AP. A modified population analysis profile(PAP) method to detect hetero-resistance to vancomycinin Staphylococcus aureus in a UK hospital. J AntimicrobChemother 2001;47:399e403. Erratum appears in J Anti-microb Chemother 2001;48:161.

194. Howe RA, Wootton M, Walsh TR, Bennett PM,MacGowan AP. Heterogeneous resistance to vancomycinin Staphylococcus aureus. J Antimicrob Chemother 2000;45:130e131.

195. National Institute for Clinical Excellence. Infection con-trol: prevention of healthcare associated infection in pri-mary and community care. London: NICE; 2003.

Glossary of terms

Bacteraemia: presence of bacteria in the blood.Bloodstream infection: the presence of microbes in the blood

with significant clinical consequences, e.g. shock.Carrier of MRSA: a person who harbours MRSA with no overt ex-

pression of clinical disease, but who is a potential source ofinfection. Recognized carrier sites for MRSA include thenose, throat and certain skin sites including the perineum,groin, axilla and buttock. The carriage of MRSA can be tran-sient, intermittent or chronic.

Clinical trial: A statistical method of determining the efficacyof a drug, or other intervention, in comparison with a pla-cebo, an existing drug or a conventional intervention (see‘Double-blind randomized controlled trial’).

Cohort nursing: a group of patients with a disease or infectionwho are separated from patients who do not harbour the dis-ease or infection and who are nursed in a geographically dis-tinct area or with physical separation in the same room.Isolation in separate rooms is preferable to cohort nursing.

Ideally, the same nursing staff should provide daily carefor the same cohort for the duration of the isolation period.

Colonization with MRSA: the presence and multiplication ofMRSA at a body site without tissue invasion, damage or clin-ical disease.

Double-blind randomized controlled trial: a method of experi-mentation in which neither the patient nor the experimenter(doctor) knows beforehand which treatment has been as-signed to the patient, i.e. drug, intervention or placebo. Ineach arm (control/placebo or treatment), the subjectsshould be assigned at random, taking care that they arematched as far as is possible for sex, age range and conditionto be treated.

Endemic disease: the continued presence of a disease-causingorganism with or without infection in a given hospital, a givengroup of patients in a hospital, or a geographical area de-spite standard control procedures (see ‘Epidemic’).

Epidemic: the outbreak of or acquisition of a disease-causing or-ganism spreading widely among people at the same time ina hospital or community (e.g. in a residential facility) or ina geographical area with a frequency that is clearly in excessof normal expectancy. Certain phage types of MRSA areknown to spread easily among and within hospitals and aredesignated, e.g. EMRSA 15, EMRSA 16, etc. (see ‘Endemicdisease/phage type’).

Epidemiology: the study of the distribution and determinants ofan infection or disease and/or colonization in specific popu-lations with particular reference to the reservoirs, sources,routes of transmission and portals of entry.

Fomites, n. pl, sing., fome, Lat. tinder, kindling: inanimate ob-jects that when contaminated with a viable pathogen cantransfer the pathogen to a host. Examples include door han-dles, telephones, computer keyboards or any shared item.Note: the word is almost invariably used in the plural andis equally invariably mis-pronounced as ‘foamights’. The sin-gular is pronounced ‘foamay’ and the plural ‘foamitays’.

Hospital-acquired infection or healthcare-associated infection:also known as nosocomial infection. This includes infectionacquired in a variety of institutions and not just acutehospitals.

Heteroresistance: often referring to glycopeptide (usually van-comycin) drugs. These are isolates of S. aureus with hetero-geneous glycopeptide-intermediate resistance of S. aureus(hGISA or heteroGISA). These appear to be susceptible toglycopeptides (typically with an MIC of <4 mg/L), but con-tain subpopulations of cells at frequencies of approximately10�6 cells that exhibit decreased susceptibility (i.e. vanco-mycin MIC of >4e16 mg/L).

Health Protection Agency: formerly the Public Health Labora-tory Service. The Head Office is in Colindale, London, UK.

Infection control committee: this normally consists of a chair-person, the infection control doctor, infection controlnurse(s), the consultant in communicable diseases control,occupational health physician or nurse, clinician representa-tives and the chief executive or their representative. Othermembers may be co-opted as appropriate, e.g. a pharmacistor engineer. The infection control committee is responsibleto the hospital’s chief executive and provides specialist ad-vice, formulates and monitors the implementation of poli-cies, and determines and monitors the progress of theannual infection control programme.

Infection control team: these are designated staff responsiblefor day-to-day hospital infection control. The team usuallyconsists of an infection control doctor, normally a consultantmedical microbiologist, a consultant medical microbiologistif the infection control doctor is from another speciality,and an infection control nurse or nurses. The infection

http://www.who.int/infectious-disease-report/2000/index.html


control team has direct access to the hospital’s chief execu-tive or their representative and is responsible to the infec-tion control committee.

Infection with MRSA: the entry and multiplication of MRSA inthe tissues of the host where tissue damage is caused. Thisresults in clinical disease that may be local (e.g. surgicalsite wound) or systemic (e.g. bloodstream infection) (see‘Bloodstream infection’).

Incidence: the frequency with which new cases of a given dis-ease presents in a defined population within a specific periodof time (see ‘Prevalence’).

Isolation of patients: separation of patients with a disease or in-fection in an individual room in order to prevent or limit thedirect or indirect transmission of the disease or infection toother people who are susceptible.

Isolation room or unit: a single room or unit often with its ownhandwashing and toilet facilities and also preferably with ananteroom for healthcare workers to wash hands and don pro-tective clothing, e.g. plastic aprons. The air supply may beunder negative pressure, or in balance, with respect to thearea outside the room.

Minimum inhibitory concentration: the lowest concentration ofa drug or agent that will inhibit the growth of the organism.MICs apply to testing agents against bacteria, protozoa and fungi.

Morbidity: the state of being ill and suffering, the sickness ratein a community or population.

Mortality: the death of individuals in a population.Nosocomial infection: (see ‘Hospital acquired infection’).Outbreak:oftenused interchangeablywith ‘epidemic’.Maybeused

to refer to a local cluster of cases or a small, limited outbreak.Phage type: MRSA and other staphylococci can be divided into

distinct strains or types by testing their susceptibility tobacterial viruses or ‘phages’ in the laboratory. Certain phagetypes of MRSA show characteristic patterns of spread andoccur more commonly in distinct geographical areas (see‘Epidemic’).

Prevalence: (of a disease) the total number of people with thedisease at a defined point in time.

Prospective study: an assessment of data that are collected asthey are being generated (see ‘Retrospective study’).

Randomized controlled trial: subjects with a disease are allo-cated randomly to one of two or more treatments, one ofwhich may be a control (or placebo) treatment, and the out-comes of treatment are compared.

Retrospective study: the assessment of information or data col-lected in the past and not concurrently.

Virulence: the capacity of an organism to cause disease (also re-ferred to as ‘pathogenicity’).

mrsa - the hospital infection society

Documents