Article
Reference
Non-compliance with IDSA guidelines for patients presenting with methicillin-susceptible Staphylococcus aureus prosthetic joint
infection is a risk factor for treatment failure
BOUAZIZ, A, et al .
Abstract
The long-term impact of treatment strategies proposed by the IDSA guidelines for patients presenting with methicillin-susceptible S. aureus (MSSA) prosthetic joint infection (PJI) is not well-known.
BOUAZIZ, A, et al . Non-compliance with IDSA guidelines for patients presenting with methicillin-susceptible Staphylococcus aureus prosthetic joint infection is a risk factor for treatment failure. Médecine et maladies infectieuses , 2017
DOI : 10.1016/j.medmal.2017.09.016 PMID : 29122410
Available at:
http://archive-ouverte.unige.ch/unige:100717
Disclaimer: layout of this document may differ from the published version.
ScienceDirect
www.sciencedirect.com
Médecineetmaladiesinfectieusesxxx(2017)xxx–xxx
Short communication
Non-compliance with IDSA guidelines for patients presenting with methicillin-susceptible Staphylococcus aureus prosthetic joint
infection is a risk factor for treatment failure 夽
Le non-respect des recommandations de l’IDSA pour les infections de prothèses à Staphylococcus aureus sensible à la méthicilline est un risque d’échec A. Bouaziz
a, I. Uc¸kay
b, S. Lustig
c,d,e, A. Boibieux
a, D. Lew
b, P. Hoffmeyer
b,
P. Neyret
c,d, C. Chidiac
a,d,e,f, T. Ferry
a,d,e,f,∗aServicedemaladiesinfectieusesettropicales,hôpitaldelaCroix-Rousse,hospicescivilsdeLyon,93,GranderuedelaCroix-Rousse,69004Lyon,France
bHôpitauxuniversitairesdeGenève,4,rueGabrielle-Perret-Gentil,1211Geneva14,Switzerland
cServicedechirurgieorthopédique,centreAlbert-Trillat,hôpitaldelaCroix-Rousse,hospicescivilsdeLyon,69004Lyon,France
dUniversitéClaude-BernardLyon1,69100Lyon,France
eCentrederéférencedesIOAcomplexesdeLyon,Lyon,France
fInsermU1111,CNRSUMR5308,ENSdeLyon,UCBL1,centreinternationalderechercheeninfectiologie(CIRI),69007Lyon,France Received9August2016;receivedinrevisedform9November2016;accepted26September2017
Abstract
Objective.–Thelong-termimpactoftreatmentstrategiesproposedbytheIDSAguidelinesforpatientspresentingwithmethicillin-susceptible S.aureus(MSSA)prostheticjointinfection(PJI)isnotwell-known.
Patientsandmethods.–Retrospective(2000–2010)cohortstudyincludingpatientspresentingwithMSSAhiporkneePJI.AunivariateCox analysiswasperformedtodetermineifthenon-compliancewithIDSAsurgicalguidelineswasariskfactorfortreatmentfailure.
Results.–Eighty-ninepatientswithameanfollow-upof2.8yearswereincluded.Non-compliancewithIDSAsurgicalguidelineswasassociated withtreatmentfailure(hazardratio2.157;95%CI[1.022–4.7]).TheAmericanSocietyofAnesthesiologistsscore,inadequateantimicrobialtherapy, andarifampicin-basedregimendidnotsignificantlyinfluencepatientoutcome.
Conclusion.–BasedontheIDSAguidelines,ifapatientpresentingwithMSSAPJIisnoteligibleforimplantretention,completeimplant removalisneededtolimittreatmentfailure.
©2017ElsevierMassonSAS.Allrightsreserved.
Keywords:Prostheticjointinfections;MSSA;Guidelines Résumé
Objectif.–L’impactdesstratégiesproposéesparlesrecommandationsdel’IDSApourlespatientsprésentantuneinfectiondeprothèsearticulaire (IPA)àS.aureussensibleàlaméthicilline(MSSA)n’estpasconnu.
Patientsetméthodes.–Étudedecohorterétrospective(2000–2010)incluantdespatientsavecuneIPAàMSSA.UneanalyseunivariéedeCox aétéutiliséepourdéterminersilenon-respectdelastratégiechirurgicaleproposéeparl’IDSAétaitassociéàunéchec.
夽 Thisarticlehasnotbeensubmittedorpublishedelsewhere,butpreliminaryresultswerepresentedasaposterattheInterscienceConferenceonAntimicrobial AgentsandChemotherapy,SanFrancisco,September2012.
∗Correspondingauthor.Servicedemaladiesinfectieusesettropicales,hôpitaldelaCroix-Rousse,hospicescivilsdeLyon,93,GranderuedelaCroix-Rousse, 69004Lyon,France.
E-mailaddress:[email protected](T.Ferry).
http://dx.doi.org/10.1016/j.medmal.2017.09.016
0399-077X/©2017ElsevierMassonSAS.Allrightsreserved.
2 A.Bouazizetal./Médecineetmaladiesinfectieusesxxx(2017)xxx–xxx
Résultats.–Chez89patientsavecunsuivimédiande2,8ans,lenon-respectdelastratégiechirurgicaleproposéeparl’IDSAétaitassociéàun échec(HR2,157;IC95%[1,022–4,7]),alorsquelescoreASA,uneantibiothérapieinadéquateetuneantibiothérapiecontenantdelarifampicine n’influenc¸aientpaslepronostic.
Conclusion.–PourlespatientsprésentantuneIPAàMSSAneremplissantpaslescritèrespouruntraitementconservateur,l’ablationtotaledes implantsestrequise.
©2017ElsevierMassonSAS.Tousdroitsr´eserv´es.
Motsclés:Infectionsdeprothèsesarticulaires;SASM;Recommandations
1. Introduction
Prostheticjointinfections(PJI)areuncommon,buttheyrep- resent the most serious complication following arthroplasty.
Their management is based on both surgery and long-term antimicrobialtherapy.Accordingtotreatmentalgorithmsrein- forcedbyZimmerlietal.in2004,surgicaloptionsdescribedin theIDSAguidelinesincludedebridementandimplantretention (DAIR),one-or two-stage exchange,or permanentexplanta- tionorarthrodesis,dependingontheclinicalpresentation[1,2].
BasedontheZimmerlialgorithm,conservativetreatmentwith implantretentionshouldbereservedforpatientswithduration of symptoms <3 weeks, astable implant, and withintact or slightly damaged soft tissue [1]. Two previous retrospective studiesincluding60–70 patients presentingwithPJIduetoa wide range of pathogens, reported that the risk of treatment failurewassignificantlyhigherforthosetreatedwithasurgi- cal strategy otherthan that recommended byZimmerli et al.
[3,4].TheIDSAguidelines mentionaslightlydifferentcrite- rionasconservativetreatmentwithimplantretentionshouldbe reservedforpatientswithadurationofsymptoms<3weeksor ajointage<30days,andastableimplantwithoutsinustract.
Staphylococcusaureusisoneofthemostfrequentmicroor- ganisms involved in PJI and is particularly associated with treatment failure, presumably due to biofilm formation, vir- ulence of the pathogen, and the ability to transform into small-colony variants (SCV) [5–7]. Based on animal exper- iments and in vitro data, rifampicin-based regimens are recommendedfor S.aureusPJI[1,2],butrifampicincouldbe associated with gastrointestinal disorders, drug-drug interac- tions,andthedosefor patientspresentingwithPJIisdebated (600mg/dirrespectiveoftheweightinSpain,600to900mg/d intheIDSAguidelines, 20mg/kg/dintheFrench guidelines) [1,2,8,9].Methicillin-resistantS.aureus(MRSA)isconsidered moredifficult totreatas it isusually resistantto manyclini- callyimportantnon-beta-lactamdrugs,suchasrifampicinand fluoroquinolones.Moreover,severalstudiesdemonstratedthat implantretentionwasarisk factorfortreatmentfailureinPJI duetomethicillin-resistantstaphylococci[10,11].Todate,little dataisavailableonmethicillin-susceptibleS. aureus(MSSA) intermsofimpactofmedicalandsurgicalstrategiesonpatient outcome.
Weperformedaretrospectivecohort studytodetermineif thenon-compliancewithIDSAsurgicalguidelines inpatients
presentingwithMSSAprostheticjointinfectionhadalong-term impactonoutcome.
2. Methods
We performed a retrospective cohort study of all patients admitted betweenJanuary 1,2000andDecember31,2010to theGenevaUniversityHospitals,Geneva,Switzerland,andthe Hospices Civils de Lyon, Lyon, France, with a diagnosis of hip or knee PJI due to MSSA. The databases of the hospi- tals’administrativecodingsystems,bacteriologylaboratories, andtheorthopedicandinfectiousdiseasesunitswereusedfor patientselection.Thestudywasperformedaccordingtolocal ethicscommitteeguidelinesatbothinstitutions.Datawascol- lectedfrommedicalreportsusingastandardizeddatacollection tool. Tolimit missingoutcome data,patients, their family,or their healthcare providers were contacted by telephone and askedabouttheinfectionoutcome.Theburdenofthepatient’s comorbiditywasassessedusingtheAmericanSocietyofAnes- thesiologists(ASA)score.PJIswerecategorizedintoearly(≤3 months afterimplantation), delayed (>3to<12months after implantation), or late(≥12monthsafter implantation)infec- tion.BasedontheIDSAguidelines,thesurgicaltreatmentwas considered inadequate if patients did not qualify for implant retentionandifacompleteimplantremovalwasnotperformed (i.e., ifimplant retention,totalor partialone-stage exchange, or incomplete implant or cement removal was performed instead of a complete implant removal). Medical treatment was considered inadequate if patients did not receive intra- venoustherapywithantistaphylococcalactivityduringthefirst 15daysoftherapy(flucloxacillin,oxacillin,cloxacillin,orvan- comycin)followedbyacombinationofantimicrobialsbasedon rifampicin-fluoroquinolonecombinationforadurationofatleast threemonths.AstheFrenchguidelinessuggest15daysofintra- venoustherapyfollowedbyacombinationofantistaphylococcal oraldrugsthatcannotcontainrifampicin(e.g.,fluoroquinolone- clindamycin,fluoroquinolone-fusidicacid,clindamycin-fusidic acid),wedidnotconsiderthatthismedicalstrategywasinad- equate [8]. Treatment failure was defined by the need for subsequentsurgerytocontroltheinfection(ortotreatsuperin- fection)oramputationordeathattributedtothePJI.Univariate Coxanalysiswasusedtoassessinteractionsbetweentreatment strategies and the study center. Univariate Cox analysis and
Kaplan Meier curves were performed to determine risk fac- torsfor treatment failure.Statisticalanalyseswereperformed using SPPS Statistics Base 17.0 (Softonic International, San Francisco,CA,USA).
3. Results
Eighty-ninepatients (mean age: 66±15 years;55 males) wereincluded(54hipinfections).Infectionwasearly,delayed, orlatein33(37%),11(12%),and45(51%)patients,respec- tively(Table1).MostpatientspresentingwithearlyPJIhada durationofclinical signsofinfection beforethe surgery<30 days.MedianASAscorewas3(IQR:2–4).Durationofclinical symptomswas>3weeksin25patients(28%).Thesofttissue wasconsideredsignificantlydamaged(mainlylargeabscesses) in25patients(28%)andtheimplantwasunstableforthesurgeon in15patients (17%).Theimplantwasretainedin62patients (70%),whereastwo-stageexchangewouldhavebeenbetterfor 35patients accordingtothe IDSAguidelines. Amongthe33 patientspresentingwithacutePJI,retentionoftheimplantwas consideredinadequatefor16ofthem(49%)asthesurgerywas performed>30daysafterprosthesisimplantationor>3weeks afterthefirstclinicalsymptom(13patients),and/oraspoorlocal conditionswereobserved(10 patients),and/oras the implant wasunstable(10patients),and/orasafistulawasobserved(5 patients).For16patientspresentingwithlatechronicinfections, surgerywith implant retention was considered inadequate in sevenofthem.Antimicrobialtherapywasconsideredinadequate in25 patients(28%),including fivewhodid notreceivesup- pressiveantimicrobialtherapy.Duringalong-termfollow-upof 2.8±2.2years,treatmentfailureoccurredin37patients(42%), including 27 (30%) witha recurrence of MSSA persistence.
Ofnote,38ofthe56delayedorlatePJIswerenotconsidered chronicpostoperativeinfections,butofhematogenousoriginas patients developedacuteclinical symptoms. In thissubpopu- lationof patients, the rate of relapse washighas 14 patients (37%) experienced a relapse with S. aureus still in culture.
Fig.1.KaplanMeiercurvewiththecumulativeprobabilityoftreatmentfailure bysurgicaltreatmentforpatientspresentingwithmethicillin-sensitiveStaphy- lococcusaureusprostheticjointinfectionduringlong-termfollow-up.
CourbedeKaplan-Meiermontrantlaprobabilitécumuléedesurvenued’un échecselonlapriseenchargechirurgicaledespatientsprésentantuneinfection deprothèseàStaphylococcusaureussensibleàlaméthicilline
Partofthemwasprobablypostoperativeandnothematogenous PJIs.Nointeractionbetweenthetwocenterswasdetected.No difference was observedbetweenthe two centers or between patients with andwithout treatment failure regarding patient characteristics,typeofinfection,andmedicaltreatmentmodal- ities.Kaplan MeiercurvesandCox analysisrevealed that an inadequatesurgicalstrategy(i.e.,performingimplantretention or incomplete implant removal, whereas two-stage exchange wasrecommended)wastheonlyriskfactorfortreatmentfail- ure (Fig.1;hazard ratio (HR)2.157;95% CI [1.022–4.659];
P=0.050).Non-compliancewiththeIDSAsurgicalguidelines wasassociatedwithacumulativeprobabilityoftreatmentfailure of80%duringlong-termfollow-up.TheASAscore(HR,0.520;
95% CI [0.243–1.110]; P=0.091), inadequate antimicrobial
Table1
Outcomeofthe89patientsincludedinthestudyandcompliancewithguidelines,dependingonlocalization(hiporknee)andclinicalpresentation.
Pronosticdes89patientsinclusdansl’étudeetadéquationauxguidelines,selonlalocalisation(hancheougenou)etlaprésentationclinique.
n Relapse(n[%]) Implantretention (n[%])
Inadequatesurgical therapy(n[%])
Rifampicin-based regimen(n[%])
Inadequateantimicrobial therapy(n[%])
Acutea 33 10(30) 27(81) 16(49) 20(62) 11(33)
Delayed 11 4(36) 0(0) 5(46) 7(63) 3(27)
Lateb 45 23(51) 30(67) 17(38) 26(58) 11(24)
Hip 54 20(37) 20(37) 27(50) 32(59) 17(32)
Acute 24 6(25) 21(86) 13(54) 15(63) 9(38)
Delayed 5 2(40) 3(60) 3(60) 4(80) 2(40)
Late 13 12(48) 14(56) 13(52) 13(52) 6(24)
Knee 35 17(49) 24(69) 8(23) 21(60) 8(23)
Acute 9 4(44) 6(67) 5(56) 5(56) 2(22)
Delayed 6 2(33) 2(33) 2(33) 3(50) 1(17)
Late 20 11(55) 16(80) 4(20) 1(65) 5(25)
a Inthesepatients,implantretentionwasconsideredinadequatein16ofthem(49%)asthesurgerywasperformed>30daysaftertheprosthesisimplantationor
>3weeksafterthefirstclinicalsymptom(13patients)and/oraspoorlocalconditionswereobserved(10patients)and/orastheimplantwasunstable.
b Lateinfectionsincludedinfectionsofsuspectedhematogenousoriginsandlatechronicinfections.Inthe16patientspresentingwithcertainlatechronicPJI, surgerywithimplantretentionwasconsideredinadequatein7ofthem.
4 A.Bouazizetal./Médecineetmaladiesinfectieusesxxx(2017)xxx–xxx
therapy(HR,1.016;95%CI[0.441–2.336];P=0.971),anda rifampicin-basedregimen(HR,0.899;95%CI[0.791–0.899];
P=0.791)werenotfoundtosignificantlyinfluencepatientout- come.
4. Discussion
Our results highlight the importance of surgical treatment in the management of PJI due to MSSA. We observed that non-compliancewiththeIDSAsurgicalguidelines inpatients presentingwithMSSAPJIwasstronglyassociatedwithtreat- mentfailureduringlong-termfollow-up.
In2004,Zimmerlietal.publishedrecommendationsforthe surgicalandantimicrobialtreatmentofPJI[1].Theseguidelines havenotyetbeenprospectivelyvalidatedandwerebasedononly onerandomizedstudy, variouscohort studies,expertopinion, andanimalexperiments[1].However,Betschetal.andGiulieri et al.demonstrated the value of current recommendations in cohortstudiesincludingpatientsinfectedwithabroadrangeof pathogens[3,4].An inappropriatechoiceof conservativesur- gicalstrategiesandinadequateantibiotictreatmentswereboth associatedwithfailureduringfollow-up.DAIRisrecommended inpatients presentingwithacutepostoperativeandhematoge- nous PJI. Thesuspicion of ahematogenous origin isusually mainlybasedonacuteclinicalsignsofsepticarthritiswithfever, butitissometimesdifficulttoformallyexcludeachronicinfec- tion,evenifthereisnoprosthesisloosening,andespeciallyif thepatienthadnegativebloodcultures.MarkerofPJIchronic- itysuchasdelta-toxindetectionusingMALDI-TOFspectrum analysismayhelpdistinguishacutefromchronicPJI,andasa consequencemayhelpthedecisiontoadoptaconservativeora non-conservativestrategy[12].
Staphylococci(S.aureus andcoagulase-negativestaphylo- cocci) are the most frequent causes of implant-associated infections[1–5].Thesebacteriahavetheabilitytomodifytheir phenotypeinordertopersistinvivo.S.aureus,largelyassoci- atedwithrelapseandrecurrenceinpatientspresentingwithPJI, isknownto(i)formbiofilmontheimplantsurface,thuslead- ingtoan embedded,slow-growing, multi-layeredcommunity ofbacteriaand(ii)persistintracellularlyinosteoblastandfibro- blastbyproducingSCVs[6,13].Rifampicinhasdemonstrated in vitro bactericidal activity against surface-adhering, slow- growing,biofilm-producingS.aureusstrains.Forthisreason,it isconsideredthedrugofchoiceforthetreatmentofS.aureusPJI incombinationwithan appropriate anti-staphylococcalagent topreventacquisitionof resistance[1,2].However,little data has clearly demonstrated the superiority of rifampicin-based regimensincomparisonwithothercombinationsinpatientspre- sentingwithS.aureusPJI,andthereisadebateontheadequate doseforpatientspresentingwithPJI[9,14–17].
Weobservedthat neither inadequateantimicrobialtherapy nor the use of arifampicin-based regimen influenced patient outcome.Thismaybeduetothelowsamplesize,butalsoto thestrong association betweensurgicaltherapyandoutcome.
Indeed,thenon-compliancewithIDSAsurgicalguidelines(i.e., requirementtoperformcompleteimplantremovalifthepatient does not qualify for implant retention) led to a very high
cumulativeprobabilityoftreatmentfailureinourstudy(80%), thus highlighting the crucial importance of surgeryto eradi- cateS.aureus-associatedbiofilmandinfectedhostcellswithS.
aureusSCVs.Finally,mostpatientsexperiencedtreatmentfail- ureintheyearfollowingtheinitialepisodeof PJI.Thanksto anextendedfollow-up,wewereabletoshowthatsomepatients experienced treatmentfailureuptofouryearsafter theinitial PJIepisode.
Our study has limitations. First, the study was performed intwodifferentcentersinFranceandSwitzerland.Evenifno interactionbetweentreatmentstrategiesandthetwocenterswas detected,wecouldnotabsolutelyexcludebiasduetoatreatment differencebetweenthecenters.Despitetheretrospectivestudy design,wewereabletotrackthelong-termfollow-upofmost patientsandcollectedalargeamountofdatawithveryfewlost tofollow-up.
5. Conclusion
The study findings demonstrate that non-compliance with theIDSAsurgicalguidelinesinpatientspresentingwithMSSA PJIwasstronglyassociatedwithtreatmentfailureduringlong- termfollow-up.Tolimittheoccurrenceoftreatmentfailure,itis crucialtoperformcompleteimplantremovalifthepatientdoes notqualifyforimplantretention.
Contributionofauthors
TFcoordinatedthestudy.ABwrotethefirstversionofthe article.Allauthorscontributedtopatientcare,andimprovedthe article.
Funding
Therewasnosourceoffundingforthestudy.
Disclosureofinterest
Theauthorsdeclarethattheyhavenocompetinginterest.
References
[1]ZimmerliW,TrampuzA,OchsnerPE.Prosthetic-jointinfections.NEngl JMed2004;351:1645–54.
[2]OsmonDR,BerbariEF,BerendtAR,LewD,ZimmerliW,SteckelbergJM, etal.,InfectiousDiseasesSocietyofAmerica.Diagnosisandmanagement ofprostheticjointinfection:clinicalpracticeguidelinesbytheInfectious DiseasesSocietyofAmerica.ClinInfectDis2013;56(1):e1–25.
[3]BetschBY,EggliS,SiebenrockKA,TäuberMG,MühlemannK.Treatment ofjointprosthesisinfectioninaccordancewithcurrentrecommendations improvesoutcome.ClinInfectDis2008;46:1221–6.
[4]GiulieriSG,GraberP,OchsnerPA,ZimmerliW.Managementofinfection associatedwithtotalhiparthroplastyaccordingtoatreatmentalgorithm.
Infection2004;32:222–8.
[5]PeelTN,BuisingKL,ChoongPF.Prostheticjointinfection:challengesof diagnosisandtreatment.ANZJSurg2011;81:32–9.
[6]SendiP,Rohrbach M,GraberP,Frei R,OchsnerPE,ZimmerliW.S.
aureussmallcolonyvariantsinprostheticjointinfection.ClinInfectDis 2006;43:961–7.
[7]LeeJ,KangCI,LeeJH,JoungM,MoonS,WiYM,etal.Riskfactorsfor treatmentfailureinpatientswithprostheticjointinfections.JHospInfect 2010;75:273–6.
[8]Sociétédepathologieinfectieusedelanguefranc¸aise(SPILF),Collègedes universitairesdemaladiesinfectieusesettropicales(CMIT),Groupede pathologieinfectieusepédiatrique(GPIP),Sociétéfranc¸aised’anesthésieet deréanimation(SFAR),Sociétéfranc¸aisedechirurgieorthopédiqueettrau- matologique(SOFCOT),Sociétéfranc¸aised’hygiènehospitalière(SFHH), Sociétéfranc¸aisedemédecinenucléaire(SFMN),Sociétéfranc¸aisede médecinephysiqueetderéadaptation(SOFMER),Sociétéfranc¸aisede microbiologie(SFM),Sociétéfranc¸aisederadiologie(SFR-Rad),Société franc¸aisederhumatologie(SFR-Rhu).Recommendationsforboneand jointprostheticdeviceinfectionsinclinicalpractice(prosthesis,implants, osteosynthesis).MedMalInfect2010;40(4):185–211.
[9]Lora-TamayoJ,MurilloO,IribarrenJA,SorianoA,Sánchez-Somolinos M, Baraia-EtxaburuJM, etal., REIPI Groupfor the Study of Pros- theticInfection.Alargemulticenterstudyofmethicillin-susceptibleand methicillin-resistantS.aureus prostheticjointinfectionsmanagedwith implantretention.ClinInfectDis2013;56(2):182–94.
[10]SalgadoCD,DashS,CanteyJR,MarculescuCE.Higherriskoffailure ofmethicillin-resistantS.aureusprostheticjointinfections.ClinOrthop RelatRes2007;461:48–53.
[11]ParviziJ,AzzamK,GhanemE,AustinMS,RothmanRH.Periprosthetic infectionduetoresistantstaphylococci:seriousproblemsonthehorizon.
ClinOrthopRelatRes2009;467:1732–9.
[12]ValourF,RasigadeJP,Trouillet-AssantS,GagnaireJ,BouazizA,Karsenty J,etal.Delta-toxinproductiondeficiencyinStaphylococcusaureus:adiag- nosticmarkerofboneandjointinfectionchronicitylinkedwithosteoblast invasion and biofilm formation. Clin Microbiol Infect 2015;21(6):
e1–11.
[13]Trouillet-AssantS,GalletM,NauroyP,RasigadeJP,FlammierS,Parroche P,etal.DualimpactofliveS.aureusontheosteoclastlineage,leadingto increasedboneresorption.JInfectDis2015;211(4):571–81.
[14]Ferry T, Uc¸kay I, Vaudaux P, Franc¸ois P, Schrenzel J, Harbarth S, et al. Risk factors for treatment failure in orthopedic device-related methicillin-resistantS.aureusinfection.EurJClinMicrobiolInfectDis 2010;29:171–80.
[15]SennevilleE,JoulieD,LegoutL,ValetteM,DezèqueH,BeltrandE,etal.
Outcomeandpredictorsoftreatmentfailureintotalhip/kneeprosthetic jointinfectionsduetoS.aureus.ClinInfectDis2011;53:334–40.
[16]Aboltins CA,Page MA, BuisingKL, JenneyAW, DaffyJR,Choong PF, etal.Treatment ofstaphylococcal prostheticjoint infections with debridement,prosthesisretentionandoralrifampicinandfusidicacid.Clin MicrobiolInfect2007;13:586–91.
[17]PerlrothJ,KuoM,TanJ,BayerAS,MillerLG.Adjunctiveuseofrifampin forthetreatmentofS.aureusinfections:asystematicreviewofthelitera- ture.ArchInternMed2008;168:805–19.
