www .e l s e v i e r . c o m / l o c a t e / b j i d
The
Brazilian
Journal
of
INFECTIOUS
DISEASES
Original
article
Clostridium
difficile
outbreak
caused
by
NAP1/BI/027
strain
and
non-027
strains
in
a
Mexican
hospital
Rayo
Morfin-Otero
a,b,∗,
Elvira
Garza-Gonzalez
c,d,
Sara
A.
Aguirre-Diaz
a,
Rodrigo
Escobedo-Sanchez
a,
Sergio
Esparza-Ahumada
a,b,
Hector
R.
Perez-Gomez
a,
Santiago
Petersen-Morfin
b,
Esteban
Gonzalez-Diaz
a,b,
Adrian
Martinez-Melendez
d,
Eduardo
Rodriguez-Noriega
a,b,
for
the
Hospital
Civil
de
Guadalajara,
Fray
Antonio
Alcalde
Clostridium
difficile
Team
1aHospitalCivildeGuadalajara,FrayAntonioAlcalde,Mexico
bInstitutodePatologíaInfecciosayExperimental,CentroUniversitariodeCienciasdelaSalud,UniversidaddeGuadalajara,Mexico
cServiciodeGastroenterología,HospitalUniversitarioDr.JoséEleuterioGonzález,UniversidadAutónomadeNuevoLeón,Mexico
dDepartamentodePatologíaClínica,HospitalUniversitarioDr.JoséEleuterioGonzález,UniversidadAutónomadeNuevoLeón,Mexico
a
r
t
i
c
l
e
i
n
f
o
Articlehistory:
Received1July2015 Accepted4September2015 Availableonline24November2015
Keywords: Clostridiumdifficile Outbreak 027strain Mexico
a
b
s
t
r
a
c
t
Background:ClostridiumdifficileinfectionscausedbytheNAP1/B1/027strainaremoresevere,
difficulttotreat,andfrequentlyassociatedwithrelapses.
Methods:Acase–controlstudywasdesignedtoexamineaC.difficileinfection(CDI)outbreak
overa12-monthperiodinaMexicanhospital.ThediagnosisoftoxigenicCDIwasconfirmed byreal-timepolymerasechainreaction,PCR(CepheidXpertC.difficile/Epi).
Results:Duringthestudyperiod,288adultpatientswereevaluatedand79(27.4%)patients
hadconfirmedCDI(PCRpositive).C.difficilestrainNAP1/B1/027wasidentifiedin31(39%)of thepatientswithconfirmedCDI(240controlswereincluded).SignificantriskfactorsforCDI includedanyunderlyingdisease(p<0.001),priorhospitalization(p<0.001),andantibiotic (p<0.050)orsteroid(p<0.001)use.Laboratoryabnormalitiesincludedleukocytosis(p<0.001) andlowserumalbuminlevels(p<0.002).Attributablemortalitywas5%.Relapsesoccurred in10%ofpatients.RiskfactorsforC.difficileNAP1/B1/027straininfectionsincludedprior useofquinolones(p<0.03).
RiskfactorsforCDIcausedbynon-027strainsincludedchroniccardiacdisease(p<0.05), chronicrenaldisease(p<0.009),andelevatedserumcreatininelevels(p<0.003).Deathsand relapsesweremostfrequentinthe027group(10%and19%,respectively).
Conclusions:C.difficileNAP1/BI/027strainandnon-027strainsareestablishedpathogensin
ourhospital.Accordingly,surveillanceofC.difficileinfectionsisnowpartofournosocomial preventionprogram.
©2015ElsevierEditoraLtda.Allrightsreserved.
∗ Correspondingauthorat:InstitutodePatologíaInfecciosayExperimental,CalleHospital308,ColoniaElRetiro,C.P44280Guadalajara,
Jalisco,Mexico.
E-mailaddress:rayomorfin@gmail.com(R.Morfin-Otero).
1 MembersfortheHospitalCivildeGuadalajara,FrayAntonioAlcaldeClostridiumdifficileTeamweregiveninAppendixA.
http://dx.doi.org/10.1016/j.bjid.2015.09.008
Introduction
Clostridium difficile infections (CDI) are the leading
world-wide cause of healthcare-associateddiarrhea and in some countriesCDIsurpassallotherhealthcare-associated infec-tions(HCAI).1ArecentprevalencesurveyofHCAIconducted
across183hospitalsdeterminedthatC.difficilewasthemost frequentlyreportedinfectiousagent,responsiblefor12.1%of allHCAI.1
IntheUnitedStatesofAmerica(USA)during2011,15,461 CDIcaseswerereportedwith24.2%ofcaseshavinganonset during hospitalization. Incident CDI cases were estimated tobe>450,000withanestimated>29,000deaths.2 However,
theemergence oftheC. difficileNAP1/B1/027strain in2000 changedthemorbidity andmortalityrates associatedwith CDI.3,4
Since2004,the role ofother emergentC. difficilestrains causing human disease has expanded. These strains are derivedfrom39differentribotypesandsomeC.difficilestrains havebeenfoundtobetoxinA-negativebuttoxinB-positive,5
and027strainwasthesecondmostcommonisolate respon-sibleforCDI.6Ribotype078wasreportedtohaveanincreased
prevalence,7 and ribotype244 seemsto cause moresevere
disease with higher mortality rates than rates associated withribotype027.8,9 Theprevalenceofotherribotypesnow
appearstosurpassthatof027,includingribotypes037,018, and078.10–12
Epidemiologicresearch of CDI resulting from infections withdiverseC.difficilestrains,includingstrainNAP1/B1/027 in developing countries, is expanding and includes data regardinghospitalepidemiology,clonalspread,and dissem-inationacrosstherespectivecountries.13–16
Thepresentstudyreportsona12-monthevaluationofa CDIoutbreakcausedbydifferentC.difficilestrainsincluding theNAP1/BI/027strain.
Methods
Setting,studydesign,andstudypopulation
TheoutbreakdescribedinthisreportoccurredattheHospital CivildeGuadalajaraFrayAntonioAlcalde,an899-bedtertiary careteachinghospitallocatedinthecityofGuadalajara,the secondlargestcityinMexico.
This was a case–control study of adult patients with hospital-onsetCDIpresentingbetween December2013and December2014.During thestudy period288adultpatients wereevaluatedandallpatientshaddiarrheadefinedasthe passageof≥3unformedstools(Bristolscaletype5–7)within 24or 48hafteradmission.17 Casepatientswere definedas
thosewithafirstepisodeofnosocomialCDI.
Clostridiumdifficiletoxinidentification
Starting in April 2014, all stool samples were tested for
C.difficile toxins usingreal-time polymerasechainreaction
(PCR)(CepheidXpert C.difficile/Epi,Cepheid,Sunnyvale CA) toidentifytoxin-producingC.difficilestrains,includingstrain NAP1/B1/027.PriortotheavailabilityofPCR-baseddiagnostic
approaches all diarrhea specimenswere tested byenzyme immunoassay(MeridianBioscience,Cincinnati,OH,USA).All positive specimens were saved for future testing.All stool specimenswerestoredat4◦Cforfivedays,andthenfrozen
at −70◦C. AfterPCR retesting, onlypositive sampleswere includedinthefinalanalysis.
Controlpatients
PatientswithoutdiarrheaorapositiveCDItestwereselected atthe same time and ward that CDI patients were identi-fied.Controlpatientswererandomlyselectedacrossthestudy period.Controlpatientswerematchedtocasepatientsata3:1 ratio.
Definitions
Previous hospitalizationwas defined asa hospitalstay six weekspriortotheonsetofdiarrhea.Recentantibiotictherapy andsteroiduseweredefinedasexposuretothesemedicines sixweekspriortodiarrheaonset.
Clinicalseverityscoreassessmentandoutcome
Patientswere clinicallyevaluatedfordisease severityusing theSHEA/IDSAdefinitionsofmild,moderate,orsevere dis-ease.Serumcreatininelevelswereincludedinthedefinition ofseveredisease.18Inaddition,age>60years,fever>38.3◦C,
andaWBCcount>15,000wereusedtofurtherdefineclinically severedisease.19Patientswith>2findingswereconsideredto
haveseveredisease.
Apooroutcomewasdefinedasdeathwithin14daysafter CDIdiagnosis.Favorableoutcomewasdefinedbysurvival14 days afterCDI diagnosis.Relapse was definedas asecond episodeofdiarrheaafteradequateresponsetotherapy.
Therapyandfollow-up
TherapyforCDIwasadministeredfor10daysafteran ade-quateresponsetotreatmentwasachieved(definedasa50% reduction of loose stoolsafter 24h oftherapy, continuous reductionafter48hoftreatment,andnodiarrheaafter72h oftreatment).Allpatientsdischargedwherefollowedvia tele-phoneevery30days.
Statisticalanalysis
Thedatageneratedwerecoded,entered,validated,and ana-lyzedusingtheStatisticalPackageforSocialScience(SPSS), version22.0.Univariateanalyseswere usedtodescribe sig-nificant variables among cases and controls and among individualsinfectedwithstrain027andindividualsinfected withnon-027strains.P-valueswerecalculatedusingthe Chi-squaredtestortheFisher’sexacttestforcategoricalvariables andtheStudent’st-testorWilcoxonrank-sumtestfor con-tinuousvariables.Ap-value≤0.05wasconsideredstatistically significant.Multivariateanalysis:logisticregressionanalysis wascarried outconsideringCDIasdependentvariable and clinicalanddemographicdataasindependentvariables.
Table1–CharacteristicsofCDIpatientsandcontrols,severity,outcomes,relapses,andclinicalaspectsofpatients infectedwith027andnon-027strains.
Parameters CDIpatients(n=79)
n(%) Controls(n=240) n(%) p-value 027strain(n=31) n(%) Non-027strain (n=48) n(%) p-value Age(y) 18–30 23(29.1) 62(25.8) 0.28 7(23) 16(34) 0.22 31–50 24(30.4) 88(36.7) 0.15 11(35) 13(27) 0.29 51–65 21(26.6) 53(22.1) 0.20 10(32) 12(25) 0.32 >65 11(13.9) 37(15.4) 0.38 3(10) 7(14) 0.39 Gender Male 51(64.5) 148(61.7) 0.32 21(67.7) 30(62.5) 0.40 Female 28(35.5) 92(38.3) 0.32 10(32.3) 18(37.5) 0.40 Underlyingdisease Any 73(92.4) 170(70.8) <0.001 27(87.1) 46(95.6) 0.15 Malignancy 11(13.9) 18(7.5) 0.142 6(19.3) 5(10.4) 0.21 Diabetesmellitus 20(25.3) 51(21.3) 0.27 5(16.1) 15(31.3) 0.10 Chroniccardiacdisease 22(27.8) 49(20.4) 0.112 5(16.1) 17(35.4) 0.05
Chronichepaticdisease 2(2.5) 5(2.0) 0.55 0 2(4.2) 0.36
Previousepisodeof pneumonia
16(20.2) 16(6.7) 0.005 5(16.1) 11(22.9) 0.33
Chronicrenaldisease 23(29.1) 43(17.9) 0.027 4(12.9) 19(39.6) 0.009
Healthcare-associatedexposure Priorhospitalization 43(54.4) 57(23.8) <0.001 20(64.5) 23(47.9) 0.11 Priorsurgery 44(55.7) 117(48.8) 0.173 17(62.9) 27(56.3) 0.54 Priorantibiotics Any 58(73.4) 150(62.5) 0.050 23(74.2) 35(72.9) 0.55 Betalactams 43(54.4) 108(45) 0.155 17(54.8) 30(62.5) 0.32 Quinolones 16(20.3) 31(12.9) 0.142 9(39.13) 5(10.4) 0.03 Clindamycin 10(12.7) 38(15.8) 0.313 4(17.39) 6(12.5) 0.60
Prioruseofacidsuppressingmedication
Protonpumpinhibitors 67(84.8) 192(80.0) 0.21 25(80.65) 42(87.5) 0.30
H2blocker 5(6.3) 18(7.5) 0.47 1(3.23) 4(8.3) 0.34
Prioruseofsteroids 17(21.5) 16(6.7) <0.001 8(25.8) 9(18.8) 0.31 Whitebloodcellscount
>12,000/mm3 37(46.8) 63(26.3) <0.001 15(48.4) 22(45.8) 0.50 Serumcreatinine ≥1.5mg/dl 22(27.8) 53(22.1) 0.20 3(9.7) 19(39.6) 0.003 Serumalbumin<3gdl 46(58.2) 91(37.9) <0.002 18(58.1) 28(58.3) 0.58 InitialClinicalSeverity
Score≥2 69(87.3) – – 28(90) 41(85) 0.39 Outcome Poor/death 4(5) – – 3(10) 1(2) 0.16 Good/cured 75(95) – – 28(90) 47(98) 0.16 Relapses 8(10) – – 6(19) 2(4) 0.03
Results
StudypopulationThe age range of CDI patients and controls were similar (Table1).Patients>65yearsofageweretheminorityinboth groups (Table 1). There was no genderdifference between cases and controls; however, males were more frequently affected with CDI than females (Table 1). Thepresence of any underlying disease was an important risk factor for acquiringCDI,especially aprevious episode ofpneumonia orthepresenceofchronicrenaldisease(Table1).Additional riskfactors associatedwithCDIincluded prior hospitaliza-tion,antibioticorsteroiduse,andelevatedwhitebloodcell
counts(>12,000/mm3)combinedwithlowserumalbumin
lev-els(<3g/dl)(Table1).Four(5%)patientsdiedintheCDIgroup and8(10%)relapsed(Table1).TheincidenceofCDIwas1.7 per1000discharges.
NAP1/B1/027infections
StrainNAP1/B1/027wasidentifiedin31 (39%)patientswith CDI.ThereweresomedifferencesbetweenCDIresultingfrom infectionswithstrain027andnon-027strains.Thepresence of chronic cardiac disease and chronic renal disease were foundtobesignificantlymorefrequentinthenon-027group (Table1).Althoughbothgroupshadasimilarinitialseverity score,moredeathsandrelapseswereassociatedwithstrain
Table2–Multivariatelogisticregressionanalysisofrisk factorsforCDI.
Riskfactor OR CI95% p-value
Whitebloodcellscount >12,000/mm3 2.541 1.414–4.567 0.002 Priorhospitalization 4.029 2.240–7.246 0.001 SerumAlbumin<3gdl 2.026 1.138–3.608 0.016 Previousepisodeof pneumonia 4.251 1.848–9.779 0.001 Prioruseofsteroids 5.077 2.203–11.698 0.001
027 infections (Table 1). Additional risk factors associated with027andnon-027infectionswereprioruseofquinolone andabnormalserumcreatininelevel(>1.5mg/dl),respectively (Table1).
Logisticregression analysis included thesignificant risk factors forCDI prioruse ofsteroids, aprevious episode of pneumonia,andpriorhospitalization(Table2).Also identi-fiedabnormalwhitebloodcellcountandlowserumalbumin levelsasindependentriskfactorsforacquiringCDI(Table2).
Discussion
TheCDIoutbreakdescribedinthisreportoccurredfollowing introductionoftheC.difficile027strainintoourhospitalbya patientdiagnosedinDecember2013.Thisindividualhadhad multiplehealthcarecontactsintheUSA(includingseveraldue todiarrhea)priortobeingadmittedtoourneurosurgicalward. IntroductionofC.difficileintoahospitalwillusuallydevelop intoanoutbreakandpreviousstudieshavedocumented out-breaksfollowingdetectionofstrain027.3,4,20
OtherLatinAmerican countriesfrom Centraland South America havenow described the presenceand
dissemina-tion of C. difficile.15,21 The C. difficile dissemination pattern
seeninMexicowassimilartothatdescribedinhospitalsin the USA and Canada, but different from that of the Euro-peanUnionwhereC.difficile027isnotyetasprevalent.The appearance,establishment,anddisseminationofC.difficilein Mexicoseemedtooccurinlargereferralhospitalswhereahigh percentageofpatientsadmittedhadmultipleriskfactorsfor CDI.14,22
Thepresenceofaseriousunderlyingdiseaseisafrequent riskfactor forthe development ofCDI23 and the presence
ofanyunderlyingillness(particularlyapreviousepisodeof pneumonia)wasasignificantriskfactorinpatientscompared tocontrols.
Inourpopulation, communityacquiredpneumoniawas diagnosedmostfrequentlyinolderpatientswith comorbidi-tiesincludingchronicobstructivepulmonarydisease.These patientstypicallyhadmultipleprevioushealthcareexposures includingpriorhospitalizationsallowingforagreater prob-abilityofacquiringC.difficile. Becausecurrentguidelines of our hospital recommend administration of quinolones as empirictreatment forpneumonia24 this patientgroup had
beenexposedtothisdrug.
Afrequentriskfactorinourstudyincludedpatientswith chronicrenaldisease.Sinceourhospitalisaregionalcenter forthediagnosisandcareofpatientsinneedofrenal replace-menttherapyorarenaltransplantitisresponsibleforalarge
populationthatisaffectedbythisunderlyingillness.25Similar
topatientswithotherchronicdiseases,patientswithchronic renal disease have multiple healthcarecontacts (including dialysis) and have multiple previous antibiotic exposures duetoempiricordefinitivetreatmentofdifferentinfectious diseases complications,including peritonitisresultingfrom peritoneal dialysis.Concomitant administrationofsteroids occursfrequentlywhenpatientshaveseriouscomorbidities and needassistance inthe treatmentofvarious complica-tions.
TheclinicalfeaturesfoundinourCDI patientsincluded anincreasedwhitebloodcell count,elevatedserum creati-nineand reducedserumalbuminlevels.Thesefindingsare thebasisformostclinicalpredictionrulesusedtoday.18,19,26–29
PatientspresentingwithsevereCDItypicallywereolderand had anincreasednumber ofbowelmovements,had a his-tory of systemic antibiotic use, and presented with fever, abdominal distention, abnormalrespiratory rate, abnormal level ofC-reactiveprotein,priorepisodes ofCDI,increased whitebloodcellcount,elevatedserumcreatininelevel, and lowserumalbuminlevel.18,19,26–29Usingtheseclinical
predic-tionrulesmostofourpatientshadsevereCDI.
Theuse ofclinical predictionrulesinCDI arealsoused todetermineindividualsatriskofhavingpooroutcomesor a relapse.28,30–34 Themostprominentfactor predictiveofa
pooroutcomeorrelapseamongCDIpatientsdescribedinthis reportwasinfectionwithstrain02735andchronicrenal
dis-ease.
Aftereliminatingconfounders,independentriskfactorsfor CDIincludedprioruseofsteroids,previousepisodesof pneu-monia,andpriorhospitalizations(Table2).Theepidemiology
ofC.difficileinfections isconstantlychangingandprobably
explainssomeofthedifferencesfoundinourstudycompared withpreviousobservationsmadeinMexico.14,22,36
ThediagnosisofCDIwasprimarilycarriedoutusinga com-mercialPCRkit,atestthathashighsensitivityandspecificity butwithseverallimitations,includingtheinabilityofthistest toidentifyemergentC.difficilevariants.9,37,38
The choice for initial empirical therapy in our study consisted of metronidazole. Other therapeutic choices included administration oforalvancomycin as opposed to intravenousadministrationcombinedwitheitherintravenous metronidazoleorintravenoustigecyclinebasedonindividual responsetooralmetronidazole.39,40
Inanefforttocontroltheoutbreakourintervention pro-gramfocusedonidentifyingCDIcasesasquicklyaspossible, providingearlytreatment,isolatingCDIcasesinadedicated ward,andrestrictingallquinoloneuse.41–43 Thepresenceof
adiseasesuchasCDIthatistransmittedviaoral-fecal con-taminationpromptedustoreevaluatepatienthandwashing practicespriortoeachmealortheintakeoforalmedication, inadditiontoassessinghandwashingpracticesofthestaff assignedtohelpfeedpatients.Thisresultedinthe implemen-tationofanaggressivepatienthandwashingcampaign.
All patientsdischarged afteranepisodeofCDI received careful instructions on how to proceed should a relapse occur.44 Theinstructionsincludedadescriptionofsomeof
thesymptomsthatmaypresentduringarelapse,wheretoget medicalattention,andwhattoinformhealthcarepersonnel onarrivaltoclinics.
Thepresent study had several limitationsincluding the
lackofC.difficileculturestoenabletyping,limiteduseofa
computedtomographyscanforabdominalradiographic imag-ing priortocolonoscopy, colonoscopyfordiagnosis ofCDI, follow-upPCRtestingwasusedonlyinselectpatients,45and
noautopsieswereperformed.
Inconclusion,thecontrolofCDIinourhospitalnow rep-resentsaconstantchallenge.ThecontrolofCDIinahospital likeoursshouldincludeatailoredstrategydesignedto iden-tifycasesofCDIasrapidlyaspossible.Thisstudyrepresents thefirstdescriptionofanextendedCDIoutbreakcausedby diverseC.difficilestrainsincludingtheNAP1/B1/027strainin aMexicanhospital.
Funding
Nofundingwasreceivedforthiswork.
Ethical
approval
ThisstudywasperformedwiththeapprovaloftheHospital CivildeGuadalajaraFrayAntonioAlcaldeInstitutionalReview Board.WritteninformedconsentapprovedbytheEthics Com-mitteewasobtainedfromallpatients.
Conflicts
of
interest
Theauthorsdeclarenoconflictsofinterest.
Appendix
A.
Members for the Hospital Civil de Guadalajara,Fray Anto-nio AlcaldeClostridium difficile Team were:Leon-Garnica G, Castillo-MondragonA,Camacho-Rubio JR,Rodriguez-Nu ˜nez AJ, Mendoza-Mujica C, Heredia-Cervantes J, Mata-Esteban RA, Llamas-Alonso J, Lucio-Figueroa JO, Macias-Hernandez KZ,Macías-Bola ˜nosDJ.Cardenas-LaraFJ,Fernandez-Ramirez A(Instituto de Patologia Infecciosay Experimental, Centro UniversitarioCienciasdelaSalud,Universidadde Guadala-jara),Vazquez-LeónM,Gomez-QuirozP.(InfectiousDiseases UnitAttendings),EduardoOrtigosa-Medrano,Gomez-Gomez K, Vargas-Garcia LF (Infectious Diseases Unit Fellows) Gutierrez-MartinezES,Garcia-ReyesMG,Zamora-MoralesS, Casillas-PachecoMA, Martinez-CardonaL,Maga ˜na-IbarraS. Tello G. (Epidemiology) Rodriguez-Chagollan JJ, Anguiano-GaytanG,Atilano-DuranMCG,Llanos-PerezE,Gomez-Quiroz A,ZavalaM(Microbiology).
r
e
f
e
r
e
n
c
e
s
1. MagillSS,EdwardsJR,BambergW,etal.Multistate
point-prevalencesurveyofhealthcare-associatedinfections. NEnglJMed.2014;370:1198–208.
2. LessaFC,MuY,BambergWM,etal.BurdenofClostridium difficileinfectionintheUnitedStates.NEnglJMed. 2015;372:825–34.
3.LooVG,PoirierL,MillerMA,etal.Apredominantlyclonal multi-institutionaloutbreakofClostridiumdifficile-associated
diarrheawithhighmorbidityandmortality.NEnglJMed. 2005;353:2442–9.
4.McDonaldLC,KillgoreGE,ThompsonA,etal.Anepidemic, toxingene-variantstrainofClostridiumdifficile.NEnglJMed. 2005;353:2433–41.
5.vandenBergRJ,ClaasEC,OyibDH,etal.Characterizationof toxinA-negative,toxinB-positiveClostridiumdifficileisolates fromoutbreaksindifferentcountriesbyamplifiedfragment lengthpolymorphismandPCRribotyping.JClinMicrobiol. 2004;42:1035–41.
6.MartinH,WilleyB,LowDE,etal.Characterizationof
ClostridiumdifficilestrainsisolatedfrompatientsinOntario, Canada,from2004to2006.JClinMicrobiol.
2008;46:2999–3004.
7.GoorhuisA,BakkerD,CorverJ,etal.EmergenceofClostridium difficileinfectionduetoanewhypervirulentstrain,
polymerasechainreactionribotype078.ClinInfectDis. 2008;47:1162–70.
8.BaldanR,CavallerioP,TuscanoA,etal.Firstreportof hypervirulentstrainspolymerasechainreactionribotypes 027and078causingsevereClostridiumdifficileinfectionin Italy.ClinInfectDis.2010;50:126–7.
9.LimSK,StuartRL,MackinKE,etal.Emergenceofaribotype 244strainofClostridiumdifficileassociatedwithseveredisease andrelatedtotheepidemicribotype027strain.ClinInfect Dis.2014;58:1723–30.
10.DuP,CaoB,WangJ,etal.SequencevariationintcdAandtcdB ofClostridiumdifficile:ST37withtruncatedtcdAisapotential epidemicstraininChina.JClinMicrobiol.2014;52:3264–70.
11.NgamskulrungrojP,SanmeeS,PusathitP,etal.Molecular epidemiologyofClostridiumdifficileinfectioninalarge teachinghospitalinThailand.PLOSONE.2015;10:e0127026.
12.BaldanR,TrovatoA,BianchiniV,etal.Asuccessfulepidemic genotype:ClostridiumdifficilePCRribotype018.JClin Microbiol.2015;53:2575–80.
13.LopardoG,Morfin-OteroR,MoranV2nd,etal.Epidemiology ofClostridiumdifficile:ahospital-baseddescriptivestudyin ArgentinaandMexico.BrazJInfectDis.2015;19:8–14.
14.Camacho-OrtizA,Lopez-BarreraD,Hernandez-GarciaR,etal. FirstreportofClostridiumdifficileNAP1/027inaMexican Hospital.PLOSONE.2015;10:e0122627.
15.AguayoC,FloresR,LevesqueS,etal.Rapidspreadof
ClostridiumdifficileNAP1/027/ST1inChileconfirmsthe emergenceoftheepidemicstraininLatinAmerica.Epidemiol Infect.2015:1–5.
16.Quesada-GomezC,Lopez-UrenaD,Acuna-AmadorL,etal. Emergenceofanoutbreak-associatedClostridiumdifficile
variantwithincreasedvirulence.JClinMicrobiol. 2015;53:1216–26.
17.CaroffDA,EdelsteinPH,HamiltonK,PeguesDA,Program CDCPE.TheBristolstoolscaleanditsrelationshipto
Clostridiumdifficileinfection.JClinMicrobiol.2014;52: 3437–9.
18.CohenSH,GerdingDN,JohnsonS,etal.Clinicalpractice guidelinesforClostridiumdifficileinfectioninadults:2010 updatebythesocietyforhealthcareepidemiologyofAmerica (SHEA)andtheinfectiousdiseasessocietyofAmerica(IDSA). InfectControlHospEpidemiol.2010;31:431–55.
19.ZarFA,BakkanagariSR,MoorthiKM,DavisMB.Acomparison ofvancomycinandmetronidazoleforthetreatmentof
Clostridiumdifficile-associateddiarrhea,stratifiedbydisease severity.ClinInfectDis.2007;45:302–7.
20.JohnsonS,SamoreMH,FarrowKA,etal.Epidemicsof diarrheacausedbyaclindamycin-resistantstrainof
Clostridiumdifficileinfourhospitals.NEnglJMed. 1999;341:1645–51.
21.Lopez-UrenaD,Quesada-GomezC,MirandaE,FonsecaM, Rodriguez-CavalliniE.SpreadofepidemicClostridiumdifficile
NAP1/027inLatinAmerica:casereportsinPanama.JMed Microbiol.2014;63:322–4.
22.Camacho-OrtizA,Galindo-FragaA,Rancel-CorderoA,etal. FactorsassociatedwithClostridiumdifficilediseaseina tertiary-caremedicalinstitutioninMexico:acase–control study.RevInvestClin.2009;61:371–7.
23.LefflerDA,LamontJT.Clostridiumdifficileinfection.NEnglJ Med.2015;372:1539–48.
24.MandellLA,WunderinkRG,AnzuetoA,etal.Infectious DiseasesSocietyofAmerica/AmericanThoracicSociety consensusguidelinesonthemanagementof
community-acquiredpneumoniainadults.ClinInfectDis. 2007;44Suppl.2:S27–72.
25.ThongprayoonC,CheungpasitpornW,PhatharacharukulP, MahaparnP,BruminhentJ.Highmortalityriskinchronic kidneydiseaseandendstagekidneydiseasepatientswith
Clostridiumdifficileinfection:asystematicreviewand meta-analysis.JNatSci.2015:1.
26.FujitaniS,GeorgeWL,MurthyAR.Comparisonofclinical severityscoreindicesforClostridiumdifficileinfection.Infect ControlHospEpidemiol.2011;32:220–8.
27.MillerMA,LouieT,MullaneK,etal.Derivationandvalidation ofasimpleclinicalbedsidescore(ATLAS)forClostridium difficileinfectionwhichpredictsresponsetotherapy.BMC InfectDis.2013;13:148.
28.ButtE,FosterJA,KeedwellE,etal.Derivationandvalidation ofasimple,accurateandrobustpredictionruleforriskof mortalityinpatientswithClostridiumdifficileinfection.BMC InfectDis.2013;13:316.
29.NaX,MartinAJ,SethiS,etal.Amulti-centerprospective derivationandvalidationofaclinicalpredictiontoolfor severeClostridiumdifficileinfection.PloSone.
2015;10:e0123405.
30.MillerM,GravelD,MulveyM,etal.Healthcare-associated
ClostridiumdifficileinfectioninCanada:patientageand infectingstraintypearehighlypredictiveofsevere outcomeandmortality.ClinInfectDis.2010;50: 194–201.
31.EyreDW,WalkerAS,WyllieD,etal.Predictorsoffirst recurrenceofClostridiumdifficileinfection:implicationsfor initialmanagement.ClinInfectDis.2012;55Suppl.2: S77–87.
32.PetrellaLA,SambolSP,CheknisA,etal.Decreasedcureand increasedrecurrenceratesforClostridiumdifficileinfection causedbytheepidemicC.difficileBIstrain.ClinInfectDis. 2012;55:351–7.
33.SeeI,MuY,CohenJ,etal.NAP1straintypepredictsoutcomes fromClostridiumdifficileinfection.ClinInfectDis.
2014;58:1394–400.
34.RaoK,MicicD,NatarajanM,etal.ClostridiumdifficileRibotype 027:relationshiptoage,detectabilityoftoxinsAorBinstool withrapidtesting,severeinfection,andmortality.ClinInfect Dis.2015;61:233–41.
35.ScardinaT,LabuszewskiL,PachecoSM,AdamsW, SchreckenbergerP,JohnsonS.Clostridiumdifficileinfection (CDI)severityandoutcomeamongpatientsinfectedwiththe NAP1/BI/027straininanon-epidemicsetting.InfectControl HospEpidemiol.2015;36:280–6.
36.FreemanJ,BauerMP,BainesSD,etal.Thechanging epidemiologyofClostridiumdifficileinfections.ClinMicrobiol Rev.2010;23:529–49.
37.GilbreathJJ,VermaP,AbbottAN,Butler-WuSM.Comparison oftheVerigeneClostridiumdifficile,SimplexaC.difficile
UniversalDirect,BDMAXCdiff,andXpertC.difficileassaysfor thedetectionoftoxigenicC.difficile.DiagnMicrobiolInfect Dis.2014;80:13–8.
38.SchroederLF,RobilottiE,PetersonLR,BanaeiN,DowdyDW. Economicevaluationoflaboratorytestingstrategiesfor hospital-associatedClostridiumdifficileinfection.JClin Microbiol.2014;52:489–96.
39.AbouChakraCN,PepinJ,SirardS,ValiquetteL.Riskfactorsfor recurrence,complicationsandmortalityinClostridiumdifficile
infection:asystematicreview.PLOSOne.2014;9:e98400.
40.BagdasarianN,RaoK,MalaniPN.Diagnosisandtreatmentof
Clostridiumdifficileinadults:asystematicreview.JAMA. 2015;313:398–408.
41.MutoCA,BlankMK,MarshJW,etal.Controlofanoutbreakof infectionwiththehypervirulentClostridiumdifficileBIstrain inauniversityhospitalusingacomprehensivebundle approach.ClinInfectDis.2007;45:1266–73.
42.AbbettSK,YokoeDS,LipsitzSR,etal.Proposedchecklistof hospitalinterventionstodecreasetheincidenceof healthcare-associatedClostridiumdifficileinfection.Infect ControlHospEpidemiol.2009;30:1062–9.
43.DubberkeER,CarlingP,CarricoR,etal.Strategiestoprevent
Clostridiumdifficileinfectionsinacutecarehospitals:2014 Update.InfectControlHospEpidemiol.2014;35:628–45.
44.D’AgostinoRBSr,CollinsSH,PencinaKM,KeanY,GorbachS. RiskestimationforrecurrentClostridiumdifficileinfection basedonclinicalfactors.ClinInfectDis.2014;58:1386–93.
45.DubberkeER,ReskeKA,SeilerS,HinkT,KwonJH,Burnham CA.RiskfactorsforacquisitionandlossofClostridiumdifficile
colonizationinhospitalizedpatients.AntimicrobAgents Chemother.2015;59:4533–43.
