Does empirical treatment of community-acquired pneumonia with fluoroquinolones delay tuberculosis treatment and result in fluoroquinolone resistance in Mycobacterium tuberculosis? Controversies and solutions

(1)

ContentslistsavailableatSciVerseScienceDirect

International

Journal

of

Antimicrobial

Agents

j o ur na l ho me p ag e :ht t p : / / w w w . e l s e v i e r . c o m / l o c a t e / i j a n t i m i c a g

Review

Does

empirical

treatment

of

community-acquired

pneumonia

with

ﬂuoroquinolones

delay

tuberculosis

treatment

and

result

in

ﬂuoroquinolone

resistance

in

Mycobacterium

tuberculosis?

Controversies

and

solutions

Gwan-Han

Shen

a

_,

_Thomas

_Chang-Yao

_Tsao

b

_,

_Shang-Jyh

_Kao

c

_,

_Jen-Jyh

_Lee

d

_,

_Yen-Hsu

_Chen

e

_,

Wei-Chung

Hsieh

f

_,

_Gwo-Jong

_Hsu

g

_,

_Yen-Tao

_Hsu

h

_,

_Ching-Tai

_Huang

i

_,

_Yeu-Jun

_Lau

j

_,

Shih-Ming

Tsao

k

_{, Po-Ren}

_Hsueh

l,∗

a_Division_of_Respiratory_and_Critical_Care_Medicine,_Department_of_Internal_Medicine,_Taichung_Veterans_General_Hospital,_Taichung,_Taiwan

b_Division_of_Respiratory_and_Critical_Care_Medicine,_Department_of_Internal_Medicine,_School_of_Medicine,_Chung_Shan_Medical_University,_and_University_Hospital,_Taichung,_Taiwan c_Pulmonary_Division,_Department_of_Internal_Medicine,_Shin_Kong_Wu_Ho-Su_Memorial_Hospital,_Taipei,_Taiwan

d_TB_Laboratory_Section,_Department_of_Internal_Medicine,_Buddhist_Tzu_Chi_General_Hospital,_Hualien_and_Tzu_Chi_University,_Hualien,_Taiwan

e_Division_of_Infectious_Diseases,_Department_of_Internal_Medicine,_Kaohsiung_Medical_University_Hospital,_Graduate_Institute_of_Medicine,_Kaohsiung_Medical_University,_Kaohsiung, Taiwan

f_Division_of_Respiratory_and_Critical_Care_Medicine,_Department_of_Internal_Medicine,_Da_Chien_General_Hospital,_Maioli,_Taiwan

g_Division_of_Infectious_Diseases,_Department_of_Internal_Medicine,_Ditmanson_Medical_Foundation_Chia-Yi_Christian_Hospital,_Chia-Yi,_Taiwan h_Division_of_Pulmonary_and_Critical_Care_Medicine,_Department_of_Internal_Medicine,_Taipei_City_Hospital,_Heping_Fuyou_Branch,_Taipei,_Taiwan i_Division_of_Infectious_Diseases,_Department_of_Medicine,_Chang_Gung_University_and_Memorial_Hospital,_Taiwan

j_Infectious_Diseases,_Department_of_Internal_Medicine,_Show_Chwan_Memorial_Hospital,_Taichung,_Taiwan

k_Infectious_Section,_Internal_Medicine_Department,_Institute_of_Microbiology_and_Immunology,_Chung_Shan_Medical_University,_and_University_Hospital,_Taichung,_Taiwan l_Departments_of_Laboratory_Medicine_and_Internal_Medicine,_National_Taiwan_University_Hospital,_National_Taiwan_University_College_of_Medicine,_No.₇_Chung-Shan_South_Road, Taipei100,Taiwan

a

r

t

i

c

l

e

i

n

f

o

Keywords: Community-acquiredpneumonia Fluoroquinolones Tuberculosis Fluoroquinoloneresistance Multidrug-resistantMycobacterium tuberculosis

a

b

s

t

r

a

c

t

Theroleoffluoroquinolones (FQs)asempiricaltherapyforcommunity-acquiredpneumonia(CAP) remainscontroversialincountrieswithhightuberculosis(TB)endemicityowingtothepossibilityof delayedTBdiagnosisandtreatmentandtheemergenceofFQresistanceinMycobacteriumtuberculosis. Althoughtheratesofmacrolide-resistantStreptococcuspneumoniaeandamoxicillin/clavulanic acid-resistantHaemophilusinfluenzaehaverisentoalarminglevels,theratesofrespiratoryFQ(RFQ)resistance amongsttheseisolatesremainrelativelylow.Itisreportedthatca.1–7%ofCAPcasesarere-diagnosed aspulmonaryTBinAsiancountries.Alongerduration(≥7days)ofsymptoms,ahistoryofnightsweats, lackoffever(>38◦C),infectioninvolvingtheupperlobe,presenceofcavitaryinfiltrates,opacityinthe lowerlungwithoutthepresenceofair,lowtotalwhitebloodcellcountandthepresenceoflymphopenia arepredictiveofpulmonaryTB.AmongstpatientswithCAPwhoresideinTB-endemiccountrieswho aresuspectedofhavingTB,imagingstudiesaswellasaggressivemicrobiologicalinvestigationsneedto beperformedearlyon.PreviousexposuretoaFQfor>10daysinpatientswithTBisassociatedwith theemergenceofFQ-resistantM.tuberculosisisolates.However,ratesofM.tuberculosisisolateswithFQ resistancearesignificantlyhigheramongstmultidrug-resistantM.tuberculosisisolatesthanamongst sus-ceptibleisolates.Consequently,inTaiwanandalsoinothercountrieswithTBendemicity,ashort-course (5-day)regimenofaRFQisstillrecommendedforempiricaltherapyforCAPpatientsifthepatientisat lowriskforTB.

1. Introduction

Community-acquired pneumonia (CAP) is one of the lead-ingcausesofdeathworldwide.Themortalityratehasincreased

∗ Correspondingauthor.

E-mailaddress:hsporen@ntu.edu.tw(P.-R.Hsueh).

significantlyoverthepast10yearsnotonlyinTaiwanbutalsoin othercountriesintheAsia-Pacificregion[1–3].Thekeycausative pathogens of CAP are Streptococcus pneumoniae, Haemophilus influenzaeandatypicalpathogens[1–3].Accordingtothe antimi-crobialtreatmentguidelinesoftheInfectiousDiseasesSocietyof America, theAmerican ThoracicSociety, the European Respira-torySocietyandtheInfectiousDiseasesSocietyofTaiwan[1–3], thedrugsofchoiceforCAPinoutpatientsarepenicillin-related

(2)

agents if urine cultures are positive for pneumococcal antigen, andmacrolide-ortetracycline-relatedagentsifurineculturesare negativeforpathogens.However,therateofpenicillin,macrolide andtetracyclineresistanceamongstS.pneumoniaeisolatesishigh inTaiwan[4].TheantibioticoptionsforinpatientswithCAPare ␤-lactams or respiratory fluoroquinolones (RFQs) (levofloxacin, moxifloxacinandgemifloxacin).IntheIntensiveCareUnit(ICU), a␤-lactamantibioticcombinedwitha macrolideorwitha flu-oroquinolone(FQ)isappropriate[1–3].However,theincreasing resistanceofkey pathogens to␤-lactamantibiotics posesgreat challengestophysiciansinTaiwan.RFQscanbeusedinthe treat-ment of CAP in outpatients[5], inpatients and patients in the ICU.RFQshavebeenshowntohaveexcellentactivityagainstkey causativepathogensofCAPaswellasatypicalpathogens; how-ever,useofRFQsforempiricaltreatmentofCAPmightmaskthe diagnosisoftuberculosis(TB),leadingtodelayedtreatmentandFQ resistanceamongstsubsequentlyisolatedMycobacterium tubercu-losisstrains.

Thisarticle brieﬂyreviews thecommon microbial causesof CAP,theresistanceratesamongstkeypathogens,andtheproper administrationofFQsinthetreatmentofCAP.Theincidenceofand mortalityassociatedwithTBandthestatusofmultidrug-resistant M.tuberculosis(MDR-TB)inTaiwanarealsodescribed.Inaddition, wereviewthecontroversiessurroundingtheempiricaluseofFQs totreatpatientswithCAP,treatmentoptionsforpatientswitha delayedTBdiagnosis,andtheemergenceofFQresistanceamongst M.tuberculosisisolates.

2. Community-acquiredpneumonia

2.1. Aetiologyofcommunity-acquiredpneumoniainTaiwan Lauderdaleetal.collected168isolatesfrom468patientsfrom December2001toApril2002inTaiwanandfoundthatthemost commoncauseof CAPamongst adultpatientsinTaiwanwasS. pneumoniae(24%), followed byatypicalpathogens (Mycoplasma pneumoniae, Chlamydophila pneumoniae and Legionella pneu-mophila),H.inﬂuenzaeandKlebsiellapneumoniae[6].Theaetiology ofCAPwasundeterminedinca.40%ofCAPcases[6,7]. Staphylo-coccusaureuswasthecausativepathogenin2%ofCAPcases,and theoverallmortalityrateofpatientswithCAPwas8.3%[6]. 2.2. Antimicrobialsusceptibilityproﬁlesamongstrespiratory pathogens

Lin et al. foundthat amongst all S.pneumoniae strains that caused bacteraemia, only 29.2% were susceptible to penicillin, 15.1% to erythromycin, 18% to tetracycline and 33.7% to clin-damycin [8]. However, 96.4% were susceptible to cefotaxime, 97.3%tolevofloxacin and 98.4%tomoxifloxacin. Amongst non-bacteraemicstrains,only23.8%weresusceptibletopenicillin,5% toerythromycin,30%totetracyclineand30%toclindamycin. How-ever,theratesofsusceptibilityamongstS.pneumoniaeisolatesto cefotaxime,levofloxacinandmoxifloxacinwereeach100%[8].

Therateofnon-susceptibilityofS.pneumoniaetolevofloxacinin amedicalcentreinTaiwanwas1.2%in2005,peakedat4.2%in2007 andthengraduallydecreasedto3%in2010[9].Formoxifloxacin, thenon-susceptibleratewas1.3%in2005,4%in2008andthen graduallydecreasedto1%in2009and2010.Hsiehetal.alsoshowed thattheprevalenceofFQ-resistantS.pneumoniaeisolatesinTaiwan waslow,even thoughFQs arewidelyused inthat country[9]. AmongsttheFQ-non-susceptibleisolatesinthatstudy,serotype 9V(20%)wasthemostcommon,followedby19F(6.8%),23F(3.9%) and14F(1.8%)[10].Theseserotypesareallvaccine-typeS. pneumo-niae.Fig.1showstheproportionoflevofloxacinresistanceamongst

Fig.1.Proportionoflevoﬂoxacin-susceptibleStreptococcuspneumoniae isolates obtainedfrom12majorteachinghospitalsindifferentpartsofTaiwan,2010.N1–N5, ﬁvehospitalsinNorthTaiwan;M1–M2,twohospitalsincentralTaiwan;S1–S4,four hospitalsinsouthernTaiwan;andE1,onehospitalineasternTaiwan.

S.pneumoniaeisolatedfrom12majorteachinghospitalslocatedin differentpartsofTaiwanin2010.Themajority(81–100%)oftheS. pneumoniaeisolatesweresusceptibletolevoﬂoxacin[9].

ThesusceptibilityrateofH.influenzaetoamoxicillin/clavulanic acid(AMC)decreasedmarkedlyfrom95%in2002to88%in2009 inamedicalcentreinTaiwan[11].AMCshouldbeadministered withcautiontopatientswithCAP.Inaddition,Jean and Hsueh showedthattherateofextended-spectrum␤-lactamase (ESBL)-producing K.pneumoniae strainsin Taiwan was26% [4]. Wang etal.foundthattheresistanceratestoAMC,cefuroxime,cefaclor, ceftazidime, ceftriaxone and levofloxacin amongst K. pneumo-niaeisolatesassociatedwithcommunity-acquiredrespiratorytract infectionwereall≤10%[12].Amongst101community-acquired meticillin-resistant S. aureus (CA-MRSA) isolated in a medical centreinTaiwan,96%weresusceptibletolevofloxacinand mox-ifloxacin[13].

RegardingatypicalpathogensinTaiwan,theratesof suscepti-bilitytolevoﬂoxacinwerereportedtobe93.9%forM.pneumoniae, 85.7%forC.pneumoniaeand100%forL.pneumophila[14]. 2.3. Roleofrespiratoryﬂuoroquinolonesinthetreatmentof community-acquiredpneumonia

InTaiwan,patientswithCAPwhowerepreviouslyhealthyand havenotusedantibioticsinthe3monthspriortodiseaseonset arenormallygivenamacrolideordoxycyclineasoutpatient treat-ment [3]. However,the rates of non-susceptibilityto penicillin anderythromycinamongstclinicalisolatesofS.pneumoniaehave increasedmarkedlyinrecentyears[4].Therefore,cautionshouldbe exercisedbeforeadministeringmacrolidesforCAPunlessatypical pathogensarehighlysuspected.Forpatientswithco-morbidities, aRFQora␤-lactamantibioticplusmacrolideissuggested[3].For inpatientswithco-morbidities,especiallyintheICU,a␤-lactam antibioticpluseitherazithromycinoraFQissuggested[1–3].For atypicalpathogens,FQsareaseffectiveasmacrolides.Thelength ofstayinhospitalandthetimetoclinicalstabilityfavourtheuseof FQs[15].Dragoetal.showedthatthecombinationoflevoﬂoxacin withceftriaxoneproducedthehighestrateofsynergy(54%),mainly againstmacrolide-resistantisolates,whereasclarithromycin com-binedwithAMCwasshowntobeantagonisticin22%ofisolates

[16].NoantagonismwasnotedbetweenFQand␤-lactam antibi-otics[16].Theprevalenceoflevoﬂoxacin-resistantS.pneumoniae increasedmarkedlyduringtheperiod2001–2007inHongKong, especiallyamongsttheelderly[17].Themostcommonaetiologyof levoﬂoxacinresistancewassuboptimaluseofaFQinwhichsmall

(3)

Fig.2. Incidenceofandmortalityrates(per100000population)associatedwith tuberculosisinTaiwan,1979–2010.

doses(100–200mg)ofofloxacinandlevofloxacinwere adminis-teredtwoorthreetimesdaily.Accordingly,appropriatedosesof aRFQ(levofloxacin,750mg/day;moxifloxacin, 400mg/day;and gemifloxacin,320mg/day)arerecommended[1–3].

3. Tuberculosisandcommunity-acquiredpneumonia

3.1. TuberculosisinTaiwan

TheincidenceofTBwas75casesper100000populationin2002, butdecreasedto62casesper100000populationin2003becauseof thesevereacuterespiratorysyndrome(SARS)outbreak(Fig.2).In 2004,theincidencereboundedto76casesper100000population (16784newcases)[18].In2005,theTaiwanCentersforDisease Control(CDC)performeda‘StopTBProgram’totrytoreducethe incidencebyone-half.From2005to2010theTBincidencerate decreasedfrom70to57per100000populationandthemortality ratedecreasedfrom4.3to3per100000population(Fig.2).The overallnumberofpatientswhodieddecreasedfrom970in2005 to700in2010.Thelong-termtrendinTBmortalityalsodecreased from294per100000populationin1947to3per100000in2010. 3.2. Proportionoftuberculosisamongstpatientswith

community-acquiredpneumonia

Fig.3demonstratestheproportionsofTBamongstpatients ini-tiallydiagnosed as havingCAPin differentcountries. The rates rangedfrom1–3%inTaiwanto7%inIndia[6,7,19,20].Themajority (>50%)ofthosepatientswithCAPduetoTBwereofadvancedage (>65years)andhadvariousco-morbidities[6].

Fig. 3.Proportion of Mycobacterium tuberculosis as the causative agent of community-acquiredpneumonia(CAP)inseveralcountries.

3.3. Antimicrobialresistanceandmultidrugresistanceamongst Mycobacteriumtuberculosisisolates

TheTaiwanCDCreportedthattherewasamarkeddifferencein antimicrobialresistanceratesbetweenpatientswithincidentTB andpatientswithrecurrentTB[21];theratesofresistancewere, respectively,9%and18%toisoniazid,2%and10%torifampicin,2% and7%toethambutol,8%and12%tostreptomycinand14%and23% toanyﬁrst-linedrugduring2009–2010(http://www.cdc.gov.tw). Theratesofresistancetoﬁrst-linedrugstendedtodecrease dur-ing2000–2010exceptfortherateofresistancetostreptomycin, which remainedstable.TherateofincidentMDR-TBwas1%in 2010.Therateofresistancetoanydrugclassaswellasthe inci-dence of MDR-TB also showeda downward trend during that decade[22].

3.4. FluoroquinoloneresistanceamongstMycobacterium tuberculosisisolatesinTaiwan

Duringtheperiod2000–2006,theFQresistancerateamongst non-MDR-TBisolatesinTaiwanwas0.1%,thatamongstisolates frompatientswithpreviousanti-TBtreatmentwas7.9%andthat amongstMDR-TBisolateswas16.7%[23].Wangetal.evaluated FQ susceptibilityaswell asgeneticmutationsamongst isolates frompatientswhohadbeenexposedtoFQsfromJanuary2004 toDecember2005[24].Theyfoundthatmultipledrugresistance hadthestrongestcorrelationwithFQresistance(19%ofisolates)

[24].NeitherprevioususeofFQsnorthedurationofFQexposure wascorrelatedwithFQsusceptibility.Amongst theFQ-resistant isolates,35.7%hadagyrAmutation(D94GandA90V)and7.1%had agyrBmutation(N538)[24].

Amongstthe215MDR-TBisolatesfromTaiwanreportedbyYu etal., 42.8%wereresistanttooﬂoxacin[21].Therateof exten-sively drug-resistant TB (XDR-TB) was 10.3% in 2004 (12/116 isolates)and10.1%in2005(10/99isolates)[21].Accordingtothe annual reportfrom TaiwanCDC,therate ofFQ-resistant MDR-TB isolateswas29.1% (144/494)fromJuly2009toMarch2009 (http://www.cdc.gov.tw).In total,43MDR-TBisolateswerealso resistanttocapreomycin,amikacinorkanamycin.Therateof XDR-TBamongstMDR-TBisolateswas8.7%.NearlyallFQ-resistantM. tuberculosisisolateswerefoundamongstMDR-TBisolates(Fig.4)

[20–26].ThemostcommonsitesfromwhichFQ-resistantM. tuber-culosisisolateswereobtainedwerethegenitourinarytract(5.1%), pulmonarytract(1.5%)andpleuralcavity(1.0%)[22].

Fig.4. Proportionofquinoloneresistanceamongstmultidrug-resistant Mycobac-teriumtuberculosis(MDR-TB)andnon-MDR-TBisolatesinTaiwan[21–24,26].

(4)

3.5. Empiricaluseofﬂuoroquinolonesamongst community-acquiredpneumoniapatientsanddelayed tuberculosisdiagnosisandtreatment

Dooley et al. wasthe ﬁrst topoint out the possibility that empiricaltreatmentwitha FQ mightdelay thediagnosisof TB

[27].AmongstpatientswhoreceivedempiricalFQtreatment,the mediantimebetweensymptomonsetandreceiptofanti-TB med-ication was 21 days compared with a median time of 5 days amongstthosewhodidnotreceiveFQs[27,28].However,the sam-plesizeintheirstudywassmall.Ofthe17patientswhodidnot receive FQs,7 receivedanti-TB therapy and 3 patientsdid not receiveantibiotictreatment.The initialacid-fast bacillussmear waspositivein8ofthosepatients.TBwassuspectedinatleast one-halfofthepatientswhodidnot receiveFQs,soantibiotics werenot prescribed and anti-TB treatmentwas given directly. Interestingly, of those patients who received FQ monotherapy, 83%experiencedimprovementinthesymptomsofTB,and clini-calimprovementoccurredanaverageof3daysaftertheinitiation oftherapy.Otherstudieshavealsoreportedthatthedelayin ini-tiationofanti-TBmedicationwaslongeramongst patientswho receivedFQsthanamongstpatientswhodidnotreceiveFQ-based antibiotics(43.1±40.0daysvs.18.7±16.9days;P=0.04)[28–30]. Medianhealthcaredelayforpatientswhoreceivedantibioticsfor non-TBdiagnoses/indicationsprior toTBdiagnosiswas39 days versus15days (P<0.01)for patientswho hadinitiallyreceived TB therapy [26]. Notonly didadministration of a FQ result in adelayeddiagnosisofTB (median29 days),but administration ofotherantibiotics(suchas␤-lactams,macrolidesor carbapen-ems)hadasimilareffect(median31days).InthestudybyGolub etal.,only57%ofpatientsinitiallydiagnosedashavingpneumonia hadachestradiograph[31].Interestingly,34%ofthechest radio-graphssuggesteda diagnosisofTB;however,thepatientswere stillsubsequently prescribedempiricalantibiotics.More impor-tantly,bothantibioticuseandnothavingachestradiographtaken during theﬁrst healthcare visit were independently related to longerhealthcaredelaysintheoverallcohort.Mathuretal.showed that ca. 44–55% of active TB patients were given an incorrect diagnosisattheinitialpresentation,mainlybecauseof atypical radiographicmanifestations [32].A studyfromMalaysiaclearly demonstratedthat a duration ofsymptoms of>2 weeksbefore hospitaladmission [oddsratio (OR)=25.1;P<0.001), history of nightsweats(P=0.038),a chestradiographshowingupperlobe involvement(P=0.012)or cavitaryinﬁltrates (P=0.002),a total whitebloodcellcountof≤12×109_/L_on_admission_(P₌_0.029)_and

lymphopenia(P=0.040)]wassignificantlyassociatedwith culture-positivepulmonaryTB(Fig.5)[19].Inaddition,lowerlungfieldTB (LLFTB)isdifficulttodifferentiatefrompneumonia andisoften

Fig.5.Clinicalandlaboratorypredictorsofpatientswithtuberculosiswhowere initiallydiagnosedashavingcommunity-acquiredpneumonia[19].CXR,chest radiography;WBC,whitebloodcellcount.

misdiagnosedbecauseof atypicalﬁndingsonchest radiographs

[33,34].InTaiwan,ca.20%ofpulmonaryTBpatientshaveLLFTB

[33]. Multivariateanalysis conductedby Lin et al. showedthat prolongedduration ofsymptoms ≥7 days(OR=4.57; P=0.038), lackoffever >38◦C (OR=9.04;P=0.001)and theabsenceofair bronchograms(OR=12.08;P=0.007)weresigniﬁcantpredictorsof LLFTBinpatientswithLLFpneumonia[35].Acalculatedprobability of>0.36suggestsLLFTBwithasensitivityof81.8%andaspeciﬁcity of86.1%[35].

3.6. Empiricaluseofﬂuoroquinolones(FQs)amongst

community-acquiredpneumoniapatientsandtheemergenceof FQresistanceinMycobacteriumtuberculosis

Recently,Laietal.reportedthattheFQresistancerateinM. tuberculosiswasonly1.3%inTaiwanduring2005–2010[25].Prior tothattimeperiod,FQresistanceincreasedmarkedlyfrom7.7% in1995–1997to22.2% in1998–2000[26].Laiet al.foundthat approximatelytwo-thirdsofofloxacin-resistantM.tuberculosis iso-lateswereMDRand,surprisingly,theratesofFQresistancewere highestamongstadultsaged34–44years[25].Only22.2%of FQ-resistantTBisolates(8/36)weresusceptibletoallfirst-lineanti-TB agents.Theirresultsweresimilartothosereportedbyvanden Boogaard et al., who showed that the rate of FQ-resistant TB waslowinTBpatientsandwasnotrelatedtopreviousbriefFQ exposure[36].Parketal.evaluatedtheimpactofshort-term expo-suretoFQonofloxacin resistanceinhumanimmunodeficiency virus(HIV)-negativepatientswithTBandfoundthattherateof ofloxacin-resistantM.tuberculosiswaslowandthatmostcasesof ofloxacinresistancewereassociatedwithMDR-TB[37].Theyalso foundthatthefrequencyofofloxacin-resistantM.tuberculosiswas lowamongstpatientswhowereexposedtoFQsforashortperiod oftime[37].ThesefindingsfavourtheapplicationofFQsinthe regimenforCAPorTBinpatientswithshorterdiseasedurations.

4. Solutions

InTaiwan,only1–2%ofpatientswithCAPreceiveafinal diag-nosisofTB.ItisimportanttodifferentiatebetweenCAPandTBin theinitialpresentation.Ifthelesionislocatedintheupperlung field,clinicalspecimencollectionorrapidnucleicacid amplifica-tioncouldshortenthedelaytoTBdiagnosisratherthanrestrictthe empiricaltreatmentwithFQsinCAP.Ifthelesionisinthelower lobe,riskfactorsincludingadvancedage,prolongeddurationofthe lesion,lackoffeverandabsenceofairbronchogramsshouldraise thesuspicionofLLFTB.

AlthoughuseofFQsinTaiwanishigh,theincidenceofTB,the mortalityrateassociatedwithTBandtherateofdrugresistance havedecreased.TherateofFQresistanceinS.pneumoniaeislow andtherateofsusceptibilitytoFQsishighamongstH.inﬂuenzae,K. pneumoniae(includingESBL-producingstrains),atypicalpathogens andCA-MRSAinTaiwan.

Because of the high incidence of CAP caused by atypical pathogensinTaiwan,coverageofatypicalpneumoniamustbe con-sideredintheempiricaltreatmentofCAPbothinoutpatientsand inpatients.Macrolidesshouldbeusedwithcautionbecauseofhigh ratesofresistancetothatantimicrobialclassinS.pneumoniae.FQs haveagoodsynergisticeffectwithotherantimicrobialagents,with theexceptionof amacrolide combinedwitha␤-lactam, which mightshowsomeantagonistproperties.

5. Conclusions

Empirical treatment of CAP with a FQ might mask active TB, delay treatment and contribute to the development of FQ

(5)

resistance.FQresistanceinM.tuberculosisisrelatedtoFQ dura-tionandthetimingofexposure.ExposuretoaFQ for>10days andexposurefor>60daysbeforeTBdiagnosiswerebothshown tobeassociatedwithasignificantriskofdevelopingFQresistance. Consequently,inTaiwanaswellasinothercountrieswith endemic-ity ofTB, a short-course(5-day) regimenof a FQ(levofloxacin, moxifloxacinandgemifloxacin)isstillrecommendedforempirical therapyforCAPpatientsifthepatientisatlowriskforTB. Further-more,FQresistanceislesslikelytooccuramongstM.tuberculosis strainsisolatedfrompatientswithshort-termexposure(<10days) toFQ.

Funding:Nofundingsources. Competinginterests:Nonedeclared. Ethicalapproval:Notrequired.

References

[1]MandellLA,WunderinkRG,AnzuetoA,BartlettJG,CampbellGD,DeanNC,etal. InfectiousDiseasesSocietyofAmerica/AmericanThoracicSocietyconsensus guidelinesonthemanagementofcommunity-acquiredpneumoniainadults. ClinInfectDis2007;44(Suppl.2):S27–72.

[2] WoodheadM,BlasiF,EwigS,HuchonG,IevenM,OrtqvistA,etal.Guidelines forthemanagementofadultlowerrespiratorytractinfections.EurRespirJ 2005;26:1138–80.

[3]InfectiousDiseasesSocietyofTaiwan;TaiwanSocietyofPulmonaryandCritical Medicine;MedicalFoundationinMemoryofDr.Deh-LinCheng;Foundation ofProfessorWei-ChuanHsiehforInfectiousDiseasesResearchandEducation; CYLee’sResearchFoundationforPediatricInfectiousDiseasesandVaccines. GuidelinesonantimicrobialtherapyofpneumoniainadultsinTaiwan,revised 2006.JMicrobiolImmunolInfect2007;40:279–83.

[4]JeanSS,HsuehPR.AntimicrobialdrugresistanceinTaiwan.JFormosMedAssoc 2011;110:4–13.

[5]WispelweyB,SchaferKR.Fluoroquinolonesinthemanagementof community-acquired pneumonia in primary care. Expert Rev Anti Infect Ther 2010;8:1259–71.

[6]LauderdaleTL,ChangFY,BenRJ,YinHC,NiYH,TsaiJW,etal.Etiologyof com-munityacquiredpneumoniaamongadultpatientsrequiringhospitalizationin Taiwan.RespirMed2005;99:1079–86.

[7] YenMY,HuBS,ChenYS,LeeSS,LinYS,WannSR,etal.Aprospective etio-logicstudyofcommunity-acquiredpneumoniainTaiwan.JFormosMedAssoc 2005;104:724–30.

[8] LinSH,LaiCC,TanCK,LiaoWH,HsuehPR.Outcomesofhospitalizedpatients with bacteraemic and non-bacteraemic community-acquired pneumonia causedbyStreptococcuspneumoniae.EpidemiolInfect2011;139:1307–16. [9]HsiehYC,ChangLY,HuangYC,LinHC,HuangLM,HsuehPR.Circulationof

internationalclonesoflevoﬂoxacinnon-susceptibleStreptococcuspneumoniae inTaiwan.ClinMicrobiolInfect2010;16:973–8.

[10]HsiehYC,LinPY,ChiuCH,HuangYC,ChangKY,LiaoCH,etal.Nationalsurvey ofinvasivepneumococcaldiseasesinTaiwanunderpartialPCV7vaccination in2007:emergenceofserotype19Awithhighinvasivepotential.Vaccine 2009;27:5513–8.

[11]ChungKP,HuangYT,LeeLN,YuCJ,LaiCC,HsuehPR.Alarmingly decreas-ingratesofamoxicillin–clavulanatesusceptibilityamongclinicalisolatesof Haemophilusinﬂuenzaefrom2001to2009inamedicalcenterinTaiwan.J Infect2011;62:185–7.

[12]WangH,ChenM,XuY,SunH,YangQ,HuY,etal.Antimicrobial suscepti-bilityofbacterialpathogensassociatedwithcommunity-acquiredrespiratory tractinfectionsinAsia:reportfromtheCommunity-AcquiredRespiratoryTract InfectionPathogenSurveillance(CARTIPS)study,2009–2010.IntJAntimicrob Agents2011;38:376–83.

[13]ChenYH,LiuCY,LuJJ,KingCH,HsuehPR.Invitroactivityofnemonoxacin (TG-873870),anovelnon-ﬂuorinatedquinolone,againstclinicalisolatesof Staphylococcusaureus,enterococciandStreptococcuspneumoniaewith var-ious resistancephenotypes inTaiwan.J AntimicrobChemother 2009;64: 1226–9.

[14]vanRensburgDJ,PerngRP,MithaIH,BesterAJ,KasumbaJ,WuRG,etal. Efﬁ-cacyandsafetyofnemonoxacinversuslevoﬂoxacinforcommunity-acquired pneumonia.AntimicrobAgentsChemother2010;54:4098–106.

[15]GrifﬁnAT,PeyraniP,WiemkenT,ArnoldF.Macrolidesversusquinolones in Legionella pneumonia: results from the Community-Acquired Pneu-monia Organization internationalstudy. Int J Tuberc Lung Dis 2010;14: 495–9.

[16]Drago L, Nicola L, Rodighiero V, Larosa M, Mattina R, De Vecchi E. Comparativeevaluationofsynergyofcombinationsof␤-lactamswith ﬂu-oroquinolones ora macrolide in Streptococcus pneumoniae. J Antimicrob Chemother2011;66:845–9.

[17]HoPL,ChengVC,ChowKH.Decreasingprevalenceoflevoﬂoxacin-resistant StreptococcuspneumoniaeinHongKong,2001to2007.JAntimicrobChemother 2009;63:836–8.

[18] LoHY,ChouP,YangSL,LeeCY,KuoHS.TrendsintuberculosisinTaiwan, 2002–2008.JFormosMedAssoc2011;110:501–10.

[19] Liam CK,Pang YK, Poosparajah S.Pulmonary tuberculosis presenting as community-acquiredpneumonia.Respirology2006;11:786–92.

[20] SinghA.Fluoroquinolonesshouldnotbetheﬁrst-lineantibioticsto treat community-acquiredpneumoniainareasoftuberculosisendemicity.Clin InfectDis2007;45:133.

[21]YuMC,WuMH,JouR,Extensivelydrug-resistanttuberculosis,Taiwan.Emerg InfectDis2008;14:849–50.

[22] LaiCC,LiuWL,TanCK,HuangYC,ChungKP,LeeMR,etal.Differencesindrug resistanceproﬁlesofMycobacteriumtuberculosisisolatescausingpulmonary andextrapulmonarytuberculosisinamedicalcentreinTaiwan,2000–2010. IntJAntimicrobAgents2011;38:125–9.

[23]LaiCC,TanCK,HuangYT,ChouCH,HungCC,YangPC,etal.Extensively drug-resistantMycobacteriumtuberculosisduringatrendofdecreasingdrug resistancefrom2000through2006atamedicalcenterinTaiwan.ClinInfect Dis2008;47:e57–63.

[24]WangJY,LeeLN,LaiHC,WangSK,JanIS,YuCJ,etal.Fluoroquinolone resis-tanceinMycobacterium tuberculosisisolates:associatedgenetic mutations andrelationshiptoantimicrobialexposure.JAntimicrobChemother2007;59: 860–5.

[25]LaiCC,TanCK,HuangYT,LiaoCH,HsuehPR.Fluoroquinolone-resistant tuber-culosisatamedical centreinTaiwan,2005–10.JAntimicrobChemother 2011;66:2437–8.

[26] HuangTS, Kunin CM,Shin-Jung Lee S,Chen YS, TuHZ, Liu YC. Trends inﬂuoroquinoloneresistanceof Mycobacteriumtuberculosiscomplex ina Taiwanesemedical centre:1995–2003.J AntimicrobChemother2005;56: 1058–62.

[27] DooleyKE,GolubJ,GoesFS,MerzWG,SterlingTR.Empirictreatmentof community-acquiredpneumoniawithﬂuoroquinolones,anddelaysinthe treatmentoftuberculosis.ClinInfectDis2002;34:1607–12.

[28]ChenTC,LuPL,LinCY,LinWR,ChenYH.Fluoroquinolonesareassociated withdelayedtreatmentandresistanceintuberculosis:asystematicreview andmeta-analysis.IntJInfectDis2011;15:e211–6.

[29]YoonYS,LeeHJ,YoonHI,YooCG,KimYW,HanSK,etal.Impactof ﬂuo-roquinolonesonthediagnosisofpulmonarytuberculosisinitiallytreatedas bacterialpneumonia.IntJTubercLungDis2005;9:1215–9.

[30] WangJY,HsuehPR,JanIS,LeeLN,LiawYS,YangPC,etal.Empiricaltreatment withaﬂuoroquinolonedelaysthetreatmentfortuberculosisandisassociated withapoorprognosisinendemicareas.Thorax2006;61:903–8.

[31] GolubJE,BurS,CroninWA,GangeS,SterlingTR,OdenB,etal.Impactofempiric antibioticsandchestradiographondelaysinthediagnosisoftuberculosis.Int JTubercLungDis2005;9:392–7.

[32]Mathur P, Sacks L, Auten G, Sall R, Levy C, Gordin F. Delayed diagno-sisofpulmonarytuberculosisincityhospitals.ArchInternMed1994;154: 306–10.

[33]ChangSC,LeePY,PerngRP.Thevalueofroentgenographicandfiber bron-choscopicfindingsinpredictingoutcomeofadults withlower lungfield tuberculosis.ArchInternMed1991;151:1581–3.

[34]WangJY,LeeLN,HsuehPR.Factorschangingthemanifestationofpulmonary tuberculosis.IntJTubercLungDis2005;9:777–83.

[35]Lin CH, Chen TM, Chang CC, Tsai CH, Chai WH, Wen JH. Unilateral lowerlungﬁeldopacitiesonchestradiography:acomparisonofthe clin-ical manifestations of tuberculosis and pneumonia. Eur J Radiol 2011, doi:10.1016/j.ejrad.2011.03.028[Epubaheadofprint].

[36] vandenBoogaardJ,SemvuaHH,vanIngenJ,MwaigwisyaS,vanderLaan T,vanSoolingenD,etal.Lowrateofﬂuoroquinoloneresistancein Mycobac-teriumtuberculosisisolatesfromnorthernTanzania.JAntimicrobChemother 2011;66:1810–4.

[37]ParkIN,HongSB,OhYM,LimCM,LeeSD,LewWJ,etal.Impactofshort-term exposuretoﬂuoroquinolonesonoﬂoxacinresistanceinHIV-negativepatients withtuberculosis.IntJTubercLungDis2007;11:319–24.