Variables determining mortality in patients with Acinetobacter baumannii meningitis/ventriculitis treated with intrathecal colistin

(1)

ContentslistsavailableatScienceDirect

Clinical

Neurology

and

Neurosurgery

j ou rn a l h o m epa g e : w w w . e l s e v i e r . c o m / l o c a t e / c l i n e u r o

Variables

determining

mortality

in

patients

with

Acinetobacter

baumannii

meningitis/ventriculitis

treated

with

intrathecal

colistin

Bahadır

Ceylan

a

_,

_Ferhat

_Arslan

b,∗

_,

_Oguz

_Resat

_Sipahi

c

_,

_Mustafa

_Sunbul

d

_,

_Bahar

_Ormen

e

_,

˙Ismail

N. Hakyemez

f

_,

_Tuba

_Turunc

g

_,

_Yes¸

_im

_Yıldız

h

_,

_Hasan

_Karsen

i

_,

_Gul

_Karagoz

j

_,

Recep

Tekin

k

_,

_Burcu

_Hizarci

l

_,

_Vedat

_Turhan

m

_,

_Sebnem

_Senol

n

_,

_Neﬁse

_Oztoprak

o

_,

Mesut

Yılmaz

a

_,

_Kevser

_Ozdemir

p

_,

_Sinan

_Mermer

c

_,

_Omer

_F.

_Kokoglu

q

_,

_Ali

_Mert

a

a_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Faculty_of_Medicine,_Istanbul_Medipol_University,_Turkey

b_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Faculty_of_Medicine,_Istanbul_Medeniyet_University,_Goztepe,_Istanbul,_Turkey c_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Faculty_of_Medicine,_Ege_University,_Izmir,_Turkey

d_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Faculty_of_Medicine,₁₉_Mayıs_University,_Samsun,_Turkey e_Department_of_Infectious_Disease_and_Clinical_{Microbiology,}_Atatürk_Education_and_Training_Hospital,_Izmir,_Turkey

f_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Faculty_of_{Medicine,Vakıf}_Gureba_Bezmi_Alem_University,_Istanbul,_Turkey g_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Baskent_University,_Adana,_Turkey

h_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Faculty_of_Medicine,_Gazi_University,_Ankara,_Turkey i_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Faculty_of_Medicine,_Harran_University,_S¸_anlıurfa,_Turkey j_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Umraniye_Education_and_Training_Hospital,_Istanbul,_Turkey k_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Faculty_of_Medicine,_Dicle_University,_Diyarbakır,_Turkey l_Department_of_{Anesthesiology,}_Faculty_of_Medicine,_Istanbul_Medipol_University,_Istanbul,_Turkey

m_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Sultan_Abdulhamid_Education_and_Training_Hospital,_Istanbul,_Turkey n_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Faculty_of_Medicine,_Celal_Bayar_University,_Manisa,_Turkey o_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Antalya_Education_and_Training_Hospital,_Antalya,_Turkey p_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Faculty_of_Medicine,_Pamukkale_University,_Denizli,_Turkey q_Department_of_Infectious_Diseases_and_Clinical_{Microbiology,}_Faculty_of_Medicine,_Sutcu_Imam_University,_{Kahramanmaras,}_Turkey

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received15November2016

Receivedinrevisedform3December2016 Accepted13December2016

Availableonline14December2016 Keywords: Acinetobacterbaumannii Colistin Meningitis Intratechal

a

b

s

t

r

a

c

t

Aim:ToexaminethevariablesassociatedwithmortalityinpatientswithAcinetobacterbaumannii-related centralnervoussysteminfectionstreatedwithintrathecalcolistin.

Materialsandmethods:Thismulti-centreretrospectivecasecontrolstudyincludedpatientsfrom11 centresinTurkey,aswellascasesfoundduringaliteraturereview.

OnlypatientswithCNSinfectionscausedbymultidrug-resistantorextensivelydrug-resistant Acine-tobacterbaumanniitreatedwithintrathecalcolistinwereincludedinthisstudy.Thevariablesassociated withmortalityweredeterminedbydividingthepatientsintogroupswhodiedorsurvivedduring hos-pitalisation,andwhodiedorsurvivedfromAcinetobactermeningitis.

Results:Amongthe77casesenrolledinthestudy,35werefoundthroughaliteraturereviewand42were casesfromourcentres.Forty-fourcases(57.1%)weremaleandthemedianagewas48years(range:20–78 years).Thirty-sevenpatients(48%)diedduringhospitalisation.Thevariablesassociatedwithincreased all-causemortalityduringhospitalisationincludedoldage(oddsratio,1.035;95%confidenceinterval (CI),1.004–1.067;p=0.026)andfailuretoprovidecerebrospinalfluidsterilisation(oddsratio,0.264;95% confidenceinterval,0.097–0.724;p=0.01).Thereisatrend(P=0.062)towardshighermortalitywith usingofmeropenemduringmeningitistreatment.Fifteencases(19%)diedfrommeningitis.Therewere nosignificantpredictorsofmeningitis-relatedmortality.

Conclusions:Themortalityrateforcentralnervoussysteminfectionscausedbymultidrug-resistantor extensivelydrug-resistantAcinetobacterbaumanniiishigh.OldageandfailuretoprovideCSFsterilisation areassociatedwithincreasedmortalityduringhospitalisation.

∗ Correspondingauthor.

E-mailaddress:ferhatarslandr@hotmail.com(F.Arslan).

1. Introduction

Acinetobacterbaumanniiisanon-entericGram-negativebacillus withlowvirulencecharacteristics.Astheprevalenceofnosocomial http://dx.doi.org/10.1016/j.clineuro.2016.12.006

(2)

44 B.Ceylanetal./ClinicalNeurologyandNeurosurgery153(2017)43–49 infections caused by this bacterium is increasing, it is

becom-inganimportantconcern[1].Thepatientsthatmostcommonly developcentralnervoussystem(CNS)infectionsassociatedwithA. baumanniiarethosethathaveundergonebrainsurgeryand exter-nalventriculardrainageviaacatheter[2].Comparedwithother bacteria,theabilityofA.baumanniitodevelopresistanceagainst antibioticsthatareincurrentuseisquitehigh;thus,thisbacterium ranking among the most frequent causes of hospital-acquired infections[3].Onlyalimitednumberofantibioticsareeffective againstthisbacterium,andnonewantibioticsareexpectedtobe introducedintheshortterm.Inrecentyears,colistin,whichwas previouslyabandonedduetoitstoxiceffects,startedbeingused againforresistantA.baumanniiinfections[4].Theclinical effective-nessofcolistinagainstvariousinfections,includingCNSinfections, isnotknownsinceithasnotbeensubjecttodrugdevelopment tri-alsandhasnotbeenevaluatedincomparativeclinicalstudies.Ithas beensuggestedthatintravenouscolistindoesnotpenetratewell intothecerebrospinalﬂuid(CSF)[5,6].Therefore, intrathecalor intraventricularadministrationofcolistinhasbecomean increas-inglycommonmethodfor thetreatmentofmultidrug-resistant (MDR)or extensivelydrug-resistant(XDR)A. baumannii-related CNSinfections[7–17].

However,ourknowledgeconcerningthesemethodsof admin-isteringcolistinislimitedduetoalackofrandomisedcontrolled studiesonthesubject,andbythefactthattheavailablestudies arecase reportsandcaseseriesincludinglownumbersofcases [7–17].Theaimofthisstudywastoexaminethevariables associ-atedwithmortalityinA.baumannii-relatedCNSinfectionstreated withintrathecalcolistin.

2. Materialsandmethods

This multi-centre retrospective case control study included patientsdrawnfrom11centresinTurkey.Informationfrom elec-tronicpatientfilesbetween2001and 2015wasreviewed,with eligiblepatientsincludedonthat basis. The literaturewasalso reviewed to identify cases suitable for inclusion in this study. PatientstreatedwithintrathecalcolistinduetoMDRor XDRA. baumannii-relatedCNSinfectionswereincluded.Presenceof noso-comialmeningitis/ventriculitiswasdeterminedinaccordancewith thedefinitionoftheCenterforDiseaseControl[18].Accordingto thisdefinition,fulfilmentofoneofthefollowingcriteria,inapatient withafeverabove38◦CandanA.baumanniiinfectionidentified fromtheCSF,isconsideredaCNSinfection:(1)numberof leuko-cytesinCSF>10/mm3_,₍₂₎_CSF_protein_levels_>₄₅_mg/dL,_or₍₃₎_CSF

glucoselevel<40mg/dL.MDRandXDRA.baumanniiweredeﬁned inaccordancewiththeliterature[19].CaseswithCNSinfections causedby Acinetobacter thatwere susceptibletocefepime, cef-tazidime,carbapenem,andtrimethoprim-sulfamethoxazole,allof whichhavegoodCSFpenetrationandcanbeusedtotreat meningi-tis,aswellascaseswithCNSinfectionscausedbynon-Acinetobacter Gram-negativebacilli,wereexcludedfromthestudy.Thefollowing datawereextrapolatedfromthepatients’ﬁles:general characteris-tics(age,gender,andunderlyingdiseases),sourceofinfection,CSF pre-treatmentcharacteristics(numberandtypeofcells,and glu-coseandproteinlevels),bloodcultureresults,doseandduration ofintrathecalcolistin,doseanddurationofintravenouscolistin, antibioticsusedbeforeandconcomitantwithintrathecalcolistin, durationofstayintheintensivecareunit,timefromhospitalisation untildevelopmentofmeningitis,reasonofdeath,timefrom initia-tionofmeningitistreatmentuntildeath,whetherCSFsterilisation wasachieved,andtimeuntilCSFsterilisation.

To identify cases in the literature withCNS MDR and XDR A.baumanniiinfections, acomprehensivesearchofthePubMed databasewasperformed, of papers published since 2000.Only

Englishlanguagepublicationsweresearchedfor,usingthe follow-ing terms: multidrug-resistant,extensively drug-resistant, CNS, A.baumannii,ventriculitis,intrathecal,meningitis, intraventricu-lar,colistimethate,colistin,review,casereport,andmortality.The referencesofeacharticlewerereviewedtopreventinclusionof duplicatecases.Intotal,11articleswerefound,whichincluded35 caseseligibleforinclusion.

The variables associated with mortality were identiﬁed by dividing the patients into groups who died or survived dur-inghospitalisation,andwhodiedorsurvivedfromAcinetobacter meningitis.Toensurethatdeathswererelatedtomeningitis,allof thefollowingcriteriahadtobepresent[20]:(1)lackof improve-mentof clinicalandlaboratoryindicatorsofinﬂammationafter treatment,(2)lackoftwoconsecutivenegativeCSFcultures,and(3) nodisorderapartfrommeningitisthatcouldleadtodeath.Unless twoconsecutiveCSFcultureswereobtainedfromacase, discontin-uationofmeningitis-relatedtreatmentbeforedeathwasincluded amongtheinclusioncriteria.

2.1. Statisticalanalysis

TheSPSS softwarepackage (ver. 17.0;SPSSInc., Chicago,IL, USA) was used for all statistical analyses.Categorical variables wereexpressedasa numberand percentageand non-normally distributed constant variables were expressed as the median (minimum-maximum). The Kaplan-Meier method and uni- or multi-variableCoxregressionanalysiswereusedtodeterminethe independentvariablesassociatedwithmortality.APvalue<0.05 wasconsideredtoindicatestatisticalsigniﬁcance(Fig.1). 3. Results

3.1. Casecharacteristicsandriskfactors

Seventy-sevenmeningitiscases,causedbyXDRorMDR Acine-tobacterandtreatedwithintrathecalcolistin,wereincludedinthis study.Intotal,35casesfromtheliteratureand42casesfromour centreswereincluded.Forty-fourpatients(57.1%)weremaleand themedianagewas48years(range:20–78years).

ThesourceofCNSinfectionwasdeterminedtobe intraventric-ularcatheters(35,45.5%),lumbarcatheters(9,11.7%),openhead trauma(2,2.6%),neurosurgicalintervention(30,39%) (Ventricu-loperitonealshuntoperationin3cases),andhematogenousspread duringA.baumanniisepsis(1,1.2%).A.baumanniiwasobservedin thebloodof12(15.5%)patients.

3.2. CSFexaminationresults

ThemediannumberofleukocytesinCSFwas2739/mm3_(range:

10–16,000/mm3_),_the_median_{polymorphonuclear}_leucocytes_ratio

was86%(range:44–100%),themedianglucoselevelwas26mg/dL (range: 1–122mg/dL)mg/dL, and the median proteinlevel was 510mg/dL(range:53–3373mg/dL).

3.3. Informationregardingtreatment

Priortodevelopingmeningitis,32(47.1%),27(39.7%),7(10.3%), 6(8.9%), 4(5.4%) and 32(47%) patientsweretreated with car-bapenems(meropenemandimipenem),cephalosporins(cefazolin, ceftriaxone,cefepime,ceftazidime,andcefuroxime), piperacillin-tazobactam,quinolones (ciproﬂoxacinand levoﬂoxacin), amino-glycosides(amikacinandgentamicin),anddrugseffectiveagainst Gram-positive cocci (vancomycin, teicoplanin, and linezolid), respectively.

A median of 10mg/day (range: 3–40mg/day) of intrathecal colistin wasused for a duration of 16days (range: 2–47days).

(3)

Fig.1. Kaplan–Meiercurveoftheeffectofprovidingcerebrospinalﬂuidsterilisationonmortality.

Sixty-three(81.8%) casesweretreated with418mg/day (range: 240–720mg/day) of intravenous colistin for 20days (range: 2–47days)inadditiontointrathecalcolistintreatment. Intrathe-calcolistintreatmentwasadministeredusingintraventricularand lumbar cathetersin71 (92.2%) and 6(7.8%) cases,respectively. PatientswithA.baumanniibacteraemiaalsoreceivedintravenous colistin.

3.4. Outcome

Twenty-seven(64.3%)patientsfromourcentres,andten(28.6%) patientsfromthe literature,died (P=0.002). Ofall patients, 37 (48%) died during hospitalisation (Table 1). Univariateanalysis revealedthatpatientswhodiedduringhospitalisationwereolder, had a lower CSF sterilisation rate, and a higher rate of use of meropenemwithintrathecalcolistin(Table2).Inamultivariate analysis,olderageand failuretoprovideCSF sterilisationwere independentlyassociatedwithmortality(Table2).Althoughnot statisticallysigniﬁcant,mortalitywashighin patientsreceiving meropenemduringmeningitistreatment(p=0.062).Fifteencases (19.2%)diedduetomeningitis.Novariableswereassociatedwith meningitis-relatedmortality(Table3).

4. Discussion

A.baumanniiisasanopportunisticmicroorganismknownto causehealth-relatedinfections[21].XDRandMDRA. baumannii-related meningitis has been increasingly reported in recent years,particularlyinpatientswhohad undergonebrainsurgery andweretreatedwithintraventricularorintralumbarcatheters [7–16,22,23].Asthesepatientsgenerallystayintheintensivecare

unitduetosevereunderlyingdiseases,andreceivebroad-spectrum antibioticsforvariousinfections,theyareatriskofdevelopingXDR andMDRA.baumannii-relatedCNSinfections.Inapreviousstudy, theuseofcephalosporins,carbapenems,penicillin,quinolones,and aminoglycosidesforvariousinfectionswasreportedin63%,44%, 31%,31%,and25%ofpatients,respectively,priortothe develop-mentofA.baumannii-relatedCNSinfections[8].Consistentwith theliterature,therateofuseofcarbapenemsandcephalosporins foranyinfectionpriortoCNSinfectionwashighinourstudy.

Onlytwostudieshavepreviouslyreportedall-causemortality ratesinpatientswithMDRorXDRA.baumannii-related meningi-tis,of13%and16%,respectively[7,8].Mortalitywashigheramong thecasesthatweincludedfromourcountrycomparedwiththose obtainedfromtheliterature.Inapreviousstudyon Acinetobacter-related meningitis, an APACHE II score>19, use of intrathecal colistin,oldage,highnumberofleukocytesintheCSF,and non-removalofaforeignbody(intraventricularcatheter,intralumbar catheter, orventriculoperitoneal shunt) fromtheCNSwere the variablesassociatedwithall-causemortality[7,8,24].Inourstudy, theAPACHEIIinstrumentwasnotused,butoldage,whichisone ofthemostimportantvariablesintheAPACHEIIscoringsystem, wasanindependentvariableassociatedwithmortality.Duetothe retrospectivenatureofourstudy,wewereunabletoexaminethe effectsofearlyremovalofforeignbodiesfromtheCNS.

Colistin and carbapenemcombinationtherapy didnot affect survivalinpatientswithA.baumanniiinfectionssusceptibleto col-istinonly[25].However,apreviousinvitrostudysuggestedthat colistin-sulbactam,colistin-meropenem,and colistin-meropenem-sulbactam combinations had synergistic effects when time-kill curves wereexaminedfor a case that developed A. baumannii-related meningitissusceptible tocolistinonly [9].In ourstudy,

(4)

46 B.Ceylanetal./ClinicalNeurologyandNeurosurgery153(2017)43–49

Table1

Characteristicsofpatientswithmeningitiscausedbymultidrug-resistantAcinetobacterbaumannii.

Survival(n=40,52%) Non-survival(n=37,48%)

Age(years) 42.5(20–69) 53(23–78)

Gender(males,n) 23(57.5) 21(56.8)

Sourceofinfection(n)

Ventriculardrainagecatheter 17(42.5) 18(48.6)

Lumbarcatheter 7(17.5) 2(5.4)

Openheadtrauma 2(5) 0(0)

Neurosurgicalintervention 13(32.5) 17(45.9) Sepsis 1(2.5) 0(0) UnderlyingCNSdisease(n) Cerebrovasculardisease 17(42.5) 10(27) CNStumour 6(15) 9(24.3) Hydrocephalus 1(2.5) 0(0) Arteriovenousmalformation 0(0) 2(5.4)

Numberofpre-treatmentCSFleukocytes/mm3 ₂₇₃₉_(10–16000) ₂₅₀₀_(90–8900)

Pre-treatmentCSFneutrophils(%) 86(44–100) 86(70–100)

Pre-treatmentCSFglucoselevel(mg/dL) 26(2–107) 26(1–112)

Pre-treatmentCSFproteinlevel(mg/dL) 515(67–2362) 505(53–3373)

PresenceofconcomitantAcinetobacterbaumanniibacteraemia 5(14.3) 7(21.2)

Intrathecalcolistindose(mg/day) 10(3–40) 10(5–40)

Intrathecalcolistintreatmentduration(days) 19(3–40) 14(2–47)

Intravenouscolistindose(mg/day) 418(240–720) 410(300–720)

Intravenouscolistintreatmentduration(days) 21(7–40) 20(2–47)

Concomitantparenteralantibiotictreatmentformeningitis(n)

Tigecycline 3(7.5) 3(8.1)

Sulbactam 9(25.7) 6(18.2)

Meropenem 7(20) 21(63.6)

Rifampicin 2(5.7) 3(9.1)

AchievementofCSFsterilisation(n) 31(100) 22(73.3)

TimefromintrathecalcolistintreatmentinitiationtoCSFsterilisation(days) 5(1–48) 6(1–32)

Treatmentinintensivecareunit(n) 37(92.5) 35(94.6)

CNS,centralnervoussystem;CSF,cerebrospinalﬂuid.

Table2

Resultsofuni-andmultivariateCox-regressionanalysesofvariablesassociatedwithall-causemortalityinpatientswithmultidrug-resistantAcinetobacterbaumannii-related

infectionstreatedwithintrathecalcolistin.

Univariateanalysis Multivariateanalysis

Oddsratio 95%conﬁdenceinterval P-value Oddsratio 95%conﬁdenceinterval P-value

Age 1.026 1.005–1.049 0.017 1.035 1.004–1.067 0.026

CSFsterilisationachievementwithtreatment 0.121 0.05–0.290 0.0001 0.264 0.097–0.724 0.010

Useofmeropenemduringmeningitistreatment 2.404 1.167–4.951 0.017 2.387 0.956–5.964 0.062

CSF,cerebrospinalﬂuid.

thecombineduseofmeropenemorsulbactamand colistinhad

no effect on the mortality rate of our patients. This might be attributabletothefactthatthesynergisticeffectsofthese antibi-oticsinvitroarelesspronouncedinclinicalpracticecomparedto theeffectsofotherfactors.Moreover,aCSFpenetrationratioaslow as1–33%mayalsoresultinalackofeffectofsulbactamon mortal-ity[1,26].Thereisatrendtowardshighermortalitywithusingof meropenemduringmeningitistreatmentinourStudy.Thismaybe duetothattheclinicianmayhavepreferredtousemeropenemin criticallyillpatients.Aninvitroandaninvivoanimalstudy inves-tigatingrifampicinuseincombinationwithcolistinsuggesteda synergisticactionagainstAcinetobacter-relatedinfections[27,28]. However,studiesperformedoncaseswithventilator-associated pneumoniacausedbyA.baumanniiandsepsisyieldedcontroversial resultsonthesynergisticeffectsofcolistin-rifampicin combina-tions[29].Theuseof rifampicinappeared tohavenoeffecton mortalityinourstudy,whichmightbeattributedtothesmall num-berofpatientstreated withthis agent.Insomeofourcases,A. baumanniiwassusceptibletotigecycline,forwhichitwas admin-isteredincombination withcolistininsomepatients.However, theuseoftigecyclineincombinationwithcolistindidnothavean effectonmortality,whichweattributetotheweakdispersionof tigecyclineinCSF.

AccordingtotheInfectiousDiseaseSocietyofAmerica guide-lines, intraventricularcolistin should be administered at a dose of 10mg/day for 21days inmeningitis cases caused byaerobic Gram-negative bacilli [30]. In the literature, no standard dose hasbeenestablishedforresistantA.baumannii-related meningi-tis;themediandosagerangeis12mg/day(range:3–40mg/day) [9–15,22,23]. A previous study including few patients demon-stratedthatthedoseofintrathecalcolistindidnotaffectmortality [8].However,norandomisedcontrolledstudieshaveassessedthe effectsofintrathecalcolistindoseonmortalityandthedosewas notassociatedwithmortalityinourstudy.

Variousintrathecalcolistintreatmentdurationshavepreviously beenreportedforAcinetobacterrelatedmeningitis[9–15,22,23].In ourstudy,wewereunabletoinvestigatetheeffectsoftreatment durationonmortality,becauseinmanycasesthetreatmentperiod wascutshortduetopatientdeath.WeshowedthatA. bauman-niibacteraemiaconcomitantwithAcinetobacter-relatedmeningitis hadnoeffectonmortalityrate,which mightbedue tothefact thatalloftheAcinetobacterbacteraemiacasesreceivedintravenous colistin.However,noconclusionscanbedrawnregardingthis asso-ciation,asonlyasmallnumberofpatientshadbacteraemia.Inour study,CSFsterilisationwasachievedinallsurvivingpatients,but onlyin73.3%ofthosewhodied.Theseresultssuggestthatalack ofCSFsterilisationwithtreatmentmayleadtoincreased

(5)

mortal-B. Ceylan et al. / Clinical Neurology and Neurosurgery 153 (2017) 43–49 47

Characteristicsofpatientswithmultidrug-resistantAcinetobacter-relatedmeningitistreatedwithintrathecalcolistin. Gender Age (years) CSFa sterilisa-tiontime (days) Underlying disease Causeof meningitis Timebetween hospitalisation andmeningitis development

Antibioticsused1month priortomeningitis treatment Intrathecal colistindose (mg/day) Intrathecal colistin treatment duration(days) Intravenous colistindose (mg/day) Intravenous colistin treatment duration(days)

Causeofdeath Timebetween meningitis treatment initiationand death(days) Durationof stayin intensivecare unit Susceptibility ofbacteriumto antibiotics otherthan colistin ConcomitantA. baumannii bacteraemia Case1 Mb 21 7 ,0 Lumbar catheter 4 Ampicillin/sulbactam, Tigecycline,Colistin 10 14 14 300 – – 0 – –

Case2 Mb 23 None ,0 Brainsurgery 11 N 10 28 14 450 – – 0 – –

Case3 Fc 72 None SVDd Ventricular

catheter

2 N 10 9 9 450 Meningitis 9 0 – –

Case4 Mb 24 None ,0 Brainsurgery 15 Meropenem,

Vancomycin 10 21 21 300 – – 1 Gentamicin, Tigecycline – Case5 Mb 54 8 ,0 Ventricular catheter 6 Ceftriaxone 10 11 18 450 – – 23 –

Case6 Fc 24 3 Braintumor Ventricular

catheter 81 Meropenem, Ciproﬂoxacin,Linezolid 10 21 21 450 – – 135 Gentamicin, Amikacin Yes

Case7 Mb 62 None SVDd Brainsurgery 24 Piperacillin/tazobactam,

Moxiﬂoxacin, Clarithromycin

10 14 26 300 – – 79 –

Case8 Fc 51 9 ,0 Brainsurgery 9 Ceftriaxone,

Meropenem,Linezolid

10 20 7 300 – – 3 Levoﬂoxacin,

Tigecycline –

Case9 Mb 26 16 ,0 Brainsurgery 14 Rifampicin,Linezolid 10 28 28 300 – – 50 Gentamicin –

Case10 Mb 55 12 ,0 Brainsurgery 35 Meropenem,

Cefopera-zone/sulbactam, Tigecycline,Linezolid, Amikacin 10 19 30 450 – – 51 Gentamicin, Tigecycline –

Case11 Mb 23 None ,0 Headtrauma 0 Ceftriaxone,

Metronidazole, Meropenem, Vancomycin 10 10 14 450 – – 14 Amikacin, Tigecycline Yes Case12 Fc 47 1 ,0 Ventricular catheter 16 N 10 12 – – – 37 Gentamicin –

Case13 Mb 24 None Braintumor Brainsurgery 30 Linezolid,

Amphotericin-B, Colistin,Caspofungin, Daptomycin

10 21 21 300 – – 50 – –

Case14 Fc 58 None SVDd Brainsurgery 19 Ceftriaxone,

Vancomycin,tigecycline, Colistin

10 25 33 300 – – Gentamicin,

Tigecycline –

Case15 Mb 31 3 ,0 Brainsurgery 7 Meropenem 10 21 – – – Tigecycline Yes

Case16 Mb 43 None SVDd Brainsurgery 2 N 10 16 24 300 ,- – Gentamicin,

Tigecycline –

Case17 Fc 47 None ,0 Ventricular

catheter

vankomycin, Meropenem,Cefepime, Fluconazole,Linezolid

10 47 47 300 Meningitis 47 60 Gentamicin –

Case18 Fc 65 6 Braintumor Lumbar

catheter 29 Ceftazidime, Vancomycin, Meropenem 10 11 11 450 Otherdisease outof meningitis 11 37 Gentamicin, Amikacin –

Case19 Mb 73 None ,0 Ventricular

catheter

9 Cefazolin,

Piperacillin/tazobactam

10 2 2 450 Meningitis 2 14 Amikacin –

Case20 Mb 33 4 ,0 Brainsurgery 41 Doripenem,Imipenem,

Colistin

10 8 26 300 Meningitis 9 66 Gentamicin,

Tigecycline Yes

catheter

18 Meropenem,

Vancomycin

(6)

48 B. Ceylan et al. / Clinical Neurology and Neurosurgery 153 (2017) 43–49 Table3(Continued) Gender Age (years) CSFa sterilisa-tiontime (days) Underlying disease Causeof meningitis Timebetween hospitalisation andmeningitis development

Antibioticsused1month priortomeningitis treatment Intrathecal colistindose (mg/day) Intrathecal colistin treatment duration(days) Intravenous colistindose (mg/day) Intravenous colistin treatment duration(days)

Causeofdeath Timebetween meningitis treatment initiationand death(days) Durationof stayin intensivecare unit Susceptibility ofbacteriumto antibiotics otherthan colistin ConcomitantA. baumannii bacteraemia Case22 Fc 58 15 ,0 Ventricular catheter 16 Cefazolin,Meropenem, Vancomycin 10 7 31 450 Otherdisease outof meningitis 31 25 Gentamicin, Amikacin – Case23 Fc 28 3 ,0 Ventricular catheter

12 Cefazol,Ceftriaxone 10 10 13 450 Meningitis 13 24 Gentamicin,

Amikacin –

Case24 Mb 60 3 SVDd Brainsurgery 17 Meropenem 10 10 6 300 Meningitis 10 1 – –

Case25 Fc 55 5 Braintumor Brainsurgery 28 Meropenem,

Ceftriaxone,Tigecycline

10 6 16 450 Otherdisease

outof meningitis

16 36 Tigecycline –

Case26 Mb 62 8 ,0 Brainsurgery 48 Meropenem,

Trimetho-prim/sulfamethoxazole, Ciproﬂoxacin, Vancomycin,Tigecycline 10 25 25 450 Otherdisease outof meningitis 28 95 Amikacin, Levoﬂoxacin, Tigecycline –

catheter

8 Cefazol,Meropenem,

Vancomycin

10 3 3 450 Meningitis 3 5 Amikacin –

Case28 Mb 23 2 Braintumor Brainsurgery 28 Piperacillin/tazobactam,

Linezolid,Meropenem

outof meningitis

45 67 – –

catheter 14 Ceftriaxone, Meropenem, Vancomycin,Colistin 10 21 21 300 Otherdisease outof meningitis 53 50 Tigecycline –

Case30 Fc 63 None Braintumor Brainsurgery 16 Vancomycin,

Ceftazidime,Tigecycline, Colistin,Linezolid, Rifampicin

10 13 14 300 Meningitis 14 31 – Yes

Case31 Fc 52 None ,0 Brainsurgery 12 Cefazol,Cefepime 10 4 4 450 Meningitis 4 21 Amikacin Yes

catheter

14 Ceftriaxone, Cefopera-zone/sulbactam, Vancomycin, Meropenem,Linezolid

10 28 14 300 Meningitis 46 61 Tigecycline Yes

Case33 Mb 30 None Braintumor Brainsurgery 2 Meropenem,

Vancomycin

outof meningitis

150 – –

Case34 Mb 42 3 SVDd Brainsurgery 9 Meropenem 10 30 20 450 Otherdisease

outof meningitis

30 44 – –

Case35 Mb 26 None SVDd Brainsurgery 12 Meropenem,

Vancomycin

10 20 28 450 Meningitis 28 32 – –

Case36 Mb 50 None ,0 Brainsurgery 21 Cefuroxime,Ceftriaxone,

Ciproﬂoxacin, Meropenem,Colistin

10 3 15 450 Meningitis 14 30 – Yes

Case37 Mb 62 4 ,0 Brainsurgery 12 Meropenem,Linezolid,

Vancomycin, Cefoperazone/sulbactam 10 16 19 450 Otherdisease outof meningitis 65 92 Amikacin, Tigecycline – Case38 Fc 58 15 ,0 Ventricular catheter

25 Meropenem,Linezolid 10 32 28 300 Otherdisease

outof meningitis

36 56 – –

Case39 Mb 53 None Brainsurgery 18 Ceftriaxone,

Meropenem, Vancomycin 10 3 27 450 Otherdisease outof meningitis 41 34 Gentamicin –

Case40 Fc 53 None SVDd Ventricular

catheter

5 10 16 22 300 Otherdisease

outof meningitis

33 40 Tigecycline Yes

Olgu41 Mb 28 6 Braintumor Brainsurgery 29 Piperacillin/tazobactam,

Meropenem, Vancomycin 10 15 15 450 Otherdisease outof meningitis 27 60 Gentamicin, Amikacin –

Case42 Fc 48 1 Braintumor Brainsurgery 8 Piperacillin/tazobactam,

Meropenem,Linezolid 10 17 32 300 Otherdisease outof meningitis 89 99 Tigecycline – a_{Cerebrospinal}_ﬂuid. b _Male. c _Female. d _{Serebrovascular}_diseases.

(7)

ity.However,culturesofCSFsamplestakenundertheantibiotic pressurewouldnotindicateamicrobiologicalcure.Ifwetakeodds ratiosinmultivariateanalysisintoaccount,CSFsterilizationseems tohavemoreeffectonmortalitythanage.

In conclusion, we demonstrated that MDR and XDR A. baumannii-relatedCNSinfectionstreatedwithintrathecalcolistin are associated with a high risk of mortality, which is further increasedbyoldageandfailuretoachieveCSFsterilisation. Acknowledgements

WethankProf.OguzResatSipahiforassistancewithcollecting relevantpatientsﬁlesfromournationalinstitutesandProf.AliMert forcommentsthatgreatlyimprovedthemanuscript.

References

[1]H.Giamarellou,A.Antoniadou,K.Kanellakopoulou,Acinetobacterbaumannii: auniversalthreattopublichealth?Int.J.Antimicrob.Agents32(August(2)) (2008)106–119.

[2]S.Navon-Venezia,R.Ben-Ami,Y.Carmeli,UpdateonPseudomonas aeruginosaandA.baumanniiinfectionsinthehealthcaresetting,Curr.Opin. Infect.Dis.18(August(4))(2005)306–313.

[3]P.-E.Fournier,D.Vallenet,V.Barbe,S.Audic,H.Ogata,L.Poirel,etal., ComparativegenomicsofmultidrugresistanceinAcinetobacterbaumannii, PLoSGenet.2(January(1))(2006)e7.

[4]H.Giamarellou,G.Poulakou,Multidrug-resistantGram-negativeinfections: whatarethetreatmentoptions?Drugs69(Octobert(14))(2009)1879–1901. [5]S.L.Markantonis,N.Markou,M.Fousteri,N.Sakellaridis,S.Karatzas,I.

Alamanos,etal.,Penetrationofcolistinintocerebrospinalﬂuid,Antimicrob. AgentsChemother.53(November(11))(2009)4907–4910.

[6]C.Antachopoulos,M.Karvanen,E.Iosiﬁdis,B.Jansson,D.Plachouras,O.Cars, etal.,Serumandcerebrospinalﬂuidlevelsofcolistininpediatricpatients, Antimicrob.AgentsChemother.54(September(9))(2010)3985–3987. [7]G.Fotakopoulos,D.Makris,M.Chatzi,E.Tsimitrea,E.Zakynthinos,K.Fountas,

Outcomesinmeningitis/ventriculitistreatedwithintravenousor

intraventricularplusintravenouscolistin,ActaNeurochir.(Wien)158(March (3))(2016)603–610,discussion610.

[8]T.Khawcharoenporn,A.Apisarnthanarak,L.M.Mundy,Intrathecalcolistinfor drug-resistantA.baumanniicentralnervoussysteminfection:acaseseries andsystematicreview,Clin.Microbiol.Infect.16(July(7))(2010)888–894. [9]C.-H.Lee,Y.-F.Tang,L.-H.Su,C.-C.Chien,J.-W.Liu,Antimicrobialeffectsof

variedcombinationsofmeropenem,sulbactam,andcolistinona

multidrug-resistantAcinetobacterbaumanniiisolatethatcausedmeningitis andbacteremia,Microb.DrugResist.14(September(3))(2008)233–237. [10]J.Ng,I.B.Gosbell,J.A.Kelly,M.J.Boyle,J.K.Ferguson,Cureofmultiresistant

Acinetobacterbaumanniicentralnervoussysteminfectionswith intraventricularorintrathecalcolistin:caseseriesandliteraturereview,J. Antimicrob.Chemother.58(November(5))(2006)1078–1081.

[11]N.AlShirawi,Z.A.Memish,A.Cherfan,A.AlShimemeri,Post-neurosurgical meningitisduetomultidrug-resistantAcinetobacterbaumaniitreatedwith intrathecalcolistin:casereportandreviewoftheliterature,J.Chemother. (FlorenceItaly)18(October(5))(2006)554–558.

[12]E.Paramythiotou,D.Karakitsos,H.Aggelopoulou,P.Sioutos,G.Samonis,A. Karabinis,Post-surgicalmeningitisduetomultiresistantAcinetobacter baumannii.Effectivetreatmentwithintravenousand/orintraventricular colistinandtherapeuticdilemmas,Med.MalInfect.37(February(2))(2007) 124–125.

[13]A.Cascio,A.Conti,L.Sinardi,C.Iaria,F.F.Angileri,G.Stassi,etal.,

Post-neurosurgicalmultidrug-resistantAcinetobacterbaumanniimeningitis successfullytreatedwithintrathecalcolistin.Anewcaseandasystematic reviewoftheliterature,Int.J.Infect.Dis.14(July(7))(2010)e572–e579.

[14]M.Hoenigl,M.Drescher,G.Feierl,T.Valentin,G.Zarfel,K.Seeber,etal., Successfulmanagementofnosocomialventriculitisandmeningitiscausedby extensivelydrug-resistantAinetobacterbaumanniiinAustria,Can.J.Infect. Dis.Med.Microbiol.24(3)(2013)e88–90.

[15]I.Karaiskos,L.Galani,F.Baziaka,E.Katsouda,I.Ioannidis,A.Andreou,etal., Successfultreatmentofextensivelydrug-resistantAcinetobacterbaumannii ventriculitisandmeningitiswithintraventricularcolistinafterapplicationof aloadingdose:acaseseries,Int.J.Antimicrob.Agents41(May(5))(2013) 480–483.

[16]R.Dersch,E.Robinson,L.Beume,S.Rauer,W.-D.Niesen,Fullremissionina patientwithcatheter-associatedventriculitisduetoAcinetobacterbaumannii treatedwithintrathecalandintravenouscolistinbesidescoinfectionswith othermultidrug-resistantbacteria,Neurol.Sci.36(April(4))(2015)633–634. [17]B.López-Alvarez,R.Martín-Láez,M.C.Fari ˜nas,B.Paternina-Vidal,J.D.

García-Palomo,A.Vázquez-Barquero,Multidrug-resistantAcinetobacter baumanniiventriculitis:successfultreatmentwithintraventricularcolistin, ActaNeurochir.(Wien)151(Nov(11))(2009)1465–1472.

[18]A.P.Lozier,R.R.Sciacca,M.F.Romagnoli,E.S.Connolly,

Ventriculostomy-relatedinfections:acriticalreviewoftheliterature, Neurosurgery62(February(Suppl.2))(2008)688–700.

[19]A.-P.Magiorakos,A.Srinivasan,R.B.Carey,Y.Carmeli,M.E.Falagas,C.G.Giske, etal.,Multidrug-resistant,extensivelydrug-resistantandpandrug-resistant bacteria:aninternationalexpertproposalforinterimstandarddeﬁnitionsfor acquiredresistance,Clin.Microbiol.Infect.18(March(3))(2012)268–281. [20]T.Tängdén,P.Enblad,M.Ullberg,J.Sjölin,NeurosurgicalGram-negative

bacillaryventriculitisandmeningitis:aretrospectivestudyevaluatingthe efﬁcacyofintraventriculargentamicintherapyin31consecutivecases,Clin. Infect.Dis.52(June(11))(2011)1310–1316.

[21]L.L.Maragakis,T.M.Perl,Acinetobacterbaumanniiepidemiology,

antimicrobialresistance,andtreatmentoptions,Clin.Infect.Dis.46(April(8)) (2008)1254–1263.

[22]B.López-Alvarez,R.Martín-Láez,M.C.Fari ˜nas,B.Paternina-Vidal,J.D. García-Palomo,A.Vázquez-Barquero,Multidrug-resistantAcinetobacter baumanniiventriculitis:successfultreatmentwithintraventricularcolistin, ActaNeurochir.(Wien)151(November(11))(2009)1465–1472.

[23]P.DeBonis,G.Lofrese,G.Scoppettuolo,T.Spanu,R.Cultrera,M.Labonia,etal., Intraventricularversusintravenouscolistinforthetreatmentofextensively drugresistantAcinetobacterbaumanniimeningitis,Eur.J.Neurol.23(January (1))(2016)68–75.

[24]A.RodríguezGuardado,A.Blanco,V.Asensi,F.Pérez,J.C.Rial,V.Pintado,etal., Multidrug-resistantAcinetobactermeningitisinneurosurgicalpatientswith intraventricularcatheters:assessmentofdifferenttreatments,J.Antimicrob. Chemother.61(April(4))(2008)908–913.

[25]M.E.Falagas,P.I.Rafailidis,S.K.Kasiakou,P.Hatzopoulou,A.Michalopoulos, Effectivenessandnephrotoxicityofcolistinmonotherapyvs.

colistin-meropenemcombinationtherapyformultidrug-resistant

Gram-negativebacterialinfections,Clin.Microbiol.Infect.12(December(12)) (2006)1227–1230.

[26]B.-N.Kim,A.Y.Peleg,T.P.Lodise,J.Lipman,J.Li,R.Nation,etal.,Management ofmeningitisduetoantibiotic-resistantAcinetobacterspecies,LancetInfect. Dis.9(April(4))(2009)245–255.

[27]E.J.Giamarellos-Bourboulis,E.Xirouchaki,H.Giamarellou,Interactionsof colistinandrifampinonmultidrug-resistantAcinetobacterbaumannii,Diagn. Microbiol.Infect.Dis.40(July(3))(2001)117–120.

[28]A.Pantopoulou,E.J.Giamarellos-Bourboulis,M.Raftogannis,T.Tsaganos,I. Dontas,P.Koutoukas,etal.,Colistinoffersprolongedsurvivalinexperimental infectionbymultidrug-resistantAcinetobacterbaumannii:thesigniﬁcanceof co-administrationofrifampicin,Int.J.Antimicrob.Agents29(January(1)) (2007)51–55.

[29]J.Vila,J.Pachón,A.baumanniiresistanttoeverything:whatshouldwedo? Clin.Microbiol.Infect.17(July(7))(2011)955–956.

[30]A.R.Tunkel,B.J.Hartman,S.L.Kaplan,B.A.Kaufman,K.L.Roos,W.M.Scheld, etal.,Practiceguidelinesforthemanagementofbacterialmeningitis,Clin. Infect.Dis.39(November(9))(2004)1267–1284.