The efficacy and nephrotoxicity associat

Dosing ofintravenous colistin was prospectivelyreviewedbyclinicalpharmacists.Theinstitutional guidelineforcolistindosingwasasfollows3: Loading dose Colistin Base Activity [CBA], mg= C-Ta

The efficacy and nephrotoxicity associated with colistin use in an

intensive care unit in Vietnam: Use of colistin in a population of lower

a

b

c

f

g

1 Introduction

Multi-drug resistant gram-negative bacteria, such as

MDR-Acinetobacter baumannii,carbapenemase-producing

Enterobacteria-ceae, MDR-Pseudomonas aeruginosahavespreadrapidlyworldwide,

includingAsia.1Colistin,whichisproducedin vivoafterhydrolyzation

ofitsprodrugcolistimethatesodium,hasbeenincreasinglyemployed foroveradecadeasakeydrugforthetreatmentoftheseMDR-GNB.2

Colistinisknownforitsnephrotoxicitywhichinitially resultedin abundanceofitsclinicalusein1970s.2Majorityofrecentstudieson theclinicaluseofcolistinwereconductedinEuropeorNorthAmerica, andtherehasbeendebateontheappropriatedosinganditsrelationto theefficacyandnephrotoxicityofcolistin.2Informationoncolistinuse pertainingtotheAsianpopulationislimited.Recently,theinterim guideline to administercolistin incritically ill patientsbased on pharmacokinetic,pharmacodynamic,andtoxicodynamicprinciples

A R T I C L E I N F O

Received 11 March 2015

Received in revised form 21 March 2015

Accepted 26 March 2015

Keywords:

Colistin

Hospital acquired infection

ICU

Nephrotoxicity

Asia

Lower body weight

S U M M A R Y

Background:TherehasbeenagrowingneedforcolistinasakeydrugforthetreatmentofMDR-GNB infection.InformationoncolistinuseinAsianpopulationislimited

Methods:Aretrospectiveobservationalstudywasconductedtoassesstheefficacyandnephrotoxicityin criticallyilladultpatientswhoreceivedintravenouscolistinforMDR-GNBinfectionintheintensivecare unit(ICU)atBachMaiHospitalinHanoi,Vietnam.Colistinwasadministeredaccordingtothedosing guideline that was based on pharmacokinetic, pharmacodynamic and toxicodynamic principles, adjustedbybodyweightandcreatinineclearance

Results:Twenty-eighteligiblepatientswereincluded.Themeanpatientagewas6020.4years.The meanbodyweightwas538.6kg.Themeandaily doseofcolistinwas4.11.6MIU,andthemean cumulativedoseofcolistinwas48.222.8MIU.Colistintherapieswereclassifiedasclinicallyeffectivein19 (67.9%)cases.Six(21.4%)patientsdevelopednephrotoxicityduringthestudyperiodaccordingtoRIFLE criteria

Conclusion:Apersonalizeddosingprotocolofcolistinwaseffective,withlownephrotoxicity,among criticallyillVietnamesepatientswithlowbodyweight.Furtherstudiesarewarrantedforassessingthe efficacyandtoxicityinalargercohort

ß2015TheAuthors.PublishedbyElsevierLtdonbehalfofInternationalSocietyforInfectiousDiseases ThisisanopenaccessarticleundertheCCBY-NC-NDlicense(

http://creativecommons.org/licenses/by-nc-nd/4.0/)

* Corresponding author 1-21-1 Toyama, Shinjuku, Tokyo, 162-0052, Japan,

Disease Control and Prevention Center, National Center for Global Health and

Medicine, Tokyo, Japan Tel.: +1 80 6871 9083.

ContentslistsavailableatScienceDirect

j o urn a l hom e pa ge : ww w e l s e v i e r c om/ l o ca t e / i j i d

http://dx.doi.org/10.1016/j.ijid.2015.03.020

1201-9712/ß 2015 The Authors Published by Elsevier Ltd on behalf of International Society for Infectious Diseases This is an open access article under the CC BY-NC-ND

administrationofcolistinhasnotbeenwellevaluatedworldwide,

evenlesssoinAsiancountrieswherepeopletendtohavelowerbody

weightthaninEuropeorNorthAmerica.Inthisstudy,weevaluated

the efficacy andnephrotoxicity of personalized administrationof

colistinincritically-illpatientsadmittedtoICUinVietnam

2 Methods

2.1 Study Design and Patient Population

This was a retrospective observational studyto assess the

efficacyandnephrotoxicityincriticallyillpatientswhoreceived

intravenouscolistinatBachMaiHospital(BMH)betweenAugust

15,2013andJanuary152014.BMHhas2000bedsandserves

as a tertiary care hospital in Hanoi, Vietnam The studywas

approved by Bach Mai Hospital institutional review board

Adultpatientsagedgreaterthan18yearswereincludedinthe

studyiftheywereadmittedtotheintensivecareunit(ICU)and

receivedintravenouscolistinforhospitalacquiredinfectiondue

to MDR-GNB with positive microbiological culture Hospital

acquiredinfection(HAI)wasdeterminedaccordingtoCDC/NHSN

definitions4 and according to multiple physicians’ evaluation

Patientswereexcludediftheywerepregnantorbreast-feeding

or were receiving renal replacement therapy (intermittent

hemodialysisor continuousrenalreplacement therapy)before

theinitiationofcolistin.Patientswereexcludediftheyreceived

colistinforlessthanfivedays,toensureadequateexposureto

thedrug

2.2 Microbiology

BMHhasasinglecentralizedmicrobiologylaboratory.Standard

identificationand susceptibilitytesting of clinicalisolates were

performed in accordance with the Clinical and Laboratory

Standards Institute (CLSI) criteria.5 The minimum inhibitory

concentrations (MICs) of collistin were determined by E-test

(Sysmex-bioMerieux, Tokyo, Japan) according to the

manufac-turer’sinstructions

2.3 Colistin administration

The colistin product used in this study was Coly-Mycin1

producedby Sanofi-Aventis Dosing ofintravenous colistin was

prospectivelyreviewedbyclinicalpharmacists.Theinstitutional

guidelineforcolistindosingwasasfollows3:

Loading dose (Colistin Base Activity [CBA], mg)=

C-Tar-get2Totalactualbodyweight(kg)

Maintenance dose (CBA, mg)=C-target(1.5CrCl

[Creati-nineclearance,mL/min]+30)

Maintenancedosewasinitiated24hoursafterloading dose

infusion.C-targetwascalculatedasfollows.C-targetwasequal

to the identified colistin MIC for the causative organism of

HAI.ThedosescalculatedbasedonCBA (mg)weredividedby

33.3toconvertthemtoMIU(millioninternational units).The

total dailydosage wasdivided into twodoses for twice-daily

administration

Eachbottleofcolistinwasdissolvedin50mLofnormalsaline

solution(0.9%NaCl)andwasinfusedimmediatelyover30minutes

to2hoursfollowingitsdissolution.Clinicalpharmacistsrechecked

andrecalculated themaintainingcolistindose accordingtothe

patient’smeasured renal functionduring colistintherapy Body

weights and CrCl were measured within 2 days of colistin

administration.Nebulizedcolistinwasnotusedthroughout the

studyperiod

2.4 Data Collection

The following parameters were retrieved from the medical records of patients in the study: age, sex, weight, underlying diseases, baseline serum creatinine concentration, Charlson’s score,6AcutePhysiologyandChronicHealthEvaluation(APACHE)

IIscore,7ClinicalPulmonaryInfectionScore(CPIS),8andSequential OrganFailure Assessment (SOFA)score on ICU admission.9 The information on the use of other nephrotoxic drugs (NSAIDs, furosemide,contrastagent,angiotensin-convertingenzyme inhi-bitors)wasalsocollected

2.5 Clinical assessment

Clinicalassessmentswereconductedat3timepoints:thefirst waspriortousingcolistin;thesecondwasafterday5ofcolistin treatment;thelastpointwasafterdiscontinuingcolistin.Multiple physicians involved in the patients’ care evaluated the clinical effectivenessofcolistintherapyateachtimepoint,basedonthe resolution, persistence orworsening of symptoms and signs of infection

2.6 Microbiological assessment

Microbiological culture samples werecollected at two time points,thefirstwaspriortoadministeringcolistinandthesecond wasafterday5ofcolistintreatment.Samplesweretransferredto themicrobiologydepartment,andsamplecultureresultandMICs weredetermined.Microbiologicalefficacywasevaluatedbasedon the comparison of two consecutive culture results; i.e., if the secondculturewasnegative,thenitwasevaluatedas microbio-logicallyeffective

2.7 Nephrotoxicity assessment

Dailyserumcreatininelevelwasrecordedfromthefirstdayof colistin therapy until discharge or death Nephrotoxicity was definedbasedontheincreaseintheserumcreatinine concentra-tionof50percentasperRIFLE(risk,injury,failure,loss,and end-stagekidneydisease)criteria.10

2.8 Statistical Analysis

AllanalyseswereperformedusingSPSS20.Bivariateanalyses wereperformedusingtheFisher’sexacttestortheChi-squaretest forcategoricalvariablesandthet-testortheMann-WhitneyUtest forcontinuousvariables.AllP-valuesweretwo-sided,apvalueof lessthan0.05wasconsideredtoindicateastatisticallysignificant difference.Throughoutthetext,thepercentagesdisplayedarethe

‘‘validpercent’’,whichindicatesthepercentexcludingthemissing datafromthedenominator

3 Results During thestudyperiod,28eligiblepatientswereidentified Themeanagewas60(20.4;range:19-88)years,and18(64%)were

(8.6;range:35.5-75)kg.Eight(28.6%)patientshadpreexistingrenal failurepriortotheadministrationofcolistin,whichwasdefinedbya serumcreatinine(Scr)value>1.2mg/dl.Themajority(n=26,92.9%)

ofpatientshadventilator-associatedpneumonia(VAP),and2(7.1%) patientshadblood-streaminfections

Acinetobacter baumannii were most frequently isolated (n=24 [85.7%]; 23 from sputum, 1 from blood), followed by

Pseudomonas aeruginosa (n=3 [10.7%]; 3 from sputum), and

Klebsiella pneumonia(n=3 [10.7%]; 2from sputumand 1 from

Table 1

Patient characteristics based on the clinical response to intravenous colistin therapy (n = 28)

(n = 28)

Clinically effective (n = 19, 68%)

Clinically ineffective (n = 9, 32%)

P value (Effective group

vs ineffective-treatment group) Demographics

Body weight (kg), median (IQR),

[mean  SD]

53.5 (45.5-58.5) [53  8.6] 57 (51-59) [54.5  6.8] 49.5 (41.5-54.3) [49.8  11.4] 0.05 Charlson comorbidity index,

median (IQR)

Length of ICU stay prior to colistin

therapy (days), median (IQR)

Severity of illness

Site of Infection

Microbiology

Colistin MIC ofAcinetobacter

Colistin therapy

Average daily dose (MIU),

median (IQR), [mean  SD]

4.0 (2.7-5.6) [4.1  1.6] 4.4 (3.1-6.2) [4.4  1.7] 3.2 (2.6-4.1) [3.5  1.0] 0.12 Average daily dose per kg (MIU),

median (IQR), [mean  SD]

0.08 (0.05-0.11) [0.08  0.03] 0.09 (0.06-0.11) [0.08  0.03] 0.08 (0.05-0.11) [0.07  0.03] 0.50 Total cumulative dose (MIU),

median (IQR), [mean  SD]

39 (33-57) [48.2  22.8] 50 (33-72) [54.1  24.2] 37.5 (26.5-43.5) [35.7  13.1] 0.12 Total cumulative dose per kg

(MIU), median (IQR), [mean  SD]

0.84 (0.64-1.17) [0.91  0.38] 0.86 (0.67-1.33) [0.98  0.38] 0.80 (0.4-1.13) [0.76  0.35] 0.27 Duration of colistin therapy

(days), median (IQR)

Combination therapy with

carbapenem, n (%)

Use of concomitant nephrotoxic agents

Renal function

Pre-existing renal failure

(Scr > 1.2 mg/dl), n (%)

Scr, prior to colistin therapy, mg/dl,

median (IQR), [mean  SD]

0.8 (0.8-1.43) [1.2  0.9] 0.9 (0.8-1.2) [1.1  0.6] 0.8 (0.65-2.45) [1.5  1.4] 0.63 Scr, worst during therapy, mg/dl,

median (IQR), [mean  SD]

1.05 (0.8-2.25) [1.7  1.4] 1.2 (0.8-2.3) [1.5  0.9] 0.9 (0.75-3.85) [2.1  2.2] 0.73 Scr, upon discharge from ICU, mg/dl,

median (IQR), [mean  SD]

0.9 (0.7-1.5) [1.4  1.3] 0.9 (0.7-1.4) [1.2  0.7] 0.8 (0.6-3.7) [1.9  2.1] 0.79 CrCl, prior to colistin therapy, ml/min,

median (IQR), [mean  SD]

62.3 (34.8-76.1) [62.6  37.8] 67.6 (51.7-83.5) [80  40.5] 50 (21.1-66.2) [44.9  24.8] 0.06 CrCl, worst during therapy, ml/min,

median (IQR), [mean  SD]

52 (25.5-70) [53.7  30] 61 (34-79) [60.4  31.5] 48 (16-57.5) [39.4  21.9] 0.09 CrCl, upon discharge from ICU,

ml/min, median (IQR), [mean  SD]

60 (35.8-87.3) [63.5  34.1] 64 (47-99) [71  33.6] 55 (16.5-74) [47.7  31.2] 0.08 Renal failure upon discharge

(Scr > 1.2 mg/dl), n (%)

Outcome

Total ICU length of stay (days),

median (IQR)

Nephrotoxicity during colistin

therapy per RIFLE criteria a

, n (%)

Increase of Scr > 150% as compared

to baseline upon discharge, n (%)

Abbreviations ACEI, Angiotensin-converting enzyme inhibitors; APACHE, Acute Physiology and Chronic Health Evaluation; BSI, blood stream infection; CBA, colistin base activity; CrCl, Creatinine clearance; CPIS, Clinical Pulmonary Infection Score; ICU, intensive care unit; IQR, interquartile range; MIU, Million International Units; Scr, serum creatinine; SD, standard deviation; SOFA, Sequential Organ Failure Assessment; VAP, ventilator associated pneumonia.

blood) The colistin MIC50 and MIC90 of A baumannii were

0.125mg/L, and 0.5mg/L, respectively, and MIC ranged from

0.064-0.75mg/L

Themeandailydoseofcolistinusedinthewholecohortwas

4.11.6MIU(136.553.3mgCBA;median: 4.0MIU[133.2mg

CBA],IQR[interquartilerange]:2.7-5.6MIU[89.9-186.5mgCBA])

The mean dailydose of colistin per kg was 0.080.03 MIU/kg

(2.661.0mgCBA/kg;median:0.08MIU/kg[2.66mgCBA/kg],IQR:

0.05-0.11MIU/kg[IQR:1.67-3.66mgCBA/kg]).Themeandurationof

colistintherapywas 12.55.2 days(range:5-23 days),andthe

mean cumulative colistin dose was 48.222.8 MIU [1605.1

759.2mgCBA)](median:39MIU[1298.7mgCBA],IQR:33-57MIU

[1098.9-1898.1mgCBA]).Themeancumulativecolistindoseperkg

was 0.910.38 MIU/kg [30.312.65mg CBA/kg] (median:

0.84 MIU [27.97mg CBA/kg], IQR: 0.64-1.17 MIU/kg

[21.31-38.96mgCBA/kg]).Inallcases,colistinwasusedascombination

therapy and most frequently combined with carbapenem in 25

(89.3%)cases

Nephrotoxicity during colistin therapy was identified in 6

(21.4%)patients accordingtotheRIFLE criteria.Allsix patients

identifiedasAKIperRIFLEcriteriawerecategorizedas‘‘risk’’under

RIFLEcriteria.Among8patientswithpre-existingrenalfailure,3

(37.5%) developed AKI per RIFLE criteria Among 20 patients

withoutpre-existingrenalfailure,3(15%)developedAKIaccording

toRIFLEcriteria

Concomitantnephrotoxicity agents were usedin 11 (39.3%)

patients: 1 patient received angiotensin converting enzyme

inhibitor,and10patientsreceivedfurosemide.Othernephrotoxic

agents,suchasnon-steroidalanti-inflammatorydrug,

vancomy-cin,aminoglycoside,orcontrastagentwasnotusedconcurrently

withcolistininanypatient

3.1 Comparison between the clinically effective-treatment group and

the ineffective-treatment group

Colistin therapies were evaluated as clinically effective in

19cases (68%)and clinicallyineffective in 9 (32%) cases.The

patients’ characteristics based on the clinical response to

intravenouscolistintherapy,weresummarizedinTable1.The

patientsoftheclinicallyeffective-treatmentgroupwereyounger

than patients without response (mean age 51.819.6 vs

7412.8,p=0.01).Thepatientsoftheineffective-treatmentgroup

comprised the population that showed a greater incidence of

comorbidities with higher median Charlson comorbidity index

(3 [IQR: 1-4] vs 5 [IQR: 3-6], p=0.03) and had more organ

dysfunctionindicatedbyhighermedianSOFAscore(6[IQR:4-8]

vs8[IQR:8-10],p=0.02).ThemedianCPISscorewassignificantly

lower in the clinically effective-treatment group than in the

ineffective-treatmentgroup(6[IQR:5-7]vs8[IQR:6-8],p=0.01)

Thepatientsinclinicallyeffective-treatmentgrouphadmarginally

highermedianbodyweightthantheineffective-treatmentgroup

(57[IQR:51-59]vs.49.5[41.5-54.3],p=0.05).ThemedianMICofA.

baumanniiislowerinclinicallyeffective-treatmentgroupthanin

ineffective-treatmentgroup(0.13[0.09-0.16]vs.0.38[0.13-0.50],

p=0.02).Theothervariables,includingpatients’demographics(e.g

sex,APACHEIIscore,lengthofICUstaypriortocolistinuse),severity

ofillness (e.g severe sepsis,septic shock),site ofinfection, and

colistin therapy (daily doses, duration, cumulative dose), and

concomitant nephrotoxic agent use were similar between two

groups.Theprevalenceofpre-existingrenalfailuredefinedasScr

>1.2mg/dl prior to colistin therapy were similar between two

groups.Thefrequenciesofthedevelopmentofnephrotoxicity,which

weredefinedbyRIFLEcriteria,weresimilarbetweenthetwogroups

Nodeathswereobservedintheclinicallyeffective-treatmentgroup

during the study period; however, 5 (55.6%) patients without

responsedied(p 0.01).The14-day mortalitywashigherinthe

clinicallyineffective-treatmentgroupthantheeffectivetreatment group(6[66.7%]vs.2[10.5%],P<0.01)

3.2 Comparison between nephrotoxicity and non-nephrotoxicity group

Characteristicsofpatientswhodevelopednephrotoxicityand whodidnotdevelopnephrotoxicityduringcolistintherapybased

on RIFLE criteria were also compared (Table 2 Patients with nephrotoxicity had higher median body weight than patients withoutnephrotoxicity(57.8[IQR:53.8-65.3]vs.51.3[IQR: 43.8-58.1],p=0.05).However,thecolistindosesweresimilarbetween

2groupsbecauseofdecreasedCrClinthenephrotoxicitygroup Durationofcolistintherapydidnotdifferbetweenthetwogroups Othercharacteristicssuchas age,sex,severityofillness,site of infectionsweresimilarbetweenthetwogroups.Theprevalenceof pre-existing renal failure was similarbetween the two groups Outcome parameterssuchas microbiologicalandclinical effec-tiveness,in-hospitaland14-daymortalitydidnotdifferbetween thetwogroups

4 Discussion

Toourknowledge,thisisthefirststudytodescribethedetailsof clinicalexperiencesofcolistinuseinVietnam.Thevastmajorityof reports regarding colistin use in clinical settings have been publishedfromEurope andNorthAmerica,andthedataonthe clinicaluseofcolistininAsiaisscarce

Thepreviouslyreportedincidencesofnephrotoxicityduring colistin therapyvaryfrom 6%to 55%,2 anddepend on various factorssuchasdosing,populationdifferences(e.g.comorbidity, severity,clinicalsetting),anddefinitionofnephrotoxicity.Inthis study, weusedRIFLEcriteriatoidentifynephrotoxicity,which wasvalidated tocorrelatewith prognosis.11WeusedonlyScr changesforapplyingtheRIFLEcriteriawithoutreferencetoGFR, because previous reports suggested that the changes in Scr concentrationsdonotcorrelatewiththepercentdecreasesinGFR

intheRIFLEclassification.12Theincidenceofnephrotoxicityin this study (21.4%) was lower than the incidence reported in previousstudies(range:31-54.6%)thathaveevaluated nephro-toxicity duringcolistin therapy usingRIFLEcriteria.13 Patients receivedrelativelylower doseofcolistinin ourstudythanthe previous studies,14–17 even if their actual body weights are considered The mortality and prevalence of non-responding patients were not higher in our study,14,17,18 even though Charlson comorbidity indexscores, SOFA scores, andpatients’ ages in our study were apparently not lower than previous studies.14,17,18Weusedactualbodyweightinplaceofidealbody weightsincetherewerenoobesepatientsinourstudycohort,in whom the dissociationbetween actualand ideal bodyweight couldbesignificant

Weexcludedpatientsiftheyreceivedcolistinforlessthanfive daystoensureadequateexposuretothedrug.Inourstudy,no patient developed nephrotoxicity within 5 days after starting colistintherapy.However,previousstudiesreportedthe devel-opment of nephrotoxicity of various incidences (15.5%-100%) within7daysofcolistinadministration,15,18andusedinclusion criteriaofreceivingcolistinlongerthan48to72hours.14–18There was no patientwho died within 5 daysafter starting colistin therapy

Inthisstudy,weusedadosingstrategyincludingtheloading dose and maintenance dose, which wassuggested by a recent study byGaronziketal.3The reportofGaronziket alincluded patientsfromThailand;however,italsoincludedpatientsfromthe U.S.,andthustherangewasmuchwiderthanourstudy(median bodyweight59.1kg[range:30.0–106.4]vs.53.5kg[35.5-75]).We

evaluated the efficacyand safety of a colistin dosing strategy,

includingtheloadingdoseandthemaintenancedose,3whichwas

adjustedbybothbodyweightandCrCl,inAsianpopulationwith

lower body mass Further studies are warranted to assess the

efficacyand toxicityin a largercohort Due tothe worldwide

spreadofmulti-drugresistantpathogens,apersonalizedapproach

iscrucialfortheappropriateuseandevaluationofefficacyand

safetyofcolistintherapyinclinicalsettings

Acknowledgments

This workwassupported bytheJ-GRID (Japan Initiativefor

GlobalResearchNetworkonInfectiousDiseases),MEXTJapan

Conflict of interest statement:Noconflictsofinteresttodeclare

The study was approved by Bach Mai Hospital institutional

reviewboard.(ApprovalNo.38)

References

1 Jean SS, Hsueh PR High burden of antimicrobial resistance in Asia.International journal of antimicrobial agents2011;37:291–5.

2 Yahav D, Farbman L, Leibovici L, Paul M Colistin: new lessons on an old antibiotic. Clinical microbiology and infection: the official publication of the European Society of Clinical Microbiology and Infectious Diseases2012;18:18–29.

3 Garonzik SM, Li J, Thamlikitkul V, Paterson DL, Shoham S, Jacob J, et al Population pharmacokinetics of colistin methanesulfonate and formed colistin in critically ill patients from a multicenter study provide dosing suggestions for various catego-ries of patients.Antimicrobial agents and chemotherapy2011;55:3284–94.

4 Horan TC, Andrus M, Dudeck MA CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting.American journal of infection control2008;36:309–32.

5 The Clinical and Laboratory Standards Institute Performance Standards for Antimicrobial Susceptibility Testing; Twentieth Informational Supplement Approved Standard M100-S23 Wayne, PA: CLSI; 2013.

6 Charlson ME, Pompei P, Ales KL, MacKenzie CR A new method of classifying prognostic comorbidity in longitudinal studies: development and validation.

Journal of chronic diseases1987;40:373–83.

Table 2

Patient characteristics based on nephrotoxicity to intravenous colistin therapy (n = 28)

(n = 6, 21%)

Non-nephrotoxicity group (n = 22, 79%)

P value (Nephrotoxicity group vs non-nephrotoxicity group) Demographics

Body weight (kg), median (IQR), [mean  SD] 57.8 (53.8-65.3) [59.9  8.1] 51.3 (43.8-58.1) [51.1  8.0] 0.05

Severity of illness

Site of Infection

Microbiology

Colistin MIC ofAcinetobacter baumannii,mg/L, median (IQR) 0.13 (0.03-0.17) 0.13 (0.13-0.37) 0.31

Colistin therapy

Average daily dose (MIU), median (IQR), [mean  SD] 3.3 (2.2-5.8) [3.8  1.8] 4 (3-5.5) [4.2  1.5] 0.43

Average daily dose per kg (MIU), median (IQR), [mean  SD] 0.07 (0.04-0.1) [0.07  0.03] 0.09 (0.06-0.11) [0.08  0.03] 0.22

Total cumulative dose (MIU), median (IQR), [mean  SD] 42 (32-89.3) [57.2  33.3] 39 (33-57) [45.7  19.3] 0.70

Total cumulative dose per kg (MIU), median (IQR), [mean  SD] 0.76 (0.56-1.5) [0.96  0.54] 0.85 (0.7-1.16) [0.9  0.34] 0.96

Use of concomitant nephrotoxicity agents

Renal function

Scr, prior to colistin therapy, median (IQR) 1.2 (0.8-2.0) [1.5  1.0] 0.8 (0.7-1.3) [1.2  0.9] 0.26

Scr, worst during therapy, median (IQR) 2.7 (1.3-3.7) [2.7  1.7] 1.0 (0.8-1.4) [1.4  1.2] 0.03

Scr, upon discharge from ICU, median (IQR) 1.9 (1.2-3.7) [2.4  1.8] 0.8 (0.7-1.1) [1.1  1.0] 0.02

CrCl, prior to colistin therapy, median (IQR) 62.2 (25.4-119.7) [77.4  66.6] 62.3 (37.9-76) [58.6  26.4] 0.80

CrCl, worst during therapy, median (IQR) 28.5 (13.3-89) [46.7  43.2] 54 (44.3-68) [55.6  26.4] 0.40

CrCl, upon discharge from ICU, median (IQR) 42 (17.8-81) [52.2  46.1] 61.5 (44.8-90.3) [66.6  30.7] 0.31

Outcomes

Abbreviations ACEI, Angiotensin-converting enzyme inhibitors; APACHE, Acute Physiology and Chronic Health Evaluation; BSI, blood stream infection; CrCl, Creatinine clearance; CPIS, Clinical Pulmonary Infection Score; ICU, intensive care unit; IQR, interquartile range; MIU, Million International Units; Scr, serum creatinine; SD, standard deviation; SOFA, Sequential Organ Failure Assessment; VAP, ventilator associated pneumonia.

a

Nephrotoxicity was defined according to RIFLE criteria 10

7 Knaus WA, Draper EA, Wagner DP, Zimmerman JE APACHE II: a severity of

disease classification system.Critical care medicine1985;13:818–29.

8 Pugin J, Auckenthaler R, Mili N, Janssens JP, Lew PD, Suter PM Diagnosis of

ventilator-associated pneumonia by bacteriologic analysis of bronchoscopic

and nonbronchoscopic ‘‘blind’’ bronchoalveolar lavage fluid. The American

review of respiratory disease1991;143:1121–9.

9 Vincent JL, Moreno R, Takala J, Willatts S, De Mendonca A, Bruining H, et al The

SOFA (Sepsis-related Organ Failure Assessment) score to describe organ

dys-function/failure On behalf of the Working Group on Sepsis-Related Problems of

the European Society of Intensive Care Medicine. Intensive care medicine

1996;22:707–10.

10 Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P Acute Dialysis Quality

Initiative w Acute renal failure - definition, outcome measures, animal models,

fluid therapy and information technology needs: the Second International

Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group.

Critical care2004;8:R204–12.

11 Ali T, Khan I, Simpson W, Prescott G, Townend J, Smith W, et al Incidence and

outcomes in acute kidney injury: a comprehensive population-based study.

Journal of the American Society of Nephrology: JASN2007;18:1292–8.

12 Pickering JW, Endre ZH GFR shot by RIFLE: errors in staging acute kidney injury.

Lancet2009;373:1318–9.

13 Pulluru HKK, Pogue JM, Marchaim D, Hayakawa K Risk factors for and incidence

of nephrotoxicity associated with intravenous colistin in the modern era: a systematic review.Interscience Conference on Antimicrobial Agents and Chemo-therapy.2012.

14 Dalfino L, Puntillo F, Mosca A, Monno R, Spada ML, Coppolecchia S, et al High-dose, extended-interval colistin administration in critically ill patients: is this the right dosing strategy? A preliminary study. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America

2012;54:1720–6.

15 Deryke CA, Crawford AJ, Uddin N, Wallace MR Colistin dosing and nephrotoxi-city in a large community teaching hospital.Antimicrobial agents and chemo-therapy2010;54:4503–5.

16 Doshi NM, Mount KL, Murphy CV Nephrotoxicity associated with intravenous colistin in critically ill patients.Pharmacotherapy2011;31:1257–64.

17 Pogue JM, Lee J, Marchaim D, Yee V, Zhao JJ, Chopra T, et al Incidence of and risk factors for colistin-associated nephrotoxicity in a large academic health system.

Clinical infectious diseases: an official publication of the Infectious Diseases Society

of America2011;53:879–84.

18 Paul M, Bishara J, Levcovich A, Chowers M, Goldberg E, Singer P, et al Effec-tiveness and safety of colistin: prospective comparative cohort study.The Journal of antimicrobial chemotherapy2010;65:1019–27.