Fabrizio Fabrizi , Roberta Cerutti , Vivek Dixit , Piergiorgio Messa review and meta-analysis The impact of antiviral therapy for HCV on kidney disease: asystematic Original

(1)

w w w . r e v i s t a n e f r o l o g i a . c o m RevistadelaSociedadEspañoladeNefrología

Original

The impact of antiviral therapy for HCV on kidney disease: a systematic review and meta-analysis

Fabrizio Fabrizi

^a,∗

, Roberta Cerutti

^a

, Vivek Dixit

^b

, Piergiorgio Messa

^a,c

aFoundationIRCCSCa’GrandaOspedaleMaggiorePoliclinico,Milano,Italy

bDivisionofGastroenterology,UCLASchoolofMedicine,CA,USA

cUniversitySchoolofMedicine,Milano,Italy

a r t i c l e i n f o

Articlehistory:

Received6April2019 Accepted15July2019

Availableonline6December2019

Keywords:

Chronicrenalinsufﬁciency HepatitisC

Interferon Meta-Analysis RenalDialysis

a bs t r a c t

Background:ControversypersistsabouttheroleofhepatitisCasariskfactorfordeveloping kidneydiseaseinthegeneralpopulation.Someauthorshaveevaluatedtheeffectofantiviral therapyforHCVontheriskofkidneydisease.

StudyAimsandDesign:Asystematicreviewofthepublishedmedicalliteraturewasper- formedtoassesswhetherantiviraltherapyforHCVhasanindependentimpactonkidney survivalintheadultgeneralpopulation.Arandomeffectsmodelwasusedtogeneratean overallestimateoftheriskofkidneydiseaseafteranti-HCVtherapyacrossthepublished studies.Meta-regressionandstratiﬁedanalysiswerealsocarriedout.

Results:Fifteenstudieswereeligible(n=356,285patients)andseparatemeta-analyseswere conductedaccordingtotheoutcome.Poolingstudiesbasedonviralresponses(n=7;34,763 individualpatients)demonstrated a relationshipbetweensustainedviral responseand lower frequencyofkidneydisease;theoverallestimateforadjustedriskofkidneydis- easewas2.50(95%CI,1.41;4.41)(p=0.0016)andbetween-studyheterogeneitywasfound (p-valuebyQtest=0.004).Aggregationofstudiescomparingtreatedvsuntreatedcohorts (n=8,n=333,312patients)revealedanassociationbetweenanti-HCVtherapyandlowerrisk ofkidneydisease.Theoverallestimateforadjustedriskofkidneydiseaseacrosstheeight studieswas0.39(95%CI,0.25;0.612)(p=0.0001).Meta-regressionshowedthattheeffective- nessofantiviraltherapyinreducingthefrequencyofkidneydiseasediminishesascirrhosis (p=0.02)andHBVinfection(p=0.0001)increaseamongHCV-infectedindividuals.

Abbreviations: ACEi,angiotensin-convertingenzymeinhibitor;AH,arterialhypertension;aHR,adjustedhazardratio;ALT,alanine aminotransferase;ARB,angiotensinIIreceptorblocker;AST,aspartateaminotransferase;CKD,chronickidneydisease;CI,confidence intervals;CNI,calcineurininhibitors;CRF,chronicrenalfailure;DAAs,direct-actingantiviralagents;DM,diabetesmellitus;EOT,endof treatment;ESRD,end-stagerenaldisease;eGFR,estimatedglomerularfiltrationrate;GN,glomerulonephritis;HCV,hepatitisCvirus;HD, haemodialysis;ICD-9-CM,InternationalClassificationofDiseasesNinthRevisionClinicalModification;IFN,interferon;ITT,intention- to-treatanalysis;MDRD,modificationofdietinrenaldisease;NA,notavailable;NOS,Newcastle/OttawaScale;NSAID,non-steroidal anti-inflammatorydrug;RBV,ribavirin;SVR,sustainedvirologicalresponse.

∗ Correspondingauthor.

E-mailaddress:fabrizio.fabrizi@policlinico.mi.it(F.Fabrizi).

https://doi.org/10.1016/j.nefro.2019.07.007

(2)

Conclusions:AntiviraltherapyforHCVlowerstheriskofkidneydiseaseamongHCV-infected individuals.Studiestounderstandthemechanismsunderlyingthisassociationareongoing.

ThisisanopenaccessarticleundertheCCBY-NC-NDlicense(http://creativecommons.

org/licenses/by-nc-nd/4.0/).

ImpactodelaterapiaantiviralparaVHCenlasenfermedadesrenales:

revisiónsistemáticaymeta-análisis

Palabrasclave:

Insuﬁcienciarenalcrónica HepatitisC

Interferón Meta-Análisis Diálisisrenal

r e s u m en

Antecedentes:SigueexistiendocontroversiaacercadelroldelahepatitisCcomofactorde riesgodedesarrollodeenfermedadesrenalesenlapoblacióngeneral.Algunosautoreshan evaluadoelefectodelaterapiaantiviralparaelVHCenelriesgodeenfermedadrenal.

Objetivosydise ˜nodelestudio:Serealizóunarevisiónsistemáticadelaliteraturamédicapub- licada,paraevaluarsilaterapiaantiviralparaelVHCteníaunimpactoindependienteenla supervivenciarenalenlapoblaciónadultageneral.Seutilizóunmodelodeefectosaleato- riosparagenerarunaestimacióngeneraldelriesgodeenfermedadrenaltraslaterapia anti-VHC,entrelosestudiospublicados.Tambiénserealizaronanálisisdemeta-regresión yestratiﬁcados.

Resultados:Quinceestudiosresultaronelegibles(n=356,285pacientes),realizándosemeta- análisis separados con arreglo al resultado. Los estudios agrupados basados en las respuestasvirales(n=7;34.763pacientesindividuales)demostraronunarelaciónentrela respuestaviralsostenidaylamenorfrecuenciadeenfermedadrenal;laestimacióngen- eralparaelriesgoajustadodeenfermedadrenalfuede2,5(95%IC,1,41;4,41)(p=0,0016), encontrándoseunaheterogeneidadentreestudios(valorpconpruebaQ=0,004).Lasuma deestudioscomparativosdecohortestratadasvsnotratadas(n=8,n=333.312pacientes) revelóunaasociaciónentrelaterapiaanti-VHCyelmenorriesgodeenfermedadrenal.La estimacióngeneralparaelriesgoajustadodeenfermedadrenalentreochoestudiosfuede 0,39(95%IC,0,25;0,612)(p=0,0001).Lameta-regresiónreﬂejóquelaefectividaddelaterapia antiviralparareducirlafrecuenciadeenfermedadrenaldisminuyeamedidaqueaumentan loscasosdecirrosis(p=0,02)einfecciónporVHB(p=0,0001)entrelosindividuosinfectados deVHC.

Conclusiones:Laterapiaantiviral paraelVHC disminuye elriesgode enfermedadrenal entrelosindividuosinfectadosdeVHC.Soncontinuoslosestudiosparacomprenderlos mecanismossubyacentesaestaasociación.

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

Introduction

Hepatitis C is one of the main causes of liver cirrhosis worldwide¹.Inadditiontoitseffectsintheliver,chronichep- atitisCvirusinfectionshowsextra-hepaticactivityinvarious organsandtissuesincludingkidneys¹.Recentevidencesug- gestsalinkbetweenhepatitisCvirusinfection andchronic kidneydisease².ChronicHCVandCKDaremajorpublichealth issuesallovertheworld;globally,anestimated71millionpeo- plehave chronichepatitisC infection and therate ofCKD intheadultgeneralpopulationofindustrializedcountriesis around10-15%³.

Conventional risk factors for developing chronic kid- neydiseaseincludediabetesmellitus,arterialhypertension, smoking,metabolicsyndromeandincreasingage;however, thesefactors donotfully explainthe current frequencyof chronic kidney disease in the adult general population of

developedworld.Asystematicreviewwithmeta-analysisof cross-sectional/longitudinalstudies(n=15;2,299,134unique patients)recentlydemonstratedarelationshipbetweenpos- itiveanti-HCVserologicstatusandincreasedriskofchronic kidney disease;the summary estimateforadjusted hazard ratiowas1.54(95%ConﬁdenceInterval,1.26;1.87,P=0.0001)⁴. Novel data show that sustained viral response with interferon-based regimens leads to an improvement of hepatic and extra-hepatic outcomes, such as reduction rates of stroke, acute coronary syndrome, and better kidney survival⁵. In a prospective French study, 668 cirrhotic patients(50.5%)underwentantiviraltreatmentforHCVand achieved SVR;theyhad a lower riskofcardiovascular (HR, 0.42; 95% CI, 0.25-0.69; P=0.001) and infectious (HR, 0.44;

95% CI,0.29-0.68;P<0.001)events⁶. Inaretrospective analysis performed at the VA Greater Los Angeles Healthcare System(n=523HCV-infectedpatients),patientswhoreached

(3)

SVR12hadalowerdeclineinGFRcomparedwiththosewho didnotachieveSVR12,3.1mL/min+0.75mL/minper1.73m² vs.11mL/min+2.81mL/minper1.73m²(P=0.002)⁷.However, theseresultshavenotbeenconﬁrmedbyothers⁸.

Theaimofthisstudywastoreviewtheavailableevidence ontheimpact ofantiviral therapiesforHCVonthe riskof kidneydiseaseintheadultgeneralpopulationbyperforming asystematicreviewoftheliteraturewithameta-analysisof clinicalobservationalstudies.

Material and methods

ThisworkisinagreementwiththePreferredReportingItems forSystematicreviewsandMeta-Analysesstatement(Supple- mentalFile1)⁹.

SearchStrategyandDataExtraction

Twoauthors(F.F.,andV.D.)independentlyreviewedEnglish- language citations from the National Library ofMedicine’s Medline database from 1989 through June 30, 2018. Data on HCV status were not available before 1989, when the ﬁrst assay for HCV was manufactured. We conductedour searchbyfourMedlinedatabasesengines(Embase,Grateful Med, Ovid,and PubMed). Our Medline search was limited to human studies. We applied the following algorithm in medicalsubjectheadingandinfreetextwords:(“HEPATITIS C” or”HCV” or “HEPATITIS C VIRUS INFECTION” or “HCV INFECTION” or”ANTIBODY POSITIVE SEROLOGIC STATUS”

or”ANTI-HCV”)and (“CHRONIC KIDNEY DISEASE”or “CKD”

or “END-STAGE RENAL DISEASE” or “ESRD” or “GLOMERU- LONEPHRITIS” or”GN” or”LOW GLOMERULAR FILTRATION RATE”or”lowGFR”or”KIDNEYDISEASE”or”KIDNEYEVENTS”) and (“INTERFERON” or”IFN” or”pegylated-INTERFERON”

or”peg-IFN” or “RIBAVIRIN” or”RBV”) and (”DIRECT-ACTING ANTIVIRALAGENTS”or“DAAs”)and(“SUSTAINEDVIROLOG- ICALRESPONSE”or”SUSTAINEDVIRALRESPONSE”or”SVR”or

“CURE”)and(“HAZARDRATIO”or“HR”or“INTERNATIONAL RATE RATIO” or “IRR” or “ODDS RATIO” or”OR”). An addi- tionalsearchwasperformedwithelectronicsearchesofthe CochraneLibrary;manualsearchesofselectedspecialtyjour- nalsweredonetoidentifyallpertinentliterature.Reference lists from qualitative topic reviews and published clinical studieswerealsosearched.Itwaspreviouslydemonstrated thataMedlinesearchalonemightnotbesensitiveenough¹⁰. Data onstudy design,study period,patient characteristics, HCVprevalence,antiviraltherapytowardsHCV,andkidney diseaseoutcomeswereabstracted.Authorsofselectedpapers were contactedtoobtain missingdata and onlydatafrom individuals with known HCV status were included in the meta-analysis.Consensuswasachievedforalldata.Studies werecomparedtoeliminateduplicatereportsforthesame patients, which included contact with investigators when necessary.Eligibilityandexclusioncriteriawerepre-speciﬁed.

Oursearchwaslimitedtohumanstudiesthatwerepublished intheEnglishliterature.

InclusionCriteria

Studies were included if they met the following inclusion criteria: (1) they presented original data from cohort and

longitudinalstudies;(2)theoutcomeofinterestwasclearly deﬁnedastheriskofoccurrence(ordeterioration)ofkidney diseaseafterantiviraltherapyamongHCV-positivepatients;

(3)theyprovidedadjustedriskestimatesandtheirconﬁdence intervals.Weconsideredbothcase-controlstudiesandcohort studiesaseligibleforinclusionintheanalysis.Ifdataonthe samepopulationwereduplicatedinmorethanonestudy,the mostrecentstudywasincludedintheanalysis.

IneligibleStudies

Studieswereexcludediftheyreportedinadequatedataonthe rateofkidneydiseaseintheHCV-infectedpopulationafter antiviraltherapytowardsHCV(e.g.,incompleteinformation onHCVstatusorantiviraltherapyorrenaloutcomes).Unpub- lishedstudies,studiesthatwere onlypublishedinabstract formorasinterimreportswereexcluded;lettersandreview articleswerenotconsideredforthissystematicreview.

QualityAssessment

The quality ofthe 15 studies was appraised using ascale adapted from the ‘Newcastle/Ottawa Scale (NOS)’¹¹. The Newcastle-Ottawa scale is a scoring system that assesses everyaspectofanobservationalepidemiologicstudyfroma methodological pointofview.When astudy includedrele- vantinformationthatcouldbeassociatedwiththeNOS,one pointwasadded.Sevenitemsincross-sectionalstudiesand eightitemsincohortandcase-controlstudiesthatcouldbe relatedtotheNOSwereidentiﬁed.Therefore,cross-sectional studiesassigned8-10,6-7,4-5,or0-3points(stars)wereevalu- atedasverygood,good,satisfactoryorunsatisfactorystudies, respectively.Similarly,cohort/case-controlstudieswith7-9,5- 6,4and0-3points(stars)wereidentiﬁedasverygood,good, satisfactoryorunsatisfactory,respectively.Wecarriedoutsub- groupanalysesbasedonthosestudiesprovidedwithverygood quality.Dataextractionandqualityscoringwereperformed independentlybytworeviewers(F.F.andV.D.)andtheresults weremergedbyconsensus.Thecompleteprotocolforquality scoringisavailableon-line(seeSupplementalFile2).

Outcomesmeasures

Theprimaryendpointwastoprovideadjustedestimatesof therisk(and95%CIs)ofkidneydiseaseamongHCV-infected populationafterantiviraltherapyforHCV.Multivariateanaly- siswasmadetoestimatetheindependenteffectofanti-HCV therapy on the frequency of kidney disease after adjust- mentforpotentialconfounders(covariates)(e.g.,age,gender, race/ethnicity, diabetes mellitus, and others). Cox propor- tionalhazardregressionanalysiswascarriedouttoassessthe effectofantiviraltherapyforHCVperseontheriskofkidney diseaseafteradjustmentfordifferentialfollow-uptimeand distributionofpotentialconfounders.

DataSynthesisandAnalysis

Weweightedthestudy-speciﬁcloghazardratiosbytheinverse of their variance to obtain a pooled effect estimate and its 95% conﬁdence intervals. For each study, we used the

(4)

Records identified through database searching (n=4472)

Additional records identified through other sources

(n=61)

Included Identification

Screening

Eligibility

Studies included in quantitative synthesis (meta-analysis), n=15 studies (13 articles)

Full-text articles assessed for eligibility (n=197)

Records screened (n=4533)

Records excluded (n=4336)

Full-text articles excluded (n=184) N=6 neurologic outcomes N=8 skin

N=9 diabetes mellitus N=9 cardiac outcomes N=63 mixed cryoglobulinemia N=81 glomerular disease N=8 others

(N=50 narrative reviews, N=8 systematic reviews, N=3 editorial comments)

Figure1–Flowdiagramofstudyselection.

estimate of the effect measure that was adjusted for the largestnumberofconfounders.Wepresentbothﬁxed-effects andrandom-effectspooledestimatesbutuseandreportthe latterwhenheterogeneitywaspresent.Weusedtherandom- effectsapproach,asdescribedbyDerSimonianand Laird¹², CochraneQ–testwasusedforquantifyingtheheterogeneity¹³. The I² statistic, which is the percentage of total variation acrossstudiesduetoheterogeneityratherthanchance,was also calculated¹⁴. The null hypothesis of this test is the absenceofheterogeneity.Weexploredtheoriginofhetero- geneitybystratiﬁedanalysisandmeta-regressiontechnique.

Subgroup or stratified analysis and meta-regression were pre-specified.Stratifiedanalysiswasmadebyrestrictingthe analysistosubgroupsofstudiesdefinedbystudycharacteris- tics(suchascountryoforigin,responsetoantiviraltherapy, andothers).Meta-regressionwasperformedinordertolook at simultaneously the effects of multiple (continuous and categorical) covariatesonthe outcome ofinterest.Weper- formedrandom-effectsmeta-regressionusingthemethodof momentsormaximumlikelihoodapproacheswhereappro- priate.Publication bias was assessedby the Egger test for funnel-plotasymmetry.Allanalysesweredonewiththesta- tisticalpackageComprehensiveMeta-Analysis(CMA),version 2.0(Biostat Inc., USA, 2005).The 5%significancelevel was adoptedfor␣risk.Everyestimatewasgivenwithits95%Con- fidenceIntervals.

Results

LiteratureReview

AsshowninFigure1,weretrieved4,533articlesand197full- textpaperswereassessedforeligibility.Thelistofthe 197 full-textpapersisreportedintheSupplementalFile3.Fifteen studiesmetourinclusioncriteria;thesewerepublishedin13 papers[Figure1]andcarriedoutin3continents(n=356,825

patients)^15–27.Thus,somestudiesreporteddataonmorethan onekidneydiseaseoutcome,buteachcohortwasrepresented once inanymeta-analysis. Therewas a100%concordance betweenreviewerswithrespecttoﬁnalinclusionandexclu- sionofstudiesreviewedbasedonthepredeﬁnedinclusionand exclusioncriteria.

We included studies where the diagnosis of HCV infection was done by testing for anti-HCV antibody in serum15,16,18,21,22,26,27.InformationonHCVserologicalstatus wascollectedatthetimeofenrolment.Surveysbasedonthe diagnosisofHCVinfectionbyadministrativedata(ICD-9-CM codes)werealsoconsidered17,19,20,23–25.

Someauthorsdeﬁnedthepresenceofkidneydiseaseby ICD-9-CM codes17,19,20,23–25; others adopted assessment of eGFR15,16,18,21,22,26,27.

ManystudiesadoptedtheCoxproportionalhazardmodel to examine the independent association of antiviral ther- apywithoutcomes;onestudyusedthemultivariablePoisson regression model²⁶. Conventional regression analyses were conductedintwopapers^16,18.Someauthorshaveperformed the propensity score methods to obtain better control for confounding19,24,25,27.

Patientcharacteristics

Tables1–4reportsomesalientdemographicandclinicalchar- acteristicsofsubjectsenrolledinthe includedstudies.The mean age ofpatient cohortsranged from 39.8to 58years.

Thegenderdistributionrangedfrom49.6%to97.1%male.Five studies werefrom Asia;sixand fourfrom U.S.andEurope, respectively.Theaveragefollow-uprangedbetween1.8to8.2 years.Table5gaveinformationonstudydesignandanti-viral therapy.

AslistedinTable6,therewaslargeheterogeneityinthe deﬁnitionoftheoutcome.Themajorityofstudiesaddressed theincidenceofkidneydiseaseafterantiviraltherapyamong patients withHCV infection15–17,19–27. One reportevaluated

(5)

Table1–Studiesincludedinthemeta-analysis(demographicandclinicalcharacteristicsatbaseline)(SVR-basedresults).

Authors Country Patients,n Follow-up,yrs Anti-HCVpositivepts,n Age,yrs SVR

AraseY,etal.,2011 Japan 650 6.5 650 57.4+11.7 210(32%)

BleM,etal.,2014 Spain 175 3.8(1-9) 175 55(33-84) 75(42.8%)

LeoneS,etal.,2016 Italy 340 NA 340 39.8 (35.8-43.5) 102(30%)

BerenguerJ,etal.,2017 Spain 1,625 5.4(3.5-7.1) 1,625 40(37-43) 628(38.6%)

MahaleP,etal.,2017(1) USA 31,143 5.1(2.9-7.2) 31,143 50.8 10,575(33.9%)

KovariH,etal.,2017 Switzerland 626 8.2 626 49(43-53)/50(45-55)* 345(55.1%)

SatapathyS,etal.2018 USA 204 5.5 204 55+7 145(71%)

∗ SVRvs.noSVR

Table2–Studiesincludedinthemeta-analysis(demographicandclinicalcharacteristicsatbaseline)(treatedvs.

untreated).

Authors Country Patients,n Follow-up,yrs Anti-HCVpositivepts,n Age,yrs

SatapathyS,etal.2012 USA 552 7 552 50+10.6

HsuY,etal.2014 Taiwan 2,822 3.78+2.2/3.59+2.1* 2,822 54.9+8/55+8*

ChenY,etal.,2015 Taiwan 8,747 3.2+2.1/4.7+1.8* 8,747 54+11/55.7+15*

HsuY,etal.,2015 Taiwan 37,152 3.3+2.4/3.2+2.4* 37,152 51.8/51.7*

MahaleP,etal.,2017(2) USA 158,097 5.1 158,097 50.8/52.3*

LinM,etal.,2017 Taiwan 2,583 3.8+2.7/3.6+2.6* 2,583 58.1+10/58.4+11

ParkH,etal,,2017(1) USA 55,818 1.8 55,818 52(47;57)/54(49;58)/56(51;60)/55(48;59)^**

ParkH,etal.,2017(2) USA 55,751 1.8 55,751 52(47;57)/54(49;58)/56(51;60)/55(48;59)^**

∗ Treatedvs.untreatedcohorts

∗∗ Patientsondual,triple,all-oraltherapy,andno-treatment,respectively

Table3–Studiesincludedinthemeta-analysis(demographicandclinicalcharacteristics).

Authors Gender,males Hypertension Diabetics,n Cirrhotics,n

AraseY,etal. 405(62.3%) 152(23%) 149(22.9%) 650(100%)

SatapathyS,etal.(2012) 377(68.3%) 217(39.3%) 105(19%) 33(5.9%)

HsuY,etal.(2014) 1,838(65%) 1,212(42.9%) 2,822(100%) 646(22.9%)

BleM,etal. 129(73.7%) 48(27%) 62(35%) 16/96(17%)

ChenY,etal. 4,343(49.6%) 3,825(43.7%) 2,897(33.1%) 1,535(17.5%)

HsuY,etal.(2015) 21,890(58.9%) 8,036(21.6%) 5,153(13.8%) 5,792(15.5%)

LeoneS,etal. 262(77.1%) 66(19.4%) NA 54(15.9%)

BerenguerJ,etal. 1,219(75%) NA 47(2.9%) 445(38.6%)

MahaleP,etal(1) 29,980(96.3%) 14,397(46.2%) 5,436(17.5%) 7,850(26.4%)

MahaleP,etal.(2) 156,253(97.1%) 82,081(51.5%) 33,135(20.7%) 37,842(24.8%)

LinM,etal. 1,937(63.2%) 1,829(59.7%) 1,295(42.2%) 340(11.1%)

ParkH,etal.(1) 33,667(60.3%) 20,954(37.5%) 9,136(16.3%) 2,459(4.4%)

ParkH,etal.(2) 33,667(60.3%) 20,954(37.5%) 9,136(16.3%) 2,459(4.4%)

KovariH,etal. 462(73.8%) NA NA NA

SatapathyS,etal.(2018) 138(68%) 85(42%) 52(25%) 0

Table4–Studiesincludedinthemeta-analysis(demographicandviralcharacteristics).

Authors HBV HIV Caucasian,n HCVgenotype1,n

AraseY,etal. 0 0 0 389(59.8%)

SatapathyS,etal.(2012) NA 42(7.6%) 22(4%) NA

HsuY,etal.(2014) 0 NA 0 NA

BleM,etal. NA NA NA 153(87%)

ChenY,etal. 0 NA 0 NA

HsuY,etal.(2015) 0 NA 0 NA

LeoneS,etal. 175(51.5%) 340(100%) 319(93.8%) 125(36.7%)

BerenguerJ,etal. 55(3.4%) 1,625(100%) NA 805(49.5%)

MahaleP,etal.(1) 321(1%) 736(2.4%) 17,075(54.8%) 21,498(69%)

MahaleP,etal.(2) 2,037(1.3%) 5,086(3.2%) 72,197(44.9%) 87,979(54.7%)

LinM,etal. 0 0 0 NA

ParkH,etal.(1) 1,193(2.1%) 1,187(2.1%) NA NA

ParkH,etal.(2) 1,193(2.1%) 1,187(2.1%) NA NA

KovariH,etal. 34(5.4%) 626(100%) NA 204(32.5%)

SatapathyS,etal.(2018) 0 0 149(73%) NA

(6)

Table5–Studiesincludedinthemeta-analysis(designandtherapy).

Authors Studydesign Outcome Population,type Antiviraltherapy,type

AraseY,etal. Retrosp,Co CKD Innercityresidents IFN(Recombinant-IFN␣2aor

IFN␣2b,Natural-IFN␣orIFN␤) aloneorwithribavirin SatapathyS,etal.(2012) Retrosp,Co CKD Metropolitanarearesidents IFN-basedtherapy HsuY,etal.(2014) CaseControl ESRD Nationalhealthinsurance

programUsers

Peg-IFNplusRBV

BleM,etal. Retrosp,Co CKD Livertransplantrecipients Peg-IFN␣2aorPeg-IFN␣2bplusRBV

ChenY,etal. CaseControl CKD Nationalhealthinsurance

programUsers

IFN(IFN␣,Peg-IFN␣-2a,Peg- IFN␣-2b)

aloneorplusRBV HsuY,etal.(2015) CaseControl ESRD Nationalhealthinsurance

programUsers

Peg-IFN(␣2aor2b) plusRBV

LeoneS,etal. Retrosp,Co CKD HIV/HCVco-infected

patients

Peg-IFN(orstandardIFN)plusRBV

BerenguerJ,etal. Prosp,Co ESRDandCRF HIV/HCVco-infected patients

Peg-IFN␣2aorPeg-IFN␣2bor StandardIFN␣plusRBV

MahaleP,etal.(1) Retrosp,Co GN Veterans AnyIFNaloneorplusRBV

MahaleP,etal.(2)

LinM,etal. Casecontrol ESRD Nationalhealthinsurance

programUsers

Peg-IFNaloneorplusRBV

ParkH,etal.(1) Retrosp,Co CKD Healthinsuranceprogram

Users

1)IFN␣,IFN␤

Peg-IFN␣-2a,Peg-IFN␣-2b (aloneorplusRBV) 2)pegIFN/RBVandDAAs 3)DAAswithorwithoutRBV

ParkH,etal.(2) GN

KovariH,etal. Prosp,Co CKD HIV/HCVco-infected

patients

1)PegIFNplusRBV

2)PegIFNplusRBVandDAAs SatapathyS,etal.(2018) Retrosp,Co ESRD Livertransplantrecipients 1)IFN-basedtherapy

2)IFN-basedtherapyplusDAAs 3)DAAs

AraseY,etal.[15]HRadjustedforage,gender,bodymassindex,eGFR,HCVviralload,HCVgenotype,AST,ALT,plateletcount,typeofIFN, uricacid,triglycerides,totalcholesterol,diabetesmellitus,hypertension,frequencyofcontrastimagingperyear,antiviralcombinationwith ribavirin

SatapathyS,etal.[16]ORadjustedforarterialhypertension,bodyweight,HCVRNA,age,gender,HIVstatus,diabetesmellitus,intravenous drugabuse,alcoholabuse,race,HCVgenotype1.

HsuY,etal.[17]HRadjustedforage,gender,arterialhypertension,hyperlipidemia,chronicobstructivelungdisease,peripheralarterialocclusive disease,cirrhosis,thyroiddisorder,aspirin,NSAID,ACEi,ARB,statin,metforminandinsulinconsumption

BleM,etal.[18]RRadjustedforage,gender,arterialhypertension,eGFR,diabetes,cirrhosis,donorage,liverﬁbrosis,timesinceEOT,CNIlevels 1yraftertreatment,ﬁbrosis

ChenY,etal.[19]HRadjustedforage,gender,comorbidities(diabetesmellitus,arterialhypertension,coronaryarterydisease,hyperlipidemia, cirrhosis),geographicregionofresidence,socio-economicstatus,urbanizationlevel,enrolleecategory,numberofmedicalvisits,Charlson comorbidityindexscore

HsuY,etal.[20]HRadjustedfordemographiccharacteristics(age,gender),comorbidities(arterialhypertension,hyperlipidemia,diabetes mellitus,chroniclungdisease,peripheralarterydisease,cirrhosis,thyroiddisorder),Charlsonindex,hospitallevel,useofaspirinandNAIDs LeoneS,etal.[21]HRadjustedforbaselinecovariates(age,calendaryear,modeofHIVtransmission,HCVgenotype,HBsAgstatus),time- updatedcovariates(bodymassindex,systolicbloodpressure,diastolicbloodpressure,cirrhosis,cholesterol,tryglicerides,CD4T-cellcount, HIVRNA,numberofAIDSevents,antiretroviraltherapyuse-abacavir,tenofovir,ritonavir,atazanavir,lopinavir,indinavir),diabetesmellitus, chronicvasculardisease

BerenguerJ,etal.[22]HRadjustedage,gender,previousAIDS-deﬁningconditions,HIVtransmissioncategory(injectiondruguseornot),nadir CD4T-cellcount,combinationantiretroviraltherapy,HIVRNAstatus,liverﬁbrosis-4score,HCVgenotype,andcumulativeexposuretoselected antiretroviraldrugs

MahaleP,etal.[23]HRadjustedforage,gender,race/ethnicity,periodofmilitaryservice,numberofoutpatientvisits,BMI,smoking,alcohol intake,arterialhypertension,diabetes

LinM,etal[24]HRadjustedforpropensityscore(includinggender,age,income,urbanizationlevel,hospitallevel,medicalhistory,medications) ParkH,etal.[25]HRadjustedforage,gender,diabetes,hypertension,dyslipidemia,chronicobstructivepulmonarydisease,coronaryartery disease,peripheralvasculardisease,cerebrovasculardisease,heartfailure,ACEIuse,ARBuse,contraindicationstoIFN-basedtherapy,hepatitis A,hepatitisB,alcohol/drugabusedisorders,cirrhosis,decompensatedcirrhosis,hepatocellularcarcinoma

KovariH,etal.[26]IRRadjustedforHIVacquisitioncategory,age,HIV-RNA,smoking,alcoholuse,activeintravenousdruguse,anddurationof HIVandHCVinfection,HIVtype1RNA

SatapathyS,etal.[27]HRadjustedforpropensityscore(includingageatlivertransplant,gender,race/ethnicity,eGFR3monthaftertransplant, bodymassindex,arterialhypertension,diabetesmellitus,proteinuriadegree,LDL-cholesterolvalue,useofmTORinhibitorortacrolimusfor immunosuppressionafterLT)

(7)

Table6–Allstudiesincludedinthemeta-analysis(outcomeanddeﬁnitioncriteria).

Authors Outcome Deﬁnition

AraseY,etal. CKD eGFR<60mL/min/1.73m²

SatapathyS,etal. CKD eGFR<60mL/min/1.73m²

orserumcreatinine>1.5mg/dL

HsuY,etal.(2014) ESRD Dialysisorrenaltransplant

BleM,etal. CKD eGFR<60mL/min/1.73m²

ChenY,etal. CKD Stage1-5CKD

HsuY,etal. ESRD Dialysisorrenaltransplant

LeoneS,etal. CKD eGFR<60mL/min/1.73m²

BerenguerJ,etal. Renalevents Dialysis,renaltransplant,chronicrenalfailure

(requiringdialysisornot)

MahaleP,etal.(1) Glomerulardisease *

MahaleP,etal.(2) Glomerulardisease

LinM,etal. ESRD Newonsetdialysis

ParkH,etal.(1) CKD Stage3-5CKD

ParkH,etal.(2) Glomerulardisease *

KovariH,etal. Kidneyevents Dialysisorrenaltransplant

SatapathyS,etal. CKD eGFR<30mL/min/1.73m²

∗ glomerulardiseasewasdefinedeitherbythehistologicfindingofglomerulonephritisonkidneybiopsy,orbythepresenceofatleasttwo ofthefollowingclinicalsigns:proteinuria,hematuria,andreducedestimatedglomerularfiltrationrate(eGFR)<60mL/min/1.73m2without alternativecauseforchronickidneydisease

theimpactofantiviraltherapyontheprogressionofkidney diseaseamongHCV-infectedindividuals¹⁸.

Thedeﬁnitionofkidney diseaseadoptedbytheauthors was chronickidney disease (n=7),end-stage renal disease (n=3), glomerular disease (n=3), and kidney events (n=2) (Table 6). Chronic kidney disease and end-stage renal dis- easewereassessedaccordingtotheKidneyDiseaseOutcomes QualityInitiative/KidneyDisease:ImprovingGlobalOutcomes classiﬁcation.

Summaryestimateofoutcome(SVR-basedstudies)

Somestudies(n=7)expressedtheirresultsaccordingtothe viralresponseafteranti-viraltherapy[Table7].Poolingstud- ies based onviral responses (n=7;34,763unique patients) demonstratedthattheriskofkidneydiseasewasgreaterin patientswho didnotobtain thanthose who obtainedsus- tainedviralresponseafteranti-HCVtreatment;thesummary estimateforadjustedriskwas2.5(95%CI,1.41;4.41)(P=0.0016) andbetween-studiesheterogeneitywasfound(PvaluebyQ test=0.004).AccordingtotheEgger’sregressionintercept,no publicationbiasoccurred(P=0.58).Thequalityscoresranged between4and8stars.

Summaryestimateofoutcome (Treatmentvs.No-treatmentstudies)

Eightstudies(n=333,312patients)comparedpatientcohorts whoreceivedvs.thosewhodidnotreceiveantiviraltreatment forHCV.Thesummaryestimateforadjustedriskoflowerinci- denceofkidneydiseaseacrosstheeightstudieswas0.39(95%

CI,0.25;0.612)(P=0.0001)[Table8].Testsforhomogeneityof theadjustedriskacrosstheeightstudiesgaveaQvalueof 52,608(P=0.0001),I²=86.6;thehomogeneityassumptionwas rejected.Thefunnelplotconcerningthepublicationbiasis

reportedinFigure2.TheEggertestshowedpublicationbias (P=0.015).Thequalityscoresrangedbetween5and8stars.

Stratiﬁedanalysisandmeta-regression

StratiﬁedanalysesareshowninTables7–8,nosubstantialdif- ferenceinpooledadjustedeffectestimatesacrossdesignswas foundevenifthehomogeneity assumptionwasrejectedin somesubsets.

Aggregation of studies comparing treated vs. untreated cohorts(n=7;321,522uniquepatients)reportedthattheinci- denceofkidneydiseasewaslowerinpatientswhoreceived antiviral therapy comparedto those who had notreceived it, the summaryestimate foradjusted HRof reducedinci- denceofkidneydiseasewas0.44(95%CI,0.25;0.63)(P=0.0001) (Figure3).Between-studiesheterogeneitywasnoted(Pvalue byQtestP=0.0001).AccordingtotheEgger’sregressionintercept,therewaspublicationbias(P=0.037).Figure4showsthe adjustedriskoflowerincidenceofCKDamongHCV-infected patientsreceivingIFN-basedtherapy(n=6reports).

Wemademeta-regressionanalysisinthesubsetofstudies (n=8,n=333,312uniquepatients)comparingpatientcohorts who received antiviral therapy for HCV vs. those who did not[Table 2]. Aslisted inTable 9, meta-regression demon- stratedasigniﬁcantimpactoffrequencyofcirrhosis(P=0.02), HBVinfection(P=0.0001),arterialhypertension(P=0.002),DM (P=0.051),andgender(male)(P=0.05)ontheoutcomeofinter- est(adjustedriskofreducedincidenceofrenaldiseaseamong HCV-positive patients after antiviral therapy ended). Other covariatesincludedinthemodel(age,durationoffollow-up, size,referenceyear)hadnoeffectontheattitudeofantiviral therapyinreducingthe incidenceofkidneydiseaseamong HCV-infectedpatients.

Figure5reportsthattheadjustedHRoflowerincidence ofkidneydiseaseafterantiviraltherapyamongHCV-infected

(8)

Table7–Summaryestimateforadjustedeffectestimateamongvarioussubgroupsofinterest(studiesbasedonviral response:No-SVRvs.SVR).

N Adjustedeffectestimate (random-effectsmodel)

Qvalue(by Chi-squaredtest)

I²

No-SVRvs.SVR (allstudies)

7 2.50(1.41;4.41) (P=0.0001)

19.07 (P=0.004)

14.2

StudiesadoptingHR 6 2.46(1.34;4.53)

(P=0.003)

18.6 (P=0.002)

19.2

Outcome:CKD(stage 3-5)

3 2.36(1.22;4.57) (P=0.001)

2.77(NS) 0.73

Outcome:CKD(stage 4-5)

2 6.88(2.75;17.2) (P=0.0001)

1.06(NS) 22.1

StudiesfromEurope 4 2.3(1.29;4.13)

(P=0.005)

2.76(NS) 12.3

StudiesfromUS 2 3.26(0.5;21.3)

(P=0.21)

12.9(P=0.003) 32.3

Studieswithquality score>6

2 1.36(1.04;1.78) (P=0.02)

0.9(NS) 12.5

Studieswithquality score<6

5 3.01(1.61;5.62) (P=0.005)

8.26(NS) 14.2

Table8–Summaryestimateforadjustedeffectestimateamongvarioussubgroupsofinterest(Treatmentvs.

No-treatmentstudies).

N Adjustedeffectestimate (random-effectsmodel)

Qvalue

(byChi-squaredtest)

I2

Treatmentvs.No-treatment (allstudies)

8 0.39(0.259;0.612) (P=0.001)

52.608 (P=0.0001)

86.69

StudiesadoptingaHR 7 0.44(0.25;0.63)

(P=0.0001)

47.5 (P=0.0001)

87.36

Case-controlstudies 4 0.24(0.15;0.39)

(P=0.0001)

6.06 (NS)

50.4

Outcome:CKDstudies 3 0.451(0.226;0.9)

(P=0.024)

6.47 (P=0.039)

69.1

Outcome:GNstudies 2 0.82(0.69;0.96)

(P=0.016)

2.12 (NS)

0.00

Outcome:ESRDstudies 3 0.213(0.127;0.35)

(P=0.0001)

3.87 (NS)

48.3

StudiesfromAsia 4 0.24(0.15;0.39)

(P=0.0001)

6.06 (NS)

50.4

StudiesfromUS 4 0.706(0.527;0.946)

(P=0.002)

7.736 (P=0.05)

61.2

StudiesusingICD-9code 7 0.44(0.25;0.63) (P=0.001)

47.5 (P=0.0001)

87.36

StudiesadoptingIFN-based therapy

6 0.29(0.14;0.59) (P=0.001)

51.4 (P=0.0001)

90.2

Studieswithquality score>6

6 0.459(0.29;0.72) (P=0.001)

35.09 (P=0.0001)

85.7

Studieswithquality score<6

2 0.29(0.19;0.43) (P=0.001)

0.78 (NS)

0.00

Studiesusingpropensity scorematchinganalysis

3 0.528(0.32;0.85) (P=0.01)

11.7 (P=0.008)

74.53

patients (whenadjusted forthe other covariates) becomes moreimportantwhenthefrequencyofdiabetesincreases.

Discussion

In the last decade, numerous authors have addressed the relationship between hepatitis C and kidney disease. The REVEAL-HCVcohort study,acommunity-based prospective study conducted in Taiwan, is the latest report on the topic-theadjustedprevalenceoddsratioofCKD(deﬁnedby

consecutiveproteinuriaoranestimatedglomerularﬁltration rateunder60mL/min/1.73m²)was4.99(95%CIs,2.25;11.06) forthosewithhighHCVRNAandgenotype2².Asmentioned above,HCVinfectionandCKDarecommonlyobservedina largeproportionoftheadultgeneralpopulation;itisdifﬁcult, however,toestablishwhetherasimpleassociationexistsor whetheradditionalpathogenicmechanismsdirectlyorindi- rectlylinkchronicHCVinfectiontoCKD.

Therecentpublicationofreportsevaluatingtheimpactof antiviral therapiesforHCVupon renalsurvivalgaveusthe opportunitytodeepenthecorrelationbetweenHCVandCKD.

(9)

Funnel plot of precision by log hazard ratio

14 12 10 8 6 4 2 0

-2,0 1,5 -1,0 -0,5 0,0 0,5 1,0 1,5 2,0

Log hazard ratio

Precision (1/std err)

Figure2–FunnelplotofprecisionbyLogRatio(cohortswhoreceivedantiviraltherapyvs.thosewhodidnot;n=8studies, n=333,312uniquepatients).

aHR and 95% CI=0.44 (0.25;0.63) (p=0.0001) aHR=1.0

Hsu y (2014) Chen y Hsu y (2015) Mahale p (2) Lin M Park H (1) Park H (2)

Case-control studies High-quality studies Studies (outcome GN) Studies (outcome ESRD)

Hazard ratios Studies based on

propensity score matching

0 1 2

Figure3–EstimatedHRand95%conﬁdenceintervalsforeachstudy(cohortswhoreceivedantiviraltherapyvs.thosewho didnot;n=7studies,n=321,522uniquepatients)

TheverticallinerepresentsthepooledaHRofriskofkidneydisease(random-effectsmodel)aftercomparison.

AdjustedHR0.44(95%CI,0.25;0.63)(P=0.0001).Heterogeneitystatistics:Q-value(byChi-squaredtest)=47.5,df=6, P=0.0001;I²=87.4%.

Thecurrentmeta-analysis(n=15studies,n=356,825patients) suggeststhatantiviraltherapypersefavoursrenalsurvival intheHCV-infectedpopulation.Thesummaryestimate for the adjustedrisk ofkidney disease was 0.43 (95% CI,0.26;

0.69)(P=0.0001)fromsurveys(n=8)comparingpatientswho receivedvs. those who didnotreceive antiviral treatment.

These conclusions appear reliable despite some between- studiesheterogeneityoccurredinsomecomparisons.

Thereisgrowingevidenceinmedicalliteraturesuggest- ing thatHCV infection isnot aliver-focuseddisease but a systemicillnessgivingseveralextra-hepatic (includingkid- ney)manifestations.Kidneysappearanimportanttargetof theextra-hepatic activity ofHCV;HCVcan giveglomerular

damagethroughactivationanddepositionofcryoglobulins²⁸. Also, HCV may cause tubulo-interstitial damage via a direct cytopathic effect or antibody immune complex²⁹. Additionally,non-immunologicalpathways(oxidativestress, pro-inﬂammatorycytokines,liverorperipheralinsulinresis- tance,amongothers)helpthedevelopmentofrenaldisease byvascularinjury^30–33.Endothelialdysfunctionatkidneylevel can bepromoted by viralreplication in theatherosclerotic plaque³⁴.

Thus,theresultsgeneratedbythissystematicreviewand meta-analysislendfurthersupporttothenotionthatHCVis ariskfactorfortheacquisition(orprogression)ofCKDinthe adultgeneralpopulation.Priorstudieshavefoundaconsis-

(10)

Adjusted HR 1.75

1.50

1.25

1

0.75

0.50

0.25

Mahale 2 (2017) Chen (2015) Hsu (2015) Hsu (2014) Satapathy

(2012)

Lin (2017)

Figure4–AdjustedriskoflowerincidenceofCKDafterthe endofIFN-basedantiviraltherapyamongHCV-infected patients(n=6studies)(Oddmanout).

Regression of Log hazard ratio on Diabetes

Diabetes

0,0 20,0 40,0 60,0 80,0 100,0 120,0

Log hazard ratio

1,50 1,00 0,50 0,00 -0,50 -1,00 -1,50 -2,00 -2,50 -3,00 -3,50 -4,00 -4,50

Figure5–Meta-regression:Regressionline(and95%CI)of LoghazardratioonDiabetes.

tentrelationshipbetweenanti-HCVpositiveserologicstatus andriskofCKDinthedevelopedworld;andthesehavebeen systematicallyreviewed⁴ but theactivity ofanti-HCVviral therapyontheoutcomeshasnotbeenreviewed.Theeffectof anti-HCVantiviraltherapyontheriskofrenaldiseaseinthe generalpopulationhasnotbeenreviewedtodate.

The ﬁndings obtained in the current meta-analysis are inkeeping withthose obtainedbyother investigatorswho

addressedtheinﬂuenceofHCVeradicationonextra-hepatic outcomes including chronickidney disease.Inaretrospec- tivecohortofpatientswithstage3CKD,regressionmodels reportedthatsustainedviralresponsewasassociatedwitha 9.3(95%CI,0.44to18)mL/minper1.73m²increaseineGFR duringthe6-monthpost-treatmentfollow-upperiod³⁵.

SomeauthorshavealreadysuggestedthatIFN-basedtreat- mentsofchronichepatitisCpromotealong-terminhibition ofoxidativestresswithconcomitantimprovementofnecro- inflammatory activity and fibrosis^36,37. In other words, it remains unclear whether the benefit on hepatic or extra- hepatic outcomes conferred from IFN can be attributed to the antiviral activity per se or to the immunomodula- tory/antiproliferative properties of IFN. The impact of the newerdirect-actingantiviralagentsremainstobeaddressed.

Theresultsfromthecurrentmeta-analysispresentsome limitations.First,manyreportsshowretrospectiveandcohort design-fromatheoreticalpointofview,arandomizedcon- trolledtrialwithplacebogivesthebestevidenceontheefficacy ofanintervention.However,alargesampleandalongfollow- upareneededtoperformaRCTinthissettingasthefrequency of events is low; also, the current availability of safe and effectivedrugs(DAAs)forthetreatmentofHCVmakesthe randomisationtoplacebonotethicallyacceptable.Secondly, multivariateanalysiswasperformedinallstudiesbutresidual confounding(confoundingremainingafteradjustment)can- notbeexcludedasfullinformationwasnotgivenonvarious confounders.DataonHCVvirology,lifestyle,illicitdrugabuse, andfamilyhistorywereoftenmissed.Thirdly,ourstratified analysisandmeta-regressionwasnotabletocapturefullythe heterogeneityamongstudieswehaveobserved.Thegreathet- erogeneityobservedinsomecomparisonssuggestedthatall thestudiesincludedintheanalysisarenotfunctionallyiden- ticalandthisprecludedtheadoptionofafixed-effectsmodel.

Moreover,individualdatafromeachstudy(‘meta-analysisat patientlevel’)werenotavailable;thus,itwasimpossibleto performourownadjustmentsevenifthestudiesenrolledin thismeta-analysisadjustedfornumerousfactors(‘covariates’) thatcouldprovetobepotentialconfounders.Finally,aswith allmeta-analyses,thisstudyhasthepotentiallimitationof publicationbias;underpublicationbias,onlyresultsshowing eitherstatisticalsigniﬁcanceand/orthedesireddirectionof theeffectsarepublished.Inthiscase,thepublishedliterature ismerelyaselective(andtoooptimistic)partofallexisting knowledge³⁸.Ourapproachtoavoidthepublicationbiasisto relyontheretrievalofdatafromasmanysourcesaspossible.

However,wehavenotincludedtrialspublishedasabstracts;

Table9–Meta-regressionanalysis:Impactofcontinuousvariablesonadjustedrisk*(n=8studies,n=333,312unique patients*(Randomeffectsmodel(MM),Z-Distribution,Loghazardratio).

Intercept RegressionCoefﬁcient Standarderror 95%CI Zvalue Pvalue

Intercept -1.8187 0.6105 -3.01;-0.622 -2.98 0.0029

Cirrhosis 0.0816 0.0354 0.012;0.151 2.3 0.0212

HBV 0.9066 0.2131 0.48;1.324 4.25 0

Males -0.0365 0.0192 -0.074;0.0001 -1.91 0.0568

Diabetics -0.0138 0.0071 -0.027;0.001 -1.95 0.051

Hypertension -0.0437 0.0145 0.0153;0.072 3.02 0.0025

∗ (AdjustedriskofreducedincidenceofrenaldiseaseamongHCV-positivepatientsafterantiviraltherapyended)

(11)

evidencepresentedinabstractformatisoftenwithouthigh qualityandcangivegreatertreatmenteffect³⁹.

In conclusion, this meta-analysis shows the reduction ofkidney complicationsamongHCV-infectedpatients who attainedviralresponseaftertherapy. Weneedstudies pro- vided with prospective design and uniform outcomes; the purposeofantiviraltherapyinthecurrenteraofDAAsisto treatnotonlyliverdiseasebutalsoextra-hepaticcomplica- tions(includingkidneyoutcomes),irrespectiveofliverdisease staging.Earlyinitiationofantiviraltherapytoreducetheexpo- sureofHCVamonginfectedindividualsisencouraged.

Funding

Nosourcesoffundingwereusedforthepreparationofthis manuscript

Conﬂict of interest statement

FabrizioFabrizi:consultantoradvisortoAbbVieandMerck&

Co

references

1. CourougeM,Vallet-PichardA,PolS.HCVandthekidney.

LiverInt.2016;36:S28–33.

2. LaiT,LeeM,YangH,YouS,LuS,WangL,etal.,forthe REVEAL-HCVStudyGroup.HighhepatitisCviralloadand genotype2arestrongpredictorsofchronickidneydisease.

KidneyInt.2017;92:703–9.

3. GlassockR,WarnockD,DelanaveP.Theglobalburdenof chronickidneydisease:estimates,variabilityandpitfalls.Nat RevNephrol.2017;13:104–14.

4. FabriziF,DonatoF,MessaP.AssociationbetweenhepatitisC virusandchronickidneydisease:Asystematicreviewand meta-analysis.AnnHepatol.2018;17:364–91.

5. GoossensN,NegroF.Cardiovascularmanifestationsof hepatitisCvirus.ClinLiverDis.2017;21:465–73.

6. NahonP,BourcierV,LaveseR,AudureauE,CagnotC, MarcellinP,etal.ANRSCO12CirVirGroup.Eradicationof hepatitisCvirusinfectioninpatientswithcirrhosisreduces riskofliverandnon-livercomplications.Gastroenterology.

2017;152:142–56.

7. AbyE,DongT,KawamotoJ,PisegnaR,BenhammouN.Impact ofsustainedvirologicresponseonchronickidneydisease progressioninhepatitisC.WorldJHepatol.2017;9:1352–60.

8. RossiC,SaeedS,CoxJ,VachonML,Martel-LafferiereV, WalmsleyS,etal.fortheCanadianCo-InfectionCohort Investigators.HepatitisCviruscuredosenotimpactkidney functiondeclineinHIV-coinfectedpatients.AIDS.

2018;32:751–9.

9. MoherD,LiberatiA,TetzlaffJ,AltmanD,PrismaGroup.

Preferredreportingitemsforsystematicreviewsand meta-analyses:thePRISMAstatement.JClinEpidemiol.

2009;62:1006–12.

10.PoynardT,ConnH.Theretrievalofrandomisedclinicaltrials inliverdiseasesfrommedicalliterature.Acomparisonof MEDLARSandmanualmethods.ControlClinTrials.

1985;6:271–9.

11.WellsG,SheaB,O’ConnellD,PetersonJ,WelchV,LososM, etal.TheNewcastle-Ottawascale(NOS)forassessingthe

qualityofnonrandomisedstudiesinmeta-analyses.Ottawa HealthResearchInstitutewebsite.Availableat:

http/www.ohri.ca/programs/clinicalepidemiological/oxford.asp.

Accessed16August2012.

12.DerSimonianR,LairdN.Meta-analysisinclinicaltrials.

ControlClinTrials.1986;7:177–88.

13.PetittiD.Approachestoheterogeneityinmeta-analysis.Stat Med.2001;20:3625–33.

14.HigginsJ,ThompsonS,DeeksJ,AltmanD.Measuring inconsistenciesinmeta-analysis.BrJMed.2003;327:557–60.

15.AraseY,SuzukiF,KawamuraY,SuzukiY,KobayashiM, MatsumotoN,etal.Developmentrateofchronickidney diseaseinhepatitisCviruspatientswithadvancedﬁbrosis afterinterferontherapy.HepatologyResearch.2011;41:946–54.

16.SatapathyS,LingisettyC,WilliamsS.Higherprevalenceof chronickidneydiseaseandshorterrenalsurvivalinpatients withchronichepatitisCvirusinfection.HepatolInt.

2012;6:369–77.

17.HsuY,LinJ,HoH,KaoY,HuangY,HsiaoN,etal.Antiviral treatmentforhepatitisCvirusinfectionisassociatedwith improvedrenalandcardiovascularoutcomesindiabetic patients.Hepatology.2014;59:1293–302.

18.BleM,AguileraV,RubinA,Garcia-ElizM,VinaixaC,PrietoM, etal.Improvedrenalfunctioninlivertransplantrecipients treatedforhepatitisCviruswithasustainedvirological responseandmildchronickidneydisease.LiverTransplant.

2014;20:25–34.

19.ChenY,HwangS,LiC,WuC,LinA.ATaiwanesenationwide cohortstudyshowsinterferon-basedtherapyforchronic hepatitisCreducestheriskofchronickidneydisease.

Medicine(Baltimore).2015;94:e1334.

20.HsuY,HoH,HuangY,WangH,WuM,LinJ,etal.Association betweenantiviraltreatmentandextra-hepaticoutcomesin patientswithhepatitisCvirusinfection.Gut.2015;64:495–503.

21.LeoneS,ProsperiM,CostarelliS,NastaP,MaggioloF,Di GiambenedettoS,etal.Incidenceandpredictorsof cardiovasculardisease,chronickidneydisease,anddiabetes inHIV/HCVco-infectedpatientswhoachievedsustained virologicalresponse.EurJClinMicrobiolInfectDis.

2016;35:1511–20.

22.BerenguerJ,Rodriguez-CastellanoE,CarreroA,Von WichmannM,MonteroM,GalindoM,etal.,theGESIDA HIV/HCVCohortStudyGroup.EradicationofhepatitisCvirus andnon-liver-relatednon-acquiredimmunedeﬁciency syndrome-relatedeventsinhumanimmunodeﬁciency virus/hepatitisCviruscoinfection.Hepatology.

2017;66:344–56.

23.MahaleP,EngelsA,LiR,TorresH,HwangL,BrownE,etal.

Theeffectofsustainedvirologicalresponseontheriskof extrahepaticmanifestationsofhepatitisCvirusinfection.

Gut.2018;67:553–61.

24.LinM,ChenT,LinW,LiuC,HsiehY,ChiuW,etal.Add-on neurologicalbeneﬁtsofantiviraltherapyinHCVpatients withchronickidneydisease-anationwidecohortstudy.BMC Gastroenterology.2017;17:99.

25.ParkH,ChenC,WangW,HenryL,CookR,NelsonD.Chronic hepatitisCincreasestheriskofchronickidneydiseasewhile effectivetreatmentdecreasestheincidenceofCKD.

Hepatology.2017Sep5,http://dx.doi.org/10.1002/hep.29505 [Epubaheadofprint].

26.KovariH,RauchA,KouyosR,RougemontM,CavassiniM, SchmidP,etal.fortheSwissHIVCohortStudy.HepatitisC infectionandtheriskofnon-liver-relatedmorbidityand mortalityinHIV-infectedpersonsintheSwissHIVCohort Study.ClinInfectDis.2017;64:490–7.

27.SatapathyS,JoglekarK,MolnarM,AliB,GonzalezH,Vanatta J,etal.Achievingsustainedvirologicresponseinliver

(12)

transplantrecipientswithhepatitisCdecreasesriskof declineinrenalfunction.LiverTranspl.2018;24:1040–9.

28.FabriziF,PlaisierE,SaadounD,MartinP,MessaP,CacoubP.

HepatitisCvirusinfection,mixedcryoglobulinemia,and kidneydisease.AmJKidneyDis.2013;61:623–37.

29.KasunoK,OnoT,MatsumoriA,NogakiF,KusanoH, WatanabeH,etal.HepatitisCvirus-associated

tubulointerstitialinjury.AmJKidneyDis.2003;41:767–75.

30.SaraﬁdisP,RuilopeL.Insulinresistance,hyperinsulinemia, andrenalinjury:mechanismsandimplications.AmJ Nephrol.2006;26:232–44.

31.StattermayerA,RutterK,BeinhardtS,ScherzerT,Stadlmayr A,HoferH,etal.AssociationoftheIL28Bgenotypewith insulinresistanceinpatientswithchronichepatitisC.J Hepatol.2012;57:492–8.

32.AdinolﬁL,ZampinoR,RestivoL,LonardoA,GuerreraB, MarroneA,etal.ChronichepatitisCvirusinfectionand atherosclerosis:clinicalimpactandmechanisms.WorldJ Gastroenterol.2014;20:3410–7.

33.MoucariR,AsselahT,Cazals-HatemD,VoitotH,BoyerN, RipaultP,etal.InsulinresistanceinchronichepatitisC:

associationwithgenotypes1and4,serumHCVRNAlevel, andliverﬁbrosis.Gastroenterology.2008;134:416–23.

34.BoddiM,AbbateR,ChelliniR,GiustiB,GianniniC,PratesiG, etal.HepatitisCvirusRNAlocalizationinhumancarotid plaques.JClinVirol.2010;47:72–5.

35.SiseM,BackmanE,OrtizG,HundemerG,UfereN,ChuteD, etal.Effectofsofosbuvir-basedhepatitisCvirustherapyon kidneyfunctioninpatientswithCKD.ClinJAmSocNephrol.

2017;12:1615–23.

36.ConjeevaramH,WahedA,AfdhalN,HowellC,EverhartJ, HoofnagleJ,Virahep-CStudyGroup.Changesininsulin sensitivityandbodyweightduringandafterpeginterferon andribavirintherapyforhepatitisC.Gastroenterology.

2011;140:469–77.

37.SerejoF,EmeritI,FilipeP,FernandesA,CostaM,FreitasJ, etal.OxidativestressinchronichepatitisC:theeffectof interferontherapyandcorrelationwithpathologicalfeatures.

ClinJGastroenterol.2003;17,664-650.

38.CrawfordJ,BriggsC,EngelandC.Publicationbiasandits implicationsforevidence-basedclinicaldecisionmaking.J DentalEducation.2010;74:593–600.

39.HaidichA.Meta-analysisinmedicalresearch.Hippokratia.

2010;14Suppl.1:29–37.