Am J Transplant. 2017;17:3159–3171. amjtransplant.com  | 3159© 2017 The American Society of Transplantation and the American Society of Transplant Surgeons

Received:10March2017  |  Revised:11July2017  |  Accepted:1August2017DOI: 10.1111/ajt.14459

O R I G I N A L A R T I C L E

Long- term outcome of renal transplantation from octogenarian donors: A multicenter controlled study

Piero Ruggenenti1,2 | Cristina Silvestre3 | Luigino Boschiero4 | Giovanni Rota5 |  Lucrezia Furian3 | Annalisa Perna2 | Giuseppe Rossini6 | Giuseppe Remuzzi1,2,7  |  Paolo Rigotti3

Abbreviations:aMDRD,abbreviatedModificationofDietinRenalDisease;BMI,bodymassindex;CI,confidenceinterval;CKD-Epi,CKDEpidemiologyCollaboration;ECD,expandedcriteriadonor;eGFR,estimatedglomerularfiltrationrate;ESKD,end-stagekidneydisease;GFR,glomerularfiltrationrate;HR,hazardratio;IQR,interquartilerange;IRCCS,IstitutodiRicoveroeCuraaCarattereScientifico;KDPI,kidneydonorprofile index;KDRI,kidneydonorrisk index;mTOR,mechanistictargetofrapamycin;NIT,NordItaliaTransplant;OPTN,OrganProcurementandTransplantationNetwork;UNOS,UnitedNetworkforOrganSharing.

GiuseppeRemuzziandPaoloRigotticontributedequallyaslastauthors.

1IRCCS–IstitutodiRicercheFarmacologicheMarioNegri,ClinicalResearchCenterforRareDiseasesAldoeCeleDaccò,Bergamo,Italy2NephrologyandDialysisUnit,AziendaSocioSanitariaTerritoriale(ASST)PapaGiovanniXXIII,Bergamo,Italy3KidneyandPancreasTransplantUnit,UniversityHospitalofPadua,Padua,Italy4KidneyTransplantationUnit,DepartmentofSurgery,AziendaOspedaliero-Universitaria(AOUI)diVerona,Verona,Italy5KidneyTransplantationCenter,UnitofPediatricSurgery,AziendaSocioSanitariaTerritoriale(ASST)PapaGiovanniXXIII,Bergamo,Italy6OrganandTissueTransplantImmunologyUnit,FoundationIRCCS“Ca’GrandaOspedaleMaggiorePoliclinico”,Milan,Italy7DepartmentofBiomedicalandClinicalSciences,UniversityofMilan,Milan,Italy

CorrespondenceGiuseppeRemuzziEmail:giuseppe.remuzzi@marionegri.it

Toassesswhetherbiopsy-guidedselectionofkidneysfromveryoldbrain-deaddo-norsenablesmoresuccessfultransplantations,theauthorsofthismulticenter,obser-vationalstudycomparedgraftsurvivalbetween37recipientsof1or2histologicallyevaluatedkidneysfromdonorsolderthan80yearsand198reference-recipientsofnon–histologically evaluated single grafts fromdonors aged60years and younger(transplantationperiod:2006-2013at3Italiancenters).Duringamedian(interquar-tile range) of 25 (13-42) months, 2 recipients (5.4%) and 10 reference-recipients(5.1%) requireddialysis (crude anddonor age- and sex-adjustedhazard ratio [95%confidence interval] 1.55 [0.34-7.12],P=.576and1.41 [0.10-19.54],P =.798, re-spectively). Shared frailty analyses confirmed similar outcomes in a 1:2propensityscore study comparing recipientswith74 reference-recipientsmatchedby center,year,donor,andrecipientsexandage.Serumcreatininewassimilaracrossgroupsduring84-monthfollow-up.Recipientshadremarkablyshorterwaitingtimesthandidreference-recipientsandmatchedreference-recipients(7.5[4.0-19.5]vs36[19-56]and40[24-56]months,respectively,P<.0001forbothcomparisons).Mean(±SD)kidneydonorriskindexwas2.57±0.32inrecipientsvs1.09±0.24and1.14±0.24in reference-recipients andmatched reference-recipients (P < .0001 for both com-parisons).Adverseeventsweresimilaracrossgroups.Biopsy-guidedallocationofkid-neys fromoctogenariandonorspermits furtherexpansionof thedonororganpoolandfasteraccesstoakidneytransplant,withoutincreasingtheriskofprematuregraftfailure.

K E Y W O R D S

clinicalresearch/practice,clinicaltrial,donorsanddonation:donorevaluation,graftsurvival,kidneytransplantation/nephrology,organallocation,organprocurement,organprocurementandallocation,pathology/histopathology

3160  |     RUGGENENTI ET al.

1  | INTRODUCTION

To reduce the progressively increasing gap between the number ofavailableorgansforkidneytransplantationandthenumberofpatientswho need transplants, in 2002 the American United Network forOrganSharing(UNOS)proposedincreasingthekidneydonorpoolbyconsideringkidneysfromexpandedcriteriadonors(ECDs).1However,theaverage3-yearsurvivalofECDkidneysis70%lowerthanthesur-vivalof“ideal”kidneysfromyoungdonors.2,3 Moreover, despite this policy,>40%ofECDkidneys recovered in theUnitedStatesduringthe past decade have never been transplanted.4Thisdiscardratecouldbereducedbyallocatingtodualtransplantationkidneysthatarecon-sideredunsuitableforasingletransplantationbecauseofatleast2ofthefollowingcriteria:(1)donorage>60years,(2)estimatedcreatinineclearance <65mL/min, (3) rising serum creatinine to >2.5mg/dL atthetimeoforganrecovery,(4)comorbiditiessuchashypertensionordiabetes,or(5)glomerulosclerosis>15%and<50%.5Thisstrategyre-ducedtheproportionofdiscardedkidneysto20%butwasassociatedwithexcesspatientdeathandgraftfailure.5Pretransplantationbiopsyevaluationwasnotmandatory, however, andpatients could receivekidneysthat,duetoseverestructuralchanges,couldfailprematurelyposttransplantation.6

In June 2013, the Organ Procurement and TransplantationNetwork(OPTN)approvedanewallocationpolicythatstratifiesde-ceaseddonorsaccordingtoakidneydonorprofileindex(KDPI)thattakes into account donor age, height,weight, ethnicity, history ofhypertensionanddiabetes,causeofdeath,serumcreatinine level,hepatitisCvirusstatus,anddonationaftercirculatorydeathstatus.7 Thisscaleisaimedtopredictthekidneydonorriskindex(KDRI)—thatis,thefailureriskofagraftfromagivendeceaseddonorcom-paredwiththefailureriskofagraftfromanaveragedonorofthepreviousyear.8Despite this novel approach, however, thediscardratesofkidneysrecoveredfor transplantationbyusingthekidneydonor profile index (KDPI) scale (18.3%) did not differ apprecia-bly from thediscard ratepreviouslyobservedduring theECDera(18.1%).9 Moreover,mostofthekidneysobtainedfromdonorswiththehighestKDPIvaluescontinuedtobediscarded.4

Notably,similarlytoECD,theKDPIscalehasbeenimplementedtopredicttheriskofgraftfailureonthebasisofclinicalparametersonly,independentofhistologicdatafromapretransplantationkid-neybiopsy.AccordingtoKDPI,theriskofprematurefailureofakid-neyfroma60-to70-year-oldwhitedonorwithhypertensionand/ordiabeteswhoisbraindeathfromcerebrovasculareventsispredictedtoexceedby2-to2.5-foldthefailureprobabilityofagraftobtainedfromanaveragedonoridentifiedduringthepreviousyear.7 In sharp contrast with these figures, we previously found that the short-term10andlong-term11outcomesofkidneysfromdonorsolderthan60yearsofage—inmostcases,withahistoryofhypertensionand/ordiabetes,renaldiseaseand/orcerebrovasculardeath—whichhadbeen systematically allocated for single or dual transplantation ordiscardedbasedonpredefinedstandardizedhistologiccriteria,weresimilar to thoseof kidneys fromyoungdonors selected accordingtostandardclinicalcriteria.Similarresultswereobtainedwhenthe

samebiopsy-guidedstrategywasextendedtodonorswhowere70or older.12Anotherstudy,however,foundthat2-yeargraftlossex-ceeded30%whenkidneys fromdonorsolder than75yearswereusedforsingleordualtransplantationaccordingtothebiopsyfind-ings.13 These data suggest that there is an upper donor age limitthat shouldnotbeexceeded toavoidanexcess riskof transplantfailure.Alternatively,datacouldbeexplainedbypoorly restrictivehistologic criteria used for kidney allocation to the single or dualtransplantation.13Toformallyaddressthis issue,wecomparedtheoutcomesof37recipientsof1or2histologicallyevaluatedkidneysfromdonorsaged80orolderwiththeoutcomesof198recipientsofsinglekidneysfromdonorsaged60oryounger,whichhadbeenallocatedaccordingtothesameNordItaliaTransplant(NIT)networkstandardcriteriabutwithoutpreimplantationhistologicevaluationandwereexpectedtohaveasurvivalratesimilartothatofkidneysfromanaveragedonor.

2  | MATERIALS AND METHODS

Thismatched cohort study14 involved235 consecutivepatients re-ferred to transplantation centers in Padua, Verona, and Bergamo(Italy) between2006and2013.Toassess the impactofdonorageontransplantationoutcomes,wecomparedtheoutcomesof37pa-tientswhohadreceived1or2kidneysfrombrain-deaddonorsaged80yearsorolderwhohadbeenhistologicallyevaluatedbefore im-plantation10-12(“recipients”)withtheoutcomesof198patientswhohadreceived1kidneyfrom“ideal”brain-deaddonorsaged60yearsoryoungerwhohadbeenselectedandallocatedfortransplantationaccordingtothesamestandardcriteriaoftheNordItaliaTransplant(NIT) network (“reference-recipients”) but without biopsy evalua-tion.15,16Tominimizethepotential roleofconfoundingfactors thatcould affect outcomedata in addition todonor (and recipient) age,theoutcomesofthe37recipientswerealsocomparedwiththeout-comesof74ofthe198reference-recipientswhohadbeenmatchedwith corresponding recipients by using a propensity scoremodel17 (“matchedreference-recipients”).

Allpatientsprovidedwritten informedconsent toundergorenaltransplantation according to theNITguidelines.15,16,18Patientswhowere to receivehistologically evaluated kidneysprovided additionalwritten consent to receiving either 1 or 2 kidneys, depending onthe results of a preimplantation biopsy. The biopsy-guided organselection and allocation programwas approved by the NIT organi-zation.18Thisprogramaimed toofferpatientsaged50orolder theoptionoforgantransplantationthatwasanadditiontotheoptionofthetransplantationofasinglekidneyfromayoungdonorwithoutapretransplantationbiopsy.Ateachrenaltransplantcenter,recipientsand reference-recipientswere treated, as per the centers’ standardprocedure,withasimilar immunosuppressiveprotocolcombiningin-ductiontherapywithbasiliximaband/orthymoglobulinsandmainte-nancetherapywithcalcineurininhibitorsand/ormechanistictargetofrapamycin(mTOR)inhibitors,mycophenolatemofetilorazathioprine,withorwithoutsteroids,andweremanagedbythesamesurgicaland

     |  3161RUGGENENTI ET al.

medicalteamaccordingtothesamestandardizedmonitoringprotocol.AccordingtoNITguidelines,thetransplantationwasperformedonlywhentheantidonorPRAtestwasnegative.Inallpatients,serumcre-atininelevelsweremeasuredaccordingtotheisotopicdilutionmassspectrometrystandardizedmethod.

The primary efficacy variable was graft function loss requiringchronic dialysis therapy. Secondary efficacy variables were a com-bined endpoint of need for dialysis or death and changes in serumcreatinine levels frommonth3posttransplantationtostudyend.Alldata, includingadverseevents,weremonitoredandrecordedbytheMonitoringUnit of theAldo eCeleDaccòClinical ResearchCenterof the IRCCS–MarioNegri Institute forPharmacologicalResearch,Bergamo.DatawereusedaccordingtoNITstandardregulationsfordataregistrationanduseandforthepreservationofpatients’anonym-ityandprivacy.15,16,18

2.1 | Donor and kidney evaluations

All potential brain-dead donors—including octogenarian donorsand the donors of reference-recipients and matched reference-recipients—were identified, screened, and selected according topredefinedprotocolsoftheNITnetworkbasedonthesamedemo-graphic, anthropometric, clinical, and laboratory parameters and aninstrumentalworkupthatincludedultrasoundassessmentofrenalpa-renchymaandurinarytractandanechocolor-Dopplerevaluationoftheaorta,renalvasculartree,andkidneyperfusion.15,16,18Computedtomography scanning or angiographic evaluations were consideredonly in selected cases with specific indication. Information aboutdonor ethnicity, comorbidities, kidney function, and cause of deathwasrecordedanddataweresubsequentlyusedtocalculatetheKDPIandKDRIofeachstudydonorbyusingtheOPTNcalculator(availableat https://optn.transplant.hrsa.gov/resources/allocation-calculators/kdpi-calculator/).

According to standard NIT guidelines, kidneys were discardedthat, during donor ultrasound evaluation or direct evaluation afterexplantation, showed severe macroscopic parenchymal abnormali-ties (including diffuse scarring or hypoplasia, or neoplasticmasses)or major vascular abnormalities (eg, severe, diffuse atheroscleroticchangesorcalcificationsoftheaortaorrenalvasculartreethatinthesurgeon’sjudgmentpreventedthepossibilityofsuccessfulanastomo-seswithrecipientarteryvesselsorwereathighriskofposttransplan-tationthrombosis).Detectionofurologicabnormalitiesconditioningobstructionthatcouldbeaddressedandofbenignrenalcystswasnota contraindication to transplantation. Eligible kidneys from donorsaged 60years oryoungerwere not evaluated histologically.Tissuesampleswereobtainedwitha16-gaugeneedlebysurgeonsfromtheinferiorpoleofbothkidneys fromoctogenariandonorsat the timeofbenchevaluation,werefixedinformalinandparaffinembedded,andwereevaluatedbygeneralpathologistswhowereoncallatpar-ticipating centers. Changes in vessels, glomeruli, tubules, and con-nectivetissue inbiopsyspecimenswerequantifiedonascale from0 (nochanges) to3 (severechanges)byusinga standardizedscorepredefinedbyaninternationalpanelofpathologists10-12thatreliably

predictedthescoringofthestructuralchangesobservedatautopticevaluationof thewholekidneys19andwasvalidated inprospectivecontrolledstudies.10-12Whenkidneyshadascorebetween0and3,theywereusedfor2singletransplantations.Whenonekidneyhadascorebetween0and3andtheotherkidneyhadascoreof4orgreater,orwhenbothkidneyshadascorebetween4and6, the2kidneyswere transplanted together into the same recipient. If onekidneyhadascorebetween4and6andtheotherkidneyhadascoreof7orgreater,orbothkidneyshadascoreof7orgreater,the2kid-neyswerediscarded.10-12Indualtransplantrecipients,the2kidneyswereimplantedthrough2bilateralincisionsorasingleincisionaspercenterpractice.Durationofsurgeryaveragedapproximately5hoursfor dual transplantations with bilateral incisions, 4 hours for dualtransplantationswithunilateralincisions,and2hours30minutesforsingletransplantations.

2.2 | Statistical analyses

Descriptivestatisticsat renal transplantation (baseline)of recipientsand reference-recipients were based on frequency and percentageanalysisforcategoricalvariablesandonmean(SD)ormedian(inter-quartilerange[IQR])valuesforcontinuousvariables.DonorandpatientbaselinecharacteristicswerecomparedbyusingthePearsonχ2 test, theFisherexacttestforcategoricalvariables,ortheWelch2-samplettestorWilcoxonrank-sumtestforcontinuousvariables.Normalityassumption was assessed by using the Shapiro-Wilk test. Kaplan-Meiercurveswereshownfordescriptivepurposesonly.Cumulativeincidenceswerecomparedwith theuseofhazard ratios (HRs), andtheir95%confidenceintervals(CIs)wereobtainedbyusingCoxre-gression analysis.Participantswhodidnot experience theeventofinterestwereright-censoredonthelastdayoftheobservationperiod.UnadjustedandadjustedHRswereobtainedtocomparesurvivaluntilthefirsteventbetweentherecipientsandreference-recipients.Theproportional hazards assumptionwas checked by using Schoenfeldresiduals.Foranalysesofposttransplantationrenalfunctionrecovery,serumcreatinine levelsofpatientswithgraft losswerecarried for-wardtotheendoftheobservationperiod.Thesameapproachwasusedforposttransplantationglomerularfiltrationrate(GFR)thatwasestimatedbyusingboththeabbreviatedModificationofDietinRenalDisease (aMDRD) and the CKDEpidemiology Collaboration (CKD-Epi)predictionequations.

Matched cohort analyseswere carried out by using a propen-sity score–based algorithm,17 identifying 74 of the 198 reference-recipients,whowerematchedwiththe37correspondingrecipientsinthecontextofa1:2matchedcohortdesign (matchedreference-recipients). Variables at the time of transplantation considered formatching included the transplant center, the year of transplanta-tion (±1yearvs corresponding recipients), donor sex, recipient sex,recipient age, mismatches, and donor:recipient body mass indexratios. Associations between baseline covariates and the group ofinterest (ie, recipients or reference-recipients) were obtained byusing logistic regression in the pooled baseline data.Next, a step-wise selection algorithm using the SASmacro “OneToManyMTCH”

3162  |     RUGGENENTI ET al.

was considered to reach the final propensity scoremodel. Survivalanalysiscomparing37recipientsandthecorresponding74matchedreference-recipientswascarriedoutwitha shared frailtymodelbyusingSTATA.20Two-sidedP-values<.05wereconsideredstatisticallysignificant.Recordeddatawereanalyzedwith theuseofSAS,ver-sion 9.2 (SAS Institute, Cary, NC) and STATA software, version 13(StataCorp,CollegeStation,TX),attheLaboratoryofBiostatisticsoftheClinicalResearchCenter.

2.3 | Role of the funding source

Thiswasa fullyacademic, internally fundedstudy.Nosponsorwasinvolvedinstudydesign,inthecollection,analysis,andinterpretationofdata,inthewritingofthereport,orinthedecisiontosubmitthemanuscriptforpublication.

3  | RESULTS

Eightof58potentialkidneydonorsaged80orolderwerenotconsid-eredforkidneydonationbecausesevererenalvascular,parenchymal,orurinarytractabnormalities(seeMaterialsandMethods)werede-tectedduringthestandardultrasoundevaluationthatisroutinelyper-formedinanypotentialkidneydonoraccordingtoNITguidelinesandindependentofdonorage(Figure1).Thus,100kidneyswererecov-eredfrom50octogenariandonors.Fourteenkidneysfrom7ofthesedonorswerediscardedbeforepretransplantationbiopsybecauseofsevereanddiffuseatheroscleroticchangesoftheaortaandrenalar-teryandtheramificationsfoundduringthemacroscopicevaluation,and12kidneysfrom6otherdonorswerediscardedbecausethebi-opsyevaluationrevealedseveretissuedamage.Thus,74kidneysfrom37 (74%) of 50 available donors were suitable for transplantation.Kidneysfrom33ofthesedonorswereallocatedtodualtransplanta-tionin33recipients.Kidneysfromtheremaining4donorswereallo-catedtosingletransplantation:4kidneysweretransplantedatstudycentersandtheoutcomesofthe4recipientswereconsideredhere,whereastheother4kidneysweretransplantedelsewhere.Therecipi-entsofthese4kidneyswerenotincludedinpresentanalysestoavoidtheconfoundingeffectofdifferentmonitoringandtreatmentstrate-gies at centers out of the studynetwork. Thus, 37 recipientswereavailableforcomparativeanalyseswith198reference-recipientsand74matchedreference-recipientswhoreceivedasinglekidneytrans-plantfromdonorsaged60oryoungeratthesamecentersandduringthe same observation period.

3.1 | Donor and patient characteristics

3.1.1 | Donors

Octogenariandonorswere34.4 ± 9.4and32.8 ± 7.0yearsolderthandonorsofreference-recipients(P<.0001)andmatchedreference-recipients (P<.0001), respectively (Table1).Octogenariandonorswere more frequently female and weighted less than donors of

reference-recipients, and their kidneys had longer cold ischemiatimesbefore implantation than thekidneys fromyoungerdonors.Otherparametersconsidered, includingethnicity,kidneyfunction,and the distribution of comorbidities and causes of death, weresimilar across groups; in particular, theywere verywell balancedbetween octogenarian donors and donors ofmatched reference-recipients, with the exception of the per-protocol difference indonor ages. Of interest, estimated GFR (by aMDRD or CKD-Epiequations)wasinthenormalrangeinmostofoctogenariandonorsand was only slightly, but nonsignificantly, lower than in donorsof reference-recipients ormatched reference-recipients (Table1).Noteworthy,KDPIwas100%in34ofthe37octogenariandonorsandrangedfrom97%to98%and99%intheother3octogenarians.Conversely, KDPIwas <95% in all donors of reference-recipientsandmatchedreferencerecipients,respectively (Figure2).Thedif-ferenceinKDPIdistributioninoctogenariandonorscomparedwiththeotherdonor groupswashighly significant (P < .0001 for both comparisons, Figure2).Consistently,meanKDPIwas significantlyhigherandKDRIwasmorethandoubleinoctogenariandonorsthanintheother2donorgroups(Table1,Figure3A),whereasthepro-portionofpatientsprogressingtoend-stagekidneydisease(ESKD)wasmuchthesameinthe3groups(Figure3B).

3.1.2 | Patients

Recipientswere17.9 ± 9.3 and 16.4 ± 8.7yearsolderthanreference-recipients (p<.0001)andmatchedreference-recipients (p<.0001)and had significantly more HLA mismatches compared with theircorresponding donors. Time on dialysis was significantly shorterforrecipientsthanforreference-recipientsandmatchedreference-recipients,andthisdifferencewasfullyexplainedbythesignificantlyshorter time on awaitlist for recipients comparedwith recipientsinthe2referencegroups.Consistently, timeondialysisbefore in-clusion inawaitlistwassimilar in the3groups (Table2,Figure4).Second transplantations tended to be less frequent in recipientsthaninreference-recipients;thisnonsignificantdifference,however,wasfullybluntedwhenmatchedreference-recipientswereconsid-ered for comparative analyses vs recipients. A significantly higherproportion of kidneys from female donorswere transplanted intomalepatientsamongrecipientscomparedwithreference-recipients.Consistently,thedonor:recipientbodyweightandmassindexratiosweresignificantlylowerforrecipientsthanforreference-recipients.Thesedifferenceswherebluntedinthecontextofthematchedco-hort comparisons. Recipients spent a 5- to 6-fold shorter time ona waitlist compared with patients in both control groups. Othercharacteristics, including follow-up duration,were similar in the 3groups.Distribution of causes of ESKDwas verywell comparablein particular between recipients andmatched reference-recipients(Table2).Distributionofmedicationsadministeredforinductionandmaintenanceimmunosuppressionwassimilarinthe3groupswithanonsignificanttrendtoa lessfrequentuseofcalcineurin inhibitorsand amore frequent use ofmTOR inhibitors in recipients than inreference-andmatchedreference-recipients(Table3).

     |  3163RUGGENENTI ET al.

3.2 | Graft and patient survival

Duringamedian(IQR)of25(13-42)months,2(5.4%)ofthe37recipientsand10(5.1%)ofthe198reference-recipientsrequiredchronicrenalre-placementtherapybydialysis(HR[95%CI]1.55[0.34-7.12],P=.576).

Outcomesweresimilar(HR[95%CI]1.41[0.10-19.54],P=.798)evenafter prespecified adjustment for donor age and sex (Figure5 andTable3).Thus,graftsurvivalwassimilarbetweengroupsdespiteKDPIandKDRI being significantly higher for octogenarian donors than foryoungdonorsofreference-recipients(Table1,Figure3).Tworecipients

F IGURE  1 Flowchartofselectionandallocationofkidneysfromoctogenariandonors

58 potential donors 80 years age or older identified

8 potential donors discarded for severe renalor vascular abnormalities at ultrasoundevaluation before explanation

26 kidneys from 13 donors discarded:- 14 for renal of vascular abnormalities at

pre-transplant evaluation- 12 for too severe histological changes at

pre-implantation kidney biopsy

50 potential donors eligible for kidney explantation

100 kidneys explanted

74 kidneys from 37 donors available for transplantation

66 kidneys from 33 donors allocated to dual transplantation into 33 recipients at the transplant Centers of Bergamo, Padua and Verona

8 kidneys from 4 donors allocated to single transplantation into 8 recipients

4 kidneys from 4 donors transplanted into 4 recipients at the transplant Centers of Bergamo, Padua and Verona

4 kidneys from 4 donors transplanted into 4 recipients in other Centers

37 recipients of single or dual transplants available for

outcome analyses

3164  |     RUGGENENTI ET al.

(5.4%)diedbytheageof63and73ofstrokeandsepsis,respectively.Fourreference-recipients(2.0%)diedbytheageof39,44,59,and62frommetastaticcoloncancer,sepsis,bacterialpneumonia,andlungcan-cer,respectively(Table3).Thus,4recipients(10.8%)and14reference-recipients(7.1%)requireddialysisordied(HR[95%CI]2.31[0.75-7.07],P=.144).Similarresultswereobtained(HR[95%CI]3.66[0.48-27.87],P=.210)whentheanalyseswereadjustedfordonorageandsex.

In addition to the 2 recipients, 4 (5.4%) of the 74 matchedreference-recipients also required dialysis (HR [95%CI] 1.25 [0.24-6.56], P=.795) (Figure6, Table3). Again, graft survival was similarbetweengroupsdespiteKDPIandKDRIbeingsignificantlyhigherforoctogenariandonorsthanfordonorsofmatchedreference-recipients(Table1, Figure3) Only 1 matched reference-recipient (1.3%) died,aged62oflungcancer.Thus,inadditiontothe4recipients,5(6.7%)matchedreference-recipientsalsorequireddialysisordied(HR[95%CI]2.07[0.57-7.49],P=.270).

3.3 | Posttransplantation kidney function recovery

At3months posttransplantation serum creatininewas higher in re-cipientsthaninreference-recipientsandmatchedreference-recipients(1.69 ± 0.68 vs 1.51 ± 0.56 vs 1.38 ± 0.41mg/dL, respectively),and the difference between recipients and matched reference-recipients was significant (P<.05). Consistently, estimated GFR(eGFR) (by aMDRD equation) was significantly lower in recipientsthan in reference and matched reference-recipients (46.89 ± 19.13 vs 55.42 ± 22.81vs57.99 ± 20.15, P < .05 and P<.01,respectively).However,onfollow-up,serumcreatininelevelswererelativelystablein recipients,whereas they tended toprogressively increase inbothreferencegroups (Figure7,AandB).However,differencesbetweengroupswereneversignificant.Consistently,eGFR(aMDRDandCKD-Epi)wasstableinrecipientsandprogressivelydeclinedinbothrefer-encegroups(datanotshown).

TABLE  1 Donor characteristics at the time of transplantation

Donors of recipients (n = 37)Donors of reference- recipients (n = 198)

Donors of matched reference- recipientsa (n = 74)

Age,y 82.0±2.1 47.6±10.0‡ 49.1±8.5‡

Agerange,y 80-86 16-60 19-60

Malesex,n(%) 10(27.0) 118(59.6)† 23(31.1)

Ethnicity,n(%)

White 37(100) 191(96.5) 72(97.3)

AfricanAmerican 0 1(0.5) 1(1.4)

Other/unknown 0 6(3.0) 1(1.4)

Comorbidities,n(%)

Hypertension 27(73.0) 45(22.7)‡ 18(24.3)‡

Diabetes 6(16.2) 7(3.5)* 4(5.4)

HepatitisCvirus 0 1(0.5) 0

Bodyweight,kg 66.9±10.0 76.4±19.8‡ 70.6±15.1

BMI,kg/m2 25.0±3.0 29.0±35.5 24.8±4.2

Serumcreatinine,mg/dL 0.83(0.24) 0.92(0.44) 0.86(0.52)

eGFR, mL/min/1.73 m2b 85.3±35.3 99.8±39.7 102.5±43.6

Coldischemiatime,h 16(13-18) 14(11-17)* 14(12-18)

Causeofdeath,n(%)

Cerebrovascular/stroke 28(75.7) 125(63.1) 53(71.6)

Headtrauma 8(21.6) 43(21.7) 15(20.3)

Anoxia 0 22(11.1) 5(6.8)

Centralnervoussystemtumor 0 2(1.0) 1(1.4)

Other/unknown 1(2.7) 6(3.0) 0

KDPIscore(%) 99.8(0.6) 55.6(21.0)‡ 60.1(20.8)‡

KDRIscore(%) 2.57(0.32) 1.09(0.24)‡ 1.14(0.24)‡

Donorcharacteristicsareaccordingtogroupsofpatientswhoreceived1or2histologicallyevaluatedkidneysfromdonors≥80y(“recipients”)or1non–histologicallyevaluatedkidney from≤60-ydonorsconsideredasawhole (“reference-recipients”)or in thecontextof the1:2matched-cohortdesign(“matchedreference-recipients”).aMatchingbythepropensityscoremodel.bAccordingtoabbreviatedModificationofDietinRenalDisease(aMDRD)equation.*P < .01, †P < .001, ‡P<.0001vsrecipientswithdonors≥80y.Dataaregivenasmean±SDormedian(IQR)orn(%).

     |  3165RUGGENENTI ET al.

3.4 | Safety

Therewererelativelyfewfatalandnonfatalseriousadverseevents,and they were distributed similarly across groups (Table3). The 5casesofgraft lossowingtochronicallograftnephropathywereob-servedinreference-recipients.

4  | DISCUSSION

Thisstudyindicatesthatthesurvivalofkidneygraftsrecoveredfromdonorsaged80orolderandselectedandallocatedforsingleordualtransplantationaccordingtobothstandardclinicalcriteriaandbiopsyfindingsbeforetransplantationwassimilartothatofsinglegraftsfromdonors aged 60 or younger selected and allocated based on stand-ard clinical criteria only. The 1:2 matched-cohort study comparing

theoutcomesofrecipientsofoldergraftswiththoseofrecipientsofyoungergrafts,whohadbeen identifiedandmatchedbasedonpre-definedcharacteristics(includingtransplantcenter,yearoftransplan-tation, donor and recipient sex, and recipient age), produced similarfindings.Notably,posttransplantationgraftandpatientsurvival,aswellasrenalfunctionrecovery,weresimilaracrossallconsideredgroups,despitegraftsfromoctogenariandonorsbeingalmost35yearsolderthanthose fromyoungerdonorsandrecipientsofoctogenariankid-neys being approximately 17years older than recipients of youngerkidneys.Notably,in34ofour37octogenariandonors,theKDPIwas100%. In the remaining3donors, it rangedbetween97%and99%.Consistently,theirKDRIexceededbymorethan2-foldtheKDRIofdo-norsaged60yearsoryounger.Thus,itisconceivablethataverysmallminority,probablynone,ofthesedonorswouldhavebeenconsideredfororganexplantationonthebasisoftheKDPI-basedkidneyallocationsystem.Thus,ourpresentfindingsprovidetheevidencethatextending

F IGURE  2 Distributionofdonorsofrecipients,reference-recipients,andmatchedreference-recipientsaccordingtotheirkidneydonorprofileindex(KDPI).Thedistributionwassignificantlydifferentbetweendonorsofrecipientsanddonorsoftheother2controlgroups(P < .0001 forbothcomparisons),whereasKDPIdistributiondidnotdifferbetweendonorsofreference-recipientsandmatchedreference-recipients

100

90

I ran

ge RecipientsReference-RecipientsMatched-Reference-Recipients

70

80

g to

KD

P

50

60

acco

rdin

30

40

don

ors

10

20

erce

nt o

f0

P

0-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-95 96-99 100

KDPI (%) ranges

F IGURE  3 Mean(SD)kidneydonorriskindex(KDRI)(A)predictedonthebasisofkidneydonorprofileindex(KDPI)andpercentageofpatientsprogressingtoend-stagekidneydisease(ESKD)duringthestudy(B)inrecipients,reference-recipients,andmatchedreference-recipients.KDRIwassignificantlyhigherforrecipientsthanforthe2othergroups(P < .0001 for both comparisons),whereastheincidenceofESKDwasidenticalamongthe3groups

p<0 0001

3.0 15.0p<0.0001

.A B

(KDRI)

2.5

KD

(%)

12.5

k In

dex

2.0 10.0

g to

ES

nor R

is

1 0

1.5gr

essi

n

5 0

7.5

idne

y D

o

0.5

1.0

ient

s pr

o

2.5

5.0

K

0

Pat

0Reci ients Reference- Matched-p

Recipients Reference-Recipients

3166  |     RUGGENENTI ET al.

abiopsy-guidedpolicyofgraftselectionandallocationtooldorveryolddonorsisexpectedtofurtherandsubstantiallyexpandthedonorpoolandthenumberoftransplantations,withoutaffectingthepooloforganspotentiallysuitablefor“standard”singletransplantationsand,atthesametime,withoutincreasingtheriskofprematuregraftfailure.

Actually,anincreasednumberofavailableorganstranslatedinto5-to6-foldshortermeanwaitlisttimeforrecipientsofkidneysfromocto-genariandonorscomparedwithrecipientsofyoungerkidneydonors.Thisfindinghasmajorimplications,becauserecipientsofkidneygrafts,includingthosefromolderdonors,havesubstantiallyreducedmortality

Recipients (n = 37)Reference- recipients (n = 198)

Matched reference- recipientsa (n = 74)

Age,y 65.7±6.2 47.8±9.8‡ 49.3±9.6‡

Agerange,y 54-81 23-76 27-76

Malesex,n(%) 26(70.3) 131(66.2) 47(63.5)

Ethnicity,n(%)

White 37(100) 190(96.) 69(93.2)

AfricanAmerican 0 7(3.5) 4(5.4)

Hispanic 0 1(0.5) 1(1.4)

CauseofESKD,n(%)

Glomerulardisease 17(45.9) 75(37.9) 31(41.9)

Hypertension/nephroangiosclerosis

3(8.1) 17(8.6) 10(13.5)

Diabetes 2(5.4) 4(2.0) 2(2.7)

ADPKD 8(21.6) 31(15.7) 10(13.5)

Congenital/interstitialdisease

2(5.4) 37(18.7) 9(12.2)

Systemicdisease 0 6(3.0) 1(1.4)

Other/unknown 5(13.5) 28(14.1) 11(14.9)

HLA-DRmismatchesvs donor, n

5(4-5) 4(3-4)‡ 4(4-5)*

Bodyweight,kg 72.6±13.3 69.7±13.6 69.2±12.5

BMI,kg/m2 25.0±3.2 24.3±4.0 24.2±3.8

Bodyweightratio(donor:recipient)

0.95±0.23 1.13±0.38† 1.04±0.29

BMIratio(donor:recipient) 1.02±0.19 1.23±1.58† 1.05±0.24

Donor-to-recipientsexmismatches, n/nb

18/2 48/35* 31/7

Timeondialysis,mo 18.0(11.0-33.0) 48.5(31.0-72.0)‡ 48.0(30.0-71.0)‡

Timeonwaitlist,mo 7.5(4.0-19.5) 36.0(19.0-56.0)‡ 40.0(24.0-56.0)‡

Timeondialysisbeforewaitlistinclusion,mo

11.5(2.0-18.5) 12.0(6.0-21.0) 14.0(6.0-22.0)

Dualtransplantations, n(%)

33(89.2) 0 0

Secondtransplantations, n(%)

0 30(15.1)# 3(4.0)

Follow-up,mo 24(12-36) 26(14-48) 25(18-48)

Characteristicsofpatientswhoreceived1or2histologicallyevaluatedkidneys fromdonors≥80y(recipients)or1non–histologicallyevaluatedkidneyfromideal≤60-ydonorsconsideredasawhole(reference-recipients) or in the context of the 1:2 matched-cohort design (matchedreference-recipients).ESKD,end-stagekidneydisease;BMI,bodymassindex;ADPKD,autosomaldominantpolycystickid-neydisease.aMatchingbythepropensityscoremodel.bFemale donor–to–male recipient/male donor–to–female recipient.*P < .01, †P < .001, ‡P<.0001vsrecipientswithdonors≥80y.Dataaregivenasmean±SDormedian(IQR)orn(%).

TABLE  2 Characteristicsofpatients

     |  3167RUGGENENTI ET al.

ratesand improved lifeexpectancycomparedwith transplantcandi-datesonmaintenancedialysistreatment.2Shorteningthewaitingtimeforatransplantmayalsohavemajorclinicalimplicationbecauseepro-longedtimeonawaitlistisoneofthestrongestmodifiableriskfactorsforpooroutcomeposttransplantation.21

All patients had a negative donor-specific PRA test at the timeof transplantation; the proportion of second transplantations wassimilaramonggroups (inparticularbetweenrecipientsandmatchedreference-recipients);andthenumberofHLA-DRmismatchesandthe

distribution of immunosuppressivemedicationswere also similar inthe3groups.Thus, study findingswereunlikelyconfoundedbydif-ferentimmunologicriskindifferentgroups.Withtheexceptionoftheexpectedagedifference,mainanthropometric,clinical,andlaboratorycharacteristicswerealsosimilarinthe3patientgroups.

Ourpresentfindingsareinharmonywithdatafromarecentretro-spectiveanalysisof442singleordualtransplantationsofkidneysfrommarginaldonorsthathadbeenselectedandallocatedonthebasisofapretransplantationdonorbiopsybyusingthesamehistologicscore

F IGURE  4 Totaltimeondialysis(left),timeonawaitlist(middle),andtimeondialysisbeforeinclusioninawaitlist(right)inrecipients,reference-recipients,andmatchedreference-recipients.Alldataaremean±SEM.Totalwaitingtimeandtimeonawaitlistweresignificantlylongerforreference-recipientsandmatchedreference-recipients(P < .0001 for both comparisons),whereastimeondialysisbeforeinclusiononawaitlistwassimilarinthe3groups

Total Time on Dialysis

Time on a Waiting List

Time on Dialysis BeforeInclusion in a Waiting List

p<0.0001

p<0.0001 p<0.0001

p<0.0001

60

40

50

30onths)

20(m

10

0Recipients Reference-

RecipientsMatched-Reference-Recipients

Immunosuppressive therapyRecipients (n = 37)

Reference- recipients (n = 198)

Matched reference- recipientsa (n = 74)

Induction,n(%)

Basiliximabalone 13(35.1) 119(60.1) 41(55.4)

Thymoglobulinalone 18(48.6) 49(24.7) 17(23.0)

Basiliximabpluslow-dosethymoglobulin

6(16.2) 27(13.6) 14(18.9)

None 0 2(1.0) 1(1.4)

Unknown 0 1(0.5) 1(1.4)

Maintenance

Steroids 34(91.9) 169(85.4) 61(82.4)

Cyclosporineortacrolimus 24(64.9) 185(93.4) 67(90.5)

Mycophenolatemofetilorazathioprine

25(67.6) 166(83.8) 63(85.1)

Sirolimusoreverolimus 24(64.9) 32(16.2) 11(14.9)

Belatacept 0 7(3.5) 5(6.8)

Dataaregivenasn(%).Inductionandmaintenanceimmunosuppressivetherapyofpatientswhoreceived1or2histologicallyevaluatedkidneysfromdonors≥80y(recipients)or1non–histologicallyevaluatedkidneyfromideal≤60-ydonorsconsideredasawhole(reference-recipients)orinthecontextofthe1:2matched-cohortdesign(matchedreference-recipients).aMatchingbythepropensityscoremodel.

TABLE  3  Inductionandmaintenanceimmunosuppressivetherapyofpatients

3168  |     RUGGENENTI ET al.

weoriginallyimplemented10andtestedinclinicalstudies10-12includ-ing the present one. Data showed that biopsy-guided allocation ofmarginalkidneyshalvedtherelativediscardrateandalloweda>25%absoluteincreaseintherateofrecoveryofkidneyswithaKDPIscore>80.22Thesefindingswereestimatedtocorrespondtoanoverallin-creaseintransplantationofapproximately4%consideringtheentiredonor pool.22Inthatstudy,however,donorageaveraged60yearsandonly5(1.1%)ofthedonorswereaged80orolder.AccordingtodataoftheOPTN/UNOSRegistry,88%ofkidneysfromdonorswithKDPIrangingfrom90%to100%andpotentiallyavailableforsingleordualtransplantationwerediscardedbetween2002and2012.4Thisisex-plainedbythefactthatselectionandallocationofkidneysfromthesedonorswithoutpreimplantationhistologicevaluationareexpectedtotranslateintoanunacceptableexcessriskofprematuregraftfailure.Notably,KDPIexceeded96%inallouroctogenariandonors.

On theotherhand, finding that75%ofkidneys recovered fromoctogenariandonorsweresuitable for transplantationconfirms thatbiopsy-guidedorganallocation isanefficientstrategy to further re-duce the number of discarded kidneys, evenwhenvery old donorsareconsidered.Thesebenefitslargelyoffsettheextratimeandcosts

required to select and allocate kidneys based on biopsy findings.Notably,almosthalfofthedonorswithunsuitablekidneyswereiden-tifiedbeforeexplantationthroughastandardscreeningprotocolthatisappliedbyanycenteroftheNITnetworkandisbasedontheuseofsimpleand inexpensiveprocedures, includinganabdominalultra-soundevaluation,whichiseasilyaccessibleinanyintensivecareunitandisakeycomponentofscreeningprotocolsofanypotentialdonor,independentofage, ineverydayclinicalpractice.Kidneys fromonly13—lessthanone-fourth—ofthe59consideredoctogenariandonorswereeventuallydiscardedafterexplantationbecauseofmacroscopicvascularabnormalitiesorhistologicchangesthatweretoosevere.

Indeed,anadditionaladvantageofpreimplantationbiopsyevalua-tionofolderkidneys10-12isthatitmayprotectpatientsfromreceivinggraftswithstructuralchangesthataretoosevereandmaybeassoci-atedwith the donor’s hypertension, diabetes, or other concomitantdiseasesorjustreflectrenalageing23andthatmaypredictpoorkid-ney survival.24 Thismay explainwhy posttransplantation functionalrecovery of octogenarian kidneyswas similar to that of young kid-neysfromidealdonorsandoldgraftsappearedtobeprotectedfromchronicallograftnephropathy.Indeed,chronicinjuryinrenalallografts

F IGURE  5 Kaplan-Meiersurvivalcurvesforprogressiontoend-stagekidneydiseaseduringa80-monthperiodin37recipientsof1or2kidneysfromdonorsaged80orolderallocatedbasedonpreimplantationhistologicevaluationandin198reference-recipientswhoreceived1kidneyfromdonorsaged60oryoungerthatwasnotevaluatedhistologicallybeforeimplantation.[Colorfigurecanbeviewedatwileyonlinelibrary.com]

37 20 6 2 2Recipients198 140 61 32 23Reference-Recipients

Patients at risk

0.00

0.25

0.50

0 20 40 60 80 Months

HRa (95% CI): 1.55 (0.34 to 7.12), p=0.576HRb (95% CI): 1.41 (0.10 to 19.54), p=0.798

Prop

ortio

n of

rena

l tra

nspl

ant

reci

pien

ts w

ith g

raft

loss

aCox unadjusted, bCox adjusted for donor age and gender.

Recipients

Reference-Recipients

F IGURE  6 Kaplan-Meiersurvivalcurvesforprogressiontoend-stagekidneydiseaseduringa80-monthperiodin37recipientsof1or2kidneysfromdonorsaged80orolder allocated based on preimplantation histologicevaluationandin74matchedreference-recipientsidentifiedbythepropensityscoremodelinthecontextofa1:2matched-cohortstudywhoreceived1kidneyfromdonorsaged60oryoungerthatwasnotevaluatedhistologicallybeforeimplantation.[Colorfigurecanbeviewedatwileyonlinelibrary.com]

37 20 6 2 2Recipients74 54 23 13 10Matched-Ref.-Recipients

Patients at risk

0.00

0.25

0.50

0 20 40 60 80 Months

HRa (95% CI): 1.25 (0.24 to 6.56), p=0.795

Prop

ortio

n of

rena

l tra

nspl

ant

reci

pien

ts w

ith g

raft

loss

aFrailty Cox

Recipients

Matched-Reference-Recipients

     |  3169RUGGENENTI ET al.

issustained,atleastinpart,byanimbalancebetweengraftfiltrationpowerandrecipientmetabolicdemand,whichresultsincompensatoryhyperfiltrationofglomerulisurvivingreperfusioninjury,rejection,anddrug toxicity.25Asobserved inexperimental26 andhuman27 chronic nephropathies characterized by reduced nephronmass, these (mal)adaptivechangesmayacceleraterenalfunctiondeteriorationandpro-gressiveglomerulosclerosisuptoterminalrenalfailure.25Theseeventscouldconceivablybeprevented,evenwhenkidneysfromextremelyolddonorsareused, thanks tobiopsy-guidedappropriatedosingofsuppliednephronswithsingleordualtransplantation.28Consistently,patientandgraftoutcomesinrecipientgroupsweresimilardespitethelargemajorityof femaledonors, themuchmore frequentallocationofkidneysfromfemaledonorstomalerecipients,andthelowerdo-nor:recipientbodyweightratioobservedinrecipientsofoctogenariankidneyscomparedwith the2controlgroups.These findingsappeartobeincontrastwithwell-establishedevidencethatshort-andlong-termgraftsurvivalisrelativelypoorwhenkidneysfromfemaledonorsare transplanted into male recipients29and/orallograftsizeissmallrel-ativetorecipientbodyweight.30Conceivably,theseadverseoutcomescanbeprevented ifappropriatenephronnumbers for transplant re-cipients are ensured by pretransplantation biopsy evaluation.Thesefindingsmayhaveimportantclinicalimplications,becausefemaledo-norsaccountforthemajorityofolderdonors,whereasmostpatientswaiting for a kidney transplant aremale.31The clinical relevanceofadequatenephrondosingishighlightedbyresultsofarecentimport-ant report32thatcomparedtheoutcomesofrecipientsofsingleordualkidneygrafts(histologicallyevaluatedinmost,butnotall,cases)fromECDdonorscategorizedaccordingtodonorage.Incontrastwithourpresentdata,graftsurvivalwasapproximately10%lowerinrecipientsofkidneysfromoctogenariandonors(65.9%)thaninrecipientsofkid-neysfromyounger(50to79years)donorsconsideredasawhole(ap-proximately75%).Alargepartofthisdifferencewasdrivenbyagraft

failureratethatapproximated50%inrecipientsofsingletransplantsfromoctogenariandonors.Independentofthepotentialroleofallo-cationof someorganswithoutprevioushistologicevaluation, thesedataweremostlikelyexplainedbythefactthatkidneyswithascoreof4wereallocatedtosingleratherthantodualtransplantations.10-12 Conceivably, thenephronmass suppliedwitha single,octogenariankidneywithascoreof4isnotsufficientforthemetabolicdemandoftherecipient,whichmaytranslateintomaximizedhyperfiltrationofre-sidualnephronsandacceleratedgraftfunctionexhaustion,uptoter-minalfailure.Ontheotherhand,allocationofthesekidneystoasingletransplantationmay further increase the organ pool and transplantoptionsforolderrecipients.Thus,themostperformantbiopsy-guidedallocationstrategytoenhancethenumberofsuccessfultransplanta-tionsfromoldandveryolddonorsshouldbetestedinthecontextofacontrolled,prospectivestudy.

4.1 | Safety

Only2 recipients of kidneys fromoctogenariandonors diedwith afunctioningkidneyaged63and72,respectively.Thus,themortalityratewasrelativelylowcomparedwithmortalityreportedinpreviousseriesofrecipientsofsingleordualECDgrafts.2,3The incidenceofnonfatal complicationswas also lowand similar indifferentpatientgroups,andnograftwaslostbecauseofcomplicationswiththepreim-plantationbiopsy,whichconfirmedthatbiopsy-guidedtransplantationof1or2kidneysfromextremelyolddonorstorelativelyoldrecipientswasasafeandwell-toleratedprocedure.Moreover,thecoldischemiatime(timebetweenprocurementoftheorganandtransplantation)ofoctogenariankidneysexceededthe ischemia timeofyoungkidneysby only 2 hours, a difference thatwas fully blunted in the contextofmatched-cohortcomparisons.Combined,thesefindingsunderlinethat no kidney fromanoctogenariandonorwasdiscardedbecause

F IGURE  7 Posttransplantationserumcreatininechangesduring84monthsoffollow-upin37recipientsof1or2kidneysfromdonorsaged80orolderallocatedbasedonpreimplantationhistologicevaluationandin198reference-recipients(top)ormatchedreference-recipientsidentifiedbythepropensityscoremodel(bottom)whoreceived1kidneyfromdonorsaged60oryoungerthatwasnotevaluatedhistologicallybeforeimplantation

Reference-Recipients

Recipients

Seru

m c

reat

inin

e(m

g/dl)

2.00

3.00

1.50

3.50

2.50

Recipients

Matched-Reference-Recipients

Seru

m c

reat

inin

e(m

g/dl)

2.00

3.00

1.50

3.50

2.50

1.00

1.00

3170  |     RUGGENENTI ET al.

ischemiatimewastoolong,indicatingthatevaluatingapreimplanta-tionbiopsyspecimeniscompatiblewithroutineorganprocurementandallocation.Thismayhaveclinicalimplicationsbecausethedura-tionofcoldischemiatimeisamajordeterminantingraftoutcomes,inparticularwhenmarginalkidneysareusedfortransplantation.2,3

4.2 | Limitations and strengths

Thiswasaprospectiveoutcomeanalysisofdatathat,however,hadbeen already retrospectively recorded for other (clinical) purposes.Thus,somedataconcerningthenumberofoctogenariansthatwerenot considered for donation and of sensitized patients or patientswith donor-specific antibodies aswell as information about dimen-sionsoftransplantedkidneyswerenotavailable.Proteinuriawasalsoreported inaminorityofcases.Moreover,becauseof the relativelysmallnumberofpatientsandshortfollow-up,theresultsmustbecon-sideredwithcaution.However,theycanpavethewayto largerandmorepowerfulstudiesaimedtooptimizetheuseofolderdonorstoin-creasetransplantactivitywithoutaffectingtransplantationoutcomes.Ontheotherhand,patientswere identified, treated,andmonitoredbasedonpredefinedandstandardizedprotocolsthatweresimilarforallconsideredcohorts.TheseprotocolsaresharedbyNITcentersandarethesameprotocolsthatareappliedtoanyaveragekidneytrans-plantrecipientineverydayclinicalpractice.Thisenhancesthegener-alizabilityofour findings to the realworld.Moreover,noadditional,time-consuming,andexpensivetestssuchascomputedtomographyscanning or angiographic evaluations were routinely required andwereperformedonlyon thebasisof specific indications,as foranyaveragedonor.Thus, theevaluationofoctogenariandonorsdidnotimply extrahumanwork and costs that couldbedirectly related tothedonorage.Ontheotherhand,theuseofoctogenariandonorsisprogressively increasingwith encouraging results also in liver trans-plantation.33-36Thus,octogenariandonorscouldbeevaluatedforbothkidney and liver transplantation.Thiswill further increase the cost-effectiveness of organ procurement fromvery old donors.Of note,octogenariandonorsandtheiroldrecipientshadarelativelylowprev-alenceofclassiccardiovascular(andrenal)riskfactorssuchasobesity,diabetes,orhypertension.Thismost likelyexplainedtheir longevity.Conceivably,thisalsoexplainedwhykidneyfunctionofoctogenariandonorswasalmostsimilartothatofidealdonorsandwhyrecipientsofoctogenariankidneyshadaposttransplantationrateofcardiovasculareventsthatwassimilartothatofyoungerrecipients.Thisfurtheren-hancesthecost-effectivenessofdonorpoolexpansionwiththeuseofoctogenariandonors.

Biopsysamplescanbeevaluatedbyageneralpathologistwithoutspecifictraininginrenalpathology,whichenhancesthefeasibilityoftheprocedure.Theuseofapropensityscoremodelinthecontextofamatched-cohortdesignlimitedtheroleofpotentialconfoundingfac-tors.Thesamplesizewasnotcalculatedaprioribasedonanexpecteddifferenceacross groups in theprimaryoutcomevariable; however,thestudysizewassimilartothatofstudiesreportedpreviously inasimilarcontext.10-12Long-termfollow-upandcarefulpatientmonitor-inganddatarecordingweremajorstrengths.

5  | CONCLUSIONS

Kidneysfromdonorsaged80oroldercanprovideexcellentgraftsurvivalandrenal function recovery forup to7yearsafter trans-plantation,providedtheyareallocatedassingleordualtransplantsaccording to biopsy findings before transplantation and that kid-neyswithmoresevere,chronicchangesarediscarded.Thesedataconfirm that there should not be any predefined upper age limitto screening and evaluation of a potential brain-deceased kidneydonor.Thestudyfindingshighlightasimpleandsafeprocedurethatcanenablefurtherexpansionofthedonororganpool toenhancetheopportunitiesforsuccessfultransplantation,evenforrecipientsofkidneysfromextremelyolddonors,andatthesametimeensurethatthedualtransplantationprocedureisrestrictedtoorgansthatarenotsuitableforsingletransplantationsandwouldotherwisebediscarded. Whether transplant activity can be further optimizedwith an integrated use of the biopsy score and the KDPI/KDRIscalesisworthinvestigating.

ACKNOWLEDGMENTS

TheauthorsthankNadiaRubisandAlessandroVillafordatamoni-toring,GiovanniAntonioGiuliano forhishelpwithdatahandling,AntoniettaChiancaforhercontributiontoimplementingthefrailtymodels for the matched-cohort study, and Angela Russo for hercontributiontotheiconographyofthemanuscript.TheauthorsarealsogratefultoAnnaritaPlatiandallthephysiciansandnurseswhowereinvolvedinpatientcareandmonitoringatthestudycenters,aswell as the pathologists and technicianswhowere involved inprocessingandevaluatingthekidneybiopsysamples.ThestaffoftheNord ItaliaTransplantnetworkorganizeddonor identificationandkidneyallocation.

DISCLOSURE

TheauthorsofthismanuscripthavenoconflictsofinteresttodiscloseasdescribedbytheAmerican Journal of Transplantation.

AUTHOR CONTRIBUTIONS

Paolo Rigotti, Giuseppe Remuzzi, and Luigino Boschiero had theoriginal idea.PieroRuggenentiwrotetheinitialdraftandthefinalversion of themanuscript and contributed to data interpretation.Annalisa Perna performed the statistical analyses. Paolo Rigotti,CristinaSilvestre,LucreziaFurian,LuiginoBoschiero,andGiovanniRota performed the transplantations and contributed to patientcareanddatarecording.GiuseppeRossinicontributedtoorganal-locationanddatarecording.Alltheauthorshadfullaccesstodata,critically revised the manuscript, and approved the final version.Giuseppe Remuzzi is the corresponding author and had final re-sponsibilityforthedecisiontosubmitthepublication.Nomedicalwriterwasinvolved.Thestudywasfundedinternally,andnospon-sororcompanywasinvolvedinconductingthestudy.

     |  3171RUGGENENTI ET al.

ORCID

Giuseppe Remuzzi http://orcid.org/0000-0002-6194-3446

REFERENCES

1. RosengardBR,FengS,AlfreyEJ,etal.ReportoftheCrystalCitymeet-ingtomaximizetheuseoforgansrecoveredfromthecadaverdonor.Am J Transplant.2002;2:701-711.

2. Ojo AO, Hanson JA, Meier-Kriesche H, et al. Survival in recipi-ents of marginal cadaveric donor kidneys compared with otherrecipientsandwait-listed transplant candidates. J Am Soc Nephrol. 2001;12:589-597.

3. AubertO,KamarN,VernereyD,etal.Longtermoutcomesoftrans-plantationusingkidneysfromexpandedcriteriadonors:prospective,populationbasedcohortstudy.BMJ. 2015;351:h3557.

4. Tanriover B, Mohan S, Cohen DJ, et al. Kidneys at higher risk ofdiscard: expanding the role of dual kidney transplantation. Am J Transplant.2014;14:404-415.

5. GillJ,ChoYW,DanovitchGM,etal.OutcomesofdualadultkidneytransplantsintheUnitedStates:ananalysisoftheOPTN/UNOSda-tabase. Transplantation.2008;85:62-68.

6. RuggenentiP,PericoN,RemuzziG.Waystoboostkidneytransplantviability:arealneedforthebestuseofolderdonors.Am J Transplant. 2006;6:2543-2547.

7. https://optn.transplant.hrsa.gov/resources/allocation-calculators/kd-pi-calculator/risk.AccessedJuly29,2016.

8. Rao PS, Schaubel DE, Guidinger MK, et al. A comprehensive riskquantification score for deceased donor kidneys: the kidney donorriskindex.Transplantation.2009;88:231-236.

9. BaeS,MassieAB,LuoX,AnjumS,DesaiNM,SegevDL.Changesindiscardrateaftertheintroductionofthekidneydonorprofileindex(KDPI).Am J Transplant.2016;16:2202-2207.

10. Remuzzi G, Grinyo J, Ruggenenti P, et al. Early experience withdual kidney transplantation in adults using expanded donor cri-teria. Double Kidney Transplant Group (DKG). J Am Soc Nephrol. 1999;10:2591-2598.

11. Remuzzi G, Cravedi P, PernaA, et al. Long-term outcome of renaltransplantation from older donors. N Engl J Med.2006;354:343-352.

12. RigottiP,EkserB,FurianL,etal.Outcomeof renal transplantationfromveryolddonors.N Engl J Med.2009;360:1464-1465.

13. ColliniA,KalmarP,DhamoA,RuggieriG,CarmelliniM.Renaltrans-plant from very old donors: how far can we go? Transplantation. 2009;87:1830-1836.

14. http://people.vetmed.wsu.edu/jmgay/courses/GlossClinStudy.htm#-ClinicalStudyTypes.AccessedSeptember4,2017.

15. NITp. Il Nord Italia Transplant program. Milano, 30 giugno 2006.http://cm.argonet.it/websites/policmi/staging/home_ctit.nsf/wDoc-Tree/IDCW-5UDFQ4?opendocument.AccessedSeptember4,2017.

16. CentroInterregionalediRiferimento.Noteinformativealpazienteinattesa di trapianto di rene e rene-pancreas. http://www.policlinico.mi.it/nitp/area_paziente/trapianti_carte_servizi/1202rene_rene-pancreas.pdf.AccessedSeptember4,2017.

17. Barnieh L, James MT, Zhang J, Hemmelgarn BR. Propensity scoremethods and their application in nephrology research. J Nephrol. 2011;24:256-262.

18. PierobonES,SandriniS,DeFazioN,etal.Optimizingutilizationofkidneys from deceased donors over 60 years: five-year outcomesafterimplementationofacombinedclinicalandhistologicalallocationalgorithm.Transpl Int.2013;26:833-841.

19. MazzuccoG,MagnaniC,FortunatoM,TodescoA,MongaG.There-liability of pre-transplant donor renal biopsies (PTDB) in predictingthekidneystate.Acomparativesingle-centrestudyon154untrans-plantedkidneys.Nephrol Dial Transplant.2010;25:3401-3408.

20. Gutierrez RG. Parametric frailty and shared frailty survivalmodels.Stata J.2002;2:22-44.

21. Meier-KriescheHU,KaplanB.Waitingtimeondialysisasthestron-gest modifiable risk factor for renal transplant outcomes: a paireddonorkidneyanalysis.Transplantation.2002;74:1377-1381.

22. GandolfiniI,BuzioC,ZanelliP,etal.TheKidneyDonorProfileIndex(KDPI) of marginal donors allocated by standardized pretransplantdonorbiopsyassessment:distributionandassociationwithgraftout-comes. Am J Transplant.2014;14:2515-2525.

23. GlassockRJ, RuleAD.Aging and the kidneys: anatomy, physiologyand consequences for defining chronic kidney disease. Nephron. 2016;134:25-29.

24. KosmoliaptsisV,SaljiM,BardsleyV,etal.BaselinedonorchronicrenalinjuryconfersthesametransplantsurvivaldisadvantageforDCDandDBDkidneys.Am J Transplant.2015;15:754-763.

25. BrennerBM,MilfordEL.Nephronunderdosing:aprogrammedcauseofchronicrenalallograftfailure.Am J Kidney Dis.1993;21:66-72.

26. AndersonS,MeyerTW,RennkeHG,BrennerBM.Controlofglomer-ularhypertension limitsglomerular injury inratswithreducedrenalmass. J Clin Invest.1985;76:612-619.

27. RuggenentiP,CravediP,RemuzziG.Mechanismsand treatmentofCKD.J Am Soc Nephrol.2012;23:1917-1928.

28. MackenzieHS,TulliusSG,HeemannUW,etal.Nephronsupplyisamajordeterminantoflong-termrenalallograftoutcomeinrats.J Clin Invest.1994;94:2148-2152.

29. Neugarten J, Srinivas T, Tellis V, Silbiger S, Greenstein S. The ef-fect of donor gender on renal allograft survival. J Am Soc Nephrol. 1996;7:318-324.

30. NicholsonML,WindmillDC,HorsburghT,HarrisKP.Influenceofal-lograftsizetorecipientbody-weightratioonthelong-termoutcomeof renal transplantation. Br J Surg.2000;87:314-319.

31. https://optn.transplant.hrsa.gov/data/view-data-reports/nation-al-data/#.AccessedJune7,2016.

32. MessinaM,DienaD,Dellepiane S, et al. Long-termoutcomes anddiscardrateofkidneysbydecadeofextendedcriteriadonorage.Clin J Am Soc Nephrol.2017;12:323-331.

33. Ghinolfi D, Marti J, De Simone P, et al. Use of octogenarian do-nors for liver transplantation: a survival analysis. Am J Transplant. 2014;14:2062-2071.

34. GastacaM,GuerraM,AlvarezMartinezL,etal.Octogenariandonorsin liver transplantation. Transplant Proc.2016;48:2856-2858.

35. Chedid MF, Rosen CB, Nyberg SL, Heimbach JK. Excellent long-termpatientandgraftsurvivalarepossiblewithappropriateuseofliversfromdeceasedseptuagenarianandoctogenariandonors.HPB (Oxford).2014;16:852-858.

36. DariusT,MonbaliuD, Jochmans I, et al. Septuagenarian and octo-genarian donors provide excellent liver grafts for transplantation.Transplant Proc.2012;44:2861-2867.

How to cite this article:RuggenentiP,SilvestreC,BoschieroL, etal.Long-termoutcomeofrenaltransplantationfromoctogenariandonors:Amulticentercontrolledstudy.Am J Transplant. 2017;17:3159–3171. https://doi.org/10.1111/ajt.14459