Acute kidney injury patterns in acute heart failure

(1)

www.revportcardiol.org

Revista Portuguesa de

Cardiologia

Portuguese Journal of Cardiology

ORIGINAL ARTICLE

Acute kidney injury patterns in acute heart failure:

The prognostic value of worsening renal function and its timing

João Presume

^a^,^b^,∗

, Gonc ¸alo J.L. Cunha

^b

, Bruno M.L. Rocha

^b

, Luís Landeiro

^a

,

Q1

Sara Trevas

^a

, Marta Roldão

^a

, M. Inês Silva

^a

, Margarida Madeira

^a

, Sérgio Maltês

^a^,^b

, Catarina Rodrigues

^a

, Inês Araújo

^a

, Cândida Fonseca

^a^,^c

aHeartFailureClinic,InternalMedicineDepartmentIII,HospitaldeSãoFranciscoXavier,CentroHospitalarLisboaOcidental, Q2

Lisbon,Portugal

bCardiologyDepartment,HospitaldeSantaCruz,CentroHospitalarLisboaOcidental,Lisbon,Portugal

cNOVAMedicalSchool,FaculdadedeCiênciasMédicas,UniversidadeNovadeLisboa,Lisbon,Portugal

Received2November2021;accepted2June2022

KEYWORDS Acuteheartfailure;

Q3

Acutekidneyinjury;

Cardiorenal syndrome;

Worseningrenal function

Abstract

Introduction:Acute decompensated heart failure (ADHF)admissionsare frequently compli- cated by different patterns of serum creatinine (SCr) elevation. We aimed to assess the prognosticimpactofworseningrenalfunction(WRF)basedonthetimingofitsoccurrence.

Methods:Thiswasaretrospectivecohortofpatients admittedforADHF.StandardWRFwas definedasanincreaseinSCrof≥0.3mg/dlduringhospitalization.WRFtimingwasclassifiedas early(within48hoursofadmission)orlate(>48hours).Acutekidneyinjury(AKI)atadmission was defined asa riseinSCr of≥0.3mg/dl fromoutpatient baseline measurementto first measurementatadmission.Theprimaryendpoint wasacompositeofall-causemortalityor hospitalizationforcardiovasculareventsatone-yearfollow-up.

Results:Overall,249patientswereincluded(meanage77±11years,62%withpreservedleft ventricularejectionfraction).EarlyWRFoccurredin49patients(19.7%)andwasassociated withahigherriskoftheprimaryoutcome(HR2.49;95%CI1.66---3.73),whereaslateWRFwas not(p=0.411).AfterstratiﬁcationforthepresenceofearlyWRFand/orAKIatadmission,only patientswithearlyWRFbutnoAKIatadmissionandpatientswithbothAKIatadmissionand earlyWRFshowedahigherriskoftheprimaryoutcomeaftermultivariateCoxregression.

∗Correspondingauthor.

E-mailaddress:[email protected](J.Presume).

https://doi.org/10.1016/j.repc.2022.06.015

3

4

5

6

7

8 9 10

11 12 13 14

15

16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

32

(2)

Conclusion: EarlyWRFwasassociatedwithahigherriskoftheprimaryoutcome.Thetiming ofWRFseemstobeanimportantfactortotakeintoaccountwhenconsideringtheprognostic impactofcreatininevariationsduringhospitalizationforADHF.

Padrõesdelesãorenalagudanainsuﬁciênciacardíacaaguda:ovalorprognósticodo agravamentodafunc¸ãorenaleoseutiming

Resumo

Introdução: Admissõespor insuficiênciacardíaca aguda(ADHF)são frequentementecompli- cadasporelevaçãocreatininasérica(sCr),asquaispodemterpadrãovariável.

ObjetivoAvaliaroimpactoprognósticodoagravamentodafunc¸ãorenal(WRF)combaseno timingdasuaocorrência.

Métodos: Estudo retrospetivo de coortede doenteshospitalizados porADHF.WRF standard foidefinidacomoumaumentonaSCr≥0,3mg/dLduranteinternamento,foisubclassificada consoanteotimingdasuaocorrênciaemprecoce(quandoocorreunasprimeiras48horasdesde aadmissão)outardia(quandoocorreuapósas48h).Lesãorenalaguda(AKI)àadmissãofoi definidacomoumaumentodaSCr≥0,3mg/dLdesdeumvalorbasalambulatórioatéaprimeira determinaçãohospitalar.Oendpointprimáriofoiumcompostodemortalidadeporqualquer causaouhospitalizaçãoporeventoscardiovasculares,comumanodefollow-up.

Resultados: Foramincluídos249doentes(médiade77±11anos,62%comfraçãodeejeçãodo ventrículoesquerdopreservada).WRFprecoceocorreuem49doentes(19,7%)eassociou-sea maiorriscoparaooutcomeprimário(HR2,49;95%CI1,66-3,73),enquantoaWRFtardianão demonstrou essaassociação(p=0,411).ApósestratificaçãoparaapresençadeWRFprecoce e/ouAKIàadmissão,apenasosdoentescomWRFprecocemassemAKIàadmissão,bemcomo os doentes comambas (WRFprecoce e AKI àadmissão), demonstraram maiorrisco para o outcomeprimárioapósregressãomultivariáveldeCox.

Conclusão: WRFprecoce parece estar associada amaiorrisco para o outcome primário. O momentodaocorrênciadaWRFduranteointernamentopareceserumimportantefatorater emcontaquandoseconsideraoimpactoprognósticodasvariac¸õesdecreatininaemdoentes admitidosporADHF.

Introduction

Heart failure (HF) is a major cardiovascular syndrome associated with a signiﬁcant risk of mortality and hospitalization.^1---3 Congestion and/or hypoperfusion can lead to organ injury, which is associated with increased mortality.⁴ Renal dysfunction is one of themost frequent noncardiaccomorbiditiesinHF.⁵Increasedserumcreatinine (SCr)levelsareverycommonin acutedecompensatedHF (ADHF),withanincidencerangingfrom20%to50%.^6,7

Several groupshave studied theprognosticsigniﬁcance of worsening renal function (WRF) following initiation of diuretictherapy.However,studieshaveyieldedconﬂicting results.^8---11 The timing of creatininerise mayidentify dif- ferentsubgroupsofpatientswithdifferentprognoses.The aimofthisstudywastoassesstheprognosticimpactofthe timingofWRFinpatientswithADHF.

Methods

Westudiedasingle-centerretrospectivecohortofpatients admittedtoanHFunitduetoADHFwitha‘warmandwet’

clinicalproﬁle(typeB)accordingtothe2016EuropeanSoci- etyofCardiologyHFguidelines,¹betweenJanuary2014and August2018.Patientswithchronickidneydisease(CKD)on hemodialysis,needforrenalreplacementtherapy,needfor inotropictherapy,nooutpatientSCrmeasurementinthesix monthsbeforeadmission,noserialSCrassessmentavailable during hospitalization, or discharge in less than 48 hours wereexcluded.

During hospitalization patients underwent treatment withintravenousfurosemideandstartedguideline-directed medicaltherapy as soon as indicated. Data pertaining to the index hospitalization, patient characteristics, laboratory study results and medication use, as well as events

2

33 34 35 36 37 38

39 40

41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

67

68 69 70 71 72 73 74 75 76 77 78 79 80 81 82

83

84 85 86 87 88 89 90 91 92 93 94 95 96 97

(3)

Figure1 Representationofheartfailurepatienttimelineand thedifferenttypes ofcreatininevariation.AKI:acute kidney injury;HF:heartfailure;WRF:worseningrenalfunction.

during follow-up,were extracted fromelectronicmedical records.

WRF was definedas an increase in SCr of ≥0.3 mg/dl based on the standard definition (from admission to any timeduring hospitalization).^9,10 WRF timing wasclassified as early, when occurring within 48 hours, or late, when observedafter48hoursofhospitalization(Figure1).Acute kidneyinjury(AKI)atadmissionwasdefinedasariseinSCr of ≥0.3mg/dl fromoutpatientbaselinemeasurement(up tosixmonthsbeforetheacuteepisodeasanoutpatient)to thefirstmeasurementafterpatientarrival.

Patientswerethen stratiﬁedintofourgroups according tothepresenceofAKIatadmissionandthedevelopmentof earlyWRF:group1(A−/W−)---noAKIatadmissionandno early WRF;group 2(A+/W−)--- loneAKI (AKIat admission withnoearlyWRF);group3(A−/W+)---loneearlyWRF(no AKIatadmissionbutwithearlyWRF);andgroup4(A+/W+) ---AKIatadmissionandearlyWRF.

Theprimaryendpointwasacompositeofall-causemortalityorhospitalizationforcardiovascularevents,truncated atoneyearafteradmission.

Estimated glomerular ﬁltration rate (eGFR) was cal- culated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation.¹² CKD was deﬁned as eGFR<60ml/min/1.73m².

Categorical variablesarepresented asfrequencies and percentages, andcontinuousvariablesasmeans andstan- dard deviations, or medians and interquartile ranges for variables with skewed distributions. Differences between the experimental groups were assessed by an analysis of variance(ANOVA)model,followedbytheTukey-Kramertest when findings with ANOVA were significant. Kaplan-Meier survivalcurveswerecalculatedforeachpatientgroup.Uni- variateandmultivariateanalysiswithCoxregressionwere performedtoassesstheprognosticvalueofdifferentparam- eters.Allreportedp-valuesaretwo-tailed, withap-value of0.05indicating statisticalsignificance.Theanalysiswas performedusingIBMSPSSStatistics,version25(2017).

Results

Baselinepopulationcharacteristics

Atotal of249 patientswereincluded, ofwhom47% were

Table1 Baselinecharacteristicsofthe study population Q4 (n=249).

Age,years,mean±SD 77±11

Males,n(%) 116(46.6)

LVEF,mean(±SD) 51±17

Reduced,n(%) 66(26.5)

Mid-range,n(%) 28(11.2)

Preserved,n(%) 155(62.2)

HFetiology

Ischemic,n(%) 76(30.5)

Hypertensive,n(%) 113(45.4)

Valvular,n(%) 29(11.6)

Other,n(%) 31(12.4)

Comorbidities

Hypertension,n(%) 203(81.5)

AF,n(%) 149(59.8)

BaselineSCr,mg/dl,median[IQR] 1.08[0.85---1.41]

BaselineeGFR(ml/min/1.73m²), mean±SD^a

58.4±22.4 BaselineeGFR<60ml/min/1.73

m²,n(%)^a

139(55.8)

Diabetes,n(%) 92(36.9)

Laboratoryresultsatadmission

Hemoglobin,g/dl,median[IQR] 11.8[10.3---13.2]

Creatinine,mg/dl,median[IQR] 1.26[0.95---1.69]

Urea,mg/dl,median[IQR] 61[44---96]

NT-proBNP,pg/ml,median[IQR] 4740[2554---10950]

Sodium,mmol/l,median[IQR] 140[137---142]

Potassium,mmol/l,median[IQR] 4.2[3.8---4.7]

Medicationatadmission

Beta-blocker,n(%) 150(60.2) ACEinhibitor/ARB,n(%) 166(66.7)

MRA,n(%) 58(23.3)

Furosemide,n(%) 180(72.3)

Outpatientdailyoralfurosemide (mg),median[IQR]

40[0---60]

Oralanticoagulation,n(%) 108(43.4) Antiplatelettherapy,n(%) 79(31.7) Lengthofhospitalstay,days,

median[IQR]

7[5---10]

ACE: angiotensin-converting enzyme; AF: atrial ﬁbrillation;

ARB: angiotensin receptor blocker; CKD: chronic renal disease; COPD: chronic obstructive pulmonary disease; eGFR:

estimated glomerular ﬁltration rate; HF: heart failure; IQR:

interquartile range; LVEF: left ventricular ejection fraction;

MRA: mineralocorticoid receptor antagonist; NT-proBNP: N- terminalpro-B-typenatriureticpeptide;SCr:serumcreatinine;

SD:standarddeviation.

aeGFRwascalculatedusingtheChronicKidneyDiseaseEpi- demiologyCollaborationformula.

most frequent HF etiologies were hypertensive (45.4%) and ischemic (30.5%). Overall, 155 patients (62.2%) had preserved left ventricular ejection fraction (EF) and the majority had hypertension (81.5%), atrial ﬁbrillation (AF) (59.8%)orCKD(55.8%).Atadmission,patientshadamedian SCr and serum urea of 1.26 [0.96---1.68] mg/dl and 61

98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135

136

137

138

140 141 142 143 144 145

(4)

Table2 Associationofstandard,earlyandlateworseningrenalfunctionwiththeprimaryoutcomeatoneyearafterhospital- ization(univariateCoxregression).

n(%) HR(95%CI) p

StandardWRF 90(36.1) 1.689(1.152---2.477) 0.007

EarlyWRF(≤48h) 49(19.7) 2.487(1.659---3.728) <0.001

LateWRF(>48h) 41(16.4) 0.795(0.453---1.395) 0.411

CI:conﬁdenceinterval;HR:hazardratio;WRF:worseningrenalfunction.

Figure2 Kaplan-Meiercurvesfortheprimaryoutcome(all-causemortalityorhospitalizationforcardiovascularevents)truncated atoneyear.EarlyWRF:≤48h;lateWRF:>48h;WRF:worseningrenalfunction.

natriuretic peptide (NT-proBNP) was 4740 [2548---11400]

pg/ml.Patientswerehospitalizedforamedianof7[5---10]

days. Over a median follow-up of 351 [73---366] days, 81 patients died (nine in-hospital, 72 during follow-up), and 125 were admitted due to a cardiovascular event.

Medicationatadmissionandat dischargeforpatientswith reducedEFisdescribedinSupplementaryTableS1.

Characterizationofworseningrenalfunction Overall,90patients(36.1%)developedWRF.Thesepatients were signiﬁcantly older and had a higher SCrat baseline andadmissionandlongerhospitalstay thanthosewithout WRF.PatientswithWRFhadanincreasedincidenceofthe

composite endpoint in comparison to those who did not (hazardratio[HR]1.69[1.15---2.48];p=0.007)(Table2).

Early WRF (<48 h) was associated with a signiﬁcan- tly higher incidence of the primary outcome (HR 2.49 [1.66---3.73]; p<0.001), whereas late WRF was not (HR 0.80[0.45---1.40];p=0.411),comparedtopatientswhodid not develop WRF (Table 2). Survival over follow-up is depicted in Figure 2. These subgroups are characterized inSupplementary Table S2.Patientswithearly WRF were older,hadalongerhospitalstay,andhadahigherproportion ofpatientswithpreserved EF.Shorterfollow-upanalysisis describedinSupplementaryTableS3,whichrevealsaworse outcomefortheearly-WRFsubgroupat one,threeandsix months.

147 148 149 150 151 152 153

154

155 156 157

158 159 160 161 162 163 164 165 166 167 168 169 170 171

(5)

Table3 Baselinecharacteristicsofeachgroup accordingtothepresenceofacutekidneyinjury atadmissionand/orearly worseningrenalfunction(≤48h).

G1(A−/W−) (n=137)

G2(A+/W−) (n=63)

G3(A−/W+) (n=36)

G4(A+/W+) (n=13)

p

Age,years,mean±SD 75±12 79±10 82±9 82±6 0.003

Males,n(%) 62(45.3) 29(46) 19(52.8) 6(46.2) 0.882

LVEF,mean±SD 49±16 48±18 61±14 53±10 0.001

Reduced,n(%) 43(31.4) 19(30.2) 3(8.3) 1(7.7)

Mid-range,n(%) 14(10.2) 8(12.7) 3(8.3) 3(23.1)

Preserved,n(%) 80(58.4) 36(57.1) 30(83.3) 9(69.2)

HFetiology 0.585

Ischemic,n(%) 43(31.4) 22(34.9) 9(25.0) 2(15.4)

Hypertensive,n(%) 59(43.1) 26(41.3) 21(58.3) 7(53.8)

Valvular,n(%) 17(12.4) 7(11.1) 4(11.1) 1(7.7)

Other,n(%) 18(13.1) 8(12.7) 2(5.6) 3(23.1)

Comorbidities

Hypertension,n(%) 113(82.5) 53(84.1) 28(77.8) 9(69.2) 0.575

AF,n(%) 82(59.9) 35(55.6) 23(63.9) 9(69.2) 0.755

Baselinecreatinine,median[IQR] 1.04[0.84---1.29] 1.11[0.87---1.49] 1.10[0.85---1.57] 1.57[1.22---1.84] 0.003

Type2diabetes,n(%) 51(37.2) 27(42.9) 9(25) 5(38.5) 0.370

Outpatientdailyoralfurosemide, median[IQR]

40[0---60] 40[0---60] 40[0---60] 40[0---70] 0.687

Laboratoryresultsatadmission,median[IQR]

Hemoglobin,g/dl 12.0[10.6---13.7] 11.9[10.2---12.7] 11.6[10.4---13.1] 9.7[8.9---10.7] 0.001 Creatinine,mg/dl 1.07[0.85---1.34] 1.65[1.40---1.98] 1.10[0.86---1.32] 2.07[1.67---2.65] <0.001 Urea,mg/dl 52[40---73] 81[61---123] 57[43---79] 110[74---145] <0.001

NT-proBNP,pg/ml 4380

[1974---10178]

6952

[3418---20387]

3500[2316---6724] 4380

[3171---17035]

0.001 Sodium,mmol/l 140[138---142] 139[134---142] 141[139---142] 139[132---141] 0.090 Potassium,mmol/l 4.1[3.6---4.5] 4.3[4.0---4.8] 4.4[3.9---4.7] 4.3[4.0---4.9] 0.023 Lengthofhospitalstay,days,median

[IQR]

7[5---8] 7[4---11] 8[7---11] 12[6---14] 0.021

A+:AKIatadmission;A−:noAKIatadmission;AF:atrialﬁbrillation;AKI:acutekidneyinjury;CKD:chronickidneydisease;COPD:chronic obstructive pulmonarydisease; IQR:interquartilerange; LVEF:left ventricularejectionfraction;NT-proBNP:N-terminalpro-B-type natriureticpeptide;SD:standarddeviation;W+:presenceofearlyWRF;W−:noearlyWRF;WRF:worseningrenalfunction.

Outcomesstratiﬁedbygroupaccordingto admissionacutekidneyinjuryandtimingof worseningrenalfunction

Differentpatternsofcreatininevariationwereobservedand classiﬁedintofourgroups.

A total of137 patients (55.0%)were included in group 1 (A−/W−), 63 (25.3%) in group 2 (A+/W−), 36 (14.5%) in group 3 (A−/W+) and13 (5.2%) in group 4 (A+/W+). A detailed overviewof baselinecharacteristics according to groupisshowninTable3.Group2(A+/W−)hadsigniﬁcan- tlyhigheradmissionNT-proBNPthangroups1(A−/W−)and 3(A−/W+);group3(A−/W+)weresigniﬁcantlyolderthan group1 (A−/W−)andhadahigherproportionof patients withpreservedEFthangroups1(A−/W−)and2(A+/W−);

andgroup4(A+/W+)hadhigherbaselineSCr,loweradmis- sionhemoglobinandlongermedian hospitalstaythan the other groups. Survivalover follow-upfor each of thefour subgroupsisdepictedinFigure3.

After multivariate Cox regression, this stratification showeda statistically significant association withthe pri- maryoutcome bothfor loneearly WRF(A−/W+)(HR2.34 [1.37---4.00];p=0.002)andearlyWRFwithAKIatadmission (A+/W+) (HR 2.43 [1.18---4.00]; p=0.016) (Table 4). How- ever, lone AKI at admission lost itsstatistical significance (p=0.166)aftermultivariateanalysis.

Discussion

Themainﬁndingsofthecurrentanalysiscanbesummarized asfollows: (i)in patients hospitalized for ADHF, standard WRFwasassociatedwithall-causemortalityor cardiovas- culareventsatoneyear;(ii)afteradjustmentforasetof baseline characteristics and admission laboratory results, thepresence of early WRF (either alone or in association withAKIat admission)remainedanindependentpredictor oftheprimaryoutcome.

172 173 174

175 176 177 178 179 180 181 182 183 184 185 186 187 188 189

190 191 192 193 194 195 196

197

198 199 200 201 202 203 204

(6)

Figure3 Kaplan-Meiercurvesfortheprimaryoutcome(all-causemortalityorhospitalizationforcardiovascularevents)truncated atoneyear.AKI:acutekidneyinjury;earlyWRF:≤48h;WRF:worseningrenalfunction.

OurfindingsregardingstandardWRFaresimilartothose previously reportedby other studies.^8,11,13 In particular, a study by Formanetal. assessing 1004ADHF patients, the majorityof whomhadheart failurewithreducedejection fraction(HFrEF),showedanassociationbetween WRFand in-hospitaldeath.¹³However,otherstudiesshowedconflict- ingresults.AprospectivestudybyMetraetal.thatincluded 318patientsrevealedthatstandardWRFwasnotassociated withdeathor HF hospitalization,⁹ although a sizablepro- portionofthesepatientsweretreatedwithdopamine(22%) orinotropes(9%),whichmayhaveinfluencedSCrvariation in responsetodiuretic therapyandrenal hemodynamics.⁹ Similarly, Cowie et al. studied 299 patients with acute decompensated HFrEF and showed no statistically significant association between standard WRF and mortality at sixmonths.¹⁰Themajorityofthesepatientsdevelopedlate WRF (median time toWRF was four days),which, in our study, was not associated with worse outcomes.¹⁰ Other studieshavebeenpublishedonWRF,usingawidevarietyof definitions,hinderingthecomparabilityoftheresults.^11,14---16 AnotherimportantfindingofourworkisthatearlyWRF wasassociatedwithworseoutcomesnotonlyatoneyearof follow-up,butalsoatshorterfollow-ups,incomparisonto lateWRF.Thisunderscorestheimportanceofthisadverse event in patient hospitalization, which may be explained by the presence of more severe disease, associated with lower cardiac and/or renal reserve, limiting the compen- satoryresponsethatisusuallytriggered.

To thebestof ourknowledge, thisis the ﬁrststudy to assesstheprognosticimpactofWRFtiminginpatientswith ADHF.Weaimedtoanalyzewhetherearlydeteriorationof renalfunction,speciﬁcallywithin48hoursofadmission,was amarkerofworseprognosis.Weemployeda48-hourcut-off forWRF,asthisexcludedseveralconfoundingfactorsassoci- atedwithSCrvariationsduringhospitalization.Forexample, excessivediuretictherapyorinitiation/uptitrationofrenin- angiotensin-aldosteroneinhibitors frequentlyoccur during hospitalizationandmayraiseSCr.ThistypeofWRF(often labeledpseudo-WRF)doesnotseem tobeassociatedwith worseoutcomes.¹⁷ Thisideaisfurthercorroboratedbyour study,sincewefoundnoassociationbetweenlateWRF(>48 hours after admission) and worse outcomes. Hence, the timingofSCrelevationseemstohave animpactin differ- entiatingriskanditspathophysiology.

Furthermore,stratificationof patientsaccordingtoAKI at admission and early WRF status shows that lone early WRF(A−/W+)wasassociatedwithahigherriskforthepri- maryoutcome.Althoughthiswasinapopulationthatwas older,hadlongerhospitalstayandhadahigherproportion ofpreservedEF,theprognosticimpactremainedaftercon- trollingforvariousrelevantfactors.Conversely,loneAKIat admission(A+/W−)wasnotanindependentpredictorofthe primaryoutcomeafter multivariateCox regression.These results seem todiffer fromthe work of Shirakabe etal., who concluded that, in a cohort of 1083 ADHF patients, lone WRF (in the firstfive days) was not associated with

205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232

233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259

(7)

Table 4 Hazard ratios (univariate and multivariate analysis) for the primary outcome, according to selected patient characteristics.

Compositeoutcomeat1year

Univariateanalysis Multivariateanalysis

HR(95%CI) p HR(95%CI) p

NoAKIatadmission+noearlyWRF 1.000 1.000

AKIatadmission+noearlyWRF 1.653(1.032---2.648) 0.036 1.409(0.868---2.287) 0.166 NoAKIatadmission+earlyWRF 2.833(1.711---4.692) <0.001 2.339(1.368---3.999) 0.002 AKIatadmission+earlyWRF 3.592(1.843---7.001) <0.001 2.429(1.181---4.995) 0.016

Age 1.028(1.009---1.049) 0.005 1.017(0.995---1.040) 0.127

Gender 1.183(0.808---1.731) 0.388

LVEF 1.011(0.999---1.023) 0.078 1.001(0.986---1.015) 0.912

PreservedLVEF 1.000

Mid-rangeLVEF 0.604(0.302---1.206) 0.153

ReducedLVEF 0.639(0.397---1.028) 0.065

Outpatientdailyfurosemidedose 1.005(1.000---1.011) 0.060

Hypertension 1.393(0.818---2.371) 0.222

AF 1.282(0.860---1.912) 0.222

Baselinecreatinine 1.223(0.562---2.662) 0.611

Type2diabetes 1.037(0.703---1.531) 0.854

Hemoglobin^a 0.854(0.775---0.941) 0.001 0.909(0.813---1.016) 0.909

NT-proBNP^a 1.000(1.000---1.000) 0.287 1.000(1.000---1.000) 0.655

Urea^a 1.005(1.001---1.010) 0.017

Sodium^a 0.977(0.937---1.019) 0.279

Potassium^a 1.050(0.817---1.348) 0.704

Chloride^a 0.980(0.949---1.012) 0.222

Ischemicetiology 0.955(0.628---1.451) 0.828

Lengthofhospitalstay 1.027(0.997---1.057) 0.073

AF:atrialﬁbrillation;AKI:acutekidneyinjury;CI:conﬁdenceinterval;CKD:chronickidneydisease;COPD:chronicobstructivepulmonary disease;HR:hazardratio;LVEF:leftventricularejectionfraction;NT-proBNP:N-terminalpro-B-typenatriureticpeptide.

a Atadmission.

higher risk of all-causedeath.⁸ However,as well asusing abroaderWRFdeﬁnition,thiscohortofpatientshadmore severedisease,andwereadmittedtoanintensivecareunit withfrequentneedforintravenousinotropes/vasopressors, andhadaconsiderablylongerhospitalstay.⁸

EachofthesesubtypesofWRFseemstoidentifyadiffer- entsetofpatients.Changesinrenalfunctioninpatientswith HFarecomplexandmultifactorial.Multiplemechanismscan playapartinWRF,includingrenalhypoperfusion,activation of therenin-angiotensin-aldosterone andsympatheticsys- tems,andvenouscongestion.^18,19Theclinicalimportanceof eachmechanism islikelytovaryfrompatienttopatient.⁸ AKIatadmissionmayalsoberelatedtoprogressionofrenal diseaseinthemonthsprecedinghospitalizationaswellasto thecumulativecongestionthesepatientsusuallydevelop.⁴ Ontheotherhand,earlyWRFmayberelatedtoreducedcar- diacoutputandlowcardiacand/orlowrenalreserve,which resultinelevationofSCrafterinitiationofdiuretictherapy.⁴ Consequently, the prognosis varies, depending mostly on themainunderlyingmechanismofrenaldysfunctionrather than SCr, which is a limited surrogate biomarker of renal function.

Inouropinion,acuterenaldysfunctionshdifferentiated accordingtovariousfactors.Aswellasconsideringabsolute variationinSCr,thetimingofitsappearanceshouldbetaken

intoaccount.WespeculatethatearlyWRFafterinitiation ofdiuretictherapymaybeamarkerofmoresevereunderly- ingcardiacand/orrenaldisease,thusidentifyingpatientsin whomtheusualcompensatorymechanismscannotcounter- actthehemodynamiceffectsofdiuretics,unveilingreduced organreserve.

Limitations

Thepresentstudyhassomelimitations.Mostpatientswere not admitted directly to the HF clinic. There was some variationbetweenpatients inthetimefromarrivalatthe emergency department until admission to the HF clinic.

Nonetheless,thevast majority were admittedtoourunit within24hours.Also,informationregardingdiureticdoses administeredwasnotavailable.

Mechanisms associated with creatinine variation during hospital stay were not directly measured, since the datawerecollectedretrospectively.Studyingthesefactors prospectivelywould shedmore light ontheir mechanisms andprognosticimpact.

This was a single-center retrospective study. Thus, it should be viewed as hypothesis-generating, due to the possible presenceof unmeasured confoundingfactorsand selectionbias.

260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284

285 286 287 288 289 290

291

292 293 294 295 296 297 298 299 300 301 302 303 304 305 306

(8)

Thelownumberofpatientsinﬂuencesthepowerofthis study.Consequently,associationswithsmallereffectsizes maynotbeapparentinthisanalysis.

Nonetheless,ourstudysupportstheﬁndingsofprevious investigations,reinforcingtheimportanceoftimingpatterns ofWRFasaprognosticpredictor.

Conclusion

Inthisstudy,thepresenceofWRF,particularlywhenoccur- ring within 48 hoursof initiation of diuretic therapy,was associated with a higher risk of death fromany cause or hospitalizationforcardiovascularevents.ThetimingofWRF appears to be an important characteristic to take into accountwhenconsidering theprognosticimpactofcreati- ninevariationduringhospitalizationforADHF.

Authors’ contributions

Conceptionanddesign:JP,GC,BR,IA,CF;datacollection:

JP,GC,LL,ST,MR,MIS,MM,SM;dataanalysis:JP,GC,BR;

drafting:JP,GC,BR;revising:JP,GC,LL,ST,MR,MIS,MM, SM,BR,CR,IA,CF.Allauthorsreadandapprovedtheﬁnal manuscript.

Conﬂicts of interest

Theauthorshavenoconﬂictsofinteresttodeclare.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.repc.2022.06.015.

References

1.PonikowskiP,VoorsAA,AnkerSD,etal. 2016ESC Guidelines for the diagnosis and treatment of acuteand chronic heart failure: The Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology(ESC). Developedwiththespecial contributionof theHeartFailure Association(HFA)oftheESC.EurHeart J.

2016;37:2129---200.

2.GouveiaMRDA, Ascenc¸ão RMSES,FiorentinoF,etal. Current costsofheartfailureinPortugalandexpectedincreasesdueto populationaging.RevPortCardiol.2020;39:3---11.

3.GouveiaM,Ascenc¸ãoR,Fiorentino F,et al.Thecurrentand future burden of heart failure in Portugal. ESC Hear Fail.

2019;6:254---61.

4.HarjolaV-P,MullensW,BanaszewskiM,etal.Organdysfunction, injuryandfailureinacuteheartfailure:frompathophysiology todiagnosisandmanagement.EurJHeartFail.2017;19:34.

5.Schefold JC, Filippatos G, HasenfussG, et al. Heart failure andkidneydysfunction:epidemiology,mechanismsandmana- gement.NatRevNephrol.2016;12:610---23.

6.DammanK,TestaniJM.Thekidneyinheartfailure:anupdate.

EurHeartJ.2015;36:1437---44.

7.DiLulloL,BellasiA,RussoD,etal.Cardiorenalacutekidney injury: epidemiology, presentation, causes, pathophysiology andtreatment.IntJCardiol.2017;227:143---50.

8.ShirakabeA,HataN,KobayashiN,etal.Worseningrenalfunc- tiondeﬁnitionisinsufﬁcientforevaluatingacuterenalfailure inacuteheartfailure.ESCHearFail.2018;5:322---31.

9.MetraM,NodariS, ParrinelloG,et al.Worseningrenalfunc- tion in patientshospitalised for acute heartfailure: clinical implications and prognostic signiﬁcance. Eur J Heart Fail.

2008;10:188---95.

10.Cowie MR, Komajda M, Murray-Thomas T, et al. Prevalence andimpactofworseningrenalfunctioninpatientshospitalized with decompensated heart failure: results of the prospec- tive outcomes study in heart failure (POSH). Eur Heart J.

2006;27:1216---22.

11.Kociol RD, Greiner MA, Hammill BG, et al. Long-term outcomesofMedicarebeneﬁciarieswithworseningrenalfunc- tion during hospitalization for heart failure. Am J Cardiol.

2010;105:1786---93.

12.Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular ﬁltration rate.J Nephrol Ther.

2009;150:604---12.

13.Forman DE, Butler J, Wang Y, et al. Incidence predictors atadmission,and impactofworseningrenalfunction among patients hospitalized with heart failure. J Am Coll Cardiol.

2004;43:61---7.

14.PatelUD,GreinerMA,FonarowGC,etal.Associationsbetween worsening renal function and 30-day outcomes among Medi- care beneﬁciaries hospitalizedwith heart failure.Am Heart J. 2010;160, http://dx.doi.org/10.1016/j.ahj.2010.03.033, 132---138.e1.

15.AtherS,BavishiC,McCauleyMD,etal.Worseningrenalfunc- tionisnotassociatedwithresponsetotreatmentinacuteheart failure.IntJCardiol.2013;167:1912---7.

16.Akhter MW, Aronson D, Bitar F, et al. Effect of elevated admission serum creatinineand itsworseningon outcomein hospitalizedpatientswithdecompensatedheartfailure.AmJ Cardiol.2004;94:957---60.

17.MullensW,DammanK,TestaniJM,etal.Evaluationofkidney function throughouttheheartfailure trajectory---a position statementfromtheHeartFailureAssociationoftheEuropean SocietyofCardiology.EurJHeartFail.2020;22:584---603.

18.Carubelli V,MetraM,LundLH.Negotiatingrenaldysfunction whentreatingpatientswithheartfailure.ExpertRevCardiovasc Ther.2018;16:113---22.

19.EllisonDH,FelkerGM.Diuretic treatmentinheartfailure.N EnglJMed.2017;377:1964---75.

307 308 309 310 311 312

313

314 315 316 317 318 319 320

321

322 323 324 325 326

327

328

329

330 331 332

333

334 335 336 337 338 339 340 341 342 343 344 345

346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400