Since January 2020 Elsevier has created a COVID-19 resource centre with
free information in English and Mandarin on the novel coronavirus
COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the
company's public news and information website.
Elsevier hereby grants permission to make all its COVID-19-related
research that is available on the COVID-19 resource centre - including this
research content - immediately available in PubMed Central and other
publicly funded repositories, such as the WHO COVID database with rights
for unrestricted research re-use and analyses in any form or by any means
with acknowledgement of the original source. These permissions are
granted for free by Elsevier for as long as the COVID-19 resource centre
EuropeanJournalofInternalMedicine000(2020)1–10
ContentslistsavailableatScienceDirect
European
Journal
of
Internal
Medicine
journalhomepage:www.elsevier.com/locate/ejinme
Original
article
Use
of
hydroxychloroquine
in
hospitalised
COVID-19
patients
is
associated
with
reduced
mortality:
Findings
from
the
observational
multicentre
Italian
CORIST
study
The
COVID-19
RISK
and
Treatments
(CORIST)
Collaboration
1 ,∗Q1
a
r
t
i
c
l
e
i
n
f
o
Keywords: Hydroxychloroquine COVID-19 Disease severity Mortality Inflammationa
b
s
t
r
a
c
t
Background: Hydroxychloroquine(HCQ)wasproposedaspotentialtreatmentforCOVID-19.
Objective: Weset-upamulticenterItaliancollaborationtoinvestigatetherelationshipbetweenHCQtherapyand COVID-19in-hospitalmortality.
Methods: Inaretrospectiveobservationalstudy,3,451unselectedpatientshospitalizedin33clinicalcenters inItaly,fromFebruary19,2020toMay23,2020,withlaboratory-confirmedSARS-CoV-2infection,were ana-lyzed.Theprimaryend-pointinatime-toeventanalysiswasin-hospitaldeath,comparingpatientswhoreceived HCQwithpatientswhodidnot.WeusedmultivariableCoxproportional-hazardsregressionmodelswithinverse probabilityfortreatmentweightingbypropensityscores,withtheadditionofsubgroupanalyses.
Results: Outof3,451COVID-19patients,76.3%receivedHCQ.Deathrates(per1,000person-days)forpatients receivingornotHCQwere8.9and15.7,respectively.Afteradjustmentforpropensityscores,wefound30% lowerriskofdeathinpatientsreceivingHCQ(HR=0.70;95%CI:0.59to0.84;E-value=1.67).Secondary analy-sesyieldedsimilarresults.TheinverseassociationofHCQwithinpatientmortalitywasparticularlyevidentin patientshavingelevatedC-reactiveproteinatentry.
Conclusions: HCQusewasassociatedwitha30%lowerriskofdeathinCOVID-19hospitalizedpatients.Within thelimitsofanobservationalstudyandawaitingresultsfromrandomizedcontrolledtrials,thesedatadonot discouragetheuseofHCQininpatientswithCOVID-19.
1. Introduction
1
The aminoquinoline hydroxychloroquine (HCQ) has been
exten-2
sivelyused inthetreatmentof malariaandis currentlywidely used
3
totreatautoimmunediseaseslikerheumatoidarthritis(RA),systemic
4
lupuserythematosus(SLE)andanti-phospholipidsyndrome(APS),due
5
toitsimmunomodulatoryandanti-thromboticproperties[1] .More
re-6
cently, a promising role of HCQ has been suggested in viral
infec-7
tions[2] ,sinceitdirectlyinhibitsviralentryandspreadinseveralin
8
vitroandinvivomodels.Duetotheseproperties,HCQhasbeenused
9
inEbolavirusdisease[3,4] ,humanimmunodeficiencyvirus(HIV)
in-10
fection[5] ,SARS-CoV-1infectionandtheMiddleEastRespiratory
Syn-11
drome(MERS)[6 ,7 ]andgainedworldwideattentionasapossible
ther-12
apyinCOVID-19patients[8] .
13
HCQmightinhibittheintracellularglycosylationofACE2,the
re-14
ceptorusedbytheSARS-CoV-2virustoenterthecells,resultingina
15
∗Correspondingauthorat:DepartmentofEpidemiologyandPrevention,IRCCSNeuromed,Viadell’Elettronica,86077Pozzilli(IS),Italy.
E-mailaddress:licia.iacoviello@moli-sani.org
1 ThemembersofTheCOVID-19RISKandTreatments(CORIST)CollaborationarelistedinAppendix1attheendofthearticle.
reducedligandrecognitionandinternalizationofthevirus[7] andexert- 16
ingapossibleprotectiveroleinSARS-CoV-2infection.Moreover,dueto 17
itsimmunomodulatory,anti-inflammatoryandanti-thromboticeffects, 18
HCQcouldalsomodulatetheseverityofthedisease.However,theexact 19
mechanismforthepotentialbenefitinCOVID-19islargelyspeculative 20
[9] andmightbecounterbalancedbyadverseeffects,mainlycardiovas- 21
cular[10 ,11 ],sothatthenetbalanceofthisdrug’suseremainstobe 22
established. 23
TheAmericanFoodandDrugAdministration(FDA)allowedChloro- 24
quine(CQ)phosphateandHCQtobeprovidedtocertainhospitalized 25
patients becausethese drugsmaypossibly helppatients withsevere 26
COVID-19[12] .TheEuropeanMedicinesAgency(EMA)authorizedthe 27
useofCQandHCQforCOVID-19inclinicaltrialsorasemergencyuse 28
[13] ,while theItalianDrugAgency(AIFA)statedin thisemergency 29
phasethattherapeuticuseofHCQmightbe consideredin COVID-19 30
patients,bothinthosewithmildpresentationmanagedathomeandin 31
https://doi.org/10.1016/j.ejim.2020.08.019
Received28June2020;Receivedinrevisedform19August2020;Accepted20August2020 Availableonlinexxx
hospitalizedpatients[14] .Inclinicalpractice,HCQratherthan
chloro-32
quinehasbeenusedbecauseofitsmorepotentantiviralpropertiesand
33
bettersafetyprofile[15] .
34
However,inthelightofarecentpublication[16] ,thatwaslater
re-35
tracted[17] ,onthelackofsafetyandefficacyofHCQinthetreatment
36
forCOVID-19patientstheExecutiveGroupoftheSolidarityTrial
de-37
cidedtoimplementatemporarypauseoftheHCQarmwithinthetrial
38
asaprecaution,whilethesafetydataisbeingreviewed[18] .Similarly,
39
theItaliandrugAgencyAIFAdecidedtosuspendtheauthorizationto
40
useHCQforCOVID-19treatmentoutsideclinicaltrials[19] .
41
Recent reviews of clinical trials or observational studies [20–
42
24] havereportedinsufficientandoftenconflictingevidenceonthe
ben-43
efitsandharmsofusingHCQtotreatCOVID-19andconcludedthatas
44
such,itwasimpossibletodeterminethebalanceofbenefitstoharm.
45
Untilnow,althoughseveraltrialshadbeenstartedontheuseofCQand
46
HCQinCOVID-19,onlyfewofthemhavebeenpublished[25] onsmall
47
numbersofpatientsoronsurrogateendpointsorinexposedsubjectsfor
48
prophylaxisuse[26] .
49
Whilewaitingtheresultsfrom ongoingrandomizedclinicaltrials
50
(RCT)todefinetheefficacyinpreventinghardendpointsofthis
treat-51
mentsowidelyusedduringtheemergencyphaseoftheCOVID-19
pan-52
demic,poweredretrospectiveobservationalstudiesperformedin
differ-53
entgeographicalanddiseaseconditionsmaystillbeusefultoshedlight
54
onthisdebate.Tworetrospectiveobservationalstudies,bothconducted
55
intheNewYorkmetropolitanregion,didnotreportanysignificant
as-56
sociationbetweenHCQuseandratesofintubationordeath[27 ,28 ].
57
NodataarepresentlyavailablefromlargecohortsofpatientsinItaly,
58
whichrepresentsoneofthemostaffectedcountriesintermsof total
59
deathsforCOVID-19intheworld[29] .Weundertookamulticenter
Ital-60
iancollaboration[30] toinvestigatetherelationshipbetween
underly-61
ingriskfactorsandCOVID-19outcomes,andtoevaluatetheassociation
62
betweendifferentdrugtherapyanddiseaseseverityand/ormortality.
63
Wereportheretheresultsobtainedin3,451 hospitalizedCOVID -19
64
patientsreceivingornotHCQtreatment.
65
2. Materialandmethods
66
2.1. Setting
67
Thisnationalretrospectiveobservationalstudywasconceived,
co-68
ordinatedandanalysedwithintheCORISTProject(ClinicalTrials.gov
69
ID:NCT04318418,30].Thestudywas approvedby theinstitutional
70
ethicsboardoftheIstitutodiRicoveroeCuraaCarattereScientifico
71
(IRCCS)Neuromed,Pozzilli,andofallrecruitingcentres.Dataforthe
72
presentanalyseswereprovidedby33hospitalsdistributedthroughout
73
Italy(listedinthesupplementaryfile).Acceptancetoparticipateinthe
74
projectortoprovidedataforthepresentanalysiswasnotrelatedto
75
theuseofCQ/HCQ.Eachhospitalprovideddatafromhospitalized
pa-76
tientswhohadapositivetestresultfortheSARS-CoV-2virusatany
77
timeduringtheirhospitalizationfromFebruary19toMay23,2020.
78
Thefollow-upcontinuedthroughMay29,2020.
79
2.2. Datasources
80
Wedevelopedacohortcomprising3,971patientswith
laboratory-81
confirmedSARS-CoV-2infectioninanin-patientsetting.The
SARS-CoV-82
2statuswasdeclaredbasedonlaboratoryresults(polymerasechain
re-83
actiononnasopharyngealswab)fromeachparticipatinghospital.
Clin-84
icaldata were abstractedat one-timepoint from electronicmedical
85
recordsorcharts,andwerecollectedusingeitheracentrallydesigned
86
electronic worksheet or a centralized web-baseddatabase. Collected
87
dataincludedpatients’demographics,laboratorytestresults,
medica-88
tionadministration,historicalandcurrentmedicationlists,historical
89
andcurrent diagnoses,andclinicalnotes. Inaddition,specific
infor-90
mationonthemostseveremanifestationofCOVID-19occurredduring
91
hospitalizationwasretrospectivelycaptured.Maximumclinicalsever- 92
ityobservedwasclassifiedasmildpneumonia;orseverepneumonia;or 93
acuterespiratorydistresssyndrome(ARDS)[31] .Specifically,weob- 94
tainedthefollowinginformationforeachpatient:hospital;dateofad- 95
missionanddateofdischargeordeath;age;sex;thefirstrecordedinpa- 96
tientlaboratorytestsattheentry(creatinine,C-reactiveprotein);past 97
andcurrent diagnoses(myocardialinfarction, heartfailure,diabetes, 98
hypertension,respiratorydiseaseandcancer)andcurrentdrugthera- 99
piesforCOVID-19– HCQ,lopinavir/ritonavirordarunavir/cobicistat, 100
remdesevir,tocilizumaborsarilumab,corticosteroids,heparin,andfor 101
comorbidities(insulin,anti-hypertensivetreatments,aldosteronerecep- 102
torantagonists,diuretics,statins,sacubitril/valsartan).Adiagnosisof 103
pre-existing cardiovascular disease was basedon history of myocar- 104
dialinfarctionorheartfailure.Chronickidneydiseasewasclassified 105
as: stage1: kidneydamagewithnormalor increasedglomerular fil- 106
tration rate (GFR) (>90 mL/min/1.73 m2); stage 2: mild reduction 107
in GFR (60-89 mL/min/1.73 m2); stage 3a: moderate reduction in 108
GFR(45-59mL/min/1.73 m2);stage3b:moderatereductionin GFR 109
(30-44 mL/min/1.73 m2); stage 4: severe reduction in GFR (15-29 110
mL/min/1.73m2);stage5:kidneyfailure(GFR<15mL/min/1.73m2 111
ordialysis).Forstatisticalanalysis,stages3aand3bandstages4and 112
5werecombined.GFRwascalculatedbytheChronicKidneyDisease 113
EpidemiologyCollaboration(CKD-Epi)equation.Patientsweredefined 114
asreceivingHCQiftheywerereceivingitatadmissiontohospitalor 115
receiveditduringthefollow-upperiod.AccordingtotheAIFAguidance 116
[14] ,HCQwasadministeredatdoseof400mgx2/dayorx4/daythe 117
firstday,and200mgx2/dayfromtheseconddayonwardsforatleast 118
5toamaximumof10days,accordingtotheclinicalevolutionofthe 119
disease. 120
2.3. Statisticalanalysis 121
Thestudyindexdatewasdefinedasthedateofhospitaladmission. 122
IndexdatesrangedfromFebruary19,2020toMay23,2020.Thestudy 123
endpointwasthetimefromstudyindextodeath.Thenumberofpa- 124
tientswhoeitherdied,orhadbeendischargedalive,orwerestilladmit- 125
tedtohospitalasofMay29,2020,wererecorded,andhospitallengthof 126
staywasdetermined.Patientsalivehadtheirdatacensoredonthedate 127
ofdischargeorasthedateoftherespectiveclinicaldatacollection.Data 128
werecensoredat35daysoffollowupinn=330(8.3%)patientswitha 129
followupgreaterthan35days. 130
Oftheinitialcohortof3,971patients,350patientswereexcluded 131
fromtheanalysisbecausetheyhadatleastonemissingdataatbaseline 132
orlosttofollowuponHCQuse(N=94),otherdrugtherapiesforCOVID- 133
19(n=265),timetoevent(n=59),outcome(death/alive,n=8),COVID- 134
19 severity(n=4),age(n=4withmissingdataandn=2withage<18 135
years)orsex(n=2).Oftheremaining3,621patients,170patientsdied 136
orweredischargedwithin24hoursafterpresentation,andwerealso 137
excludedfromtheanalysis. 138
Attheend,theanalysedcohortconsistedofn=3,451patients.Inpa- 139
tientsnotincludedintheanalysis(n=520),asuniquedifferencewiththe 140
analysedgroup,theprevalenceofdiabetics(19.9%vs14.8%,P=0.0066) 141
and,toalessextent,ofmen(62.3%vs58.3%,P=0.081)washigher.Out 142
of3,541patients,295(8.5%)hadatleastamissingvalueforcovariates. 143
Distributionofmissingvalueswasasfollows:n=178forC-reactivepro- 144
tein;n=69forGFR;n=74forhistoryofischemicdisease;n=64forhis- 145
toryofchronicpulmonarydisease;n=51fordiabetes;n=51forhyper- 146
tensionandn=56forcancer.Weusedmultipleimputationtechniques 147
(SASPROCMI,n=10imputeddatasets;andPROCMIANALYZE)tomax- 148
imizedataavailability.Assensitivityanalysis,wealsoconductedacase- 149
completeanalysison3,156patients. 150
Coxproportional-hazardsregressionmodelswereusedtoestimate 151
theassociationbetweenHCQuseanddeath.Sincemultipleimputation 152
was applied,thefinalstandarderrorwas obtainedusingtheRubin’s 153
rulebasedontherobustvarianceestimatorinCoxregression[32] .The 154
European Journal of Internal Medicine 000 (2020) 1–10
feldresiduals,andnoviolationwasidentified.Toaccountforthe
non-156
randomizedHCQadministrationandtoreducetheeffectsof
confound-157
ing,thepropensity-scoremethodwasused.Theindividualpropensities
158
forreceivingHCQtreatmentwereassessedwiththeuseofa
multivari-159
ablelogistic-regressionmodelthatincludedage,sex,diabetes,
hyperten-160
sion,historyofischemicheartdisease,chronicpulmonarydisease,GFR,
161
C-reactiveprotein,hospitalsclusteringanduseofotherdrugtherapies
162
forCOVID-19(lopinavir/ritonavirordarunavir/cobicistat,remdesivir,
163
corticosteroids,tocilizumaborsarilumab).AssociationsbetweenHCQ
164
treatmentanddeathwasthenappraisedbymultivariableCox
regres-165
sionmodelswiththeuse ofpropensity-score andfurthercontrolling
166
forhospitalsclusteringasrandomeffect(frailtymodel).Theuseofa
167
frailtymodelwas chosenassuggested in[33] .Theprimary analysis
168
usedinverseprobabilitybytreatmentweighting;thepredicted
proba-169
bilitiesfromthepropensity-scoremodelwasusedtocalculatethe
sta-170
bilizedinverse-probability-weighting weight[34] . Stabilized weights
171
werenormalizedso thatthey addedup the actualsample size.
Sec-172
ondaryanalysesusedpropensity-scorestratification(n=5strata)or
mul-173
tivariableCoxregressionanalysisormultivariable logisticregression
174
analysescomparingdeathversusalivepatients,oraccountedfor
hos-175
pitalsclusteringviastratificationorbyrobustsandwichestimator.
Pre-176
establishedsubgroupanalyseswereconductedaccordingtoageorsex
177
ofpatients,degreeofCOVID-19severityexperiencedduringthe
hospi-178
talstay,C-reactiveproteinatbasalorotherdrugtherapiesfor
COVID-179
19.Hospitalswereclusteredaccordingtotheirgeographical
distribu-180
tion, as illustrated in Table 1 . To quantify the potential for an
un-181
measuredconfoundertorenderapparentstatisticallysignificanthazard
182
rationon-significant,theE-value wascalculated[35] . Analyseswere
183
performedwiththeaidoftheSASversion9.4statisticalsoftwarefor
184
Windows.
185
3. Results
186
Weincludedinthefinalcurrentanalyses3,451patientswhowere
187
hospitalizedwithconfirmedSARS-CoV-2infectionat33clinicalcentres
188
acrossItalyandeitherdied,hadbeendischarged,orwerestillin
hospi-189
talasofMay29,2020.Ofthesepatients,2,634(76.3%,rangeamong
190
hospitals53.2%to93.6%)receivedHCQ.Timingofthefirstdoseof
191
HCQafterpresentationtothehospitalwas1dayforthelargemajority
192
ofcentres,and2to3daysfortheothers.HCQwasadministeredinall
193
centresatthedoseof400mg/day(inonecentrehoweveritwasusedat
194
thedoseof600mg/dayandinanotheratthedoseof600mg/daybut
195
onlyinpatientsyoungerthan65years).Durationoftreatmentranged
196
from5to15days(with10daysasthemodalvalue).Thedrugusedwas
197
HCQinallhospitals.
198
BaselinecharacteristicsaccordingtoHCQuseareshowninTable1.
199
PatientsreceivingHCQweremorelikelyyounger,menandhadhigher
200
levelsofC-reactiveproteinandlesslikelyhadischemicheartdisease,
201
cancerorstages3aorgreaterchronickidneydisease(Table1).Patients
202
receivingHCQmorelikelyreceivedanotherdrugforCOVID-19
treat-203
ment(78.4%;lopinavir/ritonavirordarunavir/cobicistat,remdesevir,
204
tocilizumaborsarilumab,corticosteroids),incomparisonwithnon-HCQ
205
patients(46.3%;P<0.0001;Table 1 ).
206
Theunadjusteddifferencesanddifferencesadjustedbypropensity
207
scoresbetweenHCQ-treatedandnon-HCQtreatedpatientsforeach
vari-208
ableincludedinthepropensityscoreareshowninFig.1.Allthe
pre-209
treatmentdifferencesdisappearedafteradjustmentbypropensityscore
210
weighting.TheC-statisticofthepropensity-scoremodelwas0.74.
211
3.1. Primaryoutcome
212
Outof3,628patients,576died(16.7%),2,390weredischargedalive
213
(69.3%)and485(14.1%)werestillatthehospital.Themedian
follow-214
up was14 days (interquartilerange8to22;range 2to35;55,388
215
person-days).Deathrate(per1,000person-days)was8.9inHCQand
216
15.7innon-HCQpatients(Table2).Atunivariableanalysis,hazardra- 217
tioformortalitywas0.56(95%CI:0.47to0.67).Intheprimarymul- 218
tivariableanalysiswithinverseprobabilityweightingaccordingtothe 219
propensityscore,HCQusewasassociatedwitha30%(95%CI:16%to 220
41%)reductionindeathrisk(Fig.2,Table2,E-value=1.67).Secondary 221
multivariableanalysesyieldedverysimilarresults(Table2),aswellas 222
case-completeanalysesrestrictedtothe3,156patientswithoutmissing 223
data(Table2).Consideringsecondarymultivariableanalysesoverall, 224
HRformortalityassociated withHCQranged between0.64to0.70, 225
accordingtotypeofanalyses.Controlofhospitalsclusteringwithdif- 226
ferentapproachesalsoyieldedsimilarresultsfortheprimaryanalysis 227
(HR=0.71,95%CI:0.59to0.85whenhospitalsclusteringwasstratified 228
forandHR=0.69,95%CI:0.54to0.88withtherobustsandwichesti- 229
mator). 230
SubgroupanalysesarepresentedinTable3.HCQuseremainedcon- 231
sistentlyassociatedwithreducedmortalityinalmostallsubgroups.The 232
inverseassociationofHCQwithinpatientmortalityisslightlymoreevi- 233
dentinwomen,elderlyandinpatientswhoexperiencedahigherdegree 234
ofCOVID-19severity.Itwasabsentin-patientwithC-reactiveprotein 235
<10mg/LandclearlyconfinedtopatientswithelevatedC-reactivepro- 236
tein(Table3). 237
4. Discussion 238
Inalargecohortof3,451patientshospitalizedforCOVID-19in33 239
clinicalcentersalloverItaly,coveringalmostcompletelytheperiodof 240
thehospitalizationforCOVID-19,theuseofHCQwasassociatedwith 241
asignificantbettersurvival.In-hospitalcrudedeathratewas8.9per 242
1,000person-day forpatientsreceivingHCQand15.7forthosewho 243
didnot.Afteradjustmentforknownpossibleconfounders,weobserved 244
a30%reductionintheriskofdeathinpatientsreceivingHCQtherapy 245
ascomparedwiththosewhodidnot. 246
Ourfindings provideclinicalevidenceinsupportof guidelinesby 247
ItalianandseveralinternationalSocietiessuggestingtouseHCQther- 248
apy in patients with COVID-19. However,the observed associations 249
shouldbeconsideredwithcaution,astheobservationaldesignofour 250
studydoesnotallowtofullyexcludingthepossibilityofresidualcon- 251
founders.Largerandomizedclinicaltrialsinwell-definedgeographical 252
andsocio-economicconditionsandinwell-characterizedCOVID-19pa- 253
tients,shouldevaluatetheroleofHCQbeforeanyfirmconclusioncan 254
bereachedregardingapotentialbenefitofthisdruginpatientswith 255
COVID-19. 256
Over 76%of patients receivedHCQ either aloneor in combina- 257
tion with other drugs. They were more likely to be younger, men 258
andwithhigherlevelsofCreactiveproteinatentry,whilelesslikely 259
hadpre-existingcomorbiditiessuchasischemicheartdisease,cancer 260
andseverechronickidneydisease,ascomparedtopatientsnotreceiv- 261
ingthedrug.Weadjustedouranalyses forpossible confounders,in- 262
cludingage,sex,diabetes,hypertension,historyofischemicheartdis- 263
ease,chronicpulmonarydisease,chronickidneydisease,C-reactivepro- 264
tein andadditionaltreatments for COVID-19,and tookinto account 265
possible differences across centresbyeitheradjustmentor stratifica- 266
tion. Tominimize biasduetotheobservationaldesign, weuseddif- 267
ferentanalyticalapproachesaimingatcreatinganoverallbalancebe- 268
tweencomparisongroups.Finally,wetriedtolimitbiasduetomiss- 269
ing data by using a multiple imputation approach, but in no case, 270
the resultwas changed. Despite all these precautions, we recognize 271
thepossibility,however,ofresidualunmeasuredconfoundersaffecting 272
results. 273
Systematic reviews of small clinical trialshad reported contrast- 274
ing results that were however scarcely reliable becauseof poor de- 275
signs [20–25] . TheHCQ dosestested in a Chinese randomizedclin- 276
ical trial[25] were approximately doubleas comparedto thatused 277
in our study(1200 mg vs 800 mg as loadingdose, 800mg vs400 278
mg as maintenancedose) for twice thetime (14-21 days versus 7- 279
10 days). National guidelines in Italy suggest to use HCQ 200 mg 280 3
Table1
GeneralcharacteristicsofCOVID-19patientsatbaseline,accordingtohydroxychloroquineuse.
Hydroxychloroquine
Characteristic No (N = 817) Yes (N = 2,634) P-value unadjusted ∗
Age -median (IQR-yr.) 73 (58-83) 66 (55-77) < .0001
Gender - no (%) < .0001 Women 361 (44.2%) 940 (36.7%) Men 456 (55.8%) 1,694 (64.3%) Diabetes - no (%) 0.71 No 633 (77.5%) 2,090 (79.3%) Yes 162 (19.9%) 515 (19.6%) missing data 22 (2.7%) 29 (1.1%) Hypertension - no (%) 0.31 No 378 (46.3%) 1,294 (49.1%) Yes 416 (50.9%) 1,312 (49.8%) missing data 23 (2.7%) 28 (1.1%)
Ischemic heart disease - no (%) < .0001
No 610 (74.7%) 2,190 (83.1%) Yes 179 (21.9%) 398 (15.1%)
missing data 28 (3.4%) 46 (1.8%)
Chronic pulmonary disease - no (%) 0.21
No 666 (81.5%) 2,225 (84.5%) Yes 127 (15.5%) 369 (14.0%) missing data 24 (2.9%) 40 (1.5%) Cancer - no (%) 0.036 No 694 (84.9%) 2,338 (88.8%) Yes 101 (12.4%) 262 (9.9%) missing data 22 (2.6%) 34 (1.3%) CKD stage ∗∗ - no (%) < .0001 Stage 1 241 (29.5%) 970 (36.8%) Stage 2 281 (34.4%) 991 (37.6%) Stage 3a or stage 3b 180 (22.0%) 487 (18.5%) Stage 4 or stage 5 89 (10.9%) 143 (5.4%) missing data 26 (3.2%) 43 (1.6%) C Reactive Protein - no (%) 0.0003 < 1 mg/L 104 (12.7%) 256 (9.7%) 1-3 mg/L 120 (14.7%) 301 (11.4%) > 3 mg/L 549 (67.2%) 1,943 (73.8%) missing data 44 (5.4%) 134 (5.1%)
Lopinavir or Darunavir use < .0001
No 621 (76.0%) 1,203 (36.7%) Yes 196 (24.0%) 1,431 (64.3%)
Tocilizumab or Sarilumab use < .0001
No 755 (92.4%) 2,160 (82.0%) Yes 62 (7.6%) 474 (18.0%) Remdesivir use 0.0015 No 808 (98.9%) 2,551 (96.9%) Yes 9 (1.1%) 83 (3.1%) Corticosteroids use < .0001 No 596 (73.0%) 1,655 (62.8%) Yes 221 (27.0%) 979 (37.2%) Clusters of hospitals < .0001
Northern regions (except Milan) (n) 169 (20.7%) 616 (23.4%) Milan (m) 161 (19.7%) 525 (19.9%) Center regions (except Rome) (c)) 303 (37.1%) 747 (28.4%) Rome (r) 94 (11.5%) 390 (14.8%) Southern regions (s) 90 (11.0%) 356 (13.5%)
(n)includehospitalsofNovara,Monza,Varese,Pavia,CremonaandPadova;(m)include Hu-manitasClinicalandResearchHospital,CentroCardiologicoMonzino,andhospitalsofSan DonatoMilanese(Milano)andCiniselloBalsamo(Milano);(c)includehospitalsofModena, Ravenna,Forlì,Firenze,Pisa,ChietiandPescara;(r)includeNationalInstituteforInfectious Dis-eases“L.Spallanzani” andUniversità CattolicadelSacroCuore;(s)includehospitalofNapoli, Pozzilli(Isernia),AcquavivadelleFonti(Bari),Foggia,Taranto,Catanzaro,CataniaandPalermo
∗Chi-squaretest.∗∗Stage1:Kidneydamagewithnormalorincreasedglomerularfiltrationrate
(GFR)(>90mL/min/1.73m2);Stage2:MildreductioninGFR(60-89mL/min/1.73m2);Stage
3a:ModeratereductioninGFR(45-59mL/min/1.73m2);Stage3b:ModeratereductioninGFR
(30-44mL/min/1.73m2);Stage4:SeverereductioninGFR(15-29mL/min/1.73m2);Stage5:
Kidneyfailure(GFR<15mL/min/1.73m2ordialysis).
twice daily for at least 5-7 days in patients over 70 years and/or
281
withco-morbidities(chronicobstructivepulmonarydisease,diabetes,
282
cardiovasculardisease)evenwithmildrespiratorysymptomsorwith
283
radiographically documented pneumonia or in severe patients [36] .
284
Thelowerdosesof HCQusedin ourcenters,assuggested byItalian
285
official guidelines [19 ,36 ], mayhave been both more effective and 286
safer. 287
Tworecentlypublishedlargeobservationalstudies,bothfromlarge 288
hospitalsinNewYorkCity,showednoassociationbetweenHCQuse 289
European Journal of Internal Medicine 000 (2020) 1–10
Table2
IncidenceratesandhazardratiosfordeathinCOVID-19patients,accordingtohydroxychloroquineuse.
Multiple imputation analysis (N = 3,451)
Death (N = 576) Patient at risk (N = 3,451) Person-days Death Rate (x1,000 person-days)
Hydroxychloroquine
No- no. (%) 190 (23.3%) 817 (100%) 12,084 15.7 Yes- no. (%) 386 (14.7%) 2,634 (100%) 43,304 8.9
Hazard ratio for death (HCQ versus non HCQ) HR (95% CI)
Crude analysis 0.56 (0.47 to 0.67) Multivariable analysis ∗ 0.70 (0.58 to 0.85)
Propensity score analysis, inverse probability weighting ∗∗ ( primary analysis ) 0.70 (0.59 to 0.84)
Propensity score analysis, stratification (n = 5 strata) ∗∗ 0.67 (0.56 to 0.81)
Odds ratio for death (HCQ versus non HCQ) OR (95% CI) Propensity score analysis, inverse probability weighting ∗∗ 0.67 (0.54 to 0.82)
Case Complete Analysis (N = 3,156)
Death (N = 510) Patient at risk (N = 3,156) Person-days Death Rate (x1,000 person-days)
Hydroxychloroquine
No- no. (%) 170 (22.9%) 741 (100%) 11,050 15.4 Yes- no. (%) 340 (14.1%) 2,415 (100%) 39,274 8.7
Hazard ratio for death (HCQ versus non HCQ) HR (95% CI)
Crude analysis 0.56 (0.46 to 0.67) Multivariable analysis ∗ 0.71 (0.59 to 0.86)
Propensity score analysis, inverse probability weighting ∗∗ 0.64 (0.53 to 0.76)
Propensity score analysis, stratification (n = 5 strata) ∗∗ 0.68 (0.56 to 0.82)
Odds ratio for death (HCQ versus non HCQ) OR (95% CI) Propensity score analysis, inverse probability weighting ∗∗ 0.67 (0.54 to 0.82)
Abbreviations: HR, hazard ratios;CI, confidence intervals. ∗Controlling forage, sex,diabetes, hypertension, historyof
ischemic heart disease, chronic pulmonary disease, chronic kidney disease, C-reactive protein, lopinavir/ritonavir or darunavir/cobicistat,tocilizumaborsarilumab,remdesivirorcorticosteroidsuseasfixedeffectsandhospitalsclusteringas randomeffect.∗∗Includinghospitalsclusteringasrandomeffectcovariate.
studyofGelerisetal.[27] ,thepercentageuseofHCQwaslowerthanin
291
Italy;moreover,inbothUSstudies[27 ,28 ]thedrugwasmorefrequently
292
administeredtopatientswithpreviousillnessesandamoresevere
pre-293
sentationofthedisease.Ourcohortincludedmilderpneumoniapatients
294
thantheUSpopulation,duetobetween-countrydifferencesin
indica-295
tionstothedrugforthebeginningoftherapy(e.g.,mildpneumoniain 296
ItalyversusonlyseverepneumoniaandARDSintheUS).Concomitant 297
useofotherdrugsforCOVID-19wasverylowinonestudy[27] andwas 298
notreportedintheotherstudy[28] .Inourcohort,patientsreceiving 299
HCQweremorelikelytreatedwithanotherdrugforCOVID-19treat- 300
Fig.1. TheunadjustedstandardizeddifferencesandstandardizeddifferencesadjustedbypropensityscoresbetweenHCQ-treatedandnon-HCQtreatedpatientsfor thevariablesincludedinthepropensityscore.Alldifferencesforthematchedobservationsarewithintherecommendedlimitsof–0.25and0.25,whichareindicated byreferencelines.
Fig.2. Survivalcurvesaccordingtohydroxychloroquineuse.Thecurvesareadjustedbypropensityscoreanalysis(inverseprobabilityfortreatmentweighting)and hospitalindexasrandomeffect,andaregeneratedusingthefirstimputeddataset.Theotherimputeddatasetsaresimilarandthusomitted.
Table3
Hazardratiosformortalityaccordingtohydroxychloroquineuseindifferentsubgroups.
Hydroxychloroquine NO (N = 817) Hydroxychloroquine YES (N = 2,634)
Subgroups No. death/patient at risk No. death/patient at risk HR (95% CI) ∗
Women 80/361 116/940 0.63 (0.46 to 0.86)
Men 110/456 270/1,694 0.74 (0.60 to 0.93)
Age < 70 years 22/357 93/1,542 0.76 (0.50 to 1.16)
Age ≥ 70 years 168/460 293/1,092 0.68 (0.56 to 0.83)
Highest degree of COVID-19 severity experienced at hospital
Mild pneumonia or less 28/424 40/1,358 0.70 (0.41 to 1.18) Severe pneumonia 80/253 172/764 0.76 (0.58 to 0.99) Acute respiratory distress syndrome 82/140 174/512 0.68 (0.52 to 0.90)
Use of other COVID-19 treatmentsˆ
No 101/439 64/570 0.63 (0.45 to 0.88)
Yes 89/378 322/2,064 0.77 (0.61 to 0.99)
C-Reactive Protein at basal ∗∗
< 10 mg/L 56/412 125/1,138 1.23 (0.86 to 1.77)
≥ 10 mg/L 123/361 241/1,362 0.59 (0.47 to 0.73)
Abbreviations:HR,hazardratios;CI,confidenceintervals;∗Propensityscoreanalysis,inverseprobabilityweighting,including
hos-pitalclusteringasrandomeffectcovariate;multipleimputedanalysis.
ˆLopinavir/ritonavirordarunavir/cobicistatortocilizumaborsarilumaborremdesivirorcorticosteroids.
∗∗MissingdataforN=178.Frequenciesandhazardratiosarebasedonacasecompleteanalysis(N=3,273)withoutmissingdatafor
C-reactiveProtein;multipleimputedanalysis(N=3,451)yieldedverysimilarresults.
ment(78.4%),incomparisonwithnon-HCQpatients(46.3%).Anyway,
301
ourfindingsareadjustedforconcomitantotherdrugsuse.
302
WhiletheUSstudieswereconfinedtoonehospitalonlyoradefined
303
relativelysmallareaintheCountry,ourstudyincluded33hospitals
304
distributedalloverItaly,coveringregionswithahighnumberofcases
305
andahighintra-hospitalmortalityandregions withalowerburden
306
ofthedisease.TheparticipatingItalianclinicalcentershavedifferent
307
healthcarefacilities,differentsize,specialization,andownership,and
308
thereforequitecloselyrepresentthereal-lifeItalianapproachto
COVID-309
19.Moreover,theydifferedforthepercentageofuseofHCQandfor
310
therateofin-hospitalmortalitythatrangedbetween34.1and1.5per
311
1,000persons/day.Toconsiderthisvariability,weadjustedthe
analy-312
sisforrecruitingcenterandperformedanumberofsubgroupanalyses.
313
Inallcircumstances,theassociationbetweenHCQuseandareduced
314
riskofdeathofabout30%wasmaintained.Quiteinterestingly,the
in-315
verseassociationofHCQwithinpatientmortalitywasmoreevidentin 316
elderly,inpatientswhoexperiencedahigherdegreeofCOVID-19sever- 317
ityorespeciallyhavingelevatedC-reactiveprotein,suggestingthatthe 318
anti-inflammatorypotentialofHCQmayhavehadmoreimportantrole 319
ratherthanitsantiviralproperties.HCQ,indeed,besideanantiviralac- 320
tivity,mayhavebothanti-inflammatoryandanti-thromboticeffects[8] . 321
Thiscanjustifyitseffectinreducingmortalityrisk,sinceSars-Cov-2can 322
inducepulmonarymicrothrombiandcoagulopathy,thatareapossible 323
causeofitsseverity[37 ,38 ]andthelackinpreventingSARS-CoV-2in- 324
fectionafterexposure[26] 325
Nevertheless,largerandomizedclinicaltrialsontheefficacyofHCQ 326
onhardend-pointsarestilllackingandthelargestobservationalstudy 327
showing no effect in reducing mortalityhas beenretracted [16 ,17 ], 328
Agencies have suspended clinical trials on the efficacy of HCQ on 329
European Journal of Internal Medicine 000 (2020) 1–10
inclinicaltrials,intheabsenceofanample,sereneandbalanced
dis-331
cussionatinternationallevel.
332
Veryrecently,alargeRCThasbecomeavailableasapre-print
pub-333
lication[39] ,reportingnobeneficialeffectofHCQinpatients
hospital-334
izedwithCOVID-19.However,thedoseofHCQusedinthattrialwas
335
almostthedoubleofthatadministeredinourreallifeconditions.A
re-336
ducedmortalitywasalsoobservedbyotherobservationalstudiesusing
337
loworintermediatedosesofHCQ[40 ,41 ].
338
Moreover, in our studypatients takingHCQ more frequently
re-339
ceivedotheranti-COVIDdrugs,whoseinteractioninreducingmortality
340
cannotbecompletelyruled-out.Ofnote,despitethehigherdosageused,
341
theRCTdidnotshowanyexcessinventriculartachycardiaor
ventric-342
ularfibrillationintheHCQarm(39).
343
Therefore,itwillbe veryimportanttocompareresults ofstudies
344
withdifferentmodeofuseanddosesofHCQ,differentcharacteristics
345
oftreatedanduntreatedpatientsanddifferentacademicorreal-world
346
conditions.
347
4.1. Strengthsandlimitations
348
Amajorstrengthofthisstudyisthelarge,unselectedpatientsample
349
from33hospitals,coveringtheentireItalianterritory.Patientsampling
350
coveredalltheovert epidemicperiod inItaly. Severalstatistical
ap-351
proacheswereusedtoovercomebiasesduetotheobservationalnature
352
oftheinvestigation.
353
Thisstudyhashowever,severalrecognizedlimitations.Thestudy
354
populationpertainstoItaly,andtheresultsobtainedmaynotbe
appli-355
cabletootherpopulationswithapossiblydifferentgeographicaland
356
socio-economicconditionsandnaturalhistoryofCOVID-19.Duetothe
357
retrospectivenatureofthestudy,someparameterswerenotavailable
358
inallpatients,andallin-hospitalmedicationsmighthavebeennotfully
359
recorded.Moreover,althoughguidelinesontheuseofHCQin
COVID-360
19patientshadbeenpublishedinItalysincethefirstphaseofthe
pan-361
demic,individualcenterscouldhavedeviatedfromrecommendations
362
anduseddifferentdosesortreatmentschemes.Wehavenoinformation
363
ontheHCQdosesusedindividuallynoroftheirpossibleassociationwith
364
azithromycin.Moreover,adverseeventspossiblyrelatedtodrugtherapy
365
werenotcollected,thuswecannotexcludebiasduetotherapy
inter-366
ruptionbecauseofsideeffects;wedonotknowwhethersomedeaths
367
couldhavebeenduetocardiovascularcomplicationsofHCQ.However,
368
recentdataonItalianwardsshowedthatCOVID-19patientsreceiving
369
HCQandazithromycinhadaQTc-intervallongerthanbeforetherapy,
370
butdidnotexperience,duringtheirhospitalstay,anyarrhythmic
com-371
plications,suchassyncopeorlife-threateningventriculararrhythmias
372
[42] ,afindingalsoreportedbytheRCTmentionedabove(39).
373
Finally,thepossibilityofunmeasuredresidualconfoundingcannot
374
becompletelyruled-out.However,theE-valueforthelowerboundary
375
oftheconfidenceintervalofourmainresultis1.67,indicatingthatthe
376
confidenceintervalcouldbemovedtoincludethenullbyastrong
un-377
measuredconfounderassociatedwithbothHCQtreatmentanddeath
378
withariskratioof1.67-foldforeach,aboveandbeyondallthe
mea-379
suredconfounders.Weakerconfounders,however,couldnotdoso.
380
5. Conclusions
381
Ourstudy,includingalargereallifesampleofpatientshospitalized
382
withCOVID-19alloverItaly,showsthatHCQuse(200mgtwice/day)
383
wasassociatedwitha30%reductionofoverallin-hospitalmortality.
384
Intheabsenceofclear-cutresultsfromcontrolled,randomizedclinical
385
trials,ourdatado notdiscouragetheuse of HCQin inpatientswith
386
COVID-19.Giventheobservationaldesignofourstudy,however,these
387
resultsshouldbetransferredwithcautiontoclinicalpractice.
388
Sourceoffunding
389
None.
390
DeclarationofCompetingInterests 391
None. 392
CRediTauthorshipcontributionstatement 393
AugustoDiCastelnuovo:Conceptualization,Datacuration,Inves- 394
tigation, Supervision, Writing - review & editing, Writing - original 395
draft.SimonaCostanzo:Investigation,Supervision,Writing-review 396
& editing.AndreaAntinori:Investigation,Supervision,Writing-re- 397
view&editing.NausicaaBerselli:Investigation,Supervision,Writing 398
-review&editing.LorenzoBlandi:Investigation,Supervision,Writing 399
-review&editing.RaffaeleBruno:Investigation,Supervision,Writing 400
-review&editing.RobertoCauda:Investigation,Supervision,Writ- 401
ing-review&editing.GiovanniGuaraldi:Investigation,Supervision, 402
Writing-review&editing.LorenzoMenicanti:Investigation,Super- 403
vision,Writing -review& editing.IlariaMy:Investigation,Supervi- 404
sion,Writing-review&editing.GiustinoParruti:Investigation,Su- 405
pervision, Writing-review&editing. GiuseppePatti:Investigation, 406
Supervision,Writing-review&editing.StefanoPerlini:Investigation, 407
Supervision, Writing -review& editing.Francesca Santilli:Investi- 408
gation,Supervision,Writing-review&editing.CarloSignorelli:In- 409
vestigation,Supervision,Writing-review&editing.EnricoSpinoni: 410
Investigation,Supervision,Writing-review&editing.GiulioG.Ste- 411
fanini:Investigation,Supervision,Writing-review&editing,Formal 412
analysis.AlessandraVergori:Investigation,Supervision,Writing-re- 413
view & editing. WalterAgeno: Investigation, Supervision, Writing- 414
review& editing.Antonella Agodi:Investigation,Supervision,Writ- 415
ing-review&editing.LucaAiello:Investigation,Supervision,Writ- 416
ing-review&editing.Piergiuseppe Agostoni:Investigation,Super- 417
vision,Writing-review&editing.SamirAlMoghazi:Investigation, 418
Supervision,Writing-review&editing.MarinellaAstuto:Investiga- 419
tion,Supervision,Writing-review&editing.FilippoAucella:Inves- 420
tigation, Supervision,Writing-review&editing.GretaBarbieri:In- 421
vestigation,Supervision,Writing-review&editing.AlessandroBar- 422
toloni:Investigation, Supervision,Writing -review& editing. Mari- 423
alauraBonaccio:Investigation,Supervision,Writing-review&editing. 424
PaoloBonfanti:Investigation,Supervision,Writing-review&editing. 425
FrancescoCacciatore:Investigation,Supervision,Writing-review& 426
editing.LuciaCaiano:Investigation,Supervision,Writing-review& 427
editing.FrancescoCannata:Investigation,Supervision,Writing-re- 428
view&editing. LauraCarrozzi:Investigation,Supervision,Writing- 429
review&editing.AntonioCascio:Investigation,Supervision,Writing 430
-review&editing.ArturoCiccullo:Investigation,Supervision,Writ- 431
ing-review&editing.AntonellaCingolani:Investigation,Supervision, 432
Writing-review&editing.FrancescoCipollone:Investigation,Super- 433
vision,Writing-review&editing.ClaudiaColomba:Investigation,Su- 434
pervision,Writing-review&editing.FrancescaCrosta:Investigation, 435
Supervision,Writing-review&editing.ChiaraDalPra:Investigation, 436
Supervision, Writing-review& editing.GianBattistaDanzi: Inves- 437
tigation, Supervision,Writing-review& editing.DamianoD’Ardes: 438
Investigation,Supervision,Writing-review&editing.KatleendeGae- 439
tanoDonati: Investigation,Supervision, Writing -review& editing, 440
Writing -originaldraft. PaolaDel Giacomo:Investigation,Supervi- 441
sion,Writing-review&editing.FrancescoDiGennaro:Investigation, 442
Supervision, Writing-review& editing.Giuseppe DiTano:Investi- 443
gation,Supervision,Writing-review&editing. GiampieroD’Offizi: 444
Investigation,Supervision,Writing-review&editing.TommasoFilip- 445
pini:Investigation,Supervision,Writing-review&editing.Francesco 446
Maria Fusco:Investigation,Supervision, Writing-review& editing. 447
Ivan Gentile:Investigation,Supervision,Writing -review& editing. 448
AlessandroGialluisi:Investigation, Supervision,Writing -review& 449
editing.GiancarloGini:Investigation,Supervision,Writing-review& 450
editing.ElviraGrandone:Investigation,Supervision,Writing-review 451
& editing.Leonardo Grisafi:Investigation,Supervision,Writing- 452
review&editing.GabriellaGuarnieri:Investigation,Supervision, 453 7
Writing-review&editing.SilviaLamonica:Investigation,
Super-454
vision,Writing-review&editing.FrancescoLandi:Investigation,
455
Supervision,Writing-review&editing.ArmandoLeone:
Investi-456
gation,Supervision,Writing-review&editing.GloriaMaccagni:
457
Investigation,Supervision,Writing-review&editing.Sandro
Mac-458
carella:Investigation,Supervision,Writing-review&editing.
An-459
dreaMadaro:Investigation,Supervision,Writing-review&
edit-460
ing.MassimoMapelli:Investigation,Supervision,Writing-review
461
&editing.RiccardoMaragna:Investigation,Supervision,Writing
-462
review&editing.LorenzoMarra:Investigation,Supervision,
Writ-463
ing-review&editing.GiulioMaresca:Investigation,Supervision,
464
Writing-review&editing.ClaudiaMarotta:Investigation,
Super-465
vision,Writing-review&editing.FrancoMastroianni:
Investiga-466
tion,Supervision,Writing-review&editing,Methodology.Maria
467
Mazzitelli:Investigation,Supervision,Writing-review&editing.
468
AlessandroMengozzi:Investigation,Supervision,Writing-review
469
&editing.FrancescoMenichetti:Investigation,Supervision,
Writ-470
ing-review&editing.MariannaMeschiari:Investigation,
Super-471
vision,Writing-review&editing.FilippoMinutolo:Investigation,
472
Supervision,Writing-review&editing.ArturoMontineri:
Investi-473
gation,Supervision,Writing-review&editing.RobertaMussinelli:
474
Investigation, Supervision, Writing - review & editing. Cristina
475
Mussini: Investigation,Supervision, Writing - review& editing.
476
MariaMusso:Investigation,Supervision,Writing-review&
edit-477
ing.AnnaOdone:Investigation,Supervision,Writing-review&
478
editing.MarcoOlivieri:Investigation,Supervision,Writing-review&
479
editing,Software.EmanuelaPasi:Investigation,Supervision,Writing
480
-review&editing.FrancescoPetri:Investigation,Supervision,
Writ-481
ing- review& editing. BiagioPinchera: Investigation, Supervision,
482
Writing-review&editing.CarloA.Pivato:Investigation,Supervision,
483
Writing-review&editing. VenerinoPoletti:Investigation,
Supervi-484
sion,Writing-review&editing.ClaudiaRavaglia:Investigation,
Su-485
pervision,Writing-review&editing.MassimoRinaldi:Investigation,
486
Supervision,Writing-review& editing.AndreaRognoni:
Investiga-487
tion,Supervision,Writing-review&editing.MarcoRossato:
Investi-488
gation,Supervision,Writing-review&editing.IlariaRossi:
Investiga-489
tion,Supervision,Writing-review&editing.MariannaRossi:
Inves-490
tigation,Supervision,Writing-review&editing.AnnaSabena:
Inves-491
tigation,Supervision,Writing-review&editing.FrancescoSalinaro:
492
Investigation,Supervision,Writing-review&editing.Vincenzo
San-493
giovanni:Investigation,Supervision,Writing-review&editing.Carlo
494
Sanrocco:Investigation,Supervision,Writing-review&editing.Laura
495
Scorzolini:Investigation,Supervision,Writing-review&editing.
Raf-496
faellaSgariglia:Investigation,Supervision,Writing-review&editing.
497
PaolaGiustinaSimeone:Investigation,Supervision,Writing-review&
498
editing.MicheleSpinicci:Investigation,Supervision,Writing-review
499
&editing.EnricoMariaTrecarichi:Investigation,Supervision,Writing
500
-review&editing.AmedeoVenezia:Investigation,Supervision,
Writ-501
ing-review&editing.GiovanniVeronesi:Investigation,Supervision,
502
Writing-review&editing,Formalanalysis.RobertoVettor:
Investiga-503
tion,Supervision,Writing-review&editing.AndreaVianello:
Investi-504
gation,Supervision,Writing-review&editing.MarcoVinceti:
Investi-505
gation,Supervision,Writing-review&editing.LauraVocciante:
Inves-506
tigation,Supervision,Writing-review&editing.RaffaeleDeCaterina:
507
Conceptualization,Investigation,Supervision,Writing-review&
edit-508
ing,Writing-originaldraft.LiciaIacoviello:Conceptualization,Data
509
curation,Investigation,Investigation,Supervision,Writing-review&
510
editing,Writing-originaldraft.
511
Acknowledgments
512
Wethanktheparticipatingclinicalcentresincludedinthiscohort.
513
ThisArticleisdedicatedtoallthepatientswhosufferedordied,often
514
insolitude,duetoCOVID-19;theirtragicfategaveusmoralstrengthto
515
initiateandcompletethisresearch.
516
TheAuthorsaloneareresponsiblefortheviewsexpressedinthisAr- 517
ticle.Theydonotnecessarilyrepresenttheviews,decisions,orpolicies 518
oftheInstitutionswithwhichtheyareaffiliated. 519
Appendix1 520
Augusto Di Castelnuovoa, Simona Costanzob, Andrea Antinoric, 521
Nausicaa Bersellid, Lorenzo Blandie, Raffaele Brunof,g, Roberto 522
Caudah,i,GiovanniGuaraldij,LorenzoMenicantie,IlariaMyk,Giustino 523
Parrutil, Giuseppe Pattim, Stefano Perlinin,o, Francesca Santillip, 524
Carlo Signorelliq, Enrico Spinonim, Giulio G. Stefaninik, Alessandra 525
Vergorir,WalterAgenos,AntonellaAgodit,LucaAiellou,Piergiuseppe 526
Agostoniv,w,Samir AlMoghazix, MarinellaAstutot, FilippoAucellay, 527
Greta Barbieriz, AlessandroBartoloniaa, MarialauraBonacciob, Paolo 528
Bonfantiab,ac, Francesco Cacciatoread, Lucia Caianos, Francesco 529
Cannatak, Laura Carrozziae, Antonio Cascioaf, Arturo Cicculloh, 530
Antonella Cingolanih,i, Francesco Cipollonep, Claudia Colombaaf, 531
Francesca Crostal, Chiara Dal Praag, Gian Battista Danziah, Dami- 532
ano D’Ardesp, Katleen de Gaetano Donatih, Paola Del Giacomoh, 533
Francesco Di Gennaroai, Giuseppe Di Tanoah, Giampiero D’Offiziaj, 534
TommasoFilippinid,FrancescoMariaFuscoak,IvanGentileal,Alessan- 535
dro Gialluisib, GiancarloGinis, ElviraGrandoney, Leonardo Grisafim, 536
Gabriella Guarnieriam, Silvia Lamonicah, FrancescoLandiu,Armando 537
Leonean, Gloria Maccagniah, SandroMaccarellaao, Andrea Madaroap, 538
Massimo Mapelliv,w, Riccardo Maragnav,w, Lorenzo Marraan, Giulio 539
Marescaaq,ClaudiaMarottaai,FrancoMastroianniap,MariaMazzitelliar, 540
Alessandro Mengozziz, Francesco Menichettiz, Marianna Meschiarij, 541
Filippo Minutoloas, ArturoMontineriat, Roberta Mussinelliq, Cristina 542
Mussinij, Maria Mussoau, AnnaOdoneq, Marco Olivieriav, Emanuela 543
Pasiaw,FrancescoPetriab,BiagioPincheraal,CarloA.Pivatok,Venerino 544
Polettiax, Claudia Ravagliaax, Massimo Rinaldiap, Andrea Rognonim, 545
Marco Rossatoag, Ilaria Rossip, Marianna Rossiab, Anna Sabenan, 546
Francesco Salinaron, VincenzoSangiovanniak, CarloSanroccol,Laura 547
Scorzoliniay, Raffaella Sgarigliaaq, Paola Giustina Simeonel, Michele 548
Spinicciaa, Enrico Maria Trecarichiar, Amedeo Veneziaap, Giovanni 549
Veronesis, Roberto Vettorag, Andrea Vianelloam, Marco Vincetid,az, 550
LauraVoccianteaq,RaffaeleDeCaterinaai,LiciaIacoviellob,s 551 aMediterraneaCardiocentro,Napoli.ItalybDepartmentofEpidemi- 552
ologyandPrevention,IRCCSNeuromed,Pozzilli(IS).ItalycUOCIm- 553
munodeficienze Virali, National Institute for Infectious Diseases “L. 554
Spallanzani”,IRCCS.Roma.ItalydSectionofPublicHealth,Department 555
of Biomedical, MetabolicandNeuralSciences,Universityof Modena 556
andReggioEmilia,Modena.ItalyeIRCCSPoliclinicoSanDonato,San 557
DonatoMilanese.ItalyfDivisionofInfectiousDiseasesI,FondazioneIR- 558
CCSPoliclinicoSanMatteo,Pavia.ItalygDepartmentofClinical,Sur- 559
gical, Diagnostic,andPaediatricSciences,Universityof Pavia,Pavia. 560
Italy hFondazionePoliclinico UniversitarioA. Gemelli IRCCS,Roma. 561
ItalyyiUniversità CattolicadelSacroCuore-DipartimentodiSicurezza 562
eBioeticaSedediRoma,Roma.ItalyjInfectiousDiseaseUnit,Depart- 563
mentofSurgical,Medical,DentalandMorphologicalSciences,Univer- 564
sityofModenaandReggioEmilia,Modena.ItalykHumanitasClinical 565
andResearchHospitalIRCCS,Rozzano-Milano.ItalylDepartmentofIn- 566
fectiousDisease,AziendaSanitariaLocale(AUSL)diPescara,Pescara. 567
Italy mUniversityof Eastern Piedmont, MaggioredellaCarità Hospi- 568
tal,Novara.ItalynEmergencyDepartment,IRCCSPoliclinicoSanMat- 569
teoFoundation,Pavia.ItalyoDepartmentofInternalMedicine,Univer- 570
sityof Pavia, Pavia.Italy pDepartmentof MedicineandAging,Clin- 571
icaMedica,“SS.Annunziata” HospitalandUniversityofChieti,Chieti. 572
ItalyqSchoolofMedicine,Vita-SaluteSanRaffaeleUniversity,Milano. 573
ItalyrHIV/AIDSDepartment,NationalInstituteforInfectiousDiseases 574
“LazzaroSpallanzani”-IRCCS,Roma.ItalysDepartmentofMedicineand 575
Surgery, University of Insubria, Varese. Italy tDepartment of Medi- 576
cal and Surgical Sciences and Advanced Technologies “G.F. Ingras- 577
sia”,UniversityofCatania;AOUPoliclinico-VittorioEmanuele, Cata- 578
nia. Italy uUOC. Anestesiae Rianimazione. Dipartimento di Chirur- 579
European Journal of Internal Medicine 000 (2020) 1–10
diologicoMonzinoIRCCS, Milano.ItalywDepartmentofClinical
Sci-581
ences andCommunity Health, Cardiovascular Section, Universityof
582
Milano,Milano.ItalyxInfezioniSistemichedell’Immunodepresso,
Na-583
tionalInstitute for Infectious Diseases L.Spallanzani, IRCCS, Roma.
584
ItalyyFondazioneI.R.C.C.S“CasaSollievodellaSofferenza”,San
Gio-585
vanniRotondo,Foggia.ItalyzDepartmentofClinicaland
Experimen-586
tal Medicine,Azienda Ospedaliero-Universitaria Pisana,and
Univer-587
sity of Pisa, Pisa. Italy aaDepartment of Experimental and Clinical
588
Medicine. University of Florence, Firenze. Italy abUOC Malattie
In-589
fettive, Ospedale San Gerardo, ASST Monza, Monza. Italy acSchool
590
ofMedicineandSurgery,Universityof Milano-Bicocca,Milano. Italy
591
adDepartmentofTranslationalMedicalSciences.UniversityofNaples,
592
Federico II,Napoli. ItalyaeCardiovascularandThoracic Department,
593
AziendaOspedaliero-UniversitariaPisana,andUniversityofPisa,Pisa.
594
ItalyafInfectiousandTropicalDiseasesUnit-DepartmentofHealth
Pro-595
motion,Mother andChildCare, InternalMedicineandMedical
Spe-596
cialties (PROMISE) -Universityof Palermo,Palermo. Italy agClinica
597
Medica 3, Department of Medicine- DIMED, Universityhospital of
598
Padova,Padova. ItalyahDepartmentof Cardiology,Ospedale di
Cre-599
mona,Cremona.ItalyaiMedicalDirection, IRCCSNeuromed,Pozzilli
600
(IS).ItalyajUOCMalattieInfettive-Epatologia,NationalInstitutefor
In-601
fectiousDiseases L.Spallanzani, IRCCS, Roma.ItalyakUOCInfezioni
602
Sistemicheedell’Immunodepresso,AziendaOspedalieradeiColli,
Os-603
pedaleCotugno.Napoli. ItalyalDepartmentof ClinicalMedicineand
604
Surgery.UniversityofNaples“FedericoII”,Napoli.ItalyamRespiratory
605
PathophysiologyDivision,DepartmentofCardiologic,Thoracicand
Vas-606
cular Sciences,University of Padova, Padova. Italy anUOC di
Pneu-607
mologia, P.O. San Giuseppe Moscati, Taranto. Italy aoASST Milano
608
Nord- Ospedale Edoardo Bassini, CiniselloBalsamo.Italy ap
COVID-609
19Unit.EEOspedaleRegionaleF.Miulli,AcquavivadelleFonti(BA).
610
ItalyaqUOCMedicina -POS.Mariadi LoretoNuovo-ASL Napoli1
611
Centro.NapoliarInfectiousandTropicalDiseasesUnit.Deparmentof
612
MedicalandSurgicalSciences“MagnaGraecia” University,Catanzaro.
613
ItalyasDipartimentodiFarmacia,Università diPisa,Pisa,Italy.atU.O.
614
C.MalattieInfettive eTropicali,P.O.“SanMarco”,AOU
Policlinico-615
Vittorio Emanuele, Catania.Italy auUOC Malattie Infettive-Apparato
616
Respiratorio,NationalInstituteforInfectiousDiseases“L.Spallanzani”,
617
IRCCS, Roma. Italy avComputer Service, University of Molise,
Cam-618
pobasso.Italy.awMedicinaInterna.OspedalediRavenna.AUSLdella
619
Romagna,Ravenna.ItalyaxUOCPneumologia.DipartimentodiMalattie
620
ApparatoRespiratorioeTorace.OspedaleMorgagni-PierantoniForlì,
621
Forlì. Italy ayUOC Malattie Infettive ad Alta Intensità di Cura,
Na-622
tionalInstituteforInfectiousDiseases“L.Spallanzani”,IRCCS,Roma.
623
ItalyazDepartmentofEpidemiology,BostonUniversitySchoolofPublic
624
Health,Boston.USA.
625
Supplementarymaterials
626
Supplementarymaterialassociatedwiththisarticlecanbefound,in
627
theonlineversion,atdoi:10.1016/j.ejim.2020.08.019 .
628
References
629
[1] Shukla AM, Wagle Shukla A. Expanding horizons for clinical applications of 630
chloroquine, hydroxychloroquine, and related structural analogues. Drugs Context 631
2019;8:2019 9-1Published 2019 Nov 25. doi: 10.7573/dic.2019-9-1 .
632
[2] Savarino A, Boelaert JR, Cassone A, Majori G, Cauda R. Effects of chloro- 633
quine on viral infections: an old drug against today’s diseases? Lancet Infect Dis 634
2003;3(11):722–7. doi: 10.1016/s1473-3099(03)00806-5 .
635
[3] Dowall SD, Bosworth A, Watson R, et al. Chloroquine inhibited Ebola virus replica- 636
tion in vitro but failed to protect against infection and disease in the in vivo guinea 637
pig model. J Gen Virol 2015;96(12):3484–92. doi: 10.1099/jgv.0.000309 .
638
[4] Akpovwa H. Chloroquine could be used for the treatment of filoviral infections and 639
other viral infections that emerge or emerged from viruses requiring an acidic pH 640
for infectivity. Cell Biochem Funct 2016;34(4):191–6. doi: 10.1002/cbf.3182 .
641
[5] Savarino A, Shytaj IL. Chloroquine and beyond: exploring anti-rheumatic drugs to 642
reduce immune hyperactivation in HIV/AIDS. Retrovirology 2015;12:51 Published 643
2015 Jun 18. doi: 10.1186/s12977-015-0178-0 .
644
[6] de Wilde AH, Jochmans D, Posthuma CC, et al. Screening of an FDA-approved com- 645
pound library identifies four small-molecule inhibitors of Middle East respiratory 646
syndrome coronavirus replication in cell culture. Antimicrob Agents Chemother 647 2014;58(8):4875–84. doi: 10.1128/AAC.03011-14 . 648 [7] Devaux CA, Rolain JM, Colson P, Raoult D. New insights on the antiviral effects of 649 chloroquine against coronavirus: what to expect for COVID-19? Int J Antimicrob 650 Agents 2020, Mar 12. doi: 10.1016/j.ijantimicag.2020.105938 . 651 [8] Quiros Roldan E, Biasiotto G, Magro P, Zanella I. The possible mechanisms of action 652 of 4-aminoquinolines (chloroquine/hydroxychloroquine) against Sars-Cov-2 infec- 653 tion (COVID-19): a role for iron homeostasis? [published online ahead of print, 2020 654 May 13]. Pharmacol Res. 2020;158:104904. doi: 10.1016/j.phrs.2020.104904 . 655 [9] Vincent MJ, Bergeron E, Benjannet S, et al. Chloroquine is a potent inhibitor of 656 SARS coronavirus infection and spread. Virol J. 2005;2:69 Published 2005 Aug 22. 657 doi: 10.1186/1743-422X-2-69 . 658 [10] Mercuro NJ, Yen CF, Shim DJ, Maher TR, McCoy CM, Zimetbaum PJ, Gold HS. 659 Risk of QT interval prolongation associated with use of hydroxychloroquine with or 660 without concomitant azithromycin among hospitalized patients testing positive for 661 coronavirus disease 2019 (COVID-19). JAMA Cardiol 2020 May 1. doi: 10.1001/ja- 662
macardio.2020.1834 . 663
[11] Saleh M, Gabriels J, Chang D, et al. The effect of chloroquine, hydroxychloroquine 664 and azithromycin on the corrected QT interval in patients with SARS-CoV-2 Infec- 665 tion. Circ Arrhythm Electrophysiol. 2020 Apr 29. doi: 10.1161/CIRCEP.120.008662 . 666 [12] Food and Drug Aministration (FDA) Fact sheet for health care providers: emergency 667 use authorization (EUA) of hydroxychloroquine sulfate supplied from the strategic 668 national stockpile for treatment of COVID-19 in certain hospitalized patients. Posted 669 April 2020;27. Available at: https://www.fda.gov/media/136537/download (Ac- 670 cessed May 31, 2020) . 671 [13] European Medicines Agency. COVID-19: chloroquine and hydroxychloroquine 672 only to be used in clinical trials or emergency use programmes. Available at: 673
https://www.ema.europa.eu/en/news/covid-19-chloroquine-hydroxychloroquine- 674
only-be-used-clinical-trials-emergency-use-programmes . (Accessed June, 26, 2020). 675 [14] Agenzia Italiana del Farmaco (AIFA). Idrossiclorochina nella terapia dei pazi- 676 enti adulti con COVID-19. Available at: https://www.aifa.gov.it/documents/ 677
20142/1123276/idrossiclorochina_29.04.2020.pdf/386d6ea3-c79b-6437-f457- 678
23d33df74256 . (Accessed June, 26, 2020). 679 [15] Yao X, Ye F, Zhang M, et al. In vitro antiviral activity and projection of optimized 680 dosing design of hydroxychloroquine for the treatment of severe acute respira- 681 tory Syndrome coronavirus 2 (SARS-CoV-2). Clin Infect Dis 2020 Mar 9 ciaa237. 682 doi: 10.1093/cid/ciaa237 . 683 [16] Mehra MR, Desai SS, Ruschitzka F, Patel AN. Hydroxychloroquine or 684 chloroquine with or without a macrolide for treatment of COVID-19: a 685 multinational registry analysis. Lancet 2020. published online May 22. 686
https://doi.org/10.1016/S0140-6736(20)31180-6 . 687 [17] Mehra MR, Ruschitzka F, Patel AN. Retraction-Hydroxychloroquine or chloroquine 688 with or without a macrolide for treatment of COVID-19: a multinational registry 689 analysis [retraction of: Lancet. 2020 May 22;:]. Lancet 2020;395(10240):1820. 690 doi: 10.1016/S0140-6736(20)31324-6 . 691 [18] World Health Organization. Hydroxychloroquine and COVID-19. Available at: 692
https://www.who.int/emergencies/diseases/novel-coronavirus-2019/global- 693
research-on-novel-coronavirus-2019-ncov/solidarity-clinical-trial-for-covid-19- 694
treatments (Accessed June, 26, 2020). 695 [19] Agenzia Italiana del Farmaco (AIFA). AIFA sospende l’autorizzazione all’utilizzo 696 di idrossiclorochina per il trattamento del COVID-19 al di fuori degli studi 697 clinici Available at: https://www.aifa.gov.it/-/aifa-sospende-l-autorizzazione-all- 698
utilizzo-di-idrossiclorochina-per-il-trattamento-del-covid-19-al-di-fuori-degli-studi- 699
clinici . (Accessed May 31, 2020). 700
[20] Chowdhury MS, Rathod J, Gernsheimer J. A rapid systematic review of clinical trials 701 utilizing chloroquine and hydroxychloroquine as a treatment for COVID-19. Acad 702 Emerg Med 2020 May 2. doi: 10.1111/acem.14005 . 703 [21] Cortegiani A, Ingoglia G, Ippolito M, Giarratano A, Einav S. A systematic review on 704 the efficacy and safety of chloroquine for the treatment of COVID-19. J Crit Care 705 2020;57:279–83. doi: 10.1016/j.jcrc.2020.03.005 . 706 [22] Shah S, Das S, Jain A, Misra DP, Negi VS. A systematic review of the prophylactic 707 role of chloroquine and hydroxychloroquine in coronavirus disease-19 (COVID-19). 708 Int J Rheum Dis 2020;23(5):613–19. doi: 10.1111/1756-185X.13842 . 709 [23] Das S, Bhowmick S, Tiwari S, Sen S. An updated systematic review of the therapeutic 710 role of hydroxychloroquine in coronavirus disease-19 (COVID-19) [published online 711 ahead of print, 2020 May 28]. Clin Drug Investig. 2020 10.1007/s40261-020-00927- 712 1. doi: 10.1007/s40261-020-00927-1 . 713 [24] Hernandez AV, Roman YM, Pasupuleti V, Barboza JJ, White CM. Hydroxychloro- 714 quine or chloroquine for treatment or prophylaxis of COVID-19: a living system- 715 atic review [published online ahead of print, 2020 May 27]. Ann Intern Med. 2020 716 10.7326/M20-2496.. doi: 10.7326/M20-2496 . 717 [25] Tang W, Cao Z, Han M, et al. Hydroxychloroquine in patients with mainly mild to 718 moderate coronavirus disease 2019: open label, randomised controlled trial. BMJ 719 2020 May 14:369 m1849. doi: 10.1136/bmj.m1849 . 720 [26] Boulware DR, Pullen MF, Bangdiwala AS, et al. A randomized trial of hydroxychloro- 721 quine as postexposure prophylaxis for covid-19 [published online ahead of print, 722 2020 Jun 3]. N Engl J Med. 2020 NEJMoa2016638. doi: 10.1056/NEJMoa2016638 . 723 [27] Geleris J, Sun Y, Platt J, et al. Observational study of hydroxychloroquine in hospital- 724 ized patients with covid-19. N Engl J Med 2020;382(25):2411–18. doi: 10.1056/NE- 725
JMoa2012410 . 726
[28] Rosenberg ES, Dufort EM, Udo T, et al. Association of treatment with hydroxy- 727 chloroquine or azithromycin with in-hospital mortality in patients with COVID- 728 19 in New York State [published online ahead of print, 2020 May 11]. JAMA. 729 2020;323(24):2493–502. doi: 10.1001/jama.2020.8630 . 730 [29] COVID-19 Coronavirus pandemic worldmeters. [ https://www.worldometers.info/ 731
coronavirus/#countries (Accessed June, 26, 2020). 732 9