Stability of refrigerated whole blood samples for osmotic fragility test

Pleasecitethisarticleinpressas:SalvagnoGL,etal.Stabilityofrefrigeratedwholebloodsamplesforosmoticfragilitytest.HematolTransfus w w w . h t c t . c o m . b r

Hematology,

Transfusion

and

Cell

Therapy

Original

article

Stability

of

refrigerated

whole

blood

samples

for

osmotic

fragility

test

Gian

Luca

Salvagno,

Davide

Demonte,

Francesco

Dima,

Chiara

Bovo,

Giuseppe

Lippi

UniversityofVerona,Verona,Italy

a

r

t

i

c

l

e

i

n

f

o

a

b

s

t

r

a

c

t

Background:Theosmoticfragilitytest(OFT),conventionallyusedforassistingthediagnosis ofmanyerythrocytedisorders,isamanualandtime-consuminganalysisnotdaily per-formedinmanymedicallaboratories.Thisstudy wasaimedatdefiningthestabilityof wholebloodsamplesusedforassessingerythrocyteosmoticresistance.

Methods:Twenty-oneconsecutiveroutinewholebloodsamplescollectedinto5.4mgK2EDTA weretestedimmediatelyaftercollection(day0)andatdifferenttimeintervalsafterward (day1,2,3,4,7,10and14)afterstorageat4◦C.TheOFTwasperformedwiththeOsmored Monotest(1.3%glycerol;Eurospital,Trieste,Italy).Resultsatthedifferenttimepointswere comparedwiththoseobtainedatday0andwiththereferencechangevalue(i.e.,33%).

Results:Themedianvalueofbothhyperosmolarandhyposmolarresistanceincreasedfrom baseline,reachingstatisticalsignificanceatday7forhyperosmolarresistanceandatday 1forhyposmolarresistance,respectively.Themedianrelativeincreaseofhemolysis per-centagevaluesbecomegreaterthanthereferencechangevalueatday3forhyposmolar resistance, whilethis limitwasneverovercome forhyperosmolarresistance.A signifi-cantinverseassociationwasfoundbetweenthemeanincreaseinhyperosmolarresistance andthebaselinevalueofhyperosmolarresistance(r=−0.92),meancorpuscularvolume (MCV;r=−0.46)ormeancorpuscularhemoglobin(MCH;r=−0.44),aswellasbetweenthe meanincreaseinhyposmolarresistanceandthebaselinevalueofhyposmolarresistance (r=−0.86),orpatientage(r=−0.56).

Conclusions: ThesamplestabilityseemscriticalfortheOFT.Wholebloodspecimensshould notbestoredrefrigeratedat4◦Cfor>2daysbeforetesting.

©2019Associac¸ ˜aoBrasileiradeHematologia,HemoterapiaeTerapiaCelular.Published byElsevierEditoraLtda.ThisisanopenaccessarticleundertheCCBY-NC-NDlicense (http://creativecommons.org/licenses/by-nc-nd/4.0/).

∗ _{Correspondingauthorat:SectionofClinicalBiochemistry,UniversityHospitalofVerona,PiazzaleL.A.Scuro10,37134,Verona,Italy.} E-mailaddress:giuseppe.lippi@univr.it(G.Lippi).

https://doi.org/10.1016/j.htct.2019.06.001

2531-1379/©2019Associac¸ ˜aoBrasileiradeHematologia,HemoterapiaeTerapiaCelular.PublishedbyElsevierEditoraLtda.Thisisan openaccessarticleundertheCCBY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/4.0/).

Pleasecitethisarticleinpressas:SalvagnoGL,etal.Stabilityofrefrigeratedwholebloodsamplesforosmoticfragilitytest.HematolTransfus

Introduction

The osmotic fragilitytest (OFT) is conventionally used for assessing erythrocyte resistance to hemolysis, thus assist-ing in the diagnosis ofseveral erythrocyte disorders, such ashemolyticanemia,hereditaryspherocytosis,elliptocytosis andthalassemia.1Introducedatfirstinclinicallaboratories by Daciein 1954, the OFT analyzes erythrocyte hemolysis usingaseriesofhypotonicsolutionsofsodiumchloride(NaCl). In1980,Zanellaetal.introducedamodifiedversion ofthe classicDaciemethod,byaddingglyceroltoanisotonic phos-phatebufferedsaline.2_{Thelightwavelengthcommonlyused}

formeasurement is 540nm, withonly hemoglobin (asthe major erythrocyte protein) contributing to absorption. Ery-throcyte osmotic fragility is then defined by shifts in the hemolysis curve,by correlatinglight absorbance and NaCl concentration.2

TheOFTisanessentiallymanual,labor-intensiveand time-consuminganalysis,soitcannotbeperformeddailyinmost clinical laboratories.1 _{Whole blood samples may hence be}

storedfordaysbeforetesting.3_{Severallinesofevidencenow}

demonstrate that mostlaboratory errors occur inthe pre-analytical phase,primarilydue tothe lackofstandardized proceduresforsamplecollection(i.e.,transport,handlingand preparationbeforetesting)andstorage.4_{Withincertainareas}

oftesting,suchaslaboratoryhematology,theconsequencesof errorsmightoftenbeserious,especiallythoserelatedto spe-cializeddiagnostictests,sincetheymayleadtothegeneration offalsepositiveornegativetestresultsandmisdiagnosis,thus jeopardizingpatientsafetyandproducinganincrementalbut avoidableexpenseforthehealthcaresystem.Theclinical con-sequencesoflaboratoryerrorsmaybedetrimentalbecause theymaycontributetothedisruptionofbothclinicaland ther-apeuticdecisions.5AccordingtotheInternationalStandards Organization(ISO)15189:2012document,medicallaboratories shouldalwaysverifybloodsamplestabilityandassess poten-tialinfluenceofsomepreanalyticalvariables(i.e.,timeand temperatureofstoragebeforeanalyses).6_{RegardingtheOFT,}

thistestmaybeinfluencedbyseveralpreanalyticalvariables relatedtotheerythrocyteanaerobicmetabolism.7_Thisstudy

wasthereforeaimedatverifyingthestabilityofwholeblood samples used for assessing the erythrocyte osmotic resis-tance.

Materials

and

methods

Twenty-oneconsecutiveroutinewholebloodsamples(mean patient age, 68±9years; 6women and 15 men), collected into5.4mgK2EDTAtubes(VacutestKima,Padova,Italy)and

referred tothe locallaboratory for assessmentof the ery-throcyteosmotic resistance,were tested immediately after collection (day0) and at differenttime intervals afterward (1,2, 3, 4,7, 10 and 14 days). During the study period, all samples were kept refrigerated at 4◦C. TheOFT was per-formedwiththeOsmoredMonotest(1.3%glycerol;Eurospital, Trieste,Italy),whichusesbothhyperosmotic(normalvalue, >70%)andhyposmotic(normalvalue,<30%)solutions.Results atthe differenttimepointsofstoragewerecomparedwith thoseobtainedatday0withtheMann-Whitneypairedtest.

TherelativevariationoftheOFTresultsovertimewasalso comparedwiththereferencechangevalue(i.e.,33%), calcu-lated from the local analytical variabilityofthe assay (i.e., 1.3%)andfromrecentlypublisheddataonthebiological vari-ationintheerythrocyteosmoticstability(i.e.,12%)inhealthy, physicallyactivesubjects.8_{Potentialassociationsweretested}

withSpearman’scorrelation.Thestatisticalanalysiswas per-formedwithAnalyse-it(Analyse-itSoftwareLtd,Leeds,UK). Statisticalsignificancewassetatp<0.05.Thestudywasbased onanonymizedroutinesamplesreferredforroutineOFT,so thatwritteninformedconsentwasunnecessary.Theresearch wasclearedbythelocalEthicalCommittee(University Hos-pitalofVerona,n.971CESC,July25,2016)andconductedin accordancewiththeHelsinkiDeclaration,asrevisedin2008.

Results

The variation ofhyperosmolar and hyposmolar resistance (hemolysispercentage)inwholebloodsamplesstoredforup to14daysat4◦CisshowninTable1.Themedianvalueof bothhyperosmolarandhyposmolarresistanceprogressively increasedfrom baseline, reachingstatisticalsignificanceat day 7forhyperosmolar resistanceand atday 1for hypos-molarresistance,respectively.Themedianrelativeincrease inhemolysispercentagevaluesbecomehigherthanthe ref-erence change value (i.e., 33%) at day 3 for hyposmolar resistance, while this limit was never reached during the studyperiodforhyperosmolarresistance(Fig.1).Asignificant inverseassociationwasfoundbetweenthemeanincreasein hyperosmolarresistanceovertimeandthebaselinevalueof hyperosmolar resistance (r=−0.92;p<0.001),mean corpus-cularvolume(MCV;r=−0.46;p=0.034)ormeancorpuscular hemoglobin(MCH;r=−0.44;p=0.048),aswellasbetweenthe mean increaseinhyposmolarresistanceovertimeand the baselinevalueofhyposmolarresistance(r=−0.86;p<0.001), or age ofpatients (r=−0.56;p=0.008)(Fig. 2). Unlike these findings,themeanincreaseinthehyperosmolarresistance was notassociatedwiththeage (r=−0.41;p=0.067)orsex ofpatients (r=0.00;p=1.00)or hemoglobinvalue(r=−0.14; p=0.559),whilethemeanincreaseinhyposmolarresistance wasnotassociatedwiththesexofpatients(r=0.00;p=1.00), hemoglobin (r=0.30; p=0.189), MCV (r=−0.29; p=0.194) or MCH(r=−0.18;p=0.444).

Discussion

In2011,theWorldHealthOrganization(WHO)reinforcedthe conceptthatappropriatemanagementofpatientsamplesis criticalfortheaccuracyandreliabilityoflaboratorytesting,as wellasforthesafeguardingoftheclinicaldecisionmaking.9

Therefore, laboratories must be proactive in ensuring that allsamplesmeetthenecessaryrequirementsforproducing accuratetestresults,sincesamplemanagementispartofthe processcontrolandoneoftheessentialaspectsofthequality managementsystem.9_{Bloodsamplequalityremainsa}

main-stayforthequalityofthetotaltestingprocess becausethe receiptofunsuitablesamplesmaybeassociatedwith diagnos-ticdelay,missedorwrongdiagnoses,butmayalsogenerate

Pleasecitethisarticleinpressas:SalvagnoGL,etal.Stabilityofrefrigeratedwholebloodsamplesforosmoticfragilitytest.HematolTransfus

Table1–Variationinhyperosmolarandhyposmolarresistance(hemolysispercentage;medianvaluesandinterquartile

range)inwholebloodsamplestoredforupto14daysat4◦C.

Daysofstorage Test Baseline 1 2 3 4 7 10 14 Hyperosmolar resistance (hemolysis percentage) 84(32) 85(31) p=0.488 87(33) p=0.425 84(31) p=0.409 84(31) p=0.414 90(25) p=0.031 94(21) p=0.002 99(13) p=0.001 Hyposmolarresistance (hemolysis percentage) 20(18) 23(21) p<0.001 25(20) p<0.001 26(20) p<0.001 27(26) p<0.001 38(23) p<0.001 43(19) p<0.001 55(22) p<0.001

Fig.1–Variationofhyperosmolarandhyposmolarresistance(hemolysispercentage;medianvaluesandinterquartile

range)inwholebloodsamplestoredforupto14daysat4◦C.Thehorizontaldottedlinesdefinethereferencechangevalue inosmoticresistance.

remarkableeconomicconstraintsonhospitalandlaboratory budgets.10

Accordingtotheongoingreorganizationofmedical labo-ratoryservices,increasinglyintegratedwithinlargenetworks of facilities (i.e., the so-called “hub-and-spoke” paradigm), strengthenedeffortswillbenecessaryinthemonitoringof thequalityofdiagnosticbloodsamples.11 _{Reliableevidence}

suggeststhatthepreanalyticalphaseisthemostcriticalpart ofthetotaltestingprocess,sinceupto70%ofdiagnosticerrors

arerelatedtoinappropriateproceduresforsamplecollection, handling,transportation,andstorage.10

TheOFT,oneoftheoldesthematologylaboratorytestsused forassessingthephysicalstateoferythrocytes,remainsan essentialanalysisforscreeningredbloodcell(RBC)disorders. Briefly, the OFT measuresthe surface/volume ratio of ery-throcytesasitexistswhentheyaresuspendedinanisotonic medium,thusreflectingtheextentofmembraneredundancy whenRBCsareinequilibriumwithinanisotonicsaltsolution

Pleasecitethisarticleinpressas:SalvagnoGL,etal.Stabilityofrefrigeratedwholebloodsamplesforosmoticfragilitytest.HematolTransfus Fig.2–Spearman’scorrelationbetweenthevariationinhyperosmolarorhyposmolarresistanceandtheirsignificant predictors.

withorwithoutglycerol.12_{However,thebloodsample}

stor-ageconditionsmayhaveaprofoundinfluenceonerythrocyte shape,volumeandintegrity,duetothecontinuous anaero-bicmetabolismofthesebloodelements.Rapoportoriginally describedthatmembraneinjuriesmayoccurduring

erythro-cytestorage,whichmaythenleadtosphericaltransformation, lowermembraneredundancyandincreasedvulnerabilityata relativelyhighosmoticstrength.13_{Beutleretal.also}

demon-stratedthatosmoticfragilityvariationisnotonlyduetoloss ofmembraneintegrity,butalsotothepresenceofosmotically

Pleasecitethisarticleinpressas:SalvagnoGL,etal.Stabilityofrefrigeratedwholebloodsamplesforosmoticfragilitytest.HematolTransfus activesubstances(i.e.,lactate)withinthecell,whichwould

thenmaketheintracellularspacehyperosmolarwhenthecell isimmersedinisotonicsaline.12

Takentogether,ourfindingsdemonstratethatprolonged storageofwholebloodmayhaveasubstantialinfluenceon thediagnosticaccuracyoftheOFT,especiallyfor hyposmo-larresistance.Therefore,wholebloodspecimensshouldnot bestoredrefrigeratedat4◦Cfor>2daysbeforetesting.These resultsareinaccordancewiththosepublishednearly25years agobySaxenaet al.,14 _{whoshowedthatstorageat4}◦_Cfor

up to15 daysof heparinizedsamples ofwhole blood was associated with a modest increase in erythrocyte osmotic fragility(assessedusinghypotonicsalinesolutions),although noexperimentwereperformedforevaluatingthestabilityof hyposmolarresistance.

Inconclusion,theresultsofourstudydemonstratethat sample stability seems critical for the OFT, especially for hyposmolar resistance. This finding could help laboratory professionals to define clear policies for sample storage requirementsforOFTtestinginlaboratoryhandbooks.

Conflicts

of

interest

Theauthorsdeclarenoconflictsofinterest.

Authorship

contributions

GLS and GL conceived and designed the study, analyzed thedata,performedthestatisticalanalysisand draftedthe manuscript;DDandFD,performedtheexperiments;GLSand GLinterpretedthedataandwrotethemanuscript.Allauthors criticallyrevisedthemanuscriptandapprovedthefinal ver-sion.

r

e

f

e

r

e

n

c

e

s

[1]. CiepielaO.Oldandnewinsightsintothediagnosisof hereditaryspherocytosis.AnnTranslMed.2018;6:339.

[2].ZanellaA,IzzoC,RebullaP,ZanusoF,PerroniL,SirchiaG. Acidifiedglycerollysistest:ascreeningtestforspherocytosis. BrJHaematol.1980;45:481–6.

[3].FariasMG.Advancesinlaboratorydiagnosisofhereditary spherocytosis.ClinChemLabMed.2017;55:944–8.

[4].LippiG.Governanceofpreanalyticalvariability:travellingthe rightpathtothebrightsideofthemoon?ClinChimActa. 2009;404(1):32–6.

[5].LippiG,GuidiGC,MattiuzziC,PlebaniM.Preanalytical variability:thedarksideofthemooninlaboratorytesting. ClinChemLabMed.2006;44:358–65.

[6].InternationalStandardsOrganization.ISO15189:2012. Medicallaboratories—requirementsforqualityand

competence.Geneva,Switzerland:InternationalOrganization forStandardization;2012.

[7].WalskiT,Chludzi ´nskaL,KomorowskaM,WitkiewiczW. Individualosmoticfragilitydistribution:anewparameterfor determinationoftheosmoticpropertiesofhumanredblood cells.BiomedResInt.2014;2014:162102.

[8].ParaisoLF,Gonc¸alves-E-OliveiraAF,CunhaLM,deAlmeida NetoOP,PachecoAG,AraújoKB,etal.Effectsofacuteand chronicexerciseontheosmoticstabilityoferythrocyte membraneofcompetitiveswimmers.PLoSOne. 2017;12:e0171318.

[9].WorldHealthOrganization.Laboratoryqualitymanagement system:handbook,Version1.1.Geneva:WorldHealth Organization;2012.

[10].LippiG,vonMeyerA,CadamuroJ,SimundicAM.Blood samplequality.Diagnosis(Berl).Diagnosis(Berl). 2019;6(1):25–31.

[11].LippiG,SimundicAM.Laboratorynetworkingandsample quality:astillrelevantissueforpatientsafety.ClinChemLab Med.2012;50:1703–5.

[12].BeutlerE,KuhlW,WestC.Theosmoticfragilityof erythrocytesafterprolongedliquidstorageandafter reinfusion.Blood.1982;59:1141–7.

[13].RapoportS,WingM.Dimensional,osmotic,andchemical changesoferythrocytesinstoredblood.i.bloodpreservedin sodiumcitrate,neutral,andacidcitrate-glucose(acd) mixtures.JClinInvest.1947;26:591–615.

[14].SaxenaRK,SeshadriV.Measurementofosmoticresistanceof normalandpathologicalhumanredbloodcells.IndianJ PhysiolPharmacol.1983;27:1–6.