Int J Lab Hem. 2017;1–8. wileyonlinelibrary.com/journal/ijlh  | 1© 2017 John Wiley & Sons Ltd

Received:12December2016  |  Accepted:22May2017DOI: 10.1111/ijlh.12705

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

Reticulocyte count and extended reticulocyte parameters by Mindray BC- 6800: Reference intervals and comparison with Sysmex XE- 5000

M. Buttarello1  | A. Rauli2 | G. Mezzapelle2

1DepartmentofLaboratoryMedicine,AziendaOspedaliera-UniversitàdiPadova,Padova,Italy2ClinicalPathologyLaboratory,HospitalofAdria,Adria,Italy

CorrespondenceMauroButtarello,DepartmentofLaboratoryMedicine,AziendaOspedaliera-UniversitàdiPadova,Padova,Italy.Email:mbuttarello@yahoo.it

AbstractIntroduction: Inthisstudy,analyticperformance(imprecision,carryover,timestability)anddiagnosticefficiencyofMindrayBC-6800analyzertoquantifyreticulocytesandextended reticulocyte parameterswas evaluated.Moreover, reference intervals onadult populationweredetermined.Resultswere comparedwith thoseobtainedbySysmexXE-5000analyzer.Methods: Onehundredandeighty-fourhealthyadultsofbothsexes,and368subjectsaffectedbyvariouspathologicconditions(nutritionalanemiasbeforeandaftertreat-ment,hemolyticandposthemorragicanemias,acuteandchronicinflammations,malig-nancyundertherapy,andbetathalassemiatrait)wereselected.Results: Referenceintervalswereasfollows:reticulocytes(×109/L):23.2-93.2;imma-turereticulocytefraction:0.015-0.14;meanreticulocytehemoglobinequivalent(RHE)(pg):30.9-35.7;meanreticulocytevolume(fL):97.8-118.Imprecisiononreticulocytecountatallconcentrationswasclosetoanalyticgoalbasedonwithin-subjectbiologi-calvariation.Carryover(2.3%)wasnegligible,andtime-stabilitywasexcellentupto8 hours.Conclusion: WhencomparedwithXE-5000,BC-6800showsagoodoverallcorrela-tiononcountingdespiteevidenceofdifferenceintheupperlimitofreferenceinter-vals (93.2 vs 101.3). Comparison of diagnostic efficiency of extended parametersshowsagoodtotalagreementofRHE(91.6%).

K E Y W O R D S

immaturereticulocytefraction,MindrayBC-6800,referenceintervals,reticulocyte,reticulocytehemoglobincontent

1  | INTRODUCTION

Reticulocytecountisclinicallyimportantbothforpathophysiologicalclassificationofanemias,andforearlyidentificationofnormalizationoferythropoiesisbythemarrowaftertherapeutic intervention(iron,folate,cobalamin,etc.),orasawayofcheckingearlyregenerationaftermarrowor stemcell transplant.Manualmicroscopicmethod is rela-tivelyinaccuratebecauseofthesubjectivemorphologicaldefinitionofreticulocyteandisveryimprecisewithacoefficientofvariation(CV)

between16%athighleveland68%atlowconcentration.1 These lim-itationsmakethemethodunreliableaboveallatlowvalues,thatis,insituations inwhich onemust define reduced erythropoietic activityofthemarrow(valuesunderthelowerlimitofthereferenceinterval),orinwhichsmallbutsignificantvariationsthatappearintheearlyre-coveryofpostaplasiaorpostmarrow transplantmustbemonitored.Automated analyzers represent a revolution for reticulocyte countusingdyestobindreticulocyteRNAandflowcytometerstoperformrapidcounts.Thesemethodsallowforobjectivecounts,andbecause

2  |     BUTTARELLO ET AL.

theyanalyzetensofthousandsofcells,theyreducesamplingerrorandarethereforemoreprecisemainlyatlowvalues(CVbetween3%and25%).2Afurtheradvantageistheidentificationofotherreticulocyteparameters,suchastheimmaturereticulocytefraction(IRF)orthere-ticulocyteindicesthatis,tosay,themeanhemoglobincontent(RHE),andthemeanreticulocytevolume(MRV),whichareusefulinseveralclinical conditions.3-9 Problems still exist that essentially dependonthedifferingsensitivityofdyesusedtostainRNA,onthetechnologyused to identify positive cells (fluorescence, light scattering, absor-bance),andonsoftwaresthataremoreorlesscapableofseparatingreticulocytesfrommatureerythrocytesasthereisaphysiologiccon-tinuumbetween thesepopulations.Thesedifferences contribute totheincompleteagreementofcountswiththeconsequentnecessityofmethod-specificreferenceintervals.2,4

The aim of our studywas to evaluate the analytic performance(imprecision, carry-over, time stability) of the hematology analyzerMindrayBC-6800,andthereference intervalscalculationonreticu-locytecountandreticulocyteparameters.Toevaluatepossibleinter-changeability, the values obtained have been comparedwith thoseobtained by Sysmex XE-5000, our standard equipment, for whichthereisalargeamountofliteratureonclinicalusefulnessofextendedreticulocyteparameters.10-13

2  | MATERIALS AND METHODS

2.1 | Hematology analyzers

2.1.1 | Mindray BC- 6800

MindrayBC-6800(Mindray,Shenzhen,China)isanautomatedhema-tology analyzer able to performa complete blood count (CBC) andleukocytedifferentiation.Itcanalsocountreticulocytesusingadedi-catedchannelwhereafluorescentdye(asymmetriccyanine)canbindtocytoplasmicRNAtoallowforreticulocytesseparationfrommatureredcells.AsfluorescentsignalsareproportionatetotheRNAcontent,theanalyzersubdividesreticulocyteintothreefractionsaccordingtotheirmaturitylevel.IRFisdefinedasthesumofpopulationsathighandmedium immaturity. It alsoprovides,with a forward light scat-termeasure,somereticulocyte indexessimilartothoseforerythro-cytes suchas themean “reticulocytehemoglobinequivalent” (RHE)expressedinpicograms,andtheMRV.

2.2 | Sysmex XE- 5000

TheSysmexXE-5000(Sysmex,Kobe,Japan)isanautomatedanalyzerwhich,inadedicatedchannel,usingafluorescentdye(polymethine),isabletocountreticulocytes.Thecounter,usingaforwardlightscat-termeasureandsidefluorescenceintensity,iscapableofdifferenti-atingreticulocytes fromother redcells, tocalculate the IRF,andtoobtaintheRHE(Ret-He).

Bothanalyzerswerecalibratedaccordingtothemanufacturer’sin-structionsandcheckedtwiceadayusingcommercialtri-levelcontrolprovidedbythecompanies.

2.3 | Samples

Healthy subjects (N=184) were selected in accordance with theH44-A2 specifications to calculate reference intervals.14 Subjectsweremen(N=81)andwomen(N=103)aged19-71(mean44).Theywereexcluded if any laboratory resultsofCBC,markers for liverandrenalfunctions,andironstatus,wereoutsidethereferencein-tervalsusedinour laboratory.Othercriteriaofexclusionwereasfollows: pregnancy, clinical evidence of recentmedical disorders,andblooddonationduringtheprevious6months.Inaddition,sam-ples from368 subjects affectedby various pathologic conditions(nutritionalanemiasbeforeandaftertreatment,hemolyticandpos-themorragicanemias,acuteandchronicinflammations,malignancyunderchemotherapy,andbetathalassemiacarriers)wereanalyzedinfourdifferentdays.Forthepossibleeffectduetothepresenceof red cell inclusions, three sampleswithbasophilic stippling andfivewith Jollybodieswere identified.All sampleswere collectedin themorning, under fasting conditions with the consent of in-formeddonorsbyvenipunctureinto3mLK2EDTAevacuatedtubes(Vacutest,Kima,Italy).Forsomepatients,samplesweretakenfromleftoverbloodcollectedforCBC.Theanalysisonbothinstrumentswasperformed inautomatedmodewithin45minutesofdrawingsamples.

The study is in accordwithethical standardsestablishedby theinstitutioninwhichtheexperimentwasperformed.

2.4 | Imprecision and carryover

Imprecisiononreticulocytecountwasestimatedwiththeimpreci-sionprofile2,15extendedtoawiderangeofvalues(5to230×109/L)by performing repeated analysis (8 counts) on 17 samples rein-serted in random order. Calculation of carryover was performedbyanalyzingapairof samples:A samplewithahigh reticulocyteconcentration(156.7×109/L)analyzedthreetime(i1,i2,i3)immedi-atelyfollowedby1sampleatalowconcentration(35.5×109/L)alsoanalyzed3 times consecutively (j1, j2, j3) using the formula(j1−j3)/(i3−j3)×100.

16

2.5 | Method comparison

Results of the comparison with Sysmex XE-5000 on reticulocytecount,IRF,andRHEweresummarizedgraphicallywithBland-Altmanplot with 95% confidence intervals which represent the limits ofagreement.16ForRHE,theevaluationwasbasedalsoontheassess-mentofclinicalagreementonselectedgroupsofpatients.

2.6 | Time stability

Forthestudyofpreservationtimeofbloodsamples,wecomparedthevalueofparametersofinterestbyretestingsamplescomingfromthesamegroupof368subjects (between45minutesand8hoursfromsampling:<45minutes,+2,+4,+8hoursmaintainingsamplesatroomtemperature).

     |  3BUTTARELLO ET AL.

2.7 | Statistical analysis

Reference intervals were calculated with a nonparametric method(middle95thpercentileofthedistribution).Toevaluatepopulationdis-tribution characteristics, a Shapiro-Wilk testwas used. Formethodcomparison,Bland-Altmanmodelwasperformed.Behaviorofparam-etersofinterestasafunctionoftimewerecomparedusinganalysisofvariance(ANOVA).Allsignificancetestingweretwo-tailed,andaP value<.05wasconsideredsignificant.

3  | RESULTS

3.1 | Reference intervals

Reference intervals (mean and range) are summarized in Table1.Distribution of absolute reticulocyte count and IRF are log-normal,whileforRHEandMRVitisapproximatelyGaussian.WhencomparedtoXE-5000, the interval for reticulocyte count isdifferent inmeanand upper threshold (−5.4 and −8×109/L, respectively), but lowerthresholdsarecoincident.ForIRF,thevaluesareessentiallysuperim-posable,whileforRHE,thereisapositivedifferenceinlowerthresh-old (+1.5pg),butthemeanandupperthresholdaresimilar.MRVisnotcomparableasitisnotavailableonXE-5000.

3.2 | Imprecision

The imprecisionprofileofBC-6800 in reticulocyte count is charac-teristically shapedwith a CV% rapidly increasing while the param-eterconcentrationislowering.Forvaluesclosetothelowerlimitofthereference interval,CVisnear7%,whilefortheupper limit, it is4.5% (Figure1).These resultsareclose to theanalyticalgoalbasedon within-subject biological variation.17-19 The carryover (2.3%) isnegligible.

3.3 | Method comparison

Table2showsthecomparisonbetweenXE-5000andBC-6800withdatarelatedtoBland-Altmanstatistics(meandifferencewithconfi-denceintervals).Figure2A,Bshowsthegraphicaldisplayforreticu-locytecountandIRF.Thecomparisonoftheseparameterswasalsoevaluatedwiththebiparametricmatrix:reticulocytecountvsIRF3,4

asshowninFigure3.Toverifytheagreementbetweenmethods,itisusefultoseewhereBC-6800placedpatientspreviouslydefinedas“aplasticorwithearlymarrowrecovery”byXE-5000.ThisisshownbyblackdotsinFigure3.Fromthepositionofthesedots,itisevi-dentthatallpatientswereplacedinthesameareabybothanalyzers.Thisindicatesthat,inspiteofnumericaldifferencesinreferencein-tervals,inclinicaluse,resultsobtainedwiththematrixbythetwoan-alyzersaresuperimposable.Thesampleswithredcellinclusionsdidnotdemonstratesystematicdifferencesonthecountsofbothana-lyzers(reticulocytesfrom37to55×109/LatBC-6800andfrom35to60×109/LatXE-5000.InFigure2A,therearefoursampleswithmarkedly higher results with BC-6800 than with XE-5000. Thesebelongedtotwopatientswithbetathalassemiamajor,1withbetathalassemic trait,and1withchronichemolyticanemia. In the first

Mindray BC- 6800 Sysmex XE- 5000

Mean 2.5% 97.5% Mean 2.5% 97.5%

Reticulocyte(109/L) 44.99 23.18 93.2 49.44 23.08 101.3

Immaturereticulocyte(fraction)

0.053 0.015 0.14 0.059 0.012 0.15

Reticulocytehemo-globinequivalent(pg)

32.79 30.90 35.70 32.30 29.40 35.98

Meanreticulocytevolume(fL)

108.87 97.79 117.98 N.A. N.A. N.A.

TABLE  1 Referenceintervalsonreticulocyteandreticulocyteparameters(N=184)

F IGURE  1 BC-6800:imprecisionprofileofreticulocyteabsolutecount

02468

101214161820

0 50 100 150 200 250

CV %

Reticulocytes (109/L)

TABLE  2 Comparisonofmethods:analysisofdatabyBland-Altmanstatistics(N=552)

Parameters Mean difference

95% Confidence intervals P

Reticulocyte(109/L) −1.8 −2.64,−0.95 <.001

Immaturereticulocyte(fraction)

−0.006 −0.008,−0.005 <.001

Reticulocytehemoglobinequivalent(pg)

0.99 0.87,1.10 <.001

4  |     BUTTARELLO ET AL.

F IGURE  2 Bland-Altmanplotscomparingthereticulocytecount(A),theimmaturereticulocytefraction(B),andthereticulocytehemoglobinequivalent(C)ofBC-6800andXE-5000.Thedottedlinesrepresent95%limitsofagreement

–30

–20

–10

0

10

20

30

40

50

60

70

0 50 100 150 200 250

Diffe

renc

e (B

C680

0 - X

E500

0)

Mean of RET (109/L)

–0.1

–0.08

–0.06

–0.04

–0.02

0

0.02

0.04

0.06

0.08

0.1

0 0.05 0.1 0.15 0.2 0.25 0.3

Diffe

renc

e (B

C680

0 - X

E500

0)

Mean of IRF (fraction)

–4

–2

0

2

4

6

8

10

15 20 25 30 35 40 45

Diffe

renc

e ( B

C680

0 - X

E500

0)

Mean of RHE (pg)

(A)

(B)

(C)

     |  5BUTTARELLO ET AL.

twosamples,therewasasimilarpercentageoferythroblastsamongthetwoanalyzers(23.5and18.7vs24.7and20.3)whileintheothertwotheerythroblastswereabsent.AlsotheIRFvaluesweresimilaramong the twomethods (from 0.08 to 0.13 vs 0.05 to 0.13) andnootherpossible interferencewashighlighted inthebloodsmear.In 2 of these samples (beta thalassemiamajor andhemolytic ane-mia),BC-6800 reported the flag “cellulardebris.” Figure2C showstheplotofRHEwithapositivemeandifference(0.99).Thiscomplieswith a tendency to produce higher resultsmainly at low value ofthe reference interval.Figure4 shows theclinicalusefulness (abil-itytodistinguishbetweennormalandabnormalconditions)ofRHEwhencomparedtoRet-HeofXE-5000,keepinginmindthatclinicaluseofresultsisparticularlyusefulatsubnormalvalues(diagnosisofirondeficiencyerythropoiesis).10-13Table3showsthedisagreementinsamplesclassification(31fromatotalof368ones).Thesediffer-encesareidentifiedbyblackdotsinFigure4.

F IGURE  3 TwoparametersmatrixcomparingtherelationshipofimmaturereticulocytefractionvsabsolutereticulocytecountofBC-6800andXE-5000.TheblackdotsshowthesamplesthatXE-5000placedinthe“aplastic”category.Theverticalandhorizontallinesrepresentthereferenceintervals

0

0.05

0.1

0.15

0.2

0.25

0.3

0 20 40 60 80 100 120 140 160 180 200

IRF

(frac

tion)

XE-5000 Reticulocytes (109/L)

BC-6800 Reticulocytes (109/L)

0

0.05

0.1

0.15

0.2

0.25

0.3

0 20 40 60 80 100 120 140 160 180 200

IRF

(frac

tion)

F IGURE  4 Plotofreticulocytehemoglobinequivalent:BC-6800vsXE-5000.Theverticalandhorizontallinesrepresentthereferenceintervals.Theblackdotsindicatethedisagreementinsamplesclassification

15

20

25

30

35

40

45

15 20 25 30 35 40

RHE

(pg)

-BC6

800

RHE (pg) - XE5000

6  |     BUTTARELLO ET AL.

3.4 | Time stability after venipuncture

Regarding the time laps inmeasuring samples stored at room tem-perature,Table4showsresultsobtained.Theevaluationwasstoppedafter8hoursfromvenipunctureasitisareasonablelimitthatshouldnotbeexceeded.Infact,fora longertime,notalltheotherparam-eters of theCBC are stable regardless of storage temperature.20-22 Forreticulocytecount, therearenosignificantdifferencesbetween45minutesand8hours.Forotherparameters(IRF,RHE,MRV),dif-ferencesarestatisticallysignificantalreadyafter4hours,butalwaysclose to the “optimal” bias based on biological variability (0.125[CVi2+CVg2]1/2),whereCViandCVgrepresentthewithinsubjectandbetweensubjectvariability)17,19 .

4  | DISCUSSION

Thelargediffusionofautomatedmethodsdependsonthepossibilityofreplacingmanualvisualcounts,whicharesubjective,highlyimpre-ciseandtedious,andontheadvantagesderivingfromapreciseandobjectivecountmainlyat lowvalues,necessaryforthediagnosisofhypoplasticanemiasorfortheearlymonitoringofbonemarrowre-generation.Afurtheradvantageistheidentificationofotherreticu-locyteparameters,suchastheIRFwhichisusefulinclassificationofanemias based onmarrow response, mainly using two-dimensionalmatricesofIRFvsabsolutereticulocytecount,andindifferentialdiag-nosisofreticulocytopeniaindistinguishingbonemarrowaplasiafromearlyerythropoieticresponse.OtherusesincludemonitoringtherapyefficacyinnutritionalanemiabecausetheincreaseinIRFprecedestheincreaseintotalreticulocytecountbyseveraldays.3,4,23OtherusefulparametersavailableonlywithautomatedanalyzersarereticulocyteindexessuchasRHEandMRV.Theseindexesareimportantbecauselowvaluesindicateiron-deficienterythropoiesis,eveninconditionsinwhichtraditionalbiochemicalmarkerssuchasferritinandtransferringsaturationareinadequateasininflammationsoranemiafromachronicdisease.Moreover, they are useful inmonitoring an early responsetotherapybecausetheychangesignificantlyafteronly48-72hoursfrom administration of iron, cobalamin or folate in nutritional ane-mias.9,24Increaseofdifferentanalyzersonthemarket,amongwhichMindrayBC-6800representsoneofthelastintroduced,bringsinthepressingneed for resultscomparisonandharmonization.The refer-enceintervalinabsolutevaluebetweenthetwoanalyzersisdifferent,

TABLE  3  Instrumentagreementofreticulocytehemoglobinequivalent:distributionalclassification(N=368)

Mindray BC- 6800

Sysmex XE- 5000 Low Normal High Total

Low 105 4 0 109

Normal 19 210 2 231

High 0 6 22 28

Total 124 220 24 368

TABLE 4 Timestabilityevaluation(N=368)

Para

met

ers

T0T2

T4T8

Opt

imal

bia

sM

ean

Mea

nM

ean

diff

eren

ceP

Mea

nM

ean

diff

eren

ceP

Mea

nM

ean

diff

eren

ceP

Reticulocytes(109 /L)

50.8

850

.66

0.22

.413

50.33

0.55

.052

50.42

0.46

.124

3.9%(1.98)*

Immaturereticulocytefraction

(fraction)

0.07

00.

068

0.00

2<.051

0.06

70.003

<.001

0.06

70.003

<.001

2.4%(0.002)*

Reticulocytehemoglobin

equivalent(pg)

31.97

31.90

0.07

.081

31.81

0.16

<.001

31.80

0.18

<.001

0.49%(0.15)*

Meanreticulocytevolume(fL)

106.

2210

5.99

0.23

.078

105.

720.

50.0

0210

5.65

0.57

<.001

0.41%(0.43)*

*Parenthesis:absolutevalues.

     |  7BUTTARELLO ET AL.

mainly on the upper limit, and the disagreement needs for specificreferenceintervals.Resultsdisagreealsowithpreviouspublishedre-sultsobtainedwithMindrayBC-6800onadultpopulation,especiallyontheupperlimit:93.2vs115×109/L.25Thismaydependonthese-lectedpopulation, but thedifference is so large to requires furtherstudy before reaching a final conclusion.Also comparison betweenmethodsonallanalyzedsamplesshowsanegativemeandifference(−1.8×109/L). Inthisstudy,thenumberofsampleswitherythrocyteinclusionswasmodest(threewithbasophilicstippling,fivewithJollybodiesandnoonewithinclusionsfrommalarialparasite),andthere-fore,amoreextensiveevaluationwouldbenecessarytoassesstheeffectofallpossibleinterferences.Forthefoursampleswithmarkeddifferencesincounting,nointerferencefromerythroblastsorahigherproportion of young reticulocytes can be hypothesized consideringthe similar values between the two analyzers. However, these dif-ferencesmaydependon thedifferentways inwhich the software,incertainsamples,determinestheseparationbetweenreticulocytesandmatureerythrocytes.Itshouldbenoted,however,that,fordiag-nosticpurposes,bothanalyzersshowedan increase inreticulocytesalthoughmorepronouncedforBC-6800thanforXE-5000.

Theanalyticgoalforimprecisionwhichiscurrentlywidelysharedis based on biological variability.Among various clinical conditions,monitoring has the most restrictive criteria (CVa<0.5 CVi). Studiespublishedonbiologicalvariabilityofreticulocytecountprovideessen-tiallyconcordantresults:CVibetween9.5and11%,26-28withanalyt-icalgoalsfrom4.75%to5.5%,respectively.Withimprecisionprofilewecanidentifyimprecisioninconnectionwithcriticalconcentrations.Thismethodwhenextendedtoawide rangeofconcentrationsandwithasufficientnumberofsamples(tolimittheeffectofpossibleout-liers)representsavalidalternativetoANOVAperformedonduplicatesofsamplesofdifferentconcentration(ie,low,normal,andhigh),16orofmanyconsecutivemeasurementsonthesamesample.Figure1showsthatthegoalforBC-6800isreachedatnormalandhighconcentra-tions.At lowconcentration, imprecision(from5%to10%) isgreaterthanthegoal,but it isnegligiblebyaclinicalpointofview,andit isbetterthanthatofmanypreviousgenerationofanalyzers.2Stabilityofallparametersovertimeafter8hoursfromsamplingissatisfactorywithvariations lowerorapproximate to themore restrictivecriteria(so-calledoptimalbias)basedonbiologicalvariation.17OnevaluationofclinicalusefulnessofRHE,totalagreementwithXE-5000isgood(91.6%).Atlowvalues,wheretheresultsareparticularlyusefulindi-agnosisofirondeficiencyerythropoiesis,arepresent23discrepancies(4 lowwithXE-5000butnormalwithBC-6800and19normalwithXE-5000,butlowwithBC-6800).Inthiscase,allvalueswereincludedinalittle intervalbetween29.5and30.9pg.Lessinterestingfromadiagnosticpointofview(increasedinmacrocyticnutritionalanemia),arethehighvaluesofRHE,buttheirrapiddecline,likewisetothatofMRV,isusefulinmonitoringearlyresponsetotherapyasreportedinFigure5.Thisfigureshowsalsotheoverlapbetweeniron-deficiencyandbetathalassemiatraitbecauseinthislastconditionRHEisalwaysreducedindependentlyfromironstores.13,29Themainlimitationofthisstudyisthatmethodsformeasuringreticulocyteextendedparametersarenotstandardized,andreferencematerialforthesedeterminations

islacking.Thismakesitimpossibletoexplorespecificcausesfortheobserveddiscrepanciesbetweenthetwoanalyzers.

Inconclusion,BC-6800providesaprecisereticulocytecountatallconcentrationswithanegligiblecarry-over.Stabilityofallparametersisexcellentup to8hours.ClinicalagreementwithXE-5000 isgoodmainlyusingbivariatematricesonclassificationofanemia,andRHEin diagnosing iron-deficiency conditions. The incomplete numericalagreementoncountand insomeparameterssuchasRHEmakes itnecessarytheuseofspecificreferenceintervalsandclinicaldecisionvalues.

CONFLICT OF INTERESTS

Theauthorsdeclarethattheyhavenocompetinginterests.

REFERENCES

1. ButtarelloM,BulianP,VenudoA,RizzottiP.LaboratoryevaluationoftheMilesH.3automatedreticulocytecounter.AcomparativestudywithmanualreferencemethodandSysmexR-1000.Arch Pathol Lab Med.1995;119:1141-1148.

2. ButtarelloM,BulianP, FarinaG, et al. Flowcytometry reticulocytecounting. Parallel evaluation of five fully automated analyzers: anNCCLS-ICSHapproach.Am J Clin Pathol.2001;115:100-111.

3. Davis BH. Immature reticulocyte fraction (IRF): by any name, auseful clinical parameter of erythropoietic activity. Lab Hematol. 1996;2:2-8.

4. ButtarelloM, Bulian P, FarinaG, PetrisMG,TemporinV,Toffolo L.Fivefullyautomatedmethodsforperformingimmaturereticulocytefraction.Comparisonindiagnosisofbonemarrowaplasia.Am J Clin Pathol.2002;117:871-879.

5. BrugnaraC,LauferMR,FriedmanAR,BridgesK,PlattO.Reticulocytehemoglobincontent (CHr):early indicatorof irondeficiencyandre-sponsetotherapy.Blood.1994;83:3100-3101.

F IGURE  5 Plotofreticulocytehemoglobinequivalentvsimmaturereticulocytefraction.Theinsetsrepresentsomewell-differentiatebehaviorincertainareasoftheplot:(A):betathalassemiatraitwithorwithoutirondeficiency;(B):irondeficiencyatdiagnosis;(C):macrocyticnutritionalanemiaatdiagnosis.Superimposedarethetrendsforthreepatientsaftertherapy:•)irondeficiency;▲)betathalassemiatraitwithirondeficiency;*)cobalamindeficiency

8  |     BUTTARELLO ET AL.

6. d’OnofrioG,ChirilloR,ZiniG,CaenaroG,TommasiM,MicciulliG.Simultaneous measurement of reticulocyte and red cell indices inhealthysubjectsandpatientswithmicrocyticandmacrocyticanemia.Blood.1995;85:818-823.

7. BrugnaraC.Reticulocytecellularindices:anewapproachinthediag-nosisofanemiasandmonitoringoferythropoieticfunction.Crit Rev Clin Lab Sci.2000;37:93-130.

8. PivaE,BrugnaraC,ChiandettiL,PlebaniM.Automatedreticulocytecounting:stateoftheartandclinicalapplicationsintheevaluationoferythropoiesis.Clin Chem Lab Med.2010;48:1369-1380.

9. ButtarelloM.Laboratorydiagnosisofanemia:aretheoldandnewredcellparametersusefulinclassificationandtreatment,how?Int J Lab Hematol.2016;38(suppl1):123-132.

10. Garzia M, Di Mario A, Ferraro E, et al. Reticulocyte haemoglobinequivalent; an indicator of reduced iron availability in chronic kid-ney diseases during erythropoietin therapy. Lab Hematol. 2007;13: 6-11.

11. UrrechagaE,BorqueL,EscaneroJF.PotentialutilityofthenewSysmexXE5000redbloodcellextendedparametersinthestudyofdisordersofironmetabolism.Clin Chem Lab Med.2009;47:1411-1416.

12. UrrechagaE,BorqueL,EscaneroJF.Erythrocyteandreticulocytein-dicesintheassessmentoferythropoiesisactivityandironavailability.Int J Lab Hematol.2013;35:144-149.

13. ButtarelloM,PajolaR,NovelloE,MezzapelleG,PlebaniM.EvaluationofthehypochromicerythrocyteandreticulocytehemoglobincontentprovidedbytheSysmexXE-5000analyzerindiagnosisofirondefi-ciencyerythropoiesis.Clin Chem Lab Med.2016;54:1939-1945.

14. Clinical and Laboratory Standard Institute (CLSI). Method for Reticulocyte Counting. Approved Guideline H44-A2. (2nded.).Wayne,P.A.:ClinicalandLaboratoryStandardInstitute;2004.

15. SadlerWA,SmithMH.Useandabuseofprecisionprofile:somepit-falls illustratedbycomputingandplottingconfidence intervals.Clin Chem.1990;36:1346-1350.

16. BriggsC,CulpN,DavisB,etal. ICSHguidelines for theevaluationofbloodcellanalysersincludingthoseusedfordifferentialleucocyteandreticulocytecounting.Int J Lab Hematol.2014;36:613-627.

17. FraserCG,HyltoftPetersenP,LibeerJC,RicosC.Proposalforsettinggenerally applicable quality goals solely based on biology.Ann Clin Biochem.1997;34:8-12.

18. FraserCG,HyltoftPetersenP.Analyticalperformancecharacteristicshouldbejudgedagainstobjectivequalityspecifications.Clin Chem. 1999;45:321-323.

19. Sandberg S, Fraser CG, HorvathAR, et al. Defining analytical per-formancespecifications:consensusstatementfromthe1stStrategic

Conference of the European Federation of Clinical Chemistry andLaboratoryMedicine.Clin Chem Lab Med.2015;53:833-835.

20. ZiniG. Stability of complete blood count parameterswith storage:toward defined specifications for different diagnostic applications.(Editorial).Int J Lab Hematol.2014;36:111-113.

21. Buoro S, Mecca T, Seghezzi M, et al. Assessment of blood sam-ple stability for complete blood cell count using the Sysmex XN-9000 andMindray BC-6800 analyzers. Rev Bras Hematol Hemoter. 2016;38:225-239.

22. DavesM,Zagler EM,CeminR, et al. Sample stability for completebloodcellcountusingtheSysmexXNhaematologicalanalyser.Blood Transfus.2015;13:576-582.

23. ButtarelloM,PlebaniM.Automatedbloodcellcounts.Stateof theart.Am J Clin Pathol.2008;130:104-116.

24. ButtarelloM,TemporinV,CeravoloR, FarinaG,BulianP.The newreticulocyteparameter(Ret-y)oftheSysmexXE2100.Itsuseinthediagnosisandmonitoringofposttreatmentsideropenicanemia.Am J Clin Pathol.2004;121:489-495.

25. GrilloneR,GrimaldiE,ScopacasaF,ConticelliM,DenteB.Evaluationof the reticulocyte counting by the Mindray BC 6800 automatedhematology analyzer: comparison with ABX Pentra 120, CoulterLH 750, and microscopy. (Letter to the editor). Int J Lab Hematol. 2015;37:e3-e6.

26. Ricos C, Alvarez V, Cava F, et al. Current databases on biolog-ical variation: pros, cons, and progress. Scand J Clin Lab Invest. 1999;59:491-500.

27. SandbergS,RustadP,JohannesenB,StolsnesB.Within-subjectbio-logicalvariationofreticulocytesandreticulocyte-derivedparameters.Eur J Haematol.1998;61:42-48.

28. ZhangP,TangH,ChenK,ChenY,XuD.Biologicalvariationsofhe-matologic parameters determined by UniCel DxH 800 hematologyanalyzer.Arch Pathol Lab Med.2013;137:1106-1110.

29. Noronha JFA, Grotto HZW. Measurement of reticulocyte and redblood cell indices in patients with iron deficiency anemia and β-thalassemia.Clin Chem Lab Med.2005;43:195-197.

How to cite this article:ButtarelloM,RauliA,MezzapelleG.ReticulocytecountandextendedreticulocyteparametersbyMindrayBC-6800:ReferenceintervalsandcomparisonwithSysmexXE-5000.Int J Lab Hem.2017;00:1–8.https://doi.org/10.1111/ijlh.12705