Evaluating the hip-flask defence using analytical data from ethanol and ethyl glucuronide. A comparison of two models

Evaluating

the

hip-

flask

defence

using

analytical

data

from

ethanol

and

ethyl

glucuronide.

A

comparison

of

two

models

G.

Höiseth

a,b,c,

*

,

G.H.

Nilsson

d

,

R.

Lundberg

d

,

M.

Forsman

d

,

C.

Kronstrand

e

,

I.

Nyström

d

,

C.

Oscarsson

d

,

E.

Ericsson

d

,

M.D.

Cherma

d

,

J.

Ahlner

d,f

,

F.C.

Kugelberg

d,f

,

R.

Kronstrand

d,f a_{OsloUniversityHospital,DepartmentofForensicSciences,Oslo,Norway}

b

CenterforPsychopharmacology,DiakonhjemmetHospital,Oslo,Norway

c

UniversityofOslo,InstituteofClinicalMedicine,Oslo,Norway

d

NationalBoardofForensicMedicine,DepartmentofForensicGeneticsandForensicToxicology,Linköping,Sweden

e

LinköpingUniversity,FacultyofHealthSciences,Linköping,Sweden

f

LinköpingUniversity,DepartmentofBiomedicalandClinicalSciences,DivisionofDrugResearch,Linköping,Sweden

ARTICLE INFO

Articlehistory: Received11May2020

Receivedinrevisedform5July2020 Accepted7July2020

Availableonline26August2020

Keywords: Ethanol Ethylglucuronide Hip-flaskdefence

ABSTRACT

Aim:Claimedintakeofalcoholafteratrafficincident,calledthehip-flaskdefence,canbeobjectively assessedbydifferentmethods.Oneofthemistheuseoftwoconsecutiveethanolconcentrationsinurine andtheratiobetweenethanolconcentrationsinurineandblood.Anotheroneistheconcentrationsof ethylglucuronide(EtG)andethylsulphate(EtS)inbloodandtheirratiotoethanol.Theexperimental basisforboththesemodelsisfromsingledosestudiesonly.Theaimofthisstudywasthereforeto describethekineticsofethanol,EtGandEtSafteringestionoftworepeateddosesofethanolandto investigatetheusefulnessofthedifferentmodelsfortheassessmentofthehip-flaskdefence. Methods:Thirty-fivesubjectsingestedafirstdoseof0.51gofethanolperkilobodyweight,andtwohours lateraseconddose(thehip-flaskdrink)of0.25,0.51or0.85gofethanolperkilobodyweight.Tenurine and17bloodsampleswerecollectedandanalysedforethanol,EtGandEtSusingfullyvalidatedmethods. Itwasinvestigatedifallsubjectsfulfilledthecriteriaforrecentdrinking,accordingtothetwodifferent models,whenusingthesamplescollected180–240minutesafterstartoffirstdosedrinking.Accordingto thefirstmodel,increaseinurinaryethanolconcentrationsandaratioUAC/BACbelow1.3indicated recentdrinking.Accordingtothesecondmodel,increaseinbloodEtGconcentrationsandaratioethanol (g/kg)/EtG(mg/L)above1indicatedrecentdrinking.

Results:Allsubjectsinthehighdosegroupfulfilledallcriteriaforrecentdrinking.Onesubjectinthe mediumdosegroupandninesubjectsinthelowdosegroupfailedtoshowincreasingUACand/oraUAC/ BACratiobelow1.3.Onesubjectinthelowdosegroupfailedtoshowincreasingconcentrationsofblood EtG,butallsubjectsshowedaratioethanol/EtGabove1.

Conclusions:Thepresentstudyshowed,bytheuseofexperimentaldata,thatbothtwomodelsusedto investigatethehip-flaskdefencecanbeused,butonlywhenthehip-flaskdoseissufficientlyhigh. ©2020TheAuthor(s).PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense

(http://creativecommons.org/licenses/by/4.0/).

1.Introduction

Althoughuseof non-alcoholdrugs increases,ethanol is still widely detected in impaired drivers, and this is also the drug associatedwiththehighestincreaseinriskoftrafficaccident,both fatal and non-fatal[1–6]. Afterfor instance a car accident, the suspectsometimesclaimsthatheorshehadnotdrunkanyalcohol orjustasmalleramountbeforedriving,butingestedalcoholafter

the end of driving, before apprehension by the police. This is named the hip-flask defence [7–10]. Forensic toxicologists are oftenconvenedasexpertwitnesses insuchcases,toassessthe credibilityoftheexplanation.Anobjectiveestimationofthetime ofalcoholintakeisthenimportant.

Bloodorurinesamplesaretypicallycollected1 2hoursafter end of driving [11–13]. Different methods like analyses of congenersandmeasurementofethanolintwoconsecutiveblood samples have been proposed for cases involving the hip-flask defence[7,8].Otherpossiblemethodsincludetheuseofethanol ratios between blood and urine and between two consecutive urinesamples[7]andtheanalysesofthenon-oxidativeethanol metabolites ethylglucuronide (EtG)and ethylsulphate (EtS)in

* Correspondingauthorat:OsloUniversityHospital,Department ofForensic Sciences,POBox4950,Nydalen,N-0424,Oslo,Norway.

E-mailaddress:gudrho@ous-hf.no(G. Höiseth).

http://dx.doi.org/10.1016/j.forsciint.2020.110409

0379-0738/©2020TheAuthor(s).PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).

ContentslistsavailableatScienceDirect

Forensic

Science

International

blood [12], and thesetwo methods willbe investigatedin the present article.A urinary alcohol concentration(UAC) to blood alcoholconcentration(BAC)ratiolowerthan1.3 andincreasing UACintwoconsecutiveurinesamplesisconsideredindicativefor recent drinking. Regarding blood EtG and EtS, low absolute concentrations [12], a high ratio between ethanol (g/kg) and EtG/EtS(mg/L)andincreasingEtGorEtSvaluesintwoconsecutive samples indicate that intake occurred shortly before sample collection.

These scientific interpretations are based on experimental pharmacokinetic studies of healthy volunteers, administering mostly low to moderate doses of ethanol [14_–18]. An obvious weaknesswhentransferringthistothecasesofhip-flaskdefenceis the frequent use of higher doses in real life cases, making extrapolationnecessary.Also,experimentalstudiesareperformed withsingledoseintake,asituationnotnecessarilycomparableto theuserpatterninreallifecaseswherealcoholmighthavebeen drunk also before driving. Data from studies using repeated ingestion,aswell as higherdoses, couldthereforeimprovethe accuracyintheinterpretationofhip-flaskdefencecases.

Theaimofthisstudywasthereforetodescribethekineticsof ethanol,EtG and EtS in blood and urine after ingestionof two repeateddosesofatotalrelativelylargedoseofethanol.Wealso wantedtoinvestigatetheusefulnessofthedifferentmodelsforthe assessmentofthehip-flaskdefence.

2.Materialsandmethods 2.1.Studyprotocol

Thirty-sevenhealthyvolunteersreceived0.51gethanolasbeer (5%)perkilobodyweightatstartoftheexperiment,overatotalof 1h(dividedintofourportionsadministeredduring15mineach). Twohoursafter start of theexperiment, an additional doseof ethanolwasingestedover15forthelowdoseand30minforthe mediumandhighdose(0.25g/kg,0.51g/kgor0.85g/kgingestedas beer,wineorspirits).Breakfastwasprovidedduringthefirst1h drinkingsession.About30minaftertheseconddrink(threehours afterstartof theexperiment),thesubjectswereprovided with lunch. During the entire experimental day the subjects were providedwithwater,coffee,tea,fruitandbiscuitsadlibitum.The timelineoftheexperimentisshowninFig.1.

Oversevenhours,atotalof10urinesampleswerecollectedin tubes (NUNC TM,Roskilde, Denmark), and 17 blood samples(2 tubesateachsamplingtimeforanalysisofethanolandEtG/EtS) weredrawnin5mLvacutainertubes(BDVacutainer1,Plymouth, United Kingdom), with 143 IU (0.286mg) heparin and 20mg fluoride.ThesubjectdemographicsareshowninTable1.Thestudy protocol was approved by the regional ethics committee in Linköping(Dnr:2015:41/31and2017:213/32).

2.2.Analysesofethanolinbloodandurine

Allbloodandurinesamplesfromthestudywereanalysedfor ethanol. Blood samples were analysed for ethanol the day of collectionand urinesamplesanalysedattheend ofeach study week (within 5 days). The method used has previously been validatedforethanolquantificationinbloodorurinesamplesfrom legal cases[19,20]. In brief,a 100

m

Laliquot of studysamples, calibration samples and control samples were diluted with 1000

m

Linternalsolutioninhead-spacevials,sealedwithcrimp capsandplacedonanautosamplerforethanolanalysisby head-space gas chromatography and _flame ionization detection. All technicalinformationaswellasthevalidationdataofthemethod isdescribedinpreviouspublications[19,20].

Thecalibrationrangewasdeterminedto0.0948–4.7393g/kg, accuracy was 100–103%, intraday imprecision <0.8% and total imprecision1.7%.

2.3.AnalysesofEtGandEtSinblood

AllbloodsampleswereanalysedforEtGandEtS.Sampleswere stored refrigerated and shipped refrigerated to the laboratory before analysis. Storage and shipping time together varied betweenapproximatelyoneandfourweeksandstability experi-mentswereperformedtoinvestigatestabilityduringthreemonths at4C. Intheseexperiments,10sampleswerereanalysedafter threemonthsofstorageat4C.

ThemethodusedhaspreviouslybeenvalidatedforEtGandEtS quantificationinblood[21].Inbrief,toa100

m

Laliquotofstudy samples, calibration samples and control samples, internal standardwas addedbeforeproteinprecipitation(PPT)with ice-coldacetonitrile(ACN).Thesupernatantswerefilteredthrougha 96-well phospholipidremoval plate.The filtered samples were evaporated todryness and reconstituted in 150

m

L water/ACN/ formic acid. Analysiswas performedbyultra-highperformance liquidchromatography-tandemmassspectrometry.Alltechnical information as well as the validation data of the method is described in submitted publication. The calibration range was 0.089–22mg/LforEtGand 0.025–6.3mg/LforEtS,accuracyand precisionwere<10%forbothanalytes.

2.4.Interpretationofcases

Forallincludedparticipants,itwasdeterminedwhetherthe dataindicatedrecentintakeofethanol,accordingtotwodifferent methodsforassessmentofthehip-flaskdefence:useofethanol ratios betweenblood and urine and between two consecutive urine samples and the analyses of the non-oxidative ethanol metabolitesEtGandEtSinblood.

For all subjects included in the study, samples collected at 180min after start of _first drinking and the following sample (210min for blood and 240min for urine) were assessed. This correspondsto60,90and120minaftertheseconddose(hip-flask drink)and thesetime pointswereconsideredrealisticforwhat wouldhavebeencollectedafteranaccidentorothersuspected drunkdriving.Thiswouldimplythatweassumeintakeofafirst alcoholdoseapproximatelytwohoursbeforeaveryshortdriving, ingestion of thehip-flaskdrink rightafter driving,and sample collection60and90or120minafterthis.Wethenassessedifthe samplesinvestigatedwoulddetecttherecentintakeofthesecond dose.

AnUAC/BACratioinsamplescollected180minafterstartof drinkingandthedifferencebetweenUACat240minandUACat 180minwascalculatedforallsubjects.AUAC/BACratiobelow1.3 andincreasingUACvalueswereconsideredindicativeforrecent drinking.

CorrespondinglyforEtGandEtS,aratiobetweenbloodethanol (g/kg)andEtG/EtS(mg/L)insamplescollected180minafterstart of drinking and the difference between blood EtG and EtS at 210min and 180min were calculated for all subjects. A ratio betweenethanol(g/kg)andEtG/EtS(mg/L)above1andincreasing EtGandEtSvalueswereconsideredindicativeforrecentdrinking. OnlydetailsfortheEtGcriteria,nottheEtScriteria,areshown. 2.5.Statistics

Kineticaversion5.1 (Thermo FisherScientificInc., Waltham, MA,USA)wasusedforpharmacokineticprocessingoftheEtGand EtSdata.Themaximumconcentration(Cmax)andthetimetoreach

Cmax(Tmax)fordifferentdoseswerecalculatedforeachsubjectin

theKineticasoftware.IBMSPSS1_{Softwareversion19.0(SPSSInc.,}

Chicago, IL,USA) was used for statistical analyses. Median and range values are reported. Figures were prepared in excel. DifferencesinBAC,UACand Tmaxfor ethanolwereinvestigated

usingKruskalWallisnon-parametrictest.

3.Results

Ofthe37subjects,two(number3and18)wereexcludeddueto onlypartialingestionofthesecondalcoholdose.Furtherresults are therefore from 35 subjects (14 in the low dose, 16 in the mediumdoseand5inthehighdosegroup).

Table1

Demographicsanddosesofthesubjects.

Group(2nddrink) Subject(#) Gender(Male/Female) Age(years) Weight(kg) Height(cm)

0.25g/kgBeer 1 F 25 79.2 175 2 F 22 69.0 174 3* F 24 84.0 178 4 M 23 63.8 163 5 M 20 71.5 179 0.25g/kgWine 6 M 22 76.7 179 7 M 21 82.0 191 8 M 22 80.3 186 9 F 21 67.0 173 10 M 23 77.1 178 0.25g/kgVodka 11 F 24 73.0 170 12 F 25 63.8 178 13 F 25 84.6 181 14 F 22 67.7 167 15 F 23 60.1 163 0.51g/kgBeer 16 F 24 65.0 167 17 M 23 85.0 193 18* F 23 62.0 173 19 M 27 82.1 185 20 M 21 87.0 189 0.51g/kgWine 21 M 24 91.0 189 22 M 28 72.0 185 23 F 21 55.0 165 24 M 20 70.0 183 25 M 21 79.7 185 0.51g/kgVodka 26 M 24 111.9 189 27 M 24 82.4 185 28 M 21 86.5 188 29 M 23 89.6 184 30 M 21 78.0 189 31 F 25 84.6 186 32 F 22 72.2 163 0.85g/kgVodka 33 M 22 75.0 178 34 M 24 83.0 178 35 M 24 87.0 185 36 M 22 77.9 183 37 M 25 81.0 185 *

Subject3and18wereexcludedduetopartialintakeofsecondalcoholdose.

Table2

CmaxandTmaxvaluesforBACandUACinthedifferentdosinggroups.

BloodMax1 UrineMax1 BloodMax2 UrineMax2

Dose(g/kg body weight) Cmax(g/kg) median(range) Number Tmax(min) median(range) Number Cmax(g/kg) median(range) Number Tmax(min) median(range Number Cmax(g/kg) median(range) Number Tmax(min) median(range) Number Cmax(g/kg) median(range) Number

Tmax(min)median

(range)Number 0.51+0.25 0.41(0.29–0.64) 90(60–105) 0.58(0.35–0.80) 90(90–120) 0.58(0.48–0.77) 165(150–210) 0.72(0.59–0.92) 210(180–240) 14 14 12 12 14 14 14 14 0.51+0.51 0.39(0.29–0.61) 75(75–105) 0.58(0.44–0.84) 90(90–90) 0.90(0.62–1.27) 180(165–210) 1.10(0.82–1.54) 240(240–240) 15 15 15 15 16 16 16 16 0.51+0.85 0.41(0.30–0.44) 83(75–105) 0.59(0.47–0.64) 90(90–90) 1.27(1.26–1.37) 180(180–210) 1.69(1.59–1.74) 240(240–240) 4 4 5 5 5 5 5 5

3.1.Resultsfrombloodandurineethanolmeasurements

Based onresultsfromcomparisonswiththenon-parametric KruskalWallistest,therewerenodifferencesinmaximumBAC, maximumUACorTmaxforbloodorurinebetweenbeer,wineor

vodkainthesamedosinggroup.Resultsfromthedifferenttypesof ethanol are therefore presented together, according to dosing group.AsummaryoftheresultsforethanolisshowninTable2, whereCmaxandTmaxvaluesforethanolinbloodandurineafter

bothfirstandsecondingestionareseen.

Fig.2depictsindividualBAC,UACandUAC/BACratiosforthe subjectsinthemediumdosinggroup(0.51+0.51g/kg).Sixofthe subjectshad no measurable BAC30minafter thebeginning of drinking.Allsubjectsexceptnumber 28reachedamaximumin BACbeforetheseconddosewasingested.Additionaldataforlow andhighdosesareshowninSupplementalFigs.1and2.

3.2.ResultsfrombloodEtGandEtSmeasurements

Thereanalysesof samplesafter threemonthsshowedsmall deviations inconcentrations, witha 20%decreasein blood EtG from a sample stored at 4_{C as the largest difference. As the}

shippingperiodwas muchshorterthanthreemonths,bothEtG andEtSwereconsideredstableforthepurposeofthestudy.

CmaxandTmaxvaluesforEtG andEtS inthedifferentdosing

groupsareseeninTable3.Fig.3depictsindividualconcentrations ofEtGandEtSaswellastheratiosbetweenethanolandEtGforthe subjectsinthemediumdosinggroup(0.51+0.51g/kg).Additional dataforlowandhighdosesareshowninSupplementalFigs.3and 4.

3.3.Interpretationofcasesusingthetwomodels

All subjects were assessed according to fulfilment of the describedcriteriaforrecentdrinking(seesupplementalTable1for alldata).Allsubjectsinthehighdosegroupfulfilledcriteriafor recentdrinkingusingthesamplescollected180and210–240min afterstartofdrinkingthefirstdrink(60–120minafterthe hip-flaskdrink).However,onesubjectinthemediumdosegroupand ninesubjectsinthelowdosegroupfailedtofulfiloneormoreof thecriteria.ThevaluesforUAC/BACratio,increaseinUAC,blood ethanol/EtGratioandincreaseinEtGbloodconcentrationsforall thesesubjectsareseeninTable4.

AsseeninTable4,eightsubjectsdidnotfulfilthecriteriafor recentdrinkingregardingtheincreaseinUACfrom180–240min afterstartof intake.RegardingtheratiobetweenUACand BAC 180minafterstartofintake,eightsubjectsalsofailedtofulfilthis criterion.

AsalsoseeninTable4,onesubjectdidnotfulfilthecriteriafor recentintakeregardingincreasingconcentrationsofbloodEtGand EtS(datanotshown).ItshouldbenotedthattheincreaseinEtS concentrationswasgenerallylesssteepthantheincreaseinEtG concentrations,makingEtSsomewhatlesssuitableforassessment ofthiscriteria.Regardingtheratiobetweenethanol(g/kg)andEtG/ EtS(mg/L),allsubjectsfulfilledcriteriaforrecentintake. 4.Discussion

Thepresent studyshowed pharmacokineticsof ethanol, EtG and EtS after ingestion of two doses of ethanol and used the experimentaldatatoinvestigatethehip-_flaskdefencebytheuseof twopharmacologicallydifferentbutobjectivemethods.Thefirst methodis the ratiobetween urine and blood ethanol concen-trationsandthedifference inurinaryethanol concentrationsin

twoconsecutive voids(forsimplicitycalledtheethanolmodel). ThesecondmethodisthedifferenceinEtG/EtSconcentrationsin twoconsecutivebloodsamplesandtheratiobetweenethanoland EtG/EtSinblood(forsimplicitycalledtheEtGmodel).

The possibility to objectively assess recent drinking has previously been indicated successful by the use of the two methods, but only when a single dose of ethanol is ingested [14,22]. When a suspect claims that alcohol was exclusively ingested after driving, theinterpretation is therefore relatively straightforward[7,12].However,suspectsoftenclaimthatalcohol wasingestedbothbeforeandafterdriving,andtheexperimental basisforsuchascenariohasbeenverylimited[20].

Thepresentresultsshowedthatthesamecriteriacanbeused whentwodosesareingested,butonlywhentheseconddoseis sufficientlyhighcomparedtothefirstdose.Therefore,ifasuspect claimsanintakeafterdrivingthatiscomparableorhigherthan the intake before driving, the criteria for recent drinking are assumed to be present. It should however be noted that the present study investigated one specific scenario, i.e. one dose ingestedtwo hoursbeforea veryshortdriving andthesecond dose ingested directly after driving, one to two hours before samplecollection.In reallife cases, avast numberof different scenariosregardingtimeandsizeofthetwodosescouldoccur. The presentresults mustthereforebeusedwithcautionwhen transferredtoreallifecases.

Oneaimofthepresentstudywastocomparethetwodifferent models,theethanolmodelandtheEtGmodel.Forthemediumand highdosegroup,theyshowedequallysatisfactoryresults,butfor the low dose group,the ethanol model failed todetect recent drinkinginmorethanhalfoftheincludedsubjects.Thesecasesare however assumed to represent the smallest problemin traffic cases,asingestionofasmallamountofalcoholafterdrivingwill havesmallerimpactonthecalculatedbloodalcoholconcentration andthereforebeoflessforensicinterest,unlessthemeasuredBAC isclosetothelegallimit.Wethereforeconcludethatbothmodels canbeusedtoassessthehipflaskdefence.FortheEtGmodel,EtS canbeassessedinadditiontoEtG,butsinceEtSshowsalesssteep riseinconcentrations,EtGisthemoreusefulofthetwo.

In a previous smaller study, only the ethanol model was investigated, andit was concluded that theincrease in urinary ethanol concentrations in two consecutive voids was more sensitivethantheUAC/BACratio[20].Thepresentstudydidnot supportthis,asanequalnumberofpatientsfailedtofulfil both thesecriteria.

Regardingpharmacokinetics,thepresentstudyadded knowl-edgeaboutCmaxandTmaxvaluesforethanolandEtG.Forethanol,a

medianCmaxof0.4g/kgafteringestionof0.51gofethanolperkilo

body weightwas in accordance withor somewhat lower than previous studies investigating the same dose [17,23], probably causedbytheslowingestioninthepresentstudy.Whenthesame dose was repeated, the maximum concentrations more than doubled, to a median of 0.9g/kg. This couldbe caused by the quicker ingestion of the second dose compared to the first, reducing_firstpassmetabolismofethanol[24].

Table3

CmaxandTmaxvaluesforEtGandEtSinthedifferentdosinggroups.

Dose(g/kgbody weight)

CmaxEtG(mg/L)median(range)

Number

CmaxEtS(mg/L)median(range)

Number

TmaxEtG(min)median(range)

Number

TmaxEtS(min)median(range)

Number 0.51+0.25 0.50(0.30–0.71) 0.25(0.17–0.45) 300(240–360) 240(180–300) 14 14 14 14 0.51+0.51 0.80(0.37–1.03) 0.35(0.25–0.44) 360(300–420) 270(240–300) 16 16 16 16 0.51+0.85 1.31(0.88–1.54) 0.55(0.44–0.80) 420(360–420) 300(270–420) 5 5 5 5

For EtG,maximum concentrations afteringestion ofa given dose of ethanol have previously been published. For example, maximumconcentrationsof1.06mg/Lwasseenafteringestionof 1.0g of ethanol per kilo body weight [23]. The present study showedmedianmaximumconcentrationsof0.8mg/Lwhenatotal dose of 1.02g of ethanol per kilo body weight was ingested, indicatingthatlowerpeakconcentrationswillbeseenwhendoses aredividedintwo.Thiscouldbeexplainedbyforinstancehigher relativefirstpassmetabolismofethanolwhensmallerdosesare ingested[24],butalsototheslowerintakeofthefirstdoseinthe presentstudy,makingmaximumBAClower.

The strength of thepresent studyis the combination of an experimental studywitha practical approach toassessment of cases.Also, a strictprotocolwas applied and allsamples were analysedforethanoland EtG/EtSusingfullyvalidatedmethods. Themainweaknessofthepresentstudyistheinvestigationofonly onespeci_ficscenario,andinclusionofonlyyoungparticipantswith body mass index, volume of distribution, and elimination not representativeofawiderpopulation.Also,fortheEtGmethod,one of the criteria for recent drinking, i.e. the increase in EtG concentrations between the two samples collected 180 and 210minafterstartoffirstdrinking,couldpossiblybearesultof thefirstdoseEtGincrease,althoughatthistimepoint,aplateau levelcouldalsobepresent[14,16].Fortheratioethanol(g/kg)/EtG (mg/L),theseresultscouldnotbearesultofthefirstdoseingestion [14,24]andwillreflectthecombinationofthetwodoses,aswe aimedtostudy.

In conclusion, the present study showed that the two investigatedmethodstoevaluatethehip-flaskdefencecouldalso beusedwhenalcoholwasalreadyonboardfromapreviousintake, butonlyifthelastdoseissufficientlyhigh.

Authorstatement

GHsupervisedanalyses,participatedinanalyzingthedata,and wasresponsibleforwritingthemanuscript

MFactedascoordinator,supervisedthesubjects,obtainedand analyzedsamples,analyzeddataandparticipatedinthe prepara-tionofthemanuscript

GN,RL,IN,CO,andEEsupervisedthesubjects,obtainedand analyzedsamples,analyzeddataandparticipatedinthe prepara-tionofthemanuscript

CKappliedforethicsapprovalandfundingandreviewedthe manuscript

MCandJAactedasmedicalsupervisorsandparticipatedinthe preparationofthemanuscript

FKdesignedthestudyandparticipatedinthepreparationofthe manuscript

RKactedasprincipalinvestigator,designedthestudy,applied forethicsapprovalandfunding,analyzeddataandparticipatedin thepreparationofthemanuscript

Allauthorsapprovedthesubmittedversionofthemanuscript.

AppendixA.Supplementarydata

Supplementarymaterialrelatedtothisarticlecanbefound,in the online version, at https://doi.org/10.1016/j.forsciint.2020. 110409.

References

[1]H.Gjerde,P.T.Normann,B.S.Pettersen,T.Assum,M.Aldrin,U.Johansen,L. Kristoffersen, E.L. Oiestad,A.S. Christophersen, J. Morland, Prevalence of alcoholanddrugsamongNorwegianmotorvehicledrivers:aroadsidesurvey, Accid.Anal.Prev.40(2008)1765–1772.

[2]H. Gjerde, A.S. Christophersen, P.T. Normann, J. Morland, Toxicological investigationsofdriverskilledinroadtrafficaccidentsinNorwayduring 2006-2008,ForensicSci.Int.212(2011)102–109.

[3]H.Gjerde,P.T.Normann,A.S. Christophersen,S.O.Samuelsen,J. Morland, Alcohol,psychoactivedrugsandfatalroadtrafficaccidentsinNorway:a case-controlstudy,Accid.Anal.Prev.43(2011)1197–1203.

[4]S.A.Legrand,H.Gjerde,C.Isalberti,T.VanderLinden,P.Lillsunde,M.J.Dias,S. Gustafsson,G.Ceder,A.G.Verstraete,Prevalenceofalcohol,illicitdrugsand psychoactivemedicinesinkilleddriversinfourEuropeancountries,Int.J.Inj. Contr.Saf.Promot.(2013).

[5]S.T.Bogstrand,H.Gjerde,Whichdrugsareassociatedwithhighestriskfor beingarrestedfordrivingundertheinfluence?Acase-controlstudy,Forensic Sci.Int.240c(2014)21–28.

[6]H.Gjerde,M.C.Strand,J.Morland,Drivingundertheinfluenceofnon-alcohol drugs—anupdatepartI:epidemiologicalstudies,ForensicSci.Rev.27(2015) 89–113.

[7]R.Iffland,A.W.Jones,Evaluatingallegeddrinkingafterdriving–thehip-flask defence.Part1.Doublebloodsamplesandurine-to-bloodalcoholrelationship, MedSci.Law.42(2002)207–224.

[8]R.Iffland,A.W.Jones,Evaluatingallegeddrinkingafterdriving—thehip-flask defence.Part2.Congeneranalysis,Med.Sci.Law43(2003)39–68. [9]D.Brahams,Backcalculationofalcohol:the"hip-flaskdefence",Lancet2

(1988)1504.

[10]A.W.Jones,ToptendefencechallengesamongdrinkingdriversinSweden, Med.Sci.Law31(1991)229–238.

[11]S.M.Havig,G.Hoiseth,M.C.Strand,R.A.Karinen,G.W.Brochmann,D.H.Strand, L.Bachs,V.Vindenes,THCandCBDinbloodsamplesandseizuresinNorway: doesCBDaffectTHC-inducedimpairmentinapprehendedsubjects?Forensic Sci.Int.276(2017)12–17.

[12]G.Hoiseth,G.O.Berg-Hansen,J.Morland,Evaluationofthehip-flaskdefence bydeterminationofethylglucuronideandethylsulphateconcentrationsin blood,ForensicSci.Int.257(2015)398–402.

[13]A.W.Jones,F.C.Kugelberg,Relationshipbetweenbloodand urinealcohol concentrationsinapprehendeddriverswhoclaimedconsumptionofalcohol afterdrivingwithandwithoutsupportingevidence,ForensicSci.Int.194 (2010)97–102.

[14]G.Hoiseth,J.P.Bernard,R.Karinen,L.Johnsen,A.Helander,A.S.Christophersen, J.Morland,Apharmacokineticstudyofethylglucuronideinbloodandurine: applicationstoforensictoxicology,ForensicSci.Int.172(2007)119–124. [15]G.Hoiseth,L.Morini,A.Polettini,A.Christophersen,J.Morland,Bloodkinetics

ofethylglucuronideandethylsulphateinheavydrinkersduring alcohol detoxification,ForensicSci.Int.188(2009)52–56.

[16]A.M.Lostia,J.L.Vicente,D.A.Cowan,Measurementofethylglucuronide,ethyl sulphateandtheirratiointheurineandserumofhealthyvolunteersaftertwo dosesofalcohol,AlcoholAlcohol.48(2013)74–82.

[17]A.W.Jones,Ethanoldistributionratiosbetweenurineandcapillarybloodin controlledexperimentsandinapprehendeddrinkingdrivers,J.ForensicSci.37 (1992)21–34.

[18]A.W.Jones,Referencelimitsforurine/bloodratiosofethanolintwosuccessive voidsfromdrinkingdrivers,JAnal.Toxicol.26(2002)333–339.

Table4

Evaluationparametersforthe9subjectsthatdidnotfulfilcriteriaforrecentdrinking. Subject UEth240-UEth180 Indicativeofrecent intake? UEth180/ BEth180 Indicativeofrecent intake? BEtG210-BEtG180 Indicativeofrecent intake? Beth/ BEtG180 Indicativeofrecent intake? 1 0.06 No 1.34 No 0.14 Yes 1.4 Yes 2 0.14 No 1.34 No 0.08 Yes 2.2 Yes

4 0.01 Yes 1.36 No 0.08 Yes 2.4 Yes

5 0.15 No 1.54 No 0.06 Yes 2.0 Yes

6 0.09 No 1.32 No 0.1 Yes 1.3 Yes

7 0.08 No 1.37 No 0.08 Yes 2.1 Yes

8 0.06 No 1.32 No 0.07 Yes 2.9 Yes

9 0.01 No 1.17 Yes 0.1 Yes 2.2 Yes

11 0 No 1.21 Yes 0 No 2.3 Yes

[19]A.W. Jones,J. Schuberth,Computer-aidedheadspacegaschromatography appliedtoblood-alcoholanalysis:importanceofonlineprocesscontrol,J. ForensicSci.34(1989)1116–1127.

[20]C.Kronstrand,G.Nilsson,M.D.Cherma,J.Ahlner,F.C.Kugelberg,R.Kronstrand, Evaluating the hip-flask defence in subjects with alcohol on board: an experimentalstudy,ForensicSci.Int.294(2019)189–195.

[21]L.V.Sidqey,L.Kristoffersen,Quantitativedeterminationofethylglucuronide and ethyl sulfate in postmortem and antemortem whole blood using phospholipidremoval96-wellsplatesandUHPLC–MS/MS,J.Anal.Toxicol. (2020).

[22]A.W.Jones,Urineasabiologicalspecimenforforensicanalysisofalcoholand variabilityintheurine-to-bloodrelationship,Toxicol.Rev25(2006)15–35. [23]G.Hoiseth,B.Yttredal,R.Karinen,H.Gjerde,J.Morland,A.Christophersen,

Ethyl glucuronide concentrations in oral fluid, blood, and urine after volunteers drank0.5 and1.0 g/kgdoses ofethanol, J. Anal. Toxicol.34 (2010)319–324.

[24]A.Norberg,A.W. Jones,R.G.Hahn,J.L.Gabrielsson, Roleofvariability in explainingethanolpharmacokinetics:researchandforensicapplications,Clin. Pharmacokinet.42(2003)1–31.