The association of urinary polycyclic aromatic hydrocarbon biomarkers and cardiovascular disease ...

127 Pages·2016·4.7 MB·English

University of Nebraska - Lincoln
DigitalCommons@University of Nebraska - Lincoln
Public Health Resources Public Health Resources
2016
The association of urinary polycyclic aromatic
hydrocarbon biomarkers and cardiovascular
disease in the US population
Omayma Alshaarawy
Michigan State University, [email protected]
Hosam A. Elbaz
The University Of Michigan
Michael E. Andrew
Biostatistics and Epidemiology Branch, Health Effects Laboratory Division, National Institute for Occupational Safety and
Health, Centers for Disease Control and Prevention, Morgantown, WV
Follow this and additional works at:http://digitalcommons.unl.edu/publichealthresources
Alshaarawy, Omayma; Elbaz, Hosam A.; and Andrew, Michael E., "The association of urinary polycyclic aromatic hydrocarbon
biomarkers and cardiovascular disease in the US population" (2016).Public Health Resources. 426.
http://digitalcommons.unl.edu/publichealthresources/426
This Article is brought to you for free and open access by the Public Health Resources at DigitalCommons@University of Nebraska - Lincoln. It has
been accepted for inclusion in Public Health Resources by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln.
EnvironmentInternational89–90(2016)174–178
ContentslistsavailableatScienceDirect
Environment International
journal homepage: www.elsevier.com/locate/envint
The association of urinary polycyclic aromatic hydrocarbon biomarkers
and cardiovascular disease in the US population
OmaymaAlshaarawya,⁎,HosamA.Elbazb,MichaelE.Andrewc
aDepartmentofEpidemiologyandBiostatistics,MichiganStateUniversity,EastLansing,MI48824,UnitedStates
bDepartmentofInternalMedicine,DivisionofGastroenterology,UniversityofMichigan,AnnArbor,MI48109,UnitedStates
cBiostatisticsandEpidemiologyBranch,HealthEffectsLaboratoryDivision,NationalInstituteforOccupationalSafetyandHealth,CentersforDiseaseControlandPrevention,Morgantown,
WV26506,UnitedStates
a r t i c l e i n f o a b s t r a c t
Articlehistory: Background:Polycyclicaromatichydrocarbons(PAHs)arepotentatmosphericpollutantsproducedbyincom-
Received7October2015 pletecombustionoforganicmaterials.Pre-clinicalandoccupationalstudieshavereportedapositiveassociation
Receivedinrevisedform7January2016 ofPAHswithoxidativestress,inflammationandsubsequentdevelopmentofatherosclerosis,amajorunderlying
Accepted3February2016 riskfactorforcardiovasculardisease(CVD).Theaimofthecurrentstudyistoestimatetheassociationbetween
Availableonlinexxxx
levelsofPAHbiomarkersandCVDinanationalrepresentativesampleofUnitedStates(US)adults.
Methods:Weexaminedadultparticipants(≥20yearsofage)fromthemergedUSNationalHealthandNutrition
Keywords:
ExaminationSurvey2001–2010.Logisticregressionmodelswereusedtoestimatetheassociationsofeachuri-
Cardiovasculardisease
naryPAHbiomarkerandCVD.Post-exploratorystructuralequationmodelingwasthenusedtoaddresstheinter-
Polycyclicaromatichydrocarbons
NHANES dependent response variables (angina, heart attack, stroke and coronary heart disease) as well as the
Structuralequationmodeling interdependenciesofPAHbiomarkers.
Results:PAHbiomarkerswerepositivelyassociatedwithcardiovasculardiseaseinmultiplelogisticregression
models,althoughsomeassociationswerenotstatisticallyrobust.Usingstructuralequationmodeling,latent
PAHexposurevariablewaspositivelyassociatedwithlatentCVDlevelvariableinthemultivariableadjusted
model(β=0.12;95%CI:0.03,0.20).
Conclusion:AmodestassociationbetweenlevelsofPAHbiomarkersandCVDwasdetectedinUSadults.Further
prospectivestudieswithadequatesamplesizeareneededtoreplicateorrefuteourfindings.
©2016ElsevierLtd.Allrightsreserved.
1.Introduction producedasbyproductsofsmoking,indoorsandoutdoorsfuelburning
andfoodgrilling(Liuetal.,2008;Rameshetal.,2004;Simko,2005).
Despite advances in prevention, diagnosis and treatment, Pre-clinicalstudieshavereportedapositiveassociationbetweenexpo-
cardiovascular disease (CVD) remains the number one cause of suretoPAHs,oxidativestressandatherosclerosis(Jengetal.,2011;
mortalityinUnitedStates(U.S.)adults(Kochaneketal.,2014).Car- Curfsetal.,2005).Epidemiologicalstudieshavefoundapositiveassoci-
diovasculardiseasereferstonumerousconditions,manyofwhich ationbetweenbiomarkersofPAHexposureandseruminflammatory
arerelatedtoatherosclerosiswhichdevelopswhenplaquebuilds markersofCVD(Alshaarawyetal.,2013).Inaddition,apositiveassoci-
up in arterial walls, narrowing the arteries, and decreasing or ationbetweenoccupationalexposuretoPAHsandCVDmorbidityand
sometimes completely blocking tissue blood flow (Goff et al., mortality has been reported (Brucker et al., 2014; Burstyn et al.,
2014).Suchmechanismscanmanifestthemselveseventuallywith 2005).HoweveritisnotclearifexposuretoPAHsisassociatedwithcar-
adverseclinicaloutcomessuchascoronaryheartdisease,angina diovasculardiseaseinthegeneralpopulation.
pectoris,heartattack,andstroke. Withadvantagesofrecentnationallyrepresentativesamplesurveys
Epidemiologicalevidencesuggeststhatexposuretoparticulatemat- andstandardizeddatacollectionapproacheswithrelativelylargesam-
terpresentinambientairmayposeanincreasedCVDriskinhumans ples,weexaminedtheassociationofPAHexposureandCVDinthe
(Bhatnagar,2006).Polycyclicaromatichydrocarbons(PAHs)arepotent UnitedStates(US)generalpopulation.Inaddition,weusedStructural
atmosphericpollutantsthatoccurinoil,coal,andtardeposits,andare Equation Modeling approach trying to account for the interdepen-
denciesthatmustbeconfrontedwhendifferentPAHbiomarkersare
studiedallatonce.Wealsoareaccountingforthesharedpathogenesis
⁎ Correspondingauthorat:DepartmentofEpidemiologyandBiostatistics,909FeeRoad
oftheresponsevariables,namelytheCVDevents(angina,heartattack,
B601WestFeeHall,MichiganStateUniversity,EastLansing,MI48824,UnitedStates.
E-mailaddress:[email protected](O.Alshaarawy). strokeandcoronaryheartdisease).
http://dx.doi.org/10.1016/j.envint.2016.02.006
0160-4120/©2016ElsevierLtd.Allrightsreserved.
O.Alshaarawyetal./EnvironmentInternational89–90(2016)174–178 175
2.Methods thespikingsolutionsand/orreagents.TwoQCLsandtwoQCHswere
preparedandanalyzedatthebeginningandtheendofeachrun;their
2.1.Studypopulation concentrationswerecomparedwithacceptancecriteriatoassurethe
properoperationoftheanalysis.Relativeretentiontimeswereexam-
TheNationalHealthandNutritionExaminationSurveys(NHANES) inedfortheinternalstandardtoensurethechoiceofthecorrectchro-
consistofaseriesofsurveysdesignedbytheNationalCenterforHealth matographicpeak.Moredetailsonqualitycontrolprocedurescanbe
StatisticstocontinuouslymonitorthehealthstatusoftheU.S.civilian foundintheonlinesupplementalmaterials.
non-institutionalizedpopulation(UnitedStatesCentersforDisease Seven urinary low molecular weight PAH analytes (1-hydroxy-
ControlandPrevention.NationalCenterforHealthStatistics(NCHS), napthol,2-hydroxynapthol,2-hydroxyfluorene,3-hydroxyfluorene,
2010).TheNHANESsurveyincludesastratifiedmultistageprobability 1-hydroxyphenanthrene,2-hydroxyphenanthrene,3-hydroxyphenan-
sample.Selectionisbasedoncounties,blocks,householdsandindivid- threne)andoneurinaryhighmolecularweightanalyte(1-hydroxy-
ualswithinhouseholds,andincludedoversamplingofnon-Hispanic pyrene)wereconsistentlyavailableinNHANES2001–10.Fewother
BlacksandMexicanAmericansinordertoprovidestableestimatesof biomarkersarenotincludedinthecurrentstudybecausetheywere
thesegroups.Dataarecollectedforatwo-yearsurveycycle. notavailableinallyears.Allanalytesweremeasuredinthesameunit;
Inthecurrentstudywemerged2001–02,2003–04,2005–06, ng/L.UrinaryanalytesofPAHswerecorrectedforcreatinineconcentra-
2007–08and2009–10datacycles,whereeightanalytesofPAHs tiontoreducetheirvariabilitybydifferencesinurinaryoutputs.Urinary
havebeenconsistentlymeasured.Therewere27,584participants levelsofOH-PAH(ng/L)weredividedbyurinarycreatininelevel
whowere≥20yearsofage.UrinaryPAHbiomarkerswereonlymea- (mg/dL) multiplied by 0.01, i.e., [(ng/L) ÷ (mg/dL ∗ 0.01)] and
suredinasubsampleNHANES(n=7848).Participantswithmissing expressedasnanogrampergramofcreatinine(ng/gcreatinine).
informationonserumcotinineandothercovariateswerenotinclud-
edinthefinalmodel.Thisresultedin7301participantsincludedin 2.4.Covariates
thefinalanalyses.
InNHANES,informationonage,gender,ethnicself-identification
2.2.Outcome:Cardiovasculardisease (ESI),alcoholdrinking,andpoverty-incomeratio(PIR)wereobtained
duringastandardizedquestionnaireduringahomeinterview.Informa-
Astandardizedmedicalconditionquestionnairewasadministered tiononanthropometric,physicalandlaboratorycomponentswereob-
duringthepersonalinterviewonabroadrangeofhealthconditionsin- tainedduringthemedicalcenterexamination.Bodymassindex(kg/
cludingcoronaryheartdisease,anginapectoris,heartattackandstroke. m2)wascalculatedasweightinkilogramsdividedbyheightinmeters
Participantswereasked“hasadoctororotherhealthprofessionalever squared.Serumtotalcholesterol(mg/dL)wasmeasuredenzymatically.
toldyouthatyouhave:coronaryheartdisease(CHD)/anginapectoris/ Serumcotinine(ng/mL)wasmeasuredbyanisotopedilution-highper-
heartattack/stroke?”(Thesewere4separatequestionswiththesame formanceliquidchromatographyatmosphericpressurechemicalioni-
wordingstyle).AparticipantwasconsideredaCVDcaseifshe/hean- zationtandemmassspectrometry.
swered “yes” to any of the previous questions (0: no event occur-
rence/1:occurrenceofatleastoneofthe4clinicalevents). 2.5.Statisticalanalysis
2.3.Mainexposure:UrinarylevelsofPAHbiomarkers Exploratorydataanalysistechniqueswereusedtoassesstheunivar-
iatedistributionofthestudyvariables.UrinaryPAHbiomarkerlevels
Detailedurinespecimencollectionandprocessinginstructionsare werelog-transformedasaresultoftheirskeweddistribution.Weran
discussedintheNHANESLaboratory/MedicalTechnologistsProcedures logisticregressionmodelstocalculatetheoddsratio([OR]and95%con-
Manual(UnitedStatesCenterforDiseaseControlandPrevention.Na- fidenceinterval[CI])ofself-reportedCVD,foreachurinaryOH-PAH,
tionalCenterforHealthStatistics(NCHS),2006).Specificanalytesmea- controllingforage(years)andsex.Finalmodelswereadditionallyad-
suredinNHANESaremonohydroxy-PAH(OH-PAH).Theprocedure justedforESI(non-HispanicWhite,non-HispanicBlack,Hispanics,and
involvesenzymatichydrolysisofurine,withextraction,derivatization allothers),education(lessthanhighschool,highschool,andabove
andanalysisusingcapillarygaschromatographycombinedwithhigh highschool),PIR,past-yearalcoholdrinking,BMI(normal,overweight,
resolutionmassspectrometry(GCHRMS).Thismethodusesisotopedi- obese),totalcholesterol(mg/dL),andserumcotinine(ng/mL).
lutionwithcarbon-13labeledinternalstandards.Ionsfromeachanalyte StructuralEquationModeling(SEM)wasthenusedtoconstructa
andeachcarbon-13labeledinternalstandardaremonitored,andthe seriesofmodelstoestimatetheassociationofPAHsandCVDlevel.
abundanceofeachionismeasured.Theratiosoftheseionsareused Buncheretal.discussedtheapplicationofSEMinEnvironmentalEpide-
as criteria for evaluating the data. By evaluating these analytes in miologyearlyin1991(Buncheretal.,1991).Structuralequationmodel-
urine,ameasurementofthebodyburdenfromPAHexposureisobtain- ing is a group of statistical methods that can model relationships
ed(Adetaileddescriptionisavailableonlineathttp://www.cdc.gov/ betweenoneormoreindependentvariablesandoneormoredepen-
nchs/data/nhanes/nhanes_03_04/l31pah_c_met.pdf). dentvariables(Ullman,2006).Inthecurrentanalysis,PAHexposureis
Thelimitofdetection(LOD)wasdefinedasthehigherLODcalculat- alatentconstructthatisnotdirectlymeasuredbutratherassessedindi-
edbytwomethods:(i)indirectrelationtomethodblankspreparedin rectlybyPAHbiomarkers.InsteadofsimplycombiningPAHbiomarkers
parallelwiththeunknownsamples,as3timesthestandarddeviationof bytakingthesumandassigningequalweighttoeachbiomarker,they
themethodblanks,and(ii)accordingtotheinstrumentaldetection areemployedasindicatorsofalatentconstructandhenceallowsfores-
limitdefinedasthelowestpointinthecalibrationcurve(0.5pg/uL,or timationandremovalofthemeasurementerror(Ullman,2006).Limita-
5pg/mLin2-mLurinesamples)verifiedtogiveasignalwiththeS/N tionsofincludingmultipleindicatorsofthesameexposureinregression
equaltoorN3.Thelimitofdetectionforeachbiomarkerandeach models(suchascollinearity)areaccountedforintheSEMapproach
datacyclecanbefoundinthecorrespondingNHANESPAHlabfilespro- (Gefenetal.,2000;Suhr,2006).SimilarlyalatentconstructforCVD
videdassupplementarymaterials.b1%ofthestudypopulationhasPAH levelwascreatedviaself-reportedphysiciandiagnosis of coronary
levelsbelowlowerdetectionlimit. heartdisease,angina,heartattackand/orstroke.Forbothconstructs,
Checksweremadeonthestabilityoftheanalyticalsystem.Blanks wedefinedfactorloadingsof0.4orgreaterasstronglycorrelatedindi-
andstandards,aswellasqualitycontrolmaterials,wereaddedtoeach catorswiththelatentconstruct.Thispartofthemodelthatrelatesthe
day'srunsequence.Theblankandstandardwereanalyzedatthebegin- measuredvariablestothecorrespondinglatentconstructisthemea-
ningofeachruntocheckthesystemforpossiblecontaminationorin surementpartofthemodel.Thehypothesizedassociationbetween
176 O.Alshaarawyetal./EnvironmentInternational89–90(2016)174–178
thetwoconstructsisconsideredthestructuralpartofthemodel, Table2
regressingthelatentCVDlevelconstructastheresponsevariable Selectedpercentilesofurinarybiomarkersofpolycyclicaromatichydrocarbons(ng/gcre-
atinine)amongparticipantsincludedinthefinalanalysis.DatafortheUnitedStatesadults
onPAHsconstruct.Intheinitialmodels,weusedmaximumlikeli-
≥20yearsofagebasedontheNationalHealthandNutritionExaminationSurvey2001–
hoodestimationwithrobuststandarderrors(MLR).BecauseMLR
2010.
doesnotprovidefitindices,inourpost-exploratorystepweuseda
Biomarkers 10th Q1 Median Q3 90th
robustweightedleastsquaresestimator(WLSMV).Wedetermined
percentile percentile
goodnessoffitofthehypothesizedmodelstothedatabyseveralfit
indices,includingcomparativefitindex(CFI)≥0.95androotmean 1-Hydroxynaphthalene 515 901 2081 6941 17,532
squareerrorapproximation(RMSEA)≤0.05.(HuandBentler,1999; 22--HHyyddrrooxxyynflaupohrethnaelene 915073 1154726 3121247 7640594 151,761370
BrowneandCudeck,1992). 3-Hydroxyfluorene 33 50 86 279 909
1-Hydroxyphenanthrene 62 90 141 231 361
3.Results 2-Hydroxyphenanthrene 26 39 61 102 178
3-Hydroxyphenanthrene 36 53 85 158 297
1-Hydroxypyrene 26 44 83 159 313
Table1presentsmaincharacteristicsofthestudypopulation.About
onehalfofthestudypopulationwasfemale(51.2%).Themajorityofthe
study population was non-Hispanic White (71.1%). The arithmetic CVDconstructs,stratifiedbysex,ESIandsmokingstatus.Overall,PAH
meanofserumcotininewas64.1ng/mL.Table2presentsselectedper- wasassociatedwithCVDinthestratifiedsubgroups,althoughsomeas-
centilesofurinaryOH-PAHbiomarkers. sociationswerenotstatisticallyrobust.Wedidnotdetectsubgroupvar-
ThemainestimatesofthestudyarepresentedinTable3.Individ- iationsintheestimates.
ual polycyclic aromatic hydrocarbons were positively associated
with self-reported CVD. However, only 1-hydroxynaphalene, 2- 4.Discussion
hydroxynaphalene,2-hydroxyfluoreneand3-hydroxyfluorenewere
statisticallyrobust. Exposuretopolycyclicaromatichydrocarbonsmeasuredbyeight
Fig.1presentsthebaselineconceptualmodelregressingthelatent urinaryPAHbiomarkersexhibitedapositiveassociationwithcardiovas-
constructofCVDlevelastheresponsevariablesonthelevelofPAHex- culardiseaseinadults≥20yearsofageindependentofserumcotininea
posure,measuredbyeighturinaryPAHbiomarkers,controllingforage markerofnicotineexposureaswellasotherpotentialconfounders.Our
(years)andsex.ResultsindicatedapositiveassociationbetweenPAHs resultsareconsistentwithfindingsfrompreviousoccupationalstudies
andCVD(β=0.12;95%CI:0.07,0.17).Thefactorloadingswereall whichreportedpositiveassociationsbetweenexposuretoPAHs,andis-
≥0.4suggestingastrongcorrelationbetweenthemeasuresandeach chemicheartdiseasesandcardiovascularmortalityinoccupationsthat
correspondingconstruct. includelikelyexposuretoPAHs(Bruckeretal.,2014;Burstynetal.,
Table4presentstheassociationofPAHexposurelevelandCVDlevel 2005; Thériault et al., 1988; Tüchsen et al., 1996). Xu et al. used
constructsadditionallyadjustingforESI,education,PIR,alcoholdrink- NHANES2001–04andreportedapositiveassociationbetweenPAHbio-
ing, serum cotinine, total cholesterol and BMI. Our results did not markersandself-reportedCVDusinglogisticregressionmodes(Xu
changeappreciably(β=0.12;95%CI:0.03,0.20),suggestingapositive etal.,2010).Inthecurrentstudy,alatentCVDconstructapproachwas
associationbetweenPAHsandCVDindependentofthepotentialcon- selectedinadditiontoutilizingthestandardbinaryapproach(0:no
foundersstudiedhere.OurresultswereconsistentusingeitherMLRor eventoccurrence/1:occurrenceofatleastoneofthe4clinicalevents)
WLSMVestimator.Table4alsopresentstheassociationofPAHand toallowforconsiderationofthesharedpathogenesisinthedevelop-
mentofCHD,angina,heartattackandstrokebymeansofamaincom-
Table1 mon mechanism, namely atherosclerosis (Soler and Ruiz, 2010). A
Baselinecharacteristicsofthestudypopulation(n=7301).DatafortheUnitedStates latentPAHvariablewasalsousedtoaccountfortheinterdependencies
adults≥20yearsofagebasedontheNationalHealthandNutritionExaminationSurvey ofthePAHbiomarkersashumansareusuallyexposedtomixturesof
2001–2010. PAHs(Elovaaraetal.,1995).Duetothehighcostofdetectingparent
Characteristics Meanvalues±stderrorofmeanor PAHlevelsinhumans,themostcommonlyusedbiomarkersofPAHex-
samplesize(weightedpercentages) posureareurinaryOH-PAHbiomarkers.UrinaryPAHbiomarkershave
Age(years) 46.2±0.3 been previously found to correlate well with levels of exposure to
Sex(%) PAHsinthegeneralpopulation(Castano-Vinyalsetal.,2004).
Male 3550(48.8) ThemechanismsunderlyingthepositiveassociationofPAHexpo-
Female 3751(51.2) sureandCVDremainunknown.UponexposuretoPAHs,detoxification
Ethnicself-identification(%)
Non-HispanicWhite 3693(71.1)
Non-HispanicBlack 1378(10.7)
Hispanics 1919(12.8) Table3
Allothers 311(5.4) TheassociationofPAHbiomarkersandcardiovasculardisease.DatafortheUnitedStates
Education(%) adults≥20yearsofagebasedontheNationalHealthandNutritionExaminationSurvey
Lessthanhighschool 2079(18.3) 2001–2010.
Highschool 1773(25.1)
OH-PAH(ng/g Age-sexadjustedodds Multivariable-adjusted
Abovehighschool 3449(56.6)
Income-povertyratiob1 1325(12.6) creatinine) ratio(95%CI) oddsratio(95%CI)a
Past-yearalcoholdrinking(%) 4527(67.6) 1-Hydroxynaphthalene 1.13(1.07,1.21) 1.11(1.04,1.18)
Bodymassindex(kg/m2) 2-Hydroxynaphthalene 1.31(1.19,1.45) 1.22(1.08,1.38)
b25 2184(32.7) 2-Hydroxyfluorene 1.31(1.18,1.45) 1.27(1.12,1.43)
25–29.9 2542(33.4) 3-Hydroxyfluorene 1.21(1.11,1.33) 1.18(1.06,1.32)
≥30 2575(33.9) 1-Hydroxyphenanthrene 1.12(1.00,1.25) 1.07(0.94,1.21)
Serumcotinine(ng/mL) 64.1±2.8 2-Hydroxyphenanthrene 1.22(1.09,1.37) 1.11(0.98,1.26)
Totalcholesterol(mg/dL) 198.8±0.8 3-Hydroxyphenanthrene 1.13(1.00,1.27) 1.10(0.96,1.25)
AnyCVD(%) 732(7.9) 1-Hydroxypyrene 1.20(1.08,1.33) 1.12(1.00,1.26)
Coronaryheartdisease(%) 333(3.7)
a Adjustedforage(years),sex,ESI(non-HispanicWhite,non-HispanicBlack,Hispanic
Angina(%) 218(2.4)
andallothers),education(lessthanhighschool,highschoolandabovehighschool),PIR,
Heartattack(%) 307(3.3)
past-yearalcoholdrinking,BMI(normal,overweightandobese),totalcholesterol(mg/dL)
Stroke(%) 235(2.4)
andserumcotinine(ng/mL).
O.Alshaarawyetal./EnvironmentInternational89–90(2016)174–178 177
Fig.1.ModeldepictingthehypothesizedPAHexposure-CVDassociation.DatafortheNationalHealthandNutritionExaminationSurvey2001–2010(n=7301)
occursleadingtotheformationofhighlyreactiveintermediatesthatcan Severalofthestudylimitationsmeritattention.Ofmainconcernis
interactwiththeDNA(Curfsetal.,2004).SucheffectsofPAHsexposure thecrosssectionalnatureofNHANESwhichdoesnotallowtodraw
onplaquebuildupinanimalswerefoundtobedosedependent(Penn temporalorcausalinferencesregardingtheassociationofPAHsand
etal.,1981).Pre-clinicalstudieshavealsosuggestedthatPAHsmight CVD.Inaddition,ourstudydoesnothavethedatatoestimate the
alsoexerttheiratherogeniceffectviastimulationofaninflammatory sourcesofexposuretoPAHs.Urinarybiomarkermeasurementsreflect
processinvolvinganincreasedinfluxofproinflammatorycellsinto recentPAHexposureasnon-smokingsourcesofPAHscanvaryday-
theseplaques(Curfsetal.,2005).Theroleofinflammationasarisk to-dayinthegeneralpopulation.Finally,CVDwasascertainedbyself-
factorforCVDdevelopmenthasbeenwellestablished(Pearsonetal., report.Accordingly,somerecallbiasmayexist.However,thesebiases
2003;Goffetal.,2013).Populationbasedstudiessupportedanassocia- arelikely tobenon-differentialbiases,whichwouldminimizeany
tionbetweenPAHandinflammation(Alshaarawyetal.,2013;Clark observed association. Notwithstanding these limitations, the study
etal.,2012).Arecentstudyalsosuggestedanassociationbetween findingsareofinterestbecauseoftheinclusionofarelativelylargena-
PAHsandanumberofobesity-relatedcardiometabolichealthriskfac- tionallyrepresentativemultiethnicsample,theNHANESstandardized
tors(Ranjbaretal.,2015). datacollectionapproaches,andtheabilitytoadjustforpotential
confounders.
Inconclusion,thisepidemiologicalevidencefromthecurrent
studytendstoconfirmwhatpriorresearchfound–namely,aposi-
Table4
tiveassociationbetweenPAHexposureandCVD.Anyclaimofath-
Estimatedassociationofpolycyclicaromatichydrocarbonexposureoncardiovasculardis-
easelevelinadults(≥20+years),stratifiedbyparticipantcharacteristics.Dataforthe erogeniceffectsofPAHexposureispremature,butgivenincreased
UnitedStatesbasedontheNationalHealthandNutritionExaminationSurvey2001–2010. prevalence of PAH exposure and CVD in the U.S. and elsewhere,
thereisareasontostudytheirlinkages.Moreprobingexperimenta-
Characteristics Age-sexadjusteda Multivariableadjustedb
β(95%CI) β(95%CI) tionofaclinicaltranslationalcharacterisneeded,includingresearch
onpotentialmechanisms.
Fullsample 0.12(0.07,0.17) 0.12(0.03,0.20)
Sex
Male 0.06(−0.01,0.13) 0.08(−0.02,0.17)
Female 0.18(0.11,0.25) 0.16(0.07,0.26) Acknowledgmentsandfundingsources
ESI
NH-White 0.13(0.07,0.19) 0.13(0.03,0.23)
TheauthorswouldliketothankDr.JamesC.Anthonyforhisvalu-
NH-Black 0.10(0.01,0.19) 0.15(0.04,0.27)
Hispanic −0.04(−0.12,0.04) 0.01(−0.10,0.13) ablesuggestions.ThisworkiscompletedduringOApostdoctoralep-
Smoking idemiologyfellowship,supportedbytheNationalInstituteonDrug
Never 0.08(−0.04,0.20) 0.11(−0.01,0.23) Abuse(T32DA021129)andDr.JamesC.AnthonyNIDASeniorScien-
Former 0.03(−0.09,0.15) 0.02(−0.10,0.14) tistAward(K05DA015799),andbyMichiganStateUniversity.The
Recentlyactive 0.24(0.09,0.39) 0.25(0.13,0.37)
contentisthesoleresponsibilityoftheauthorsanddoesnotneces-
AWLSMVestimatorwasused(probitregressionareestimated).Modelfitforeachofthe sarilyrepresenttheofficialviewsofMichiganStateUniversity,the
threemodels:RMSEAb0.05;CFI≥0.95.
NationalInstituteonDrugAbuse,theNationalInstituteforOccupa-
a Adjustedforageandsex,exceptforstratifiedvariable.
b Adjustedforage,sex,ESI,education,PIR,alcoholdrinking,serumcotinine,totalcho- tionalSafetyandHealth,ortheCentersforDiseaseControlandPre-
lesterolandBMI,exceptforstratifiedvariable. vention.Theauthorsdeclarenoconflictsofinterests.
178 O.Alshaarawyetal./EnvironmentInternational89–90(2016)174–178
AppendixA.Supplementarydata Ullman,J.B.,2006.Structuralequationmodeling:reviewingthebasicsandmovingfor-
ward.J.Pers.Assess.87(1),35–50(Aug,PubMedPMID:16856785.eng).
Gefen,D.,Straub,D.,Boudreau,M.-C.,2000.Structuralequationmodelingandregression:
Supplementarydatatothisarticlecanbefoundonlineathttp://dx. guidelinesforresearchpractice.Commun.Assoc.Inf.Syst.4(1),7.
doi.org/10.1016/j.envint.2016.02.006. Suhr,D.,2006.TheBasicsofStructuralEquationModeling.UniversityofNorthColorado.
Hu,L.,Bentler,P.,1999.Cutoffcriteriaforfitindexesincovariancestructureanalysis:con-
ventionalcriteriaversusnewalternatives.Struct.Equ.Model.6(1),1–55(PubMed
References PMID:WOS:000208063500001.English).
Browne,M.W.,Cudeck,R.,1992.Alternativewaysofassessingmodelfit.Sociol.Methods
Res.21(2),230–258.
Kochanek,K.D.,Murphy,S.L.,Xu,J.,Arias,E.,2014.MortalityintheUnitedStates,2013.
NCHSDataBrief178,1–8(Dec,PubMedPMID:25549183.eng). Théraimauoltn,gGp.Pr.i,mTarreymablluamy,inCu.Gm.,pArromdsutcrtoinogn,wB.oGr.k,e1r9s8.8A.mR.isJ.kInodf.isMcheedm.1ic3h(6e)a,rt65d9is–e6a6s6e
Goff,D.C.,Lloyd-Jones,D.M.,Bennett,G.,Coady,S.,D'Agostino,R.B.,Gibbons,R.,etal.,
(PubMedPMID:3389361.eng).
2014.2013ACC/AHAguidelineontheassessmentofcardiovascularrisk:areportof
Tüchsen,F.,Andersen,O.,Costa,G.,Filakti,H.,Marmot,M.G.,1996.Occupationand
theAmericanCollegeofCardiology/AmericanHeartAssociationTaskForceonPrac-
tice Guidelines. Circulation 129 (25 Suppl. 2), S49–S73 (Jun, PubMed PMID: icsucphaetmioincshaetaprtodteisnetaiaslehinighthreisEku.rAompe.aJ.nInCdo.mMmeudn.i3t0y:(a4)c,o4m0p7a–r4a1t4iv(eOscttu,dPyuobfMoecd-
24222018.eng).
PMID:8892545.eng).
Bhatnagar,A.,2006.Environmentalcardiology:studyingmechanisticlinksbetween
pollutionandheartdisease.Circ.Res.99(7),692–705(Sep,PubMedPMID: Xu,X.,Cook,R.L.,Ilacqua,V.A.,Kan,H.,Talbott,E.O.,Kearney,G.,2010.Studyingassocia-
tionsbetweenurinarymetabolitesofpolycyclicaromatichydrocarbons(PAHs)and
17008598.eng).
cardiovascular diseases in the United States. Sci. Total Environ. 408 (21),
Liu,G.,Niu,Z.,VanNiekerk,D.,Xue,J.,Zheng,L.,2008.Polycyclicaromatichydrocarbons 4943–4948(10/1/).
(PAHs)fromcoalcombustion:emissions,analysis,andtoxicology.Rev.Environ.
Contam.Toxicol.192,1–28(PubMedPMID:18020302.eng). Solerm,Eic.Ph.,eRaurtizd,iVs.eCa.,s2es0:1s0i.mEpiliadreitmieisolaongdydanifdferriesnkcfeasc.toCrusrro.fCcaerrdebiorla.lRisecvh.e6m(i3a)a,n1d38is–c1h4e9-
Ramesh,A.,Walker,S.A.,Hood,D.B.,Guillen,M.D.,Schneider,K.,Weyand,E.H.,2004.Bio-
(Aug,PubMedPMID:21804773.PMCID:PMC2994106.eng).
aInvta.iJl.aTboilxitiycoaln.d23ri(s5k)a,s3s0e1s–sm33e3nt(PoufboMraelldyPinMgIeDst:e1d5p5o1l3y8c3y1cl.iEcpaurobm2a0t0ic4/h1y0d/3ro0c.aernbgo)n.s. Elovaara,E.,Heikkila,P.,Pyy,L.,Mutanen,P.,Riihimaki,V.,1995.Significanceofder-
malandrespiratoryuptakeincreosoteworkers:exposuretopolycyclicaromatic
Simko,P.,2005.Factorsaffectingeliminationofpolycyclicaromatichydrocarbonsfrom
6sm37o–k6e4d7m(Juela,tPufoboMdesdaPnMdIDliq:u1i5d94s5m1o1k9e.Eflpauvbo2ri0n0g5s/.0M6/o1l0..Nenugtr)..FoodRes.49(7), h52yd(r3o)c,a1rb9o6n–s2a0n3d(Murainr,arPyuebxMceredtiPoMnIoDf:17-h7y3d5r3o9x4y.pPyMreCnIeD.:OPccmucp1.1E2n8v1ir8o7n..EMpeudb.
1995/03/01.Eng).
Jeng,H.A.,Pan,C.H.,Diawara,N.,Chang-Chien,G.P.,Lin,W.Y.,Huang,C.T.,etal.,2011.
Castano-Vinyals,G.,D'Errico,A.,Malats,N.,Kogevinas,M.,2004.Biomarkersofexposure
Polycyclicaromatichydrocarbon-inducedoxidativestressandlipidperoxidationin
relationtoimmunologicalalteration.Occup.Environ.Med.68(9),653–658(Sep, topolycyclicaromatichydrocarbonsfromenvironmentalairpollution.Occup.Envi-
ron.Med.61(4),e12(Apr,PubMedPMID:15031403.PMCID:Pmc1740739.Epub
PubMedPMID:21126960.eng).
2004/03/20.Eng).
Curfs,D.M.,Knaapen,A.M.,Pachen,D.M.,Gijbels,M.J.,Lutgens,E.,Smook,M.L.,etal.,2005.
Polycyclicaromatichydrocarbonsinduceaninflammatoryatheroscleroticplaque Curfs,D.M.J.,Lutgens,E.,Gijbels,M.J.J.,Kockx,M.M.,Daemen,M.J.A.P.,vanSchooten,F.J.,
2004.Chronicexposuretothecarcinogeniccompoundbenzo[a]pyreneinduceslarg-
phenotype irrespective of their DNA binding properties. FASEB J. 19 (10),
1290–1292(Aug,PubMedPMID:15939734.Epub2005/06/09.eng). eJ.rPaanthdopl.h1e6n4ot(y1p)i,c1a0ll1y–d1i0ff8er(e1n/t/)a.theroscleroticplaquesinApoE-knockoutmice.Am.
Alshaarawy,O.,Zhu,M.,Ducatman,A.,Conway,B.,Andrew,M.E.,2013.Polycyclicaromat-
ichydrocarbonbiomarkersandserummarkersofinflammation.Apositiveassocia- Penn,A.,Batastini,G.,Soloman,J.,Burns,F.,Albert,R.,1981.Dose-dependentsize
tionthatismoreevidentinmen.Environ.Res.126,98–104(Oct,PubMedPMID: idnimcreetahsyelsbenozf(a)aaonrtthicracleensei.oCnasncfeorlRloews.i4n1g(2c)h,r5o8n8i–c592ex.posure to 7,12-
23972896.Epub2013/08/27.eng).
Pearson,T.A.,Mensah,G.A.,Alexander,R.W.,Anderson,J.L.,Cannon3rd,R.O.,Criqui,M.,et
Brucker,N.,Charão,M.F.,Moro,A.M.,Ferrari,P.,Bubols,G.,Sauer,E.,etal.,2014.Athero- al.,2003.Markersofinflammationandcardiovasculardisease:applicationtoclinical
scleroticprocessintaxidriversoccupationallyexposedtoairpollutionandco-
morbidities.Environ.Res.131(0),31–38(5//). andpublichealthpractice:astatementforhealthcareprofessionalsfromtheCenters
forDiseaseControlandPreventionandtheAmericanHeartAssociation.Circulation
Burstyn,I.,Kromhout,H.,Partanen,T.,Svane,O.,Langård,S.,Ahrens,W.,etal.,2005. 107(3),499–511(Jan28,PubMedPMID:12551878.Epub2003/01/29.Eng).
Polycyclicaromatichydrocarbonsandfatalischemicheartdisease.Epidemiology
16(6),744–750(Nov,PubMedPMID:16222163.eng). GoffJr.,D.C.,Lloyd-Jones,D.M.,Bennett,G.,O'Donnell,C.J.,Coady,S.,Robinson,J.,etal.,
2013.2013ACC/AHAguidelineontheassessmentofcardiovascularrisk:areportof
UnitedStatesCentersforDiseaseControlandPrevention.NationalCenterforHealthSta-
theAmericanCollegeofCardiology/AmericanHeartAssociationTaskForceonPrac-
tistics(NCHS),2010N.NationalHealthandNutritionExaminationSurveyData.U.S.
ticeGuidelines.J.Am.Coll.Cardiol.(Nov12,PubMedPMID:24239921.Epub2013/
DepartmentofHealthandHumanServices,CentersforDiseaseControlandPreven-
11/19.Eng).
tion(Availablefrom:http://wwwn.cdc.gov/nchs/nhanes/search/nhanes09_10.aspx).
Clark,J.D.,Serdar,B.,Lee,D.J.,Arheart,K.,Wilkinson,J.D.,Fleming,L.E.,2012.Exposureto
UnitedStatesCenterforDiseaseControlandPrevention.NationalCenterforHealthSta- polycyclicaromatichydrocarbonsandseruminflammatorymarkersofcardiovascu-
tistics(NCHS),2006a.NationalHealthandNutritionExaminationSurveyLaboratory lardisease.Environ.Res.117,132–137(Aug,PubMedPMID:22626472.PMCID:
Protocol.DepartmentofHealthandHumanServices,CentersforDiseaseControland
PMC3444300.eng).
Prevention,Hyattsville,MD:U.S.(Availablefrom:http://www.cdc.gov/nchs/nhanes/
Ranjbar,M.,Rotondi,M.A.,Ardern,C.I.,Kuk,J.L.,2015.Urinarybiomarkersofpolycyclicar-
nhanes_citation.htm).
omatichydrocarbonsareassociatedwithcardiometabolichealthrisk.PLoSOne10
Buncher,C.R.,Succop,P.A.,Dietrich,K.N.,1991.Structuralequationmodelinginenviron-
mentalriskassessment.Environ.HealthPerspect.90,209–213(Jan,PubMedPMID: (9),e0137536.
2050063.PMCID:PMC1519490.eng).
Laboratory Procedure Manual

Analyte: Monohydroxy-Polycyclic Aromatic
Hydrocarbons (OH-PAHs)

Matrix: Urine

Method: Isotope Dilution Gas
Chromatography/High Resolution Mass
Spectrometry (GC/HRMS)

Method No: 6703.02

05/31/2012
Revised:

as performed by:

Organic Analytical Toxicology Branch
Division of Laboratory Sciences
National Center for Environmental Health

contact:
Andreas Sjodin, Ph.D.
PAH Biomarker Laboratory
Phone: [phone]
Fax: [phone]
Email: [email protected]

Dr. James Pirkle, Director
Division of Laboratory Sciences

Important Information for Users
The Centers for Disease Control and Prevention (CDC) periodically refines these laboratory methods. It is the
responsibility of the user to contact the person listed on the title page of each write-up before using the analytical
method to find out whether any changes have been made and what revisions, if any, have been incorporated.
PAHs in Urine
NHANES 2007-2008

This document details the Lab Protocol for NHANES 2005–2006 data.

A tabular list of the released analytes follows:

Data File Name Variable Name SAS Label
URXPO1 1-napthol (ng/L)
URXPO2 2-napthol
URXPO3 3-fluorene
URXPO4 2-fluorene
URXPO5 3-phenanthrene
PAH_E
URXPO6 1-phenanthrene
URXPO7 2-phenanthrene
URXP10 1-pyrene
URXP17 9-fluorene
1. Clinical Relevance and Summary of Test Principle

a. Clinical Relevance

Polycyclic aromatic hydrocarbons (PAHs) are a class of ubiquitous environmental
contaminants formed during incomplete combustion processes. Many of them have
been identified as suspected human carcinogens. Common routes of occupational
exposure may include work involving diesel fuels and coal tars such as paving and
roofing. Possible environmental exposures include smoking, diet, smog and forest
fires. Threshold levels for carcinogenicity have not been determined for most PAHs.
Application of this method to analyze samples obtained from participants in the
National Health and Nutrition Examination Survey (NHANES) will help determining
the reference range of these chemicals in general U.S. population, aged 6 years and
higher.

b. Test Principle

The specific analytes measured in this method are monohydroxy-PAH (OH-PAH).
The procedure involves enzymatic hydrolysis of urine, extraction, derivatization and
analysis using capillary gas chromatography combined with high resolution mass
spectrometry (GC-HRMS). This method uses isotope dilution with carbon-13
labeled internal standards. Ions from each analyte and each carbon-13 labeled
internal standard are monitored, and the abundances of each ion are measured.
The ratios of these ions are used as criteria for evaluating the data. The analytes
measured in this procedure are shown in Table 1. By evaluating these analytes in
urine, a measurement of the body burden from PAH exposure is obtained.

Table 1. Analytes measured, their parent compounds, and their abbreviations.
No. Parent PAH Metabolite/Analyte Abbreviation
1 Naphthalene 1-hydroxynaphthalene 1-NAP
2 Naphthalene 2-hydroxynaphthalene 2-NAP
3 Fluorene 9-hydroxyfluorene 9-FLU
4 Fluorene 2-hydroxyfluorene 2-FLU
5 Fluorene 3-hydroxyfluorene 3-FLU
6 Phenanthrene 1-hydroxyphenanthrene 1-PHE
7 Phenanthrene 2-hydroxyphenanthrene 2-PHE
8 Phenanthrene 3-hydroxyphenanthrene 3-PHE
9 Phenanthrene 4-hydroxyphenanthrene 4-PHE
10 Pyrene 1-hydroxypyrene 1-PYR

2. Safety Precautions
a. Reagent toxicity or carcinogenicity

Some of the reagents necessary to perform this procedure are toxic. Special care
must be taken to avoid inhalation or dermal exposure to the reagents necessary to
carry out the procedure.

b. Radioactive hazards

There are no radioactive hazards associated with this procedure.

c. Microbiological hazards

Although urine is generally regarded as less infectious than serum, the possibility of
being exposed to various microbiological hazards exists. Appropriate measures
must be taken to avoid any direct contact with the specimen (See Section 2.e.). CDC
recommends a Hepatitis B vaccination series and a baseline test for health care and
laboratory workers who are exposed to human fluids and tissues. Observe
Universal Precautions.

d. Mechanical hazards

There are only minimal mechanical hazards when performing this procedure using
standard safety practices. Laboratory analysts must read and follow the
manufacturer’s information regarding safe operation of the equipment. Avoid direct
contact with the mechanical and electronic components of the mass spectrometer
unless all power to the instrument is off. Generally, mechanical and electronic
maintenance and repair must only be performed by qualified technicians. The
autosampler and the mass spectrometer contain a number of areas which are hot
enough to cause burns. Precautions must be used when working in these areas.

e. Protective equipment

Standard safety precautions must be followed when performing this procedure,
including the use of a lab coat/disposable gown, safety glasses, appropriate gloves,
and chemical fume hood. Refer to the laboratory Chemical Hygiene Plan and CDC
Division of Laboratory Sciences safety policies and procedures for details related to
specific activities, reagents, or agents.

f. Training

Formal training in the use of a high resolution mass spectrometer is necessary.
Users are required to read the operation manuals and must demonstrate safe
techniques in performing the method. Anyone involved in sample preparation must
be trained in for all sample preparation equipment, chemical handling, and have
basic chemistry laboratory skills.

g. Personal hygiene

Follow Universal Precautions. Care must be taken when handling chemicals or any
biological specimen. Routine use of gloves and proper hand washing must be