Document oxp17qR8d2e66Vx2ERy5RMq7

722 FREES MMoortrtaalliittyy ooff eemmppllooyyeeeess ooff aa ppeerrtf:lluuoorrooooccttaanneessuullpphhoonnyyll fflluuoorriiddee mmaannuuffaaccttuurriinngg ffaacciilliittyy 8HAlexander, G W Olen, J MBurris, J HMondel,J Mandel B H Alexander, G W Olsen, J M Burris, l H Mandel, J S Mandel xatd 0180722729 Occup Environ Med 2003;60:722-729 Ssdspthle AeEip iFposOosbSkimToneeb gAcvedeoobrseknoioTg1ro4r t3thed:cTcyr lFofOeso]r) ncsoelid nbloeyeeltoofbicyao ydacy phhloowmes oedanecalpoon prerssomfrcet wapnipfihooobrtde,roSsoaattcdcaonPeshShepeohheYfoonc!eyacxovrpideod:se See end of article for IRINLT ryvate esargos nhtporedd e eTd orle rlyTroeforksLcnhoonrd fe au~ors' affiliations Aim: fo evaluate the mortality experience of a cohort of employees of a perfluorooctanesulphonyl fluoride (POSF) based fluorochemlcal production facility. Melhods: A retrospective cohort mortalil7 study followed all workers with at least one year of cumulative employment at the facility. The jobs held by cohort members were assined to one of three exposure subgroups; high exposed, low exposed, and non-exposed, based on biological monitoring data for perfJuorooctane sulphonate (PFOS). Results: A total of 145 deaths were identified in the 2083 cohort members. Sixty five deaths occurred among workers ever employed in high exposed jobs. The overall mortality rates for the cohort and the Correspondence to: Dr B H Alexander BTTR University of Minnesota, TRL School of Public Heallh, Eres Division of Environmental IRE, and Occupational Health, MMC 807 Mayo Building, 4.20 Delaware St, SE LEO Sun, Minneapolis, MN 55409, BERTRAM USA; balex@umn.edu jo mir A,ccepted 23 December 2002 exposure subcohorts were lower than expected in ihe general population Two deaths }ram liver cancer were observed in the ~orkers with at least one year of high or low exposure (standardised mortality ratio Son 00 v0 Gla be 110 Tw ok odh h Fels Cres vo re] ce (SMR) 3.08, 95% CI 0.37 to 11.10). The risk of death from blr~dder cancer was increased for the enti~e Ee rr cohort (three observed, SMR 4.81, 95% CI 0.99 to !4.06). All three bladder cancers occurred among Solon 0 blk opsi SUR 127.91256371 oti oscov workers who held a high exposure job (SMR 12.77, 95% CI 263 to 37.35). The bladder cancer cases poh ote eeriotoao oh. kab monk, rd Howoh ond wana primarily worked in non-production jobs, including mainlenance and incinerator and wastewater treatment plant oerations. Continwok colors nigh expen bshdan creerum of dock omBod Conclusion: Workers employed in high exposure jobs had an increased number of deaths ['ram bladder Cte hover onowheat i oe coscnbs cbt forchamcl crporae, cancer: however it is not clear whether these three cases can be attributed tc fluorochemical exposure, an inonTdopbeavnienegrcbrsdShaid gcoeabosoriLoyf oro hnrcsWskor.cSt/oo eTodn unknown bladder carcinogen encountered during the course of maintenance work, and/or non- occupational exposures. With only three observed cases the possibility of a chance finding cannot be ruled out. red rink canons ne tec wed for el dad. be co 10 fect my 0 1LIOritlalcd argMlJc C0lllpOLlllds have bt'ell used fin FFea: onTMot brio, ander Tomb TERE dccadc~ in a wide ~ariety of industrial and commercial A obama og Roel he qos of FEOS. in ton upton applications, such as specialty Iubricanls, semiee eon mambo coda cx ot nd Wile, wt ts md canduclo[ nlanu[a(luring, protectixrc barriers ar coatings, tl WORT CELSO Be pcan BES rl co Shot te come surfactants, fiw retardanls, and non-co[tdtlclive coalallts. ee Onl Chm bons lcs VO 1 dau nc of 1103 1 bos Pcd]uorooctancsulphonyl fluoride (POSF, CsF~TSOzF) is the a hr oe it Pn i Shes Sr Spat 0 Ilo(he sa lic lei (NOEL) prccursar for oue line of these specialty chemicals [Ning POSF as the basic building block, unique d]emisates ee created by deriver[sing POSF tlxough tl~e sulphonyl moiety using conventional hydrocarbon reactions to make amides, Comm esha mit ofchhe n1l sh0o]esB3Y5 e0scnhosvewrienied he hlseeTouosoef nhearmiecloor oxazolidones, sdanes, carboxylates, and alkoxylates as commercial products, magi of whirl] are ill the lOflll at high n S r he n To so HaEeRtn2a FOS dpb ap TceesCogan lliODctllal weight pol),l]tcrs. The majcu application5 arc not full~ understood, it may be cue to an effect an fall}' acid transport and llletatlollSlll, iltembratle functiOll, peroxiso[lle proliferation, and/or mitochondrial bioenergedcs)-: Recemly lhc presence of PEOS in nowoccupationally exposed popalalions and wildlife, marine mammals, and piscivolocis birds raised concerns about the en~ironmelllal el[eels ol t'FOS,' "~ ' lhe background levels of FFOS azc low and appear to bc bchm the no observable effect level (NOEL) in loxicological studies; hmvever, PFOS is now recognised a pervasive compouud that will persist in the cnvironmcm and can accumulate m wildIife. These environmenlal centerns have prompted tl~e phase-out of the production of ?OSF based chclnical~ b? the major producer (3M Company). surface treatments, paper and packaging proteclams, and performance chemicals.' Depm~ding on die degree of poly- a Shee oer oS mcrisation, stuh POSF based compocmds, or their residuals, EAR en ih ROA. is mi on wah su FEOS oh A can degrade or n]clabolisc to pcdluowoctanesutphonatc TO Ca 0 a unbtmes i Th FOS nals yo rer en om ln pr (PFOS, CgFITSO$ ) to an undetermined degree. The POSF AihemeeS ens adoy I LE Se age ewan wos et based fluorochelnicals tiave been produced almost exdusivoly a~ ane facil[ty m the Uaited States and mnc m Belgium. Stn Foor iol plese hes 0s wing aki ah Sibu, wnpabished wep 1905 Studies of laboratow animals indicate d~at the liver is the primary target organ tar Pt:DS.' [n rats and cynomolgus A tS md aA ly glSb et p~i~nates, high doses of PFOS induced enlargement of liver Sr crs i al ct TL Sn 08 a To i cd ie ,ind npparcnl ahcrations in illetabolic proces%'s, iedtacd serum dmlcstcrol levels) ' PFOS w,l~ not found to bc pea a ss bi a! a dcvclopmcniM toxicant m rats or rabbits ~ Higher maleraal doses el PFOS increased neonatal lnortalJt}fl absmpt~ons, ommin atdem T rc~orpdons, and reduced weight ~am in rat pups ~ There were ThTE SY ty ct i) tn Ilo effects nn lmslnatai neuralogJcal develo[)inell[ or on etme ebint we eapmoaio, Smtr btph Ke lc~tility and ocstrous cycling in offspring in multigeneralion kl i nn SO mt Lh A eTLT ~tudies. Mahiple genotnxicity assays indicate PFOS does prescnI a hazanl fl(llll intmactiaa with genetic material.' en he tnoso ss i SnSm on, rd ly 0 Ald~otigh Ihc mechaldSln of loxiciiy iN laboratc)~y animals Exposure to higher concenlrations of these occurs in employees of tl~e manulacturing fatalities) Medical surveillance ef workers has not associated abnmmal clinical che~nistry rcstdts with serum PFOS levels.*' A martalit? study of workcrs employed at one plant reporWd an overall lower all cause and cancer cause specitic mortalily rate than expecled, thongh the population was wlatively young (Mandel and Johnson, unpublished report, 1995). Hereto wc report the results o[ an update of that cohort mortality study, with specific emphasis on exposures leading to increased serum PFOS levels Thia sludy pFolucol was &bbrevi,alion:,: el, confidence interval; ICD International Class f ca of Disease; NDI, National Death Index; N-EtFOSE, Nethylperlluorooctane sulphonamide alcohol; PFOA, perfluorooctaqoic add" PFOS, perfluorooclane sulphonate; POSF, perfl'uorooctanesulphonyl fluoride; SMR, standardised mortality ratio Ex3h1i4b5it 33114455..00000011 - Mortali'y of fh~orochemical workers 723 + bacon)f0e ) od C e+ TE a-- rC fnp podsT sned duPcetnsqucooroncbtaen.esulphonyt.pfluoeridet(PaOSFl} rbeas2ed0kpro ducts can Be metabolised to pe~uoroodane sulphoee en SR LT hate (PFOS), to an undetermined degree, which is Boooctrumeitoe rae vy ofpespoi se. FOS,swaicnk od o AONn etn i~ieorasciscteunmtualantde paervvaasriievtey inofthsepeecnievsir.onPmFOenSt, aandwecaank re i bien eroxisome proliferator, targets the liver, resulting in F~ oo cigar ord Shiono mac 5RE ee eoatic enlargement, and alteration of metabolic process,mot bololy redchion in sen ho uy act ad kos fdy co on processes, most notably a reduction in serum cholesterol_ Fahoome LoL HL EC The cohort had a standardised mortality ratio (SMR) for bladerconor hoswo sigiay nce,which 1 he on 3 rs 1d 1) at bladder cancer that was significantlyincreased, which ITinnhPeMcaoymF2pO0a0n0yitthhoaatt ifptrweodanushcpeahstatishinensggeo opurtoode uf cthtse aprnrnoooduutcnitcioend of POSF based materials because of the persistence and pervasiveness of POSF in the environment. wcrc ab'tractcd into all electronic databa:~c, rccordiilg lilt worker's ~amc, s(iciat ~cCtlril~ nlllnber, employee idcntitic<> liui] l/llill[)C[, ddtc OI birtlL alld the elates ol filly el]try i111 work hislory re'cord, includin~ lavofl~ and lea~c~ of <lbscilce, the lob departmenl codes, and job title. To be eligible inclusion ill the cohort, a worker i/ad to ac'crut' al [e~lM was based on three deaths. PFOS, a final metabolite of POSF based chemistD,, is not genotoxic, is soluble in Erm urine at the concentrations present in employees and failed to produce urinary tract tumours of i'nflammation 1yoo othgard seror ploSy, r~ two year rat studies ~nd a six month primate study. i()$ class ol ctimtJlalive ciiq)loyn~cill a[ lhe site by Dctcmbcr 1997. Periods of absence duc 1o illness, nlilitai)' I TL Icavo, matornil) lear<-, oi lay~dt did 11o1 conllibule I~ even 190 th It doeof cAI l)c'cember 1997 as their last date of elnpk)ynient. -[here were no significant associations between ey employment atthe fluorochemical manufacturing plant dnd n]ortality'frc~ J~r di~e~se::(~:i~]]o; ~:ar~cer), Shcbe r edie e do although the nurnbei" of observed and expected deaths were few. + Oher ropuin] pases ad perl bobs Olher occupational exposures and personal habits nd ny were not examined in this study. The newly c'iltllllCl<llcd cohort wa~ linked to rccoid~ IFOIll the original cohort to update empl(iyment information, and = to verify ilalilo$, sodal security numbers, dates of birlh, and dates of dcath for prcviousl) idcmificd decedents. 1% 11 SD, Discrepancies idl'lii ified in ll~c i ecoi ds ~crc ~ csolt cd through c~c'dit ~cpoifing dgcncics, and the Social Sc'curit) a en Adlninistr.lt{on service !or epidenfiological res&~ich ee he do ea sot fo lhe latter reports rite [11o5[ recent account acIi~il) of all individual and whether they arc rccoidcd as deceased rcvicwcd and approxcd by the UnivcrsiW of Mti/[to<~ota TT Ts 8 Ut tluman Subjects Research CommiHee. werioos METHODS Study population SR wr Tlic siinty pol)ulath)n worked <It a nlanufa(luring fadlity Social Security Death Index. yop ond dermis abi pe Follow up and determination of vital status FIi!;ible coll~lri nll'lilbel~ were lolhlwcd from tb.c da) the'7 rn, nn tng [)cccnibc] 1998 or their date of dead> Vital rodold> Lat Eu were ix'rformc'd tor all eligible cohort lllelllbers 1oi t~ll(illl i I)ctdlUl, Aldbdma, that began production in 1901 aild death ceriilicate was noc obtaincd iH the o]igmal stud} consi~t~ of two l~]anl~: ~twcffihy IJlin (film plant) ])elerlllinaHon o[ vita[ Sl{lltb WLI> Hiaclc by $cd~diiHg thc 'qK'cialty n/alerials (che'.nical plani )Tttc film plant is lot'a[0d al)llroximaicly 300 ~ards wcbl of tile chcmical plalll ,lilt[ ploducc~ a valic[~ ol polycstci anti ilOll l)olycstci l'illltS Fhc lhlcc inajol pioducl grutlp~ (lt'{Ollc'd [0 ,1~ Ioctl~ factories) Ee a the dwmica] plant are proi0cmc c:hcmical~, pcilorli;aucc d/clnical$, olld []tlt)loc'/aMoli/c:l> TliclC" ale illorc Ilion 700 men A RA diflcmn[ protegee% ~xith more Lh<tll 90qg being batd~ Nalion<d Dcattl Index (NI)I) for all worker,~ m the original study and netx workers incltlded in tile cohort lhe Sicial Scctlrily AdnlilliMraIiotl dala and or the Social Sco.llilx IYcath Iitdcx (SSl)I! ~xc~c searched to verify the xitdl Malt> wae shea par Ot ~Olkt'lS WIIo [crlnilidtcd Clllp[o} utcllt bel7~ic 1070 The record~ of coholt lllcqI/bors idenlified a~ ier eye pe i]lrough the NDI or SSDI were n/antialI) reviewed proce~cs. Eaw malerials and intcllncdiaies [or each product xdlid n/arch, and a cop} of the death certificate wa> requested group alay flow th~oLlgh many difJerem p[oduction buildings lronl the stale ol record. The death ccrtificales were coded by bcforc packaging and shipment P~rfluorooctanesulphonyi a licensed nosologist tlsing the rules ior the [nlerllational fluoride (POSt, C~F~:SO2F) is the major sulphonaled Classification ~f Disease ( ICI)I version in eflecc at a ins, ne a ey fluorodlemica[ i/lal/ufacturcd, and is produced via all electrochemical cell piocc'$~. B{ological IIlOIlJLillhlg nlcasurcmcnl) o1" PFO~ has been used io assess the employ- dcalh, ~xEich allo~xcd {he ~sc of actual tunadju~cd) mortality reterence data. Decedent~ identified through NI)I lot ~vhich a death certificaw could iloi be obtained, or if Llle ees" exposure experience to POgP [}aged fluorochemicals cause of death could ~ot be coded, were included in the btu thot Enip]oyt:c~ max be exposed b) one ol all inhalation, skin cont,qcWabsorption, a~ad ingestion doe ml loutc~ to analysis inthc"othcr' cause otdcath category. POgF based tluorochcmical materials. ]lie primary route eXl)O~urc nlay be dilferent tor each employee and dcpcnds on nt Sm fr several different Licit)is, such as process conditions, job tasks, work location, per>onal hygiene, personal habits, and gellerd[ work practices. Aifl~ough POSF ba~ed chemicals are the plilllaly []tlOrOttlelllic3iS produced al tilts plant, cxpogt~rc ocher fhloroclici~lical~, inchiding pcrfluorohcxancsull)h~)i@ lhlorl>chenlical~, a~c' likely FEE rt it pk on Cohort enumeration le~,mt~ nbtaincd [toni Iw pl,lni The record, of all employcc~ "the relative differences in serum PFOS bs job id J,Ss, ih ar, pos 1998.'" lIl Ibis study, a total of 232 culployees were randomly ~clcclcd ior scrum >ainpling ~{lh 80% pallidpaiion tn 12o cheiTfical plant; 60 l'{hll plant). ~oitltl; /Vt'l't" eXH,/c't0d tl~iIlg at1 toe pairin~ extraction prcx 0dtliC and ~crc quanlilalivcly andl]'scd lot ,cium EE 1'I;O5 u~ing high a ERLE Io I.I I; lot Iihn planl culplo)ecs 11 v,</> 0 1 pl)ni t'a~",, C[ 0 I ~o 0.1L Tho majori~) of film plant job~ ha~ workpla(e dxpoMlre la fluoroct~emkals. -lhcii een 33114455..00000022 [r---- 724 Alexander, Olsen, Burris, et nl Conon goat i a in, Ch Ta cxl~osruc in ptoxmtitx, tu the ,.heroical plaut. Chemical plant wee afd io SE a Gh sn, jobs wctc classified into eight categories: ccll operators, Chemo open. maa modes (oh chenlicn] operators, lUaillle~ance workers { prilllaril} has aT chk, Ton, RC aR mccham~s and electricians), mill operators, waste tream~en{ aoisgu,ioanndpanmminiceumTanmai . The Iso. managers, and administrative assistants. "lhe highcsl gcu t~sing standa~vl lilt table methods vdth the PC Life Table ok Sor CHEB Sot nd Be Analysis Sysl('nl (t'CI TAS) software developed by the Sal oa of Oia Sir mt Holly Natianal [nstitntc> of Occupational Safely and Health NOW TH wk of Un ry ws wn cab om (NI()SHL~ The race of the wurkers was not available lrom HY Po AR 1 Ps ce the colnpaIly records, bnt the populalion of p[aill employees A ebro ts plcdolnJnantl} Cau:asJan; therefore the relcrencc t-atc~ for Coa etd ot th nbs (;aucasians were used for this aualysis. illCtliC inca{l level uf serum PFOS was observed in cell Operdors (20 go. followed by he wade operators he emir coor and the suber of cv #51 pou operalurs (2 0 ppoll. Iollowed by the waste operators UT mn ra (L310, ml crams Sc oe ooat, AToveaPond. 9 al (1.5 ppml, cl:cmical operators (l.5 ppm), and main~c~aucc ork 13 omy Sapa mogs (D477. I mak wa ond a Coote EA Ah workers (1 3 ppm) Supervisors/managers (09 ppm), m~ll rin (04pm, on WTR U0 S50 Whe he xn ABORTED operatars (0.~, ppmL eogincer!Iab workers (04 ppm), and mine ie 11 PR Tec Jom hd am 0 can oS, sh administralive assistanls (0.4 plnnl had lower geometric asco THOS eh, MAL Cas acs oy oc Dt expr, ot har ped of on mean serum I'FOS levels. Mill opma~o~s ale often entry level on tan ian toch STH HRS SEAR Whey ada ach eo ol positions that usoally progress to chemical operators within a emore mun Ter coving eons an Gn 1 few }'ears of hire. dpi, cx toot roche. RLU: 4s RIOT opt we id As mcntioued prcviausl~, exposure m other /luorochet11J A Te gg icals, including PFOA, are pussible. Until 1998, PFOA was not macnn sco, oe, FOR Roar WE Apt re snail by kaif plat manufactured at this lacility. Rsther, PFOA exposures were io te rode 8 Bec She es pov pop Te os of tn of5 pro the result o[ it beiug produced as a by-product of the ame cl vc of 1 ne 2 cr Bn cis ms ant Ft Sc en electrolytic cell production or in its use as an emulsilier in poner oda Athol ODS gy I cane and os ih, Bw i sr, FEO flumopolymc~ prodnction. Although observed at slightiy Tomer em PEON To coin WE Sho ts gh cnt rion on lower levels, the serum PFOA levels correlated ~vith {heir Som 08 rae in he Tn he oefdsoh ho spr se~um PFOS levels ons pee press as ened usta ncn cos ons oF moofsri. Arch eed Because prodtlct~ou processes have renlained co~stant over ne Soe xa mate dob os ups 2h od by do pont time, a simple cxposmc matrix was developed based on the rk Boor cc of he ay cs WAR he bod work hi~tory records ot the study cohort With the knawlcdgc Se ma lc sc ROS Ih oman ot the major job specific serum PFOS levels, a company SE TI IO 2SI RGIS tote ten industrial hygienist and cpidcmiologist assigned each umquc Io am Sep cons hk ey st 4% ke, Moni, oaof083 soresss jub and dcpamncnt coznbinatian in the work history ~eco~ds neofsh loa th fs Epa cine oo rele traOfthew: 7% 03)work to due of tl~e follo~ving three major exl~sure catego~ics: 0Fkugsspun POST ad Frchencs IOC 0 med" ugh Gabi") No workplace exposure to POSF based fluorochcmicals PgnPu ) ost ed ne ADo eeronME 13 45013i 819 wiegdh cnonr Cexpponadi$c1 (encorupasscs tilnr plant jobsl. ov ove nb ep id ome i comida al Low potential workplace exposure to POSF based lluora hem nddi mgrly, comedy ml Te ih apo sro on chemicals {includes such jobs as engineers, quality comrol technician% enviromnental heahh, and safety workers. adminisnati,cc asgis~ants, and managers). i po tpi h cope POST ad Ste he Sk he ch Iiigh potential workplace cxpo~mc to POSF onrchimcal icudes cl apts. cheical O13. (pica suo, Tht wet 15deaths Aad fluarochenficals all,eludes cell operators, chemical os. rc RTs UR as WO ts oon 43 eth hh pon Fo. 37 [ors. illabllcnalic~~ warkers, lni]] operators, waste tt npr ow tre frp 31 mocap i, Dh tors, and crow supervisors) based opera opela- etic thes re srkw,i kd as tenon. Sole Shed and coded for 13) (oy of th Hereafter these three c,~tegories xvill be relerred Io as the non- Ee rr cxpo:,cd, lou exposed, aud high exposed subcohorts, rcspcc- Tin. Al RC SFIS were mae wiht Ks Sy ot oneness lively. All expasure assignments were made without knowl- CIR uf monn cms Ax sdb canon 0% POR Ht (0 edge of tire murtality outcomes. An additional classification min cure nnd the names tow pc SSI of onc for cumulative exposure assigned the non-exposed, low ne a Wa cpm pt ie Boos wb Ter etc 742ues who woke ninofa cxposurc, and high exposure jobs relative laOSF based iob one vs of 1, nd 10 ee, dod on t yrik nhrbo cxpore on Te a exposure value of 1, 3, and lO, respectively, based on man da Wh Yas ems oh oh wis mare er = highs bt cxporsob.Theater biomonitoring data lhe years spent m each job were ued bt he ai weighs to dp3 quasine 4 IL AE multiplied by the relative weights to develop a qnantitativc RU UF COA Gs CE Ld er i i, dee wt ent exposure urctric for examining dose-response rclatians. Anais ESigonon wdaee hor wmaert in 1ighCoombrinse gohb.a0d an Analysis The mortality expcricnce of the cohort was compared to that Sh ne of Nona An addon in oon To a come an ce spec molly ics for the ol the state of Mabama. An additional relerence population 2 ea ue We Sd kh Een hn et es him xpd om te using 23 regional counties was also used to check the resuhs St a aan we Trh ee en plat of ba 185 snd nd 3a attd rule out large variations within the state. The reference a ih ardete fo Nal Cone or Heh nied SNE 03,591053001370 420A smi data, which are derived from National Center for Health So om fh iy an atch ta al nd qtati,+tics dala, +yore obtaiucd front the Mortality Populatiot~ Data Syslem (MPDS) centre a{ the University of Pittslmrgh. The Efe dtd ae ge (in year) Fk, Te aE aud fe cnc of dea ce fone a Expiod fo These rcfercucc data arc age {five ycar), gender, race, and i ed (ne eo) ek nd col i heh ra a J xp (OA ~,dend~r periled tfive ye;u) specific, and are coded usiug the orn oC to i to hah Prd nal 0) Shoe Wh CORON 16 TAT talcs t'o~ the ICD version in effect for the calendar period nd md aed vod At SaRndd Who cl oe ht 63 of eT nth Hh Age, geuder, and calendar period adjusted standardised a a {SRD wet Compute or 1 ne amd xp ale 6 4d ME ow oRne ASD (1 ntartality rattus {SMRI were conlputed lot all cause arrd EE La f eh on he tat et dna Sa show. Sama ota tes a hee specific causes of death using the Alabama reference data. TSR Sd 90 Condit Ea re comcl Gath 0 1 bp: of es eh be lhe SMRs and 95% confidence intervals were computed The all cause and cause specific SMRs wetc~omputcd fo~ Ihe entire cuhort and the subcohorls ol ever ~igh ezpogure. ever 1o~$ cxp{)sufc, and non-exposed. This iltilial stln/lnary analysis was followed by an exposure specific analysis in ~vh~ch the exposed gubcohorts were restricled to worker5 wt~o had at least one year of cnmulative employment in jobs with high or low exposure. For this analysis the period of Iolluw up bcgan whcn the individual reached onc ycal of cumulative exposure, ~hereby conn[ing person-time and deaths thai occurred after the nlini111~lll exposure was accrued Causes of death potentially related to Raorochemical exposure were analysed by duration o[ employmenl in the three exposure subgroups. The causes of death of a priori interest were cancers of the digestive system, in parlicular li~vr cancer and liver cirrhosis, because in animals, EFOS drcula[es through the enterohepatic circulation and con~cnIrates in the liver.* '~ OIher causes of death thai appean?d to be in exce<s ia one or more of the fluorochemical exposed groups wine also c~ aluatcd by duration of exposure. RESULTS Of the 3512 p,,o~kers identified, a total of 2083 workers met the one }'ear enrohnent criteria. Of these, 47% (982) worked at son'~e time in jobs where exposure to POSF based fluorochemicals was considered high (tabIc Apploximatcl} 14% t289) worked Ln low cxposmc a~cas, but never held a job in the high exposure areas, and 812 139%~ wine considmed to have no or minimal wolkplacc exposure Io fiuorochemicals The high exposure group was modestly yourtger than the other workers, but had a longci aterage tenure at the plant than the remair'tder of tire cohort. Male workers made up 83% ol the cohort and 84% of the higlt exposure ~ubcohor~. There were 145 deaths identified in the cohort, with 65 deaths in the high exposure group, 27 in the low exposnre group, and 53 in the non-exposed group. Dcatb ccrtilicales were obtained and coded for 139 (90%1 ol the dcccdcut>. The six lnlSsing death certificates wcm iu the high exposed group (n - ~}, the low exposure group (n = 2), and the non-exposed group (n - 1). The cohort accrued 50 972 person },ears of follow up. There were 782 workers who worked a ntinitr~um o[ one year in high exposure jobs and 1065 workers who worked a minin:tum of or'~e year i~'~ high or low exposure jobs. The latter subcohort included ',~ orkers who held high and low exposure jobs for tess than one year, but had a combined exposure period of one year. Fifty three deaths ,,',,ere antony those working one year or more in high exposure jobs, and au additional 29 deaths were in the combined high and low exposure group. The all cause and cause specific mortality rates for the entire cohort "~,'ere louver than expected compared to tbc general population of Alabama; 145 observed and 27,0 expected (SMR 0.63, 95% CI 0.53 to O741 (~able 2i. A similar pattern was observed for all deaths from cancer; 39 observed, 54 expected (SMR 0.72. 95% CI O.51 to 0.98). The all caus,' and all cancel cause', i3l death were lewcl ~[lalt expected hu the high expusure {table 3 ), low expost~re /table 4 ), and exposed {table 5) subcuhorrs. When restricted to worker~ who accrued at least one year of en~f~loyment in the higb exposure (table 6), and high or low exposure sub.~rolB>; (data not shov,'n L Ihe standardised mortality ratios lot all causes death and all malignant neoplasms were well below unity. www.occenvmedcom 33114455..00000033 Mortality ot fluorochemical workers 725 . [451G tne pt 1 et Table 1 Characterislics of eligible coho~ members by PFOS exlaosure category Totat 982 289 812 2083 M (%) | e& n gL BB F (%) Age at ~Io,~ up [EL Sa fe fe Be Median Inlerquarlile range [EET ew we He | Year ef birtfi Median ee Bas Wo, Hw Uo | I,~terquartile range Year~ of employment ET fa Mw Ta He /,edian I~terquartile range Per~on-}'ears of follow up | BEST a Bae Pa Bw Median I~terqua~ile range ou & > 5 i Dee~hs 826 (84) 156 49.7 43.9-56.8 1948 1941-54 16.7 4.5-26 1 24.8 19.4"31 ~1 65 204 (71) 85 52.9 46.1 --60.2 1945 ! 937-52 10.4 2.7-22.7 26.3 20.8 31.9 27 T Er --T "Ever employed in high exl:x~sure iob. EEE tEv~" employed i, low ex~uro jo!~% ~ never a high exposure job. $qNo w minimal wcrkplac~ {~uorochemical e,'q:~sure. 700 [861 I 12 51 4 46.4-56.9 1947 t 941 "52 9.9 2.7-25 2 26.7 21.4-30.1 53 1730 (83) 353 50.9 45.1,57. 1947 1941 "53 13.2 3.4-25.3 25.9 20.4-30.8 145 Five deaths from cirrhosis ot liver were identified it: the conducled for the deaths flonl liver cancer. Both eases wcre entire coho[t, t~vo of vdfich occurred it. thc high cxposure group; however Ibis did no[ exceed lhe number expected confirIned to be prhnary hepatocellular carcinomas Tt~ree deaths were anributed to malignant neoplasms {2.5). Two dead, s from liver (ahem were obsm~cd in the the bladder {0.62 expected in the entire cohort, SMR 4.81, entire cohort, with 1.24 expected (SMR 1.61, 95% CI 0.20 to 93% C[ to 0.99 to 14.05). Tire workers who died from bladder 5.82). OIlc livi cancer case was observed in the subcohort employed in a high exposure jobs (0A expected, table 6L and cancel were in the sub cohort that worked in jobs with high exposure for at least one year (0.19 expected SMR 16.12, 95% the other held a ]o~x exposure job for at least one year The CI 3.32 to 47.41, table 6). All three cases of bladdc~ cancer SMR lot liver cancm a~nong workers who held high o~ 1o~ were male and each had worked in high exposure jobs for exposure jobs for at least one year was k08 (95% C[ 0.)7 to least live years; the SMR for five or more years of employ- 11.10). Thc worker employed in the high ex[~sure job meat in a high exposure job was 25.5 (3 observed, 0.12 worked at the [acility for just over one year. The worker expectedi with only three cases of bladdc~ cancer, all with employed in the lower exposure jobs worked for approxi- cxpericnce in high exposure jobs, an exposure response ma[ely 10 years bc[o[c ~ctiring. Medical validation b} follow~ analysis usiug the weighted cumulative exposure metric back to the physician of record on the death certificate was not informative The results for bladder cancer hi rclatinn i Comp ai hyne h rd Table 2 Cause specific deaths and standardised mortality ratios for all causes and Co sticked causes of death for all cohort members T--w oe RR R eG n All dea~s TFoegmome. T3 m% Bow OsWwE All malignan~ neoplasms Digestive organs a~a peritoneum I EE om ~us ~r~ in,fine eg i 2288 Bilia~ ~ and liver R~irat~ ~m iad io oE Hn ~us, Wa~, lung Br~ Em Po oi BH gfi~ 0r~ns ~ an~ o~r u~ Bee. 1 BOE OEE .~g~~ma SE L~ic an~ h~iefic CrEaryE 3PE o@ op oEme ~v~u~r~se All h~d ~s~ N~-mal~nant respir~ow di~ = Ie 3i BBhEE SEPiN Cir~sis ~ liv~ rn psf HEHE All acci~ re BOE oH EE ~tor vehicle acciden~ 2 ious HEE All 0~ ~idents Viden~ = : UE OEE Sdcides H~icides mii *Cause nol listed if not observed. 39 54.26 0,72 0.51 to 0.gg 5 9.78 0.51 0.17Io 1,19 2 1.15 ].76 012J ~ 635 1 3.35 0.30 0.01 ~ 1.~ 2 1.24 1.61 0,~ ~ 5.82 15 21.01 0.71 0.~ b 1:18 15 ~.33 0.74 0.4t ~'I.22 2 1.28 1.57 O.19 ~ 5,~ 3 1.89 1.59 0,33 b 4.65 3 0.62 4.B1 .O.~b 14.~ 3 1 .~ 1.67 0.34 ~4.88 4 5.68 0.70 Q.I 9 ~ 1,80 5 7.13 0.70 0.23 b 1,64 35 62.53 O.~ 0.39 ~ 0.78 1 10.6 0.09 O.~ b 0.52 5 5.g9 0.85 0.27 to 1 98 27 36.65 0.74 0.49 b 1.07 15 22.05 0.68 0.38~ 1.12 12 14.61 0.82 0.42 to I ,~ 10 19.27 0.52 0.25 b 0.95 5 11.42 0.~ 0.14 ~ 1.02 5 7.85 0.~ 0.21 ~ 1.49 ---- 33114455..00000044 726 Alexander, Olsen, Burris, el al relmCosne phfbbansserorroysctrltdc d 1 Table 3 Cause specific deaths and standardised mortality ratios for all causes and selected causes of death for cohort members ever employed in a high exposure job | an Cause" | = Ee A~I deaths Fobra Your wom Cancers ESE m.VOR g OBEER All malignant neoplmms Digestive organs and peritoneum Ee I] Oesophagus Biliary passages and livor ERE PoE EE Respiratory system Bronchus, trachea, lung Urinary organs Beno 1 HF SEE Bladder and other urinary organs EBeEam H HIRE Malignant rnelanoraa Lymphatic and haematapaietic Non-malignant causes Cerebcovascular disease Si in R i E4 FEE All h~,art disease a POE wim Cirrhosis of liver AJI accident~ He Tou iEE Motor vdaicle accidents All other accidents &= POE SEE Violence = Poh BEE Suicides Homicides O/osecved 65 18 2 1 1 7 7 3 3 2 1 2 14 2 17 9 8 3 1 2 Expe~'d 93 56 ~ .5MR 0.69 2~54 3.91 0.46 0.50 824 7.97 0.75 0 23 0.76 2.31 0.84 0.51 2.16 2.00 0.85 0.88 4.02 12.77 2.62 0.~-3 276 24.78 246 15.87 9.60 6.26 8.38 4.97 3A2 0.72 0.56 0.81 1.07 0.94 1.28 0.36 0.20 0.59 95% CI 054 to 0.89 050 to 1.32 006 to 1.85 0.05 to 12.02 O0S to | 1.10 0.34 to 1.75 0.35 to 1.81 0.83 to 11 2.63 to 37.35 0.32 to 9.46 0.01 to 2-40 0 09 to 2.62 0.31 to 0.95 0.10 to 2.94 0.62 to 1.72 0.43 to 1.78 0 5.5 to 2.52 0.07 to 1.05 0.01 to 1.12 0.07 to 2.11 nti *Cause nol lisled if not observed. fluorochenfical expo~me did not change whcn thc reference population wa~ the local counlie~ rather thap, the entire state expected or occurred too infrequently to evaluate wilh confidence of Alabama The job classifications of the workers who died Some limitat[ons must be considered when interpreting from bladder cancer were classified a> high exposme; the results of this mortaliW analysis. Although several howevc~ the} were nol exclusively chemical production jobs. methods of follow up were employed to identify decedents The majoruy of their work histor} entailed work in in this cohort, the possibility remains that some deaths were iuaiutenance work or in the plant urclnerator or wastewater not ascertained. A dcatl~ certificate was not oDtained fo~ ~ix t~catmcnt plant. knovvn decedents; thus they were not included in the cause specific death analysis. The extent to which these limitations DISCUSSION ItIJS sltlt~.y evaluated tire mortality cxpclTiCllCC Of workers with al least one yea~ of employment at a facility that would affect the results is unknown, but would have the largcsl elDct on the analysis it the unknm~n and ut~coded causes of death were at[iJbutcd to lclativcl} rare cause5 oI produced perfluornoclanesultd~onyl [luoride based fluuro- death, including liver, bladder, kidney, and progtate cancer. C]lillicals. Exposures to these chemicals ale ol pal ticulal ~other limitation tn this stud}: is the lack ol employee interesl because an end stage metabolitc, PFOS, has been shown to bc pmvasivc and biopersistcnt, leading to bioaccu specific exposure data for PFOS and other flt~orochemicah. Employment reconls were asctt to classity Ihc exposure llltllatioll in ]/Ulllan and IlOll-hLllnatl species~ ' ~ The o~rct'all status of cohorl nlelllbcr5. The ~elative exposure cs*h33ates IllOrlalJtv tales were below tl:ose expected fm most causes. were guided ~vith biomonitoring data from n previous An excess ol death from bladder cancer was detected and exposure assessment, which reduced lll[sclgss[fkTa~R)n, buI limited 1o workers who held high POSF exposnre jobs for al did not eliminate {t. Information of potentially co~3~ouuding least five years Other causes of deafl~ were below the number factors, Mr e~amplc, smoking, were not available fo~ thi~ [ 0mkt4CSomrm pto e ony eScn o Table 4 Cause specific d~aths and standardised mortallb' ratios [or all causes and Ee - selected causes of death [or cohort members ever employed in a low exposure job, but never a high exposure job |B T rx~s~*?' "~ ' " - ':~ " All deaths Fp All rnalignan~ neoptasms Ee D~jestive organs and peritoneum Large intestine Spee Biliary passages and live~ Respiratory system ER ey Bronchus, trachea, lung SE Nan-m&gnant causes Cerebro~ascular disease GErs All heart disease Cirr~sis o~ I~ver aFEAll accidents violence per Suicides Homicides *Cause rot listed aii ined wa 27 43.43 0.64 1 up sm 6 I 1 47 0.52 fH 2 2.02 0.99 I 0.70 1.43 I 0.25 3.94 i 8@ 8 ~ 4.60 0.87 ma95% a 0.42 to 0 93 sm 0.19 to 1.14 HR 0.12 to 3.57 0.04 to 7.94 HEH 0. I 0 to 21.88 aE 0.24 to 2.22 4 4 46 0.90 0 24 to 2 29 PoE mm 2 1 61 1.25 0.15 to 4.,50 Pook om 7 12 85 0.54 0.22 to 1.12 PoE EE 0.93 1.07 0.03 to 5.95 I 4~33 0.23 0.01 to 1.28 PoE OE OBE 3 2.34 128 0.26 to 3.74 1 1.45 0.69 0.02 to 383 2 0.89 2.24 0.27 k~ 8.08 eesti www.occenvmed.com 33114455..00000055 Mortality of fluorochemical workers 727 [|STabdl5e Cou spocicduo andshyrdidmaltyro fo ol cocnnd | Table 5 Cause specific dealhs and slandardised mortality ratios for all causes and selected causes of death for cohort members who only worked in non-exposed jobs Cov Cadse* - Obed Lpeond ~ecled WSMkR 9955%5C0I Observed All d~ths 53 Cancers Brows... 1 UB LEE | Al! malignant neoplasms 15 Digeslive organs and perir,.,neum 1 Oesophagus 1 i . Sa an fun Respiratory system 4 EE [i Bronchus, trachea, lung 4 Breast 2 | Boe. 1 8 BEET | Malignant melc~oma 1 Lymphatic and haematopoletic 3 ET ae 3 wm Non-malignant causes Cerebrovascular disease 1 sme. 4 8B Eu All heart disease 1 st ET 1 EE EE Non-malignant respiratory dis~a~ 1 Cirrhosis o} liver 2 A ! 220 All accidents 9 /v~tor vehicle accidents 6 oma I! Loo AIr other acdde~ts 3 Violence 4 Ho 1 3 moun St~icides 3 Homicides 1 B8.31 20.45 3.72 0.44 7.90 7.64 0.39 0.72 2.19 2.62 23.72 3.gl 2.37 14.67 8.82 5J14 7.79 4.59 3.20 0 60 0.73 0.27 2.25 0 51 0.52 5.1 | 1.38 1.37 0.38 0.59 0.26 0.04 0.61 0.68 0.51 0.51 0.65 0.31 0 45 ta 0 79 0.41 to 121 0.01 to 1 49 0.Ca5 to 12.51 0 14 t~ 1 30 0.14 to 134 0.62 to 1845 0.03 to 7 67 0.28 to 400 0.01 to 2.12 0.32 to 0.99 0.01 to 1 ..46 0.10 to 3.05 0.28 ta !17 0.25 to 1 48 0.11 to 1.50 0.14 to 1.3t 0.13 to 1 91 0.01 to 174 *Cause not UL listed if not obse.r~t. ---------- cohort. Finally, the locus of this study was fluorochemical furthcl dcaths from bladder cancer for this excess to fully exposure, sa the ip.fluence of other workplace exposures ,/,'as not assessed. The nulnbe~ of deaths from bladder cancer was more than attenuate to null. Current toxicological evidence does not iudicatc llk~t tbc bladder is a target of PFOS. Two year feeding bioassays expected, based on the prevailing rate~ for Alabama and the of N-e~hylperfl uorooctanc sulphonamide alcohol (N- regional counties. In the cltrreut analysis three cases wc~c obscr,,cd; however the fact that all three cases worked for at leasl five years iu high exposure jobs warrants further evaluatinn. Possible explanations for the excess include: (1 ~ EtFOSEI and PFOS have trot shown an increased risk bladder Itllrlours.~ ~5 t,, The |ormer colnpollnd call lltetabolisc to an undetenmned degree to I:'I:OS Most bladder carcinogens are genotoxic and/or precipitate in the urine. PFOS and tile cancers are due to exposnle to PFOS and other related chemistries are neither genotoxic nor insoluble in fluorochemica[s; {2} the cancers are attributable to another, nrine at the levels measnred in employees.~ undetermined occupaticnal exposure; (~} the cancers are A knmvn risk factor for bladder cancer is tobacco related t{~ non-occnpational factors such as smoking or other smokitrg.'- Data on sntokJilg habit ~'~c~c not available for peisonal habits; and (4', chancc, with only three cases the this anah. sis. However, in this cohort the rates of other possibility' that the linding is due Io chance cannot entirely be smoking related cauceLs, including lung cancer, wore not m]ed oul. However, given the magnitude of the risk estimate increased, suggesting that lh~" cohnrt as a whole did not it would take many year~ o[ additional follow up wile no smoke appreciably more than the general populatiou. Ee WS Table 6 Cause specific slandardised mortality ratios for workers employed for at least rorinatigh oor one year n a high exposure job" a TT eeRT All deaths C~neers Tee... 4% BEE All malignant neoplasms Digestive organs and p~rit~n~m EIT !] 2 4 5% ~SOl~ng~ fBi i 11B2 2BoBmf Bili~,ap/passg.ge~ and liver 53 73.26 0.72 0.54 to 0.95 14 16.67 0.84 0.46 to 1.41 2 3.05 0.66 0.08 to 2.37 1 0.37 2.73 0,07 to 15.t6 1 .0.39 2.57 0.06 to 14.26 Re~*'~a~/ s),stem Bro~hus,-~, lung erin i .u~g~, : , . . ~:a~d ~ urln~ ~'~ . . . -!..- . f 1 Br ie Ce~ehfoyascular disease Lo All head disease = Gn4~sis o~ liver 6 6.45 0.93 0.34 to ~-.03 - 6 6.24 0.96 0.35 i~ 2.09 1 El 2 3 0.59 - 5.u. L0,5~0.14.93 3 .O.19 " . ~16:12 ~2 ld 1of %BRq8 e 22 21 2.14. om 0.93 onway 0.11 to 3.37 I 2 19.52 0.61 0.32 to 1.07 1 5 88 2 1.93 1.03 0.13 to 3.73 All accidents 13 12.46 1.04 0.56 to 1.78 /v~olor vehicle accidents 6 All otl'er accidenls 7 Violence 3 = I Sukide~ 1 H~lades 2 NEE *Caus~ not listed if not obm~-~d. 87.53 4.93 6.63 3.93 2.71 80.80 1.42 0.45 0.25 0.74 = 0.29 to 1.74 0.57 to 2.92 0.09 to 1.32 0.01 to 1 .st1 0.09 to 2.67 tn 33114455..00000066 a Aste Gen, Barn. tal 728 Alexander, Olsem Burris, et al lu addition: to smoking, bladder i ilu~el has bcc~'~ a-,satiated ih td Sat Sat tEe arenaeh chshlcrBeod(ews itetESAdO s aneyectnacchat ~vilb suvcral occupational cxpobmes. Eiilplogulcnt H1 tt~c RA ll ot -- lcxtilc indusLry il~ particular has long beeu associaled wilh ~sk at hladde~ cancer due to the use oI aniline dyes and other Sel net oh wk a Th oe ith he i, Sev agents Sevcial d~emicals currently o~ historically used m the aT Te TI eek Ar k'x~ilc indusuy have been classilied by the lntcrnationnl Agency lot Rcsearcll uu CallCCr {IARC) as class t ((arcino- ca, nel hr yma Wikam Cxpoue ot14 gemcL clas~ 2A {probably carcinogenic}, and class 2B ei isis Ja Thr we 10 eta oean wre {possibly carcinogenicL The ntajo~ity of the chemicals sm Ho pare ome ih ch is dai ofooti ih xpos cvalaatcd fall imo the 2B category, including plasticiscrs a so, JA xd te on uid nr sucln as acekamide and dil2-ethyl hexylI-phthalate, llamc Aro "both in Ics cae a hho of Gi woe 3 retardalttS such as d~iourea. Ins {2,5-dibromo propyl phosoon kos ve ol fees fo Gh dy bane FFOS phatcl, and antimony trioxide.'~ Occupational exposure to Mii ee some Gower ed 1 he Int 1 svc cpt sc cs aromatic amines in the textile and other industries are Tr co. bon ve. hn 11h cr 1 eas rs possibly responsible for up to 25% of bladder cancer in some Ee re Ti ep. he hue pose 0 ih dvs of 105 ed conntrics'~ Polgcgclic aromalic hydrocarbuns (PAHs) cxponn mw, he cern 3 ns and pen cu iy ty surc has also been linked to bladder cancer ~'' These te ok wes, 1k ssh ht dc rue: ls associations have been repurted m the aluminium industry, a rea hl dnl oe 0 TOS The and occupatton5 where exposure to coal tar pitch is collllnon. el SC hn, fe trues bn rade, mas 1 eam 108 Several other occupational exposures that are not indusily i. su chaos and ch 81100. hers rod, thet em lol specific including metal cutting fluids, diesel exhaust, Fh Be cog 2 in. Wi the Bhs wd 10pm polybmminated biphenyIs, and perchloroethylene and other i ae ots Mus oa sn Sug FOS dc mo solvents, have been associaled, albeit inconsistently, wJl~l i ct cto Does 3 hh fom on ih Kn bladder cancer in some populations.~' ~, hte adh as 105 CALEUTEL ELE ig oofs NEFOSE nd 103 in rs Although the bladder cancc~ cases m this cohort occurred A oF omar my ht hs mars tess in etl Mons anlong workers who held high PFOS exposed jobs, the ayo ht dv 12 ch ay" LAIOSE td 1703 specific jobs of the caseg suggesl that other exposures "ad acon Fouche pti plkcnon bt self have been presenl The bladder cance~ cases vvorkcd mostly sm brwewe o a Fo ni oIfron Tho ul,of hc in maintenance or at the incinerator alqd wastewater me Chi oda ny et boty tes ee chaired 2.3 run irealment plant. The exposures in these jobs are not limited rk my of Sn hl oe a bs nts, to those encountered in a specific chemical production aedb wok HOR Wik FFO5 3 rks 10 ro bps process. Maintenance workers may work in lnanv areas ol wae aun sod hie Fee profs i he 11 4 wo yar ows the plant. The wastewatcr treatlncnt and incinerator workers are the ting dof ead stat od bo TRG Hed ot a cand a i of en are on the ~eceiviag end of the waste stream, and have lhe era 0 Ant 3FA LB Se des laden Tae opportunity to encounter a ~:umber of chemical cxposurcs. ecntotcteo noah fhoose Joe Wiats p1ioSnos en tao buries upnedlesnliec Following the initial release of Ihcsc findings to the chemical e~:pused subcohurts; approxinlatcly three tinrcs that expeclcd. Both rases were uerificd as prilllal} caucers by medkal record fallacy-back. Though more than expected, Ihese results are ditticult to inlcrpret hecause tile},' are based Oil only [~O kdhC5 Jilt CflSC with the highest I]uorochelncial exposure had a Sigh PFOS exposed job Ior only 14 months general helper m departnlents with low exposure lot 11 years. There were [10 livel CdlICCl cases among the vvorkcrs with much longer duration of employment in high exposnre jobs, thus a dose response relation conld hal be evalualed. Death from liver cancer an~ cirrhosis of the liver were a priori outconqes oI interest for t[~is study because PFOS concen:ratcd ill the liver); and several repeat dose studies have consistently d~own that the liver is lhe lmmary targel organ ~ Liver tissue response to high doses of PFOS included the enlargement of li~er and apparenl alterations in metabolic processes ~vith the reduction in serum cholesterol levels observed as the earliest clinical respoose to PFOS These ellects occurred in cynonlolgus primates at scrmn PFOS levels at 100 ppm, whereas production workers serum levels have averaged I 2 ppm, with the highest around 10 ppm.' Multiple gcuotoxicity assays suggested PFOS does not present a hazard from interaction with genetic material. Two year fecding bioassays of N-EtFOSE and PFOS in rats have shown marginal increases in hepatocellular adenomas at the highest dose in each stud}'." '~ N-EtFOSE and PFOS produced weak peroxisome proliferation, but neither compound ~sas tound to be genotoxic) Thus, the results of the two }ear bioassay sludies were characterised as a nollgenotoxic threshold response Jar liver Iulnours. ~I*OA, utl[ike PFOS, is a moderate to strung hepatic peroxisome proliferator in :[~e rat > :~ A two year bioassay at PFOA m the diel of rats caused an increase of hepatic, pancreas, and Eeydig cell adenomas.'" '~ There is considerable dehate whether pc~oxisomc prolifcrators that produce liver cumpany and worker~, a reviev, uf the current and past TE at povesit es ormgers Piedconions udlevels of aman expose presence of knu~u bladder carcinogens at this facility was oda A tofko of Joe lsCS onan Lanes of ei? sla cum conducted. A list of known m potential bladder carcinogens a opel ware Wao Th hi ave manscae pits awid w ad fi was developed, based ou ~eview of the medical literature. nh nl ee ll Aga fine id Arve Pe Forty five s~ch compounds ~vere identified. ~he historic raw al a es Ap,UT) (a fd gs ni ho bb nlaterial jEveu[oty [ccord5 kept by the company ~vere aid I CAS manker to Sty om of Eeof oe can cma searched by CAS number to identify any use at these med! Toe ds rd MTS PEKRISE eneo lc US eh campounds These records covered all materials purchased awl es echo we. am mane. Qh comeing cuss may lode Serbdos lor production use. In addition, plant records were reviewed antl nde en SAN Ct ren Bete Hione and key personnel were illtelA'iexved In ascertain more etmek on wing oot se oti hed o actedwile sesarcomiol bo specific information on chetnical use. Of the 45 compounds i: som meas aang 0 3,107 ate since Heyes, ie 12 uo sl initially identified, five were recognised as being nsed at the mun a Fut AT sl G1 pint cos opm Bacio cons. manulacturing plant site. Four materials were fonnd to have te berms ho ah Eo toat 1 dn bo Che bccn part o[ formcL inactive processes. These include 4,4 pci choehline hondasb ind nl ert of dl om ai methylenc-dianiline, orlhotoluidine, benzidine sails, and Senthil Te eo em s n net nn hv butyl benzyl phthalate. The use oI these materials ended [n ry Sovwmimoraleainry fui the t960s and 1970s. The chemicals were no~ widely used i wr rai pore mrt sn a ks tir cn Sonnet lhe plant; however the available exposure mouitori~ag and ifmt edBiv nh 2% ce Sf aman oar ty ubc ilfforma~ion ~'as ve~' limited. Exposure comrols con- Ha soot waged 400] cures. i anly une Snes Soest bn 5 sisted of ventilation controls and employee lraiuing for safe ang aoe, On er evlmte SRSHU Ror epost of ats cbr ch kd handling practices One mamrial (melamine) is currently iu ono ae oben) potato 0 et i wet en Tu use m an epoxy capsule (non-fluorochemicalt prodnct line. amineect for pst atened Sc cred to iy 0 col ih Melamine has been used D~ the past decade iu the chemical ALAS Ce, I plant Jn epoxy capsule manufacturing. Respiratory protection ny SULT Ct its 5 id A and local exhaust venlilalion have been the prinlary exposure ns inctod exposures 10 miami were bo bad on comrol methods lot melamine. Qualitalivc exposure Ulelll5 indicated exposures to melamiuc ~ve~e low based on {pe aa OR AckowsptunTs shall exposnrc task duraHous and expusure conlrol practices. ai ae oo and oo heTr a ok Foy Un BA Chloroprcnc was also used in several nla~tdacturi~g pro- ccsgcs in the chemical plant in the 1960s and 1970s. es ain a pra Vath CT omni tod to nd wo, of ot Tere WH Chloroprelqe is cunsidclcd a possible htunan carcinogen; -- atte 4k mn bl mt WUaTEtDvd he om however, the evidence for a rtde in bladder cancer equivocal.:* tuinour> in rats are carcinogenic to humans tinder pated conditions and levels at hmnan exposure" K~own CaUSeS o[ prinlar}' hcpatoccllular Calcinollla in humanq incl~Jde Ihe hepatitis g and C viruses, and aflaloxins pr~uced hv Aspet?lilhes fl~vrt~s and Asl,c~sKllus primarily in foodstuffs '* tlepatitis B virus is Ibuught to be req~o~siblc [or up to NSvo ol liver cancers in endemic arcas; howe~er the prevalence of infection in the USA is relatively low. other contributing causes mat include cirrhosis dne to excessive alcohol cousumption. Expusurc to high concentrations of ~inyl chloride is associated with angiosarcoma of the liver, a vcr} rare cancer. However, there arc no clear occupational causes of primary hepatocelhllar carcinoma. It is possible that lhal this finding is duc to chance. In sunnnary, an excess ocuunence of death from bladder cancer was observed in this population; however, it is not clear whether these cases can be attributed to fluorochemical exposure, an unknowu bladder carcinogen encounlcrcd during the coarse of tllaintenaucc vvo~ k, and/or non-occupational exposures. With only three cases observed in the population the possibility of chance cannot be entirely ruled out. t'l~e renditions of a priori interest were fewe~ than expected or occurred too infrequently t0 cvaluatc with confidence. The relaltvely voung age aud small size of the PUY Hor hot SpEIR rays. cohort cuncutly precludes a detailed analysis by exposure. particularly for less cummon diseases. ACKNOWLEDGEMENTS Tile authors wish to thank NazlQ, Pengra, Mark Bolk, nbe<k, alut Diner Kampa for a~israme wilb die IllOltalilv follutv up. data This work was suppoitcd by the 3M Company. 33114455..00000077 tl boca te Mortality oJ fluorochemical ~'orkers tig Authors' affiliations [UESEyfMssnr , Shokm o PcHr c. B H Alexander, Universih/of Minnesota, School o~ Public Hec~lth, Division of Environmental and Occupational Health, MMC 807 Mayo Ee i Sr Building, 420 Delaware St, SE Minneapolis, MN 55409, USA G W olsen, J M Burris, J H Mandel, 3M Company Medical ES EE fd Depar'ment, Saint Paul, MN, USA STR Dy hs sd theo J S Mandel, Emory University, Rollins School of Public Health, Tbe Depar~men! of Epidemiology, Atlanta, GA, USA .729 Sp 15 Seacat A/~, Thomfurd P J, bulenhof~ JL Terminal obse~ations in Spra~ue B Sy e pnhE DoPey rals a~er h~time diela~ exposure to ~lassium RNC SA sulfonate Toxicologist 200266: ar 16 ~omford P J, Seacat ~, Butenkaff JL Terminal obse~atlons in Sprague EE nt Dawley ra~s after liflime dieta~ exposure to N ethyl EEA a re perfluorooctanesulfonamide emanoh [ox~co!og;st 2002,66:185 om Ra rrrepeee rmen,te- 17 Silverman DT, Marzipan AS, Dev~sa SS Bladder cancer In Sche~enfeld D Fraumeni )F, cds Cancer epidemiolo~y and prevention New York Universi~ Press, 19961156-79 18 Inlernafio~al Age~ ~r Res~r~h on Cancer IARC monographs on REFERENCES. fimo dm fe REFERENCES Ry stp tend 7Ep esdGr Cond I 3M Company SIDS init;at assessn~ent report: perfluor~octane suJphonic acd J a NAN and i~s salts U.SEPA Pubhc Dcx:ket AR 226 Saint Paul 3M Company, 2000 sar SHEE ier ir 2 Seacat AM Butenhoff JL. Comparative sub chronic toxicity of perfluorooctane sulphonate IP[OS) and N elh'~ perfluoro~x'lanesulphor, arnide ethanol (N. Ee RT EtFOSE) in Ihe =at Toxico!ogist2001;60:348 3 So, cat AM Thomford PJ J Ht( et al Subchronic toxicil), studies on perfluorooctanesulphate potassium salt in cynomolgus -nonkeys Taxicol So, 7MdmL 2002;68249 64 Sond LI Cp Fo See lc dc 4 Case M[, York FG, Christian MS Rat and rabbil oral developmental toxicology studies with ~o perfluorinated compounds Int J Toxicol BET cists 2 BOE 2001,20101-9 5 Luebker DJ, Hanson KJ, Bass N, et al Interactions of fiuorochemicals with rat on li~er Jatty acid~bindlng prot~ir~ Toxicology 2002;176:175-85 A 6 wallace KB, luebkcr DJ, Butenhoff Jl, et al Perfluorooctanesulphonale and 2 (n-e~ylp~rl]uoroaclanesulphonamindeke~fiyl alcohol are p~r~xisome NE CLA------ prolibrator~ in rat~, but not guinea pigs Toxicologist ~001;60348. 7 Sta~ov A, Butcnhoff JL, Seacat AM, et al Structural determinants of eR 27Pe Aa.Det 1 a1 Ger aves sdspn a co mito~hondrial dysfunction caused b~, in villa exl~osufe to selected Ee, ERE Tr p~rfboro0tanyl compounds Toxicologist 2001 ;~0:1658 B Hanson KJ, C]emen LA, Efle~on ME, et al Compound specitlc, quantitative characterization of organic fluorochernicals in biological matrices Environ rt Ee A Sh Techna12001 ;3S 766-70 9 Giesy JP Kannan K Global distribution of per~uoroctane suJ~phonate and related pedtuorinated compounds in wildlife Environ Sci Tecnno WEE ERR 2001;35 1339 42. 10 Olsen GW Logan PW, Simpson CA, et a! An occupational exposure EE Cal = A et assessment at a parfiuorooctunesulfuny~ fluoride production sit~: biomonltorlng Am Indust Hyg Assoc J In press n SbiERENE rn, TEER 11 OIscn GW Logan PW, Simpso~ CA, et al FIuoroch~mico~ EA assessment of Decatur chemi~ai and firm plant employees USEPA Public Docket AR-226 Saint Paul~ 3M Company, 1999. 1 EERT es 2 Sen cpm, 12 Ol~en GW, Bums JM, Mandel JH, e~ al Serum perfluorooctane suffanate and hepatic and lipid clinical chernistr~ tests in }]uorochemical production JR A A ------ employees J Occup Environ Med1999;41:799-806 13 National Instltutes for O~cupat~onal Safety and Health. PC LTAS life table Sm I Comet Cope Wao 4 pr analysts ~ystern rrar use on the PC Cincinnati US Department of Health ard 18een8 Sn5)Oh FF utr ev ofexten nd ge 34LdnWESheKk Le coms ScbmnldO Fram, oh Hunnan Services, 1998 14 Johnson JO, Gib~on SJ, Ober RF Exlenl and route of excretion and tissue Er RoR e RATE pape SteeTnet Clem hn distribut;on af total ~-arb~n 14 ~n rats after a sing,le i v d~se of FC 94-~ C Projecl 890031020~ Saint Paul, MN: Riker Laboratories Inc, 1979 evaluation o~ carcm~enic risks to humans 5ome flame re~rdants and ~tilu ~hemicals ande~s~r~ m lhe textile manufac~ring indus~ Lyon IARC, 1990. 19 Vineis P, Pirastu R A~omatic amines and cancer Cancer Cause~ Control ~ 997;8346 55 20 Boffe~ P, Jourenkova N, Gustavs~on P. Cancer risk from occu~tiona] and enviranm~tul exposure to ~[y~clic aromatic hyd~o~ar~ns Cancer Causes Control 1997;8~444 72 21 Ward EM, Burneff CA, Ruder A, el aJ Industries and cancer Can~er Control 1997;8:356-70 22 B~assi S, Maria F Pearce N, e~ al Bladder cancer and occupationa~ ex~sure b ~lycyclic aromatic hydr~ar~ns ~ntlCanc~r1989;~:648~51 23 Hours M, Dananche B, Fevo~e J et al Bladder cancer and occupational ex~sures Stand J Work Envir~ Heal$ 1994;20322-30 24 S~in~k G, Plato N, Norell SE~ et al. Uro~elial cancer and some indust~- relat~ chemicals: an evaluation of the epid~iol~ic literature Am J Ind M~ 1990:17:371-91. 25 An~n-CuDer H, L~-Fe~dslein A, Taylor TH. ~cupation and b~adder cancer risk. Am J Epid~io~ 1992;136:8~94 26 Weiss NS Cancer in ~elation to ~cupoflonal e~osure t~ perch[oroethylene Cancer Causes Control ~ 995;6~257-66 27 Po~u S, Aubnti V, Donato F, etaL B~adde~ cancer and occupation: a cas~ conlrol ~dy in no~ern Italy ~cup Environ Med 1996;53:6-10. 28 Interna~onal hg~ for Re~h on Can~er IAEC mon~raphs on ~e evaJuatio~ of carcin~n/~ risks ~u human~: Re evaluation of some organic chemicals hydrazine and hydr~en peroxide [Pa~ One] Lyon IA~C, ]999 29 Haughom ~ Spydevo~d O The n~echanism underlying ~e hypolipemic e~ect of ~rfluor~lanoic acid, perfluorooctane ~ulphonic acid (PFOSA) and cloflbric acid Bi~h~m B~ophys A~a 1992;1128 65-72 30 Sibinskl ~ Two-year oral (diet) toxici~/carcinogenici~ of fluorochemical FC- J 43 in ra s ~int Pau : R ker Laboratories, 1987 31 fi~e[ ~, Liu RCM, Hu~ ME, eta~ Eff~ts o~ ammonium perfluorooctanoae on [eydig cell ~nction: in vitro, n viva. and ex viva studies Toxicol Appl Pharmacc11995;134:18 25 32 O~urn JD, Frame SR, Bell RH, etal Mechanisms for ~e pancreaeafic oncogenic eff~ts of peroxlsome proiife~ato~ Wye~-I 4,643 Taxicol Appl Pharmacel 199 / : 145 33 Cagey RC, DeLuca J Elcan~ E eta! Do ~roxisame p~oliferating compunds ~se a h~tacarciongenic hazard to humans R~ul Tox~co~ Pharmacal 1998;27:47 60 ~4 ~ndon~, McG~ynn ~ Liver cancer ~n Schoffenfeld D, Fraumeni JF, ads Cancer epidemiology and preventJon. New York: Oxford Universi~ 1996:772 93 33114455..00000088