Document XO6oEdb73dme6L5bV8ed5pNvK

Page 1 o f 2 Shawn Pennybacker ~ From: R. Dean Hartley Sent: Thursday, April 19, 2001 11:41 AM To: Shawn Pennybacker Subject: FW: interesting paper Get a hard copy o f this paper in the near future. -----Original Message----From: martyn smith [mailto:martynts@uclink.berkeley.edu] Sent: Thursday, April 19, 2001 12:48 AM To: ahawes@alexanderlaw.com; rmetzger@toxictorts.com; dhartley@hartleyobrien.com Subject: interesting paper Scand J Work Environ Health 2000 Dec;26(6):482-9 lRelated Articles,JBooks Acute myeloid leukemia and clonal chromosome aberrations in relation to past exposure to organic solvents. Albin M, Bjork J, Welinder H, Tinnerberg H, Mauritzson N, Johansson B, Billstrom R, Stromberg U, Mikoczy Z, Ahlgren T, Nilsson PG, Mitelman F, Hagmar L. Department of Occupational and Environmental Medicine, Lund University Hospital, Sweden. maria.albin@ymed.lu.se OBJECTIVES: The effects of occupational and leisure-time exposures on the risk of acute myeloid leukemia (AML) were investigated with emphasis on clonal chromosome aberrations (CCA) and morphological subtypes. METHODS: Consecutively diagnosed cases of AML (N=333) and 1 population referent per case were retrospectively included in the study. Information on worktasks, companies, and leisure-time activities was obtained with telephone interviews. Exposure probability and intensity were assessed by occupational hygienists. Associations were evaluated with logistic regression. RESULTS: Exposure to organic solvents was associated with an increased risk of AML [low exposure: OR 1.5 (95% confidence interval (95% CI) 1.O-2.3, moderate-high exposure: OR 2.3 (95% CI 1.O-5.0)]. For exposure to solvents, but not to benzene, the OR was 1.2 (95% CI 0.69-2.0) for "low" and 2.7 (95% CI 1.O-7.3) for "moderate-high" exposure. The observed effects increased with intensity and duration of exposure. The estimated effects were higher for patients >60 years of age at the time of diagnosis. The effect of exposure to organic solvents was not differential with regard to morphology [except possibly erythroleukemia: OR 4.2, 95% CI 1.O-17 or the presence of CCA in general]. No increased risk for AML with complex CCA or with total or partial losses of chromosomes 5 or 7 were observed, but a higher risk was found for AML with trisomy 8 (OR 11, 95% CI 2.7-42) as the sole aberration. CONCLUSIONS: Exposure to organic solvents was associated with an increased risk of AML. This association was not due to benzene exposure alone and may be modified by age. Furthermore, specific associations with trisomy 8, and possibly also erythroleukemia, were suggested. PMID: 11201395 [PubMed - indexed for MEDLINE] Martyn T. Smith, Ph.D. Professor of Toxicology Division of Environmental Health Sciences School of Public Health 216 Earl Warren Hall University of California Berkeley, California 94720-7360 4/19/01 Apr-14-01 12:21pm Frorn- Acute m y e W leukemia (Ah&) is characrcJized by acquin=d. nonraadamclonal ~hromosorrreahmmalities, which have provedto be clinicaliy and biologicallyim- pomar sin& they provide diagrrostic. prognosck, e t b krgk and pahogen& infans;irau (1.2). soarr?experi- --'Imcatunsreadl isffceurdeinest indicate Qat different exposures m y types of clonal chmmomme abnomtali- rk(3-a Punbermom, AML followingd o orckm- 0cher;rPy differscytogenetically from de nova AML (6, 7).The ptrzms fCtrcIom1ChrWMPa b n m i i u e s ia agenu have been associared with total or partial losses of c h m ~ r r m e5s and 7 add ddcrionsaf the short arms of chmomnxs 3 and 12. WhmM topairomernse-lr inhibitMEhave been linked to srruciuralrewrangemtsof chrcrmosomebands 1lq23 and 21q22&9). In addition. id peripheral lymphocytes of healthy workers exposed to high coTlccntrariOnr of bcnscne, Qsedependenr griw (mainly trisomy)of ckom~somz8 (10)add deleubmof chromosomes 5 and 7 C11) havebeen found. Known risk factors for AML include radio- and chemorherapy. bcnzeneexposunk and smoking (12). In addition. exposure to o t k r organic solvznrs (13). ex- I Apr-18-01 12:21pm Fron- T-415 P.03/11 F-465 Afhher a1 mmdy low fiequoncy tlectrwlagnutic fields (ElMF) (14-18), hair dyes (1!+21), pesticides (22). petroleum products and exhusr gases (23). dmalbomt viruses (241, and fcmuy work (25) have beenindicated as risk factors,albeit with confiiainglesults(12,26). stlldies ofsu&ysted o c c u ~ n arilskfacms far spccific subtypes of AML, as defined by rnorpbbgy br karyotype, have aoo1been consistent. Exposure was not associami wirh any specific marphology in one study (271,urherw- others have indicated i m a s e d risks for acute myelomonocyticlcukemja &r benzene (28) and peaicide (29) exposum. i n d r j s k for acute promy- Czlocyticleukemia in elecuicians(30).and a possibleassocktion between exposure tosolvents.insecdcides,ot petroleumproducts and erythroleukemia(31). With regard IO c l m l chromosormalmormalirics, in some studies, AML.patients ~xpaPodu)sdwms 01 pemicides (31- 33)or 'chemicalu" (27.34) had a kighm proportion of clomal chqnc&ameabnonnalitie (cspacially cornpbte or p i a l lossdchnomosorae5 or 7. aod trisomy 8). but not in orhas (I d35.36).It has alsobeen ruggesredthat AML io ~ s o c i a rwi i~rh previous roxic occupational exposam hasa woise prognosis and qmsenti a disdnct clirlicd.as milas qtogenezic,entity (32). Limt is h o w n abaur rhc effccr modificarion of tx- posunby a&, genckr, and smoking in AML.However, sidlcrcl al(373ObscIvGd 86 exp6snrc-nspmstbehvcen SmQkmg andrisk for AML only fcx subjecrs above tbe age of 60years at &e ofdiagnosh. The pmwt study was unlnaJreato expkue prevaient apcwrcs. such BS organic sclvents, ELhdF,and la& dyes. suggcsrsdas risk hctonin preview studies wirh regardto o v d l risk for AML M m w , other objec- tives were IO study chararmisticclonal cimmosorneabnormalities@ markers of exposure, diffemtial risksfor mctrpldogicaisubtypes,and effecc moditicafi011 by age. $onflex,a d ~ ~ n ~ k i ~ g . cases Our study is based on a scsjes of 372 umse~udvecases d Ah&. d i a p o d h e m 1976 and 1993;the sub$cts were at least 20 years of age when cytogenetically dvestigaredat &e Dcpartmant of Clinical Generics,Ludd, which pelfcmls all cytegenctjcanalysesiaswcheanswedcn.AU referralsfor such analyses are ma& wirhh the pubiic Wdr system. "%e hatoIogy depar&m~81 he Lund Univemiry Huspital and at thc n&y b ~ p i t din Helsingbcug, thmughwt &e study paioti, included cytogCnetic Sfnalysesas a routine in the diagnmliccrtIrmtioxl of hemamlogical dgnancies, lodicading that the coverage of incidem cases is mclsf complete far rhese catchmcat areas (Lund-Helhgbmg). Intmkw mformah a was obtajntdfbr333 caas(9096) (table 1). The non- parcicipadng caseswcrc OW(median age 68years] and fLum cdknwamm 09%). The medical meo& for eachcase werereviewedby 1 Of3 Jmnadogisn(38)."hediagnosesw m based on tbc moiphoiogical hnd clinical findings present 81 the time 61the CyIPgeneric investigation. The m o ~ o g i ca1 subqpcs according to the French-Amerjcan-IBritkh FAB) cl;rsPiflcacianwe= as follows: miaimally M i x enrfawd (MO)1 case, myeloblastic without melmUim - Apr-18-Ol 12:22pw From- T-415 P.04/11 F 4 6 5 a Referents Statistics Swedcn seleaed 3 referents foreach case. -----Imatched wich e p e c t to gender, year ofbinh. and CyIun- ty of living fmnthesmdy popukion (souchemS w a n ) within the calends year in which each case was diag- * nosed. Ode referent in each matched set was randomly selected 10 be interviewed.Whcn referenrsor h k eext- I of kin could not k inrwizwud, orherreferems from rhe match& set were sekcced M t o g e t k 351referentswrra ineluded in the study (parriciphn rate 72%) (ubkI). The nonparticipatingrekrenrs wert alder (median age 68 years) and mmc o m women (60%). The s ~ b j e a sw r c r o t r o s w v d y cnrollcd in rhe study and conlacted with a letter describing the purpasc 484 Scand J work Envimn Hegllh 2o00,df26,no 6 dtht study and rhcmain question9 to be addmseddur- ing the inmiew. A life-hhg aecupati&nalbtrmy was &mined during chc Mod 1995-1997 in a rmuculred rekpbne 'mrview made by 1 of3 occupa- ticmadheal* nurSCS who were a w m of thecase-refemt st~hlli.The histmy inc!eW all jobs held far at h r 1 ycar. k l d i g w m h k ciepamm, d nameofcom- pany. No specific e s r p o s ~were asked fclr ia &e her- view, bur follow-up questions w e triggered iag cbzapp.iicacionufpesdeidesinwork as ilgardeRuor horricuhrisr, lumberjack, zdway w o r k (track main- renancc), f m m , oraamhmri;cype of WrgQ inwoslr,as a driver or un a m d m t ship;handlingofc y t o d c s in work as a rmmz; and working I a mea counter in work aa ashop Wpm oc shop assismm. Infbrrrmrionon extra jobti held f a at least 1 year wns collmtcd in tbe same way as for the regnlarjob. Tbe following hobby rriviries were explicitly in- quid abom:motor repair, mi& pluuing withoil. fur- nimm renovattoa,CcuJs~tionbfmOderkich%,animal h d i a g . hrmring, wd gardening. Follow-up questions were asked & h u rxrivity periodand average hours per week spcnr in ihbe aaivity. A follow-upquesrionOII reg- ular pcslicide use was triggered by gnrdsniag as a hob- by.Furthennore, qucsrions WCIS asked abouc awar hair dye UK.and smoking habits. Tlze medical histmy pre- ceding&a rim of the casediagnosis iacluded quesrianr &our radiotherapy. chtmtttewy,and ct- wim imp. lf rhz 3ubjjcct was deceasedMtaa ill to penicipe, a nexc-&kin was selsred in the fdhar;lg orda:spusc, parenr., siilmg. and cbild. Infomciw hadLDbe OMIiOcd fmrhtnext-&-kin more oftzn for cases rhan fm mfer- e m (tahle 1). The proporriOn of nexrsf-kin interviews increased as the age of thc subjtccs increased (eg. next- of kin imerviews were used & 933% of the cases and 23.5% ofthe refmnrs o mthan60y%Wsof agck S p - es and children were rbz most frrquently interviewed amng the next-&kin abath thecases (48% and 36%, respectively) and ehe r e f m r s (27% and 6396,rcspzc- tively). Exposure assessmerlt Exposure aSS2SSm2ntwas pesfwmcd by 1of 3 OCCUpational hygienists, CUI tho basis of the suuctured relep b inaetviewsmadeby the ocqat-malhedrhnurs- es. Tbe occuparional hygienists werc U M W ~of the casI?-rcfemntstatus, The ssesfllocncs For all cases,and for rb ccurespondingrefeenu, were rtsrrictcdUI erposure periodsof at least 1 year duringthe 20 years before [he diagnwis of the case. All ccdings w p e supervised by double-checks petformed by one of the hygienists (m. Apr-14-01 12:23pm From- Albin QIai Exposure assessments were performed for estab- judged from coinparisonswith rhe medical recordsavail- lished, and suggcsred.cccupaficnaland leisure-timerisk ableforthe cases (&le I ). ThusThe effect cstimareswcrc lacton for AMI.(listed in table 2). Quantitative defini- likely to be biased and therefore have not been presai- tions of the diffaenr exposurr intensity catcpories were 4. uscdfor organic solvents, benzene. and U F ."W. `modeme",and "high"exposure 10 organic solvenrscor- Statisisticalanalfies responded to t--5%, 5-1546, and >15%, of the present When Only -& mMCMg ~CCOISsuch Bs @md gCn- Swr?dishaccupational exposure limits;,reipmvely. For der are uscd, confounding can narmslly be rno* c,E- knzene. rhc comspondingcargwics WEE 0.05-02, ciently controlled in a suarified analysis rhan in a 0 2 4 . 5 , and ;.OS rngld. Exposure to ELMFwas c b - matched mnalyUu(45). In om smdy, all effat estimates rified m x m i i i g u)S-hour arithmetic averages reported were therefore obtainedby uneandiritmllogis& rcgresl for differem occuparions (441,with the fallowing cate- sion (36),sfzttified[or age into 3 b a d eaiipbrics (G4, gories: 0.2%0.30.0.3043.50. and rO.50 pT. For #I 55-69,md 270 yean) md gender. Comprpison with -. Mher occupational exposures. thedifferentexposurein- estimates from a conditional logistic regression model tensity caregcnies ryere &heated by examplesofw u - (46)showed hah e stratification sufficiemlyaccounted patioas or rasls. Pot txposurc to aromacic organic sol- for the cfftct of the matching vaiablcs. Year of diagno- vents (exchdiq borneat), ChloridptedO I p i c solvents. L and county ofresidence did not influence the e f k t and cytostatic drug. only dichoramous assessmznc (ex- cstimces and were therefore noT included. When rhe posedhnexposed)was pdble. Expowretocattle7po~& [rends were tested,the m&icienr from a logistic mods1 rry mB fnesh mew was consideredtogether due to the hy- with h e exposureintensity (L=bobby-low, kmderate. potherisofaannmon causal factor (animalhe v i m - 3=bigh)add dumian (as a continuous variabk). respec- IS)-The main occuparion in this exposure catego- was tively, was IS^ as Ihr wt mrixic, ppovided drar tbtdata fkllner. were furly cmsisun! v,+krih such a model (47). Thzpop Fm rtgular hair dye use and smoking. classificatiou ulrruon atrribucabh risk, with 95% cadfidence ialrarals was dincrly bawdM the specificquestionsin he inrer- obtained witb tbc ManuA-Hacnszel qpmach, was esti- view. Cases add refmatswere classified3s smokers if mated whcn appropriate (48,49). rbeyhadmoLEdrwarlensr1 yearduringtheMyears Tn wdcr to miucc cxponse midasdACadon, only before &e dugnosisofthe case. Mure-time expornma expos^ assesswnts wilh bigh cbniideoctfaroccupa- were asses& for9 tbe agmars,excepr EIUIIF.ionizing tional exposive~were used indK:a o a l y s ~ as l l b j a were &ion and cb-puric drugs. classifiedas uncxpd fora particulara g a only if they For &e occuparional expawms, rhc ~mfidzncein he were unexposed with bigk confidenceduring rbt tntirr: mpmureoodinps was assessed (high CDafidonCc or not) &szssmeni period. subjects were clarified a i exposed on the basis af the quality of informadm on worktasks 5&cy were e x p e d 4 t h bigb confidence for at bast 1 and erpoams in Ehe iareiviev data and the confidence y e w duing rhc assessmeni period. Subjects that could of possibleexposure for differentworkrafks. not bc classlftd as txpased Munexposedwith highcon- kmte~rcbbiliy was assessed a f m the exposure frdcncc were excluded fromrheanalyses. A kss restric- assessments were finabed for all the interview forms. tiveappwachdidMI change tbe results subsranrially. and Tht main occupatiamlbygienisr (HW) reaq$essed198 they herefon have nor been presuftcd However, the randomly sd~cltdirh\riew fomts. unawareof the arig- appmch implies that the psesented exposumdurations inal exposure -KS. h rbe ctrigiaal Lusessmenr47 and intentities are minimumvalues. of these subjerns were dasSi3ed as ha- certainoccu- pational OT Icim-t* exposue IO organic solvents, 70%of tbem were ctwrifid in same way in the itassesmen&The 193 iPerViews were alsor o s ~ ~ a bcyd Retvllr aZndaecapMidhy~.Tbeconcardaawbetwcm the 2 r c a p ~ e s s mw~as~s~iightly benn (8wL;shpn bc- tweenrbenew and old assesuncnts;thisdifference indi- caw Lt SIighS sbift ia asasmml with riw. However, lb#wxenot&fkmws io &ecoac-bttween he wssaodrribartf. OntytbeoldassesrmentEw c n mcd in the anai]nar of wpi%ufceffects. Wecbwetoincludeinfmmrion on jwevjous chemo- arm- h m beU~h e ir#crvicw d tht medical d s (casesd y l in cbe chsihuian. sincetbe inter- view was insensitive i m detecting such trcaunenls as .. ...I :' -_ t . Apr-18-01 12:23pm From- .# T-415 P.06/11 F-465 -. 5 63 237 33 28 -25 10 65 288 7 a73 0.23-2.4 67 252 92 1.0 0.66-13 35 1196 0.56-1.7 28 1.1 6.8&+1 -IS 69 262 Ll82 0.35-1.9 0.98 0.64-15 indicatedan i-mg risk with increasing intensity of exposure [ m e 3, wsr fervend (hobby-low, maderape, high),Pd).Ql]. Trpicrla t m p t.m s in rhc modcmc-high intenshy caregory WCR painters and a b b e industry workas. slrd in rbeIow-intensitycalcgory carpentersand mtor mechanicsprevailed. S u p d asvaciations were observed for eqosure to kozene. bur not fcrr chlorinatedhyhnsbons. No in- creasing Lislr by intensity of benzene txpowc WBS ob- w e d . E f f e fFomexposum to organic sthcnts and to benzene were separated by rzsmcrionof rhe analyses rn mbjccrs c q w d 10 sokenrs othzr than benzene. The odds ratio was 1.2 (95% CI 0.69-2.6) Tor the hobby- low category and 27(%% CI 1.0-7.3) for them d x - *high categoty;this m u I t indicates that the observed cffcct was not due to be~llencexposwe aloae. Separa- tion of the &acts from exposum IOall organic solvents from rh3z of sromatic solvenu orher rhaa W n c was not gossiblcbdeauhthey w e too closely associated. Analyses by dunrion of only occupationalexposure (ic,excluding kisure-rimzexposures) to all organic sol- vents &Q indicated 2 positive assmiation with risk far AML [1-7 0 2 case^, I5 rr4-u): OR 0.!38,45% CT 0.44-2.2: 8-14 ye- (IS cases, 15 refmenu): OR 1.2. 95% CI -2.5; 215 y a ' ( 3 0 C- 17 nfet- 486 &and J Wwk Engiron NmNn 2#0, w26,cIa6 Albin er d 0355-3141 Apr-19-01 12:Zlpm FroM- ~ b mz t prevalmt qusure wag white spirit. For this agenk thcokd-wouldconespolsd madcffccrlcv- el of<lS rnghd, which is remarlrabiy LOW. In addition, an effectfrom leisure-timc exposurr only [OR1.8,95% cIO.92-33. WC in WBS indirMed, E- txposurc, which is Bn established risk as- rnfar AML, did not fully explain &? e m s risk for ex- pururc:to organic wlvents. +IkobservedleveIs of ben- zene e x @ m were low(72% ofthe exposedcases were e x p d to iarensities belnw 0.07 ppmj. uul our modest- ly elevatedO R estimates fw benzene exposrule are thus rlw ia conflictwith rRuseof formcr srudies. We did not observean increased risk in association with oven11c x p m topesticides.Thc fquent ust of Uhxssion prpxy mpdnJcals fw thecases didDOE p e d a classlfi- catitioa of tbc type of peticicle. Our frnd-5 of no in- ?he salientfeature of the ~ 6 s e a ustudy was thz ccmsist- creaved risk with regard to the u ~ oef hair d y u am in a t &ahan bemeenexp- m organic salvcslrs and acrordaaEewith rheresults of 2 f m m r studies {51,52), 1 risk for AML,c~pecidlyia cases wirh rrisamy a as rhe while 3 other mdjes[l9-21) have b mmo&stly sup sole abwarion.Cmrrary to our hypochis, che a s s ~ c i a - portive of such a risk. Revibus studies of &e risk far uoas were essentially Rondiffemwial with regard to the A M L associared c k h cxpogurc to ELMF are conflicting O V pres~ence dclond ch.romosolllt!&mrmaliticr, as well as co LOWur paltiidlosses ofchmrnosows5 and7, The hypothesis of an associarimh w e m suchexposure andtrLomy 8 was less sadng, having been Wed only and indicate a modmrdy incmacdri* if any (26). C~IK resulrs ate compatib[c with thistinding. We may, IO somr?extent.have underestimated the in- Ensity and duration oexposure among the subjars by oll a&w earlierpublicarioay (10,31,331. Inou1study, 7 out of 13 wirh 4 3 as a sole anomaly hada history of exposuretoorganic sdvmts (table4). a pnvakncesimilar to that rcporced by Dwico et a1 (33). It Is. in his contrxf. mtew0Ifby rbat a mriew of mament-nhtcd AML ody wing m p s m jgsessmeftts with high confidwcc. Bias resmiting fronsuch miscldfication may be in eid m direction when mre than 2 exposmo categoriesare used in the analysis (53). However, for hoch fht cases and referenu classified as expossd to organic sohrer~s. did not reveal a s @ ~ t I y higherincidenceof uisomy 8 in such AML.w h camparedwirh de novoAML {7]. The possibility rbat expwure TO o r p i c solvents thus comulrsin difmntkaryotypic patterns thandoes exporn to radio- or chcrnotkapy is fuitber suppond by rhe the average number of years assessedwirh high c M - dencewm I8 (out of20 possible); this fiading indicares that this is a minor problem. In addrim. we chose. rn the basis of pnviws&s of treaaneat-rehd AMlL (13)and AMLrelatedtohighbenzew exposum(S),to present lwk of an associaion bwbea exposure to solv c m and itbwmalitieS of ~ h r o ~ s a m5d asnd 7, aber- rations that are commod id mstmc?nt-dated AML (7). focus exposure assessmaton the 20 years before d q nosis. If kss inwnse, aad kss toxic, exposure is associacd wirh P longer lsrznCy t h e rhno 20 p&So,ur assess- It remains to he sezn v t h e h other myeloid malignan- men&would u n k r r s ~ m w&vam cxpomwdurations cies with trisomy 8 as a soh anomaly, especially myelo- An i m p m r spcct of rhc vaIidiry af the study ist&e dysphdc Syndtomcs and chronic myclopdiferaiivc dis- diffcrrndalUSE of proxy~ e q w d c mfsor cases amirefa- orders [1). are also associated with exposure to organic enrs; rbis us(! violacesthe camparable accuwy principle.. . solvents. for;rhcselection ofrdcmns (55) and may lead TO negir- The suggested higher risk for eryrhmhukmia rhaa rive damation bias. However, when there is no indica- for other subtypes of AML in relation to exposunsto all rimb asubstantial difference ia in far ma ti^^qnality.the organic solventsis inmesting,but it bas previously only dominsrlr usz:of pmxy ".pbAdentp for Lht:caacsand selfbeen repxd in relatian to a similar small number of r q x m EMxcfc=nts is justified (W.Thenfemts in our cases frona pooled study (3I>and a case-seriesof kn- study rewned tht s ~ m icnediaanumher ofjobs held as z e n e - e x w wcarbrr (91- rht casts (table 1). In additian, no d&rcncc in the Our dbsrrvations of censirtontty increased risb,.even n u m k of yzars assessedwih high mfidenct WSxecn Awn low-kveicltposlsrr [not exceeding5% ofthe Swe+ ish accuparional cxpc~urclimits) to organic solvents are a s o u t f e ofworry. The results also indicate rhat expo- beween &C E B S ~ S&r&mntr f& expsuro ta OTgsolvents. Thus no substantialMerencz in infannation quality ws indicmd, b u ~tke ~ ~ S U X EaSre crude. W e WIZ to osganic solvents m y concribwe IOan impomnr refrained fmm addingrrsponden~S f a N S US acovdate in ~ o p t i o dof &e A M L cases in&e generalpopulatiw. tht togietic r e p s i o n model becaosc such adjustments 488 scan6J Work ~nvimnHeal* MM. VOI2s. o 8 . Apr-lfl-01 12:Zlpm From- Atbin dl ar F-46 5 1 do not secewidy nedpce tbe net bias (57). An effect e+ data entry. Asra Pcrsson and Gunnel Nilsson rzuicved timate fororganic svIvencs lestricfad io self-ftspondenls the mcdical records. did not indjcrte any subsranrialbias. SrilI, rhe dear dif- The pmjm was sapponed by grants fromtbr Swcd- ference in rhe pmpcmiun d ~ ~ ~ - d -i nk ~ivni c w ms ay ish C m c i l for Work Lift &search, rhe Swedish Can- have Hecred the affect csricnares. Thjs possibility espe- cer Society, the Medical Fxulry of h a d Uoiverrity, cially app1iesto cases diagnowd for DMrs people. Lunds URivessity Hospital, Crunnar.Arvid and Elisakth Although Iheassess men^ of exposure was made wirh- Nilsfoms Rt$earc.h Foundation, and PRBEM Resmch out b l c d g o~f c a s c d m r st~m5h, r ethical m o n sI Founddon. The pmjcct w e a p p ~ v e dby the Ethics' the nurseswho c m t d but tbc inmiews hadh i s bowl- Cbmmirlccof Lund University. edge. This situcion may have inrroduced a ipositive) bias, bot, in OUT opinion, it i s likely io have becd m a l l since the interviews were nrherArmly svucNIzd References All refmals far cytogeneticinresrigsrionsare made wilhin h e Swedish public bed&spun, and there is dwcforc no reason to suspect any socioecmxnically bared sekicm biasamong tbe cms, althoughcases di- agnosed at older ages, whhl mrarive i d d therapy is kss frequent,me likely u)be urrdtmprr+lented.A lower pUr;cipar;oll RAILamong the mfuents rkan amoqs the cases may, however. h8VC introdtlceddifferentialseis- daa into che uudy with rcgud ta,for example, socio- ocmmmicfaam, aad &erefox resulred in biased e f f a esfixwcs. Sucha seltction w d d be likely to invMJuce a bias in the same directioa for bropd carzgaritsofex- posure. We foundno iabicatimsof such a gencnl shift inthtefFendmawsforindasajstl-utban(g~esnd diesel fir4 and exhausr gases add ELf+=Golr s@cultUr- f-m(pesticides, frefh-m~t-mimalcontact, fresh wood) exposmw (table 2). bar u CBMOI btmlcd out- The resulrs of a nasmafof expsrs\ae to organic solvtms agreedwofl with rhcnsuhs aftbe initial il~ses5- m e a Them wcre rn difftxeruxs in the rtpeement be- cwen rhe c&sSandreferen&. The medim durationof occupationatexposurewith- in dose categaks was dightly higherforcases di-d at olderages(261 yws),and the propdon of oexr-of- kin inrerpiewswas higharhpdmIhe younge~age p u p , but it is ques~0a;lblewhether or nor thio differme LX- plains the sugge!md effM mabrfica~mby age. Oned- temarjveintaprera*wn Lrh;rrAhfLinddQccgohganic solvcn~occprsa ~ a molder agefhim AML withothered- o~.~possibiliyisan~g~cffectfros,slr- p u r e in that it h a beEI snggatd thttse effect a f a k-stagecarcimgm would b~ m4e prmpouacedat an ddor age (58). - Apr-14-01 1Z:ZGpm From- 1 -4 I t - Apr-18-01 12:27n Fron- T-415 P.11/11 &Oh e! at F465 U i ?