Document 0qnM2a8waQQEBneyXQDbp76pd

212 t) TMK Vf.w ENGLANDJOUANAL OF MEDICINE Jan 24 I** CANCER MORTALITY IN WORKERS EXPOSED TO 2,3,7,8-TETRACHLORODlBENZO-- DIOXIN * V Marilyn A. Finorknit, Pm.D., William E. Halfuun, M.D., David A. Marlow, B $ . Laurie A. Piaotelli, M.5., Patricia A. H onchar, Ph.D., Makir H. Sweenpy, Ph.D., Aurr. L. CREirr., Ph.D., Patricia A. Dh.l, A.B., K vli Steenland. Ph D and A n i h o n y J . Sukuua, M .D . nt fr o W /V ) r*ot Background In both RfwmaJ and epldem*> jotyc studies, exposure to dioxin (2.3.7,S-tetrecNoredlbeazo-p~dioxin, or TCDD) has been associated with an Increased risk of cancer. Methods. We conducted a retrospective oohoct study of mortality among the 5172 worker at 12 plants in tit# United States that produced chemicals contaminated vwth TCDO. Occupational exposure was documented by re viewing job descriptions a/.d by measuring TCDO in se rum from a sample of 253 workers. Causes of death were taken from death certificates. Rnufa. Mortality from several cancers previously as sociated with TCDD (stomach, liver, and nasal cancers, Hodgkin's disease, and non-Hodgkin's lymphoma) was not significantly elevated in this cohort, Mortality from softtissua sarcoma was increased, but not significantly (4 deaths; standardized mortality ratio [SMR), 336; 95 per cent confidence interval. 92 to 665). In the aubcohori of 1520 workers with > 1 year of exposure and >20 years d latency, however, mortality was significantly increasad for oft-tissue Mrooma (3 death5; SMR, 922.95 peroent conMsnce Interval. 190 to 2595) and for cinctrs of the respi ratory system (SMR, 142; 95 percent confidence interval. 103 to 192). Mortality from all cancers combined was lightiy but significantly elevated in the overall coho'* (SMR, 115; 95 percent oonfkjence interval. 102 to 130) and was higher In the subcohort with >1 year of exposure nd 20 years of latency (SMR, 148; 95 percent confi dence IntervalJ2J to 176). ConchjtbfM, ^ h l s study of mortakty among workers with occupational exposure to JQDO does not confirm me high relative rlske reported for many cancers In previous studies. Conclusions about an increase in the nsk of softtissue sarcoma are limited by small numbers and misciasaffloatlon on death certificatea. Exoese mortality from all cancers combined, cancers of the respuatory tract, and oft-tissu# sarcoma may result from exposure to TCDD. although we cannot exclude the possible contribution o' Motors such aa smoking and occupational exposure to other chemicals. (N Engl J Med 1991; 324212-ft.) SEVERAL epidemiologic and toxicologic studies have suggested an association between 2,3,7,& tetrachlorodibcnzo-^-dioxin (TCDD), or the chemicalj it contaminates, and soft-tissue sarcoma,1*4Hodg kin' dijesse,4 non-Hodgkin's lymphoma,M stomach cancer,' 10 nasal cancer, and cancer of the liver.1*'1* In oilier studies of these cancers, no significant associ ations with TCDD exposure were found.u``*The car cinogenicity of TCDD has been demonitrated in ind was earned into subsequent production pro.tucs.*1 One de rivative, '2,4,5-mehlorupheiioxyacetk ud, v u widely used m the United States to kill brush and was a constituent at defolianu loch aa Agent Orange. Other derivative induded the herbicides 2-{2,4,$-irichIorophr.noxy)propionic arid (Stive*) and 2.(2,4,3tndilurophcnoxy)-ethyl 2,2-dichluropropiaoaic (Erbon), the inectidde 0,0-dimeihyt 0-{2,4,!i-trirhlorophcnyl1priophofnfhiotr (Ronnd), and the banentide Z.ll'-methyleue-L^. I.kuiililw ophenol] (hntachlomphcnej. toeotiflosrtton of Exposed Workers studies of rats, mice, and hamsters; histiocytic lym phoma, fibrosarvomas, and turnon of liver, skin, Worker Drum 12 companies were included in the itudy cohort if a penonnd or payroll recurd documented that they had Keen al lung, thyroid, tongue, hard paUxc, and nasal turbi igned to a production or miinteuancr job jo a proceai inw ivui| nates have been found.1111-* TCDD acts as a pro- TCDD cooumination (n 5000), or if they had been identified in moier, `,,J and may also initiate carcinogenet3S.,/-n, t previously published inidy on the basis of expwurr to TCDD (e 172).H Personnel records for 202 workers did not reveal the f '. To evaluate the eiTcci of occupational exposure to duration of tboir assignment to processes involving TCDD ountamiTCDD, particularly with respect to the cancers listed taboo; they were therefore included in the analysis of overall mor above, we conducted a retrospective cohort study of mortality among U.S. chemical workers assigned to the production of substances contaminated wnli TCDD. tality but excluded from analyse according to duration uf exposure Sbtty-tcveft women are not included in this rrpon. there were 10 deaths among them, including a single death from canter (lung oaACor). At Mrh plant, w mad* a thorough review of operating cundi- M ethods Mamtfication of CompanJas tkms, job duties, and i nxirtis of TCDD levels to lodustnal-hygiriic samples, intermediate reactants, products, and wastes. Thra review provided dear evidence of potential dally exposure tu TCDD The production of TCDD-euntaminatad substances at ilx various In 1978 the National Institute for Occupational Safety and Health began an effort that wuuld eventually identify the exposed workers at all U.S. chemical com|wnies that had made TCDDcontaminated products between 1942 and 1984. TCDD was gener ated as a contaminant in the production of 2,4,3-trkhloropheuul plants involved similar raw materials, p ro a n rs. and jnh duties ** However, there were differences between job sod beiwrrn plants in the extent of TCDD exposures. Occupational exposure tu suhttancat oontaminated with TCDO was confirmed by measuring erum TCDD levels, as adjusted fur lipids, in 253 surviving mem bers of the study enhort Hwh i- u plants wlio were alio parucip^ms in a related cmes sectional medical itudy * From the Industrywide Siudiei Brandi, Dtawn of SutvciIUm. Hus/d F*luatlmii, and Fwtu Studies. S'snonal Inu.ios* for Ocxwpatiotvsi Safety and HaaiOi. Camara tor Disaaw Control. 47ft Columbia Pky.. Ciauxtuu. OH522ft. wham rtpriw requests sfcwld be sddrMied to Dr. Fwifcrtu Supported in pan by the Ajene) for T*w Substances and Due** Rag- iscy Ufe-Table Analysis Vita) status was determined as of December 31, 1987. Crom rec ords of the Social Security Administration nr Internal Revenue Service. or from the National Death lade*. All death certificates C 00351 0 0 0 4 G7 Voi. H t No 4 (Ta VCER MORTAUTV IN WORKERS EXCUSED TO TCDD -- FINGER HUT ET Al. 213 were independently elat'ifird by two nn90lu5i.1t cumding to the rules uf the reviiion nCthe fnirm aim t C lu n fa tm ofD itaat (ICDJ til effret at fht dam of dratli.*7 Life-iahl* analysis used to c^alMtf monaliiy in the cohort.*" At each plant, the number nf pcrson-yeari at risk waa calculated as the interval bciwrcu the first yitcaatimlly dncumenicd ynmem to a proem invnUuiit TODD enntaminsttun and the date of death or December 31. tft#7. whirbfver nefurred first. Those wliuic vital natui was unknown werr assumed to be alive at the end of the study. Standardized mortality ratios (SMR*) were computed by dividing the observed numbrr of deaths by the expected num ber and multiplying by 100, after smtifictuun to adjust for the confounding eftects of age, race, and year of death. Two* aided 95 percent confidence interval were computed Gar<ach.cause* specific SMR, with use of the flyar approximation for eight dcaihr nr more and Fuher'i exact method for fewer rhan eight deaths.*' The U.S. population w u used as the reference group because the IS plant* west located in II tu ie i thruughout the country. Analytes According to Duration of Expoart and Employment Duration oT uposuiv defined as the number ol yens the worker was employed in process* iawivingTCDl) conuminarinn and wat calculated with data from penonnef record* We used duration uf exposure as a surrogate for cntnulative exposure to TCDD on the basis of the high tom iaiioa of tftr kgsrjthm nf serum TCDD levels with the iugarfthia of the numbri of yean assigned in processes involving TCDt) oiaramitiadnn In uur sample Of 253 wurfcera (Prarron' produCl-thuHttui coefficient r 0.7V) (Fig. 1), and un the assuni|iiion that the prndueiiun processes were Similar in the 12.plant*.1* Because of (he concentration of person-years in the short-dunlion categories, duration of exposure was stratified before analysis into categoric* of < 1, I to < 3,3 10 <13, and /3 yean (TWe 1). Monality was also examined accordutg to time since first exposure (latency) i periods of 0 to < 10, 1010 < 20, and 20 years since fust exposure. To examine mortality in a subgroup with substantial "exposure and adequate time fur eanect to develop, we identified a group of worker* who had 1 yen ur mwt *rf eipM un to processes involving TCDD contamination and at least 20 yean ol latency. One year was choieti as a cutoff point for this high-exposure luhcohort because in the sample uf workers whrwr a m TCDD level were meusurrd, 100 percent of those afpwed h r mwrthaw woeyear had serum TCDD level* higher litas the mean level la the Ufiexpoied reference group (7 pg per gnm (of Bpsd). for this subCobon. the number of person-years at risk was calculated from tha date the person attained both 20 years of latency and 1 year of 'Cxoaeuro. Mott of the 12 plant* were large U.3. chemical manu&cturing aita that produced thousands of chemical*. Complete documenta tion of each worker's exposures wa* impartible. A separate mcaiuro called "duration of employment," defined as the tout time that each worker ms employed at a study pitot, was therefore used. Because of the long tuul employment at the plants, analyses according to duration of employment were stratified into periods of <3, 3 to <10,10 to <15, IS to <20, 20 to <25,25 to <50, and 50 years (Table 1). Fur these analyses, latency was defined as time since first employment. When die SMR* showed an apparent trend aatyuated with dura* tion of exposure or employment and when the observed numb ra of deaths were sufficiently large, we eundueted internal comparisons using directly mjidsrdi/.rd rate ratios and teats fix trend." Fur the standardised rate ratift), die cause-specific monality rate In each of the categories of longer duration was compared with the rate in the category of shone* duration, after iiniificauuit of the rates fur the potential confounding effect or age, race, and calendar time. Rx*tiers The cohort uf 51 72 mild workers from 12 plants had 116,746 persou-years of observation. Table 1 de scribes the vital atatus, race, latency, and duration of. 10.000 $ t.ooo *8 ! too 1 10 0.000S . 11 * 1 * * Vvw* \ ... . - . V i i v * ! i- y - ' A \ * ". * r - 0.72 P<o OOOt 0.005 0.05 0.51.0 SO 50.0 Y arn ol Exposure figure 1 'Je m m Lavala of FOOD, Adjusted for Lipidi, in 253 Aooordlng to Yaars ol Exposure. exposure and employment uf the workers. Overall mortality for ail cause of death was similar to nation* al rates in thiXJtiited States (1032 deaths; SMR, 99; 95 percent confidence interval, 93 to 105). Mortality from heart .disease wax also similar to ^national rates Table t. Vital Status and Demographic and Employ ment Characteristics cl the Study Cohort. VklMSU Vitto to * AKm Daad Unkaowt Toni Dssda" ahiQDM Nonwkiii **a ' Tats) Dato cartlflosa* obtained kmat WWW Nonrhki Uakaowa Tbtal Dumb* of sparo (ytiT <1 lw<3 5 <15 IS Total OuraioD of wnoloyflwnt (yrit <4 Sto <10 10 to <15 IS to <20 30 to <25 2510 <50 50 Total Yean rta fini smura (Uaeeyjt <10 10 <30 30 Tote} Yosn ilm Isa cxpowtT <10 10to <2ft 20 Toul soo at) 1053120) 77 ID 3173 000) SIS (94) 07(6) 1033(100) 10)7(93) 4590(19) sts o i 197 (4) 3172 000) 2697(54) 1437(29) 639(1 207(4) 4970 (100) 2123 (43) 501 (10) 603 02) 403(1) 391 (X) . 415 (1) 530(11) 4970(100) 271(5) 1663 (53) 3056(61) 4170 (100) 4 (9| 1716 (36) 3726 (69i 497UIIUU) "A irfO aM atoll. 1917. C00352 fwtoro SOI totto m to to w*i*iaa w puiw to r TCSOMsniHM v m iw*ow (mm m a tn 04R it THE NEW ENGLAND JOURNAL OF MEDICINE J*u. 24. lOyl (393 death*; SMR. 9C*, 9.rr percent confidence interval, 8 7 to 106). There were significant reductions in the mortality rates for diseases of the circulatory system (67 deaths; SMR, 77; 93 percent confidence interval, 60 to 98), primarily because of fewer deaths from stroke, and (or diseases of the digestive system (38 deaths; SMR, 70; 93 percent confidence interval, 49 to 96), primarily because of fewer deaths from cirrhosis. There were alio significantly fewer deaths from alco holism and personality disorders (2 deaths; SMR, 23; 93 percent confidence interval, 3 to 87). The low mor* tality from circulatory disease may be a reflection of the "healthy worker" effect -- cohorts of workers die at lower rates than the general population, particular ly of causes other than cancer.11The reduced number of deaths from cirrhosis and alcoholism implies that this cohort consumed less alcohol than the general population. Reduction may also have occurred simpK by chance, since numerous comparisons were made between the cohort and the U.5. population. Fatal injuries were significantly more frequent in the cohort (106 deaths; SMR,* 12B; 95 percent confidence inter val, 104 to 154), but they did not appear to he associat ed particularly with exposure to TCDD. Moruliu from all cancers combined (263 deaths; SMR, 115; '.'5 percent confidence interval, 102 to 130) was signifi cantly elevated in the cohort. Cgneers of a Priori Interest The term utoft-tistue sarcoma" describes the group of rare malignant neoplasms arising from supporting tissue other than bone.3*We restricted uur analysis of mortality due to soft-tissue sarcoma to cases of softtissue sarcoma.listed as the underlying cause of death Tabte 2. Cancer Mortality In the Bruir Cohort and In Woricars with Mora Than 20 Yoara of Latency. 1 , 1. i * * i In g fC k iirti ' ICO a * - Gwmt Conor, (N -SlTijT W y e * ! it X Yl t f L t m t (M I flwMtnatni 1n urUMtuu < UitH 1turn tate temte et)crad MINI Sam 4tls temei nfna4 sms; tart talk* atfcrta upKiad SMSjJ All crasi Brasi nd phir)oi Run Otterpc/U WiwtiwoquM Etepfctp* Slanaeb StMfl btMJiw md eotm tciian Lira a Mliciy EmbMO Fcfwwwtji isd uwjwlfied RnpintoijpiTten - Urynt Trsetes. brada. 140-30* 140-149 146-m 143-149 I50-IS9 150 151 193-ISS 154 135, 156 137 Ite. 159 IftU-ldS 161 163 265 229.9 5 7.0 3 3.4 2 1.9 67 7 y 3.9 10 9.T 25 20.4 9 IU 6 13 10 11.9 3 1.1 96 MS 7 13 9 10.1 Il3(l02t30) 70133-166) 1 (IS- 259) 105<13*379) 113(17 U3) 152(70-2SW; 103(50-IVO) 132(79-111) 9(29-209) 116(42-252) *4(40-15 194(22-6M) 113(92-IW) 211 iiu-454, 111 (19-(37) 4* 46. 102(76-136) 114 71.0 146(121-176)** 2 14 145(11-324) 2 2.2 9Q(l!-)2t 2 07 296(36-1040) 0 (.2 Ot--) 0 0.4 0(-) 2 0.6 339(40-1190) ?3 11. 111 (39-119) 1.2 169(20-602) 2N 20.1 140(93-202) 4 20 200(55-31?) 3 1.7 171(37-521) 4 2.9 13*(31-333) 3 4.3 )17(36-274) 13 7.3 ITI (95-304) 1 1.0 100(3-357) 1 1.0 100(3-557) 1 2.4 41(1-232) 0 0.2 0(--> 19 11.4 (01(62-161) 2 0/ 297(34-1074) 17 17.3 96(16-15 2 1.7 115(14-41 4l 1.7 59(1-327) 4.0 100(27-233) 0 0.4 0(-) 43 30.3 M2(10J-NJ\ 3 l.l 261(35-76.1) 40 21.1 139(99-1*91 sndfu* Mile c*niulet|ai Proem* UristfyocgM) Kldatr lliddtf mdother LmWaOc iMhtaKorak % teiui 3-117 IT 113 III (65-177) 115 IT 13.9 122(71-195 TISS-II9 119,0-1119.2 - 11.4 14(16-231) 5.7 1400-27 IN. 9 17 7 (72-291) 119.3-119.9 20O-30S 24 ' 211' 109(70-162) 2 3.2 63(1-22) 2 10 67(1-237) 3 2.4 122(26-373) 3 1.2 233(32-742) 0 1.3 (-) 4 3.9 103(21-260) 9 60 149(61-213) 9 5.9 132(70-290) 6 4.0 (49(53-324) 2 1.9 106(13-314) 4 2.2 116(31-476) 6.4 125(14-247) Hodfkk'i dnuje 301 Noo-Ho4ti*`i (jrmpfcnaufT 200,202 LjnrSwu/cnma 300 ntlteJeuROG&sft 3 119(25-34 10 7.3 137(66-254) s J-3 142(46-333) 0 0.2 0(--) 2 IJ 135(16-4) 0 0.6 0(-) t 0.4 276(7-1534) 2 2.1 93(11-337) l 0.9 107(3-594) Otter baphutetf Multiplelyetonatt Ltetuna sadMutai Otter Hm 203 30 304-301 170-173, s 3.7 133(43-31)) s 3.0 164(33-313) 6 1.9 67(24-14) 39 29.6 .131 (94-ISO) 2 0.9 211(36-779) 0 0.6 0C--1 2 1.6 126(15-457, 5 15 17(22-202) 1 1.4 71(2-363) 3 U 262(54-766) 2 2.6 77(9-277) II 9.0 201 (111-316)** Sttn Bniamdaerrai span Boat Goanicihc duce and tentila Otter md rapvdAtl 190-199 m. m 191.193 170 171 194-199 4 4V 12(13-111 s 7.3 66(22-160) 2 0.9 227(27-1)9) *`4 1.2 33(92-6631 24 141 163(104-241]** 0 0.9 oc--> 0 0 1.3 0.1 OH 0 1--> 0 OJ O H 5 3.1 119(12-372) 2 1.2 155(19-339) 2 1.9 106(13-364) 1 0.2 521 (13-2903) 3 0.3 922(190-26931 10 1.1 196(94-361) ' f t f te hmamumt Ckttqkm*m y flo w a. M w wiet *Mtmmmtmot p*m u m*4. m-- m rtm i at h f h H i SCtctaSnUrttatari ter*6mi te 4mmi rfmijjwnMpravaMMiimqTTIH>cta*M*#ia<*,nt tuiUUi Awtat frate HW t a t a et yen tifiW . &.); mm* n o te ( )* **$*jy*t. 107; U.7W prw w rw i tuli. ta ta 4 ,-v i n |o d , t . |; mat**<X j* n Utjkryea, lt.3; l).134,nua;wrt rite KM! MX uJu bum t e tta kf ta te ttp x m U nriuptM W IO). 0*1 tifltnttxi KOQS. ftteta^ew IM W te*rJtate m rwpta tawIMO;mtate mtate Owrm iNnOni V , h iw ntew t m 4J potm tenM*** utuntb 000353 IS 0 0 0 4 G ' ! 3f i * o" k .Ml VU. m No. 4 CA\*CUl MORTALITY lit WORKERS EXPOSED TO TCDD -- HNOERHU 1` F.T.M.. on death certificate* and assigned to the ICD category "malignant neuplosms or connective and other soft tissue." In the cohort mortality from soft-tissue sarco ma was nonsignificantly higher than in the reference population (four deaths; SMR, 338; 95 percent confi dence interval, 92 to 865) (Table 2). The deaths oc curred at 2 of the 12 plants, with a significant increase at 1 plant (two deaths; SMR, 1512; 95 percent confi dence interval, 183 to 5-162). A review of tissue speci mens from the four men whose deaths were attributed to toft-tissue sarcoma showed that only two were in fact soft-tissue sarcomas (Case* I and 4, Table $).u Mortality from soft-tissue sarcomas was increased sig nificantly in the subcohori of 1520 workers with 1year or more of exposure and at least 20 years of latency (the high-exposure subcohort) (three deaths; SMR, 922; 95 percent confidence interval, 190 to 2695). Two other deaths in the cohort (Cases 5 and 6) were attrib uted to soft-tissue sarcoma according to hospital rec ords, and one of them (Case 5) was confirmed by review of a tissue specimen. These two deaths did not contribute to mortality due to ofi-tissue sarcoma in our life-table analysis, because the deaths were as.signed other ICD codes. We are aware of a seventh death from soft-tissue sarcoma, which occurred in a group of 139 workers will) cfaloracne who were ex cluded from the cohort because they did not meet the entry criteria. In the cohort, the SMRs for the other cancers of a priori interest were noftsignificantly increased (Table 2).-There were no deaths from nasal cancer, although approximately one was expected. In the high-expo sure subcohort, the SMRs were oonsignificamly high er !or Hodgkin's disease and stomach cancer and low er for non-Hodgkin's lymphoma and cancer of the liver, biliary passages, and gallbladder (Table 2). A Posteriori Plntfngs A small but significant increase in mortality due to all cancers combined was observed in the entire cohort (SMR, 115; 95 percent confidence Interval, 102 to 130). In the high-exposure suheohon the SMR w* 146 (95 percent confidence interval, 121 to 176) ( IV .^6$, ^ 12 plants, mortality from all cancers combined -was increased; at one of these plants the increase was statistically significant. Mortality was significantly higher than expected in the category of cancers of unspecified sites, which included those of rare sites not included in a category of the lile-tablr analysis and those for which no primary site was listed on the death certificate. Hospital records, which were obtained for 96 percent of these cancers, revealed no particular clustering according to site. The cohort had a nonsignificant increase in mortal ity from cancers of the trachea, bronchus, and lung (ICD code 162; SMR, H I; 95 percent confidence in terval, 69 to 137). Mortality from cancers of the respi- riifotysysiem (ICD code 160 to 165) was significantly higher than expected in the high-exposurc suheuhon (SMR, 142; 95 percent confidence interval, 103 to 192) (Table 2). To eitimate the effect of smoking on the increase in lung cancer, the expected number of lung cancers was adjusted according to the smoking preva lence found in lifetime histories obtained in 1987 by interviewing 223 workers from two plants.'9This ad justment increased the expected number of lung cancers in the overall cohort by 5 percent and in the high-exposure subcohort by l percent, which reduced the SMR in the full cohort to 105 (95 percent confi dence Interval, 65 to 130) and in the high-exposure suhcohorr'to 137 (95 percent confidence interval,.98 to 187). Analyses According to Duration of Exposure and Employment The study cohort worked a mean of 2.7 years in processes involving TCDD contamination and I2.fi years at the plants. The high-exposurc subcohort worked a mean of 6.8 yean in processes involving TCDD contamination and a mean of 19.2 years in total employment at the plants. The numben of deaths due to the rare cancers of T dS i horn Stot-Ttuut Saroom t among Cam ra*j NO. tm o r m Tffaitf lam tnr TlU Hm Y&ua lim n Y1U ur O bn UTVor W Cut Bum tMkP.Wit! 1 lW6~ I67t TC? and ttso 197S 21 MFH MTH MFH Mr 2 I94A-1972 IU?*U 1,4,f-T lew 7. 1972 34 tip eu iw ee LifHMCOUft Caniflum. 4ifrti*<K><adt s 19SO-I97S TCP IMS ijs l?S 13 Ftbnufcowa ftbfnwcoma Rent) amffc<mat 4 1931- I 9S2 TCP HOI M.f 19) MFH Mm um St 1*43-1913 TC? or kuftvxsat 1Maowo 1940 UakMwa CveiautuiMiil Myxoid nwnten. LammyeuiToau 2.4> T m tm om * 41 1941-1964 TCP 1949 U okisn IKS 1 M ttn ric otsofr- ftbrounoflia Not MffQAftt *Cum i ) Ihm mm a..Itikwftm." nw m i ptriauil, SmuiooI mftHMiie*IT Aon Oil lapunaatImt. mcwadrfOMat)TOOT**** J.O -T , ML M MIK a r fu ia S m i hito pom. tThjw&o*AfUairwaia4Mtt. fS a a a i ** la w nco*t o<n r taTCOO a a ivla0h. *4 (hi m m an m* < i*. < IV*feutli Max,ipaLr, t*abMa J 4*. ICPdaaMa*1. 3-, tc--i--4a * *>iwihrea, tanwm i 1Qa^> *aa m wftfcaaS ta * a a a a* i>r*** -- ... fit 10 4  * - r- -J* J4 24. Tads 4 MortalityfromAHCancars and fromCancorf cl toTrachea, Bronchus, am Lung, Accordingtoutortcv Ptnoa snd Duration of Exposure to Processes Involving TCDDOontsminetlonA 1 Said* AfiOBM <10 Yt in<30 Yt *30 Yr Tttl M Tndm*. broodut, mi lung <10 v 10to*'30 Vr *20 Yr TohI sax im*S SMX MS* SMS 10 1 21 IM 4* 102 10 VI 100 S 71 IS ft 3V tost IS U7t 127 3 77 6 6V 1? 96 to too 1 VS a 19 IS m 25 in 109 i n a A sua * ohNiW on (M U U 44j cm' TM M Hi* j 71 ts 122 yt US St I2S 123 0T 0 340T IS US 21 69 162 70 m 129T 25 ut 132 not 129 i 79 s * 110 IS 140 24 ISI 106 00 1 137 9 136 10 134 136 7 4 21 101 57 123 IS HZ *OuMiX*I*h*trtea, ittwtiitAgf uywiriii to*uto| TCDD *4m tmtaMtt a m i nunlt. TV*n o t otntmd Sued vwUUh UimAee *T6n stiytiy 6we *wi 1*U L SM ifem i iw ten* .J**. tfCC.CS. * |P<U.0(. ft priori interest were too small to permit meaningful logical diagnosis and rolsdassificatiott on death certifi analyses according in duration. For all cancers com* cates. Consequently, the interpretation of the in- bitted and for cancers of the trachea, bronchus, and* ' creased mortality from soft-tissue sarcoma in mu- lung. Table 4 shows the distribution of mortality with study is limited by the small number of cases and tin* increasing duration of exposure to products contami . Act that the cause of death was sometimes iniscUsw- nated with TCDD. The standardized rate ratios were fied on the death certificates of the workers (Tabic 3) increased in the strata of longer duration for both and in the U.S. comparison population." these categories, but significant linear trends were not * Several case-control studies have found significant found. Mortality increased with increasing latency for fourfold increases in nnn-Hodgkin's lymphoma in per both these categories of cancer. Table 5 shows the sons reporting exposure to phenuxy herbicides or distribution of mortality for the same categories with chlorophenols, some of which contained TCDD.0* increasing duration of employment. Significant linear The magnitude of the increase in mortality in the co trends were not observed for either category with in hort described here (SMR, 137; 95 percent confidence creasing length of employment, although standard Interval, 66 to 254) suggests a smaller increase in thi* ized rate ratios were higher than expected in several risk, or no increase at all. Mortality was not signifi strata of employment >20 years. Mortality increased cantly higher d u n expected for other cai/ u s of . with iutieaimg latency for both categories of eanur. priori in terest* liver and stomach cancers and Uorit- arum Uvals of TCOO kin1 disease. No deaths from nasal cancer were ob served. The inconsistency between the results report The mean serum TCDD level, as adjusted for lipids, ed here and thou of earlier epidemiologic studies is in the sample of 233 workers from two plants was 233 .accentuated by the longer and probably greater expo pg per gram of lipid (range, 2 to 3400) (Fig. 1). A sure of this cohort to phenoxy herbicides and ehlorn- mean level of 7 pg per gram was found in the compari phenols contaminated with TCDD. son group of 79 unexposed persons, alt of whose levels Mortality from cancer* of the trachea, bronchus, 'were under 20, a range found in other unexposed pop and lung was nonsignlficantiy higher in the cohort. ulation*.1*The mean for 119 workerswith one year or Among the workers with 20 yean or more of latency. more of exposure was 418 pg per gram. Alt the work mortality from respiratory cancer was significantly in ers had received their last occupational exposures 13 creased in the high-exposure subcohort, which had to 37 years earlier. I year or more of exposure (SMR, 142; 95 percent D iscu ssio n confidence interval, 103 to 192) but not in the subco hort with less than 1 year of exposure (SMR, 103; 95 TCDD, widely known as dioxin, has acquired the percent confidence interval, 62 to 161) (Table 2: reputation of a potent carcinogen. Our study, al SMRs for lung cancer arc known to be somewh.ti though limited in Its ability to detect increased num higher in bluc-collar groups than in the general U.S. bers of rare cancers, found little increase In mortality population because of more dgarette smoking in the from the cancers associated with TCDD In previous blue-collar groups.91However, the increased number studies in humans. The exception was an increase in of lung cancers in the high-exposurc subcohort was soft-tissue sarcoma. The difficulties of evaluating soft- probably not due to confounding by smoking, for sev tissue sarcomas, in a cohort study of mortality have eral reasons. First, other diseases related to smoking been described.93 These include variability in patho were not more common than expected in this subeo- C00355 on 047 U ll ** .. * . * -- * w Vcrf. n< No * Ca M.'LR MORI AL1TY IN WORKERS EXPOSED TO TCpI> -- MNOtRHUT KT AL. ; T A d S. Mortalityfrom AH Cancan and from Cancan of (ha Trachea, fiivnohua, and Lung. According to latency Piriod and Duration of Employment at the Study Plant.* C*tnaL*rMn tm m AUcmcot <10 Yr 10 k><20 Yr >29 Yr Tout SAX TfIChtt. bRMKfUU, a a d l | <10 Yr 10to <30 Yr 20 Yr Total SMt Oua*tiOMur EMiMUBunra) g - IV-1 lira a a su m ua 4mA* Mtt*4 CMS Mm fllwwit SMS MfHi *wr-t4 SMR 4mM tbMfttf IMS JMth1 damn*a SMR dead), s*ur-- SMR ikutu n--PKJ SMR 0-- IU 4--11 80*0*4 SMR 10 IS 1 II 0 0 0 0 0 U Q 0 0 0 II 64 21 114 3 120 12 10) so 0 0 0 0 . 0 0 40 103 40 13 IS 140 6 70 IS 91 M 134 31 no 54 l)3t 193 1251 71 IX) 21 104 11 19 23 91 34 134 31 110 54 iWr 232 110 UW w 01 70 . W 14 113 J 103 S 12 11 102 19 9S too 1 74 0V 2 Si * 40 tt. D0 3 ijy 2 05 7 103 91 0000u 0 0 0 4 122 0 0 (I 0 0 0 3 59 12 133 IS n o t 19 126 7 11 U 133 It ISOt 19 120 m '/i174 > 14T M 4 V4 44 9 07 It7 IS M2 n,i iiH Tr\ti 0.9 0.6 K tiM nltt mmatt r IK S at. At tnMMi of h IVfRMm fewrUf ?uju n c m Io m w m *MUhUf m i --R # yea. SM imam itaatniia na mw is-.O.oi. hort; mortality from nontnalignani rrspintory disease (ICD codes 470 to 478 and 490 to 519)) which is often associated with smoking, was lower than expected (15 deaths; SMK, 96; 95 percent confidence interval, 54 to 156). Second, in the exposed population with 20 years of latency, whose members presumably shared similar smoking habits, the increase was confined to the high- exposure subcuhori. Third, on the basis of empirical evidence frum other studies, Siemiatycki et al. have shown that between a blue-collar population and the general U.S. population, ronfuunriing by smoking ii unlikely to account for an excess risk of more than 10 to 20 percent. Finally, a limited adjustment in the risk of lunj cantcrJ" 'wbased nn the smoking prevalence of surviving workers at only two plants* did not substan tially change our results.3* Although confounding by smoking is unlikely to explain the higher rate of respi ratory cancer in the high-exposure subcohort, it re mains possible that the increase wasklue to confound ing by occupational exposures odier than TCDD. For example, asbestos may have contributed to mortality from lung cancer in the cohort, since two deaths were due to mesotheliomas. ' An unexpected finding was the small but significant increase in mortality from all cancers combined. The observed increase is consistent with a carcinogenic ef fect of TCDD. For all cancers combined, mortality was significantly higher than expected in the entire cohort, more pronounced in the high-exposure subco* hon, and increased at 9 of 12 plants. With mortality from cancers of the trachea, bronchus, and lung ex cluded, mortality from all remaining cancers com bined was still higher than expected in the overall cohort (SMR, 117; 95 percent confidence interval, 100 to 136) and in the high-exposurc subcohon (SMR, 150; 95 percent confidence interval, 118 to 189). Con sequently, the increased risk for all cancers combined is not explained by smoking or by increased mortality due to cancer of the trachea, hronchut, and lung. The generation of tumors in a number of organs in animals exposed to TCDD11119 and the demonstration that TCDD promoted tumors in two organs31'1" make it biologically plausible (hat TCDD may produce tu mors in more than one organ in humans. Moreover, a significantly Increased SMR for all cancers combined is unusual in occupational studies of chemical work ers. Results similar to ours were observed In a study of German workers exposed to TCDD after a 2,4,5-trichlorophenol reactor accident in 1953. A subgroup oi workers with chloraenc (used as a surrogate for expo sure) and at least 20 years of latency had an SMR of 201 (90 percent confidence interval, 122 to 315) for all cancers combined, based on 14 deaths*'H iis is the only other industrial cohort with both substantial ex posure to TCDD and a long period of latency during which mortality was examined. Workers from U.S. production cohorts described in previous studies were included in the current study if they met our entry crilexia.4042 Two observations argue against a carcinogenic ef fect of TCDD. First, there was not a significant linear trend of increasing mortality with increasing duration of exposure to products contaminated with TCDD (Table 4). However, our use of duration of exposure may have misdatsified the cumulative dusc of some workers. Zn addition, a doio-rejporuc. relation is gen erally viewed as strong evidence for an association when it is present, but as fairly weak evidence against an association when it is absent.1*1Second, our study dld not directly assess the effect of exposure to TCDD alone. The workers were exposed concurrently to the chloropltenols and phenoxy herbicides that were con taminated with TCDD. In addition, they may have been oeposed to numerous other chemicals while em ployed at the plants. Because the exposure of our cohort was substantial ly higher than that of most nouoccupational popula tions, the estimates of effect in this study may provide an upper level of risk to be anticipated in humans. Fur several types of cancer previously associated with 000356 t-.lt! 21 THE NEW ENGLAND JOURNAL OP MEDICINE Jin . 2 4 ,19iM TCDD, wft frmnri no incrdaR$ above expected levels. Soft-tissue la rc n m a was an exception-, ninefold in crease w u found among workers who were exposed for 1 year or more and who had at least 20 years of latency. Interpretation of the increased SMR is limit* ed, however, by the small number of cases and be cause this cause of death was sometime misdassified on the death certiorates of the workers and In the na tional comparison population. Continued surveillance of the cohort may provide a firmer estimate of risk. Mortality front all cancers combined was 15 percent higher than expected in the overall cohort. The subco hort with 1 year or more of exposure and 20 years or more of latency had a 46 percent increase in *11 can cer* combined and a 42percent increase in cancers of the respiratory tract Although the study could not completely exdude the possible contribution of ocher occupational carcinogens or smoking, the increased. mortality, especially in the subcohort with one year or more of exposure, is consistent with the status of TCDD as a carcinogen. We are nrirbifd tu die National Institute for OrcupaiicniJ Safety and Health aiatiitiral drrks, Steve Green, Joyce Godfrey, uud oth er!, fur their trdmical contributions; to representatives yf the com panies and unions lor assistance in gathering the data fur the study; to nur colleague* at the Center for Environmental Health aud In jury Control, Centers for Disease Control, for analyato of the serum samples; and to Lawrence Fine, David Brown, and the memhen of our ue-ribban review panel for their helpAil advice. RxnmaNCES 1. HardeHL..SrtnhmMii A. Ca**<yotrd xudy;aofutim umomw tadaipo* nae to pMnoxvacetic andor ehluroptettl*. hr J Cancer 1979; 39:711*7. 2. Erikjuq M. HstidJ L. Lim w sponn dieaim as a dab fauor ft* oft tissue tvuraa: population baaad caae-contml study. I Nad Cmoa Im ipgo; *2:4*640. 3. EriklaonM, HardSti L, UcfgN, Matfei T, AtsIwNO.Seft-dmiseumonui sod exposure tn ctwmical ubiiancer a cue-referent study. Ir 1 tod Mad 19SI; M'^7-33. 4. KwdcU L. Citkiseo M. TUs aMiaUloe bemeen soft tiswa urcomaa and exposure to pbwwyicodc adds: a ate caliefcrert a a tf. Case* 19*1; 62:632-6. 5. Harden L. Bengiiaoa NO. HpidtmWogfe study of weioacooomie fa u n and cli&kal findings la Hodgkin's dhiM*. md reenalyiii.of previous data regarfing chemical sure. l r J Caoear 19*3; 41J 17*23. #. HanJMI L, eiikiaoa M, Unnsr P. Uadgrea C, MaligMai lymphoma and eapoaun tochnuiali. capedally oryaale iol*au, chknphenota acd fhcaooy andc a case-control Mudy. l r I Caacar IMl; 43:169-74. 7. WoodsJS, PoltarUSewremMLHearer 14, KulaMer 1 0 .leftttiue sarcoma sod ws-Hodgkin's lymphoma la nlsiloe sophonoiyharbieideand chlorinated ptiaaot txpmure la wcatan Washiapon, J K*U Caacet lass 1917; 7l;l99-9tO. I. Poftsm B, Dtftlaadar A, Rudrfksaon M, Brega HN, Ohtsen CG. Aiaboa O. Matignut ty p h o a n sad occupational exposures. Bri tad Mad 19*9; 4dJ16.au. 9. Aialaoo O. SiadeO L Afttonra X, Edliag C, Kogamdi C, Ritag H. Herbicide Mponn ud hum* mortality: an updatMspMendotugk feweatigatiea onSaMUb nilread carvrfcen. StandJ WwfcEavironHub* 19*0; 9:73-9. Jtt TOcreAM. Frenrel-Dayrae X, LtokR. Morality xMy of penona aapoaed to dkoia In a trieWwopheael-pHwi** acridesi tore ocewnd la do BASP AQ os November 17, 19)3. Am J tod MM 1912; 3:17949. II. KankR L, Jofcaatam B. Autsoo O. Gptoamolopnl My of aaaal ud BaanpiMuynguI eancar sad their relation n pteooxy add or chtonpfaangl exposure. AmJ lad Med 19*2; 3347-J7. 12. kuciba X. Kayto 0. Beyer J, at al. Breaks of two-year chronictoaieity andeoeofwicliy studyof2,3.7.<Asochioi'odibaro*p-d>cxtnla rats. To*. ICO) Aft* NvamwcoS 1971; 4A279-30J. 13. Nattoaal Toaieology Piegreni (NTPJ, Cvrino|ntil) biousay of 2,3.7.1utfacUoradibeafttwdioxin (CAS No. 1746-014) to Osboma-Mindai rats aad B6C3PI mica (givagc study) 1912. Washington. D.C.: Cownuneai Ntodag OfUcc. 19. (DHHS pubSicasion ao. (Sill) 12-1765.) H. Smith AH. Puna HE. nwter 00. Ones HI. Teague CA. Howirt )K 5of: tiiwa aaieoma aad aaaoaun to phcaoaybcctikidvi m i chiorontwruii ir. Naw Zaabod. J N1 Cttoar toat 1964; 73:1111.7. IS. Wlkluad K> Hols L- Sort dssue saivusna nak la Swedith isnsailtural sne lorestry w to tn . J Nail Uneer In I9tt6; 76.229 34 . n s , oncpjwB ha, aiiuui ah, imgi'; i.a mon-rKOSktn Slynipiai ^ MiMisreung:iaaipudedcuc<tmtrul stud). 1mi Cancer l9|7; 39:133- I t. . Dieb I, Huho LE. Mifc of malignant lytnphoma to Swedith paadcida ipplton. Br J Cancar 19*7; 36:5034. II. Otoea JH. Jtnsaa OM. Ntul caneer and dKompbenoU. Iannt I9M. 1~. tf. Kstiril L;Beegsaon S , lotisoo V, Eridtisoo S, Uruoa L. Aatiolegical aspsoto m prtaB y'bw caocsr wito ptdal regard to tkoboi, orguic aolventa and acvta Itnennlitani poiphyria -- as epidemiological inwvtitS' lion. B ri Caucv 1914; S0.Jw.97. 30. haoMS. Subhano V, J*iuad iO. IcatpelU DC. Caretnogemviiyof2.3.7.*. lacichlorodjhT sp-dlciin to the Syriao inliton hsoiaur. Caicinoi&vii. 1911; 9:14774. 31. Wand A. P ita D, Oover l . Tbaour pneMign by TCDO is skin of HM/I bditea ntoa. Ns&ire 19*2; 300:271-3. 22- fUesHC.OeldiwfldhyT,CimpbaUKA, PolandA. Qaasnlutive evaluation Ofth* pnototion by 3,7 .l^ffnchtoradlbann^p-dioito of bapatocaruiiuvftoesto fb n jaftyWnamina. Cueer Baa 19*0; 40:3416-JD. 33. BspoaUo M. Henan T. (hydn F. Dutii. 19*0. Ctocioaati: bduxiHji Bavboeaaeul Unaaidi Ubcmory. 19*0. (BPA dBbtieetton 4007 K(u 1V7.) 34. fusMad U t, Harutng V*. K uua Bcatih ftocu 2,4J-T asto lia luaic CQCBsndf--IS. 1AMA I9U; 331:2372-10. IS. .Ptopslgis 14, llalpafb W, Mwiow D, es at. Mortal^ aotoog (J.3. wnAara mptoyad Ifl On pseduetios of chemiealt oumareinaiad with 2J.7,|< unvbJorodlbtaco-p4ioato (TCOO). NIOSH Flotl Rtpon PB9U17<71 ffpriflfScfd. V.; National Ttobofcal Infamation Service. 1990. 26. Sweeney MH, Afaifeui MA, Coooally U). as iL Frogrsn of toe NIOSH boss-sectionat midteal dtidy of woriers uxupattoaaT.y noted to ctuw- cali oontatnlastsd with 2 J,7 4 'TCZ)0. CtetHxphct 1919; 19.-973-7. 37. World Htoatiti Organiutioe. loumational elaulfteatton of diteascs. manual of too ituruatioMl stailstlrif elutifleaunn of disuses, injunct, and caum of daato. Omavai WM llaalto Oigwiatoam, 1977. 33. SteenUnd K. Baautoont 1, Spatob3. ct at. Nowdsnotopmanu totoe NIU3H Ufatabla aatiyiis sysum. J Owup Mad 1990,3M09i>t, 29. Ratoaut X, BoiseJ. Cpidcauslogic andyaii with a prvgnmnuhJa talcula* tor. WsattiegHe. D C.; Govemmans Printing Odtoa, 1979:30-). (DHEW ptblieadui ao. (NIHi Tl-'f*'/ 30. XotoiiuuM.Mtricnatiktotrii'li'gy Ru<um uuie, ...* tJ4K22,,l )3a> 21. ftot Ai. Cbtllar f t . Law monaJiry rvtea ni tndusolal cctuc itudiea due to aatoctioa (be wurk mA am inl m toe toduatry. I r J Pi Soc Mod 1976. XU03-30. 32. Salt HD. Saraaaa of loft drew. CA i*7S; 3*94-93. 33. FtagehaiMA.IUpwtAWe.HoDehirPA, SotitoAB.OrotoDH. XukmII WO. An efvahiatioa of mporu of dtoaia eipoaure u d soft ttoiue sirvoma fiatooloiy imoog etantoal - n t i n to too Uoitod Suua. Seand J Work t i n t Health 1914; 10499*303. M. PlUnoaDair,Ra|wbtoMA,ltObmDlt,eito.Levalinfpolychtorinat. ad s^haaco-p-dtoaifli and dibootofvasa to v i m aapoaed to 2,3,7,a. (MlemdtoOMfrMtoria. Am Med MM 19*9; 16:135^6. 39, IWcyC. JtoaeftSlU.GIaetierL.AcaBieyofeaoordeatocaniflctU'iim' kisffacsoacaacaraartilifydaiiuka. AmJPuWlc Health I96l;71:242 M. 36. Sk&iityeldJ, WecfcaldcrS, DvwvR. Cardii li, Greenwood C, RicUMtn* L. Dagrea of coefouadtog Maareiatad to sawUog. ethnic group, and hkio* osaomie u n to ertmaw of toe samtoatioa btwn ompauiM srl m m . i Oecop MM 1MI; 3Qc617-23. 37. Aulsofl 0. Asnada on coafandtag to occuprteaal hwirn epidcimji^\ Seand J Wash fcwboe Haelto ti ll ; 4:91-102. 3*. I manland K, Beaumont I. Hslpacta W. MathWi of cobom for smokmit wramrional ooMt aortatoy sutotos. Seand J Wort Envirrw Itcilto I9t-. 10:1434. 3*-. Cabar A, Mm i w P, Haber P. TLiity-faw- y r morullty fnik .,.r ,. BASP employ aipuMd to 2.3.74-7CDD aftor Uw 1933 aecukni 1 Anh Oocup Environ Haalto 1990; 62:129-37. 40. Zack JA. laskiad 1. Tba mortality aapsrlaaoe of woman oipoicd 1*. tncblondlbmmliaria la a vkhtorephanul psoow to r n :.) Oecup 19*0:22:11-4. 4 |, Zadt JA, Gtffey Wft, A aumlity tody of wushsn employed 1 tor Mori' s o Company ptom to Ntoo, Weal Virgini. Environ Sci m ivs* 36:373*91. 1 42. OaMO, Olaoa BA.Cook U , Bead OG-Cntom mortality xud> at chemi cal worlun wlto posantial npnwro to toe htgbn chlonnaiad dmm- J Ocevp Med 19*7; 29:4224. 42. Monsoc BA. GbonMoaal itomolagy. Iw a Baton. Fla. OK' t>. 1990:11 C00357 6 4-73