Document JrzvQRd7Ok0yy0aw1eX5j55ma

I MiihiU"H Ht'WtJnh. VS (IVX^) V7- I<KJ pNoMcr Ihomedkul Press f (TcP~HG^> 30 fi (/) INTERNATIONAL COMMISSION FOR PROTECTION AGAINST ENVIRONMENTAL MUTAGENS AND CARCINOGENS o ro CJ1 I CP BMC Working Paper TG1/2 */79 Mutagenicity and teratogenicity of vinyl chloride monomer (VCM) -1* Epidemiological evidence i J oh a n nes Clem me.se n ! Slotkht>/tn\^uJc '/'M < nju'nlui^fit, J)< innink (Receded U' September I VS 1) (Accepted 21 September Ivxl) The rarity and special character of angiosarcoma hepatis and its association with exposure to vinyl chloride monomer (VC'M) has occasionally led to the error that a case of this tumor is tantamount to such exposure. Apart from the possibility that a number of such cases may in the past have been taken for cholangiocarcinomas, the experience of a British team of histopathologists going over cases from Britain during 1963-1973 has revealed some over-estimate of the frequency of this lesion, with the additional experience that only I of the r/greer/ 14 eases could be confidently associated with exposure to vinyl chloride (Baxter et al.. 1977). It follows that exception must be taken to the conclusion that single eases of angiosarcoma hepatis from the surroundings of polyvinyl factories may be taken as evidence ol an escape directly from the plants or otherwise. It would, however, be important if mutagenic or teratogenic effects of VC M be demonstrable cither in the surroundings of factories or in the domestic' environment of workers employed in PVC-producing facilities, and this possibility is the subject of the following review. < i f * ll has been agreed to publish this document as a working paper fir Tjsk (iroup t of the In(ern.icnftal Commission for Protection against Environmental Mutagens ,inJ t aremogens <K PP.NU i I he views r, \expressed are those of the author and Jo not neeosanlv represent those of the C ommiNM.-n Ihcv un i> published to stimulate discussion and comments which will be vscUmned hv the author All correspondence and reprint requests >h*>uld be addressed to ilk scvrctarv of !< PEM( Paul 11 M } l.oilman. Ph I), Medkal lJiologis.al I-iihoratorv 1NO. PO Hos 4N A A Ri)swi)K ( I he Nether lands) Tel. 15-1 .^K777, telex 3Sf'3-1 pmlno nl (K'PLMt Jnumcni 1?' IVM-I5N it ' -r*r 0U^-lim,X2 (KMK) '(KXH)/$02 75 - Elsevier biomedical Pres. J * * *_ - litaa. *, . . f i 9S t i An investigation of a possible increased occurrence of birth defects in the surroundings of PVC plants or among the children of workers has been attempted by Infante (1976) and Infante et al. (1976). The studies were based on data from 4 Ohio communities, of which 3, Ashtabula, Paincsvillc and Avon Lake, had at least 1 polymerization facility in operation, Ashtabula since 1954, Avon Lake since 1946, and Paincsvillc since 1946, with a 2nd plant opened in 1967. The 4th community. North Ridgcvillc, was near to Avon Lake and was without PVC production. Population numbers ranged from 24000 in Ashtabula to 12000 in Avon Lake. Between the census years 1960 and 1970 the population of Avon Lake increased by 30%. wheteas the populations of Ashtabula and Paincsvillc remained about the same. According to birth-certificate data for 1970-1973, it appeared that while the rate R&S 024451 of malformations for the entire State per 1000 live births was 10.14, the rates in the index communities were : Ashtabula, 17,37; Paincsvillc. 18.10; and Avon Lake, 30.33. For these 3 cities with PVC production facilities the differences between observed and expected numbers of malformations in each city were significant at the /'<0.01 level according to the x2 test. Nevertheless, the highest rates for malforma tions were found in North Ridgcvillc, 27.26; and in Geneva, 12 miles from Ashtabula, 25.40, both cities without PVC plants (Infante, 1976.) Infante, furthermore, found an excess of CNS defects among still-births and live births front the index cities, with the exception of Avon Lake. Most of the excess, however, was attributable primarily to Paincsvillc and secondarily to North Ridgc villc. It was therefore obvious that the findings did not link PVC production with the said anomalies. Edmonds et al. (1975), observing that the increase reported was not uniform and appeared more prominent in the Paincsvillc area, analyzed data collected through the Center for Disease Control's hospital-based Birth Defects Monitoring Pro gramme (BDMP). For 2 hospitals located in cities with polymerization plants. Pottstown in Pennsylvania and Paincsvillc in Ohio, they compared CNS- s- < malformation rates for white infants born during 1970-1974 with the rates for white infants in each State. No increase was seen in the Pennsylvania hospital, but an increase, primarily in anencephaly and spina bifida, was noted in the Painesville hospital amounting to 22 cases, or twice the expected. After inspection of the birth-defect registry for Ohio, 1 more case could be included in the study, bringing the total to 15. Interviews with parents revealed that none of them had worked at cither of the 2 PVC plants in Painesville. However, 2 of the fathers of controls had worked at 1 of the plants. A significantly larger proportion of control mothers than case mothers worked (including housewives) within a 10-mile radius of the PVC plant, which was probably a chance occurrence (95% confidence level). It was concluded that, although the follow-up confirmed a.moderate increase in CNS malformations in Painesville, Ohio, no association had been found with vinyl chloride exposure. A further study (Edmonds, 1976) of data from hospitals in Pottstown, Pennsyl vania, and in Painesville, Ohio, revealed no difference between the cases and the ? rc''*fjz---mrX controls in possible exposure to VCM. Residential histories showed no difference between the cases and controls when compared at various distances from the PVC plants. In Kanawha County the author found a difference in the pattern of residents for cases and controls living within 3 miles, but the available data suggested no association with VCM. In a personal communication, Selikoff reported that he had estimated fetal deaths among wives of workers exposed to VCM at 7--14 per 100 pregnancies, although without the use of controls. Referring to this statement. Infante et al. (1976a) attempted a comparison of pregnancy outcome among wives of VCM workers before and after exposure, compared with wives of PVC workers and a similar number of rubber-factory workers matched as a group to the VCM 'workers by age. A total of 95 VCM-polymerization workers and 158 ruhber and PVC'-fabrication workers was interviewed in October 1974. Interviews were conducted with workers, not with their wives. Questions about pregnancy outcome were contained in a much larger interview questionnaire, which was the initial item of a cross-sectional health survey including physical examination. No data were obtained on maternal age. but, on the basis of paternal age. fetal death rates for the primary VCM exposure group were age-adjusted to the control group separately before and after the husband's exposure. u' `)T It should be mentioned that Paddle (1976) asked for tabulation of the raw data r; before analysis finding that age adjustments appeared to have influenced figures misleadingly. From these data published without delay by Infante et ah. (1976b). in combina tion with those of the first publication, the following observations stand out. before exposure, the mean paternal age at conception for study pregnancies was 26.4 years and for controls 23.0. with crude fetal death rates of 10.1 an 6.9%, respectively. Because fetal loss is known to increase with increasing parental age. the fetal death rales for the primary VCM exposure group were age-adjusted to the control group. This reduced the rate for the study group, before exposure, from 10.1 to 6.1%. Contrarily, the raw rate, after exposure, of 16.5% for the study group versus 8.8% for controls was only slightly influenced by age adjustment, showing 15.8% for the study group, so that Paddle's objection was fully justified. The asymmetry in age, before exposure, between study group and controls, is r paralleled with an asymmetry in numbers of families which, before the husband's exposure, numbered 70 for study families against 62 after exposure, versus 95 and 159, respectively, for control families. Further objections were raised by Downs et al. (1977) in a critical review prepared for the plastics industry. They pointed out that matching by age should have been based on age prior before employment not on age at the interview, and that age adjustment to the controls' standard, made separately before and after exposure, does not justify comparison of these 2 values. The use of Mantel-Haenszels lest, which requires independence of the 2 rates being compared, is also found incorrect. Therefore, when the evidence is weighted, it seems that there is no demonstration so far of an effect of VCM as alleged by infante et al. 37 P CO O to -b cn to * qgts,& rL--p^ .,'?, lafr<^Ajt4fciCip,wi, .,, fLV" lri *i ! rJ 100 References Maxtor. P.J,, P.P, Anthony. Me Sween ct al, (1977) Angiosarcoma of the hscr tn Groat Bn tain 1963*1973. Dr Med. J., 2. 919-921. Downs, T.D., R,A. Stallones. R.P. Prankowski ct al (1977) Vmsl Chloride. Hirih Defects ard Petal Wastage, The Society of Plastic Industries. 1977, Udmonds, L. (1976) llirth defects and vinyl chloride. Proceed on Women and the Workplace. Cnf. Jan 17-19th. Washington. DC. pp 114-139. Pdmonds, L.D., II. Palk and J.I;. Nissim (1975) Congenital malformations and vinsl chloride, Lancet. 2. io9k. Infante. P.P. 11976) Oncogenic and mutagenic risks in communities \siih polswnsl chloride pn-ducnon facilities, Ann. N.Y. Acad. Sci. 271,49-57. Infante, P.P.. J.K. Wagoner, Me, Michael ct al. (1976a) Genetic risks oT sms I chloride. Lancet, 1.734 -735 and I2K9-I290. Infante, P.P., J.K. Wagoner and R.J. Waxwcilcr (I976h) Carcinogenic, mutagenic and teratogenic risks associated with vinyl chloride. Mutation Res.. 41. I3I-I4L Paddle, Ci.M. (1976) Genetic risks of vinyl chloride. Lancet. I, 1079 r % 3 ? 4 k r {iXt R&S 024453 i