Document X7RNyRO0Ra6BmmyLk39dYKBqd

American Journal of Industrial Medicine 23:313-319 (1993) Mortality and Cancer incidence Among PVC-Processing Workers in Sweden Ingvar Lundberg, mo, Annika Gustavsson, BSc, Bo Holmberg, PhD, Gustavo Molina, mo, and Peter Westerholm, md The mortality pattern and the cancer incidence were investigated among 717 men who had been employed for at least 3 months during 1964-1974 in three Swedish PVCprocessing plants. The mortality was followed 1964-1986 and the cancer incidence 1964-1984, Expected figures were calculated from Swedish national rates. Among Swedish citizens, the observed mortality and cancer incidence was close to the expected in most diagnoses. Among immigrants, mostly from Finland, there was a marked excess of circulatory deaths. This finding was probably due to the higher incidence of coronary mortality in Finland compared to Sweden. In the whole cohort, five cases of malignant melanoma had occurred as compared to 1.5 expected (SMR = 3.4, 95% confidence limit 1.1-7.9). This may be due to chance but merits further investigation since an increased incidence of malignant melanoma has previously been found among Norwegian PVC-manufacturing workers. 1993 Wiley-Liss, Inc. Key words: vinyl chloride, cohort study, circulatory diseases, malignant melanoma, immigrant workers INTRODUCTION The association between exposure to vinyl chloride monomer (VCM) and liver angiosarcoma among PVC-manufacturing workers has been established through a number of epidemiological studies [IARC, 1987]. The large majority of the angio sarcomas have occurred among workers with VCM exposures of several hundreds of ppm and more [Wu et al., 1989; Simonato et al., 1991], No other form of cancer has been consistently reported among VCM-exposed workers (Doll, 1988], The PVC-processing industry has had a lower VCM exposure than the VCMand PVC-manufacturing industry (Holm et al., 1982] and has been the subject of few Department of Occupational Health, Karolinska Hospital, Stockholm, Sweden (I.L.). Department of Occupational Medicine, National Institute and Occupational Health, Solna, Sweden (A.G., P.W.). Department of Occupational Toxicology, National Institute of Occupational Health, Solna, Sweden (B.H.). School of Medicine, University of Conception, Conception, Chile (G.M.). Address reprint requests to Ingvar Lundberg, Department of Occupational Health, Karolinska Hospital, S-171 84 Stockholm, Sweden. Accepted for publication June 10, 1992. 1993 Wiley-Liss, Inc. i 314 Lundberg et al. epidemiological studies. In the mid 1970s, a cohort was established of PVC-processing workers in Sweden. The intentions were to study the mortality and cancer incidence pattern among PVC processing workers as well as to examine any possible effects of exposure to low levels of VCM. The cohort was followed in the register of causes of death and in the cancer incidence register until 1976. An increased risk of ischemic heart disease was found but no excess of any type of cancer [Molina et al., 1981]. We have now made a new follow-up concerning causes of death and cancer incidence among the cohort members in three of the four factories included in the original cohort. The results of a follow-up of the workers in the fourth factory has been published [Hagmar et al., 1990], MATERIALS AND METHODS Exposure PVC pipes, tubes, and filaments as well as sheetings for food packaging, cars, wallpaper, etc. were produced in the three factories. The work activities of the cohort members were classified in three categories according to potential VCM exposure. The group with "high" exposure consisted of employees at the mixing departments of the factories. In these departments, the PVC powder, the basic raw material in the production, was mixed with other ingredient in the plastic at high temperature. Other departments in which the plastic was heater were classified as having "intermediate" exposure, while the remaining departments were classified as having "low" exposure. Individuals were assigned to the highest exposure cateogory in which they had worked for at least 1 year. The 8-hour time-weighted average VCM exposure in the mixing departments never exceeded 10 ppm during the period before 1975, and was much lower after that. In the other departments it was below 5 ppm before 1975 and below 1 ppm after that [Bruder and Straby, 1976; Holm et al., 1982], The workers were also exposed to differents phtalates used as plasticizers, but there was no attempt to quantify this exposure. Asbestos was never used in the processing in the three factories. Study Group The cohort comprised all male workers who had been employed for at least 3 months during the period 1964-1974 in three PVC-processing factories in the south ern part of Sweden; 717 workers met these inclusion criteria. Of these, 519 had been Swedish citizens at least since 1977 (Swedes) and 198 were foreign citizens or had become Swedish citizens after 1977 (immigrants). Of the cohort, 67% had "low," 23% had "intermediate," and 10% had "high" VCM exposure; 46% of the cohort had been employed for at least 5 years. Method The cohort members were followed during the period 1964-1986 for causes of death and during the period 1964-1984 for cancer incidence in the Swedish national cancer register. The observed numbers of deaths and cancers were compared to the expected, calculated by multiplication of the person-years observed within 5-year age categones during every calender year of the study period with the cause- and genderspecific national death and cancer incidence rates, and by adding, the fractional contributions. The 95% confidence limits were calculated for the quotient between i' R&S149917 Mortality and Cancer Among PVC-Processing Workers 315 TABLE I. Observed and Expected Mortality Among the Swedish Workers in Three PVC-Processing Plants in Sweden" ICD 8 no. Cause of dealh Obs Exp Obs/exp 1-999 140-209 155 162 172 390-458 410-414 460-519 800-999 All causes 105 103.1 All cancers 23 24.2 Liver 0 0.6 Lung 3 4.6 Malignant melanoma 3 0.5 Circulatory diseases 49 51.7 Ischemic heart disease 38 37.2 Respiraiory diseases 6 5.2 Violent deaths 15 10.6 1.0 1.0 0 0.7 6.3 1.0 1.0 1.2 1.4 95% Cl 0.8-1.2 0.6-1.4 0.0-6.1 0.1-1.9 1.3-18 0.7-1.3 0.7-1.4 0.4-2.5 0.8-2.3 "At least 3 months of employment during 1964-74 was required for cohort membership. observed and expected numbers, based on a Poisson distribution. The Epilin com puter program was used for the calculations [Anjou et al., 1981]. RESULTS Causes of Death, and Cancer Incidence, Among Swedes Among the Swedish workers, 105 had died compared to 103 expected (Table I). The number of deaths in the major disease groups were also close to the expected. Three men had died from malignant melanoma compared to 0.5 expected from national rates. There was no excess risk for other forms of cancer. There was no increase in mortality in any diagnosis with increasing duration of employment, in creasing VCM exposure category, or increasing time since first exposure. The dif ferences in mortality between the three factories were small. The incidence of all cancer diagnoses, apart from melanoma, was close to the expected (Table II). The three cases of melanoma observed in the death register could also be found in the cancer register. There was no increase of any form of cancer with increasing duration of employment, increasing VCM exposure category, or increas ing time since first employment. Causes of Death, and Cancer incidence, Among Immigrants The immigrants showed an increased mortality compared to Swedish males, particularly from ischemic heart disease (Table III). However, the increased mortality was evenly distributed over different strata according to duration of employment, VCM exposure, and time since first employment. One immigrant had died from malignant melanoma. Among the immigrants, ten cases of cancer had been reported to the cancer register. Two of these cases were malignant melanomas as compared to 0.3 cases expected based on Swedish national rates. Both these immigrants came from Nordic countries. Malignant Melanoma There were five cases of malignant melanoma found in the cancer register in the whole cohort; 1.5 cases were expected from the Swedish general male population without any requirement concerning time since first employment (observed/expected 316 Lundberg et al. TABLE II. Observed and Expected Cancer Incidence Among the Swedish Workers With and Without a Requirement for at Least 10 Years Between First Employment in PVC-Processing Plant and Diagnosis _______ (CD 7 no. 140-209 155 162 190 Cancer localization All cancers Liver Lung Malignant melanoma No latency Obs Exp 33 35.6 0 0.9 3 4.1 3 1.2 10 years latency Obs Exp 22 25.7 0 0.7 3 3.0 3 0.7 TABLE III. Observed and Expected Mortality Among Immigrant Workers in Three PVC-Processing Plants in Sweden* [CD 8 no. Cause of death Obs Exp Obs/Exp 1-999 All causes 30 16.0 140-209 All cancers 5 3.5 155 Liver 0 0.1 162 Lung 0 0.7 172 Malignant melanoma 1 0.1 390-458 Circulatory diseases 15 6.1 410-414 Ischemic heart disease 13 4.4 1.9 1.4 0 0 7.7 2.5 3.0 JAt least 3 months of employment during 1964-74 was required for cohort membership. 95% Cl 1.3-2.7 0.5-3.4 0.0-44 0.0-5.6 0.2-43 1.4-4.1 1.6-5 1 = 3.4; 95% confidence limits = 1.1-7,9). Similarly, four workers had died due to this tumor versus 0.6 expected (observed/expected = 6.6; 95% confidence limits = 1.8-17). The melanomas were diagnosed 9-17 years after the start of employment. They had occurred in all three factories. Two of the melanomas were found among workers in the "intermediate" exposure category, and three among workers in the "low" exposure category. Two of the workers with melanoma had been employed in the factories under study for only about 1 year. The melanomas were localized to widely different parts of the body. DISCUSSION Mortality Differences Between Swedes and Immigrants The immigrant workers, who mainly came from Finland, showed increased mortality compared to Swedish males. Most immigrants were young. Less than 6% of the person-years under observation among the immigrants belonged to the age groups above 60. Between 20 and 60 years of age, the death rates for all causes were around 1.5 times higher in Finland than in Sweden during the follow-up period. For ischemic heart disease, the death rates were 2-3 times higher than among Swedish males [Central Statistical Office of Finland, 1964-1986]. The observed deaths among the immigrants were therefore probably close to what should have been expected if the Finnish male population had constituted the reference category. Thus, it seems likely that the increased mortality from ischemic heart disease among the immigrants was not due to work environment exposures. Mortality and Cancer Among PVC-Processing Workers 317 Comparison With Previous Investigations The mortality and cancer incidence has been followed in another Swedish PVC-processing factory [Hagmaret al., 1990]. In this factory deaths from, as well as incidence of, all cancers combined was increased compared to regional rates. There was also an increased mortality from, and incidence of, respiratory tract tumors that tended to increase with increasing asbestos exposure. There was no relation between the estimated individual cumulative VCM exposure and the risk of respiratory cancers and there was no case of hemangiosarcoma of the liver. A multicenter study on cancer risks among European VCM-exposed workers showed excess risk for liver angiosarcoma and no other tumor [Simonato et al., 1991], A recent review on VCM carcinogenicity concluded that VCM does not cause other tumors than liver angiosarcoma, with the possible exception of lung cancer after very heavy long-term exposures [Doll, 1988], Our study is too small to exclude excess risks of most diseases, and particularly of relatively uncommon diseases such as cancers at individual sites, associated with low-level exposure to VCM. The fact that we did not find any excess risk for most cancer diagnoses must therefore not be interpreted as an evidence of lack of effect but rather as inconclusive results. Yet, together with our studies performed in the PVCprocessing industry, our study could possibly contribute to the knowledge on this issue. However, that asbestos was not used in the processing is compatible with the finding of no increased risk of lung cancer in the three plants investigated. Malignant Melanoma During the follow-up period, the age-standardized incidence of melanoma was somewhat higher in the regions where the investigated factories were located than in the general Swedish male population. On the other hand, the incidence of melanoma is somewhat lower among Swedish blue collar workers than among Swedish males [Vagero and Persson, 1986]. The reported incidence of melanoma in the other Nordic countries is not higher than the incidence in Sweden. We did not try to estimate the influence of these contradictory tendencies on the expected value, but it appears that the studied cohort shows some excess of melanoma. Exposure to sunlight is an important cause of melanoma. However, the cohort members should not have been more exposed to sunlight than other sectors of the population. There are no established risk factors for malignant melanoma in the occupational environment, but exposure to polychlorinated biphenyls [Bahn et al., 1976], laboratory work [Aquavella et al., 1982; Austin et al., 1981; Thomas and Decoufld, 1979], oil refinery work [Rushton and Alderson, 1981; Thomas et al., 1980], and work with exposure to electricity as well as work in the electronics industry [Olin et al., 1985; Sorohan et al., 1985; Vagero et al., 1985] have been observed to be associated with melanoma in some studies. Such exposures were uncommon in the factories studied. An increased incidence of melanoma was also detected in a Norwegian factory for PVC production [Storetvedt Heldaas et al., 1987]. The authors implicated VCM as the cause of the melanomas. In our study, VCM is an unlikely cause of the melanomas detected since the tumors occurred among workers in departments with very low VCM exposure. However, the Norwegian finding may support a role for some exposure associated with PVC handling in the etiology of melanoma. On the 318 Lundberg et al. other hand, the short employment duration of two of the melanoma cases in our study, as well as the localization of the melanomas to different pans of the body, does not immediately support a common occupational etiology. CONCLUSIONS The mortality and cancer incidence patterns among the Swedish PVC-processing workers was close to the expected. The immigrants, mainly from Finland, showed excess ischemic heart disease as compared to Swedish national rates. This was probably due to population characteristics, rather than to occupational exposures, as the incidence of this disease is higher in Finland than in Sweden. An excess incidence of malignant melanoma was detected among Swedish as well as immigrant workers, based on five cases. This finding may be due to chance but deserves further investigation since excess melanomas have also been detected in workers at a Nor wegian plant for PVC production. ACKNOWLEDGMENTS The study was financially supported by the Swedish Work Environment Fu; .1. REFERENCES Anjou C-G, Andersson K, Axelson O, Edling C, Samueisson P-G, Wigertz O (1981): A computer system for occupational health epidemiology. In Gremy F, Degoulet P, Barber B, Salomon R (eds): "Proceedings From the Third Congress of the European Federation of Medical Informatics." Berlin: Springer-Verlag. Aquavella J, Tietjen G, Wilkinson G, Key C, Voelz G (1982): Malignant melanoma incidence at the Los Alamos National Laboratory. Lancet i:883-884. Austin D. Reynolds P, Snyder M, Biggs M, Stubbs H (1981): Malignant melanoma among employees of Lawrence Livermore National Laboratory. Lancet ii:712-7!6. Bahn A, Rosenwaike I, Henman N, Grover P, Stellman J, O'Leary M (1976): Melanoma after exposure to PCB's. N Engl J Med 295:450. Bruder U, Straby A (1976): Vinylklondexpositionen i den PVC-bearbetande industrin (Vinyl chloride exposure in the PVC processing industry). In Swedish. Rapport frin en undersokning gjord i samverkan mellan Svenges Plastforbund och Arbetarskyddsstyrelsen. Undersokningsrapport TK 1/76, Arbetarskyddsstyrelsen. --- Central Statistical Office of Finland (1964-1986). "Causes of Death in Finland." Helsinki: Central Statistical Office of Finland. Doll R (1988): Effects of exposure to vinyl chloride. An assessment of the evidence. Scand J Work Environ Health 14:61-78. Hagmar L, Akesson B, Nielsen J, Andersson C, Linddn K, Atteweii R, Moller T (1990): Mortality and cancer morbidity in workers exposed to low levels of vinyl chloride monomer_at a polyvinyl chloride processing plant. Am J Ind Med 17:553-565. Holm L. Westlin A, Holmberg B (1982): Technical control measures in the prevention of occupational cancer: An example from the PVC industry. In: "Prevention of Occupational Cancer--Interna tional Symposium. ' ' Geneva: Occupational Safety and Health Series, No. 46, International Labour Office. International Agency for Research on Cancer (1987): "IARC Monographs on the Evaluation of Carci nogenic Risks to Humans. Overall Evaluations of Carcinogenicity: An Updating of IARC Mono graphs Volumes 1 to 42. Supplement 7." Lyon: IARC. Molina G, Holmberg B, Elofsson S, Holmiund L, Maasing R, Westerholm P (1981): Mortality and cancer rates among workers in the Swedish PVC processing industry. Environ Health Perspect 41: 145-151. Mortality and Cancer Among PVC*Processing Workers 319 Olin R, Vagero D, Ahlbom A (1985V Mortality experience of electrical engineers. Br J Ind Med 42:211-212. Rushton L, Alderson M (1981): An epidemiological survey of eight oil refineries in Britain. Br J Ind Med 38:225-234. Simonato L, L'Abbd K, Andersen A. Belli S. Comba P, Engholm G, Ferro G, Hagmar L, Langard S. Lundberg l. Pirastu R. Thomas P. Winkelmann R, Saracci R (1991): A collaborative study of cancer incidence and mortality amomz vinvl chloride workers. Scand J Work Environ Health 17:59-69. SorahanT, Waterhouse JAW, McKicrnan MJ, Aston RHR (1985): Cancer incidence and cancer mortality in a cohort of semiconductor workers. Br J Ind Med 42:546-550. Storetvedt Heldaas S, Andersen A, Langard S (1987): Incidence of cancer among vinyl chloride and polyvinyl chloride workers: Further evidence for an association with malignant melanoma. Br J Ind Med 44:278-280. Thomas T, Decoufld P (1979): Mortality among workers employed in the pharmaceutical industry. J Occup Med 21:619 -623. Thomas T, Decoufld P, Moure-Eraso R (1980): Mortality among workers employed in petroleum refining and petrochemical plants. J Occup Med 22:97-103. V3gero D, Ahlbom A. Olin R. Sahisten 8 (1985): Cancer morbidity among workers in the telecommu nications industry. Br J Ind Med 42 I9I-I95. Vagero D, Persson G (1986): Occurrence of cancer in socioeconomic groups in Sweden. Scand J Soc Med 14:151-160. Wu W, Steenland K, Brown D. Wells V. Jones J. Schulte P. Halperin W (1989): Cohort and case-control analyses of workers exposed to vinyl chloride: an update. J Occup Med 31:518-523.