Publication of the International Union Against Cancer
Int. J. Cancer: 109, 425– 429 (2004) © 2004 Wiley-Liss, Inc.
TESTICULAR CANCER AND OCCUPATIONAL EXPOSURE TO POLYVINYL CHLORIDE PLASTICS: A CASE-CONTROL STUDY Lennart HARDELL1,2, Nils MALMQVIST3, Carl-Go¨ran OHLSON3*, Håkan WESTBERG3 and Mikael ERIKSSON4 1 ¨ rebro University Hospital, O ¨ rebro, Sweden Department of Oncology, O 2 ¨ rebro University, O ¨ rebro, Sweden Department of Natural Sciences, O 3 ¨ rebro University Hospital, O ¨ rebro, Sweden Department of Occupational and Environmental Medicine, O 4 Department of Oncology, University Hospital, Lund, Sweden Exposure to polyvinyl chloride (PVC) plastics as a risk factor for testicular cancer was investigated. In total, 981 cases who were 20 –75 years old and had reported to the Swedish Cancer Registry during 1993–1997 were included. One matched control from the population registry was used. Exposure was assessed by a questionnaire that was supplemented over the phone. Furthermore, an occupational hygienist qualified all exposures. In all, 791 matched pairs completed the questionnaire. Overall exposure to PVC plastics gave the odds ratio (OR)ⴝ1.35, 95% confidence interval (CI)ⴝ1.06 –1.71, increasing with >10 year latency period to ORⴝ1.45, 95% CIⴝ1.06 –1.98. No dose-response relationships were found but rather an inverse relationship with the highest odds ratios in the lowest exposure category. In conclusion, no clear association with testicular cancer and exposure to PVC could be found in our study in contrast to a previous observation. © 2004 Wiley-Liss, Inc. Key words: case-control study; exposure matrix; polyvinyl chloride; testicular cancer
An increasing incidence of testicular cancer has been reported from several Western countries during the last decades.1 In Sweden, the annual age-adjusted incidence of testicular cancer increased significantly by 2.0% (p⬍0.02) during the period 1981 to 2000.2 It is the most common cancer among young males. Pre- and perinatal exposures have attracted the largest interest to explain the increasing incidence. Cryptorchidism is an established risk factor, and an increased risk has also been reported for the descendent testis,3,4 suggesting common risk factors. Prenatal exposures that have been discussed to be of etiologic significance are environmental pollutants with estrogenic potency, i.e., xenoestrogens.1 Support for the influence of persistent organic pollutants on the risk for testicular cancer during the pre- and peri-natal period was obtained in a recently published Swedish study.5 Testicular cancer has usually not been regarded to be an occupational disease. However, in a case-control study on testicular cancer and various occupational exposures, we observed a 6-fold increase in the risk for testicular cancer, mainly seminoma, among men exposed to polyvinyl chloride (PVC) plastics.6,7 Vinyl chloride, the monomer of PVC plastic, is a well-established risk factor for angiosarcoma but not for testicular cancer.8 No excess of testicular cancer was observed in 2 cohort studies of workers exposed to vinyl chloride.9,10 Furthermore, no association between exposure to PVC plastics and testicular cancer was observed in a Danish case-control study of 3,745 cases identified in a Danish pension register.11 Other agents most extensively tested for estrogenic properties are phthalates, bisphenol A and nonylphenols. Phthalates are extensively used in PVC as plasticizers and are widely spread in the environment. The phthalates reported to have the strongest estrogenic potencies are butyl-benzyl-phthalate (BBP) and di-n-butyl phthalate (DBP), being 106–108 times less potent than the natural hormone 17--oestradiol.112,13 The most commonly used phthalate, di-(2-ethylhexyl)phthalate (DEHP), has an even weaker, if any, estrogenic effect compared to the effect of 17--oestradiol.14 The aim of our study was to confirm, or reject, the increased risk of testicular cancer after exposure to PVC plastics as reported in
the previous study. In the present study all cases and controls were ascertained during a later time period than in the previous one and no overlapping of the studies occurred. MATERIAL AND METHODS
Our study was designed as a matched case-control study with the whole Swedish male population in the years 1993–1997 as the study base. The study was approved by the responsible research ethics committee. Subject ascertainment All cases of testicular cancer (ICD 7 code 178) aged 20 –75 years in the study base were identified in the Swedish Cancer Registry. In total 1,061 cases were identified. Of these 37 were deceased (only living cases were included), 18 had a benign tumour and 3 had cancer in situ. Furthermore, only germ cell tumours were included. Thus, 1 case with adenoidcystic cancer, 2 with adenocarcinoma, 1 with mesothelioma, 3 with carcinosarcoma and 8 with unspecified tumour in the testis were excluded. The physicians who had responsibility for the cases were contacted and asked for permission to include the patient in the study. Four cases were judged not to be capable to participate due to, e.g., Down’s syndrome. Furthermore, 3 cases had unknown addresses. Consequently, the remaining study group consisted of 981 cases. Of these, 592 cases had seminoma and 389 nonseminoma tumours (376 embryonal cancer, 11 teratoma and 2 choriocarcinoma). The controls were sampled from the Swedish Population Registry by selecting at random 1 subject to each case in the same 5-year age groups, i.e., 20 –24, 25–29 years etc, without regard to area of residence. Thus, the primary sample included 981 cases and 981 matched controls. Assessment of exposure A letter of introduction and an 18-page questionnaire were sent to the cases and the controls. Reminders were sent to those who did not respond within 3 weeks and also a second time to the remaining nonresponders. The questionnaire was answered by 889 cases (91%) and 870 controls (89%). In total 791 matched pairs remained for the statistical analyses. The mean age for the cases was 36.4 years (range 20 –74 ) and for the controls 36.6 (range 17–74). Grant sponsor: Swedish Council for Work Life Research; Grant number: ¨ rebro Cancer Fund; Grant sponsor: Nyckel98-0074; Grant sponsor: O fonden *Correspondence to: Department of Occupational and Environmental ¨ rebro University Hospital, SE-701 85 O ¨ rebro, Sweden. Medicine, O Fax: ⫹46-19-12-04-04. E-mail:
[email protected] Received 2 May 2003; Revised 2 October 2003; Accepted 15 October 2003 DOI 10.1002/ijc.11709
426
HARDELL ET AL.
Lifetime working history and specific exposures were asked for. The questionnaire was basically the same as the one used in our previous study6 but with more detailed information on, e.g., plastics exposure. The questionnaire contained questions about the work places, the branch of trade, the occupations and the work tasks during different periods of time. However, the items on exposure to plastics were extended to cover the exposure in detail, e.g., PVC production, manufacture of PVC products and handling of PVC products. Similar questions on exposure to other plastics, e.g., polyethene, polypropylene, polystyrene, polyamide, acetal plastics and polycarbonate, were also included. Furthermore, the questionnaire asked for working history and exposure to certain other agents such as organic solvents, pesticides, insect repellents, asphalts, use of computers etc. The questionnaires were scrutinized by a specially trained interviewer and supplemented over the phone if necessary. These interviews, as well as coding of the questionnaires, were made without knowing about case or control status. The information in the questionnaires was then evaluated by an occupational hygienist with a special competence for assessing exposure in various industrial settings. A qualitative classification of the exposure was made according to the amount of softening agents (mainly DEHP) added in the PVC material, e.g., rigid or soft PVC products in carpets, sheets and cables. A quantitative classification was made of the mean exposure intensity in different periods based on both the information in the questionnaires and the information on the production of different enterprises provided by the branch association. Every work task was classified according to type of PVC contact, i.e., type of production, manufacture or handling of PVC products and with regard to dust levels and potential risk of dermal absorption. Hence, the exposure matrix applied in our study conforms with the exposure assessments used in many retrospective cohort studies.15 The exposure intensity of a work task with PVC contact was then classified according to an ordinal scale in 5 levels, except for the unexposed (category 0): insignificant exposure (category 1), low exposure (category 2), moderate exposure (category 3), high exposure (category 4) and very high exposure (category 5). Exposure categories 1 and 2 represent low-intensity handling or contact with PVC products, exposure category 3 represents a regular handling of PVC products, category 4 represents production of PVC products and category 5 represents production of PVC plastics (see Appendix). Every more or less exposed period as reported by the subjects was then given the proper exposure level (i.e., categories 1–5). As most exposed subjects had more than 1 type of exposure, their exposure periods were considered and weighted in 3 different measures: median intensity, maximum intensity and cumulative exposure. The median intensity was calculated by ranking the periods in order of the exposure categories (disregarding the length of the periods in years) and selecting the period in the middle as the characteristic exposure for the particular case or control. Maximum intensity was the highest exposure category of any exposure for each subject. Cumulative exposure was calculated by multiplying the median exposure category by the total number of years of exposure. Thereby a score was obtained for each subject.
Statistical methods A conditional logistic regression model for matched studies including age as matching factor was employed to obtain odds ratios (OR) and 95% confidence intervals (CI) (Stata/SE 8.1 for Windows). The exposure categories of median intensity and maximum intensity, respectively, were categorized into 4 dummy variables (the fifth category containing only 1 subject was not included) with category 0 as the reference. Cumulative exposure (as median category times years) was classified into 4 classes (scores 1–10, 11–25, 26 –50 and 51ⱖ) with category 0 as the reference. The analyses were made with regard to latency time, taken as the time between first exposure and the year of diagnosis, i.e., ⬎1 year, ⬎5 years and ⬎10 years, respectively. Cases and the correspondingly matched controls with an exposure time shorter than 1 year before diagnosis were classified as unexposed. In analyses with latency times of ⬎5 or ⬎10 years, all subjects with shorter latency times (2–5 years and 2–10 years, respectively) were excluded. Exposure time was dichotomized by the median exposure time of all controls, the cut-off time being 8 years, instead of using the 3 arbitrarily chosen exposure time classes used in Table I. A Mantel-Haenszel analysis for stratified data was used as a complement to the regression model, with age being stratified into 20 –29, 30 –39, 40 – 49, 50 –59, 60 – 69 and 70⫹ years of age. RESULTS
Some exposure to PVC, i.e., having ever been exposed, was reported by 160 subjects. However, the occupational hygienist judged that 359 subjects had had such exposure. Information on median intensity exposure in 3 exposure time classes for the 791 matched pairs is given in Table I. In total, a somewhat higher proportion of the cases had been exposed to PVC compared to the controls, 25 % vs. 20 %. Exposure time ⬎ 10 years was most common in categories 2 and 3. Exposure category 5 contained only 1 subject, a control who had worked as an operator in PVC production. This category was omitted from the further analyses. Odds ratios of the total material based on conditional logistic regressions increased with increasing latency time, OR 1.35, 1.39 and 1.45 respectively (Table II). The highest odds ratios were observed for the lowest exposure category, OR 2.5, 3.5 and 1.75, with increasing latency period, respectively. The lowest odds ratios were observed for the highest exposure category, OR 0.64, 0.67 and 0.67, respectively. Stratification by exposure time did not exhibit any consistent pattern. The same pattern was observed with cumulative exposure and with maximum exposure. The same analyses were carried out by considering nonseminoma and seminoma separately, but the results did not differ substantially from the other analyses. However, the odds ratios for nonseminoma were somewhat higher than for seminoma and increased with increasing latency time in all exposure categories. No increase of the odds ratios with increasing latency time was observed among seminoma. In addition, analyses of the total material were carried out with regard to the calendar time of the exposure, i.e., with regard to decades of exposure. The odds ratio for exposure in the 1950s was slightly higher than the odds ratios for exposure in the 1960s and later.
TABLE I – THE DISTRIBUTION OF THE EXPOSURE INTENSITY FOR 791 MATCHED PAIRS BY CASE/CONTROL AND EXPOSURE TIME IN 3 CLASSES1 Category
Exposure intensity
Cases/controls
0 1 2 3 4 5
Unexposed to PVC Insignificant exposure Low exposure Moderate exposure High exposure Very high exposure2
591/631 20/9 90/75 80/62 10/13 0/1
Exposure time, years 1–5
⬎5–10
⬎10
— 11/4 25/29 35/17 6/7 0/1
— 6/4 28/19 13/7 3/4 0/0
— 3/1 37/27 32/38 1/2 0/0
1 Numbers of all exposed cases and controls are given.–2Production of PVC. This exposure category omitted in the analyses.
427
TESTICULAR CANCER AND PVC EXPOSURE TABLE II – CONDITIONAL LOGISTIC REGRESSION, ODDS RATIOS (OR) AND 95% CONFIDENCE INTERVALS (CI) FOR TESTICULAR CANCER BY DIFFERENT CATEGORIES, 1– 4, OF MEDIAN INTENSITY EXPOSURE WITH DIFFERENT LATENCY PERIODS1 Category
Total Exp yr ⱕ 8 Exp yr ⬎ 8 Total, 1–3 Exp yr ⱕ 8 Exp yr ⬎ 8 1 Exp yr ⱕ 8 Exp yr ⬎ 8 2 Exp yr ⱕ 8 Exp yr ⬎ 8 3 Exp yr ⱕ 8 Exp yr ⬎ 8 4 Exp yr ⱕ 8 Exp yr ⬎ 8
⬎1 year latency
⬎5 years latency
⬎10 years latency
Ca/co
OR (CI)
Ca/co
OR (CI)
Ca/co
OR (CI)
200/159 109/79 91/80 190/146 101/68 89/78 20/9 14/6 6/3 90/75 43/42 47/33 80/62 44/20 36/42 10/13 8/11 2/2
1.35 (1.06–1.71) 1.45 (1.07–1.97) 1.23 (0.87–1.72) 1.42 (1.11–1.82) 1.57 (1.14–2.17) 1.26 (0.89–1.78) 2.50 (0.97–6.44) 2.20 (0.81–6.00) 4.82 (0.52–44.9) 1.26 (0.88–1.81) 1.05 (0.66–1.67) 1.62 (0.94–2.80) 1.46 (1.01–2.11) 2.22 (1.27–3.88) 1.01 (0.61–1.67) 0.64 (0.25–1.64) 0.56 (0.19–1.66) 1.00 (0.14–7.10)
178/137 87/57 91/80 169/126 80/48 89/78 18/6 12/3 6/3 80/63 33/30 47/33 71/57 35/15 36/42 9/11 7/9 2/2
1.39 (1.07–1.80) 1.57 (1.11–2.24) 1.23 (0.88–1.74) 1.46 (1.12–1.91) 1.72 (1.18–2.50) 1.26 (0.89–1.80) 3.50 (1.15–10.6) 3.67 (1.02–13.1) 3.00 (0.31–28.8) 1.32 (0.89–1.95) 1.02 (0.59–1.76) 1.70 (0.98–2.96) 1.42 (0.96–2.10) 2.39 (1.25–4.56) 1.01 (0.62–1.67) 0.67 (0.24–1.87) 0.57 (0.17–1.95) 1.00 (0.14–7.10)
123/97 48/29 75/68 116/91 43/25 73/66 8/6 5/3 3/3 51/39 13/11 38/28 57/46 25/11 32/35 7/6 5/4 2/2
1.45 (1.06–1.98) 1.76 (1.09–2.85) 1.28 (0.88–1.87) 1.52 (1.10–2.11) 1.90 (1.14–3.16) 1.35 (0.91–1.99) 1.75 (0.51–5.98) 1.67 (0.40–6.97) 2.00 (0.18–22.1) 1.50 (0.89–2.51) 1.28 (0.53–3.11) 1.61 (0.87–2.95) 1.52 (0.98–2.35) 2.44 (1.13–5.31) 1.17 (0.68–2.01) 0.67 (0.19–2.36) 0.50 (0.09–2.73) 1.00 (0.14–7.10)
1 In the dose-response calculations median exposure time among the controls, 8 years, was used as cut-off time. Numbers of all exposed cases and controls are given.–The columns are not mutually exclusive as the lower latency periods contain subjects in higher latency periods.
To further elucidate the effect of age on the risk, a MantelHaenszel test was performed with the material stratified on age and with ⬎ 5 years latency. The results did not differ significantly from those of the conditional regression analysis. Exposure to soft plastics in categories 3 and 4 was used as an exposure measure with all unexposed as reference. This analysis yielded an increased but statistically insignificant odds ratio for exposure to soft plastics: OR⫽1.48, 95% CI⫽0.94 –2.34 (54 cases, 37 controls). Exposure to rigid plastics was not associated with any particularly increased risk: OR⫽1.06, 95% CI⫽0.55–2.01 (23 cases, 26 controls). DISCUSSION
The main finding of this case-control study that contained 791 matched pairs was a slightly increased risk of testicular cancer after exposure to PVC-containing materials. Further analyses of the risks with different measures of exposure based on detailed assessments of the individual exposures did not indicate a doseresponse relationship but rather an inverse relationship with the highest odds ratios in the lowest exposure category. However, the odds ratios increased somewhat with increasing latency time. This inverse relationship between exposure to PVC and odds ratio does not clearly indicate an increased risk for testicular cancer in spite of the overall increased risk as strengthened by taking latency into account. The same pattern was observed with cumulative exposure and with maximum exposure. However, many hormone-disrupting chemicals exhibit an inverted U-dose response curve. Such chemicals disrupt hormones at low doses but not at high doses. This has been demonstrated for Bisphenol A,16 vinclozolin and DDE17 and certain phytoestrogens.18 The mean age for the cases was 36 years in the study, which may influence the dose-response calculations since the cut-off was 8 years (median number of exposure years among controls) and not allowing longer exposure time among the younger subjects. In the total material only 9 cases and 4 controls had an exposure time ⬎8 years in the age group ⬍30 years. Thus low numbers may have influenced the dose-response calculations. However, the results of the Mantel-Haenszel analysis were similar to the ones obtained with the conditional logistic regression and indicate a possible influence of age in the dose-response calculations. The risk of testicular cancer was obviously increased for those who had worked with PVC-containing materials in general, and the fact that the odds ratios increased somewhat with increasing latency time, mainly among nonseminoma, lends some support to
the possibility of a true association with PVC exposure. However, this increase cannot be attributed to exposure to phthalates, or even to PVC dust in general, since the odds ratios were highest in the lowest exposure category and lowest in the highest category. An alternative explanation to this finding could be the use of some particular agent in the processes that caused the spurious effect of an increased risk with low dose exposure. The occurrence of an interacting risk factor would not necessarily increase the risk proportionally, leading to an absence of a dose-response relationship. The observation that exposure to soft PVC was associated with a slightly higher odds ratio than exposure to rigid PVC could be taken as pointing in this direction. One possible explanation of the negative association between risk and degree of exposure could be a lack of statistical power due to insufficiently large study groups. However, a power calculation with regard to the 1,582 cases and controls showed that the probability (i.e., 1-) was 99% of detecting a doubled risk at the 5% significance level and 90% of detecting a risk of 1.4. The narrow confidence intervals also support the conclusion that the power should have been sufficiently large to detect even a slightly increased risk. On the other hand, the power for the high exposed group was considerably lower, since this group (category 4) only contained 10 cases and 13 controls. The probability was only around 60% for this sample size of detecting an odds ratio of 2 among the highly exposed, given the expected exposure for 13 individuals out of 791 controls. Three aspects of the exposure were regarded to be of particular importance, i.e., PVC production, manufacture of PVC products and handling of PVC products. The exposure evaluation made by the industrial hygienist showed only a moderate agreement with the exposure reported in the questionnaires, but information from the latter was not used in the analyses. The individual exposure estimates made by the industrial hygienist were focused on the concentration of PVC-containing dust in the breathing zone and not on exposure to specific substances, e.g., phthalates. This could have hampered the possibility of detecting an association between testicular cancer and phthalates but increased the possibility of detecting associations with exposure to PVC dust in general. The risk of dermal absorption was considered to be of relevance in some work situations, but it could not be integrated into the exposure classification. Another possible risk factor for testicular cancer that could be of some interest is plastic welding at workplaces classified as low exposed. Plastic welding implies exposure to electromagnetic fields, although not established as risk factors
428
HARDELL ET AL.
for testicular cancer.19 However, electromagnetic fields have come up as a possible risk factor in a couple of studies.20,21 The possibility of a promoting effect of xenoestrogens, e.g., phthalates, on the growth of a testicular tumor has been discussed elsewhere.7 The most commonly used phthalate, DEHP, has a very low, if any, estrogenic potency, but there are other phthalates that are used in PVC materials which have a stronger estrogenic potency, albeit still very low compared to 17--oestradiol. The exposure levels in the PVC industry are not well known, but there is information on the exposure levels of different phthalates in a few studies. The urine concentrations of MEHP (monoetylhexylphthalate), the main metabolite of DEHP, was analyzed in PVC workers and the conclusion was that MEHP could be used as an exposure measure for DEHP exposure22 and the air concentrations of DEHP varied between 0,009 and 1.3 mg/m3.23 The half-time of DEHP is only 12 hr and after 48 unexposed hr, DEHP would be below the detection limit. Concentrations of DEHP with various work tasks, e.g., extruding, calendering, welding and molding, were measured in 12 work sites and the concentrations varied between 0.02 and 0.5 m3.24 Exposure to DIDP (diisododecylphthalate), DEHP and BBP (benzylbutylphthalate) varied between 0,1 and 2.8 mg/m3 at production of PVC sheets.25 In a study of the cancer incidence among workers producing PVC carpeting, the exposure levels were estimated to be 0.5–3 mg/m3 for a highexposed group.9 Consequently, the available information on exposure to phthalates indicates that the exposure levels have been modest but not negligible in past. In our previous study,6 the exposure criterion would have corresponded to moderate (category 3) and high exposure (category 4) in the present study. Somewhat increased risk was found in category 3 in the present study, although not of the same magnitude as in the previous one. One explanation of this discrepancy might be
that the previous study happened by chance to catch a cluster of cases of testicular cancers with exposure to PVC in a certain geographical area. To test this notion, the analyses were repeated for a part of the present study base, restricted to the geographical areas represented in the previous study. However, similar findings as in the whole study were thereby obtained. A time clustering effect could still exist in the previous study. Our findings are in accord with the lack of association between PVC exposure and testicular cancer reported in the study of Hansen.11 The manufacture of PVC products may involve exposure to the carcinogenic monomer vinyl chloride, but the air concentration levels in general are considered to have been well below 1 ppm after 1975.9 It is important to note that no published study has demonstrated an increased risk of testicular cancer from vinyl chloride.8 –10 It might be added that in a study of cases of stillbirths or infant deaths, some malformations and low birth weight, an increased risk was found for mother’s exposure during pregnancy to PVC but not for other plastics.26 In conclusion, the earlier observation that exposure to PVC could be a risk factor for testicular cancer was not corroborated by our study. The puzzling observation of a somewhat elevated risk in lower exposure categories and by increasing latency nevertheless suggests additional studies in this respect to be carried out in other countries as the male population of Sweden is now exhausted as a study base for quite some time. ACKNOWLEDGEMENTS
The authors thank Irene Larsson for her collection of all the data and Michael Carlberg and Ing-Liss Bryngelsson for conducting the statistical analyses; Mats Eriksson supplemented the questionnaires over the phone.
REFERENCES
1.
2. 3. 4. 5.
6. 7. 8. 9.
10. 11. 12.
13.
Toppari J, Skakkebaek NE, Larsen JC, eds. Male reproductive health and environmental chemicals with estrogenic effects. Milj¨oprojekt 290. Danish Environmental Protection Agency, Copenhagen, Denmark 1995. The National Board of Health and Welfare. Centre for Epidemiology. Cancer Incidence in Sweden 2000. Health and diseases 2002:5. Stockholm 2002. Henderson BE, Benton B, Jing J, Yu MC, Pike MC. Risk factors for cancer of testis in young men. Int J Cancer 1979;23:598 – 602. Schottenfeldt D, Warshauer ME, Sherlock S, Zauber AG, Leder M, Payne R. The epidemiology of testicular cancer in young adults. Am J Epidemiol 1980;112:232– 46. Hardell L, van Bavel B, Lindstr¨om G, Carlberg M, Dreifaldt AC, Wijkstr¨om H, Starkhammar H, Eriksson M, Hallquist A, Kolmert T. Increased concentrations of polychlorinated biphenyls, hexachlorobenzene and chlordanes in mothers of men with testicular cancer. Environ Health Perspect 2003;111:930 – 4. Hardell L, Ohlson CG, Fredrikson M. Occupational exposure to polyvinyl chloride as a risk factor for testicular cancer evaluated in a case-control study. Int J Cancer 1997;73:828 –30. Ohlson CG, Hardell L. Testicular cancer and occupational exposures with a focus on xenoestrogens in polyvinyl chloride plastics. Chemosphere 2000;40:1277– 82. IARC Monographs on the Evaluation of Carcinogens in Humans. Polychlorinated Dibenzo-para-Dioxins and Polychlorinated Dibenzofurans, vol 69. Lyon, France: IARC,1997. Hagmar L, Åkesson B, Nielsen J, Andersson C, Lind´en K, Attwell R, M¨oller T. Mortality and cancer morbidity in workers exposed to low levels of vinyl chloride monomer at a polyvinyl chloride processing plant. Am J Ind Med 1990;17:553– 65. Langård S, Rosenberg J, Andersen A, Heldaas SS. Incidence of cancer among workers exposed to vinyl chloride in polyvinyl chloride manufacture. Occup Environ Med 2000;57:65– 8. Hansen J. Risk for testicular cancer after occupational exposure to plastics. Int J Cancer 1999;82:911–2. Jobling S, Reynolds T, White R, Parker MG, Sumpter JP. A variety of environmentally persistent chemicals, including some phthalate plasticizers, are weakly estrogenic. Environ Health Perspect 1995;103: 582–7. Harris CA, Henttu P, Parker MG, Sumpter JP. The estrogenic activity of phthalate esters in vitro. Environ Health Perspect 1997;105:802–11.
14. Moore NP. The oestrogenic potential of the phthalate esters. Reprod Toxicol 2000;14:183–92. 15. Stewart PA, Lemanski D, White D, Zey J, Herrick RF, Masters M, Rayner J, Dosemeci M, Gomez M, Potter L. Exposure assessment for a study of workers exposed to acrylonitril. I. Job exposure profiles: a computerized data management system. Appl Occup Environ Hyg 1992;7:820 –25. 16. Rubin BS, Murray MK, Damassa DA, King JC, Soto AM. Perinatal exposure to low doses of Bisphenol A affects body weight, patterns of estrous cyclicity, and plasma LH levels. Environ Health Perspect 2001;109:675– 80. 17. Baatrup E, Junge M. Antiandrogenic pesticides disrupt sexual characteristics in the adult male guppy (Poecilia reticulata). Environ Health Perspect 2001;109:1063–70. 18. Almstrup K, Fern´andez MF, Petersen JH, Olea N, Skakkebaek NE, Leffers H. Dual effects of phytoestrogens result in U-shaped doseresponse curves. Environ Health Perspect 2002;110:743– 8. 19. Hardell L, Holmberg B, Malker H, Paulsson L-E. Exposure to extremely low frequency electromagnetic fields and the risk of malignant diseases: an evaluation of epidemiological and experimental findings. Eur J Cancer Prev 1995;4:3–107. 20. Davis RL, Mostofi FK. Cluster of testicular cancer in police officers exposed to hand-held radar. Am J Ind Med 1993;24:231–3. 21. Floderus B, Stenlund C, Persson T. Occupational magnetic field exposure and site-specific cancer incidence: a Swedish cohort study. Cancer Causes Control 1999;10:323–32. 22. Dirven HA, Brock PH, Jongeneelen F. Determination of four metabolites of the plasticizer di(2-etylhexyl)phthalate in human urine samples. Int Arch Occup Environ Health 1993:64:555– 601. 23. Dirven HA, Brock PH, Arends A, Nordkamp H, Lepper A, Henderson P, Jongeneelen F. Metabolites of the plasticizer di(2-etylhexyl)phthalate in urine samples of workers in polyvinyl chloride processing industries. Int Arch Occup Environ Health 1993;64:549 –54. 24. Vainiotalo S, Pf¨affli P. Air impurities in the PVC Plastic Processing Industry. Ann Occup Hyg 1990;34:585–90. ˚ kesson B, Skerfving S. Phthalate ester exposure: air levels 25. Nielsen J, A and health of workers processing polyvinylchloride. Am Ind Hyg Assoc J 1985;46:643–7. 26. Ahlborg G Jr, Bjerkedal T, Egenaes J. Delivery outcome among women employed in the plastics industry in Sweden and Norway. Am J Ind Med 1987;12:507–17.
TESTICULAR CANCER AND PVC EXPOSURE
429
APPENDIX: – EXPOSURE INTENSITY IN 5 CATEGORIES AND EXAMPLES OF OCCUPATIONS AND WORK TASKS
Category 0: unexposed. Category 1: insignificant exposure. Category 2: low exposure. Category 3: moderate exposure. Category 4: high exposure. Category 5: very high exposure.
No occupational contact with PVC: architect, judge, physician, painter, asphalt worker PVC had been handled at the work place, but the subject had not been working with any PVC containing materials: e.g., cleaner (floor-polishing), farm worker Regular handling of PVC containing products: electrician, electro-technician, caretaker (electrical repairing), carpenter (demolition of plastic carpeting), salesman, car mechanic, car electrician Regular handling of PVC containing products: Carpet layer, electrician, cable layer, pipe layer, plastics worker, pipe fitter Manufacturing PVC products: tarpaulin worker, cable worker, injection moulding, plastics worker PVC production: machine-tender/operator
