Document 0JvgXdG0Y652LJkBRaB96QLYJ

,/y ELSEVIER Toxicology Letten 77 (1995) 379-385 Toxicology Letters VI Changes in lymphocyte single strand breakage and liver function of workers exposed to vinyl chloride monomer Chung-Li Du**, Min-Liang Kuob, Hsiao-Ling Chang, Tzong-Jen Sheub, Jung-Der Wang4 `Institute of Occupational Medicine and Industrial Hygiene. College of Public Health, National Taiwan University. No. I, Sec. 1. Jen-Ai Rd, Taipei. Taiwan bInstitute of Toxicology, College of Medicine, National Taiwan University. No. I. Sec. I. Jen-Ai Rd, Taipei. Taiwan Accepted 13 January 1995 Abstract Vinyl chloride monomer (VCM) is a suspected human carcinogen. Its metabolite, chloroethylene epoxide, is able to alkylate the DNA molecule and to produce single strand breakage (SSB). A total of 244 workers from 4 polyvinyl chloride (PVC) manufacturing factories were recruited to assess the SSB of their peripheral lymphocyte DNA. The method of alkaline unwinding' and hydroxyapatite chromatography was used to detect and calculate frequencies of SSB. In addition, hepatitis B and C markers and the liver function of the workers were also examined. The worker's cumulative exposures to VCM were retrospectively constructed from the current monitoring data and each worker's job history. Multiple linear regression models were constructed to predict the worker's level of SSB and liver functions based on various exposure indices and variables, such as age, sex, smoking, drinking, and hepatitis markers. The results showed that current smoking and drinking status, and the presence of VCM exposures on the previous day were 3 major determinants of the level of SSB. Among the liver function tests, only y-glutamyl transpeptidase (GGT) was associated with current VCM exposures. In contrast, aspartate aminotransferase (AST), alkaline phosphatase (ALP) and alanine aminotransferase (ALT) were mainly affected by the presence of hepatitis B surface antigen (HBsAg) or anti-hepatitis C virus (anti-HCV). We conclude that GGT should be considered to be included in the regular health screening of VCM workers, and that the SSB method may not be suitable for long-term monitoring of cumulative expo sure because of the quick DNA repair mechanism in humans. Keywords: Single strand breakage (SSB); Peripheral lymphocyte; Alkaline unwinding and hydroxyapatite chroma tography; Gamma glutamyl transpeptidase (GGT); Time-weighted average (TWA) 1. Introduction Vinyl `chloride monomer (VCM), mainly as a raw material for the manufacturing of polyvinyl chloride (PVC) resins, was considered to have low 1 Corresponding author. toxicity until 1949 when workers with hepatitis were noticed. Hepatomegaly, splenomegaly, Raynaud's phenomenon, scleroderma and acroosteoiysis were subsequently reported to be associ ated with VCM exposure {1-6], VCM was suspected to be a potential carcinogen since the 1970s due to several animal and epidemiological 0378-4274/95/SO9.JO 1995 Elsevier Science Ireland Ltd. All rights reserved SSDI 0378-4274(95)03321-B NOTICE THIS MATERIAL MAY Bt PROTECTED SY COPYRIGHT LAW `TiTlE 17 US. CODE) V, ASl 000013951 380 C.'L, Du et at. / Toxicology Letters 77 (1995) 379-385 studies. Since then, agreement on the development of liver angiosarcoma in VCM-exposed workers was gradually achieved, and increased odds ratios in brain and lung cancer were usually mentioned [7-9]. The mechanism of the VCM carcinogenic effect is usually thought to be through an epoxide generation, which causes DNA alkylation [10]. However, a normal human body has many repair mechanisms to DNA damage, such as dealkyla tion. DNA adduct may be excised and no carcino genic effect will show up. Under the condition of heavy burden from excessive VCM exposure, re pair defect ensues and lesions may be expressed. These ranged from DNA-base-pair adduct, substi tution, addition, deletion, DNA strand separation or sister chromatid exchange (SCE) to chromo somal or gene abnormality [11]. Cytogenetic study such as SCE and chromosomal aberration (CA) that reflect chronic exposure were tested on these workers [12], and many of them showed positive results. However, these tests were time consuming, and since Maki-Paakkanen et al. [13], reported a simultaneous elevation of frequencies of CA and single strand breakage (SSB) among styrene workers we suspect that SSB might be an earlier genotoxic biomarker than CA or SCE. Thus, we performed a DNA breakage test to study the early genotoxicity upon VCM exposure. Routine liver function tests were also done to search for any early indicator of liver damage. 2. Materials and methods Four PVC manufacturing factories with 244 workers in Taiwan were studied. We first perform ed personal and area samplings to measure the VCM concentration [14]. This was followed by physical examination 3 months later. A compre hensive physical examination which included a questionnaire about personal and working history was performed in the morning of a workday. Blood was drawn for tests of hemoglobin, white blood cell, serum protein, and alanine aminotrans ferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), y-glutamyl transfer ase (GGT) as well as creatinine, hepatitis B antigen (HBsAg), anti-hepatitis C virus (anti-HCV), and DNA breakage test of peripheral lymphocyte. HBsAg and anti-HCV were examined by radioim munoassay (RIA, Abbott Laboratory, IL) and enzyme-immunoassay (EIA, Abbott Laboratory, Wiesbaden, Germany), respectively. To determine SSB, the method was used as described by Erixon and Ahnstrom [15]. Briefly, lymphocytes were iso lated from Ficoll-Hypaque and lysed in 0.03 M NaOH, then kept in dark for 5 min, until the cell membrane ruptured. The solution was then neutralized to pH 6.8 with 20 mM NaH2P04. After a short sonication, sodium dodecyl sulfate (final concentration 0.25) was added. The DNA samples were then put on the hydroxylapatite col umns at 60C. The single and double strand DNA was eluted with 0.08 M and 0.25 M potassium phosphate buffer, respectively. DAP1 (4,6diamidino-phenylindoie 2HC1) was added and the amount of DNA was measured by fiuorometry. The fraction of double strand DNA (FDs) was calculated, and logFDs was taken as represen tative of level of SSB [15,16]. The personal expo sure to VCM was measured and medium time-weighted average (TWA) was assigned to each job. Because every job was performed with at Table 1 Median exposure concentrations of different categories of exposure and jobs in PVC factories Catergory Job description of exposure High Medium Low Unloading PVC Adding catalyst Cleaning tank (water jet) Cleaning tank (manual)* Unloading VCM Foreman Control room Supervisor Engineering control Stripping tank Secretary Drying operation Laboratory testing Hygienist Machinist PVC storage Gatehouse `Short-term exposure. Medium TWA concentration (ppm) 73.8 63.7 38.2 21.9 4.85 1.76 1.73 1 34 1.34 1.34 1.3 0.99 0.88 0.68 0.67 0.43 0.36 Table 2 Demographic No. of worker Current mean Age (yean) Height (cm) Body weight ( BM! Systolic preset Diastolic press Smoking (pact Alcohol drinki Duration of w `Mean (star "P < 0.0J least 4 perse ly construct with a medi The work according tc mulative exj exposure co the duratior mation also lative expo: Table 3 Laboratory res Hemoglobin (g White blood ce Creatinine (mg Total protein ( Albumin (g/dl) Globulin (g/dl) AST (IU/1) ALT (IU/1) GGT (IU/1)** ALP (IU/1) % with hepatiti % with anti-hej Level of single `Mean (stain "P < 0.01 AS I 000013952 C.-L Du el at. / Toxicology Letters 77 (1995) 379-385 Table 2 Demographic and physical data of workers in 3 exposure categories Exposure categories High Medium No. of workers Current mean exposure (ppm) Age (years) Height (cm) Body weight (kg) BMI Systolic pressure (mmHg)** Diastolic pressure (mmHg) Smoking (pack/year) Alcohol drinking (g/week) Duration of work (months) 16 56.3 (22.6)' 39.8 (9.3) 167.0 (2.6) 68.0 (9.2) 24,4 (3.3) 121.3 (13.6) 83.0 (9.6) 106.3 (204.3) 40.0(1! 0,0) 158 (105) . 'Mean (standard deviation). **P < 0.05 (general linear model test). 104 1.8 (1.0) 40.2 (9.2) 167.2 (6,0) 65.6 (9.0) 23.4 (2.7) 115.2 (13.9) 76.7 (9.3) 85.8 (144.4) 42.6 (294.6) 177 (114) Low 122 0.67 (0.26) 39.5 (9.5) 168.2 (5.8) 66.3 (9.4) 23.4 (2.8) 116.0(12.1) 76.9 (9,5) 82.8 (147.2) 34 1 (142.4) 170 (117) 381 least 4 personal measurements, we have successful* ly constructed 15 different exposure groups each with a medium TWA value (Table 1). The worker's current exposure was calculated according to his present time activity data. The cumulative exposure was the summation of each job exposure concentration (in ppm 8-h TWA) times the duration (in months) of that job. Log transformation also was tried for both current and cumulative exposure indices during our multivariate regression analysis to seek a better fit. These data were analyzed with the SAS program [17], using general linear and multiple regression models. 3. Results There are totally 244 workers examined, ineluding 7 females. One of the subjects showed clinically overt hepatitis B signs and symptoms and he was excluded from our study. The average Table 3 Laboratory results of 3 exposure groups Hemoglobin (g/dl) White blood cell (x lOCXVpl) Creatinine (mg/dl) Total protein (g/dl) Albumin (g/dl) Globulin (g/dl) AST (IU/1) ALT (IU/1) GGT (IU/1)" ALP (IU/1) % with hepatitis B surface antigen % with anti-hepatitis C virus Level of single strand DNA breakage 'Mean (standard deviation). **P < 0.01 (general linear model test). Exposure groups High 16.1 (1.0)' 6.5 (1.6) 7.6 (0.4) 4.6 (0.2) 3.0 (0.3) 1.2 (0.2) 26.2(12.3) 29.5 (21.0) 61.9(92.1) 172.3 (33.6) 18.8 0 0.124(0.087) Medium 15.8 (1.1) 6.5 (1.4) 7.6 (0.4) 4.5 (0.2) 3.1 (0.4) 1.2 (0.1) 23.3 (11.0) 23.1 (20.2) 28.6(18.2) 166.0 (42.3) 22.1 2.9 0.125(0.095) Low 15.9 (1.2) 6.5 (1.4) 7.6 (0.4) 4.5 (0.3) 3.1 (0.3) 1.2 (0.1) 21.8 (8.7) 20.8 (14.3) 31,1 (26.2) 168.8 (42.0) 17.2 3.3 0.126(0.099) I r- >1 >/ <'L 7- ' V v 7 ft 382 C.-L Du et al. / Toxicology Utters 77 (1995) 379-385 Table 4 Multivariate regression models or serum enzyme activities: the listed figures are regression coefficients (S.E.) of different independent variables Scrum enzyme Intercept BMI Current VCM exposure* Alcohol drinking HBsAg AST ALT GGT ALP N.S. -48.3 -57.3 136 1.2(0.43) 2.9 (0.82) 3.7 (0.73) N.S. N.S. N.S. 0.57 (0.14) N.S. N.S. N.S. N.S. N.S. 8.1 (3.1) 13.4 (5.8) N.S. 14.6 (7.1) `Current exposure: measured TWA (in ppm) value 3 months prior to blood tests. N.S. indicates non-significant for P < 0.0S. Anti-HCV 27.4 (7.5) 36.0(14.1) N.S. N.S. Model R2 0.12 0.11 0.19 0.03 age is 40 years old, and the mean duration of em ployment is 14 years. We also classified these workers into 3 groups according to their current TWA exposure concentration (> 10 ppm, < 1 ppm, and in between). The basic demographic data is summarized in Table 2. It revealed no signi ficant difference in these 3 groups. The com parison of biochemical testing results is shown in Table 3, which indicated that only GGT is significantly higher in the high exposure group, while the level of SSB was rather close among these 3 groups. A multiple regression analysis was performed with liver enzymes as the dependent variable and drinking status, HBsAg, anti-HCV and various exposure indices as the independent variables. We found that ALT and AST were determined by anti-HCV, HBsAg, and body mass index (BMI), while VCM exposure and drinking Table 5 Regression models of level of single strand breakage using different exposure indices and other predicting variables Meaning of exposure index Exposure index Intercept Age (years) Current smoking Current drinking Model (cig./day) (g/week) R2 Current exposure concentration (in ppm) Log transformation of current exposure concentration (log(ppm)) Cumulative exposure (ppm/month) XI = 1, X2 0 mean exposure concentration of the day before SSB test > 1 ppm X2 = 1, XI = 0 mean exposure concentration of the day before SSB test < '1 ppm XI = 0 and X2 = 0, no exposure the day before SSB teat X * 1, working on the day before SSB test X = 0, absent from work the day before SSB teat N.S. N.S. N.S. XI = 0.05 (0.03) X2 = 0.06 (0.03) 0.05 (0.03) 0.14(0.03)* 0.14 (0.03) 0.13 (0.03) 0.08 (0.04) 0.08 (0.04) N.S. N.S. N.S. N.S. N.S. 0.0018 (0.0008) 0.002 (0.0008) 0.00007 (0.00003) 0.06 0.00007 (0.00003) 0.07 0.0016 (0.0008) 0.0018 (0.0008) N.S. 0,07 0.00007 (0.00003) 0.08 0.0018 (0.0008) 0.00007 (0.00003) 0.08 *The listed figures are regression coefficients (S.E.M.); ail sre significant with P < 0.03. N.S., nonsignificant had no ity was BMI. fi to be borderl multiva regressi current cant efl and dri exposui SSB. hi betweer before 1 ed that factors who we VCV corr. . centrati (r2 = 0.1 4. Discti Meth elude d strand t new bas a vf` bre..* blooo study, it mice to VCM cc 80% of Anothei to meth; first inc sixth da control In our increase rent after sir SSB is si posure ' possible drinking - , --l*J!l| C.-L DuetaL/ Toxicology Utttrt 77 (I99S) 379-383 383 had no influence, as shown in Table 4. GGT activ ity was influenced by current exposure levels and BM1. Although alcohol drinking has been known to be a major determinant, it only reached borderline significance (P = 0.09) in our multivariate analysis (not shown in Table). The regression model of the level of SSB showed that current smoking and drinking status had signifi cant effect, but cumulative amounts of smoking and drinking did not. The current and cumulative exposure of VCM had no effect on the level of SSB. However, there is a significant difference between those who were absent from work the day before blood was drawn and those who still work ed that day after control of confounding by other factors (Table S). When we take the 31 persons who were examined for both personal sampling of VCM and DNA breakage test, we found that the correlation between the level of SSB and the con centration ofcurrent TWA value is relatively small (r2 = 0.014). 4. Discussion Methods of detecting DNA damage repair in clude detection of damaged bases, detection of strand breaks, and detection of incorporation of new bases [18], DNA unwinding is thought to be a very specific method to detect single strand breaks (19], Single strand DNA exists in normal blood lymphocyte and repairs continually. In one study, it was shown that after 24 h of exposure of mice to VCM, the level of SSB is correlated with VCM concentration, and after 20 h post-exposure, 80% of DNA damage had been repaired (20], Another study showed that after single exposure to methylmethane sulphonate, the level of the SSB first increased, then decreased gradually. At the sixth day, the level of SSB is similar to that of the control group [21]. In our study, we have not found any significant increase of SSB associated with cumulative or cur rent VCM exposure, but instead, we found that after stopping exposure for just 1 day, the level of SSB is significantly less than in those in which ex posure was not stopped after controlling other possible confounding factors such as smoking and drinking. This may suggest that the repair of DNA damage in humans is very fast. Although animal experiments showed a linear relationship between the exposure dose and level of SSB, we cannot find a clear linear relationship between VCM exposure and DNA breakage. The reasons may be as follows. First, the exposure assigned is probably not accurate enough because the current exposure dose was estimated through a cross-sectional area sampling, and not through long-term constant per sonal samplings. Moreover, although there has been no major change in working processes, the exposure 3 months ago may not directly reflect the current exposure level. Second, other cocfcundiiig factors may exist, such as the T/B lymphocyte ratio, because it was usually thought that only Tlymphocytes had an ability to repair [22]. Third, the time of drawing blood is in the morning of the workday instead of at the end of shift, so the lym phocytes have had a rest time for at least nearly 10 h which may be long enough for a major repair, and we were unable to document a clear doseresponse relationship. Therefore, we recommend future study of SSB should have the workers' per sonal air sampling performed on the same day and the blood drawn either at the end of the shift or compared before and after the shift to avoid the above problems. The studies in the 1970s showed a higher preva lence of abnormal AST, ALT and ALP in VCM- exposed workers [23-26], However, Sugita et al. [27] reported that VCM probably had no effect on these enzymes after control of confounding by other risk factors. Through a multivariate regres sion analysis we have also found that AST, ALT were associated with hepatitis B and C alone, and had nothing to do with the VCM exposure. GGT was thought to be associated with VCM exposure, especially under high exposure condi tions [28,29], We have also found that GGT eleva tion was correlated with high current exposure levels. Since it was also known that BMI, alcohol and diabetes might influence the level of GGT [30], the interpretation of GGT elevation should be carried out with caution. However, our multivariate analysis has already controlled these potential confounding factors and still showed a consistent elevation of GGT along with a higher level of current (but not cumulative) exposure. 384 C.-L Du tt al./ Toxicology Utters 77 (1995) 379-383 Besides, as workers in the high and medium expo sure groups were comparable with the low expo sure group in al) demographical characteristics, and all subjects under study were not known to have any pancreas, heart, kidney or lung diseases, the association of the GGT elevation with a higher current exposure to VCM cannot be explained by any other alternative hypotheses. Although Sugita et al. [27] could not find such a correlation, they had no sampling data available. In this study, we have constructed a personal TWA value as well as cumulative exposure indices for each worker. Thus, our classification of exposure indices may be more accurate than in other previous studies, and is more likely to detect such an association. Therefore, we conclude that the elevation of GGT along with a high current exposure to VCM may be real. The mechanism of GGT elevation among VCM workers has not been clear so far. It is recognized that GGT catalyzes the transfer of the y-glutamyl group of peptides, such as glutathione, to other amino acids. In mice exposed to carcinogens, reduction of glutathione was usually accompanied by an elevated GGT [31], which was interpreted as a feed-back mechanism of DNA repair. Another study proposed that the elevation of GGT might be due to a selective advantage of the cell in response to the toxic environment created by the carcinogen [32]. Thus, it might be speculated that VCM-associated GGT elevation is probably an early indication of DNA repair or a cellular response to carcinogenic stress, and GGT might be useful in the long-term follow-up health screening of VCM workers. In conclusion, we found that the DNA unwin ding method may be a sensitive method to reflect short-term DNA damage and repair status, but it cannot be used as an indicator for long-term moni toring of cumulative exposure. We recommend methods reflecting a more persistent damage, such as SCE or CA be tested to compare the results in the future. Also, due to the association between current VCM exposure and GGT elevation, it is possible that some degree of genotoxicity may exist in early stage of carcinogenesis for VCM workers. We therefore recommend the GGT test be performed on the regular screening of liver function for these workers. Acknowledgement We are indebted to the National Science Council for the support of this study, NSC 81 -0421-F-002-539-Z. References HI Wilson, R.H., McCormick, W.E., Tatum, C.F. and Creech, J.L. (1967) Occupational acroosteolysis. A re port of 31 cases. J. Am. Med. Assoc. 201, 577-581. [2] Harris, D.K. and Adams. W.G.F. (1967) Acro-osteolysis occurring in men engaged in the polymerization of vinyl chloride. Br. Med. J. 5567, 712-714. [3] Juhe, S., Lange, G.E., Stein, G. and Veltman. G. (1973) On the so called Vinyl Chloride disease. Dutsch. Med. Weschr. 98. 2034-2037. [4] Dinman, B.O., Cook. W.A., Whitehouse, W.M., Magnuson, H.J. and Ditcheck, T. (1971) Occupational acroosteolysis. 1. An epidemiological study. Arch. Environ. Health 22, 61-73. 15] Cook, W.A., Giever. P.M., Dinman. B.D. and Magnuson, H.i. (1971) Occupational acroosteolysis. II. .... , An industrial hygiene study. Arch. Environ. Health 22, 74-82. 16] Dodson. V.N., Dinman, B.D., Whitehouse. W.M.. Nasr. A.N.M. and Magnuson, H.J. (1971) Occupational acroosteolysis. III. A clinical study. Arch. Environ. Health 22. 83-91. [7] Wong, O., Whorton, M.D., Foliart, D.E. and Ragland. D. (1991) An industry-wide epidemiologic study of vinyl chloride workers, 1942-1982. Am. J. Ind. Med. 20. 317-334. [8] Wu, W., Steenland, K,, Brown. D., Wells. V., Jones, J.. I J Schulte, P. and Halperin, W. (1989) Cohort and casecontrol analyses of workers exposed to vinyl chloride: an update. J. Occup. Med. 31. 518-523. [9] Heldaas, S.S., Langard, S.L. and Andersen, A. (1984) In cidence of cancer among vinyl chloride and polyvinyl chloride workers. Br. J. Ind. Med. 41, 25-30. [10] Bolt, H.M., Filter, J.G. and Laib. R.J. (1982) Covalent binding of haloethylenes. In: R. Synder, D.V. Parke, J.J. Kocsis, D.J. Jellow, G-G. Gibson and C.M. Witmcn I (Eds.), Biological Effects, Plenum Press, New York. pp. 667-683. [11] Brusiek, D. (1987) Fundamentals of genetic toxicity. In: D. Brusiek (Ed.). Principles of Genetic Toxicology, New York and London. Plenum Press, pp. 47-48. [12] Sinues. B., Sara, A. and Bernal, M.L. (1991) Sister chro matid exchanges, proliferating index, and micronuciei in biomonitoring of internal exposure to vinyl chloride monomer in plastic industry workers. Toxicol. Appl. Pharmacol. 108. 37-45. [13] Mkki-Paakanea. J., Walks, S., Ostcrman-Golkar, S. and Norppa, H. (1991) Single-strand breaks, chromosome aberrations, sister-chromatid exchanges, and micronuciei in blood lymphocytes of workers exposed to styrene dur i | ing th Muta, [14] Du, C pliug chiori Natio [15] Erixoi break norm* deterr lapati' [16] Wall; in Dr metha the a 319-3 [17] SASI 6.03 e [IS] Cleavt of Dh In: ft Pros Amsit [19] Tanaii L. and suits c DNA Antics [20] Wallet single vinyl c [21] Engels accur D Car.. [22] Boern disapp and T Mutat ASI 000013956 C.-L. Duel aU Toxicology Utters 77 (1993) 379-383 385 ing the production of reinforced plastics. Environ. Mol. Mutagen. 17, 27-31. (14) Du, C.L (1993) Comparison of personal and area sam pling strategies in assessing workers' exposure to vinyl chloride monomer. Master thesis. College of Medicine, National Taiwan University, Taipei, pp. 1-42. [13] Erixon, K. and Ahnstrom, G. (1979) Single-strand breaks in DNA during repair of UV-induced damage in normal human and Xeroderma Pigmentosum cells as determined by alkaline DNA unwinding and hydroxylapatite chromatography. Mutat. Res. 39, 257-271. [16] Walles, S.A.S and Erixon, K. (1984) Single-strand breaks in DNA of various organs of mice induced by methyl methanesulfonate and dimethyl-sulfoxide determined by the alkaline unwinding technique. Carcinogenesis 5, 319-323. [17] SAS Institute Inc. (1988) SAS Procedures Guide, release 6.03 ed. Cary N.C. [ 18] Cleaver, J. E. (1978) Methods for studying excision repair of DNA damaged by physical and chemical mutagens. In: B.J. Kilbey (Ed.), Handbook of Mutagenicity Test Procedures. Elsevier Scientific Publishing Company, Amsterdam, pp. 19-37. [19] Tanaingher, M., Bordone, R., Russo, P., Grilli, S., Santi, L. and Parodi, S. (1987) Major discrepancies between re sults obtained with two different methods for evaluating DNA damage: alkaline elution and alkaline unwinding. Anticancer Res. 7, 669-680. [20] Walles, S.A.S and Holmberg, B. (1984) Induction of single-strand breaks in DNA of mice after inhalation of vinyl chloride. Cancer Lett. 25, 13-18. [21] Engelse, L.D. and De-Brij, R.J. (1980) Persistence and accumulation of (potential) single-strand breaks in liver DNA of rats treated with ethyl methanesulphonate. Cancer Lett. 11, 199-208, [22] Boerrigter, M.E.T.I and Vijg, J. (1991) Induction and disappearance of DNA single-strand breaks in human B and T lymphocytes after exposure to ethylnitrosourea. Mutat. Res, DNA Repair 253, 49-55. [23] Velman, G., Lange, C.E., Juhe, S,, Stein, G. and Bachner, U. (1975) Clinical manifestations and course of vinyl chloride disease. Ann. N.Y. Acad. Sci. 246, 6-17. [24] Waxweiler, R.J., Falk, H., McMichaet, A., Mattov, J.S., Grivas, A.S. and Stringer, W.T. (1977) A cross-sectional epidemiological survey of vinyl chloride workers. Na tional Institute for Occupational Safety and Health (DHEW publication No. 77-177), Cincinnati. [25] Lilis, R.. Anderson, H., Nicholson, W.3., Daunt, S., Fischbein, A.S. and Selikoff, U. (1975) Prevalence of disease among vinyl chloride and polyvinyl chloride workers. Ann. N.Y. Acad. Sci. 246, 22. [26] Wyatt, R.M., Kotchen, J.M., Hochstrasser, D.L., Buchanan, 3.W, Jr., Campbell, D.R. and Slaughter, J.C. (1975) An epidemiological study of blood screening tests and illness histories among chemical workers involved in the manufacture of polyvinyl chloride. Ann. N.Y. Acad. Sci. 246, 80-87. [27] Sugita, M,, Masuda, Y. and Tsuchiya, K. (1986) Early detection and signs of hepatoangiosarcoma among vinyl chloride workers. Am. J. Ind. Med. 101, 411-417. [28] Nosko, M., Zajkov, H., Uzunova, D. and Spasovski, M. (1976) Biochemical disturbance in workers exposed to vinyl chloride. Letopisi na Higienno, Epidemiologicartata Sluzba. 20, 47-50. [29] Langauer-Lewowicka, H., Kurzbauer, H., Uzunova, D. and Spasovski, M. (1983) Vinyl chloride disease -- neurological disturbance, (nt Arch. Occup. Environ. Health 52, 151-157. [30] Rosalki, S.B. (1975) Gamma-glutamyl transpeptidase. Adv. Clin. Chem. 17. 53-107. [31] Fiala, S., Mohindru, A., Kettering, W.G., Fiala, A.E. and Morris, H.P. (1976) Glutathione and gamma gluta myl transpeptidase in rat liver during chemical car cinogenesis. J. Natl. Cancer. Inst. 57, 591-598. [32] Hanigen, M.H. and Pitot, H.C. (1985) Gamma-giutamy! transpeptidase -- its role in hepatocarcinogenesis. Car cinogenesis 6, 165-172.