Document 2q5wJZqM1rqVaELNa5Q36jEy6

American Journal of Industrial Medicine 10:411-417 (1986) Early Detection and Signs of Hepatoangiosarcoma Among Vinyl Chloride Workers Minoru Sugita, md, Yoshinori Masuda, md, and Kenzaburo Tsuchiya, md Health examinations of 108 workers exposed to vinyl chloride monomer (VCM) at a Japanese chemical plant were carried out in 1979. The polymerization of vinyl chloride was started at the plant in 1949. In this study, the highest concentration of VCM in autoclaves was determined to be 250 ppm in 1961. However, the workers at the plant had been exposed to higher concentrations of VCM several times before 1960. More recent VCM exposure was considered negligible. Examinations assessed data on age, height, weight, obesity index, sake consumption, VCM exposure concentration, latent period, cumulative exposure, ICG (indocyano green test), serum bilirubin, GOT (glu tamic oxaloacetic transaminase), GPT (glutamic pyruvic transminase), Al-P (alkaline phosphatase), GGTfy-glutamyl transpeptidase), ZTT (zinc turbidity test), LDH (lactate dehydrogenase), cholesterol, TTT (thymol turbidity test), A/G (albumin globulin ratio), and thrombocytes. Variation in VCM exposure did not affect (1) tests of pigment excretion from the liver, such as ICG; (2) thrombocytes; and (3) enzyme activity (such as GPT); nor bilirubin or flocculation reaction in serum. Key words: tiny) chloride monomers, liver function tests, ICG, thrombocytes INTRODUCTION Many studies [Block, 1974; Creech and Johnson, 1974; Nicholson et al, 1975; Waxweiler et al, 1976; Bertazzi et al, 1980] of workers exposed to vinyl chloride monomer (VCM) that reported a high incidence of hepatoangiosarcoma were re viewed. Some investigators [Marsteller et al, 1975; Popper and Thomas, 1975; Veltman et al, 1975] considered that portal hypertension among the workers could be regarded as a precursor of hepatoangiosarcoma. Early diagnosis of hepatoangiosar coma has been studied [Krammer and Mutchler, 1972; Veltman et al, 1975; Wyatt et al, 1975; Bertazzi et al, 1980], but definitive approaches to diagnosis have been elusive. Department of Public Health. School of Medicine, Tokai University, Boseidai, Isehara-shi, Kanagawaken Japan (M.S.). Health Administration Center. Keio University 4-1-1 Hiyoshi. Kohoku-ku, Yokohama-shi, Kanagawaken Japan (Y.M.). University of Occupational and Environmental Health, 1-1 Iseigaoka. Yahatanishi-ku, Kitakyushu-shi Japan (K.T.). Address reprint requests to Minoru Sugita. MD, Dept, of Public Health, School of Medicine, Tokai University, Boseidai, Isehara-shi. Kanagawa-ken 259-11, Japan. Accepted for publication May 15, 1986. 412 Sugita, Masuda, and Tsudma Wc investigated liver function of workers exposed to VCM at a Japanese chemical plant. Hepatoangiosarcoma was not found among them, and no statistically significant differences in mortality rate were detected between the VCM-exposed workers and a control group [Masuda, 1980]. No VCM-exposed workers were shown to suffer from liver dysfunction. Statistical analyses, including multivariate analyses of liver function data, were carried out and early diagnosis of liver dysfunction among VCM workers was sought. METHODS In 1979, health examinations of 108 workers who had been exposed to VCM since 1949, at a Japanese chemical plant, associated with the polymerization of vinyl chloride monomers [Masuda, 1980], were carried out. The highest concentration of VCM in the autoclaves was recorded as 250 ppm in 1961. However, the workers had been exposed to considerably higher concentrations several times before 1961 [Ma suda, 1980], Since the plant design was improved and workers were no longer required to service the autoclaves, the concentration of VCM in the working environ ment has become undetectable and recent exposure to VCM is considered negligible. Health examination included the following: age (years), height (cm), weight (kg), obesity index, sake consumption (go/week = 180 ml/week), VCM exposure concentration index, latent period (first exposure to 1979), cumulative exposure (years), ICG test, serum bilirubin, GOT, GPT, Al-P, GGT, ZTT, LDH, total choles terol, TTT, A/G and thrombocytes. The obesity index used by us, was w'cight/{height - (height + 151)/3], where the denominator is the formula for determining the standard weight of Japanese according to Matsuki et al [1970]. The workers were divided into four groups by VCM exposure concentration. The VCM exposure concentration index was as fol lows: index 4 included workers involved in cleaning autoclaves and dehydration of polymerized vinyl chloride polymer (VCP); index 3 included workers involved in polymerization; index 2 included workers involved in drying of polymerized VCP i and in laboratory work related to polymerization; and index 1 included workers involved in tasks other than the above, such as production and analysis of VCM. Higher exposure concentration index indicated greater VCM exposure. More detailed information on VCM concentration in the working environments was not available. Blood samples were analyzed by the same technician for four months; all analyses for each test (for example, GPT) were done using the same method. Liver function data of healthy subjects, also analyzed by the one technician, were within normal values; correlation coefficients between analyzed items did not differ greatly from values given in other reports [Wakabayashi et al. 1976]. RESULTS AND DISCUSSION Table I presents means, standard deviations, numbers of samples, maximums, and minimums of the health examination indicators. The age distribution was skewed: range, 23-57 years with a mean of 48.0 years. The means of height, weight, and the obesity index are 164.2 cm, 59.8 kg, and 101.5% respectively. Sake consumption showed a range of 0-17.5 go/week, with a mean and a standard deviation of 7.77 and 5.36 go/week, respectively. The mean VCM exposure concentration index was 2.6, Hepatoartgiosarcoma Among Vinyl Chloride Workers 413 TABLE 1. Mean. Standard Delation (SD) Number of Samples, Maximum and Minimum 1. Age (years) 2. Height (cm) 3, Weight (kg) 4. Obesity index (5?) 5. Sake consumption (go/week)* 6. Exposure concentration index 7. Latent period ivears) S. Cumulative exposure duration (years) 9. ICG (90 10. Bilirubin (mg 100 ml) 11. GOT (units) 12 GPT (units) 13. Al-P (units) 14. GGT (units) 15. ZTT (units) 16. LDH (units) 17. Cholesterol (mg 100 ml) 18. TTT (unitsi 19 A/G 20. Thrombocvies (10J mm3) Mean 48.0 164.2 59.8 101.5 7.77 2.6 20.5 9.75 9.9 0.88 27.3 20.9 7.4 56.2 7 26 191. 192. 1.9 2.0 26.4 SD 6.6 6.4 8.3 11.1 5.36 1.1 8.5 7.26 3.9 0.28 21.1 17.8 2.3 73.3 1.92 42. 33. 0.9 0.3 5.9 No. of samples 108 108 108 108 10S 108 108 108 95 108 108 108 108 108 108 108 108 108 108 104 Maximum 57 177 80 136 17.5 4 30.3 27.3 24 1.6 191 150 20 565 19.1 304 310 8.0 3.0 42.9 Minimum 23 146 38 78 0 1 1.9 1.0 1 0.4 14 7 4 7 4.1 72 123 0.7 1.3 14.9 go = ISO ml. with 22. 27. 29. and 30 workers distributed among indices 1-4, a nearly uniform distribution. The latent period and the duration of cumulative exposure ranged from 1.930.3 years and 1.0-27.3 years, respectively. The means of the blood chemistry results did not vary significantly from normal values. Some extraordinary values, ie. GOT 191 units and GPT 150 units for subject A, Al-P 20 units and GGT 565 units for subject B, and ZTT 19.1 units and TTT 8 units for subject C, were found. These values exceeded the mean plus four standard deviations; subjects A, B, and C are regarded as having had disordered liver functions. These six results cannot be derived from the normal distributions of liver function data of healthy subjects and were therefore excluded from the analyses. However, it should be mentioned that the statistical results obtained without using die six findings do not differ greatly from the statistical results obtained using the six. Correlation coefficients between 20 examination items were calculated by num ber of workers, age groups (20-39, 40-49, or 50-59 years), VCM exposure concen tration index (4, 3, 2, or 1) and cumulative exposure duration (0-4, 5-14, or 15-30 years). The distribution by age groups was 11, 43, and 54 workers for 20-39, 4(M9, and 50-59 years, respectively. Distribution by cumulative exposure period was 37, 48, and 23 workers for 0-4, 5-14, and 15-30 years, respectively. Correlation coefficients between VCM exposure concentration index, latent period, or cumulative exposure duration and other examination items (for example GPT) were calculated by age group, exposure concentration index, and cumulative exposure duration: Table II shows statistically significant correlations. No consistently significant examination findings were observed. Examination results that were calculated by stratum and were ASI 00004794 414 Sugita, Masuda, and Tsuchiya TABLE II. Significant Correlation Coefficient* Stratum Exposure concentration index Latent period Cumulative exposure duration Total sample Age Exposure concentration index Cumulative exposure duration (years) 20-39 40-49 50-59 High (4) Medium (3) Slight (2) Low (1) 0-4 5-14 J5-30 A/G ICG - , BIL + BIL + + ICG GOT - Age ++, GOT +, GPT +, LDH +, A/G - Age +, Al-P -- Age + +, A/G -- GPT +,LDH + Age + +, Choi + Age + +, A/G -- Age 3- +, ZTT 4*, TTT 4-, THR Age + Age + +, GPT +, A/G AI-P + Age +. LDH + Age +, Al-P -- BIL + LDH 4 + Age + Al-P + GOT + , LDH + + *+ + , +, positive correlation,-----, negative correlation: +, - p <0 05; ++,------,p < 0,01 Choi, cholesterol; THR, thrombocytes. significant twice are are ICG (negative correlation) and bilirubin (positive correla tion). Negative correlation of the ICG indicated that an increase in VCM exposure reduced the value of ICG. Consequently, the workers of this study can be regarded as having had no disorder of pigment excretion from the liver, owing to VCM exposure. This is not consistent with other reports [Kramer and Mutchler, 1972; Veltman et al, 1975] that investigated pigment excretion from liver using, for example, the BSP (bromosulfalein) test. No repons that showed a noteworthy relationship betwen bilirubin and VCM exposure were found. Our study only found a significant relationship between biliru bin and VCM exposure in two groups. Significant positive correlations were detected between age and latent period or cumulative exposure duration. Examination results that had more than one significant correlation were GOT(+), GPT( + ), LDH( + ), and A/G(-), each sign indicating the direction of the correlation coefficient. How ever, no significant examination result was found in the group with the highest VCM exposure concentration index. Accordingly, the detection of several significant corre lations between examination results and the latent period or cumulative exposure duration may not be reliable. It should be mentioned that GOT, GPT, LDH, and A/G values were not within normal values in some elderly workers. Correlation analysis revealed, however, that no worker in this study consistently had marked liver dysfunction. The correlation coefficients between the 12 examination items did not differ from those of another report [Wakabayashi et al, 1976], which studied laboratory data of healthy subjects. Age, the VCM exposure concentration index, and cumulative exposure duration were selected as three factors for analysis of variance (ANOVA), and two-way layout ANOVA [Scheffe, 1959]. These were calculated three (3C2 ~ 3) times, assigning two factors to two of the three factors and a dependent variable to each examination item. The results of the ANOVA are not shown because few notable relations were found. However, there was a tendency for some examination items to have relatively high F values in the ANOVA for the age factor. This suggests that Hepatoangiosarcoma Among Vinyl Chloride Workers 415 some elderly workers had liver dysfunction consistent with the findings described above. Stepwise regression analysis was performed with each of the 12 laboratory examination items. For each analysis, a dependent variable was assigned to a labora tory examination item and independent variables were assigned to all other items except age and the dependent variable. Table III shows accepted independent variables by dependent variable. The VCM exposure concentration index is associated with a negative coefficient when the dependent variable is ICG. This is equivalent to saying that the correlation coefficient between the ICG and the VCM exposure concentration index is negative and statistically significant. The results of the stepwise regression analysis do not differ greatly from those of other reports [Wakabayashi et al, 1976: Kitamura, 1980], in which laboratory examination data of healthy subjects was studied. Since unusual findings among the workers were not detected by ANOVA or stepwise regression analysis, principal component analysis was carried out fShioya and Asano, 1967; Okuno et al, 1971] to investigate latent relationships. According to this analysis, three factors with eigen values 2.0 or greater were selected from among the 20 examination items by principal component analysis. The summation of the communalities for the three factors was 0.383. Some values of the factor loadings for laboratory examination items were high, whereas the values for the VCM exposure indices w'ere not elevated. This suggests that VCM exposure indices are in some category other than the laboratory examination items, and that results of the health examinations of the workers w'ere not consistently related the VCM exposure concen tration index. The last analysis conducted was the canonical correlation coefficient [Okuno et al, 1978], The first variable group of the analysis consisted of 12 laboratory exami nation results, ie, liver function tests and thrombocytes, and the second variable group consisted of 8 nonlaboratory examinations, ie, age, height, w'eight, obesity index, sake consumption, VCM exposure concentration index, latent period, and cumulative exposure duration. In this analysis, it was found that the sum of not only the VCM TABLE III. Selected Independent Variables by Stepwise Regression Analysis*________ ___________________ Dependent variable Independent variables ICG BIL GOT GPT Al-P GGT ZTT LDH Choi TTT A/G THR Exposure concentration-----, ZTT + +, Cho) + 4Choi 4* 4-, A/G + + Weight - -, GPT + +, GGT + +, LDH + 4Weight 4- +, GOT 4 4-, LDH Choi Sake 4-, GOT 4t 4*, Al-P 4Choi TTT ++, A/G -GOT ++, GPT Obesity +, ICG + , BIL + + GGT +, ZTT 4 4. Choi 4 4 Sake 4-, latent period --, BIL 4 4. ZTT -- ICG BIL + + . GGT + *Chol, cholesterol: THR. thrombocytes: ++. +. positive regression coefficient;-----. negative regression coefficient; 4-4-. partial F value a 7: +,-,!> partial F value g 4. 416 Sugita, Masuda, and Tsuchiya exposure concentration index but also latent period, height, and sake consumption were related to A/G. This does not demonstrate that the VCM exposure concentration index necessarily affects the liver function of VCM workers. This resembles the stepwise regression analysis in which the dependent variable is A/G. Additional findings about sake consumption are as follows: (1) sake consump tion is not significant on ANOVA where the factor is age, VCM concentration index, or cumulative duration; and (2) the correlation coefficient between sake consumption and GGT or A/G is statistically significant. CONCLUSIONS Relationships between a VCM exposure concentration index and (1) reduction of pigment excretion from liver, such as ICG; (2) decrease of thrombocytes; or (3) elevation of Al-P, GGT, and other enzymes were not detected among workers exposed to VCM. No remarkable changes of bilirubin, transaminases, flocculation reaction tests or others were found. Thus, no specific disorder of liver function was detected among the VCM workers in this study. At the present time, it is not known whether or not these VCM workers will suffer from hepatoangiosarcoma at a later time. Certainly, early abnormalities were not detected, with the diagnostic tests used. ACKNOWLEDGMENTS This study was planned and co-authored by Dr. Yoshinori Masuda, who passed away on June 5, 1980. The work was completed by Sugita et al following the wishes of Dr. Masuda. May he rest in peace. REFERENCES Bcrtazzi PA, Villa A, Foa V, Saia B, Fabbri L. Mapp C, Marcer C, Manno M, Marchi M, Marian! F. Bottasso F (1980): An epidemiological study of \inyl chloride exposed workers in Italy. Pro XIX Intern Congress Oeeup Health 1:379-397. Block JB (1974): Angiosarcoma ofliver following vinyl chloride exposure. JAMA 229:53-54. Creech JL, Jr, Johnson MN (1974): Angiosarcoma of liver in the manufacture of polyvinyl chloride. J Occup Med 16.150-151. Kitamura M (1980)' What is normal value--relation between laboratory data and diagnosis logic. Nihon Rinsho 38 (spring extra prtnt)-4-14 (in Japanese). Kramer CG, Mutchler JE (1972): The correlation of clinical and environmental measurements for workers exposed to vinyl chloride. Am Ind Hyg Assoc J 33:19-30. Marsteller HJ. Lelbach 3VK, Muller R, Gcdigk P (1975): Unusual splenomegalie liver disease as evidenced by peritoneseopv and guided liver biopsy among polyvinyl chloride production work ers. Ann NY Acad Sci 246:95-134. Masuda Y (1980): Long-term mortality study of vinyl chloride and polyvinyl chloride workers in a Japanese plant. Pro XIX Intern Congress Occup Health 1:403-409. Matsuki S, Kataoka K, Atsuji H, Sobara H, Suzuki Y (1970): A comment on Broca's index for obesity judgement. Clin Nutr (Rinsho Eiyo) 36:331-334 (in Japanese). Nicholson WJ, Hammond EC, Seidman H, Selikoft IJ (1975): Mortality experience of a cohort of vinyl chloride--polyvinyl chloride workers. Ann NY Acad Sci 246:225-230. Okuno T, Kume H, Haga T, Yoshizawa T (1971): "Method of Multi-Variate Analysis." Tokyo: NikkaGiren, pp 323-372 (in Japanese). Okuno T, Haga T, Yajima K, Okuno C, Hashimoto $, Furukawa Y (1978): "Method of Multi-Variate Analysis (Continued)." Tokyo: Nikka-Giren, pp 147-174 (in Japanese), Hepatoangiosarcoma Among Vinyl Chloride Workers 417 Popper H, Thomas LB (1975): Alterations of liver and spleen among workers exposed to vinyl chloride. Ann NY Acad Sci 246:172-194. Schcffe H (1959): "The Analysis of Variance." New York: Wiley, pp 112-119. Shioya M, Asano C (1967): "Method of Multi Variate Analysis: Lecture of Information Science," A.5.3 Tokyo: Kyoritsu-Shuppan, pp 132-178 (in Japanese). Vcltman G. Lange CE. Jiihe S. Stein G. Bachner U (1975). Clinical manifestations and course of vinyl chloride disease. Ann NY Acad Sci 246.6-17. Wakahayashi M. Hosomi H, Mori H, Tasaki T (1976): Normal values and multi variate analysis. Nihon Rinsho 34 (autumn extra print): 1782-1798 (in Japanese). Waxweiler RJ, Stringer W, Wagoner JK. Jones J (1976): Neoplastic risk among workers exposed to vinyl chloride. Ann NY Acad Sci 271:40--48, Wyatt RH, Kotchen J.M. Hochstrasser DL. Buchanan JW, Jr, Campbell DR, Slaughter JC (1975): An epidemiologic study ofblood screening lest and illness histones among chemical workers invoked in the manufacture of polyvinyl chloride. Ann NY Acad Sci 246:80-87. ASI 00004798 V