Document mpV3bYeEaq3X2Mxya2V8mdrjk

696 British Journal of Industrial Medicine 1991;48:696-701 Exposure to mixtures of solvents among paint workers and biochemical alterations of liver function Jong-Dar Chen, Jung-Der Wang, Jenn-Pyng Jang, Yann-Yuh Chen Abstract Because the possible effects on GGT activity of The objective of this study was to determine non-occupational factors were controlled for, biochemical alterations of liver function it is concluded that increased GGT activity among paint manufacturers and sprayers among exposed workers may be due to a higher associated with exposure to organic solvents. exposure to the mixture of solvents. Two paint manufacturing factories and 22 various kinds of spray painting factories (16 Organic solvents are widely used in various indus- car painting, two aircraft painting, three video tries such as paint manufacturing, spray painting, terminal painting; and one trailer painting) and shoemaking. Hepatotoxic effects of some of these were included. Air concentrations of organic solvents, for example, carbon tetrachloride and solvents were collected by personal samplers chloroform, were early recognised. Because some and analysed by gas chromatography. A total solvents produce a chronic hazardous health effect, of 180 workers were given a comprehensive however, it usually takes decades to document such a physical examination, a questionnaire, a liver problem. For example, vinyl chloride was once function test, and a test for hepatitis B surface thought to be safe and had been used for 42 years antigen. The questionnaire contained ques- until it was found to cause liver angiosarcoma.1' An tions regarding detailed personal medical his- occupational physician has to maintain vigilance on tory, intake of alcohol, and use of medicine. exposure in the workplace even though the exposure Mixtures of solvents were used throughout the might be thought to be safe for a certain period.' In factories, and xylene and toluene were the the meantime, more epidemiological studies of these major components found in almost all air agents in the environmental situation should be samples with average contents of 46% and 29% performed to help the policy makers to formulate on a weight basis of 67 air samples. No strong levels ofexposure that do not have detrimental effects hepatotoxic solvents were detected. Workers on health. were classified according to the different Today, with the increasing consumption ofalcohol exposure patterns and different air concentra- and medicine, workers are often exposed to the tions of breathing zones as: high (eight hour combined action of occupational chemicals, alcohol, time weighted average (8 h TWA) hygienic and medicine. Because the mechanism of solvent effects of solvents 0-25-9-83, median 1-66), short medicine alcohol interaction is usually not clear, term high (8 h TWA hygienic effects of solvents difficulty may occur in interpreting the results of an 0-3-38, median 0-12), and low (8 h TWA epidemiological study involving the interaction.34 hygienic effects ofsolvents all below 0 38). After To partially solve this problem, a study design that applying a multivariate model to control the distinguishes several mutually exclusive categories of non-occupational factors (alcohol, medica- potentially hazardous exposure to obtain a reference tion, age, and hepatitis B viral infection), category may limit the possible confounders.' increase in y-glutamyl transferase (GGT) Besides, one should take a falsification attitude and activity was found to be associated with consider alternative explanations for the abnormal severity of exposure to the mixture of solvents. liver function, such as that caused by viral hepatitis, alcohol, or medicines.5 Many investigators have reported that the spray painting operation and paint Center for the Research of Environmental and Occupational Disease, Graduate Institute of Public Health and Department of Internal Medicine, National Taiwan University, College of Medicine J-D Wang, J-P Jong, Y-Y Chen manufacturing industry usually do not produce damage to the liver.' As the exhaust ventilation is generally not well designed, or installed, in Taiwan, however, we are still concerned that paint manufacturers and sprayers might be suffering from health Department of Family medicine, Provincial Tao- hazards in the workplace. The purpose of this study Yuan General Hospital, Taiwan, Republic of China J-D Chen was to determine biochemical alterations in liver function among paint manufacturers and sprayers Exposure to mixtures of solvents among paint workers and biochemical alterations of liver function 697 and their association with exposure to the organic solvents used in Taiwan. As the prevalence of hepatitis B carriers is as high as 15-20% in Taiwan9 the opportunity was readily available for us to evaluate the possible interaction between the effects of hepatitis B infection and solvents in this study. Material and methods Taipei city has two factories for paint manufacturing, three factories for video terminal painting, two factories for aircraft painting, and one factory for trailer spray painting. All workers who had been working in these factories for more than one year were studied. Also, 200 car painting factories are found in Taipei city and a random sample of 16 of thenm were included. In total 180 workers were studied. All the workers were given a comprehensive physical examination, a questionnaire, a liver function test, and a test for hepatitis B surface antigen (HBsAg). The questionnaire contained questions regarding a detailed personal medical history, an extensive occupational history including previous occupational exposure and duration of employment, consumption of alcohol, and medicines taken in the past two weeks. All the interviews were performed by two standardised interviewers in the workplace. We took a walk through survey for each factory before air sampling. Air concentrations of organic solvents were collected by personal samplers and analysed by gas chromatography. We took two to four samples at each factory. A total of 67 air samples were collected, each representing at least four hours of continuous exposure during a complete workday. During the walk through survey we found that car painters had the most serious exposure when working in a poorly ventilated painting booth; they must spend 30-90 minutes in the painting booth every workday. The fluctuation of solvent exposure for car painters could not be neglected, so nine further air samples were randomly chosen from the 16 car painting factories. All nine air samples were collected from the breathing zone when car painters were actually working in the painting booth. The workers were exposed to a mixture of solvents and eight different solvents were detected in the air. The average contents of xylene and toluene in these air samples were 46% and 29%, respectively. The air concentration was expressed as the threshold limit value eight hour time weighted average (TLVTWA) (table 1). Table 2 gives the solvent exposure of car painters with serious exposure in the painting booth. The hygienic effect is adopted as a measure of total solvent exposure and is defined as the sum of the fraction of the TLV that each solvent represents. We divided workers into three groups according to the different exposure patterns and different categories of air concentrations in our analysis. These three groups are defined as: Exposure index 0: This is the low exposure group, including three video terminal painting (35 workers) and two aircraft painting (seven workers) factories. They were exposed to a low concentration of organic solvents; the individual solvents detected in the Table 1 Detected solvents in the air samples of solvent exposed workers. Allfigures are expressed as TL V-TWA (No of samples = 67, sampling period = four hours) Solvents Range (ppm) No of samples that Mean (SD) (ppm) TLV-TWA (ppm) exceeded the TLV-TWA Xylene Toluene Acetone Benzene Methyl isobutyl ketone Methyl ethyl ketone Ethyl acetate Butyl acetate 0-365 0-540 0-124 0-20 0-68 0-70 0-29 0-41 30 (75) 19 (72) 8 (24) 2 (4) 2 (10) 2 (9) 1 (4) 2 (7) 100 100 750 10 50 200 400 150 7 1 0 3 1 0 0 0 TLV = Threshold limit value recommended by the American Conference of Governmental Industrial Hygienist (ACGIH)." Table 2 Detected solvents in air samples of car painters working in painting booths. Figures are expressed as 15 minute time weighted average (No of samples = nine, sampling period = 15 minutes) Solvents Toluene Xylene Ethylacetate Methyl isobutyl ketone Benzene Butyl acetate Range (ppm) 12-948 25-511 0-483 8-369 7-138 0-258 Mean (SD) (ppm) 388 (391) 231 (192) 115 (165) 76 (115) 53(40) 50 (81) No of samples (%) that TLV-STELt (ppm) exceeded the TLV-STELt 150 5 (55) 150 4 (44) 400t 1 (11) 75 2 (22) 25 8 (88) 200 1 (11) tTLV-STEL recommended by the American Conference of Governmental Industrial Hygienists." tNo recommended value of TLV-STEL, 400 ppm is the value of TLV-TWA. 698 Chen, Wang, Jang, Chen personal samplers were all below a quarter of the hygienic effect concentrations recommended by the American Conference of Governmental Industrial Hygienists (ACGIH)." The 8 h TWA hygienic effects of solvents ranged from 0 to 0-38; and the median level was 0-12. Exposure index 1: This is the short term high exposure group, including 80 workers who were selected from 16 car painting factories. They were exposed to a low concentration of organic solvents; the 8 h TWA hygienic effects of solvents ranged from 0 to 3 38; the median value was 0 12. They spend 3090 minutes in the poorly ventilated painting booth every workday; the 15 minute TWA hygienic effects of solvents while working in the painting booth ranged from 1 32 to 24 52; the median value was 10 40. Because such a high level of exposure is usually limited to less than 90 minutes among car painters, they are defined as the short term high exposure group. Exposure index 2: This is the high exposure group, including two paint manufacturing (19 workers) and one trailer painting (39 workers) factory. Most of them were exposed to xylene and toluene. The air concentrations of xylene and toluene detected in the breathing zone were usually above those recommended by ACGIH. The 8 h TWA hygienic effects of these solvents ranged from 0-25 to 9-83; the median value was 1-66. Evaluation of biochemical alterations in liver function was performed by measuring aspartate aminotransferase (AST), alanine aminotransferase (ALT), y-glutamyltransferase (GGT), and alkaline phosphatase (ALP) activities, and serum bile acid (SBA) concentrations. Concentrations of SBA were measured by an enzymatic method (Enzabile, Oslo, Norway). Hepatitis B surface antigen was studied with a commercially available solid phase radioimmunoassay (Austria-II, Abbott Laboratories, Chicago, Ill). A 10 ml venous blood sample was drawn from a cubital vein of each worker in the morning (between 0800 and 1000); all the workers were asked to fast overnight. The blood samples were put in a 4C refrigerator in the workplace and immediately centrifugated to obtain the serum after returning to our laboratory. The serum was kept frozen at - 20C until analysed (within four days). Having found that the distributions of the activities of liver enzymes and SBA concentration are not Gaussian, the non-parametric Wilcoxon rank sum test was used for group comparison. The prevalences of abnormal liver function tests and SBA concentrations between different groups were analysed by the Mantel extension for the test of trend."2 The activities of ALT, AST, GGT, ALP, and SBA were also analysed by multivariate regres- sion analysis. Results A total of 180 workers were investigated, divided into three groups according to the different exposure patterns and different categories ofair concentrations. In general, workers with high exposure were older and employed longer compared with the other two groups (table 3). Three workers who had a history of blood transfusion were excluded from the analysis to prevent possible confounding by non-A, non-B hepatitis. Table 4 presents results on the influence of nonoccupational factors (alcohol, medicines, HBsAg) on liver function. It showed that workers who had an alcohol consumption of greater than 40 gm a day or took medicine (except vitamins and antacids) in the past two weeks were associated with an increase in GGT activities; but this did not reach statistical significance. Significant increases in ALT activity and in SBA concentration (p < 0 05), were noted in HBsAg carriers. The standard deviations of liver function tests were comparatively large (table 4) owing to the inclusion of some workers with clearly pathological levels at the time of the study. After workers were stratified individually by age, duration Table 3 Eight hour hygienic effect of solvents, age, duration of employment, smoking and alcohol consumption, medicines, and HBsAg carriers of workers with different indices of exposure Index of exposure 0 (low) I (short term high) 2 (high) p Value of ANOVA Total no of workers examined Eight hour TWA hygienic effect of solvents Range Median Age (y; mean (SD)) Duration of employment (y; mean (SD)) No of smokers (%) No with alcohol consumption exceeding 40 g/day (%) No of blood transfusions (%) No with medicines in past two weeks (%) No of HBsAg carriers (%) 42 0-0 38 0-12 30-3 (7 6) 5-7 (7-4) 20 (47 6) 0 1 (2 4) 1 (2-4) 9 (21-4) tThe 15 minute TWA hygienic effect of solvents in painting booth. 80 0-338 (13-24-5)t 0-12 (10-40)t 26-4 (8-7) 6-4 (6-7) 52 (65 0) 4 (5-0) 1 (1-3) 6 (7 5) 19 (23 8) 58 0-25-9-83 1-66 41.0 (12-4) 17-6 (13-3) 40 (70-0) 3 (5-1) 1 (1-7) 8 (13-8) 10 (17-2) <0 05 <0 05 Exposure to mixtures of solvents among paint workers and biochemical alterations of liver function 699 Table 4 Serum activities of liver enzymes (ALT, AST, GGT) and concentrations of SBA among workers stratified by alcohol consumption, medicines, and HBsAg ALT (y/1/) AST (p/1) Alcohol consumption: >40 gm/day (n = 7) <40 gm/day (n = 170) p Value Medication: Yes (n= 15) No (n = 162) p Value HBsAg carriers: Carriers(n = 38) Non-carriers (n = 139) p Value 12 (13) 9 (8) NS 14(14) 12 (13) NS 15 (11) 11(14) < 0 05 18 (13) 17 (15) NS 18 (8) 18 (13) NS 20(14) 17(12) NS Three workers who had had previous blood transfusion were excluded before analysis. Results are means (SD); NS = non-significant. GGT (p/1) 18 (9) 14 (21) < 0-05 21(20) 13 (21) NS 12(8) 15 (23) NS SBA (pmol/l) 7 (11) 5 (4) NS 9 (15) 7 (11) NS 10(19) 6(7) < 0 05 of exposure, consumption of alcohol, HBsAg, and use of medicine, we found that GGT activities increased universally among groups of short term high exposure and high exposure compared with the low exposure group. We applied a multiple linear regression model to take account of possible confounding effects. The results showed that GGT activity was associated with both exposure and consumption of alcohol. This association persisted even if the alcohol indicator was expressed as grams ofalcohol (g) consumed per day (table 5). Because the age and the duration of exposure are highly correlated with the severity of exposure (Pearson correlation coefficients r = 0 38 and 0 40, respectively), we limited the workers to only those who were less than 30 years of age, and the association still persisted; the age and duration of exposure did not influence GGT activity. The concentrations of SBA and ALP activity were associated with the HBsAg state (table 5). After excluding those who had had previous blood transfusion (three workers) or whose alcohol con- sumption was greater than 40 g a day (seven workers), the prevalence rate of abnormal enzyme activity and SBA concentration (more than 10 Mmol/l) were 6-5% (11/170) and 12-4% (21/170). If we consider the frequencies of abnormal enzyme activity and SBA concentration among different indices of exposure, however, no statistically significant trend was found because of the small numbers when stratified by HBsAg carrier state (table 6). Discussion No strongly hepatotoxic solvents were detected in this study. The major components of solvents were xylene and to a lesser degree toluene. The median air concentrations of 8 h TWA hygienic effects of solvents for exposure indices 0, 1, and 2 were 0 12, 0-12 (10-40 in painting booth), and 1 66 respectively. These concentrations were similar to those reported by some other investigators of paint industry workers7 13; toluene, however, dominated the solvent mixtures used by car painters in Finland.6 The total prevalence rate of abnormal liver function in our Table 5 Effects of solvent exposure, alcohol consumption, HbsAg, and medicines on serum activities of liver enzymes and concentration of SBA among solvent exposed workers by multiple linear regression analysis. (No of workers = 177) Independent variable Exposure index Alcohol consumption HBsAg Medication Categorisation Dependent variable ALT AST GGT ALPt SBA GGT (age < 30) (n = 102) Intercept 15-2 11 0 7-2 133-3 4-7 6-2 0: low exposure 1: short ten high exposure 2: high exposure bl (SE) -0-28 (1-35) 1-14 (1-31) 4-88 (2-07)* 6-62 (7-42) 0-75 (1-15) 3-60 (1-14)* 0: < 20 g/day 1: > 20 g/day < 40 g/day 2: > 40 g/day b2 (SE) 0-46 (2-18) 2-60 (2-12) 9-47 (3-36)* -7-47 (11-98) 0-28 (1-87) 5 70 (2-77)* 0: negative 1: positive b3 (SE) 4-11 (2-37) 3-40 (2-29) -0-81 (3-63) 29-26 (12-9)* 4-40 (2-02)* 0-79(1-73) 0: no 1: yes b4 (SE) 2-57 (3-55) -9-17 (3-43) 5-00 (5 43) -13-26 (19-11) 2-36 (3-02) 3-81 (2-80) *The p value of t statistic for coefficient of regression is < 0-05. tBecause the record of ALP activity of 19 workers from the first paint manufacturing factory (exposure index 2, n = 10) and first car painting factory (exposure index 1, n = 9) were lost, only 158 workers were analysed for ALP activity. 700 Chen, Wang, Jang, Chen Table 6 No of workers with abnormal liver function (defined as ALT >35 p1l, or AST> 35 p11, or GGT >60 pl1) or raised SBA concentration among different exposure indices stratified by HBsAg State of HBsAg Index of exposure Mantel-extension 0 1 2 for the trend Conventional liver function test: HBsAg(+)(n = 38) HBsAg (-)(n = 132) Total(n = 170) Odds ratio SBA concentration: HBsAg(+)(n = 38) HBsAg (-) (n = 132) Total(n = 170) Odds ratio 2/8 0/20 2/10 NS 1/33 3/55 3/44 NS 3/41 3/75 5/54 1.0 0-52 1 34 NS 0/8 5/20 1/10 NS 4/33 7/55 4/44 NS 4/41 12/75 5/54 1.0 1-64 0 93 NS Workers who had had previous blood transfusion or intake of alcohol exceeding 40 g/day were excluded. study was 6 5%, and no association was found between the frequency of abnormality and exposure. The serum activities of liver enzymes and concen- tration of SBA seemed to be influenced by nonoccupational factors (table 4). Some authors argued that measurement of SBA concentration is more sensitive than conventional liver function tests such as ALT, AST, and GGT activities and bilirubin concentraton.'3 14 We also found that the prevalence rate of abnormal SBA concentration (12 4%) was higher than found by the conventional liver function tests (6 5%). Serum concentrations of bile acid, however, were not different among our three categories of exposure (table 5). After applying a multivariate model to control for all non- occupational factors, GGT activity was associated with the severity of exposure to the mixtures of solvents. This association persisted even after we limited the workers to those who were less than 30 years of age (table 5). Because the possible effects of alcohol, HBsAg, medicines, and age on GGT activity were under statistical control, we conclude that increased GGT activity among exposed workers may be due to exposure to the mixture of solvents. We consider that GGT activity is probably more useful as a measure than SBA concentration in workers' health surveillance when they are exposed to a mixture of solvents, especially in countries where use of alcohol is not prevalent and xylene is the major solvent used. Activity of GGT has been generally regarded as a sensitive test for cholestasis and liver cell damage, it is also susceptible to certain drugs and alcohol.'5 Our data showed that GGT activities increased independently with both an increased consumption ofalcohol and exposure to mixtures of solvents. Although it has been reported that long term exposure to low concen- trations of solvents may lead to steatosis,1617 the magnitude of increase in GGT activities for workers in our series seems to be still at a sub-clinical level. Moreover, we found no trend for change in alkaline phosphatase activity (table 5), so we tentatively concluded that the increase in GGT activity may be a phenomenon of enzyme induction. We recommend that a longitudinal study with a long term follow up should be performed to define more clearly the possible hepatotoxic effect among paint workers. This can be done in combination with several other dynamic and static diagnostic tools for the early recognition of possible hazard caused by exposure to the mixture of solvents. Are HBsAg carriers more susceptible to exposure to a mixture of organic solvents? As a whole, ALT and SBA activities were significantly higher among HBsAg carriers in our study (table 4). But after being stratified by the index ofexposure, the prevalence rate of abnormal liver functions and SBA concentration showed no significant trend among different exposure groups. Although we found an increased SBA concentration among HBsAg carriers after controlling other factors (table 5) no specific trend with the exposure index occurred. We conclude, therefore, that for a moderate level of exposure to the mixture of non-chlorinated organic solvents (for example a median 8 h hygienic effect up to 1-66), HBsAg carriers seem not to be more susceptible than workers who do not have HBsAg. This study was supported by a special grant offered from the Taipei City Government. We are also indebted to Drs Ming-Yang Lai and Pei-Ming Yang for their constructive comments. Requests for reprints to: Dr Jung-Der Wang, Graduate Institute of Public Health, National Taiwan University, College of Medicine, No 1, Section 1, Jen-Ai Road, Taipei, Taiwan, Republic of China (10018). 1 Axelson 0. Solvents and the liver. Eur J Clin Invest 1983;13: 109-11. 2 Greech JL, Johnsson MN. Angiosarcoma of the liver in the manufacture of polyvinyl chloride. J Occup Med 1974;16: 150-1. Exposure to mixtures of solvents among paint workers and biochemical alterations of liver function 701 3 Dossing M. Occupational toxic liver change. J Hepatol 1986;3:131-3. 4 Dossing M. Metabolic interaction between organic solvents and other chemicals. In: Riihimaki V, Ulfvarson U, eds. Safety and health aspects of organic solvents. New York: Alan R Liss 1986:97-105. 5 Wang JD. Concepts and methods of epidemiologic research. Taipei: National Taiwan University College of Medicine Academic Press, 1989:9-25. (In Chinese.) 6 Kurppa K, Husman K. Car painters' exposure to a mixture of organic solvents. Serum activities of liver enzymes. Scand J Work Environ Health 1982;8:137-40. 7 Lungberg I, Hakansson M. Normal serum activities of liver enzyme in Swedish paint industry workers with heavy exposure to organic solvents. Br J Ind Med 1985;42:596-600. 8 Kurppa K, Tola S, Hernberg S, et al. Industrial solvents and the liver. Lancet 1983;i:129. 9 Chen DS, Sung JL. Hepatitis B virus infection and chronic liver disease in Taiwan. Acta Hepatogastroenteral 1978;25:423-30. 10 Sung JL, Chen DS, Lai MY, et al. Epidemiology study on hepatitis B virus infection in Taiwan. Chinese Journal of Gastroenterology 1984;1: 1-9. 11 American Conference of Governmental Industrial Hygientists. Threshold limit values for chemical substances in the work environment adopted by ACGIH. Cincinnati, Ohio: ACGIH Inc, 1988. 12 Mantel N. Chi-square tests with one degree of freedom: extension of Mantel-Haenszel procedure. Journal of the American Statistical Association 1963;58:690-700. 13 Franco G, Fonte R, Tempini G, Candura F. Serum bile acid concentrations as a liver function test in workers occupationally exposed to organic solvents. Int Arch Occup Environ Health 1986;58:157-64. 14 Edling C, Tagesson C. Raised serum bile acid concentrations after occupational exposure to styrene: a possible sign of hepatotoxicity? Br J Ind Med 1984;41:257-9. 15 Anonymous. Serum bile acid in hepatobiliary disease. Lancet 1982;ii: 1 37-40. 16 Sotaniemi EA, Sutinen S, Arranto AJ, et al. Liver injury in subjects occupationally exposed to chemicals in low dose. Acta Med Scand 1982;212:2007-215. 17 Dossing M, Arlien-Soborg P, Petersen LM, et al. Liver damage associated with occupational exposure to organic solvents in house painters. Eur J Clin Invest 1983;13:151-7. Accepted 28 January 1991 Correspondence and editorials The British Journal of Industrial Medicine welcomes correspondence relating to any of the material appearing in the journal. Results from preliminary or small scale studies may also be published in the correspondence column if this seems appropriate. Letters should be not more than 500 words in length and contain a miniimum of references. Table and figures should be kept to an absolute minimum. Letters are accepted on the understanding that they may be subject to editorial revision and shortening. The journal now also publishes editorials which are normally specially commissioned. The Editor welcomes suggestions regarding suitable topics; those wishing to submit an editorial, however, should do so only after discussion with the Editor.