Document Ne4Eoe7ZJYeZ89nVpbY4MDoJw

ROENTGENGGRAPHIC LUNG CHANGES, ASBESTOSIS AND MORTALITY IN A BELGIAN ASBESTOS-CEMENT FACTORY L.M. XJCQOET & L. van der LINDEN Department of Pathophysiology3 Academisoh Ziekenhuis Pellenberg, University of Leuven, Belgium J. LEPOOTEE Etemit N. V., Kapelle op due Bos3 Belgium iI 1 ii ! j ! The Etemit factory at Kapelle op de Bos employs about 2000 workers and 400 staff members and processes annually about 35,000 tons of chrysotile, 3000 tons of crocidolite, and 1000 tons of amosite for the manu facture of a variety of asbestos-cement building materials and pipes. In the present study, findings of yearly roentgenographic examina tions of the lung and the mortality of workers are related to the age of the individual and to the duration and level of dust exposure. We looked for possible effects of exposure to asbestos on the occurrence of pneumoconiosis, pleural changes and malignancies. !i i iS Ia f ii 11 METHODS Dust concentration We considered the following five typical working areas, which are given in decreasing order of dustiness: area 4, which involves the handling of asbestos fibres, such as carrying asbestos bags, milling asbestos, preparing the asbestos-cement mixture; area 3, where asbestoscement products are finished by sawing, drilling, filing, etc.; area 2, which is situated between the previous two and where the asbestos-cement sheets and pipes are moulded, pressed, dried, and lifted off the mould; i -783- HWBUI0009417 784 LACQUET ET AL. area 1 where very little dust exposure is expected, such as in offices; and area 0, which represents work outside the asbestos industry, with negligible dust levels. Actual fibre counts obtained with the filter-membrane methods are available for 1970 through 1976; dust concentrations were much higher in previous years, but the actual values can only be estimated. Dust concentrations are believed to have followed a more or less logistic decay, with an inflection point at the year 1960, which is satisfied by the formula Oy - eQ!(1 + 1.162/-1960), where a is concentration in fibres/ ml, y is a number (year) between 1928^-1977, and oQ is 0.4, 16, 24 or 100 fibres/ml for areas 1, 2, 3 or 4, respectively (i. e., roughly 10 times present-day levels). Fibre-years The degree of individual exposure to asbestos was expressed in fibres per ml times duration in years, or 'fibre-years '. Since employees may have worked consecutively in different areas, fibre-years for any one individual were calculated as Z (fibres/ml)^^, where i is the type of area (0, 1, 2, 3, 4) and y is the year. 3 Roentgenologic study In 1975, 2650 employees were enlisted, of which 1973 (74.5%) had both a history taken and a chest roentgenogram made; they are distributed as follows: 1963 male workers (85.5% of those enlisted), 214 female workers (81.4%); 78 male (25.1%) and 2 female (1.8%) staff members. The yearly chest roentgenogram is compulsory for all workers, while the staff submit to this examination on a voluntary basis. The roentgenologic findings were coded according to the extended system elaborated by the International Union Against Cancer. Irregular small opacities were graded according to the full 12-point scale of the UICC/Cincinnati (1970) Classification. The pleural changes were classified as: (a) pleural adhesion; (b) blurred contour of heart and/or diaphragm; (c) pleural thickening; or (id) calcified pleural plaque. Three grades (1 = minimal, 2 = moderate, 3 = marked) were noted but were pooled for analysis because of the small number observed. Pleural adhesions are exudative sequelae, e.g., obliteration of the costophrenic angle. Pleural thickening comprises 'hyaline' plaques and probably a few cases of post-exudative diffuse thickening, as it is not always possible to distinguish these two on a routine chest roentgenogram. Also noted were changes suggest ive of cancer, tuberculosis, sarcoidosis, heart disease, active or inactive tuberculosis and trauma (e.g., rib fractures, thoracotomy). n HWBUI0009418 ROENTGENOGRAPHIC LUNG CHANGES IN BELGIAN FACTORY 785 Diagnosis of asbestosis The criteria for making the diagnosis of asbestosis were three fold: the presence of irregular small lung opacities with profusion of at least 1/0; and the presence of at least two of the following ab normalities : ill-defined outline of heart and/or diaphragm, clubbing of the fingertips, crackling rales audible over the lung bases, carbon monoxide transfer or vital capacity less than 76% of the predicted value. A decrease of vital capacity was disregarded if it could be attributed to some obvious restrictive lesion other than asbestosis or to obstructive lung disease. The mere presence of pleural plaques was not used as a criterion for asbestosis. Asbestosis and mortality study The population studied consisted only of male workers who, within the 15-year period 1963-1977, worked in the factory for at least 12 months. All causes of death were checked through family doctors and/or social workers who visited the relatives (Belgian authorities never release individual information from death certificates) Expected mortalities by age group and by cause of death were cal culated from the yearly mortality rates for Belgium, available in the tables of the World Health Organization for 1965 through 1975; data for other years were obtained through extrapolation. RESULTS Table 1 shows the number of chest roentgenograms with small lung opacities and with pleural changes, by group of exposure. We saw no small opacities with profusion greater than 2. We pooled all grades of pleural changes because the numbers were small; grade 3 was scored twice and only on account of calcified plaques. We found 29 cases of asbestosis during the 15-year observation period (Table 5). Total mortality, by age and cause of death, is given in Table 6; and details of malignant neoplasms are given in Table 7. Among the cases of respiratory cancer there were three cancers of the upper airways, 17 cancers of the lung and one mesothelioma of the pleura. HWBUI0009419 786 LACQUET ET AL. Table 1. Number of roentgenograph!c lung changes found in 1958 employees examined in 1975 Fibre-years 0-49 50 - 99 100 - 199 200 - 399 400 - 799 800 -1599 Total Small opacities 0/- 0/0 0/1 1 2 128 886 46 2 0 3 225 29 2 0 5 302 35 9 0 5 184 40 11 1 0 22 6 2 1 1 11 r" 1 1 _0 142 1630 157 27 2 Pleural changes Adhesion 1 ,2 Blurred contour 1 ,2 36 0 15 0 34 3 30 2 71 3 0"" 125 6 Thickening 1 ,2 15 7 23 16 5 1 67 Calcified plaque 1,2,3 4 7 6 6 0 0 23 DISCUSSION Fibre-years We realize that our estimates of dust concentrations prevailing in the past are only good guesses at best. Nevertheless, it is unlikely that they would be inaccurate by more than one order of magnitude, and, more importantly, the figures proposed have at least relative accuracy the dustiness in areas where dry asbestos fibres are being handled is certainly greater than in other areas, and the dustiness was certainly higher during the first 30 years than during the last 15 years of the factory's existence. Roentgenograpkia changes We excluded 15 cases from the study, either because of a history of sarcoidosis or of exposure to silica dust which would lead to inter pretation of roentgenographic changes as being due solely to asbestos exposure. The profusion of small lung opacities and of benign pleural changes correlate strongly with fibre-years, as well as with age. We therefore examined, by age group, the linear trend in proportions (i.e., number with changes per number X-rayed, by fibre-years groups). The trends are significant for small lung opacities grades 1 and 2 (Table 2) and grade 0/1 (not shown, P = 0.001); the trend is also significant for pleural.adhesion (Table 3), and for pleural thickening (Table 4); the trend is not significant for calcified plaques (not shown). To rule out the effect on our analysis of post-inflammatory pleural changes unrelated to asbestos exposure, we reexamined the given trends in proportion after eliminating 175 cases with evidence of old or recent HWBUI0009420 ROENTGENOGRAPHIC LUNG CHANGES IN BELGIAN FACTORY 787 Table 2. Prevalence {%) of :s-t-u lung opacities grades 1 and 2 Fibre-yearsa Age group (yrs) .15-34 35-44 45-54 55-65 Total 0 - 49 50 - 99 100 - 199 200 - 399 400 - 799 800 - 1599 0 1.1 0 0 .0 0 1.1 2.9 00 2.5 7.5 0 4.5 2.2 9.8 00 5.9 15.4 0 0 33.3 0 0.2 0.8 2.6 5.0 9.7 7.1 Total % 0 0.7 1.7 5.3 1.5 a Median taken as the score Linear trend with pooled regression coefficients: P = 0.001 Table 3. Prevalence (%) of pleural adhesions Fibre-years'2 Age group (yrs) 15-34 35-44 45-54 0 - 49 50 - 99 100 - 199 200 - 399 400 - 799 800 - 1599 1.2 3.2 6.7 0 4.2 6.4 0 3.3 12.3 0 4.5 11.7 0 0 17.6 0 0 33.3 Total % 1.1 3.5 9.8 55-65 11.7 14.3 14.9 15.9 30.8 22.2 15.0 a Median taken as the score Linear trend with pooled regression coefficients: P = 0.002 Total 3.4 5.8 9.7 12.4 22.6 21.4 6.4 HWBUI0009421 788 LACQUET ET AL. Table 4. Prevalence (%} of pleural thickening Fibre-yearsa Age group (yrs) 15-34 35-44 45-54 0 - 49 50 - 99 100 - 199 200 - 399 400 - 799 800 - 1599 0 0.5 3.3 00 6.4 0 2.5 8.0 00 9.5 00 17.6 00 33.3 Total % 0 0.9 7.1 55-65 8.5 2.9 10.4 3.7 15.4 11.1 7.3 a Median taken as the score Linear trend with pooled regression coefficients: P = 0.01 Total 1.4 2.7 6.6 6.6 16.1 7.1 3.4 chest trauma (e.g., rib fracture), tuberculosis (e.g., hilar calcification), or empyema. This did not alter the conclusions (for pleural adhesion, P = 0.008; for pleural plaques, P = 0.005). Asbestosis The number of cases with diagnosed asbestosis increases highly sig nificantly with the degree of exposure, expressed in fibre-years (Table 5). We observed no case of asbestosis for an exposure below 100, and only one case in the range 100-200 fibre-years. Two workers who had never handled free asbestos fibres (area 4) developed asbestosis, but their degree of exposure (starting in 1930) had nevertheless been high: 435 and 350 fibre-years, respectively (on account of work in area 3). The present study thus indicates that the handling of asbestos-cement is less dangerous than the handling of free asbestos fibres; however, since less dust is generated by the former, we cannot conclude from the present data whether asbestos-cement dust is less fibrogenic per se than free asbestos. Mortality Total mortality was not significantly different from that which would be expected in a Belgian population of matched age and sex (Table ).8 There is, however, a significant excess mortality due to external causes. There is also an excess mortality, not really significant, due to non-malignant respiratory disease, which probably must be attributed to the seven deaths due to asbestosis (Table 6). . ') HWBUI0009422 ROENTGENOGRAPHIC LUNG CHANGES IN BELGIAN FACTORY Table 5. Incidence of asbestosis from 1963 to 1977 Fibre-;years0 Man-years New cases No. % 0 - 49 50 - 99 100 - 199 200 - 399 400 - 799 800 - 1599 1600 - 3200 Total 11,340 4094 5791 6629 1179 307 26 29,366 00 00 1 0.02 7 0.11 8 0.68 10 3.26 _3 11.54 29 0.1 a Median taken as the score Linear trend in proportions: P < 10"6 789 Table 6. Deaths observed in male workers by age and by cause of death, 1963-1977 Age group Man-years 15 - 24 25 - 34 35 - 44 45 - 54 55 - 65 Total Expected ? 3463 5142 6261 9340 5160 29,366 External cause Malignant cause Mon-malignant cause Cerebro- cardiovascular Respiratory 3 7i 83 1 6 1 13 11 12 33 16 37 5 14 43 29.0 o.or 48 55.4 NS 60 77.8 NS 20" 15.9 0.06s Gastro intestinal 6 5 1.1 9.0 0.10s Other Unknown or . poorly specified cause 1 34 47 8 11 9.2 16.7 NS 0.04s Total 3 10 18 58 112 . 201 213 NS ' Include asbestosis in 7 cases ' In the standard normal approximation of the binomial distribution ' In a Poisson distribution with the expected number as the mean MS - not significant The mortality due to all malignant neoplasms does not exceed expectation (Table 7), but there is a significant excess of deaths due to, and only to, gastrointestinal cancer (Table 8) . A similar excess of gastrointestinal cancer, and not of respiratory cancer, was found by Van de Voorde et al. (1967) for workers employed in this same factory. t*U I i 5 f :S i !{ .i HWBUI0009423 790 LACQUET ET AL. Table 7. Deaths due to malignant neoplasms observed in male workers, by age and site (29,366 man-years) Age group Respiratory 15 - 24 25 - 34 35 - 44 45 - 54 55 - 65 Total Expected P* 2 6 14" -- 22 22.3 NS Gastrointestinal 3 14 17 11.8 0.04 Nervous Lymphoid & haematopoietic Other 4 .... . 4 2.1 0.10 1 1 ------- - 2 3.9 NS 2 -- 2 2.4 NS " Includes one case of pleural mesothelIoma In a Poisson distribution, with the expected number as the mean NS - not significant Not sped fi ed 1 1 12.9 Total 1 2 n 34 ...-- 48 55.4 Table 8. Deaths of male workers due to respiratory and gastrointestinalcancer, by exposure group Fibre-yearsa Man-years Respiratory cancer 1963-1977 Observed Expected" Gastrointestinal cancer Observed Expected" 0 - 49 50 - 99 100 - 199 200 - 399 400 - 799 800 - 1599 1600 - 3200 11,340 4094 5791 6629 1179 307 26 6 3 5 4 1 2 1 5.16 2.43 4.60 7.47 1.95 0.57 0.17 4 1 1 7 2 2 0 2.76 1.32 2.46 3.93 1.00 0.29 . 0.04 Total 29,366 Linear trend in proportion 22 22.35 P = 0.22 17 P = 0.16 11.80 a. Median taken as the score b Adjusted for age distribution within fibre-years groups G Pleural mesothelioma ) HWBUI0009424 / ROENTGENOGRAPHIC LUNG CHANGES IN BELGIAN FACTORY 791 Table 9. Distribution by exposure group of cancer cases and of matched internal controls (4 per case) Fibre-yearsa Respiratory cancer Cases Controls Cases/Total Gastrointestinal cancer Cases Controls Cases/Total 0 - 49 6 20 50 - 99 3 17 100 - 199 57 200 - 399 4 32 400 - 799 18 800 - 1599 24 1600 - 3200 JL Total 22 88 Linear trend P = 0..11 in proportions 0.23 0.15 0.42 0.11 0.11 0.33 1.00 4 16 15 1 14 7 20 28 25 --- 17 68 P = 0.23 0.20 0.17 0.07 0.26 0.20 0.29 a Median taken as the score b Pleural mesothelioma Either this excess is due to asbestos exposure, or the workers studied differ from the general Belgian population. An argument against an effect of asbestos exposure is that we found no significant relationship ).between excess of gastrointestinal cancer and fibre-years (Table 8 The validity of comparing the mortality in a population of enlisted workers with national statistics might be questionable. To eliminate the so-called 'healthy worker effect' we made comparisons with internal case-controls; the principles of the method were described by Liddell et al. (1977). In the present analysis, for each male worker who died of respiratory or gastrointestinal cancer, we selected four controls, at random, out of the group of all male workers alive at least one year after the case worker had expired, and. matched for age 1 year and for date of enlistment 1 year. We compared the distribution of the cases and the controls with respect to dust exposure in fibre-years, which could differ only in the fibres/ml and not in the years: there was no significant difference (Table 9). This means that dust exposure did not significantly affect mortality due to respiratory cancer (even though the one case of mesothelioma is almost certainly related to heavy exposure and is included in the analysis) . HWBUI0009425 792 LACQUET ET AL. CONCLUSIONS The following conclusions were drawn with regard to the workers in the asbestos-cement factory that we investigated: (1) there is a strong dose-response relationship for asbestosis; no case was observed for an exposure of less than 100 fibre-years; (2) there is a dose-response relationship for s-t-u small lung opacities, pleural adhesions and pleural thickening visible on a roentgenogram; (3) general mortality is not in excess of the national expected figures, but there is an excess of accidental deaths; (4) the mortality due to malignant tumours is not in excess of the expected figures; (5) there is no excess mortality due to respiratory cancer; (6) there is an excess mortality due to gastrointestinal cancer, which shows no dose-response relationship when compared with the expected figures or with internal case-controls. SUMMARY Annual chest radiographs, work history and mortality of 1,973 workers in an asbestos-cement factory were correlated with age and with duration and level of dust exposure. Degree of radiographic lung change was significantly related to fibre-years of exposure in the case of small lung opacities, pleural adhesions and pleural thickening. For 29 cases of asbestosis diagnosed between 1963 and 1977, a highly signif icant dose-response relationship was found. In comparison with national mortality rates, there was an excess of deaths due to cancer of the gastrointestinal tract, although there was no relationship to fibreyears . J HWBUI0009426 ROENTGENOGRAPHIC LUNG CHANGES IN BELGIAN FACTORY 793 ' RESUME Les auteurs ont mis en correlation les radiographies thoraciques annuelles, les antecedents professionals et la mortalite de 1 973 travailleurs d'une usine d'amiante-ciment avec 1'age de ces sujets et la duree et le niveau de leur exposition aux poussieres. Le degre de lesion pulmonaire radiographique s'est avere lie de maniere signi ficative aux annees-fibre d*exposition dans le cas des petites opacites pulmonaires, des adhesions pleurales et de l'epaississement pleural. Pour 29 cas drashestose diagnostiques entre 1963 et 1977, une relation dose-reponse tres nette a ete observee. Comparativement aux taux nationaux de mortalite, on notait un excedent de deces par cancer des voies digestives, mais sans relation avec les annees-fibre. RKhyKKNrF-S Liddell, F.D.K., McDonald, J.C. & Thomas, D.C. (1977) Methods of cohort analysis: appraisal by application to asbestos mining. J. R. stat. Soc.3 Ser. A, 140, 469-491 UICC/Cincinnati (1970) Classification of the radiographic appearances of pneumoconioses. Chest, 58, 57-67 Van de Voorde, H., Meulepas, E., Gyselen, A. & Koppen, 0, (1967) Doodsoorzaken bij de bevolking woonachtig rond en bij de arbeiders werkzaam in een asbesfverwerkende. nijverheid in het noorden van Brabant. Aeta tubere. pneumol. Belg., 58, 924-942 ::P' ! iii r 'J ;; I if ;j i| 1 ij j{ :jj jj i! [! HWBUI0009427 INTERNATIONAL AGENCY FOR RESEARCH ON CANCER INSTITUT NATIONAL DE LA SANTE ET DE LA RECHERCHE MEDICALE BIOLOGICAL EFFECTS OF MINERAL FIBRES EFFETS BIOLOGIQUES DES FIBRES MINERALES Volume 2: Clinical and Radiological Observations on Asbestos-Related Pathology - Epidemiology of Asbestos-Related Disease - Scientific Basis for Environmental Control of Fibres General Discussion on Asbestos - Dust Physics and Chemistry of Other Mineral Fibres - Clinical and Radiological Observations on Mineral-Fibre-Related Pathology -- Epidemiology of Groups Exposed to Other Mineral Fibres -- General Discussion on Other Mineral Fibres Proceedings of a Symposium organized by IARC, I'lnstitut National de la Same et de la Recherche Medicale (National Institute of Health and Medical Research), and the MRC Pneumoconiosis Unit, Penarth, UK, held at the International Agency for Research on Cancer, Lyon, France 25-27 September 1979 EDITOR-IN-CHIEF J.C WAGNER TECHNICAL EDITOR FOR IARC W. DAVIS IARC Scientific Publications No. 30 INSERM Symposia Series Volume 92 INTERNATIONAL AGENCY FOR RESEARCH ON CANCER LYON 1980 HWBUI0009428