Document 6b9LM2ELxOE4bYZVbaV52e25g

Radiological changes after withdrawal from asbestos exposure M. R. BECKLAK.E1, F. D. K. LIDDELL1, J. MANFREDA1, AND J. C. MCDONALD* From the 1 Department of Epidemiology and Health, McGill University, Montreal, and the 2TUC Centenary Institute of Occupational Health, London School of Hygiene and Tropical Medicine abstract A group of 135 ex-employees of the Quebec chrysotile production industry comprising men who were bom between 1901 and 1920, who had left the industry between 1950 and 1961, and had been radiographed before leaving, attended in 1972 for a medical examination and further chest radiograph. This report concerns 86 men whose withdrawal film was within 12 months of leaving and who had not been exposed subsequently to asbestos or other fibrogenic dust. Seven separate assessments were made of the paired radiographs, side by side in known temporal order. The assessments were classified as showing `increase' or `no change'. In 66 film-pairs from men with two years' employment or longer, at least four assessments were of parenchymal increase in six, of pleural increase in 13, and of both parenchymal and pleural increase in two, total 21 (or 31 %). There was similar agreement of no change in 24 pairs (36%), and evidence was equivocal in the i iii remaining 21 pairs. Parenchymal increase was not agreed in any of the 20 film-pairs from men with shorter employment, but pleural increase was seen in four of these men, a proportion similar U> that in those exposed longer. We conclude that the parenchymal changes observed after leaving the- industry, most of which were `attacks' rather than `progression', were attributable to the earlier occupational exposure to chrysotile; there were no important differences in age or smoking habit between those with and without parenchymal change. Exposure to airborne dusts containing asbestos appearances in men who had not been exposed tty fibres may lead to depressed lung function, respira asbestos or other fibrogenic dust over a substantial tory signs and symptom's, and radiological changes. period. Second, to determine whether any incidence Of these, small irregular opacities in the parenchyma or progression observed could have been predicted and pieural thickening on chest radiography are the from the work history, taking into account age and more specific signs of asbestosis, if not always the smoking habit first detectable. In many industries it has been the i practice to advise a worker, whose periodic chest Materials and methods radiograph shows early pneumoconiosis, to withdraw from further exposure in the hope that this will delay The Quebec asbestos mortality study (McDonald et or prevent deterioration. However, although Gilson al,, 1971, 1974; McDonald and Lidcfelf, 1978)' (1973) stated that `it is generally agreed that pro embraces all 11 379 of those subjects born between: gression occurs without further exposure when 1891 and 1920 who worked in the chrysotile- asbestosis has reached a degree of severity which producing industry of Quebec for at least a month. permits confident diagnosis', he also remarked that They have been followed from first employment (as in less severe cases `it is still not known whether early as 1904) to the end of 1975, with a high trace there is functional or radiographic improvement rate (more than 98 % of those known to be alive in after removal'. The present study was undertaken 1935). A work history, with dates of start and with two objectives. First, to see whether there was termination of each job, was obtained for each any incidence or progression of pneumoconiotic person, and a measure of accumulated exposure to- airborne respirable asbestos dust was calculated from Received for publication 10 April 1978 Accepted for publication 28 July 1978 assessments of concentration by job and year and expressed in terms of million particles per cubic .i `I I { i B0C2783 6113 18138 J 24 M. R. Berklake, F. D. K. Liddell, J. Manfreda, and J. C. McDonald foot x years, or mpcf.y (Gibbs and Lachance, 1972). From this cohort, we identified 381 men still alive in 1972, who had been born after 1900 and who had left the industry for any reason in the period 1950-61. Of these, 287 were-residing in accessible districts of the Province of Quebec, and for 260 a full size postero-anterior chest radiograph (the `withdrawal radiograph') taken somewhere near the time of leaving the industry, was available. We wished to include not only subjects with heavy exposure and radiographic abnormality, in whom progression might have been expected, but also men with light exposure, in whom progression would be less likely. For this reason the 260 men were classified by the reading of the withdrawal radiograph (Rossiter et al., 1972) and by accumulated dust exposure (Table 1), Table 1 Radiographic findings and dust exposure in 260 possible subjects Accumulated dust exposure {mpcf.y) < 10 10, < 100 100, < 200 200, < 400 400, < 800 800 4- All Findings at time of withdrawal radiograph Segaxive Positive Alt .. 74 86f IS 16 9 6 206 17 20 3 ~6 7 1 54 91 106 18 22 16 7 260 At least one reader (ofsix) recorded profusion 0/1 or greater of small opacities and/or any pleural change. -^Excluded from study; see text. and the 86 subjects with negative radiographic findings and with dust exposure between 10 and 100 mpcf.y were excluded, for reasons of economy. We had hoped to examine all those remaining, but failed in 39 cases, because of the subject's illness or refusal or non-attendance despite repeated promises. We were able to examine 135 men in Montreal and other urban centres in Quebec. The examination consisted of a questionnaire on smoking habits, a request to recall a complete work history, and a full size (14 in x 17 in (a: 0-35 ni x 0-43 m)) p-a chest radiograph. (Although not of concern in this report, the questionnaire also elicited information regarding respiratory symptoms, lungs were auscultated, and lung function was tested.) The questionnaire was that of the British Medical Research Council, adapted as in our previous studies of workers in the Quebec asbestos industry (McDonald et al., 1972). The responses were classified by two of us (FDKL and JCMcD), blind as to radiograph readings, according to current smoking habit as non-smoker, ex-smoker, smoker, or heavy smoker; the latter was defined as one who smoked 30 or more cigarettes a day or an equivalent amount of tobacco. The reported work histories were compared with those collected previously from personnel records, and in most cases were compatible if not identical. Discrepancies were investigated, and the man's own report led to a few clarifications of the recorded history. In s|x cases, however, doubts whether the man who had been examined was the same one whose radiograph had been read in 1967 were sufficient to warrant the subject's exclusion. The reported work histories also showed that 14 men had been exposed to asbestos or other fibrogenic dust since withdrawal from the industry; these, too, were omitted. After study of the intervals between actual withdrawal from the industry and the available withdrawal radiograph, we decided to exclude a further 29 subjects for whom that interval was greater than 12 months. This left 86 men who met all our criteria. For each of these 86 subjects, we calculated the following. Before the withdrawal radiograph Gross service: the period from start of first job in the industry to the date of the withdrawal " - radiograph. Net service: the total of periods of actual employment, within gross service. Total dust exposure: the total of annual dust exposures, summed ovei the period of gross service; the units were mpcf.y. Between available radiographs Interval: the period from the withdrawal radio graph to the radiograph in 1972. Gross service: the period from the withdrawal radiograph to actual withdrawal from the industry, by definition not more than 12 months. Total dust exposure during any service after the withdrawal radiograph. The paired films, with identifying information obliterated, were read side by side in known temporal order into the complete ILO U/C 1971 classification of radiographs of the pneumoconioses (International Labour Office, 1972). The more recent film was read first, and then the earlier film. The pairs of films had been assembled into six batches, which were presented to each reader in a different sequence. ILO standard films were available to all readers. Seven assessments of each film-pair were made by eight experienced readers: six worked quite inde pendently; the other two read together, yielding one assessment independent of the rest. The readers were Dr L. J. Bristol (USA)*; Dr P. H. Cartier (Canada)*f; Dr R. G. Fraser (Canada)t; Dr J. C. Gilson (UK)*; Dr T. R. Grainger (Canada)*!; Dr J. C. McDonald (Canada)*!; and,working together BQC2784 Radiological changes after withdrawalfrom asbestos exposure Table 2 Radiographic change between films Sit Parenchyma* Pleura Withdrawal radiograph Any reading 0b or 0/0 0/1 or above Any reading 0 for al) features Not all features 0 1972 radiograph No more than one step higher Two or more steps higher Two or more steps higher 0 The identical reading Not all features 0 At least one step higher on any graded feature, or appearance of an additional feature Maximum small opacities* ix. the higher reading of profusion of small rounded and small irregular opacities. 25 Evaluation No change Increase: attack Increase: progression No change Increase: attack Increase: progression Drs J. A. Dick and J. S. Washington (UK). The five 10-9 yr, making the age at radiography 41-0 yr. The marked with an asterisk (*) had taken part in our average gross service between this radiograph and previous studies (Rossiter et al., 1972) and in the actual withdrawal was 0-3 yr, with an imposed development of the UICC/Cincinnati (1970) and maximum of 12 months. The interval from the ILO U/C (1971) classifications (UICC Committee, withdrawal film until the radiograph in 1972 1970; International Labour Office, 1972); only four averaged 16-9 yr, so that the average age in 1972 was readers (t) were aware that the subjects had not been 57-9 yr. Total exposure up to the withdrawal radio occupationally exposed to dust in the interval graph ranged from zero to 3542 mpcf.y, with an between the taking of the two films. average of 236 mpcf.y. The average exposure after Every parenchymal and pleural assessment of a the first film before actual withdrawal was 3 mpcf.y, film-pair was classified as either `no change' or but ranged up to 47 mpcf.y. `increase', the latter subdivided (a) as parenchymal The distributions of average radiological scores on if>r pleural or both and (b) as `attack'or `progression', the withdrawal radiographs (see Table 4 for defini s indicated in Table 2.The very few recorded regres tions) were highly skew, the standard deviations sions were included as no change which, for reasons being at least as great as the mean scores. In other discussed later, also embraced assessments of a words, for a high proportion of the men the average single-step increase in profusion of small opacities. score, whether parenchymal or pleural, was close to In this study, increase, either parenchymal or pleural, zero, but several scores reached the maximum level, is taken as `agreed' if recorded in at least four of the unity. Despite this variation and the very wide varia- seven assessments; similarly, no change is considered tion in total exposure the correlations between (a) the to be agreed if there were at least five such assess radiological scores and (6) two measures of exposure ments. (gross service and the square-root transformation of The study population of 86 subjects is described total dust exposure) were all small, none being in Tables 3 and 4. There were rather more men from greater than 0-25. Table 3 Smoking habit and mining area Smoking habit Non-smoker Ex-sraoker Smoker Heavy smoker AH Town Asbestos (Quebec) 2 (4-2) 8 (16-7) 24 (50-0) U (29*2) 48 (100) Thetford Mines (Quebec) 7 (18 4) 9 (23-7) 17(44-7) S (13 -2) 38 (100) Percentages within each column in parentheses. the town of Asbestos, Quebec (56%) than from the area of Thetford Mines, Quebec (44%), and the former were heavier smokers. On average, the age at start of service was 30T yr, and the interval from then to the time of the withdrawal radiograph was Results The main findings are summarised in Table 5. At least four of the seven assessments were of parenchymal increase in six film-pairs (7-0%), of pleural increase in 17 pairs (19-8%), and of both parenchymal and pleural increase in two pairs (2-3%). Thus there was some agreement that there had been increase in the films of 25 men (29-1 %). In 34 film-pairs (39-5 %), there was agreed no change: all seven assessments were of no change in 14; six were of no change in 11; and five were of no change in nine pairs. Hie evidence was equivocal in the remaining 27 pairs (31-4%). Twenty of the men had experienced less than two years of gross service before the withdrawal radiograph (average service: eight months): none of their film-pairs showed , agreed parenchymal increase, but four pairs showed agreed pleural increase. BGC2785 6113 18140 26 M. R. Becklake, F. D. K. Liddell, J. Manfreda, and J. C. McDonald Table 4 Ages, lengths ofservice, exposure, radiological scores before withdrawal, and interval between radiographs Findings Minimum Maximum Mean SD Age at start of service (yr) Before withdrawal radiograph Gross service (yr) Net service Or) Total exposure (mpcf.y) At time of withdrawal radiograph Age (yr) Average radiological score parenchyma (a) pleura (6) Between radiographs Interval (yr) Gross service (jr)* Total exposure (mpcf.y> 149 0 0 0 30-4 0 0 11-3 -10 0 47-3 35-4 32-S 3542 54-9 I 1 231 H> 47 30-1 10-9 so 236 41-0 023 0-1$ 16-9 0-3 3 (S-S) (9-9) (7-6) (324) (33) (0-24) (0-15) (3-1) (0-4) (9) () If the profusion of both rounded and irregular small opacities was recorded as 0/~ or 0;Q, that assessment was scored 0. If either form of profusion was recorded 0/1 or higher, the score was taken as I. The arithmetic mean of the seven scores was taken as the average parenchymal score. () If aU pleural features were assessed as zero by any reader, a score 0 was allocated: the existence of any pleural change was scored I. The arithmetic mean of the seven scores was taken as the average pleural score. A very few of these averages were approximate because of a data processing error, which could have had little effect, and which it was not feasible to correct. A negative value means that the earlier film was taken after withdrawal. Table 5 Radiographic change Findings Cross service before withdrawal radiograph < 2 yr 2 >T T AU Number of subjects 20 Average length of gross service before withdrawal radiograph (>t) 0-7 Increase agreed (by at least 4 readers) Parenchyma! and pleural -- (0) Parenchymal only -- (0) Pleural only 4 (20 0) 66 14 2 (3-0) 6 (9-n 13 (19-7) 86 1! 2 (2-3) 6 (7-0) 17 (19-8) Total Equivocal `No change' agreed (by at least 5 readers) 4 (20 0) 6 (30-0) 10(50-0) 21 (31 8) 21 (31-8) 24 (36 4) 25 (29-1) 27(31-4) 34 (39 5) Grand total 20 (100) 66(100) 86 (100) Percentages within each column in parentheses. subjects), but no other association with smoking habit could be found. The men in whose film-pairs parenchymal increase was agreed, had started their serviceonaverage 10 yr earlier than the men in all the other groups (Table 6, *l)and,asaresult,hadgivenlonger gross service. Their total dust exposure had also been considerably greater on average than that in the other groups; although their net service was only about one-third longer, they had worked in higher dust concentra tions. About a third of the agreed parenchymal increases could be considered progression from a less than normal withdrawal radiograph (Table 6, *2); about 85% of pleural increases were in the nature of attacks (Table 2). There was no suggestion that pleural increase was related to any measure of dust exposure. In Table 6, four groups of men are compared: eight whose film-pairs showed agreed parenchymal increase with or without pleural increase; 17 showing agreed pleural increase only; 27 where assessments were equivocal, and 34 of agreed no change. Differences between these groups in age at the time of the withdrawal film, interval from that radio graph to the second in 1972, and service after the withdrawal radiograph were small and non-systematic. However, of the 38 men who had worked at Thetford Mines, 16 (or 42%) showed change compared with only nine out of 48 at Asbestos (19%) (Table 6, *3). Of the eight non-smokers among the 86 subjects, pleural increase was agreed in five (62-5%, compared with 22T % in all Discussion At face value, our results indicate that a substantial proportion (9*3 %) of men showed an increase in parenchymal small opacities after withdrawal from asbestos exposure. As the increases were confined to the more heavily exposed group and were not explained by differences in age and smoking habit, it suggests to us that they were attributable to their work. However, other factors, such as environmental pollution in the mining areas, cannot be ruled out because they were not examined. Two questions still follow: first, were the radiological increases the results of actual changes in the parenchyma or simply caused by differences in radiographic tech nique; second, were the increases more than would BQC27S6 Radiological changes after withdrawal from asbestos exposure Table 6 Radiographic change in relation to service, exposure, age and radiological scores Findings Number of subjects From Asbestos From Thctford Mines Proportion at Thetford Mines Non-smokers Age at start ofsen-ice (yr) Before withdrawal radiograph Gross service (yr) Net service (yr) Total exposure (ropcf.yr) At time of withdrawal radiograph Age (yr) Average radiological score Parenchymal: see (a). Table 4 Pleural; see (b)f Table 4 Between radiographs Interval (yr) Gross service (yr) Increase agreed by at least 4 readers Parenchymal Pleural s* 17 36 9 $ 11 62% 16 64% 2* 21-0 64% Cl) JS 314 1*3) 200 10-3 498 41-0 0-36 0-23 15 6 0-4 C2) 9-5 8-0 149 410 0-21 0-20 17-2 0-3 Equivocal 27 17 10 37% -2 31-9 10-t 7* -262 42-0 0-37 0-20 171 0*3 39 22 36% (*4) 27 So change agreed by at least 5 readers 22 12 % I 30-2 10-0 7-5 197 eo-2 0-11 0-08 16*9 0-4 * P value for difference between adjacent columns: (*1) 0003; (*2) 0015f; (*3) 0-016; (*4) 0000004. 0 Including 2 film pairs in which pleural change was also agreed. After allowing for iniercorrelations in contrast (*I). m e expected in the usual lag period between inhalation of dust and radiological appearance of disease? Were we planning this study again, we would certainly take more seriously the question of radio logical technique, although to have achieved full comparability over up to 25 years between radio graphs would not have been easy. Because of this problem, we adopted a conservative approach to the definition of parenchymal increase; namely, that there should be a difference of at least two steps on the 12-point scale of profusion. Moreover, as the readings were made side by side in specified temporal order, the readers were in a position to recognise differences in technique and to allow for them. It was evident from their comments that they did so, but we do not know how well. Side by side reading has the disadvantage, of course, that the reader may be influenced by what he believes about the relation between the films, and so may introduce precon ceptions into his reading. The readers varietf in their knowledge of the provenance of the films, but all were completely ignorant of the exposure history of each subject; furthermore, they assessed all 135 filmpairs of the men examined. For these various reasons, we are satisfied that the parenchymal increases observed are not attributable to technical differences or other bias. The second question is more difficult to answer. As we had only two films for each man we cannot distinguish a pattern of parenchymal change, con fined to the first few years after withdrawal, from one of continuous progression. It would be necessary to have at least two observations on each subject after withdrawal, the first preferably at about five years after the end of exposure. It has been suggested that a comparison of the present series with a group of workers, matched in age and work history but who had continued employment in the industry, might be informative. It is difficult to see how appropriate allowance could be made for the effects of subse quent dust exposure in the comparison group unless, of course, they fared no worse than those who had left. This study was supported by the'Medical Research Council of Canada and the Institute of Occupational and Environmental Health of the Quebec Asbestos Mining Association. We thank all the film readers for their co-operation. Much assistance was given us by Dr G. Utz in the field work and by several critics of earlier drafts of this report. References Gibbs. G. \V., and Lachance, M. (1972). Dust exposure in the chrysotile asbestos mines and mills of Quebec. Archives of Environmental Health. 24, 189-197. ECC-737 r 6113 18142 m 28 M. R. Becklake, F. D. K. Liddell, J. Manfreda, and J. C. McDonald Gilson, J. C. (1973). Progress in epidemiology. In Biological Effects of Asbestos, p. 8. Edited by P. Bogowslci, J. C. Gilson, V. Timbrell, and J. C. Wagner. IARC Scientific Publications, No. 8, Lyon. International Agency for Research on Canceq: Lyon. International Labour Office (1972). ILO V C International Classification of Radiographs of the Pneumoconioses, 1971. Occupational Safety and Health Series No. 22 (revised). ILO: Geneva. McDonald, J. C., and Liddell. F. D. K. (1978). Mortality in Canadian miners and millers exposed to chrysotile. Presented at Conference on The Scientific Basis for the Public Control of Environmental Health Hazards, New York Academy of Sciences, June 24th 1978. Annals of the New York Academy of Sciences (m press). McDonald, J. C., McDonald, A. D.. Gibbs, G. W., Siemiatycki, J., and Rossiter, C. E. (1971). Mortality in the chrysotile asbestos mines and mills of Quebec. Archives of Environmental Health, 22, 677-686. McDonald, J. C., Becklake, M. R., Fournier-Massey, G., and Rossiter, C. E. (1972). Respiratory symptoms in chrysotile asbestos mine and mill workers of Quebec. Archives of Environmental Health, 24, 358-363. McDonald, J. C., Becklake, M. R.. Gibbs, G. \V., McDonald. A. D., and Rossiter, C. E. (1974). The health of chrysotile mine and mill workers of Quebec. Archives of Environ mental Health, 28, 61-68. Rossiter, C. E., Bristol, L. J.. Cartier, P. H., Gilson. J. C., Grainger. T. R., Sluis-Cremer. J. K.., and McDonald. J. C. (1972). Radiographic changes in chrysotile asbestos mine and mill workers of Quebec. Archives of Environmental Health, 24, 388-400. UICC Committee (1970). UlCC/Cincinnati classification of the radiographic appearances of pneumoconioses. Chest, 58, 57-67. BGC2788 6113 18143 British Journal of Industrial Medicine February 1979 Vol. 36 No. 1 Contents Dependence of acoustic attenuation of hearing protectors on incident sound level a. m. martin page 1 The mortality of men in the Rhondda Fach, 1950-1970 a. l. cochrane, t. i. l. haley, r. moore, and d. hole page 15 Radiological changes after withdrawal from asbestos exposure Jm. r. becklake, f. d. k. liddell, j. manfreda, and j. c. mcdonald page 23 , _____ Pulmonary function in firefighters: acute changes in ventilatory capacity and their correlates a. w. musk, T. J. SMITH, J. M. PETERS, AND E. MCLAUGHLIN page 29 Histamine release from platelets for assay of byssinogenic substances in cotton mill dust and related materials S. K. AINSWORTH, R. E. NEUMAN, AND R. A. HARLEY page 35 Squamous cell skin cancer in the North-west of England, 1967-69, and its relation to occupation c. i. whitaker, W. R. LEE, AND J. E. DOWNES page 43 Occupational lead exposure in Denmark: screening'with the haematofluorometer P. grandjean page 52 Bromine in blood, EEG and transaminases in methyl bromide workers M. m. verberk, t. rooyakkers-beemster, M. DE VLIEGER, and A. G. M. VAN VLIET page 59 Dimethylformamide and alcohol intolerance w. h. lyle, t. w. m. spence, w. m. mckinnely, and k. duckers page 63 Book reviews page 67 Notices page 70 Information section page 71 ^ k ASTM CODEN: BJIMAG 36(1) 1-84(1979) ISSN 0007-1072 British Medical Association Tavistock Square London WC1 B0C27S9 t 6113 18144