Document OJqpoQNyG1on0R5w0yyN63e41

Epidemiology--Silica DUST EXPOSURE INDICES AT THE EARLIEST APPEARANCE OF PNEUMOCONIOSIS EDWARD MOORE, Ph.D* John R. Martin, M.D.* Alison C. Edwards, OSc * Henry A. Anderson, M.D.f * Edward A. MacLaughlin, M.B4 Division of Community Medicine and Behavioural Sciences and ^Discipline of Radiology, Health Sciences Centre, Memorial University of Newfoundland, St. John's, A1B 3V6, Canada fSection of Environmental and Chronic Disease Epidemiology, Wisconsin Division of Health, Madison WI 53701, USA INTRODUCTION In 1981 a number of workers with a radiologic appearance compatible with pneumoconiosis were identified in a cross sectional survey of an iron ore mine and beneficiation plant in Labrador West, designated as the Labrador West Dust Study (LWDS).1 In some, pneumoconiosis had been diagnosed in the years preceding the survey but in die case of others, die LWDS marked die first appearance of radiologic abnormality. Various indices for exposure to dust have been calculated for each of these workers at the point in time when pneumoconiosis was first suspected or diag nosed from their periodic surveillance chert radiographs. These indices were compared for their power to predict pneumoconiosis (ILO category ^ 1), using workers placed in ILO category 0 in the LWDS as a comparison group. One of die indices, cumulative quartz, has been used to assess the validity of ACGIH recommendations for exposure to silica.2 METHODS Case Selection One thousand nine hundred and fifty workers (1950) com pleted all phases ofthe LWDS. Forty-four (44) qualified for inclusion in category 1 or higher in die ILO classification (1980) of radiographs for pneumoconiosis by the method described elsewhere.1,3 Exclusions 1. For workers in all ILO categories. a. Exposure to rock dust at previous places of employ ment in excess of 25% of employment with present company. b. Location in administrative or "not dusty" jobs as de fined elsewhere.1 2. For workers scoring ^ ILO categoiy 1 LWDS. a. Presence ofconfounding factors in radiographic inter pretation. b. Absence of antecedent normal chest radiographs. The 36 workers remaining in ILO category ^ 1 comprised two groups. The first Group A--were workers where die LWDS confirmed a diagnosis of pneumoconiosis made for 672 die first time between 1972-79. The second Group B, were workers where the diagnosis was made for die first time in die LWDS (1981). In both groups periodic surveillance chest films had been read by a single observer without reference to die ILO classification. Consequently, in Group A the earliest appearance of pneumoconiosis has been defined as the first mention of radiographic changes compatible with or suspicious of this diagnosis. In Group B however, die first manifestation of changes suggestive of pneumoconiosis oc curred in the LWDS film which had been read to the ILO classification. For this group die earliest appearance of pneumoconiosis refers to a score of 1/0 or higher on the ILO scale. Exposure Indices Five exposure indices were created for each worker as described elsewhere.4,5 In brief, they were years of ex posure to detectable amounts of (hist and four quantitative dust exposure indices. The latter were functions of years of exposure and dust concentrations (based cm Harvard nylon cyclone measurements) at all die locations where a worker had been employed. The four measurements were cumulative and "peak" respirable dust and quartz indices. "Peak" in dices, were devised to take account ofperiods in a worker's dust concentration-time profile when dust levels were much higher than usual. These elevations would receive no special weighting in calculating cumulative indices. Hence these special "peak" parameters were computed to assess their importance. For each index ofexposure, the 1709 workers were analyzed and the observed prevalence values i.e. (number ofdiagnosed cases)/(number in workforce) were calculated and tabulated for a set of sub-intervals appropriate to each index of ex posure. Multivariate logistic regression analysis was per formed on all ofdie five indices to give predicted prevalence figures. RESULTS AND ANALYSIS The characteristics of die workers scoring ILO category 1 or higher in the LWDS, and who were included in this analysis, are listed in Table I by group. In die combined groups, when radiologic abnormality was first noted, the mean months of exposure at the present company ofemploy- Table I Worker Characteristics at Appearance of Radiologic Abnormality Months of Exposure Present Company Group A Group B Group A & B Months Since Previous Normal X-ray Group A Group B Group A & B Age Group A Group B Group A & B Max 224 211 224 26 96 96 52 61 61 Min 85 61 61 2 5 2 27 24 24 Mean 135 150 147 13 33 23 38 40 39 Epidemiology--Silica Median 135.5 156.0 138.5 12.0 15.5 12.5 35.5 37.0 36.0 Group A Group B Group A&B LWDS ILO Classification (1981) Subcategory (n) 1/0 (4); 1/1 (5); 1/2 (4); 2/1 (2); 2/2 (5); 2/3 (2); 3/3 (2) 1/0 (11); 1/1 (1) 1/0 (15); 1/1 (6) ; 1/2 (4); 2/1 (2); 2/2 (5); 2/3 (2); 3/3 (2) Shape Round Round Round n% 17 71 4 33 21 58 Group A: Diagnosed as pneumoconiosis prior to LWDS n = 24 Group B: Diagnosed as pneumoconiosis at LWDS n = 12 Group A & Bs n * 36 673 Epidemiology--Silica ment was 147 months with a maximum of 224 and a minimum of 61 months. The mean of the interval between the normal and the abnormal chest radiographs was 23 months with a maximum of96 and a minimum of 2 months. The explanation for this wide spread was that five workers, all in Group B, had moved from "dusty" jobs where surveillance films are customary to locations where they are not. At die time of the LWDS, the profusion of the small opacities ranged from 1/0 to 3/3 and the shape ofthe primary small opacities was described as round in 38% of the 36 diagnosed cases. A stepwise multivariate logistic regression analysis was per formed on foe five exposure indices. This gives foe logistic response function which is foe probability that an employee is in ILO category ; 1. The most important variable chosen by the regression analysis was "peak" quartz (PQ)--in feet, this was foe only variable chosen if a p-value of 3% is kept as foe criterion for foe `improvement chi-square'. This gave foe function: Probability (Category ^1) = expf-5.12 + 0.00343(PQ)1 1 + exp[--5.12 + 0.00343(PQ)1 The observed and predicted values for this function are shown in Table H. Iffoe 5% value is relaxed to 6%, then cumulative quartz (CQ) and years of exposure (YRS) are included as foe next variables in foe response Auction. However, further analysis does not allow foe inclusion of foe remaining two indices--cumulative and "peak" respirable dust--for an acceptable response function. This is in accord with our Case-Control Study where quartz was identified as foe dominant differentiating mineral index for exposure between category 0 and categories 1, 2 and 3.4 The rest of this present analysis concerns only foe three significant indices; viz, years of exposure, cumulative quartz and "peak" quartz. Figures 1-3 show foe distributions for foe healthy workers as opposed to foe "cases' for each of foe foree indices. These figures along with foe median values in Table III show that "peak" quartz is foe index which distinguishes best between foe distributions of foe `healthy' worker and foe `cases'. Values for the ratio: (number of cases)/(number of total workers) for sub-intervals of the force significant exposure indices are shown in Figures 4-6. The ratios rise with in creasing exposure, but for reasons which are not apparent, this rise was not sustained in years ofexposure after 12 years. DISCUSSION The workers scoring ILO category 1 or higher in the LWDS comprised two groups--one diagnosed prior to, and die other during the LWDS. The groups differed in the method used for determining foe earliest manifestation of pneumoconiosis; foe significance of this is being examined. Probability (Category ^1) = expr-4.87 + 0.003(PQ) + 0.18(CQ) - 0.11(YRS)1 1 + exp[--4.87 + 0.0Q3(PQ) + 0.18(CQ) - 0.11(YRS)] Table n Observed and Predicted Values for Logistic Response Function PQ Obs%: Pred%; 200 0.5 1.1 400 4.0 2.0 600 7.5 4.0 800 8.5 7.8 1000 20.0 14.3 1200 15.1 24.5 1400 38.1 38.8 PQ: "Peak " Quartz? (mg/m3)3.yr 674 Table IH Median Values for Exposure Indices for `Healthy* Workers and `Cases' Epidemiology--Silica Years of Exposure CQ PQ Healthy Workers n = 1673 5.9 1.8 34.2 Cases 11.4 7.6 912.0 n = 36 CQ: Cumulative Quartz; mg/m^yr PQ: "Peak" Quartz; (mg/m^)^.yr Figure 1. Relative frequency polygons for healthy workers and cases--years of exposure. Figure 2. Relative frequency polygons for healthy workers and cases--cumulative quartz. 675 Epidemiology-Silica Figure 3. Relative frequency polygons for healthy workers and cases--"Peak** quartz. (mg.nv3yr) "Ratio** Values - Cases________ (Cumulative Total Workers Quartz) Figure 5. Ratio values for cumulative quartz. 'Ratio" Values 'Cases * Total Workers (Yrs of Exposure) Figure 4. Ratio values for years of exposure. -Ratio" Values - *Cases * Total Workers ('Peak' Quartz) Figure 6. Ratio values for "Peak" quartz. 676 Dividing die workforce into two populations--those with pneumoconiosis (cases) and those without pneumoconiosis (healthy workers)--''peak' ' respirable quartz is the index which most clearly separates the two (Figure 3). This is substantiated by the logistic regression which singles out only this same index for prediction purposes at die 5% level. Thus the "peak" exposure index is a better predictor for pneumo coniosis than the cumulative exposure index evaluated on the same dust concentration time profile. This result is further confirmation of our previous observation that short lasting periods or "peaks" when dust levels are much higher than usual, contribute significantly to the development of pneumoconiosis.4 4 `Threshold limit values... represent conditions under which it is believed that nearly all (our emphasis) workers may be repeatedly exposed day after day without adverse effect."2 The ACGIH Threshold Limit Value--Time Weighted Average (TLV--TWA) for exposure to silica is 0.1 mg/m3. Assuming that a worker is exposed to this TLV-TWA for say forty years, about die maTimnm duration of employment that could be expected for such workers, then his cumulative quartz index would be 40 X 0.1 = 4 mg/m3 yr. In our study, for workers with cumulative quartz indices of Epidemiology--Silica 4 mg/m3 yr or less, die average number of years of ex posure was 7 years and as Figure 5 shows, die prevalence of pneumoconiosis was 6/1176 cases or 0.5 %. Thus, to date, in die conditions prevailing at this mining complex, 99.5% of those workers who have been exposed to silica concen trations resulting in a cumulative quartz index of ^4 mg/m3 yr, appear to have suffered no adverse respirable effects. REFERENCES 1. Martin, J.R., Muir, D.C.F., Moore, E., Edwards, A.C., et al.: Pneumoconiosis in Iron Ore Surface Mining in Labrador. J. Occp. Med. (in press). 2. TLV's Threshold Limit Values and Biological Exposure Indices for 1987-1988. American Conference ofGovernmental Industrial Hygienists (ACGIH). Cincinnati, Ohio (1987-88). 3. Guidelines forthe use ofILOInternational Classification ofRadiographs ofPneumoconioses. International Labour Office, Occup Saf Hlth Series No. 22 (Rev. 80); ILO. Geneva (1980). 4. Moore, E., Muir, D.C.F., Martin, J.R., Edwards, A.C.: A CaseControl Study to Investigate the Association Between Indices of Dust Exposure and the Development of Radiologic Pneumoconiosis. Arch. Environ. Hlth. 42:351-3 (1987). 5. Moore, E., Knight, G.: Interpretation of Historic Impinger Data. American Industrial Hygiene Conference. Detroit, MI, U.S.A., (May 1984). 677 Epidemiology--Silica SILICA DUST, RESPIRATORY DISEASE AND LUNG CANCER --RESULTS OF A PROSPECTIVE STUDY M. NEUBERGER M. Kundi A. Rutkowski W. Grundorfer Department of Preventive Medicine, Institute of Environmental Hygiene University of Vienna, A-1095 Vienna, Kinderspitalgasse IS, Austria INTRODUCTION Mineral particles are not generally considered to be car cinogenic, except for a few specific species such as asbestos, and die carcinogenicity of crystalline silica to humans has been a matter of controversy2*4 since results of retrospec tive studies and results on persons selected for silicosis have been questioned.6 Our investigation was initiated by the hypothesis that chronic irritation of the bronchial mucosa tty inspirable particles increases lung cancer incidence. Mean while other and more detailed pathogenetic mechanisms have been discussed in this context, considering respiratory clearance, carrier effects and other combined effects.2 Our contribution will be new results on die first cohort of dust exposed workers which has been followed prospectively from a preventive checkup in the 1950s, during die second part of life up to death (and in about 50% up to autopsy). Details of the source population,10 setting up of cohorts5 and first results of follow-up8*9 have been given earlier. METHODS A mobile team of the occupational health care unit started screening examinations in 1950 which included an occupa tional and a smoking history.10 All persons with a history of dust exposure were chest X-rayed and are the source of our exposed cohort. 1630 men given the first chest X-ray in 1950-1960 because of silica and/or heavy ``inert'* dust exposure were born before 1911 and resident in Vienna. An equal number of Viennese men without occupational dust ex posure was selected from die occupational health care ex amination files by matching year of birth, year of first ex amination and smoking.5 At a later control of exposure histories at die pension insurance board 5 men were found not to meet the eligibility criteria and were therefore ftliminatftH from the study together with their matched counterparts (dustwoikers: 1 because ofconfounding asbestos exposure, 1 because of lack of dust exposure, 1 because of being bom 1911; references: 2 because of dust exposure). Thus we followed 1925 workers exposed to non fibrous dust in the metal, glass, ceramics, stone, construction, cleaning agent and a few other industries and 1625 non-dust-exposed workers from a great variety of branches (publishing and printing, chemical, construction, textile, leather, food, elec trical industry, etc.). Diagnoses at death were traced and encoded without knowledgement ofdie exposure. Death certificates in Vienna 678 are based on autopsy in more than 50% ,3 but a small pro portion of diagnoses had to be clarified with the reporting hospital or physician. The best available informations were used for comparison on dust and non-dust-exposed. The of ficial diagnosis was used for comparison with the general population of Vienna, which gives conservative estimates of standard mortality ratios (SMR), because mortality (especial ly cancer mortality) is higher in Vienna than in Austria, and some of the cohort members moved to rural districts alter retirement. RESULTS 1621 dust-exposed (DUST) and 1621 reference subjects (NO DUST)=99,8% could be traced up to the end of 1986.1442 DUST (89%) and 1384 (NO DUST) (85,4%) died. The underlying cause of death is given in Table I for 1439 (99,8%) DUST and 1379 (99,6%) NO DUST. Observed cases (0) in DUST divided by expetted cases in NO DUST are given under Relative Risk (RR). This age-adjusted relative risk of DUST was significantly raised (p <0,001, Poisson, two-sided) for lung cancer, stomach cancer, chronic obstructive lung disease (emphysema, bronchitis, asthma) and silicosis/fibrosis/tuberculosis ofthe lung. Gastrointestinal diseases (mainly liver cirrhosis) were found more frequent, too (p <0,05). Respiratory diseases were also found increased in OUST as secondary cause on death and additional diagnoses on death certificates (mainly silicosis and chronic obstuctive lung disease, but also pneumonias). Table II shows official diagnoses ofunderlying cause ofdeath and standard mortality ratios. In DUST overall mortality, lung cancer, stomach cancer, chronic obstructive lung disease, silicosis/fibrosis/tuberculosis and acute/infectious respiratory diseases was found higher (p <0,001, Poisson, two-sided) than expected in Viennese men of same age. In NO DUST overall mortality was the same as in the general population; silicosis/fibrosis/tuberculosis, cardiovascular diseases and accidents were lower than expected and lung cancer was slightly higher (p <0,05). For 775 foundry workers, 475 grinders and other metal workers, 191 glass and ceramic workers, 87 stone cutters and construction workers and 65 other dust-exposed workers observed lung cancels and stomach cancers are given in Table m (32 workers could not be included in this stratification because the main branch of occupational dust exposure was Table 1 Main Causes of Death: Dust Exposed Compared to Reference Cohort Epidemiology--Silica Best available diagnose DUST 0 NO DUST 0 RR cancer of lung n other respirat.o. n stomach H intestine H other digestive o. other sites It not localised chron. obst. lung dis. tuberculosis, silicosis acut. infect, respir.d. heart disease cerebrovascular dis. other vascular dis. gastrointestinal dis. accidents, suicide other disease unknown, ill-defined 183 4 80 37 50 67 1 83 78 50 362 125 132 93 41 55 3 142 4 48 50 45 84 2 47 15 55 404 157 145 84 39 58 5 146*** 114 190*** 88 125 92 202*** 513*** 110 106 97 107 127* 109 112 111 * p < 0,05, ** p<0,01, *** p<0,001 ill-defined on entry file). Table IQ shows increased lung cancer mortality in all subgroups of DUST with the highest SMR in stone and glass/ceramics (p <0, 001). The SMR for stomach cancer was above 100 in all strata, but significantly only in those employed in foundries and other metal industries. Life table analyses showed reduced survival of DUST not before 7 years of observation and 58 years of age (Figure 1). At the median the survival difference between DUST and NO DUST was 3 years. Survival probability from lung cancer at age 70 was 91,4% in DUST (95% confidence in terval: 89,9-92,9) and 71,7% in NO DUST (95% confidence interval: 93,2-95,6) which is significant (Figure 2). DISCUSSION In Viennese workers we found a relation between exposure to respirable particulates and mortality from lung cancer and chronic diseases ofthe lung. The rate ofchronic obstructive lung deseases was 2-times higher and the rate of lung cancer was 1,5-times higher than in non-dust-exposed workers with comparable smoking habits. The 17% lung cancer increase in 679 Epidemiology--Silica Table n Main Causes of Death: Dust Exposed and Reference Cohort Compared to General Population Official diagnose (ICD, 9.Rev.) DUST 0 SMR NO DUST 0 SMR *** 0to0 --1 1 CANCER OF LUNGS (162) 180 * OTHER RESPIRATORY 0.(161, ETC.) 4 * STOMACH (151) 78 ' INTESTINE (152-154) 36 ' OTHER DIGESTIVE (140-150, 155-159) 48 " OTHER 70 CHRONOBSTRLUNG DIS. (490-493) TUBERCULOSIS, SILICOSIS, FIBROSIS (Oil, 502, etc.) ACUTE, INFECT.RE SPI RAT.D.(460-487) 77 75 52 CARDIOVASCULAR D. (390-459) GASTROINTESTINAL D. (520-579) ACCIDENTS, SUICIDE (800-999) 628 95 42 ALL CAUSES 1442 + p^OS, ** p^0,01, *** p<0,G01 67 167*** 84 115 92 205*** 299*** 135* 97 109 86 123*** 142 117* 4 60 47 89 50 100 44 92 85 95 45 101 17 63* 55 120 707 92* 85 87 42 76* 1384 100 cancer DEATHS mTable Lung and Stomach Cancer in Dust Exposed Industries Compared to General Population FOUNDRY 0 SMR METAL 0 SMR iGLASS/CERAMICS 0 SMR STONE 0 SMR OTHER 0 SMR LUNG stomach 85 163*** 135* 28 236*** 15 293*** 6 153 40 176*** 21 149* 6 116 4 173 3 172 * p<0.05, *# p<0.01, *** ?< 0.001 680 non-dust-exposed workers compared to the general popula tion could be due to smoking and occupational exposures (vapours etc.) except dust. In dust-exposed die increased lung cancer rate was found in all subgroups, which have very dif ferent confounding exposures (e.g. polycyclic aromatic hydrocarbons in foundries, possibly arsenic in glass industry, nitrosamines in some metal grinders, etc.). We discussed elsewhere8*9 the minor importance of confounding ex posures for our study, such as die very limited use of asbestos in Vienna and its metal industry (no signs of asbestos ex posure were found on chest X-rays of dust-exposed and no mesotheliomas occurred), the lack of sources of radiation in the plants investigated (only in a few stone- and tunnelworkers radiation might have been a co-factor) and die negligible use ofcarcinogenic metals and compounds in iron foundries at the time of the cohort recruitment. Confound ing exposures after registration seem to be of minor impor tance, too, because work histories obtained in 1982 from the national pension insurance board showed a low inter industry and interplant mobility of dust-exposed (because of registration after age 40) and no second jobs with exposure to accepted carcinogens (one lung cancer case was exposed to glass wool as an insulator and a few other cases with suspected confounding exposure such as a sepiolite grinder are still alive or died from nonmalignant diseases). Most earlier studies on lung cancer in dust workers have been biased by selection and competing causes of death (silicosis). In coal workers a protective effect of coal dust has been discussed.7 The seemingly conflicting results on lung cancer in silica and "inert" dust-exposures could perhaps be ex plained by a multistage model of carcinogenesis: Genotoxic substances (from tobacco smoke as well as from foundry air or other sources) could function as initiators and particles as adsorbens and promotors. In this case it would not be sur prising if lung cancer increases with total dust load and not necessarily e.g. with PAH-concentrations, even though both might be involved. From available measurements in our study8 we cannot decide whether silica or total respirable or inspirable dust was responsible for die increase of lung cancer, but we suspect that also "inert" dust exposure (if heavy and long lasting) can promote chronic obstuctive lung diseases and lung cancer. Bombardment of airways by dust particles, even "inert" ones, could eventually sustain high levels of polymorphonuclear recruitment, alveolar macrophage activation and finally result in a disturbed balance of protease and antiprotease activity.1 From die age in which we observed excess lung cancer deaths (Figure 2) we conclude that screening examinations should be continued after retirement from a job with heavy dust ex posure. Screening for chronic obstructive lung disease must begin much earlier. Most important, however, is primary prevention by reduction of dust exposures, even "inert" ones. SUMMARY From 1625 men examined in 1950-1960 with a history of occupational exposure to silica and other non-fibrous par ticulates in foundries, other metal, glass, ceramic, brick, stone and some other dusty industries 1621 were traced up to 1986, 1442 died, 183 from lung cancer (SMR 168), 80 Epidemiology--Silica JM5 9usfvork*r SU -- " -- -1426 CwtrtUf SU 2.2 3.6 66.3 $6.9- 66.7 64.1 61.6- 46.4 ( 76y> 96% K! Figure 1. Survival ofdust-exposed (solid line) and reference subjects (dashed line). Figure 2. Survival from lung cancer (dust-exposed: solid line, reference subjects: dashed line). from stomach cancer (SMR 167), 83 from chronic obstruc tive lung disease (SMR 205) and 78 from silicosis/fibrosis /tuberculosis (SMR 299). From 1625 non-dust-exposed workers from the same source population, matched in 1950-1960 to dust exposed workers for age, smoking and begin ofobservation, 1621 were traced, 1384 died, 142 from lung cancer (SMR 117), 48 from stomach cancer (SMR 89), 47 from chronic obstuctive lung disease (SMR 101) and 15 from tuberculosis/fibrosis (SMR 681 Epidemiology--Silica 63). Life table estimates show that primary and secondary prevention should not be focused on pneumoconioses alone. REFERENCES 1. Becklake, M.R.: Chronic airflow limitation: its relationship to work in dusty occupations. Chest 88, 4:608-617 (1985). 2. Cburg, A., Wiggs, B.: Mineral particles, mineral fibers and lung cancer. Environ. Res. 37:364-372 (1985). .. 3. Feigl, W., Friedl, H.P.: Daten aus dcr Osterreichischcn gesun- dheitsstatistik ond ihre Verwertbarkeit fir epidemiologische Unter- suchungen. OsterrArztezeitung 34:1313-1316 (1979). 4. Goldsmith, D.F. etal.: Silica, Silicosis snd Cancer. Cancer research monographs, Philadelphia, Praeger (1986). 5. Griindorfer, W., Popper, L.: Dust exposure and incidence ofbronchial carcinoma. (German) International Congress ofOccupational Health, Vienna, Proreedings, 3:173-176 (1966). 6. Kennaway, E.L., Kennaway, N.M.: The incidence of cancer of the hmg in coal miners in England and Wales. British Journal ofCancer, 7:10-18 (1953). 7. Miller, B.G., Jacobsen, M.: Dust exposure, pneumoconiosis and mor tality of coalminers. Brit. J. Iod. Med. 42:723-733 (1985). 8. Neuberger, M. et al.: Cancer mentality of dust workers and controlsresuhs of a prospective study. In: Prevention ofOccupational Cancer, Occupational SafetyandHealth Series 46:235-241, International Labour Office, Geneva, 1982. 9. Neuberger, M. et al.: Long term effect ofoccupational dust exposure. Jap. J. Industrial Health, 30, 5 (1988). lO.Popper, L., Tuchmann, E.: Occupational health care ofthe Viennese health insurance for laborers and employees. (German) International Congress oo Occupations/ Health, Vienna, Proceedings, Suppl. 2 (1966). ACKNOWLEDGEMENT: This work was supported by die Medical Science Research Fund of the Mayor of the City of Vienna. 682 Epidemiology--Silica EPIDEMIOLOGIA DE LA SILICO-TUBERCCILOSIS EN MINEROS ASTURIANOS: TASA DE NUEVOS CASOS BACTERIOLOGICAMENTE POSITIVOS. PERIODO 1971-1985. J. A. MOSQUERA Jefe del Servicio de Neumologia del INSTITUTO NACIONAL DE SIUCOSIS (INS). Oviedo. SPAIN. ABSTRACT La poblacion diana la forman S0.470 7.454 mineros--del carbon de Asturias. La muestra estudiada prospecdvamente abarca a 3.612.2 1.779 primeras visitas que son estudiados en el Instituto Nacional de Silicosis cada ano. A todo sospechoso de tuberculosis se le practica sistematicamente: 4 extensiones de esputo (Metodo Ziehl-Nielsen) y 3 cultivos (Medio de Lowenstein) y si precisan, otros tests diagnosticos mas invasivos. Durante el periodo 1971-1985, 1.136 pacientes fueron diagnosticados de tuberculosis por presentar positivos los tests bacteriologicos. Durante los 7 primeros anos de observation, entre los 508 tuberculosos, la distribution de la categoria de neumoconiosis (ILO 1971) y edades era 7.8% tenian Categoria 0 y 51.1 13.6 anos; 9.8% con Categorias 1,2,3 y 54.3 13.3 anos; 82.3% con Fibrosis Masiva Progesiva y 53.5 10.6 anos. Entre 1978-85, la distribucion de neumoconiosis y edades eran respectivamente: 33.3% y 48.7 14.7 anos; 14.1% y 55.5 10.5 anos; 62.6% y 58.4 10.2 anos. La tasa de nuevos casos bacteriologicamente positovos file' de 150.1 30.9 por 10s mineros y ano, no modificahdose significativamente a lo largo de los 15 anos, a pesar de la eficacia de los tratamientos. Esta tasa es 3 veces superior a la de la poblacion no minera de la zona, pero no es diferente a la encontrada en otras poblaciones ocupacionales similares. INTRODCICCION La Tuberculosis Pulmonar es responsable en en nuestra region de la evolution de las categorias simples de neumoconiosis (1,2,3) a Fibrosis Masiva Progresiva (FMP) en el 9.7% de los casos.1 Por ello el control de la Silicotuberculosis reduciria la morbilidad y mortalidad que condiciona la FMP. METODOS El periodo de estudio comprende desde el 1 de Enero de 1971 hasta el 31 de Diciembre de 1985, dividido en dos periodos 1971-1977 y 1978-1985. La poblacion diana, objeto de este estudio, es la poblacion minera de Asturias. Esta region, localizada en el Norte de Espana tiene yacimientos de hulla y antracita. La cuantia media anual, con una desviacion standard ( 1SD), de esta poblacion se compone, a lo largo de estos anos de 50.470 7.454 mineros. La poblacion muestreada anualmente (1SD) asciende a 3.612.2 1.779.1 mineros. Los diagnosticos de neumoconiosis se realizaba en base a exposition laboral y presencia de opacidades redondas de profusion igual o mayor a P 1/1 (Gasification ILO 1970).2 La lectura radiologica se efectuo'por 3 lectores con experiencia en neumoconiosis, usando la escala de profusiones ampliada.3 Las discrepancias de lecturas eran promediadas. El diagnostico de tuberculosis pleuro-pulmonar se bacia al observar cuadro clinico-radiologico compatible y presencia de material alcohol-acido resistente con el metodo de ziehl Nielsen y/o identification del Mycobacterium Tuberculosis en medio de Lowenstein-Jensen. A todo sospechoso de tuber culosis se le practicaba sistematicamente 4 extensiones de esputo con el metodo Ziehl y 3 cultivos sucesivos en medio de Lowenstein; si era preciso, el paciente era sometido a otros metodos diagnosticos, (broncoscopia, puncion transpulmonar, etc.). A efectos epidemiologicos eran excluidos los casos en que el diagnostico se sustentaba solo por metodos histologicos. Las resistencias "in vitro" del Mycobacterium Tuberculosis eran estudiadas mediante el metodo de Canetti, modificado.45 Todos los pacientes tuberculosos eran seguidos por una seccion especifica de Neumologia del INS, que los trataba y controlaba su evolucion. Durante el periodo 1971-1977, el tratamiento antituberculoso era el "standard" de 18 meses, con seguimiento, clinico, radiologico y bacteriologico cada 3 meses hasta 2 anos despues de finalizado el tratamiento Desde 1978 a 1985 los tratamientos usados fueron 2 regimenes de curso corto; a) Hidracida (I) 300 mg./dia, Rifampicina (R) 600 mg./dia, Etambutol (E), 1200 mg/dia, durante 9 mg./dia, Rifampicina 600 mg./dia y Pirazinamida (Z) 1500 mg./dia, durante 2 meses, seguido de Hidracida y Etambutol durante 7 meses mas. (SIZR)2, (IE)7. (6). Se practico' el mismo tipo de seguimiento pero a intervalos de 683 Epidemiology--Silica 1 roes. Tambien durante este periodo se hizo chequeo de contactos intimos. En caso de retratamientos en enfermos tuberculoses, se usaba el regimen standard de 18-24 meses, intentando que la com bination de drogas utiles, conocidas por antibiograma, sumase 1.5. Esto se hacia adjudicando valor 1 a I,R y 0.5 a Z y aminoglicosidos.7 La cuantificacion de la enfermedad tuberculosa, se hacia midiendo la tasa anual de los nuevos casos bacteriologicamente posidvos. Esta se obtenia refiriendo el numero de nuevos enfermos, hallados por cultivo y/o estensio'n, a la poblacion diana que existia a 31 de Diciembre del ano correspondiente. RESULTADOS 1.136 pacientes fueron diagnosticados de tuberculosis pulmonar despues de excluir 2 casos en que crecio' una Mycobacteria no tuberculosa. Durante 1971-77, entre los 507 tuberculosos la distribution de las categorias de neumoconiosis y edades eran: 7.8% tenian categoria 0 y 51.1 13.6 anos; 9.8% con categorias 1,2 y 3 y 54.3 13.3 anos; 82.3% con FMPy 53.5 10.6 anos. Entre 1978-85, la distribucion de neumoconiosis y edades eran respectivamente de 33.3% y 48.7 14.7 anos para la categoria 0, 14.1% y 55.5 10.5 anos para las neumoconiosis simples y 62.6% y 58.4 10.2 anos para la FMP. La tasa de resistencias primarias a las drogas anrituberculosa fueron para los mineros sin neumoconiosis (categoria 0) del 12.9% y 11.9% para las categorias 1,2 y 3. Las cifras de curation para los pacientes que completan los tratamientos, al fmalizar los mismos eran: a) Tratamiento standard 121/123 (90.9%). b) Tratamiento curso cento (E1RZ)2 (IE)7 26/27 (96.2%). c) Tratamiento curso corto (IRE)9 39/40 (97.5%). d) Retratamiento con "standard** en pacientes sin resisten cias: 48/49 (97.9%). e) Retratamiento con "standard** en pacientes can resisten cias a 1 droga: 25/28 (89.2%). f) Retratamiento con "standard" en pacientes con resisten cias a 2 drogas 14/17 (82.3%). g) Retratamientos con "standard** en pacientes con resistencias a multiples drogas 9/10 (90.0%). DISCUSION La tasa de neuvos casos bacteriologicamente posidvos se ha mantenido estable a lo largo del periodo de observation, con un valor medio ( 1SD) de 150.1 30.9 por 10s mineros y ano. Esta tifra triplica a la de la poblacion no minera de la zona. Las tasas registradas en otras poblationes mineras ban variado desde cifras de 49.6 x 105 y ano8 hasta las encontradas por Popovac et al9 que alcanzan valores de 8000 x 10s y ano. Nuestros resultados son casi coincidentes con la tifra de 146 x 105 y ano encontrada por Lander en Dinamarca.10 La persistencia de la tuberculosis ocurre a pesar de la alta eficacia de los tratamientos y la no excesiva alta prevalencia de resistencias primarias. BIBLIOGRAF1A 1. Munoz J., Sala J., Mendez Lanza A., Cabezudo M.. Cairetero } Mosquera J: Neumoconiosis complicada. Arch dc Broocoocumol, 14:175-178 (1978). 2. ILO U/C 1971 International Classification of Radiographs of the Pneumoconioses. Medical Radiograph and Photograph. 48:67-110 (1972). 3. Liddell FDK: Assessment of radiological progression of simple pneumoconiosis in individual miners. BrJ ofIndust Med. 31:185-195 (1974). 4. Vestal AL.: Procedures for the isolation and identification of Mycobacteria. USDepartment ofHealth, Education and Welfare Publ. oo(cdc) 77-82-30. Center for Disease Control. Atlanta: pp 41-115 (1975). 5. Moreno A., Cimadevilla R., Yanez B., Leiva P., Fleitas A., Santos Ml. Libro del I Congreso de la Sociedad Espahola de Enfemxdades Infecciosas y Microbiologia Clinics, pp 122-122. (1984). 6.4^ Encontros. Jornada Galako-Duriense. Sociedade Portuguesa de Patologia Respiratoria. Colmeiro A., Munoz J.: Efficacia de los regimenes tuberculosos de curso corto en los pacientes cod la asociacioh Tuberculosis Pulmonary Neumoconiosis con Fibrosis Masiva Progresiva. Braga. Portugal. (1986). 7. Lester W.: Treatment ofdmg-resistence tuberculosis. UM;XVH 3-41 (1971). 8. Prowse K., Cavanagh P.: Tuberculosis in the Potteries. LANCET 2:357-359 (1976). 9. Popovac D., HaThin M.: Silikotuberfculoze n Sadara Trapcinih Rndnika. Flucne Bol Toberk. 29:17-21 (1977). 10. Lander F., Sherson DL.: Incidensen af Silikotubertulose--bland stdberiarbefdese. Ugeskr. Laeger. 149:180-182 (1987). 684 Epidemiology--Silica EPIDEMIOLOGICAL STUDY OF SILICOSIS IN HARDROCK MINERS IN ONTARIO EX C. F. MUIR A. Sebestyen J. Julian C. Bemholz D. K. Verma H. Shannon McMaster University, Hamilton, Canada ABSTRACT An epidemiological survey of silicosis in hardrock miners in Ontario was undertaken. Radiographs had been taken annually, and these were classified by 5 readers for silicosis. The point in time at which miner's radiographs were considered to have passed the 1/1 category for small round opacities was taken as identi fying the onset of silicosis. A side-by-side comparison of konimeter and gravimetric dust sampling was carried out in both goldand uranium mines, the resulting relationship being used to convert historical konimeter data from the mines to equivalent respirable quartz concentrations. By using work records, a cumulative respirable quartz exposure index was calculated for each miner. A Kaplan-Meir survival curve analysis was used to derive risk estimates for the development of silicosis in relation to cumulative respirable quartz exposure. No Paper provided. 685 Epidemiology--Silica RADIOGRAPHIC ABNORMALITIES IN VERMONT GRANITE WORKERS EXPOSED TO LOW LEVELS OF QUARTZ WILLIAM G. B GRAHAM Sheila Weaver Taka Ashikaga David Hemenway Robert O'Grady Depts. of Medicine, Biostatistics, and Engineering, University of Vermont, and the Vermont State Health Department. Burlington, VT, USA ABSTRACT Whether exposure to levels of granite dust below the current OSHA limits leads to radiographic abnor malities after a lifetime of exposure has not been settled, hi 1953, we carried out an X-ray survey of the Vermont granite industry. Quarry and stone shed workers who had been exposed to the low dust levels prevailing in the Industry since 1940 were offered chest X-rays. Films were read by three "B" readers, and were considered abnormal if 2 or 3 of the readers assigned a profusion score of 1/0 or greater. 976 workers out of a total of approximately 1400 participated. 65 (6.9%) of the films were judged abnormal, but the profusion scores were low, only 4 ofthe films being assigned scores of6/1 or greater. Only 7 (0.7%) had "pqr" or rounded opacities as the major abnormality. The remaining 61 had irregular opacities, large ly lower lobe in location, which are of doubtful significance, but may be related to smoking and aging. In addition, total dust concentrations were measured using personal samplers; dust levels were similar to previous measurements. Mean concentrations were 601 365 micrograms/cuhic meter. Using previously estimated values for percentage quartz of 10%, the mean quartz concentration was below the current OSHA standard of 100 micrograms/cubic meter, although 11 % of the samples were above this value. If exposure levels have remained approximately the same over the past 45 years, we conclude that current dust con trols, which conform to OSHA standards, have essentially eliminated silicosis. BACKGROUND Studies of workers' health in the Vermont granite industry have provided a great deal of information on die health ef fects of quartz dust inhalation since 1969, when die first com prehensive study of the industry was published.4 Based on a high incidence ofsihcotuberculosis, control measures which reduced granite dust levels below 10 million parts per cubic foot (mppcf) were accomplished between 1935 and 1940. At that time, the Vermont Division of Industrial Hygiene (DIH) began annual radiographic surveys to assess the ef fect of dust reduction on the prevalence of silicosis. Subse quent studies showed that as workers with established silicosis retired or left the industry because of illness, and as new workers were hired, the percentage of the work force with radiographic abnormalities declined.1*3 In 1964, it was stated that no new cases of silicosis had been detected in workers exposed only to the lower dust levels prevailing after 1940. hi 1974, a study analyzing the results ofthe DIH radiographic survey of 1970-1 suggested that radiographic abnormalities were present in approximately 30% of the workforce.7 Most ofthese were oflow grades ofprofusion and comprised both irregular and rounded shadows. However, 67 films (5% of the total) were assigned profusion scores of 6 or 3, which was believed to be consistent with definite silicosis. Three possible criticisms might be made ofthis study. First, die films were interpreted only by a single non-certified reader. Second, workers who were exposed only to the low dust levels prevailing after 1940 were not analyzed as a separate group. It was therefore not possible to know whether levels less than 10 mppcf had eliminated silicosis, or were, on the contrary responsible for some of the radiographic ab normalities. Finally, a certain percentage of the films were interpreted as showing opacities in workers with zero dust years of exposure. The present study attempts to clarify the issue whether the dust levels present in die industry since 1940 have caused radiographic abnormalities, and if so, of what type and ex tent. We present die results of an industry-wide radiographic survey done in 1953, which includes only workers exposed to granite dust after the institution of dust controls in 1935. Results of dust sampling for total respirable dust will also be mentioned briefly. METHODS All workers employed in 1953, including quarry and stone shed workers, were offered 14 x 17 chest radiographs taken in a mobile van provided by the Appalachian Laboratory for Occupational Safety and Health (ALOSH). Work histories were recorded or updated on all participants, including oc 686 cupational category, the shed where employed, duration of employment and smoking histories. Forced spirometries were also carried out. The chest radiographs were interpreted by three "B" readers using the ILO format (1980). The defini tion of an abnormal film was that either 2 or 3 ofthe readers assigned a profusion scores of 1/0 or greater, either of the rounded (pqr) or irregular (stu) type. One worker with definite silicosis was excluded because his major work ex perience occurred in Canada. Gravimetric dust sampling of respirable size particles were collected using personal breathing zone samplers at flow rates of 6 liters/minute. RESULTS 972 workers out ofa total work force of approximately 1400 were x-rayed. Ofthose workers not having X-rays, 102 were absent on the day of the survey and the remainder (326 or 23% of the work-force) refused. Only 28 (2.88%) of those x-rayed were interpreted by 2 or 3 of die three readers as showing abnormalities consistent with pneumoconiosis. In only 7 films did all 3 readers agree that an abnormality was present. 21 of the 28 films judged abnormal had as the primary abnormality an irregular or stu type of opacity; this was true of the secondary readings as well. Only 7 (0.7% of die total cohort) showed rounded opacities (pqr) of the type typically seen in early silicosis. The grades of profu sion were extremely low, only four films being judged as have profusion scores of2/1 or higher and these films showed irregular (stu) types of opacities. The location of the stu changes, even at low grades of profusion, tended to be in the lower lung zones. No large opacities or egg shell calcifications were observed. Comparison of die workers with abnormal radiographs com pared to the remainder of die workforce showed that the former were older on average (53.6 yrs vs. 41.5 yrs.), and spent longer in granite (30.9 yrs vs. 17.5 yrs.), had smoked longer (27.9 yrs. vs. 19.2 yrs.), and more heavily (32.3 yrs. vs 26.99 yrs.) P values for die first three variables show significant differences at the .001 level. Several of the workers with abnormal films had notably low exposures to dust: one was a lumper for 6 years, and for the remainder of his 37 years had been a draftsman with little exposure. Another had worked exclusively in an office as a draftsman without any exposure. Average dust concentrations of417 samples was 601 368 micrograms/cubic meter, which is similar to values observed by previous workers.5 Epidemiology--Silica DISCUSSION These results indicate that radiographic abnormalities con sistent with silicosis have occurred at a very low level (0.7% of cohort) at quartz exposures which are in conformance with the current OSHA limit of 100 micrograms/cubic meter. ,This is particularly surprising inasmuch as approximately 10% of the samples, using die 10% quartz value of previous workers, were over 100 micrograms/cubic meter. Further, the changes observed tend to be a very low grades of profu sion. The predominant type of opacities judged to be pres ent were of the stu or irregular type, seen in 21 of the 28 workers judged to have abnormal films. The significance of these changes is uncertain, but may be associated with peribronchial fibrosis associated with pathological changes of chronic bronchitis which has been described in older, heavily smoking patients.6 In summary, this study provides an overview of the prevalence and type ofradiographic abnormalities which have developed in Vermont granite workers over a 45 year period of observation, when dust controls were effectively main tained. This essential elimination of radiographic silicosis has occurred at quartz levels which are in conformance with cur rent OSHA standard. REFERENCES 1. Ashe, H.B., Bergstrom D. E.: Twenty-six Years Experience with Dust Control in the Vermont Granite Industry. Ind. Med. Surg. 33:2.73-78 (1964). 2. Eisen, E.A., Smith, T.J., Wegman, D.H., Louis T.A., Froines, J.: Estimation of Long Tom Dust Exposures in the Vermont Granite Sheds. Am. Iod. Hyg. AssocJ. 45 (2):89-94 (1984). 3. Hosey, A.D., Ashe, H.B., Trasko, V.M.: Control ofSilicosis in Ver mont GraniteIndustry. Washington D. C.: U.Su. Department ofHealth, Education and Welfare. (Public Health Service Publication No. 557) 1957. 4. Russell, A. E., Britten, R.H., Thompson, L.R., Bloomfield, J.J.: The Health ofWorkers in Dusty Trades. Exposure ofsiliceous dust (granite industry). Washington, D.C. (Public Health Bulletin No. 157). USGPO. 5. Eisen, E.A., Smith T.J., Wegman, D.H., Louis, T.S., Froines, J.: Estimation ofLong term Dust Exposures in the Vermont Granite Sheds. Am. Ind. Hyg. Assoc. J. 45: 89-94 (1984). 6. Weiss, W.: Cigarette Smoke, Asbestos, and Small Irregular Opacities. Am. Rev. Respir. Dis. 130:293-301 (1984). 7. Theriault, G. P., Peters, J.M., Johnson, W.M.: Pulmonary Function and Roentgenographic Changes in Granite Dust Exposure. Arch. En viron. Health. 28:23-27 (1974). Supported by National Institutes ofOccupational Safety and Health (5 ROl 0H0135-04) 687 Epidemiology--Silica A STUDY OF SILICOTIC CHINESE GRANITE QUARRY WORKERS IN SINGAPORE W. H. PHOON, MBBS, M Sc (Occ. Med.), DIH, MFOM, AM S. E. Chia, MBBS, M Sc (Occ. Med.) Department of Industrial Health Ministry of Labour Republic of Singapore SYNOPSIS Although silicosis is less prevalent today than in the past owing to better dust control at workplaces, cases still occur in a number of countries. In Singapore, the greatest hazard of silicosis is in die granite quarries. In 93 Chinese granite quarry workers diagnosed as having silicosis and followed up, there was an increased prevalence of chest X-rays with large opacities for category 3 profusion compared to category 1 although this was not statistically significant. The duration of exposure to silica dust did not seem to affect die extent of opacities in the chest X-rays as workers with more than 20 years exposure in the quarries did not have a higher category of profusion of opacities on chest X-ray than those with shorter exposure. There was no statistical significance in the difference in prevalence of cyanosis, clubbing and crepitations among the 3 profusion categories of chest X-ray. Although nearly 25 % of the cases had pulmonary tuberculosis, this did not seem to be related to the profusion category on chest X-ray. 58% of the cases had FVC results which were less than 80% of the predicted values, including all the 6 cases with category 3 profusion on chest X-ray, compared with about 52% ofthose in category 1. This was statistically significant (p = 0.002). INTRODUCTION It is well-established that occupational exposure to silica dust can cause silicosis.1 Although die disease is less prevalent today than in the past owing to better dust control at workplaces, new cases still occur in a number of countries.2 This paper describes the radiological, clinical and ventilatory function findings in silicotic workers on followup in Singapore. MATERIALS AND METHODS Silicosis is a notifiable and compensable occupational disease in Singapore. Confirmation of the diagnosis is made by the Department of Industrial Health, Ministry of Labour, after due investigation. The diagnosis is made mainly on: a) a confirmed history of occupational exposure to dust containing free silica b) a chest X-ray picture consistent with silicosis, with a grading of at least category 1/1 ofthe ILO international classification c) a clinical picture consistent with the disease, and ex clusion of other similar diseases. Silicosis cases are followed up 3 yearly by die Department ofIndustrial Health. A clinical examination, full size posteroanterior chest X-ray and ventilatory function tests (FEVj and FVC) are done. The ventilatory function tests are done on the Autospiror Hl-498. The best of 3 readings is taken and corrected to BTPS. Bach reading is compared to the predicted value for die local population, specific for ethnic group, sex, age and height of die subject.3 688 The chest X-rays were read by one of die authors together with a consultant radiologist. The films were compared to die standard ILO films and graded according to the ILO international classification of radiographs of pneumoconi oses.4 Although the workers were asked for respiratory symptoms (cough, breathlessness, sputum), this study has excluded them because die authors felt that the replies might be too subjective for reliable analysis. Instead, this study has included the objective clinical findings of cyanosis, club bing and crepitations. All cases were asked for a history ofpulmonary tuberculosis. This was then checked with the tuberculosis registry of the National Tuberculosis Control Unit to confirm the history. The period of 1984 to 1986 was taken. As the silicosis cases are followed up 3 yearly, this period would cover all the cases on followup. New cases during this period were also in cluded. However, the final number included in this study has excluded 82 who had died, and 63 others who were unable to come for examination, were unable to cooperate sufficiently for the FEVt and FVC tests, and those who were lost to followup. Four of the cases whose latest chest X-rays were graded as category 0 were also excluded. Thus a total of 140 were included in this study. In the analysis ofthe data, Pearson chi-square test was used, with the significance level set at 0.05. Fisher's Exact Test was used where cell values were less than 5. RESULTS A total of 140 cases were followed up in the period 1984 to 1986. Most of them were men. Chinese male workers comprised by far the largest single group (Table I). Of the 1 IS Chinese men, 93 were granite quarry workers, 22 of whom were still working in the quarries at die time of die follow-up. Seventeen had been exposed to a powder con taining a high percentage of free silica which was used as a "filler" in die milling of rubber (Table II). Two of the workers worked in a quarry where such powder was ob tained; the others worked in rubber factories which used the powder. Epidemiology--Silica Since the vast majority of the cases were Chinese men who had worked or were still working in the granite quarries, it was decided to study them as a fairly homogeneous group. Table m shows that their mean exposure duration as at die date of follow-up was 22.8 years, with a standard deviation of 9.9 years. Their average age was 59.1 years with a S D of 10.2 years. In nearly 68 % ofthis group, the profusion grading of opaci- Tablel Distribution of Silicotic Workers Followed-Up 1984-86, by Ethnic Group and Sex Ethnic Group Male Female Total Chinese Malay Indian Total 115 15 130 7- 7 3-- 3 125 15 140 Table II Distribution of Chinese Male Silicotic Workers by Type of Exposure Type of Silica Exposure Still Exposed Ceased Total Exposure Granite quarry Rubber filler Others Total 22 2 1 25 71 93 15 17 45 90 115 689 Epidemiology--Silica ties on their chest X-rays was category 1, with only 6 out of 93 with category 3 (Table IV). The average age of the workers in the 3 categories was similar. Workers with more than 20 years ofdust exposure did not have a higher category of profusion of opacities than those with shorter exposures (Table V). Table in Duration of Exposure of Chinese Male Silicotic Granite Quarry Workers Duration of exposure* (years) Bo. 1611 16 21 + 5 6 7 23 52 Total X SD (*as at date of followup) 93 22.8 yrs 9.9 yrs Table IV Chest X-ray Category and Age of the Chinese Male Quarry Workers Chest X-ray Profusion Age (years) n X SD Category 1 Category 2 Category 3 63 59.6 10.2 24 58.0 11.0 6 58.2 7.8 Total 93 59.1 10.2 690 A total of 11 out of the 93 cases showed large opacities on their followup chest X-ray films (Table VI). There was an increasing prevalence of chest X-rays with large opacities from the category 1 profusion group to category 3. However this was not statistically significant at the 0.05 level. As the symptoms of cough and dyspnoea were subjective, we analyzed for the more objective signs of cyanosis, clubbing of nails, and crepitations on clinical examination. There was no statistical difference in die prevalence ofthese signs among the 3 profusion categories of chest X-ray (Table VII). Epidemiology-Silica Table VIH shows an apparent increasing prevalence of past occurrence of pulmonary tuberculosis with profusion category. But this was not statistically significant at the 0.05 level. There was no statistical significance in the prevalence of "abnormal" FEV)5 results among the 3 chest X-ray categories (Table IX). But the difference in the prevalence of "abnormal" FVC5 results was significant between categories 1 and 3 (p = 0.002). All the 6 workers with category 3 chest X-rays had FVC values which were less than 80% of those predicted for their age and height. Table V Duration of Exposure in Quarries and Chest X-ray Category Exposure in Quarries Chest X-ray Profusion ^ 20 years > 20 years Total Category 1 Category 2 Category 3 26 13 2 37 63 11 24 46 Total 41 52 93 Chest X-ray Profusion Category 1 Category 2 Category 3 Total Table VI Prevalence of Large Opacities on Chest X-ray With large opacities No. nZ 63 5 7.9 24 4 16.7 6 2 33.3 93 11 11.8 691 Epidemiology--Silica Chest X-rty Profusion Table VH Clinical Signs and Chest X-ray Category Vo. Cyanosis Clinical signs Clubbing Crepitations Category 1 63 Category 2 24 Category 3 6 2 1 0 6 5 0 10 2 1 Total 93 3 11 13 Table vm History of Pulmonary Tuberculosis and Chest X-ray Category Chest X-ray Profusion Vo. Pulnonary tuberculosis No. X Category 1 Category 2 Category 3 63 24 6 12 19 8 33.3 3 50 Total 93 23 24.7 All the 6 persons in category 3 had stopped smoking com pared with 9 out of 20 smokers in category 2, and 15 out of 52 smokers in category 1. DISCUSSION In Singapore, the greatest hazard of silicosis is in die granite quarries where the blasting and crushing of granite are very dusty operations. The second commonest type of exposure was to powder containing a high percentage of free silica which was used in the milling of rubber. Such exposure is much less common now. Of the 93 Chinese granite quarry workers with silicosis followed up, 68% had a chest X-ray profusion grading of category 1, with only 6 in category 3. Eleven showed large opacities. This compares with die 9 out of the 144 silicotics 692 who had large opacities in the followup study reported by Koskinen.6 There was an increasing prevalence of chest X-rays with large opacities from the category 1 profusion group to category 3 but this was not statistically significant. Theriault et al.7 were able to obtain a dose-response curve of granite dust on roentgenograms by plotting the percent age of people with opacities on their chest X-rays against an increasing exposure expressed in dust-years. In our study, the duration of exposure to silica dust did not seem to affect the extent of opacities in die chest X-ray. Workers with more than 20 years exposure in the quarries did not have a higher category of profusion of opacities on chest X-ray than those with shorter exposure. The pathogenesis of silicosis involves the interaction between Table IX FEVj and FVC Results, and Chest X-ray Category FEV^ FVC Chest X-ray No. Profusion normal* abnormal** normal* abnormal** Epidemiology--Silica Category 1 63 Category 2 24 Category 3 6 22 7 1 41 17 5 30 33 9 15 06 Total 93 30 63 39 54 * normal is ^ 80Z of the predicted value ** abnormal is < 80% of the predicted value silica dust and pulmonary macrophages. Inflammatory and fibrotic reactions involved in cell-mediated and humoral im mune responses also participate in the pathogenesis.8 The immunological basis for the development of silicosis, which would have contributed to our findings, could also explain the well-documented wide variation among individuals in the way workers respond to the same exposure to silica dust.9 As the symptoms of cough and dyspnoea were largely sub jective, we analyzed for die more objective signs ofcyanosis, clubbing and crepitations on clinical examination. There was no statistical significance in the prevalence of these signs among die 3 profusion categories of chest X-ray. This could be because the physical signs associated with generalized pulmonary fibrosis, including asbestosis, i.e. clubbing of nails and basal inspiratory crepitations, are less frequent find ings in silicosis.10 Twenty three (or nearly 25 %) of the 93 cases had pulmonary tuberculosis. This compares with 23 (16%) of 144 cases in the series reported by Koskinen.6 However, tuberculosis is endemic in Singapore. It has been said that the risk of con tracting tuberculosis increases with die severity of silicosis, and that previous tuberculosis, whether treated or not, prob ably increases the risk and severity of silicosis.11 But although our cases showed an apparent increasing prevalence of past occurrence ofpulmonary tuberculosis with profusion category on chest X-ray, this was not statistically significant at the 0.05 level. Ventilatory function may be affected in silicosis. This is mainly restrictive in nature, and is indicated by a FVC reading which is less than 80% of the predicted value for the age and height.6 Fifty-four or about 58% of our cases had FVC readings which were less than 80% ofthe predicted values (Table 10). In the series reported by Koskinen, 46% of 144 followup cases were in this category.6 The decrease in lung function may be greater with the coalescence of shadows on the chest X-ray.11 In our series, those with category 3 profusion on chest X-ray all had abnormal FVC readings, compared with about 52; of those in category 1. This was statistically significant (p -- 0.002). Smoking could be a potential confounder although it is less likely to affect the FVC results. Category 3 chest X-rays also showed more large opacities than category 1. It was interesting to note that all the 6 persons in category 3 had stopped smoking on followup, compared with only 15 of the 52 smokers in category 1, although their mean ages were similar. This might have been partly due to the increas ing respiratory difficulty experienced by those in category 3. REFERENCES 1. Koskela, R.S., Klockars, M., Jarvinen, E., Kolari, P.S., Rossi, A.: Mortality and Disability among Granite Workers. Scand. J. Work En viron. Health 13:18-25 (1987). 2. Vigliani, E.C.: Silicosis. In: Parmeggiani, L. (ed.). Encyclopaedia of Occupational Health and Safety V2:2037-2041. International Labour Office, Geneva. (1983). 3. Zee, K.O.: Ventilatory Function in Normal Industrial Workers in Singapore. Annals of(be Academy ofMedicine Singapore 12:587-95 (1977). 4. International Labour Office. Guidelines for the Use of ILO Interna tional Classification ofRadiographs ofPneumoconioses. ILO, Geneva (1980). ' 5. American Thoracic Society: Evaluation of Impairment/Disability Sec ondary to Respiratory Disorders. Am. Review of Resp. Dis. 133:1205-1209 (1986). 6. Koskinen, H. Symptoms and Clinical Findings in Patients with Silicosis. ScandJ. Work Environ. Health 11:101-106(1985). 693 Epidemiology--Silica 7. Theriault, G.P., John, M.P., Lawrence, J.F.: Pulmonary Function and Roentgenographic Changes in Granite Dust Exposure. Arch. Environ. Health 28:23-27 (1974). 8. Gerald, S. Davis: The Pathogenesis of Silicosis. State ofthe Art. Chest 89:166S-169S (1986). 9. WHO: Recommended Health-Based Limits in Occupational Exposure to Selected Mineral Dusts. (Silica, coal). Technical Report Series 734, WHO, Geneva (1986). 10. Morgan, W.K.C., Seaton, A.: Occupational Lung Diseases. Saunder, Philadelphia (1973). 11. WHO: Early Detection ofOccupational Diseases. WHO, Geneva (1986). 694 Epidemiology-Silica & Asbestos REVISED ESTIMATES OF PULMONARY FUNCTION LOSS IN VERMONT GRANITE WORKERS: RESULTS OF A LONGITUDINAL STUDY W. GRAHAM T. Ashikaga S. Weaver University of VT Burlington, VT, USA ABSTRACT Previous studies have suggested that excessive losses of FVC and FEVj.o were occurring in Vermont granite workers despite the fact that quartz levels existing in the industry were below the current OSHA standards. We re-examined these losses in granite workers over an eight year period, testing the workforce from 1979 to 1987 on alternate years. All workers including stone shed, quarry and office were offered forced spirometry using a 10 L. Collins water-sealed spirometer. In the peak year of participation (1983), 887 workers out of a total of approximately 1400 were tested. Estimates of longitudinal loss were based on 711 workers who participated in at least 3 of the semi-annual surveys. The mean age of this group was 42.9 years, and the mean years employed was 19.3 yrs. 21.4% were non-smokers (NS), 34.2% ex-smokers (ES) and 44.4% current smokers (CS). Average annual losses of FVC were .025 .055 L. (CS: .032 L.; NS: .014 L.; ES: .024 L.). Average annual losses of FEVj.o were .036 .040 L. (CS: .044 L.; NS: .027 L.; ES: .033 L.). Analysis of covariance indicated that losses were related to the initial values for FVC or FEV10, height, age, and smoking history. The losses of both FVC and FEV10 were not correlated with years employed in the granite industry. The losses of pulmonary function were significantly smaller than those estimated previously, e.g., .070-.080 L. in FVC, and .050-.070 L. in FEV10. We conclude that current dust levels in the Vermont granite industry do not accelerate pulmonary function loss. See Table of Contents, Part n, for Paper. 695 Epidemiology--Silica & Asbestos LONG FUNCTION WITH ASBESTOS-RELATED CIRCUMSCRIBED PLAQUES EDWARD A. GAENSLER, MJX Peter J. Jederllnic, M.IX Theresa C. McLoud, M.U Departments of Medicine and Surgery, Boston University School of Medicine and Department of Radiology, Harvard Medical School, Boston, MA, USA Early studies of asbestosis made no mention of plaques, and pleural calcifications were not identified with asbestos ex posure until 1955.1012-22 In 1965 Selikoff36 found a long la tent period of 20 years and a high prevalence of 44%. Thereafter, pathologic studies showed no difference between plaques with and without calcification, and often plaques were found without microscopic asbestosis.20,23 Radiographs showed plaques in only a small proportion of cases who had such lesion at autopsy.20,37 Epidemiological ly, it emerged that plaques can be caused by relatively slight household and neighborhood exposure.8 Clinically, plaques early on were described as ` `harmlos-skurriler Schonheitsfehler," that is, a harmless beauty mark,5 since they were neither precancerous lesions nor caused symptoms or loss of function.5,7,23 A number of studies since 196824 have dealt with the func tional consequences of asbestos-related pleural disease but most often no clear distinction was made between cir cumscribed plaques and diffuse pleural thickening. Because fibrothorax of whatever cause may have serious physiologic consequences,13,16,27,29,38 in this study we made a strict distinction between plaques and diffuse thickening. Also, we addressed the confounding effect of smoking; and for con trols we used both normal subjects studied by the same pro tocol and employees matched for age and years of employ ment but without plaques. METHODS Clinical Material We studied 1,764 persons during annual industrial surveys between 1966 and 1988 at two large shipyards, three papermills and one asbestos plant. Details of employment and type of exposure have been described elsewhere.9,15,27 Contra/ subjects included two groups: 100Normal unexposed males (group I) 40 years or older without discernable lung disease and without prior asbestos exposure, who presented for pre-employment examination (Table I)- The second group consisted of 154 persons exposed for 15 ormoreyears (group II) who had normal roentgenograms and who were selected from our survey group by matching for age and years since first exposure with group m which had plaques only and no diffuse thickening or asbestosis. Survey Studies On-site examination included a medical and detailed occupa 696 tional history, a physician-administered respiratory question naire and chest physical examination. Lung function studies included forced vital capacity (FVC), the forced expiratory volume in one second (FEVi) and other flow derivatives, and the single breath diffusing capaci ty (Dl) with alveolar volume (VA) calculated from single breath helium dilution. Instrumentation, unchanged over the years, has been described in detail.17 FVC and FEVj were selected from the best of 3 efforts2 and DL values were ac cepted if VA (BTPS) was at least 90% of FVC (BTPS). Predicted values were calculated from Morris et al.,31 and for Dl from our own data.18 Chest roentgenograms, PA, lateral, and on at least one oc casion oblique views, were obtained within one week of ex amination. Reading was according to the 1980 ILO scheme21 by two "B" readers, one of whom was unaware of the nature and type of exposure. Films were read pro spectively without recourse to other films ("apart reading*'), and were reviewed later by display of the entire series of each case in order of date ("side-by-side" reading). Definitions for this Study Excluded were persons with significant non-asbestos intrathoracic disease, most often chest surgery, trauma, ex tensive pleural and parenchymal scarring from tuberculosis or residuals from infarction or pneumonia. Persons with heart disease we excluded only with marked cardiomegaly and/or evidence of chronic passive congestion. Chronic obstructive lung disease (COLD) was diagnosed for this study only when the ratio of FEVj /FVC was more than 2 SD below the normal predicted.1 We ignored lesser degrees of COLD and evidence of "small airways disease" which was observed in virtually all smokers and ex-smokers. Circumscribed plagues were distinguished from diffuse thickening by detailed study of routine and oblique films aided by review of history, outside records, and CT scans in some cases.27 Descriptions of Fletcher and Edge11 were useful, but the ILO Film was of no help because we believe that the single example of "diffuse thickening" also represents a circumscribed plaque.21 Large plaques were bilateral with width "b" and extent "2" or larger. Large diffuse thickening could be unilateral but had to be of width and extent 2b or larger. Asbestosis was graded according ILO major categories 0,1,2 and 3.21 Years since first ex posure included prior asbestos exposure. Nonsmokers had Group Epidemiology--Silica & Asbestos Table I Clinical Material for Plaque Study Definition NO COLD AlSO Also COLD Diffuse I II III IV-VI Normal Unexposed, age over 40 Exposed > 15 Yrs,Normal X-ray (Matched For Age & Exposure With Group III) Circumscribed Plaques Only Plaques and Asbestosis 1,2 or 3 100 129 197 151 0 25 21 25 0 0 12 50 smoked less than 2 pack-years and had stopped at least 20 years earlier. RESULTS Among 1,764 persons in the survey we found 218 (12.3%) with circumscribed pleural plaques as the only abnormality (Group HI), and there were 176 (10.0%) who had plaques and asbestosis (Groups IV-VI) (Table I). Additionally, 158 exposed persons with normal roentgenograms were matched for age and years since first exposure with the group with plaques (Tables I, II). COLD was found in 16.2% of those with normal X-rays, 9.6% with plaques only, and 14.2% among those with plaques and asbestosis. For some com parisons with normal unexposed persons die COLD group was excluded (Tables I--m. Figure 2). Mean age was similar for all exposed groups, but the 100 unexposed controls were 3 years younger. First exposure was 27 years ago for groups n-VI, and in this study this figure was virtually the same as total years exposed. Lung function tests for die 100 controls were about 4% lower than predicted (Table II) because the Morris equations are for nonsmokers.31 For the 40 nonsmokers in this group all function tests averaged almost exacdy 100% (Table HI). Mean values for the exposed controls (Group H) were slightly lower (p < .05). With circumscribed plaques as the only abnormality (Group m) all three screening tests actually were slighdy higher than the normal subjects (p < .05) or exposed persons without plaques (p <.01) (Table II). With progressing asbestosis there was the predictable precipitous decline, with Dl most severely affected. Inclusion of persons with COLD reduced mean fiinction by only 2%-3% (Table II) because there were few such persons (Table I) and usually their obstructive disease was slight. COLD was no more common among per sons with plaques than those without them (Table I). However, among all smokers and ex-smokers, including "normal" controls, both FEVj and DL were significandy worse (p < .01) than among nonsmokers (Table m). The fact that circumscribed plaques have no measurable ef fect on function was further documented by separation ac cording to width and extent. Table IV and Figure 2 indicate no functional difference between large and small plaques. Calcified plaques were seen in 80 persons (17.5%). This group was older and had longer employment by about 4 years. Nevertheless, the screening tests were virtually iden tical compared to the group with uncalcified plaques. Diffuse pleural thickening initially was recorded from apart readings in 158 cases (Table V). Subsequent detailed study led to exclusion of 96 cases. Among these, subpleural fat pads were recognized more often following publication of the beautiful illustrations by Sargent et al.35 Among the in cluded 62 cases (3.5%) diffuse thickening was most often the residue of a benign effusion (Table V). Diffuse thickening, unlike plaques, caused a significant loss of lung function and, unlike plaques, this loss was strongly related to extent and thickness, with bilateral cases most markedly impaired (Figure 2). Lung volume (FVC) was most severely affected, and DL/VA often was larger than predicted as has been noted by others.26,29,38 DISCUSSION The prevalence of asbestosis is declining rapidly, and most of the 254 cases (14.4%) in this series were the result of first exposure more than 38 years ago.15 Circumscribed pleural plaques were more frequent (23.6%), and many were recognized among persons first exposed less than 38 years ago. This was, in part, because plaques can arise from lesser exposure and, in part, because of improved recognition of early lesions. Therefore, clinical and functional implications of plaques have become of increasing interest. Published material does not provide a good overview of the physiologic effects of circumscribed plaques because, initial ly, attention was focused on calcifications,24 and later the effect ofplaques was obscured by inclusion of diffuse pleural thickening under the general term of "pleural changes."3,6 This is not surprising because neither the 1958 nor the 1971 ILO schemes provided for separate quantification of these pleural reactions.27. Initial physiologic and pathologic studies of benign asbestos effusion showed that these often bilateral and often recur ring bloody effusions frequently result in marked functional impairment and fibrothorax, sometimes so severe as to re quire decortication.16,29 An epidemiologic study of effusions showed persisting radiographic changes: Among 34 persons there remained a blunted costophrenic angle in 91.4% and measurable diffuse thickening in 54.3%.9 The serious con- 697 Epidemiology--Silica & Asbestos sequences of imprisoned lung, so well described in the days of tuberculosis and empyema,13 have now been rediscovered in the asbestos-exposed under such fancy terms as "lung en currasse," "lung entrapment,''^"pleural hyalinosis complicata,"32 or "squashed lung."38 However, unlike pleural thickening after empyema or trauma, in asbestos cases the cortex may increase over the years,28'38 probably from recurring subclinical effusions. Rounded atelectasis from effusion is also described in the asbestos exposed4'30 and also may be associated with functional im pairment. In all of these cases there was marked dyspnea, severe reduction of all lung volumes and D^, and sometimes ventilatory failure. However, in contrast to pulmonary fibrosis, Dl/VA (sometimes called KCO) was normal, indi- Table H Pleural Plaques: Age, Years Since First Employment and Lung Function Group A9 years since First Eapl. r vw % Predicted % Predicted % Predicted Excluding Obstruction, and Excluding Diffuse Thickening I II III IV V VI 100 129 197 112 31 8 51.8 10.7 55.3 8.1 53.8 7.6 56.1 7.6 59.5 8.1 56.8 6.6 0 27.9 + 7.5 27.4 5.9 27.4 6.1 29.0 + 7.0 23.4 7.9 94.5 12.6 91.0 i 13.6 96.1 i 11.4 82.3 i 11.3 73.6 + 16.7 53.9 i 16.8 96.9 13.3 97.3 + 13.3 101.8 13.6 87.5 12.5 77.6 17.7 58.7 + 20.1 95.3 i 17.4 93.8 + 17.9 101.6 13.9 80.2 16.5 58.1 13.6 45.8 11.1 Including Obstruction. I 100 51.8 10.7 II 154 55.5 7.7 III 218 54.0 7.5 IV 128 56.3 i 7.3 V 40 59.5 7.3 VI 8 56.8 6.6 0 28.3 i 7.4 27.4 6.3 27.4 i 6.8 28.9 t 7.1 23.4 7.9 94.5 12.6 90.5 14.3 94.5 12.7 81.6 12.5 74.4 17.1 53.9 i 16.8 96.9 t 13.3 93.6 17.2 98.7 16.8 84.9 15.4 75.2 19.1 58.7 16.8 95.3 i 17.4 91.0 19.1 100.4 14.7 78.4 t 17.0 58.4 t 14.0 45.8 + 11.1 Table m Pleural Plaques: Effect of Smoking Excluding Diffuse Thickening, Including Obstruction Group I II III IV V VI I II III IV V VI I II III IV V VI NO. FVC % 60 110 163 107 36 7 FEVX % 60 110 163 107 36 7 dl % 60 110 163 107 36 7 Smokers 92.4 90.5 93.8 82.4 75.6 55.3 93.4 91.1 97.4 84.9 75.4 60.1 91.6 86.5 98.8 77.3 57.8 47.7 NO. 40 44 55 20 4 1 39 44 55 20 4 1 39 44 55 20 4 1 Nonsmokers 97.8 90.5 96.8 77.6 63.8 44.0 102.4 99.7 102.6 84.6 73.0 49.0 101.0 102.2 105.0 86.3 64.1 32.4 698 Epidemiology--Silica & Asbestos Number 100 1i 54 218 t28 40 6 Age 51.8 55.5 54.0 56.3 59.5 56.8 Yr. Expos 0 28.3 0 FVC % 27.4 27.4 28.9 23.4 DL % % FEV1/FVC Figure 1. Lung function with plaques and asbestosis. Forced vital capacity (FVC) and diffusing capacity (Dl) as percent of predicted, and die FEVj/FVC ratio of 100 normal unexposed males over age 40 (NORM), exposed males with normal roentgenograms matched for age and years since first exposure (CTRL), 218 males with circumscribed hyaline pleural plaques only (PLAQ 0), and 176 persons with plaques and varying degrees of asbestosis (PLAQ 1,2,3). This representation includes persons with chronic obstructive lung disease while persons with diffuse pleural thickening have been excluded. eating that reduced Dl was the result of reduced lung volume and not of impaired respiratory gas exchange. From the foregoing it is evident that inclusion of but a single case of this nature in a group largely with circumscribed plaques would have a significant effect on average values of lung function. A larger series of diffuse thickening was reported by McGavin and Sheers.26 They, like Britton6 before them, devised a grading scheme to measure radiographic extent of diffuse thickening and found impairment of FVC and Dl closely related to severity score. Calcifications alone had no significant effect on function.24-25 SUMMARY AND CONCLUSIONS In 218 persons with circumscribed pleural plaques, but without diffuse thickening or apparent asbestosis, we found lung function with respect to volume, flow and gas exchange no different than that of 154 persons matched for age and years since first exposure who had no visible plaques, and no different than that of 100 unexposed normal subjects. In clusion of persons with chronic obstructive lung disease in the exposed groups did not alter results significantly. There was no difference among persons with and without calcifica tion of plaques, and no difference between large and small plaques. With plaques and increasingly severe asbestosis there was die predictable progressive functional impairment. Diffuse pleural thickening, unlike plaques, caused signifi cant functional loss, especially with regard to lung volumes. This is because pleural plaques involve only the parietal pleura, do not cause adhesive pleuritis, and are patchy and interrupted structures that do not interfere with thoracic mo tion. Diffuse thickening, on the contrary, involves both visceral and parietal pleurae forming an uninterrupted fibrous peal with granulation tissue that extends to involve cortical interlobar septa, and seriously interferes with motion of both 699 Epidemiology--Silica & Asbestos Table IV Circumscribed Plaques: Effect of Size and Extent (Cases with Diffuse Thickening or Obstruction Excluded) Group No. A9e Years Since First Espl- FVC % Predicted FBV % Predicted D^ % Predicted Ssall Plaques l< 2b1 HI 100 S3 6 + 6.9 IV 27 55.9 7.5 V7 VI 4 62.4 6.8 S9.3 5.1 27.5 5.5 28.5 i 7.2 31.4 6.9 26.5 t 8.5 96.0 11.6 81.3 + 9.0 74.7 23.2 58.8 18.7 102.5 13.1 89.4 8.8 74.3 i 16.7 65.8 22.6 101.0 14.4 75.1 + 17.8 58.8 9.5 47.5 10.7 Large PI amiAA <> 2b) III 97 54.0 i 8.3 IV 35 56.4 5.9 V 11 58.4 1 9.5 VI 4 54.3 7.7 27.4 i 6.4 27.9 t 5.3 29.5 7.0 20.2 6.9 96.3 + 11.1 80.3 + 10.6 71.4 19.3 49.0 i 15.7 101.2 t 14.1 86.5 12.0 76.0 i 19.3 51.8 t 17.2 102.2 1 13.4 84.7 t 17.2 49.2 12.1 44.0 13.0 120-1 SMALL LARGE DIFFUSE THICKENING SLIGHT EXTENSIVE Per Cent of Predicted Number 100 ^ FVC % 12 12 B FEV1 % DL % Figure 2. Lung function: plaques and diffuse thickening. The effect on lung function of asbestosrelated pleural disease. Forced vital capacity (FVC), forced expiratory volume in one second (FEVi) and diffusing capacity (DJ were all in the normal range for persons with circumscribed plaques, and there was no difference between small plaques (<2b) and large plaques (>2b). Persons with diffuse pleural thickening had a significant functional deficit, and this was greater with extensive diffuse pleural disease. 700 Table V Causes of Diffuse Thickening Initial Coding from Apart Readings Excluded After Further Study 96 Actually Confluent Plaques 40 Malignancy 5 Infection, Trauma, Surgery 22 Subpleural Fat Pads 18 Included as Diffuse Thickening 62 After Benign Effusion Related to Asbestosis Unexplained 41 15 6 Epidemiology--Silica & Asbestos 158 lungs and thorax. The cause of diffuse thickening in most cases could be traced to one or more episodes of benign asbestos effusion. Our studies confirm the old dictum of Bohlig et al5 that plaques are an epidemiologic leading fossil for asbestos exposure, but otherwise are merely a beauty mark and without functional consequences. We also agree with McGavin and Sheers26 that diffuse thickening at times may lead to significant pulmonary insufficiency and then represents an industrial injury even when there is no asbestosis. REFERENCES 1. Arbeitsphysiologische und Arbeitspatbologiscbe Studien, European Community for Coal and Steel, G. Coppe, Ed., Luxemburg, (1961) p.145. 2. ATS Statement--Snowbird Workshop on Standardization ofSpirometry. R.M. Gardner, Chmn, Am. Rev. Resp. Dis. 119: 831-838 (1979). 3. Becklake, M.R., Fournier-Massey, G., McDonald, J.C., Siemiatycld, J., Rossiter, C.E.: Lung Function in Relation to Chest Radiographic Changes in Quebec Asbestos Workers. BollPhysiopatb. Respir. 6:637 (1970) 4. Blesovsky, A.: The Folded Lung. Brif. J. Dis. Cbest 60:19 (1966) 5. Bohlig, H., Dalquen, P., and Hain, E.: Epidemiologie asbestbedingter Gesundheitsschkden. Der Internist 13:318-325 (1972) 6. Britton, M.G.: Asbestos Pleural Disease. Brit. J. Dis. Cbest 76:1-10 (1982) 7. Dalquen, P., Hinz, I., and Babbert, A.F.: Pleuraplaques, Asbestose und Asbestexposition. Eine eptdemiologische Studie aus dem Hamburger Raum. Pneumonologie 143:547-558 (1970) 8. Epler, G.R., FitzGerald, M.X., Gaensler, E.A., and Carrington, C.B.: Asbestos-related Disease From Household Exposure. Respiration 39:229-240 (1980) 9. Epler, G.R., McLoud, T.C., Gaensler, E.A.: Prevalence and Incidence of Benign Asbestos neural Effusion in a Working Population. JAMA 247:617-622 (1982) 10. Fehre, W.: ber doppelseitige Pleuraverkalkungen infolge beniflicher Staubeinwirkungen. Fortschr. Roentgenstr. 85:16-25 (1956) 11. Fletcher, D.E., Edge, J.R.: The Early Radiological Changes in Pulmonary and Pleural Asbestosis. Clin. Radiol. 21:355-365 (1970) 12. Frost, J., Georg, J, & Moller, P.L.: Asbestosis with Pleural Calcifica tion among Insulation Workers. Danish Med. Bull. 3:202-204 (1956). 13. Gaensler, E.A.: Lung displacement: Abdominal Enlargement, Pleural Space Disorders and Deformities of the Thoracic Cage. In: Handbook ofPhysiology, Section 3: Respiration, Vol. W.O. Fenn and H. Rahn, Eds. Washington: Am Physiol Soc, pp 1623-1661 (1965). 14. Gaensler, E.A., Carrington, C.B., McLoud, T.C.: Thoracic Surgical Problems in Asbestos-related Disorders. An. Thoracic Surg. 40:82-96 (1985). 15. Gaensler, E.A., Jederlinic, P.J., McLoud, T.C.: Radiographic Pro gression of Asbestosis with and without Continued Exposure. Trans actions of VUth International Pneumoconiosis Conference (1988) (see elsewhere in this volume). 16. Gaensler, E.A., and Kaplan, A.I.: Asbestos Pleural Effusion. An. Int. Med. 74:178-191 (1971) 17. Gaensler, E.A., Macklem, P., Chemiack, R., Permutt, S., and Fer ris, B.: Epidemiology Standardization Project m. Recommended Stan dardized Procedures for Pulmonary Function Testing. Am. Rev. Resp. Dis. U8:(#6, part 2) 55-88 (1978) 18. Gaensler, E.A., and Smith, A.A.: Attachment for Automated Single Breath Diffusing Capacity Measurement. Chest 63:136-145 (1973) 19. Hedenstierna, G., Alexandersson, R., Kolmodin-Hedman, B., Szamosi, A., Tollqvist, J.: Pleural Plaques and Lung Function in Construction Workers Exposed to Asbestos. Eur. /. Respir. Dis. 62:111-122 (1981) 20. Hourihane, D.O'B., Lessor, L., Richardson, P.C.: Hyaline and Calcified Pleural Plaques as an Index of Exposure to Asbestos. A Study of Radiological and Pathological Features of 100 Cases with a Consideration of Epidemiology Brit. MJ. 1069-1074 (1966). 21. International Labour Office Guidelines for the Use ofthe ILO Interna tional Classification ofRadiographs ofPneumoconioses, Revised Edi tion (1980). International Labour Office Occupational Safety and Health Series. No. 22 (rev 80) Geneva: (1980). 22. Jacob, G. Bohlig, H.: Roentgenological Complications in Pulmonary Asbestosis. Fortschr. Roentgenstr. 83:515-525 (1955) 23. Kiviluoto, R.: Pleural Calcification as a Roentgenologic Sign of NonOccupational Endemic Anthophyllite-Asbestosis. Acta. Radiolog. Suppi. 0 194 1-67 (1960). 24. Leathart, G.: Pulmonary Function Tests in Asbestos Workers. Trans. Soc. Occup. Med. 18:46 (1968) 25. Luroley, K.P.S.: Physiological Changes in Asbestos Pleural Disease fir: Walton W.H. ed. Inhaled Particles IV. Oxford: Pergamon Press 781 (1977). 26. McGavin, C.R., Sheers, G.: Diffuse Pleural Thickening in Asbestos Workers: Disability and Lung Function Abnormalities. Thorax 39: 604-607 (1984). 27. McLoud, T.C., Woods, B.O., Carrington, C.B., Epler, G.R., and Gaensler, E.A: Diffuse Pleural Thickening in an Asbestos Exposed Population: Prevalence and Etiologies. Am. J. Roentgen 144:9-18 (1985). 28. McMillan, G.H.G., Pethybridge, R.J., and Sheers, G.: Effect of Smok ing on Attack Rates of Pulmonary and Pleural Lesions Related to Ex posure to Asbestos Dust. Brit. J. Industr. Med. 37:268-272 (1980). 29. Miller, A., Teirstein, A.S., Selikoff, I.J.: Ventilatory Failure Due to Asbestos Pleurisy. Am J. Med. 911-919 (1983). 701 Epidemiology--Silica & Asbestos 30. Mintzer, R.A., Cugell, D.W.: The Association of Asbestos-induced Pleural Disease and Rounded Atelectasis. Chest 81:457 (1982). 31. Morris, I.F., Koskd, A., Johnson, L.C.: Spiroraetric Values for Healthy Nonsmoking Adults. Am. Rev. Resp. Dis. 103:57-67 (1971). 32. Naratil, M., Dobias, J.: Development of Pleural Hyalinosis in Long Term Studies of Persons Exposed to Asbestos Dust. Environ. Res. 6:455-472 (1973). 33. Pattern, W.E., Watson, T.R., Jr. and Gaensler, E.A.: Pulmonary Func tion Before and at Intervals after Surgical Decortication of the Lung. Suig. Gyoec. & Obst. 95:477-496 (1952). 34. Rom, W., Thornton, J., Miller, A., Lilis, R., and Selikoff, I.J.: Ab normal Spirometry in Shipyard Workers With Pleural Disease. Am. Rev. Resp. Dis. 115:Part 2 239 (1977). 35. Sargent, E.N., Boswell, W.D., Ralls, P.W., Markovitz, A.: Subpleural Fat Pads in Patients Exposed to Asbestos: Distinction from Non-calcified Pleural Plaques. Radiology 152: 275-277 (1984). 36. Selikoff, U.: The Occurrence of Pleural Calcification Among Asbestos Insulation Workers. An. N.Y. Acad. Sci. 132:351-367 (1965). 37. Wain, S.L., Roggli, V.L., Foster, W.L. Jr.: Parietal Pleural Plaques, Asbestos Bodies, and Neoplasia: Chest 86: 707-713 (1984). 38. Wright, P.H., Hanson, A., Kreel, L., and Capel, L.H.: Respiratory Function Changes After Asbestos Pleurisy. Thorax 35:31-36 (1980). Supported in part by a Program Project Grant (HL 19717), a Career Award(HL 1173), and a Training Grant (HL5567), all from the National Heart, Lung and Blood Institute, U.S. Public Health Service. 702 Epidemiology--Silica & Asbestos PREDICTIVE SIGNIFICANCE OF LESSER DEGREES OF PARENCHYMAL AND PLEURAL FIBROSIS. PROSPECTIVE STUDY OF 1,117 ASBESTOS INSULATION WORKERS, JANUARY 1, 1963-JANUARY 1, 1988. MORTALITY EXPERIENCE IRVING J. SELIKOFF, M.D, Herbert Seidman, MBA Ruth Lilis, M.D Yehuda Lerman, M.D Division of Environmental and Occupational Medicine, Mt. Sinai School of Medicine of the City University of New York and the Department of Epidemiological Research of the American Cancer Society, New York, NY, USA abstract In 1963, 1,117 asbestos insulation workers were examined. X-ray findings were categorized according to the Saupe classification and reported. Subsequently, the films were recategorized when the International Labour Office Classification was introduced, using die 1980 Classification. The entire cohort has been main tained under observation. We have investigated all deaths that have occurred in the 25 years and will pre sent the mortality experience of this group in relation to the radiological findings in 1963, with particular reference to 0/0, 1/0, 0/1, 1/1 interstitial, as well as the presence or absence of pleural fibrosis. No Paper provided. 703 Epidemiology--Silica & Asbestos SPIROMETRIC ABNORMALITIES IN 2573 ASBESTOS INSULATORS WITH LONG TERM EXPOSURE: EFFECTS OF SMOKING HISTORY AND RADIOGRAPHIC ABNORMALITIES ALBERT MILLER, M.D. Ruth Litis, M.D. James Godbold, Ph.D. Eva Chan, Ph.D. Irving J. Selikoff, M.D. Divisions of Environmental and Occupational Medicine (Community Medicine) and Pulmonary Disease (Medicine) Mount Sinai School of Medicine, City University of New York New York, NY USA INTRODUCTION The 2573 insulators who are the subjects ofthis report com prise one of the largest reported populations occupationally exposed to asbestos. They were selected on the basis of a long duration from onset of exposure (DURON; 87% were ^ 30 years) in order to provide sufficient time for evolu tion ofdisease, whether it be pleuropulmonary fibrosis caused by asbestos or chronic airways obstruction attributable to smoking and perhaps to occupational dusts. This population is therefore well suited to provide information concerning: 1. The prevalence of the various pulmonary function im pairments in a large well defined group occupationally exposed to asbestos. 2. The effect of such influences as radiographic abnor malities, DURON and smoking on pulmonary function. In this regard, the large number of lifetime nonsmokers (n = 506) allows characterization of the effects on pulmonary function of asbestos inhalation alone, uncon founded by cigarette smoking. METHODS Details concerning subject selection, medical evaluation and radiographic reading are given in a companion paper.Sub jects were studied in 19 cities in North America over a two year interval. Spirometric tests adhered to current guidelines;2 at least 3 acceptable efforts were obtained on each subject, who was standing and wore a noseciip. A com puterized rolling seal spirometer was used; efforts were monitored by maximum expiratory flow-volume curves recorded in real time and all data, including the flow volume arrays, were stored on digital tape. Predicted values were those published by this laboratory us ing the same equipment in a random sample of the popula tion of a large industrial state, adjusted for the effects of smoking in current smokers (which were significant on all tests except FVC).3 "Nonsmokers" (NS) smoked less than one cigarette a day, had smoked ^ ten cigarettes a day for ^ six months or smoked only cigars and pipes, which are not inhaled. Cur 704 rent smoker (SM) exceeded these limits. * `Ex-smokers' ' (XS) exceeded these limits and had discontinued smoking ^ two years previously. Impairments were defined as follows: Normal (Nl): Nl. FVC, FEV,/FVC and MMF; Restrictive (Rest): FVC <95% lower confidence inter val (Cl) Restrictive (Rest): FVC <95% lower confidence inter val (Cl), ,FEV|/FVC nl (^0.65 age ^60, ^0.70 age 30-59); Overt Obstructive (Obs): FVC nl, FEVj/FVC below limits defined above; Small Airways Dysfunction (SAD): FVC nl, FEV, /FVC normal, FET^s* (also called mid-expiratory time) ^0.78 sec;5 Combined, primarily restrictive (Comb Rest): Both FVC and FEVj/FVC i ; i FVC i FEVj/FVC; Combined, primarily obstructive (Comb Obs): i FEVj/ FVC > i FVC. RESULTS Mean Values Table I shows mean demographic, exposure and pulmonary function variables for NS and those with a smoking history. There is no difference in age, DURON or years exposed. FVC and FEV! are reduced in the NS, but not FEVj/FVC or the flows FEF^s* or FEF75%. FVC and FEVi are reduced to a greater extent in the smokers but FEVt /FVC and flows are only minimally reduced using smoking specific predicted values. Prevalences of Functional Impairment By Smoking History (Table II): Of the 2573 workers studied, 506 (19.7%) were NS, 861 (33.5%) SM and 1206 (46.9%), XS. This last group, which includes only those who have discontinued smoking for at least 2 years, reflects the effect of educational efforts to discontinue smoking among asbestos workers. Epidemiology--Silica & Asbestos Table I Demographic, Exposure and Pulmonary Function Variables by Smoking History (Mean and SD) Nonsmokers (n = 506) Current and Ex-Smokers (n = 2067) Age (yrs) Height (cm) Duron (yrs) Yrs exposed Dur smoking (yrs) Pack years FVC (% pred) FEV-i (% pred) fef25-75% (* pred) fef75% (* pred) FEV-j/FVC x 100 FEV-j/FVC (% pred) 58.6 ( 8.6) 173.2 ( 6.9) 36.3 ( 7.6) 32.0 ( 8.3) 0 0 86.5 (16.6) 85.5 (17.4) 96.6 (36.2) 101.2 (45.5) 79.2 ( 7.6) 100.6 ( 9.5) 57.3 ( 8.0) 174.0 ( 6.6) 34.9 ( 7.0) 31.6( 7.5) 31.6 (12.0) 40.6 (26.1) 82.0 (17.3) 02.3 (20.6) 86.6 (40.1) 97.7 (49.3) 73.4 (10.5) 97.1 (13.5) Table II Spirometric Impairments in 2573 Asbestos Insulators ^20 Years from Onset of Exposure (by Smoking History) NS (19.7%)* SM (33.5%) XS (46.9%) All Smoking Categories Normal pf Restrictive 222 (43.9)* 156 (30.8) 146 (17.0) 270 (31.4) 376 (31.2) 413 (34.2) 744 (28.9) 839 (32.6) Obstructive 17 (3.4) 144 (16.7) 92 (7.6) 253 (9.8) Small Airways 98 (19.4) 145 (16.8) 204 (16.9) 447 (17.4) Combined 13 (2.6) 156 (18.1) 121 (10.0) 290 (11.3) Combined, rest Combined, obst. 11 (2.2) 2 (0.4) 110 (12.8) 46 ( 5.3) 92 (7.6) 29 (2.4) 213 (8.3) 77 (3.0) All Impairments 506 861 1206 2573 * Percentages are shown in parentheses. Percentages after each smoking category are of the total population (e.g., 19.7% of the population were NS). Percentages within each smoking category are of each impairment within that smoking category (e.g., 43.9% of NS had nl pf). The prevalence of several impairments varies by smoking history. Normal pulmonary function was most likely in NS (43.9%) and least in SM (17.0%); XS were intermediate (31.2%). Overt obstruction was most likely in SM (16.7%) and least in NS (3.4%). While restrictive impairment by itself did not vary in frequency by smoking category, combined impairment was far more common in SM (18.1%), again with XS intermediate. Frequency of small airways dysfunc tion was similar in all smoking categories. By radiographic abnormality (Table m): Normal pulmonary function was most likely (43.4%) when the chest radiograph was normal and least likely (21.2%) when both parenchymal and pleural disease was present. While frequencies of small airways dysfunction did not vary by radiographic abnormality, restrictive and combined im pairments were most common when parenchyma and pleura were both abnormal and obstruction was more likely when only die parenchyma was abnormal. Interestingly, when fre quency of restriction was compared in isolatedparenchymal vs. isolatedpleural disease, it was greater in the latter (34.4% vs. 22.3%). Despite the greater frequency of normal function when the radiograph was normal (43.3%), 21.3% of those with a nor mal film had restriction and 7.7% had combined impairment, meaning that 29% had a reduced FVC. Similarly, while a normal film was more likely when pulmonary function was normal (25.1%), 10.4 percent of workers with restriction had a normal film. 705 Epidemiology--Silica & Asbestos Table m Spirometric Impairments in 2573 Asbestos Insulators ^20 Years from Onset of Exposure (by Radiographic Abnormality) Normal pf Restrictive Obstructive Small Airway Combined All Impairments "Normal" (16%)* Parenchymal Only (11.7%) 187 (43.3)* 105 (34.9) 88 (21.3) 67 (22.3) 36 (8.7) 37 (12.3) Both (Parpleu) (48.2%) 263 (21.2) 471 (38.0) 121 (9.8) Pleural Only (24.1%) 189 (30.5) 213 (34.4) 59 (9.5) All Radiograph Categories 744 (28.9) 839 (32.6) 253 (9.8) 77 (18.6) 25 (7.7) 64 (21.3 28 (9.3) 202 (16.3) 182 (14.7) 104 (16.8) 55 (8.9) 447 (17.4) 290 (11.3) 413 301 1239 620 2573 * Percentages are shown in parentheses. Percentages after each radiographic abnormality are of the total population (e.g., 16% of the population had normal films). Percentages within each radiographic abnormality are of each impairment within that radiographic abnormality (e.q., 43.3% of subjects with normal films had normal pulmonary function). The percentage of NS was relatively greater in subjects with normal pulmonary function no matter in which radiographic category they fell whereas the percentage of SM was in creased in those with obstructive or combined impairments within each radiographic category. By Duration from Onset of Exposure (Table IV): Frequency of normal pulmonary function fell with increased duration of exposure while frequency of restriction and of combined impairment increased. Small airways dysfunction did not change and obstruction actually decreased. Regression Analysis of Pulmonary Function FVC (analyzed as percent predicted, to adjust for age and height) and the actual ratio FEVj/FVC were analyzed for the contributions of such independent variables as radiographic category (any par, any pleu and parpleu interac tion), DURON and pack years (Table V). Parenchymal in volvement was not significant for FVC, although pleural and combined parenchymal-pleural involvements were. Pleural involvement (alone or combined) was not significant for FEVj/FVC: The predominant influence on FEVj/FVC was pack years, each pack year diminishing the FEVj/FVC (x 100) by 0.11 so that 41 pack years (the mean of all subjects with a positive smoking history) would diminish the FEVj/FVC (x 100) by 4.5. By contrast, DURON 35 years (the mean of all subjects in the study) would diminish FEVj/FVC (x 100) by two-thirds of the smoking effect (2.9) and the presence ofparenchymal disease on chest film would diminish die ratio negligibly (0.013). Each year from onset diminished the FVC by 0.436 percent ofpredicted, so that DURON 35 years would diminish FVC by 15.3 percent of predicted, whereas each pack year 706 diminished FVC by 0.096 percent of predicted. Cumulative decrement for 41 pack years was 3.9 per cent of predicted or 25 % of the effect of DURON. The presence of pleural involvement on radiography diminished the value by 4.7 per cent of predicted and the presence of parenchymal plus pleural involvement diminished the value by 7.8 percent of predicted. DISCUSSION The 2573 asbestos insulators are a large enough group to allow analysis of the effects on lung function of such indepen dent variables as years from onset of exposure, cigarette smoking (pack years) and radiographic abnormalities. We used percent predicted FVC, the most easily and universal ly measured single test of pulmonary function, as our index of restrictive impairment. The largest effect was that of DURON, followed by combined pleuropulmonary involve ment, smoking and isolated pleural involvement. As ex pected, smoking was the predominant influence on FEVj/FVC, used as an index of airways obstruction. The effect of DURON is probably attributable to aging (which obviously parallels duration) since the FEVj/FVC ratio is not adjusted for age as is percent predicted FVC. In computing die prevalence of impaired pulmonary func tion, common practice has been to use as the numerator die number of subjects with abnormal values for any of the ma jor pulmonary function parameters, e.g., FVC or FEVj. This tends to obscure differences in the types of impairments and the relationship of these impairments to various ex posures, e.g., asbestos vs. cigarette smoking. A reduced FVC is not necessarily indicative of restriction (airways obstruction with air trapping and an elevated RV may reduce Epidemiology--Silica & Asbestos Table IV Spirometric Impairments in 2573 Asbestos Insulators ^20 Years from Onset of Exposure (by Duration from Onset of Exposure) < 29 Yrs. n = 331 (12.9%) 30-39 Yrs. n = 1593 (61.9%) > 40 Yrs. n = 649 (25.2%) Normal Restrictive Obstructive Small Airways Combined 127 (38.4) 84 (25.4) 39 (11.8) 56 (16.9) 25 ( 7.6) 470 (29.5) 487 (30.6) 171 (10.7) 287 (18.0) 178 (11.2) 147 (22.7) 268 (40.3) 43 (6.6) 104 (16.0) 87 (13.4) Table VA Regression Analysis of Percent Predicted FVC Percent Predicted FVC (n = 2667) Table VB Regression Analysis of FEVj/FVC FEV,/FVC x 100 (n = 2573) Intercept Duron Pack Yrs. Any pleu Parpleu interact Any par R2 106.19 -0.436 (pQ.0001, F 92.4) -0.096 (p0.0001, F 74.2) -4.676 (pO.0001, F 28.1) -3.091 (pO.0001, F 14.8) NS 11.7 Intercept Pack Yr. Any par Duron Any pleu Parpleu interact 81.949 -0.111 (p0.0001, F 265.8) -0.013 (p0.0018, F 9.73) -0.084 {p0.002, F 9.56) NS NS R2 11.1 the FVC as well) nor are reduced FEVj and flow rates specific for obstruction (restrictive disease will generally result in a reduction in FEV( and flow rates proportional to the reduction in FVC). Instead, a mutually exclusive classification of impairments based on a combination of spirometric measurements, in cluding the FEVi/FVC ratio, was employed. We have published die prevalence ofthese impairments in 351 patients with chronic pulmonary sarcoidosis.4 Preliminary results from our sampling of die population of the state of Michigan5 show far lesser frequencies of restriction (7% NS, 9% XS, 12% CS) and combined impairment (0% NS, 5% XS, 5% CS) but similar or greater frequencies ofobstruc tion (13% NS, 17% XS, 21% CS). Not surprisingly, normal pulmonary function was most likely in NS (44%) and least likely in SM (17%) while obstruction was most likely in SM (17%) and least in NS (3%). FRE QUENCY OF RESTRICTIVE IMPAIRMENT DID NOT VARY BY SMOKING HISTORY, consistent with the predominant effects on FVC of asbestos exposure (measured as years from onset) and radiographic evidence of pleuropulmonary or pleural fibrosis. However, combined im pairment was far more frequent in SM (18%) than NS (2.6%), with XS intermediate. This pattern results from the addition of obstruction (attributable to smoking) to the restric tion of pleuropulmonary fibrosis. Such combined impairment is frequently seen in advanced sarcoidosis, bronchiectasis, cystic fibrosis or silicosis (progressive massive fibrosis).6'7 It is of interest that the most common impairment in both SM and XS (as well as NS) was restriction. It is recognized that what we call combined impairment can also result from reduction in FVC secondary to air trapping in severe obstruc tion. True combined impairment and air trapping can be distinguished only by full lung volumes. These obviously can not be measured using physiologic methods on such large numbers of subjects under survey conditions. (We hope to measure full lung volumes using a planimetric method8 on the posteroanterior and lateral chest films of these workers). We attempted to separate combined impairments into "predominant restrictive," and "predominant obstructive" by the relative decreases in FVC vs. FEVj/FVC. RESTRICTION WAS IDENTIFIED AS THE PREDOMINANT ELEMENT IN THE COMBINED IMPAIRMENT IN ALL SMOKING CATEGORIES ffeble II) The large number of NS in our study permits us to characterize the pulmonary function patterns attributable to inhalation of asbestos fibers alone. The 3 percent prevalences each of overt obstruction and of combined impairment do not provide strong evidence for obstructive impairment resulting from such inhalation (Table II). Halfofthe patients with these impairments (14 of 30) did not show radiographic evidence of pulmonary fibrosis. These 30 (of 506) NS may well have an independent cause of obstruction (e.g., asthma). The decrease in frequency of obstruction with increasing 707 Epidemiology--Silica & Asbestos duration ofexposure is the reciprocal of the increase in com bined impairment, resulting from the superimposition of restriction in subjects with obstruction. The greater contribution of radiographically identified pleural fibrosis (by itself) than of interstitial fibrosis (by itself) to restrictive impairment is seen in the regression analysis (Table V) and in the higher frequency ofrestrictive and com bined impairments in those with isolated pleural vs. isolated parenchymal involvement (Table DI). These findings con cerning pulmonary function are similar to those concerning dyspnea in this population1*9 and demonstrate the important effect of pleural fibrosis on lung function. These conclusions concerning pleural vs. parenchymal fibrosis may not be generalizable to other exposed populations for the follow ing reasons: 1. This population was selected for long duration from onset and for heavy occupational exposure. 2. It is to a certain extent a survivor population in which those with the most severe interstitial fibrosis may have died (either of respiratory insufficiency or of bron chogenic carcinoma). 3. The largest proportion of subjects (48.2% of all sub jects and 52.4% of those with abnormal films) had pleuropulmonary disease. However, pleural disease is more prominent than interstitial fibrosis in many occupations with less intense exposure, and in family exposures. The important effect of pleural disease on lung function may be relevant to these groups. Pulmonary function tests are used not only to quantify and characterize impairment in asbestos-related disease but to detect evidence of disease which is not apparent clinically or radiologically. In this regard, it is noteworthy that21.3% ofthose with a normal film had restriction and an additional 7.7% had combined impairment and that of those with only pleural disease, 34.4% had restriction and an additional 8.9% had combined impairment. Further testing, especially of gas exchange (Dl, Vd/V?) would undoubtedly uncover addi tional individuals with intrinsic lung disease despite normal lung fields on chest radiograph. REFERENCES 1. Lilis, R., Miller, A., Godbold, J., Chan, E., Klein, S., Selikoff, I.J.: Radiographic Abnormalities in a Large Group ofInsulators with Long Term Asbestos Exposure: Effects of Duration from Onset ofExposure and Smoking. VII International Pneumoconioses Conference; Pittsburgh, Pennsylvania, August (1988). 2. Gardner, R.M., Chairman, Standardization of Spirometry--1987 Up date. Am. Rev. Respir. Dis. 136:1285-1298 (1987). 3. Miller, A., Thornton, J.C., Warshaw, R., Bernstein, I, Teirstein, A.S., Selikoff, I.J.: Mean and Instantaneous Expiratory Flows, FVC and FEVi: Prediction Equations for Nonsmokers and Smokers from a Ran dom Sample of Michigan, a Large Industrial State. Bull. Eur. Physiopath. Resp. 22:589-597 (1986). 4. Miller, A., Teirstein, A.S., Pilipski, M., Brown, L.K.: The Spectrum of Airways Obstruction in Sarcoidosis. Proc. XI World Congress on Sarcoidosis and Other Granulomatous Disorders, Milan, Italy, September (1987). 5. Miller, A., Thornton, J.C., Anderson, H.A., Selikoff, I.J.: Prevalence of Clinical Respiratory Abnormalities by Sex and Smoking History in a Representative Sample of the Adult Population of Michigan. Chest, in press. 6. Wanner, A.: Interpretation of Pulmonary Function Tests, in Sackner, M.A. (fid); Diagnostic Techniques in Pulmonary Disease, Part/, Volume 16 of Lung Biology in Health and Disease, Marcel Dekker, New York (1980). 7. Miller, A., Teirstein, A.S., Jackler, I., Chuang, M., Siltzbach, L.E.: Airway Function in Chronic Pulmonary Sarcoidosis with Fibrosis. Am. Rev. Resp. Dis. 109:179-189 (1974). 8. Harris, T.R., Pratt, P.C., Kilbum, K.H.: Total Lung Capacity Measured by Roentgenograms. Am. J. Med. 50:756-763 (1971). 9. Lilis, R., Leman, Y., Malkin, J., Selikoff, I.J.: Interstitial Pulmonary Fibrosis and Pleural Fibrosis: Comparative Prevalence and Symptoms in Insulation Workers with Over 30years from Onset ofExposure. Proc. VI International Pneumoconioses Conference; Bochum,FRG, October (1983), International Labour Organization, 697-715. 708 Epidemiology--Silica & Asbestos MORTALITY AND CANCER INCIDENCE AMONG SWEDISH CERAMIC WORKERS WITH SILICOSIS GORAN TORNLING*t C. Hogstedt*t A. Gustavssont P. Westerholm Departments of Occupational Medicine and tThoracic Medicine Karolinska Hospital, S-104 01 Stockholm ^National Institute of Occupational Health, Solna The Swedish Trade Union Confederation, Stockholm. Sweden The question whether quartz exposure increases the risk for lung cancer was raised 50 years ago by Anderson & Dible.1 More recently two literature reviews considering both animal experiments and epidemiological investigations have been published, one talking for3 and the other against4 such an association. In most published epidemiological investigations the subjects have, beside quartz, been exposed to known lung carcinogens such as polycyclic aromatic hydrocarbons in the foundry industry and ionizing radiation in mining, which make die interpretation difficult.5 This investigation, which has become a part of a multicenter initiative by IARC, was undertaken in order to study the risk for lung cancer in the Swedish ceramic industry, where no confounding lung car cinogens are known to occur. MATERIAL AND METHODS The Swedish Pneumoconiosis Registry has compiled case notifications from 1931 and onwards. The study population consists of those 314 males from the ceramic industry who had been accepted as compensated cases of silicosis. Since there were only 36 females they were excluded from analysis. The vital status ofdie study persons was established by link ing die ten-digit identification number based on time ofbirth for each person with the census register for all living per sons in Sweden, the death register of all deceased persons in Sweden, and the emigration register. By these procedures all but one of the persons could be identified as alive or deceased during the study period. During the study period (1951-1985) for the mortality analysis the study population has accumulated 5695 personyears. Cause ofdeath from the death certificates was recorded on all deceased persons in the study population. The Swedish Cancer Registry was established in 1958 and receives notifications on more than 95 % of all malignancies. During the study period (1958-1983) for the cancer incidence calculations die study population had accumulated 4247 person-years. The cancer morbidity was established by link ing the identification numbers of die individuals in die study population with the National Cancer Register. The expected number of deaths and malignancies was calculated by multiplying person-years of observation within five-year age categories during each year of the study periods by site- and gender-specific national rates. The calculations of standardized mortality/morbidity rates (SMR) with 95 % confidence intervals (Cl) based on a Poisson distribution was performed by a computer program developed at the Univer sity of Linkoping (EPHJN program package). RESULTS The overall mortality in the study population was increased (SMR= 138; Cl 120-157) due to an excess in mortality from respiratory tuberculosis (SMR=1932; Cl 1144-3054) and other non-malignant respiratory diseases (SMR=746; Cl 577-947). There was no overall increased incidence of malignant diseases in the study population (SMR=94; Cl 67-126). However, nine cases of lung cancer were observed vs. 4.8 expected (SMR=188; Cl 85-356). With a latency time re quirement of 10 years from discovery of silicosis SMR was 236 (Cl 107-448), and with a latency time of 20 years there was a further increase (SMR=267; Cl 98-582). Lung cancer was diagnosed 36-72 years after first quartz exposure and 11-32 years after that silicosis had been detected. DISCUSSION The results from this study on Swedish ceramic workers with silicosis demonstrate an increased mortality in non-malignant, but not in malignant, respiratory diseases. However, the lung cancer incidence was doubled. The size ofthe cohort is small, and we have no data on smok ing habits in the study population. The results are, however, in accordance with other studies on silicotics from the ceramic industry. An increased risk among pottery workers of dying from lung cancer has been reported, especially if they had been exposed to talc,7*9 but also in individuals with no talc exposure.8 Talc was, however, only used in four of the 19 factories in our study, and the lung cancer cases were not accumulated to these factories. A recent Italian casereferent study among ceramic workers controlling for age, period of death and smoking did also show an increased lung cancer risk, especially among individuals with silicosis.2 The mechanism for a carcinogenic effect of quartz is not clear. A direct carcinogenic effect is supported by the cytotoxic effect of quartz in vitro.6 It has been shown 709 Epidemiology--Silica & Asbestos among ceramic workers that silicotics have a greater risk for lung cancer than non-silicotics.2 This could be due to a higher quartz exposure, but a causal relationship between the silicotic lesions and die cancer should be considered. Fibrotic lesions in the lung might impair die pulmonary clearance mechanisms for various carcinogenic substances, and a high incidence of bronchial carcinoma has been reported among individuals with various fibrotic lung diseases without relationship to dust exposure.10 REFERENCES 1. Anderson, C.S., Dible, J.H.: Silicosis and Carcinoma ofthe Lung. J. Hyg. 38:185-204 (1938). 2. Forastiere, F., Lagorio, S., Michelozzi, P., Cavariani, F., Arcaco, M., Borgia, P., Perucci, C., Axelson, 0.: Silica, Silicosis and Lung Cancer among Ceramic Workers: a Case-Referent Study. Am. J. Ltd. Med. 10:363-370 (1986). 3. Goldsmith, D.F., Guidotti, T.L., Johnston, D.R.: Does Occupational Exposure to Silica Cause Lung Cancer?. Am. J. Iod. Med. 3:423-440 (1982). 4. Heppleston, A.G.: Silica, Pneumoconiosis and Carcinoma ofthe Lung. Am. J. Iod. Med. 7:285-294 (1985). 5. IARC.: Monographs on the Evaluation of the Carcinogenic risk of Chemicals to Humans Vof 42. Silica and some Silicates. IARC, Lyon (1987). 6. Langcr, A.M., Nolan, R.P.: Physiochemical Properties ofQuartz Con trolling Biological Activity. Cancer-Research Monographs. 2:125-135 (1986). 7. Thomas, T.L.: A Preliminary Investigation ofMortality among Workers in the Pottery Industry, lot. J. Epidemol. 11:175-180 (1982). 8. Thomas, T.L., Stewart, P.A.: Mortality from Lung Cancer and Respiratory Diseases among Pottery Workers exposed to Silica and Talc. Am. J. Epidemiol. 125:35-43 (1987). 9. Thomas, T.L., Stewart, P.A., Blair, A.: Nonfibrous Dust and Cancer: Studies at the National Cancer Institute. CancerResearch Monographs. 2:441-450 (1986). 10. Turner-Warwick. M.: Immunology ofthe Lung. Edward Arnold, Lon don (1978). ACKNOWLEDGEMENT: The study was supported by grant 86-1391 from the Swedish Work Environment Fund. 710