Document vOqynoDKOY2GpL4vj9Jd6gY9

GRACE Construction Products Division July 9, 1984 TO: H. A. Eschenbach R. C. Walsh R. J. Bettacchi R. T. Frohlich J. W. Wolter FROM: A. N. Crawford- SUBJ: Article From The American Thoracic Society (ATS) RE: "Pulmonary Changes After Exposure To Vermiculite Contaminated With Fibrous Tremolite" The above referenced article is attached for your information. I received this copy from the Vice President of Research KNAUF Fiberglass Co., Dr. John D. Koch, in Shelbeyville, Indiana. Dr. Koch received this in a packet of information sent to all members on the Medical & Scientific Committee of the Thermal Insulation Manufactures Association (TIMA). Knauf is evaluating our new experimental vermiculite despersion and they are concerned about the potential exposure to tremolite. Regards, ANC/dp A. N. Crawford 10002235 Pulmonary Changes after Exposure to Vermiculite Contaminated with Fibrous Tremolite1'3 J. E. LOCKEY, S. M. BROOKS, A. M. JARABEK, P. R. KHOURY R- T. MCKAY, A. CARSON, J. A. MORRISON, J. F. WIOT, and H. B. SPITZ Introduction Vermiculite is the geological name given to a group of hydrated laminar aluminum-iron-magnesium silicates. It has the unique property of expanding as much as 12 times its original size with the application of heat between 427 and 1,093 C (1, 2). Unexpanded vermiculite is mined mainly in Mon tana, Virginia, and South Carolina in the United States, and in South Africa, and it is shipped to approximately 47 regional expander plants located in 30 states. The domestic uses of expanded vermiculite relate to its fire resistance, insulation, and ion exchange proper ties, but, additionally, it is used as a soil additive, animal feed bulking agent, and as a carrier for various chemicals, including herbicides, insecticides, fun gicides, and fertilizers (2). Investigations of some of the unex panded vermiculite ore have demon" strated contamination of the ore with fibrous minerals (3). Montana ore con tains a fibrous form of the amphiboic tremolite. Virginia and South Carolina vermiculite ore contain a type of tremo lite that, when milled, tends to form cleavage fragments that have fibrous characteristics with low length-to-width aspect ratios. South African ore is cur rently felt to be free of amphibole or cleavage fragment contamination (4). A rurally located company that pro cessed mainly Montana vermiculite ore to its expanded form for use as an inert carrier for herbicides and fertilizers reported a cluster of 12 cases of pleural effusions of unknown origin among their employees over a 12-yr period. Environmental sampling of work areas revealed airborne fibers believed to be tremolite. There was concern that the observed cluster of pleural effusion cases represented manifestations of exposure to the fibrous contamination of the vermiculite. The present study was undertaken to assess the respira 952 SUMMARY Worfcan axposad to vurmlcuBta contaminated with fibrous tromollls war** surveyed lor lha pratanca of raaplralory aymplonu by questlonnslre, and for pneumoconiosis by chasl radiograph. Pulmonary function was maaunad by apiromatry and single-breath carbon monox ide diffutlng capacity (OLco*b). Fiber axpotura Indexes, axpraaaad at flberlml-yr, wara derived tor aach arorfcer from avellable Indutfrfal hygiene data and work histories. The attlmatad cumula tive axpotura for the work force ranged from 0.01 to 39 fiber/ml-yr. Dltcrlmlnanl analyala demonatratad algnlllcant dorralafaa with ahortnaat of breath and plaurttlc chaat pain to cumulative liber exposure. The radiographic changaa wara limited to pleural changaa and Involved 4.4% of the population. Parametric and dlacrtmlnant analyala demonetrated a algnlllcant correlation with radiographic changaa and cumulative fiber axpoaura Thera wara no corralatlona between apiromatry or DLcoah and fiber axpotura. ExoauraJovannleulltacontamlnatadjirlth_tlbroua tremolite can ceute pleural changaa In occupationally expoaed workara. Thla la aupported by the previouely Idantltiad 12 caaaa of benign piaural attualona In thla working population and the aeaodatlon of piaural radiographic changaa and plaurttlc cheat eymptoma with cumulative fiber axpotura. The lack ol algnlllcant parenchymal radiographic, aplromatrtc, and Dlco*h changaa moat likely redacts the low cumulative fiber axpotura. am rev respih DS iat4; I:l52-t5e tory status of current workers exposed to vermiculite contaminated with fi brous tremolite in this plant facility. Methods Study Population The study population surveyed included all employees with a past history of vermiculite exposure and a control group of employees without such exposure. There were a total of 330 employees asked to participate in the study; 9 refused and 9 were not available because of vacation or illness, giving a total of 512 employees (97%) interviewed. Medical Examination All employees were interviewed by trained personnel using a modified American Thoracic Society (ATS) Respiratory Ques tionnaire (3). The major modifications of the questionnaire were the inclusion of questions pertaining to previous employ ment with asbestos and other fibrous mineral exposure, questions about time employed within various locations in the facility, and questions relating to pleuritictype chest pain and illnesses with pleural manifestations. A limited physical examination was per formed on each employee for the presence of late inspiratory rales (crackles) in 4 dif ferent chest locations and for the presence of nail clubbing. Spirometry was performed using an Ohio-Med 822 dry rolling seal spirometer (Ohio Instruments, Pine Brook, NJ) with a Spirotech 200 microprocessor (Spirotech, Inc, Atlanta, GA). Tests were accomplished according to ATS criteria using trained tech nicians (5). Employees were retested at a later date if they had had a respiratory in fection within the preceding 3 wk. Tests performed were forced vital capaci ty (FVC), forced expiratory volume in one second (FEV,), ratio of FEV, to FVC ex pressed as a percentage (FEV,/FVC%), and forced expiratory flow during the middle half of the FVC (FEF,M,). AH results were temperature corrected to bits. Results were expressed as measured values and as per centage of predicted using the normal values of Knudson and coworkers (6). {Received in original form June 29, 1983 and in revised form January 10, 1984) ` From the Department of Environmental Health, Division of Clinical Studies, University of Cincinnati, Cincinnati, Ohio. ' Presented in pan at the Annual Meeting of the American Thoracic Society, Detroit, May 1981. * Requests for reprints should be addressed to James E. Lockey, M.D., M.S., Rocky Mountain Center for Occupational and Environmental Health, Building 512. University of Utah, Sail Lake City, UT 84112. 10002236 AT CHANOES AFIE* EXPOSURE TO VEAMICUUTE 953 Carbon Monoxide Diffusing Capacity Single-breath carbon monoxide diffusing capacity (Dicoib), alveolar volume (Va), and diffusing capacity per unit of lung vol ume (Dl/Va) were obtained using a Collins Modular Lung Analyzer* (Warren E. Col lins Co., Braintree, MA). All diffusion measurements were conducted according to ATS criteria using trained technicians (S). The diffusion carbon monoxide (CO) ana lyzer, helium meter, and Collins water seal spirometer were calibrated every 3 h during use. Employees were retested at a later date if they had had a respiratory infection with in the preceding 3 wk or had smoked within 1 h prior to testing. Tests were accomplished in a seated position with a noseclip. Repeat tests were performed after a minimum wait of 5 min. Diffusion studies were accepted only if inspired volume (Vi) was within 10% of the best FVC (atps). Acceptable breathholding time was between 8 to 12 s. Values of DLcosb (ml CO (sTPD]/min/mmHg) and Dl/Va (ml CO [sTPDj/min/mmHg/L [btpsJ) used were the means of 2 acceptable values within 7% of each other. The test results were expressed as measured values and as percentages of predicted using the normal values of Samet and coworkers (7). Chest Radiographs Chest radiographs taken in the posteroanterior projection were obtained on all employees. The radiographs were reviewed by 2 board-certified radiologists (B readers) using a modification of the 1LO U/C 1971 International Classification for Pneumo coniosis. The modification includes the addition of grading criteria for radiographic changes associated with asbestos exposure. Radiographs were interpreted independent ly, with random interspersion of control films and without the radiologist's knowl edge of the employee's work history. Any difference in interpretation was resolved by consensus reading by a third reader. Radio graphs that could not be interpreted be cause of poor quality were repeated. Environmental Measurements The company first began using vermiculite in their facility in 1957. Industrial hygiene sampling for airborne fibers using mem brane fillers at a sampling rate of 2 L/min was initiated in 1972. Particles with a length greater than 5 pm, a diameter less than 3 (im, and an aspect ratio of 3:1 or greater were counted as fibers. Before 1976, sampling was accomplished by industrial hygiene personnel following an employee with a sampling device, but after 1976, fiber levels were obtained by industrial breathing zone sampling. Exposure indexes expressed as fibers/ml were developed for each department, based on an 8-h time-weighted average (TWA). A separate index was developed before and up through 1973 and for the period beginning with and continuing after 1974. There was a substantial reduction in airborne fiber levels after the implementation of improved environmental controls in 1973 to 1974. The industrial hygiene values used to estimate the < 1973 exposure index per departmenr were mean fiber values for the years < 1973 or the mean values from the year industrial hygiene values were first available. The > 1974 exposure index was developed in a similar manner. The industrial hygiene measurements initiated in 1972 became more comprehensive in ensuing years. The < 1973 exposure index most likely underes timates prior employee fiber exposure. Collection filters were analyzed by polarized light microscopy with dispersion staining. Additionally, fiber analysis was done by scanning electron microscopy with energy dispersive X-ray analysis and trans mission electron microscopy with selected area electron diffraction. After reviewing the industrial hygiene and manufacturing process data, it was apparent that the employees could be divided into 3 main exposure groups repre senting 9 departments. Group I, with limited or no exposure to airborne fibers, included workers in the chemical process, research, and management departments. Industrial hygiene measurements indicated their exposures were similar to background levels for the local community. Group II Lad low-level fiber exposure and included central maintenance, packaging, and the warehouse workers. Group III, with highlevel fiber exposuie, included vermiculite expanders, plant maintenance, and the pilot plant. The 8-h TWA exposure indexes are summarized in table 1. The chemical processing facility that employs the majority of workers in the comparison group (Group I) was completed in 1969 and was located one-quarter mile from the vermiculite facility. Chemicals used in this facility were the same as those TABLE 1 DEPARTMENT FIBER EXPOSURE INDEXES BASED ON AN 6-HOUR TIME-WEIGHTED AVERAGE* < 1973 Exposure Index > 1974 Exposure Index Group 1 Chemical process Research Front office Group II Central maintenance Packaging Warehouse Group ill Vermiculite expanders Plant maintenance Pilot plant 0 049 0.049 0.049 0415 0.250 0.110 1.511 1.264 1.264 0.049 0 049 0.049 0131 0.031 0110 0.375 0.212 0.212 * Fiber eipoture index lot departments in study popvia lion Group I are controls. Group II are low fiber exposure departments, and Group ill are high fiber exposure depart menu Values are libervmi used in the vermiculite facility, except ver miculite was not used as an inert carrier. There was minimal rotation of job positions between the chemical and vermiculite facilities. The work areas with the highest airborne fiber exposure were the vermiculite ex panders area and the vermiculite railroad car and truck unloading areas. Fiber levels increased when there were more vermiculite expanders in operation and when Montana vermiculite ore was used rather than vermi culite from other sources. Fiber levels in the unexpanded vermiculite ore unloading area were recorded as high as 103 fibers/ml for a 5-min sampling period. During unloading of the ore, high levels of fiber dust were generated, but the peak levels rapidly de creased to concentrations less than 5 fi bers/ml by IS to 20 min. Exposure indexes were expressed in 3 ways: (/) cumulative fiber exposure (fiber/ ml-yr), (2) time period from first exposure (latency), and (J) exposure groups, ije. Groups I, II, and HI. Cumulative fiber ex posure for each individual employee was calculated from exposure values and length of employment in each particular depart ment. The exposure index was based on an 8-h TWA and a maximum of a 365-day work year. Extensive overtime had been scheduled at the facility, but more precise estimates of past total work days per year were not available. Employees with a cumu lative fiber exposure of less than 1 fiber/mlyr were found to have fiber exposure equiv alent to the community population exposed to ambient air. These employees acted as a comparison group for the exposed popula tion. Statistical Analysis Discriminant analysis was performed on the results of the questionnaire, physical exami nation, and radiographic data. The discrim inant analysis attempted to establish a rela tionship between a nominal dependent vari able (radiographic results) and independent variables (cumulative fiber exposure, smok ing in pack-years, and age). Spirometry and diffusion data were analyzed using analysis of covariance after adjusting for height and smoking in pack-years. The spirometry and diffusion data were also divided into smok ing and exposure groups and reanalyzed in a similar manner. Mean difference between smoking groups and between exposure groups within smoking groups were ana lyzed after adjustment for age and height. The association between the radiographic data and cumulative fiber exposure was examined using a pair-matched analysis. Each employee with radiographic findings consistent with commercial asbestos fiber exposure was matched by age with a second employee with a normal radiograph. A paired t test and a nonparametric test, Wilcoxon's signed-rank test, were performed on the difference in cumulative fiber exposure. 10002237 10CKEY. (ftOOKV JAAAtEk. ft 41 Results The mean age of all participants in the study was 37.5 yr (range, 19 to 66 yr). The ethnic distribution of participants was 496 (96.9%) white, 12 (2.3%) black, 2 (0.4%) Oriental. 1 (0.2%) American Indian, and 1 (0.2^0) Ameri can-born Hispanics. There were 480 males and 32 females. The high fiber exposure group (Group III) tended to be older than the other 2 exposure groups (p < 0.01). Of the entire group, 44.4% were current smokers, 20.3% were ex-smokers, and 35.3% were never smokers. There was no significant dif ference in smoking history between ex posure groups. Cumulative fiber expo sure was significantly higher in Group 111 than in Groups I and II (p < 0.01). Years of employment were significantly greater in exposure Group HI than in exposure Groups 1 and 11 (p < 0.05). These data are summarized in table 2. Questionnaire Data History of respiratory illness with time lost from work, chest injury or opera tion, phlegm production, wheezing with colds, and previous fibrous miner al exposures in industry or hobbies were not significantly related to age, smoking in pack-years, or cumulative fiber exposure. There was a significant association between history of pneu monia, confirmed by a physician, and age (p < 0.05). The prevalence of chron ic cough, defined as cough 4 to 6 times/ day, 4 or more days/week for 3 consec utive months for at least 2 yr, was sig nificantly related to smoking in packyears (p < 0.05). Apart from colds, wheezing or whistling present for at least 2 yr, and without associated short ness of breath with wheezing or current history of asthma confirmed by a phy sician, was significantly related to smoking in pack-years (p < 0.05). Chron ic airway obstruction, defined as wheez ing most days or nights and/or Grade 3 dyspnea (stop for breath when walking at your own pace on the level), and/or FEV,/FVC% 60% or less, was signifi cantly related to smoking in pack-years (p < 0.05). The prevalence of 2 or more attacks of shortness of breath with wheezing without a current history of asthma confirmed by a physician was significantly related to smoking in pack-years (p < 0.05) and strongly re lated to cumulative fiber exposure (p < 0.1). Discriminant analyses indicated short ness of breath Grade I (shortness of TABLE 2 AGE. SMOKING HISTORY. CUMULATIVE FIBER EXPOSURE. AND YEARS OF EMPLOYMENT By EXPOSURE GROUP Group 1 Group II Group III - NS EX CS NS EX CS NS EX CS Number ot Employees 49 Age. yr Mean SE 338 17 Pack-years Mean SE 0 Fiber/ml-yr Mean SE 0.35 0.67 Years of employment Mean 6.6 SE 1.1 22 39.5 2.5 13.1 2.9 0.57 1.01 11.3 1.6 41 39.1 1.8 18.6 2.1 0.50 0.74 10.5 1.2 63 34.5 1.5 0 1.15 0.60 8.4 1.0 36 107 399 34.9 2.0 1.1 14.2 2.3 1.56 0.79 18.7 1.3 0.97 0.46 13.3 8.9 1.3 0.7 69 402 1.4 0 6.51 0.57 12.2 0.9 46 42.8 1.7 14.3 2.0 7.55 0.70 13.0 1.1 79 375 1.3 17.8 1.5 6.05 0.53 10.7 09 Otfmition of obbrovitttons: NS * tmot tmofctr, EX s ta-smoMf. CS * current smofctt. breath when hurrying on the level or walking up a slight hill) and Grade II (walk slower than persons your own age on the level because of breathlessness) were significantly related (p < 0.05) to both smoking history in pack-years and cumulative fiber exposure. The preva lence of pleuritic chest pain lasting 6 h or more with physician evaluation (fig ure 1) was significantly related to cumulative fiber exposure (p < 0.05). Physical Examination The physical examination finding of crackles heard on auscultation of the chest was related to cigarette smoking in pack-years (p < 0.05), whereas club bing of the nails was positively related to age (p < 0.05). Pulmonary Function Tests There were no differences in mean spi rometry values or percent predicted values in relation to cumulative fiber exposure, time since first exposure, or exposure group. Predicted values for blacks were considered to be normally 10% lower than a corresponding white population. The prevalence of "restric tive lung defect" defined as FEV,/FVC ratio of equal to or greater than 70% and FVC less than 80% predicted was not shown to be significantly related to cumulative fiber exposure. The FEV,, FEV ,/FVC%, and FEF,,.,, showed significant differences only between smoking categories (p < 0.01). No sig nificant difference in FVC was found for the various smoking categories. There were no differences in mean Dicosb or Dl/Va values or predicted values in relationship to cumulative fiber exposure, time since first expo sure, or job category. A significantly low DLcosb was noted in smokers (p < 0.01). 15% -- 10%- HISTORY OF PLEURITIC CHEST PAW 1 ! i--i r--i mfln ttNS la CS N 49 41 Group I )NS (K CS ) >07 Group I Job Category n 7fns fa CS M 44 Grou> Fig 1. Percentage of workers reporting pleuritic chest pain by exposure group and by cigarette smoking history Discriminant analysis Indicates a significant association between cumulative fiber exposure and history of chest pain (p < 0 05) For definition of groups, see table 1 Chest Radiographs The results of the radiographic survey included all employees with available interpretable films; 501 of 512 (97.9%) employees were reviewed. There were 479 (95.6%) with no significant radiographic changes, 11 (2.2%) with costophrenic angle blunting only, 10 (2%) with significant pleural changes (thick ening, plaques, and/or calcifications), and 1 (0.2%) individual showed paren chymal changes of bilateral, small, ir regular opacities (table 3). The mean cumulative fiber exposure for the latter 11 employees (10 with pleural and 1 with parenchymal change) was 12.07 (range, 0.01 to 39.9 fiber/ml-yr). The 10002238 AMI CHANCES AFTER EXPOSURE TO VfRMICULIIE V00 TABLE 3 CHEST RADIOGRAPH CHANGES TABLE s RADIOGRAPHIC CHANGES BY EXPOSURE GROUP'* Type Grade* Group I G'Oup II Group HI Costophrenic angle blunting only. 11 employees Pleural changes. 10 employees Bilateral pleural thickening Bilateral pleural plaques Bilateral pleural thickening Lett pleural thickening Bilateral pleural thickening Bilaleral pleural plaques A/1 1/1 C/2 A/1 B/1 1/1 Results (n) f/.) Age- (r. (V.) Age^- <> <%) Age- Normal 104 97.2 36 4 * 10 8 195 96 1 34.9 * 11.9 179 94.2 39 1 x 120 Costophrenic angle blunting only 1 1.0 55.5 5 2.5 49 7 x 13.0 5 2.6 41.0 x 8.2 Pleurallparenchymat changes* 2 1.9 48 7 st 8.3 3 1.5 52.5 x 3.8 6 3.2 55.1 x 4.0 Combined changes 3 2.6 51.0 8 3.9 50.7 11 SB 46.7 Total 107 204 190 Bilateral pleural thickening Left pleural plaque Lett pleural thickening Left pleural plaque B/2 1/1 * l s control roup, II * low fiber exposure group. Ill * iwgn fiber exposure group. 1 Values ere mean * SO A/1 1 One employee with bilateral small irregular s-type opacewi This employee was in exposure Group III. 0/1 Right pleural thickening Right pleural plaque Unilateral pleural thickening Letl pleural calcilication Bilateral pleural thickening Bilateral pleural plaques Right diaphragm calcification Parenchymal changes Bilateral small, irregular opacities Type s C/1 1/1 A/1 Grade 1 B/1 1/1 Grade 2 Perfusion 1/1 ees with greater than 10 fiber/ml-yr exposure, 12.5 % had costophrenic angle blunting or pleural/parenchymal changes (table 4). Employees with greater than 10 yr of employment from initial employment in exposure Groups II or HI had an increased prevalence of radiographic changes. Of 48 employees difference in cumulative fiber exposure (table 6) was noted between the 22 employees with abnormal radiographs and the control group (p < 0.041). The results of the nonparametric analysis (Wilcoxon's sign-rank test) indicated a strong trend toward an association be tween cumulative fiber exposure and * Grade of pleural thickening width. A. < 5 mm; 6.6*10 mm; C. > 10 mm el widest pan of pleural shadow Grade of pleural thickening by extent: 1 * definite pleural thick ening m one or more places, such that the total length d> not exceed one half of the projection of one lateral wah; 2 * pleural thickening greater than Grade 1 Certainty of plague 0/1, possible present; 1/0. probably present, in, definitely present. Grade of pleural calcification by total lengih: Grade 1 <20 mm, Grade 2 s 20-100 mm; Grade }s > 100 mm. mean cumulative fiber exposure for employees with costophrenic angle blunting was 5.4 (range, 0.2 to 27.5 fiber/ml-yr). There was an increased prevalence of radiographic changes in employees with 1 to 10 and greater than 10 fiber/ml-yr cumulative exposure in comparison with the control group. Of 48 employ with 20 yr or more of employment, 11.1% had costophrenic angle blunting or pleural/parenchymal changes. Em ployees ever employed in the low (Group 11) or high (Group III) fiber exposure groups had a higher prevalence of radi ographic changes than did the control group (table 5). The employees with radiographic changes showing costo phrenic angle change only or pleural changes were older (p < 0.01) than employees with normal radiographs. Because age could act as a confounding factor in the observed radiographic changes, an age-matched control study was conducted. Each of 22 employees with an abnormal chest radiograph was matched by age to an employee with a normal chest radiograph. A significant significant pleural/parenchymal abnor malities (p < 0.098). An association was noted using discriminate analysis be tween cumulative fiber exposure (p < 0.05), smoking in pack-years (p < 0.05), and radiographic changes. Discussion This cross-sectional epidemiologic study revealed a number of findings showing a significant association with cumula tive fiber exposure: dyspnea on exer tion, pleuritic chest pain, and pleural changes on chest radiograph. Further more, an apparent dose-response rela tionship between cumulative fiber ex posure and radiographic changes was suggested. The occurrence of dyspnea among workers occupationally exposed to as table * RADIOGRAPHIC CHANGES BY CUMULATIVE FIBER EXPOSURE bestos has been documented in previ ous studies (8-10). The presence of dyspnea on exertion was associated Fiber/ml-yr with cumulative tremolite fiber expo Results Control * (n) (V.) 1-10 <") <%> > 10 <") (V.) sure This is in the face of a relatively low cumulative fiber exposure and a relatively short latency period for the Normal 247 97.6 190 95.0 42 67.5 work force The association among cig Costophrenic angle blunting only Pleural/parenchymal changest Combined changes 4 1.6 5 2.5 2 4.2 2 0.9 5 2.5 4 84 6 24 10 5.0 6 12.5 arette smoking and the presence of chronic cough, wheezing, and spirometric evidence of air-flow obstruction is expected and provides validation of the questionnaire Total 253 200 48 a Less than 1 fiber/ml-yr toiai exposure 1 One employee with bilateral smn irteguiar s-type opacities This employee's cumulative e posure was 39 (iper/mi-yr The observed association between the presence of pleuritic chest pain and cumulative tremolite fiber exposure is an interesting finding. The presence of 10002239 4 LOCH It. ROOK&. MMttk. tl *4. TABLE 6 DIFFERENCES IN EXPOSURE (FIBER/ML-YR) PAIR MATCHED ANALYSIS (NORMAL VERSUS ABNORMAL RADIOGRAPH) AND PARAMETRIC ANALYSIS (PAIRED f TEST)* Exposure Fibei/ml-yt Coslopltremc Angle Blunting Only (n s n pairs) Pleural/Parenchymal Change^ (n = f f pairs) Both (n = 22 pairs) Mean difference SD P 3.0 9.7 0.232 6.6 14.7 0.062 6.6 12.5 0.041 * Cases were ach matched by age with employees who had normal radiographs 1 One employee with bilateral small irregular s-type opacities pleuritic chest pain was documented if the employee was evaluated by a physi cian, and no specific medical or surgi cal cause for chest pain was identified. Pleuritic chest pain was examined be cause it can be a clinical manifestation of inflammation involving the parietal pleura. Irritation of pleura surface by inhaled fibers is a postulated mecha nism for the occurrence of pleural effu sion and plaque formation. A recent hypothesis suggests that the fibers reach the parietal pleura by lymphatic drainage. Macrophages within the pari etal pleural engulf fibers and stimulate submesothelial fibroblasts with subse quent pleura plaque formation (11, 12). In either case, the resulting inflamma tion from pleural irritation may cause pleuritic chest pain, possible bloody pleural effusions and, eventually, pleu ral thickening or pleural plaque forma tion. The pleural changes noted on chest radiographs of workers exposed to air borne fibrous tremolite are consistent with the pleural changes seen with com mercial asbestos exposure (11, 13-16). In the present study, the prevalence of these pleural changes increased with duration of exposure, time since initial exposure, and high exposure group, and were dose related. The absence of sig nificant parenchymal changes (ie., asbesiosis-fibrosis) and a higher percen tage of pleural changes most likely reflect the short interval since initial exposure and the low cumulative fiber exposure. In any event, the presence of a greater prevalence of pleural disease among the population exposed to vermiculite contaminated with fibrous tremolite is supported by the following observations: (I) cluster of 12 cases of benign pleural effusion, (2) higher prevalence of pleural changes on chest radiographs, (i) age-matched control study revealing greater tremolite fiber exposure among workers with pleural changes on chest radiographs, and (l) higher prevalence of pleuritic chest pain in exposed employees. The lack of association between sim ple spirometric and DLcosb measure ments and fiber exposure most likely reflects the low cumulative fiber expo sure and short interval period. Simple spirometric measurements have been shown to be sensitive indicators of the toxic effects of cumulative asbestos exposure. The DLcosb changes are not asbestos-dose related and are less sensi tive than spirometry (17-19). The level of cumulative fiber exposure needed to cause a change in spirometric values is greater than the exposure levels reported in the present study. Weill and col leagues (18) reported decrease in lung function after 100 mppcf-year dust ex posure, while Becklake and colleagues (19) showed an effect at a cumulative dust exposure index of 10 to 100 mppcfyear. Berry and Lewinsohn demonstrated a 12.1% reduction for FEV, and 10.6% reduction for FVC per 100 fiber/ccyears (20). The cumulative fiber exposure in this study was low compared with that in other studies. Only 9.6% of the em ployees had greater than 10 fiber/ml-yr exposure; 10.7% had been employed 20 yr or more since initial exposure. The highest cumulative fiber exposure for any employee was 39 fiber/ml-yr. It is likely that the exposure level reported in this study underestimates the actual cumulative fiber exposure. No indus trial hygiene data were available before 1972; the lower fiber values after 1974 reflect improved environmental con trols within the facility. Additionally, personal sampling, which more ade quately reflects individual exposure, was not introduced until 1976. Nonmalignant pleural changes asso ciated with asbestos exposure include pleural plaques with or without calcifi cation, bilateral pleural thickening, progressive pleural fibrosis (21-23), pleural effusions, and pleural thicken ing, with costophrenic angle involve ment or costophrenic angle blunting alone as residual findings of a previous pleural effusion (15, 24). The preva lence of pleural changes commonly exceeds that of asbestos-related inter stitial fibrosis (25, 26). The prevalence of pleural plaques, calcification, and thickening is related to duration of asbestos exposure, time since onset of exposure (10, 17, 27-29), and possible cumulative fiber exposure (30, 31). The occurrence of pleural effusions may be an early manifestation of asbestos exposure (15, 32). The majority of asbestos-related pleural changes have a latency period of more than 20 yr from initial expo sure (28). However, pleural changes are seen in workers with short durations of asbestos exposure. Studies of asbestos insulation workers (21), Navy shipyard workers (26), household contacts of amosite asbestos workers (10), and Quebec chrysotile mine and mill work ers (24) support this observation. Benign asbestos-related pleural effu sions are now recognized to cause pleu ral change including greater than 50% rate of residual pleural thickening and greater than 90% of residual costo phrenic angle blunting. The occurrence of asbestos-induced pleural effusion appears to be dose related and repre sents, in one study, the most common manifestation of asbestos exposure in the initial 20 yr since first exposure (15). In the current study, 7 employees were still employed with previously docu mented benign pleural effusions. Of these 7, four had residual pleural changes, including 2 with unilateral costophrenic angle blunting, 1 with bilateral costophrenic angle blunting, and 1 with bilateral pleural thickening and pleural plaques. These 4 were in cluded in the total of 22 employees with pleural changes noted in chest radio graphs. The prevalence of pleural thickening and pleural calcification in a rural pop ulation of midwestern dairy farmers without occupational exposure to fibrogenic dust was found to be 0.9% and 0.0%, respectively. The prevalence in a New Jersey urban population was 1.2% for pleural thickening and 0.0% 10002240 rfT CHAMGE5 am* EXPOSURE TO VERMICUUll ,or pleural calcification (21). The pre local deposits of minerals containing dictive value for bilateral pleural thick fibrous tremolite are used to make ening as an indicator of previous possi whitewash and stucco, revealed an in ble asbestos exposure with exclusion of creased prevalence of pleural thicken known medical or surgical etiologies ing and calcification and interstitial was recently found to be 81% (23). pulmonary fibrosis (39). A careful review of all chemical and Animal studies in rats with fibrous physical agents used at the plant facility tremolite administered by intraperito- failed to identify any substance known neal injection have shown it to be both to be associated with pleural radio- fibrogenic and tumorgenic (40). Al graphic changes. A few of the chemi though reported animal studies with cals could cause acute pulmonary in vermiculite are limited, 2 studies on the jury after exposure to high concentra use of South African vermiculite (gen tions. The study and control populations erally believed to be free of tremolite) were evenly matched for exposure his did not show Fibrogenic or tumorgenic tory, except for the presence or absence changes after intrapleural or intratra of exposure to vermiculite contami cheal injection (41, 42). nated with tremolite. Exposure to fibrous tremolite that There are over 150 minerals that exist contaminates certain vermiculite ores in fibrous form and that are generally can cause pulmonary abnormalities. expected to contain fibrous minerals Identification of other minerals that (3) . The amphibole tremolite is a sili have similar fiber contamination and cate mineral found in commercial talc recognition of the potential health deposits from New York state and in implications are critical issues in occu vermiculite deposits within the United pational and pulmonary medicine. Just States (4, 33). The size and shape of tre as important is the recognition that not molite after undergoing crushing and all amphiboles exist in fibrous forms. grinding in a milling process is largely The biologic activity of tremolite that dependent on the original crystalline forms cleavage fragments with low as structure of the mineral. Tremolite can pect ratios may be different from fi occur in a fibrous form with high as brous tremolite with high aspect ratios. pect ratios identical to commercial as There are enormous health, economic, bestos fibers or as cleavage fragments and regulatory implications at stake in with lower aspect ratios (34, 35). The these issues. commercial talc deposits in New York state contain varying amounts of tre molite and anthophyllite, both as cleav age fragments and as true fibers (33). The vermiculite deposits from Mon Acknowledgment The writers gratefully acknowledge the kind assistance of Dr. Gary Liss, Dr. James Donovan, Dr. James Blanchard, Joan Beck tana contain a fibrous form of tremo ham, Bonnie Mosley, and Carol Perry for lite, while the vermiculite deposits in technical assistance, and Mary Ann Brock Virginia and South Carolina contain a man. Juanita Fogle, Mary Campbell, and type of tremolite that forms predomi Mary Bishop for typing the manuscript. nantly cleavage fragments when milled (4) . Vermiculite itself does not exist in References a fibrous form, but plates of vermicu 1. Otis LM. Vermiculite In: Mineral facts lite can roll up on themselves as a scroll and problems. Washington, D.C.: U.S. De and resemble fibers (36). partment of the Interior, Bureau of Mines, Tremolite is not mined or used as a I960 (bulletin no. $85). commercial asbestos. Therefore, no 2. Meisinger AC. Vermiculite In: Mineral working populations were exposed only to tremolite. On the other hand, natural exposure to tremolite does occur in cer tain rural populations. Pleural changes, including plaques and calcifications, have been reported in rural populations of Czechoslovakia, Bulgaria, Finland, and Greece. These changes have been attributed to tremolite and anthophyllite fibers deposited in the local soil (37, 38). A recent epidemiologic report on 7,000 inhabitants around the town of facts and problems. Washington, D.C.: U.S. Department of the Interior, Bureau of Mines, 1980 (bulletin no. 671). 3. Bank W. Asbestiform and/or fibrous minerals in mines, mills, and quarries. Washington, D.C.: U.S. Department of Labor, Safety and Health Administration, 1980 (publication no. 1980-603-102/34). 4. Lockey JE, Moatamed F. Health impli cations of nonasbestos fibers. In: Gee JBL, ed. Contemporary issues in pulmonary dis ease Vol 1: current topics in occupational lung disease New York: Churchill Living Cermik in southeast Turkey, where ston, Inc., (in press). 5. Ferris BC. Epidemiology standardiza tion project. Am Rev Respir Dis 1978; 118: 10. 6. Knudson RJ, Slatin RC. Lebowitz MD, Burrows B. The maximal expiratory flowvolume curve. Am Rev Respir Dis 1976; 113:587. 7. Samet JM, Epler GR. Gaensler ES, Rosner B. Absence of synergism between exposure to asbestos and cigarette smoking in asbestosis. Am Rev Respir Dis 1979; 120: 75. 8. Selikoff IJ. Churg J, Hammond EC. The occurrence of asbestosis among insula tion workers in the United States. Ann NY Acad Sci 1965; 132:139-55. 9. Murphy RLH, Ferris BG, Burgess WA, Worchester J, Gaensler EA. Effect of low concentration of asbestos: clinical, environ mental, radiologic, and epidemiologic ob servations in shipyard pipe coverers and controls. N Engl J Med 1971; 285:1271-8. 10. Anderson HA. Asbestosis among household contacts of asbestos factory workers. Ann NY Acad Sci 1979; 330:38799. 11. Casey KR, Rom WN, Moatamed F. Asbes tos-related disorders. In: Brooks SM, Lockey JE, Khoury P, eds. Clinics in chest medicine, occupa tional lung disease. 1. Philadelphia: W.B. Saun ders Co.. 1981; 2:179-202. 12. Hillerdal G. The pathogenesis of pleural ptacques and pulmonary asbestosis: possibilities and impossibilities. Eur J Respir Dis 1980; 61: 129-38. 13. Becklake MR. Asbestos-related diseases of the lung and other organs: their epidemiology and applications for clinical practice. Am Rev Respir Dis 1976: 1)4:187-227. 14. Murphy RLH, Gaensler EA. Ferris BG, Fitzgerald M, Solliday N, Morrisey W. Diagnosis of "asbestosis"; observations from a longitudinal survey of shipyard pipe covers. Am J Med 1978; 65:488-98. 15. Epler GR, McLoud TC, Gaensler EA. Pre valence and incidence of benign asbestos pleural effusion in a working population. JAMA 1982; 247:617-22. 16. Becklake MR. Asbestos-related diseases of the lungs and pleura. Am Rev Respir Dis 1982; 126:187-94. 17. Jones RN, Diem JE, Glindmeyer H. Weill H. Progression of asbestos radiographic ab normalities: relationships to estimates of dust exposure and annual decline in lung function. In: Wagner JC, ed. Biological effects of mineral fibres. Vol 2. Lyon, France: International Agency for Research on Cancer, 1980:537-43. 18. Weill H, Ziskind MM, Waggenspack C, Rossiter CE. Lung function consequences of dust exposure in asbestos cement manufacturing plants. Arch Environ Health 1975; 30:86. 19. Becklake MR, Fournier-Massey G, Rossiter CE, McDonald JC. Lung function in chrysotile asbestos mine and mill workers of Quebec. Arch Environ Health 1972; 24:401. 20. Berry G, Lewinsohn HC. Dose-response relationships for asbestos-related disease: impli cations for hygiene sundards. Part 1: morbidity. Ann NY Acad Sci 1979; 330:185-94. 10002241 21. Anderson HA, Selikoff IJ. Pleural reactions to environmental agents. Environmental pharma cology of the lung. Fed Proc 1978; 37:2496-500. 22 Hillerda! G. Non-malignant asbestos pleural disease. Thorax 1981; 36.669-75. 23. Albelda SM. Epstein DM. Gefter WB. Miller WT. Pleural thickening: its significance and relationship to asbestos dust exposure. Am Rev Respir Dis 1982; 126:621-4. 24. Rossiter CE. Bristol LJ. Cartier PH, et at. Radiographic changes in chrysolite asbestos mine and mill workers in Quebec. Arch Environ Health 1972; 24:388. 25. Harris PG, Mackenzie FAF, Sheers G, Kemp 3H, Oliver TP, Wright DS. Radiological survey of men exposed to asbestos in naval dock yards. Br J Ind Med 1972; 29:274-9. 26. Felton JS. Radiographic search for asbestosrelated disease in a naval shipyard. Ann NY Acad Sci 1979; 330:341-52. 27. Sheers G, Templeton AR. Effects of asbes tos in dockyard workers. Br Med J 1968; 3:574-9. 28. Selikoff IJ. Hammond EC. Asbestos-assoc iated diseases in United States shipyards. CaCancer J Clin 1978; 28:87-99. 29. Polakoff PL, Horn BR, Scherer OR. Pre valence of radiographic abnormalities among northern California shipyard workers. Ann NY Acad Sci 1979. 330:333-9. 30. Lacquet LM, van der Linden L. Roentgenographic lung changes, asbestosis and mortality in a Belgian asbestos-cement factory In: Wagner JC, ed. Biological effects of mineral fibres Vol. 2. Lyon, France: International Agency for Re search on Cancer. 1980:783-93. 31. Hourihane DO'B, Lessof L, Richardson PC. Hyaline and calcified pleural placques as an index of exposure to asbestos. A study of radio logical and pathological features of 100 cases with a consideration of epidemiology. Br Med J 1966; 1:1069-74. 32. Sheers G. Asbestos-associated disease in employees of Devonpon dockyard. Ann NY Acad Sci 1979; 330:281-7. 33. Dement JM. Zumwalde RD. Occupational exposures to talc containing asbestiform miner als. In: Lemen RA, Dement JM, eds. Dusts and disease. Park Forest South, Illinois: Pathotox Publishers, Inc., 1979:287-305. 34. Campbell W'J, Blake RL, Brown LL. Cather EE, Sjoberg TJ. Selected silicate minerals and their asbestiform varieties. Mineralogical defini tion and idenlincation-characterization. Wash ington, D.C.: U.S. Department of the Interior, Bureau of Mines. 1977 (information circular no. 8751). 35. Unger AM, Rohl AN, Woirf MS. Selikoff IJ. Asbestos, Fibrous minerals, and acicular cleavage fragments: nomenclature and biological properties. In: Lemen RA. Dement JM, eds. Dusts and disease. Park Forest South. Illinois Pathotox Publishers. Inc.. 1979.1-22. 36. Chatfield FJ. Lewis GM. Examination of vermiculite for the presence of asbestos Fibers. Mississauga, Ontario. Canada: Ontario Research Foundation. 1979 (report no. 22056-2). 37. Wagner JC, Elmes PC. The mineral Fibre problem. In: McDonald JC. ed, Recent advances in occupational health. Edinburgh: Churchill Livingston. Inc., 1981:1-14. 38. Bazas T, Bazas B, Ritas D, Gilson J, McDonald J. Pleural calcification in northwest Greece. Uncet 1981; 2:254. 39. Yazicioglu S, llcayto R. Balci K, Sayli BS. Yorulmaz B. Pleural calcification, pleural meso thelioma, and bronchial cancers caused by tremolite dust. Thorax 1980; 35:564-9. 40. Smith WE, Hubert DD, Sobel JH, Marquet E. Biologic tests of tremolite in hamsters. In: Lemen RA, Dement JM, eds. Dusts and disease. Park Forest South, Illinois: Pathotox Publishers, Inc, 1979:335-9. 41. Goldstein B. Rendall REG. The relative toxicities of the main classes of minerals. In: Pneu moconiosis: Proceedings of the International Conference. Johannesburg, 1970:2129. 42. Hunter B, Thomson C. Evaluation of the lumorgenic potential of vermiculite by intra pleural injection in rats. Br J Ind Med 1973; 30: 167. 10002242