PULMONARY ASBESTOSIS is a chronic scarring disease that is recognized radiographically by fine irregular opacities. The International Labour Office (ILO) used the expanded scale of 1980 to determine that the spectrum of severity extends from none (i.e., 0/0) to severe (i.e., 3/+). There is an impression that pulmonary asbestosis is a restrictive disease, as defined physiologically.[2-8] It appears that this concept was based on the observation that vital capacities were reduced, and confirmation of pulmonary asbestosis was supposedly con firmed by the finding of decreased total lung capacities, as measured by gas dilution. The results of numerous studies have shown that airways obstruction is associated with pulmonary asbestosis[6-10] and pleural asbestosis--even in the absence of cigarette smoking.[7-12]
The question remains whether asbestosis, throughout its course, is an obstructive disease or whether it is obstructive initially but becomes restrictive in advanced stages. I sought an answer by comparing the patterns of physiological impairment and clinical findings in two subgroups of men selected from 12,856 asbestos-exposed American workers who had been studied, with standard protocol, by one medical team for asbestosis and other pulmonary diseases. The first subgroup had radiographically advanced pulmonary asbestosis (RAA), which is defined as an ILO profusion of irregular opacities greater than 2/2. The second subgroup was defined by a total lung capacity (TLC) and a forced vital capacity (FVC)--both below 80% of predicted--thus meeting the physiologic criteria for restrictive disease (PRD).[13-16]
Approximately 3,445 of 12,856 (27%) asbestos-exposed male workers had radiographic signs of pulmonary asbestosis.[10-12] Radiographically advanced asbestosis was found in 88 workers; this condition was defined by a profusion of irregular opacities greater than 2/2. Given that 3 of the 88 workers had reduced lung volumes from lung resections, they were not analyzed further. The physiological measurements of the 85 men were adjusted for height, age, and duration of cigarette smoking. They were compared with 52 men selected from the 12,856 men for whom both TLC and FVC were below 80% of predicted values (i.e., physiological restrictive disease).[13-16] The groups did not overlap.
Recruitment. The 12,856 asbestos workers comprised a national sample of male construction workers from 69 sites across the United States, from Los Angeles, California, to Jacksonville, Florida. Most were active or retired members of either the United Association of Journeymen and Apprentices of Plumbers and Pipefitters of the United States and Canada or the International Brotherhood of Boilermakers, Blacksmiths, Welders, Forgers, and Iron Ship Makers. Others were from metal trades or chemical unions. All had experienced occupational asbestos exposure for at least 5 y; in addition, exposure had begun at least 15 y earlier. Union business agents sent letters of invitation to members explaining the asbestos and pulmonary disease examination. Members who volunteered made appointments with our office. The study was approved by the Human Subjects Institutional Review Board of the University of Southern California. Informed consent was obtained from each worker.
All men completed questionnaires and had chest physical examinations, chest radiographs, spirometry, and measurement of alveolar carbon monoxide by one medical team. Questionnaires were completed by trained interviewers and included an occupational history for the nature and duration of asbestos exposure and medical, pulmonary, and cardiovascular histories. The interviewers applied the criteria to define chronic bronchitis and asthma on the basis of DLD-78. Chronic bronchitis was diagnosed when phlegm had been produced most days of the week for at least 3 mo/y for 2 successive years. Asthma was diagnosed by episodic wheezing, which was relieved spontaneously or with medication with "normal breathing" between attacks. Chests were examined for size; shape; deformities; and the presence of normal, decreased, or absent breath sounds, rales, and wheezing. Extremities were examined for cyanosis, clubbing, and edema.
Posteroanterior and lateral chest radiographs were obtained on standard 35.6- x 42.3-cm films with a Picker portable x-ray machine at a 173-cm target distance (KV 120-130--a suitable grid) and a Kodak processor. To insure that films were exposed at full inspiration for the measurement of TLC, technicians repeated the PA radiograph if the diaphragm was above the 9th posterior intercostal space. Unsatisfactory and borderline x-rays were repeated until film quality met ILO criteria. One experienced physician (KHK), who used the 1980 ILO criteria for pneumoconiosis, read the films for profusion of irregular opacities typical of pulmonary asbestosis and for type and extent of pleural disease and other abnormalities. Periodically during the study, blinded sets of 100 films were reread for temporal variation. Such variation was confined to ILO profusions of 0/1, 1/0, and 1/1; was randomly above or below original readings; and never exceeded 5% of films. Emphysema was diagnosed by the presence of three or more of the following radiographic criteria: (1) hyperlucent lungs, (2) low flat diaphragms, (3) bullae on the posteroanterior film on the lateral view, and (4) low flat diaphragms and a retrosternal space wider than 2.5 cm.
One experienced technician used planimetry to measure the areas of the right and left lung, excluding the heart on the posteroanterior, and lateral radiographic images and the lateral area, including the heart. Total lung capacity (TLC) was calculated from the regression equation developed by Harris et al. Residual volume (RV) was obtained by subtracting FVC from spirometry from TLC.
Spirometry was done on a rolling-seal spirometer (Ohio 822), with the subjects standing and using a nose clip; they also followed American Thoracic Society (ATS) recommendations, including reproducible efforts. Spirometers were calibrated repeatedly for volume and clock speed at each examination site. Recording expiratory flow until it ceased or for 10 s guaranteed complete emptying of the lungs. Technicians corrected pulmonary-function values to BTPS, and they were compared with height and age-adjusted predicted values; therefore, pulmonary-function values were expressed as percentage of predicted. In addition, I used population-based regression equations[13,21] to adjust for current and excigarette smoking, the mean percentage predicted of forced expiratory volume in 1 s ([FEV.sub.1.0]), flows, and TLC. The equations used follow: ([FEV.sub.1.0] = -4.908 + 0.0566 Ht cm. - 0.0233 age - 0.0094 duration of smoking; Log [FEF.sub.25-75] = 0.3005 + 0.0081 Ht cm. - 0.0096 age - 0.0052 duration of smoking;. Log [FEF.sub.75-85] = -1.0752 - 0.0120 Ht cm. - 0.0222 age - 0.0112 duration of smoking; and L = 0.0971 + 0.0156 Ht cm. - 0.001 7 Wt kg. + 0.0014 duration of smoking.) I made no adjustment for duration of smoking for FVC. The RV was calculated as TLC minus FVC.
Questionnaires, physical examinations, and ILO readings were recorded in an optically coded format and machine read into a microcomputer. Data were analyzed with Stata statistical programs (Stata Corporation [College Station, Texas]) following transfer to an IBM-compatible computer. I compared the means of pulmonary functions--after adjusting for height, age, and duration of smoking--by t test for unequal-sized groups. I considered a p value less than .05 to be statistically significant.
Radiographically advanced asbestosis was rare; 85 (0.66%) of the 12,856 men studied had it, and 2.5% of this small group had asbestosis. The 85 men were 15 y older, had 8 more years of asbestos exposure, and had smoked for 12 y longer than the averaged study-population. Thirty-nine of the RAA group were exsmokers, and only 4 men had never smoked cigarettes. Fifty-eight percent of the men had asthma or chronic bronchitis, as evidenced by their medical histories. Twenty-seven percent had fine rales, and 23.5% had clubbed terminal digits. Their average profusion of irregular opacities was 3/2. Pleural plaques or thickening was present in 33% of the men. only 4 had emphysema. Their mean TLC was 7.45 I ([+ or -] 1.08 I standard deviation [SD]) that, as percentage of predicted, was 105.5 [+ or -] 14.4 I. Mean FVC was reduced to 82.6 [+ or -] 1 7.4% predicted, it ranged from 32% to 127% predicted, and RV/TLC (times 100)was elevated to a mean of 54.4 [+ or -] 9.2. Expiratory airflows were decreased significantly, and was 65.5 [+ or -] 13.3, compared with a predicted [FEV.sub..01]/FVC of 73.7 [+ or -] 10.5. Six men had severe airway obstruction (i.e., [FEV.sub.1.0]/FVC [is less than] 50), whereas only 6 men had no obstruction. No TLC was less than 80% predicted; therefore, restrictive disease was absent from the group.
The second group of 52 men (i.e., 0.4% of the total men) had reduced TLC, which indicated restrictive disease. They were 8.4 y younger, had smoked less, and had 3 y less asbestos exposure than those with severe asbestosis (Table 1). Their FVC was reduced to 61.5 [+ or -] 14.1% predicted, and TLC was reduced to 72.6 [+ or -] 6.2% predicted. Thus, their volumes were significantly lower than those of the RAA group (p [is less than] .0001). Only 50% of the second group of 52 men had radiographic pulmonary asbestosis, which was of low average ILO profusion (i.e., 1/0). Pleural asbestos disease was present in 27% of these men and occurred only slightly less frequently than in the 85 men with RAA. Although the 52 men had restrictive disease, small-airways obstruction was evidenced by reduced mid-flows (Table 1).
Table 1.--Demographic Data, Prevalences of Airways Disease, and Asbestosis Flows and Lung Volumes in Workers with Advanced Asbestosis(*)
Notes: TLC = total lung capacity, FVC = forced vital capacity, FEF: forced expiratory flow, [FEV.sub.1.0] = forced expiratory volume in 1 second, RV = residual volume, and NS = not significant.
(*) International Labour Office (ILO) category > 2/2.
([dagger]) p values obtained by t test for unequally sized groups.
([double dagger]) Compared with workers with advanced asbestosis.
([sections]) p < .05, compared with referent population.
(#) p < .0001.
Radiographically advanced asbestosis and physiological restrictive disease were mutually exclusive in appropriately selected subgroups of 12,856 American workers who had previous exposure to asbestos for 20+ y. Men with RAA, therefore, had normal total lung capacities, and none had a reduced TLC. Their RV/TLCs were elevated, and, therefore, reductions in vital capacity were accompanied by reciprocal increases in residual volume. The body plethysmographic and the radiographic methods measure the entire TLC, rather than a partial TLC. In contrast, gas-dilution measurements for TLC fail to measure the gas trapped behind obstructed airways in asbestosis--as is also the case in emphysema.[12,15] Men with advanced asbestosis had airway obstruction, which was evidenced by a reduced [FEV.sub.1.0]/FVC and air trapping, and elevated RV/TLC, the effect of which was a reduction in VC. Accompanying these findings were excessive frequencies of chronic bronchitis and asthma. A question comes to mind: How are these physiologic changes best explained?
Could concomitant emphysema have effectively balanced the "lung shrinkage" from asbestosis, resulting in normal TLCs? Given that emphysema was diagnosed radiographically in only 4 men with RAA, it is highly unlikely that the other 81 men had radiographically undetectable emphysema sufficiently severe to overcome the postulated "shrinkage" of asbestosis. However, effects of smoking on lung volume could still be opposing volume loss inasmuch as only 4 of the 85 men with advanced asbestosis had never smoked. To find 20 men with RAA who had never smoked, I would have to study 64,280--five times as many men as were studied here; this is impractical, and ultimate resolution of this question is, therefore, impossible. Advanced asbestosis occurs mainly in current smokers or exsmokers.
In a study of 2,611 asbestos-exposed insulators, of whom 87% were exposed 30 y or more, Miller et al. found higher prevalences of ILO profusions above 2/2 (i.e., 7.5% in 1,557 subjects with typical opacities), compared with my 0.66% of 12,856 (i.e., 2.5% of the 3,445 with typical opacities). One can attribute these differences to selection criteria: (a) selection of insulators versus building and shipyard tradesmen, (b) longer time since first exposure (i.e., 30 y versus 15 y), and (c) greater durations of exposure (i.e., nearly 30 y versus 5 y). After the RAA groups were identified, their ages were 11 y greater than in the entire population of the present study, but years since first exposure (i.e., 32.8 y and 34.8 y, respectively) and duration of smoking (i.e., 37.1 y and 31.6 y, respectively) varied less. Both populations provided me with opportunities to answer questions about the physiologic function in RAA.
Restrictive disease is rare in asbestosis. In the current study, only half of those with a reduced TLC had pulmonary asbestosis, and their profusion of opacities was low. The reduced TLC and FVC in the group was associated with reduced mid-flows and with air trapping, both of which were appropriate for mild asbestosis. If physiological restrictive disease were strongly correlated with asbestosis, most of this group should have had radiographically advanced asbestosis.
How can these new observations be reconciled with earlier observations of an apparently reduced TLC in pulmonary asbestosis? A review of studies[2-8] revealed that TLC was invariably measured, using the FRC obtained by helium dilution, which depends upon gas equilibration in the lung. In addition, TLC requires a reproducible end-tidal position if expiratory reserve volume is measured; therefore, FRC - ERV = RV and VC + RV = TLC.[14,15] The helium rebreathing method underestimates FRC when there is air trapping, as is the case in asbestosis and emphysema. It is also likely that when spirometry is conducted with the expectation of finding a restrictive pattern, forced expirations are terminated too early, thus underestimating the slowly emptying (obstructed) portion of the FVC. This approach reduces the TLC calculated from VC + RV, when RV is derived from functional residual capacity (FRC - ERV + RV). Measurement of total lung capacities from chest radiographs taken in full inspiration and measurement of FVC after complete expiration results in enhanced dependability of TLC, FVC, and RV, thus eliminating the need to define the expiratory pause point of tidal volume to measure expiratory reserve volume (ERV).
Could the radiographic lung areas enclosed for the calculation of lung volume by planimetry exceed the actual areas? Probably not, because in a variety of lung diseases, radiographic measurements match those made by body plethysmography.[13,16] The important error with this method--that of underinflation--diminishes the radiographic TLC. Possibly the lungs were inflated by other substances that replaced air. If the replacement material was as dense as water, shadows would have been detected in chest x-rays, and postmortem lung weights would have been increased; neither has been described in asbestosis. The only radiolucent alveolar filling material was fat or alveolar (lipo)proteinosis. Although in silicosis some alveoli are filled by alveolar (lipo)proteinosis, which reduces TLC measured by dilution below TLC measured from radiographs, alveolar (lipo)proteinosis has not been reported in asbestosis cases.[25-29]
I conclude that diminished lung volume occurs rarely in severe asbestosis diagnosed radiographically. When it occurs, its cause should be sought. Asbestosis is characterized by progressive obstruction of small airways that quietly transfers volume from vital capacity into the compartment (residual volume), which does not exchange with outside air during breathing. The implications for patient care are that the physician should treat this disease to improve clearance of airway secretions and maintain patency of airways.
The concept that asbestosis would be an obstructive airways disease was expressed by Gloyne in his original description. The use of inappropriate methods for measuring lung volume led to the erroneous idea that asbestosis reduced lung volume and was a restrictive disease. This medical error has been identified and corrected. Asbestosis is an airways obstructive disease. It follows that administrative and legal descriptions and requirements for finding restrictive disease in asbestosis must be corrected so that equitable judgments can be rendered. This is important here and abroad because, although the United States is on the downward leg of the prevalence of asbestosis, the remainder of the world may not have reached the peak.
Peer review of this study was supervised by John F. Finklea and Arthur L. Frank.
Submitted for publication June 29, 1999; accepted for publication August 26, 1999.
Requests for reprints should be sent to Kaye H. Kilburn, M.D., University of Southern California, School of Medicine, Environmental Sciences Laboratory, 2025 Zonal Ave., CSC 201, Los Angeles, CA 90033.
[1.] International Labour Office. U/C International classification of radiographs of pneumoconiosis. In: Occupational Safety and Health Series. Geneva, Switzerland: International Labour Office, 1980.
[2.] Wright GW. Functional abnormalities of industrial pulmonary fibrosis. AMA Arch Ind Health 1955; 11:196-203.
[3.] Williams R, Hugh-Jones P. The significance of lung function changes in asbestosis. Thorax 1960; 15:109-19.
[4.] Thomson ML, McGrath MW, Smith WJ, et al. Some anomalies in the measurement of pulmonary diffusion in asbestosis and chronic bronchitis with emphysema. Clin Sci 1961; 21:1-13.
[5.] Wallace WFM, Langlands JHM. Insulation workers in Belfast. I. Comparison of a random sample with a control population. Br J Ind Med 1971; 28:211-16.
[6.] Kleinfeld M, Messite J, Kooyman O, et al. Effect of asbestos dust inhalation on lung function. Arch Environ Health 1966; 12:741-46.
[7.] Becklake MR, Fournier-Massey G, McDonald JC, et al. Lung function in relation to chest radiographic changes in Quebec asbestos workers. I. Methods, results and conclusions. Bull Physiopathol Respir 1970; 6:637-59.
[8.] Becklake MR, Fournier-Massey G, Rossiter CE, et al. Lung function in chrysotile asbestos mine and mill workers of Quebec. Arch Environ Health 1972; 24:401-09.
[9.] Kilburn KH, Warshaw RH, Einstein K, et al. Airways disease in non-smoking asbestos workers. Arch Environ Health 1985; 40: 293-95.
[10.] Kilburn KH, Warshaw RH. Abnormal lung function associated with asbestos disease of the pleura, the lung and both: a comparative analysis. Thorax 1991; 46:33-38.
[11.] Kilburn KH, Warshaw RH. Pulmonary functional impairment associated with pleural asbestos disease circumscribed and diffuse thickening. Chest 1990; 98:965-72.
[12.] Kilburn KH, Warshaw RH. Severity of pulmonary asbestosis as International Labour Organization profusion of irregular opacities in 8,749 asbestos-exposed American workers. Arch Int Med 1992; 152:325-27.
[13.] Kilburn KH, Miller A, Warshaw RH. Total lung capacity in asbestosis: a comparison of radiographic and body plethysmographic methods. Am J Med Sci 1993; 305:84-87.
[14.] Miller A. The uses of pulmonary function tests in respiratory epidemiology and occupational lung disease. In: Miller A (Ed). Pulmonary Function Tests in Clinical Occupational Lung Disease. New York: Grune and Stratton, 1986; pp 425-48.
[15.] Ries AL, Clausen JL. In Wilson AF (Ed). Lung Volumes, in Pulmonary Function Testing: Indications and Interpretations. New York: Grune and Stratton, 1985; pp 69-85.
[16.] Altose MD, Crapo RO, Warner A. The determination of static lung volumes: report of the section on respiratory pathophysiology. Chest 1984; 86:471-74.
[17.] Ferris BG Jr. Epidemiology standardization project. Am Rev Respir Dis 1978; 118:7-54.
[18.] Sultinen S, Christoforidis AJ, Klugh GA, et al. Roentgenographic criteria for the recognition of nonsymptomatic pulmonary emphysema. Am Rev Respir Dis 1965; 91:69-76.
[19.] Harris TR, Pratt PC, Kilburn KH. Total lung capacity measured by roentgenograms. Am J Med 1971; 50:756-63.
[20.] ATS Statement. Standardization of spirometry--1987 update. Am Rev Respir Dis 1987; 34:1285-98.
[21.] Miller A, Thornton JC, Warshaw RH, et al. Mean and instantaneous expiratory flows, FVC and [FEV.sub.1]: prediction equations from a probability sample of Michigan, a large industrial state. Bull Eur Physiopathol Respir 1986; 22:589-97.
[22.] Kilburn KH, Warshaw RH, Thornton JC, et al. Predictive equations for total lung capacity and residual volume calculated from radiographs in a random sample of the Michigan population. Thorax 1992; 49:519-23.
[23.] Miller, A, Lilis R, Godbold J, et al. Relationship of pulmonary function to radiographic interstitial fibrosis in 2,611 long-term asbestos insulators. Am Rev Respir Dis 1992; 145:263-70.
[24.] Begin R, Ostiguy G, Cantin A, et al. Lung function in silica-exposed workers: a relationship to disease severity assessed by CT scan. Chest 1988; 94:539-45.
[25.] Gloyne SR. The morbid anatomy and histology of asbestosis. Tubercle (London) 1993; 14:550-58.
[26.] Hourihan DOB. A Study of the Pathology of Diffuse Mesotheliomas and an Analysis of Their Association with Asbestos and Asbestosis. Thesis for MD National University of Ireland, 1965.
[27.] Wright JL, Churg A. Morphology of small-airway lesions in patients with asbestos exposure. Hum Pathol 1984; 15:68-74.
[28.] Churg A, Golden J. Current problems in the pathology of asbestos related disease. Pathol Ann 1982; 17:33-66.
[29.] Craighead JF, Abraham JL, Churg A, et al. Pathology of asbestos associated disease of the lungs and pleural cavities: diagnostic criteria and proposed grading scheme. Arch Pathol Lab Med 1982; 106:543-91.
COPYRIGHT 2000 Heldref Publications
COPYRIGHT 2000 Gale Group