Document qdxKBkL9R5L95Vv04wXqeYqyE

yt hility tjf tu Isia! washing ilK performed if Hie inhibitor growth tf tuber U.illest possibli iin aliquot fl ulosis. ;ink Drs. llnwanf irtie C. Rower se A. Schafer fm s manuscript. ilnsis. Am J Ml Vk\ et al: Tlir n:wpsy in tin* rii.iL* y. Am Rev l)R: Cyliiln>;y in ? bronchial brush putum. Chest fu Numbers of Asbestos Bodies in Urban Patients with Lung Cancer and * Gastrointestinal Cancer and in Matched ! Controls* / / // ------------------- TV Kndrew M. Churg, M.D. and Martha L. Warnock, M.D.t i i! <* f *t compared the numbers of asbestos bodies extracted Imm the lungs of 10J patients with lung cancer and M patients with gastrointestinal malignant neoplasms lo the numbers of bodies extracted from lungs of con trol patients matched for age, sex, smoking habits, and, a Mime cases, occupation. All patients were urban f iorllcrs over the age of 40 vears, and none was a prinary asbestos worker. No differences in the counts of whcslus bodies were observed between the tested and control populations. The numbers of asbestos bodies did correlate well with occupation; the highest counts were found in mate manual laborers. We conclude that in the urban population studied herein, the numbers of asbestos bodies alone do not correlate with the presence of pulmonary or gastrointestinal carcinoma; however, micoatcd asbestos fibers are also known to he present in the lung, and the possibility that such tumors may he related to the numbers of these libers in lungs re mains to be explored. et al: Rrnnrhu! bronchoscope m lesions. Chest (u s ia the fiU*n|it* ^cutive cases anl best CS:521-5-1- the diagnosis an*l -itive tuhcretilom iRoG in age and sputum osis. Scand J Rr'p tions in tu!>creu s. 1957 Monehosenpy in j ' hronehnscopir* value of castrii tuberculosis. Am *cts of ancsthiiii mg bronchoscope. topical anesthetic obtained by 1mm* 19(S2 ; al: Should filter <1? Am Rev llcspir rohial activity ol one. J Infect Do he sienificance ol *m tultcrculmit in ' 18-523, 1975 nputer files aid losis patients. 11 :cimens. Am Res ' Pxccpt for cigarette smoke, the causes of the rising ^ incidence of cancer of the lung arc largely un- 1-tcrmined. Recent epidemiologic studies indicate hit carcinoma of the lung has increased in fre- jiicncy from 25/100,000 male subjects in 1951 to #1/100.000 male subjects in 197.3' and that lung i nicer is especially prevalent in industrial workers } i lower socioeconomic class and in certain geo- tuphic areas in the United States (Chicago, where Ithis study was performed, is in one of the counties itli the highest death rate from lung cancer in the | n'uiitry).--3 These data cannot be accounted for by f .'.itterns of smoking hut may correlate with indusl exposure or industrial air pollution.'1 It is well known that inhalation of large amounts ( asbestos may cause carcinoma of the lung, espe cially in smokers. Smaller amounts may be asso} -ated with mesothelioma.3 Epidemiologic studies have also suggested that gastrointestinal cancers are | iwociated with exposure to asbestos.6 The use of asbestos is now widespread in many "iliistries and products, and there is increasing evi- !'iice of contamination with low levels of asbestos `imm the Department of Pathology, Pritzkcr School of | Mnlicinc. Chicago. | xniiniirtcd by grants ICS 01122 and Ca 1-4590 from the I'lilihc 1 leal ill Service and by a grant from the American 'nicer Society, Illinois Division. "I.....ntly at the Department of Pathology, University of California. San Francisco. _ M iMiiscript received September 15; revision accepted Novemi yi * trtrinf requests: Dr. Warnock, Department of Pathology, WsU'.VJS. University of California School of Medicine, San l':niitro9J143 CHEST. 76: 2. AUGUST, 1979 2, AUGUST, 1979 in air and water supplies.'1'7 as well as the possible hazard to the general population from such unlikely sources as road-paving materials* and even drugs." One manifestation of such contamination is the presence of both asbestos bodies and uncoated as bestos fibers in the lungs of almost everyone in the general population. There has been considerable controversy in the past about whether such bodies actually contain asbestos;10 we have recently shown that almost all such bodies do contain cores of amphibole asbestos."-13 Several qualitative studies have failed to demonstrate an association between the presence of asbestos bodies in urban individuals and bronchogenic carcinoma.13'13 hut in a prelimi nary quantitative study, we1G found that 27 percent of patients with lung cancer had more than 50 as bestos bodies per gram of lung, whereas only 5 percent of the unmatched control population had that number. Since then, wc17 have discovered that in our urban postmortem and surgical population, sex and occupation arc important factors in deter mining the numbers of asbestos bodies found in the lungs; hence, an appropriate control population must be carefully matched for these variables. In the present report, wc have counted asbestos bodies in 103 urban patients with bronchogenic carcinoma and compared them with 103 urban patients matched for age, sex. smoking history, and, where possible, occupation. A similar comparison was per formed for 50 patients with gastrointestinal carci noma and matched controls. NUMBERS OF ASBESTOS BODIES IN URBAN PATIENTS 143 I ASARCO ALV 0001445 Materials and Methods Samples Samples of lung w ere obtained from 363 patients for whom light-microscopic counts of asbestos bodies; detailed smoking, occupational, and residential histories; and complete histo logic material were available. Occupational, smoking, and residential histories were obtained by interviews with tire patients or relatives. Interviews were conducted without knowledge of the count of asbestos bodies by a social worker using pages 4 and 5 of the questionnaire on respiratory symp toms of the Medical Research Council, supplemented with questions on occupational history, hobbies, and use of insulat ing and construction materials at home. Two hundred fiftytwo of these patients have been analyzed previously for correlation of occupation and counts of asbestos bodies.17 All patients were seen at the University of Chicago Medical Center between February 1974 and May 1977. Samples of lung were taken from 103 patients with lung cancer (66 from autopsies and 57 from patients undergoing resection for lung cancer), front 50 patients dying with cancer of the esopha gus, stomach, or colon, and from 210 control patients dying during the same period of time. The control population was composed of patients with either nonneoplastic disease or neoplasms other than cancer of the lung or gastrointestinal tract. Only urban patients who resided for more than 20 years in a city with a population of more than 100,000 and who were ov er the age of 40 years were included in the study. In order to determine the relationship of our counts of asbestos bodies to counts in patients with heavy exposure to asbestos, we analyzed samples of lung from six patients with asbestosis. Preparation of Pulmonary Tissue Lung fixed in formaldehyde solution (Formalin) was di gested with bleach (sodium hypochlorite, 5225 percent), and the digest was collected on membrane filters with a pore size of 0.45(r (Milliporel, according to the method of Smith and Naylor1' as modified by Churg et al.17-10 Three samples with a wet weight of approximately 4 gm each were taken from each lung from autopsy (one sample from the upper and two from the lower lobe). Two samples were taken from resected lobes. Samples were taken from parenchyma away from large bronchi, tumor, consolidation, scar, or pleura. The values for the samples counted were averaged, and the result was expressed as the number of asbestos bodies per gm of wet weight of lung. For the samples from lungs with asbestosis, the size of the sample was decreased to 1 to 2 gm, and bodies in ten random areas measuring 0.04 sq mm were counted; the result was multiplied by an appropriate factor. The results for wet lung max' be roughly consorted to bodies per gram of dry weight by multiplying by a factor of 10.17 Asbestos bodies were counted without knosvledge of either the patient's disease or occupation. Data on Patients Slides and protocols from each of the 103 cases of lung cancer were reviewed to confirm the diagnosis and exact location of the tumor. Cases in which the possibility of a nonpulmonary primary carcinoma was raised were eliminated from the study. All surgical cases in which the preoperative work-up failed to disclose another primary cancer were included. Tumors were classified using the classification of the World Health Organization. Tumors were examined without knowledge of the counts of asbestos bodies or knowl edge of occupational and environmental histories. Protocols 144 CHURG, WARHOCK I and slides from the 50 patients with gastrointestinal cance were reviewed to confirm the diagnosis. Protocols for all i the control patients were reviewed to confirm the absence lung and gastrointestinal cancer. Because we11 have four that a level of approximately 100 asbestos bodies per gra. appears to be a cutolf between persons with clear-cut se ondarv occupational exposure to asbestos and persons with i evident occupational exposure, we examined the type ar distribution of lung cancer in patients with greater or Ic than this number of bodies. Case Matching and Statistical Methods All data for the 363 patients were stored on comput disks. Patients with lung cancer were matched against co trols by sex, age within ten years, and pack-years of cig rettes within ten. For this study, patients with less than pack-years of smoking or patients who smoked cigars or pip were designated nonsmokers. Matching trials were perform) with and without the patients with gastrointestinal cancers the control group. Since occupation has been shown influence the number of asbestos bodies in the lung in nu subjects,1; an additional matching of patients with lu: cancer and control subjects using aee. sex. smoking, ai occupation was performed. For this comparison, patients wi white-collar occupations were matched with patients wi white-collar occupations, but manual laborers were not su classified as to type of work. Matching was done largely I the computer to decrease bias; however, five men had to ! matched by hand because they had smoked very heavily (i to 220 pack-years). In these instances, persons with 100 more pack-years of smoking were considered suitab matches. Three women were similarly matched by hand. T1 patients with gastrointestinal cancer were matched by ti same criteria for sex, age, and smoking against the contr group that excluded patients with lung cancer. Since ai given matching trial could not make use of all of the contr cases, the computer program was designed to permit ditfere. sets of control patients to be matched against the patier with cancer. Multiple matching trials using this capabili were performed for each tested group. Because the counts of asbestos bodies were marked skewed, with most counts being less than 300 (even for ma manual laborers), statistical tests assuming a normal distrib' tion were not considered to be suitable; and nonparametr tests were chosen to analyze the data. The sign test ar Wilcoxon's matched-pairs signed-ranks test were perform' on tire paired data.10 These tests utilize information abo the direction and the magnitude of differences betvver paired values, respectively. In order to compare the grou; more specifically on the basis of actual measured valu (rather than, for example, on the means or signs of pairr differences) while avoiding assumptions about distribute we disregarded all matching and treated the groups wicancer and the control group as independent samples for tl Kolmogorov-Smirnov test.70 Because our hypothesis was th the counts of asbestos bodies in the experimental group we "higher" than in the control group, we used a one-tailed tr throughout Results Pulmonary Cancers Counts of asbestos bodies for the patients wit lung cancer varied from zero to 9,200/gm of luo and varied for the controls from zero to 1,980/gmc CHEST, 76: 2, AUGUST, 137 ASARCO ALV 0001446 i.nail c.itic_cn Is for all of ab encc n| we found per firam ear-cut see* *ons with no ]J j j j Croup Table 1--Statistical AnalyJsis o'f Cases Analysis of Matched Cases No. of WilMatched Results of Sign Test l'airs I' Value* ('a > Control** CuCControlf 1* Value Analysis of Cases Disregarding Matching (Kolmogorov-Smirnov Test) X5 D Value 1,1/~2) F Value uter or 1cn l.'inc cancer Mm vs nil controls (matched for nee and smoking) Men vs controls without gustro intestinal cancers (matched n computer tgainst con* fur age and smoking) Mm vs controls without gastro irs of ciga- intestinal cancer (marched less than 5 fur ago, smoking, ami .ars or pipe* occupation) performed U'unicn vs all controls .1 cancers in Women vs controls without shown to 1 c:i1rnmtcsfinal cancer me in mate with lung oking, and itients will tients with 1 tn-troinipstinal cancer Mm vs controls without luntr ivinccr Women vs controls without lung c:mcer nil on no 34 34 3i 10 (1.17 0.21 0.12 o. in 0.21 0.37 0 12 34 36 31 18 10 14 12 35 0.5 0.12 2.0 >0.3 32 0.36 0.14 2.9 >0.2 20 0.44 0.18 4.0 >0.1 15 0.36 0.00 0.2 >0.9 It 0.24 0.12 1.0 >0.5 16 0.57 0.16 1.6 >0.3 7 0.18 0 21 1.7 >0.3 largely h> i had to !* jravily (up itli 100 or 1 suitahlr hand. The m*cI by die control nee any the control t different c patients capability markedly n for male .1 dixtrilro* parametric test and performed tion about between he group* il value* of paired ^tribution, tups with les for the * was that nup were tailed test its with lung imoT 'Wilcoxon's matched-pairs sign**d-mnk* te< t. *MYir> with greater numbers of; shrsto> bo< lies in patient with eaneer than in control. tPairs with greater numbers of: sliestos bot lies in control subject than in patient with cancer. lung. Only two of the patients with lung cancer had uMiits over 2,000/gm; these were 9,000/gm and l.-IKl/gm, respectively. The average age of the women with lung cancer was 58 years and of the men was 61 years. The median count of asbestos Wlics for women with lung cancer was 11/gm and Inr matched controls was 13/gnt; the median count tor men with lung cancer was 33/gm and for their matched controls was 44/gm. Matching of patients with lung cancer and control mhjccts was initially based on age. sex, and smoking Hi'tnry. Two different control groups were selected. The first excluded patients with gastrointestinal rancors because it was thought that this group might have high counts of asbestos bodies and thus mini mize differences between tested and control groups; lor similar reasons, patients with lung cancer were not used as controls for patients with gastrointestinal cancers. The second control group included patients *ith gastrointestinal cancers. In a separate matching trial, 60 of the 69 men with King cancer were matched by occupation, as well as ice and smoking history. Examination of the counts c( asbestos bodies for each matched pair by the sign t 4 indicated no significant differences between patirtils with lung cancer and control subjects. The Mimhor of pairs with higher counts in the patient with cancer than in the control subject approximate ly equaled the number where the reverse was true (Tabic 1). Similarly, the results for Wilcoxon's test showed no significant differences (Table 1). The X2 values which are derived from the D values for the Kolmogorov-Smirnov test for each of the com parisons for men and women were not significant (Table 1 and Fig 1). The multiple matching trials using different subgroups of the control population all gave similar results. Results of classification of tumors by type and location in the lung for patients with counts above Number of Asbestos Bodies ZG. Lung Fir.ung 1. Results of Kolmogorov-Smirnov test comparing 60 men with lung cancer (light line) and their 60 matched control subjects (heavy line), treated as two independent random samples. Cumulative distributions of each sample are plotted at intervals of ten asbestos bodies. These distribu tions for men demonstrate also marked skewing of data. Arrows indicate largest deviation between curves. Because chance deviation of this size is not unlikely, this comparison provides no evidence that samples were from different popu lations. fST, 1975 CHEST, 76= 2, AUGUST, 1979 KUMBERS f ASBESTOS BODIES IN URBAN PATIENTS 145 ASARCO ALV 0001447 J iiij i j 3 i ? i * Table 2--Type o/ Lung Cancer /fccorrfing to Count o/ Ro/ltps Group and No. of Asbestos Bodies * No. of Cases (Perrcnt) Squamous Cell Adeno barge Cell Small Cell Careinoma carcinoma Careinoma Carcinoma Women <l00/gm >100/gm 6 (21) 3(75) 12(41) 1(25) 6(21) 0 5(17) 0 Men CtOO/gm** > 100/cm 15(34) 10(40) 18(41) 12 148) 5(11) n 6(14) 3(12) Value of 100 asbestos bodies per pram of lung was arbitrarily selected as cutoff between counts l>elnw that value (environ mental exposure) and counts above that value (occu pational exposure).17 *()ne adenosquamous tumor 1ms been omitted. and below 100 bodies per gram are given in Tables 2 and 3. No differences between the two groups are apparent for type of tumor or location, with the single exception of the high frequency of squamous carcinoma and low frequency of adenocarcinoma in the women with high counts. Gastrointestinal Cancers Counts of asbestos bodies for the patients with gastrointestinal cancers varied from zero to 559/gm. The median count of asbestos bodies for women with cancer was 27/gm and for matched controls was 15/gm; the median count for men with cancer was 20/gm and for matched controls was 40/gm. Twenty-three of 31 male patients and ten of 19 female patients were smokers. The average age of the women was 63 years and of the men was 65 years. The locations of the gastrointestinal cancers are given in Table 4. Matching of patients with gastrointestinal cancer and control subjects was based on age, sex, and smoking. All patients with lung cancer were ex cluded from the control group. Examination of the Table 3--Location of Lung Cancer According to Count of Asbestos Bodies * No. of Can s (Pereent) Group and No. of AsIwMos Bodies Women < 100/gm of lung >100/gin of lung Men <100/gm of lung > 100/jsm of lung Up|Kr and Middle Lolies Lower Lobes 10 (70) 3 (75) 7(301 1 (25) 23 (64) 18(82) 13(36) 4 (18) Origin of six tumors in wnmrn and 11 in men could not. be attributed to specific lobar bronchus. 146 CHURG, WARNOCK Table 4--Gastrointestinal Cancer by Site Group Men Women Total No. of Cases (Percent) Esophagus 7(22) 4 (21) 11(22) Stomach 12(30) 5 (26) 17(34) Cnlun 12(30) 10(53) 22(11) paired counts of asbestos bodies showed no signifi cant differences between patients with gastroin testinal cancer and controls for the sign test or \Vflcoxon's test (Table 1). The values for the Kolmogorov-Smimov test for each of the comparisons for men and women were not significant (Table 1). Smoking and Age os Counts of Asbestos Bodies Figures 2 and 3 indicate that asbestos bodies 4s not increase with age or pack-years of cigarrtln smoked for male manual laborers. Comparison ni counts of asbestos bodies and cigarettes smoked per day also failed to show a correlation. In patients with asbestosis, the numbers of adics tos bodies varied from 8,000/gm to 520,000/gm in our samples (Table 5). In most of the cases, t limited amount of lung was available for analysis Marked discrepancies in replicate counts were noted n n. r 3k a 'A Ul( ?I LSI na :os- \ iui 'un Jab Pati Pali Figure 2. Relation of number of asbestos bodies per pan of lung to age of male manual laborers. There is no evidence for increase in numbers of asbestos bodies with age. Six pa tients with more than 1,200 asbestos bodies per gram o( lung were excluded; all but two of these were less than 53 years old. ?ati?ati ?ati' Pali. CHEST, 76: 2, AUGUST, 1979 I ASARCO ALV 00014-4-8 by Site 1200- rut ` Colon 12 (311) 10 (53) 22 (4-11 900- ved no signifi- with gastroin- i : l 600-1 in test or Wil- a ` or the Kolnio- nnparisons for [Table 1). is Bodies JCO stos bodies do s of cigarettes Comparison <i( les-smoked per i hers of asbes520,000/gm in t the cases, a !e for analysis, uts were noted -t *** * " * 57 115 172 230 FttCK-YEARS SMOKED (I'K.enE 3. Relation of number of asbestos bodies per gram of kmc to total pack-years smoked for male manual laborers. I lu re is no evidence of increase in numbers of asbestos Imdics with increasing pack-years smoked. Smokers with fro pack-years include pipe and cigar smokers. Six patients ilh more than 1.200 asbestos bodies per gram of lung were rwluded; all smoked less than 60 pack-years. in cases 5 and 6, suggesting that sampling is impor tant in scarred lungs, where distribution and clear ance may be abnormal. By contrast, in our sample population, only seven of our counts were over 800 .ulicstos bodies per gram; the vast majority of our patients had counts that were two or three orders of magnitude below counts in the patients with asbeslosis. Discussion In this study, we failed to find differences in the numbers of asbestos bodies between patients with lung or gastrointestinal cancers and matched con- Table 5--Numbers of Asbcsta* Bodies in Patients with Asbestosis 77 Sto bodies per cram re is no evidence vith age. Six paies per gram Ol e less than 53 Patient i 1 Patient 2 Pal lent 3 Patient 4 I'alirnt 5 Patient 6 Asbestos Jlixlios |x*r Gram of Lung ,----------------------- -------------------V Sample l Sample 2 370,000 430,000 340,000 520,000 12,000 27,000 23,000 62,000 15,000 360,000 8,000 170,000 2. AUGUST, 1979 CHEST, 76: 2, AUGUST, 1979 trols. The lack of statistically significant differences between groups with cancer and control groups, the absence of large numbers of patients with cancer who bad counts greater than the matched control subject, and the low median values for numbers of asbestos bodies in groups with cancer and control groups all indicate to us that, for practical purposes, asbestos bodies in this matched population are not associated with these cancers. / We did, however, find an association between occupation and counts of asbestos bodies; 60 percent of the construction workers, 41 percent of the steel workers, and 25 percent of the other male manual laborers had more than 100 asbestos bodies per gram of lung, whereas only 7 percent of men and women with white-collar occupations had more than 100 asbestos bodies per gram. Somewhat similar results have been obtained by Sclikoff and Ham mond-1 and by Doniach et al=i using semiquantitativc methods. In our study the number of asbestos bodies seems to be unrelated to smoking in male manual laborers (Fig.3). Also, the numbers of as bestos bodies do not increase with age after 40 years (Fig 2), although others have found that such an increase does occur;21 however, careful matching for age is still necessary because asbestos-induced neoplasms are related to the time since the onset of exposure, with a period of latency of 15 to 50 years between initial exposure and appearance of the cancer. There are many epidemiologic studies on the inci dence of lung cancer in workers in the asbestos industry. These have recently been reviewed by Becklake.5 Several of these reports have attempted to grade the degree and length of exposure to as- , hestos; and. in general, an increased incidence of \ lung cancer was not found in persons in the groups with lowest exposure, even after long latent periods.5 Since our patients almost certainly have less exposure to asbestos than these workers, our results are in keeping with the epidemiologic data. The type and distribution of lung cancer in as bestos workers were examined by Kannerstein and Cluirg.23 They studied lungs from 50 male patients with occupational exposure and lungs from 50 male control patients and found that there were no differ ences in histologic type, but the asbestos-associated tumors tended to occur more frequently in the lower lobes. In the present study, all tumors occurred more frequently in the upper lobes, quite the reverse of the distribution seen with asbestos-associated can cers. Attempts to determine possible asbestos-related risk for workers in nonasbestos industries and for members of the general population have been rela- NUMBERS OF ASBESTOS BODIES IN URBAN PATIENTS 147 I ASARCO ALV 0001449 tively few. In one of these studies, Whitwell et al24 compared the numbers of asbestos fibers visible by phase-contrast microscopic examination in lungs of 100 patients with lung cancer, 100 control subjects, and 100 patients with mesothelioma and found that patients in the group with lung cancer and the control group had similar distributions of fiber counts. Whitwell et al24 found little overlap be tween their control group and their patients with mesothelioma who had asbestosis, although there was considerable overlap of patients with mesothe lioma who did not have asbestosis and controls. Most of the controls had lived in Liverpool, Eng land, but 22 percent of the patients with lung cancer had lived in rural areas. No matching for occupation or place of residence was attempted, despite the fact that Whitwell ct al24 noted large discrepancies in counts according to occupation. They concluded that levels of exposure to asbestos that do not cause asbestosis are not etiologically related to lung can cer. While the counts from many of our patients fall within the values cited by others, it is difficult to compare results from different laboratories, particu larly since few details of sampling, preparation of tissue, and methods of counting are provided. In our laboratory, we have found that in lungs from the general population, two or three samples from the same lung yield a relatively narrow range of results (mean number of bodies 50 percent). In lungs from primary asbestos workers who have asbestosis, the range in a single lung is considerably wider and often approximates one order of magnitude (Table 5). Such discrepancies are probably related more to a nonuniform distribution in scarred lung, rather than to technical details; hence, one must be cau tious about interpreting counts from only one or two areas in a scarred lung. Nevertheless, we have observed a very great separation between values found in the general population, only seven of which were over 800 asbestos bodies per gram, and those in asbestos workers. We wish to emphasize that even a strong positive correlation between numbers of asbestos bodies and carcinomas would not, by itself, define an increased incidente of cancer in these patients; such a conclu sion would require studies of incidence based on large populations with varying levels of asbestos bodies. Nevertheless, our negative findings should not yet be interpreted to mean that there is no danger to the general population from asbestos. In asbestos workers the number of uncoated fibers in the lung is far greater than the number of asbestos bodies; similarly, the number of fibers visible by light-microscopic examination in such workers is only a fraction of the number demonstrable by elec tron-microscopic examination.2 Whether this is also true of the general population has not been deter mined. When this study was begun, we had no informa tion concerning the type of cores found in fer ruginous bodies in the general population. We have now examined over 500 ferruginous bodies by elec tron optical techniques and have found that over 95 percent contain a core of amphibole asbestos;"'12 we have also demonstrated that it is possible routine ly to pick out those which do not contain asbestos by simple light-microscopic examination.12 Thus, there is no doubt that the counting of asbestos bodies in lungs of the general population serves as one marker of exposure to asbestos; however, it is known that uncoated asbestos fibers (generally chrysotile as1 bestos) too small to form asbestos bodies may he ' found in most lungs25 and that most of the asbestos used in this country is chrysotile. rather than am phibole. Therefore, in order to establish an associa tion between the total pulmonary burden of asbestos and carcinoma of the lung in the general population, it will be necessary to measure total asbestos fibers in a controlled study similar to this one. ACKNOWLEDGMENTS: We thank Mr. Richard Blouph of the Biomedical Computation Facilities. University of Chi cago, for assistance in preparing this report. Dr. J. Churg and Dr. V. Ko supplied the samples of lung from patients with asbestosis. References 1 Vincent RG, Pickren JW, Lane WW, et al: The changing histopathology of lung cancer: A review of 1.G82 cases. Cancer 39:1617-1655, 1977 2 Mason TJ, McKay FW, Hoover R. et al: Atlas of Cancer Mortality for U.S. Counties: 1950-1969 (publication NTH 75-780). US Department of Health, Education, and Wel fare, 1976 3 Brown SM, Selvin S, Winkelstcin W: The association of economic status with the occurrence of lung cancer. Cancer 36:1903-1911, 1975 4 Selikoff IJ, Nicholson WJ, Langer AM: Asbestos air pollution. Arch Environ Health 25:1-13, 1972 5 Bccklake MR: Asbestos-related diseases of the lung and other organs: Their epidemiology and implication for clinical practice. Am Rev Respir Dis 114:187-227, 1976 6 Selikoff IJ, Hammond EC, Scidman If: Cancer risk of insulation workers in the United States. In Bogovski P, Cilson JG, Timbrel! V, et al (cds): Biological Effects of Asbestos (publication No. 8). Lyon, France, IARC Scien tific Publications, 1973, pp 209-216 7 Nicholson WJ: Analysis of amphibole asbestiform fibers in municipal water supplies. Environ Health Perspect 9:165172,1974 8 Rold AN, Langer AM. Selikoff IJ: Environmental asbestos pollution related to use of quarried serpentine rock. Sci ence 196:1319-1322, 1977 9 Nicholson WJ, Maggiorc CJ, Selikoff IJ: Asbestos con tamination of parenteral drugs. Science 177:171-173, 1972i 148 CHURG, WARNOCK j CHEST, 76: 2, AUGUST, 1979 ASARCO ALV 0001450 y trable by decher this is also >t been deter- id no informafound in forition. Wc have >odies by decid that over `J5 e asbestos;"-" issible routinetin asbestos In- Thus, there stos bodies in as one marker is known tli.il chrysotile nv mdies may lie if the asbestos thcr than am i.sh an associnIon of asbestos ral population, isbestos fibers te. | M Cwi P, deTreville RTP, Cralley LJ, et ah Pulmonary I ferruginous bodies. Arch Pathol 83:539-516. 1968 | I! Clung A, Warnock ML: Analysis of tlie cores of fer| riiginous (asbestos) bodies from tlie general population: I I. Patients with and without lung cancer. Lab Invest * 17:280-286, 1977 Oitirg A, Warnock ML, Creen N: Analysis of the cores of ferruginous (asbestos) bodies from the general popula tion: 2. True asbestos bodies and pseudo asbestos bodies. Lib Invest-10:31-38, 1979 II \iz7.e II: Exposure to asbestos and tlie genesis of pleural plaques and neoplasia. Arch Pathol 95:213-214, 1973 11 flreedin PH, Buss Dll: Ferruginous (asbestos) bodies in the lungs of rural dwellers, urban dwellers and patients with pulmonary neoplasms. South Med J 69:401-404, 1976 15 Anjilvcl L, Thurlbeck WM: The incidence of asbestos limlics in the lungs at random necropsies in Montreal. Can Med Assoc; 95:1179'-11S2. 1906 !* Warnock ML, Cliurg AM: Association nf asbestos and bronchogenic carcinoma in a population with low asbestos exposure. Cancer 35:1236-1242, 1975 I* Churg AM, Warnock ML: Correlation of quantitative asbestos body counts and occupation in urban patients. Arch Pathol Lab Med 101:629-634, 1977 IS Smith MJ, Naylor B: A method for extracting ferruginous -chard Blouglt nl liversity of Chi- Dr. J. Cluirc -nwl n patients with bodies from sputum and pulmonary tissue. Am J Clin Pathol 58:250-254, 1972 19 Churg A, Sakoda N, Warnock ML: A simple method for preparing ferruginous bodies for electron microscopic examination. Am J Clin Pathol 68:513-517, 1977 20 Siegel S: Nonparametric Statistics for the Behavioral Sciences. New York, McGraw-Hill Book Co, 1956, pp BB SS, 127-136 21 Selikoff IJ, Hammond EC: Asbestos bodies in the New York City population in two periods of time. In Shapiro HA (ed): Pneumoconiosis: Proceedings of tlie Interna tional Conference, Johannesburg, 1969. London, Oxford Univ ersity Press. 1970, pp 99-105 22 Doniach I, Swettenliani KV, Hathorn MKS: Prevalence of asbestos bodies in a necropsy series in East London: Association with disease, occupation, and domiciliary ad dress. Br J Ind Med 32:16-30, 1975 23 Kannerstein M, Churg J: Pathology of carcinoma of the lung associated with asbestos exposure. Cancer 30:14-21, 1972 24 Whitwcll F, Scott J. Crimshaw M: Relationship between occupations and asbestos-fibre content of the lungs in patients with pleural mesothelioma, lung cancer, and other diseases. Thorax 32:377-386, 1977 25 Longer AM, Selikoff IJ, Sastre A: Chrysotile asbestos in tlie lungs of persons in New York City. Arch Environ Health 22:348-361, 1971 il: Tlie changing of 1,682 cases. Atlas of Cancrt publication Ntll cation, and Wri ng association nf of lung cancer. M: Asbestos air 1972 -.-f the lung ami implication for 187-227, 1970 : Cancer risk n( In Bognvski I', higical Effects nf ice, IARC Scicn- stiform fillers in ' Perspcct 9:16.'- nmental aslicslm entine rock. Sci- <: Asbestos con177:171-173, Current Concepts in Pediatric and Adult Allergy The R. A. Cook Institute of Allergy will present a Postgraduate Course, "Current Con cepts in Pediatric and Adult Allergy" September 26-28 in New York City. For informalion, contact Dr. Stanley R. Fine, Roosevelt Hospital, Department of Allergy, 428 West 59th Street, New York, New York 10019. 13th Annual Respiratory Therapy Symposium The University of Kansas Medical Center will present the 13th Annual Respiratory Therapy Symposium in Kansas City, September 10-11. For further information, please contact the Division of Continuing Education, University of Kansas Medical Center, 30th and Rainbow Blvd, Kansas City, Kansas 66103. .. AUGUST, 1979 CHEST, 76: 2, AUGUST, 1979 NUMBERS OF ASBESTOS BODIES IN URBAN PATIENTS 149 \ ASARCO ALV 0001451