Document b5OOwNDgrkYe0O4e5DG8bvxK6

Downloaded from oem.bmj.com on 29 June 2007 OEM ONLINE Mesothelioma: cases associated with non-occupational and low dose exposures G Hillerdal Occup. Environ. Med. 1999;56;505-513 Updated information and services can be found at: http://oem.bmj.com/cgi/content/abstract/56/8/505 References Rapid responses Email alerting service These include: 11 online articles that cite this article can be accessed at: http://oem.bmj.com/cgi/content/abstract/56/8/505#otherarticles You can respond to this article at: http://oem.bmj.com/cgi/eletter-submit/56/8/505 Receive free email alerts when new articles cite this article - sign up in the box at the top right corner of the article Notes To order reprints of this article go to: http://www.bmjjournals.com/cgi/reprintform To subscribe to Occupational and Environmental Medicine go to: http://www.bmjjournals.com/subscriptions/ Downloaded from oem.bmj.com on 29 June 2007 Occup Environ Med 1999;56:505-513 REVIEW 505 Mesothelioma: cases associated with non-occupational and low dose exposures Gunnar Hillerdal Abstract Objectives--To estimate the importance of low dose exposure to asbestos on the risk of mesothelioma. Methods--A review of the literature. Results and conclusions--There is no evi dence of a threshold level below which there is no risk of mesothelioma. Low level exposure more often than not contains peak concentrations which can be very high for short periods. There might exist a background level of mesothelioma occur ring in the abscence of exposure ot asbes tos, but there is no proof of this and this "natural level" is probably much lower than the 1-2/million/year which has been often cited. (Occup Environ Med 1999;56:505-513) Keywords: low exposure; asbestos; mesothelioma Department of Lung Medicine, Karolinska Hospital, Stockholm, Sweden G Hillerdal Correspondence to: Dr Gunnar Hillerdal, Department of Lung Medicine, Karolinska Hospital, Stockholm, Sweden. Accepted 8 March 1999 Mesothelioma is an incurable disease which is almost exclusively due to inhalation of asbestos fibres. Asbestos has been extensively used in industry and construction in the 20th century, especially during and after the second world war, and even if the mineral is no longer used in most rich western countries the total world production remains high. There is a worldwide pollution with asbestos, as indicated by the finding of the mineral in samples of Greenland ice1 and on the Yorkshire Moors,2 and every citizen in the world has been exposed to some extent. Consequently, asbestos fibres can be found in most lungs at necropsy.3 It is thus understandable that there is concern about the risk of mesothelioma for the general popula tion. However, it should be remembered that mesothelioma is a rare disease with incidence in industrialised countries ranging from 1 to 5/million/year among women and values for Table 1 Proportionate mortality from mesothelioma in various cohorts Cohort Amphibole miners Crocidolite miners Insulation workers Patients with asbestosis Deaths (n) 1225 118 4951 283 40 59 Deaths due to mesothelioma (%) 2 4 6 9 9 10 10 Reference 4 5 6 7 8 9 10 men 5-10 times higher (see table 3). Even in cohorts with a very heavy exposure to asbestos most people will die from other causes. In peo ple with certified asbestosis--that is, with a heavy exposure--up to 10% will develop mes othelioma; among insulators in the United States and Canada, also a heavily exposed group, 9.3% of the deaths have been due to this disease; and in amphibole miners in South Africa or Australia, this figure is 2-4% (table 1). Clearly, with exposure concentrations several magnitudes lower, as occurs in the gen eral population, the risk is very small, often impossible to measure. A discussion of the risks from low exposure must include the dose-response curve; the existence or non-existence of a threshold, and thus a background concentration; and should try to define low exposure and estimate to what degree that really means a low concentration. From conflicting findings and opinions at tempts must be made to make a meaningful conclusion. The different types of asbestos seem to differ considerably in their ability to cause mesothe liomas. Chrysotile is considered by many authors to be a weak carcinogen in humans,11 whereas the two amphiboles crocidolite and tremolite are much more dangerous according to many studies.12 The third of the more important amphiboles, anthophyllite, was long considered not to cause mesothelioma, but such tumours have now been reported al though the risk seems to be small.13 There is, none the less, a minority opinion that chrysotile is in fact responsible for most of the pleural mesotheliomas in society14 or should at least be considered to carry the same risk.15 This discussion, however, falls outside the present review and is not important for the conclusions drawn here. Definition and diagnosis of mesothelioma Mesotheliomas are, by definition, tumours that arise from mesothelial cells and can thus arise from any body cavity: the pleura, the perito neum, the pericardial sac, and even the tunica vaginalis testis. Pleural mesotheliomas are the most common and pathologically the best defined ones. Pleural mesotheliomas have a male to female rate of about five to one, whereas for peritoneal tumours this ratio is 1.5 to 1.16 Thus, either the aetiology is different, the 506 Downloaded from oem.bmj.com on 29 June 2007 Hillerdal Table 2 Possible risk factors and mediators of risk of mesothelioma (other than asbestos) Factors Erionite Chronic inflammation Radiation Beryllium Vegetable fibres Hereditary factors Immunological factors Dietary factors Viruses Comments Very high incidence of mesothelioma environmental exposure in Turkey Pleural scars (tuberculosis, pleurisy, therapeutic pneumothorax) Single cases after Thorotrast injection or radiotherapy; causality not proved One case in atomic bomb survivor Two doubtful cases described No proof in humans Familial cases (explained by common asbestos exposure?) Correlation with parental cancer Rapidly progressive cases in patients with HIV infection Provitamin A, p-carotene may decrease the risk Mesotheliomas in animals Simian virus 40 DNA sequences reported in mesotheliomas References 19 20 21 22 23 18 24-26 27 2829 3031 32 diagnosis is more difficult in women, or women exposed to asbestos are more likely to develop a peritoneal than a pleural mesothelioma. In fact, one type of mesothelioma occurring almost exclusively in women is the so called multicystic mesothelioma. This tumour has a better prognosis, is more sensitive to cytostatic drugs, and seems to have no connection with exposure to asbestos.17 The pathological diagnosis of mesothelio mas requires experience, and confusion can occur with both benign pleural lesions and with metastatic pleural diseases.18 Many countries now have "mesothelioma panels", which has meant a great improvement in the diagnosis. In most national cancer registries, most likely both overdiagnosis and underdiagnosis occur. Suggested causes of mesothelioma other than mineral fibres Apart from mineral fibres, radiation (for exam ple, through the contrast medium Thorotrast) has been suggested to cause human mesothe lioma. However, in a large retrospective study of more than 250 000 women treated for mammary carcinoma, a quarter of them initially with radiotherapy, no association between radiation and mesothelioma could be found.22 As can be seen from table 2, other causes or contributing factors have also been suggested. Viruses can cause mesotheliomas in animals, but this has not been described in humans. DNA sequences associated with Simian virus 40 (SV 40) transforming factors have been reported in a high proportion of mesothelio mas from some countries.32 This suggests that there is a connection between SV40, which was a contaminant of live polio vaccines in 1959-61, and later development of mesothe lioma. Thus, SV 40 might be a cofactor to asbestos in some patients with mesothelioma , but the results have not been confirmed and are still disputed. In summary, then, as far as is known today, factors other than mineral fibres can only explain a very small proportion of mesothelio mas, and can for practical purposes be disregarded. Thus, a malignant mesothelioma can be regarded either as caused by asbestos or belonging to a normal background level--that is a spontaneously occurring tumour. The rela tive imnportance of these two factors has been debated and will be further explored in this review. Incidence of mesothelioma There is a large variation in the incidence of mesothelioma in different countries and in most places a steadily rising number of cases with time. In table 3, the incidence or mortality from mesotheliomas in different countries at various times can be seen. As mortality for practical puprposes is the same as the inci dence for this disease, both figures have been used in the table. Some of the differences between the countries are probably due to diagnostic difficulties, but most of the varia tions can be explained by the use of asbestos in the particular society some decades earlier. Dose-response and latency time Most researchers agree that there is a positive dose-response curve for mesothelioma--the heavier the exposure to asbestos, the greater the risk. This is found in cohort studies as well as in analyses of amphibole asbestos fibres in the lungs.45-48 It was realised early that time since first exposure was of great importance, and therefore the "cubic residence-time model" was suggested by Doll and Peto in their report in 198549: I (T) = c x F x (T4-(T-D)4 Where, I (T)=incidence at the time T after exposure; c=a constant depending on the process, F=intensity of exposure, and D=duration of exposure. This equation has been used in many studies with an acceptable fit for normal occupational exposure concen trations . F in the equation is the total exposure--a combination of fibre concentra tions and exposure time--usually measured in fibre-years. (1 fibre-year = a mean of 1 fibre/ml air for 1 working year). Thus, the doseresponse curve is supposed to be linear, but the result is heavily influenced by the time factor. Unfortunately, the exact value of F is often uncertain, even in well defined highly exposed cohorts. The equation is thus rarely useful especially with low doses, in which F is usually a crude guess only. As can be seen from table 1, even with heavy exposure only up to 10% of a cohort will die from mesothelioma--so the for mula is not applicable in these cases either. With very heavy exposure, most patients will die from pulmonary insufficiency due to asbestosis before there has been sufficient time to develop a mesothelioma. Even more troublesome is the time factor T, which quickly becomes very important in this equation. If the equation is correct, the risk Mesothelioma Downloaded from oem.bmj.com on 29 June 2007 507 Table 3 Incidence or mortality of mesothelioma in various countries and areas over time (! million inhabitants /year) Country or area United States North America Nantes-Saint-Nazaire, France Texas Selected cities, United States United States Barcelona, Spain Great Britain Finland Great Britain Nantes-Saint-Nazaire, France United Kingdom Denmark Nantes-Saint-Nazaire, France Great Britain Australia Great Britain Great Britain Australia Year 1968-81 1972 1956-74 1976-80 1970s 1986 1983-90 1968-71 1990-94 1972-76 1975-84 1983 1978-80 1985-92 1968-71 1982-88 1982-86 1987-91 1994 Male 2.1 2.8 5.2 5.8 4.4-11.1 7-13 8.3 8.4 10 12.6 17.2 17.5 14.7 19.4 20.7 28.3 30.5 44.0 49.9 Female 0.8 0.7 0.2 2.1 1.2-3.8 1-2 4.7 2.3 2.9 2.8 0.8 3.2 7.0 4.0 4.3 3.3 4.9 6.4 4.8 Reference 33 34 35 36 37 38 39 40 41 40 35 16 42 35 40 43 40 40 44 would increase steeply with time, making early childhood exposure of great importance.50 However, there are clear indications that mineral fibres clear from the lung, albeit with different half lifes for the different types of asbestos. Chrysotile has the shortest half life, and crocidolite is generally accepted to have the longest. The half life of crocidolite has been estimated to 7-8 years.51 This clearing of fibres would, at least theoretically, tend to actually decrease the risk of mesothelioma and other diseases with time. In conclusion, the value of the cubic residence-time formula is in practice low and it should not be used for extrapolations, at least not at the extreme ends of exposure. The latency time varies in different cohorts, and is dependent on how long a cohort is followed up. In 370 necropsy cases from Italy, latency time could be calculated in 312.52 Latency time was also dependent on exposure, varying from 29.6 years for insulators (with the highest exposure) to 51.7 in women with domestic exposure. The threshold value and the background concentration There have been strong arguments for the existence of a threshold value (a minimal expo sure required for development of a mesothelioma).53-55 In most studies, several patients with mesothelioma do not report any occupational or other exposure to asbestos,33 56-60 and thus there seems to be a small spontaneous basal or background inci dence of the tumour. In a large study from England, consisting of 185 cases and 159 con trols who were very carefully interviewed, 5% of the cases (and 27% of the controls) seemed not to have any kind of exposure to asbestos.59 This included domestic and even residential exposure. However, it is of course possible that some of these background cases might in fact be due to occupational, domestic, or even envi ronmental exposure, unknown to (or forgotten by) the patients themselves. There are authors who claim that the presumed background level must be very low, and retrospective searches for the tumour in the medical literature yield no convincing cases of mesothelioma before 1946,61 although such negative evidence is of questionable value. McDonald and McDonald, in a recent review, estimated the background level to be 1-2/ million/year; they came to this figure by extrapolating backwards from epidemiological studies from various countries.62 Malignant mesothelioma can occur in children,63 and such cases can be considered as proof of non-asbestos (spontaneous) aetiology, as the latency time with necessity must be very short in these cases. Asbestos fibres have, how ever, been reported in the lungs of children, even in stillborn ones,64 65 showing that asbestos fibres spread in the human body and even pen etrate through the placenta. Even if a latency time of only a few years is extremely rare in mesothelioma related to asbestos,49 it can occur. There are some published examples of latency times of only 5 years.66 Mesotheliomas also occur in animals, from baboons67 and domestic dogs68-70 to fish.71 Dogs are exposed environmentally to asbestos just like their human masters, which might explain some of the tumours,68 69 72 but in fish it would be difficult to blame asbestos for the tumour. Thus, as in other animals, there is probably a background level of spontaneous mesothelio mas in humans. Levels of exposure Although many authors write about low level exposure to asbestos, there is rarely a definition of this term. In fact, in many articles low level exposure seems to be synonymous to nonoccupational exposure, which, as described later, is certainly not true in many cases. Occu pational as well as non-occupational exposure can be anything from very heavy to very low. Occupational exposure to asbestos It must be realised that occupational exposure to asbestos occurs or has occurred not only in the "classic" industries, such as asbestos mines and factories, shipyards, insulating business, asbestos cement industry, building and con struction etc, but also in very many other occu pations and trades. Examples are pulp and paper industry,73 oil refineries,74 electrical industry,75 jewellery workers,76 sugar refineries,77 and cigarette filter workers.78 Sea men and fishermen can have been exposed to asbestos used as insulation in their boats. In the reprocessed textile industry, bags heavily con taminated with asbestos could be reused for various other purposes, for instance covering heaps of rags; in an Italian investigation of such an industry, mesotheliomas and lung cancer were found to be fairly common among rag sorters.79 Given the extensive use of the mineral, many people have been occupationally exposed to asbestos. This exposure can have been only brief but perhaps intense during that short period. In many or most instances the workers have no idea of the exposure and it can be impossible or almost impossible to elucidate it. Also, the level of exposure is often very difficult to estimate, should the information be avail able. Mesothelioma Downloaded from oem.bmj.com on 29 June 2007 509 Table 4 Reported mesothelioma cases after exposure to "asbestos in place" Occupation or exposure School teachers Attended school Office clerk Female office worker Asbestos insulation at home Comments 9/487 patients with mesothelioma 1 case each 4 cases 6/262 patients with mesothelioma Reference 99 100-103 104 102 102 105 106 It has been claimed that up to 1000 prema ture deaths from lung cancer or mesothelioma will occur in the future among school children from schools where asbestos was used in the walls50--calculations which, however, had to be built on extrapolations and assumptions. Sev eral case reports have been published on patients with mesothelioma, in which the only exposure to asbestos that was reported was "in place" (table 4). From various cohorts with such exposure, significant increases in radio logical findings from the lungs--such as pleural plaques--have also been reported, indicating exposure ot asbestos, but these results are not undisputed and there is a probable overdiagno sis, as control groups are missing.107 ENVIRONMENTAL MESOTHELIOMAS FROM LOCAL DEPOSITS OF FIBROUS MINERALS "Endemic pleural plaques" were first described from Finland and since then many such findings have been reported. In these areas, there are small local pockets of asbestos which sometimes have been quarried, often for generations, for some local use. The most common use is whitewashing of houses with tremolite, which has resulted in an extremely high incidence of mesothelioma in some villages (table 5). When the exposure is due to whitewashing of the houses the risk will disap pear when this procedure is stopped, but due to the long latency time this will take many decades.122 A non-asbestos fibre, the zeolite erionite, has been found in some Turkish villages. Roads, buildings, etc, can contain this fibre in small amounts. Erionite is even more dangerous than crocidolite and the incidence of mesotheliomas in these unfortunate villages is extremely high. Endemic plaques are of interest also in other countries, as many people born in these places and living there in their childhood and youth now have moved to other places, taking with them not only the plaques but also the risk of mesothelioma.131 Concentrations of exposure OCCUPATIONAL CONCENTRATIONS The concentration of exposure which the first workers exposed to asbestos have experienced can only be guessed. Estimated or recreated values from the past suggest fibre concentra tions from 25 up to occasional values of 1000 2000 fibres/ml. With his own recalculations, Harries in 1970 estimated the fibre concentra tions in the dockyard in 1951 as follows: sprayed asbestos insulation 171-322 fibres/cc; stripping asbestos 334; sweeping 353; adjacent passage 83; and in the passageway to the shower 25. Bagging debris 564; pipe lagging 194-200; removal of pipe lagging 171. Snap samples showed values to up to 1000-2000 fibres/cc.132 These values are similar to the ones published by McMillan in 1983, who recreated values from the past: engine room 88; delag ging in boilers room 171; bagging debris 353 f/ml.133 Measurements from working places in the 1960s often showed peak doses of 20 fibres/ml and much less in more recent years. Where asbestos is still used, many countries have adopted a concentration of1-2 fibres/ml as the upper legal concentration of exposure. These figures should be compared with the few avail able non-occupational measurements (table 6). A problem with the legal concentration is that most asbestos use today occurs in developing countries, many of which have adopted stand ards which they cannot enforce. As a result, actual exposures may be much higher than the standard in these countries. NON-OCCUPATIONAL CONCENTRATIONS The fibre concentrations in domestic exposure might in fact be as high as in occupational exposure. Brushing clothes might give peaks of >100 fibres/ml.53 Ordinary vacuum cleaning is not effective in removing asbestos fibres, which Table 5 Local deposits of mineralfibres (asbestos or erionite), occurrence ofplaques, and of malignant mesothelioma Country or area Afghanistan Austria Bulgaria Corsica Cyprus Czechoslovakia Finland Greece: Metsovo Type offibre Tremolite Tremolite Anthophyllite Tremolite Tremolite Tremolite Unknown Anthophyllite Tremolite SW Aridea, Macedonia New Caledonia x2 South Africa Turkey USSR Amosite crocidolite Tremolite Erionite Unknown *12 cases/145 000 inhabitants/10 y. fOccurrence in local area. Plaques (% of investigated inhabitants) -- 5.3 2.8 women 5.6 men 41 (>50 y) 2.7-6.6 6.5-9.0 46.9 24.2 (>40 y) 2.5-6.6 1.2-25 65 Locally high Mesothelioma risk (x/106/y) -- Not increased Not increased High High -- Not increased 280 140 (1985-94) High 8.3* 300f High High Comment Case report only Vineyard and field workers Tobacco growers General pollution General pollution Farmers White washing houses White washing houses Population around mine White washing houses Farmers Reference 108 109 110 111 112 113 114 115 116 117118 119-121 122 123 124 125 4 126-128 19, 128, 129 130 Downloaded from oem.bmj.com on 29 June 2007 510 Hillerdal Table 6 Fibre concentrations in air and lungs with non-occupational exposure to asbestos Type of exposure Fibres in air (f/ml) Fibers in lungs (jx106/g dry tissue) Reference Domestic exposure (paraoccupational) Near asbestos mines, factories, etc: In Wittenoom: When mine operating After closure of mine Farmers near mine in Canada Local asbestos findings: Greece: In the yard of an abandoned house In a newly whitewashed room Corsica New Caledonia: While sweeping floor Road dust clouds Asbestos in place: Teacher's aid Female office worker 0.5 0.01-0.21 0.01 0,02-17.9 * 78 0.06-0.67 5.3-319.5 0.0049 1.2-26.8 21 11 0.0043 31 (TEM) 11 102 48 134 123 114 124, 125 102 103 *39 ng/m3 (100 times higher than controls). TEM = transmission electron microscopy. Fibres in the lungs of patients with non-occupational mesothelioma As has been mentioned, asbestos fibres can be found in the lungs of the general population without any known exposure. In Germany, the upper normal limit was estimated to be 300 000 fibres/g dry lung. In a large study of 324 malignant mesotheliomas, from which 46 lung samples were available, it was found that even at a fibre concentration of 100 000-200 000 fibres/g, there was a fivefold increased risk of mesothelioma, which was significant.136 In this study (which has unfortunately only been pub lished in German) as in many others, the mean number of fibres in the lungs of patients with malignant mesothelioma is much higher than the normal values, but there are usually patients with values that lie within the normal level.25 137-141 can remain for years in the house and be airborne again whenever disturbed. Thus, domestic exposure is not low exposure. Environmental concentrations in the villages where whitewashing occurs, low values are reported when there is no disturbance, but in a newly whitewashed room and while sweeping floors, the concentrations can be quite consid erable (table 6). With asbestos in place, as long as the asbes tos is undisturbed the concentrations in the air are very low (zero or hardly measurable: <0.001 fibres/ml), but once deterioration takes place values can go up to 15 fibres/ml, and when being removed there can be even higher values. In schools in the United States, the mean concentration was estimated to be 0.003 fibres/ml, and in federal buildings 0.006.135 Thus, with so called non-occupational expo sure, the typical exposure is a low or very low, almost unmeasurable, background concentra tion, but occasional high exposure when there is a disturbance of some kind. It follows, firstly, that retrospective estimation of cumulative exposure from history alone is an impossibility in most cases; but secondly, and perhaps more importantly, that any person living or working in (or even temporarily visiting) buildings where asbestos has been used in construction or otherwise might well have been exposed to high concentrations of airborne asbestos fibres once or many times in their lives, and in most instances unknowingly. This includes most of us! A better way of estimating lifelong exposure might be analysis of fibres in the lungs, but as already mentioned fibres do clear from the lungs. How big the differences in clearance are between people is unknown. Thus, the correla tion between lifetime cumulative exposure and fibre concentrations in the lungs is not excellent, but the findings from the lungs prob ably give a better estimation of exposure than even a careful retrospective analysis of the patient's history, at least in low grade exposure. In most studies, there is a clear dose-response relation between exposure and the number of fibres in the lungs.11 12 46 48 51 Discussion Any asbestos fibre found in a lung must have been inhaled. As far as is known, no truly unexposed group can be found in the world. There is no proof of a threshold value--that is, a minimal lower limit below which asbestos fibres cannot cause the tumour--and thus it is plausible that even such low exposure can cause mesothelioma (even if the risk is extremely low). Patients with mesothelioma whose lungs show fibre concentrations within the normal range cannot be dismissed as back ground cases,--that is, not due to asbestos. The only way to prove such a hypothesis would be to compare the incidence of mesothelioma in a group with such background exposure with the incidence in a truly non-exposed group. This is not possible, as no such group can be found. It is nevertheless possible that there is a background level of mesothelioma,--that is, that the tumour can occur even in the complete absence of asbestos (or erionite) fibres. How ever, the data reviewed here indicate that if so, this background level must be very low-- probably much <1 case/million people/year. This figure comes from studies of industrial ised countries, where background exposure to asbestos is unavoidable. What the true figure is can only be guessed. What, then, are the consequences for the public health? From the studies of nonoccupational exposures it seems probable that the occasional high level exposure situations are the ones that are most important. Although the background, hardly measurable, concentra tions of fibres in the air cannot be completely dismissed, the cumulative risk of these expo sures is probably minor--and what is more, there is no way to reduce these concentrations. It is the high concentration situations which should be avoided. By knowing where asbestos occurs, such risks could be identified. Any source of pollution by asbestos which releases significant amounts of fibres should be elimi nated as soon as it is discovered, using correct equipment and techniques. Correct techniques are also necessary whenever rebuilding or tear ing down of structures containing asbestos to avoid asbestos pollution of the environment. If 512 Downloaded from oem.bmj.com on 29 June 2007 Hillerdal 67 Fortman JD, Manaligod JR, Bennett BT. Malignant mesothelioma in an olive baboon (Papio anubis). Lab AnimalSci 1993;43:503-5. 68 Glickman LT, Domanski LM, Maguire TG, et al. Mesothe lioma in pet dogs associated with exposure of their owners to asbestos. Environ Res 1983;32:305-13. 69 Harbison ML, Godleski JJ. Malignant mesothelioma in urban dogs. Vet Pathol 1983;20:531^0. 70 Smith DA, Hill FW. Metastatic malignant mesothelioma in a dog. J Comp Pathol 1989;100:97-101. 71 Herman RL. Mesothelioma in rainbow trout, Salmo gairdneri Richardson. Journal ofFish Diseases 1985;8:373-6. 72 Mirabella F. Studio epidemiologico del mesotelioma spontaneo canino: sua possibile utilita' in riferimento alla patologia umana. Med Lavoro 1987;78:28-32. 73 Jarvholm B, Malker H, Malker B, et al. Pleural mesothe lioma and asbestos exposure in the pulp and paper industries: a new risk group identified by linkage of official registers. Am J Ind Med 1988;13:561-7. 74 Finkelstein MM. Asbestos-associated cancers in the Ontario refinery and petrochemical sector. AmJ Ind Med 1996;30: 610-15. 75 Imbernon E, Goldberg M, Bonenfant S, et al. Occupational respiratory cancer and exposure to asbestos: a case-control study in a cohort of workers in the electricity and gas industry. Am J Ind Med 1995;28:339-52. 76 Kern DG, Hanley KT, Roggli VL. Malignant mesothelioma in the jewelry industry. Am J Ind Med 1992;21:409-16. 77 Maltoni C, Pinto C, Valenti D, et al. Mesotheliomas follow ing exposure to asbestos used in sugar refineries: report of 11 Italian cases. In: A Mehlman, A Upton, eds. Advances in modern environmental toxicology. Vol XXII. The identification and control ofenvironmental and occupational diseases: hazards and risks of chemicals in the oil refining industry. Prinecteon, NJ: Princeton Scientific, 1994:629-34. 78 Talcott JA, Thurber WA, Kantor AF, et al. Asbestosassociated disease in a cohort of cigarette-filter workers. N EnglJ Med 1090;321:1220-3. 79 Quinn MM, Kriebel D, Buiatti E, et al. An asbestos hazard in the reprocessed textile industry. Am J Ind Med 1987;11: 255-67. 80 Kane MJ, Chahinina AP, Holland JF. Malignant mesothe lioma in young adults. Cancer 1990;65:1449-55. 81 Magnani C, Terracini B, Ivaldi C, et al. A cohort study on mortality among wives of workers in the asbestos cement industry in Casale Monferrato, Italy. Br J Ind Med 1993;59:779-84. 82 Schneider J, Woitowitz HJ. Tumors linked to paraoccupational exposure to airborne asbestos. Indoor Built Environment 1996;5:67-75. 83 Botha JL, Irwig LM, Strebel PM. Excess mortality from stomach cancer, lung cancer, and asbestosis and/or mesothelioma in crocidolite mining districts in South Africa. Am J Epidemiol 1986;123:30-40. 84 Musk AW, de Klerk N, Eccles JL, et al. Wittenoom, Western Australia: a modern industrial disaster. Am J Ind Med 1992;21:735^7. 85 Harington JS. Mesothelioma among workers in the Quebec chrysotile mining and milling industry. Am J Ind Med 1992;22:925-6. 86 Churg A. Lung asbestos content in long-term residents of a chrysotile mining town. Am Rev Respir Dis 1986;134:1257. 87 Hain E, Dalquen P, Bohlig H, et al. Katamnestische Untersuchungen zur Genese des Mesotelioms. Bericht fiber 150 Falle aus dem Hamburger Raum. Internationales Archiv fur Arbeitsmedizin 1974;33:15-37. 88 Smith RJ. Briefing. Science 1979;204:285. 89 Longo WE, Rigler MW, Slade J. Crocidolite asbestos fibers in smoke from original Kent cigarettes. Cancer Res 1995;55:2232-5. 90 Moorcroft JS, Duggan MJ. Rate of decline of asbestos fibre concentration in room air. Ann Occup Hyg 1984;28:453-7. 91 Suta BE, Levine RJ. Non-occupational asbestos emissions and exposures. In: L Micahels, SS Chissick, eds. Asbestos. Vol 1. Properties, applications, and hazards. New York: John Wiley, 1979:171-205. 92 Cunningham HM, Moodie CA, Lawrence GA, et al. Chronic effects of ingested asbestos in rats. Arch Environ Contam Toxicol 1977;6:507-13. 93 Sebastien P, Masse R, Bignon J. Recovery ofingested asbes tos fibers from the gastrointestinal lymph in rats. Environ Res 1980;22:201-16. 94 Cook MP, Olson GF. Ingested mineral fibers: elimination in human urine. Science 1979;204:195-8. 95 Boatman H. Use of quantitative analysis of urine to assess exposure to asbestos fibers in drinking water in the Puget Sound area. Environ Health Perspect 1983;53:131-9. 96 Webber JS, Syrotynski S, King MV. Asbestos-contaminated drinking water: its impact on household air. Environ Res 1988;46:153-67. 97 Greenberg M. Malignant mesothelioma in paper mill workers: where might the asbestos have come from? AmJ Ind Med 1996;30:641. 98 Molloy LB. Asbestos in place in metropolitan New York. Ann NYAcad Sci 1991;643:614-21. 99 Anderson HA, Hanrahan L, Schirmer J, et al. Mesothelioma among employees with likely contact with in-place asbestos-containing building materials. Ann N Y Acad Sci 1991;643:550-72. 100 Dodson RF, O'Sullivan M, Corn C. Technique dependent variations in asbestos burden as illustrated in a case of nonoccupational mesothelioma. Am J Ind Med 1993;24:235^0. 101 Huncharek M. Occult asbestos exposure. Am J Ind Med 1991;20:713-14. 102 Roggli VL, Longo WE. Mineral fiber content of lung tissue in patients with environmental exposures: household contacts v building occupants. Ann N Y Acad Sci 1991;643:511-18. 103 Srebro S, Roggli VL. Asbestos-related disease associated with asbestiform tremolite. Am J Ind Med 1994;26:80919. 104 Lilienfeld DE. Asbestos-associated pleural mesothelioma in school teachers: a discussion of four cases. Ann NYAcad Sci 1991;643:454-8. 105 Stein RC, Kitajewska JY, Kirkham JB, et al. Pleural mesothelioma resulting from exposure to amosite asbestos in a building. Respir Med 1989;83:237-9. 106 Dodoli D, Del Nevo M, Fiumalbi C, et al. Environmental household exposures to asbestos and occurrence of pleural mesothelioma. Am J Ind Med 1992;21:681-7. 107 Hillerdal G. Pleural plaques: incidence and epidemiology, exposed workers and the general population. Indoor Built Environment 1997;6:86-95. 108 Voisin C, Marin I, Brochard P, et al. Environmental airborne tremolite asbestos pollution and pleural plaques in Afghanistan. Chest 1994;106:974-6. 109 Neuburger M, Grfindorfer W, Haider M, et al. Umweltbedingte endemische Pleuraplaques. Zentrblatt fur Bakteriologie und Hygiane, Abteil Original 1978;167:391-404. 110 Neuburger M, Kundi M, Friedl HP. Environmental asbes tos exposure and cancer mortality. Arch Environ Health 1984;39:261-5. 111 Burilkov T, Babadjov L. Ein Beitrag zum endemischen Auftreten doppelseitiger Pleuraverkalkungen. Praxis fur Pneumologie 1970;24:433-8. 112 Zolov C, Bourilkov T, Babadjov L. Pleural asbestosis in agricultural workers. Environ Res 1967;1:287-92. 113 Rey F, Boutin C, Steinbauer J, et al. Environmental pleural plaques in an asbestos exposed population of Northeast Corsica. Eur RespirJ1993;6:978-82. 114 Rey F, Viallat JR, Boutin C, et al. Les mesotheliomes environnementaux en Corse du nord-est. Rev Mal Respir 1993; 10:339-45. 115 McConnochie K, Simonato L, Mavrides P, et al. Mesothe lioma in Cyprus: the role of tremolite. Thorax 1987;43: 342-7. 116 Marsova M. Beitrag zur Atiologie der Pleuraverkalkungen. Zeitschrift fur Tuberkulose 1964;121:329-34. 117 Kiviluoto R. Pleural calcification as a roentgenologic sign of non-occupational endemic anthophyllite asbestosis. Acta Radiol 1960;(suppl 194). 118 Raunio V. Occurrence of unusual pleural calcification in Finland. Studies on atmospheric pollution caused by asbestos. Annales de Medicina Interna Fenniae 1966; 55(suppl 47). 119 Constantopoulos SH, Malamou-Mitsi VD, Goudevenos JA, et al. High incidence ofmalignant pleural mesothelioma in neighbouring villages of Northwestern Greece. Respira tion 1987;51:266-71. 120 Bazas T, Oakes D, Gilson JC, et al. Pleural calcification in northwest Greece. Environ Res 1985;38:239-47. 121 Langer AM, Nolan RP, Constantopoulos SH, et al. Associ ation of Metsovo lung and pleural mesothelioma with exposure to tremolite-containing whitewash. Lancet 1987;i: 965-7. 122 Sakellariou K, Malamou-Mitsi V, Haritou A, et al. Malignant pleural mesothelioma from non-occupational exposure in Metsovo (north-west Greece): slow end of an epidemic? Eur RespirJ 1996;9:1206-10. 123 Sichletidis L, Daskalopoulou E, Chioros D, et al. Five cases of pleural mesothelioma with endemic pleural calcifications in a rural area in Greece. Med Lavoro 1992;83:259-65. 124 Goldberg P, Luce D, Billon-Galland MA, et al. Role potentiel de l'exposition environnementale et domestique a la tremolite dans le cancer de la plevre en Nouvelle Caledonie. Rev Epidemiol Sante Publ 1995;43:444-50. 125 Luce D, Brochard P,Quenel P, et al. Malignant pleural mesothelioma associated with exposure to tremolite. Lancet 1994;ii:1777. 126 Baris YI, Sahin AA, Ozesmi M, et al. An outbreak of pleu ral mesothelioma and chronic fibrosing pleurisy in the vil lage of Karain/Urgfip in Anatolia. Thorax 1978;33:181-92. 127 Baris YI, Bilir N, Artvinli M, et al. An epidemiological study in an Anatolian village environmentally exposed to tremolite asbestos. BrJ Ind Med 1988;45:838-40. 128 Selcuk Z, Coplfi L, Emri S, et al. Malignant pleural mesothelioma due to environmental mineral fiber exposure in Turkey. Analysis of 135 cases. Chest 1992;102;790-6. 129 Artvinli M, Baris YI. Environmental fiber-induced pleuropulmonary diseases in an Anatolian village: an epidemio logic study. Arch Environ Health 1982;37:177-81. 130 Ginzburg EA, Silova MV, Kornejeva MJ, etal. Rontgenologie der nichtberufs-bedingten Asbestose der Pleura. Radiologia Diagnostica 1973;14:307-12. 131 Boman G, Schubert V, Svane B, et al. Malignant mesothe lioma in Turkish immigrants to Sweden. Scand J Work Environ Health 1982;8:108-12. 132 Harries PG. The effect and control of diseases associated with exposure to asbestos in a naval dockyard [MD Thesis]. Alverstoke, Gosport, UK: Institute of Naval Medicine, 1970. 133 McMillan GHG. The health ofwelders in naval dockyards. J Occup Med 1983;25:727-30. 134 Churg A. Analysis of lung asbestos content. Br J Ind Med 1991;48:649-52.