Document 1QnZbDyQ47O4wyk0YjmayRzzX
Ksspidon
Clinical Investigations
Respiration DOI: 10.1159/000163443
Received: December 31, 2007
Accepted after revision: July 11,2008 Published online: October 9, 2008
Mortality from Occupational Exposure to Relatively Pure Chrysotile: A 39-Year Study
L. Sichletidis D. Chloros D. Spyratos A.-B. Haidich I. Fourkiotou M. Kakoura D. Patakas
Laboratory for the Study of Environmental Diseases, Pulmonary Clinic of Aristotle University of Thessaloniki, Thessaloniki, Greece
Key Words Asbestos Epidemiology Lung cancer Mesothelioma Chrysotile
Abstract Background: Asbestos exposure is related to serious ad verse health effects. However, there is disagreement about the relationship between chrysotile exposure and mesothe lioma or lung cancer. Objectives: Our aim was to investigate the mortality rate among workers exposed to relatively pure chrysotile in an asbestos cement factory. Patients and Methods: In an asbestos cement plant opened in 1968, we prospectively studied all 317 workers. A quantity of 2,000 tons of chrysotile, with minimal amphibole contamination, was used annually until 1 January 2005. Asbestos fiber con centration was measured regularly. Date and cause of death were recorded among active and retired workers. Results: Asbestos fiber concentration was always below permissible levels. Fifty-two workers died during the study. The cause was cancer in 28 subjects; lung cancer was diagnosed in 16 of them. No case of mesothelioma was reported. Death was attributed to cardiovascular diseases in 23 subjects and to liver cirrhosis in 1. Overall mortality rate was significantly lower than that of the Greek general population, standard ized mortality ratio (SMR) was 0.71 (95% CI 0.53-0.93). Mor tality due to cancer was increased (SMR 1.15, 95% CI 0.77-
1.67), mainly due to lung cancer mortality (SMR 1.71, 95% CI
0.98-2.78), but not significantly. Conclusions: Occupational
exposure to relatively pure chrysotile within permissible lev
els was not associated with a significant increase in lung can
cer or with mesothelioma. Decreased overall mortality of
workers indicates a healthy worker effect, which - together
with the relatively small cohort size - could have prevented
small risks to be detected.
Copyright 2008 S. Karger AG, Basel
Introduction
The term asbestos is used to describe a group of natu ral mineral fibers, which present extensively in the earth's crust and have silica as their main component [1]. Based on their physical properties, asbestos fibers are divided into 2 main types: (1) serpentines (flexible curly fibers; chrysotile is the main representative) and (2) amphibole forms (straight, needle-like fibers; crocidolite, amosite, anthophyllite, tremolite and actinolite are included in this group). Asbestos fibers inhalation, even in small quantities, is considered dangerous. The Commission of the European Communities recently adopted a directive (2003/18/EC) that obliges member states to prohibit the marketing and use of chrysotile asbestos [2]. The prohibi tion became effective on 1 January 2005.
KARGER
Fax +41 61 306 12 34 E-Mail karger@karger.ch www.karger.com
2008 S. Karger AG, Basel 0025-7931/08/0000-0000$24.50/0
Accessible online at: www.karger.com/res
Dr. Lazaros Sichletidis G. Papanicolaou Hospital GR-57010 Exochi, Thessaloniki (Greece) Tel. +30 23 1099 2363, Fax +30 23 1330 7316, E-Mail sichlet@med.auth.gr
This directive was received in the light of recent stud ies, which concluded that chrysotile, even though less toxic than amphiboles, could induce mesotheliomas and lung cancer [3-6]. On the other hand, studies on asbestos cement workers have shown that mesothelioma is associ ated with the use of amphibole asbestos [7] and that ex posure to moderate chrysotile concentrations does not increase the risk of lung cancer [8, 9]. It was argued that the frequent contamination of chrysotile deposits by am phibole fibers might contribute to the carcinogenic ef fects of occupational exposure [10, 11]. A recent review article concluded that available data could not prove a causative relationship between exposure to pure chrysotile and mesothelioma [12],
The reasons for these conflicting conclusions may be a number of confounding factors such as chrysotile pu rity, job type, exposure duration and intensity, smoking history as well as epidemiologic methodology [13].
To study mortality among workers in an asbestos ce ment plant, where chrysotile with minimal amphibole contamination was used, we undertook a long-standing, prospective cohort study. Specifically, we investigated workers' mortality under low concentrations of relatively pure chrysotile fibers, a subject that has not been studied extensively.
We believe that the present study contributes to a more rational policy regarding asbestos usage and surveil lance.
Subjects and Methods
Occupational Environment In an asbestos cement products plant that has operated since 1968, asbestos was used until 1 January 2005, when the European Union banned chrysotile asbestos usage; 317 subjects worked there. This plant is located near Thessaloniki, Greece, and its sur face area is approximately 244,000 m2. Two thousand tons of raw asbestos were used annually. Only chrysotile asbestos that was imported from Canada (Johns Manville Corporation, Lake As bestos Mine) and Russia (Ural Asbestos Mining and Ore Dressing Company, Orenburgasbest) up to 1983 and then obtained from a mine in Zidani, Greece, was used in the plant. The average chrys otile contamination with amphiboles was 0.5% while, in rare cas es, the maximum concentration approached 3%. Asbestos cement products contained chrysotile in a percentage of about 8-12%, whereas 25,000-30,000 tons of asbestos cement was produced an nually.
Measurement ofAirborne Fiber Concentration Airborne asbestos fiber concentration in the workplace was measured regularly by the Environmental Control Laboratory of the factory. Measurements took place in 32 different points 4 times annually, and the mean measurement period was 8 h every
time. A specific air pump (Casella ARC 124; Casella CEL, Bed ford, UK) was used for air suction, while air supply was fixed at 1 liter/min. Millipore type AA filters with 0.8-^m pore diameter were used (Millipore Corporation, Bedford, Mass., USA). Then the filters were checked with contrast phase microscope (X450). Asbestos fibers were counted in 100 different optical fields of fil ter surface. Only the fibers with length >5 ^m, diameter <3 ^m and ratio length/diameter 63/1 were counted [14]. Airborne fiber concentration was expressed as fibers/cm3. Fiber concentration was reduced from 6.49 to 1.7 fibers/cm3 during the time period 1968-1983. From 1984, concentration was always below 1 fiber/ cm3 (mean values reduced from 0.8 to 0.07) according to directive 83/477/EU [14],
Concentration of asbestos fibers in the plant ranged from 6.49 to 3.51 fibers/cm3 before 1980, while after that year stricter safety measures were undertaken causing a reduction in fiber concen tration to 1.70-1.13 fibers/cm3 during the period 1981-1983. After 1983, concentration was constantly below 1 fiber/cm3 as the Eu ropean Union directive imposed.
Subjects We followed 317 workers from 1968, when the factory started its operation, until 31 December 2006. Occupation for at least 5 years was an inclusion criterion for the study. On 31 December 2006, there were still 41 workers in the plant, which now bases its production on nonasbestos materials. Of the remaining 276 sub jects, 224 had retired and were still alive, while 52 workers had died. There was a permanent occupational medical doctor who fol lowed workers annually (clinical examination, spirometry, radio graphs). Smoking history was recorded at the beginning of em ployment and then annually. All retired persons were followed on a regular basis. Date and cause of death were recorded, among ac tive workers as well as retired subjects, based on death certificates, occupational medical records and personal communication be tween occupational doctor and general practitioner. The Medical Ethics Committee of the G. Papanicolaou Hospital in Thessalon iki approved the study protocol. The mean age at hire was 31.9 8 9.1 years for the 265 subjects who were still alive at the end of the study and 39.7 8 7.9 years for the workers who had died. Total follow-up time was 9,537 person years. Taking into account only the person years observed after age 40, the most risky period for developing lung cancer, we cal culated that 72.5% (6,918 person years) of our follow-up time re ferred to a critical period for these subjects to develop cancer. The mean age and years of occupation 8 standard deviation for the 265 subjects who were still alive at the end of the study was 63.1 8 12.1 and 19.3 8 7.6 years, respectively. Workers who had died had a mean age of 63 8 9.5 years, while they had worked in the plant for 16.7 8 5.6 years. Among the workers who were still alive at the end ofthe study, 123 were current smokers and 65 were ex-smokers with a mean exposure of 24.5 8 24.6 pack years. Among subjects who died, 33 were active smokers and 10 ex-smokers with a smoking history of 38.4 8 30.3 pack years. Quantitative exposure to asbestos was 42.9 8 27.8 fibers/cm3 X years of exposure for those who were still alive and 52.4 8 23.7 for workers who had died. Detailed de scriptive statistics for workers are shown in table 1.
2 Respiration
Sichletidis/Chloros/Spyratos/Haidich/ Fourkiotou/Kakoura/Patakas
Table 1. Descriptive statistics of age distribution, smoking history and chrysotile exposure of workers
Alive mean 8 SD median range
interquartile range
Deceased mean 8 SD median range
interquartile range
Age at hire, years Age at the end of the study, years Smoking, pack years1 Occupation, years Fibers/cm3 X years
31.98 9.1 63.18 12.1 34.78 22.4 19.3 8 7.6 42.98 27.8
31 63 44.5 20 44.5
1 Only smokers and ex-smokers were considered.
15-56 14-93 0.6-85.2
5-33 0.6-85.2
25-38 55-72 15.3-71.1 14-25 15.6-66.6
39.78 7.9 63.08 9.5 47.68 26.4 16.78 5.6 52.48 23.7
41 63.5 41.5 16.5 54.7
23-55 42-79
2-115 5-30 1.4-84.1
36-45 56-70 31.5-60 12-19.8 37.4-71.9
Table 2. Observed and expected number of deaths and SMR adjusted for age and sex among workers during the follow-up period 1968-2006
Cause of death
All causes Malignant neoplasms Lung neoplasms Diseases of circulatory system Other causes of death1
1 Cirrhosis.
Observed
52 28 16 23
1
Expected
73.19 24.24
9.35 29.82 19.12
SMR
0.71 1.15 1.71 0.77 0.05
95% CI
0.53-0.93 0.77-1.67 0.98-2.78 0.49-1.16 0.01-0.29
Statistical Analysis Standardized mortality ratios (SMR) were calculated as ratios between observed and expected numbers of deaths, adjusting for age and sex, according to life tables for the Greek male population of 2003 (General Secretariat of National Statistical Service of Greece: www.statistics.gr). Confidence intervals (CI) were calcu lated using exact methods [15]. Follow-up for each individual started at 1968 or at the date of engagement if they started work after 1968. The follow-up was up to 31 December 2006 or the death date if that occurred before that date. Exposure time, pack years smoked and asbestos fiber concen tration were considered as candidate predictors for death since first exposure to chrysotile. Predictive analyses used Cox models, after ascertaining that proportionality of hazards was not vio lated [16]. Unadjusted and adjusted hazard ratios for smoking and age are presented. The adjustment for smoking was undertaken by adding the pack years smoked as a covariate in the model. Mul tivariate models were built using backward elimination of vari ables according to likelihood ratio criteria, starting with all vari ables with p < 0.1 in univariate models. Also, the forward selec tion of variables according to likelihood ratio criteria was applied to see if the final model differed. Three outcomes were consid ered: all-cause deaths, malignant neoplasm deaths and lung can cer deaths. All tests of statistical significance were two-sided and performed using Stata Statistical Software (version 8.0; Stata Cor poration, College Station, Tex., USA) and SPSS (version 14.0; SPSS Inc., Chicago, Ill., USA).
Results
Fifty-two deaths were recorded during the study. Twenty-three of them were attributed to cardiovascular diseases, 28 to malignant neoplasms and 1 to liver cirrho sis. Of 28 cancer deaths, 16 were due to lung cancer (12 squamous cell, 2 adenocarcimonas and 2 small cell lung cancer), while the origin ofthe remaining were as follows: 4 stomach, 2 pancreas, 2 nervous system, 2 hematologic system, 1 colon and 1 skin.
The overall mortality among male workers was sig nificantly decreased (SMR 0.71, 95% CI 0.53-0.93; ta ble 2). However, death from malignant neoplasms was in creased in males but not significantly different compared to the Greek male population (SMR 1.15, 95% CI 0.77 1.67) mainly due to lung cancer mortality (SMR 1.71, 95% CI 0.98-2.78). Relative risk for lung cancer was estimated to be 6.61 (3.30-11.84) for smokers compared to non smokers and 4.55 (1.24-11.67) for ex-smokers compared to nonsmokers. Conversely, mortality rate due to circula tory diseases or to other causes was decreased (SMR 0.77 and 0.05, respectively).
Mortality from Occupational Exposure to Chrysotile
Respiration
3
Table 3. Predictors of death since first chrysotile exposure; unadjusted and adjusted analysis for age and smoking
Mean a
Hazard ratiob
p value
Hazard ratiob
Years of chrysotile exposure All-cause deaths Malignant neoplasm deaths Lung cancer deaths Alive
Pack years of smoking All-cause deaths Malignant neoplasm deaths Lung cancer deaths Alive
Fibers/cm3 X years All-cause deaths Malignant neoplasm deaths Lung cancer deaths Alive
16.71 (5.57) 16.57 (5.63) 15.38 (5.99) 19.26 (7.62)
38.44 (30.30) 48.54 (28.09) 50.13 (26.46) 24.47 (24.59)
52.36 (23.72) 56.76 (21.75) 57.89 (21.89) 42.91 (27.82)
0.92 (0.89- 0.96) 0.92 (0.88- 0.97) 0.90 (0.84- 0.97)
1.15 (1.05--1.26)c 1.27 (1.14--1.42)c 1.28 (1.12--1.48)c
0.99 (0.88--1.12)c 1.07 (0.91--1.27)c 1.10 (0.99--1.37)c
<0.001 0.003 0.005
0.002 <0.001
0.001
0.887 0.374 0.362
0.92 (0.89-0.96) 0.91 (0.87-0.96) 0.89 (0.83-0.96)
1.15 (1.05- 1.26)c 1.26 (1.13- 1.41)c 1.28 (1.12- 1.48)c
1.02 (0.90- 1.16)c 1.06 (0.89- 1.28)c 1.08 (0.85- 1.38)c
a Figures in parentheses are standard deviations. b Adjusting for age and smoking; figures in parentheses are 95% CI. c For every 10-unit increase; figures in parentheses are 95% CI.
p value
<0.001 0.001 0.002
0.003 <0.001
0.001
0.789 0.495 0.495
Workers with greater chrysotile exposure had better survival (table 3). Specifically, the risk of death from all causes and from malignant neoplasms was 8% lower for every year of chrysotile exposure (p = 0.001 and 0.003, respectively), while the risk of death from lung cancer was 10% lower for everyyear ofchrysotile exposure (p = 0.005). These associations remained after adjusting for age and smoking (p < 0.001, 0.001 and 0.002, respectively). Nota bly, the deceased workers had a heavier history of asbestos exposure than the living workers (tables 1 and 3).
Sichletidis et al. [17] studied 21,854 subjects of the gen eral population in northern Greece and found high prev alence of smoking (47.8% for males) among them. Spe cifically the subgroup of age 61-80 years, which was com parable with our population at the end ofthe study, showed 27.2% smokers, 32.9% ex-smokers and 39.9% nonsmokers. The corresponding percentages for our study group were 48.6% (n = 154), 23.7% (n = 75) and 27.8% (n = 88), respec tively. We should notice that consumption of cigarettes (pack years) of the living workers was less than that of the general population, while those who died had similar pack years as the general population.
As anticipated, smoking contributed significantly to all-cause mortality, all cancer mortality and specifically lung cancer mortality (p < 0.002 for all 3 comparisons; table 3). For every 10 pack years of smoking, the risk of death from all causes increased 1.15 times and the risk of
death from all cancers increased 1.27 times, while the risk of death from lung cancer increased 1.28 times. These as sociations remained after adjusting for age (p = 0.003, 0.001 and 0.001, respectively). Asbestos fiber concentra tion was not significantly associated with time to death since first exposure in any of the 3 outcomes (p = 0.887, 0.374 and 0.362, respectively).
In multivariate modeling, only smoking (p < 0.001) was a significant independent predictor for all 3 mortal ity outcomes. Additionally, the hazard ratio for lung can cer was 16.36 (95% CI 8.64-30.96, p < 0.001) after adjust ment for age and smoking.
Discussion
The present long-standing, prospective study showed that subjects exposed to relatively pure chrysotile asbestos, with airborne fiber concentrations much lower than the permissible upper limits, had no increased risk for meso thelioma and lung cancer. SMR for lung cancer was calcu lated to be 1.71 and attributed almost exclusively to ciga rette smoking. No case of mesothelioma was reported.
The major advantages of the study are that, for the pe riod 1968-2006, we followed a specific population under low level of relatively pure chrysotile exposure, while smoking history was recorded in a quantitative manner.
4 Respiration
Sichletidis/Chloros/Spyratos/Haidich/ Fourkiotou/Kakoura/Patakas
According to older studies, chrysotile asbestos was considered responsible for mesothelioma and lung can cer. Studies among workers in the textile industry (3,211 in the UK [4] and 1,247 in the USA [18]) and asbestos ce ment production (7,996 in Denmark [19] and 2,565 in the USA [20]), where chrysotile asbestos had variable con tamination with amphiboles, showed that SMR ranged between 1.17 and 2.25 for lung cancer. Moreover, propor tional mortality rate (per 1,000) for mesothelioma ranged between 2.5 and 20.7. Chrysotile contamination to a vari able degree was a constant characteristic of all studies that showed increased risk for lung cancer or mesotheli oma [21] .
In a recent study concerning 515 workers in China who had been exposed to pure chrysotile and followed for 25 years, the relative risk for lung cancer was calculated to be 6.6 compared to control workers, while only 2 cases of mesothelioma were diagnosed [6]. We should point out that exposure was heavy with airborne fiber concentra tion ranging from 4.5 to 7.6 fibers/cm3.
On the contrary, Newhouse and Sullivan [22] did not find increased risk for lung cancer (SMR 1.03) among 9,104 workers in the UK who had been exposed to chrysotile containing low concentration of amphiboles. A meta-analysis of 71 asbestos cohorts exposed to free as bestos fibers does not support the hypothesis that pure chrysotile causes mesothelioma [7]. In a clinicopathological study in the USA concerning 1,445 cases of malig nant mesothelioma with known exposure to asbestos, 268 had fiber burden analysis of lung tissue and chrysotile fibers were detectable in 36. In only 2 of these, chrys otile was the exclusive detectable asbestos type [11].
Hodgson and Darnton [23] tried to calculate the quan titative risk of mesothelioma and lung cancer in relation to asbestos exposure. They identified all cohort mortality studies for which quantified data on exposure were avail able and concluded that the specific risk of mesothelioma was in the ratio 1:100:500for chrysotile, amosite and crocidolite, respectively. For lung cancer, the conclusions were less clear-cut, suggesting a relative risk between chrysotile and the 2 kinds of amphibole fibers between 1:10 and 1:50.
It is suggested that the frequent contamination of chrysotile by amphibole fibers to variable degrees [10] as well as the intensity of exposure to chrysotile when used uncontaminated [6] are the most critical factors that pro voke malignant diseases of the chest. The same position was supported by Bernstein and Hoskins [24] in a recent review on the topic. They concluded that heavy and pro longed exposure to chrysotile could produce lung cancer,
while low exposures to pure chrysotile do not present a detectable risk to health.
This remark is supported by in vitro studies which showed that chrysotile clearance from the lungs is rapid, with clearance half times ranging between 0.3 and 11.4 days for fibers >20 pm in length. On the other hand, in halation clearance half time was 536 days for crocidolite and even longer for tremolite [25] .
In the present study, we did not record any case of me sothelioma and the mortality rate of the workers due to lung cancer did not differ from the general population. Based on the study of Hughes et al. [26], which dealt with asbestos cement workers with similar exposure level to chrysotile (40 fibers/cm3 X years), we estimated that the expected number of mesotheliomas would have been less than 1.
We corrected lung cancer mortality using duration of exposure, intensity of exposure (fibers/cm3 X years) and smoking history. Uni- and multivariate models were used and concluded that only smoking history was an inde pendent factor contributing to lung cancer mortality risk. The healthy worker effect could explain the lower overall mortality among the workers group compared with the control population. It is possible that individualized fac tors related to asbestos resistance might conceal the neg ative relationship between chrysotile exposure and rela tive risk for malignancies.
The decrease in risk with years of chrysotile exposure is a phenomenon that is difficult to explain. Diachronic reduction of fiber concentration in the factory has al ready been described in detail. The number of ex-smok ers grew in a parallel way. These 2 facts might be the main reasons for the observed decrease in risk with years of chrysotile exposure. On the other hand, it is true that de ceased workers had a heavier history of asbestos exposure than living workers.
Our results are in accordance with current studies about the health effects of chrysotile [12, 21]. Unfortu nately, usage of chrysotile is declining steadily in devel oped countries. Epidemiologic and toxicological data about organic and synthetic fibers, which may substitute chrysotile, showed that these fibers had physical and bio logical properties resembling amphiboles [27, 28].
We conclude that occupational exposure to relatively pure chrysotile under strict permissible levels does not increase the relative risk for mesothelioma and lung can cer. Decreased overall mortality of workers indicates a healthy worker effect, which - together with the relative ly small cohort size - could have prevented small risks to be detected.
Mortality from Occupational Exposure to Chrysotile
Respiration
5
References
1 Browne K: Asbestos-related disorders; in Parkes WR (ed): Occupational Lung Disor ders, ed 3. London, Butterworth-Heinemann, 1994, pp 411-504.
2 Directive 2003/18/EC of the European Par liament and of the Council of 27th March 2003 amending Council Directive 83/477/ EEC on the protection of workers from the risks related to exposure to asbestos at work. Off J Eur Union 2003;L 97:48 -52.
3 Dement JM, Harris RL Jr, Symons MJ, et al: Exposures and mortality among chrysotile asbestos workers. Part II. Mortality. Am J Ind Med 1983;4:421-433.
4 Peto J, Doll R, Hermon C, Binns W, Clayton R, Goffe T: Relationship ofmortality to mea sures of environmental asbestos pollution in an asbestos textile factory. Ann Occup Hyg 1985;29: 305-355.
5 Stayner LT, Dankovic DA, Lemen RA: Oc cupational exposure to chrysotile asbestos and cancer risk: a review of the amphibole hypothesis. Am J Public Health 1996;86: 179-186.
6 Yano E, Wang ZM, Wang XR, Wang MZ, Lan YJ: Cancer mortality among workers ex posed to amphibole-free chrysotile. Am J Epidemiol 2001;154:538-543.
7 Yarborough CM: Chrysotile as a cause of mesothelioma: an assessment based on epi demiology. Am Rev Toxicol 2006;36:165187.
8 Gardner MJ, Winter PD, Pannett B, Powell CA: Follow up study ofworkers manufactur ing chrysotile asbestos cement products. Br J Ind Med 1986;43:726-732.
9 Neuberger M, Kundi M: Individual asbestos exposure: smoking and mortality - a cohort study in the asbestos cement industry. Br J Ind Med 1990;47:615-620.
10 McDonald JC, McDonald AD: Chrysotile, tremolite and carcinogenicity. Ann Occup Hyg 1997;41:699-705.
11 Roggli VL, Sharma A, Butnor KJ, Sporn T, Vollmer RT: Malignant mesothelioma and occupational exposure to asbestos: a clinicopathological correlation of 1445 cases. Ultrastruct Pathol 2002;26:55-65.
12 Yarborough CM: The risk of mesothelioma from exposure to chrysotile asbestos. Curr Opin Pulm Med 2007;13:334-338.
13 Huuskonen MS, Karjalainen A, Tossavainen A, et al: Asbestos and cancer in Finland. Med Lav 1995;86:426-434.
14 Council Directive 83/477/EEC of 19 Septem ber 1983 on the protection of workers from the risks related to exposure to asbestos at work (second individual directive within the meaning of article 8 of directive 80/1107/ EEC). Off J Eur Union 1983;L 263:25-32.
15 Breslow NE, Day NE: Statistical Methods in Cancer Research. IARC Scientific Publica tion No. 82. Lyon, International Agency for Research on Cancer, 1987, vol 2: The Design and Analysis of Cohort Studies.
16 Collett D: Modelling Survival Data in Medi cal Research. Boca Raton, Chapman & Hall, 1993.
17 Sichletidis L, Chloros D, Tsiotsios I, et al: High prevalence of smoking in Northern Greece. Primary Care Respir J 2006; 15:92 97.
18 Dement JM, Brown DP, Okun A: Follow-up study of chrysotile textile workers: cohort mortality and case-control analyses. Am J Ind Med 1994;26:431-447.
19 Raffn E, Lynge E, Juel K, Korsgaard B: Inci dence of cancer and mortality among em ployees in the asbestos-cement industry in Denmark. Br J Ind Med 1989;46:90-96.
20 Hughes JM, Weill H: Asbestos exposure: quantitative assessment of risk. Am Rev Respir Dis 1986;133:5-13.
21 McDonald JC, McDonald AD: The epidemi ology of mesothelioma in historical context. Eur Respir J 1996;9:1932-1942.
22 Newhouse ML, Sullivan KR: A mortality study ofworkers manufacturing friction ma terials 1941-1986. Br J Ind Med 1989;46:176179.
23 Hodgson JT, Darnton A: The quantitative risks of mesothelioma and lung cancer in re lation to asbestos exposure. Ann Occup Hyg 2000;44:565-601.
24 Bernstein DM, Hoskins JA: The health ef fects of chrysotile: current perspective based upon recent data. Regul Toxicol Pharmacol 2006; 45:252-264.
25 Bernstein DM: Asbestos; in Salem H, Katz SA (eds): Inhalation Toxicology, ed 2. New York, Taylor and Francis Group, 2006, pp 647-667.
26 Hughes JM, Weill H, Hammad YY: Mortal ity of workers employed in two asbestos ce ment manufacturing plants. Br J Ind Med 1987;44:161-174.
27 Hesterberg TW, Hart GA: Synthetic vitreous fibers: a review of toxicology research and its impact on hazard classification. Crit Rev Toxicol 2001;31:1-53.
28 Topinka J, Loli P, Dusinska M, et al: Muta genesis by man-made mineral fibers in the lung of rats. Mutat Res 2006;595:174-183.
6 Respiration
Sichletidis/Chloros/Spyratos/Haidich/ Fourkiotou/Kakoura/Patakas