Home  |  Contact Us  |  Site Map Français  |  Español

Science and publications

An Executive Summary

Workshop On Health Risks Associated With Chrysotile Asbestos

St. Helier, Channel Islands, 14-17 November 1993


Chrysotile as mined can be harsh or soft and vary in trace metal and associated mineral content and other physico-chemical properties. Industrial manipulation and treatment may change fibre dimensions and other properties. At high temperature it may alter to another mineral (non-asbestos). Thus, health risks must be evaluated in the light of the mineral as mined with its associated minerals, other contaminants and as used. A key feature of chrysotile is the magnesium in its outer structure which in the lung is leached, leaving magnesium depleted fibres of reportedly, reduced biological activity.

Fibrous tremolite occurs in some chrysotile deposits and is found in high concentrations in the lungs of some Québec chrysotile miners and millers. Participants suggested that tremolite in various chrysotile deposits should be more thoroughly characterised, guided by the results of epidemiological studies.


In contrast to its past widespread use in friable products such as insulation and textiles, chrysotile is now used in four principal product categories - asbestos cement, friction materials, roof coatings and cements and gaskets. In these products asbestos is 'locked-in" or encapsulated within a matrix.



Fibre diameter and length are primary determinants of penetration into the respiratory system, access to translocation pathways, surface properties and biopersistence which, in turn affect the potential for biological response. Surface properties and biopersistence are also complementary parameters which determine biological response. Data concerning the dimensions of airborne chrysotile fibres are limited and do not always include statistically valid measurements of the very long fibres. Reliable measurements of long fibres are needed, in particular for the textile industry, as fibre dimension differences have been proposed as one of the more likely explanations for apparent large differences in lung cancer risks between the textile and other industrial sectors using chrysotile.


Measurements of chrysotile exposures using a variety of instruments and methods began in the 1930s and developed to phase contrast optical microscopic fibre counting using the "membrane filter method" in the mid 1960s. Refinements in the membrane filter method in the UK have resulted in a several-fold increase in the number of fibres counted. Some US participants considered the effect of method changes to be less in the USA.

Workplace exposure standards are based on epidemiological studies in which past particle count measurements were converted to their membrane filter equivalent. Such conversions are subject to uncertainty. Approximate conversions to membrane fibre concentrations can be made as long as they are industry specific and as far as possible, process specific. However, conversions have been done for only a few chrysotile industries.

Concentrations in some of the industries studied epidemiologically, have been reported, in the past, to have measured in hundreds of fibres/ml. By contrast, concentrations in well controlled plants in Japan in 1992, ranged 0.07 - 0.66 f/ml , with 98 % of values below 0.3 f/ml. Group mean concentrations in building air inhaled by occupants were less than 0.0005 f/ml greater than 5 µm (90th percentile) and concentrations of fibres > 5 µm in length measured while workers ran cables through an asbestos sprayed ceiling ranged 0.001 - 0.228 f/ml with a time weighted average concentration (TWA) of 0.0167 f/ml. Very few data exist on levels of exposure to chrysotile in households. Chrysotile in lung tissue may be used as a marker of exposure, but must take into account fibre alteration and clearance. Asbestos bodies are not a good indicator of human exposure to chrysotile.



With the exception of the textile industry, the slopes of the exposure response curves for lung cancer in the various chrysotile industry sectors were shallow, with no detectable risk of an extremely low level risk of lung cancer associated with exposure to chrysotile asbestos at and below lieftime cumulative exposures of 30 fibres / ml-years. No chrysotile related increased risk was detected at considerably higher exposures in the mining sector.

The mining industry

In the mortality study of approximtely 11,000 Québec chrysotile miners and millers born 1891 - 1920, the standardized mortality ratio (SMR) for men with exposures of less than 300 million particles per cubic foot years (mppcf-yrs) was 1.27, but there was no trend of increasing lung cancer risk with increasing exposure. A case-control study of all 548 lung cancer deaths observed between 1950 - 1988 showed that the risk of lung cancer at concentrations of less than 15 mppcf (approximately 50 f/ml) was "vanishingly small". Smoking was an important risk factor for many causes including lung cancer and ischemic heart disease. The interaction between chrysotile exposure and smoking was more than additive but less than multiplicative. There was no evidence of an increased risk of laryngeal or stomach cancer in the Québec chrysotile mining cohort.

Asbestos-cement industry

Fourteen cohorts of asbestos cement workers have been studied. Standardized mortality ratios were 2 or less in 13 of them. Exposure-response relationships exist for lung cancer, mesothelioma and asbestosis. One study has shown that the excess lung cancer risk was restricted to persons with radiological evidence of asbestosis of which there was little evidence below a cumulative exposure of 30 f/ml-years. This suggests that a non-threshold model for lung cancer may not be appropriate. Additional studies to confirm or otherwise this important observation are needed. The use of amphibole fibres and their importance in evaluating health risks in this industry sector and the potential for silica exposure as a possible confounder were noted. The absence of lung cancer and mesothelioma risks in workers exposed to reportedly high concentrations of chrysotile in a UK asbestos cement plant in which silica was not used, as well as in two similar plants in Zimbabwe was also noted.

Friction manufacturing industry

There have been three main studies of friction manufacturing workers. These show that, if there are any effects on mortality due to work in the manufacture of friction materials, the effects must be small. A UK study of more than 13,000 workers, followed 1941 - 1988 showed no excess of mortality from lung cancer. There was also no excess of chrysotile related cancer at any site. A study of 3641 men in the USA showed that respiratory cancer was not related to cumulative chrysotile exposure. In two small factories in Canada, fewer lung cancer cases had worked in departments where asbestos was used compared to controls. Overall, there was 1 confirmed mesothelioma in the friction material manufacturing industry for whom the only known exposure was chrysotile.

The textile industry

In an update of the mortality of the cohort of 1200 South Carolina textile workers there were 41 lung cancers (SMR=2.25). The rate of increase in lung cancer risk with fibre exposure was steeper than that of other cohorts using mainly chrysotile, with a doubling of lung cancer risk at approximately 30 fibres/cc-years. This finding was in line with previous studies at the plant and the lung cancer experience was similar to that of textile plants in the UK and Pennsylvania, both of which had experienced greater amphibole exposures. Although a study of oils at the Carolina plant failed to show them to have a significant effect on lung cancer risk, participants agreed that two hypotheses deserved further examination. These were that the differences were due to differences in the size distributions of the fibres (longer in textiles) or that they were in some way related to the use of oils on the fibre. There were two confirmed mesotheliomas of unknown etiology in the S. Carolina textile plant. Crocidolite yarn has been used in the plant.

Building occupants

The estimation of lung cancer risks for building occupants involves assumptions, in particular the applicability of the dose-response slope to use to estimate risk. The majority view was that the choice should be influenced heavily by the group to which the risk estimate is to apply. Since air in US buildings contains short fibres, mainly chrysotile and very little amphibole, the mining/milling slope or an aggregate of mining/asbestos cement and friction data was considered most appropriate. Risks based on these shallower slopes gave annualized risk estimates of the order of 0.01/million (well below the risk of being electrocuted). Using steeper slopes also resulted in "negligible risks". It was a clear opinion of many participants that the use of a linear non-threshold model may have no basis at low levels of exposure.



There was agreement by the majority of participants that differences exist in the risk of mesothelioma associated with different fibre types. This was based on an extensive literature. A minority view was that the risk difference was not as important. This was based on the use of ratios of mesothelioma to excess lung cancers on the assumption that this corrects for exposure. This was not accepted by the majority who expressed the view that there was at least a 10 fold lower risk of mesothelioma in workers exposed to chrysotile compared to amphiboles. The view was expressed that at current chrysotile levels mesothelioma would be unlikely to occur. Such a conclusion is supported by the observation that in Québec miners and millers, the only large group for which lung fibre analyses exist, mesotheliomas were associated with very high concentrations of both chrysotile and tremolite in the lung, similar to those found in cases of asbestosis. Independent data showed that high chrysotile exposures were necessary as the combined chrysotile and tremolite concentrations in the lungs of mesothelioma cases in miners and millers were 100 times greater than the amosite concentrations in the lungs of shipyard workers with mesothelioma.

Based, on the available evidence, asbestiform tremolite was suggested as being responsible for some or most of mesotheliomas related to chrysotile exposure in the Québec chrysotile mining industry. However, the actual role of tremolite still remains to be clarified as an alternative hypothesis that chrysotile or more specifically sequestered chrysotile is responsible cannot be ruled out at this time. As the potential of chrysotile to induce mesothelioma is related to the particular chrysotile as mined and distributed, it was suggested that the extensive epidemiological evidence on mesothelioma should be used to quantify risks for various situations. While most participants considered continuous deposition or long retention of chrysotile (with or without tremolite) necessary for mesothelioma induction, a minority opinion was that long term retention was not required.

Concerning downstream risks, a study in Germany involving 324 mesothelioma cases found no evidence of any association between mesothelioma and work as brake repair mechanics.


Non-malignant conditions associated with exposure to chrysotile dusts include pulmonary parenchymal and pleural fibrosis, small airway abnormality and conditions affecting the large airways such as chronic bronchitis and chronic airflow limitation. The prevalence rates for all the markers of morbidity (radiological change, pulmonary function, etc.) have been shown to increase with increasing exposure. They increase more steeply when exposure is in the textile industry than in the mining industry. The presence of long fibre tremolite or other amphibole fibres may result in steeper exposure response slopes.

In Sweden, persons with pleural plaques identified during a general population survey were studied prospectively. The incidence of lung cancer was higher than in the general population. However the fibre types involved and levels of exposure were not known. Several studies have shown no excess lung cancer in persons with plaques beyond that accounted for by exposure level. Extensive calcification may occur in cases of pleural plaque without appreciable thickening and without any increased incidence of lung cancer. In Québec, large differences in the rates of pleural calcification between Thetford and Asbestos have been recognised for many years. It was hypothesized that calcification here might be associated with fibrous tremolite exposure.



Animal inhalation studies have shown that chrysotile asbestos can cause fibrosis, benign and malignant pulmonary tumours. Studies reporting high tumour rates also report high levels of asbestosis. Two inhalation studies have shown both fibrosis and the risk of pulmonary tumours to be dose related but the number of animals used have been inadequate to demonstrate the shape of the exposure response curve. While all 8 species of experimental animal studied with chrysotile develop lung fibrosis, only rats develop pulmonary tumours.

Diffuse mesotheliomas of the pleural and peritoneal cavities can occur in animals without exposure to mineral fibres. Inoculation of fibres into the pleural cavities may result in mesotheliomas but the fibre burdens are extremely high. Tumour incidence depends on the dose, source and preparation of the fibres employed. Very few mesotheliomas are produced in inhalation studies compared to inoculation although a non-asbestos fibre, a fibrous zeolite known as erionite has yielded nearly 100 % tumours in one study. Hamsters which have been reported to be sensitive models of mesothelioma production (based on refractory ceramic fibre studies) did not produce mesotheliomas with chrysotile used as a control substance.

Penetration, deposition and clearance of chrysotile cannot be directly compared with other mineral fibre types because of its unique physical and chemical characteristics. However, it was reported from Germany that carcinogenicity studies in rats using inhalation and intra-cavitary injection of chrysotile, amosite and crocidolite gave no clear indication of a lower carcinogenic potency per chrysotile fibre than per amphibole fibre if equal fibre numbers and fibre sizes were applied, although the chrysotile content of the lungs was relatively low. Mesothelioma rates after inhalation of extremely high concentrations were also similar (5%) for chrysotile and amphiboles although the lung chrysotile content was low. These exposures were 100 times those in textile plants to induce the same lung tumour incidence.

Because factors that affect deposition and retention are not fully taken into account using intra-tracheal and other artificial routes of administration, results cannot be directly extrapolated to humans. Mechanistically, injection experiments have indicated that long thin fibres more readily induce mesotheliomas. However, this has been difficult to confirm in inhalation studies, although fibres of less than 5 um in length appear to cause neither fibrosis nor pulmonary tumours. It was suggested that to produce neoplasia, fibres of length longer than 20 µm may be needed. This view contrasted with another in which short chrysotile fibres (< 5 µm) were considered to have carcinogenic potency, but that it was low.

Ingestion studies with chrysotile have been universally negative. One study examined the larynx and had no pathological changes. Significant numbers of kidney tumours have never been recorded. The role of chrysotile contaminants in the etiology of lung related disease in animals has not been clarified. Because of interspecies differences and other factors, participants were clear that human data should have priority over animal data when available.



Pulmonary macrophages are an important part of the lung's defence against particles deposited by inhalation. On a mass basis, chrysotile appears to be as toxic as amphibole fibres while on a fibre number basis, it appears to be less toxic to alveolar macrophages. Short chrysotile fibres are cleared from the lung rapidly and longer ones at a slower rate. Short fibres generally exhibit less macrophage toxicity and less release of cytokines than longer ones. The pulmonary retention half-time for chrysotile for primates has been estimated at 105 and 90 days. Despite the fast clearance of chrysotile, asbestosis was reported to be the same severity as in animals exposed to amphibole fibres which did not clear. The use of half-times was questioned because the process is complex with splitting of fibres and transport of fibres into sequestered compartments of the lung where clearance is slow.

Transformation of rat pleural mesothelial cells has been observed in vitro following treatment with chrysotile. Canadian chrysotile was more mutagenic than crocidolite on a per weight basis; in contrast, on the basis of the number of fibres with critical dimensions as defined by Stanton and colleagues, crocidolite was more efficient. Experiments using other dusts mixed with chrysotile showed increases in long term effects and also increased translocation to the pleura.

Whether chrysotile is an initiator of lung cancer or mesothelioma in human cells is unclear as evidence of chromosomal abberations in human bronchial epithelial cells have been for the most part negative. Cell proliferation may be a more relevant phenomenon, with no-observed-effect levels in all responses using low doses of both chrysotile and crocidolite. However, it seems probable that the increased clearance and dissolution of chrysotile may render it less potent than the amphibole fibres.



Studies of human lungs indicate that for virtually all types of exposure, the relative proportion of amphibole retained far exceeds that in the original dust and the proportion of chrysotile far less. Chrysotile is cleared from the lung parenchyma quickly with the bulk of fibres removed from human lungs within weeks - months after inhalation. Amphibole half-lives are of the order of years to decades.

Graham W. Gibbs


1. The workshop was organized by the Scientific Committee on Mineral Fibres of the International Commission on Occupational Health (ICOH) in collaboration with the WHO/ILO/UNEP International Program on Chemical Safety (IPCS) and was attended by 36 invited scientists and 11 scientific observers from 12 countries. Prior to the meeting, some 70 other scientists were invited to submit relevant information which was made available at the workshop.

Papers on selected topics were prepared prior to the workshop. Key issues were introduced in 10 minute presentations followed by 1/2-hour discussion periods. Rapporteur's summaries were scrutinized and corrected by participants at the workshop and all papers presented at the meeting were peer reviewed by the Journal prior to publication in the August 1994 issue of the Annals of Occupational Hygiene (AOH).

In recent months, I received several requests for an executive summary of the workshop. This summary is an attempt to respond to these requests. It has not been reviewed by the meeting participants. However, I am grateful to four participants who offered constructive comment on my draft efforts.

Readers should refer to the published report for technical detail and in the event of any discrepancies between this summary and the published report, the latter should be considered correct. This summary does not necessarily reflect the writer's opinions or interpretation of the available data, but is an attempt to condense the papers, discussion and opinions conveyed at the workshop.

July 20 1994

Back to index of publications

International Chrysotile Association © All rights reserved | Contact us | Sitemap