Reconstructing historical exposure to asbestos: the validation of 'educated guesses'.

Background: In both the epidemiological and legal context, the causal attribution of asbestos-related lung diseases requires retrospective exposure assessment (REA).

Aims: To assess the correlation between the retrospective assessment of occupational and anthropogenic environmental exposure to asbestos and its content in the lung tissue.

Methods: Based on the available exposure information, a team of occupational physicians retrospectively assessed cumulative exposure to asbestos in 24 subjects who died of asbestos-related diseases. The asbestos lung content was analysed using analytical scanning electron microscope (SEM-EDS). The Log10 asbestos fibre count in the autoptic samples was predicted as a function of the Log10 estimated cumulative exposure using univariate regression analysis.

Results: The median count of asbestos fibres by grams of dry weight (ff/gdw) in the lung tissue was 81 339 (range 0-2 135 849.06); it was 287 144 (range 0-2 135 849.06) among the occupationally exposed, and 29 671 (range 0-116 891) among the subjects who only had anthropogenic environmental and/or household exposure. Amphiboles, and particularly amosite (52%) and crocidolite (43%), were detected in all the study subjects. Chrysotile was not detected in any of the samples. Overall, the retrospective estimate of lifetime cumulative exposure to asbestos showed a moderate correlation with the total asbestos fibre count in the autoptic lung, with the regression model explaining 38-55% of the total variance.

Conclusions: Detailed information on occupational, environmental and household exposure circumstances would be indispensable for experienced industrial hygienists and/or occupational physicians to reliably assess past exposure to amphiboles or mixed types of asbestos.