The present study analysed the asbestos lung burden in necroscopic samples from 55 subjects free from asbestos-related diseases, collected between 2009 and 2011 in Milan, Italy. Multiple lung samples were analysed by light microscopy (asbestos bodies, AB) and EDXA-scanning electron microscopy (asbestos fibres and other inorganic fibres). Asbestos fibres were detected in 35 (63.6%) subjects, with a higher frequency for amphiboles than for chrysotile. Commercial (CA) and non-commercial amphiboles (NCA) were found in roughly similar frequencies. The estimated median value was 0.11 million fibres per gram of dry lung tissue (mf g -1 ) for all asbestos, 0.09 mf g -1 for amphiboles. In 44 (80.0%) subjects no chrysotile fibres were detected. A negative relationship between asbestos mass-weighted fibre count and year of birth (and a corresponding positive increase with age) was observed for amphiboles [-4.15%, 95% confidence interval (CI) = -5.89 to -2.37], talc (-2.12%, 95% CI = -3.94 to -0.28), and Ti-rich fibres (-3.10%, 95% CI = -5.54 to -0.60), but not for chrysotile (-2.84%, 95% CI = -7.69 to 2.27). Residential district, birthplace, and smoking habit did not affect the lung burden of asbestos or inorganic fibres. Females showed higher burden only for amphiboles (0.12 versus 0.03 mf g -1 in males, P = 0.07) and talc fibres (0.14 versus 0 mf g -1 in males, P = 0.03). Chrysotile fibres were shorter and thinner than amphibole fibres and NCA fibres were thicker than CA ones. The AB prevalence was 16.4% (nine subjects) with concentrations ranging from 10 to 110 AB g -1 dry, well below the 1000 AB g -1 threshold for establishing occupational exposure. No AB were found in subjects younger than 30 years. Our study demonstrated detectable levels of asbestos fibres in a sample taken from the general population. The significant increase with age confirmed that amphibole fibres are the most representative of cumulative exposure.