We prospectively analyzed benign asbestos-related pleural and parenchymal abnormalities on high-resolution CT scans and correlated them with clinical diagnoses in 100 asbestos-exposed workers. All subjects had high-resolution CT scans in conjunction with conventional CT at the time of clinical evaluation. To evaluate for asbestosis, we ranked high-resolution CT scans as high, intermediate, or low probability of asbestosis on the basis of the multiplicity and extent of observed parenchymal changes. By linear regression analysis, the most distinctive high-resolution CT features of asbestosis included thickened nondependent interstitial short lines and parenchymal bands. In 45 subjects satisfying clinical criteria of asbestosis, high-resolution CT probability of asbestosis was high in 38 (84%), intermediate in five (11%), and low in two (4%). In 20 (36%) of 55 subjects without clinical asbestosis, parenchymal abnormalities indicative of a high probability of asbestosis were observed on high-resolution CT. High-resolution CT probability scores had a strong positive correlation with chest radiographic profusion scores (p less than .0001) and asbestos-related pleural thickening (p less than .0001). Significant inverse correlations were seen with forced vital capacity (p less than .006) and single-breath diffusing capacity (p less than .03), both functional measures of restrictive interstitial lung disease. Neither clubbing nor rales were sufficiently prevalent to have statistical correlation with high-resolution CT scores. High-resolution CT is sensitive in detecting both pleural and parenchymal abnormalities in the asbestos-exposed subject. Asbestos-related pleural changes are observed more frequently on high-resolution CT than on conventional CT or chest radiography. The probability of asbestosis based on high-resolution CT parenchymal features has a significant correlation with existing clinical determinants of disease, and high-resolution CT can detect abnormality when other methods are not diagnostic.