Pleural Plaques in Asbestos-exposed Workers: Reproducibility of a New High-resolution CT Visual Semiquantitative Measurement Method

Abstract

To assess the reproducibility of a new high-resolution computed tomography (CT) visual semiquantitative method for pleural plaques in asbestos-exposed workers. We performed thin-section CT in 752 chrysotile asbestos mining workers and ex-workers. Institutional review board approval and signed written informed consent from subjects were obtained. Two readers independently evaluated the 752 CT scans and identified 57 workers (mean age +/- SD, 61.8 years +/- 8.1; range, 37 to 81 years) who had pleural plaques and no other pleural or parenchymal abnormality. Three independent radiologists then quantified the plaque burden in these 57 workers using a scoring system based on the evaluation of the maximum thickness of parietal pleural plaques and percentage of parietal pleural surface involvement. We also calculated the proportion between the number of CT slices with diaphragmatic plaques and the total number of slices in which the diaphragm was seen (pdiaph). The intraobserver and interobserver agreements were analyzed using weighted Kappa coefficient. Interobserver agreements were good for the pleural plaque score (k = 0.61, 0.75, and 0.79) and ranged from good (k = 0.61) to excellent (k = 0.86) for the pdiaph. Intraobserver agreements ranged from good to excellent for the pleural plaque score (k = 0.79 and 1.00) and for the pdiaph (k = 0.79 and 0.93). The method proposed for high-resolution CT pleural plaque quantification in asbestos-exposed workers has a high reproducibility.