Assessment of commercially available in-plane bismuth breast shields for clinical use in patients undergoing thoracic computed tomography

Abstract

Objective: We aimed to assess qualitatively the effects of a bismuth breast shield by measuring image noise and computed tomography (CT) number changes with 0-, 1-, 2-, and 3-cm shield-to-phantom distances. We also sought to assess the dose reduction achieved by the shield and to evaluate its effect on image quality. Methods: A cylindrical body phantom was scanned using an adult thoracic CT protocol with 0-, 1-, 2-, and 3-cm foam spacers placed between the shield and the phantom, measuring the noise and CT numbers (in Hounsfield units [HU]) of the image data. We also used the shield with 3-cm spacer over the left breast in 180 female patients referred for chest CT. Dose measurements were performed using thermoluminescent dosimeters. The image quality was assessed following European guidelines. Results: A 0-cm shield-to-phantom distance significantly increased noise and CT numbers of the image data. The 3-cm shield-to-phantom distance effectively lowered shield-induced image noise; however, the HUs remained significantly increased over all shield-to-phantom distances (p < 0.001). In the patient study, the average absorbed doses to the shielded and non-shielded breasts were 13.6±3.1 mGy and 24.04±4.7 mGy, respectively; a 43.4% dose reduction. Conclusion: Combining a bismuth shield with a 3-cm shield-to-breast foam spacer significantly reduced radiation exposure without qualitative or quantitative deterioration of images in terms of image noise. However, increases in the HUs of the images persisted.