Effects of breast shielding during heart imaging on DNA double-strand-break levels: A prospective randomized controlled trial

Purpose: To examine the effect of breast shielding on blood lymphocyte deoxyribonucleic acid (DNA) double-strand-break levels resulting from in vivo radiation and ex vivo radiation at breast-tissue level, and the effect of breast shielding on image quality. Materials and Methods: The study was approved by institutional review and commpliant with HIPAA guidelines. Adult women who underwent 64-section coronary computed tomographic (CT) angiography and who provided informed consent were prospectively randomized to the use (n = 50) or absence (n = 51) of bismuth breast shields. Peripheral blood samples were obtained before and 30 minutes after in vivo radiation during CT angiography to compare DNA double-strand-break levels by γ-H2AX immunofluorescence in blood lymphocytes. To estimate DNA double-strand-break induction at breast-tissue level, a blood sample was taped to the sternum for ex vivo radiation with or without shielding. Data were analyzed by linear regression and independent sample t tests. Results: Breast shielding had no effect on DNA double-strand-break levels from ex vivo radiation of blood samples under shields at breast-tissue level (unadjusted regression: β = .08; P = .43 versus no shielding), or in vivo radiation of circulating lymphocytes (β =-.07; P = .50). Predictors of increased DNA double-strand-break levels included total radiation dose, increasing tube potential, and tube current (P <.05). With current radiation exposures (median, 3.4 mSv), breast shielding yielded a 33% increase in image noise and 19% decrease in the rate of excellent quality ratings. Conclusion: Among women who underwent coronary CT angiography, breast shielding had no effect on DNA double-strandbreak levels in blood lymphocytes exposed to in vivo radiation, or ex vivo radiation at breast-tissue level. At present relatively low radiation exposures, breast shielding contributed to an increase in image noise and a decline in image quality. The findings support efforts to minimize radiation by primarily optimizing CT settings.