Separation of bone from iodine- and gadolinium-based contrast agents using dual energy CT

This study aims to evaluate the separability of bone from iodine- and gadolinium-based intravenous contrast agents using dual energy CT techniques in a phantom. The phantom was prepared containing varying concentrations of iodine-based contrast, gadolinium-based contrast, and calcium hydroxyapatite (to simulate bone). Thirteen iodine concentrations from 0.1 to 12 mg/mL, twelve gadolinium concentrations from 0.72 to 34.42 mg/mL, and four calcium concentrations from 0 to 200 mg/mL were used. These phantoms were scanned on a dual source CT using two different source spectra, producing one set of data at 80 kVp and another at 140 kVp. On each resulting image, the mean HU was measured at every concentration level for iodine, gadolinium, and calcium, and plotted on a graph of HU value at 80 versus 140 kVp. Linear regression was used to produce a best-fit line for each material. These lines were compared to test for a difference of slopes between calcium and iodine as well as between calcium and gadolinium. Each material exhibited a linear relationship between the HU values at 140 and 80 kVp (R2 = 0.99) and demonstrated a unique slope to this line. The slope for iodine was 2.00, for gadolinium was 1.63, and for calcium was 1.55. The slopes of the calcium and iodine lines were significantly different (p < 0.05), while the slopes of the calcium and gadolinium lines were not significantly different (p > 0.05). Our results suggest that while it is technically feasible to separate iodine from bone, gadolinium-based contrast does not appear to be as readily separable from bone as iodine. This result is surprising as the atomic number and k-edge of calcium (Z = 20, k-edge = 4 keV) are closer to iodine (Z = 53, k-edge = 33 keV) than to gadolinium (Z = 64, k-edge = 50 keV).