Subtraction CT Angiography of the Lower Extremities:: Single Volume Subtraction versus Multi-Segmented Volume Subtraction

Abstract

To validate the hypothesis that a multisegmented approach during subtraction computed tomography (CT) angiography of the lower extremities can improve bone removal efficiency by suppressing regional motion. The institutional review board of our hospital approved this retrospective study. One hundred and one consecutive patients that had undergone the lower extremity CT angiography were included in this study. Subtraction CT angiography was performed using two different methods, namely, by single volume subtraction and by multisegmented volume subtraction. Multisegmented volume subtraction was conducted by dividing the whole volume of the CT data into three segments along the z axis of the lower extremities, performing a subtraction process for each segment, and combining segments to form as single subtracted volume. The bone removal efficiencies of the two methods was assessed by analyzing bone subtraction scores on maximum intensity projection (MIP) images for each bone segment in a blinded fashion. In addition, overall MIP image qualities were compared by displaying MIP images produced using the two methods side by side. Differences between bone subtraction scores were tested using Wilcoxon's signed rank test. Multisegmented volume subtraction MIP images demonstrated significantly better bone removal for the following bone segments: pelvis (P < .0001), hip (P = .0002), thigh (P = .0258), knee (P = .0004), ankle (P = .0008), metatarsal bone (P < .0001), and toes (P < .0001). Overall bone subtraction score and subjective image qualities determined by performing side-by-side comparisons were better for the multisegmented volume subtraction method. Bone removal performance and overall MIP image quality can be increased by adopting multisegmented volume subtraction during subtraction CT angiography of the lower extremities.