Shun Nakajima1, Hiroyuki Ito1, Tetsuya Mitsuhashi1, Yutaka Kubo2, Kazuhiro Matsui1, Isao Tanaka3, Rika Fukui3, Hisako Omori1, Takashi Nakaoka4, Hiroshi Sakura1, Eiko Ueno3, Haruhiko Machida3. 1. Department of Medicine, Tokyo Women's Medical University Medical Center East, 2-1-10 Nishiogu, Arakawa-ku, Tokyo 116-8567, Japan. 2. Department of Medicine, Tokyo Women's Medical University Medical Center East, 2-1-10 Nishiogu, Arakawa-ku, Tokyo 116-8567, Japan. Electronic address: ykubogm@dnh.twmu.ac.jp. 3. Department of Radiology, Tokyo Women's Medical University Medical Center East, 2-1-10 Nishiogu, Arakawa-ku, Tokyo 116-8567, Japan. 4. Department of Medicine, Tokyo Women's Medical University Medical Center East, 2-1-10 Nishiogu, Arakawa-ku, Tokyo 116-8567, Japan. Electronic address: nakaoka.takashi@twmu.ac.jp.
Abstract
BACKGROUND AND AIMS: Coronary computed tomography (CT) angiography allows non-invasive classification of non-calcified coronary plaques (NCCPs) based on Hounsfield unit (HU) values. This methodology, however, is somewhat limited for reliable classification of NCCPs. Therefore, we evaluated the effective atomic number (EAN) for classifying NCCPs by single-source dual-energy CT with fast tube voltage switching (SSDECT). METHODS: We prospectively enrolled 18 patients undergoing both SSDECT and intravascular ultrasonography (IVUS). Monochromatic images at 70 keV and EAN images were reconstructed from SSDECT data sets. Regions of interest (ROIs) within NCCPs were placed on IVUS-matched SSDECT images, and mean HU values and EANs for soft and fibrous plaques, classified using IVUS, were compared with an unpaired t-test. RESULTS: We placed 96 ROIs in 29 soft plaques and 37 ROIs in 15 fibrous plaques in 12 coronary arteries of 11 patients. The mean HU value in soft plaques (58.2 ± 32.8 HU) was significantly lower than that in fibrous plaques (103.9 ± 48.3 HU) (p < 0.001). The mean EAN in soft plaques (8.7 ± 0.5) was also significantly lower than that in fibrous plaques (9.6 ± 0.5) (p < 0.0001). Area under the curve for EAN (0.91) was significantly higher than that for HU value (0.79) in receiver operating characteristic curve analysis (p = 0.046). With a cutoff EAN of 9.3, sensitivity was 90% and specificity, 87%; whereas with a cutoff HU value of 55.0 HU, sensitivity was 62% and specificity, 93%. CONCLUSIONS: EAN measurement by SSDECT can be clinically useful for accurately classifying soft and fibrous coronary plaques.
BACKGROUND AND AIMS: Coronary computed tomography (CT) angiography allows non-invasive classification of non-calcified coronary plaques (NCCPs) based on Hounsfield unit (HU) values. This methodology, however, is somewhat limited for reliable classification of NCCPs. Therefore, we evaluated the effective atomic number (EAN) for classifying NCCPs by single-source dual-energy CT with fast tube voltage switching (SSDECT). METHODS: We prospectively enrolled 18 patients undergoing both SSDECT and intravascular ultrasonography (IVUS). Monochromatic images at 70 keV and EAN images were reconstructed from SSDECT data sets. Regions of interest (ROIs) within NCCPs were placed on IVUS-matched SSDECT images, and mean HU values and EANs for soft and fibrous plaques, classified using IVUS, were compared with an unpaired t-test. RESULTS: We placed 96 ROIs in 29 soft plaques and 37 ROIs in 15 fibrous plaques in 12 coronary arteries of 11 patients. The mean HU value in soft plaques (58.2 ± 32.8 HU) was significantly lower than that in fibrous plaques (103.9 ± 48.3 HU) (p < 0.001). The mean EAN in soft plaques (8.7 ± 0.5) was also significantly lower than that in fibrous plaques (9.6 ± 0.5) (p < 0.0001). Area under the curve for EAN (0.91) was significantly higher than that for HU value (0.79) in receiver operating characteristic curve analysis (p = 0.046). With a cutoff EAN of 9.3, sensitivity was 90% and specificity, 87%; whereas with a cutoff HU value of 55.0 HU, sensitivity was 62% and specificity, 93%. CONCLUSIONS: EAN measurement by SSDECT can be clinically useful for accurately classifying soft and fibrous coronary plaques.
Authors: Ulrika Asenbaum; Richard Nolz; Stefan B Puchner; Tobias Schoster; Lukas Baumann; Julia Furtner; Daniel Zimpfer; Guenther Laufer; Christian Loewe; Sigrid E Sandner Journal: Sci Rep Date: 2020-08-17 Impact factor: 4.379