Peijun Liu1, Lu Lin1, Cheng Xu1, Yechen Han2, Xue Lin2, Yang Hou3, Xiaomei Lu4, Mani Vembar5, Zhengyu Jin1, Yining Wang1. 1. Department of Radiology, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. 2. Department of Cardiology, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. 3. Department of Radiology, Shengjing Hospital of China Medical University, Shenyang, China. 4. Clinical Science, Philips Healthcare, Beijing, China. 5. CT Clinical Science, Philips Healthcare, Cleveland, OH, USA.
Abstract
BACKGROUND: To evaluate the segmental myocardial extracellular volume (ECV) fraction and to define a threshold ECV value that can be used to distinguish positive late gadolinium enhancement (LGE) segments from negative myocardial segments using dual-layer spectral detector computed tomography (SDCT), with magnetic resonance imaging (MRI) as a reference. METHODS: Fifty-six subjects with cardiac disease or suspected cardiac disease, underwent both late iodine enhancement on CT (CT-LIE) scanning and late gadolinium enhancement on MRI (MRI-LGE) scanning. Each procedure occurred within a week of the other. Global and segmental ECVs of the left ventricle were measured by CT and MRI images. According to the location and pattern of delayed enhancement on MRI image, myocardial segments were classified into 3 groups: ischemic LGE segments (group 1), nonischemic LGE segments (group 2) and negative LGE segments (group 3). The correlation and agreement between CT-ECV and MRI-ECV were compared on a per-segment basis. Receiver operating characteristic (ROC) curve analysis was performed to establish a threshold for LIE detection. RESULTS: Among the 56 patients, 896 segments were analyzed, and of these, 73 segments were in group 1, 229 segments were in group 2, and 594 segments were in group 3. In segmental analysis, CT-ECV in group 3 (27.0%; 24.9-28.9%) was significantly lower than that in group 1 (33.2%; 30.7-36.3%) and group 2 (34.9%; 32.3-39.8%; all P<0.001). Good correlations were seen between CT-ECV and MRI-ECV for all groups (group 1: r=0.920; group 2: r=0.936; group 3: r=0.799; all P<0.001). Bland-Altman analysis between CT-ECV and MRI-ECV showed a small bias in all 3 groups (group 1: -2.1%, 95% limits of agreement -11.3-7.1%; group 2: -0.6%, 95% limits of agreement -13.1-11.9%; group 3: 1.0%, 95% limits of agreement -12.7-14.7%). CT-ECV could differentiate between LGE-positive and LGE-negative segments with 83.1% sensitivity and 93.3% specificity at a cutoff of 31%. CONCLUSIONS: ECV values derived from CT imaging showed good correlation and agreement with MR imaging findings, and CT-ECV provided high diagnostic accuracy for discriminating between LGE-positive and LGE-negative segments. Thus, cardiac CT imaging might be a suitable noninvasive imaging technique for myocardial ECV quantification. 2022 Quantitative Imaging in Medicine and Surgery. All rights reserved.
BACKGROUND: To evaluate the segmental myocardial extracellular volume (ECV) fraction and to define a threshold ECV value that can be used to distinguish positive late gadolinium enhancement (LGE) segments from negative myocardial segments using dual-layer spectral detector computed tomography (SDCT), with magnetic resonance imaging (MRI) as a reference. METHODS: Fifty-six subjects with cardiac disease or suspected cardiac disease, underwent both late iodine enhancement on CT (CT-LIE) scanning and late gadolinium enhancement on MRI (MRI-LGE) scanning. Each procedure occurred within a week of the other. Global and segmental ECVs of the left ventricle were measured by CT and MRI images. According to the location and pattern of delayed enhancement on MRI image, myocardial segments were classified into 3 groups: ischemic LGE segments (group 1), nonischemic LGE segments (group 2) and negative LGE segments (group 3). The correlation and agreement between CT-ECV and MRI-ECV were compared on a per-segment basis. Receiver operating characteristic (ROC) curve analysis was performed to establish a threshold for LIE detection. RESULTS: Among the 56 patients, 896 segments were analyzed, and of these, 73 segments were in group 1, 229 segments were in group 2, and 594 segments were in group 3. In segmental analysis, CT-ECV in group 3 (27.0%; 24.9-28.9%) was significantly lower than that in group 1 (33.2%; 30.7-36.3%) and group 2 (34.9%; 32.3-39.8%; all P<0.001). Good correlations were seen between CT-ECV and MRI-ECV for all groups (group 1: r=0.920; group 2: r=0.936; group 3: r=0.799; all P<0.001). Bland-Altman analysis between CT-ECV and MRI-ECV showed a small bias in all 3 groups (group 1: -2.1%, 95% limits of agreement -11.3-7.1%; group 2: -0.6%, 95% limits of agreement -13.1-11.9%; group 3: 1.0%, 95% limits of agreement -12.7-14.7%). CT-ECV could differentiate between LGE-positive and LGE-negative segments with 83.1% sensitivity and 93.3% specificity at a cutoff of 31%. CONCLUSIONS: ECV values derived from CT imaging showed good correlation and agreement with MR imaging findings, and CT-ECV provided high diagnostic accuracy for discriminating between LGE-positive and LGE-negative segments. Thus, cardiac CT imaging might be a suitable noninvasive imaging technique for myocardial ECV quantification. 2022 Quantitative Imaging in Medicine and Surgery. All rights reserved.
Authors: Jesús G Mirelis; Javier Sánchez-González; Esther Zorio; Tomas Ripoll-Vera; Rafael Salguero-Bodes; David Filgueiras-Rama; Esther González-López; María Gallego-Delgado; Rodrigo Fernández-Jiménez; María Jesús Soleto; Juana Núñez; Gonzalo Pizarro; Javier Sanz; Valentín Fuster; Pablo García-Pavía; Borja Ibáñez Journal: Rev Esp Cardiol (Engl Ed) Date: 2017-03-22
Authors: Philip Haaf; Pankaj Garg; Daniel R Messroghli; David A Broadbent; John P Greenwood; Sven Plein Journal: J Cardiovasc Magn Reson Date: 2016-11-30 Impact factor: 5.364
Authors: Daniel R Messroghli; James C Moon; Vanessa M Ferreira; Lars Grosse-Wortmann; Taigang He; Peter Kellman; Julia Mascherbauer; Reza Nezafat; Michael Salerno; Erik B Schelbert; Andrew J Taylor; Richard Thompson; Martin Ugander; Ruud B van Heeswijk; Matthias G Friedrich Journal: J Cardiovasc Magn Reson Date: 2017-10-09 Impact factor: 5.364