Related Experiment Video
Updated: Jul 14, 2025
In Vivo Quantitative Assessment of Myocardial Structure, Function, Perfusion and Viability Using Cardiac Micro-computed Tomography
Published on: February 16, 2016
Differentiating Macrovascular and Microvascular Ischemia Using Fractal Analysis of Dynamic Myocardial Perfusion
Florian Michallek1, Satoshi Nakamura, Tairo Kurita
1From the Department of Radiology, Charité-Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Berlin, Germany (F.M., M.D.); Department of Advanced Diagnostic Imaging, Mie University Graduate School of Medicine, Tsu, Japan (F.M., K.K.); Department of Radiology, Mie University Graduate School of Medicine, Tsu, Japan (S.N., H.S.); Department of Cardiology and Nephrology, Mie University Graduate School of Medicine, Tsu, Japan (T.K.); Department of Advanced MRI Collaborative Research, Tohoku University Graduate School of Medicine, Sendai, Japan (H.O.); Department of Cardiology, Tohoku University Graduate School of Medicine, Sendai, Japan (K.N.); Saiseikai Matsuyama Hospital, Matsuyama, Japan (R.O.); Takasaki General Medical Center, Takasaki, Japan (T.S.); National Hospital Organization Kagoshima Medical Center, Kagoshima, Japan (H.N.); Peking Union Medical College Hospital, Beijing, China (Y.-N.W.); Kobe University Graduate School of Medicine, Kobe, Japan (T.I.); German Center for Cardiovascular Research, Berlin, Germany (M.D.); and Deutsches Herzzentrum der Charité (M.D.), Berlin, Germany.
Insights
Fractal analysis of 4D-CTP effectively differentiates coronary artery disease (CAD) and coronary microvascular disease (CMD), even in complex cases like ischemia and nonobstructed coronary arteries (INOCA). This multicenter study validates its accuracy for diagnosing these conditions noninvasively.
Area of Science:
- Cardiovascular Imaging
- Radiology
- Medical Diagnostics
Background:
- Coronary artery disease (CAD) and coronary microvascular disease (CMD) are significant causes of myocardial ischemia.
- Noninvasive differentiation of obstructive CAD and CMD remains challenging.
- Dynamic myocardial stress computed tomography perfusion imaging (4D-CTP) offers a potential noninvasive solution.
Purpose of the Study:
- To validate fractal analysis of 4D-CTP in a multicenter setting.
- To assess the diagnostic accuracy of fractal analysis in differentiating CAD and CMD.
- To evaluate performance in subgroups with ischemia and nonobstructed coronary arteries (INOCA) and mild to moderate stenosis.
Main Methods:
- A multicenter trial (AMPLIFiED) included patients with suspected chronic myocardial ischemia.
- Patients underwent dual-source CT angiography, 4D-CTP, and CT delayed-enhancement imaging.
- A combined reference standard (invasive angiography with FFR, CT-MBF) defined CAD, CMD, and normal perfusion.
Main Results:
- Fractal analysis accurately differentiated CAD, CMD, and normal perfusion (multiclass AUC 0.92).
- In patients with ischemia, accuracy for CAD vs. CMD was 89% (AUC 0.83).
- In INOCA and mild-moderate stenosis subgroups, fractal analysis achieved high accuracy (88-89%) and AUC (0.94-0.95).
Conclusions:
- Fractal analysis of 4D-CTP is a validated, accurate method for differentiating CAD and CMD in a multicenter setting.
- The technique demonstrates high diagnostic performance in subgroups with INOCA and mild to moderate stenosis.
- This noninvasive approach holds promise for improved diagnosis of coronary artery pathologies.
Objectives:
Fractal analysis of dynamic myocardial stress computed tomography perfusion imaging (4D-CTP) has shown potential to noninvasively differentiate obstructive coronary artery disease (CAD) and coronary microvascular disease (CMD). This study validates fractal analysis of 4D-CTP in a multicenter setting and assesses its diagnostic accuracy in subgroups with ischemia and nonobstructed coronary arteries (INOCA) and with mild to moderate stenosis.
Materials And Methods:
From the AMPLIFiED multicenter trial, patients with suspected or known chronic myocardial ischemia and an indication for invasive coronary angiography were included. Patients underwent dual-source CT angiography, 4D-CTP, and CT delayed-enhancement imaging. Coronary artery disease, CMD, and normal perfusion were defined by a combined reference standard comprising invasive coronary angiography with fractional flow reserve, and absolute or relative CT-derived myocardial blood flow. Nonobstructed coronary arteries were defined as ≤25% stenosis and mild to moderate stenosis as 26%-80%.
Results:
In 127 patients (27% female), fractal analysis accurately differentiated CAD (n = 61, 23% female), CMD (n = 23, 30% female), and normal perfusion (n = 34, 35% female) with a multiclass area under the receiver operating characteristic curve (AUC) of 0.92 and high agreement (multiclass κ = 0.89). In patients with ischemia (n = 84), fractal analysis detected CAD (n = 61) over CMD (n = 23) with sensitivity of 95%, specificity of 74%, accuracy of 89%, and AUC of 0.83. In patients with nonobstructed coronary arteries (n = 33), INOCA (n = 15) was detected with sensitivity of 100%, specificity of 78%, accuracy of 88%, and AUC of 0.94. In patients with mild to moderate stenosis (n = 27), fractal analysis detected CAD (n = 19) over CMD with sensitivity of 84%, specificity of 100%, accuracy of 89%, and AUC of 0.95.
Conclusions:
In this multicenter study, fractal analysis of 4D-CTP accurately differentiated CAD and CMD including subgroups with INOCA and with mild to moderate stenosis.
