Normal values of high-resolution transmural perfusion distribution metrics for automated quantitative pixel-wise myocardial perfusion cardiovascular magnetic resonance.

No Thumbnail Available

All Authors

Kamani, CH.
Brown, L.
Anderton, T.
Tomoaia, R.
Soo, C.
Gulsin, GS.
Broadbent, DA.
Yeo, JL.
Wood, AL.
Saunderson, CED.

LTHT Author

Kamani, Christel
Broadbent, David

LTHT Department

Cardio-Respiratory
Cardiac MRI
Oncology
Medical Physics & Engineering
Radionuclide & MRI Physics

Non Medic

Clinical Scientist

Publication Date

2025

Item Type

Journal Article
Multicenter Study

Language

Subject

Subject Headings

Abstract

BACKGROUND: The myocardial blood flow (MBF) transmural distribution between the subendocardial (ENDO) and subepicardial (EPI) layers under resting and hyperemic conditions can aid in the diagnosis of several forms of heart disease. Recently proposed automated in-line myocardial perfusion cardiovascular magnetic resonance (CMR) allows pixel-wise quantification of ENDO- and EPI-MBF, but normal values for these parameters are lacking. We therefore aimed to establish normal values for transmural distribution of MBF in a healthy population. METHODS: 138 healthy participants from two centers underwent adenosine stress and rest myocardial perfusion CMR. Global and myocardial slice-specific stress/rest ENDO- and EPI-MBF values were derived using pixel-wise in-line automatic post-processing, and transmural perfusion metrics (ENDO and EPI myocardial perfusion reserve [MPRENDO, MPREPI]; stress and rest ENDO-to-EPI gradient [sGRAD and rGRAD]) were computed using the Gadgetron software. RESULTS: The study cohort comprised 84 males and 54 females (mean age: 50+/-36) with no cardiovascular disease or risk factors. In the entire cohort, MPRENDO (3.3+/-1.2) was significantly lower (p<0.001) than MPREPI (3.9+/-1.2). sGRAD (0.98+/-0.09) was significantly lower (p<0.001) than rGRAD (1.11+/-0.07). "While there were no sex-specific differences in the majority of these metrics, all correlated inversely with increasing age. We propose specific values for each slice. These are conditional to the pulse sequence, acquisition timing and analysis method used in this work, as mean +/- SD values at the basal, mid and apical level for MPRENDO (3.7+/-1.1, 3.3+/-0.9, 3.6+/-1.0), MPREPI (4.0+/-1.1, 3.9+/-1.1, 4.0+/-1.1), sGRAD (1.00+/-0.13, 0.92+/-0.09, 1.06+/-0.18) and rGRAD (1.10+/-0.09, 1.09+/-0.07, 1.18+/-0.11). CONCLUSION: Normal global and myocardial slice-specific values of MPRENDO, MPREPI, sGRAD and rGRAD using in-line automated MBF quantification from first pass myocardial perfusion CMR are presented. While there were no sex-specific differences in any of these metrics, all correlated inversely with increasing age. Understanding the MBF dynamics of the myocardial layers in healthy subjects will help to characterize MBF alterations in patients with coronary artery disease or microvascular dysfunction.

Journal

Journal of Cardiovascular Magnetic Resonance