Acute Microstructural Changes after ST-Segment Elevation Myocardial Infarction Assessed with Diffusion Tensor Imaging

• Published in: Radiology Vol. 299; no. 1; pp. 86 - 96
• Main Authors: Das, Arka, Kelly, Christopher, Teh, Irvin, Stoeck, Christian T, Kozerke, Sebastian, Chowdhary, Amrit, Brown, Louise A E, Saunderson, Christopher E D, Craven, Thomas P, Chew, Pei G, Jex, Nicholas, Swoboda, Peter P, Levelt, Eylem, Greenwood, John P, Schneider, Jurgen E, Plein, Sven, Dall'Armellina, Erica
• Format: Journal Article
• Language: English
• Published: United States 01.04.2021
• ISSN: 0033-8419; 1527-1315
• DOI: 10.1148/RADIOL.2021203208

Background Cardiac diffusion tensor imaging (cDTI) allows for in vivo characterization of myocardial microstructure. In cDTI, mean diffusivity and fractional anisotropy (FA)-markers of magnitude and anisotropy of diffusion of water molecules-are known to change after myocardial infarction. However, little is known about regional changes in helix angle (HA) and secondary eigenvector angle (E2A), which reflects orientations of laminar sheetlets, and their association with long-term recovery of left ventricular ejection fraction (LVEF). Purpose To assess serial changes in cDTI biomarkers in participants following ST-segment elevation myocardial infarction (STEMI) and to determine their associations with long-term left ventricular remodeling. Materials and Methods In this prospective study, 30 participants underwent cardiac MRI (3 T) after STEMI at 5 days and 3 months after reperfusion (National Institute of Health Research study no. 33963 and Research Ethics no. REC17/YH/0062). Spin-echo cDTI with second-order motion-compensation (approximate duration, 13 minutes; three sections; 18 noncollinear diffusion-weighted scans with values of 100 sec/mm [three acquisitions], 200 sec/mm [three acquisitions], and 500 sec/mm [12 acquisitions]), functional images, and late gadolinium enhancement images were obtained. Multiple regression analysis was used to assess associations between acute cDTI parameters and 3-month LVEF. Results Acutely infarcted myocardium had reduced FA, E2A, and myocytes with right-handed orientation (RHM) on HA maps compared with remote myocardium (mean remote FA = 0.36 ± 0.02 [standard deviation], mean infarcted FA = 0.25 ± 0.03, < .001; mean remote E2A = 55° ± 9, mean infarcted E2A = 49° ± 10, < .001; mean remote RHM = 16% ± 6, mean infarcted RHM = 9% ± 5, < .001). All three parameters (FA, E2A, and RHM) correlated with 3-month LVEF ( = 0.68, = 0.59, and = 0.53, respectively), with acute FA being independently predictive of 3-month LVEF (standardized β = 0.56, = .008) after multivariable analysis adjusting for factors, including acute LVEF and infarct size. Conclusion After ST-segment elevation myocardial infarction, diffusion becomes more isotropic in acutely infarcted myocardium as reflected by decreased fractional anisotropy. Reductions in secondary eigenvector angle suggest that the myocardial sheetlets are unable to adopt their usual steep orientations in systole, whereas reductions in myocytes with right-handed orientation on helix angle maps are likely reflective of a loss of organization among subendocardial myocytes. Correlations between these parameters and 3-month left ventricular ejection fraction highlight the potential clinical use of cardiac diffusion tensor imaging after myocardial infarction in predicting long-term remodeling. © RSNA, 2021