Hyperacute Incidental Late Myocardial Enhancement in Ischemic Stroke using Chest Spectral CT: Relationship with Etiology

PurposeHyperacute cardiac imaging of patients with acute ischemic stroke (AIS), though desirable, is impractical. Using delayed-enhancement, low-dose, non-gated, chest spectral computed tomography scans (DESCT), we explored the prevalence and patterns of incidental myocardial late iodine enhancement (LIE) and embolic sources, and the relationship with stroke etiology. MethodsSince July 2020, DESCT was performed after cerebrovascular CT angiography (CTA) among patients with suspected AIS undergoing CT using a dual-layer spectral scanner, without additional contrast administration. Images were analyzed using monoenergetic reconstructions and iodine density maps, and the myocardial extracellular volume fraction (ECV, %) was calculated. ResultsEighty consecutive patients with AIS were included. DESCT identified a cardiac thrombi in 6 patients (7.5%), and a complex aortic plaque in 4 (5%) cases; reclassifying 5 embolic strokes of uncertain source (28% of ESUS) to cardioembolic (CE, n=3) and non-CE (n=2) etiologies. LIE, most commonly ischemic (82%), was identified in 38 (48%) patients. We did not identify significant relationships between AIS etiology and the presence, pattern, and extension of LIE (p>0.05); ECV (p=0.56), severe aortic (p=0.25) or valvular (p=0.26) disease, or the extent of coronary calcification (p=0.39). Patients with evidence of major cardiovascular DESCT findings had higher rate rates of all-cause death at 90 days (42% vs. 19%, p=0.037). ConclusionIn this study, hyperacute cardiac imaging of AIS with DESCT identified a high prevalence of incidental cardiac disease predominantly involving LIE of ischemic etiology and mostly not related to the stroke etiology.

Computed tomography for myocardial characterization in ischemic heart disease: a state-of-the-art review

This review provides an overview of the currently available computed tomography (CT) techniques for myocardial tissue characterization in ischemic heart disease, including CT perfusion and late iodine enhancement. CT myocardial perfusion imaging can be performed with static and dynamic protocols for the detection of ischemia and infarction using either single- or dual-energy CT modes. Late iodine enhancement may be used for the analysis of myocardial infarction. The accuracy of these CT techniques is highly dependent on the imaging protocol, including acquisition timing and contrast administration. Additionally, the options for qualitative and quantitative analysis and the accuracy of each technique are discussed.

P5679Utility of late iodine enhancement computed tomography with image subtraction in the evaluation of cardiac resynchronization therapy response

Background Despite the impressive results of the large CRT trials, it has been observed that, on an individual basis, about 30% of patients fail to respond to cardiac resynchronization therapy (CRT). The evaluation of left ventricular (LV) dyssynchrony, myocardial scar, and coronary venous anatomy by image subtraction in Late Iodine Enhancement Computed Tomography (LIE-CT) has the potential to comprehensively characterize non-responders. Purpose To assess the feasibility and the utility of image subtraction in LIE-CT in CRT patients and compare findings between responders and non-responders. Methods Monocentric prospective study of CRT patients at least 6 months after implantation who underwent post-procedural CT between March and October 2018. CRT-responders were defined as patients with an absolute increase in LV ejection fraction >5%. CT-derived residual global and segmental dyssynchrony metrics, extent and location of myocardial scar, coronary venous anatomy, and position of LV lead relative to scar and segment of latest mechanical contraction were analyzed. Results Among the 29 patients (mean age 71±12 years; 72% men), 18 were responders (62%). All CT metrics evaluating residual dyssynchrony such as wall motion indexand wall thickness indexwere worse in non-responders (p<0.0001 for both). In LIE-CT, predictive factors of CRT-non-response were an LV lead localized in an region of myocardial scar (p=0.0007), in a region with akinesia or dyskinesia (p=0.007), and with myocardial thickness <6mm (p=0.002). Percentage of fibrosis of the myocardial mass and the presence of fibrosis in postero-lateral region were not predictive of CRT-non-response (p=0.9 and p=0.3, respectively). Of the 11 non-responder patients, 8 (73%) had at least one other coronary venous branch visualized by CT; and among those, 3 (38%) were located in an non-akinetic area with late segmental contraction. Wall Motion and LIE-CT Conclusion Image subtraction in LIE-CT in patients who had CRT is feasible and allows better characterization of CRT-non-responders, who have a greater amount of residual dyssynchrony than responders. Distribution of fibrosis in relation to the LV lead and presence of alternative venous branches may help patient management.