Why high intensity plaque is bright on MRI?

Abstract

Although CT is an excellent imaging modality for detecting lipid pools and calcification in the atherosclerotic plaque, it cannot distinguish hemorrhage from fibrous tissue. MRI has a unique capability of identifying subacute hemorrhage as high signal intensity area on T1-weighted MRI. Non-contrast T1-weighted MRI has been used to demonstrate high intensity plaque (HIP) in the coronary arterial wall. HIP was shown to be an important feature of a vulnerable plaque that increases the risk of future cardiac events. However, due to the lack of histopathological analysis of in vivo tissue samples from patients with HIP, the pathological features of HIP are still unknown. In the current issue of American Heart Journal Plus, Ehara et al. investigated the association between the PMR on non-contrast T1-weighted MRI and coronary intraplaque hemorrhage by analyzing plaque specimen obtained by directional coronary atherectomy. An antibody against glycophorin A was used to identify sites of previous plaque hemorrhage. They found a strong positive correlation between PMR and glycophorin A score. The results in the current study by Ehara reinforce the importance and implications of non-contrast T1-weighted MRI for the evaluation of HIP from a pathological perspective. Recent technical advances in MRI such as goldenangle radial sampling, self-gating, compressed sensing with deep learning reconstruction, and simultaneous acquisition of anatomical reference images and T1 weighted images may substantially reduce the total imaging time of coronary plaque MRI.

Most acute myocardial infarctions are caused by the rupture of coronary plaques and subsequent thrombus formation in the arterial lumen. Intravascular imaging, such as intravascular ultrasound (IVUS) and optical coherence tomography (OCT), is highly useful in assessing the compositions of atherosclerotic plaque that are closely associated with future occurrence of coronary events. However, the indications of intravascular imaging are limited to patients with acute coronary syndrome (ACS) or those at high risk for coronary artery disease (CAD). It is becoming increasingly important to identify vulnerable coronary plaques that are at high risk of future cardiac events by using non-invasive imaging techniques such as coronary CT, PET and MRI. Coronary CT is now most widely used for identifying high risk plaques, using high risk features such as positive remodeling, low attenuation plaque, napkin ring sign and spotty calcification. Although CT is an excellent imaging modality for detecting lipid pools and calcification in the atherosclerotic plaque, it cannot distinguish hemorrhage from fibrous tissue especially in small structures including coronary arteries.

MRI has a unique capability of identifying subacute hemorrhage as high signal intensity area on T1-weighted MRI. Early after bleeding, oxyhemoglobin (<1 day) and deoxyhemoglobin (1–3 days) are isointense on T1 weighted MRI. However, in the subacute phase (3 days to several weeks), methemoglobin exhibits paramagnetic effects similar to gadolinium MR contrast agents, shortening the T1 relaxation time of surrounding water protons and enhancing the MR signal from the water protons. Hemosiderin or ferritin in chronic hemorrhage is no longer paramagnetic and appears to be iso- or hypo-intense on T1-weighted MRI. Lipids in the necrotic core have a relatively short T1 relaxation time. However, lipids themselves do not enhance the signal of water protons, which account for the majority of MR signal from plaque tissue.

For the past decade, non-contrast T1-weighted MRI has been used to demonstrate high intensity plaque (HIP) in the coronary arterial wall. HIP on T1-weighted MRI was shown to be associated with ultrasound attenuation and remodeling demonstrated by IVUS and low plaque attenuation on CT. Noguchi investigated the prognostic value of non-contrast T1-weighted MRI in 568 patients with suspected or known CAD, and reported that HIP with a plaque-to-myocardium signal intensity ratio (PMR) ≥1.4 was associated with future coronary events with Hazard ratio of 3.96 [1]. These studies suggested that HIP is an important feature of a vulnerable plaque that increases the risk of future cardiac events. However, due to the lack of histopathological analysis of in vivo tissue samples from patients with HIP, the pathological features of HIP in the coronary arteries are still unknown, including the important point of what component of the plaque exhibits high signal intensity on T1-weighted images in patients. In a recent study by Kuroiwa, T1-weighted MR images were obtained in formalin-fixed hearts, and pathological findings in high PMR (≥1.4) and low PMR (<1.4) plaques were compared [2]. The frequency of intraplaque hemorrhage was significantly higher in high PMR plaques than in low PMR plaques. However, because MR images were obtained in autopsied hearts, T1 relaxation and signal intensities of coronary plaque and myocardium may be altered compared with in vivo measurements.

In the current issue of American Heart Journal Plus, Ehara et al. investigated the association between the PMR on non-contrast T1-weighted MRI and coronary intraplaque hemorrhage by analyzing plaque specimen obtained by directional coronary atherectomy [3]. An antibody against glycophorin A, a protein specific to erythrocyte membrane, was used to identify sites of previous plaque hemorrhage. They found a strong positive correlation between PMR and glycophorin A score. In addition, a weaker but significant correlation was found between PMR and macrophage scores. The results in the current study by Ehara reinforce the importance and implications of non-contrast T1-weighted MRI for the evaluation of HIP from a pathological perspective.

The importance of hemorrhage as indicated by glycophorin A score in coronary plaque vulnerability was previously investigated by Kolodgie et al., by analyzing autopsy samples of individuals who died suddenly from coronary events [4]. The glycophorin A score corresponded to the size of the necrotic core, and also paralleled the increase in the density of macrophages. Based on the close association among intraplaque hemorrhage, increased size of necrotic core and lesion instability, Kolodgie raised the possibility that intraplaque hemorrhage supplies cholesterol derived from erythrocyte membranes to the plaque, and the accumulation of erythrocyte membranes and Fe²⁺ or Fe³⁺ iron degraded from hemoglobin may stimulate inflammation and increase the instability of coronary atheroma.

It should be noted that it is critically important to obtain bright-blood 3D whole-heart coronary MR angiography in addition to free-breathing black-blood 3D T1 weighted MRI in order to properly interpret HIP, because of limited spatial resolution of black blood T1 weighted MRI and low contrast in the coronary arterial segments without HIP. However, the long acquisition time and operator dependency have limited widespread use of coronary plaque MRI. Recent technical advances such as golden-angle radial sampling, self-gating, compressed sensing with deep learning reconstruction, and simultaneous acquisition of anatomical reference images and T1 weighted images [5] may substantially reduce the total imaging time of coronary plaque MRI and eliminate complex planning in MRI acquisition. Recently, the feasibility of coronary artery PET/MR imaging was reported by using ¹⁸F-fluorodeoxyglucose (^18F-FDG) for plaque inflammation or ¹⁸F-sodium fluoride (¹⁸F-NaF) for microcalcification [6]. PET/MR has a potential to reveal multiple markers of coronary plaque biology, and may become a valuable tool for identifying patients with high risk for coronary events.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.