Quantitative assessment of myocardial scar in delayed enhancement magnetic resonance imaging

Abstract

Purpose: To characterize the extent and distribution of left ventricular myocardial scar in delayed enhancement magnetic resonance imaging (MRI). Materials and Methods:Delayed enhancement images from 18 patients were categorized into three groups based on myocardial scar appearance: discrete myocardial infarction (N ϭ 10), diffuse fibrosis (N ϭ 4), and circumferential endocardial scarring (N ϭ 4). Images were segmented manually by two observers (twice by one observer) to identify nonviable myocardium. Scar was characterized by the following morphologic parameters: the relative area of nonviable myocardium (Percent Scar); a measure of scar cohesion (Patchiness); and the extent to which scar traversed the ventricle wall (TransϾ50). Results:The three scar parameters successfully discriminated between patient groups, although no one parameter was able to differentiate between all groups. The average bias between readers was approximately 3% for each parameter, and the average bias between repeated measurements was 1%. In addition, five patients exhibited regions of nonhyperenhanced nonviable myocardium that were expected to show hyperenhancement based upon their location within the infarct zone and appearance on cine images. Conclusion:Quantitative characterization of myocardial scar showed good interobserver and intraobserver agreement. However, the appearance of nonhyperenhanced scar in chronic ischemia is problematic for segmentation of delayed enhancement images. CHARACTERIZING THE LOCATION and extent of myocardial injury due to infarction or chronic ischemia of the left ventricle (LV) is essential for proper clinical management of affected patients (1,2). It has been shown that patients with regions of hypocontractile yet viable myocardium benefit from reperfusion therapy or revascularization (3-5); furthermore, that the relative contributions of viable and nonviable tissue within a given segment of myocardium are related to the ultimate success of reperfusion (4 -6). However, most noninvasive techniques for assessing myocardial viability lack sufficient spatial resolution or tissue characterization capabilities to discriminate between viable and nonviable tissue within the LV wall (e.g., stress cine MRI, stress echocardiography, nuclear radiographic techniques); the most widely accepted reference standard for identifying viable myocardium remains recovery of function after revascularization (7).Delayed hyperenhancement (DE) of nonviable myocardium in contrast-enhanced magnetic resonance imaging (MRI) has been described in numerous previous studies (4 -6,8 -12), and the close correspondence between the area of hyperenhancement and infarct extent and distribution has been well established (10,13). Furthermore, DE-MRI has sufficient resolution to accurately distinguish viable (normal or ischemic) from nonviable LV myocardium (14).Qualitative or semiquantitative analysis of DE-MRI images can be subjective, which limits direct comparison of results between sequential studies and might limit its clinical application...

Segmentation of non-viable myocardium in delayed enhancement magnetic resonance images

“…Evaluation was accomplished by comparing results from each technique with results from manual thresholding, which has recently been established as a reliable technique for segmentation of DE images [10], and serves as the de facto gold standard for scar quantification in the absence of direct histological correlation.…”

“…With this software, users employ a suite of tools including an interactive thresholder and paint/erase brushes to label non-viable tissues, as described previously [10].…”

“…The interobserver and intraobserver agreement for the manual thresholding technique have been evaluated previously [10], in a study which dem onstrated no statistically significant differences in Percent Scar either between or within readers. Also, agreement by Bland-Altman analysis was good, with a bias of only 1% Percent Scar both between and within readers.…”

“…Previous studies using DE-MRI have typically relied on either visual inspection of images with a semi-quantitative analysis [3][4][5][6], or manually drawing contours around non-viable tissue [9,10], which can be time prohibitive. Semi-automatic segmentation techniques, based on the image intensity characteristics of viable myocardium, have been used extensively [7,8,[11][12][13][14], but have not been validated previously in the literature.…”

Head‐to‐head comparison between delayed enhancement and percent infarct mapping for assessment of myocardial infarct size in a canine model

“…Thus, any volume element is deemed to be either infarcted or healthy but voxels with a mixture of healthy and infarcted myocardium are not quantified accurately. This limits direct comparison of results among sequential studies and might limit the method's clinical application (24), as the viability change can be minor in any single voxel, but summing them up to obtain the percent infarct in all patchy infarcted voxels in the border zone can reveal a major contribution to transmurality when using PIM. Also, standardized, quantitative descriptions of myocardial scar would be essential for comparisons between studies, either in following an individual's response to therapy or in making comparisons among patients regarding diagnosis, therapeutic response, or prognosis (24).…”