Quantitative assessment of myocardial fibrosis by digital image analysis: An adjunctive tool for pathologist “ground truth”

Myocardial fibrosis (MF), meaning the excessive deposition of extracellular matrix (ECM) components in the myocardium, is a common pathological process in a wide spectrum of chronic heart diseases [1,2]. Being associated with the disruption of normal myocardial structure, it is behind the mechanistic base for adverse cardiac remodeling [3]. Indeed, it is a key contributor to heart failure and its progression, having recognized prognostic implications in both ischemic and non-ischemic cardiac conditions [4]. In this regard, several attempts have been recently made in trying to integrate cardiac fibrosis assessment into distinct clinical scenarios, namely for heart failure management, patients risk stratification and even therapeutic intervention before symptoms development [5].

Endomyocardial biopsy (EMB) histopathology remains the gold-standard method to diagnose MF. Additionally, in theory, the application of histomorphometry parameters at specific fibrosis-stained sections is the most accurate technique for MF quantification [6]. However, EMB samples may not be representative of a non-uniform, sometimes patchy, pathological process. Besides, it has low feasibility, as determined by limited availability, invasiveness and need for expertise in cardiac pathology interpretation [1]. Thereby, non-invasive imaging methods have been developed.

Multiparametric cardiac magnetic resonance (CMR) is currently the best imaging modality that offers a direct, whole heart assessment of myocardial fibrosis. Myocardial T1 mapping and associated techniques yield improved myocardial characterization through the ability to quantify signal intensity for each voxel in the myocardium [7]. Extracellular volume (ECV) fraction, as derived from combined pre- and postcontrast T1 mapping, seems to be particularly sensitive to extracellular space expansion. Previous single centre studies in patients with severe aortic stenosis (AS) have shown good correlations between ECV and histological diffuse fibrotic burden, as assessed by quantitative morphometry [7,8]. As for other clinical contexts, ECV obtained using CMR is poised as an ideal imaging surrogate marker for predicting diffuse MF [9]. Nevertheless, correlation results from distinct clinical cohorts are far from uniform, and some of these studies did not find associations between the ECV and collagen volume fraction in EMB samples [10]. Myocardial infiltration, oedema, and inflammation remain important potentially confounding sources of increased ECV, as imaging markers do not measure fibrous tissue directly, but the total interstitial space instead. Additionally, there are qualitative aspects related to the composition of fibrous tissue, such as the type of collagen fibres and molecular organization, with potential impact on non-invasive assessment of extracellular myocardial component [1]. Finally, and no less than important, these correlation studies with histopathology rely on the accuracy of MF quantification provided by the pathologist, which is totally operator-dependent and highly sensitive to the experience level [11]. Recently, digital algorithms for automatic morphometry started to be developed, aiming to improve the reproducibility of MF quantification. Such complementary tools would fulfil an unmet need, increasing the precision of MF invasive quantification.

Our aim is to assess if the use of an automated artificial intelligence software for the quantification of MF on EMB, might improve individual pathologist´s quantification and agreement, in terms of precision.

留言 (0)

沒有登入
gif