Enterprise AI Analysis

Integrated motion-corrected extracellular volume fraction mapping reveals subtle extracellular remodeling in remote myocardium following chronic myocardial infarction

Published: 12 January 2026 | Authors: Wenzhi Wang, Tianyu She, Yuan Wang, Fei Wang, Menglu Wang, Shichuan Xu, Xiaoyi Duan, Liping Yang

Emerging evidence indicates that diffuse fibrosis in remote myocardium contributes to post-infarction ventricular remodeling, yet conventional late gadolinium enhancement (LGE) lacks sensitivity for extracellular matrix (ECM) quantification. This study establishes a semi-automated, motion-corrected extracellular volume fraction (ECV) mapping protocol to noninvasively characterize ECM expansion in non-infarcted myocardium after chronic myocardial infarction (CMI), addressing critical gaps in early detection of subclinical remodeling. Materials This prospective dual-center study enrolled 28 patients with CMI and 22 age-matched healthy controls. All participants underwent 3T cardiac magnetic resonance (CMR) imaging (Achieva, Philips Medical Systems) using a 32-channel phased-array cardiac coil. The imaging protocol included cine imaging, native T1 mapping, LGE, and post-contrast T1 mapping. Native and post-contrast T1 maps were acquired using a motion-corrected Modified Look-Locker Inversion Recovery (MOLLI) sequence. LGE images were obtained 10–15 min after intravenous injection of gadopentetate dimeglumine (0.1 mmol/kg). ECV maps were automatically generated using hematocrit (Hct)-adjusted T1 values. Myocardial segments were classified as infarct zone (IZ), remote zone (RZ), or normal zone (NZ) based on LGE thresholds (5 standard deviations above normal myocardium). Image analysis was performed using CVI42 software (Circle Cardiovascular Imaging) by two blinded radiologists, with reproducibility assessed via intraclass correlation coefficients (ICC) analysis. Results The study cohort comprised 28 CMI patients (64% male, age 52 ±4 years) with left ventricular ejection fraction 52±6% and 22 healthy controls. Image quality analysis revealed 90.7% of T1 maps (136/150 slices) were gradable (Grades III-IV), with 9.3% excluded due to artifacts. In CMI patients, IZ exhibited significantly higher native T1 (1388±76 ms vs. 1247 ±44 ms, P < 0.001), lower post-contrast T1 (521 ±28 ms vs. 632±31 ms, P<0.001), and elevated ECV (49%±8% vs. 32%±5%, P<0.001) compared to RZ. Despite comparable native T1 values between RZ and NZ (1247±44 ms vs. 1236±32 ms, P=0.15), RZ ECV was significantly higher than NZ (34 ±5% vs. 26±3%, P<0.05). Notably, ECV did not differ between normo-kinetic and dyskinetic RZ segments (29%±4% vs. 30%±4%, P=0.47). Automated ECV demonstrated excellent reproducibility, with intra-observer ICC of 0.95 and inter-observer ICC of 0.93 for RZ, outperforming manual region-of-interest analysis (ICC 0.82 and 0.75, respectively). Conclusions Automated ECV mapping detects subclinical ECM expansion in remote myocardium post-CMI, independent of regional wall motion abnormalities. This technology provides a quantitative tool for early identification of diffuse fibrosis, potentially guiding targeted therapies to mitigate ventricular remodeling.