Discrimination of sustained VT in structural heart disease using LGE-CMR and computational simulation: moving beyond LVEF

Abstract

Background Sustained ventricular tachycardia (VT) is a major cause of sudden cardiac death in structural heart disease (SHD). Risk stratification based on left ventricular ejection fraction (LVEF) remains suboptimal. This study aimed to develop an exploratory model integrating late gadolinium enhancement cardiac magnetic resonance (LGE-CMR) scar characteristics with computational VT simulation to discriminate sustained VT, compared against LVEF. Methods In this single-centre retrospective study, consecutive patients with suspected SHD who underwent LGE-CMR were enrolled. Patients with LGE-positive underwent quantitative scar analysis, personalized digital heart-twin reconstruction, and virtual electrophysiological simulation. Results Among 82 patients (mean age 59.1 years; 82.9% male), 15 experienced sustained VT. Lower LVEF was associated with sustained VT (P = 0.008; C-statistic: 0.730), as was visual LGE positivity (P = 0.023; C-statistic: 0.634). Among LGE-positive patients, those with sustained VT had a larger core scar [10.0 [5.9, 14.8]% vs. 2.2 [1.1, 4.7]%; P < 0.001], larger grey zone [14.0 [10.9, 14.7]% vs. 5.7 [2.3, 9.2]%; P < 0.001], higher VT inducibility [50.0 [22.9, 66.9]% vs. 0 (0, 21.1)%; P < 0.001], and a greater number of inducible VT circuits [2 [2, 4] vs. 0 [0, 2]; P < 0.001]. An integrated LGE-VTsim index, derived from these four parameters, effectively discriminated sustained VT (C-statistic: 0.816) and remained independently associated with sustained VT after adjusting for LVEF (OR: 1.029; 95% CI: 1.008–1.051; P = 0.007). Conclusion The LGE-VTsim index, integrating quantitative LGE-CMR scar assessment with computational simulation, represents a promising exploratory parameter in identifying patients at risk of sustained VT in SHD, beyond LVEF.