A CMR-based predictive model incorporating LGE >5%, GCS-LV, and ls-HDF predicted ventricular arrhythmias with an AUC of 0.72 (95% CI 0.63-0.81), outperforming the LVEF <35% threshold.
Cohort (n=589)
Yes
Does a comprehensive CMR-based predictive model improve risk stratification for ventricular arrhythmias in patients with excessive trabeculation of the left ventricle compared to conventional LVEF thresholds?
A comprehensive CMR-based predictive model incorporating strain, LGE, and hemodynamic forces outperforms conventional LVEF thresholds for predicting ventricular arrhythmias in patients with excessive left ventricular trabeculation.
Hazard Ratio: 3.23 (95% CI 1.55–6.71)
p-value: p=0.002
Abstract Background Excessive trabeculation of the left ventricle (ETLV) represents a heterogeneous myocardial phenotype, still incompletely understood and with a debated clinical significance. Its relationship with adverse outcomes, particularly ventricular arrhythmias, remains controversial and insufficiently characterized. In this context, improving risk stratification through advanced imaging and functional parameters may help clinical decision-making and optimize patient outcomes. Purpose To develop a cardiac magnetic resonance–based predictive model for ventricular arrhythmias in patients with excessive trabeculation of the left ventricle (ETLV) Methods We conducted a retrospective longitudinal international study. A total of 589 patients with ETLV from 17 centres across Europe and North America were recruited. Core-lab CMR analysis was performed, including functional parameters (LVEF, strain, burden of LGE, hemodynamic forces, etc.) and morphologic variables of hypertrabeculation. The endpoint was a composite of ventricular arrhythmias (VA) including sudden cardiac death (SCD), sustained ventricular tachycardia (sVT), non-sustained ventricular tachycardia (nsVT) and appropriate ICD therapy. Cox regression analysis was performed. Results The mean age of participants was 43.2 years (18.2), and 254 (43%) were women. The average left ventricular ejection fraction (LVEF) was 47.4% (14.0), and late gadolinium enhancement (LGE) was observed in 65 patients (11.0%). Genetic testing was positive in 125 patients (49.6% of those tested). Over a mean follow-up period of 4.08 years (3.26), 53 patients (9.0%) experienced ventricular arrhythmias, including 5 cases of sudden cardiac death (0.8%), 4 cases of sustained ventricular tachycardia (1.9%), 12 appropriate ICD discharges (2.0%), and 41 episodes of non-sustained ventricular tachycardia (6.9%). Table 1 summarizes cardiac magnetic resonance (CMR) variables according to VA occurrence. Notably, none of the morphological markers of hypertrabeculation were associated with arrhythmic events. In multivariate analysis, the variables independently associated with VA were left ventricular global circumferential strain (GCS-LV) (HR 1.06, 95% CI 1.01–1.11, p = 0.033), presence of LGE 5% (HR 3.23 95% CI 1.55–6.71, p=0.002), and lateral-septal hemodynamic forces (ls-HDF) (HR 0.66, 95% CI 0.43–0.96, p=0.030). The predictive model demonstrated good discriminative ability, with an area under the curve (AUC) of 0.72 (95% CI 0.63–0.81), outperforming the conventional LVEF 35% threshold, which had an AUC of 0.66 (95% CI 0.57–0.74). Conclusions Comprehensive CMR analysis enables accurate risk stratification for ventricular arrhythmias, surpassing current guideline-based approaches. Morphological hypertrabeculation markers were not predictive for arrhythmic events. This model may improve ICD allocation and clinical outcomes.Baseline CMR variables Kaplan-Meier curves
Perez et al. (Thu,) conducted a cohort in Excessive trabeculation of the left ventricle (ETLV) (n=589). Presence of LGE > 5% was evaluated on Composite of ventricular arrhythmias (VA) including sudden cardiac death (SCD), sustained ventricular tachycardia (sVT), non-sustained ventricular tachycardia (nsVT) and appropriate ICD therapy (HR 3.23, 95% CI 1.55-6.71, p=0.002). A CMR-based predictive model incorporating LGE >5%, GCS-LV, and ls-HDF predicted ventricular arrhythmias with an AUC of 0.72 (95% CI 0.63-0.81), outperforming the LVEF <35% threshold.