Background: Arrhythmic risk stratification in patients with non-ischemic dilated cardiomyopathy (DCM) remains challenging. The LGE-dispersion mapping is a novel method for the quantification of tissue heterogeneity through the Global Dispersion Score (GDS). We sought to evaluate the usefulness of GDS in arrhythmic risk stratification of DCM patients. Methods: consecutive non-ischemic DCM patients underwent cardiac magnetic resonance imaging. GDS was calculated in LGE images. During a follow-up of 3.3 years (2 to 6 years) the combined endpoint of sudden cardiac death and appropriate implantable cardioverter-defibrillator intervention was considered. Results: The final population included 510 patients (mean age was 56±15 years). LVEF was > 35% in 241 patients (47%). LGE was present in 225 patients (45%). Median extent of LGE was 12% of LV mass (interquartile range -IQR- 6-20%). Among patients with positive LGE, GDS was 0.14 (IQR 0.08-0.20). During follow-up 81 patients had malignant ventricular arrhythmias (8 SCD, 73 appropriate ICD interventions). At Kaplan-Meier analysis, patients with GDS > 0.10 had worse prognosis than those with lower values of GDS (p < 0.0001). At multivariate analysis, GDS > 0.10 (HR 2.9, 95% CI 1.7-5, p = 0.0002) was an independent predictor of events. The prognostic value of GDS was confirmed in subgroups of patients with LVEF ≤ 35% and >35%. Conclusion: GDS is a useful marker to identify DCM patients at higher risk for malignant arrhythmic events regardless of LVEF and extent of LGE.
