Performance of the DERIVATE Risk Score 2.0 for predicting major arrhythmic events in non-ischemic cardiomyopathy: A...

Sudden cardiac death (SCD) risk stratification in non-ischemic cardiomyopathy (NICM) remains challenging. The recently proposed Derivate Risk Score 2.0 (DRS 2.0) seeks to enhance risk prediction using three variables: sex, cardiac magnetic resonance-derived left ventricular ejection fraction (CMR-LVEF), and the presence of midwall late gadolinium enhancement (LGE). The aim of this study was to assess the accuracy of the DRS 2.0 for predicting Major arrhythmic events (MAACE) in a Portuguese NICM cohort.

Patients with the clinical diagnosis of NICM who underwent CMR for clinical purposes were included in this single-center retrospective study. Patients with obstructive coronary artery disease, infiltrative, inflammatory or hypertrophic cardiomyopathy, or significant valvular or congenital heart disease were excluded. The DRS 2.0 score was calculated for each patient based on male sex, CMR-derived LVEF and on the presence and pattern of midwall LGE as reported. The primary endpoint was major arrhythmic cardiac events (MAACE), defined as SCD, sustained ventricular tachycardia (VT), or appropriate ICD therapy (shock or ATP). Follow-up was performed using clinical records and device interrogations.

A total of 385 patients were enrolled (age 62.5 years (±14.9); 63% male, median CMR-LVEF 37% (27-43), midwall LGE present 51.4%). One hundred two patients had an ICD (26.5%), 5 had an ICD previous to CMR and the remaining 97 patients received one during follow up. The median DRS2.0 score was 6.0 (IQR 4.0-9.0). Risk distribution according to the DRS 2.0 was 20% low risk (n=76), 53% intermediate risk (n=203), and 28% high risk (n=106). During a mean follow-up of 3.0±1.4 years, 26 MAACE (6.8%) occurred, including 7 SCD, 9 sustained VT, and 10 appropriate ICD therapies. The annualized MAACE rates were 0.4%/year in low risk, 1.0%/year in intermediate-risk, and 6.0%/year in the high-risk patients. Patients classified as high risk by DRS 2.0 represented 28% of the cohort yet accounted for 73% of all MAACE (HR 8.00, 95% CI 3.36-19.04, p<0.001). The DRS 2.0 showed good discrimination (C statistic 0.76, 95% CI: 0.66-0.85, p<0.001). Calibration was excellent, as indicated by a non-significant Hosmer-Lemeshow test (χ²=0.802, p=0.371).

DRS 2.0 showed robust discrimination and calibration in our Portuguese NICM cohort, effectively identifying a small subgroup that accounted for most arrhythmic events. This score appears applicable in our population and may assist in refining SCD risk stratification and management.