Pedro M. Lopes

🧠 Beyond LVEF: improving arrhythmic risk stratification in non-ischemic cardiomyopathy A new validation study explores the performance of the DERIVATE Risk Score 2.0 (DRS 2.0) in predicting major arrhythmic events in patients with non-ischemic cardiomyopathy (NICM) 💡 Why is this relevant? Sudden cardiac death (SCD) risk stratification in NICM remains challenging. For decades, decision-making has been largely based on: 👉 LVEF <35% But: many patients with low EF never experience arrhythmias others with relatively preserved EF still develop malignant events 📌 We clearly need better, multidimensional risk models. 🧬 What is DRS 2.0? A simple CMR-based score integrating: male sex CMR-derived LVEF presence and extent of midwall LGE 👉 A combination of function + tissue characterization 📊 Key findings from this Portuguese validation cohort (385 patients, median LVEF 37%) Follow-up: ~3 years Major arrhythmic events (MAACE): 6.8% Risk stratification: 🟢 Low risk → 0.4%/year 🟡 Intermediate risk → 1.0%/year 🔴 High risk → 6.0%/year 👉 Strikingly: 28% of patients (high-risk group) accounted for 73% of all arrhythmic events Hazard ratio: 8.0 for high-risk vs others Discrimination: C-statistic 0.76 Calibration: excellent 🔍 What drives arrhythmic risk? lower LVEF larger LV volumes midwall and septal LGE extent of fibrosis 👉 Not just presence, but pattern and burden matter. 🧬 Does genetics have a role in this? 👉 Probably yes — and this is the next step. Although genetics was not included in DRS 2.0, we know that: specific variants (e.g. LMNA, FLNC, DSP) confer high arrhythmic risk genotype can influence fibrosis pattern and disease progression 📌 Future models will likely integrate: CMR + clinical variables + genotype ⚠️ Clinical implications ✔️ LVEF alone is not enough ✔️ CMR should be central in risk stratification ✔️ DRS 2.0 identifies a small high-risk subgroup with disproportionate event burden ✔️ Potential impact on: 👉 ICD decision-making 🧠 Take-home message This study reinforces a key shift: 👉 from EF-based ➡️ to substrate-based risk stratification And soon… 👉 to integrated phenotype + genotype models doi.org/10.1016/j.repc.2026.…