Feed

Sudden cardiac death (SCD) remains a major cause of mortality in nonischemic cardiomyopathy (NICM) despite advances in heart failure therapy and implantable cardioverter-defibrillator (ICD) technology. Risk stratification remains challenging because left ventricular (LV) ejection fraction (LVEF) alone incompletely captures arrhythmic substrate, missing many patients at risk while leading others to receive ICDs without subsequent appropriate therapy. Cardiac magnetic resonance (CMR), particularly late gadolinium enhancement (LGE), has emerged as a central tool for identifying myocardial fibrosis and refining arrhythmic risk across NICM phenotypes. However, CMR imaging alone does not fully define arrhythmogenicity, especially in diseases with layered, septal, or intramural scar. We present a case-based framework illustrating how multimodality imaging integrates into SCD risk stratification across hypertrophic cardiomyopathy, cardiac sarcoidosis, genetic arrhythmogenic cardiomyopathy, and cardiac amyloidosis. These cases demonstrate several core principles. First, CMR can identify arrhythmogenic substrate and often refines risk beyond conventional clinical markers. Second, multimodality imaging can provide additive information, but findings must be interpreted within disease-specific context, as the prognostic implications differ across phenotypes. Third, invasive electrophysiologic evaluation, including programmed stimulation and electroanatomic mapping, can further stratify arrhythmic risk when imaging findings are incomplete or discordant with the clinical presentation. Finally, genetic testing complements imaging and invasive assessment by refining disease classification and, in selected cases, contributes to arrhythmic risk assessment by considering variant classification and wholistic clinical phenotype. Together, these observations support a substrate-based rather than function-based approach to SCD risk stratification in NICM. Multimodality imaging is foundational, but optimal assessment requires integration of imaging, electrophysiology, genetics, and clinical phenotype. Future progress will depend on longitudinal risk modeling, improved characterization of electrically active substrate, and incorporation of patient-specific tools into primary prevention ICD decision-making.