Personalizing Risk Stratification for Sudden Death in Dilated Cardiomyopathy: The Past, Present, and Future

Results from the DANISH Study (Danish Study to Assess the Efficacy of ICDs in Patients with Non-ischemic Systolic Heat Failure on Mortality) suggest that, for many patients with dilated cardiomyopathy (DCM), implantable cardioverter defibrillators (ICD) do not increase longevity. Accurate identification of patients who are more likely to die of an arrhythmia and less likely to die from other causes is required to ensure improvement in outcomes and wise use of resources. Until now, left ventricular ejection fraction (LVEF) has been used as a key criterion for selecting patients with DCM for an ICD for primary prevention purposes. However, registry data suggest that many patients with DCM and an out-of-hospital cardiac arrest do not have a markedly reduced LVEF. Additionally, many patients with reduced LVEF die from non-sudden causes of death. Methods to predict a higher or lower risk of sudden death include the detection of myocardial fibrosis (a substrate for ventricular arrhythmia), microvolt T-wave alternans (MTWA; a marker of electrophysiological vulnerability) and genetic testing. Mid-wall fibrosis is identified by late gadolinium enhancement cardiovascular magnetic resonance imaging in around 30% of patients and provides incremental value in addition to LVEF for the prediction of SCD events. MTWA represents another promising predictor, supported by large meta-analyses that have highlighted the negative predictive value of this test. However, neither of these strategies have been routinely adopted for risk stratification in clinical practice. More convincing data from randomized trials are required to inform the management of patients with these features. Understanding of the genetics of DCM and how specific mutations affect arrhythmic risk is also rapidly increasing. The finding of a mutation in LMNA, the cause of around 6% of idiopathic DCM, commonly underpins more aggressive management due to the malignant nature of the associated phenotype. With the expansion of genetic sequencing, the identification of further high-risk mutations appears likely, leading to better informed clinical decision-making as well as providing insight into disease mechanisms. Over the next 5-10 years we expect these techniques to be integrated into the existing algorithm to form a more sensitive, specific and cost-effective approach to the selection of DCM patients for ICD implantation