Cardiac MR to Improve Clinical Risk Prediction in Defibrilator Patients

Implantable cardioverter defibrillators (ICDs) are the mainstay of primary prevention of sudden cardiac death (SCD) in high risk patients. In the US, over 5.7 million people have advanced left ventricular (LV) dysfunction and by current treatment guidelines over 800,000 patients qualify for a primary prevention ICD. The main criterion for patient selection is the presence of reduced ejection fraction. SCD risk is not static, and we and others have demonstrated that as many as 25% of patients with primary prevention ICDs no longer meet criteria for implantation at the time of PG exchange. While initial selection of patients for primary prevention ICD implantation are well established, there are no evidence based criteria to decide who will benefit from pulse generator (PG) replacement at the end of the battery life. Low levels of late gadolinium enhancement (LGE) prior to implant and improvement in LV function after implant are associated with reduced risk of mortality and appropriate shocks. Identifying effective criteria for PG replacement is a clinical and public health priority and a necessary step prior to large-scale clinical trials. There are two major subgroups of patients with ICDs who are unlikely to benefit from PG replacement: a) Patients with a very low probability of arrhythmic SCD and b) Patients with a high risk of death from competing causes. We hypothesize that cardiac magnetic resonance imaging (CMR) performed around the time of PG replacement together with clinical history, serum biomarkers and electrocardiographic (ECG) features will allow us to characterize subgroups of patients with distinct risk profiles who may not benefit from PG replacement. We will perform high-resolution CMR at the time of PG replacement among participants in PROSe-ICD, a unique, extensively phenotyped multicenter cohort of patients with systolic heart failure who received a primary prevention ICD. A subset of this cohort underwent CMR imaging at enrollment. The average follow up of the entire cohort is ~6 years and PROSE-ICD participants are entering a phase where many of them will require a PG replacement. We will supplement PROSe-ICD patients with patients undergoing primary prevention ICD PG replacement at Johns Hopkins and the University of Maryland. Utilizing this cohort and novel MR imaging methods, we aim to identify structural CMR characteristics along with clinical, serum and ECG biomarkers of PROSe-ICD patients who are at low risk for experiencing a potentially lethal ventricular tachyarrhythmia and at high risk for competing mortality. We will also determine if changes in structural CMR features between the pre-implant and the follow-up CMR in the subgroup of patients with a baseline study are associated with appropriate ICD shocks. CMR, in conjunction with clinical, ECG and serum protein metrics is the most promising tool to improve SCD risk prediction. This work has profound implications for improving the precision of risk assessment and the contemporary management of patients at risk for SCD and eligible for ICD implantation or PG replacement.