Search

Introduction:Hypertrophic cardiomyopathy (HCM), a disease of progressive myocardial fibrosis, is a significant cause of sudden cardiac death (SCD) in the adult population. Patients with HCM are typically asymptomatic but can develop ventricular arrhythmia (VA) which can lead to SCD. Current risk stratification criteria (ACCF/AHA and ESC) fail to accurately identify all patients at risk for VA and in need of an implantable cardioverter defibrillator (ICD), a procedure that is not without risk. Previously, virtual-heart technology has been used to stratify SCD risk in the ischemic population.Hypothesis:We hypothesize that HCM-specific virtual-heart technology improves SCD risk stratification compared to current clinical practice.Methods:In this retrospective study, we used clinical data from 19 HCM patients who have ICD, but with only 9 experiencing VT. We reconstructed personalized left ventricular computational models from short-axis LGE-MRI acquired prior to ICD implantation. Non-fibrotic myocardium, diffuse fibrosis, and dense scar were classified by intensity-based thresholding. A human HCM-specific ionic model was used to incorporate the electrophysiological changes caused by fibrotic remodeling and fiber orientation was incorporated based on the patient-specific geometry (Fig 1). Rapid pacing was then performed to assess VT inducibility. We compared the accuracy of our approach to conventional risk criteria, the proportion of fibrotic remodeling, and the amount of hypertrophy for each patient.Results:Our virtual-heart technology was 68% accurate in predicting VA inducibility, an improvement of 21% from clinical risk scores. The proportion of fibrotic myocardium and percent of wall thickness > 15 mm were not significantly associated with clinical VA.Conclusions:Personalized virtual-heart technology improves VA risk stratification and accurately identifies patients with HCM in need of ICD compared to current clinical risk scores.