Even though finite element human torso models have been increasingly used to study ventricular defibrillation field and to optimize defibrillation electrode systems, effect of human anatomical variation on defibrillation thresholds (DFTs) has not been well investigated. In order to investigate the anatomical variables that affect ventricular DFTs, finite element torso models were built from MRI and fast CT images of two patients. One patent had a dilated left ventricle (LV) and left atrium, and LV hypertrophy, 30% ejection fraction and a relatively small RV. The other patient had large cardiac dimensions (upper bound of normal) but no heart disease. The primary defibrillation electrode systems, i.e. SVC-RV, Can-RV and Can+SVC-RV, were analyzed and compared in the two models. The DFTs calculated based 95% ventricular mass with voltage gradient greater than 5 V/cm were also compared with the clinical data. The FEA results show that anatomical variation can affect ventricular DFTs of certain lead configurations. Published clinical DFTs with these configurations correlate better with the normal heart model suggesting that it may be a better representation of an average ICD patient. The DFTs with Can+SVC-RV and Can-RV systems are robust and relatively insensitive to these anatomical differences.
This is joint work with Rahul Mehra.