Patient-specific computational modeling of internal defibrillation

Conventional transvenous defibrillation is performed with an implantable cardioverter-defibrillator (ICD). The defibrillation shock is delivered from the right ventricle (RV) electrode to the superior vena cava (SVC) electrode and the metallic housing (CAN) of the ICD. The primary goal of this study was to assess the predictive capacity of patient-specific computational models by comparing simulated and clinical defibrillation thresholds (DFT). Nine 3-D patient-specific models of the thorax and in situ electrodes were created from segmented CT images taken shortly after implantation of the cardioverter-defibrillator. The electric field distribution during defibrillation was computed using the finite volume method. The critical mass hypothesis was used to extract the DFTs from the calculated field distribution. The simulated DFTs were well matched to the clinically determined thresholds in 4 of the 9 patients examined. Although the critical mass criterion was not able to consistently predict individual patient thresholds, the fact that it was predictive in four of nine cases is significant. Inspection of the distribution of the weak field in all nine patients revealed a relationship between the degree of dispersion of the weak field and the clinical DFT that could be exploited to identify high DFT patients prior to implant.; The secondary goal of the project was to use the library of patient-derived computational models to determine the biventricular defibrillation fields and thresholds and compare to conventional transvenous defibrillation. Biventricular defibrillation uses an additional electrode placed on the left ventricular (LV) free wall. The defibrillation shock consisted of monophasic pulse delivered from the LV electrode to the SVC and CAN electrodes, followed by a biphasic pulse delivered from the RV electrode to the SVC and CAN electrodes. At equal shock strength, the simulations show that biventricular configuration reduces the DFT in all patients by 14–50%. An increase in DFT of 10–21% was observed in four models, when the leading edge of the biphasic shock was equal to the trailing edge of the monophasic shock. These mixed results are consistent with clinical reports and suggest that patient-specific computational models may be able to identify those patients who could benefit from biventricular configurations.