Ultrasound validation of mathematically modeled irreversible electroporation ablation areas

Objective. Currently, the prediction of the dimensions of irreversible electroporation (IRE) ablation is modeled using algorithms derived from mathematical and ex vivo models. These algorithms have not been validated using in vivo studies. The aim of this study was to assess the correlation between the mathematical prediction model to and ultrasonographic and histopathologic findings for in vivo ablations in a porcine model. Methods. IRE ablations were performed on porcine liver and spleen with probe spacings ranging from 0.6 to 2.6 cm. Pre and 2-hour postablation ultrasound (US) images were recorded and validated with confirmation by histopathology. Three dimensions of the regions of ablation were recorded, and ablation volumes were calculated and correlated with theoretic mathematical models for each given probe spacing. Results. In vivo axial and anterior-posterior (AP) distances of ablation were greater than predicted for nearly all probe spacings (P < .05). Ablation volumes were significantly less than predicted for the all probe spacings when modeled using both a cylinder and an ellipsoid. Geometrically, mathematically derived regions of ablation demonstrated more central tapering ("necking") and diminished volumes compared to their in vivo counterparts. The relationships between probe spacing and AP dimensions of ablation were less linear (r(2) = 0.57) than the relationships observed via ultrasonography. Conclusion. The current mathematical models predict regions of ablation observed in vivo poorly. They underestimate dimensions of ablation and, by extension, the volumes of ablation. Further work should be done to improve models for ablative planning, and physicians should recognize the limitations of existing models when planning ablative treatments.