The thermal latency phenomenon seems to be the cause of significantly greater lesion depth after short-duration high-power RF pulses. Balancing the applied total energy when the voltage and duration are changed is not the optimal strategy since short pulses can also cause overheating.

Purpose: Most modeling studies on radiofrequency cardiac ablation (RFCA) are based on limiteddomain models, which means the computational domain is restricted to a few centimeters of myocardium and blood around the active electrode. When mimicking constant power RFCA procedures (e.g., atrial fibrillation ablation) it is important to know how much power is absorbed around the active electrode and how much in the rest of the tissues before reaching the dispersive electrode. Methods: 3D thorax full models were built by progressively incorporating different organs and tissues with simplified geometries (cardiac chamber, cardiac wall, subcutaneous tissue and skin, spine, lungs and aorta). Other 2D limited-domain models were also built based on fragments of myocardium and blood. The electrical problem was solved for each model to estimate the spatial power distribution around the active electrode. Results: From 79 to 82% of the power was absorbed in a 4 cm-radius sphere around the active electrode in the full thorax model at active electrode insertion depths of between 0.5 and 2.5 mm, while the impedance values ranged from 104 to 118 X, which were consistent with those found (from 83 to 103 X) in a 4 cm radius cylindrical limited domain model. Conclusion: The applied power in limited-domain RFCA models is approximately 80% of that applied in full thorax models, which is equivalent to the power programed in a clinical setting.