摘要Two clinical ablation protocols,2C3L and stepwise,have been routinely used in our group to treat atrial fibrillation (AF),but with a less than 60％ long-term arrhythmia-free outcome achieved in patients.The goal of this study was to examine the underlying mechanism of low success in clinical outcome.MRI images from one patient were used to reconstruct a human atrial anatomical model,and fibrotic tissue was manually added to represent the arrhythmia substrate.AF was induced with standard protocols used in clinical practice.2C3L and stepwise were then used to test the efficacy of arrhythmia termination in our model.The results showed that re-entries induced in our model could not be terminated by using either 2C3L or the stepwise protocol.Although some of the induced re-entries were terminated,others emerged in new areas.Ablation using only the 2C3L or stepwise method was not sufficient to terminate all re-entries in our model,which may partially explain the poor long-term arrhythmiafree outcomes in clinical practice.Our findings also suggest that computational heart modelling is an important tool to assist in the establishment of optimal ablation strategies.