Atrial fibrillation (AF) is the most common cardiac arrhythmia, affecting an estimated 5.5 million people worldwide, including nearly 2.5 million in the United States. Approximately 350,000 new cases are diagnosed each year, with the number of sufferers increasing annually as the nation's population ages.1 By the year 2050, AF will be an even more significant public health concern, affecting an estimated 10 million people in this country alone.2 Our laboratory developed the gold standard operation for AF over 20 years ago, and despite the advent of radiofrequency ablation technology, the outcomes of the Cox Maze (CM) IV procedure remain widely variable and unpredictable in the individual patient. Clinical experience and experimental evidence suggest that further refinement of established surgical technique should thus be abandoned in favor of a new patient-specific theoretical approach. The critical mass hypothesis states that a certain minimal size of atrial tissue, representing the minimum path length necessary for reentry, is required for the induction and maintenance of AF.17 Our laboratory and others have demonstrated the validity of this theory in both in-vitro and in-vivo models, and have shown that methods such as epicardial electrode mapping and electrocardiographic imaging (ECGI), together with CT scanning, can be used to obtain patient-specific heart- torso geometry and cardiac electrophysiologic activation maps upon which a patient-specific operation can be based.19,20,22-26 We hypothesize that, by using these techniques, we can develop a model to reliably predict the probability of sustained AF based on the area or width of atrial tissue left intact after surgical ablation, and that this model can then be used to create subject-specific lesion sets that will render the atria unable to sustain AF, yet still permit activation during sinus rhythm.15 We will develop our model in a Hanford mini-swine model of acute AF (n=6), whereby following a preoperative CT scan, a radiofrequency ablation device will be used to sequentially partition the atria into relatively equal segments according to a universal pattern until sustained AF is undetectable by epicardial mapping. The model will be refined to account for the electrophysiologic changes that occur with chronic AF in a second Hanford mini-swine model (n=6) that will undergo rapid atrial stimulation at a rate of 190 5 bpm for one month prior to the procedure. The efficacy of this model to eliminate AF will be tested in a third group of chronic mini-swine (n=6) by using subject-specific CT and ECGI data obtained preoperatively to construct a lesion set for each animal, and subsequently recording any sustained AF following the procedure with epicardial electrodes. Finally, we will use individual CT and ECGI data to test the ability of our model to predict the sustainability of AF in human patients (n=24) with recurrent AF after a surgical intervention. By developing and refining a model that subsequently demonstrates efficacy in predicting human pathophysiology, we will be one critical step closer to achieving our long-term goal of creating a patient-specific surgical cure for AF.
Although it has been known for decades that a critical area of atrial tissue is required to sustain fibrillation, most surgical and catheter ablation lesion sets performed thousands of times per year are based on a theory that is now known to be wrong in many, if not all, patients with AF. The proposed research will attempt to establish a method to reliably quantitate the maximum area of atrial tissue needed to support AF in an individual subject, and use this data to create subject-specific lesion sets that will render the atria unable to sustain the arrhythmia. The knowledge gained from this project will aid in the development of a novel patient-specific approach to AF surgery that uses individual anatomic and electrophysiologic parameters measurable preoperatively to make AF unsustainable, while minimizing the extent of the lesion set required to cure AF, in an individual patient.