Our laboratory is equipped with an EP MedSystems recording system (EP MedSystems, Inc., West Berlin, New Jersey), intracardiac ultrasound, and both Carto (Biosense Webster Inc., a Johnson & Johnson company, Diamond Bar, California) and EnSite NavX (St. Jude Medical, St. Paul, Minnesota) mapping systems. AF Approach For AF ablation, a classic activation map-guided semi-circumferential pulmonary vein isolation approach with the LASSO catheter (Biosense Webster Inc.) is used. Ablation lesions are placed at the outer pulmonary vein ostium, and if possible, pulmonary veins are isolated at the level of their common trunk. Additional right and left atrial linear ablations are performed only if sustained and organized right and/or left atrial flutters are induced and entrained after isolation of the pulmonary veins. Other organized arrhythmias are also targeted for ablation. However, left atrial linear ablation is avoided due to concerns of creating new arrhythmia substrates with incomplete lines. The majority of AF procedures are performed with the NavX system (St. Jude Medical), which allows the flexibility to use various catheters as well as visualization with the LASSO catheter. Preprocedural CT scans with three-dimensional (3D) reconstruction are used to understand PV anatomy; however, image integration is not routinely performed. Pulmonary venography and intracardiac echocardiography are used to define the location of the vein ostium. Residual organized tachycardias are routinely mapped with pacing maneuvers and entrainment mapping. Electroanatomical mapping is used with caution since it is often difficult to assign early and late electrograms in rapid atrial tachyarrhythmias, which frequently results in misleading activation maps. Results Our first 100 AF cases were performed under conscious sedation with a 4 mm non-irrigated RF catheter. Then we switched to an 8 mm tip catheter, and all of these cases were done under general anesthesia, which allows for esophageal visualization by barium esophagogram. Since routine intra-operative barium opacification of the esophagus was performed, the close relationship between the ablation site and esophagus particularly in the lower pulmonary veins became apparent. In cases of direct proximity of the ablation catheter to the esophagus, endoesophageal deflection with a TEE probe was attempted but found to be rather cumbersome. Although the team at Tampa General Hospital has never encountered problems with esophageal injury in patients, a hybrid cryo-RF ablation approach is routinely used for all lesions delivered in close proximity to the esophagus. This strategy is based on the safety profile of cryoablation in clinical practice as well as animal experiments that have shown that cryo energy delivered directly to the esophagus does not result in significant tissue disruption and that the tissue in the periphery of the ice ball does not become necrotic. Sixty hybrid cryo-RF cases for atrial fibrillation have been performed thus far. The team s initial impression about the efficacy of this ablation approach is good, and long-term follow-up results are currently being evaluated. We feel the positive initial impressions may be related to being able to deliver complete left atrial lesions with cryo in proximity to the esophagus. The use of a hybrid cryo-RF ablation approach in conjunction with barium esophageal opacification has only prolonged procedure time by a half-hour to 1 hour, with a mean procedure time of approximately four hours, but given us a greater level of peace of mind for the patient. Conclusion The creation of ablative transmural lesions close to vital structures surrounding the left atrium remains problematic. Concerns of injuring the esophagus limit our ability to safely deliver radiofrequency energy in the posterior left atrium, which can result in tissue recovery and incomplete PV isolation. Esophageal temperature monitoring and real-time imaging using barium contrast or computerized 3D mapping techniques may help in esophageal localization. These techniques permit modification of type, intensity and location of energy delivery as well as endoesophageal deflection, but their efficacy and safety in preventing esophageal injury have not been established. During our cryo-RF hybrid cases, we typically complete a full three- to four-minute cryo application, even in proximity to the esophagus. We have not observed any evident esophageal injury during the follow-up in these 60 patients. Our initial willingness to use this technique is based on animal experiments involving the delivery of cryo-energy directly to the esophagus that show no tissue disruption. Our continued use of cryo-energy delivery in close proximity to the esophagus is based on the safety profile of cryoablation in clinical practice. Acknowledgements I would like to acknowledge my lab staff: Lyn Hamilton, RCVT, Steffanie Dunbar, RCIS - Testamur NASPExAM, Denise Hardy, RCIS, and John Makal.