On the accuracy of CartoMerge for guiding posterior left atrial ablation in man

• Published in: Heart rhythm Vol. 4; no. 5; p. 595
• Main Authors: Zhong, Hua, Lacomis, Joan M, Schwartzman, David
• Format: Journal Article
• Language: English
• Published: United States 01.05.2007
• Subjects: Aged; Analysis of Variance; Aorta, Thoracic - diagnostic imaging; Artificial Intelligence; Atrial Fibrillation - diagnostic imaging; Atrial Fibrillation - physiopathology; Atrial Fibrillation - surgery; Catheter Ablation; Electrophysiologic Techniques, Cardiac - methods; Female; Heart Atria - diagnostic imaging; Heart Atria - physiopathology; Heart Atria - surgery; Heart Conduction System - diagnostic imaging; Heart Conduction System - physiopathology; Humans; Image Processing, Computer-Assisted; Intraoperative Period; Male; Middle Aged; Pulmonary Veins - diagnostic imaging; Research Design; Tomography, X-Ray Computed - methods; Treatment Outcome
• ISSN: 1547-5271
• DOI: 10.1016/j.hrthm.2007.01.033

Recent reports suggest that the CartoMerge system is useful for guiding human posterior left atrial (PLA) endocardial ablation. To assess the accuracy of the CartoMerge system during PLA ablation. Sixteen patients undergoing PLA catheter ablation were studied. In each patient, registration of preoperative computed tomographic (CT) and intraoperative electroanatomic left atrial images was performed to create CartoMerge images. Encircling of right and left pulmonary venous vestibules with ablation points was then performed guided solely by intracardiac echocardiography, with point locations saved on a CartoMerge image to which the operator was blinded. The accuracy of the CartoMerge image was then assessed by measuring the distance from the location of each ablation point on the image to its actual anatomic location. In five patients, accuracy of registration of each of three left atrial CT images (just prior to mitral valve opening, at end-diastasis, at end-atrial contraction) with the electroanatomic image was compared. In two patients, accuracy of registration using left atrial image data alone was compared with that which used both left atrial and thoracic aorta image data. In each patient, inaccuracy of the CartoMerge image was apparent, the magnitude of which was similar for right- and left-vestibule ablation points. Accuracy was significantly improved when the end-atrial contraction CT image was used for registration. The inclusion of thoracic aorta image data did not improve accuracy. The CartoMerge system is inaccurate. Inaccuracy may be reduced by using CT and electroanatomic images obtained at the same point in the atrial mechanical cycle.