Comparative study of registration accuracy between 64-slice spiral CT and contrast enhanced MR in left atrium imaging for integrated cartography system

Abstract

Objective To evaluate the registration accuracy and the efficacy of integrated cartography (CARTO) system with 64 muhislice spiral CT scan and contrast enhanced MRI for the assessment of left atrium (LA) geometry and to compare the registration accuracy between CT and MRI. Methods Forty consecutive patients with drug-refractory atrial fibrillatin (AF) underwent catheter ablation under the guidance of a three-dimensional (3 D) CARTO system (Biosence Webster, Inc. , Diamond Bar, CA, USA). Gadolinium-enhanced MRI (1.5 T,n=20) and contrast-enhanced high-resolution CT(64-slice,n=20) imaging were performed within 24 h prior to the ablation procedures. The DICOM data acquired by CT/ MRA were input into CARTO system and the LA images were extracted for CARTO image registration. Clinical AF ablation was quided by LA geometry. The registrated values of general samples, CT group, MR group were analyzed and comparison was made between CT and MR groups with the mean deviation of registrated values. Results The segmented 3D MRI and CT LA reconstructions were accurately registerted to the real-time mapping space with a combination of landmark registration and surface registration. The registered 3D CT and MR LA reconstruction were successfully used to guide deployment of RF applications encircling the pulmonary veins (PVs). The distance between the surface of the registered LA reconstruction and multiple electroanatomic map points with an error of (2. 3±0. 5) rnm in general group, (2. 2±0.4) mmin CT group, and (2.5±0.6) mm in MR group respectively. The mean deviation of the registration were (2.2±0. 4) mm in CTA group and (2. 4±0. 5) mm in MRA group respectively. There was no significant difference in the LA registration error between the two groups (t=1.37, P=0. 081).Conclusions Both three-dimensional CT and MR images integrated into an CARTO system can be successfully used to perform catheter ablation for AF and there was no difference in registration accuracy between the two groups. Key words: Atrial fibrillation; Catheter ablation; Tomography; X-ray computed; Magnetic resonance imaging; Electrophysiologic techniques; cardiac