BACKGROUND: CartoSound™ (CS) module is useful in integrating 3-dimensional (3D) left atrial (LA) image with CT image. Integration method, however, has not been established. We reported the accuracy of LA electroanatomical (EA) and CT image integration by registering LA roof (LAR) and posterior wall (LAPW). METHODS: The consecutive 56 atrial fibrillation patients undergoing pulmonary vein isolation were studied. In the initial 29 patients, before the transseptal puncture, 3D CS LAR and LAPW image was created by registering a mean of 10 contour lines between the right and left pulmonary veins. After transseptally inserting a mapping catheter into LA, 3D EA image of LAR and LAPW was obtained by sampling a mean of 40 points. LA CT image was taken at the full-inspiratory position and 0% of R-R interval. After visual alignment of CS or EA and LA CT image, the 2 images were integrated with surface registration program. In the latter 27 patients, both CT and CS images were taken while matching the respiratory phase at the end-tidal position and cardiac cycle at 50% of R-R interval. RESULTS: In the initial 29 patients, the mean distances between EA and CT images and between CS and CT images were 1.53 ± 0.27 and 1.59 ± 0.23 mm, respectively (P = NS). In the latter 27, the mean distance was decreased to 1.08 ± 0.14 mm (P < 0.0001). CONCLUSIONS: CS system is useful in image integration with 3D CT. Matching both respiratory phase and cardiac cycle between CS and CT image acquisition improves the image integration accuracy.
BACKGROUND: CartoSound™ (CS) module is useful in integrating 3-dimensional (3D) left atrial (LA) image with CT image. Integration method, however, has not been established. We reported the accuracy of LA electroanatomical (EA) and CT image integration by registering LA roof (LAR) and posterior wall (LAPW). METHODS: The consecutive 56 atrial fibrillationpatients undergoing pulmonary vein isolation were studied. In the initial 29 patients, before the transseptal puncture, 3D CS LAR and LAPW image was created by registering a mean of 10 contour lines between the right and left pulmonary veins. After transseptally inserting a mapping catheter into LA, 3D EA image of LAR and LAPW was obtained by sampling a mean of 40 points. LA CT image was taken at the full-inspiratory position and 0% of R-R interval. After visual alignment of CS or EA and LA CT image, the 2 images were integrated with surface registration program. In the latter 27 patients, both CT and CS images were taken while matching the respiratory phase at the end-tidal position and cardiac cycle at 50% of R-R interval. RESULTS: In the initial 29 patients, the mean distances between EA and CT images and between CS and CT images were 1.53 ± 0.27 and 1.59 ± 0.23 mm, respectively (P = NS). In the latter 27, the mean distance was decreased to 1.08 ± 0.14 mm (P < 0.0001). CONCLUSIONS:CS system is useful in image integration with 3D CT. Matching both respiratory phase and cardiac cycle between CS and CT image acquisition improves the image integration accuracy.