BACKGROUND: Roof-dependent atrial tachycardia (roof AT) sometimes occurs after pulmonary vein isolation (PVI) of atrial fibrillation (AF). This study aimed to investigate the relationship between the anatomy of the residual left atrial posterior wall and occurrence of roof AT. METHODS: A total of 265 patients with AF who underwent PVI were enrolled. After the PVI, induced or recurrent roof AT was confirmed by an entrainment maneuver or activation mapping using a three-dimensional (3D) mapping system. To identify the predictors of roof AT, the minimum distance between both PVI lines (d-PVI) was measured by a 3D mapping system and the anatomical parameters, including the left atrial (LA) diameter, left atrial volume index (LAVi), and shape of the left atrial roof, were analyzed by 3D computed tomography. RESULTS: Roof AT was documented in 11 (4.2%) of 265 patients. A multivariable analysis demonstrated that the d-PVI, Deep V shape of the LA roof, and LAVi were associated with roof AT occurrences (d-PVI: odds ratio: 0.72, confidence interval [CI]: 0.61-0.86, P < 0.001; Deep V shape: odds ratio: 0.19, CI: 0.04-0.82, P = 0.03; LAVi: odds ratio: 1.05, CI: 1.02-1.07, P = 0.001). A receiver-operating characteristic curve analysis yielded an optimal cut-off value of 15.5 mm and 55.7 mL/m2 for the d-PVI and LAVi, respectively. CONCLUSION: The shorter d-PVI at the LA roof, greater LAVi, and Deep V shape were associated with the occurrence of a roof AT.
BACKGROUND: Roof-dependent atrial tachycardia (roof AT) sometimes occurs after pulmonary vein isolation (PVI) of atrial fibrillation (AF). This study aimed to investigate the relationship between the anatomy of the residual left atrial posterior wall and occurrence of roof AT. METHODS: A total of 265 patients with AF who underwent PVI were enrolled. After the PVI, induced or recurrent roof AT was confirmed by an entrainment maneuver or activation mapping using a three-dimensional (3D) mapping system. To identify the predictors of roof AT, the minimum distance between both PVI lines (d-PVI) was measured by a 3D mapping system and the anatomical parameters, including the left atrial (LA) diameter, left atrial volume index (LAVi), and shape of the left atrial roof, were analyzed by 3D computed tomography. RESULTS: Roof AT was documented in 11 (4.2%) of 265 patients. A multivariable analysis demonstrated that the d-PVI, Deep V shape of the LA roof, and LAVi were associated with roof AT occurrences (d-PVI: odds ratio: 0.72, confidence interval [CI]: 0.61-0.86, P < 0.001; Deep V shape: odds ratio: 0.19, CI: 0.04-0.82, P = 0.03; LAVi: odds ratio: 1.05, CI: 1.02-1.07, P = 0.001). A receiver-operating characteristic curve analysis yielded an optimal cut-off value of 15.5 mm and 55.7 mL/m2 for the d-PVI and LAVi, respectively. CONCLUSION: The shorter d-PVI at the LA roof, greater LAVi, and Deep V shape were associated with the occurrence of a roof AT.