Yusei Nishikawa, Hiroyuki Takaoka, Tomonori Kanaeda, Haruhiro Takahira, Sakuramaru Suzuki, Shuhei Aoki, Hiroki Goto, Katsuya Suzuki, Satomi Yashima, Manami Takahashi, Makiko Kinoshita, Haruka Sasaki, Noriko Suzuki-Eguchi, Koichi Sano, Yoshio Kobayashi
Heart and vessels 38(5) 721-730 2022年12月19日
Recently, myocardial extracellular volume (ECV) analysis has been measurable on computed tomography (CT) using new software. We evaluated the use of cardiac CT to estimate the myocardial ECV of left ventricular (LV) myocardium (LVM) to predict reverse remodeling (RR) in cases of atrial fibrillation (AF) after catheter ablation (CA). Four hundred and seven patients underwent CA for AF in our institution from April 2014 to Feb 2021. Of these, 33 patients (8%) with an LVEF ≤ 50% and who had undergone CT were included in our study. We estimated the LVM ECV using commercial software to analyze the CT data. RR was defined as an improvement in LVEF to > 50% after CA. LVEF increased to > 50% in 24 patients (73%) after CA. In all 24 patients, LVM ECV, LV end-diastolic and end-systolic volumes (LVEDV and LVESV), and the n-terminal fragment of pro-B-type natriuretic peptide (NT-proBNP) were significantly lower than in the other nine patients (P = 0.0037, 0.0273, 0.0443, and < 0.0001). On receiver operating characteristic curve analysis, the best cut-off of ECV, LVEDV, LVESV and NT-proBNP for the prediction of RR were 37.73%, 120 mL, 82 mL, and 1267 pg/mL, respectively. We newly defined the ENL (ECV, NT-proBNP, and LVEDV) score as the summed score for the presence or absence (1 or 0; maximum score = 3) of ECV, NT-proBNP, and LVEDV values less than or equal to each best cut-off value, and found that this score gave the highest area under the curve for the prediction of RR (0.9583, P < 0.0001). The ENL score may be useful for predicting RR in patients with AF undergoing CA.