ABC | Volume 112, Nº4, April 2019

Original Article Ciuffo et al LA Remodeling and Dyssynchrony Arq Bras Cardiol. 2019; 112(4):441-450 Table 2 – Left atrial (LA) functional parameters by groups Paroxysmal AF (n = 102) Persistent AF (n = 44) p Mean 95% CI Mean 95% CI LA structure Minimum LA volume index, mm 3 /m 2 19.0 ± 7.8 18.5 – 21.4 23.0 ± 10.1 19.5 – 26.5 0.062 Maximum LA volume index, mm 3 /m 2 38.8 ± 10.5 36.8 – 40.8 39.6 ± 11.7 35.6 – 43.7 0.691 LA Function Total LAEF, % 49.5 ± 10.0 47.6 – 51.4 44.0 ± 12.6 39.6 – 48.3 0.008 Passive LAEF, % 22.9 ± 7.3 21.6 – 24.3 20.7 ± 8.3 17.8 – 23.5 0.128 Active LAEF, % 34.6 ± 10.8 32.5 – 36.6 29.5 ± 14.1 24.6 – 34.3 0.026 S max , % 28.9 ± 8.9 27.2 – 30.5 26.0 ± 11.8 22.0 – 30.1 0.132 SR 1.1 ± 0.4 1.1 – 1.2 1.1 ± 0.5 0.9 – 1.3 0.347 SR e -1.1 ± 0.5 -1.2 – -1.0 -0.8 ± 0.4 -1.0 – -0.7 0.010 SR a -1.4 ± 0.5 -1.5 – -1.3 -1.1 ± 0.6 -1.3 – -0.9 0.011 LVEF, % 58.4 ± 6.0 57.0 – 59.8 53.4 ± 10.3 49.4 – 57.4 0.004 Median IQR Median IQR p Dyssynchrony Mean TPS, ms 397.8 374.5 - 420.2 403.5 369.9 - 429.0 0.538 SD-TPS, % 2.9 2.1 – 3.9 3.6 2.3 – 4.9 0.036 Log - SD-TPS, % 1.0 0.7 – 1.4 1.1 0.8 – 1.6 0.036 Mean SD-TPS preA , ms 795.3 692.4 - 884.9 846.7 760.6 - 967.4 0.046 SD-TPS preA, % 4.6 3.0 – 8.6 3.7 2.9 – 5.4 0.227 Log - SD-TPS preA , % 1.5 1.1 – 2.2 1.3 1.1 – 1.7 0.177 LGE extent (% LA surface) 11.6 6.0 – 17.6 13.8 7.6 – 28.4 0.061 Log LGE extent (% LA surface) 2.4 1.8 – 2.9 2.6 2.0 – 3.3 0.061 Data are presented as median (interquartile range [IQR]) or mean ± standard deviation (SD). CI: confidence interval; LAEF: LA emptying fraction; S max : maximum longitudinal LA strain; SR: peak longitudinal strain rate; SR e : early diastolic strain rate; SR a : late diastolic strain rate; LVEF: left ventricular ejection fraction; TPS: time to peak strain; TPS preA : time to peak pre-atrial contraction strain; LGE: late gadolinium enhancement. Table 3 – Univariable and multivariable analyses Model 1 Unadjusted Model 2 Clinical variables Model 3 Model 2 + Vmin + Smax β p β p β p Log SD-TPS, % 0.66 < 0.001 0.57 0.001 0.60 0.001 Log SD-TPS preA , % 0.19 0.034 0.21 0.020 0.18 0.045 Model 2, adjusted for age, sex, type of atrial fibrillation, body mass index, history of cardiac failure, hypertension, obstructive sleep apnea. Model 3, covariables included in Model 2 in addition to minimum left atrial volume and maximum longitudinal strain. V min : minimum left atrial volume; S max : maximum longitudinal strain; SD: standard deviation; TPS: time to peak strain; TPS preA : time to peak pre-atrial contraction strain. obstructive sleep apnea, hypertension, V min , and S max , both indices SD-TPS and SD-TPS preA remained significantly associated with LA-LGE (SD-TPS, β : 0.60, p = 0.001; SD‑TPS preA , β : 0.18, p = 0.045) (Table 3). Figure 4 displays the relationship between LA-LGE and LA intra-atrial dyssynchrony. Therewas no significant multiplicative interaction between AF type and LA intra-atrial dyssynchrony (interaction term for SD‑TPS: 0.008, p = 0.258 and SD-TPS preA : 0.003, p = 0.158). The LA-LGE analysis was performed in a consistent manner in all cases and took 60 ± 20 minutes per case, also depending on the image quality. Dyssynchrony: inter-reader and intra-reader reproducibility Interobserver and intraobserver variabilities of LA analysis for the MTT method were assessed in 15 randomly select subjects (Table 4, Figure 5). All parameters showed excellent intraobserver reproducibility (ICC 0.86 and 0.85 for SD-TPS and SD-TPS preA respectively, p < 0.001) (Figure 5) without significant systematic bias. In addition, both parameters showed good to excellent interobserver reproducibility (ICC 0.86 and 0.74 for SD-TPS and SD‑TPS preA , respectively, p < 0.001) (Figure 5). 445