Assessment of left atrial systolic dyssynchrony in paroxysmal atrial fibrillation and heart failure using cardiac magnetic resonance imaging: MESA study

Methods We performed a nested case-control analysis with 1:2 matching for 39 cases of paroxysmal AF (n=28, in sinus rhythm during cardiac magnetic resonance (CMR)) and HF (n=14, AF+HF; n=3) and 78 controls with similar demographic and clinical characteristics at the baseline (Table 1). LA circumferential (short axis) and longitudinal strain rate (horizontal long axis) were measured using Multi-modality Tissue Tracking (Toshiba, Japan) from short and long-axis cine CMR images. Circumferential LA systolic dyssynchrony among 18 LA segments (6 segments x 3 slices) was evaluated as; Standard Deviation (SD) of time to pre atrial contraction Strain rate (PreA Sr) and Peak systolic strain rate (Peak Sra) (Figure 1). Similarly, longitudinal LA dyssynchrony parameters (among 6 segments) were: SD-Time to pre-atrial contraction strain rate (PreA Sr) and SD-Time to peak systolic strain rate (PeakSra). Wilcoxon-rank sum test (non-parametric) or two sample t-test (parametric) were used for comparison between the groups.