Ventricular Changes Seem Worse in Persistent vs Paroxysmal AF

Key Takeaways

  • The authors found that patients with persistent atrial fibrillation (AF), compared with those with paroxysmal AF, had worse systolic circumferential left ventricle (LV) mechanics and intraventricular dyssynchrony.

  • The findings were based on echocardiographic assessment of strain and strain rate derived from velocity vector imaging (VVI).

Why This Matters

  • Accurate cardiac mechanic assessments have been historically difficult using traditional imaging modalities.

  • Early-stage LV mechanical impairment had been found in patients with paroxysmal AF using VVI-derived strain and strain rate.

  • This study adds intraventricular mechanical dyssynchrony as to possible measures associated with worsening AF.


Study Design

  • The study compared 84 patients with nonvalvular AF (49 with paroxysmal AF, 35 with persistent AF) and 43 healthy control subjects.

  • Routine echocardiographic images were used to measure global longitudinal strain (GLS) and global circumferential strain (GCS).

  • Intraventricular dyssynchrony was measured using the standard deviation of all 12 segments.


Key Results

  • Participants with persistent AF, compared with controls, showed significantly lower GLS (-12.23% vs -18.71%, P < .001) and GCS (-18.46% vs -28.75%, P < .001), with heart rate emerging as an independent risk predictor.

  • Similarly, participants with paroxysmal AF, compared with controls, showed significantly lower GLS (-17.10% vs -18.71%, P < .05) and GCS (-24.43% vs -28.75%, P < .01).

  • GLS and GCS were lower in patients with persistent AF than those with paroxysmal AF (P < .001 for both parameters).

  • Intraventricular dyssynchrony was seen in both persistent AF and paroxysmal AF groups but was more severe in those with persistent AF (P < .05).

  • Left ventricular ejection fraction did not differ significantly among the three groups.

  • Age emerged as an independent risk predictor for LV dyssynchrony.


  • The study is small and requires corroboration in larger studies.

  • Pixel-tracking VVI requires above-average image quality for optimal data extraction.

  • Actual VVI values may be altered by speckle pattern throughout the cardiac cycle and differences between regional resolutions.