Left atrial strain parameters to predicting elevated left atrial pressure in patients with atrial fibrillation.

Objectives: To assess the ability of left atrial (LA) strain parameters to discriminate patients with elevated left atrial pressure (LAP) from patients with atrial fibrillation (AF).

Methods And Results: A total of 142 patients with non-valvular AF who underwent first catheter ablation (CA) between November 2022 and November 2023 were enrolled in the study. Conventional and speckle-tracking echocardiography (STE) were performed in all patients within 24 h before CA, and LAP was invasively measured during the ablation procedure. According to mean LAP, the study population was classified into two groups of normal LAP (LAP < 15 mmHg, n = 101) and elevated LAP (LAP ≥ 15 mmHg, n = 41). Compared with the normal LAP group, elevated LAP group showed significantly reduced LA reservoir strain (LASr) [9.14 (7.97-11.80) vs. 20 (13.59-26.96), p < .001], and increased LA filling index [9.60 (7.15-12.20) vs. 3.72 (2.17-5.82), p < .001], LA stiffness index [1.13 (.82-1.46) vs. .47 (.30-.70), p < .001]. LASr, LA filling index and LA stiffness index were independent predictors of elevated LAP after adjusted by the type of AF, EDT, E/e', mitral E, and peak acceleration rate of mitral E velocity. The receiver-operating characteristic curve (ROC) analysis showed LA strain parameters (area under curve [AUC] .794-.819) could provide similar or greater diagnostic accuracy for elevated LAP, as compared to conventional echocardiographic parameters. Furthermore, the novel algorithms built by LASr, LA stiffness index, LA filling index, and left atrial emptying fraction (LAEF), was used to discriminate elevated LAP in AF with good accuracy (AUC .880, accuracy of 81.69%, sensitivity of 80.49%, and specificity of 82.18%), and much better than 2016 ASE/EACVI algorithms in AF.

Conclusion: In patients with AF, LA strain parameters could be useful to predict elevated LAP and non-inferior to conventional echocardiographic parameters. Besides, the novel algorithm built by LA strain parameters combined with conventional parameters would improve the diagnostic efficiency.