Usefulness of left atrial speckle tracking echocardiography in predicting recurrence of atrial fibrillation after radiofrequency ablation: a systematic review and meta-analysis

Anne Bjerg Nielsen,Kristoffer Grundtvig Skaarup,Mats Christian Højbjerg Lassen,Kasper Djernæs,Morten Lock Hansen,Jesper Hastrup Svendsen,Arne Johannessen,Jim Hansen,Samuel Kiil Sørensen,Gunnar Gislason,Tor Biering-Sørensen
DOI: https://doi.org/10.1007/s10554-020-01828-2
2020-04-04
The International Journal of Cardiovascular Imaging
Abstract:The usefulness of peak atrial longitudinal strain (PALS) in identifying patients at high risk of atrial fibrillation (AF) recurrence after radiofrequency ablation (RFA) has been investigated in several small AF populations. The aim of this systematic review and meta-analysis was to investigate whether PALS predicts recurrence of AF after RFA treatment. MEDLINE, EMBASE and the Cochrane Library were searched. Studies investigating the value of PALS in predicting successful RFA in AF patients were selected. Patients underwent echocardiography prior to RFA. Risk of bias was evaluated using the Quality in Prognosis Studies (QUIPS) Tool. Twelve studies and a total of 1025 patients suffering from paroxysmal or persistent AF were included. Odds ratios (OR) were assessed in a random and fixed effects model for univariable and multivariable pooled analyses respectively. PALS was found to be a significant predictor of AF recurrence after RFA across study populations (Univariable: OR: 1.17, CI95% [1.03–1.34], p = 0.018, per 1% decrease) (Multivariable: OR: 1.16, CI95% [1.09–1.24], p < 0.001, per 1% decrease). Patients with recurrence had significantly lower PALS prior to RFA than patients who maintained sinus rhythm (15.7 ± 5.7% vs. 23.0 ± 7.0%, p = 0.016). A pooled analysis of weighted mean differences (WMD) also showed a significant difference in PALS between the two groups (WMD: − 6.57, CI95% [− 8.49: − 4.65], p < 0.001). Lower values of PALS are associated with an increased risk of AF recurrence after RFA. PALS provides prognostic value in clinical practice.
What problem does this paper attempt to address?