PO-02-191 LEFT ATRIAL STRAIN CAN PREDICT THE OCCURRENCE OF ATRIAL FIBRILLATION: A SYSTEMATIC REVIEW AND META-ANALYSIS

Abstract

Despite the broadening spectrum of heart rhythm monitoring techniques, the most appropriate modality and duration to detect atrial fibrillation (AF) are not consensual. 2D speckle tracking for left atrial (LA) strain analysis seems feasible to detect atrial cardiomyopathy (ACM), which represents a substrate for AF. We aimed to perform a systematic review and meta-analysis on global longitudinal atrial strain (GLAS) and contraction strain rate (Sra) differences as primary outcomes measured in sinus rhythm (SR) between patients with and without future AF during follow-up (FU). The random effects model was used. For continuous outcomes, we calculated weighted mean differences (WMD) to compare the two groups. We identified eight eligible articles consisting of 1,230 patients. Six articles were eligible for quantitative GLAS analysis, while three were eligible regarding Sra. The baseline parameters of future AF and SR patients were comparable in most of the papers. The GLAS value in patients developing future AF was significantly lower compared to patients with no AF (WMD: -5.55% 95% CI -7.06 to -4.03 %). Pooled WMD of Sra between the groups was -0.44 1/sec, 95% CI: -0.56 to -0.33 1/sec. Baseline GLAS and Sra values of future AF patients were significantly lower than those of patients who remained in SR during FU. GLAS and Sra seem to be sensitive parameters, which can be implemented in a predictive model for AF enabling us to find patients who need prolonged heart rhythm monitoring, and it can serve also as guidance for anticoagulation indication setting.