Decreased left atrial global longitudinal strain measured in sinus rhythm can help predict atrial fibrillation: a meta-analysis

Abstract Background Despite the broadening spectrum of heart rhythm monitoring techniques, there is no wide consensus on the most appropriate modality and duration to detect atrial fibrillation (AF). Two-dimensional (2D) speckle tracking for left atrial (LA) strain analysis seems feasible to detect atrial cardiomyopathy, which might represent a substrate for AF. Purpose We aimed to perform a meta-analysis on the differences in baseline global longitudinal left atrial strain (GLAS) values as a primary outcome, measured exclusively in sinus rhythm (SR) between patients who developed AF in the future and patients who remained in SR during follow-up (FU). Methods Random-effect model was used. We calculated weighted mean differences (WMD) of baseline GLAS values of future AF patients and of those remaining in SR. A risk for bias was also assessed of the eligible studies. Results 6 articles were eligible for quantitative GLAS analysis in 1106 patients. Baseline parameters of the future AF and SR patients were comparable in most of the papers. Pooled WMD value of GLAS between the future AF and SR group was -5.55 %, with a 95% confidence interval (CI) of -7.06 to -4.03 %, and a prediction interval of -8.02 to - 3.07%. On assessment, 33% of the studies had a moderate risk for bias. Conclusion Baseline GLAS values of future AF patients were significantly lower than that of patients remained in SR during FU. GLAS is a quite sensitive parameter derived from simple, readily-available apical 2D echo images, which can be incorporated in a predictive clinical model of AF. GLAS could have a potential to separate those patients in SR who need prolonged rhythm monitoring from those who need anticoagulation even without AF detected at baseline.