The prognostic value of exercise-induced ventricular arrhythmias in patients with and without coronary artery disease: A meta-analysis

Abstract Background Significance of exercise-induced ventricular arrhythmias (EIVAs) is controversial. This meta-analysis aimed to determine the prognostic value of EIVAs in patients with and without coronary artery disease (CAD). Methods Relevant studies were searched on Pubmed though December, 2015. Pooled odds ratio (OR) of endpoints (all-cause death, cardiac death or cardiac events) for all included studies was calculated at first to explore the the significance of EVIAs in unselected population. Then, sensitivity analysis based on CAD status of population was performed to determine ORs of endpoints in CAD population, non-CAD population and mixed population, respectively. Results A total of 14 studies examining 23,002 patients were included, with 5 studies invloed CAD population, 4 invloed non-CAD population, and 5 involed mixed population (%CAD ranged from 51.2% to 76.8%). EIVAs in unselected population were associated with a pooled OR of 1.626 (95%CI 1.334 to 1.983, p < 0.001) of endpoints when compared with those without EIVAs. Sensitivity analysis further indicated that pooled ORs of endpoints were 1.395 (95%CI 1.061 to 1.833, p = 0.017) in CAD population, 1.933 (95%CI 1.567 to 2.384, p < 0.001) in non-CAD population, and 1.402 (95%CI 1.198 to 1.640, p < 0.001) in mixed population. Heterogeneous among studies was identified. Meta-regression analysis found that study quality, mean follow-up period, percentage of lost, percentage of diabetes were associated with ORs of endpoints. Conclusions EIVAs were associated with increase risk of worse outcomes, no matter the patients had CAD or not. However, more studies are required to confirmed this finding due to the variation of current evidences.