Weekly physical activity and incident atrial fibrillation in females - A dose-response meta-analysis

For years, physical activity (PA) has been considered a mixed blessing in terms of the risk of incident atrial fibrillation (AF). Previous analyses have had equivocal results regarding the cut-off of PA level beyond which AF risk increases, if such a limit really does exist. Data regarding females in particular have been scarce.

We performed a dose-response meta-analysis to investigate the relationship between weekly PA and the risk for AF in females. Major electronic databases were searched for studies assessing the association between leisure time PA and the risk for incident AF in females from the general population. The linearity of the dose-response curve was assessed using the restricted cubic spline model.

A total of 15 studies, which involved 1,821,422 females, were included in the final analysis. AF incidence was 3.7%. Dose-response analysis revealed an inverse nonlinear relationship between weekly PA and the risk for incident AF (p for linearity <0.0001). No significant heterogeneity was documented (I² = 37%). Cautious interpretation is needed for PA exceeding 50 metabolic equivalents of task- hours per week (METs- h/w), due to limited available data for these high levels of PA.

According to this analysis, physicians can safely advise females to perform up to 50METs- h/w of moderate or vigorous PA, to reduce the risk for future AF. Interestingly, significant benefit can be attained even at low levels of regular weekly PA.