Thromboembolic risk and effect of oral anticoagulation according to atrial fibrillation patterns: A systematic review and meta-analysis.

Oral anticoagulation (OAC) is recommended in both paroxysmal atrial fibrillation (pxAF) and nonparoxysmal AF (non-pxAF), but disagreement exists in classes of recommendation. Data on incidence/rate of stroke in pxAF are conflicting, and OAC is often underused in this population. The objectives of the meta-analysis were to investigate different impact on outcomes of pxAF and non-pxAF, with and without OAC. Two reviewers searched for prospective studies on risk of stroke and systemic embolism (SE) in pxAF and non-pxAF, with and without OAC. Quality of evidence was assessed according to GRADE approach. Stroke combined with SE was the main outcome. Meta-regression was performed to evaluate OAC effect on stroke and SE incidence rate. We identified 18 studies. For a total of 239 528 patient-years of follow-up. The incidence rate of stroke/SE was 1.6% (95% confidence interval [CI]: 1.3%-2.0%) in pxAF and 2.3% (95% CI: 2.0%-2.7%) in non-pxAF. Paroxysmal AF was associated with a lower risk of overall thromboembolic (TE) events (risk ratio: 0.72, 95% CI: 0.65-0.80, P < 0.00001) compared with non-pxAF. In both groups, the annual rate of TE events decreased as proportion of patients treated with OAC increased. Non-pxAF showed a reduction from 3.7% to 1.7% and pxAF from 2.5% to 1.2%. Major bleeding rates did not differ among groups. Stroke/SE risk is significantly lower, although clinically meaningful, in pxAF. OAC consistently reduces TE event rates across any AF pattern. As a whole, these data provide the evidence to warrant OAC irrespective of the AF pattern in most (virtually all) patients.