Meta Analysis:Oral Anticoagulants Versus Antiplatelet Therapy For The Primary Prevention Of Stroke In Patients With Non-valvular Atrial Fibrillation

Objective:To systematically evaluate the efficacy and safety of long-term oral anticoagulant treatment versus anti-platelet therapy for the primary prevention of stroke in patients with non-valvular atrial fibrillation.Search methods:We searched the Cochrane Central Register of Controlled Trials (November2011), EMBASE database (November2011) and China Biology Medicine disc,(CBM disc, November2011) systematically for the relative randomized controlled clinical trials both in English and Chinese.Included criteria:1. randomized controlled trials;2. participants are patients with non-valvular atrial fibrillation and without the history of stroke and TIA;3.long-term (more than4weeks) adjusted-dose oral anticoagulant treatment was compared with anti-platelet therapy for the effects and safety in primary prevention of stroke and systematic embolic events.Excluded criteria:1. non randomized controlled trials;2. participants are patients with concomitant mitral stenosis or prosthetic cardiac valves or have previous history of stroke/TIA;3. trials in which the anticoagulants were used in fixed-dose;4.trials in which the anticoagulants and anti-platelet agents were used in combination.Data collection:Data refers to the information of trial design; selection and exclusion criteria, intervention methods, efficacy and safety of the therapy were extracted independently by one author and checked over by another author. We use ITT results whenever possible.Statistical analysis:Eggers method was used to assess the bias among the trials. Funnel plots was also used to assess the bias, x2tests and the I2statistic derived from the x2values were used to test heterogeneity between the trials. The outcomes were statistically combined by use of the Yousef-Peto fixed-effects model for dichotomous clinical outcomes and adverse events, and by inverse variance methods for continuous or ordinal outcomes. Effects were expressed as odds ratios (ORs) and absolute risk differences with95%CIs and p values or as weighted mean differences (WMD) and standardized mean differences (SMD) by use of the Stata software.Results:Eleven randomized trials, including11350patients were included. Trials were of similar inclusion criteria and outcome measures. Retrieval bias tested among the trials was low (Egger’s method P=0.979, Begg’s method P=0.721). The results of primary combined outcomes yielded that oral anticoagulants were associated with lower risk of all stroke events (ischemic and hemorrhagic)(OR0.63,95%CI0.51-0.78). The combination of the secondary outcomes showed that the risk of stroke related mortality and disability rate can be reduced more effectively by oral anticoagulant treatment (OR0.69,95%CI0.54-0.89). Meanwhile, the ischemic stroke (OR0.47,95%CI0.37-0.59) and systematic embolism incidence (OR0.45,95%CI0.24-0.87) can also be reduced significantly by oral anticoagulant agents. However, there was no significant differences between the two groups in all cause mortality (OR0.97,95%CI0.84-1.13). In the aspect of safety, intracranial hemorrhage events occurred more frequently in anticoagulant group (OR1.69,95%CI1.07-2.69), but the difference of extra-cranial hemorrhage incidence was not statistically significant between the two groups (OR0.95,95%CI0.74-1.22).Conclusion:Compared with anti-platelet therapy, adjusted-dose oral anticoagulants treatment can reduce the risk of stroke as well as systematic emboli more effectively. Anticoagulant treatment was associated with a higher risk of intracranial hemorrhage, but the difference of the extra-cranial hemorrhage incidence was substantial between the two groups.