Dronedarone and the incidence of stroke in patients with paroxysmal or persistent atrial fibrillation: a systematic review and meta-analysis of randomized trials.

BACKGROUND: Stroke is the most feared complication of atrial fibrillation (AF). Dronedarone is an antiarrhythmic drug with multichannel-blocking properties. Recently, a post hoc analysis of a large randomized trial has suggested a reduction of stroke risk in patients with paroxysmal or persistent AF receiving dronedarone.

OBJECTIVE: We performed a systematic review and meta-analysis on the effect of dronedarone on the occurrence of stroke or transient ischemic attack (TIA) in patients with paroxysmal or persistent AF.

METHODS: We searched PubMed, EMBASE, and ISI Web of Knowledge as well as abstracts of major conferences for randomized trials comparing dronedarone with placebo in patients with paroxysmal or persistent AF. The endpoint was the occurrence of stroke or TIA during follow-up. Fixed effect risk differences (RDs) were calculated with the Mantel-Haenszel method. We also performed random effects analysis with the DerSimonian Laird method.

RESULTS: Four trials were included in the analysis; a total of 5967 patients were analyzed, 3183 receiving dronedarone 400 mg twice daily and 2784 receiving placebo. 160 strokes or TIAs were reported in the four trials: 67 in the dronedarone group (2.1%) and 93 in the placebo group (3.3%). In the fixed effect model, patients in the dronedarone group had a significantly lower risk for the occurrence of stroke or TIA during follow-up compared with patients in the placebo group. The RD of the incidence of stroke or TIA in all trials between patients randomized to dronedarone and those randomized to placebo was -0.0094 (95% confidence interval [CI] -0.0178, -0.0011; p = 0.027). The ATHENA trial had by far the highest statistical weight (79.5%). There was no evidence of heterogeneity (χ2 = 2.41, p = 0.300). In the random effects model, the statistical weight of the ATHENA trial was much lower (45.1%) and the RD for stroke or TIA between the dronedarone and the placebo groups did not reach statistical significance (RD -0.0064, 95% CI -0.0144, 0.0016; p = 0.120).

LIMITATIONS: First, stroke was not a prespecified outcome measure in the included trials. Second, we did not analyze trials studying patients with permanent AF; very recent data show an adverse outcome in patients with permanent AF receiving dronedarone.

CONCLUSIONS: The meta-analysis indicates a reduced risk of stroke or TIA in patients with paroxysmal or persistent AF receiving dronedarone. These findings are largely due to the results of the ATHENA trial. Further research on this topic is necessary.