Outcomes with dronedarone in atrial fibrillation: What differences between real-world practice and trials? A meta-analysis and meta-regression analysis

Dronedarone was found to have divergent safety profiles in randomized controlled trials (RCT) in term of cardiac death and overall mortality. We decided to evaluate all available evidence on the cardiovascular safety of this drug. A systematic search was made of the MEDLINE and the Cochrane Central Register of Controlled Trials from January 2003 through April 2016 for RCT comparing dronedarone to placebo/active control, to provide the most accurate estimate of the effects of this agent and observational cohort studies (OBS) reporting clinical outcomes in patients treated with dronedarone, according to current guidelines, to obtain a real-life comparator for the findings summarized by RTC analysis. The literature search yielded 2335 papers and after careful review we identified 12 RCT and 7 OBS studies. RCT meta-analysis showed that, despite high heterogeneity, dronedarone was not associated with increased all-cause mortality [OR (Odds Ratio) 1.36, 95%CI (Confidence Interval) 0.79-2.33; p=0.732, I2=57.0%] or cardiovascular mortality [OR 1.51 95%CI 0.74-3.08; p=0.860, I2=64.4%]. OBS studies had a trend toward a better survival with respect to RCT [ES (Effect Size) 2.03, 95%CI 0.53-3.53 vs. ES 3.03, 95%CI 1.23-4.83; p=0.115], reaching the significance when restricted to the cardiovascular mortality [ES 0.52, 95%CI 0.36-0.69 vs. ES 1.86, 95%CI 0.62-3.09; p<0.001]. Two variables, co-adiministration of digoxin and prevalence of non-permanent AF completely abolished the dishomogeneity among the analyzed RCT studies. In conclusion, use of dronedarone for prophylaxis of AF recurrences is not associated with an increased risk of death, either cardiovascular or total, and combination with digoxin should be avoided.