| Background: Atrial fibrillation(AF) is one of the most common arrhythmia in clinical practice. Amiodarone, a class III antiarrhythmic drug, is widely used for atrial fibrillation(AF) conversion, despite evidence about the role of amiodarone remains controversial in clinical practice. It is difficult to reach the exact conclusion on the effect of amiodarone in conversion of AF to sinus rhythm because that for most of the published randomized controlled trials(RCTs),the sample sizes are limited and designs are varied. Meanwhile, the safety of amiodarone needs to be further assessed. Heterogeneity of treatment effects between studies will influence the clinical conclusions of a systematic review. Meta-regression is a possible technique to explain the causes of heterogeneity.Objective: According to the principle of Cochrane systemic review, we conducted this review to appraise and summarize the results of RCTs on amiodarone for converting AF to sinus rhythm after 24h and 4 weeks observations. The purpose of the meta-analyses was to assess the effectiveness and safety of amiodarone on atrial fibrillation(AF) comprehensively and objectively to provide reliable and scientific evidence for guiding clinical practice and rational use of antiarrhythmic drugs. Meta-regression analysis was undertaken to explore the possible sources of heterogeneity and relationship between the effect and the characteristics of the studies.Methods: Two independent reviewers performed literature searches. Databases included Cochrane Controlled Trials Register( 2nd Quarter 2005),MEDLINE( 1966.1-2005.12), EMBASE( 1980-2005 week 41),WeiPu, WanFang, and CNKI, and manual searching as the supplement. Literatures of randomized controlled trials which compared amiodarone with nonactive drugs or class I antiarrhythmic drugs (propafenone,quinidine,flecainide, et al.) for conversion of AF were identified. Nonactive controls included placebo, digoxin, verapamil, or diltiazem. The primary outcome of interest was the rate of conversion over 24h and a 4-week period. We also collected information on adverse drug reactions in each relevant trials. For all of the selected trials that meet the inclusion criteria, assessment of methodological quality,data extraction and data syntheses analyses were conducted by two reviewers. The quality of each study was evaluated by examining four items: concealment of treatment allocation;method of randomization;masking of patients and caregivers;completeness of follow-up. Treatment effects were calculated as risk ratios(RRs) and their 95% confidence intervals (C.I.s), we used a random-effects model and a fixed-effects model to combine the data extracted from eligible studies. Statistical heterogeneity were identified by Cochrane's Q test. Meta-regression analyses were used to explore the possible explanations for heterogeneity and effect modifiers. Meta-analyses were repeated among high-quality studies to prove the sensitivity of the results. Funnel plots were used to examine the publication bias, and asymmetry was tested by Begg's rank correlation and Egger's weighted regression.Results:1. We conducted a global literature search in the electronic databases, complemented by hand-searching of references.,59 studies met inclusion criteria including 34 English language articles and 25 Chinese language articles.2. Assessment of study quality showed that only 9 English language studies of all included studies described the method of randomization, 5 studies implemented concealment allocation, 10.17 percent of the articles applied double-blinded method. In general, the report quality of English studies were better than that of Chinese language studies.3. 38 clinical trials compared amidarone with nonactive control group with a total of 3479 participants, 1852 in amiodarone group and 1627 in the control group. Both fixed-effects model and random-effects model were used to combine the results, the pooled RRs were 1.531(1.428, 1.641) and 1.660(1.456, 1.893) respectively, which indicated amiodarone was better than nonactive drugs for conversion of AF to sinus rhythm during a period of 4 weeks or less(.P=0.000).4. 26 studies compared amiodarone with class I antiarrhythmic drugs with a total of 1806 participants, 934 were allocated to amiodarone and 926 to class I antiarrhythmic drugs. The pooled RR was 1.039(95%C.I.,0.994~1.086) with fixed-effects model and 1.038(95%C.I.,0.974~1.106) with random-effects model, respectively. Both showed that the effect of amiodarone was similar compared with class I antiarrhythmic drugs.5. We found statistical heterogeneity with Cochrane's Q test among the conversion effects which compared amiodarone with nonactive control group over 4 weeks. The AF duration and the way of administration may be the factors affecting the calculated risk ratio of atrial fibrillation.6. Sensitivity analyses showed the results of this systematic review were not affected by study quality or statistical model. The funnel plots did not show evidence of publication bias, neither by statistical analyses with Begg's rank correlation and Egger's weighted regression.7. Compared with nonactive drugs, more hypotension, gastrointestinal side effects and phlebitis over the infusion site were found in the amiodarone group. There was no statistically significant difference between amiodarone and class I antiarrhythmic drugs among bradycardia and other supraventricular arrhythmia. Hypothyroidism and corneal disposition were more likely occurred in amiodarone group.It seemed very few serious side effects reported among amiodarone users.Conclusions: Amiodarone is more efficacious than nonactive control group in conversion of AF to sinus rhythm. More longer duration of AF, the patients will get more benefits treated by amiodarone. Amiodarone appears to be-similar compared with class I antiarrhythmic drugs in AF conversion, but the risk of adverse drug reactions is lower. These results showed that relatively the efficacy and safety of amiodarone are good, and it is favorable to be an alternative for patients with AF in whom class I antiarrhythmic drugs can not be used.
|
PDF Full Text Request