Atrial fibrillation: paroxysmal or persistent: no antiarrhythmic agents is clearly better than another for maintaining sinus rhythm

Clinical bottom line (level 1a)

  1. Patients with paroxysmal or persistent atrial fibrillation but currently in sinus rhythm who take quinidine compared with propafenone are more likely to remain in sinus rhythm (NN T = 5 at months) .
  2. Patients who take amiodarone compared with disopyramide are more likely to remain in sinus rhythm (NN T = 2 at months) .
  3. No other anti-arrhythmic is clearly better than another for maintaining sinus rhythm.
Miller et al: Journal of Family Practice 2000; 49 : 1033-1046
McNamara et al: AHRQ Evidence Report 2001; 12 : -
Expires November 2003

The study

Systematic review of all randomised controlled trials of
  • Patients: previous episodes of paroxsymal or persistent (lasting > 48 hours and not terminating spontaneously) atrial fibrillation and currently in sinus rhythm
  • Intervention: anti-arrhythmics: quinidine, procainamide, disopyramide, flecainide, propafenone, amiodarone, sotalol, ibutilide/dofetilide, beta-blockers, calcium-channel blockers (verapamil, diltiazem), digoxin compared with another anti-arrhythmic agent
  • Outcome: maintenance of sinus rhythm

    Articles found in all languages - though only English ones were included using CENTRAL (consisting of Medline, Embase and hand-searching of 100 journals), 1948 to May 1998 (search terms: detailed in text ) and searching Pubmed for related articles, Cardiovascular Randomized Controlled Trial Registery. Reference lists of relevant meta-analyses, recent review articles and major clinical trials were searched. Investigators in the field and search co-ordinators of relevant Cochrane Review groups were contacted to identify unpublished studies. Abstracts of major cardiology conference meetings for 1997-1998 were also checked and recent contents of journals frequently cited in the search results database.

    Selection criteria: selected by 3 clinicians
    Appraisal criteria: detailed in text - included study population, bias, confounding, outcomes, follow-up, statistical quality: by 2 independent reviewers
    Articles excluded if:
    • article did not address management of AF or atrial flutter
    • no human data
    • included post-operative AF data that could not be separated out
    • adults not part of study population
    • no original data in study
    • no randomisation
    • unable to separate out AF or flutter from other arrhythmias


    36 studies found comparing anti-arrhythmic agents and placebo (25 addressing conversion to sinus rhythm with follow-up ~ 24 hours, and 15 addressing maintenance of sinus rhythm with follow-up of 1-15 months); and 16 studies comparing 2 anti-arrhythmic agents
    Where studies were found to be heterogeneous, pooled estimates were derived using a random-effects model (propafenone v. control; amiodarone v. control)

    The evidence

    maintenance of sinus rhythm
    Outcome Time to outcome CER OR
    (95% CI)    		 NN T
    (95% CI)
    quinidine v. flecainide months /
    (30%)    		 0.7
    (0.4 to 1.2)     		-14
    (NNT = 7 to infinity;
    NNH = 25 to infinity)
    quinidine v. propafenone months /
    (30%)    		 0.3
    (0.1 to 0.7)     		5
    (4 to 14)
    quinidine v. sotalol months /
    (30%)    		 0.9
    (0.5 to 1.5)     		46
    (NNT = 8 to infinity;
    NNH = 11 to infinity)
    quinidine v. amiodarone months /
    (30%)    		 0.9
    (0.1 to 16.5)     		46
    (NNT = 4 to infinity;
    NNH = 2 to infinity)
    amiodarone v. disopyramide months /
    (30%)    		 3.2
    (1.0 to 9.6)     		4
    (2 to infinity)
    disopyramide v. propafenone months /
    (30%)    		 1.8
    (0.6 to 5.1)     		7
    (NNT = 3 to infinity;
    NNH = 10 to infinity)
    flecainide v. propafenone months /
    (30%)    		 0.9
    (0.4 to 2.2)     		46
    (NNT = 7 to infinity;
    NNH = 5 to infinity)
    propafenone v. sotalol months /
    (30%)    		 0.7
    (0.4 to 1.1)     		14
    (NNT = 7 to infinity;
    NNH = 49 to infinity)
    long-acting v. short-acting quinidine months /
    (30%)    		 3.5
    (0.9 to 13)     		3
    (NNT = 2 to infinity;
    NNH = 46 to infinity)
    flecainide v. cibenzoline months /
    (30%)    		 1.4
    (0.5 to 4.0)     		13
    (NNT = 2 to infinity;
    NNH = 8 to infinity)

    Comments

    1. 12 non-English language articles were found, but full articles were not retrieved or combined in the meta-analysis.
    2. Most studies were small (< 100 patients), and study populations and antiarrhythmic administration varied greatly.
    3. 11 studies included patients with atrial flutter
    4. No studies comparing amiodarone with control were found.
    5. The control rate for maintenance of sinus rhythm was assumed to be 30% (Rates ranged from 0% to 90% in studies).

    Citation

    1. Miller MR, McNamara RL, Segal JB, et al: efficacy of agents for pharmacological conversion of atrial fibrillation and subsequent maintenance of sinus rhythm: a meta-analysis of clinical trials. Journal of Family Practice 2000; 49 : 1033-1046
    2. McNamara RL, Bass EB, Miller MR, et al: Management of New Onset Atrial Fibrillation. Evidence Report/Technology Assessment No. 12 (prepared by the Johns Hopkins University Evidence-based Practice Center in Baltimore, MD, under Contract No. 290-97-0006). AHRQ Publication Number 01-E026. Rockville, MD: Agency for Healthcare Research and Quality.. AHRQ Evidence Report 2001; 12 : -
    Search Terms: from AHRQ website
    Contributor: Chris Ball, November 2001
    Reviewer:

    Clinical Question.
    Patient atrial fibrillation
    Intervention or Exposure anti-arrhythmic
    Comparison another antiarrhythmic
    Outcome maintenance of sinus rhythm
    		•