Rate- and Rhythm-Control Therapies in Patients With Atrial Fibrillation A Systematic Review

Background: The comparative effectiveness of treatments for atrial fibrillation (AF) is uncertain. Purpose: To evaluate the comparative effectiveness of rate-and rhythm-control therapies. Data Sources: English-language studies in PubMed, EMBASE, and the Cochrane Database of Systematic Reviews between January 2000 and November 2013. Study Selection: Two reviewers independently screened citations to identify comparative studies that assessed rate-or rhythm-control therapies in patients with AF. Data Extraction: Reviewers extracted data on study design, participant characteristics, interventions, outcomes, applicability, and quality. Data Synthesis: 200 articles (162 studies) involving 28 836 patients were included. When pharmacologic rate-and rhythm-control strategies were compared, strength of evidence (SOE) was moderate supporting comparable efficacy with regard to all-cause mortality (odds ratio [OR], 1.34 [95% CI, 0.89 to 2.02]), cardiac mortality (OR, 0.96 [CI, 0.77 to 1.20]), and stroke (OR, 0.99 [CI, 0.76 to 1.30]) in older patients with mild AF symptoms. Few studies compared rate-control therapies and included outcomes of interest, which limited conclusions. For the effect of rhythm-control therapies in reducing AF recurrence, SOE was high favoring pulmonary vein isolation versus antiarrhythmic medications (OR, 5.87 [CI, 3.18 to 10.85]) and the surgical maze procedure (including pulmonary vein isolation) done during other cardiac surgery versus other cardiac surgery alone (OR, 7.94 [CI, 3.63 to 17.36]). Limitation: Studies were heterogeneous in interventions, populations, settings, and outcomes. Conclusion: Pharmacologic rate-and rhythm-control strategies have comparable efficacy across outcomes in primarily older patients with mild AF symptoms. Pulmonary vein isolation is better than antiarrhythmic medications at reducing recurrences of AF in younger patients with paroxysmal AF and mild structural heart disease. Future research should address uncertainties related to subgroups of interest and the effect of different therapies on long-term clinical outcomes.