Surgical Maze procedure as a treatment for atrial fibrillation: a meta-analysis of randomized controlled trials.

Surgical or modified Maze procedures have been promoted to treat atrial fibrillation (AF); however, few randomized controlled clinical trials (RCTs) examine their outcomes. The purpose of this meta-analysis is to compare the efficacy of surgical Maze procedures performed concomitantly with referral cardiac surgery versus pharmacologic therapy for the treatment of AF. We searched MEDLINE, Cochrane database, FDA web-portal, and clinicaltrials.gov for all RCTs comparing surgical Maze procedures with medical therapy for sinus rhythm maintenance. Primary outcomes were either freedom from AF within 12 months postprocedure off antiarrhythmic drug (AAD), or freedom from AF while taking an AAD. Secondary outcomes included operative mortality, all-cause mortality, hospital length of stay, and postoperative complications. Both fixed- and random-effects models were used for a meta-analysis of 9 randomized controlled trials (n = 472, of which 249 underwent a Maze procedure and 213 underwent referral surgery alone). The surgical Maze procedure significantly increased the odds of freedom from AF within 12 months compared with cardiac surgery alone (OR 5.22, 95% CI 1.71-15.88). There was significant heterogeneity among the trials for freedom from AF (chi-square = 15.98 for 4 degrees of freedom, P= 0.003). Among the two studies that fully reported AAD use, there was no evidence of improved survival free from AF and AAD therapy (OR 1.78, 95% CI 0.73-4.34). Among patients with valvular AF, surgical Maze procedures are associated with a decrease in AF one year postprocedure without significant increase in mean length of hospital stay, perioperative complications, operative, or all-cause mortality. Large RCTs defining rates of freedom from AF without AADs postprocedure, are still needed to evaluate outcomes and determine the appropriate role for surgical Maze procedures in the management of AF.