Cardioversion in atrial fibrillation. Results and complications in 1,152 prospective patients. Study Group of the Working Society of Leading Cardiologic Hospital Physicians

BACKGROUND: Pharmacological and direct-current cardioversion of atrial fibrillation are often performed interventions. Little is known about results and complications of cardioversion in daily practise. PATIENTS AND METHODS: Demographic, procedural and outcome data from patients with cardioversion of atrial fibrillation were collected in a prospective, multicenter registry of 61 hospitals. RESULTS: Between July 1994 and December 1994 1152 patients with a mean age of 64 +/- 11 years were registered on an intention-to-treat basis. 62% were male. The most prevalent underlying disorders were coronary artery disease (34.7%), valvular heart disease (18.1%), and cardiomyopathy (6.9%). 16.4% of patients had lone atrial fibrillation. New onset atrial fibrillation was reported in 21%, paroxysmal in 32% and chronic in 47% of patients. The mean duration of atrial fibrillation was 7 +/- 26 weeks (range 1 day to 7 years, median 5 days). In 3.8% of patients no cardioversion attempt was made and follow-up was not possible in another 5.5%. 19.2% of patients cardioverted spontaneously. Direct current cardioversion was attempted in 39.7% and pharmacological cardioversion in 31.8% of patients. Cardioversion was successful (sinus rhythm at discharge) in 96.4% of spontaneous cardioversion, in 73.1% of direct current cardioversion and in 84.4% of pharmacological cardioversion. Success of cardioversion was significantly related to duration of atrial fibrillation, NYHA functional class and left atrial diameter (p < 0.001). In 55 (4.8%) cases complications were reported of which 14 were fatal. Five cases of sudden death occurred, all of which were related to quinidine therapy for pharmacological cardioversion. Five cases of embolism were reported. Two were not associated with cardioversion attempts and 3 occurred within 24 hours after successful direct current cardioversion. Two of these patients were effectively anticoagulated at the time of cardioversion. A total of only 62% of patients with atrial fibrillation of more than 48 hours duration were anticoagulated for cardioversion with coumadine or i.v. heparin. CONCLUSIONS: The main risks of cardioversion are fatal proarrhythmic events in pharmacological attempts to restore sinus rhythm. The risk of embolism is despite low rates of effective anticoagulation low.