Background: It has been assumed that cardioversion in patients with atrial fibrillation lasting less than 48 hours is associated with a low risk for thromboembolism. However, no clinical data support this assumption.
Objective: To determine the incidence of cardioversion-related clinical thromboembolism among patients presenting with atrial fibrillation lasting less than 48 hours.
Design: Patients were prospectively identified on admission, and clinical data on the duration of atrial fibrillation were recorded. Data on cardioversion and thromboembolism were obtained retrospectively from hospital and outpatient records.
Setting: Academic medical center.
Patients: 1822 consecutive patients admitted to the hospital for atrial fibrillation were screened. Three hundred seventy-five adults (mean age ±SD, 68 ± 16 years) with atrial fibrillation that had lasted less than 48 hours were identified. One hundred eighty-one patients (48.3%) had a history of atrial fibrillation; 23 (6.1%) had a history of thromboembolism.
Results: 357 patients (95.2%) converted to sinus rhythm during the index admission; spontaneous conversion occurred in 250 patients (66.7%) and active pharmacologic or electrical conversion was done in 107 patients (28.5%). Three patients (0.8% [95% CI, 0.2% to 2.4%]), all of whom had converted spontaneously after ventricular rate control was begun, had a clinical thromboembolic event: One had a stroke, 1 had a transient ischemic attack, and 1 had a peripheral embolus. None of, these 3 patients had a history of atrial fibrillation or thromboembolism, and all had normal left ventricular systolic function.
Conclusion: Among patients presenting with atrial fibrillation that was clinically estimated to have lasted less than 48 hours, the likelihood of cardioversion-related clinical thromboembolism is low. These data support the current recommendation for early cardioversion in these patients.