BACKGROUND

Chronic atrial fibrillation unassociated with rheumatic valvular heart disease (NRAF) considerably increases the risk of thromboembolism. Recent studies have provided new evidence concerning the risk-benefit ratio of anticoagulant therapies in patients with AF.

OBJECTIVE

To evaluate the incidence of primary end points (ischemic stroke, systemic embolism, bleeding complications to oral anticoagulant or antiplatelet therapy) and secondary end points (death, TIA) in patients with NRAF.

METHODS AND RESULTS

Between November 1992 and June 1993, 694 patients with chronic NRAF were enrolled in the Trieste Area Study on Nonrheumatic Atrial Fibrillation (TASAF), an ongoing prospective community study with a follow-up period of 2 years. The preliminary results of the enrolled study population show: an elevated mean age (71 +/- 9 years), the prevalence of males (383/694), high prevalence of overt or previous heart failure (23%), of mitral regurgitation confirmed at echocardiography (30%) and of previous myocardial infarction (11%). Many of the enrolled patients had a history of hypertension (58%). With regard to the etiology of the underlying heart disease, the following should be emphasized: a high incidence of cardiac hypertrophy (with or without history of hypertension) (28%) and of degenerative cardiopathy (20%); unclassifiable cardiopathy (14%); and lone AF (13%). Echocardiographic findings: left ventricular dysfunction (17%); mitral annular calcification (27%); and good mean left ventricular function (EF 0.50 +/- 0.15). Retrospectively there were 96 clinically documented embolic events in 78 subjects while in 34 patients there were 38 episodes suspected for embolism or TIA. Nine patients suffered 1 recurrence of embolism; three patients suffered 2 recurrences; one patient had 3 recurrences; and 4 patients had one suspected recurrence of TIA. In 35 cases the embolic events clustered around the time of the onset of the arrhythmia. In the other 99 subjects the embolic complication appeared after the onset of AF: range 1-266 months. The group of patients with true embolic events in comparison with patients without embolism or with suspected embolism or TIA had same variables predictive of thromboembolic complications: arrhythmia duration (p = 0.09) and previous myocardial infarction (p = 0.03); in contrast mitral annular calcification (p = 0.06), history of hypertension (p = 0.09) and cardiac hypertrophy (with or without hypertension) (p = 0.07) demonstrated only a slight trend of statistical significance. Comparing the clinical characteristics and echocardiographic findings of patients without embolism with those of patients with tru embolism, or suspected embolism, or TIA the variables predictive of thromboembolic events were: arrhythmia duration (p = 0.007), history of hypertension (p = 0.01), cardiac hypertrophy (with or without hypertension (p = 0.02) and mitral annular calcification (p = 0.01), at the same time, age showed only a trend of statistical significance (p = 0.06). Among the 616 patients without a history of embolism only 3% were treated with oral anticoagulant agents and 28% with antiplatelet therapy, while among the 78 subjects with documented embolism only 28% were receiving anticoagulant therapy and 58% were receiving antiplatelet agents.

CONCLUSIONS

NRAF is an important risk factor for thromboembolism. Some clinical characteristics and echocardiographic findings increase the risk. Physicians still hesitate to use oral anticoagulants and antiplatelet agents in their patients for the prevention of embolic complications.