The natural history of lone atrial fibrillation. A population-based study over three decades.

From 1950 to 1980, 3623 patients from Olmsted County, Minnesota, were found to have atrial fibrillation. Ninety-seven of these patients (2.7 percent), who were 60 years old or younger at diagnosis, had lone atrial fibrillation (atrial fibrillation in the absence of overt cardiovascular disease or precipitating illness), and their data were reviewed to determine the incidence of thromboemboli. Twenty of these patients (21 percent) had an isolated episode of atrial fibrillation, 56 (58 percent) had recurrent atrial fibrillation, and 21 (22 percent) had chronic atrial fibrillation. The total follow-up period was 1440 person-years, with a mean of 14.8 years per patient. The mean age at diagnosis was 44 years. Nineteen cardiovascular events occurred in 17 patients; 4 patients had strokes thought to be due to emboli from atrial fibrillation, and 4 had myocardial infarctions without overt evidence of previous coronary artery disease. The probability of survival at 15 years was 94 percent among the patients with lone atrial fibrillation. At 15 years, 1.3 percent of the patients had had a stroke on a cumulative actuarial basis. On an actuarial basis, there was no difference in survival or in survival free of stroke among the patients with the three types of lone atrial fibrillation (i.e., isolated, recurrent, and chronic). We conclude that lone atrial fibrillation in patients under the age of 60 at diagnosis is associated with a very low risk of stroke. This suggests that routine anticoagulation may not be warranted.