PURPOSE: Atrial fibrillation is an important risk factor for stroke. We ana
lyzed stroke risk over time in patients discharged from the hospital with a
diagnosis of incident atrial fibrillation as compared with the risk of str
oke in the Danish population.
SUBJECTS AND METHODS: In a random sample of half of the Danish population,
we identified 13,625 men and 13,577 women, aged 50 to 89 years, with a hosp
ital diagnosis of atrial fibrillation and no prior diagnosis of stroke duri
ng 1980 to 1993. Data on other medical conditions were also available from
1977 to 1993, but medication data were not available. Patients were followe
d from the diagnosis of atrial fibrillation until the first diagnosis of st
roke (nonfatal or fatal cerebral ischemic infarct and cerebral hemorrhage),
death, or the end of 1993. The risk of stroke in these patients was compar
ed with the risk in the Danish population using Poisson regression modeling
to estimate relative risks (RR) and 95% confidence intervals (CI).
RESULTS: For men with atrial fibrillation, the stroke rates increased by ag
e, from 13 per 1,000 person-years in those ages 50 to 59 years, to 22 per 1
,000 person-years in those ages 60 to 69 years, to 42 per 1,000 person-year
s In those ages 70 to 79 years, to 51 per 1,000 person-years in those ages
80 to 89 years. Age-specific stroke rates were similar id women with atrial
fibrillation. Patients with a hospital diagnosis of atrial fibrillation ha
d an increased risk of stroke (RR = 2.4; 95% CI, 2.3 to 2.5 in men and RR =
3.0; 95% CI, 2.9 to 3.2 in women) compared with the Danish population. Str
oke risk was greatest during the first year after discharge and decreased t
hereafter. Hypertension, diabetes, and peripheral atherosclerosis were also
associated with an increased risk of stroke among patients with atrial fib
rillation. Ischemic heart disease and heart failure were risk factors in me
n only. There was no reduction in the risk of stroke from 1980 to 1993.
CONCLUSIONS: Men and women with atrial fibrillation are at a substantially
increased risk of stroke, particularly in the first year after the diagnosi
s. Am J Med.2000;108:36-40. (C)2000 by Excerpta Medica, Inc.