Lost life years attributable to stroke among patients with nonvalvular atrial fibrillation: a nationwide population-based follow-up study

Aim: We assessed the number of lost life years attributable to stroke among patients with a hospital diagnosis of nonvalvular atrial fibrillation.

Methods: We identified all patients, aged 40-89 years, with an incident hospital diagnosis of atrial fibrillation or flutter in the Danish National Registry of Patients from calendar year 1980 to 2002, and no previous or concomitant diagnosis of stroke or heart valve disease. All patients were followed in the Danish National Registry of Patients for occurrence of an incident diagnosis of stroke of any type (ischemic and/or hemorrhagic) and in the Danish Civil Registration System for emigration or change in vital status. We used multivariate Cox regression analysis with stroke as a time dependent covariate to estimate excess mortality associated with incident stroke. The baseline hazard function for mortality was computed and used for modeling lost life years by sex, age, and time to incident stroke after diagnosis of atrial fibrillation, adjusted for conditions of comorbidity and calendar year of diagnosis of atrial fibrillation.

Results: The mean loss of life years attributable to incident stroke within 20 years after a first diagnosis of atrial fibrillation was most frequently less than 5 years, but a mean of up to 10 years of lost life years was observed. The largest number of lost life years was observed in women, in younger patients, and in those who had a stroke early after the diagnosis of atrial fibrillation. The relative loss of life years was up to 90% of the estimated expected remaining lifetime without stroke within 20 years after the diagnosis of atrial fibrillation, and was highest in the elderly.

Conclusion: Stroke causes a substantial loss of life years in patients with atrial fibrillation.