Mortality trends in patients diagnosed with first atrial fibrillation - A 21-year community-based study

Objectives The purpose of this study was to assess the mortality trends of atrial fibrillation (AF) in a community. Background Limited data exist regarding the mortality trends of patients diagnosed with first AF. Methods A community-based cohort of adult residents of Olmsted County, Minnesota, who had electrocardiogram-confirmed first-documented AF in the years 1980 to 2000 were identified and followed to 2004 or death. The primary outcome was all-cause mortality. Results Of a total of 4,618 residents (mean age 73 +/- 14 years) diagnosed with first AF, 3,085 died during a mean follow-up of 5.3 +/- 5.0 years. Relative to the age- and gender-matched general Minnesota population, the mortality risk was increased (p < 0.0001) with a hazard ratio (HR) of 9.62 (95% confidence interval [CI] 8.93 to 10.32) within the first 4 months and 1.66 (95% Cl 1.59 to 1.73) thereafter. Cox proportional hazards modeling showed no change in overall age- and gender-adjusted mortality (HR for the year 2000 vs. 1980: 0.99; 95% Cl 0.86 to :1.13; p = 0.84), even after adjustment for comorbidities. In secondary analyses, no changes in mortality were seen for early (within first 4 months) or late (after 4 months) mortality for the entire group or within the subgroup of patients who did not have cardiovascular disease at baseline. Conclusions In this cohort of patients newly diagnosed with AF, mortality risk was high, especially within the first 4 months. There was no evidence for any significant changes over the 21 years in terms of overall mortality, early or late mortality, or mortality among patients without pre-existing cardiovascular disease. (c) 2007 by the American College of Cardiology Foundation