BACKGROUND: Few epidemiologic cohort studies have evaluated atrial flutter (flutter) as an arrhythmia distinct from atrial fibrillation (AF).

OBJECTIVE: The purpose of this study was to examine the clinical correlates of flutter and its associated outcomes to distinguish them from those associated with AF in the Framingham Heart Study.

METHODS: We reviewed and adjudicated electrocardiograms (ECGs) previously classified as flutter or AF/flutter and another 100 ECGs randomly selected from AF cases. We examined the clinical correlates of flutter by matching up to 5 AF and 5 referents to each flutter case using a nested case referent design. We determined the 10-year outcomes associated with flutter with Cox models.

RESULTS: During mean follow-up of 33.0 +/- 12.2 years, 112 participants (mean age 72 +/- 10 years, 30% women) developed flutter. In multivariable analyses, smoking (odds ratio [OR] 2.84, 95% confidence interval [CI] 1.54-5.23), increased PR interval (OR 1.28 per SD, 95% CI 1.03-1.60), myocardial infarction (OR 2.25, 95% CI 1.05-4.80) and heart failure (OR 5.22, 95% CI 1.26-21.64) were associated with incident flutter. In age- and sex-adjusted models, flutter (vs referents) was associated with 10-year increased risk of AF (hazard ratio [HR] 5.01, 95% CI 3.14-7.99), myocardial infarction (HR 3.05, 95% CI 1.42-6.59), heart failure (HR 4.14, 95% CI 1.90-8.99), stroke (HR 2.17, 95% CI 1.13-4.17), and mortality (HR 2.00, 95% CI 1.44-2.79).

CONCLUSION: We identified the clinical correlates associated with flutter and observed that flutter was associated with multiple adverse outcomes.