Determinants of arterial stiffness in patients with atrial fibrillation.

BACKGROUND: Arterial stiffness has emerged as a strong predictor of cardiovascular disease, end-organ damage and all-cause mortality. Although increased arterial stiffness has been described as a predictor of atrial fibrillation, the relationship between arterial stiffness and atrial fibrillation is uncertain. AIM: We assessed arterial stiffness in patients with atrial fibrillation compared with that in a control group. METHODS: We enrolled 151 patients with atrial fibrillation who underwent pulmonary vein isolation (mean age 71.1±9.8 years) and 54 control patients with similar cardiovascular risk profiles and sinus rhythm, matched for age (mean age 68.6±15.7 years) and sex. Aortic distensibility as a measure of arterial stiffness was assessed by transoesophageal echocardiography. Patients with atrial fibrillation were followed over a median of 21 (15 to 31) months. RESULTS: Compared with control patients, patients with atrial fibrillation had significantly lower aortic distensibility (1.8±1.1 vs. 2.1±1.1 10-3mmHg-1; P=0.02). Age (hazard ratio 0.67, 95% confidence interval 0.003 to 0.03; P=0.02) and pulse pressure (hazard ratio -1.35, 95% confidence interval -0.07 to -0.03; P<0.0001) were the strongest predictors of decreased aortic distensibility in the study cohort. This effect was independent of the type of atrial fibrillation (paroxysmal/persistent). During follow-up, decreased aortic distensibility was a predictor of cardiovascular and all-cause hospitalizations, as well as recurrences of atrial fibrillation, with a higher incidence rate of events in patients in the lowest aortic distensibility quartile (P=0.001). CONCLUSIONS: Aortic distensibility was significantly reduced in patients with atrial fibrillation, with age and pulse pressure showing the strongest correlation, independent of the type of atrial fibrillation. Additionally, decreased aortic distensibility was associated with cardiovascular and all-cause hospitalizations, as well as recurrences of atrial fibrillation, which showed a quartile-dependent occurrence.