Inflammation and Atrial Fibrillation

Abstract & Commentary

This study reports the association between inflammation as measured by C-reactive protein (CRP) and the prevalence and development of atrial fibrillation. The data are from the Cardiovascular Health Study, a large population-based study of cardiovascular disease that included 5201 men and women, aged 65 or older who were enrolled from Medicare eligibility lists in 4 cities in the United States. After initiation of the study, a second cohort of 687 African-American participants were enrolled, giving a total of 5888 subjects. Of those, 5806 had available CRP levels at baseline. Participants were followed up every 6 months through alternating telephone interviews and annual clinic visits. Discharge diagnoses and hospital records were obtained for all hospital admissions from the Medicare database. The presence at baseline or subsequent development of atrial fibrillation was identified either by either self-report, electrocardiographic documentation, or hospital discharge diagnosis. CRP levels were related to prevalence at baseline and subsequent incidence using a logistic regression analysis after logarithmic transformation of the CRP values. Other risk factors for atrial fibrillation considered in the multivariate analysis included age, gender, race, body mass index, left ventricular dysfunction, systolic and diastolic blood pressures, history of hypertension, coronary heart disease, diabetes mellitus, cerebrovascular disease, and congestive heart failure. The ratios for atrial fibrillation were calculated either by modeling CRP level as a continuous variable or by categorizing CRP levels into quartiles.

Of the 5806 patients included in the cross-sectional study, 315 had atrial fibrillation at baseline. The risk for atrial fibrillation was progressively higher with increasing CRP quartile. CRP remained an independent predictor of baseline prevalence after multivariate analysis. The adjusted odds ratios for baseline atrial fibrillation were 1.00, 1.33, 1.45, and 1.75 in the 4 CRP quartiles. After subjects with baseline atrial fibrillation were excluded, 5491 patients were followed longitudinally to assess if CRP levels could predict future atrial fibrillation. The suggested hazard ratio was 1.31 in the patients in the highest CRP quartile. In a multivariate Cox regression analysis that treated CRP as a continuous variable, there continued to be an association between CRP and risk of future development of atrial fibrillation even after adjustment for multiple other risk factors. Aviles and associates concluded that inflammation as measured by CRP levels is a predictor for both the presence and the development of atrial fibrillation (Aviles RJ, et al. Circulation. 2003;108:3006-3010).

Comment by John DiMarco, MD, PhD

For many years, it has been recognized that factors associated with atrial enlargement, stretch, and fibrosis promote the development of atrial fibrillation. The importance of risk factors that produce atrial stretch, such as hypertension, valvular heart disease, and congestive heart failure, has long been recognized, but the role of inflammation in the pathogenesis of atrial fibrillation is a new observation. In this study, CRP is shown to be a predictor of both the presence of atrial fibrillation at baseline and the development of new atrial fibrillation in elderly patients. Inflammation is not usually a target of antiarrhythmic therapy, but conventional therapy with antiarrhythmic drugs to prevent atrial fibrillation is often unsuccessful. Clinical trials to test the hypothesis that decreasing inflammatory markers with therapy, either as the sole approach or in conjunction with antiarrhythmic drugs, will lead to a decreased prevalence and incidence of atrial fibrillation will be important.

Dr. DiMarco, Professor of Medicine, Division of Cardiology, University of Virginia, Charlottesville, is on the Editorial Board of Clinical Cardiology Alert.

Readers are Invited. . .

Readers are invited to submit questions or comments on material seen in or relevant to Clinical Cardiology Alert. Send your questions to: Christie Petrone, Clinical Cardiology Alert, c/o Thomson American Health Consultants, P.O. Box 740059, Atlanta, GA 30374. For subscription information, you can reach the editors and customer service personnel for Clinical Cardiology Alert via the internet by sending e-mail to