More than half of atrial fibrillation is likely attributable to modifiable risk factors, including high blood pressure, obesity, and smoking, according to an analysis of the Atherosclerosis Risk in Communities Study.

The finding, which was based on a cohort of middle-aged American adults from communities in the ARIC study, highlights the need for primary prevention among this population.

"Moreover, because improvement in these behaviors would also favorably affect other AF risk factors, such as diabetes mellitus and impaired glucose tolerance, the reduction in the incidence of AF would be even greater than expected through BP lowering alone," wrote Rachel R. Huxley, D.Phil., in the journal Circulation, published online March 28.

Dr. Huxley, an epidemiologist at the University of Minnesota, Minneapolis, and colleagues looked at nearly 20 years of follow-up from the ARIC survey, a prospective cohort study of atherosclerotic diseases in Forsyth County, N.C.; Jackson, Miss.; Washington County, Md.; and the suburbs of Minneapolis.

At baseline in 1987-1989, it included 15,792 men and women aged 45-64 years, selected by area probability sampling.

All patients underwent three triennial visits, and in the present study, AF cases were counted as those diagnosed at an incident study visit (not baseline assessment) by 12-lead electrocardiogram, or during follow-up with an ICD code for AF in a hospital discharge, or when AF was listed as any cause of death on a death certificate.

Those with baseline AF on electrocardiogram or history of AF were excluded, as were those with missing data.

The researchers of the current study characterized participants into one of three risk profiles. An "optimal" group had no history of cardiac disease, systolic BP less than 120 mm Hg and diastolic less than 80 mm Hg without antihypertensive drugs; a body mass index less than 25 kg/m²; fasting serum glucose less than 100 mg/dL without antidiabetic drugs or history of diabetes; and no history of smoking.

"Borderline" participants had any of the following criteria: systolic BP 120-139 and/or diastolic BP 80-89 mm Hg without antihypertensives; BMI 25-30; fasting glucose 100-125 mg/dL without use of antidiabetics and no history of diabetes; and former smoker status.

Finally, participants regarded as having "elevated" risk profiles had any of the following: history of cardiac disease (heart failure or coronary heart disease); systolic BP at least 140 mm Hg, diastolic BP at least 90 mm Hg, or use of antihypertensives; BMI greater than 30; fasting serum glucose greater than or equal to 126, or use of antidiabetic drugs; history of diabetes; or current smoker status.

Among the 14,598 subjects (55% women; 25% black; mean age, 54.2 years) over a mean 17.1 years of follow-up who were included in the current analysis, there were 1,520 cases of incident AF.

Compared with those with no risk factors, the age-adjusted incidence rates were three times higher in those with one or more elevated risk factors (2.19 vs. 6.59 per 1,000 person-years, respectively), wrote the authors (relative hazard for optimal patient group, 0.33).

For the borderline group, the incidence rate was 3.68, for a relative hazard of 0.50, compared with participants who had one or more elevated risk factors.

"Overall, the [population-attributable fraction] estimates indicated that having [one or more] elevated risk factor levels could explain 50% ... of AF events," added the authors (Circulation 2011:10.1161/CIRCULATIONAHA.110.009035).

Adding elevated and borderline levels together, that number jumped to 57%.

The authors also sought to determine which of the relevant risk factors played the biggest role in incident AF.

Elevated blood pressure, experienced by 39% of the entire cohort, accounted for roughly one in five cases of AF (21.6%). "This rose to 24.5% if borderline levels of BP, which affected another 22.7% of the cohort, were also included," wrote the authors.

"Obesity and overweight explained 17.9% of all AF cases, and diabetes mellitus and impaired glucose tolerance combined accounted for the smallest fraction [3.9%] of the AF burden in this cohort," they added.

According to the authors, this study is the second to look at the relationship between modifiable risk factors and atrial fibrillation. A 1994 analysis, using data from the Framingham cohort, found that smoking, diabetes, hypertension, and prevalent coronary heart disease combined explained 44% of the AF burden in men and 58% in women, a conclusion that Dr. Huxley called "broadly comparable" to that of the current study.

They added that the study was limited by its inability to differentiate subtypes of AF, as well as by its reliance on hospital discharge codes to ascertain more than 98% of AF cases recorded in this cohort. However, if anything, this fact likely led to the "underascertainment of cases that perhaps were not severe enough to warrant hospitalization."

The study was funded by the National Heart, Lung, and Blood Institute as well as the American Heart Association. The investigators reported having no other disclosures related to this study.

More Than Half of AF Cases Are Preventable