NEW ORLEANS — Eight easily obtained clinical variables together formed a risk score that could predict a person's risk for developing atrial fibrillation with reasonable reliability, on the basis of an analysis using data from the Framingham Heart Study.
“The next step is to show the transportability [of this risk score] to other cohorts,” Dr. Renate B. Schnabel said at the annual scientific sessions of the American Heart Association.
This is the first reported risk-assessment tool for predicting a person's likelihood for developing atrial fibrillation, and it is simple enough to be “easily applicable for clinical assessment,” said Dr. Schnabel, a researcher at Boston University and with the Framingham (Mass.) Heart Study. The formula has the potential to identify high-risk patients and help in communicating risk information to patients. Further study is needed to determine whether modifying some of the component risk factors can result in a reduced incidence of atrial fibrillation, she said.
The formula was derived with data collected from 4,764 women and men enrolled in the original Framingham Heart Study, which began in 1948, or in the Framingham Offspring Study, begun in 1971. These people were aged 46–95 at the time of enrollment, with an average age of 61. Records from more than 8,000 clinical examinations were reviewed. Incident atrial fibrillation was identified on the basis of records in the charts of these examinations, including ECG data.
The eight factors identified as significant determinants of risk for developing atrial fibrillation were age, gender, body mass index (BMI), systolic blood pressure, treatment for hypertension, PR interval, significant heart murmur, and heart failure. Together, these factors could account for 78% of incident atrial fibrillations.
The risk for atrial fibrillation was higher in men than in women, and no explanation is available to account for this gender effect, Dr. Schnabel said. Risk was also elevated with increases in age, body mass index, systolic blood pressure, and the duration of the PR interval. It was also higher in those being treated for hypertension, those who had a significant heart murmur, and those with heart failure.
An example of a low-risk person is a woman aged 60 with a BMI of 20 kg/m
In contrast, a high-risk woman would be 70 years old with a BMI of 35 kg/m
The researchers also examined whether adding three variables—left atrial size, left ventricular wall thickness, and fractional shortening—obtained from an echocardiographic examination could further improve the risk score. But there was only a slight improvement in predictive accuracy, and they were judged to not be worth including in the formula.
In future studies, the researchers will look at whether other echocardiographic findings can make a more substantial difference. They plan to post a calculator on the Framingham Heart Study's Web site (www.framinghamheartstudy.org/risk/index.html
This tool for predicting riskfor atrial fibrillation is 'easily applicable for clinical assessment.' DR. SCHNABEL