OR WAIT null SECS
BOSTON -- Two new risk factors added to the traditional risk model enhanced prediction of a woman's 10-year risk of cardiovascular disease or stroke, researchers reported.
BOSTON, Feb. 13 -- Two new risk factors added to the traditional risk model enhanced prediction of a woman's 10-year risk of cardiovascular disease or stroke, researchers reported.
Two versions of the revised model, with validated risk algorithms, reclassified 40% to 50% of women at 10-year risk into higher- or lower-risk categories, according to a report in the Feb. 14 issue of the Journal of the American Medical Association.
Only two new risk variables -- parental family history of premature coronary heart disease (before age 60) and high-sensitivity C-reactive protein (hsCRP) -- were added to the variables in the standard Adult Treatment Panel-III risk (ATP-III) score, said Paul Ridker, M.D., of Brigham of Women's Hospital here, and colleagues.
From 1956 through 1966, Framingham investigators defined what they named "coronary risk factors" (hypertension, smoking, diabetes, and hyperlipidemia), which were then codified into global risk scores. However, Dr. Ridker said, among women, up to 20% of all coronary events occur in the absence of these risk factors, whereas many women with traditional risk factors do not have a coronary event.
The authors also noted that although the understanding of cardiovascular disease has changed dramatically in the past half century, the variables included in current risk algorithms are largely unchanged from those recommended 40 years ago.
To develop and validate traditional and novel risk factors for women, the researchers assessed 35 factors among 24,558 initially healthy U.S. women 45 years or older from the Women's Health Study, a nationwide cohort started in 1992.
The women were followed up for a median of 10.2 years (through March 2004) for incident cardiovascular events (a composite of myocardial infarction, ischemic stroke, coronary revascularization, and cardiovascular death).
Data from a random two-thirds (16,400) of the women, the so-called derivation cohort, were used to develop new risk algorithms. These predictors were then tested and compared with observed and predicted outcomes in the remaining one third of the women (8,158), designated the validation cohort.
Minimization of the Bayes Information Criterion was used
in the derivation cohort to develop the best-fitting parsimonious prediction models. In the validation cohort, the researchers compared predicted versus actual 10-year cardiovascular event rates when the new algorithms were compared with models included in the ATP-III risk score.
In the derivation cohort, a best-fitting model (model A) and a clinically simplified model (model B, the Reynolds Risk Score) had lower Bayes Information Criterion scores than models in the older ATP-III score.
In the validation cohort, all measures of fit, discrimination, and calibration were improved when either model A or B was used, the researchers reported.
For example, among participants without diabetes with estimated 10-year risks according to the ATP-III score of 5% to less than 10% (n=603) or 10% to less than 20% (n=156), the best-fitting model A reclassified 379 (50%) into higher- or lower-risk categories. In each instance the revised categories more accurately matched actual event rates, the researchers said.
Similar validation effects were achieved for the clinically simplified model B (the Reynolds Risk Score), which was limited to age, systolic blood pressure, hemoglobin A1c if diabetic, smoking, total and high-density lipoprotein cholesterol, high-sensitivity C-reactive protein, and parental history of myocardial infarction before age 60.
Neither new algorithm (A or B) provided substantive information about women at very low risk according to the published ATP-III score, they said.
Despite the sample size, limitations of this analysis merit discussion, the researchers said. First, the cohort was limited to women and was largely white. Second, data on blood pressure, obesity, and family history were based on self-report. However, the Women's Health Study, source of the participants, is composed of female professionals known to provide accurate information, they said.
Regarding parental history, a conservative cut point of 60 years of age was found to be consistent with Framingham analyses, the investigators said.
Also, the combined endpoints of myocardial infarction ischemic stroke, coronary revascularization, and mortality are endpoints typically used in clinical trials.
Finally, the researchers noted that for practical reasons, the analysis was limited to blood-based biomarkers and traditional epidemiological risk factors. This study did not examine the potential for atherosclerotic imaging tests to serve as an alternative method for evaluating risk.
However, they said, the methods described here-variable selection in a derivation data set followed by prospective testing in a second validation cohort-should provide a structure for the formal evaluation of emerging risk predictors, including potential imaging tests.
In conclusion, Dr. Ridker's team wrote, "As eight to 10 million U.S women have an ATP-III-estimated 10-year risk between 5% and 20%, application of these data could have an immediate effect on cardiovascular prevention. A user-friendly
calculator for the Reynolds Risk Score can be freely accessed at http: //www.reynoldsriskscore.org, the researchers added.
In an accompanying editorial, Roger Blumenthal, M.D., of Johns Hopkins in Baltimore, and colleagues said that the study by Dr. Ridker and colleagues provides a "timely contribution to the cardiovascular-risk-prediction literature."
Importantly, they said, the model is simple including only two new variables, and the expanded prediction of the Reynolds model includes total cardiovascular events including stroke and coronary revascularization, in addition to hard coronary heart-disease endpoints.
These findings raise several critical questions including what is the impact in terms of new risk-prediction categories, altering the low-density lipoprotein cholesterol goal, and influencing the choice of whether to treat with life-long aspirin in individual patients.
The new model reclassified 5,400 women (5.4%) to high-risk with LDL-C goals of less than100 mg/dL and an optional goal of less than 70 mg/dL, Dr. Blumenthal said.
On the other hand, 13,400 (13.4%) of the women were reclassified as very low-risk women with a less than 5% risk of a major cardiovascular disease event and with an LDL goal of less than 160 mg/dL.
As a result, the editorialists said, approximately 20% of women will have different lipid treatment goals based on the Reynolds model, compared with the ATP-III guidelines, when considering a 5% to 20% 10-year risk as an intermediate risk.
Dr. Blumenthal and his colleagues noted other questions: Which of the two newly added variables has a greater influence on risk prediction? Could the Reynolds risk prediction algorithm work for men? Finally, what is the value of other markers in risk prediction, such as coronary-artery-calcification scores or exercise-testing measures that were not assessed in this study?
From a practical standpoint, they wrote, both the Framingham and the Reynolds Risk models predict only short-term 10-year risk. But Framingham data indicate that a woman who is free of cardiovascular disease at age 50, has a lifetime risk of a cardiovascular event of 39%. Lifetime risk is high, and any single risk factor left untreated will lead to atherosclerotic vascular disease.
"Future multivariable models to predict a woman's long-term (20 to 30 years) risk of developing a major atherosclerotic vascular disease event are also needed. The Reynolds Risk Score is an important contribution to preventive cardiology and provides the framework for evaluating future emerging risk factors," Dr. Blumenthal and his colleagues concluded.
Dr. Blumenthal, one of the editorial writers, reported that he has received clinical research support from Merck, Pfizer, and General Electric.