The most award winning
healthcare information source.
TRUSTED FOR FOUR DECADES.
Reynolds Risk Score: An Update to Cardiovascular Risk Assessment in Women
Abstract & Commentary
By Eileen C. West, MD, Director of Primary Care Women's Health,Clinical Assistant Professor of Internal Medicine, University of Oklahoma School of Medicine, Oklahoma City. Dr. West reports no financial relationship to this field of study.
Synopsis: For women, up to 20% of all coronary events occur in the absence of traditional major risk factors, and many women with these risk factors do not develop coronary events. The Reynolds Risk Score adds high sensitivity C-reactive protein (hs-CRP) and family history to traditional cardiovascular risk factors in order to more accurately predict cardiovascular risk in women than the ATP-III model currently in use.
Source: Ridker PM, et al. Development and Validation of Improved Algorithms for the Assessment of Global Cardiovascular Risk in Women: The Reynolds Risk Score. JAMA. 2007;297:611-619.
In the 1960s the Framingham Study helped to define hypertension, age, hyperlipidemia, smoking, and diabetes as the major factors in development of cardiovascular disease. These data were incorporated into global models for risk assessment which have received widespread attention in recent years. Unfortunately for women, up to 20% of all coronary events occur without these major risk factors, and many women with those risk factors never develop coronary events. Fifty years of research has helped hone our understanding of the biological processes of atherosclerosis. Concepts including hemostasis, endothelial dysfunction, inflammation, thrombosis and plaque instability are better defining the pathophysiology, but have not yet been incorporated into risk algorithms for women and heart disease.
Researchers from the Donald W. Reynolds Center for Cardiovascular Research at Boston's Brigham and Women's Hospital have developed a new algorithm for the assessment of global cardiovascular risk in women. Using data collected from 1992 to 2004 in the Women's Health Study (WHS), the researchers selected about 25,000 US women 45 years and older who were free of cardiovascular disease and cancer at the start of the study and followed them for an average of 10.2 years. They looked for incident myocardial infarction, ischemic stroke, coronary revascularization, and cardiovascular deaths. Plasma samples were measured for total cholesterol, HDL-C, low-density lipoprotein cholesterol (LDL-C), lipoprotein (a), apolipoproteins A-1 and B-100, hs-CRP, sICAM-1, fibrinogen, creatinine, hemoglobin A1c and homocysteine.
Two different models for risk prediction (A and B) were defined at the outset and compared to a validation cohort. Model A was more detailed than Model B. Model B was designed for easy use in the clinical setting. What they found was that both models were superior to existing models because they narrowed the large group of women who fell into two "intermediate risk" categories. The 10-year risk groups used matched those of the ATP-III risk prediction model, namely less than 5% (low risk), 5% to less than 10% (low to moderate risk), 10% to less than 20% (moderate to high risk), and 20% or higher (high risk).
Actual event rates for model A matched well with predicted rates in nearly all groups studied. Using Model A, 50% of women who fell into the 5-20% intermediate categories were reclassified more accurately as lower or higher risk. Model B is a simplified version of Model A. It too withstood statistical testing and proved to be more accurate in determining who would develop disease than the most often used existing model.
So, what new information have we obtained, and how can we use it in clinic? The analysis supports the use of the inflammatory marker hs-CRP in calculating cardiovascular risk in women. It incorporates family history of heart disease before age 60 years. The traditional risk factors systolic blood pressure, total cholesterol, HDL, smoking and hemoglobin A1c (if diabetic) are included. The new model is called the Reynolds Risk Score, and an online calculator may be found at http://www.reynoldsriskscore.org.
What didn't help so much in creating a risk model were homocysteine, fibrinogen, sICAM-1, and serum creatinine. Homocysteine and sICAM-1 do appear to predict risk, but did not meet criteria for inclusion because there was a less clear correlation between lower values and lower risk. Of interest, it seems neither body mass index, alcohol use, nor exercise frequency improved the predictive power of the model. Also of note, there was no significant contribution from menopausal status or hormone therapy. Limitations of the study include the study population which consisted of relatively well off, primarily white women health professionals. But as the authors note, all components of the models presented have been found previously to predict cardiovascular risk in men.
I follow this data with great interest, as it helps to give a good reason to order inflammatory markers in assessment for cardiovascular disease risk. As yet, the role of imaging in prediction of cardiovascular disease has not been established. More research is needed. This analysis may have two benefits: it gives more accurate risk prediction, and it presents a framework to use when designing future research with imaging tests. Look for more information in the months and years to come on this pertinent and timely model.