Comparison of cardiac risk models has implications for statin use in women

The Reynolds Risk Score was better than the Framingham-based model at determining women's cardiovascular risk, according to a large external validation cohort.

The differences are large enough to have a clinical impact for statin therapy, the researchers noted. They drew a case-cohort sample from the Women's Health Initiative Observational Study (WHI-OS), comprising 1,722 cases of major cardiovascular disease (CVD) (752 myocardial infarctions, 754 ischemic strokes, and 216 other cardiovascular disease deaths) and a random subcohort of 1,994 women without prior CVD.

Researchers estimated risk using the ATP-III score, the Reynolds Risk score and the Framingham CVD model. Results appeared online March 7 in Circulation.

The models found differing percentages of the women to be at intermediate risk of a cardiovascular event (10% or higher risk in the next 10 years): 5.5% according to ATP-III, 10.3% according to Reynolds, and 41.1% according to Framingham. Higher risk (20% or higher in the next 10 years) was predicted for 0.5%, 2.6%, and 10.6% of women by the three models, respectively.

The Reynolds Risk Score appeared relatively well-calibrated for the endpoint of major CVD, the authors noted. The Framingham CVD model, developed for a broader definition of CVD, greatly overestimated risk of major CVD. A similar pattern of overestimation of risk was seen for the ATP III model for coronary heart disease.

To directly compare discrimination of the three models, researchers recalibrated them so that the average predicted risk equaled the overall reweighted population estimate of 4%. After this, the percent of women with estimated risk of 10% or higher was then 6.6% for the ATP-III, 7.7% for Reynolds, and 6.2% for the Framingham CVD model.

The large differences in risk estimates among the models have clinical implications for statin therapy, the authors wrote. In a hypothetical population of 100,000 women, the number who would be classified at 10% or higher risk would be 5,549 with the ATP-III model, 10,304 with the Reynolds score, and 41,074 with the Framingham CVD model.

The authors concluded, “The Reynolds Risk Score significantly improved fit as compared to either the Framingham-based ATP-III CHD risk score or the newer Framingham CVD score. Within the WHI-OS, the greatest impact of the Reynolds Risk Score appeared to be among those women with 5 to 10% 10-year estimated risk according to ATP-III, a group including a large number of women destined to suffer MI or stroke, and in whom trial data indicate efficacy of statin therapy in reducing cardiovascular events.”

An editorial noted that using the Reynolds score instead of ATP-III would reclassify a substantial number of women. Of the 5% of women at ATP-III risk of 10% to 20% (intermediate/moderately high risk), 23% would be reclassified in a higher-risk category and 18% in a lesser. Of those originally predicted to have 5% to 10% 10-year (moderate) risk based on ATP-III, 5% would be reclassified as more than 20% and 29% as above 10%. Since 8 to 10 million American women have an ATP-III 10-year risk of 5% to 20%, the Reynolds score could have major impact on CVD prevention in women.

The editorialists added that the three prediction models are overly complex for clinical use, despite online calculators. And trials of statins for primary prevention didn't use Framingham or Reynolds thresholds, making LDL targets in the research and the risk factor scores discordant. Finally, while Reynolds focused on 10-year risk, the clinical issue for many women is lifetime risk.

“Having too much confidence in cardiovascular risk prediction does indeed appear to be a risky proposition,” the editorial concluded.