Do novel biomarkers add to existing scores of total cardiovascular risk?

Abstract

While cardiovascular disease and certain other conditions are considered to confer a high or very high risk of cardiovascular events, the asymptomatic population can be subdivided in different categories of total CV risk using risk models; this allows the clinician to adapt the intensity of preventive strategies accordingly. Risk models, such as that based on the US Framingham Study and the SCORE model, based on European cohorts, estimate risk according to the presence of risk factors, including age, gender, smoking habits, systolic blood pressure, and cholesterol levels. However, estimation of an individual’s cardiovascular risk remains approximate, and whether new biomarkers of risk will improve risk assessment is a key question. Several novel cardiovascular risk markers have been suggested, including lipid, inflammatory, thrombotic, and genetic biomarkers. Demonstrating that a novel biomarker is predictive of cardiovascular disease is, by itself, insufficient proof that it adds incremental value to existing risk estimation models. The Net Reclassification Improvement index provides an indication of the ability of a novel marker to improve risk estimation by classifying individuals to a more correct category. In addition, new risk models may be calibrated by measuring how closely predicted outcomes agree with actual outcomes. Traditional cardiovascular risk factors explain most of an individual’s risk. Consequently, the addition of new risk factors to existing models has provided disappointingly small effects overall. However, there addition to conventional risk estimation may be useful in correctly reclassifying individuals at intermediate risk as above or below a chosen intervention threshold.