Loading...
As more biologic markers of cardiovascular risk surface, it becomes more important to determine which add prognostic value to risk stratification, beyond that provided by traditional risk factors. These researchers used data from 3209 participants (94% without prevalent cardiovascular disease) from the 1995–1998 cycle of the Framingham Offspring Study to assess the prognostic value of 10 biomarkers: C-reactive protein (CRP), B-type natriuretic peptide (BNP), N-terminal pro-atrial natriuretic peptide, serum aldosterone, plasma renin, fibrinogen, plasminogen-activator inhibitor type 1, d-dimer, homocysteine, and urinary albumin-to-creatinine ratio.
During up to 10 years of follow-up (median, 7.4 years), 6% of participants died and 6% of those without prevalent cardiovascular disease had a first major adverse cardiovascular event. Collectively, the 10 biomarkers showed associations with mortality and with cardiovascular events. Individually, 5 biomarkers — BNP, CRP, urinary albumin-to-creatinine ratio, homocysteine, and plasma renin — made significant contributions to a multivariate-regression model for predicting death; only 2 of them — BNP and urinary albumin-to-creatinine ratio — made significant contributions for predicting adverse cardiovascular events. Including these 5 “most predictive” biomarkers in models that also incorporated conventional risk factors added only marginal prognostic value.
Wang TJ et al. Multiple biomarkers for the prediction of first major cardiovascular events and death. N Engl J Med 2006 Dec 21; 355:2631-9.
Ware JH. The limitations of risk factors as prognostic tools. N Engl J Med 2006 Dec 21; 355:2615-7.
Comment
In this Framingham analysis, poorer levels of only 5 of 10 well-studied novel biomarkers made significant contributions to a multivariate-regression model of mortality risk. Only 2 two of them — and, notably, not CRP — made significant contributions to a model predicting risk for a first major cardiovascular event. More important, these biomarkers added relatively little value to a model that incorporated traditional risk factors, which is consistent with an analysis of data from the ARIC study (Journal Watch Cardiology Aug 9 2006). In a related perspective piece, a noted statistician affirms that this group of biomarkers made a statistically significant contribution to risk prediction but that they had limited value for risk stratification of individual patients. The appropriate role of these biomarkers in the clinical care of patients remains to be determined.