Abstract 15520: Incorporating Novel Biomarkers Significantly Improves Risk Stratification for Coronary Heart Disease Patients: the Long-term Intervention with Pravastatin in Ischaemic Disease (LIPID) Trial
Background Cardiovascular risk estimation incorporating novel biomarkers has been assessed in the primary prevention setting, but not in patients with prior CHD.
Methods The predictive value of 8 novel baseline biomarkers was assessed in 7863 men and women with prior CHD in the LIPID study with a total of 1100 CHD deaths and nonfatal myocardial infarctions during 6 years of follow-up. Separate Cox regression analyses of each biomarker adjusted for 23 traditional risk factors, and then multivariable modelling was performed with risk quartiles derived.
Results BNP, CRP, cystatin-C, D-dimer, midregional pro-adrenomedullin and troponin I (Tn I) were all predictors of CHD events independent of the traditional factors, but LP(a), and Lp-PLA2 activity were not. In a multivariable model, BNP (HR 1.61, 1.34-1.92 for Q4-Q1), cystatin-C (HR 1.64, 1.36-1.99 for Q4-Q1), TnI (HR 1.50, 1.30-1.75 for >0.018 vs 0 ng/ml) and D-dimer (HR 1.44, 1.20-1.72 for Q4-Q1) all remained independent predictors (all P<0.001 for trend across categories), and collectively resulted in a net 10.3% (P<0.001) reclassification of risk. The relative treatment effect of pravastatin on CHD events was similar for each level of each biomarker. Other significant risk factors in the multivariable model included sex, HDL-C, type of acute coronary syndrome, prior revascularization, stroke, diabetes, smoking, angina, dyspnea, WBC, and treatment with pravastatin.
Conclusion Addition of the four biomarkers, BNP, cystatin C, D-dimer, and TnI, to a conventional risk model significantly improved risk estimation for recurrent CHD over 6 years in patients with known CHD. Relative pravastatin treatment effects were similar across all risk levels. The top quartile of risk identified a subgroup that obtained a large absolute benefit from pravastatin treatment (9% absolute event reduction) but still had substantial residual risk (29% over 6 years), and hence should be priority candidates for more intensive treatment.
- © 2011 by American Heart Association, Inc.