Abstract 3289: Novel Biomarkers Add Prognostic Value over Traditional Risk Factors for Nonhemorrhagic Stroke in Patients with Known Coronary Heart Disease: the Long-Term Intervention with Pravastatin in Ischaemic Disease (LIPID) Trial

Abstract

Introduction: The potential for novel biomarkers to identify risk of future nonhemorrhagic stroke in patients with coronary heart disease (CHD) has not been determined. We examined the potential of 8 novel biomarkers, measured at baseline, to predict future nonhemorrhagic stroke in 7863 patients (17% women, median age 62 years) enrolled in the LIPID study, a double-blind, placebo-controlled trial of pravastatin in patients with prior CHD. Methods and Results: Over 6 years’ median follow-up, 310 patients had a nonhemorrhagic stroke. In Cox regression analysis for each novel biomarker, BNP, CRP, cystatin-C, D-dimer (each P <0.004) and troponin I ( P <0.02) were significant, but LP(a), midregional pro-adrenomedullin, and Lp-PLA2 activity were not. Treatment with pravastatin was an independent predictor of nonhemorrhagic stroke in each of the models (all P ≤0.05). There was no evidence that this effect was modified by the baseline level of any of the novel biomarkers. In the multivariable model, BNP, CRP and cystatin-C remained independent predictors for nonhemorrhagic stroke ( Table 1 ). The effect of pravastatin also remained significant, as were the traditional risk factors: age, prior stroke, prior hypertension, diabetes mellitus, atrial fibrillation, and systolic blood pressure. This multivariable model was used to create risk groups based on the upper and lower quartiles of the score ( Table 2 ). Conclusion: BNP, CRP, and cystatin-C levels significantly improve the identification of future nonhemorrhagic stroke in patients with known CHD. Relative treatment effects of pravastatin were similar across all risk levels.