Abstract 19167: Can Variation of BNP Within the Normal Range Predict Mortality and Ischemic Events in a Secondary Prevention Population? Results From the Smart Cohort Prospective Registry Study
It is unclear whether variation of BNP levels within the normal range can predict long term mortality and cardiac ischemic events in secondary prevention population without heart failure.
Methods and Results: Baseline BNP levels were measured in 2115 subjects (mean age 61±10 years, 81% male) with stable coronary artery disease (CAD) recruited in an outpatient clinic. During a mean follow-up of 4.0 years, the primary endpoint of cardiovascular (CV) mortality occurred in 145 (6.9%) patients. Total mortality occurred in 207 (9.8%) patients. Cardiac ischemic events (fatal and non-fatal acute MI, PCI and CABG) occurred in 410 (19.4%) patients. The low mortality rate (2%/year), the low median BNP level (12 pmol/L, IQR 6-24 pmol/L) and the low median hsCRP level (2 mg/L, IQR 1-4 mg/L) reflect the low risk profile of the population. BNP levels were positively associated with age (p=0.0001), CRP (p=0.01) and creatinine (p=0.0001) and inversely correlated with BMI (p=0.0001). They were higher in females (p=0.007), in patients with hypertension (p=0.006) or using betablockers (0.04) or diuretics (p=0.0001). Univariate Cox survival analysis showed that higher BNP (5 pmol/L increase) levels were associated with higher CV mortality (Chi2=101.9; HR=1.06[1.05-1.08]; p=0.0001), total mortality (Chi2=99.4; HR=1.06[1.05-1.07]; p=0.0001) and higher rate of cardiac ischemic events (Chi2=65.3; HR=1.05[1.04-1.07]; p=0.0001). Higher BNP values (above vs below median) were associated with a higher actuarial CV mortality (10.1% vs 2.3%), total mortality (14.9% vs 4.1%), and a higher actuarial rate of cardiac ischemic events (20.1% vs 8.2%) at 4 years. In multivariable Cox analysis, BNP (5 pmol/L increase) was a strong and independent predictor of CV mortality (Chi2=9.9; HR=1.03[1.01-1.05]; p=0.001), total mortality (Chi2=8.5; HR=1.03[1.01-1.04]; p=0.003) and cardiac ischemic events (Chi2=21.0; HR=2.00[1.50-2.68]; p=0.0001).
Conclusion: Variation of Plasma BNP within the normal range in outpatients with stable CAD is an independent predictor of long-term CV mortality, total mortality and cardiac ischemic events. Measurement of BNP levels might help in detecting higher-risk patients among apparently stable relatively low risk secondary prevention populations.
- © 2010 by American Heart Association, Inc.