Abstract 12094: Usefulness of Each and Combination of N-terminal Pro-Brain Natriuretic Peptide and High Sensitive Troponin T Levels for Predicting Mortality and Heart Failure After the Discharge of Acute Myocardial Infarction

Abstract

Background: Each N-terminal pro-Brain Natriuretic Peptide (NT-proBNP) and high sensitive Troponin T (hsTnT) is reported to be useful marker for predicting cardiovascular events. However, the combination of NT-proBNP and hsTnT levels at discharge of acute myocardial infarction (AMI) for predicting future cardiac events has not been fully investigated.

Methods and Results: We investigated clinical impact of each and combined effect of the NT-proBNP and hsTnT levels collected at convalescent stage on future cardiovascular events in 1,682 AMI patients (mean age, 64.1 y.o.; Male, 77.5%) for a median follow-up periods of 1,382 days. Death and heart failure (HF) requiring hospitalization were observed in 84 and 70 patients, respectively. A multivariate Cox regression analysis adjusted with patient demographics revealed that Log2(NT-proBNP) was a significant predictor of the composite endpoint of death from any cause and readmission for HF (HR 1.44, 95%CI 1.25 to 1.66, p<0.001). ROC curve analysis revealed that Area Under Curve (AUC) of NT-proBNP for the composite endpoint was 0.740 with the sensitivity of 0.681 and the specificity of 0.688 if employed as the cut off value of 822 pg/ml. Besides AUC for mortality was 0.724, with the sensitivity of 0.726 and specificity of 0.617, respectively. As for hsTnT, AUC for mortality was 0.640 with the sensitivity of 0.678 and specificity of 0.585 if put cut off value of 0.029 ng/ml, suggesting that hsTnT itself did not surpass NTproBNP in predicting mortality or HF. Evaluating the combined effect of NT-proBNP and hsTnT for mortality, AUC of the combination of NT-proBNP and hsTnT for mortality was 0.761 with the sensitivity of 0.680 and the specificity of 0.736. Furthermore, five year mortality in patients with double positive markers was significantly higher than than those without (12.1% vs 3.2%, p<0.001), suggesting the usefulness of the combination of NT-proBNP and hsTnT for discriminating high risk patients of mortality after MI.

Conclusion: NT-proBNP rather than hs TnT collected during convalescent stage appears to be a useful predictive marker of adverse events after AMI. The combination of NT-proBNP and hsTnT might be useful for discriminating high risk patients of mortality after AMI.