Addition of N-terminal pro-brain natriuretic peptide to the Global Registry of Acute Coronary Events risk stratification to predict outcome in non-ST-segment elevation acute coronary syndrome

Zhang Ning; Liu Wenxian

doi:10.3760/cma.j.issn.1673-4777.2014.03.007

Abstract：Objective To build a composite score based on the Global Registry of Acute Coronary Events (GRACE) score and N-terminal pro-brain natriuretic peptide (NT-proBNP) concentrations to predict outcome in patients with non-ST-segment elevation acute coronary syndrome (NSTE-ACS). Methods Patients with NSTE-ACS in Beijing Anzhen Hospital affiliated to capital medical university, a composite score including the GRACE score and NT-proBNP concentrations was first randomly developed in a retrospective cohort of 409 patients with NSTE-ACS and then validated in a prediction model of other 231 patients. The mean follow- up time in a retrospective cohort were (774±217) days, and in a prediction model were (706±231)days. The primary end point was the composite of MACE, defined as cardiogenic deaths, myocardial infarction, readmission for heart failure. Results The patients were reclassified by the composite score, 105 patients were in low risk group, 209 patients were in medium risk group, and 95 patients were in high risk group. End points were reached in 26 patients (6.6%). The lgNT-proBNP in patients with NSTE-ACS had positive correlation with their GRACE risk score (r=0.507, P<0.01); The under-ROC curve area of GRACE risk score and lgNT-proBNP were 0.807 and 0.798 respectively. The composite score could be obtained as follows: GRACE+20×lgNT-proBNP+15. The patients would be reclassified by the composite score, <135 was low risk group, 135-170 was medium risk group, and >170 was high risk group. 10 patients would be reclassified at high risk using the composite score despite being classified at low risk using the GRACE score alone. Alternatively, 7 patients would be reclassified at medium risk, while being classified high risk with the GRACE score alone. 8 patients would be reclassified at low risk using the composite score despite being classified at high risk using the GRACE score alone. Finally, 2 patients while being classified medium risk of reached the end points, that was would be reclassified at high risk. 6.5% of the population in prediction model reached the end points. The use of the composite score increased the accuracy of the GRACE score, with an increase in the under-ROC curve area from 0.748 to 0.762. Conclusion Both NT-proBNP concentration and GRACE score are independently associated with outcome. The comprehensive risk score, which includes NT-proBNP concentration and the GRACE risk score, might improve the accuracy of NSTE-ACS risk stratification in clinical practice.