The Diagnostic Value Of Brain Natriuretic Peptide For Early Heart Failure After Acute Myocardial Infarction

Objectives:This study was designed to study the dynamic changes of brain natriuretic peptide(BNP)concentration over time within 48 hours after the onset of acute myocardial infarction(AMI)in patients accepting emergency percutaneous coronary intervention(PCI),including primary PCI in patients with ST-segment elevation acute myocardial infarction and emergency PCI in patients with non-ST-elevation acute myocardial infarction,and to analyze the correlation among BNP,Killip classification and left ventricular ejection fraction(LVEF),so as to evaluate the diagnostic value of BNP in heart failure in patients with AMI in the early phase.Methods:This study enrolled consecutively AMI patients admitted to the department of cardiology in Guangzhou General Hospital of Guangzhou Military Command,completing emergency PCI within 12 hours after the onset of symptom.Plasma BNP concentration and clinical cardiac function classification were attained bedside within 1 hour after PCI and 12,20,24 and 48 hours after the onset of symptom respectively.The echocardiography was carried out and cardiac function indicators was determined at 12 hours and 20?24 hours after the onset of symptom.All of the subjects were grouped by the following two different grouping motheds.According to whether heart failure was diagnosed or not(Killip class?II at any time point),the patients were divided into heart-failure group and non-heart-failure group.Another grouping mothed was according to whether BNP concentration was consistent with and Killip classification or not in evaluating cardiac function,these patients were divided into consistent group and inconsistent group.Inconsistent groups were subdivided into false positive group and false negative group.The patients with Killip class I but BNP>400pg/ml defined as false positive group(without clinical symptoms of heart failure but BNP increased)when those with Killip class>II but BNP<100pg/ml defined as false negative group(with clinical symptoms of heart failure but normal BNP concentration).Results:1.A total of 134 patients were enrolled in this study.Echocardiography was performed in 64 cases.Mean total ischemia time was 5.6±2.4 hours after the onset of symptom.Mean first time to detect BNP concentration was 6.3±2.3 hours after the onset of symptom(the average time was nearly 6 hours after the onset of symptom in order to be consistent with the subsequent detection time point and easy to understand,hereinafter referred to as "6h after the onset of symptom").There were 42 cases belonged to heart-failure group and 92 cases belonged to non-heart-failure group.Compared with non-heart-failure group,heart failure group was older(61.0 ± 12.9 vs 54.4± 11.5 years old,P=0.004)and more frequently had diabetes(42.9%vs 23.9%,P =0.026),pulmonary infection(31.0%vs 13.0%,P =0.014)and atrial fibrillation(11.9%vs 1.1%,P=0.018).2.The concentration of BNP both in the heart failure group and the non-heart-failure group increased first and then decreased within 48 hours after the onset of symptom,and reached the peak at 20?24 hours after the onset of symptom.The BNP concentration in the heart failure group were significantly higher than that in the non-heart-failure group at each time point(P<0.05).Rank Logistic regression analysis showed that gender,total ischemia time,Killip classification,and location of myocardial infarction were associated with the peak concentration of BNP(P<0.05).3.Spearman correlation analysis showed there was a weak positive correlation between BNP concentration and Killip classification within 48 hours after the onset of symptom(6 hours after onset:r = 0.23,12 hours after onset:r = 0.43,20 hours after onset:r = 0.46,24 hours after onset:r = 0.51,48 hours after onset:r = 0.56,all P<0.05).There were 37 cases(27.6%)in consistent group,including 32 cases(23.9%)in the false positive group,5 cases(3.7%)in the false negative group.The false positive group was distributed at different time points within 48 hours after onset and the false negative group was mainly concentrated within 12 hours.Logistic regression analysis showed that body mass index(BMI)(OR= 0.836,P ? 0.014)was negative related to the result of false positive,and dopamine utilization(OR=6.583,P=0.047)was positive related to the result of false negative.4.Using Killip class ? ? as the gold standard of heart failure diagnosis,the sensitivity of BNP ? 400pg/ml to diagnose early heart failure after AMI was 28.6%,47.1%,56.3%,21.4%and 9.7%,respectively at 6,12,20,24 and 48 hours from AMI onset,the specificity of BNP<100pg/ml to exclude early heart failure early after AMI was 83.3%?50.4%?18.6%?20.4%?34.5%,respectively at above time points.Receiver operating characteristic(ROC)curve analysis showed that the cutoff value of BNP at 12 hours from onset predicting heart failure within 48 hours after AMI onset was 135.7pg/ml.The area under curve(AUC)was 0.866(95%CI:0.785-0.947,P<0.001),the sensitivity was 88.2%and the specificity was 71.8%in prediction of early heart failure after AMI which was much better than that with the cutoff valure of 400pg/ml or 100 pg/ml.5.Spearman correlation analysis showed that there was no correlation between LVEF and BNP concentration and Killip classification at 12h after the onset of symptom(P>0.05),whereas there was a weak negative correlation between LVEF and BNP concentration(r =-0.3,P<0.05)and Killip classification(r =-0.2,P<0.05)at 20-24h after the onset of symptom.Conclusions:1.BNP concentrations showed a tendency to increase first and then discrease within 48h after the onset of symptom in AMI patients treated with emergency PCI.The peak concentration appeared at 20-24h after the onset of symptom.The plasma BNP concentration in patients with heart failure was significantly higher than that in patients without heart failure,but the BNP dynamic changes were consistent with the patients with or without heart failure.2.In the early stage after AMI(within 48 hours from onset),BNP>400pg/ml as the cutoff value to diagnose heart falure had low sensitivity and weak ability to detect heart failure and resulted in higher false negative rate.BNP<100pg/ml as the cutoff value had low specificity and weak ability to identify non-heart-failure patients which resulted in higher false positive rate.3.The correlation between LVEF and BNP and Killip classification within 24 hours were weak and the value of assessing cardiac function was limited.It's better not to use LVEF within 24h after the onset of symptom for cardiac function assessment.Instead,clinical manifestations and BNP dynamic changes should be combined for comprehensive assessments.4.The cutoff value of>135.7pg/ml at 12 hours after the onset of symptom is more sensitive and more specific to predict early heart failure after AMI(within 48 hours after onset of symptom)which was more accurative than the cutoff value of 400pg/ml or 100pg/ml recommended by heart failure guidelines.