Value of brain natriuretic peptide after acute myocardial infarction.

OBJECTIVE

Brain natriuretic peptide (BNP) is secreted predominantly from the ventricles in response to increased wall stress, which is known to be one of the major forces driving left ventricular (LV) remodeling. In this prospective study, we evaluated value of BNP levels in acute myocardial infarction (MI) patients for the prediction of heart failure during one year of follow-up.

METHODS

Seventy-four patients with a first ST-elevation MI were examined prospectively after 5 days and 1 month with echocardiography and blood samples for BNP were obtained. Clinical events were recorded during 12 months of follow-up. Multivariate linear regression analysis was used to analyze the value of different baseline characteristics as independent predictors of LV ejection fraction (LVEF) </= 40% and clinical heart failure. Diagnostic ability of BNP to detect LVEF </= 40% and heart failure was evaluated with receiver operating characteristic (ROC) curves.

RESULTS

Brain natriuretic peptide levels were higher in patients developing symptomatic heart failure during follow up irrespective of presence of LVEF </=40% (68.9+/-52.5 vs 21.4+/-18.4, p=0.003, for baseline BNP and 79.3+/-35.8 pg/ml vs. 22.9+/-15.8 pg/ml for one month BNP, p<0.001). Regression analysis including pain duration, peak creatine kinase-MB levels, MI localization, baseline BNP levels and baseline LV volumes yielded that baseline BNP was the most powerful predictor of one-year LVEF </=40% (Beta: 0.376, p=0.004). Multivariate analyses, testing for independent predictive information of pain duration, peak creatine kinase-MB, MI localization, thrombolytic therapy or primary percutaneous intervention, fifth day and one month LV volumes, LVEF and BNP levels, for development of clinical heart failure, showed that one month BNP was the single significant predictor (Beta: 0.675, p<0.001). There was a negative correlation between BNP levels and LVEF (r=-0.599, p<0.001, for baseline BNP level). Higher BNP levels were associated with greater increase in LV end-systolic (r= 0.531, p< 0.001) and end-diastolic volumes (r= 0.385, p= 0.001) during one year of follow-up. A baseline BNP level of >39 pg/ml identified LVEF </= 40% at one year with a sensitivity of 72.7% and specificity of 91.9% (OR=30.4, 95% CI, 6.1-152.3, p<0.001, AUC=0.852). A BNP level <39 pg/ml also increased the risk of clinical heart failure (for baseline BNP sensitivity: 60.0%, specificity 89.1%, OR=12.2; 95% CI, 2.7-54.1, p=0.001 and for one month BNP sensitivity: 80.0%, specificity 85.9%, OR=24.4; 95% CI, 4.5-134.1, p<0.001).

CONCLUSIONS

High level of BNP is a powerful marker of LV systolic dysfunction and poor prognosis after MI. Increased BNP levels are associated with progressive ventricular dilatation and development of clinical heart failure.