OBJECTIVETo evaluate the diagnostic value of NT-proBNP in aortic stenosis (AS) patients with heart failure.

METHODWe measured the whole venous blood of NT-proBNP with enzyme linked immuno sorbent assay (Biomedica, Vienna, Austria) in 40 AS patients with heart failure and 76 normal subjects and assessed the diagnostic value of NT-proBNP for heart failure.

RESULTSNT-proBNP levels were significantly higher in AS patients with heart failure compared to controls (P < 0.01). The level of NT-proBNP increased in proportion to the increase of NYHA functional classes (all P < 0.01). The level of NT-proBNP was similar between compensated heart failure group and control group (P > 0.05) and significantly (8 times) increased in decompensated heart failure group (P < 0.01 vs. control group). NT-proBNP level was also significantly higher in LVEDD > 50 mm group than that in LVEDD ≤ 50 mm group (P < 0.05) and in LVEF ≤ 60% group than that in LVEF > 60% group (P < 0.01). Patients with atrial fibrillation also had higher NT-proBNP levels compared to those with sinus rhythm (P < 0.05). The NT-proBNP value of 1360 ng/L was determined as the best cutoff value for the diagnosis of AS patients with heart failure (AUC = 0.762, P < 0.01) and decompensated heart failure (AUC = 0.997, P < 0.01), the sensitivity, specificity and accuracy were 67.50% and 100.00%, 96.05% and 96.05% and 86.21% and 95.83%, respectively. Log (NT-proBNP) was positively related with NYHA functional class and negatively related with LVEF in univariate analysis and multiple regression analyses (P < 0.05). NT-proBNP was independent correlative with NYHA functional class and LVEF.

CONCLUSIONSNT-proBNP has a fairly good diagnostic potential for the identification of AS patients with heart failure. The accuracy is 86.21% for the diagnosis of AS patients with heart failure and 95.83% for decompensated heart failure with the diagnostic cutoff value of 1360 ng/L.