ABSTRACT
BNP is a marker of systolic left ventricular dysfunction (LVSD) and heart failure. To assess BNP for the detection of diastolic dysfunction in the general population, we examined 1678 subjects within an age- and sex-stratified survey (MONICA Augsburg). BNP was measured using a commercially available RIA (Shionogi). BNP increased in subjects with diastolic dysfunction (mean 20.3+/-4.7 pg/ml vs. control 9.6+/-0.5 pg/ml, p<0.001), but to a lesser extent than in subjects with LV hypertrophy (LVH, mean 37.3+/-49.1 pg/ml, p<0.001 vs. control) or LVSD (mean 76.2+/-23.2 pg/ml, p<0.001 vs. control). Individuals with sole diastolic abnormality displayed BNP concentrations at the control level (mean 9.7+/-1.7 pg/ml). In univariate analysis, age, BMI, systolic blood pressure, left atrial size, LV mass index, diastolic dysfunction and EF displayed a significant correlation with BNP (p<0.001). However, LV mass index displaced diastolic dysfunction as a significant predictor of BNP in multivariate analysis. Upon ROC analysis, sensitivity and specificity for the detection of diastolic dysfunction by BNP were only 61% and 55%, respectively. Nevertheless, a normal BNP test virtually excluded the presence of diastolic dysfunction and concomitant LVH (NPV 99.9%). Increased BNP concentrations in subjects with diastolic dysfunction are strongly related to LVH. Population-wide screening for diastolic dysfunction with BNP cannot be recommended although a normal BNP test usually excludes diastolic dysfunction and LV hypertrophy.
