Page 114 - Biomarkers for risk stratification and guidance in heart failure

Serial NT-proBNP measurements during and early after HF admission.
Prognostic Value of NT-proBNP Parameters in Both Treatment Arms of the PRIMA Study
Because in one-half of the patients included in the PRIMA study the treating physician was not blinded to the outpatient NT-proBNP concentration, knowledge of NT-proBNP might have influenced the decision whether to admit a patient or not. However, in multivariate analysis correcting for the reference model and randomization group, inpatient change in NT-proBNP, early outpatient change in NT-proBNP, and NT-proBNP concentration 1 month after hospital discharge remained independent prognostic factors (Supplemental Table 2). Moreover, in both treatment arms all 3 NT-proBNP parameters were of prognostic importance (Supplemental Figs. 1 and 2).
DISCUSSION
In this study, we evaluated the prognostic value of serial NT-proBNP measurements
during and early after an admission for acute HF. We demonstrated that the a) 5 inpatient change in NT-proBNP concentration, b) early outpatient change, and
c) absolute NT-proBNP concentration at 1 month after discharge were each independent prognosticators and together enabled accurate short- and long-
term outpatient risk assessment. Importantly, even small changes in the early
outpatient phase (ie, <30% change in NT-proBNP) had prognostic meaning.
Inpatient NT-proBNP Measurements
Although the natriuretic peptide concentration at admission for acute HF predicts inpatient mortality,10 its prognostic effect after discharge seems to be small. This is in sharp contrast to NT-proBNP concentration at discharge and changes in NT- proBNP during admission, which both seem to be better predictors of outcome. For example, Bettencourt et al4 demonstrated that a NT-proBNP level >6,779 pg/mL at hospital admission predicted a nonsignificant trend toward hazard of readmission or death, but the NT-proBNP concentration at discharge of 4,137 pg/mL was a much stronger predictor of hazard (log rank P for cumulative hospitalization-free survival: <.001). They furthermore found that inpatient decrease in NT-proBNP values of ≥30% was related to favorable outcome. In addition, Kubler et al11 demonstrated that the optimal cutoff value for inpatient decrease in NT-proBNP was 65%. A decrease in NT-proBNP concentration in
113