Adding PIIINP to a prognostic model with clinical variables and NT-proBNP significantly improved the prediction of all-cause mortality in ADHF patients, increasing the AUC from 0.79 to 0.85 (P=0.024).
Cohort (n=217)
Does the combination of PIIINP and NT-proBNP improve the accuracy of survival prediction compared to NT-proBNP alone in patients with acutely decompensated heart failure?
The combination of the myocardial fibrosis biomarker PIIINP with NT-proBNP provides incremental prognostic value over NT-proBNP alone for predicting short-term all-cause mortality in patients with acutely decompensated heart failure.
Effect estimate: HR 1.40 (95% CI 1.08-1.81)
p-value: p=0.011
BACKGROUND: Soluble suppression of tumorigenesis-2 (sST2), Procollagen Type III N-Terminal Peptid (PIIINP) and N-terminal pro-B-type natriuretic peptide (NT-proBNP) have been verified their role in predicting survival in acutely decompensated heart failure (ADHF). However, whether their combination could improve more specific and sensitive prognostic information than NT-proBNP alone remains unclear. METHODS: This was a prospective study, in which 217 ADHF patients at admission were enrolled from November 2018 and August 2019 (mean age 66.18 years ± 13.60, 63.98% male). The blood samples were collected to measure the concentrations of NT-proBNP, sST2 and PIIINP in the first 24 h of hospitalizations. All-cause mortality was registered for all patients after they were discharge over a median period of 339 days. RESULTS: In univariate Cox analysis, the three biomarkers were predictive of short-term mortality of ADHF patients. After adjusted for some clinical variables including age, admission systolic blood pressure, peripheral edema on admission, history of chronic obstructive pulmonary disease, admission sodium < 135 mmol/L, admission hemoglobin, NT-proBNP, sST2 and PIIINP was significantly associated with the poor outcome (hazard ratio HR 1.32, 95% confidence interval CI 1.14-1.53, P < 0.01; HR 1.21, 95% CI 1.03-1.43, P = 0.020; HR 1.40, 95% CI 1.08-1.81, P = 0.011). After added with Log2 PIIINP, but not Log2 sST2, the area under the curves (AUC) in the model of clinical variables and Log2 NT-proBNP could increase from 0.79 to 0.85 (95% CI 0.0071-0.10, P = 0.024). Furthermore, compared with the model of clinical variables, Log2 NT-proBNP, the improvement in the prognostic model of clinical variables, Log2 NT-proBNP and Log2 PIIINP had statistical significance net reclassification improvement (NRI) 0.31, P = 0.018; integrated discrimination improvement (IDI) 0.068, P < 0.01. CONCLUSIONS: NT-proBNP, sST2 and PIIINP are independent prognostic factors for all-cause mortality in ADHF patients. Furthermore, the combination of NT-proBNP and PIIINP may provide incremental prognostic value over NT-proBNP in the survival of ADHF patients.
Yao et al. (Fri,) conducted a cohort in acutely decompensated heart failure (ADHF) (n=217). NT-proBNP, sST2 and PIIINP was evaluated on all-cause mortality (HR 1.40, 95% CI 1.08-1.81, p=0.011). Adding PIIINP to a prognostic model with clinical variables and NT-proBNP significantly improved the prediction of all-cause mortality in ADHF patients, increasing the AUC from 0.79 to 0.85 (P=0.024).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: