Improved prognostic power in acute breathlessness using NT-proBNP plus novel biomarker combinations
J Chew-Harris, A J Langdon, W S Kuan, I Ibrahim, J Pickering, S P Chan, C Frampton, O W Liew, R Troughton, P Adamson, L L Tan, W Lin, S B S Ooi, C J Pemberton, A M RichardsAbstract
Background
NT-proBNP reflects varying patterns of cardiac stress, rising sharply during rapid decompensation in acute heart failure (AHF) and more modestly in chronic heart failure (CHF) due to sustained myocardial strain. Concentrations are typically highest in AHF. We evaluated whether adding novel biomarkers could enhance prognostic assessment in patients presenting with acute dyspnoea, including those with underlying HF.
Methods
Patients with acute dyspnoea were prospectively recruited in New Zealand (n=416) and Singapore (n=451). Admission concentrations of suPAR (ViroGates), MR-proADM (Kryptor), BioADM (Sphingotec), and pro-enkephalin (p-ENK; Sphingotec) were measured, and prognostic performance assessed individually or combined with NT-proBNP (Roche). NT-proBNP distributions were examined across non-HF, CHF, all-HF, and de novo AHF groups. The prognosis for the composite 1-year endpoint (HF readmission and/or death) was evaluated using ROC-AUC and adjusted survival analyses incorporating three NT-proBNP tiers. Novel biomarkers were analysed using single threshold cut-offs. Analyses determined whether biomarker combinations improved risk prediction beyond NT-proBNP alone.
Results
Of 867 patients (median age 63 [IQR 51–75]; 64% male; 16.1% first-presentation HF; 39% total HF), 214 (24.7%) experienced the composite endpoint within 1 year. Elevation of biomarker levels was most pronounced in HF patients incurring the endpoint. However, NT-proBNP overlapped substantially between HF subtypes and outcome groups (Figure 1A), limiting discrimination (AUC 0.67 for AHF) (Figure 1B). Three-tier NT-proBNP survival analysis demonstrated progressive endpoint risk. Compared with 400 pg/mL (Tier 1, n=27/400 events), ≥400–2499 pg/mL (Tier 2, n=62/220) and ≥2500 pg/mL (Tier 3, n=125/247) conferred adjusted hazard ratios (aHR) of 3.0 (95%CI: 1.8–4.7) and 5.2 (3.3–8.3), respectively (Figure 1C). Adding suPAR or MR-proADM to NT-proBNP improved discrimination in HF (ΔAUC ≥0.04, P = 0.006) (Figure 2A). Combining NT-proBNP tiers with novel biomarker thresholds strengthened the prediction of HF/death at 1 year. Notably, suPAR remained independently predictive across all NT-proBNP strata; suPAR >3.8ng/mL with Tier 2 NT-proBNP yielded an aHR of 5.4 (3.1-9.7) and with Tier 3 NT-proBNP 3 to an aHR of 9.4 (5.3-16.8) (Figure 2B).
Conclusion
In acute undifferentiated breathlessness, integrating novel biomarkers, particularly suPAR with NT-proBNP, substantially enhances risk stratification. This combined biomarker strategy may improve clinical assessment across the continuum of non-HF dyspnoea, CHF and AHF.