P2616Soluble urokinase plasminogen activator receptor (suPAR) predicts 1 year mortality in patients with acute breathlessness

Abstract

Listen

Translate article

Abstract Aim Soluble urokinase plasminogen activator receptor (suPAR) is a pleotropic receptor, capable of orchestrating plaque vulnerability and vascular immune dysfunction. Its concentration in circulation may reflect on CVD associated burden. We examined the prognostic ability of suPAR to predict death in patients suspicious of ADHF. Methods suPAR measurements were undertaken with a CE-marked ELISA (ViroGates) in 444 patients presenting to hospital with the primary complaint of breathlessness to our hospital ED. A second sample at 12–48 hours post ED admission was available for 378 patients. Standard biochemistry analytes; MR-proADM (BRAHMS), NTproBNP and hsTnT (both Roche) were also measured. Statistical assessment was made using SPSS v25 (IBM), with all biomarkers treated as continuous variables and presented as median [interquartile range (IQR)]. Group comparisons were made by Mann-Whitney U test. The singular or combined clinical performances of suPAR, NT-proBNP, hsTnT and MR-proADM were assessed using receiver operator curve (ROC) area under the curve (AUC) (Z-scores) and Cox hazard regression (log-values) analyses. P-value <0.05 was considered significant. Results In the breathless cohort [median age 72 yrs (IQR: 62–81, 43% female)], 35.1% had ADHF and 94/444 patients died within the 1st yr of ED presentation. In those who died within this 1st yr, median suPAR concentrations at both time-points; 5.2 ng/mL (IQR: 2.8–5.5) vs. 5.1 ng/mL (IQR: 1.7–2.3) were higher than those who did not die (3.5 ng/mL; IQR: 2.7–5.1) (P<0.0001). Plasma suPAR, for both time points, respectively, could predict death at 30d (n=22, ROC-AUC = 0.77 and 0.76), 90d (n=41, ROC-AUC = 0.77 and 0.75) and 1 yr (ROC-AUC = 0.73 and 0.72). Improvement in 90d mortality prediction was achieved with the inclusion of suPAR in models; for e.g. using ED values, for NTproBNP; ROC-AUC of 0.67 increased to 0.71, for hsTnT; ROC-AUC of 0.69 to 0.76, and for MR-proADM, ROC-AUC of 0.72 to 0.75. Both suPAR time-points (ROC-AUC 0.70) could predict 1 yr new heart failure (HF) (n=68) but did not assist in improving HF prediction when used in combination with NT-proBNP, hsTnT or MR-proADM. After adjustment for conventional risk factors, Cox hazard regression analyses however revealed suPAR as the only biomarker capable of predicting 1 yr mortality (P=0.005) with hazard ratios of 2.8 (CI: 1.4–5.9) and 3.5 (CI: 1.1–11.3) for the ED and inpatient time-points, respectively. In terms of best window for prognostic assessment, suPAR concentrations at ED, outperformed the inpatient concentrations as the superior time-point in predicting 1 yr mortality. Conclusion suPAR exhibits excellent prognostic ability in mortality prediction, proving better than NTproBNP, hsTnT and MR-proADM in acutely breathless patients. The usage of suPAR in complementary with current candidate cardiac biomarkers could dramatically improve the prognostic tools available to guide risk management in HF. Acknowledgement/Funding New Zealand Heart Foundation, Health Research Council of New Zealand