Introduction: Patients presenting with suspected acute myocardial infarction (AMI) are stratified according to their risk of having AMI using high-sensitivity cardiac troponin (hs-cTn)-based diagnostic algorithms. Among patients where AMI is excluded, many are yet at high risk for cardiovascular events and death. However, this group is incompletely understood, and precise risk stratification represents an unmet clinical need.

Purpose: We aimed to investigate whether two blood biomarkers, soluble urokinase-type plasminogen activator receptor (suPAR) and hs-cTnI aid in the identification of patients with acute chest pain but no AMI at high-risk of all-cause mortality and to compare their prognostic capacity with the Global Registry of Acute Coronary Events (GRACE) scoring tool.

Methods: We enrolled patients with suspected AMI presenting to the emergency department. In a sub-group comprising 1,312 patients we determined the concentrations of suPAR and hs-cTnI. The final diagnoses (AMI vs. non-AMI diagnoses) were adjudicated by two independent cardiologists. Patients with AMI were excluded from this analyses. We used Kaplan-Meier plots and cox regression models with adjustment for age, sex, body-mass-index, diabetes, smoking, hyperlipoproteinemia and systolic blood pressure to investigate associations with 1-year all-cause mortality. Receiver operator characteristics curves were used to compare the GRACE score and biomarkers to predict 6-month mortality.

Results: Overall, 1,007 (76.8%) patients had no AMI at index presentation. Comparing the first and fourth quartile, crude (Figure 1) and adjusted analyses revealed plasma concentrations of suPAR (adjusted hazard ratio [adj-HR] 4.14 [95% CI 1.13, 15.16], p = 0.032) and hs-cTnI (adj-HR 7.66 [95% CI 1.56, 37.56], p = 0.012) being significantly associated with higher risk of death at one year. After additional adjustment for hs-cTnI, suPAR remained significantly associated with death (adj-HR 3.96 [95% CI 1.07, 14.72], p = 0.04). Comparing the GRACE score and suPAR, the area under the ROC curve (AUC) for 6-month mortality was similar (GRACE score AUC 0.77 (95% CI 0.67, 0.86), suPAR AUC 0.75 (95% CI 0.64, 0.87), p = 0.75). When hs-cTnI was compared to the GRACE score it showed comparable predictive accuracy as well (AUC 0.75 (95% CI 0.66, 0.84), p=0.73).

Conclusion: SuPAR and hs-cTnI independently predict 1-year all-cause mortality beyond traditional cardiovascular risk-factors in acute chest pain patients in whom AMI has been excluded. Both biomarkers alone showed similar predictive capacity as the well-established GRACE score.

Figure 1: Kaplan-Meier plots for patients without acute myocardial infarction stratified by quartiles of suPAR or hs-cTnI.