Does high sensitivity troponin add prognostic value to validated risk scores to predict in-hospital mortality in patients with acute heart failure?

Abstract

Troponin elevation correlates with an increased short and long-term mortality in patients with acute decompensated heart failure (AHF). However, it has not been included in the development of multiple validated predictive models of mortality. We aim to determine whether the addition of high-sensitivity troponin T (hs-TnT) to clinical risk scores improves the prediction of in-hospital mortality in patients with AHF. A retrospective analysis of a prospective and consecutive cohort was performed. Adult patients hospitalized between 2015 and 2019 with a primary diagnosis of AHF were included. Hs-TnT was measured on admission. OPTIMIZE-HF, GWTG-HF, and ADHERE risks score were calculated for each patient. The primary endpoint was all-cause in-hospital mortality. Discrimination of isolated hs-TnT and the risk scores with and without the addition of hs-TnT were evaluated using the area under the ROC curve (AUC-ROC). A subanalysis was performed according to left ventricular ejection fraction (LVEF). Of 712 patients, 562 (79%) had hs-TnT measurement upon admission, and was elevated in 91%. In-hospital mortality was 8.7% (n = 49). The AUC-ROC was 0.70 (95% CI 0.63-0.77) for isolated hs-TnT, and 0.80 (0.74-0.87), 0.79 (0.72 -0.86) and 0.79 (0.71-0.86) for the OPTIMIZE-HF, GWTG-HF and ADHERE scores, respectively. The addition of hs-TnT to the models did not increase the AUC: 0.72 (0.66-0.79) for the OPTIMIZE-HF + hs-TnT score (difference between AUC - 0.08 p = 0.04), 0.74 (0.68-0.80) for GWTG-HF (difference between AUC-0.04, p = 0.2) and 0.7 (0.63-0.77) for ADHERE (difference between AUC - 0.085 p = 0.07). The models presented good calibration (p > 0.05). In the sub-analysis, no differences were found between risk scores with the addition of hs-TnT in the population with LVEF < 40% and ≥ 40%. Elevated hs-TnT on admission was frequent and its incorporation into the validated risk scores did not prove an incremental prognostic benefit in patients hospitalized for AHF, regardless of LVEF. Isolated hs-TnT had a modest ability to predict hospital mortality. Additional prospective studies are needed to validate these findings.

Keywords: Heart failure; Mortality; Prognosis; Troponin.