Pre-discharge and early post-discharge troponin elevation among patients hospitalized for heart failure with reduced ejection fraction: findings from the ASTRONAUT trial

Abstract

Aims: Troponin levels are commonly elevated among patients hospitalized for heart failure (HF), but the prevalence and prognostic significance of early post-discharge troponin elevation are unclear. This study sought to describe the frequency and prognostic value of pre-discharge and post-discharge troponin elevation, including persistent troponin elevation from the inpatient to outpatient settings. Methods and results: The ASTRONAUT trial (NCT00894387; http://www.clinicaltrials.gov) enrolled hospitalized HF patients with ejection fraction ≤40% and measured troponin I prior to discharge (i.e. study baseline) and at 1-month follow-up in a core laboratory (elevation defined as >0.04 ng/mL). This analysis included 1469 (91.0%) patients with pre-discharge troponin data. Overall, 41.5% and 29.9% of patients had elevated pre-discharge [median: 0.09 ng/mL; interquartile range (IQR): 0.06–0.19 ng/mL] and 1-month (median: 0.09 ng/mL; IQR: 0.06–0.15 ng/mL) troponin levels, respectively. Among patients with pre-discharge troponin elevation, 60.4% had persistent elevation at 1 month. After adjustment, pre-discharge troponin elevation was not associated with 12-month clinical outcomes. In contrast, 1-month troponin elevation was independently predictive of increased all-cause mortality [hazard ratio (HR) 1.59, 95% confidence interval (CI) 1.18–2.13] and cardiovascular mortality or HF hospitalization (HR 1.28, 95% CI 1.03–1.58) at 12 months. Associations between 1-month troponin elevation and outcomes were similar among patients with newly elevated (i.e. normal pre-discharge) and persistently elevated levels (interaction P ≥ 0.16). The prognostic value of 1-month troponin elevation for 12-month mortality was driven by a pronounced association among patients with coronary artery disease (interaction P = 0.009). Conclusions: In this hospitalized HF population, troponin I elevation was common during index hospitalization and at 1-month follow-up. Elevated troponin I level at 1 month, but not pre-discharge, was independently predictive of increased clinical events at 12 months. Early post-discharge troponin I measurement may offer a practical means of risk stratification and should be investigated as a therapeutic target.