By Van Selby, MD

Assistant Professor of Medicine, University of California, San Francisco, Cardiology Division, Advanced Heart Failure Section

Dr. Selby reports no financial relationships relevant to this field of study.

SYNOPSIS: Among patients with acute decompensated heart failure and preserved ejection fraction, elevated troponin is associated with worse in-hospital outcomes and long-term survival, independent of other predictors.

SOURCE: Pandey A, Golwala H, Sheng S, et al. Factors associated with and prognostic implications of cardiac troponin in decompensated heart failure with preserved ejection fraction: Findings from the American Heart Association Get With The Guidelines-Heart Failure Program. JAMA Cardiol 2016;Dec 28. [Epub ahead of print].

Serum troponin is a well-established biomarker in both ischemic heart disease and heart failure with reduced ejection fraction (HFrEF). The prognostic significance of elevated troponin in patients presenting with decompensated heart failure with preserved ejection fraction (HFpEF) is not well studied.

In a retrospective analysis of the Get With The Guidelines-HF registry, Pandey et al evaluated all participants admitted for decompensated HFpEF (defined as an ejection fraction 50%) who exhibited at least one troponin level (either troponin T or I) measured from January 2009-December 2014. The primary outcome was in-hospital, all-cause mortality. Secondary outcomes included post-discharge mortality and readmission rates.

Of 96,769 patients hospitalized for decompensated HFpEF, 34,233 (35.0%) submitted to troponin level measurements and were included in the analysis. Of those, 7,732 (22.6%) demonstrated elevated troponin. Predictors of elevated troponin included higher serum creatinine, black race, older age, and ischemic heart disease. In adjusted analyses, elevated troponin was associated with increased in-hospital mortality (odds ratio [OR], 2.19; P < 0.001), longer than four-day hospital stay (OR, 1.38; P < 0.001), and eventual discharge to home (OR, 0.65; P < 0.001). Post-discharge outcomes also were worse among those with elevated troponin, with increased 30-day (OR, 1.59; P < 0.001) and one-year mortality (OR, 1.35; P > 0.001). The prognostic significance of troponin was independent of serum B-type natriuretic peptide (BNP) level and remained significant among patients without coronary artery disease. The authors concluded that troponin elevation is associated with poor in-hospital and long-term clinical outcomes among patients hospitalized for decompensated HFpEF, suggesting a role for early troponin assessment in this patient population.


HFpEF is a heterogenous disease. Recent studies have attempted to categorize HFpEF into distinct subgroups based on clinical characteristics, pathophysiology, and outcomes. Hopefully, this work will lead to individualized management of HFpEF based on a patient’s particular phenotype. For now, this phenotypic diversity has made it difficult to identify effective treatments for HFpEF. Similarly, there are few tools for prognostication in HFpEF. For this reason, the findings of Pandey et al are important. Serum troponin appears to be highly prognostic in acute HFpEF, regardless of the exact phenotype, making it a potentially useful test that is easy to implement in routine clinical care.

Current practice guidelines assign a Class I recommendation to measuring serum troponin levels in patients admitted for decompensated heart failure. In this study, only 35% of patients submitted to a troponin level check at any time during hospitalization, meaning the recommendation is not widely followed in practice. This study increases the evidence base supporting this recommendation and expands the demonstrated prognostic utility of troponin to include patients with preserved EF. It is also important to note the prognostic utility of troponin was independent of BNP, which is more commonly used in the acute heart failure setting. Checking troponin levels could help clinicians identify acute HFpEF patients who require more aggressive monitoring and treatment during hospitalization, as well as closer follow-up after discharge.

There are important limitations to this study. Unmeasured confounding is a significant concern in a retrospective, registry-based study. The study authors only could evaluate patients in whom troponin was measured, and these patients may have differed from the broader acute HFpEF population in unmeasured ways. Also, it is possible some troponin elevations were actually due to acute coronary syndromes (although these patients should have been excluded from this registry). Therefore, some of the increased morbidity and mortality in patients with elevated troponin may have been related to ischemic disease rather than HFpEF.

In light of these findings and the current Class I recommendation to measure troponin levels in patients hospitalized for acute heart failure, is it time to start checking troponin in all patients presenting with decompensated HFpEF? The authors concluded their findings “suggest a role of early troponin assessment as an important risk stratification tool during the initial evaluation” of acute HFpEF. Although better early risk stratification theoretically may lead to improved patient care and outcomes, it is important to remember this study did not specifically demonstrate the utility of this strategy. Routine measurement of troponin will increase costs and potentially could lead to increased use of diagnostic testing such as coronary angiography in patients found to exhibit elevated troponin. Whether this strategy ultimately will improve outcomes remains to be seen. For now, measurement of serum troponin in acute HFpEF is appropriate when improved risk stratification would influence clinical decision-making.