As a complement to other biomarkers, the cytokine may aid in the decision to send patients home from the ED or admit them.

Levels of growth differentiation factor-15 (GDF-15), a biomarker of oxidative stress and inflammation, are significantly associated with short-term adverse outcomes among adults presenting to the emergency department (ED) with suspected ACS, new data show.

After accounting for demographic and clinical factors, as well as concentrations of cardiac troponin and NT-proBNP, both 30- and 90-day risks of a composite of all-cause death or acute MI increased along with higher levels of GDF-15, researchers led by Bryn Mumma, MD (UC Davis Department of Emergency Medicine, Sacramento, CA), report in a paper published online this week in the Journal of the American Heart Association.

That suggests a possible role for GDF-15—whose assays have not been approved by the US Food and Drug Administration for cardiac indications—in risk-stratifying patients who do not have an acute MI, they say, and in helping to determine who can go home and who needs to be admitted for further testing.

Screening for GDF-15 is complementary to that of high-sensitivity troponin, Mumma told TCTMD. “The first step is to get the troponin to rule out an acute heart attack, and then once we’ve done that, that’s where GDF-15 comes in,” she said, noting that the novel biomarker may help enhance tools like the HEART score that estimate a patient’s risk of having a major adverse cardiac event in the next 30 days.

Low-risk patients typically get sent home from the ED and high-risk patients generally are admitted, leaving intermediate-risk patients “in this gray area,” Mumma said. “If we have GDF-15 in addition to everything else that can bump them into the low-risk bucket, then they can potentially go home without having to stay in the hospital for a cardiac stress test or a cardiac catheterization. Whereas maybe if their GDF-15 level is really high, it [might be] worth keeping them in the hospital for more testing.”

Complementing Other Biomarkers

GDF-15, Mumma et al explain, “is part of a proinflammatory cytokine and oxidative stress-dependent signaling response to ischemia, reperfusion injury, pressure overload, and heart failure.” Prior studies have shown that concentrations of the biomarker are associated with outcomes assessed at 6 months or longer in patients with ACS and/or acute chest pain. There is, however, a dearth of data from US cohorts and from studies looking at potential links to shorter-term outcomes, which may be relevant for physicians deciding what to do with patients presenting with acute chest pain.

To help fill those gaps, the investigators performed a secondary analysis of the STOP-CP observational cohort study, which enrolled patients with suspected acute MI presenting to eight US emergency departments. The analysis included 1,428 patients (median age 58 years; 46% women) who had serial measures of high-sensitivity troponin T, NT-proBNP, and GDF-15 available. One-quarter had a history of coronary artery disease, and nearly all (95%) had had chest pain lasting more than 3 hours at the time of presentation. Troponin levels were below the 99th percentile upper reference limit in 73%.

Median GDF-15 level was 1,233 ng/L at baseline and 1,363 ng/L at 3 hours. Baseline level was 2,038 ng/L in patients with an acute MI at the initial presentation and 767 ng/L in those with a troponin level below the limit of quantification. The highest median GDF-15 level observed at baseline was 3,288 ng/L in the patients who either died or had an acute MI within the first 30 days.

The first step is to get the troponin to rule out an acute heart attack, and then once we’ve done that, that’s where GDF-15 comes in. Bryn Mumma

Overall, all-cause death or acute MI occurred in 12% of the cohort during the index visit, with another 1.5% having one of those events by 30 days and yet another 1.9% between 31 and 90 days.

After adjustment for age, sex, number of cardiac risk factors, ischemic ECG findings, and levels of high-sensitivity troponin T and NT-proBNP in multiple logistic regression models, higher concentrations of GDF-15, by standard deviation (SD), were associated with greater odds of all-cause death or acute MI at 30 days (adjusted OR 1.27; 95% CI 1.06-1.54) and 90 days (adjusted OR 1.55; 95% CI 1.26-1.94), but not during the index visit (adjusted OR 1.00; 95% CI 0.83-1.16).

GDF-15 was associated with acute MI alone only at 90 days, whereas a relationship with all-cause death alone was observed at both 30 and 90 days.

In the patients who had normal or undetectable high-sensitivity troponin T levels at presentation, the group most likely to be considered for discharge from the emergency department, GDF-15 was associated with the composite endpoint at 90 days (adjusted OR 3.08; 95% CI 1.47-6.30).

Increased high-sensitivity troponin T levels correlated with risk of the composite of all-cause death or acute MI only at 30 days (adjusted OR 1.18; 95% CI 1.01-1.38).

“Together, these findings suggest that GDF-15 may have a unique prognostic role in forecasting short-term cardiac risk following ED or hospital discharge,” Mumma et al write. “This role is distinct from that of high-sensitivity cardiac troponin T, which is used primarily in diagnosing acute MI during the index visit.”

Additional Research Needed

The researchers calculated that the ideal cutoff value of GDF-15 for ruling out all-cause death or acute MI at 30 days in this cohort was 1,371 ng/L—this gave a negative predictive value (NPV) of 99.4% in patients younger than 65 and 99.1% in those 65 and older. Similar NPVs were seen for outcomes at 90 days.

However, Mumma said, additional research into the optimal threshold is needed.

“One thing that would be really helpful if we were going to put it into practice is having either a single cut point below which it’s considered normal or above which it’s considered positive, or at least being able to provide more guidance to clinicians on how to interpret a GDF-15 value or how to even potentially assign a point value to it so that we can add GDF-15 onto our existing cardiac risk-stratification schemas,” she said, noting the HEART and EDACS scores are the most commonly used.

Mumma hopes to study GDF-15 in a larger cohort, noting that the current analysis is “really the first large study of GDF-15 in a US population.”

Because the GDF-15 assay is already available in many medical centers for different populations, if the findings of this and other studies are validated in future research and optimal cut points are identified, using the biomarker for cardiac risk stratification “could be implemented relatively quickly,” Mumma added.