Prognostic value of GDF-15 in patients with suspected myocardial infarction

Background Growth differentiation factor 15 (GDF-15) is a circulating biomarker induced by tissue damage and inflammatory processes. Its prognostic implications has been evaluated for several years, mainly in the general population, but also diseased patients. Prior data suggests a strong predictive value for mortality and cardiovascular events. While prior studies have investigated GDF-15 in patient with confirmed acute coronary syndrome or myocardial infarction (M), there is currently limited data on the prognostic value of GDF-15 in unselected patients with suspected MI. However, reliable longer-term risk stratification in these patients is of upmost importance. Objective Our aim was to evaluate the predictive value of GDF-15 within an all-comer cohort of patients with suspected MI. Methods We consecutively enrolled patients presenting with symptoms suggestive of MI to the emergency department of large tertiary hospital in Germany. GDF-15 was measured in all patients batchwise using frozen samples collected at time of ED presentation. Follow-up was performed via structured telephone interview, assessing the endpoints of all-cause mortality and major adverse cardiac event (MACE) containing cardiovascular death, MI, revascularization and cardiac rehospitalization. Kaplan-Meier curves were created for each tertile of GDF-15 concentration, Further, we performed cox-regression, adjusted for age, sex, diabetes, smoking, hypertension, hyperlipoproteinemia, history of coronary artery disease, and history of heart failure with the lowest tertile as reference. Results In total, data on 3´174 individuals with median age of 63 years and 35.9% being females were available for analyses. Median follow-up time was 5.2 (5.1, 5.2) years for mortality and 4.9 (4.8, 5.0) years for MACE. Overall mortality rate was 19.0% (462) and 35.7% (902) patients experienced MACE. Patients with GDF-15 concentrations in the highest tertile had a strikingly worse outcome with a 6-year mortality of 42.6% (362) and a rate of MACE of 51.9% (417) compared to 1.7% (13) and 21.2% (186) in the lowest tertile (Figure 1). Unadjusted cox regression showed a hazard ratio (HR) of 33.8 (19.5, 58.8) for all-cause mortality and 2.9 (2.4, 3.4) for MACE for the highest tertile of GDF-15 compared to the lowest. After adjustment for cardiovascular risk-factors and pre-existing cardiovascular disease, HR for all cause-mortality was 10.2 (5.7, 18.5) for the highest GDF-15 tertile while, regarding MACE, no independent predictive value could be found (HR 1.2 [1.0, 1.5]) (Table 1). Conclusion In patients with suspected MI, elevated GDF-15 levels are a very strong and independent predictor for all-cause mortality within 6 years, while the incremental value in predicting MACE was limited.Figure 1:Kaplan-Meier curves Table 1:Cox-regression