FGF-23 is associated with increased disease severity and early mortality in cardiogenic shock

Despite the increased use of percutaneous interventions, infarction-related cardiogenic shock (CS) is still associated with high mortality. Biomarkers might be helpful to identify patients at risk, and point towards novel therapeutic strategies in CS. The phosphaturic hormone fibroblast growth factor 23 (FGF-23) has recently been introduced as a predictor for mortality in patients with chronic systolic heart failure. However, its predictive role in CS has not been investigated so far.

FGF-23 was measured in 51 patients with CS. Eighteen patients with uncomplicated acute myocardial infarction (AMI) and 940 patients with stable coronary artery disease (CAD) undergoing elective coronary angiography included in a previous study served as control groups. Compared with patients with stable CAD, FGF-23 was profoundly elevated in patients with CS, but not in patients with uncomplicated AMI (CAD: 131.1 ± 9.5; AMI: 175.3 ± 57.2; CS: 1684.4 ± 591.7 rU/ml, p<0.0001 CS vs. CAD). In patients with CS, FGF-23 correlated significantly with the SAPS II score (r=0.461, p=0.0003) and NT-pro BNP levels (r=0.489, p=0.001). Patients were stratified as "survivors" and "non-survivors" according to their 28-day mortality. The overall 28-day-mortality-rate was 37%. Non-survivors of CS showed significantly higher FGF-23 levels compared with survivors (3260.1 ± 1514.7 vs. 847.9 ± 182.4 rU/ml, p=0.028). In the ROC curve analysis, FGF-23 levels predicted 28-day mortality (area under the curve (AUC) 0.686, p=0.028), and FGF-23 level of 1180 rU/ml was identified as optimal cut-off value. In a multivariate Cox proportional hazard model adjusted for gender, blood pressure, ejection fraction and levels of creatine kinase, FGF-23 levels above 1180 rU/ml significantly predicted 28-day mortality (hazard ratio (HR) 2.74, 95% CI 1.01-7.04, p=0.037).

In CS, a tremendous increase in FGF-23 occurs, and high levels of FGF-23 are associated with poor outcome.