Long-term prognostic potential of microRNA-150-5p in optimally treated heart failure patients with reduced ejection fraction: a pilot study.

ABSTRACT

BACKGROUND: In a previous study, we found that miR-150-5p was specifically downregulated in patients with advanced heart failure (HF). Here, we investigated the long-term prognostic potential of miR-150-5p. METHODS: We studied optimally-treated HF outpatients with reduced ejection fraction. The primary outcome comprised the composite of death, urgent heart transplantation (HT) and ventricular assist device (VAD) implantation within 30 months. We used recursive partitioning analysis to identify the optimal log miR-150-5p cut-off. The association of log miR-150-5p with the primary outcome was examined using Cox regression analysis. We used the Meta-Analysis Global Group in Chronic Heart Failure (MAGGIC) risk score for adjustment in multivariable analysis. Finally, we compared the global fit of three models (MAGGIC score + miR-150-5p, MAGGIC score + NT-proBNP, and NT-proBNP + miR-150-5p) using Akaike Information Criterion. RESULTS: Recursive partitioning analysis identified the value of -2.22 as the optimal cut-off for log miR-150-5p. Thirty-month survival free of urgent HT/VAD implantation was 31% among the patients with log miR-150-5p<-2.22 and 86% among those with log miR-150-5p>-2.22. Crude hazard ratio (HR) of the primary outcome for log miR-150-5p expression level <-2.22 was 6.70 (95% CI 2.31-19.38; P<0.001). After adjusting for the MAGGIC score in multivariable analysis, the HR was 4.40 (95% CI 1.52-12.77; P=0.006). Adding log miR-150-5p to the MAGGIC score led to an increase of 0.047 in C-index. The model combining miR-150-5p and MAGGIC score had a 73% likelihood of representing the best-fit model of those evaluated. CONCLUSIONS: Our data generate the hypothesis that miR-150-5p may represent a novel risk marker in HF with reduced ejection fraction.