MicroRNAs are associated with cardiac biomarkers, cardiac structure and function and incident outcomes in heart failure.

Abstract

The association between microRNAs (miRNAs) and established cardiac biomarkers is largely unknown. We aimed to measure the association between plasma miRNAs and N-terminal pro-B-type natriuretic peptide (NT-proBNP), cardiac troponin I, soluble urokinase-type plasminogen activator receptor (suPAR), and galectin-3 with cardiac structure and function and clinical outcomes. We quantified 32 plasma miRNAs using the FirePlex miRNA assay and measured biomarkers in 139 individuals with symptomatic heart failure (HF). We used principal component (PC) analysis and linear regression to evaluate the association between miRNAs and biomarkers with ventricular size and function by echocardiography and Cox modelling for the incidence of a first composite event of HF hospitalization, heart transplant, left ventricular assist device implant, or death. The mean (standard deviation) age at baseline was 64.3 (12.4) years, 33 (24%) were female, and 122 (88%) were White. A total of 45 events occurred over a median follow-up of 368 (interquartile range 234, 494) days. Baseline NT-proBNP (β=-2.0; P=0.001) and miRNA PC2 (β=2.6; P=0.002) were associated with baseline left ventricular ejection fraction. NT-proBNP (β=20.6; P=0.0004), suPAR (β=-39.6; P=0.005), and PC4 (β=21.1; P=0.02) were associated with baseline left ventricular end-diastolic volumes. NT-proBNP [hazard ratio (HR) 1.67, 95% confidence interval (CI) 1.28-2.18, P=0.0002], galectin-3 (HR 2.02, 95% CI 1.05-3.91, P=0.036), PC3 (HR 1.75, 95% CI 1.23-2.49, P=0.002), and PC4 (HR 1.67, 95% CI 1.1-2.52, P=0.016) were independently associated with incident events. Biomarkers and miRNA PCs are associated with cardiac structure and function and incident cardiovascular outcomes. Combining information from miRNAs provides prognostic information beyond biomarkers in HF.