Multiple circulating microRNAs improve prognostication in subjects with stable heart failure

Abstract

Background and aim: Certain circulating microRNAs have been proposed as novel biomarkers in heart failure (HF). Current data mostly derives from relatively small cohorts. Accordingly, we directly compared these different circulating microRNAs in a large cohort of patients with stable HF and long-term follow-up. Methods: The study included 834 patients: mean age, 68.1±12.7 years; main etiology of HF ischemic heart disease, 43.8%; NYHA II/III (68%/24%), and mean LVEF, 35.4±13.6%. We measured 14 microRNAs in plasma samples by standard Q-PCR. Results: During a median follow-up of 2.79 years, 328 patients died. Circulating levels of miR 423-5p, miR-129-5p, miR 22\_3p and miR 320 all differed highly significantly between HF decedents versus survivors (p<0.001 for each), while circulating levels of miR-133a and miR-378 differed with p-values <0.05. miR-208a, miR-622 and miR 1254 did not differ significantly. Next, we compared these circulating microRNAs in a multivariable analysis including established prognosticators of HF like age, hemoglobin, serum creatinin and NTproBNP. In this multivariable analysis, combination of different microRNAs provided independent significant prognostication, with miR 22\_3p and miR423-5p providing most significant independent prognostic information. Conculsions: This large study shows that multiple circulating microRNAs provide additional prognostic information in a multivariable model beyond established biomarkers like NTproBNP. Recently, new platforms have emerged to clinically measure multiple circulating microRNAs in an affordable, quick and reliable manner. This opens the exciting possibility to measure a mini-microRNA profile in daily clinical care as a novel way to prognosticate HF.