Objective: For differentiating heart failure (HF) with preserved ejection fraction (HFpEF) from HF with reduced EF (HFrEF), N-terminal prohormone brain natriuretic peptide (NT-proBNP) is less accurate. Decreased expression of microRNA-19b (miR-19b) is associated with increased cardiac-fibrosis. We aim to evaluate the value of miR-19b in diagnosing HFrEF patients. Method: We included 200 HF patients and 100 healthy controls. Intergroup comparisons of miR-19b were made and correlation between miR-19b and NT-proBNP was analysed. Diagnostic values of NT-proBNP and miR-19b for HF patients versus controls and HFrEF versus HFpEF were obtained by ROC analysis and described by area under curve (AUC), sensitivity and specificity. Results: HFrEF patients (0.87, 95% CI 0.37-1.45) had significantly lower miR-19b level than HFpEF group (1.32, 95% CI 0.63-2.51) and the controls (1.82, 95% CI 0.37-1.45) (both P < .001). There was a remarkable negative correlation between miR-19b and NT-proBNP (P < .001). The additional use of miR-19b did not improve the accuracy of NT-proBNP alone in diagnosing HF patients from the controls (both AUC = 0.98, 95%CI 0.97-0.99). However, as for distinguishing the HFpEF from HFrEF, miR-19b and NT-proBNP yielded a significantly higher AUC than NT-proBNP alone (0.85, 95% CI 0.80-0.90 vs. 0.66, 95% CI 0.58-0.74) (P < .001), and the sensitivity for diagnosing HFrEF was raised from 58% to 77% and the specificity from 75% to 79%. Conclusions: On top of NT-proBNP, miR-19b added the value in diagnosing HFrEF. But in view of satisfactory accuracy of NT-proBNP in predicting HF from the healthy volunteers, miR-19b did not provide incremental value.