Diagnostic and prognostic value of circulating microRNA-21 in heart failure: A systematic review and meta-analysis

Figure 1.. PRISMA flow diagram of included studies.

Figure 2.. Forest plots summarizing the diagnostic accuracy of circulating miRNA-21 for heart failure (HF): (A) sensitivity and (B) specificity. Point estimates from six studies (seven independent datasets; Zhang et al., 2017 reported CS and PV cohorts separately) are represented as red squares with 95% confidence intervals (horizontal bars). The pooled summary estimate is illustrated by the diamond (width = 95% CI), while the dashed vertical line indicates the pooled value. Measures of between-study heterogeneity, including Cochran’s Q and I2, are provided below each panel.

Figure 3.. Diagnostic accuracy and clinical utility of circulating miRNA-21 for heart failure (HF). (A) SROC plot showing individual study estimates (circles), the bivariate summary point (diamond), fitted SROC curve, and 95% confidence/prediction regions; (B) Deeks’ funnel plot for publication bias (P ═ 0.60); (C) Fagan’s nomogram translating likelihood ratios to post-test probability (pre-test 25%: post-test 76% for LR+ ═ 10; 2% for LR-- ═ 0.06).

Figure 4.. Forest plot depicting miRNA-21 expression levels in relation to HF events. Individual squares represent study-specific fold changes in circulating miR-21 expression among HF patients compared to non-HF controls. Horizontal lines indicate 95% confidence intervals, with the size of each square proportional to the inverse-variance weight of the respective study. The diamond shape represents the pooled effect estimate derived from a fixed-effects inverse-variance model, illustrating significantly elevated circulating miR-21 levels in HF patients (overall fold change 1.61; 95% CI 1.46–1.78; P < 0.001). The analysis reveals low-to-moderate heterogeneity between studies, with an I2 value of approximately 46%. Abbreviations: HF: Heart failure; miRNA-21: MicroRNA-21; CI: Confidence interval.

Figure 5.. Subgroup analysis of circulating miRNA-21 expression levels in HF. Forest plot shows fold changes (squares) with 95% confidence intervals across subgroups defined by onset (ADHF, CHF, both), LVEF phenotype (HFrEF, HFpEF, both), sample type (plasma vs serum), quantification method, and continent; the diamond indicates the overall pooled fold change from a fixed-effects inverse-variance model. Abbreviations: HF: Heart failure; ADHF: Acute decompensated heart failure; CHF: Chronic heart failure; HFrEF: Heart failure with reduced ejection fraction; HFpEF: Heart failure with preserved ejection fraction; LVEF: Left ventricular ejection fraction.

Figure 6.. (A) Forest plot illustrating the relationship between miRNA-21 and NYHA classification. (B) Forest plot evaluating the association between miRNA-21 and heart failure-related hospitalizations. (C) Forest plot examining the relationship between miRNA-21 and heart failure-related cardiovascular mortality.