Effectiveness of vericiguat in heart failure across ejection fraction phenotypes: a comprehensive meta-analytic trial sequential analysis and meta-regression analysis

Figures

• 

  Figure 1.. PRISMA 2020 flow diagram for systematic reviews, which includes searches of databases and registers.

• 

  Figure 2.. Forest plot illustrating the effectiveness of vericiguat compared to control in heart failure across ejection fraction phenotypes. This forest plot visualizes the pooled RR with 95% confidence intervals (CI) for key outcomes – hospitalization for heart failure, all-cause mortality, cardiovascular mortality, and the primary composite outcome (death from cardiovascular causes or first hospitalization) – derived from a meta-analysis of randomized controlled trials. Additional statistical metrics, including sample size (N), P-value, Q statistic, I2 (heterogeneity), and D2, are annotated for each outcome, highlighting the robustness and variability of the results across HFrEF and HFpEF patients.

• 

  Figure 3.. (A) Trial sequential analysis of vericiguat versus control for hospitalization for heart failure, showing the cumulative Z-curve (blue) with conventional 5% thresholds (purple dashed lines) and 5% symmetric O’Brien-Fleming boundaries (red lines) across six trials (cumulative sample size: 16,866; required information size: 29,929). (B) Simplified trial sequential analysis of vericiguat versus control for hospitalization for heart failure, focusing on the cumulative Z-curve and 5% symmetric O’Brien-Fleming boundaries (red lines) with a cumulative sample size of 16,866 patients, compared to the required information size of 29,929. (C) Trial sequential analysis of vericiguat versus control for all-cause mortality, depicting the cumulative Z-curve (blue) with conventional 5% thresholds and 5% symmetric O’Brien-Fleming boundaries (red lines) across six trials (cumulative sample size: 6678). (D) Simplified trial sequential analysis of vericiguat versus control for all-cause mortality, highlighting the cumulative Z-curve (blue) and 5% symmetric O’Brien-Fleming boundaries (red lines) with a cumulative sample size of 6678 patients.

• 

  Figure 4.. (A) Trial sequential analysis of vericiguat versus control for cardiovascular mortality, showing the cumulative Z-curve (blue) with conventional 5% thresholds (purple dashed lines) and 5% symmetric O’Brien-Fleming boundaries (red lines) across six trials (cumulative sample size: 17,179; required information size: 18,4229). (B) Simplified trial sequential analysis of vericiguat versus control for cardiovascular mortality, highlighting the cumulative Z-curve (blue) and 5% symmetric O’Brien-Fleming boundaries (red lines) with a cumulative sample size of 17,179 patients, compared to the required information size of 184,229. (C) Trial sequential analysis of vericiguat versus control for the primary composite outcome (death from cardiovascular causes or first hospitalization for heart failure), depicting the cumulative Z-curve (blue) with conventional 5% thresholds (purple dashed lines) and 5% symmetric O’Brien-Fleming boundaries (red lines) across five trials (cumulative sample size: 16,685; required information size: 48,016). (D) Simplified trial sequential analysis of vericiguat versus control for the primary composite outcome (death from cardiovascular causes or first…

• 

  Figure 5.. Forest plots from the sensitivity analysis evaluating the effectiveness of vericiguat versus placebo in heart failure with reduced ejection fraction phenotype. (A) Primary composite outcome, showing a significant risk reduction (random-effects RR 0.90, 95% CI 0.85–0.96). (B) All-cause mortality (random-effects RR 0.93, 95% CI 0.87–1.00). (C) Cardiovascular mortality (random-effects RR 0.91, 95% CI 0.85–0.98). *-(D) Hospitalization for heart failure (random-effects RR 0.91, 95% CI 0.88–0.94).

• 

  Figure 6.. (A) Residual influence diagnostics from the meta-regression model for vericiguat in heart failure, displaying residual standard deviation (rstudent), Cook’s distance (cook.D), covariance ratio (cov.R), tau-squared influence (tau2.Del), Q statistic influence (Q.Del), hat values (hat), and weight influence (weight) across included studies, assessing model fit and influence. (B) Regression coefficients in the meta-regression model for vericiguat in heart failure, showing the coefficient values for intercept, age, population, sex, and subgroup (HFpEF vs. HFrEF), with bars indicating the magnitude and direction of the effect on the outcome. (C) Bubble plot of the meta-regression for vericiguat in heart failure, illustrating the effect size (log odds ratios) against predicted values (fitted), with bubble size representing study weight, highlighting the relationship between covariates and outcomes across preserved and reduced ejection fraction phenotypes.

• 

  Figure 7.. Risk of bias assessment across included studies using the Cochrane Risk of Bias Tool 2 (RoB 2).

• 

  Figure 8.. Funnel plot for assessing publication bias. Funnel plot with trim-and-fill adjustment evaluating publication bias across studies, plotting the standard error against the effect size (log odds ratio), with a symmetrical distribution indicating no significant publication bias, despite there being fewer than 10 studies.