Vericiguat and cardiovascular outcomes in heart failure revisited: a bayesian reanalysis of 12,877 patients

Introduction Heart failure (HF) remains a major global health burden. Vericiguat, a soluble guanylate cyclase stimulator, has been evaluated across HF phenotypes. However, trial results have been heterogeneous, leaving uncertainty regarding its clinical benefit. A Bayesian analysis would enable probabilistic estimation of treatment effects and more clinically interpretable quantification of benefit and harm, thereby providing a clearer and more decision-relevant appraisal of the available data. Purpose To provide probabilistic, clinically actionable estimates of vericiguat’s cardiovascular benefits and risks in HF through Bayesian meta-analysis of randomized controlled trials (RCTs). Methods We systematically searched PubMed, Embase, and the Cochrane Library for RCTs comparing vericiguat with placebo in patients with HF. A Bayesian random-effects meta-analysis was conducted, and binary outcomes were synthesized using a binomial–normal hierarchical model to estimate posterior distributions of log odds ratios (log ORs) with corresponding 95% credible intervals (CrIs). Posterior probabilities of treatment effects were calculated to quantify the likelihood and certainty of benefit or harm. Results Five RCTs comprising 12,877 patients met inclusion criteria. The mean age was 68 years, and the left ventricular ejection fraction ranged from 29% to 56%. In the Bayesian meta-analysis, vericiguat was associated with an estimated OR for HF hospitalization of 0.91 (95% CrI, 0.72–1.13), with an 86.3% probability of any benefit (OR ≤ 1.0), a 9.3% probability of a moderate effect (OR ≤ 0.80), and a 4.1% probability of a large effect (OR ≤ 0.75). For all-cause mortality, the OR was 0.94 (95% CrI, 0.75–1.24), corresponding to a 74.6% probability of any benefit, a 5.8% probability of a moderate effect, and a 1.9% probability of a large reduction. For cardiovascular death, the estimated OR was 0.92 (95% CrI, 0.70–1.25), with a 79.9% probability of any benefit, a 10.6% probability of a moderate effect, and a 4.2% probability of a large effect. Conclusion This Bayesian reanalysis suggests that vericiguat was associated with a moderate probability of benefit across HF hospitalization and mortality outcomes, although the likelihood of a large clinically meaningful effect was low. These findings support a modest overall treatment effect in broad HF populations and underscore the importance of identifying patient subgroups most likely to derive greater benefit.