Prognostic impact of secondary prevention treatment following myocardial infarction with non-obstructive coronary arteries: Bayesian versus frequentist meta-analysis

Abstract Background Myocardial infarction with non-obstructive coronary arteries (MINOCA) is a clinical entity with a plethora of causes and pathophysiologic mechanisms. Traditional secondary prevention medications may not offer benefit for patients with MINOCA, calling into question the routine use of such therapies in this population. Material and methods Literature search was conducted in MEDLINE (PubMed), Scopus by ELSEVIER and Cochrane Central Register of Controlled Trials (CENTRAL) databases until 08 March 2022. Random-effects frequentist and hierarchical Bayesian meta-analyses were performed to assess the prognostic impact of secondary prevention medications (β-blockers, RAAS inhibitors, statins, DAPT, ASA and P2Y12 inhibitors) on the occurrence of major adverse cardiovascular events (MACE) and all-cause mortality in MINOCA patients. Sensitivity analyses accounted for the effect of: i. small sample size, ii. limited study follow-up period, and iii. different ethnicity. Results Of 2,569 articles initially retrieved, 9 observational studies were deemed eligible for this analysis, encompassing a total of 14,003 patients suffering from MINOCA (mean follow-up range from 6 to 90 months, mean age = 64.9±10.4 years).Administration of ASA, DAPT and P2Y12 inhibitors was not significantly linked with MACE [pooled adjusted hazard ratios (aHRs) = 0.94, 0.88, and 0.97, with 95% confidence intervals (CIs): 0.56–1.59, 0.47–1.65, and 0.562–1.53, respectively]. RAAS inhibitors and statins were independently associated with lower risk of MACE (pooled aHRs = 0.61 and 0.55 with 95% Cis: 0.46–0.82, and 0.34–0.90, respectively) and all-cause mortality (pooled aHRs = 0.58 and 0.55 with 95% Cis: 0.37–0.91, and 0.43–0.72, respectively). B-blockers were associated with significantly lower risk of all-cause death according to the frequentist analysis (pooled aHR = 0.80, 95% CI: 0.67–0.96) and with non-significantly lower risk of MACE (pooled aHR = 0.81, 95% CI: 0.65–1.01) [Figure 1]. The results deriving from the Bayesian analyses were generally consistent with the frequentist analyses, with the exceptions of the Bayesian analyses on statin and β-blocker effects, which were affected by the pre-test probabilities set herein [Figure 2]. The overall heterogeneity observed in our analyses was relatively small, and there was not substantial publication bias. Furthermore, the performed sensitivity analyses supported the robustness of the observed outcomes. Conclusion In this meta-analysis, it appeared that predominantly RAAS inhibitors and statins across secondary prevention medications were consistently associated with lower risk of MACE or all-cause death during the follow-up period of patients that experienced MINOCA. Funding Acknowledgement Type of funding sources: None.