Survival in Patients With Suspected Myocardial Infarction With Nonobstructive Coronary Arteries: A Comprehensive Systematic Review and Meta-Analysis From the MINOCA Global Collaboration

Background: Suspected myocardial infarction (MI) with nonobstructive coronary arteries (MINOCA) occurs in ≈5% to 10% of patients with MI referred for coronary angiography. The prognosis of these patients may differ to those with MI and obstructive coronary artery disease (MI-CAD) and those without a MI (patients without known history of MI [No-MI]). The primary objective of this study is to evaluate the 12-month all-cause mortality of patients with MINOCA. Methods: Using Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, the terms "MI," "nonobstructive," "angiography," and "prognosis" were searched in PubMed and Embase databases from inception to December 2018, including original, English language MINOCA studies with >100 consecutive patients. Publications with a heterogeneous cohort, unreported coronary stenosis, or exclusively focusing on MINOCA-mimicking conditions, were excluded. Unpublished data were obtained from the MINOCA Global Collaboration. Data were pooled and analyzed using Paule-Mandel, Hartung, Knapp, Sidik & Jonkman, or restricted maximum-likelihood random-effects meta-analysis methodology. Heterogeneity was assessed using Cochran's Q and I²statistics. The primary outcome was 12-month all-cause mortality in patients with MINOCA, with secondary comparisons to MI-CAD and No-MI. Results: The 23 eligible studies yielded 55 369 suspected MINOCA, 485 382 MI-CAD, and 33 074 No-MI. Pooled meta-analysis of 14 MINOCA studies accounting for 30 733 patients revealed an unadjusted 12-month all-cause mortality rate of 3.4% (95% CI, 2.6%-4.2%) and reinfarction (n=27 605; 10 studies) in 2.6% (95% CI, 1.7%-3.5%). MINOCA had a lower 12-month all-cause mortality than those with MI-CAD (3.3% [95% CI, 2.5%-4.1%] versus 5.6% [95% CI, 4.1%-7.0%]; odds ratio, 0.60 [95% CI, 0.52-0.70], P<0.001). In contrast, there was a statistically nonsignificant trend towards increased 12-month all-cause mortality in patients with MINOCA (2.6% [95% CI, 0%-5.9%]) compared with No-MI (0.7% [95% CI, 0.1%-1.3%]; odds ratio, 3.71 [95% CI, 0.58-23.61], P=0.09). Conclusions: In the largest contemporary MINOCA meta-analysis to date, patients with suspected MINOCA had a favorable prognosis compared with MI-CAD, but statistically nonsignificant trend toward worse outcomes compared to those with No-MI. Registration: URL: https://www.crd.york.ac.uk/PROSPERO/; Unique identifier: CRD42020145356.