Study population and design

This single‑center, retrospective study evaluated all patients with ACS who were admitted to the First Department of Cardiology of the Medical University of Warsaw, Poland between January 2023 and June 2023 and underwent coronary angiography. The patients were classified into 2 groups—MINOCA and OOMI—based on the findings from ICA and optical coherence tomography (OCT) imaging. MINOCA was defined as MI with no angiographic evidence of obstructive CAD, but with an ischemic mechanism responsible for the myocardial injury (nonischemic causes, such as myocarditis or takotsubo syndrome, were excluded).

Comprehensive data, including demographic information, clinical characteristics on admission, laboratory test results, as well as OCT findings, were meticulously collected from the hospital’s medical records system.

Ethics

Due to the retrospective study design, the medical ethical committee at the Medical University of Warsaw deemed that informed consent and formal ethical approval were not required. The study was conducted in accordance with the principles outlined in the Declaration of Helsinki.

N‑terminal pro–B‑type natriuretic peptide / troponin ratio

The NT‑proBNP/high‑sensitivity troponin I (hs‑TnI) ratio was calculated for every patient using the last measurements taken before ICA.

SYNTAX score

For patients with complete data available—including left main coronary artery disease status, sex, presence of chronic obstructive pulmonary disease (COPD), age, and left ventricular ejection fraction—SYNTAX scores I and II were calculated. The calculations were performed using the official SYNTAX score calculator available at https://syntaxscore.org.

Follow‑up end points

Major adverse cardiac events (MACEs) were defined as all‑cause mortality, nonfatal MI, stroke, cardiogenic shock, pulmonary embolism, and heart failure exacerbation. Additionally, we obtained information about rehospitalization, time to the first rehospitalization, and recurrent myocardial incidents (target lesion failure [TLF], target vessel failure [TVF], target vessel revascularization [TVR], target lesion revascularization [TLR]). The patients were followed for 6 months since the admission date or until death, based on hospital medical records.

Statistical analysis

Categorical variables are presented as absolute numbers with percentages and were compared using the χ² test. If the expected size of any group was smaller than or equal to 5, the Fisher test was performed. Continuous variables are presented as mean (SD) for normally distributed data and median with interquartile range (IQR) for non‑normally distributed variables. If the variables were assumed to be normally distributed, the t test was performed; otherwise, the Mann–Whitney test was used for data comparison. Receiver operating characteristic (ROC) curve analysis was carried out to assess the diagnostic performance of the NT‑proBNP/hs‑TnI ratio. The area under the curve (AUC) was also calculated and optimal cutoff value was determined using the Youden index. Logistic regression analysis was performed to calculate odds ratios (ORs) with 95% CIs, assessing the connection between clinical variables (NT‑proBNP/troponin ratio, total cholesterol [TC], low‑density lipoprotein cholesterol [LDL‑C], systolic blood pressure [SBP], and troponins) and the likelihood of MINOCA. A 2‑tailed P value below 0.05 was deemed significant. All analyses were conducted in R package, version 4.3.1 (R Foundation for Statistical Computing, Vienna, Austria).

Baseline clinical characteristics

The study population comprised 194 patients, of whom 151 (78%) had OOMI and 43 (22%) had MINOCA. Men accounted for a higher proportion of the OOMI group than of the MINOCA group (120/151 (79%) vs 24/43 (56%); P = 0.003). However, sex distribution among the MINOCA patients was almost identical (men vs women, 24 [56%] vs 19 [44%]). There were no significant age differences between the groups. Prevalence of comorbidities was also similar, except for the history of cancer (OOMI vs MINOCA, 10% vs 23%; P = 0.04), peripheral artery disease (PAD; OOMI vs MINOCA, 97% vs 9%; P <⁠0.001), and dyslipidemia (OOMI vs MINOCA, 49% vs 26%; P = 0.01; Table 1).

Clinical characteristics on admission

Heart rate (HR), diastolic blood pressure (DBP), ejection fraction (EF), oxygen saturation levels, New York Heart Association class, Canadian Cardiovascular Society score, and ST‑segment elevation MI (STEMI) frequency did not differ between the patients with obstructive CAD and those with MINOCA. However, SBP on admission was higher in the patients with MINOCA than those with OOMI (median [IQR], 154 [140–167] mm Hg vs 140 [126–158] mm Hg; P = 0.04). Additional data are presented in Supplementary material, Table S1.

Laboratory findings

The patients with OOMI had higher levels of TC, LDL‑C, and high‑sensitivity troponin I (hs‑TnI) than the individuals with MINOCA (median [IQR], 173 [136–221] mg/dl vs 147 [123–170] mg/dl; P = 0.02; 99 [76–137] mg/dl vs 67 [56–95] mg/dl; P = 0.003; and 0.834 [0.147–7.290] ng/ml vs 0.088 [0.019–1.285] ng/ml; P <⁠0.001, respectively; reference ranges, TC <⁠200 mgl/dl; LDL‑C <⁠115 mg/dl; hs‑TnI <⁠0.014 ng/ml). Other laboratory parameters, such as high‑density lipoprotein cholesterol, triglycerides, creatinine, estimated glomerular filtration rate, C‑reactive protein, glucose, NT‑proBNP, and glycated hemoglobin were comparable in both groups (Table 2). In the logistic regression analysis TC, LDL‑C, hs‑TnI, and SBP were not significant predictors of MINOCA.

N‑terminal pro–B‑type natriuretic peptide / troponin ratio

The NT‑proBNP/hs‑TnI ratio was markedly higher in the MINOCA patients than in those with OOMI (median [IQR], 6.53 [1.28–17.25] vs 1.04 [0.15–3.65]; P <⁠0.001; Table 3). ROC analysis demonstrated good discriminatory ability of the NT‑proBNP/hs‑TnI ratio with an AUC of 0.739. The optimal threshold of 5.56 yielded sensitivity of 61% (Figure 1). The NT‑proBNP/troponin ratio was positively associated with MINOCA in the logistic regression analysis (OR, 1.15; 95% CI, 1.04–1.34; P = 0.02; Table 4).

SYNTAX score

The SYNTAX score I was significantly higher in the patients with OOMI, as compared with those with MINOCA. The average SYNTAX score I for OOMI was 15, while for MINOCA, it was 1.7 (P <⁠0.001). ROC analysis demonstrated very good discriminatory ability of the SYNTAX score I (AUC = 0.938). An optimal threshold of 4.75 yielded sensitivity of 86% (Figure 2). The SYNTAX score I was inversely related to the likelihood of MINOCA (OR, 0.34; 95% CI, 0.141–0.552; P = 0.001). However, no significant differences were found between the 2 groups regarding the percutaneous coronary intervention (PCI) SYNTAX score II, PCI 4‑year mortality, coronary artery bypass grafting (CABG) SYNTAX score II, or CABG 4‑year mortality (Table 5).

Follow‑up

There were no significant differences in the incidence of rehospitalization, TLF, TVF, TVR, TLR, MACEs, and death in 6‑month follow‑up between the patients with OOMI and MINOCA. Additional data are presented in Supplementary material, Table S2.

Limitations

The retrospective design of this study introduces limitations related to data availability and accuracy. Due to the fact that it was conducted at a single center, the findings reflect a single patient population and may be influenced by center‑specific practices and protocols. Furthermore, we only included patients who underwent coronary angiography, which excludes potential MINOCA cases where this procedure was not performed. Another limitation of our study is the low number of available OCT images, as they were performed in only 5 patients. Additionally, we excluded individuals with final hospital diagnosis of takotsubo syndrome or myocarditis; however we did not collect information on further diagnostic testing, such as cardiac magnetic resonance imaging (MRI). Since MRI is recognized as a valuable tool in differentiating MINOCA, there is a possibility of misdiagnoses in some cases.

The timing of cardiac biomarker measurements and coronary angiography may have varied across patients due to the retrospective nature of the study, introducing variability. Additionally, the 6‑month follow‑up period limits the assessment of long‑term outcomes, and follow‑up data were confined to records from a single hospital, without access to external medical history.

Conclusions

Our study presents a detailed clinical profile of MINOCA patients. Compared with individuals with OOMI, MINOCA patients are more likely to be female, have a history of cancer, and present with elevated SBP on admission. In contrast, traditional cardiovascular risk factors, such as dyslipidemia, PAD, and elevated LDL‑C and TC levels, were more commonly associated with OOMI. The NT‑proBNP/troponin ratio emerged as a promising tool for early differentiation of MINOCA, with a proposed cutoff value of 5.56. However, prospective studies with larger sample sizes are needed to validate these findings, as well as to focus on the diagnostic approaches and long‑term outcomes of MINOCA to improve patient care.