Introduction Coronary embolism (CE) is a relatively rare but significant cause of non-atherosclerotic acute myocardial infarction (AMI), representing about 3% of all AMI cases, though it is likely underdiagnosed. CE is associated with worse clinical outcomes than traditional atherosclerotic AMI, showing increased rates of cardiac death and cerebrovascular events. Atrial fibrillation (AF) is the primary underlying cause of CE, cited in 28%-73% of cases across large series. Despite AF's central role, comprehensive data detailing the clinical, biochemical, echocardiographic, angiographic characteristics, and outcomes of CE specifically linked to AF remains limited. This study aims to address this knowledge gap by retrospectively evaluating patients with AF-related CE (AF CE) and non-AF CE, characterizing their differences and identifying outcome predictors.

Methods From January 2008 to Dicember 2024, consecutive patients admitted to a tertiary care cardiology unit and meeting both the Fourth Universal Definition of AMI and either definite or probable CE per Shibata's criteria were retrospectively included. Shibata's classification involves major and minor angiographic and clinical criteria to establish the likelihood of CE. Definite CE is diagnosed with a combination of major and minor criteria, while probable CE requires fewer criteria. Cases with evidence of atherosclerotic thrombus, prior revascularization, coronary anomalies like ectasia, spontaneous coronary artery dissection, or stress cardiomyopathy were excluded.

Coronary Embolism Definition

The Shibata criteria define CE based on:

Major criteria: angiographic embolism signs unrelated to atherosclerosis, multisite CE, systemic embolism excluding left ventricular thrombus from STEMI.

Minor criteria: non-significant coronary stenosis (<25%), embolic source identified by imaging, and risk factors like AF, dilated cardiomyopathy, rheumatic valve disease, prosthetic valves, recent cardiac surgery, coagulation disorders, patent foramen ovale, or atrial septal defect.

Coronary Arteriography All patients underwent invasive coronary angiography, independently reviewed by two specialists. Stenoses were visually assessed according to recognized grading systems.

Echocardiography Transthoracic echocardiography was performed following contemporary guidelines, evaluating left atrial and ventricular size and function, and left atrial strain using speckle tracking with standardized software (Philips).

Cardiac Magnetic Resonance (CMR) CMR was performed in patients initially diagnosed with MINOCA (Myocardial Infarction with Non-Obstructive Coronary Arteries) based on standard protocols, evaluating for subendocardial or transmural late gadolinium enhancement and focal myocardial edema to support CE diagnosis.

Atrial Fibrillation Definition AF was diagnosed as per ESC guidelines, requiring over 30 seconds of rhythm without P waves and irregular RR intervals. Both history of AF and new-onset AF during hospitalization or follow-up qualified for classification into the AF CE group.

Outcomes

Two types of outcomes were evaluated:

In-hospital outcomes: composite of heart failure, cardiogenic shock, ventricular arrhythmias, stroke, or death.

Long-term outcomes: composite of reinfarction, systemic embolism, stroke, cardiac or all-cause mortality.

Ethics The study protocol was approved by an independent ethics committee, with a waiver of informed consent, justified under the principles of the Declaration of Helsinki (reference n° 3318-0000257).

Statistical Analysis Categorical variables were presented as frequencies and percentages, continuous variables as means ± standard deviations. AF CE and non-AF CE groups were compared using Chi-square or Fisher's exact tests for qualitative variables, and parametric or non-parametric tests for quantitative variables depending on distribution normality (Kolmogorov-Smirnov test). For in-hospital outcomes, univariable logistic regression was performed to identify associated variables, followed by multivariable stepwise forward logistic regression for variables with p < 0.1. Long-term outcomes were assessed using univariable Cox regression and multivariable stepwise forward Cox regression for significant variables. Kaplan-Meier survival analyses and log-rank tests evaluated long-term outcome differences. SPSS version 20 (IBM) was used, with p < 0.05 considered statistically significant.

INTRODUCTION

Coronary embolism (CE) represents a distinct and under-recognized cause of acute myocardial infarction (AMI) that does not involve atherosclerotic processes. Despite its relatively low reported prevalence of around 3%, CE remains clinically significant due to its association with adverse cardiovascular outcomes, including a higher incidence of cardiac-related mortality and cerebrovascular accidents. A wide spectrum of conditions can precipitate CE events. Among these, atrial fibrillation (AF) stands out as the most frequently implicated, followed by other contributors such as infective endocarditis, dilated cardiomyopathy, intracardiac tumors, coagulation abnormalities, and anatomical defects like patent foramen ovale.

Two major retrospective series in the field have indicated that AF accounts for a substantial proportion of CE cases, ranging from approximately 28% to 73%. Yet, despite this established link, the specific clinical, imaging, and prognostic profiles of AF-related CE remain inadequately described. Traditional understanding has primarily focused on atherosclerotic AMI, while data delineating the behavior and outcomes of embolic events of cardiac origin, especially in the setting of AF, are sparse.

In this context, the study was undertaken to systematically characterize and compare CE associated with AF (AF CE) and CE from other causes (non-AF CE). The investigators aimed to comprehensively document demographic features, risk factor profiles, laboratory results, echocardiographic parameters, angiographic findings, and subsequent clinical outcomes for both groups. Through this approach, they intended not only to fill existing knowledge gaps but also to identify prognostic indicators that could influence management and improve outcomes in this challenging clinical scenario.

MATERIALS AND METHODS

Study Population and Inclusion Criteria

Between January 2008 and Dicember 2024, consecutive patients admitted to a specialized cardiology care facility were retrospectively reviewed for inclusion in the study. Eligibility was contingent upon two primary criteria: fulfillment of the Fourth Universal Definition of AMI and classification as definite or probable CE according to the diagnostic framework proposed by Shibata et al. Patients satisfying these combined diagnostic benchmarks were incorporated into the final study cohort.

Diagnostic Criteria for Coronary Embolism

The Shibata diagnostic algorithm was employed to establish the diagnosis of CE. This system is predicated on the presence of specific angiographic, clinical, and imaging features categorized as major and minor criteria. Three major criteria include: (1) angiographic evidence of coronary artery obstruction suggestive of embolic origin rather than atherosclerotic plaque rupture; (2) embolic involvement of multiple coronary territories; and (3) concomitant systemic embolization events not attributable to left ventricular thrombi stemming from ST-segment elevation myocardial infarction (STEMI). Minor criteria comprise: (1) less than 25% luminal narrowing in coronary arteries other than the culprit lesion; (2) detection of an embolic source through non-invasive imaging; and (3) identification of predisposing thromboembolic risk factors, including but not limited to AF, valvular heart disease, previous cardiac surgeries, prosthetic valves, hypercoagulable states, and certain structural cardiac anomalies.

Definite CE was assigned when patients met two or more major criteria, one major criterion with at least two minor criteria, or three minor criteria alone. Probable CE required fulfillment of either one major criterion plus one minor criterion or two minor criteria.

Patients exhibiting angiographic or intracoronary imaging evidence of atherosclerotic thrombus, previous percutaneous coronary intervention or bypass grafting, coronary artery ectasia, plaque rupture or erosion, spontaneous coronary artery dissection, or coronary vasospasm were excluded. Likewise, patients with stress-induced cardiomyopathy were also excluded. Cardiac magnetic resonance (CMR) findings supportive of CE, including subendocardial or transmural late gadolinium enhancement (LGE) accompanied by localized myocardial edema, were utilized to corroborate CE diagnoses in patients initially labeled as MINOCA (Myocardial Infarction with Non-Obstructive Coronary Arteries).

Coronary Angiography

All included patients underwent invasive coronary angiography as part of their diagnostic workup. Each angiographic procedure was independently evaluated by two experienced cardiologists, both blinded to clinical data, to ensure consistency and minimize observer bias. Coronary artery stenosis severity was determined via visual estimation, adhering to standardized grading protocols that have been widely accepted in contemporary interventional cardiology practice.

Echocardiography

Transthoracic echocardiography was routinely performed for all patients during their hospitalization. Studies were conducted following established echocardiographic guidelines, emphasizing comprehensive assessment of cardiac structure and function. Specific focus was placed on evaluating the left atrium and left ventricle. Measurements included left atrial dimensions, areas, and volumes indexed to body surface area (Left Atrial Volume Index - LAVI), as well as left atrial ejection fraction (LAEF). Additionally, left atrial strain (LAS) was measured using two-dimensional speckle tracking echocardiography. LAS parameters included reservoir, conduit, and contraction phases, providing insights into both passive and active atrial function.

Echocardiographic image acquisition and analysis were standardized through the use of consistent imaging protocols and the same analysis software package (Philips). Studies were conducted and interpreted by experienced sonographers and cardiologists in accordance with prevailing recommendations.

Cardiac Magnetic Resonance (CMR)

Patients who were initially classified under the diagnostic category of MINOCA underwent CMR imaging. This modality, executed per international recommendations, was instrumental in clarifying ambiguous cases and confirming embolic phenomena. CMR evaluations focused on identifying focal myocardial edema and patterns of LGE consistent with ischemic injury. The detection of transmural or subendocardial LGE in the presence of regional edema was deemed indicative of embolic infarction. These findings provided supplemental diagnostic confirmation for CE in the absence of obstructive coronary artery disease.

Atrial Fibrillation (AF) Definition

AF was defined according to contemporary European Society of Cardiology (ESC) guidelines, characterized by episodes lasting longer than 30 seconds, exhibiting the absence of discernible P waves on electrocardiography and presenting with irregular RR intervals. The study included patients with both a documented history of AF and new-onset AF identified during hospitalization or subsequent follow-up evaluations.

Clinical Outcomes

Outcomes were analyzed in two distinct phases: during the index hospitalization and over a longer-term follow-up period. The in-hospital outcome composite included occurrences of heart failure, cardiogenic shock, malignant ventricular arrhythmias (ventricular tachycardia or ventricular fibrillation), ischemic stroke, and death from any cause. Long-term outcomes comprised a composite of reinfarction, ischemic stroke, systemic embolism, cardiac death, and all-cause mortality, with follow-up duration extending until the most recent patient contact or study endpoint.

ETHICAL CONSIDERATIONS

The study adhered to ethical standards established by the Declaration of Helsinki. Approval was obtained from an independent institutional review board, which evaluated the study's methodology and ethical justification for proceeding without the collection of individual informed consent. The committee granted a waiver of consent (approval code n° 3318-0000257) given the study's retrospective, observational nature, minimal risk to participants, and the importance of the research in addressing a clinical knowledge gap. All patient data were anonymized and handled confidentially to protect privacy in accordance with local data protection regulations.

STATISTICAL ANALYSIS

Analyses were conducted using SPSS software (version 20, IBM Corporation). Categorical variables were expressed as absolute numbers and percentages, while continuous variables were summarized as mean values with corresponding standard deviations. Group comparisons between AF CE and non-AF CE were performed using the Chi-square test or Fisher's exact test for categorical data, depending on expected cell frequencies.

For continuous variables, normality was assessed using the Kolmogorov-Smirnov test. Variables displaying a normal distribution were analyzed using the Student's t-test, while those not meeting the assumption of normality were compared using the Mann-Whitney U test.

Univariable logistic regression was employed to initially identify factors associated with adverse in-hospital outcomes. Variables demonstrating a p-value less than 0.1 were subsequently incorporated into a multivariable logistic regression model using a stepwise forward selection approach. This method allowed for the identification of independent predictors while controlling for potential confounding effects.

For long-term outcomes, univariable Cox proportional hazards regression analyses were conducted to explore relationships between baseline variables and subsequent clinical events. Variables with a univariable p-value <0.1 were entered into a multivariable Cox regression model, again utilizing a stepwise forward selection process. Hazard ratios (HRs) with corresponding 95% confidence intervals (CIs) were calculated for each variable.

Kaplan-Meier survival curves were generated to visualize time-to-event data for long-term outcomes, with statistical significance assessed by the log-rank test. All tests were two-sided, and a p-value of less than 0.05 was considered indicative of statistical significance throughout the analyses.

This structured and methodologically rigorous approach to data collection and analysis was designed to maximize the reliability and clinical relevance of the findings, particularly in delineating the prognostic implications of AF-related CE compared to other forms of CE.

Inclusion Criteria:

• Consecutive patients admitted to a specialized cardiology care facility were retrospectively reviewed for inclusion in the study. Eligibility was contingent upon two primary criteria: fulfillment of the Fourth Universal Definition of AMI and classification as definite or probable CE according to the diagnostic framework proposed by Shibata et al. Patients satisfying these combined diagnostic benchmarks were incorporated into the final study cohort.

Exclusion Criteria:

• Patients exhibiting angiographic or intracoronary imaging evidence of atherosclerotic thrombus, previous percutaneous coronary intervention or bypass grafting, coronary artery ectasia, plaque rupture or erosion, spontaneous coronary artery dissection, or coronary vasospasm were excluded. Likewise, patients with stress-induced cardiomyopathy were also excluded

Consecutive patients admitted to a specialized cardiology care facility were retrospectively reviewed for inclusion in the study. Eligibility was contingent upon two primary criteria: fulfillment of the Fourth Universal Definition of AMI and classification as definite or probable CE according to the diagnostic framework proposed by Shibata et al. Patients satisfying these combined diagnostic benchmarks were incorporated into the final study cohort.

Planing a multicenter study