Acute MI Arising From Coronary Embolism Rare, Often Tied to A-fib

Coronary artery embolism causes acute MI in a small percentage of patients and is most frequently due to A-fib, according to a study published online June 25, 2015, ahead of print in Circulation.  

“Long-term outcomes indicate that [coronary embolism] patients represent a high-risk subpopulation of patients with acute MI and therefore require close follow-up,” say Teruo Noguchi, MD, PhD, and colleagues of the National Cerebral and Cardiovascular Center (Suita, Japan).

Take Home:   Acute MI Arising From Coronary Embolism Rare, Often Tied to A-fib

The investigators reviewed data on 1,776 patients with new-onset acute MI who were treated at their center between January 2001 and December 2013. All underwent angiography and transthoracic echocardiography (TTE).

Coronary artery embolism—in which a thrombus arises from sources other than the coronary vasculature—was diagnosed based on histological, angiographic, and other imaging findings.

Few Coronary Risk Factors in Affected Patients

Overall, 52 patients (2.9%; mean age 66 years; 60% men) met the criteria for coronary artery embolism, which was classified as definite in 32 patients and probable in 20. Among them, 29 (56%) underwent PCI, with initial thrombectomy performed prior to all but 1 intervention and TIMI 3 flow achieved in only 19.

Patients with coronary embolism were less likely than those without the diagnosis to be smokers or to have hypertension or dyslipidemia. These patients also had fewer total coronary risk factors than those without coronary embolism (mean 1.5 vs 2.6; P < .001). Importantly, patients in the coronary embolism group were 10 times more likely to have A-fib (P < .001).

Patients without coronary embolism, on the other hand, more often presented with STEMI (P = .023) and had higher peak creatine kinase levels (P = .01). There were no differences in LVEF.

Rates of 30-day mortality were similar between those with and without embolism, but cardiovascular death was higher in those without the diagnosis (3% vs 0%; P < .001).

A-fib was the most common underlying disease in those with coronary embolism (73%). Other causes included cardiomyopathy (25%) and valvular heart disease (15%). Among patients with A-fib, 66% of cases were chronic and 34% were paroxysmal. Additionally, 58% of A-fib cases were nonvalvular. Vitamin K antagonists were given to 39% of A-fib patients (median INR 1.42; range 0.95-1.80) at the onset of coronary embolism (none received newer oral anticoagulants).

Of patients with nonvalvular A-fib, 60% had a CHADS2 score of 0 or 1 prior to embolism onset. However, reevaluation using the CHA2DS2-VASc score shifted 11 of 18 patients (61%) to a higher risk category that would benefit from oral anticoagulation.  

Over a median follow-up of 49 months after acute MI, coronary embolism or thromboembolism recurred in 5 patients (10.4%). All had A-fib (3 nonvalvular and 2 valvular), and mean INR at the time was 1.47. At 5 years, the rate of recurrent coronary embolism or thromboembolism was 8.7% and the rate of MACCE (cardiac death, MI, ventricular tachycardia/fibrillation, stroke, or recurrent thromboembolism including coronary artery embolism more than 30 days after the initial coronary embolism) was 27.1%.

On Kaplan-Meier analysis, the risks of all-cause death (HR 3.82; 95% CI 2.06-6.48) and cardiac death (HR 5.39; 95% CI 2.38-10.6) were substantially higher at 5 years in patients with vs without coronary embolism.

Two propensity-score models, which matched patients with and without coronary embolism according to different risk factors, confirmed the higher mortality in the embolism group.

Study Adds to Limited Knowledge

According to the study authors, the published literature on coronary embolism consists primarily of case reports. The largest series to date was an autopsy study of 55 patients published in 1978.

Importantly, Dr. Noguchi and colleagues note, the diagnosis of coronary embolism has been based mainly on conventional angiographic features specific for coronary occlusion. In contrast, they say, the criteria used in the current study integrate important characteristics that more appropriately diagnose the condition.

In an editorial accompanying the study, Frank D. Kolodgie, PhD, of CVPath Institute (Gaithersburg, MD), and colleagues say that while coronary thrombosis in patients younger than 50 years often is attributable to coronary artery aneurysms or other structural entities, such as a PFO or atrial septal defect, A-fib “is among those disorders contributing to nonatherosclerotic ischemic heart disease” in older patients.

They add that coronary artery embolism likely is underdiagnosed for several reasons including failure to distinguish embolism from thrombosis or to systematically search for small emboli in the distal and intramural branches of the coronaries.

As life expectancy increases, it is anticipated that A-fib incidence will increase in the older population, making it important to detect and treat the condition with anticoagulants, Dr. Kolodgie and colleagues observe.

“It’s becoming more evident that further research is needed to investigate what factors may be contributing to the increasing trends in [A-fib] incidence and prevalence and particularly its role in [acute MI] in the absence of coronary disease,” they conclude.

A-fib a Common Cause for a Rare Event

In an email with TCTMD, Elsayed Z. Soliman, MD, MSc, of Wake Forest Baptist Medical Center (Winston Salem, NC), said the study “adds to the evidence that A-fib can cause MIs,” an idea previously demonstrated in the REGARDS and ARIC studies.

A-fib in ARIC was associated with NSTEMI more so than STEMI, “suggesting that total occlusion, which could be due to coronary thromboembolism, is less likely to be the main mechanism linking A-fib to MI,” Dr. Soliman said. “This accords with the overall low rate of coronary embolism in this study, although most of these rare events were related by A-fib,” he added. “So, A-fib is a common cause for a relatively rare event.”

Dr. Soliman said the study is an important reminder for clinicians to consider the risk of MI as well as stroke in their A-fib patients. “Given that direct coronary embolism is not as common a cause for MI as it is for stroke, anticoagulants may not be as effective in preventing MI in these patients as they are for stroke,” he said.

1. Shibata T, Kawakami S, Noguchi T, et al. Prevalence, clinical features, and prognosis of acute myocardial infarction due to coronary artery embolism. Circulation. 2015;Epub ahead of print.
2. Kolodgie FD, Virmani R, Finn AV, Romero ME. Embolic myocardial infarction as a consequence of atrial fibrillation: a prevailing disease of the future [editorial]. Circulation. 2015;Epub ahead of print. 


  • The study was funded by grants from the Takeda Science Foundation, the Japan Cardiovascular Research Foundation, and the Ministry of Health, Labour and Welfare of Japan. 
  • Drs. Noguchi, Kolodgie, and Soliman report no relevant conflicts of interest.  

Related Stories:


We Recommend