Author + information
- Received April 17, 2017
- Revision received August 3, 2017
- Accepted August 8, 2017
- Published online January 15, 2018.
- Claire E. Raphael, MBBS, PhDa,
- John A. Heit, MDa,b,
- Guy S. Reeder, MDa,
- Melanie C. Bois, MDc,
- Joseph J. Maleszewski, MDc,
- R. Thomas Tilbury, MDa and
- David R. Holmes Jr., MDa,∗ ()
- aDepartment of Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota
- bDivision of Hematology, Department of Internal Medicine, Mayo Clinic, Rochester, Minnesota
- cDepartment of Pathology, Mayo Clinic, Rochester, Minnesota
- ↵∗Address for correspondence:
Dr. David R. Holmes, Jr., Department of Cardiovascular Diseases, Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905.
Coronary embolism is the underlying cause of 3% of acute coronary syndromes but is often not considered in the differential of acute coronary syndromes. It should be suspected in the case of high thrombus burden despite a relatively normal underlying vessel or recurrent coronary thrombus. Coronary embolism may be direct (from the aortic valve or left atrial appendage), paroxysmal (from the venous circulation through a patent foramen ovale), or iatrogenic (following cardiac intervention). Investigations include transesophageal echocardiography to assess the left atrial appendage and atrial septum and continuous electrocardiographic monitoring to assess for paroxysmal atrial fibrillation. The authors review the historic and contemporary published data about this important cause of acute coronary syndromes. The authors propose an investigation and management strategy for work-up and anticoagulation strategy for patients with suspected coronary embolism.
Although coronary embolism is commonly reported in case studies, it is probably under-diagnosed in clinical practice. A recent retrospective analysis suggested that up to 3% of acute coronary syndrome (ACS) cases may result from coronary embolism (1). Awareness of the characteristics of patients who present with coronary embolism and management considerations specific to this condition may improve clinical outcome in these patients. We review the published data surrounding coronary embolism, management, and clinical implications.
Coronary embolism is divided into 3 types: 1) direct; 2) paradoxical; and 3) iatrogenic, with some overlap among the categories (2–6). Direct coronary emboli commonly originate from the left atrial appendage, left ventricle, the aortic or mitral valves, or the proximal coronary artery. Embolic tissue may be thrombus, valvular material, or even neoplasm. Paradoxical emboli pass through a patent foramen ovale (PFO), atrial septal defect, or pulmonary arteriovenous malformations from the venous circulation into the systemic circulation. Most commonly the origin is from a deep vein (deep venous thrombosis). Iatrogenic emboli may occur following interventional procedures, usually valve replacements and coronary intervention.
Direct Coronary Embolism
Direct coronary embolism may result from thrombus originating from the left atrium, left ventricle, or pulmonary veins; endocarditis of the aortic or mitral valve; and, more rarely, cardiac tumors. Clinically significant coronary embolism has been reported in 1.5% of cases with infective endocarditis and microemboli to the coronary arteries were present in more than 60% of cases on postmortem examination (7,8). Although there are no studies that describe risk factors for coronary embolism, systemic embolization of infected vegetation is known to be more common in patients with endocarditis of the mitral valve, large vegetations (>10 mm in diameter on echocardiography), and in fungal or staphylococcal infections (9). Embolization of noninfective vegetations caused by autoimmune disease has also been reported (10). Although an uncommon cause of coronary embolism, cardiac myxomas can present with such, particularly when the tumors are of the villiform variety (11–13). Coronary emboli may also originate from mural thrombus within the left-sided cardiac chambers.
Paradoxical Coronary Embolism
The incidence of paradoxical coronary embolism is unknown. Case reports include coronary emboli originating from deep vein thrombosis and even foreign bodies in the venous circulation (14). Emboli gain access to the systemic circuity by passing through a PFO or atrial septal defect, and then into the coronary arteries. Patients with a PFO and major pulmonary embolus had more than 10-fold increased mortality compared with those without a PFO in a retrospective series, and this was thought to result from paradoxical emboli to either the coronary or, more commonly, the cerebral circulation (15). Closure of PFOs is easily accomplished percutaneously and closure devices have recently been approved by the Food and Drug Administration for this indication. Observational data have suggested a benefit of percutaneous PFO closure compared with medical therapy (16). Randomized trial data have been hampered by low event rates and did not demonstrate superiority compared with medical therapy on intention-to-treat analysis (17). However, closure was superior to medical therapy alone on per-protocol analysis and subgroup analysis suggested a benefit in patients with large right-to-left shunts and atrial septal aneurysms, with low rates of risk associated with device closure.
Iatrogenic Coronary Embolism
Embolism of surgical material into the coronary arteries may occur at the time of cardiothoracic surgery, during coronary intervention, commonly at the time of rotablation (18), and also during aortic (19) and mitral (20) valvuloplasty procedures. Clot may form in catheters, particularly during prolonged procedures, when full heparinization is not maintained and when catheters are not regularly flushed. Coronary intervention is now the most common cause of coronary embolism (6). In the presence of degenerative (calcific) aortic valve disease, even diagnostic coronary angiography may result in embolization of friable valvular material into a coronary artery (21).
Prevalence of Coronary Embolism
The coronary arteries appear relatively protected anatomically from embolic events compared with the other systemic circulation. Thromboemboli originating in the left ventricle or left atrium most commonly pass into the carotids or distal systemic circulation because these vessels are in a relatively linear trajectory from the left ventricular outflow tract compared with the coronaries. The acute or right angle takeoff from the aortic root likely also decreases the probability of coronary embolism when compared with other systemic embolism.
The prevalence of coronary embolism was first estimated from autopsy studies. Prizel et al. (22) presented a cohort of 55 patients, representing 13% of the acute myocardial infarctions in an autopsy series, with thrombotic material in the coronary arteries but no demonstrable acute plaque rupture, suggesting an embolic source. Most had underlying valvular heart disease, cardiomyopathy, or atrial fibrillation.
A large series of all-comers to the catheterization laboratory with a first acute myocardial infarction over a 12-year period demonstrated up to 3% prevalence of coronary embolism (1). Diagnosis was made on retrospective examination of diagnostic angiography and subsequent investigation including echocardiography, magnetic resonance imaging, and computed tomography. The authors described a scoring system for definite and probable coronary embolism (Table 1) to diagnose cases that may have been misdiagnosed at the time of acute presentation. Most cases with coronary embolism in this series had atrial fibrillation (73%); however, cardiomyopathy and valvular heart disease were also common. Fifteen percent of patients had multiple coronary emboli in different territories. Concomitant systemic and coronary embolism was present in about one-fifth of patients. Following investigation, an intracardiac source was identified in one-third of patients, most commonly within the left atrial appendage, followed by infective endocarditis.
On long-term follow-up, 10% of patients had recurrent thromboembolic disease (4% coronary embolus, 6% stroke). Recurrent events all occurred in patients with underlying atrial fibrillation. Ten-year cardiac mortality rate was approximately 50% in the coronary embolism group compared with <10% in a comparative ACS group without coronary embolism. This may relate to identification bias because small embolic events would not meet the diagnostic criteria or may suggest larger infarctions in the embolism group compared with the rest of the ACS population.
Diagnosis and Management of Acute Coronary Embolism
Presentation of coronary embolism is difficult to distinguish from atherosclerotic ACS before angiography. Therefore, initial management with antiplatelet therapy, pain relief, and urgent or emergency coronary angiography is the same as for any suspected ACS case. Angiographically, coronary embolism may present with a heavy thrombus burden, as shown in Figure 1A. Smaller emboli may be present more distally. Filling defects in multiple coronary territories is suggestive of coronary embolism. Although some patients present with a heavy thrombus load and no underlying coronary atherosclerosis, more commonly patients have a degree of bystander atherosclerosis, making diagnosis more challenging.
If there is a high thrombus burden, aspiration thrombectomy may be considered. The evidence for this is controversial because trials have suggested no significant benefit in ST-segment elevation myocardial infarction and an increased risk of post-procedural stroke (23,24). However, meta-analysis of individual patient data has suggested there may be a significant reduction in cardiovascular mortality in selected patients with high thrombus load, albeit with a small but significant increase in stroke risk (25). Thrombectomy trials have usually used aspiration devices; however, the Angiojet (Boston Scientific, Marlborough, Massachusetts) may have increased efficacy in restoration of flow and removal of thromboembolic material (26).
Aspiration allows better assessment of the underlying coronary artery. If aspiration results in full restoration of flow and a normal angiographic appearance, intravascular ultrasound or optical computerized tomography could be used to further evaluate the vessel to assess the potential for isolated plaque erosion versus embolic phenomena. If the artery appears angiographically normal, balloon angioplasty and stents are not required.
Aspirated material should be sent to pathology for further examination that may assist in narrowing the differential for thrombus origin. For instance, identification of platelet- and fibrin-rich material is consistent with acute thrombus from the left heart or proximal coronary artery (Figure 2) and may allow distinction from a venous paradoxical embolus, which would be more likely to demonstrate organization of the thrombotic material. Less commonly but vital for patient management, embolic material from a neoplasm, foreign material, or infected source (septic emboli) may be identified.
Aspiration may not remove the thrombus entirely, necessitating the use of antiplatelet and anticoagulant agents. Intracoronary thrombolytics have been used to good effect in case studies (27). Infusion of unfractionated heparin, GPIIb/IIIa agents, or bivalurudin may be used to manage distal embolus, particularly in vessels with reduced distal perfusion (3). However, in the case of more organized thrombus, for example from paradoxical emboli, these agents may not be effective.
Following acute intracoronary management, the patient should be screened for the source of embolic material and for systemic embolism. If paradoxical embolism is suspected at the time of the angiogram, right heart catheterization may identify intra-atrial communication. Transthoracic echocardiography may allow identification of left ventricular thrombus, whereas transesophageal echocardiography allows visualization of the left atrial appendage (Figure 1). Either transesophageal echocardiography or transthoracic echocardiography may suggest communication across the atrial septum and bubble studies may be used to confirm the presence of a PFO. In this case, screening of the venous circulation for deep vein thrombosis or pulmonary embolus may be considered based on clinical probability. Coronary embolism is commonly associated with other systemic embolism. Clinical evidence of stroke was present in 15% of patients presenting with coronary embolism on a retrospective series (1) and prevalence of subclinical cerebral embolism is likely even higher.
Interestingly, in the cohort study by Shibata et al. (1), most patients with atrial fibrillation and coronary embolus had a CHADS2 score of 0 or 1. Acute finding of left ventricular or left atrial appendage thrombus is usually treated with anticoagulation, and we suggest treatment with oral anticoagulation be started even if the CHADS2 or CHADS2-VASc score does not meet conventional criteria for anticoagulation (28) because these patients remain at increased risk for future thromboembolic events. The role of left atrial appendage occlusion devices in coronary embolism has not been studied; however, these may be an alternative in these patients if long-term oral anticoagulation is not suitable.
Is There a Role for Thrombophilia Testing in Patients With Coronary Embolism?
For clot to form, at least 2 of the conditions of Virchow triad (Central Illustration, top triangle) must be present. For coronary embolism, anatomic factors, such as a PFO, allow transit of clot from the venous to the arterial circulation and provide an additional risk factor.
A procoagulant state might reasonably be expected to increase risk of embolic disease; however, thrombophilia testing is seldom indicated as part of the diagnostic work-up for coronary embolism. The inherited thrombophilias have an estimated prevalence of 5% in the general population (29). Although they increase the risk of venous thromboembolism, there was no increased risk of arterial thromboembolism in long-term cohort studies of patients with inherited thrombophilia (30). Case control studies have suggested a similar prevalence of protein C, protein S, and antithrombin deficiencies in patients with cryptogenic stroke or myocardial infarction and matched control subjects, even in young patients (31–33). Development of thrombosis, both venous and arterial, is complex and genetic determinants are a smaller component of risk compared with environmental and acquired risk factors (34).
Because of the rarity of coronary embolic disease, recommendations for thrombophilia testing are extrapolated from the data for strokes of possible thromboembolic etiology and, to a lesser extent, venous thromboembolism. Routine thrombophilia testing does not change anticoagulation strategy in most cases of coronary embolism; however, it should be considered if it is likely to change management of the patient.
For most patients with coronary embolism, if the risk of bleeding is low and there are no reversible risk factors (Table 2), long-term oral anticoagulation is generally recommended. The recurrence rate for thromboembolic events was 10% (1) on follow-up so a negative thrombophilia screen may even provide false reassurance (35), because it only tests for known thrombophilia mutations. If there are concerns regarding bleeding risk with long-term oral anticoagulation, thrombophilia testing may be useful to assess individual patient risk of recurrent thromboembolism.
Young patients, both for coronary embolism and stroke, are a particular area of difficulty with minimal data in the stroke data and no data for coronary thromboembolism. Data from stroke suggests that thromboembolic disease is the underlying cause of 20% of strokes in patients younger than 50 years of age, largely in patients with first stroke at age <35 (36). Thrombophilia testing has been empirically recommended in patients younger than age 45 with a cryptogenic stroke (37); however, there are no data to support improved clinical outcome as a result of this testing and we advocate discussion of long-term oral anticoagulation as the first-line management recommendation.
Interpretation of thrombophilia testing is complex. Patients are required to be off all anticoagulation and is best performed after the acute phase reaction has resolved. Novel oral anticoagulants may cause false-positive lupus anticoagulant testing and falsely low antithrombin activity (38). Additional management dilemmas are associated with any genetic testing, including management of mutations of unknown significance and the anxiety associated with familial testing.
Full history and clinical examination should assess for procoagulant risk factors, including signs and symptoms of active cancer; vasculitis; autoimmune disease; hyperviscosity syndromes; and infection, including endocarditis and factors increasing risk of thromboembolism, including immobility and medication use (Table 2). A detailed history of recurrent miscarriage, previous venous thromboembolism, and arterial thromboembolism should be taken. Blood tests including erythrocyte sedimentation rate and vasculitis and autoimmune testing where clinically suspected should be performed. Management is then guided by clinical assessment of likelihood of recurrent embolism.
Management of Coronary Embolism: Who to Anticoagulate and for How Long
Patients presenting with coronary embolism of suspected venous origin, either from the history and examination or from histological examination of the aspirated thrombus, should undergo further work-up to assess for the source and route of paradoxical embolism, including assessment for PFO and arteriovenous malformations. Paradoxical embolism is more difficult to identify and therefore may be underdiagnosed as a cause of coronary embolism. In keeping with the venous thromboembolism guidelines, management typically is with 3 months of oral anticoagulation in patients with reversible risk factors (Table 2), such as temporary immobility, post-surgery, and postpartum Idiopathic thromboembolism or thromboembolism associated with ongoing procoagulant states, such as active cancer, are likely to benefit from long-term oral anticoagulation (38).
Patients with atrial fibrillation had a 10% recurrence rates in the series by Shibata et al. (1) and should be offered long-term oral anticoagulation unless there are contraindications. For all other coronary embolism patients, we suggest a period of oral anticoagulation for 3 months in the absence of ongoing procoagulant risk factors (Table 2). For those with persistent risk factors, long-term oral anticoagulation should be considered.
Anticoagulation should be achieved with warfarin or a novel oral anticoagulant rather than antiplatelet drugs. Patients who received a coronary stent at the time of coronary angiography are more complex. Management may either be with triple therapy or a single antiplatelet agent coupled with oral anticoagulation, such as warfarin and clopidogrel. Again, in the absence of guidelines, decisions are at the discretion of the individual clinician and assessment of bleeding risk against risk of further thromboembolic events.
Empiric Recommendations for Investigation of Coronary Embolism
• History and examination should be performed to assess for evidence of conditions resulting in either temporary or sustained increased risk of thromboembolism (Table 2).
• Patients should be investigated for atrial fibrillation, particularly paroxysmal, with electrocardiogram telemetry.
• Patients should be investigated with transesophageal echocardiography to assess for left atrial appendage thrombus or spontaneous contrast in the left atrial appendage, indicating increased predisposition to hemostasis in the left atrial appendage, PFO, and atrial septal defect.
Empiric Recommendations for Treatment of Coronary Embolism
• Patients with atrial fibrillation with a low risk of bleeding should be offered long-term oral anticoagulation, regardless of the CHADS2-VASc score.
• Patients with recurrent coronary embolism with a low risk of bleeding should be offered long-term oral anticoagulation.
• Patients with a reversible risk factor for thromboembolic disease (Table 2) at the time of coronary embolism that has subsequently resolved should receive oral anticoagulation for 3 months.
• Patients with persistent risk factors for thromboembolic disease should be offered long-term oral anticoagulation.
• Patients who underwent percutaneous coronary intervention (stenting) need to receive antiplatelet agents in addition to oral anticoagulation.
• There is no role for routine thrombophilia testing in patients with coronary embolism.
• If there are concerns regarding bleeding risk or patients are unwilling to take long-term oral anticoagulation, thrombophilia testing may aid individual risk stratification for recurrent thromboembolism.
Coronary embolus is an underrecognized cause of ACS. It occurs most commonly following coronary or surgical intervention. However, spontaneous embolization of thrombi from the left atrial appendage, left ventricle, or systemic veins (paradoxical) may occur and present as ACS. In the case of high thrombus burden or recurrent thrombus with a relatively normal underlying coronary artery, coronary embolism should be suspected. Investigation with transthoracic and transesophageal echocardiography may reveal a source of coronary emboli. Aspiration thrombectomy should be considered in the presence of a high thrombus load, whereas distal embolus may be managed with anticoagulation. Following aspiration, if the underlying vessel appears free of atherosclerosis, the patient may be managed with anticoagulation alone. Intravascular ultrasound or optical computerized tomography may be helpful in determining if there is underlying atherosclerosis and if percutaneous coronary intervention/balloon angioplasty should be considered. Patients should be screened for underlying atrial fibrillation and sources of embolic material, including left atrial appendage thrombus, left ventricular thrombus, and ateriovenous communications.
At present, the natural history of coronary embolism is not well understood; however, the cardiac mortality rate on a recent cohort study was 50% at 10 years, significantly higher than the general ACS population. This suggests these patients require further study to identify the natural history of the condition and how to optimize management to improve outcome. We suggest that patients be managed with oral anticoagulation for a period of 3 months if embolism occurred because of temporary procoagulant risk factors, and with long-term oral anticoagulation in the presence of persistent risk factors, including atrial fibrillation.
All authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- acute coronary syndrome(s)
- patent foramen ovale
- Received April 17, 2017.
- Revision received August 3, 2017.
- Accepted August 8, 2017.
- 2018 American College of Cardiology Foundation
- Shibata T.,
- Kawakami S.,
- Noguchi T.,
- et al.
- Lacunza-Ruiz F.J.,
- Muñoz-Esparza C.,
- García-de-Lara J.
- ↵Brunson JG. Coronary embolism in bacterial endocarditis. Am J Pathol 1953;29:689–701.
- Saphir O.,
- Katz L.N.,
- Gore I.
- El Sabbagh A.,
- Al-Hijji M.A.,
- Thaden J.J.,
- et al.
- Mallouppas M.,
- Vassiliou V.,
- Goddard M.,
- Rana B.,
- Braganza D.
- Konstantinides S.,
- Geibel A.,
- Kasper W.,
- Olschewski M.,
- Blümel L.,
- Just H.
- Wahl A.,
- Jüni P.,
- Mono M.-L.,
- et al.
- Carroll J.D.,
- Saver J.L.,
- Thaler D.E.,
- et al.
- ↵Chikkabasavaiah N, Rajendran R. Percutaneous coronary intervention for coronary thrombo-embolism during balloon mitral valvuloplasty in a pregnant woman. Heart Lung Circ 2016;25:e29-31.
- Dauvergne C.,
- Araya M.,
- Valenzuela J.,
- Meneses M.,
- Selman R.,
- Maluenda G.
- Jolly S.S.,
- James S.,
- Džavík V.,
- et al.
- Migliorini A.,
- Stabile A.,
- Rodriguez A.E.,
- et al.
- Nadir M.A.
- Rosendaal F.R.,
- Tans G.,
- Nicolaes G.,
- Bertina R.,
- Mannucci P.
- Martinelli I.,
- Mannucci P.M.,
- De Stefano V.,
- et al.
- Hankey G.J.,
- Eikelboom J.W.,
- van Bockxmeer F.M.,
- Lofthouse E.,
- Staples N.,
- Baker R.I.
- Mandala E.,
- Lafaras C.,
- Tsioni C.,
- et al.
- Stevens S.M.,
- Woller S.C.,
- Bauer K.A.,
- et al.
- Putaala J.,
- Metso A.J.,
- Metso T.M.,
- et al.
- Cohoon K.P.,
- Heit J.A.
- Central Illustration
- Direct Coronary Embolism
- Paradoxical Coronary Embolism
- Iatrogenic Coronary Embolism
- Prevalence of Coronary Embolism
- Diagnosis and Management of Acute Coronary Embolism
- Is There a Role for Thrombophilia Testing in Patients With Coronary Embolism?
- Management of Coronary Embolism: Who to Anticoagulate and for How Long
- Empiric Recommendations for Investigation of Coronary Embolism
- Empiric Recommendations for Treatment of Coronary Embolism