Embolic Stroke of Unknown Source: What Are the Next Steps?


  1. Embolic stroke of unknown source (ESUS) and cryptogenic stroke are not the same, but both have soft definitions.
  2. Most patients with ESUS or cryptogenic strokes do not have a recurrence.
  3. Labelling the purported etiology of stroke based on radiographic findings or specific risk factors is not valid.
  4. The presence or absence of atrial fibrillation at or around the time of stroke does not identify the cause for stroke.
  5. Recurrence of large vessel stroke (diagnosed by radiographic evidence) is not improved by any specific therapy.
  6. Lacunar small-vessel disease diagnosed by radiography is not improved by anticoagulation or platelet therapy.
  7. Anticoagulation therapy (versus placebo) in large populations of patients with atrial fibrillation with or without prior stroke reduces the risk of recurrent stroke independent of radiographic history but does not eliminate total risk of stroke; 40% of strokes are not treated by anticoagulation and are presumably unrelated to atrial fibrillation.
  8. Identifying atrial fibrillation in patients with ESUS has not been shown to reduce recurrent stroke risk.
  9. Radiographic evidence of the purported cause for stroke does not improve recurrent stroke risk based on present therapy even for those with atrial fibrillation. Similar to myocardial infarction, revascularization may benefit only those who need an acute intervention to restore blood flow but not to affect risk of recurrence.
  10. Treating cardiovascular risk using American Heart Association's Life's Simple 7,1 sleep apnea,2,3 and other predictors may be the best option to reduce risk of recurrent stroke.


The etiology of the most common cause of ischemic neurological events (i.e., strokes) is hard to classify based on poorly validated angiographic findings and a broad range of risk factors (TOAST,4,5 Causative Classification System for Ischemic Stroke,6 ASCOD7). Non-hemorrhagic strokes of undetermined etiology (cryptogenic) presumably represent ~20-40% of all strokes.8 When an ischemic stroke is labeled non-cryptogenic, the cause often remains obscure. Numerous risk calculators such as ABCD, ASCO, RRE, and Essen have been utilized to predict risk of a recurrent ischemic stroke.9-12 While validated in short-term and long-term studies, these predictors did not differentiate the cause for recurrent ischemic events. A close look at these studies' methodologies reveals that individual risk factors overlap for embolic and thrombotic strokes, making a singular diagnosis problematic. In fact, a recurrent stroke may not necessarily be due to the same cause even in a given patient.

The ultimate value of predicting risk is to identify the mechanism responsible for a future event so that an effective preventative therapy can be prescribed. With overlap of risk factors for etiologies of stroke, any patient may be at risk for multiple etiologies for stroke. Previous definitions for classification of stroke etiologies have been inconsistent, and evidence to guide therapy is scarce.

Recently, ESUS-specific criteria were defined by the Cryptogenic Stroke/ESUS International Working Group to identify strokes that were considered cryptogenic but likely due to a thromboembolus.13 But whether it is labelled ESUS or cryptogenic, the mechanism for stroke remains uncertain. Even if a stroke is likely embolic, there are multiple potential etiologies including aortic atherosclerotic plaques, pro-thrombotic states (as may exist in patients with cancer), patent foramen ovale with paradoxical thromboemboli, atrial fibrillation with atrial appendage thromboemboli, atrial thrombi unrelated to atrial fibrillation, valvular heart thromboemboli, and left ventricular clots, among other potential causes for ESUS.8 Patients often have more than one potential etiology.14

In the Athens Stroke Registry, for 2,731 patients post-stroke who were followed for 31 months, initial strokes were classified using traditional definitions.15,16 ESUS-defined patients' clinical characteristics included non-stenotic atherosclerotic plaques, moderate systolic or diastolic dysfunction, and atrial fibrillation. The ESUS patients had recurrent event rates similar to the cardio-embolic defined patients (versus much lower recurrent rates in other groups), suggesting similar etiologies. However, presence of these explanatory characteristics has not been shown to be causal. Thus, ESUS patients have multiple risks for second strokes with no clear beneficial therapy. One unfounded hypothesis is that atrial fibrillation is the leading cause of ESUS, and an opportunity to diagnose this could lead to preventative strategic anticoagulation treatment and reduce risk of further strokes.17 However, no data support this presently.

Electrocardiographic Monitoring and Cerebrovascular Accident Risk Factors

The more intense the heart rhythm monitoring in patients at risk of stroke, the more the likelihood of finding atrial fibrillation.18-22 A Taiwan registry and a large population-based study showed the relationship of the CHADS2 (and the CHA2DS2VASc) score and incidence in newly diagnosed atrial fibrillation.23,24 To support this concept, the REVEAL AF (Incidence of AF in High Risk Patients) trial and other studies were performed.25-30 The REVEAL AF trial, an observational study of patients without stroke or known atrial fibrillation, but with CHADS2 scores >3, were evaluated; atrial fibrillation was observed on implantable monitors at 30 days, 6 months, 12 months, 24 months, and 30 months in 6.2 %, 20.4%, 27.1%, 33%, and 40%, respectively. In fact, the AS5F score was developed to predict atrial fibrillation after stroke.31 These findings lead one to the presupposition that the finding of atrial fibrillation in stroke patients equates to the causality of a cardio-embolic etiology of the stroke, ignoring the fact atrial fibrillation may be an innocent bystander or merely a marker of overall stroke risk.

The Find-AF (Finding Atrial Fibrillation in Stroke)32 and CRYSTAL-AF (Study of Continuous Cardiac Monitoring to Assess Atrial Fibrillation After Cryptogenic Stroke)22 trials evaluated the heart rhythm of patients with ESUS. In the CRYSTAL-AF trial, time to first detection of atrial fibrillation with an implantable loop recorder at 6, 12, and 36 months was 8.9%, 12.4%, and 30% versus 1.4%, 2.0%, and 3% in the control arm (with electrocardiograms and Holter and Event recorders). Of the implantable loop recorder patients, 46% had an atrial fibrillation burden of 12-24 hours as a maximum during follow-up.33 Of patients with atrial fibrillation, 97% had an oral anticoagulant prescribed at time of atrial fibrillation diagnosis with no reported recurrent stroke difference. Presumably, the control group had the same incidence of atrial fibrillation (undetected); thus, a causal relationship between atrial fibrillation and stroke remains uncertain.

Further, lack of atrial fibrillation causality is supported by recurrent events in the TRENDS (A Prospective Study of the Clinical Significance of Atrial Arrhythmias Detected by Implanted Device Diagnostics),34 ASSERT (Asymptomatic Atrial Fibrillation and Stroke Evaluation in Pacemaker Patients and the Atrial Fibrillation Reduction Atrial Pacing Trial),35 and IMPACT-AF (Clinical Trial to Improve Treatment With Blood Thinners in Patients With Atrial Fibrillation)36 clinical trials, where stored electrocardiogram data demonstrated no atrial fibrillation occurred within 30 days prior to an event or >30 days prior to event and, in others, atrial fibrillation only occurred after an event. Taken together, these studies more strongly support atrial fibrillation as an overall ill-defined stroke risk and explain the lack of preventive therapy based upon atrial fibrillation event-guided therapy.

To evaluate secondary treatment options in patients with noncardioembolic stroke, the WARSS (Warfarin-Aspirin Recurrent Stroke Study) was undertaken, which failed to prove efficacy in the overall study and in the cryptogenic subgroup.37 Furthermore, in a specifically defined ESUS patient population, the NAVIGATE ESUS (Rivaroxaban Versus Aspirin in Secondary Prevention of Stroke and Prevention of Systemic Embolism in Patients With Recent Embolic Stroke of Undetermined Source)38 and RE-SPECT ESUS (Dabigatran Etexilate for Secondary Stroke Prevention in Patients With Embolic Stroke of Undetermined Source)39 trials were designed to assess oral anticoagulant efficacy versus antiplatelet therapy to reduce the risk of recurrent stroke in ESUS. In the NAVIGATE ESUS trial, efficacy of aspirin versus rivaroxaban was the same with a risk of stroke outcome of approximately 5% in 1 year, which may be no better than no therapy at all, and yet the risk of bleeding with rivaroxaban was quite a bit higher, with a hazard ratio of 2.72 (confidence intervals 1.68-4.39). With regard to the RE-SPECT ESUS trial, there was a similar recurrence rate of first stroke in both the randomized dabigatran and aspirin groups, but bleeding risks were not specifically different between the 2 groups.

Atrial fibrillation detection is possible by implantable loop recorders.40 However, Afzal et al.41 reported that in patients who had loop recorder implantation and cryptogenic stroke, 86 out of the total 100 transmissions were false positive for atrial fibrillation and other rhythm disturbances, with only 14/100 being true positives and indicating the presence of atrial fibrillation. In addition to atrial fibrillation being misdiagnosed in this study, pauses, tachycardia, and bradycardia were also diagnosed inappropriately. Atrial fibrillation was diagnosed accurately in only ~50% of individuals.

Thus, regarding detection of atrial fibrillation after ESUS, what is the reasoning for it? Enhanced and prolonged monitoring is more likely to find atrial fibrillation. Presence of atrial fibrillation detection to date has not been associated with difference in recurrent stroke rates. Oral anticoagulant versus antiplatelet therapy has not proven beneficial in patients with ESUS.

When it comes to ESUS/cryptogenic stroke, one has to question the value of any treatment, whether it be empiric or electrocardiographically guided oral anticoagulation or antiplatelet therapy. Similar to premature ventricular contractions, atrial fibrillation in left ventricular dysfunction may indicate risk but not causality. And as in the CAST (Cardiac Arrhythmia Suppression) trial, trying to guide therapy based upon monitoring may invoke harm.42

One, therefore, should consider the next step in patients with ESUS. Stop monitoring? More aggressive and accurate monitoring? Reconsideration of the risk score which we use? Considering the CHA2DS2VASc and CHADS2 scores, a score of 2 is already present in those patients with an ESUS stroke. However, oral anticoagulants are ineffective. These scores may simply not apply to patients who have had a recent stroke, at least without some evidence that atrial fibrillation was the cause for the stroke. Should we use brain natriuretic peptide to differentiate cardioembolic stroke and/or d-dimer to identify hypercoagulable states and cancer? That is not clear.

A recent white paper43 suggests the following approach:

  1. Assess risk factors including age, sex, hypertension, obesity, and diabetes.
  2. Evaluate cardiovascular dysfunction including atrial stiffness and left ventricular dysfunction.
  3. Assess for rheumatic heart disease and atrial myopathy.
  4. Determine if the stroke appears to be embolic.

If atrial fibrillation is suspected because of these risk factors, consider long-term monitoring with an implantable loop recorder. Otherwise, just consider short-term monitoring (≤72 hours).

It is the authors' opinion that 1) it is not clear that this approach will make a difference and 2) it is possible that detection of atrial fibrillation, accurate or not, may lead to a therapy that will create more harm than good with potential for greater expense and with need for health care intervention. Despite initial enthusiasm for intense monitoring in patients with ESUS, we have come to a point where we understand that the term ebolic stroke of unknown source is nothing but a hopeful moniker. The etiology for these neurological events remains cryptogenic. Therapeutic interventions have not been proven definitively to improve or affect risk for recurrent stroke, seriousness of stroke, disability from stroke, or any other specific outcome measure. Understanding what causes ESUS and what a cryptogenic stroke is may help better determine the need for assessing thrombogenicity. Perhaps we have been off base to put too much emphasis on atrial fibrillation. Other causes, such as cancer or other conditions that cause a thrombogenic state, are equally if not even more important.


  1. Ford ES, Greenlund KJ, Hong Y. Ideal Cardiovascular Health and Mortality From All Causes and Diseases of the Circulatory System Among Adults in the United States. Circulation 2012;125:987-95.
  2. Koo BB, Bravata DM, Tobias LA, et al. Observational Study of Obstructive Sleep Apnea in Wake-Up Stroke: The SLEEP TIGHT Study. Cerebrovasc Dis 2016;41:233-41.
  3. Parasram M, Segal AZ. Sleep Disorders and Stroke: Does Treatment of Obstructive Sleep Apnea Decrease Risk of Ischemic Stroke? Curr Treat Options Neurol 2019;21:29.
  4. Adams HP Jr, Bendixen BH, Kappelle LJ, et al. Classification of Subtype of Acute Ischemic Stroke. Definitions for Use in a Multicenter Clinical Trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment. Stroke 1993;24:35-41.
  5. Adams HP Jr, Woolson RF, Biller J, Clarke W. Studies of Org 10172 in Patients With Acute Ischemic Stroke. TOAST Study Group. Haemostasis 1992;22:99-103.
  6. Arsava EM, Ballabio E, Benner T, et al. The Causative Classification of Stroke System: An International Reliability and Optimization Study. Neurology 2010;75:1277-84.
  7. Amarenco P, Bogousslavsky J, Caplan LR, Donnan GA, Wolf ME, Hennerici MG. The ASCOD Phenotyping of Ischemic Stroke (Updated ASCO Phenotyping). Cerebrovasc Dis 2013;36:1-5.
  8. Ntaios G. Embolic Stroke of Undetermined Source: JACC Review Topic of the Week. J Am Coll Cardiol 2020;75:333-40.
  9. Chaudhary D, Abedi V, Li J, Schirmer CM, Griessenauer CJ, Zand R. Clinical Risk Score for Predicting Recurrence Following a Cerebral Ischemic Event. Front Neurol 2019;10:1106.
  10. Couillard P, Poppe AY, Coutts SB. Predicting Recurrent Stroke After Minor Stroke and Transient Ischemic Attack. Expert Rev Cardiovasc Ther 2009;7:1273-81.
  11. Josephson SA, Sidney S, Pham TN, Bernstein AL, Johnston SC. Higher ABCD2 Score Predicts Patients Most Likely to Have True Transient Ischemic Attack. Stroke 2008;39:3096-8.
  12. Koton S, Rothwell PM. Performance of the ABCD and ABCD2 Scores in TIA Patients With Carotid Stenosis and Atrial Fibrillation. Cerebrovasc Dis 2007;24:231-5.
  13. Hart RG, Diener HC, Coutts SB, et al. Embolic Strokes of Undetermined Source: The Case for a New Clinical Construct. Lancet Neurol 2014;13:429-38.
  14. Ntaios G, Perlepe K, Lambrou D, et al. Prevalence and Overlap of Potential Embolic Sources in Patients With Embolic Stroke of Undetermined Source. J Am Heart Assoc 2019;8:e012858.
  15. Ntaios G, Papavasileiou V, Milionis H, et al. Embolic Strokes of Undetermined Source in the Athens Stroke Registry: A Descriptive Analysis. Stroke 2015;46:176-81.
  16. Ntaios G, Papavasileiou V, Milionis H, et al. Embolic Strokes of Undetermined Source in the Athens Stroke Registry: An Outcome Analysis. Stroke 2015;46:2087-93.
  17. Glotzer TV, Ziegler PD. Cryptogenic Stroke: Is Silent Atrial Fibrillation the Culprit? Heart Rhythm 2015;12:234-41.
  18. Sposato LA, Klein FR, Jáuregui A, et al. Newly Diagnosed Atrial Fibrillation After Acute Ischemic Stroke and Transient Ischemic Attack: Importance of Immediate and Prolonged Continuous Cardiac Monitoring. J Stroke Cerebrovasc Dis 2012;21:210-6.
  19. Gladstone DJ, Spring M, Dorian P, et al. Atrial Fibrillation in Patients With Cryptogenic Stroke. N Engl J Med 2014;370:2467-77.
  20. Diener HC, Easton JD, Granger CB, et al. Design of Randomized, Double-Blind, Evaluation in Secondary Stroke Prevention Comparing the EfficaCy and Safety of the Oral Thrombin Inhibitor Dabigatran Etexilate vs. Acetylsalicylic Acid in Patients With Embolic Stroke of Undetermined Source (RE-SPECT ESUS). Int J Stroke 2015;10:1309-12.
  21. Hart RG, Catanese L, Perera KS, Ntaios G, Connolly SJ. Embolic Stroke of Undetermined Source: A Systematic Review and Clinical Update. Stroke 2017;48:867-72.
  22. Sanna T, Diener HC, Passman RS, et al. Cryptogenic Stroke and Underlying Atrial Fibrillation. N Engl J Med 2014;370:2478-86.
  23. Chao TF, Liu CJ, Chen SJ, et al. CHADS2 Score and Risk of New-Onset Atrial Fibrillation: A Nationwide Cohort Study in Taiwan. Int J Cardiol 2013;168:1360-3.
  24. Saliba W, Gronich N, Barnett-Griness O, Rennert G. Usefulness of CHADS2 and CHA2DS2VASc Scores in the Prediction of New-Onset Atrial Fibrillation: A Population-Based Study. Am J Med 2016;129:843-9.
  25. Reiffel JA, Verma A, Kowey PR, et al. Incidence of Previously Undiagnosed Atrial Fibrillation Using Insertable Cardiac Monitors in a High-Risk Population: The REVEAL AF Study. JAMA Cardiol 2017;2:1120-7.
  26. Pagola J, Juega J, Francisco-Pascual J, et al. Yield of Atrial Fibrillation Detection With Textile Wearable Holter From the Acute Phase of Stroke: Pilot Study of Crypto-AF Registry. Int J Cardiol 2018;251:45-50.
  27. Carrazco C, Golyan D, Kahen M, Black K, Libman RB, Katz JM. Prevalence and Risk Factors for Paroxysmal Atrial Fibrillation and Flutter Detection After Cryptogenic Ischemic Stroke. J Stroke Cerebrovasc Dis 2018;27:203-9.
  28. Frontera A, Carpenter A, Ahmed N, et al. Demographic and Clinical Characteristics to Predict Paroxysmal Atrial Fibrillation: Insights From an Implantable Loop Recorder Population. Pacing Clin Electrophysiol 2015;38:1217-22.
  29. Israel C, Kitsiou A, Kalyani M, et al. Detection of Atrial Fibrillation in Patients With Embolic Stroke of Undetermined Source by Prolonged Monitoring With Implantable Loop Recorders. Thromb Haemost 2017;117:1962-9.
  30. Diederichsen SZ, Haugan KJ, Køber L, et al. Atrial Fibrillation Detected by Continuous Electrocardiographic Monitoring Using Implantable Loop Recorder to Prevent Stroke in Individuals at Risk (The LOOP Study): Rationale and Design of a Large Randomized Controlled Trial. Am Heart J 2017;187:122-32.
  31. Uphaus T, Weber-Krüger M, Grond M, et al. Development and Validation of a Score to Detect Paroxysmal Atrial Fibrillation After Stroke. Neurology 2019;92:e115-e124.
  32. Wachter R, Gröschel K, Gelbrich G, et al. Holter-electrocardiogram-monitoring in Patients With Acute Ischaemic Stroke (Find-AF RANDOMISED): An Open-Label Randomised Controlled Trial. Lancet Neurol 2017;16:282-90.
  33. Brachmann J, Morillo CA, Sanna T, et al. Uncovering Atrial Fibrillation Beyond Short-Term Monitoring in Cryptogenic Stroke Patients: Three-Year Results From the Cryptogenic Stroke and Underlying Atrial Fibrillation Trial. Circ Arrhythm Electrophysiol 2016;9:e003333.
  34. Glotzer TV, Daoud EG, Wyse DG, et al. The Relationship Between Daily Atrial Tachyarrhythmia Burden From Implantable Device Diagnostics and Stroke Risk: The TRENDS Study. Circ Arrhythm Electrophysiol 2009;2:474-80.
  35. Hohnloser SH, Capucci A, Fain E, et al. ASymptomatic Atrial Fibrillation and Stroke Evaluation in Pacemaker Patients and the Atrial Fibrillation Reduction Atrial Pacing Trial (ASSERT). Am Heart J 2006;152:442-7.
  36. Martin DT, Bersohn MM, Waldo AL, et al. Randomized Trial of Atrial Arrhythmia Monitoring to Guide Anticoagulation in Patients With Implanted Defibrillator and Cardiac Resynchronization Devices. Eur Heart J 2015;36:1660-8.
  37. Sacco RL, Prabhakaran S, Thompson JLP, et al. Comparison of Warfarin Versus Aspirin for the Prevention of Recurrent Stroke or Death: Subgroup Analyses From the Warfarin-Aspirin Recurrent Stroke Study. Cerebrovasc Dis 2006;22:4-12.
  38. Hart RG, Veltkamp RC, Sheridan P, et al. Predictors of Recurrent Ischemic Stroke in Patients With Embolic Strokes of Undetermined Source and Effects of Rivaroxaban Versus Aspirin According to Risk Status: The NAVIGATE ESUS Trial. J Stroke Cerebrovasc Dis 2019;28:2273-9.
  39. Diener HC, Sacco RL, Easton JD, et al. Dabigatran for Prevention of Stroke After Embolic Stroke of Undetermined Source. N Engl J Med 2019;380:1906-17.
  40. Sanna T, Ziegler PD, Crea F. Detection and Management of Atrial Fibrillation After Cryptogenic Stroke or Embolic Stroke of Undetermined Source. Clin Cardiol 2018;41:426-32.
  41. Afzal MR, Mease J, Koppert T, et al. Incidence of False-Positive Transmissions During Remote Rhythm Monitoring With Implantable Loop Recorders. Heart Rhythm 2020;17:75-80.
  42. Echt DS, Liebson PR, Mitchell LB, et al. Mortality and Morbidity in Patients Receiving Encainide, Flecainide, or Placebo. The Cardiac Arrhythmia Suppression Trial. N Engl J Med 1991;324:781-8.
  43. Schnabel RB, Haeusler KG, Healey JS, et al. Searching for Atrial Fibrillation Poststroke: A White Paper of the AF-SCREEN International Collaboration. Circulation 2019;140:1834-50.

Clinical Topics: Anticoagulation Management, Arrhythmias and Clinical EP, Cardiac Surgery, Heart Failure and Cardiomyopathies, Invasive Cardiovascular Angiography and Intervention, Noninvasive Imaging, Anticoagulation Management and Atrial Fibrillation, Implantable Devices, SCD/Ventricular Arrhythmias, Atrial Fibrillation/Supraventricular Arrhythmias, Cardiac Surgery and Arrhythmias, Cardiac Surgery and Heart Failure, Interventions and Imaging, Nuclear Imaging, Sleep Apnea

Keywords: Arrhythmias, Cardiac, Atrial Fibrillation, Risk Factors, American Heart Association, Cardiovascular Diseases, Stroke, Myocardial Infarction, Myocardial Revascularization, Sleep Apnea Syndromes, Radiography, Anticoagulants

< Back to Listings