2024 Volume 31 Issue 12 Pages 1641-1651
Cryptogenic stroke (CS) accounts for approximately one-fourth of acute ischemic strokes, with most cases derived from embolic etiologies. In 2014, embolic stroke of undetermined source (ESUS) was advocated and the efficacy of anticoagulant therapy was anticipated. However, 3 large-scale clinical trials failed to demonstrate the superiority of direct oral anticoagulants (DOACs) over aspirin, potentially due to the heterogeneous and diverse pathologies of ESUS, including paroxysmal atrial fibrillation (AF), arteriogenic sources such as nonstenotic carotid plaque and aortic complicated lesion (ACL), patent foramen oval (PFO), and nonbacterial thrombotic endocarditis (NBTE) related to active cancer.
Transesophageal echocardiography (TEE) is one of the most effective imaging modalities for assessing embolic sources in ESUS and CS. The Mechanisms of Embolic Stroke Clarified by Transesophageal Echocardiography for Embolic Stroke of Undetermined Source/Cryptogenic Stroke (CHALLENGE ESUS/CS) registry is a multicenter registry that enrolled consecutive patients with CS who underwent TEE at 8 hospitals in Japan between April 2014 and December 2016. Their mean age was 68.7±12.8 years, and 455 patients (67.2%) were male. The median National Institutes of Health Stroke Scale (NIHSS) score was 2. Since 7 analyses have been conducted from each institution to date, novel and significant insights regarding embolic origins and pathophysiologies of ESUS and CS were elucidated from this multicenter registry. This review discusses the diagnosis and treatment of ESUS and CS, tracing their past and future directions. Meaningful insights from the CHALLENGE ESUS/CS registry are also referenced and analyzed.
Since 1988, when Mohr JP originally introduced the concept of cryptogenic stroke (CS), it had been well known that certain number of ischemic strokes had no common accepted cause and concealed specific causes such as patent foramen ovale (PFO) observed cardiac echography1). As a classification for strokes, the National Institute of Neurological Disorders and Stroke (NINDS) classification of cerebrovascular disease Ⅲ and the Trial of Org 10172 in Acute Stroke Treatment (TOAST) has been universally used for the purpose of comprehending the pathophysiology and treatment strategy2, 3). Although there were some differences among reports, strokes of undetermined etiology accounted for 12–39%, approximately one-fourth in these categories4-11).
Owing to recent advances in diagnostic imaging and understanding of pathophysiology, many strokes of undetermined etiology have an embolic mechanism, and anticoagulant therapy is expected to be efficacious. In 2014, the concept of embolic stroke of undetermined source (ESUS) was proposed12). The notion of ESUS encompasses various embolic sources, such as latent paroxysmal atrial fibrillation (AF), venous thrombosis via paradoxical cerebral embolism, arteriogenic plaques including aortic complicated lesions (ACL), and Trousseau syndrome due to occult malignancy. The epidemiological characteristics of ESUS were clarified and 2 large-scale randomized clinical trials (RCTs) using direct oral anticoagulants (DOACs) were conducted. However, the New Approach Rivaroxaban Inhibition of Factor Xa in a Global Trial vs. ASA to Prevent ESUS (NAVIGATE-ESUS) trial and the 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) trial failed to show comparable benefits of DOACs to aspirin13, 14). On the contrary, major bleeding including hemorrhagic stroke were increased in the rivaroxaban group in the NAVIGATE ESUS trial13). Moreover, apixaban for treatment of embolic stroke of undetermined source (ATTICUS randomized trial), a recently conducted trial, failed to show the superiority of Xa inhibitor to aspirin targeting in ESUS patients with at least 1 predictive factor for AF or PFO15). One possible explanation for these consequences was that ESUS possesses diverse embolic sources, as described above, and that anticoagulants are not effective for all of them. Nevertheless, the concept of ESUS itself is now considered unreliable as a therapeutic strategy based on the results of RCTs.
Transesophageal echocardiography (TEE) is one of the most useful imaging modalities for estimating embolic sources and deciding the therapeutic strategies for 1 in 3 to 7 patients with ESUS and CS16-18). Thus, we organized the Mechanisms of Embolic Stroke Clarified by Transesophageal Echocardiography for Embolic Stroke of Undetermined Source/Cryptogenic Stroke (CHALLENGE ESUS/CS) registry, a retrospective multicenter registry enrolling consecutive patients with CS who underwent TEE at 8 institutes in Japan19). From this multicenter registry, 7 analyses were performed to clarify various insights regarding embolic sources and pathophysiologies of ESUS and CS. This review outlines and discusses the current state and future directions of the diagnosis and treatment of ESUS and CS, along with important insights from the CHALLENGE ESUS/CS registry.
The CHALLENGE ESUS/CS registry is a multicenter retrospective registry that enrolled consecutive patients with CS who underwent TEE at 8 hospitals in Japan between April 2014 and December 2016. The inclusion criteria were 1) age ≥ 20 years within 7 days of the onset of stroke; 2) non-lacunar stroke on neuroradiological imaging; 3) absence of ≥ 50% arterial stenosis or occlusion in a corresponding large artery; 4) absence of major embologenic cardiac diseases; and 5) absence of other determined stroke etiologies19). These inclusion criteria were almost the same as the ESUS criteria of large-scale clinical trials, except that a small number of patients could not undergo transthoracic echocardiography (TTE) and that all patients received TEE instead. Patients with paroxysmal AF within 24 hours of admission were excluded from the study. Ultimately, 677 subjects were enrolled in the CHALLENGE ESUS/CS registry. Their mean age was 68.7±12.8 years, and 455 patients (67.2%) were male. The median National Institutes of Health Stroke Scale (NIHSS) score was 2 points19). These characteristics were consistent with the meta-analysis of previous epidemiological studies on ESUS20-24). That is, although the neurological deficit of ESUS is not severe, the frequency of recurrence is not rare in patients treated with general antiplatelet therapy.
Although the Cryptogenic Stroke/ESUS International Working Group suggested various embolic sources of ESUS, several principal embolic sources were especially highlighted in the CHALLENGE ESUS/CS registry (Table 1)12). In this article, these embolic sources are introduced to provide novel insights.
| Potential embolic sources | Insights from CHALLNGE ESUS/CS registry |
|---|---|
|
•Minor-risk cardioembolic sources Aortic and mitral valve disorders, atrial dysrhythmias and stasis, atrial structural abnormalities, ventricular dysfunctions |
•ASA had more stroke recurrences during hospitalization. |
| •Covert paroxysmal AF | •While AF detected within four days was related to spontaneous contrast echo in left atrium, AF detected in five days or later had larger infarcts ≥ 3 cm. |
|
•Cancer-associated Covert NBTE, tumor emboli from occult cancer |
•While active cancer was related to multiple vascular territory infarctions, inactive cancer had more atherosclerotic embolic sources represented with ACL. |
|
•Arteriogenic emboli Aortic arch atherosclerotic plaques, cerebral artery non-stenotic plaques with ulceration |
•CHADS2 and CHA2DS2-VASc score ≥ 2 points were associated with aortic complicated lesion ACL. •Deep and diffuse type cerebral microbleeds was related to ACL. |
|
•Paradoxical embolism PFO, atrial septal defect, pulmonary arteriovenous fistula |
•Patients with PFO and low RoPE score had larger atrial enlargement. •Patients with high risk PFO had less aortic and cardiac embolic sources regardless of their ages. |
ESUS, embolic stroke of undetermined source; CS, cryptogenic stroke; ASA, atrial septal aneurysm; AF, atrial fibrillation; NBTE, nonbacterial thrombotic endocarditis; ACL, aortic complicated lesion; PFO, patent foramen ovale; RoPE; Risk of Paradoxical Embolism
A previous autopsy study with 500 specimens of stroke revealed that aortic arch atheroma with ulcerative lesions was found in 26% of the whole stroke patients and 61% of stroke of unknown etiology25). Also, in a case-control study using TEE, the frequency of aortic arch atheroma ≥ 4 mm was 28.8% in stroke of unknown etiology, 14.4% in other stroke subtypes, and 2% in control group26). Namely, CS patients had the aortic atheromatous lesions much more frequently than other stroke subtypes. Pathologically, when the aortic plaque ruptures, cholesterol crystals and secondary formed thrombi cause cerebral and systemic embolism27, 28). In particular, aortic atheromatous lesions with plaque thickness ≥ 4 mm and mobile or ulcerative lesions were defined as aortic complicated lesions (ACLs), and were considered to be associated with a high risk of embolism (Fig.1)26, 29-32).
Complicated aortic lesion with mobile components (arrow) and ulcerative plaques (arrowhead) detected on transesophageal echocardiography.
In CHALLENGE ESUS/CS, ACL was one of the most crucial embolic sources observed in 38 % patients19). In this registry, some novel factors were associated with the existence of ACL. First, we focused on the commonly used CHADS2 and CHA2DS2-VASc scores to predict the presence of ACL in patients with CS. In patients registered in the CHALLNEGE ESUS/CS, the frequency of ACL increased along with the CHADS2 score (19%, 0 points; 33%, 1; 40%, 2; 49%, 3; 59%, 4-6; p＜0.001), CHA2DS2-VASc score (11%, 0 points; 19%, 1; 35%, 2; 39%, 3; 45%, 4; 55%, 5–8; p＜0.001) (Fig.2). Furthermore, a multiple logistic regression analysis showed that ACL was an independent risk factor for CHADS2 score ≥ 2 (plaque thickness ≥ 4 mm, odds ratio [OR] 2.25, 95% confidence interval [CI] 1.51-3.36, ulcerative or mobile plaque, OR 2.37, 95% Cl 1.38-4.06) and CHA2DS2-VASc score ≥ 2 points (plaque thickness ≥ 4 mm, OR 3.88, 95% Cl 2.07-7.27, ulcerative, mobile plaque, OR 3.25, 95% Cl 1.44- 7.34)19). Meanwhile, CHADS2 score ≥ 2 points and CHA2DS2-VASc score ≥ 2 points were independently associated with paroxysmal AF detected by electrocardiogram (ECG) monitoring and Holter ECG during hospitalization. Therefore, if the CHADS2 or CHA2DS2-VASc score was ≥ 2 points, the underlying cause of CS was ACL scrutinized with TEE or PAF during hospitalization. In other words, for high-risk patients with embolism, a bidirectional approach for determining the etiology is considered indispensable.
A: Patent foramen ovale (PFO) with microbubbles ≥ 25 (arrow) under a microbubble test with the Valsalva maneuver observed by transthoracic echocardiography. B: PFO with atrial septal aneurysm and color flow (arrowhead) observed by transesophageal echocardiography.
Second, we analyzed the underlying relationship between CS and cerebral microbleeds (CMBs) detected using brain MRI. A sub-analysis of the NAVIGATE-ESUS trial showed that the existence of CMBs did not significantly change the risk of bleeding with rivaroxaban33). However, the pathophysiology of CS patients with CMBs has never been discussed. The prevalence of CMBs depended on the types of stroke, and they were detected relatively more frequently in patients with intracranial hemorrhage (60–79%) and lacunar infarction (36–62%) than in those with atheromatous thrombotic brain infarction (ATBI, 21–46%) and cardiogenic embolism (6–30%)34-37). In CHALLENGE ESUS/CS, CMBs were found in 209 (32%) of 661 patients in whom CMBs could be evaluated with brain MRI with the GRE T2* sequence38). In the logistic regression analysis, age, male, hypertension, chronic kidney disease, deep subcortical white matter hyperintensity, and periventricular white matter hyperintensity with brain MRI (Fazekas grade 2–3) were also associated with the CMBs group. Although no significant difference was observed, ACL also tended to be associated with CMBs (OR 1.52. 95% CI 1.00–2.33). CMBs are generally classified into deep or diffuse types resulting from hypertensive and atherosclerotic microangiopathy and strictly lobar types related to cerebral amyloid angiopathy (CAA). In our sub-analysis, ACL (OR 1.78, 95% Cl 1.12–2.84) in the deep or diffuse type CMBs group and prior anticoagulant use (OR 7.88, 95% Cl 1.83–33.9) in strictly lobar type CMBs group were independently associated factors. Through this sub-analysis, some material insights were obtained. First, the prevalence of CMBs in CS was moderate, probably because of the existence of heterogeneous pathologies. Second, deep- or diffuse-type CMBs were particularly related to the presence of ACL. Finally, strictly lobar-type CMBs related to CAA pathology were associated with prior anticoagulant use. Thus, in CS patients with CMBs, anticoagulant therapy might be ineffective, not only due to the hemorrhagic risk but also due to atherosclerotic pathology38).
As rivaroxaban was not effective for aortic atheromatous lesions in the sub-analysis of NAVIGATE-ESUS39), it could be practical to utilize the CHADS2/CHA2DS2-VASC scores and CMBs to identify CS patients with latent ACLs.
Stroke is a common complication in cancer patients. In 1865, Trousseau first described Trousseau syndrome as an unanticipated thrombophlebitis that occurs with occult visceral malignancy. Recently, the term has been applied to various clinical scenarios for any kind of coagulopathy performed within the context of any malignancy40), and has been highlighted as a notable component of CS41-43). This is also defined as cancer-associated thrombosis (CAT). Cancer-associated stroke is a characteristic finding of hypercoagulation, such as multiple lesions in more than 3 vascular territories and a high concentration of D-dimer44, 45). Blood coagulation abnormalities through the mucin produced by cancer cells and tissue factors expressed on the surface of tumor cells are some of the mechanisms by which malignant tumors cause thrombosis and cerebral infarctions. However, the pathogenesis of cancer-associated strokes is multifactorial. In addition to the direct infiltration of tumor cells, cardiac origins due to NBTE and paradoxical embolism via PFO are common causes of ischemic stroke46-48).
In CHALLENGE ESUS/CS, we focused not only on the presence of active cancer but also on the relationship between inactive cancer and stroke. As a result, 41 patients (6.1%) had active cancer and 51 patients (7.5%) had inactive cancer49). A multinomial logistic regression analysis showed that multiple lesions in multiple vascular territories (OR 2.73, 95% CI 1.39–5.40) and CRP (OR 1.10, 95% CI 1.01–1.19) were identified as independent factors in the active cancer group (relative to a non-cancer group), while age (OR 1.05, 95% CI 1.01–1.08), contralateral carotid artery stenosis (OR 4.05, 95% CI 1.60–10.27), ACL in arch (OR 2.13, 95% CI 1.11–4.10), and aortic valve calcifications (OR 2.10, 95% CI 1.09–4.05) were identified as significant factors in the inactive cancer group49). Namely, the active cancer group showed clinical findings consistent with Trousseau syndrome, which presented with conventional cerebral embolism. On the other hand, the inactive cancer group was characterized by severe atherosclerotic lesions and valve calcifications, which suggested that malignancy and atherosclerosis share a mutual pathophysiology in risk factors such as diabetes and smoking, environmental factors, and molecular levels, such as oxidative stress and inflammation50). In addition, it is speculated that chemotherapy and radiation therapy could induce vascular endothelial damage leading to atherosclerosis and valve stiffness51-53). Thus, past medical history of cancer and chemoradiation therapy are also factors related to atherosclerotic changes (including ACL) on TEE.
Paroxysmal AF is one of the most important embolic sources of CS and ESUS. Previous studies reported that among the various causes of emboli, the risk of recurrence of stroke is the highest in patients with CS and ESUS54). Short-term monitoring, such as telemetry and Holter ECG during hospitalization, showed that approximately 8% of patients with CS had latent AF55, 56). The levels of D-dimer and BNP have been reported to have an association with AF detected during hospitalization57). Of the subjects enrolled in the CHALLENGE ESUS/CS registry, paroxysmal AF was found during hospitalization in 64 (9%) patients58). It was most common on the 2nd day after admission. On the other hand, there were scattered cases in which AF was identified at more than 1 week after admission. We hypothesized that AF discovered in the early and late phases after admission would have different characteristics. In the early phase, defined as within 4 days of admission, 37 patients had paroxysmal AF. In this group, the detection of moyamoya echo in the left atrium by TEE was independently associated with the detection of AF (OR 5.91, 95% CI, 2. 19–15.97). Meanwhile, in the late phase, defined as ＞5 days from admission, latent AF was found in 27 patients. Large infarcts with a major axis of ≥ 3 cm were associated with the detection of AF in the late phase (OR 3.28, 95% CI 1.35–7.97)58). These factors could be a potential indicator for recommending the placement of an implantable loop recorder.
The prevalence of PFO in CS patients is approximately 40%, and it is particularly common in young adults18, 59). The risk of paradoxical embolism (RoPE) score can classify PFO patients with CS according to the relevance to the pathogenesis of index stroke60). We hypothesized that PFO patients with low RoPE scores would be more likely to have other cardiac abnormalities, such as left atrial enlargement and high BNP in comparison to patients with high RoPE scores. Based on this supposition, 300 subjects with PFO positivity in the CHALLENGE ESUS/CS registry were divided into a high RoPE score group (＞6 points, 32 cases) and a low RoPE score group (≤ 6 points, 268 cases). We compared the clinical features and cardiogenic markers such as left atrial enlargement, high concentration of BNP, left ventricular ejection fraction, spontaneous echo contrast, E/e’ indicating left ventricular diastolic function61). BNP, left atrial enlargement, E/e’, and left atrial appendage blood flow velocity reduction were significantly associated with low RoPE scores, while left atrial enlargement was an independent predictor of low RoPE score (OR 1.15, 95% CI 1.00–1.32). In CS patients with PFO and RoPE score ≤ 6, cardiogenic markers, such as left atrial enlargement, were prevalent, suggesting the possibility of cardiogenic risk factors other than PFO61).
We analyzed the relationship between the anatomical features of PFO and other embolic sources. In this study, patients in the CHALLENGE ESUS/CS registry were divided into a high-risk PFO group (large shunt PFO defined as ≥ 25 microbubbles and PFO with atrial septal aneurysm [ASA]) (Fig.2), a right-to-left shunt (RLS) group (PFO ＜25 microbubbles without ASA), and a no-RLS group62). As a result, 91, 221, and 342 patients were classified into the high-risk PFO, RLS, and no-RLS groups, respectively. In a multinomial logistic regression analysis, male sex (OR 1.83, 95% CI 1.07–3.12) was independently associated with high-risk PFO, while hypertension (OR 0.56, 95% CI 0.33–0.97), multiple infarctions (OR 0.60, 95% CI 0.44–0.83), and other cardiac and aortic emboligenic risk factors (OR 0.51, 95% CI 0.29–0.90) were inversely associated with high-risk PFO relative to non-RLS. Importantly, in 517 patients of ≥ 60 years of age, multiple infarctions (OR 0.55, 95% CI 0.38–0.79) and other cardiac and aortic embolic risk factors (OR 0.52, 95% CI 0.29–0.96) were still inversely associated with high-risk PFO62). According to the results of the study, high-risk PFO seemed to have fewer other embolic sources and high-risk PFO itself could be a culprit of the index stroke even in patients ＞60 years of age. Although the PFO occluder device is currently only indicated for patients ＜60 years of age based on the inclusion criterion of RCTs63-65), based on these analyses, it could be expanded to patients of ＞60 years of age with anatomically and clinically high-risk PFOs.
ASA is an abnormality of the atrial septum that is associated with PFO. ASA is generally detected in 0.2–4% by TTE and 2–8% by TEE66, 67). In CS patients, the frequency of ASA increases to 6.4–39.1%18, 68). Although the coexistence of ASA and PFO increases the risk of stroke recurrence, few studies have focused on ASA, and the relationship between ASA and CS has not been fully elucidated. Among 671 subjects analyzed in CHALLENGE ESUS/CS, ASA was detected in 92 patients (14%), and patients with ASA were older (72.4±11.0 years), had a higher incidence of RLS (66%), and a lower incidence of diabetes (16%)69). Additionally, stroke recurrence during hospitalization was more common in ASA patients (8%). A logistic regression analysis revealed that ASA was independently associated with in-hospital recurrence (OR 3.26, 95% CI 1.21–8.74). This analysis revealed that ASA is not rare in patients with CS and that it has distinctive clinical features.
There are a few possible explanations for the relationship between ASA and stroke recurrence. First, some autopsy and surgical reports have found blood clots in the ASA sac70, 71). Therefore, a thrombus that cannot be detected by TEE might cause stroke recurrence. Second, the patients with ASA in the present study may have been older and had more coexisting embolic sources and systemic atherosclerosis. Lastly, the existence of ASA may induce changes in the electrophysiological substrate and physiological dysfunction of the left atrium, leading to atrial abnormality and thromboembolism72, 73). From these points of view, even isolated ASA could be a more essential embolic source than we generally thought.
First, the pragmatic concept of ESUS was constructed based on the expectation of a unified therapeutic strategy. However, none of the three large-scale randomized clinical trials, NAVIGATE-ESUS, RE-SPECT ESUS, and ATTICUS, showed the superiority of anticoagulants with DOACs to aspirin13-15). Hence, the concept of ESUS has become less relevant with time. As previously mentioned, ESUS and CS truly encompass multiple and diverse embolic origins and pathophysiologies. Thus, anticoagulants may be effective for some of the embolic sources represented by latent paroxysmal AF, while others require more specific remedies, such as lipid-modifying therapy for ACL, closure device placement for high-risk PFO, and cancer treatment for Trousseau’s syndrome.
From these perspectives, the concepts of ESUS and CS must be altered in a more practical manner. One promising strategy is to intensify the therapeutic strategies. In the Cardiovascular Outcomes for People Using Anticoagulation StrategieS (COMPASS) trial, low-dose rivaroxaban (2.5 mg) twice daily plus aspirin was more effective for secondary cardiovascular prevention among patients with atherosclerotic vascular diseases relative to aspirin or rivaroxaban monotherapy74). While the embolic origins of ESUS and CS may include not only paroxysmal AF, but also atherosclerotic vascular diseases such as ACL and nonstenotic carotid plaques, the combination of low-dose DOAC and aspirin may reduce the recurrence of embolic events relative to aspirin or DOAC monotherapy. However, the COMPASS trial also indicated that the combination therapy of low-dose DOAC and aspirin increased the risk of major bleeding events relative to aspirin or DOAC monotherapy74). Therefore, the assessment of bleeding risk in patients with ESUS and CS using hemorrhagic risk scores and the burden of CMBs would be indispensable for the implementation of these intensive combination therapies. The recently developed Ⅺa inhibitor is a prospective anticoagulant agent that was associated with fewer ischemic events and equivalent hemorrhagic complications when administered for secondary prevention of non-cardiogenic strokes in combination with antiplatelet therapy relative to a placebo in a phase 2 trial75). The use of novel anticoagulants with lower risk of bleeding can also support the use of combined therapies with antiplatelets for CS.
Another prospective reconstructive approach to the ESUS concept is to distinguish the specific pathophysiology in which anticoagulants are truly effective from other diverse embolic sources. One such attempt is the concept of atrial cardiopathy, which results in thrombus formation and embolism due to pathological changes in the atrium in the absence of AF76). AtRial Cardiopathy and Antithrombotic Drugs In Prevention After Cryptogenic Stroke (ARCARDIA) was a randomized clinical trial to compare apixaban vs. aspirin therapy for secondary stroke prevention in patients with CS and evidence of atrial cardiopathy77). As a biomarker of atrial cardiopathy, P-wave terminal force ＞5000 µV * ms in electrocardiogram lead V1, serum N-terminal pro-B-type natriuretic peptide ＞250 pg/mL, and left atrial diameter index ≥ 3 cm/m2 on an echocardiogram were applied. The trial was interrupted due to futility after conducting an interim analysis with 1015 participants enrolled, and an average follow-up period of 1.8 years. Since the rate of recurrent stroke did not differ between the apixaban group and the aspirin groups (annualized rate both 4.4%), despite expectations, apixaban could not reduce the risk of recurrent stroke relative to aspirin in patients with CS and evidence of atrial cardiopathy77).
In discussion, the confounding effects of latent AF and atherosclerotic embolic sources which could not have been excluded in the previous reports regarding atrial cardiopathy78, 79). In addition, it was unclear whether the biomarkers for atrial cardiopathy that were applied in this study were appropriate. Several other biomarkers representing pathological atrial alteration have been suggested, including midregional proatrial natriuretic peptide, premature atrial contractions, atrial fibrosis, left atrial volume, and left atrial strain, and others80, 81). In addition, the appropriate threshold for the left atrial diameter remains controversial, although the number of patients included due to the left atrial diameter was limited.
The concept of atrial cardiopathy seems imperfect, similar to that of ESUS. It is essential to complete the notion of atrial cardiopathy with the more practical use of biomarkers or to develop novel concepts that supersede atrial cardiopathy and ESUS for differentiating diverse and heterogeneous cryptogenic pathophysiologies.
In this review, which traces the past and future directions of ESUS, an innovative and pragmatic concept of CS, was discussed in relation to the CHALLENGE ESUS/CS registry, which exhibited novel insights in relation to the diverse embolic origins of ESUS and CS observed by TEE. Since 3 clinical randomized trials could not demonstrate the superiority of DOACs for the secondary prevention of ESUS, the concept of CS must be developed. One possible alteration is the enhancement of therapeutic strategies such as the combination of antiplatelet and low-dose DOAC or novel anticoagulants (e.g., Ⅺa inhibitor). Another possibility is the classification of specific pathophysiologies in which anticoagulant therapy is truly efficacious, represented by attempted treatments for atrial cardiopathy. After discovering significant insights regarding diverse embolic sources of CS through TEE, these innovations should be considered constructive challenges rather than reckless defiance.
We profoundly appreciate the staff of cooperating institutions in the CHALLENGE ESUS/CS registry, especially Dr. Tateishi and Prof. Tsujino in Nagasaki University Hospital; Dr. Kuriki and Dr. Kamiya in Showa University Koto Toyosu Hospital; Dr. Doijiri in Iwate Prefectural Central Hospital; Dr. Shimizu and Prof. Hasegawa at St. Marianna University School of Medicine; Prof. Takekawa and Prof. Hirata at Dokkyo Medical University; Prof. Koga and Dr. Yamaguchi and Prof. Ihara in National Cerebral and Cardiovascular Center; Dr. Kanemaru and Prof. Terashi and Prof. Aizawa in Tokyo Medical University Hospital; Dr. Shimada and Prof. Urabe in Juntendo University Urayasu Hospital; and Prof. Hattori of Juntendo University Faculty of Medicine.
None.
YU received lecture fees from OHARA Pharmaceutical Co., Ltd. and Daiichi Sankyo Co.
