A new risk model of assessing left atrial appendage thrombus in patients with atrial fibrillation – Using multiple clinical and transesophageal echocardiography parameters

doi:10.1016/j.ijcard.2020.04.039

International Journal of Cardiology

Volume 314, 1 September 2020, Pages 60-63

https://doi.org/10.1016/j.ijcard.2020.04.039 Get rights and content

Highlights

•
A new prediction model for left atrial appendage thrombus
•
Key risk factors for left atrial appendage thrombus formation
•
Complex roles of mitral regurgitation in atrial fibrillation
•
Integration of clinical and echo parameters for decision making before cardioversion

Abstract

Backgrounds

Predicting left atrial appendage thrombus (LAAT) in non-valvular atrial fibrillation (NVAF) patients need more precisely quantified risk models. In this study, we attempted to review the risk markers for LAAT and develop a simple and reliable model for LAAT prediction.

Methods

The study included 307 patients with NVAF who were scheduled for transesophageal echocardiography (TEE) to exclude LAA thrombus before synchronized electrical cardioversion or radiofrequency ablation for atrial fibrillation (AF). We analyzed the relationship between echo, clinical parameters and the presence or absence of LAAT.

Results

A total of 33 patients were found having LAAT (10.7%, 33/307). The age, left atrial appendage emptying velocity (LAAEV), left atrial or left atrial appendage spontaneous echocardiographic contrast (SEC), less than moderate to severe mitral regurgitation (≤mild MR), and left atrial enlargement showed association with LAAT. The multivariate logistic regression analysis revealed that LAAEV, SEC and ≤mild MR were independent risk factors of the LAAT. We used LAAEV ≤ 21.5 cm/s, SEC and ≤mild MR to construct a combined predictive model for LAAT in NVAF patients (the area under receiver operator characteristic curve: 0.88; 95% confidence interval: 0.82–0.95, P < 0.0001).

Conclusion

Comprehensive evaluation of LAAEV, SEC, and MR with associated LAAT may help risk stratifying the NVAF patients, especially if the LAA imaging quality was suboptimal for identifying thrombus. These parameters may facilitate the decision-making process at the time of TEE.

Introduction

It has been estimated that nearly 2.3 million Americans have non-valvular atrial fibrillation (NVAF) [1]. Atrial fibrillation (AF) is related to the increased risk of thromboembolic stroke with thrombus most likely located in the left atrial appendage (LAA) [2]. Transesophageal echocardiography (TEE) is regarded as the gold standard in detecting the LA thrombus with 97% sensitivity and almost 100% specificity [3]. Therefore, TEE has been used routinely to exclude the presence of LA thrombus in patients with recent-onset AF that had lasted >48 h or in the setting of sub-therapeutic anticoagulation before cardioversion [4]. However, for some patients with poor echo image quality, it is difficult to distinguish LAA thrombus from pectinate muscles, necessitating an improved risk assessment method to avoid the missed diagnosis of the potential thrombus.

As for TEE parameters, LAA peak emptying flow velocity (LAAV) has been proposed as an echocardiographic predictor of LAA thrombus [5,6]. However, there are still controversies about its predictive ability and no uniform standard for its best cut-off value [7,8]. Several studies have attempted to set up a novel method to predict the presence or absence of LAA thrombus using clinical and/or TEE parameters [9,10]. But in current practice, no commonly accepted standard has been applied for clinical use due to the complex nature of the available data. In this study, we aimed to develop a simpler and reliable model that might improve the prediction of the presence of LAA thrombus based on clinical factors and TEE parameters. This model may help with prompt decision-making in managing the cardioversion protocols of patients with NVAF.

Section snippets

Study design and patient population

This single-center retrospective study included 405 consecutive patients with NVAF undergoing TEE from July 2013 to June 2014 for the detection of LAA thrombus before cardioversion or radiofrequency ablation for atrial fibrillation. The cardiac rhythm at the time of TEE was 100% in atrial fibrillation. We did not include patients with chronic atrial fibrillation. Ninety-eight patients had no measured LAA velocity or inadequate echocardiographic view and therefore were excluded from the study.

Baseline clinical characteristics

Baseline patient characteristics, comorbidities, and medication at the time of TEE are shown in Table 1. The mean age of all NVAF patients was 67.1 years, and 73.6% of the patients were male. Among the 307 NVAF patients, thrombus within the LAA was identified by TEE in 33 (10.7%). Compared with the non-LAAT group, the LAAT group had more elderly patients, SEC and LA enlargement. LAA velocity and the frequency of significant MR were lower in patients with LAAT than in non-LAAT group. All the

Discussion

We showed following results in non-valvular atrial fibrillation (NVAF) cohort: patients with LAA thrombus had significantly lower LAA emptying velocities, more frequent spontaneous echo contrast (SEC), ≤mild MR, LA enlargement and advanced age; LAA emptying velocity ≤21.5 cm/s, SEC and ≤mild MR were independent predictors of LAA thrombus (LAAT); The capacity for predicting LAAT of combining these three factors above is stronger than using any one single factor in NVAF patients prior to

Conclusion

Our study adds to the LAAT risk stratification capability and attempts to aid in decision-making process for LAAT in patients with NVAF. Further research utilizing this method with more clinical and imaging data may help design an intelligent risk assessment system and verify patients' outcomes in the real-world clinical setting.

CRediT authorship contribution statement

Lu Chen: Methodology, Formal analysis, Data curation, Writing - original draft. Ashley Zinda: Methodology, Investigation. Nicholas Rossi: Methodology, Investigation. Xiu-Jie Han: Formal analysis, Data curation. Steve Sprankle: Methodology, Investigation. Renee Bullock-Palmer: Methodology, Investigation. Denise Zingrone: Methodology, Investigation. Mark Moshiyakov: Methodology, Investigation. Justin Szawlewicz: Methodology, Investigation. Allen Mogtader: Methodology, Investigation. David Hsi:

Declaration of competing interest

No conflict of interest relevant to this article.

References (25)

W.A. Jaber et al.
Efficacy of anticoagulation in resolving left atrial and left atrial appendage thrombi: a transesophageal echocardiographic study
Am. Heart J.
(2000)
V. Fuster et al.
ACC/AHA/ECS 2006 guidelines for the management of patients with atrial fibrillation
J. Am. Coll. Cardiol.
(2006)
R. Malik et al.
Prevalence and predictors of left atrial thrombus in patients with atrial fibrillation: is transesophageal echocardiography necessary before cardioversion?
Cardiovasc Revasc Med
(2015)
Y. Agmon et al.
Are left atrial appendage flow velocities adequate surrogates of global left atrial function? A population-based transthoracic and transesophageal echocardiographic study
J. Am. Soc. Echocardiogr.
(2002)
K.P. Cohoon et al.
The impact of gender and left atrial blood stasis on adiponectin levels in non-valvular atrial fibrillation
Int. J. Cardiol.
(2015)
R.M. Lang et al.
Recommendations for chamber quantification: a report from the American Society of Echocardiography’s guidelines and standards committee and the chamber quantification writing group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology
J. Am. Soc. Echocardiogr.
(2005)
R.T. Hahn et al.
Guidelines for performing a comprehensive transesophageal echocardiographic examination: recommendations from the American society of echocardiography and the society of cardiovascular anesthesiologists
J. Am. Soc. Echocardiogr.
(2013)
Y. Miyasaka et al.
Mild mitral regurgitation was associated with increased prevalence of thromboembolic events in patients with nonrheumatic atrial fibrillation
Int. J. Cardiol.
(2000)
R. Providência et al.
Cardiac troponin I: prothrombotic risk marker in non-valvular atrial fibrillation
Int. J. Cardiol.
(2013)
M. Anselmino et al.
Left atrial appendage thrombi relate to easily accessible clinical parameters in patients undergoing atrial fibrillation transcatheter ablation: a multicenter study
Int. J. Cardiol.
(2017)

R. Doukky et al.

Impact of diastolic function parameters on the risk for left atrial appendage thrombus in patients with nonvalvular atrial fibrillation: a prospective study

J. Am. Soc. Echocardiogr.

(2016)

E. Bertaglia et al.

NOACs and atrial fibrillation: incidence and predictors of left atrial thrombus in the real world

Int. J. Cardiol.

(2017)

Cited by (8)

Effect of mitral regurgitation on stroke risk in patients with non-rheumatic atrial fibrillation
2022, Archives of Cardiovascular Diseases
Citation Excerpt :
In the first case, several studies have demonstrated a significant reduction in the formation of atrial thrombus or systemic embolism [4–7]. However, in patients without rheumatic valve disease, some studies have pointed to a reduction in the formation of atrial thrombus and systemic embolism [8–14] and other studies have shown neutral results [15–18]. In 2011, Fukuda et al. reported a protective effect of MR on left atrial blood stasis, but this was limited to patients with severe MR [12].
Atrial fibrillation (AF) carries a thrombotic risk related to blood stasis in the left atrium. In patients with rheumatic valve disease and AF, the presence of severe mitral regurgitation (MR) has been shown to reduce the risk of atrial thrombosis and stroke. However, in patients without rheumatic disease, the results are controversial.
To analyse the association between MR and the incidence of stroke in patients with non-rheumatic AF.
We analysed data from the retrospective CardioCHUVI-AF registry, which includes 15,720 patients with AF (without mechanical prostheses or rheumatic valvular disease) in the Vigo area of Spain, during 2014–2018. We grouped the patients according to MR grades: 0–2 (n = 15,194) and 3–4 (n = 526). We performed univariate and multivariable competitive risk analyses to analyse the association between MR and stroke, with death as the competitive event.
During a median (interquartile range [IQR]) follow-up of 4.9 (2.8–4.9) years, 859 patients (5.5%) suffered a stroke. The stroke incidence was 1.3 per 100 person-years (95% confidence interval [95% CI]: 1.2–1.4), with no difference between the MR groups. In univariate analysis, no relationship was observed between MR grade and stroke (subdistribution hazard ratio [sHR]: 1.12, 95% CI: 0.79–1.60; P = 0.53); likewise after multivariable analysis (sHR: 0.98, 95% CI: 0.68–1.41; P = 0.90). This same relationship was evaluated in subgroups of interest (patients with and without: oral anticoagulation, CHA₂DS₂-VASc ≥ 2, prior heart failure, aortic valve disease, left ventricular ejection fraction ≤ 40%, and moderate–severe left atrial dilation), with results consistent with the overall population.
In our large registry of patients with non-rheumatic AF, we did not find a protective effect of grade 3–4 MR on the risk of stroke.
La fibrillation auriculaire (FA) comporte un risque thrombotique lié à la stase sanguine produite dans l’oreillette gauche. Chez les patients atteints de valvulopathie rhumatismale et de FA, il a été démontré que la présence d’une régurgitation mitrale (RM) sévère réduit le risque de thrombose auriculaire et d’accident vasculaire cérébral (AVC). Chez les patients sans maladie rhumatismale, les résultats sont controversés.
D’analyser l’association entre la RM et l’incidence des AVC chez les patients atteints de FA non rhumatismale.
Nous avons analysé les données du registre rétrospectif CardioCHUVI-AF, qui comprend 15 720 patients atteints de FA (à l’exclusion des patients porteurs de prothèses valvulaires et de valvulopathies rhumatismales) dans la région de Vigo en Espagne, de 2014 à 2018. Nous avons regroupé les patients selon les grades RM : 0–2 (n = 15 194) et 3–4 (n = 526). Nous avons effectué une analyse multivariée du risque concurrentiel pour analyser l’association entre la RM et l’AVC avec un modèle de risque compétitif, la mort étant l’événement compétitif.
Au cours d’un suivi médian (intervalle interquartile [IQR]) de 4,9 (2,9–4,9) ans, 859 patients (5,5 %) ont subi un AVC. Le taux d’incidence d’AVC était de 1,3 pour 100 années-personnes (intervalle de confiance à 95 % [IC95 %] : 1,2–1,4), sans différence entres les groupes de RM. Dans l’analyse univariée, aucune relation n’a été observée entre les grades de RM et l’AVC (rapport de risque de sous-distribution [sHR] : 1,12, IC95 % : 0,79–1,60 ; p = 0,53). Après l’analyse multivariée, aucune association significative n’a été observée entre les grades de RM et l’AVC (sHR : 0,98, IC95 % : 0,68–1,41 ; p = 0,90). Cette même relation a été évaluée par sous-groupes d’intérêt (patients avec et sans: anticoagulation orale, CHA₂DS₂-VASc ≥ 2, insuffisance cardiaque antérieure, fraction d’éjection ventriculaire gauche ≤ 40 %, maladie de la valve aortique et dilatation auriculaire gauche), avec des résultats cohérents avec l’échantillon global.
Dans notre vaste registre de patients atteints de FA non rhumatismale, nous n’avons pas trouvé d’effet protecteur d’une RM significative (grades 3–4) sur le risque d’AVC.
Effect of Mitral Regurgitation on Thrombotic Risk in Patients With Nonrheumatic Atrial Fibrillation: A New CHA<inf>2</inf>DS<inf>2</inf>-VASc Score Risk Modifier?
2021, American Journal of Cardiology
Citation Excerpt :
Our observation that MR predominantly affects LASEC formation and not LAAT formation might be related to the fact that MR jets often do not reach the LAA. In the recent and large study by Cresti et al, the incidence of LA thrombus formation was also the same in the group of patients without MR compared with the group with severe MR.19 Our findings concur with previous work showing a reduced risk for atrial thrombosis or cardioembolic events in nonrheumatic AF patients with severe MR.19–23 In all these studies, however, no appropriate correction was made for the CHA2DS2-VASc risk score; therefore, the exact adjusted ORs could not be provided. Inappropriate correction for clinical thrombotic risk factors and/or small study populations are probably the reasons why some other older studies did not find a link between MR and thrombotic risk.24,25
The current study assessed the effect of mitral regurgitation (MR) on thrombotic risk in nonrheumatic atrial fibrillation (AF). AF carries a thrombotic risk related to left atrial blood stasis. The prevalence of atrial thrombosis, defined as the presence of left atrial appendage thrombus and/or left atrial spontaneous echo contrast grade >2, was determined in 686 consecutive nonrheumatic AF patients without (adequate) anticoagulation scheduled for transesophageal echocardiography before electrical cardioversion and was related to the severity of MR adjusted for the CHA₂DS₂-VASc score. A total of 103 (15%) patients had severe MR, 210 (31%) had moderate MR, and 373 (54%) had no-mild MR; the median CHA₂DS₂-VASc score was 3.0 (interquartile range 2.0 to 4.0). Atrial thrombosis was observed in 118 patients (17%). The prevalence of atrial thrombosis decreased with increasing MR severity: 19.9% versus 15.2% versus 11.6% for no-mild, moderate, and severe MR, respectively (p value for trend = 0.03). Patients with moderate and severe MR had a lower risk of atrial thrombosis than patients with no-mild MR, with adjusted odds ratios of 0.51 (95% confidence interval 0.31 to 0.84) and 0.24 (95% confidence interval 0.11 to 0.49), respectively. The protective effect of MR was present across all levels of the CHA₂DS₂-VASc risk score and the presence of moderate-severe MR in patients with an intermediate CHA₂DS₂-VASc score (2 to 3) lowered the atrial thrombotic risk to the level of patients with a low CHA₂DS₂-VASc score (0 to 1). In conclusion, our data show that the presence of MR attenuated the atrial thrombotic risk by more than 50% in patients with nonrheumatic AF.
The Art of 2D Transesophageal Echocardiography: 2D Transesophageal Atlas with Anatomical and Surgical Correlation
2024, The Art of 2D Transesophageal Echocardiography: 2D Transesophageal Atlas with Anatomical and Surgical Correlation
Characteristics, Thrombus Resolution, and Long-Term Outcomes in Patients with Nonvalvular Atrial Fibrillation and Atrial Thrombus
2023, Clinical and Applied Thrombosis/Hemostasis
Echocardiographic predictors of thrombus in left atrial appendage—The role of novel transthoracic parameters
2022, Frontiers in Cardiovascular Medicine
Stroke risk evaluation for patients with atrial fibrillation: Insights from left atrial appendage
2022, Frontiers in Cardiovascular Medicine

View all citing articles on Scopus

View full text

A new risk model of assessing left atrial appendage thrombus in patients with atrial fibrillation – Using multiple clinical and transesophageal echocardiography parameters

Highlights

Abstract

Backgrounds

Methods

Results

Conclusion

Introduction

Section snippets

Study design and patient population

Baseline clinical characteristics

Discussion

Conclusion

CRediT authorship contribution statement

Declaration of competing interest

Am. Heart J.

J. Am. Coll. Cardiol.

Cardiovasc Revasc Med

J. Am. Soc. Echocardiogr.

Int. J. Cardiol.

J. Am. Soc. Echocardiogr.

J. Am. Soc. Echocardiogr.

Int. J. Cardiol.

Int. J. Cardiol.

Int. J. Cardiol.

J. Am. Soc. Echocardiogr.

Int. J. Cardiol.