Intracardiac echocardiography for verification for left atrial appendage thrombus presence detected by transesophageal echocardiography: the ActionICE II study

Abstract

Background

Transesophageal echocardiography (TEE) remains the gold standard for exclusion of left atrial appendage (LAA) thrombus in patients scheduled for direct electrical cardioversion (DEC) or atrial fibrillation (AF) ablation. Recently, intracardiac echocardiography (ICE) of the pulmonary artery (PA) has been shown to provide excellent LAA images and to be useful in verification of equivocal TEE findings.

Hypothesis

ICE of the PA may have a role in detecting false‐positive TEE results.

Methods

Twenty‐one patients (12 male, age 65 ± 8 years, CHADS2VASC2 score [congestive heart failure, hypertension, age ≥ 75, age 65‐74, diabetes mellitus, stroke/TIA/thrombo‐embolism, vascular disease, sex female] = 2.2; HAS‐BLED score [hypertension, abnormal renal and liver function, stroke, bleeding, labile International Normalized Ratio, elderly, drugs or alcohol] = 1.1), in whom a thrombus in the LAA was detected during TEE before DEC or AF ablation, underwent ICE of the PA.

Results

On TEE, in 7 (33%) patients, the LAA thrombus was described as “solid” and in the remaining 14 (67%) as “soft.” Disagreement between the TEE and ICE (thrombus in TEE and no thrombus in ICE) was found in 9 (43%) patients. In the solid thrombus group, ICE confirmed thrombi existence in 6 and excluded thrombi in 1 patient. In the soft thrombus group, ICE confirmed thrombi in 6 patients and excluded thrombi in the remaining 8 patients. Of the demographic and clinical variables, only the longstanding persistent type of AF was significantly associated with the presence of an LAA thrombus detected both by TEE and ICE.

Conclusions

With TEE, a false positive of an LAA thrombus may be indicated, especially when a thrombus is described as soft rather than solid. Our study suggests that ICE may be a valuable option for verification of a TEE‐based diagnosis of a thrombus.

1. INTRODUCTION

Visualization of the left atrium (LA) and the left atrial appendage (LAA) is mandatory to exclude the presence of a thrombus prior to several interventions, including atrial fibrillation (AF) ablation or AF cardioversion. Transesophageal echocardiography (TEE) is regarded as a gold standard in this setting. Compared with intraoperative findings, TEE sensitivity and specificity values reach 92% and 98%, respectively, whereas the positive predictive value is lower at 86%.1, 2

Intracardiac echocardiography (ICE) has recently become a widely used technique during invasive cardiac procedures and has been shown to accurately visualize the LA and LAA, especially when performed from the left pulmonary artery (PA).3 There are growing data in the literature suggesting that ICE can successfully replace TEE, but direct comparisons are scarce.3, 4, 5, 6 These observational studies, performed in centers in which ICE is routinely used, suggested that ICE can reveal the presence of an LAA thrombus in some patients in whom TEE did not show any significant abnormality. These findings led to growing clinical use of ICE for verification of equivocal or even normal TEE results and showed that in 5.7%7 of patients with normal or near normal TEE, there is a clear thrombus visible on ICE. However, the opposite question—whether all LAA thrombi detected by TEE are confirmed on ICE—has not been addressed.

Accordingly, the aim of the present study was to examine whether TEE‐based detection of an LAA thrombus may be a false positive as verified by ICE.

2. METHODS

2.1. Design of the study and patient flow

The study protocol was approved by the ethics committee of the Postgraduate Medical School, Warsaw, Poland. The ActionICE II (EvaluAtion of Predictive Value of Multisite Intracardiac EchoCardiography During Imaging of Structure and funCTION of Left Atrial Appendage in Comparison to Transesophageal Echocardiography II) study (grant number 501‐1‐10‐14‐11, NCT 01371279) was prospectively conducted in our department between August 2012 and February 2015. During this period, all patients undergoing TEE in our echocardiography department prior to AF ablation or elective direct current AF cardioversion were screened, and those in whom a thrombus in the LAA was detected were asked to participate in the ActionICE II study. During the study period, 396 TEE procedures in these 2 groups of patients were performed, and 30 (8%) patients had a TEE‐based diagnosis of an LAA thrombus. Out of these 30 patients, 21 gave the written informed consent to participate in the study and to undergo invasive ICE. The ICE examination was performed within 24 hours of TEE on unchanged anticoagulation therapy in examined patients.

2.2. Transesophageal echocardiography

TEE was performed according to the standard practice guidelines using a multiplane TEE probe 6T (Vivid 9; GE Vingmed Ultrasound, Horten, Norway).

The LAA thrombus was diagnosed if there was a clearly defined circumscribed echo‐dense intracavitary mass that was acoustically distinct from the endocardium and pectinate muscles of the LAA. The presence of a thrombus was corroborated in 2 or more views. Solid thrombus was defined as organized formed thrombus, whereas soft thrombus was described as sludge, gelatinous morphology, not well formed, early stage of thrombus formation.8 Care was taken to differentiate between thrombus and dense echo contrast. Patients with the latter finding were not included in the present study. The TEE‐based diagnosis of an LAA thrombus had to be independently confirmed post hoc by 2 experienced echocardiographers who were blinded to the results of ICE, which was always performed after the TEE study.

2.3. Intracardiac echocardiography

The ICE examination was performed within 24 hours after TEE and was always conducted by the same electrophysiologist (J.B.), and 1 of the echocardiographers who performed the TEE attended the ICE procedures.

The AcuNav electronic phased‐array diagnostic ultrasound catheter (Siemens Medical Solutions, Munich, Germany) (5.5–10 MHz, 8F) was introduced through a 10F or 11F hemostatic sheath, positioned in the left femoral vein and moved under fluoroscopic guidance to the right atrium. The ICE probe was then advanced into the PA. Images obtained by the ICE probe were analyzed online using an echocardiographic system (Cypress; Siemens Medical Solutions). Fluoroscopy was performed with a single‐plane digital angiographic system (Integris; Philips Healthcare, Best, the Netherlands). A detailed description of ICE methodology was described elsewhere.3

2.4. Statistical analysis

The results are presented as mean ± standard deviation or number (%). Comparisons between patients with or without an ICE‐detected LAA thrombus were performed using the Student t test or χ² test (with or without Yates correction), or Fisher test where appropriate. A P value <0.05 was regarded as significant.

3. RESULTS

The study group consisted of 21 patients (12 males, age 65 ± 8 years, CHADS₂VASC₂ score [congestive heart failure, hypertension, age ≥ 75, age 65‐74, diabetes mellitus, stroke/TIA/thrombo‐embolism, vascular disease, sex female] = 2.2 ± 1.9 and HAS‐BLED (hypertension, abnormal renal and liver function, stroke, bleeding, labile International Normalized Ratio, elderly, drugs or alcohol) score = 1.2 ± 0.9). Fifteen patients were scheduled for AF ablation, and the remaining 6 patients were admitted to the hospital for AF cardioversion.

On TEE, in 7 (33%) patients a solid thrombus was detected, and in the remaining 14 (67%) patients there was a soft thrombus. Disagreement between TEE and ICE was found in 9 (43%) patients (thrombus in TEE and no thrombus in ICE). In the solid thrombus group, ICE confirmed thrombus existence in 6 (86%) patients and excluded thrombus in 1 (14%) patient. In the soft thrombus group, ICE confirmed a thrombus in 6 (43%) patients and excluded it in the remaining 8 (57%) patients (Figure 1). Representative examples of TEE and corresponding ICE images are presented in Figure 2.

Figure 1.

Flowchart of the patients in the trial.

Figure 2.

Examples of LAA visualization. Patient with a thrombus (arrow) in the LAA in the TEE examination (A) and without in the ICE examination (B). (C, D) Projection of the thrombus in the LAA in both visualization modalities—TEE and ICE—in the same patient. Abbreviations: ICE, intracardiac echocardiography; LAA, left atrial appendage; TEE, transesophageal echocardiography.

Comparison between ICE‐positive and ICE‐negative patients is shown in the Table. There were no significant differences in the demographic and clinical variables between the groups except the proportion of patients with a persistent long‐standing AF, which was significantly higher in patients with an ICE‐detected thrombus. After excluding LAA thrombus by ICE, patients were offered cardioversion or AF ablation. Two patients from each group agreed and underwent procedures without any complications, whereas the remaining 17 patients preferred to change the anticoagulation treatment and undergo repeated echocardiographic assessment prior to cardioversion or AF ablation.

Table 1.

Comparison of demographic and clinical parameters between patients with or without thrombus in the LAA detected by ICE

	ICE Positive	ICE Negative	P Value
N	12	9
Age, y	64 ± 8	65 ± 8	0.7799
CHADS2VASC2	2 ± 1.7	3 ± 1.6	0.1875
Heart failure	3 (25%)	1 (11.1%)	0.6030
Hypertension	7 (58.3%)	8 (88.9%)	0.1778
Diabetes mellitus	2 (16.7%)	4 (44.4%)	0.3310
Stroke/TIA	0	2 (22.2%)	0.1714
Cardiovascular disease	2 (16.7%)	1 (11.1%)	1.000
Female gender	4 (33.3%)	5 (55.6%)	0.3964
HAS‐BLED	0.9 ± 0.93	1.6 ± 0.9	0.0998
Liver/kidney dysfunction	2 (16.7%)	2 (22.2%)	1.000
Major bleeding	1 (8.3%)	2 (22.2%)	0.5534
Unstable INR	1 (8.3%)	0	1.000
Alcohol/drug overtake	0	1 (11.1%)	0.4286
AF type
Paroxysmal	4 (33.3%)	5 (55.6%)	0.3964
Persistent	2 (16.7%)	4 (44.4%)	0.3310
Persistent longstanding	6 (50%)	0	0.0186
Anticoagulation therapy
LMWH	2 (16.7%)	0	0.4857
Warfarin	3 (25%)	4 (44.4%)	0.3972
ASA	3 (25%)	2 (22.2%)	1.000
NOAC	7 (58.3%)	5 (55.6%)	1.000
Rivaroxaban	4 (33.3%)	2 (22.2%)	0.6594
Dabigatran	3 (25%)	3 (33.3%)	1.000
Other drugs
ACE‐I	4 (33.3%)	4 (44.4%)	0.6731
ARB	2 (16.7%)	2 (22.2%)	1.000
β‐adrenolitics	8 (66.7%)	8 (88.9%)	0.3383
Amiodarone	1 (8.3%)	3 (33.3%)	0.2722
Sotalol	0	1 (11.1%)	0.4286
Propafenone	3 (25%)	1 (11.1%)	0.6030
Aldosterone inhibitors	2 (16.7%)	2 (22.2%)	1.000
Calcium channel blockers	3 (25%)	1 (11.1%)	0.6030
Statins	3 (25%)	8 (88.9%)	0.2722
Diuretics	4 (33.3%)	6 (66.7%)	0.1984

Abbreviations: ACE‐I, angiotensin‐converting enzyme inhibitor; AF, atrial fibrillation; ARB, angiotensin receptor blocker; ASA, acetylsalicylic acid; HAS‐BLED, hypertension, abnormal renal and liver function, stroke, bleeding, labile International Normalized Ratio, elderly, drugs or alcohol; ICE, intracardiac echocardiography; INR, International Normalized Ratio; LAA, left atrial appendage; LMWH, low‐molecular‐weight heparin; NOAC, new oral anticoagulants; TIA, transient ischemic attack, CHADS2VASC2, congestive heart failure, hypertension, age ≥ 75, age 65‐74, diabetes mellitus, stroke/TIA/thrombo‐embolism, vascular disease, sex female.

4. DISCUSSION

The main finding of our study was that in some patients, TEE can produce false‐positive findings of the presence of an LAA thrombus, and that ICE is a useful tool in verifying these false‐positive TEE results. When a solid thrombus is seen on TEE, the concordance with ICE is high (86%); however, when a soft thrombus is detected on TEE, ICE may be normal in as many as 57% of these patients.

Although TEE is regarded as the gold standard to exclude a thrombus in the LAA before scheduled ablation or AF cardioversion, the accuracy of this technique is far from 100%. Recently, ICE has been introduced in interventional cardiology and electrophysiology procedures. A few studies have shown that, when performed from the PA, ICE gives excellent LAA images and may serve as a verification tool for inconclusive TEE findings. Sriram et al have shown that ICE can detect thrombus in 6% of patients with negative TEE and in patients with equivocal TEE results.

Sriram et al reported more thrombi detected by ICE than TEE and better LAA imaging using ICE. This growing evidence of usefulness of ICE in LAA imaging led some physicians to suggest that TEE may be omitted before AF ablation when ICE is routinely used during these procedures.9 In such situation, ICE is not only used for guiding transseptal puncture and performing ablation, but also to exclude LAA thrombus prior to starting the procedure.

Although the role of ICE in verification of inconclusive or even normal TEE findings regarding LAA is growing, the problem of verification of positive TEE findings by using ICE has not yet been addressed in the literature. Although TEE specificity in detecting an LAA thrombus has been reported as high as 98%, a significant proportion of patients with abnormal TEE may in fact have no thrombus in the LAA. Our small study showed that in patients with a solid LAA thrombus, this is probably rare; however, in patients with a soft thrombus, TEE may give false‐positive results in as many as half of the patients. How to explain such a big discrepancy between TEE and ICE results? One possible explanation is that a soft thrombus detected by TEE may in fact be very dense contrast or smoke that can be revealed by ICE. Another possibility is that in the present study we used too broad a definition of soft thrombus, and some of these TEE findings may be described by others as dense contrast. However, TEE images were assessed by very experienced and certified echocardiographers, and care was taken not to include dense contrast or smoke in the soft thrombus group, and to obtain various projections to fully assess LAA. Finally, there are some technical and anatomical (location of probe) differences between TEE and ICE that may influence accuracy of detecting and characterizing a thrombus in the LAA. For example, some studies evaluating ICE in imaging LAA utilized ICE probe location not in the PA but in the right ventricular inflow tract, which again might have influenced the results.

Difficulties in making definite decisions in echocardiography where the presence of an LAA thrombus is concerned—solid, soft, smoke, or dense contrast—are well known. Even very experienced echocardiographers encounter problems in making a definite diagnosis of an LAA thrombus in some cases. There is no perfect reference method to be sure whether the thrombus in the LAA is present or absent. Only intraoperative examination may serve as a reliable reference method, but it cannot be used in clinical practice for obvious reasons. Both computed tomography and magnetic resonance have many limitations in assessing LAA and cannot serve as a reference method for TEE or ICE. This lack of a good reference method is a limitation in all studies examining TEE or ICE in various settings.

The fact that 4 patients with abnormal TEE but normal ICE safely underwent ablation or AF cardioversion may suggest that ICE correctly ruled out a thrombus in the LAA, but this is only a clinical suggestion, and the numbers are extremely low.

Among patients with an ICE‐based assessment of the LAA, only persistent long‐standing AF was significantly associated with the presence of a thrombus, whereas parameters forming the CHA2DS2‐VASc score or other variables were not. This result further underlies the importance of the type of AF in formation of a clot in the LAA.10, 11

4.1. Limitations

The TEE examinations were performed without echocardiographic contrast, but it was a limitation for both techniques. The study group was very small, and our results can be only regarded as preliminary. We used ICE as a reference method for TEE, but whether it is justified remains to be shown. All echocardiographic limitations, especially differentiation between dense smoke and early stages of thrombus, have been discussed earlier. In this study, we examined only patients with abnormal TEE results, so our findings could not be extrapolated on patients with normal or equivocal TEE findings, and we could not confirm the diagnostic role of ICE in this subset of patients.

5. CONCLUSION

Our study suggests that ICE may be a valuable option for verification of a TEE‐based diagnosis of thrombus in the LAA, especially when a soft thrombus is detected by TEE. Further prospective studies are needed to examine whether the ICE‐based strategy for LAA thrombus exclusion can be widely used in clinical practice.

Baran J, Zaborska B, Piotrowski R, Sikora‐Frąc M, Pilichowska‐Paszkiet E and Kułakowski P. Intracardiac echocardiography for verification for left atrial appendage thrombus presence detected by transesophageal echocardiography: the ActionICE II study, Clin Cardiol, 2017;40:450–454. 10.1002/clc.22675