Live 3-Dimensional Transesophageal Echocardiography

Initial Experience Using the Fully-Sampled Matrix Array Probe

Methods

Results

Successful intubation was achieved in all 211 patients without complications. Acquisition of the real-time 3D-MTEE data was accomplished in approximately 10 min. Figure 1 shows 3D images of the mitral and aortic valves in systole and diastole. Figures 2 shows examples of 3D images of the LAA and left upper pulmonary vein, visualized en-face and in the long-axis view (Figs. 2A and 2B, respectively) and the interatrial septum (Fig. 2C).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Volume-Rendered 3D-MTEE Views of Mitral and Aortic Valves

Example of systolic (left) and diastolic (right) still frames of 3D-MTEE real-time volume renderings of the MV from the left atrial perspective (A and B) and the aortic valves (C and D) from the ascending aorta. AV = atrioventricular valve; MV = mitral valve; 3D-MTEE = 3-dimensional matrix array transesophageal echocardiography.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Volume-Rendered 3D-MTEE Views of Different Cardiac Structures

Examples of the left atrial appendage (LAA) visualized en-face (A) and in the long-axis view (B) showing the ridge separating the LAA from left upper pulmonary vein (LUPV); (C) interatrial septum (IAS) from the left atrial perspective showing the foramen ovale (FO); (D) tricuspid valve displayed from the right atrial perspective showing the septal (S), anterior (A), and posterior (P) leaflets. 3D-MTEE = 3-dimensional matrix array transesophageal echocardiography.

The visualization of all MV scallops, IAS, LAA, and LV endocardium was excellent in the majority of patients, as evidenced by scores ranging from 1.74 to 1.91. The presence of atrial fibrillation affected the visualization scores of the LV endocardium, the only structure that required full-volume acquisition. In contrast, visualization quality scores for the AV and TV were lower, between 0.75 and 1.06. The percentage of patients in whom each cardiac structure was assigned an optimal 3D visualization score of 2 was 85% to 91% of all scallops for both MV leaflets, 84% of the IAS, 86% of the LAAs, and 77% of the LV endocardium. In contrast, optimal visualization of the AV cusps was possible only in 18% to 22% of the patients from both the aorta (Figs. 1C and 1D) and the LV perspectives, and in 11% of the TV leaflets from both right atrial (Fig. 2D) and ventricular perspectives. Both the visualization scores and percentages of optimal visualization were similar between institutions.

Discussion

During the last several decades, advances in echocardiography have significantly contributed to its utility as an invaluable diagnostic tool for monitoring of cardiac performance. Transesophageal echocardiography provides clinicians superior image quality and resolution as the result of its immediate proximity between the transducer and posterior cardiac structures, coupled with the absence of lung and bone tissue interference. Three-dimensional echocardiography provides unique visualization and better understanding of the relationship between cardiac structures than 2-dimensional imaging, as well as accurate measurements of valvular and ventricular function (9). Although transthoracic 3D imaging is currently performed in real time with the use of matrix-array transducers, transesophageal 3D imaging with the use of sequential multiplane acquisition has never been embraced in clinical practice. Recent advances in ultrasound transducer technology have simplified the connection between the transducer and the imaging system, resulting in a reduced size of the connecting cable and significantly lowering power consumption, thus allowing real-time 3D TEE imaging.

This study is the first to describe the feasibility and clinical utility of real-time 3D-MTEE imaging as reflected by the combined experience of 2 different institutions in a large group of patients. The high success rate of esophageal intubations with the 3D-MTEE probe is not surprising, considering that the size of the probe tip is similar to the existing TEE probe (Omni III). The 3D-MTEE transducer generally provided superb 3D visualization quality of posterior cardiovascular structures and helped identify unique spatial relationships between structures that were not previously observed with either currently available real-time transthoracic or reconstructed 3D TEE imaging technology (Figs. 1 and 2).

One of the major findings of this study was that real-time 3D-MTEE consistently provided excellent quality volume-rendered images of MV components, including anterior and posterior leaflets, as well as annulus and subvalvular structures, as evidenced by the high visualization scores. This finding suggests that 3D-MTEE imaging may become one of the modalities of choice to assess this valve during perioperative planning of MV surgery.

The IAS and LAA were also well visualized because both structures are relatively posteriorly located. The fossa ovalis was easily delineated from the left and right atrial perspectives, which could allow improved guidance of percutaneous closure of atrial septal defects and transseptal punctures. When acquiring the LAA, the left upper pulmonary vein was also readily observed. From these volume data sets, dimensions and area of both orifices and length of the LAA in long-axis can be measured, which could be useful in electrophysiology/interventional procedures such as pulmonary vein ablation or LAA device closures. Transesophageal 3D acquisition of the LV resulted in optimal visualization of the endocardium in 77% of patients despite the requirement of 4 cardiac cycles to complete data acquisition.

We found that anterior cardiac structures, such as the AV and TV, were less well visualized because of leaflet dropout artifacts, which can probably be attributed to the longer distance between these structures and the transducer and to the oblique angle of incidence of the beam combined with the thinner leaflets in these valves.

Conclusions

In summary, although the 3D-MTEE probe allows excellent visualization of native MVs, LAA, IAS, and pulmonic veins, consistent optimal imaging of the AV and TVs may require further technological advances. Optimal visualization of MV apparatus components may allow quantification of its complex geometry and thus improve the diagnostic accuracy of echocardiographic evaluation of the MV. Overall, the ease and speed of data acquisition coupled with the ability to display cardiac structures using unique 3D views is likely to result in rapid integration of 3D-MTEE into clinical practice.