In 1974, shortly after the advent of two-dimensional (2D) echocardiography (2DE), Dekker et al published the results of their graduate work at Stanford University, “A System for Ultrasonically Imaging the Human Heart in Three Dimensions.” Their vision was clear: “With the four dimensional data stored on the computer, programs to determine such properties as ejection fraction, wall thickness, aneurysms, defects, valve motions, structure velocities and heart power can be developed as well as image enhancement routines.”
In the more than 3 decades that have passed, many milestones have been achieved in three-dimensional (3D) echocardiography (3DE). Several generations of 3D or four-dimensional (3 dimensions and time) echocardiographic systems have been developed. With the newest matrix-array transducer technology and software, a four-dimensional data set can be obtained with the push of a button. A large body of scientific literature has become available: a PubMed search reveals >3000 articles published to date on the utility of 3DE. As more evidence for the utility of 3DE became available, the American Society of Echocardiography published a position paper on its use.
3DE: Challenges and Opportunities
In the real world of congenital heart disease (CHD), however, most pediatric echocardiography laboratories in both community hospitals and academic medical centers do not routinely use 3DE for clinical studies. In the scientific sessions of the American Society of Echocardiography, where heated debates flare regularly regarding the utility of 3DE versus 2DE, 3DE versus strain rate imaging, and so on, proponents of 3DE have been consistently outnumbered. Resistance to the acceptance of 3DE in the CHD community is still widespread. One must ask, why is an intuitively logical and extensively validated technology for imaging and assessing CHD still being ignored?
It is apparent that many critical questions remain to be answered before this technology gains widespread clinical acceptance: (1) Beyond prettier pictures, is 3DE truly superior to 2DE? (2) Will 3DE replace 2DE? (3) How should we use 3DE for CHD in our daily clinical and investigational work?
2DE Versus 3DE
In this issue of JASE , Huang et al describe their experiences developing 3DE-like views of secundum atrial septal defects (ASDs) using novel 2D echocardiographic views. They also compared these results with findings from 3D transesophageal echocardiography in a subset of patients. They obtained good to excellent quality 2D echocardiographic en face views in 80% of their subjects (294 of 368). The feasibility of using 2DE en face views correlated inversely with the size of the defect. The shapes of the ASDs were well delineated in 3D format and variable. The major diameters of the ASDs could be viewed in the anterosuperior-anteroinferior, posterosuperior-posteroinferior, or anterosuperior-posteroinferior direction. In their studies, conventional 2DE significantly underestimated the maximal diameter compared with the new en face views (2.27 ± 0.78 vs 2.40 ± 0.80 cm). The major and minor diameters of the 2D echocardiographic en face views correlated well with those of the 3D transesophageal echocardiographic views ( r = 0.758 and 0.689, respectively). The authors were able to obtain the en face views of the secundum ASDs by adding a few minutes to the conventional 2D echocardiograhpic protocol.
In addition to potentially improving percutaneous intervention by introducing a novel en face view for imaging and assessing secundum ASD, this article raises a number of questions that have been debated in the pediatric and adult CHD community.
2DE Versus 3DE
In this issue of JASE , Huang et al describe their experiences developing 3DE-like views of secundum atrial septal defects (ASDs) using novel 2D echocardiographic views. They also compared these results with findings from 3D transesophageal echocardiography in a subset of patients. They obtained good to excellent quality 2D echocardiographic en face views in 80% of their subjects (294 of 368). The feasibility of using 2DE en face views correlated inversely with the size of the defect. The shapes of the ASDs were well delineated in 3D format and variable. The major diameters of the ASDs could be viewed in the anterosuperior-anteroinferior, posterosuperior-posteroinferior, or anterosuperior-posteroinferior direction. In their studies, conventional 2DE significantly underestimated the maximal diameter compared with the new en face views (2.27 ± 0.78 vs 2.40 ± 0.80 cm). The major and minor diameters of the 2D echocardiographic en face views correlated well with those of the 3D transesophageal echocardiographic views ( r = 0.758 and 0.689, respectively). The authors were able to obtain the en face views of the secundum ASDs by adding a few minutes to the conventional 2D echocardiograhpic protocol.
In addition to potentially improving percutaneous intervention by introducing a novel en face view for imaging and assessing secundum ASD, this article raises a number of questions that have been debated in the pediatric and adult CHD community.
2DE Versus 3DE or 2DE and 3DE?
First, the answer to the lingering question of whether 3DE is superior to 2DE is now blurry. Three-dimensional echocardiography has been validated as a technique for diagnosing and assessing ASDs by imaging their number, location, size, and dynamics ( Table 1 ). It has also been reported that this modality is useful in evaluating complex ASDs and rare types of ASDs. With the possibility of successful percutaneous transcatheter closure of secundum ASDs in select patients, 2DE and 3DE have become imperative for selecting patients, guiding and monitoring procedures, and assessing outcomes. Repairing an ASD in a beating heart requires exquisite precision, guidance, and assessment of the procedure in 3 dimensions and in real time, which are logical applications of and an ultimate test for real-time 3DE. Although 2DE can be used to generate 3D en face views of ASDs, further validation is needed. A considerable amount of data suggests that 3DE provides incremental value for diagnosis, assessment, and interventional guidance during closure of secundum ASDs and will continue to expand the horizon for innovative therapies.
| Study | 3D echocardiographic techniques | Applications |
|---|---|---|
| Imaging and assessing simple common secundum or complex uncommon ASDs | ||
| Belohlavek et al | Reconstruction by 3D TEE | ASD anatomic details and spatial relations |
| Marx et al | Reconstruction by 3D TTE/TEE | ASD site, relative size and dynamic geometry |
| Magni et al | Reconstruction by 3DE | ASD minor and major axis diameter ( r = 0.97 and 0.91) |
| Franke et al | Reconstruction by 3D TEE | ASD dynamic size vs Qp/Qs, other structures |
| Nanda et al | 3D TEE | Sinus venosus ASD |
| Georgakis et al | 3D TEE | Complex ASD |
| Vargas-Barrón et al | 2DE/3DE | Unroofed coronary sinus |
| Chen et al | 3DE | Inferior sinus venosus ASD |
| Guiding percutaneous device closure of secundum ASDs | ||
| Magni et al | DAS-Angel Wings device | 2D TEE vs 3D TEE vs balloon sizing |
| Acar et al | Cardioseal device | 2D TEE/3D TEE: ASD shape, size, rim, residual shunt |
| Zhu et al | 3D TEE | ASD balloon sizing |
| Cao et al | Amplatzer device | 2D TEE/3D TEE for multiple ASDs |
| Acar et al | Amplatzer and Cardioseal | 3D TEE |
| Lodato et al | RT 3D TEE | ASD sizing vs 2D TEE and ICE |
| Taniguchi et al | RT 3D TEE | ASD sizing |
| Innovative therapy: beating heart ASD closure without cardiopulmonary bypass guided by 3DE | ||
| Downing et al | Prototype volumetric 3DE | In vitro study |
| Suematsu et al | Epicardial RT 3DE | 3D volume rendering |
| Suematsu et al | RT 3DE vs 2DE | Matrix-array 3DE |
| Suematsu et al | RT 3DE | In vivo swine model |
Second, it is now more intriguing to ask if 3DE can ultimately replace 2DE. In theory, all of the 2D sectional grayscale or color Doppler images can be derived from a 3D volumetric data set using the cropping tools that are available. However, we must keep in mind the strengths and limitations of 2DE and 3DE. The newest 3D echocardiographic matrix-array probe provides a smaller footprint than the first-generation matrix probe, frequencies of up to 7 MHz, and frame rates of up to 20 to 30 frames/s. Although these are amazing milestones in the evolution of 3DE, these specifications have not been able to match those of 2DE: transducers have an even smaller footprint, frequencies of up to 12 MHz, and frame rates of up to 50 to 60 frames/s. The current 3D transesophageal echocardiographic probe has a footprint that can be used only for children who weigh >30 kg. More important, we need to standardize the study protocols and procedures for 3DE and for training physicians and sonographers to ensure that they are facile and efficient when using this technology. Analogous to the evolution from M-mode echocardiography to 2DE, whereas 2DE has become one of the most well established, validated, and widely used modalities for cardiovascular imaging, M-mode echocardiography still has some valuable uses. The study by Huang et al reminds us that 2DE will continue to evolve and that new uses for CHD may also evolve.
Finally, how should we use 3DE to assess CHD in the real world? Three-dimensional echocardiographic technology obviously continues to evolve. Over time, transducer footprints will become smaller and frequencies higher, allowing this technology to be useful for newborns and infants with CHD. Cropping tools will become easier to use to visualize CHD in 3 or 4 dimensions. Volumetric data sets will increase to include the entire heart. Volumetric computational algorithms will become more automated for efficient clinical workflow. Transducers will become multifunctional to be used for all M-mode, 2D, 3D, four-dimensional, and Doppler (pulsed-wave, continuous-wave, and color) interrogations. Furthermore, methods for the qualitative and quantitative assessment of anatomy, flow, and ventricular function using real-time 3DE to study CHD in children have been systematically validated ( Table 2 ) and will continue to evolve. In the future, the same set of questions—2DE versus 3DE or 2DE as well as 3DE—will resurface when new technology becomes available, such as 2D and 3D strain analysis using speckle-tracking imaging, and so on.
