Progress in 3D echocardiography was slow in the 1980s and 1990s, due mainly to technical reasons. Recently, along with the rapid evolution in probe and computer technologies, 3D echocardiography has grown into a well-developed technique that is able to display images of the heart that contain important new tissue and morphological information. This new technique is simple and rapid and enables additional clinical information to be obtained in different fields, including valve diseases, volumes and function of the two ventricles, intracardiac masses and monitoring of new procedures.1,2
This article details the current status of 3D technology and, in particular, recent advances in 3D live methods (either transthoracic or transoesophageal echocardiography) with clinical applications in the assessment of the main heart diseases.
Methods and New Technologies
Several methods, including random and sequential scanning and free-hand techniques by both transthoracic and transoesophageal echocardiography, have been proposed in the past, all of which required an offline reconstruction. Realtime volumetrics developed in the early 1990s by the Von Ram group at Duke University3 were based on novel matrix phased-array transducer technology. However, although the first-generation instrument had several practice limitations despite its potential, promising clinical applications were demonstrated. In the early 2000s, the second-generation realtime live 3D echocardiography (and nowadays the new probe and software technologies) allowed true routine application of the method in different fields. It is simple and rapid and may be integrated and associated with the standard 2D examination. In brief, these new transducers (with 3,000/4,000 ultrasound elements) generate multidirectional beam steering and signal processing, which take place automatically in the scanning probe itself. This technology generates in the display the true pyramidal volumes of data and creates online rendered images. Because these pyramidal data sets are still limited in terms of volume (generally 60x30°), alternatively it is possible to obtain a larger volume data set of up to 100x100° (the so-called full volume data set), together acquiring up to seven (generally four or seven) sub-volumes over consecutive cardiac cycles. In addition, new software allows zoom methods to be utilised, 3D images to be acquired in the presence of arrhythmias and spatial and temporal resolution to be improved. The images acquired may be sliced immediately in several planes and rotated, in order to visualise cardiac structures from any plane and multiple perspectives.4
The ‘old technology’ 3D sequential transoesophageal acquisition (multiplane transoesophageal echocardiography and respiratory gating or new techniques with very fast acquisition without gating) is still utilised and very useful in several clinical settings. However, a new generation of transoesophageal echocardiography probes with a novel matrix array technique was recently introduced, allowing 3D presentation of cardiac structures in realtime. This new tool may potentially provide fast and complete 3D information about cardiac structures, improving spatial orientation and overcoming limitations of offline 3D technologies (acquisition and reconstruction times). Therefore, in this review we refer to realtime 3D transthoracic echocardiography, transoesophageal sequential acquisition (3D transoesophageal echocardiography) and realtime transoesophageal echocardiography.
Valve Diseases
Both qualitative and quantitative evaluation of valvular heart disease can be improved by 3D echocardiography. Anyplane and paraplane analysis of a stenotic valve allows an accurate planimetry of the smallest orifice area. Zamorano et al.5 demonstrated that 3D transthoracic echocardiography is a feasible, accurate and highly reproducible technique for assessing the mitral valve (MV) area in patients with rheumatic MV stenosis. In a consecutive series of 80 patients, the MV area was assessed by conventional echo Doppler methods and 3D transthoracic echocardiography, and the results were compared with those obtained invasively. Compared with all other echo Doppler methods, 3D transthoracic echocardiography had the best agreement with the invasively determined MV area, and intra- and interobsever variability of the method was very good. The same author6 also studied 29 patients undergoing percutaneous balloon mitral valvuloplasty. 3D transthoracic echocardiography had the best agreement with the invasively determined MV area, particularly in the immediate post-procedural period. Thus, the method may be proposed as an ideal one throughout this procedure and make invasive evaluation unnecessary in this setting.
In addition to these very important quantitative data, 3D transthoracic echocardiography may be integrated with 2D evaluation in the assessment of qualitative morphology of the MV. Commissures, leaflets, annulus calcifications and subvalvular structures may be visualised from different and unique planes, facilitating understanding of this complex apparatus. Each of the atrio-ventricular valves may in fact be depicted both from the atrium or ventricle, with access to ‘en face’ views or from any other angle. Vegetations, commissural diseases, subvalvular pathologies (tip of the leaflets/cordae/papillary muscles) and clefts may be diagnosed accurately.
MV prolapse is visualised as a bulging or protrusion on the atrial site and, in this regard, 3D echocardiography is the ideal method to demonstrate the pathology. Several studies of 3D transoesophageal echocardiography and 3D transthoracic echocardiography have demonstrated the importance of 3D methods in this field.7–15 Recent data showed that 3D (either transthoracic or transoesopahegal) echocardiography is superior to the corresponding 2D techniques in the description of MV pathology. In particular, because realtime 3D transthoracic echocardiography has an accuracy similar to that of 2D transoesophageal echocardiography, this new technique (which is also simple and rapid) may be integrated into the standard 2D examination and should be regarded as an important examination in decisions regarding MV repair.
MV prolapse is the most frequent aetiology of mitral regurgitation in industrialised countries. Since the 1970s, MV repair has become preferential to replacement and is now possible in the overwhelming majority of patients with MV prolapse. Recent studies and guidelines have underlined the importance of early surgical intervention to preserve long-term left ventricular (LV) function in severe MV regurgitation. In this regard, a non-invasive pre-operative assessment of MV anatomy is essential to define feasibility and complexity of repair in differentiating cases with simple or complex lesions and to plan the ideal surgical strategy. For all of these reasons, 3D transthoracic echocardiography and 3D transoesophageal echocardiography should be regarded as important examinations in decisions regarding MV repair, particularly in cases with complex diseases and in light of these new early surgical strategies.
Recently, a large series of cases of MV prolapse evaluated with different echocardiography methods has been published by our group.16 One hundred and twelve consecutive patients with severe mitral regurgitation due to degenerative MV prolapse underwent a complete 2D and 3D transthoracic echocardiography the day before surgery, and a complete 2D and 3D transoesophageal echocardiography in the operating room. echocardiography data obtained by the different techniques (including scallops, commisures and chordal rupture identification) were compared with MV surgical inspection. 3D techniques were feasible in a relatively short time (3D transthoracic echocardiography: 7±4 minutes; 3D transoesophageal echocardiography: 8±3 minutes), with good (3D transthoracic echocardiography 55%; 3D transoesophageal echocardiography 35%) and optimal (3D transthoracic echocardiography 21%; 3D transoesophageal echocardiography 45%) imaging quality in the majority of cases. 3D transoesophageal echocardiography allowed more accurate identification (95.6% accuracy) of all MV lesions in comparison with other techniques.
3D transthoracic echocardiography and 2D transoesophageal echocardiography had similar accuracies (90 and 87%, respectively), whereas the accuracy of 2D transthoracic echocardiography (77%) was significantly lower. Thus, these data showed that 3D transthoracic echocardiography and 3D transoesophageal echocardiography are feasible and useful methods (and not time-consuming) for identifying the location of MV prolapse in patients undergoing MV repair. They were superior in the description of pathology in comparison with the corresponding 2D techniques and should be regarded as an important adjunct to standard 2D examinations in decisions regarding MV repair.
3D live transoesophageal echocardiography technologies (miniaturisation technology of the matrix array transducer) allow similar results (even though only very preliminary data have been published) in realtime, with obvious advances in clinical applications mainly in the operating room, providing immediate data to anaesthesiologists and surgeons in the pre-and post-operative periods.
Figure 1 shows two cases of MV prolapse examined by 3D transthoracic echocardiography and 3D transoesophageal echocardiography, respectively.
The aortic valve may be evaluated easily by 3D transthoracic echocardiography or 3D transoesophageal echocardiography .17 The morphology of the valve my be defined with high accuracy, demonstrating normality of the cusps, congenital abnormalities (bicuspid aortic valve) or acquired pathologies (see Figure 2). Few data have been concentrated on the accuracy of 3D in the assessment of the severity of aortic valve stenosis. Recently, Goland et al. evaluated the reproducibility and accuracy of realtime 3D transthoracic echocardiography in 33 patients with aortic stenosis.18 The aortic valve area shown by 3D transthoracic echocardiography was compared with transthoracic 2D echocardiography planimetry, 2D transoesophageal echocardiography planimetry and, in 15 cases, with invasive measurements. Statistic analysis showed good agreement and small absolute differences in aortic valve area between all planimetric methods. Interobserver variability was better for the 3D technique. Therefore, the authors suggest that this very rapid and novel method provides an accurate and reliable quantitative assessment of aortic stenosis.
3D echocardiography offers a direct view to evaluate the leaflet surface of the tricuspid valve and a unique method to visualise the three leaflets simultaneously. Potentially, this offers the opportunity to study every tricuspid pathology by different perspectives, such as from the right ventricle (RV), the right atrium or oblique planes. Leaflet coaptation and separation may therefore be visualised easily. Even though a consecutive series of cases with tricuspid valve disease have not been published, and no data demonstrate additional clinical value of the method, several case reports have shown the importance of this technique in Ebstein’s anomaly,19 tricuspid stenosis20 and other tricuspid pathologies. This potential in different tricuspid pathologies may be furthered by the measurement of RV volumes and function (ejection fraction) by offline 3D transthoracic echocardiography analysis (preliminary data: EuroEcho, in press). In fact, in all tricuspid pathologies a crucial point is represented by RV function, and calculation of RV volumes is not feasible by conventional 2D echocardiography. Full volume analysis by 3D transthoracic echocardiography allows in a few seconds the acquisition of a complete data set of the RV, which allows offline calculation of RV volume, as well as recognition of tricuspid diseases.
Congenital Heart Diseases
The main current clinical applications of realtime transthoracic echocardiography and 3D transoesophageal echocardiography include atrial and ventricular septal defects, several complex congenital pathologies and bicuspid aortic valves.21–23 In children and young adults, due to their excellent acoustic window, 3D transthoracic echocardiography represents an ideal technique to visualise the complex anatomy of these pathologies. Position, morphology and relationship with other cardiac structures are better defined. In atrial or ventricular septal defects, the location, size, configuration, type and measurement of rims are easily imaged by 3D transthoracic echocardiography and 3D transoesophageal echocardiography in the majority of cases, thus impacting the indication of percutaneous or surgical procedures. In addition, these methods allow an in vivo 3D assessment of devices, providing new insights into positioning and interferences with the adjacent structures.24,25
Cardiac Masses
3D transthoracic echocardiography and 3D transoesophageal echocardiography are also emerging techniques for the evaluation (particularly in the pre-operative assessment) of cardiac masses, including vegetations, thrombi and tumours. The morphology, size and location of masses may be defined accurately, and a 3D approach facilitates the understanding of relationships with the adjacent anatomical structures.26–29 Identification of the attachment points of vegetations and tumours may improve clinical and procedural decisions. Initial experience with realtime 3D transoesophageal echocardiography in the operating room also supports its use in this field (see Figure 3).
Monitoring of Percutaneous or Surgical Procedures
The new generation of realtime 3D transoesophageal echocardiography and, in paediatric patients, of 3D transthoracic echocardiography probes provides a unique new imaging technique to monitor invasive percutaneous or surgical procedures. The pathomorphology of cardiac structures may be evaluated, as well as catheters and devices, during procedures.30,31 Therefore, new surgical procedures, atrial septal or patent foramen ovale percutaneous closures, aortic valve implantation, percutaneous MV repair or commisurotomy and electrophysiological procedures may be monitored by realtime 3D echocardiography, thereby increasing the safety, accuracy and efficacy of these interventions.
Even though in these fields there is only initial experience, several studies demonstrate that these new imaging techniques will become an important clinical tool. In cardiac surgery, our first experience in 120 examinations clearly shows feasibility, excellent imaging quality and an additional clinical value of these techniques mainly in valve surgery (either repair or replacement), not only in the pre-operative period but also in evaluating post-operative complications, as well as in cardiac mass resection.
Left and Right Ventricular Volumes, Mass and Function
Realtime 3D echocardiography overcomes geometrical assumptions and limitations of 2D echocardiography, and in the last decade several papers have demonstrated the advantages of 3D echocardiography versus 2D echocardiography in assessing LV volumes, global and regional systolic function and mass.32–38
More recently, new software and probes have further increased the accuracy and reproducibility of 3D transthoracic echocardiography, allowing more rapid measurements and sophisticated calculations, such as intraventricular mechanical dyssynchrony assessment.39 Segmental time–volume curves of each of the 16 or 17 segments of the LV may be obtained, calculated and plotted against time, and a systolic dyssyncrony index may be calculated to predict the efficacy of cardiac resynchronisation therapy and to evaluate results. Although 3D transthoracic echocardiography LV analysis of volumes, mass and systolic global and regional function has been evaluated widely, and the majority of new 3D ultrasound units allow online or offline measurements with several surface-rendering techniques, a new method has been introduced recently concerning RV morphology, volumes and function.
A recent study from our group40 evaluated the feasibility of a new 3D transthoracic echocardiography software adapted for RV morphology in a population of 200 cases. This method was feasible and not only allowed the evaluation of volumes and function in normal subjects and pathological patients but also clearly differentiated patients with different degrees of RV dilatation and dysfunction. The method is based on the manual (offline) tracing of systolic and diastolic borders and a phase of semiautomatic contour detection, which generates (from the 3D data set obtained in the apical view) the shape and morphology of the entire RV divided into the inflow, outflow and apical portions. This new technique may overcome limitations of the traditional 2D and Doppler parameters that are surrogates of the ejection fraction of the RV. Volumes and ejection fraction of the RV may be calculated with this 3D transthoracic echocardiography method, further improving the armamentarium of non-invasive evaluation of the RV, which has a clinical and prognostic impact on heart diseases (see Figure 4).
Limitations and Future Developments
Despite all these promising new technologies and clinical applications, 3D echocardiography has several limitations. Low temporal and spatial resolution, artefacts due to arrhythmias, or motion of the probe or of the patient (in the full volume acquisition data sets) are the main limitations. Advances in miniaturisation technology of the matrix transducer and improvement in software and online calculations will enable development of these techniques, allowing 3D echocardiography to become a routine modality in ultrasound laboratories. In addition, 3D echocardiography facilitates training and communication between experts and non-experts and between different specialists (due to the comprehensive evaluation of the cardiac anatomy) and expands the potential of non-invasive cardiology in the main cardiovascular diseases.