Three-dimensional (3D) transesophageal echocardiographic (TEE) imaging is a relatively new imaging modality that is increasingly being used to characterize a variety of cardiac pathologic features. In the present study, we reviewed the 2-dimensional (2D) and 3D TEE images from our echocardiographic database to identify patients with valve perforations. A review of the 2D TEE images resulted in the identification of 11 valvular perforations (6 aortic valves, 4 mitral valves, and 1 tricuspid valve). A review of the 3D TEE images allowed for the identification of 15 valve perforations (7 aortic valves, 7 mitral valves, and 1 tricuspid valve), including 4 perforations that could not be diagnosed using 2D imaging alone. In conclusion, 3D TEE imaging provided added benefit to traditional 2D TEE imaging because of its ability to provide en face visualization of the cardiac valves, allowing improved identification and precise anatomic localization of the perforation.
It has been suggested that 3-dimensional (3D) transesophageal echocardiographic (TEE) imaging might be superior to 2-dimensional (2D) TEE imaging in the diagnosis and evaluation of cardiac valve perforations. This is because of its ability to visualize valve leaflets en face, providing realistic anatomic views of the cardiac valves that are unattainable with traditional 2D imaging. This allows a more precise localization of the perforation and measurement of the perforation area. A paucity of data is available describing the utility of 3D TEE imaging in the assessment of valve perforations, however, and the available data have been limited to a few small case reports. We report a single-center experience in the evaluation of valve perforations using 3D TEE imaging.
Methods
We reviewed the inpatient echocardiographic data from a single center from January 2009 to March 2010. We subsequently limited our search to include patients with native heart valves, with or without previous valve repair, who had undergone 2D and 3D TEE imaging demonstrating a valvular perforation. The patients with prosthetic valve perforations were excluded from the present study. These echocardiographic images were entered into a database that also contained control 2D and 3D TEE images from patients with infective endocarditis without valve perforation. Endocarditis was chosen for the control, because it is the most frequent cause of valve perforation. The database was then reviewed by 2 expert cardiologists (TS, RJS), who were unaware of the clinical findings. They were asked to document their consensus about the presence and location of a valvular perforation when one existed. The reviewers documented their degree of confidence in the diagnosis of a perforation using a validated scoring system, as follows: high level of confidence, ≥90%; intermediate confidence, 50% to 89%; and low level of confidence, <50%. For the diagnosis of a perforation (≥90% confidence), we required visualization of the perforation in ≥2 views on the 3D TEE images and required confirmatory 2D or 3D Doppler imaging demonstrating color flow traversing the valve leaflet at the site of the perforation.
Results
A total of 1,146 patients had undergone 2D and 3D TEE imaging at our institution during the study period. Of these, 14 met our inclusion criteria and were included in the present study ( Table 1 ). A review of the 2D TEE images in the database allowed the identification of 11 valvular perforations (6 aortic valves, 4 mitral valves, and 1 tricuspid valve). A review of the 3D TEE images resulted in the identification of 15 perforations, including all the perforations identified on the 2D TEE images, plus 4 perforations that had not been identified on the 2D TEE images (7 aortic valves, 7 mitral valves, and 1 tricuspid valve; Table 1 and Figures 1 and 2 ) . We found that the en face 3D TEE images allowed more precise characterization of the size, location, and shape of the valve perforations than the 2D TEE images ( Figure 3 ). Valvular perforation had been most commonly caused by bacterial endocarditis, which accounted for 14 of the perforations identified in the present series. The final patient had experienced iatrogenic perforation of the aortic valve by a guidewire ( Figure 1 ).
| Patient Number | Age (years) | Gender | Perforation Location | 2D TEE % Confidence | 3D TEE % Confidence |
|---|---|---|---|---|---|
| 1 | 47 | Male | Right coronary cusp of aortic valve | 100% | 100% |
| 2 | 50 | Male | Noncoronary cusp of aortic valve | 90% | 95% |
| Posterior mitral valve leaflet | 90% | 95% | |||
| 3 | 59 | Male | Left coronary cusp of aortic valve | 100% | 100% |
| 4 | 68 | Male | Right coronary cusp of aortic valve | 50% | 95% |
| 5 | 70 | Male | Noncoronary cusp of aortic valve | 90% | 95% |
| Anterior mitral valve leaflet | 90% | 95% | |||
| 6 | 77 | Male | Left coronary cusp of aortic valve | 100% | 100% |
| 7 | 88 | Male | Noncoronary cusp of aortic valve | 100% | 100% |
| 8 | 46 | Male | Anterior mitral valve leaflet | 100% | 100% |
| 9 | 54 | Female | Anterior mitral valve leaflet | 25% | 95% |
| 10 | 56 | Female | Posterior mitral valve leaflet | <10% | 90% |
| 11 | 58 | Male | Anterior mitral valve leaflet | <10% | 95% |
| 12 | 63 | Female | Anterior mitral valve leaflet | 100% | 100% |
| 13 | 67 | Male | Posterior mitral valve leaflet | <10% | 75% |
| 14 | 24 | Female | Tricuspid valve | 95% | 95% |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
