7.3 3-D TEE Acquisition, Cropping, Orientation, and Artifacts
Key Takeaways
- Center and optimize the target in two orthogonal 2-D planes before choosing the smallest 3-D pyramid that contains the needed anatomy.
- Live, zoom, gated full-volume, and color modes trade field size, volume rate, spatial detail, stitching risk, and flow information differently.
- Crop from known planes, retain landmarks, verify with MPR, and label the atrial, ventricular, aortic, or surgical viewpoint explicitly.
- Stitching, dropout, blooming, shadowing, and low-volume-rate blur must be tested against uncropped data, source planes, alternate angles, and complementary 2-D or Doppler imaging.
Build the volume on a 2-D foundation
Matrix-array TEE transducers sample a pyramidal volume that can be displayed live, cropped, rotated, or interrogated with multiplanar reconstruction. Three-dimensional TEE adds en-face anatomy and spatial relationships, but it does not replace a systematic 2-D, color, and Doppler examination. The RCS sonographer assists with acquisition and processing within the TEE team's defined probe and interpretive roles. A beautiful rendering from an incomplete or artifact-limited volume is not a complete study.
Begin in a diagnostic 2-D plane. Identify the target, optimize depth, frequency, focus, overall gain, time-gain compensation, and color settings, and center the structure in both orthogonal biplane views. The complete target must remain inside both planes through the cardiac cycle; tissue outside either plane can be clipped from the volume. Select the smallest pyramid and shallowest depth that contain the clinical target because widening the volume angle generally costs volume rate, line density, penetration, or spatial detail.
| Acquisition mode | Strength | Important trade-off |
|---|---|---|
| Live or narrow-angle 3-D | Immediate single-beat feedback and high volume rate | Small field may omit surrounding anatomy |
| 3-D zoom | Focused en-face display, commonly acquired in one beat | Enlarging the region can reduce spatial or temporal detail; incorrect box placement clips anatomy |
| Gated full volume | Wider field and potentially greater detail from subvolumes | Multibeat acquisition is vulnerable to ECG, respiration, and motion stitching |
| 3-D color | Relates flow jets to valve and chamber anatomy | Color reduces volume rate and adds aliasing, blooming, and gain-dependent jet appearance |
Preview live 3-D before committing to a gated acquisition. For multibeat full volume, use stable ECG triggering and a breath hold only when safe and coordinated by the responsible team. Watch the patient, not just the acquisition bar. Arrhythmia, premature beats, ventilation, probe motion, or patient movement can shift adjacent subvolumes. Review the raw volume immediately for seam lines or duplicated anatomy and reacquire with a single-beat or smaller-volume mode when motion cannot be controlled. Preserve raw data so orientation and cropping can be revisited.
Crop deliberately and label the viewpoint
Cropping removes foreground tissue to expose internal surfaces. Start from recognizable orthogonal planes and move transverse, sagittal, or coronal crop planes gradually toward the target. Keep enough surrounding landmarks to prove location, then refine the view. Cutting a crop plane through a leaflet, septum, or appendage can manufacture a defect. Rotate the dataset or change the cut plane before declaring that tissue is absent.
Multiplanar reconstruction, or MPR, displays intersecting orthogonal planes through the same volume. Place the crosshair on the actual structure, confirm it in every plane, and keep the intersection perpendicular to the intended measurement when appropriate. MPR helps localize a regurgitant orifice, define a plane for valve analysis, and verify that a suspected gap persists in source data. Measurements follow the laboratory's validated 3-D method; a perspective rendering alone is not a measurement plane.
Orientation must be explicit because there is no universal display agreement for every structure and question. State whether a valve is viewed from the atrium, ventricle, aorta, or surgeon's perspective, and retain identifiable landmarks. A laboratory may standardize the mitral surgical view with the aortic valve at the top and left atrial appendage to the viewer's left, but the annotation and landmarks must confirm that convention. Rotate for communication, not simply for visual appeal. Unlabeled screenshots can reverse left-right relationships and mislead procedural planning.
Recognize volume-specific artifacts
Three-dimensional displays can make artifacts look like anatomy because rendering adds surfaces, lighting, and depth cues. Test a suspected finding against live 2-D, biplane, color, multiple angles, uncropped data, and MPR.
| Appearance | Likely mechanism | Corrective action |
|---|---|---|
| Step or seam across a full volume | Respiratory, rhythm, patient, or probe motion between subvolumes | Reacquire with stable trigger and safe breath hold, fewer beats, or single-beat mode |
| Hole in a thin leaflet or septum | Dropout from low gain, oblique incidence, shadow, or an aggressive crop | Raise gain carefully, change angle, back out the crop, and verify in 2-D or MPR |
| Thick fused leaflets or reduced apparent orifice | Excess gain or rendering threshold causes blooming | Reduce gain or compression effect and inspect source planes |
| Missing tissue behind calcium or prosthesis | Acoustic shadowing and attenuation | Use another multiplane angle or probe position and integrate complementary modalities |
| Blurred rapidly moving leaflet | Inadequate volume rate for the field size | Reduce depth, pyramid, line density demand, or color region while preserving the target |
| Broad or fragmented 3-D color jet | Aliasing, blooming, low volume rate, or temporal mismatch | Optimize Nyquist scale and gain, shrink the color volume, and verify with 2-D color and spectral Doppler |
A seam present only in a gated volume is not proof of a dissection or leaflet cleft. Conversely, a real abnormality should not be dismissed merely because rendering is imperfect; seek persistence across source planes and complementary views. Gain that is too low removes thin structures, while excessive gain fuses adjacent surfaces. Optimize before acquisition because postprocessing cannot recover echo information that was never sampled.
Finish with an acquisition audit
Before leaving the view, confirm that the complete target was included, volume rate and spatial detail fit the question, ECG triggering was stable, stitching is absent or documented, gain preserves thin tissue, color is not blooming, and raw data were saved. Store a labeled overview plus focused cropped and MPR views that show the orientation and finding. Record mode, limitations, rhythm, and any unconventional viewpoint. The sonographer's quality-control statement describes what the dataset supports; the physician determines the clinical interpretation.
Back out the crop
When a new hole or missing leaflet appears after cropping, return to the uncropped volume and inspect orthogonal source planes. Cropping can reveal anatomy, but it can also create a convincing false defect.
A step-like discontinuity crosses a mitral valve only in a four-beat full-volume dataset; it is absent in live 3-D and the source 2-D views. What is the most likely explanation and correction?
Match each 3-D TEE mode with its most characteristic use or trade-off.
Match each item on the left with the correct item on the right