Between December 2022 and December 2024, we enrolled 45 patients with symptomatic severe aortic stenosis, identified as high surgical risk through a thorough multidisciplinary heart team evaluation, and subsequently referred for TAVR at Ain Shams University’s Cardiology Department.

All patients underwent comprehensive pre-TAVR evaluations, including 3D TEE and MDCT. Two experienced operators, blinded to each other’s findings, measured the aortic valve apparatus dimensions using all imaging modalities. To mitigate interobserver variability, the measurements of both operators were averaged. Patients with a bicuspid aortic valve were excluded, as were those with concurrent significant and severe valvular heart disease, including severe mitral stenosis and severe mitral regurgitation.

Our assessment included the calculation of the aortic valve apparatus dimensions, specifically the diameter, perimeter, and area, in addition to measuring coronary heights utilizing the Hybrid (with the help of Flexi-Slice) mode of MDCT. We conducted a comparative analysis of the measurements obtained from the three modalities and calculated the degree of underestimation across all parameters compared to the MDCT. Clinical agreement was established based on the projected prosthesis sizing derived from the three imaging methods.

We performed both two-dimensional (2D) and 3D TEE using a high-resolution TEE transducer (Vivid E9; GE Healthcare, Milwaukee, WI, USA) in accordance with established guidelines.

Image acquisition occurred in the midesophageal three-chamber view at approximately 120–135° during mid-systole. The images were transferred to an external workstation (EchoPAC 201) for detailed analysis using the commercially available 4D Auto Aortic Valve Quantification (AVQ) software (GE, Vivid E95), specifically designed for precise measurement of aortic annulus sizing. GE 4D Auto AVQ is part of the EchoPAC platform, which holds CE/MDR certification for advanced cardiac ultrasound image analysis, including 4D quantification.

The semiautomated 4D AVQ system aids operators in aligning and segmenting the left ventricular outflow tract (LVOT) as illustrated in Figure 1 . The analysis begins by selecting the optimal imaging loop while minimizing artifacts. The software automatically aligns the LVOT during midsystole, presenting three orthogonal planes—two long-axis and one short-axis view of the annular plane—which the operator fine-tunes to accurately define the annulus level. Following this alignment, the LVOT contour is automatically segmented, and the software computes aortic annulus area dimensions, including:

• −

  Mid-systolic aortic annulus area

• −

  Annular circumference

• −

  Minimal (D-min), maximal (D-max), and mean (D-mean) diameters

Representative semiautomated 3D transesophageal echocardiography (3D-TEE) analysis using 4D Auto Aortic Valve Quantification (AVQ), demonstrating alignment of the left ventricular outflow tract (LVOT) and annular plane localization across orthogonal views. The software-generated annular contour is shown with corresponding annular diameter, perimeter, and area outputs.

Only images with a minimum frame rate of 12 frames per second and minimal artifacts were included in the analysis, independent of the patient’s cardiac rhythm .

The hybrid approach was designed to mitigate the recognized limitations of 3D TEE, particularly its systematic underestimation of aortic annular dimensions. Given the established strengths of manual Flexi-Slice multiplanar reconstruction (MPR), this modality of the 3D TEE was employed to accurately delineate the virtual aortic annulus and derive annular measurements.

During manual MPR of each aortic annulus, we identified important constraints in exclusive reliance on the semiautomated software segmentation of the LVOT. Accordingly, semiautomated measurements were not accepted without critical appraisal. All generated planes were carefully reviewed, with particular attention to precise alignment at the annular level. Manual correction of the segmented contours was performed as needed to ensure accurate tracking of the tissue–blood interface across all three orthogonal planes.

Real-time 3D TEE data sets were cropped to focus on the aortic valve apparatus, adjacent LVOT, and ascending aorta. Three-dimensional echocardiography provides additional information on spatial relationships with surrounding structures, such as the LV outflow tract and mitral annulus, without the need for mental reconstruction.

Using Flexi-Slice mode, the aortic valve was evaluated across three orthogonal planes. The double-oblique transverse view pinpoints the annular nadir by cross-checking the lowest point of each cusp, allowing precise 3D manipulation of the valve anatomy. The technique aligns sagittal and coronal planes to the lowest cusp insertion points and orients them perpendicular to the LVOT and aortic root centerline, as illustrated in Figures 2 and 3 . By accounting for the natural tilt and asymmetry of the aortic root, this approach yields more accurate annular diameter, area, and perimeter measurements.

Illustrates the multiplanar reconstruction depicting all three planes (coronal, axial, and sagittal as well as a 3D view of the sagittal view) to identify the nadir of the virtual aortic annulus. Additionally, you can take advantage of the color coding to define the type of cusp on the coronal and sagittal views, which will be automatically reflected on the same cusps in the axial view. The annulus was traced, and the THV perimeter was 69 mm and the area was 340 mm².

Flexi-Slice multiplanar reconstruction (MPR) of the 3D-TEE dataset demonstrating orthogonal planes used to identify the virtual annular plane. The annular nadirs of all three cusps are cross-checked across planes to ensure accurate annular plane alignment before planimetry of annular perimeter, area, and diameters. The maximum and minimum diameter was 24 × 25 mm with an average diameter of 24.5 mm.

The coronary heights were only evaluated via the Flexi-Slice mode, which allows us to recreate a double-oblique transverse plane, which helps us locate the height of the coronaries as illustrated in Figure 4 .

Flexi-Slice MPR used to identify the left main coronary ostium and measure coronary height relative to the virtual aortic annular plane. Coronary height is measured as the perpendicular distance from the annular plane to the ostial take-off on the aligned long-axis view. The ostium of the left main was identified, and the height of the LM coronary ostium was 15 mm.

To ensure unbiased assessment, the echocardiographer performing these manual measurements was blinded to the results from the semiautomated and MDCT methods.

All patients underwent preprocedural MDCT. All examinations were performed using a Philips Brilliance 64-slice MDCT scanner. Standard technical parameters were used: gantry rotation time 300 ms, axial coverage 40 mm (64 × 0.625 mm), 120-kV tube voltage, 850 to 900 mA intensity with our modulation, and temporal resolution 165 ms. Retrospective electrocardiogram gating was performed. Contrast enhancement was achieved with 50 to 80 mL of iohexol 400 mg/mL. All image data were transferred to an offline postprocessing dedicated workstation. The mid-systolic phase of the cardiac cycle was selected (30% of the RR interval). The thickness of the reconstructed image was 0.67 mm. The datasets were reconstructed to achieve a double-oblique transverse at the level of the virtual ring (aortic annulus) described by Jilaihawi et al, and Kasel et al ^, MPRs were oriented to display the aortic annulus at basal attachment points. Two orthogonal planes were set, bisecting the aortic valve in sagittal and coronal planes. The third orthogonal plane (double-oblique transverse view) was set to bisect the aortic annulus at the most caudal attachment points of all three native cusps, orientating/positioning the virtual ring as in the TEE short-axis view. This approach aims to standardize annular plane identification and improve measurement reproducibility, as illustrated in Figure 5 .

Manual MDCT annular analysis demonstrating standardized double-oblique reconstruction to define the virtual annular plane at the basal hinge points of all three cusps. Annular perimeter, area, and diameters are derived from the double-oblique transverse plane.

To ensure consistency in prosthesis sizing across imaging modalities, we also restricted our study to a single valve platform. This approach allowed us to determine whether the observed underestimation in measurements had a meaningful impact on THV selection.

The clinical relevance may be inferred from whether this difference resulted in a different THV size and possible undersizing of the implanted valve, which may lead to increased risk of PVAR.