Post-implant CT volumetry shows frequent divergence between segmented and nominal factory valve volume after TAVI
D Hochstein, E Ben-Assa, R Nahmias-Oz, N Nagar, O Vazhgovsky, Y Steinmetz, S Tejman-Yarden, D Dvid, M Glikson, A WolakAbstract
Background
Transcatheter aortic valve devices are manufactured to a nominal geometry, yet deployment within a patient-specific annulus and aortic root can alter frame morphology. However, systematically acquired post-implant cardiac CT datasets after TAVI are uncommon, and consequently only limited published data exist that quantitatively characterize real-world, in vivo valve geometry after implantation. Leveraging a structured post-implant CT cohort, we quantified the difference between manufacturer-derived valve volume and in vivo post-implant valve volume using STL-based volumetric analysis from CT reconstructions.
Purpose
To compare post-implant CT-derived segmented valve volume with nominal factory valve volume and quantify the magnitude, direction, and variability of volumetric deviation in vivo.
Methods
We retrospectively analyzed 75 patients who underwent post-implant cardiac CT. Three-dimensional valve models were reconstructed from CT DICOM datasets using dedicated cardiovascular imaging software, exported as STL meshes, and valve volumes were computed in mm³. These in vivo volumes were compared with the corresponding manufacturer (factory) valve volumes for each valve size. We evaluated signed and absolute volume differences and summarized relative (%) change. Paired testing assessed whether mean signed differences differed from zero.
Results
Among 75 patients, segmented valve volume exceeded factory volume in 61 (81.3%), was lower in 13 (17.3%), and was identical in 1 (1.3%). Mean segmented valve volume was 2060.0 mm³ versus 1649.6 mm³ for factory volume. Mean signed difference was +410.4 mm³ (95% CI 311.4–509.3; p<0.001). Mean absolute difference was 470.5 mm³, with a median of 401 mm³ and a range of 0–1629 mm³. Mean relative change was +25.9% (median +23.1%). Absolute deviation was substantial across all nominal valve sizes, while proportional change was greatest in smaller valves, with mean relative increases of 35.1% for ~1243 mm³ valves, 25.7% for ~1569 mm³ valves, and 20.2% for ~1977 mm³ valves.
Conclusions
Post-implant CT demonstrates frequent and quantitatively substantial divergence between nominal factory valve volume and in vivo segmented valve volume after TAVI, most often in the direction of expansion. Smaller nominal valves show the greatest proportional change. CT-based volumetry may provide a practical framework for understanding post-deployment valve geometry in real-world patientsSegmented ValvesSegmented Valves on Factory Valves