Multi-center study of computed tomography angiography-based evaluation of aortic valve area for non-invasive pre-procedural planning of transcatheter aortic valve replacement

Clin Res Cardiol (2024). DOI 10.1007/s00392-024-02406-5
C. Weber (Freiburg im Breisgau)¹, J. Studier-Fischer (Freiburg im Breisgau)¹, M. Hein (Freiburg im Breisgau)¹, N. Jander (Freiburg im Breisgau)¹, P. Ruile (Freiburg im Breisgau)¹, C. Schlett (Freiburg)², M. Potratz (Bad Oeynhausen)³, T. K. Rudolph (Bad Oeynhausen)³, J. Steffen (München)⁴, S. Deseive (München)⁴, A. Schwab (Frankfurt am Main)⁵, P. C. Seppelt (Frankfurt am Main)⁵, I. Hilgendorf (Freiburg im Breisgau)¹, M. Zehender (Freiburg im Breisgau)¹, D. Westermann (Freiburg im Breisgau)¹, C. von zur Mühlen (Freiburg im Breisgau)¹, T. Heidt (Freiburg im Breisgau)^1
1Universitäts-Herzzentrum Freiburg - Bad Krozingen Klinik für Kardiologie und Angiologie Freiburg im Breisgau, Deutschland; ²Universitätsklinik Freiburg Klinik für diagnostische und interventionelle Radiologie Freiburg, Deutschland; ³Herz- und Diabeteszentrum NRW Allgemeine und Interventionelle Kardiologie/Angiologie Bad Oeynhausen, Deutschland; ⁴LMU Klinikum der Universität München Medizinische Klinik und Poliklinik I München, Deutschland; ⁵Universitätsklinikum Frankfurt Med. Klinik III - Kardiologie, Angiologie Frankfurt am Main, Deutschland

Introduction: Transcatheter aortic valve replacement (TAVR) has become an established treatment option for patients with symptomatic severe aortic stenosis (AS) especially at increased risk for surgical complications. Transoesophageal echocardiography (TOE) is still the gold-standard to measure the aortic valve area (AVA). Little is known about the correlation of the AVA measured by TOE (AVA_TOE) compared to computed tomography angiography assessment (AVA_CTA). Consequently, direct measurement of AVA_CT could improve the diagnostic pathway and even make TOE redundant in some patients.

Methods and Results: A single-center cohort (178 patients; 93 men; mean age 80 years) of patients with moderate to severe AS who received both full-cycle CTA and TOE were included and retrospectively analyzed to compare intermodal AVA assessment. For all patients planimetry of AVA_CT and AVA_TOE was performed by two independent raters with good to excellent inter-observer correlation coefficients of 0.84 for AVA_CT and 0.98 for AVA_TOE. The mean AVA_CT was 0.97 ± 0.28 cm² compared to a mean AVA_TOE 0.89 ± 0.26 cm². The correlation coefficient of intermodal imaging was considered good at a value of 0,80. The best cut-off value for the determination between severe and non-severe AS was evaluated at an AVA_CT of 100,2 mm² by ROC-analysis (AUC 0,85; sensitivity 81,6%; specificity 78,3%). Afterwards the cut-off value was used in a second multi-center-included real-life cohort (412 patients, 217 men, mean age 81 years) with AS undergoing TAVR procedure. Here independent raters reached a correlation for AVA_CT and AVA_TOEof 0,75 and a sensitivity of 71,7% and specificity of 83,5% compared to gold-standard TOE AS severity classification retrospectively.

Conclusion: AVA_CT shows good reliability compared to gold-standard AVA_TOE in a single-center assessment. These results further depict that the best cut-off value for AVA_CT is not different from AVA_TOE. Through external validation in a multi-center-included real-life cohort with AS undergoing pre-procedural planning of TAVR AVA_CT approved a good distinction between severe and non-severe AS.These results are a further step towards a fully non-invasive pre-procedural workup of the aortic valve in TAVR screening.