Radiomics-based characterization of valvular calcium improves risk stratification for paravalvular regurgitation after TAVI.

Authors

Thamara Carvalho Morais, Antonildes Nascimento Assunção Júnior, Sérgio Figueiredo Câmara, Carla Franco Grego da Silva, Gabriela Liberato, Bruna Morhy Borges Leal Assunção, Gabriel Kanhouche, Vinícius Cardoso Serra, André Vaz, Marco Antonio Gelain, Fabio Sandoli de Brito, Alexandre Abizaid, Henrique Barbosa Ribeiro, Cesar Higa Nomura

Published in

Journal of cardiovascular computed tomography. Aug 05, 2026. Epub Aug 05, 2026.

Abstract

Significant paravalvular regurgitation (sPVR) remains a clinically relevant complication after transcatheter aortic valve implantation (TAVI) and is influenced by valvular calcium characteristics that are incompletely captured by conventional CT metrics.
To evaluate whether CT-based radiomics of aortic valvular calcium can predict sPVR after TAVI and to assess the incremental value of radiomic features beyond standard anatomic and procedural parameters.
This retrospective multicenter study included 393 patients undergoing TAVI with preprocedural contrast-enhanced, ECG-gated CT. Predictive models were developed in a development cohort (n = 195) and tested in an external validation cohort (n = 198). Aortic valvular calcium was manually segmented on CT images. A radiomics-derived score (RDS) composed of five features was generated using least absolute shrinkage and selection operator logistic regression. Model performance for predicting postprocedural sPVR (moderate or severe), assessed by transthoracic echocardiography according to VARC-3 criteria, was compared between a traditional model based on procedural and CT anatomic variables and an enhanced model incorporating the RDS.
Among 393 patients undergoing TAVI (development cohort, n = 195; external validation cohort, n = 198), the RDS showed consistent discrimination for sPVR in the development (AUC, 0.76) and validation (AUC, 0.74) cohorts. Incorporation of the RDS significantly improved model discrimination (AUC, 0.76 to 0.84; p = 0.004), risk reclassification, and model calibration. In multivariable analysis, the RDS remained independently associated with sPVR (odds ratio, 2.31 per 0.1-unit increase; p = 0.018).
CT-based radiomics of valvular calcium provides incremental information beyond conventional CT metrics for identifying patients at risk of incomplete prosthesis sealing after TAVI. These findings support the integration of radiomics into routine preprocedural CT assessment to enhance individualized risk stratification in TAVI candidates.

PMID:
42731920
Bibliographic data and abstract were imported from PubMed on 13 Sep 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.