Skip to Main content Skip to Navigation
Master Thesis

Calcifications aortiques et TAVI‎ : automatisation de la mesure et validation pronostique sur la cohorte de Rouen

Abstract : Transcatheter aortic valve implantation (TAVI) has markedly changed the therapeutic management of severe aortic stenosis in the last ten years. Yet, some procedures remain futile given their poor outcome. Thoracic aorta calcifications, a surogate of aortic stiffness , seem to predict cardiovascular events after TAVI. Purpose : In the present study, we aimed at validating a semi-automatic software to measure aortic calcification volume; the second aim was to test the prognostic value of aortic calcifications on mortality and heart failure. Methods : This is a prospective, non-interventional, single-center study. Multidetector computer tomography (MDCT) was analyzed for 331 patients implanted in Rouen University Hospital between January 2010 and December 2014. For each patient, calcification volume of ascending thoracic aorta (CAA), descending thoracic aorta (CAD) and total thoracic aorta (CAT) were determined using a semi-automatic software designed for this study. Results : The intra and inter-observer reproducibilities of the semi-automatic measurements as well as the correlation with manual measurements were excellent. Average follow-up was 706 ± 433 days. After multivariate analysis adjusting for confounding factors, CAT appeared as a predictor of cardiovascular mortality for high thresholds of calcifications, between 4000 and 5000mm3 (HR (thresholds4000) = 1,51, 95% CI [0.89 -2.56], p = 0.124; HR (thresholds5000) = 1,94, 95% CI [1.08 to 3.47], p = 0.026). CAD was an independent predictor of high NTproBNP level before TAVI (β = 0.11, p = 0.046), immediately after TAVI (β = 0.12, p = 0.051) and one year after TAVI (β = 0.19, p = 0.030). No significant relationship was found between aortic calcifications and rehospitalization for heart failure. Conclusions : CAT is easy to measure, with the software validated by our study, without adding more exam or extra cost before TAVI. This parameter seems promising for risk stratification after TAVI. Further validation in independent cohorts is warranted.
Document type :
Master Thesis
Complete list of metadatas

Cited literature [104 references]  Display  Hide  Download
Contributor : Bibliothèque Santé <>
Submitted on : Friday, February 3, 2017 - 11:05:42 AM
Last modification on : Monday, December 2, 2019 - 10:24:01 AM
Long-term archiving on: : Friday, May 5, 2017 - 12:57:14 PM


  • HAL Id : dumas-01455029, version 1


Marion Dupré. Calcifications aortiques et TAVI‎ : automatisation de la mesure et validation pronostique sur la cohorte de Rouen. Médecine humaine et pathologie. 2016. ⟨dumas-01455029⟩



Record views


Files downloads