Your browser does not support JavaScript!

Home    Search  

Results - Details

Search command : Author="Τσέτης"  And Author="Δημήτριος"

Current Record: 2 of 16

Back to Results Previous page
Next page
Add to Basket
[Add to Basket]
Identifier 000456706
Title Αιμοδυναμική μελέτη του ανευρυσματικού σάκου μετά την ενδαγγειακή αποκατάσταση ανευρύσματος με σύγχρονες τεχνικές ιατρικής απεικόνησης
Alternative Title Hemodynamic evaluation of aneurysm sac after endovascular aneurysm repair using modern imaging techniques
Author Χαραλάμπους, Σταύρος
Thesis advisor Τσέτης, Δημήτριος
Reviewer Μαρής, Θωμάς
Ιωάννου, Χρήστος
Καραντάνας, Απόστολος
Περυσινάκης, Κωνσταντίνος
Κοντοπόδης, Νικόλαος
Κεχαγιάς, Ηλίας
Abstract Introduction Abdominal aortic aneurysmal disease is a major health problem associated with a certain risk of rupture and high mortality. Endovascular aneurysm repair (EVAR) is becoming the standard of care for the treatment of abdominal aortic aneu-rysm (AAA) since its initial introduction in 1991. The Achilles heel of EVAR is endo-leak, which may be accompanied by a risk of late aortic rupture, being the main cul-prit for re-interventions. Endoleak, is defined as the presence of blood within the sac but outside the graft after EVAR. Five types of endoleaks have been described. Type 2 endoleak (T2EL) is by far the most prevalent, with an incidence as high as 44%. T2EL is char-acterized by retrograde perfusion of the aneurysmal sac from one or more aortic branches usually by inferior mesenteric and lumbar arteries and less commonly by sacral, gonadal and accessory renal arteries. It is widely accepted that the risk of aneurysmal rupture in the presence of an isolated T2EL is exceptionally low. This has led to the general practice that such lesions should be treated only if aneurysmal sac growth is observed. Nevertheless, patients with T2EL are usually under close surveillance and strict follow-up protocols in order to observe the course of the endo-leak and early detect any changes in the dimensions of the aneurysm sac. Various imaging modalities have been used for surveillance screening after EVAR. Computed tomography angiography (CTA) has been accepted as the reference standard for confirming EVAR success. However, since, there is no consensus regard-ing the optimal CT imaging acquisition protocol, post-EVAR multi-phase CT studies may involve different combinations of unenhanced, arterial and delayed phases. In addition, the required lifelong follow-up for patients with detected T2EL may be as-sociated with relatively high cumulative radiation burden and increased risk for con-trast-induced nephropathy. Therefore, current research has to seek for the optimum post-EVAR imaging protocol in order to minimize the associated radiation-related and the contrast medium-related risks, without compromising the diagnostic perfor-mance of the study. In this research study, we will evaluate the hemodynamic characteristics of T2EL using the following modern imaging technologies: A) Dual-energy CT, B) Per-fusion imaging and C) Artificial Intelligence. These technologies have been thor-oughly studied in different organs and systems, even in patients with aortic aneu-rysms after endovascular treatment. However, these studies were mainly oriented in the detection of endoleak and were not designed to isolate the suspicious or high-risk T2EL for sac expansion. Thus, current research has focused on finding prognostic factors to identify these lesions and predict their clinical consequences after diagno-sis. A. Dual-energy CT imaging Purpose: To investigate the value of dual-energy CT imaging to discriminate low- from high- risk type II endoleaks (T2EL) after endovascular aneurysm repair (EVAR). Method: Study participants were consecutive patients referred for CT at 1-month post-EVAR. CT imaging acquisition included a dual-energy CT angiography (DECTA) and a delayed single-energy CT (SECT) imaging. Patients diagnosed with T2EL were re-examined at 6-months post-EVAR to assess the aneurysm sac growth (ASG). Upon ASG recorded, patients were categorized as having low- (group A) or high- risk (group B) T2EL. DECTA image data were employed to calculate the normalized effective atomic number (NZeff), the normalized iodine concentration, the slope of HUendo-leak/HUaorta against monochromatic energy, the dual-energy index and an improvised endoleak index (EI) for each T2EL. Statistical analysis was employed to compare all above parameters regarding their ability to differentiate low- from high- risk T2EL. Results: Among 40 patients examined at 1-month post-EVAR, 14 patients were diag-nosed with T2EL. NZeff and EI were found to be significantly lower in group A. NZeff was found to have the highest power to discriminate high-risk T2EL with an area-under-curve of 86.7%, showing100% specificity and 60% sensitivity. The optimal con-trast-to-noise ratio for T2EL demonstrated a median peak conspicuity level at 54-keV. The mean effective dose from DECTA and SECT scans was 27.8% lower com-pared to the sum of three SECT acquisitions. Conclusions: NZeff and EI were found to have a significant power in predicting the aggressiveness of T2EL lesions. Virtual monochromatic images at 54-keV may en-hance T2EL detection efficiency. B. CT Perfusion imaging Purpose: To examine various CT perfusion parameters, regarding their ability to dif-ferentiate aggressive from benign type 2 endoleaks (T2ELs) after endovascular aneu-rysm repair (EVAR). Materials and Methods: Patients who had undergone an EVAR and diagnosed with a T2EL were subjected to perfusion CT imaging. Eight different perfusion-parametric maps were generated. Regions of interest were placed on endoleak, on aneurysm sac thrombus and within endograft. Maximum aneurysm sac diameter was measured at baseline and 6-months later with CT-angiography. Patients were then divided into 2 groups according to the change of aneurysm sac dimension. A T2EL was defined as high risk in case of a stable or expanding sac or low-risk if sac shrinkage was recorded. Receiver operating characteristic analysis was employed to evaluate each map’s abil-ity to identify high- from low-risk T2EL. Results: Seven patients with T2EL were analyzed. Blood flow (BF) and permeability surface (PS) perfusion-parametric maps were found to have the highest potential to discriminate between high and low risk T2ELs (p<0.05). Patients with sac shrinkage (group A) had lower BF endoleak-to-endoluminal (EE) ratio and higher PS EE ratio compared to those with stable or enlarging aneurysm sac (group B). All other perfu-sion-parametric maps had lower discriminating power. BF EE ratio cut-off criterion of 0.375 was found to provide 100% sensitivity and specificity. The mean effective dose of the examination was estimated to be 25.72 mSv. Conclusion: BF perfusion-parametric map may provide a high potential for the iden-tification of aggressive T2ELs for sac expansion after EVAR. C. Radiomics and Machine learning algorithms Background: Persistent type 2 endoleaks (T2EL) require lifelong surveillance to avoid potentially life-threatening complications. Purpose: To evaluate the performance of radiomic features (RF) derived from com-puted tomography angiography (CTA), for differentiating aggressive from benign T2ELs after endovascular aneurysm repair (EVAR). Materials and Methods: A prospective study was performed on patients who under-went EVAR, from January 2018 to January 2020. Analysis was performed in patients who were diagnosed with T2EL based on the CTA of the first postoperative month and were followed at 6 months and 1 year. Patients were divided into 2 groups ac-cording to the change of aneurysm sac dimensions. Segmentation of T2ELs was per-formed and RF were extracted. Feature selection for subsequent machine learning analysis was evaluated by means of artificial intelligence. Two support vector ma-chines (SVM) classifiers were developed to predict the aneurysm sac dimension changes at 1 year, utilizing RF from T2EL at 1- and 6-month CTAs, respectively. Results: Among the 944 initial RF of T2EL, 58 and 51 robust RF from the 1- and 6-month CTAs, respectively, were used for the machine learning model development. The SVM classifier trained on 1-month signatures was able to predict sac expansion at 1 year with an area under curve (AUC) of 89.3%, presenting 78.6% specificity and 100% sensitivity. Similarly, the SVM classifier developed with 6-month radiomics data showed an AUC of 95.5%, specificity of 90.9% and sensitivity of 100%. Conclusion: Machine-learning algorithms utilizing CTA-derived RF may predict ag-gressive T2ELs leading to aneurysm sac expansion after EVAR.
Language Greek, English
Subject Abdominal aortic aneurysm
Endoleak
Endovascular repair
Ανεύρυσμα κοιλιακής αορτής
Issue date 2023-07-28
Collection   School/Department--School of Medicine--Department of Medicine--Doctoral theses
  Type of Work--Doctoral theses
Permanent Link https://elocus.lib.uoc.gr//dlib/c/6/d/metadata-dlib-1688021301-41306-5704.tkl Bookmark and Share
Views 668

Digital Documents
No preview available

Download document
View document
Views : 1