Background: To develop a deep-learning (DL) pipeline that allowed an automated segmentation of epicardial adipose tissue (EAT) from low-dose computed tomography (LDCT) and investigate the link between EAT and COVID-19 clinical outcomes. Methods: This monocentric retrospective study included 353 patients: 95 for training, 20 for testing, and 238 for prognosis evaluation. EAT segmentation was obtained after thresholding on a manually segmented pericardial volume. The model was evaluated with Dice coefficient (DSC), inter-and intraobserver reproducibility, and clinical measures. Uni-and multi-variate analyzes were conducted to assess the prognosis value of the EAT volume, EAT extent, and lung lesion extent on clinical outcomes, including hospitalization, oxygen therapy, intensive care unit admission and death. Results: The mean DSC for EAT volumes was 0.85 ± 0.05. For EAT volume, the mean absolute error was 11.7 ± 8.1 cm³ with a non-significant bias of −4.0 ± 13.9 cm³ and a correlation of 0.963 with the manual measures (p < 0.01). The multivariate model providing the higher AUC to predict adverse outcome include both EAT extent and lung lesion extent (AUC = 0.805). Conclusions: A DL algorithm was developed and evaluated to obtain reproducible and precise EAT segmentation on LDCT. EAT extent in association with lung lesion extent was associated with adverse clinical outcomes with an AUC = 0.805.

In this study, we report our local experience of type A aortic dissections in patients with cerebral malperfusion treated with carotid stenting before or after aortic surgery, and present a systematic literature review on these patients treated either with carotid stenting (CS) before or after aortic surgery (AS) or with aortic and carotid surgery alone (ACS). We report on patients treated in our center with carotid stenting for brain hemodynamic injury of carotid origin caused by type A dissection since 2018, and a systematic review was conducted in PubMed for articles published from 1990 to 2021. Out of 5307 articles, 19 articles could be included with a total of 80 patients analyzed: 9 from our center, 29 patients from case reports, and 51 patients from two retrospective cohorts. In total, 8 patients were treated by stenting first, 72 by surgery first, and 7 by stenting after surgery. The mean age; initial NIHSS score; time from symptom onset to treatment; post-treatment clinical improvement; post-treatment clinical worsening; mortality rate; follow-up duration; and follow-up mRS were, respectively, for each group (local cohort, CS before AS, ACS, CS after AS): 71.2 ± 5.3 yo, 65.5 ± 11.0 yo; 65.3 ± 13.1 yo, 68.7 ± 5.8 yo; 4 ± 8.4, 11.3 ± 8.5, 14.3 ± 8.0, 0; 11.8 ± 14.3 h, 21 ± 39.3 h, 13.6 ± 17.8 h, 13 ± 17.2 h; 56%, 71%, 86%, 57%; 11%, 28%, 0%, 14%; 25%, 12.3%, 14%, 33%; 5.25 ± 2.9 months, 54 months, 6.8 ± 3.8 months, 14 ± 14.4 months; 1 ± 1; 0.25 ± 0.5, 1.3 ± 0.8, 0.68 ± 0.6. Preoperative carotid stenting for hemodynamic cerebral malperfusion by true lumen compression appears to be feasible, and could be effective and safe, although there is still a lack of evidence due to the absence of comparative statistical analysis. The literature, albeit growing, is still limited, and prospective comparative studies are needed.

Background: Postpartum hemorrhage is the leading cause of maternal mortality in emerging countries. This study aims to evaluate the effectiveness and safety of uterine artery embolization (UAE) using suture fragment (FairEmbo concept) in a swine model. Methods: Seven female swine uteri were embolized. The left uterine artery was embolized with 1 cm fragments of absorbable suture (Optime^® 0), and with gelatin sponge torpedoes for the contralateral side for comparison. The embolization effectiveness and the time for arterial recanalization with digital subtraction angiography (DSA) controls at D0, D7, and M1, were evaluated. Follow-up protocol also included clinical monitoring and macroscopical analyses at M1. A Mann–Whitney test (significance at P 0.05) was used for statistics. Results: A technical success was obtained for the seven arteries on each side, with no off-target embolization. The procedure time (10 min versus 3.7 min) and number of fragments (13.8 versus 5.7) required for complete occlusion were significantly greater in the FairEmbo group. All arteries were repermeabilized at M1. No necrosis was macroscopically visible at harvest at M1. Conclusion: This experimental study suggests that UAE with SBM FairEmbo method is feasible, safe, and effective in comparison with gelatin sponge procedure.

Background: Spinal cord ischemia is a major complication of treatment for descending thoracic aorta (DTA) disease. Our objectives were (1) to describe the value of angiographic cone-beam CT (angio-CBCT) and 3D road-mapping to visualize the Adamkiewicz artery (AA) and its feeding artery and (2) to evaluate the impact of AA localization on the patient surgical strategy. Methods: Between 2018 and 2020, all patients referred to our institution for a surgical DTA disorder underwent a dedicated AA evaluation by angio-CBCT. If the AA feeding artery was not depicted on angio-CBCT, selective artery catheterization was performed, guided by 3D road-mapping. Intervention modifications, based on AA location and one month of neurologic follow-up after surgery, were recorded. Results: Twenty-one patients were enrolled. AA was assessable in 100% of patients and in 15 (71%) with angio-CBCT alone. Among them, 10 patients needed 3D road-mapping-guided DSA angiography to visualize the AA feeding artery. The amount of contrast media, irradiation dose, and intervention length were not significantly different whether the AA was assessable or not by angio-CBCT. AA feeding artery localization led to surgical sketch modification for 11 patients. Conclusions: Angio-CBCT is an efficient method for AA localization in the surgical planning of DTA disorders.

Background: Embolization is a key therapeutic option in interventional radiology for the management of acute arterial bleeding and solid organ injuries. While various embolic agents exist, there is a persistent clinical need for materials that are not only highly effective but also biocompatible, easy to deliver, and cost-effective. We aim to evaluate a new eco-friendly dry foam agar-based embolization agent (ABEA) in an uncontrolled solid organ hemorrhage model. Material and Methods: Ten pigs underwent a controlled splenic injury. After a 5 min free-bleeding period, five pigs were treated with splenic artery ABEA embolization, while the remaining five received no treatment and served as the control group. Follow-up angiography was performed immediately after embolization and at 5 and 15 min in the treated pigs. Mean arterial pressures and average blood loss volumes were evaluated for 120 min. Results: The control group showed continuous blood loss, leading to a significantly higher total blood loss than the ABEA-treated group (1451 mL vs. 611 mL at 120 min, p < 0.05). Mean arterial pressure (MAP) remained below the hemorrhagic shock threshold throughout the procedure in the control group, validating the model of uncontrolled hemorrhage. In addition, a significant stabilization of MAP was observed in treated pigs, remaining above the critical level of hemorrhagic shock and differed significantly from control group values. Conclusions: Embolization with ABEA maintained MAP above critical levels and significantly reduced blood loss volume in a hemorrhagic model. These results support the technical feasibility and short-term hemostatic performance of ABEA in an acute setting. While preliminary, this proof-of-concept has provided the basis for a validated clinical study currently underway to evaluate its effectiveness and safety in human patients.