Autor:innen:
Dipl.-Ing. Margarida Costa | Technische Universität München | Germany
Dr. Ricarda Clarenbach | Helios Klinikum München West, Klinik für Gefäßchirurgie | Germany
Dr. Sarika Lacher | Helios Klinikum München West, Klinik für Gefäßchirurgie | Germany
Dr. Olaf Stanger | Helios Klinikum München West, Klinik für Gefäßchirurgie | Germany
Prof. Nassir Navab | Technische Universität München | Germany
Dr. med. Guillaume Zahnd | Technische Universität München | Germany
Dr. med. Reza Ghotbi | Helios Klinikum München West, Klinik für Gefäßchirurgie | Germany
Background:
Eversion carotid endarterectomy is an attractive alternative to patch- and stent-based procedures as the observed re-stenosis rate of patients is nearly zero. Following the hypothesis based on such phenomenological observation, a data-based methodological approach is needed to quantify the impact of eversion onto the vessel.
Research question:
The aim of the present study is to address the challenge caused by the lack of available post-eversion CTA data. Whereas preoperative CTA is systematically acquired to assess the culprit regions, such imaging technology is generally avoided postoperatively as radiations are harmful for the patients. Therefore, the limited amount of available post-eversion geometries is an obstacle to the data-based analysis of surgery outcome.
Material and methods:
The main contribution of the present work is a patient-independent simulation framework to circumvent the issue caused by the absence of postoperative CTA. The framework is composed of three main steps.
1) Geometry construction: A small set of (n=39) of carotid bifurcation volumes were downloaded from a publicly available dataset (doi.org/10.7910/DVN/LYH9UG). Based on these initial 3D examples, a larger collection of patient-independent geometries (n=60) were manually constructed using the tool MESHMIXER. Special care was taken to generate a three-fold collection of healthy, stenosed, and post-eversion geometries (n=20 each).
2) Dynamic parameters specification: for each simulated geometry, a total of 3 parameters settings was randomly applied. Parameters were peak systolic velocity (either 0.8 or 1.2 m/s), internal diameter (5.0 or 7.0 mm), and systolic blood pressure (120 or 150 mmHg), thus leading to a collection of 180 models.
3) Wall Shear Stress (WSS) computation: WSS was adopted as an established surrogate marker of vascular health. The simulation platform ANSYS was used to compute WSS values for each models over a cardiac cycle.
Results:
A blind validation of the patient-independent geometries (https://margaridamc.github.io/carotidModelViewer) was performed by an expert cardiovascular surgeon. Three scalar indices were derived from the 3D+t WSS values: Time-Averaged Wall Shear Stress (TAWSS), Oscillatory Stress Index (OSI), and Relative Residence Time (RRT). The Wilcoxon rank sum test showed a statistically significant difference between healthy, diseased, and post-eversion geometries for TAWSS and RTT but not for OSI.
Conclusion:
A proof of concept study was proposed to quantify WSS in simulated models of post-eversion carotid endarterectomy. The strength of this approach is the capacity to generate a virtually infinite amount of realistic data without exposing the patients to unnecessary CTA radiations. The feasibility of generating patient-independant data being demonstrated, future investigations will focus on analyzing thus derived WSS values, to potentially establish the superiority of eversion with respect to other procedures.