Publication Date


Document Type

Doctoral Dissertation

Academic Program

Biomedical Engineering and Biotechnology


Bioengineering and Biotechnology Program

First Thesis Advisor

Matthew J. Gounis, Ph.D.


Ischemia, Stroke, Cerebrovascular Disorders, Thrombectomy


Stroke is the third most common cause of death and a leading cause of disability in the United States. The existing treatments of acute ischemic stroke (AIS) involve pharmaceutical thrombolytic therapy and/or mechanical thrombectomy. The Food and Drug Administration (FDA)-approved recombinant tissue plasminogen activator (tPA) administration for treatment of stroke is efficacious, but has a short treatment time window and is associated with a risk of symptomatic hemorrhage. Other than tPA, the Mechanical Embolus Removal in Cerebral Ischemia (MERCI) retriever system and the Penumbra Aspiration system are both approved by the FDA for retrieval of thromboemboli in AIS patients. However, the previous clinical studies have shown that the recanalization rate of the MERCI system and the clinical outcome of the Penumbra system are not optimal. To identify the variables which could affect the performance of the thrombectomy devices, much effort has been devoted to evaluate thrombectomy devices in model systems, both in vivo and in vitro, of vascular occlusion. The goal of this study is to establish a physiologically realistic, in vitro model system for the preclinical assessment of mechanical thrombectomy devices.

In this study, the model system of cerebrovascular occlusion was mainly composed of a human vascular replica, an embolus analogue (EA), and a simulated physiologic mock circulation system. The human vascular replica represents the geometry of the internal carotid artery (ICA)/middle cerebral artery (MCA) that is derived from image data in a population of patients. The features of the vasculature were characterized in terms of average curvature (AC), diameter, and length, and were used to determine the representative model. A batch manufacturing was developed to prepare the silicone replica.

The EA is a much neglected component of model systems currently. To address this limitation, extensive mechanical characterization of commonly used EAs was performed. Importantly, the properties of the EAs were compared to specimens extracted from patients. In the preliminary tests of our model system, we selected a bovine EA with stiffness similar to the thrombi retrieved from the atherosclerotic plaques. This EA was used to create an occlusion in the aforesaid replica. The thrombectomy devices tested included the MERCI L5 Retriever, Penumbra system 054, Enterprise stent, and an ultrasound waveguide device. The primary efficacy endpoint was the amount of blood flow restored, and the primary safety endpoint was an analysis of clot fragments generated and their size distribution.

A physiologically realistic model system of cerebrovascular occlusion was successfully built and applied for preclinical evaluation of thrombectomy devices. The recanalization rate of the thrombectomy device was related to the ability of the device to capture the EA during the removal of the device and the geometry of the cerebrovasculature. The risk of the embolic shower was influenced by the mechanical properties of the EA and the design of the thrombectomy device.



Rights and Permissions

Copyright is held by the author, with all rights reserved.