Establishment of a novel hemodynamic cerebral ischemia model of atherosclerotic rabbit
Correspondence Address: Source of Support: The National Natural Science Foundation of China (NO. 30872678), Conflict of Interest: None DOI: 10.4103/0028-3886.63782
Source of Support: The National Natural Science Foundation of China (NO. 30872678), Conflict of Interest: None
Background: Atherosclerosis is the most common cause of ischemic stroke. Until now, there has been no ideal animal model for studying the hemodynamic ischemia caused by atherosclerosis in posterior circulation. Objective: This study was conducted to establish a cerebral ischemic model in an atherosclerotic rabbit by subclavian artery (SA) ligation or embolization, emphasizing the role of hemodynamics. Materials and Methods: Thirty-two male New Zealand rabbits (weight, 3.5 to 4.0 kg; provided by Animal Laboratory of Tongji Hospital of Tongji University) were randomly divided into four groups. All the groups, except the controlled group, were fed with high lipid diet for three months and two of the groups were submitted for right SA ligation or left SA embolization and the remaining one group to non-operated group. Results: A SA steal syndrome was established successfully with delayed visualization and prolonged circulation time in the posterior circulation. The circulation time was different when SA was ligated or embolized (P<0.001), and also there was statistically significant difference between controlled and non-operated groups when angiography of either, left and right, VA was performed (P=0.009, P=0.003, respectively). However, there was no statistical significant difference between the ligation and embolization group either before or after the procedure (ligation or embolization) (P=0.402, P=0.068, respectively). In the control and non-operated group, angiography of either left or right VA showed no statistically significant difference (P=0.285, P=0.197, respectively). In the groups fed with high lipid diet for three months, atherosclerotic plaques were found in the common carotid arteries (CCAs), aortic arch (AAs) and variably in the internal carotid arteries (ICAs). Conclusions: This model could, to some extent, simulate the clinical hemodynamic state of atherosclerotic stroke. and it is an ideal model for studying the hemodynamic cerebral ischemia in posterior circulation.
Keywords: Atherosclerosis, hemodynamic cerebral ischemia, model, rabbit
Atherosclerosis is the most common cause of ischemic stroke. Experimental study of ischemic stroke has made great progress due to the establishment of all kinds of animal models. These animal models involve either selective or nonselective occlusion of the intra-and/or extracerebral arteries. The animal models are of three types: focal ischemia, global ischemia, or focal with subsequent global ischemia. Such models include blocking blood flow by advancement of an intraluminal thread into the middle cerebral artery (MCA), ,,, transection of the MCA,  or infusion of foreign bodies. ,,, Additionally, models of thromboembolic stroke including photochemical sensitization of common carotid artery thrombosis (CCAT)  have also been developed. Artificial arteriovenous fistula, ,,,,, air-desiccation, balloon-overstretch injury ,,,, and silastic collar ,, were used to produce chronic brain hypoperfusion and/or atherosclerotic plaque in CCA. However no animal models have been developed to mimic the atherosclerosis associated hypoperfusion state in posterior circulation. The problem in this regard mainly lies in the difficulties to develop an atherosclerotic plaque either in vertebral artery (VA) or basilar artery (BA) in small animal models. This may be partly related to the small caliber of the VA and its deep location. In this paper we discuss two new atherosclerotic rabbit models with subclavian artery (SA) steal syndrome.
Experiments were performed in accordance to the Law of People's Republic of China on the Protection of Animals and the US National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH publication No. 85-23, revised 1996).
Thirty-two male New Zealand rabbits (weight, 3.5 to 4.0 kg; provided by Animal Laboratory of Tongji Hospital of Tongji University) were randomly divided into four groups: control group, non-operated group, ligation group and embolization group and eight rabbits were included in each group. Animals in all the groups, except the controlled group, were fed with high lipid diet (1% cholesterol, 5% lard 5% yolk and 89% basic feedstuff) for three months.
Ligation and evaluation
Rabbits in the ligation group underwent right SA ligation proximal to the origin of right VA under general anesthesia by intravenous injection of 3% pentobarbital sodium (1 ml/kg). To epilate the hair in the cervicothoracic region, 8% sodium sulfide was used. After disinfection and local anesthesia by hypodermic injection of 1% lidocaine (0.5 ml/kg), a straight-line-incision of 45 mm length was made in the midline of the sternum. Subcutaneous tissue and cervical muscle were separated layer by layer. The right CCA could be visualized after the separation of CCA sheath, and similarly brachiocephalic trunk and right SA were also exposed. The right SA was ligated distal to the origin of right CCA [Figure 1]a. During the operation procedure, all the bony structures were preserved. After the ligation procedure was over, the cervical muscle and skin were sutured. One week after the ligation of right SA, selective digital subtraction angiography of the aortic arch (AA) [Figure 1]b, left VA [Figure 1]d and both internal carotid arteries (ICAs) [Figure 1]e were performed under general anesthesia to visualize and also verify the SA stealing and hypoperfusion in posterior circulation.
Process of digital subtraction angiography and embolization
Epilation of the hair in both inguinal areas was done under general anesthesia and a straight-line-incision was made over the right femoral artery after appropriate disinfection. Cardoverine was used to sodden the right femoral artery spasm. An 18-gauge trochar (TEROMO Company, Japan) was used to puncture the right femoral artery and a 4F-artery-sheath (TEROMO Company, Japan) was then introduced by Seldinger technique. To preserve renal blood flow, the femoral sheath was negotiated to a distance under 30 mm. After heparinization (200 U/kg), a 4F-angiographic catheter (VER 135°, Cordis Johnson and Johnson Company, USA), was used to perform AA angiography. The arterial sheath and the Y-type valve connected with the 4F-angiographic catheter were kept on pressured instillation. Selective angiography in the arteries arising from AA was performed through micro catheter (Prowler-10, Cordis Johnson and Johnson Company, USA) assisted by micro guide wire (SilverSpeed-10, MTI Company, USA). The volume of contrast (Omnipaque 300 mgI/ml, GE Company, USA) injected by hand into the AA, CCA, ICA and VA was 3.0 ml, 1.0 ml, 0.8 ml and 0.6 ml respectively. And the volume in the ICA and VA for three-dimensional (3D) rotation angiography was 1.0 ml [Figure 1]a-i.
Stainless steel embolization coils (COOK Company, USA) of 3 mm diameter and 20 mm length were used to embolize the left SA proximal to the origin of left VA. Guide wire of 0.035 inch was used to push the coil to the target position by 4F-angiographic catheter under the road map. Before and after embolization, selective digital subtraction angiograms of the AA [Figure 1]f and g, and right VA [Figure 1]i were performed. Low molecular heparin (600 U/kg/d three times a day was administered subcutaneously to prevent thrombosis in the left VA.
After the angiography all the rabbits were sacrificed by overdose of anesthesia. The ICA, CCA and AA were collected and fixed in 4% formalin for pathohistological studies (HE staining).
SPSS 15.0 software package was used in the statistical analysis. Paired sample T test was used in this study. A value of P < 0.01 was considered statistically significant.
SA steal syndrome could be demonstrated by AA angiography in both ligation and embolization groups [[Figure 1]c-e, h and i]. VA angiography on the side contralateral to the side of ligation or embolization showed no perfusion of VA and BA and the contrast entered the SA distal to the point of ligation or embolization through ipsilateral VA. There was delayed visualization and the circulation time was prolonged and longer than the pre-ligation and pre-embolization state. ([Table 1], P < 0.001). There were differences between the control and non-operated groups when either of the VA angiography were performed (P=0.009, P=0.003, respectively), but the difference was not statistically significant between the ligation and embolization groups either before or after the procedure (ligation or embolization) (P=0.402, P=0.068, respectively). In both the control group and non-operated group, angiogram of the left VA and the right VA showed no statistically significant differences (P=0.285, P=0.197, respectively). In the three groups fed with high cholesterol diet, atherosclerotic plaques could be visualised in some of the ICAs (8/48), CCAs (16/48) and in all AAs (24/24) [[Figure 1]j and k] after 3D reconstruction. This finding was later confirmed by pathohistological sections [Figure 1]l.
By these experimental animal models we could successfully create SA steal syndrome by surgical ligation and endovascular embolization of SA. This novel model vividly simulates the atherosclerotic ischemia in the posterior circulation in the clinical setting. This model probably provides a specific condition to study cerebral arteriogenesis in atherosclerotic rabbits. Direct occlusion of the intracranial arteries, such as MCA occlusion by nylon-segment or electric coagulation, would result in infarction in ihe arterial territory but not hemodynamic ischemia. Artificial arteriovenous fistula would not simulate the ischemic state caused by atherosclerosis in posterior circulation, as the venous hypertension state caused by arteriovenous fistula is unusual in patients with atherosclerotic ischemia in the posterior circulation. The balloon injury and silastic collar around the CCA were useful in the anterior circulation, but they didn't work in posterior circulation because of small diameter and deep anatomic location of VA.
Though the hemodynamic cerebral ischemic models were established by different methods, hypoperfusion state was found in the posterior circulation in both the models. In terms of circulation time, there were no statistically significant differences between ligation and embolization group either before or after the procedure (ligation or embolization) (P=0.402, P=0.068, respectively). In both the ligation and embolization group, the circulation time in the posterior circulation was significantly different before and after the procedure (P < 0.001). The basilar artery was not visualized clearly, because more contrast was shunted into the contralateral SA. Angiography of both the ICAs showed more contrast in the upper segment of the BA. These observations suggest that there is hemodynamic failure and hypoperfusion in the posterior circulation. Another important finding in these experimental models was vriable occurrence of atherosclerotic plaque in different arteries. The plaques were least in the ICA, but were maximum in AA. These plaques were probably the result of shearing force in different arteries.
There are some differences between the models mostly procedure-related; operator technical difficulties, the need for more equipment and material, and procedure-related complications. Ligate the right SA is relatively easy if the operator is familiar with the anatomy of the arteries arising from AA. A straight-line-incision should be made along the middle line of the sternum, and lamina pretrachealis should be preserved. The right musculus sternocleidomastoideus, beneath which the CAA and the vagus nerve lie,should be cut. Extra care should be taken not to touch any tissues outside the right musculus sternocleidomastoideus, because of the location of jugular vein. Any injury to the jugular would cause fatal hemorrhage. The right CCA should work as a landmark, along which the right SA and brachiocephalic trunk should be exposed. The ligation point should be at the subclavian distal to the origin of the right CCA and the ligation point should be as close as one can get to the origin of the right CCA, so that the right VA would be preserved. The right SA is the recommended target vessel for ligation because it is at a much higher location over the AA than the left SA. Another reason is that the right SA is relatively shallow compared to the left. This is due to the angle between the coronal plane and the AA-located-plane. Compared with ligation, embolization of the left SA has an advantage of less injury, but it requires operator's adeptness at neuroradiological practice. To avoid delayed thrombosis of the left VA, low molecular heparin is required. The left SA is the recommended target vessel for embolization, because it is easy for the guiding catheter to reach the target
Many thanks to WU Chun-hong and YU Li-min for their technical assistance. Special thanks to professor XIA Yu-rong for her painstaking efforts in revising this paper. This work was funded by the National Natural Science Foundation of China (NO. 30872678).