Endovascular Perforation Model for Subarachnoid Hemorrhage Combined with Magnetic Resonance Imaging (MRI)

Summary

Here we present a standardized SAH mouse model, induced by endovascular filament perforation, combined with magnetic resonance imaging (MRI) 24 h after operation to ensure the correct bleeding site and exclude other relevant intracranial pathologies.

Abstract

The endovascular filament perforation model to mimic subarachnoid hemorrhage (SAH) is a commonly used model - however, the technique can cause a high mortality rate as well as an uncontrollable volume of SAH and other intracranial complications such as stroke or intracranial hemorrhage. In this protocol, a standardized SAH mouse model is presented, induced by endovascular filament perforation, combined with magnetic resonance imaging (MRI) 24 h after operation to ensure the correct bleeding site and exclude other relevant intracranial pathologies. Briefly, C57BL/6J mice are anesthetized with an intraperitoneal ketamine/xylazine (70 mg/16 mg/kg body weight) injection and placed in a supine position. After midline neck incision, the common carotid artery (CCA) and carotid bifurcation are exposed, and a 5-0 non-absorbable monofilament polypropylene suture is inserted in a retrograde fashion into the external carotid artery (ECA) and advanced into the common carotid artery. Then, the filament is invaginated into the internal carotid artery (ICA) and pushed forward to perforate the anterior cerebral artery (ACA). After recovery from surgery, mice undergo a 7.0 T MRI 24 h later. The volume of bleeding can be quantified and graded via postoperative MRI, enabling a robust experimental SAH group with the option to perform further subgroup analyses based on blood quantity.

Introduction

Subarachnoid hemorrhage (SAH) is caused by the rupture of an intracranial aneurysm and poses a life-threatening emergency, associated with substantial morbidity and mortality, accounting for approx. 5% of strokes^1,2. SAH patients present with severe headaches, neurological dysfunction, and progressive disturbance of consciousness³. Around 30% of SAH patients die within the first 30 days after the initial bleeding event⁴. Clinically, 50% of patients experience delayed brain injury (DBI) after early brain injury. DBI is characterized by delayed cerebral ischemia and delayed neurological deficits. Current studies have shown that the synergistic effects of several different factors lead to the loss of neurological function, including the destruction of the blood-brain barrier, the contraction of small arteries, microcirculatory dysfunction, and thrombosis^5,6.

One unique aspect of SAH is that the pathogenesis originates from an extraparenchymal location but then leads to detrimental cascades inside the parenchyma: the pathology begins with the accumulation of blood in the subarachnoid space, triggering a multitude of intraparenchymal effects, such as neuroinflammation, neuronal and endothelial cell apoptosis, cortical spreading depolarization, and brain edema formation^7,8.

Clinical research is limited by several factors, making the animal model a critical element in consistently and accurately mimicking the pathomechanistic changes of the disease. Different SAH model protocols have been proposed, e.g., autologous blood injection into the cisterna magna (ACM). Also, a modified method with a double injection of autologous blood into the cisterna magna and optic chiasm cistern (APC) respectively^9,10. While autologous blood injection is a simple way to simulate the pathological process of vasospasm and inflammatory reactions after subarachnoid hemorrhage, the following rise of intracranial pressure (ICP) is relatively slow, and no noteworthy changes in the permeability of the blood-brain barrier are induced^11,12. Another method, the periarterial blood placement, usually used in large SAH models (e.g., monkeys and dogs), involves placing anticoagulated autologous blood or comparable blood products around the vessel. The diameter changes of the artery can be observed with a microscope, serving as an indicator for cerebral vasospasm after SAH¹³.

Barry et al. first described an endovascular perforation model in 1979 in which the basilar artery is exposed after removing the skull; the artery is then punctured with tungsten microelectrodes, using a microscopic stereotactic technique¹⁴. In 1995, Bederson and Veelken modified the Zea-Longa model of cerebral ischemia and established the endovascular perforation, which has been continuously improved ever since^15,16. This method is based on the fact that mice and humans share a similar intracranial vascular network, known as the circle of Willis.

For postoperative evaluation and grading of SAH in the mouse model, different approaches have been proposed. Sugawara et al. developed a grading scale that has been widely used since 2008¹⁷. This method assesses the severity of SAH based on morphological changes. However, for this method, the mouse's brain tissue morphology must be examined under direct vision, and therefore, the mouse must be sacrificed for assessment. Furthermore, several methods for determining SAH severity in vivo have been established. Approaches range from simple neurological scoring to monitoring of intracranial pressure (ICP) to various radiological imaging techniques. Furthermore, MRI grading has been shown as a new, non-invasive tool to grade SAH severity, correlating to neurological score^18,19.

Here, a protocol for an SAH model caused by endovascular perforation is presented, combined with postoperative MRI. In an attempt to establish a system to objectify the amount of bleeding in an in vivo setting, we also developed a system for SAH grading and quantification of total blood volume based on 7.0 T high-resolution T2-weighted MRI. This approach ensures the correct induction of SAH and exclusion of other pathologies such as stroke, hydrocephalus, or intracerebral hemorrhage (ICH) and complications.

Protocol

The experiments were performed in accordance with the guidelines and regulations set forth by Landesamt fuer Gesundheit und Soziales (LaGeSo), Berlin, Germany (G0063/18). In this study, C57Bl/6J male (8-12 weeks old) mice with a weight of 25 ± 0.286 g (average ± s.e.m.) were used.

1. Animal preparation

1. Induce anesthesia by injecting ketamine (70 mg/kg) and xylazine (16 mg/kg) intraperitoneally. Maintain normal body temperature, contributing to quick induction of deep anesthesia. Test for adequate sedation with a pain stimulus, such as a toe pinch, and verify the absence of a reaction.
2. Carefully shave the neck hair of the mouse with a razor, clean it with 70% ethanol followed by betadine/chlorhexidine, and apply 1% lidocaine on the skin surface for local pain control.
3. Place the mouse in a supine position. Use tape to fix the limbs and tail, gently stretching the skin of the neck to the opposite side of the surgery. Simultaneously, elevate the neck slightly.
4. Use ophthalmic ointment (e.g., 5% dexpanthenol) to prevent dehydration of the eyes during the operation.

2. SAH induction

Figure 1: Step-by-step images of surgical technique. (A) Depiction of the exposed right carotid artery anatomy: the CCA and its bifurcation into ICA and ECA are identified, as well as the small branches of the ECA (OA and STA). (B) The ECA is mobilized from the surrounding tissue and ligated with two sutures before cutting it. A third ligation needs to be placed loosely near the bifurcation without occluding it. (C) The ICA and CCA are occluded temporarily (with either ligation or clips) to prevent excessive bleeding when the ECA is carefully incised. (D) The filament is inserted into the ECA and advanced into the CCA. The prearranged ligation must be tightened carefully so that no blood effusion occurs but advancing the filament remains possible. (E) The ICA and CCA are reopened, and the ECA stump needs to be adjusted to a cranial direction. By pushing the filament ~9 mm forward into the ICA, the ACA-MCA bifurcation will be reached, and the vessel is then perforated by pushing the filament ~3mm further. (F) The filament is withdrawn after ensuring a temporal re-ligation of the CCA. The prearranged ligation of the ECA is quickly occluded, and the CCA is reopened to allow reperfusion. Abbreviations: ACA = anterior cerebral artery, CCA = common carotid artery, ECA = external carotid artery, MCA = middle cerebral artery, ICA = internal carotid artery, OA = occipital artery, PPA = pterygopalatine artery, STA = superior thyroid artery. Scale bar = 2 mm. Please click here to view a larger version of this figure.

1. Open the neck skin with a sterile scalpel, from the chin to the upper edge of the breastbone (1.5 cm), and bluntly separate salivary glands from their surrounding connective tissue.
2. Separate the muscle group along one side [in this case, the right side] of the trachea, exposing the common carotid artery (CCA) sheath covered with nourishing blood vessels and venules. The CCA and the vagal nerve are located in close proximity to each other.
3. Dissociate the CCA and leave a free 8-0 silk suture around the CCA without ligating it in advance. Pay attention to the protection of the vagal nerve, as it is easily damaged (Figure 1A).
4. A triple bifurcation of the CCA, the ICA, and the ECA is visible along the lower posterior third of the diastasis. Dissect the distal end of the ECA and ligate the vessel twice as far distaally as possible.
5. Disconnect the ECA at the midpoint of the twice ligated segment, creating a vessel stump.
6. Prearrange one ligation for the filament around the ECA stump, do not close it until successful filament insertion.
7. Use a suture or micro clip to occlude the ICA and CCA temporarily (Figure 1B).
8. Make a small incision (approximately half of the ECA diameter) in the ECA using microvascular scissors. Insert a 5-0 (alternatively 4-0) prolene filament into the ECA and advance it into the CCA.
9. Close the ligature on the ECA slightly while loosening the micro clip on the ICA and CCA (Figure 1C).
10. Gently pull back on the filament and adjust the ECA stump in the cranial direction, invaginating the filament through the bifurcation into the ICA (Figure 1D).
11. Point the filament tip medially at an angle of ~30° to the tracheal midline and ~30° to the horizontal plane. Push the filament forward inside the ICA. After reaching the ACA-MCA bifurcation, resistance is encountered (~9 mm).
12. Advance the filament 3 mm further, perforating the right ACA. Promptly withdraw the filament to the ECA stump, allowing blood flow into the subarachnoid space.
13. Keep the filament in this position for about 10 s (Figure 1E). The presence of muscle tremors, ipsilateral miosis, gasping for breath, altered heart rhythm, and urinary incontinence can be supporting evidence of successful surgery.
14. Temporarily close the CCA to avoid excess blood loss. Pull out the filament instantly and ligate the ECA with the prearranged suture. Reopen the CCA and allow reperfusion and further effusion of blood into the subarachnoid space (Figure 1F).
15. After checking for bleeding leakage, disinfect the skin surrounding the wound to prevent postoperative skin infections, and suture the wound with a non-absorbable 4-0 polyester fiber suture.
16. Place the mouse in a thermal box until consciousness is regained. Wait until the animal is fully awake and ensure it has regained sufficient consciousness to maintain sternal recumbency. Do not return animals to the company of other mice until fully recovered.
17. Administer 200-300 mg/kg body weight paracetamol for postoperative pain relief.
18. Check on the mice daily after surgery.

3. MRI measurement

1. 24 h after surgery, perform MRI using a rodent scanner (Table of Materials) and a dedicated mouse head resonator- here, a 20 mm transmit/receive quadrature volume resonator was used.
2. Place the mouse on a heated circulating water blanket to ensure a constant body temperature of ~37 °C. Induce anesthesia with 2.5 % isoflurane in an O2/N2O mixture (30%/70%) and maintain with 1.5-2 % isoflurane via facemask under continuous ventilation monitoring.
3. First perform a fast reference scan acquiring 3 orthogonal slice packages (Tri-Pilot-Multi, FLASH with repetition time TR/echo time TE = 200 ms/3 ms, 1 average, flip angle FA = 30°, field of view FOV = 28 mm x 28 mm, matrix MTX = 256 x 256, slice thickness 1 mm, total acquisition time TA =30 s).
4. Then use a high resolution T2-weighted 2D turbo spin-echo sequence for imaging (imaging parameters TR/TE = 5505 ms/36 ms, RARE factor 8, 6 averages, 46 contiguous axial slices with a slice thickness of 0.35 mm to cover the whole brain, FOV = 25.6 mm x 25.6 mm, MTX = 256 x 256, TA = 13 min).
5. If the result is unclear, use an additional respiration triggered T2*weighted gradient echo sequence with the same isodistance as the T2w scan (2D FLASH, TR/TE = 600 ms/6.3 ms, FA = 30°, 1 average, 20 axial slices with 0.35 mm thickness, FOV and MTX identical to T2w, TA = 5-10 min depending on the respiration rate).
6. Transfer the data into the DICOM image format and use ImageJ software for SAH grading and volumetry of blood clots. Details on the quantification are listed as a step-by-step guide in the supplementary material (Supplementary Figure 1).

Representative Results

Mortality
For this study, a total of 92 male C57Bl/6J mice aged between 8-12 weeks were subjected to SAH operation; in these, we observed an overall mortality rate of 11.9% (n = 12). Mortality occurred exclusively within the first 6-24 h after surgery, suggesting perioperative mortality as well as SAH bleeding itself as the most likely contributing factors.

SAH bleeding grade
A total of 50 mice received MRI 24 h postoperatively to confirm SAH and ensure the detection of other co-occurring pathologies, including subacute ischemic stroke and hydrocephalus. The remaining animals were used for earlier scans to select the adequate time for postoperative MRI. Among the 50 examined mice at 24 h time point, n = 7 animals that did not present SAH (bleeding grade 0) and n = 5 mice in which additional stroke and/or ICH (bleeding grade IV) was detected. The SAH bleeding grade was quantified based on T2 weighted MRI scans as follows (Figure 2A,B):

grade 0: no SAH or hemorrhage identified (14%)
grade I: SAH thickness ≤0.80 mm (24%)
grade II: SAH thickness >0.8 and <1.6 mm (28%)
grade III: SAH thickness ≥1.6 mm (24%)
grade IV: SAH with either ICH and/or stroke (10%).

Figure 2: SAH grading system with corresponding blood volume and MRI images. (A) T2-weighted MRI axial sections depicting representative images categorizing SAH grade. Grade 0: no SAH or hemorrhage identified (14%); grade I: SAH thickness ≤0.80 mm; grade II: SAH thickness >0.8 and <1.6 mm; grade III: SAH thickness ≥1.6 mm; grade IV: SAH with either ICH and/or stroke. (B) Pie chart showing the distribution of SAH grade in the experimental mice. (C,E) Calculated SAH bleeding volume based on the formula V = A₁ + A₂ + ... + A_x) ·d, by which the bleeding area is determined via ImageJ on each slide section, and the sum of all bleeding areas is multiplied with the corresponding MRI slide thickness. (D) Total bleeding volume of each SAH grade based on the Kothari abc/2 volume estimation. Values are expressed as mean ± SEM. Abbreviations: ICH = intracerebral hemorrhage, MRI = magnetic resonance imaging. Scale bar = 5 mm. Please click here to view a larger version of this figure.

Bleeding volume
For grade I-III, bleeding volume was quantified by two different methods:

Method A: The total volume of bleeding was calculated based on the abc/2 volume estimation by Kathari et al., a modification of the equation for ellipsoid volume which has been utilized widely in the clinical setting to estimate ICH volume (Figure 2D)²⁰.

Method B: The calculated SAH bleeding volume was estimated based on the formula V = (A₁ + A₂ + ... + A_x) · d, by which the bleeding area was determined via ImageJ on each slide section and the sum of all bleeding areas were multiplied with the corresponding MRI slide thickness ('A_i' corresponds to the bleeding area on slice 'i', 'x' is the total number of slices, 'd' corresponds to the slice thickness). This method took the irregularity of the shape into account (Figure 2C,E). Expectedly, Method B showed a bigger range of values in each subgroup. However, both methods showed a significant difference in the corresponding bleeding grades that were based on the axial SAH thickness and are described in the following paragraph. Supplementary Figure 2 shows the SAH volume of all subgroups; expectedly, grade IV was of heterogeneous nature since it contained co-occurring ICH as well.

Statistical analysis and figures
Data were analyzed using GraphPad Prism for statistical analyses. One-way ANOVA analyses were used to compare multiple groups. The values are displayed as means ± standard errors and p-values of p < 0.05 were considered statistically significant. Elements of Figure 1 and Figure 2 were composed using BioRender.com.

Supplemental Figure 1: A step-by-step guide for quantifying bleeding volume with ImageJ. Import the images with ImageJ, and enter "Strg+I" to show the dimensional data. Then set the scale for the image. Identify all the images in which SAH can be seen. For method A, identify the slice with the biggest bleeding area and measure the craniocaudal length (=a) as well as the mediolateral length (=b) of the two orthogonal axes that span the ellipsoid SAH volume.The ventrodorsal dimension (=c) of the ellipsoid shape can be estimated based on the slice thickness and the number of slices on which SAH is seen [c = slice thickness x number of slices]. Calculate the volume based on the formula:V= abc/2. For method B, measure the bleeding areas on each slice separately and then calculate the volume based on the formula: V = (A₁ + A₂ + ... + A_x) · d, by which d= slice thickness. Please click here to download this File.

Supplemental Figure 2: Bleeding volumes of all subgroups. (A) Bleeding volume (mm³) in each subgroup based on method A using the formula V= abc/2. (B) Bleeding volumes (mm³) of the corresponding subgroups using method B (formula V = (A₁ + A₂ + ... + A_x) · d; d= slice thickness). Please click here to download this File.

Discussion

In summary, a standardized SAH mouse model induced by endovascular filament perforation operation is presented with minor invasion, short operative time, and acceptable mortality rates. MRI is conducted 24 h postoperatively to ensure the correct bleeding site and the exclusion of other relevant intracranial pathologies. Furthermore, we classified different SAH bleeding grades and measured bleeding volumes, allowing further subgroup analyses based on bleeding grade.

Adequate positioning of the mouse affects the success of the correct perforation. The mouse's neck should be stretched slightly to the opposite side of the operation, with the head being slightly elevated. This exposes the trifurcation and makes the puncture path easier accessible. If advancing of the filament fails, it can be helpful to withdraw the filament slightly to the trifurcation and adjust the head's position until advancing is possible without any resistance.

Intraoperative nerve protection is critical. Disturbances of the vagal nerve and cervical plexus can cause changes in respiratory and cardiac rhythms, and some mice may even die because of malignant arrhythmias. If these symptoms occur, it is essential to pause the procedure for a few minutes until the breathing and heart rate stabilize.

Reducing intraoperative blood loss is vital for improving the survival of mice. Based on our experience, double suture ligation is best applied close to the ECA. We disconnect the ECA in the middle of the two ligations to prevent blood backflow from the distal ECA stump. When the filament is inserted into the ECA, the prearranged suture should be ligated to prevent blood effusion from the incision. It is critical not to ligate the vessel too tightly as this hinders proper filament advancement.

Appropriate depth of filament insertion is essential for successful SAH induction. Due to the age of the mice used (8-12 weeks), we insert the filament ~9 mm inside the ICA and stop when resistance was encountered, then advanced ~3 mm further for perforation. Inserting the filament not deep enough could result in insufficient perforation, causing no SAH, whereas excessive insertion might lead to stroke and/or ICH (Figure 3). At the same time, the mice's original anatomy and vascular structures need to be preserved as well as possible during the operation. For example, the occipital artery (OA) or superior thyroid artery (STA), and nourishing blood vessels on the sheath, should be retained as much as possible.

Figure 3: Mouse brain anatomy and macroscopic images of SAH. (A) Schematic mouse vascular anatomy showing site of filament perforation. (B) Classical macroscopic image of successful induction of SAH. Before removing the brain, a perfusion of 1x PBS was performed. (C) Macroscopic view of the mouse in which the filament was pushed too deep, causing ICH. Abbreviations: ACA = anterior cerebral artery, ECA = external carotid artery, CCA = common carotid artery, ICA = internal carotid artery, ICH = intracerebral hemorrhage, L = left, MCA = middle cerebral artery, PPA = pterygopalatine artery, R = right. Scale bar = 3 mm. Please click here to view a larger version of this figure.

The endovascular perforation model is a commonly used animal model to study SAH but the means to ensure bleeding grade and exclude other pathologies such as stroke or intracerebral hemorrhage are not sufficiently standardized in the literature²¹. Just like any operative animal model, the success rate and robustness of SAH induction depend on the experience of the surgeon.

Currently, the endovascular perforation model is one of the most popular methods of experimental SAH induction in mice. This approach does not require craniotomy and accurately resembles the processes taking place in humans suffering from aneurysmal SAH²². Advantages include close imitation of the pathophysiology following aneurysmal SAH, regarding acute and delayed reactions²³. Additionally, mortality rates in this model have been shown to be similar to those of clinical studies in patients suffering from aneurysmal SAH²³. In comparison to blood injection models, changes in blood-brain barrier permeability are more closely mimicked, and higher rates of vasospasm are achieved in filament perforation^11,24. Blood injection models are more invasive and therefore pose a greater risk for tissue damage when compared to the less invasive endovascular perforation model. Nonetheless, it should be noted that a major advantage of blood injection methods is the easily controlled blood volume²³. The standardization of injection speed is important to consider since alterations of ICP are heavily dependent on the speed of injection²³. Apart from these classical models, the combination of elastase injection to induce aneurysm formation and hypertension by unilateral nephrectomy, ultimately leading to aneurysm rupture, poses an interesting model to study subarachnoid hemorrhage in a more pathophysiologically realistic setting²⁵. Integrating such techniques with genetically modified mice will be of interest for future studies.

Previous SAH grading systems for the filament perforation model are based on the amount of visible subarachnoid blood in different brain segments after the mouse has been sacrificed¹⁷. Consequently, these grading systems do not allow long-term studies when the blood has been already resorbed at the time of sacrification. In the clinical setting, SAH is graded based on clinical presentation as well SAH thickness on imaging, corresponding to clinical outcome^1,26,27,28. Hence, in an attempt to classify the bleeding severity noninvasively, we added a standardized MRI follow-up examination to grade SAH radiographically, by which the grading was based on pre-existing human grading scales, adapting the grading system of a previously published MRI grading system in SAH mice by Egashira et al.¹⁸. This approach also ensures quantification of total blood volume and exclusion of animals with other co-occurring intracranial pathologies (e.g., stroke, ICH, hydrocephalus). Some studies proposed intracranial pressure (ICP), cerebral perfusion, and blood pressure monitoring as evidence of successful SAH induction, which might be additional helpful tools²⁹. Indirect ways to grade the severity of SAH and potential intraparenchymal damage include combining clinical findings with histological staining for cell death markers such as p53, TUNEL or caspase-3. However, these indirect tools such as ICP monitoring as well as neurological may not distinguish neatly other pathologies such as stroke, intracranial hemorrhage, or hydrocephalus. Despite the advantages of MRI grading, there is one major drawback of this approach regarding its feasibility: MRI is not as widely available to laboratories as other methods. This limits the broad introduction of MRI grading systems in experimental SAH. When available, however, the presented MRI grading system adds a tool to standardize experimental SAH models, therefore facilitating reproducibility and comparability of the experiments²³. In this study, despite observed clinical changes during the operation, there was still a 14% rate of mice without evidence of SAH on postoperative MRI. Possibly, mice in this subgroup suffered from microhemorrhages, not detectable on MRI (similar to SAH patients with negative CT but the presence of xanthochromia in lumbar puncture). These mice were excluded in this experimental setup for further analyses. The technical reason for these "no-bleeds" on MRI could be insufficient filament insertion, resulting in no perforation (e.g., by incorrect placement into OA or pterygopalatine artery (PPA)). Additionally, the successfully perforated vessel might close again after withdrawal of the filament, preventing SAH.

In summary, a standardized model for experimental aneurysmal SAH by endovascular perforation is presented, combined with MR imaging 24 h after surgery to confirm and grade the bleeding and to exclude other relevant intracranial pathologies.

Materials

Name	Company	Catalog Number	Comments
Eye cream	Bayer	815529836	Bepanthen
Images analysis software	ImageJ		Bundled with Java 1.8.0_172
Ligation suture (5-0)	SMI		Silk black USP
Light source for microscope	Zeiss	CL 6000 LED
Ketamine	CP-pharma	797-037	100 mg/mL
MRI	Bruker	Pharmascan 70/16	7 Tesla
MRI images acquired software	Bruker		Bruker Paravision 5.1
Paracetamol (40 mg/mL)	bene Arzneimittel	4993736
Prolene filament (5-0)	Erhicon	EH7255
Razor	Wella	HS61
Surgical instrument (Fine Scissors)	FST	14060-09
Surgical instrument (forceps#1)	AESCULAP	FM001R
Surgical instrument (forceps#2)	AESCULAP	FD2855R
Surgical instrument (forceps#3)	Hammacher	HCS 082-12
Surgical instrument (Needle holder)	FST	91201-13
Surgical instrument (Vannas Spring Scissors)	FST	15000-08
Surgical microscope	Zeiss	Stemi 2000 C
Ventilation monitoring	Stony Brook		Small Animal Monitoring & Gating System
Wounding suture(4-0)	Erhicon	CB84D
Xylavet	CP-pharma	797-062	20 mg/mL