Intraluminal filament occlusion of the middle cerebral artery is the most frequently used in vivo model of experimental stroke in rodents. An alternative surgical approach to allow common carotid artery repair is performed here, which allows the reperfusion of the common carotid artery and a full reperfusion to the middle cerebral artery territory.
The ischemic stroke is a major cause of adult long-term disability and death worldwide. The current treatments available are limited, with only tissue plasminogen activator (tPA) as an approved drug treatment to target ischemic strokes. Current research in the field of ischemic stroke focuses on better understanding the pathophysiology of stroke, to develop and investigate novel pharmaceutical targets. Reliable experimental stroke models are crucial for the progression of potential treatments. The middle cerebral artery occlusion (MCAO) model is clinically relevant and the most frequently used surgical model of ischemic stroke in rodents. However, the outcomes of this model, such as lesion volume, are associated with high levels of variability, particularly in mice. The alternative MCAO model described here allows the reperfusion of the common carotid artery (CCA) and the increased perfusion of the middle cerebral artery (MCA) territory, using a tissue pad with fibrinogen-based sealant to repair the vessel, and the improved welfare of the mice by avoiding external carotid artery (ECA) ligation. This reduces the reliance on the Circle of Willis, which is known to be highly anatomically variable in mice. Representative data show that using this alternative surgical approach decreases the variability in lesion volumes between the traditional MCAO approach and the alternative approach described here.
A major cause of cerebral stroke is focal ischemia in the territory of the middle cerebral artery. Tissue plasminogen activator (tPA) is the only pharmacological treatment available with proven efficacy, despite numerous clinical drug trials targeted to ischemic stroke1,2. However, due to safety concerns and a narrow therapeutic window (< 4.5 h), only ~15% of all stroke patients are eligible to receive tPA, and the recanalization rates can be < 50%3,4.
Reproducible and clinically relevant animal models of stroke are considered essential to inform the development of new and potential stroke therapeutic treatments. However, due to concerns regarding consistency and variability in the outcomes with animal models, it remains important to refine existing in vivo models to improve the translation from preclinical studies to the clinic. The lack of translation from the preclinical experimental efficacy of potential treatments to clinical use is an ongoing concern for stroke research5. Reasons for failure of translation are likely to be multiple and may be related to, for example, the trial design, treatment delays, clinical stroke heterogeneity, and the limitations of the animal models used6. A key challenge for stroke research remains the development of safe and effective treatments.
Middle cerebral artery occlusion (MCAO) by intraluminal filament insertion is the most frequently used in vivo rodent model of experimental stroke. This model allows the restoration of the blood flow after an ischemia induction, mimicking the events that occur in human stroke7. However, in particular in mice, heterogeneous lesion volumes with varied standard deviations occur even though defined surgical protocols are applied8,9,10. It is typical to see a bimodal distribution of small striatal and large striato-cortical lesion volumes11. To induce ischemia, the filament is typically inserted through an incision of the CCA or ECA which then remain permanently ligated12. The permanent ligation of the CCA prevents the re-establishment of the blood flow into the internal carotid artery (ICA) and, subsequently, the MCA territory. This causes reperfusion to be reliant on the collateral supply within the Circle of Willis (CoW). The CoW structure has anatomical variability between individual animals, particularly in C57BL/6 mice-a strain typically used in in vivo stroke research13. An alternative method, of filament insertion through the ECA, does allow the continued perfusion through the CCA, but this method compromises the arterial supply to the ECA territory, which has been shown, in rats, to have a detrimental effect on the animal's well-being14.
The reliance on the CoW for collateral supply and reperfusion in the established MCAO model may, in part, account for the lesion volume variability following occlusion. We describe an alternative murine surgical procedure where ECA ligation is avoided and the CCA incision is repaired, thus permitting reperfusion through the CCA, independent of the CoW. The repair of the CCA incision has been previously shown in rats to result in a successful reperfusion through the CCA15. We have applied this approach successfully in mice11 and report here the protocol which results in a reduced variability in lesion volume, the main outcome measure used in experimental stroke studies.
In this protocol, we demonstrate how to undertake MCAO through CCA vessel filament insertion followed by CCA vessel repair, which involves a tissue pad and sealant application to permit reperfusion.
This protocol and the data reported were conducted in accordance with the UK Animals (Scientific Procedures) Act, 1986 (Project license 60/4315) and following institutional ethical approval. All experiments are reported in accordance with the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines16.
1. Preparation
2. Middle Cerebral Artery Occlusion Surgery
3. Post-occlusion
4. Post-operative Care
5. Magnetic Resonance Imaging and Image Processing
A total of 24 adult male C57BL/6 mice, weighing between 24–31 g at the time of surgery, were used in the study. One animal died following middle cerebral artery occlusion (MCAO) and one was excluded due to surgical complications. The data presented here are taken from a previously published work by the authors. These were used to illustrate the effect of vessel repair on MCAO outcomes11. All data are expressed as thr mean ± standard deviation. The data were statistically assessed for normality using the D'Agostino-Pearson omnibus normality test. Parametric data were compared using Student's t-test (for two means) and one-way ANOVA with the Sidak test (multiple means). Non-parametric data were compared using the Mann-Whitney U test. The variability of the parametric data was assessed using an F-test, and non-parametric data variability was assessed using Levene's test.
Typically, in MCAO procedures, the occluding filament is inserted into the CCA and the ECA is ligated to prevent this filament from passing into the ECA rather than the ICA. An avoidance of ECA ligation and the addition of analgesia showed a trend toward reduced weight loss at 48 h post-MCAO, when compared to data from previous studies undertaken by the same surgeon for the same MCAO time using ECA ligation with no analgesia, whereas the LV appeared unaffected, see Figure 1.
Mice underwent a 60 min MCAO-induced ischemia followed by reperfusion with CCA vessel repair or with the typical ligation of the CCA approach. A schematic of the ligated and unligated repaired CCA is shown in Figure 2.
Laser Doppler flowmetry was used to confirm the blood flow perfusion in the territory of the MCA at MCAO, before and after the CCA vessel repair. Figure 3 demonstrates that 5 min following the filament removal, the regional cerebral blood flow (rCBF) significantly increased in the brain region of the MCA. The perfusion was maintained up to the vessel repair, with an increase in the perfusion to the MCA territory shown following the CCA vessel repair, suggesting that the CCA repair allowed an increased blood perfusion to the ischemic territory compared to reliance on the Circle of Willis alone.
T2-weighted MRI was used to determine the total LV, and DTI scans were used to determine the core LV, 48 h after the MCAO. Figure 4A shows no significant difference in the total or core LV between the repair and ligated procedure groups. However, the data variability for both total and core LV, as assessed using Lavene's test for non-parametric or the F-test for parametric data, was significantly reduced within the CCA repair group. The total LV was broken down into cortical and subcortical LV, as shown in Figure 4B. The cortical portion was significantly less variable in the CCA repair group, whereas the sub-cortical portion of the lesion was unaffected between the two procedural groups.
A power analysis indicated that fewer animals per treatment group would be required to demonstrate a 30% reduction in the LV following MCAO using CCA repair versus the typical CCA-ligated procedure, see Table 1. An assumption of power 1-β= 0.8 and significance level α = 0.05 and a prediction of 30% reduction in the LV between the hypothetical control and test groups were used for the power analysis. Furthermore, an equal variance was assumed between the groups. Table 1 shows the number of animals required for the test and control groups when either the typical CCA-ligated method is used or the updated CCA repair method, as described here, is used. Note that the test group refers to a hypothetical treated group of animals and the control group refers to a hypothetical control group; both groups would undergo MCAO.
Figure 1: Combined analgesia treatment and the omission of ECA ligation on outcomes following MCAO. (A) Body weight, shown as a percentage of pre-MCAO weight, decreased the first 2 d following the MCAO for both groups. The ECA-unligated group (analgesia-treated with no ECA ligation at MCAO) showed a trend toward reduced weight loss on the second day following the MCAO. (B) This panel shows the lesion volume (mm3) measured by standard triphenyltetrazolium chloride (TTC) staining 48 h after the MCAO. The data shown are the mean ± standard deviation. ECA ligated: n = 17, ECA unligated: n = 10. This figure has been modified from Trotman-Lucas et al.11. Please click here to view a larger version of this figure.
Figure 2: Schematic showing the standard CCA method and the alternative CCA repair method following MCAO. (A) This schematic depicts a permanently ligated CCA using non-dissolvable sutures applied to either side of the CCA incision, resulting in the permanent ligation of the right CCA. (B) This schematic depicts the alternative CCA repair method. A small tissue pad coated with fibrinogen and thrombin sealant is used to cover the CCA incision, sealing it to allow the full perfusion of the right CCA. This figure has been modified from Trotman-Lucas et al.11. Please click here to view a larger version of this figure.
Figure 3: Regional cerebral blood flow (rCBF) parameters following MCAO. The rCBF changed 5 min after the MCAO filament removal for both the CCA-ligated and CCA-repaired groups (post-MCAO), relative to the rCBF measured during MCAO. This panel shows the rCBF data immediately before the CCA vessel repair (pre-repair) and 5 min following the CCA repair (post-repair). Significant increases in the rCBF are shown 5 min after the filament removal (post-MCAO) in both groups. An additional increase in the rCBF is shown following the CCA repair (post-repair) in the CCA repair group. No difference in the rCBF is shown between 5 min post-MCAO and pre-repair. The data shown are condensed from the analyzed time-course data reporting key time points, here as the mean ± standard deviation. CCA ligated: n = 10, CCA repaired: n = 10; **P < 0.01, ***P < 0.001, ns: non-significant. This figure has been modified from Trotman-Lucas et al.11. Please click here to view a larger version of this figure.
Figure 4: Analysis of the lesion volume obtained by MRI techniques. (A) This panel shows the lesion volume (LV; mm3) at 48 h following MCAO, the total LV taken from T2-weighted MRI images (Total LV), and the core LV taken from DTI scans and analysis (Core LV). Representative images show the total lesion volume from a T2 scan slice image with the DTI core lesion volume mask applied. The variability within the groups was significantly reduced for both the total LV (P = 0.015, CCA repair: n = 10, CCA ligated: n = 10, F-test) and the core LV (P = 0.043, CCA repair: n = 9, CCA ligated: n = 6, Lavene's test), assessed using an F-test for parametric data or Levene's test for non-parametric data. (B) This panel shows the LV at 48 h following MCAO, taken from T2-weighted MRI images and divided into cortical and subcortical lesion areas. The CCA repair significantly reduced the data variability (P = 0.03, F-test) in the cortical portion of the lesion, but no effect on the data variability in the subcortical portion of the lesion was shown. CCA repair: n = 10, CCA ligated: n = 10. The data shown are the mean ± standard deviation. #P < 0.05 (F-test), xP < 0.05 (Lavene's test). This figure has been modified from Trotman-Lucas et al.11. Please click here to view a larger version of this figure.
Approach | Lesion Volume (LV; mm3; mean ± s.d.) |
Power | Significance Level | Anticiapted difference | Group size required |
CCA Ligated (traditional approach) | 94.08 ± 53.79 | 0.8 | 0.05 | 30% | n = 58 |
CCA Repaired (new approach) | 51.73 ± 22.78 | 0.8 | 0.05 | 30% | n = 35 |
Table 1: Representative power analysis comparing traditional CCA ligation with the alternative CCA repair method explained here. This table shows the power analysis conducted to calculate the anticipated group size required to detect a significant difference in the LV between a control group, the traditional or alternative (new) approach, and a test group (predicted). The table shows the group sizes as required if a power of 0.8 is assumed, a significance level of 0.05 is applied, and if the predicted test group shows a 30% difference in the LV compared to the control group. The table shows results for both MCAO approaches (the CCA-ligated and the CCA-repaired) to determine if there is a difference in the number of animals required to gain a 30% difference in the LV. For both methods, an equal variance is assumed between the test and the control group. This figure has been modified from Trotman-Lucas et al.11.
Filament induction of transient MCAO in rodents is the most frequently used experimental stroke model, as it allows reperfusion to the affected area, mimicking the occurrence of events following clinical ischemic stroke7. Reported here is an alternative surgical approach to the traditional method of filament-induced transient MCAO in mice. The alternative approach, involving analgesia treatment, ECA ligation avoidance, and CCA incision repair, results in a reduced LV variability when assessed using both MRI and histological staining methods11.
Traditional approaches to induce MCAO largely rely on the transection, or at least ligation, of the ECA, which has been shown, in rats, to affect drinking behavior and an increase in body weight loss following the MCAO14. The protocol defined here, in mice, with the avoidance of ECA ligation and analgesia addition, suggested a reduction in body weight loss following the MCAO with no effect on the lesion volume. The use of analgesia is avoided, or at least not reported, in the majority of experimental stroke studies, due to possible confounding effects on the experimental outcome. However, avoiding analgesia completely is not always justified and there is a need to balance the welfare needs of the animals with the achievement of the scientific objectives.
Differences in animal size, strain, and cerebrovascular anatomy, in addition to filament size and type variations, are all suggested to influence stroke outcomes23,24. The alternative approach described here avoids the reliance on the CoW during reperfusion, thus reducing, at least in part, the variability seen between animals in lesion volume. CoW anatomy is highly variable in mice, in particular in the C57BL/6 strain, which is often used in experimental stroke studies. 90% of C57BL/6 mice have an incomplete CoW due to a varied posterior communicating artery (PcomA) patency, which may have an effect on the volume of ischemic damage due to the insufficient perfusion of structures outside of the MCA territory13,25. Repairing the CCA in mice, as shown here, results in the re-establishment of blood flow via the CCA to the ischemic area, as previously described in rats15. The representative data here show that the repair of the CCA increases reperfusion, although the blood flow in the CCA was not directly measured. However, it is possible for the surgeon to visualize the CCA reperfuse with blood following the vessel repair, as it returns to a pulsing and full state all along the trunk, proximal and distal to the repair location. This visual confirmation, along with laser Doppler flowmetry readings of the ischemic area, can be used to confirm the successful repair of the vessel. The time between the tissue pad application and the removal of the vessel clip from the CCA can have an impact on the resulting patency of the CCA, as reducing the time between the tissue pad application and the clip removal will prevent the tissue pad from adhering to the opposite side of the CCA. Although technically challenging, the alternative MCAO procedure explained here does not require any additional skills than those required to perform the surgical induction of MCAO in mice.
Traditionally associated with a high variability in outcome measures, experimental stroke studies may have a tendency to be underpowered. Ethical and welfare requirements in combination with economic and practical concerns may contribute to studies being underpowered. By reducing the variability in outcome and, therefore, producing more consistent lesion outcomes across an experimental group, more effective power calculations can be performed with the ultimate aim of studies being appropriately powered.
In conclusion, this alternative CCA repair procedure, in mice, results in less variability in lesion volume following experimental stroke and enables smaller experimental groups to be required for testing a treatment effect when appropriate power calculations are used.
The authors have nothing to disclose.
This work was funded by the National Centre for the Replacement, Refinement and Reduction of Animals in Research (NC3Rs; NC/M000117/1 to CG). The authors thank the staff of the Division of Biomedical Services, University of Leicester, for their care of the experimental animals and Maria Viskaduraki for her statistical advice. The representative results are adapted with permission from Disease Models & Mechanisms11.
0.7mm flexible single fibre optic probe | Moor Instruments, UK | P10d | Use with master probe code: VP10M200ST |
7-0 silicone coated monofilament | Doccol, USA | 701956PKRe | Item dependent on animal size and weight – use manufacteurer guidelines. Product code here was used for representative results shown in article. |
9.4T Preclinical MRI system | Agilent Technologies, Santa Clara, California, USA | MY11520101 | Equipped with gradient and RF coils suitable for mouse brain imaging |
Animal monitoring and gating equipment | SA Instruments, Stony Brook, New York, USA | 22124005 | MRI compatible temperature and respiration monitoring |
Bupivacaine | National Veterinary Services, Stoke-on-trent, UK | 512345 | Marcaine |
C57BL/6 Mice | Charles River, Oxford, UK | B6JSIMA49D | |
Carprofen | Norbrook Laboratories | 143658 | Carprieve 5% w/v Small animal solution for injection |
Chlorhexidine 4% hand cleanser solution | VWR International Ltd, Lutterworth, UK | MOLN10008780 | HibiScrub Antimicrobial hand cleanser, Molnlycke Health Care |
Cotton buds | National Veterinary Services, Stoke-on-trent, UK | 213512 | Any plastic body, cotton bud tip are suitable once made sterile by autoclaving. |
Dissecting stereoscope | Carl Zeiss | OPMI99 | Resident piece of equipment. Any binocular dissecting stereoscope capable of x1-x5 magnification will be suitable. |
dissolvable 6-0 sutures | National Veterinary Services, Stoke-on-trent, UK | 9544 | Absorbable Sutures Ethicon Coated Vicryl 6/0 (Ethicon code: W9981) |
Donut probe holder | Moor Instruments, UK | PHDO | Probe holder for mouse, required to be used with single fibre optic probe when used with laser doppler flowmtry machine. |
dumont #5 forceps | World Precision Instruments, Hertfordshire, UK | 500342 | |
Fibrinogen and thrombin sealant | Baxter, Berkshire, UK | 1502243 | TISSEEL Ready to use solutions for Sealant 2ml |
Gel food | Datesand group, Manchester, UK | 72065022 | Diet Gel Recovery |
Image display and measuring software package | 3D Slicer | https://www.slicer.org/ | Version 4.0 |
Image display and measuring software package | NIH, Maryland, USA | https://imagej.nih.gov/ij/index.html | NIH/ImageJ |
LDF monitor | Moor Instruments, UK | moorVMS-LDF | |
micro vannas scissors | InterFocus Ltd, Linton, UK | 15000-08 | Other microvannas spring scissors can be used as an alternative, although fine tips are required. |
Microvascular clip | World Precision Instruments, Hertfordshire, UK | 15911 | 10 G Vessel Clip |
microvascular clip holders | World Precision Instruments, Hertfordshire, UK | 14189 | |
MRI acquisition and analysis software | Agilent Technologies, Santa Clara, California, USA | VnmrJ Version 4.2 | Revision A |
no. 15 scalpel | Scientific Laboratory Supplies, Nottingham, UK | INS4678 | Sterile No15 Scalpel – manufactuer number P305. Other suppliers are available. |
Non-disolvable 6-0 suture | National Veterinary Services, Stoke-on-trent, UK | W529 | Ethicon Mersilk Sutures |
Ocular lubricant | National Veterinary Services, Stoke-on-trent, UK | 847288 | Lacrilube (5100G13) |
Optical matching gel | Moor Instruments, UK | PMG | |
Pulse Oximetry Reader | Starr Life Sciences Corp., Oakmont, PA, USA | MouseOx | MouseOx – rat & mouse pulse oximeter & physiological monitor Use with mouse thigh sensor. |
Rehydration gel | Datesand group, Manchester, UK | 70015022 | HydroGel |
Small hair clippers | vetproductsuk.com | HS61 | Contura Cordless trimmer/clippers |
Sterile 0.9 % NaCl Solution | VWR International Ltd, Lutterworth, UK |
LOCA3528286 | SODIUM CHLORIDE 0.9% W/V INTRAVENOUS INFUSION BP 500 ML IN ECOFLAC½ PLUS |
sterile petri dish | VWR International Ltd, Lutterworth, UK | 5168021 | 50mm sterile petri dish. Any brand is suitable. Minimum 50mm diameter is required. |
Topical tissue adhesive | World Precision Instruments, Hertfordshire, UK | 503763 | GLUture topical Tissue Adhesive |
Waterproof superglue | Loctite | Loctite Superglue Precision Max | Available at most hardware shops. |
White paper chip | Datesand group, Manchester, UK | CS1BPB | Pure-O'Cel |