Implantation of Human-Sized Coronary Stents into Rat Abdominal Aorta Using a Trans-Femoral Access

This protocol describes the implantation of human coronary stents into the abdominal aorta of rats with an apoE^-/- background using a trans-femoral access. Compared with other animal models, murine models carry the advantages of high throughput, reproducibility, ease of handling and housing, and a broad availability of molecular markers.

Percutaneous coronary intervention (PCI), combined with the deployment of a coronary stent, represents the gold standard in interventional treatment of coronary artery disease. In-stent restenosis (ISR) is determined by an excessive proliferation of neointimal tissue within the stent and limits the long-term success of stents. A variety of animal models have been used to elucidate pathophysiological processes underlying in-stent restenosis (ISR), with the porcine coronary and the rabbit iliac artery models being the most frequently used. Murine models provide the advantages of high throughput, ease of handling and housing, reproducibility, and a broad availability of molecular markers. The apolipoprotein E deficient (apoE^-/- ) mouse model has been widely used to study cardiovascular diseases. However, stents must be miniaturized to be implanted into mice, involving important changes of their mechanical and (potentially) biological properties. The use of apoE^-/- rats can overcome these shortcomings as apoE^-/- rats allow for the evaluation of human-sized coronary stents while at the same time providing an atherogenic phenotype. This makes them an excellent and reliable model to investigate ISR after stent implantation. Here, we describe, in detail, the implantation of commercially available human coronary stents into the abdominal aorta of rats with an apoE^-/- background using a trans-femoral access.

Percutaneous coronary intervention (PCI), combined with the deployment of a coronary stent, represents the gold standard in interventional treatment of coronary artery disease¹. The long-term success of stents, however, can be limited by the occurrence of in-stent restenosis (ISR) that is determined by an excessive proliferation of neointimal tissue within the stent^2,3. ISR may require a re-intervention either with coronary artery bypass or re-PCI. A variety of animal models have been suggested for the study of ISR, each of them featuring advantages and shortcomings. The major drawbacks of the most commonly used porcine coronary and rabbit iliac artery models, albeit developing lesions markedly similar to humans after stent implantation^4,5, are large animal and housing costs which brings up logistical difficulties especially in long-term studies, as well as limitations in handling and equipment. Furthermore, availability of antibodies to cellular proteins of swine and rabbits is limited. On the other hand, murine models provide the major advantages of high throughput and reproducibility, as well as ease of handling, housing, and therefore cost-effectiveness. Furthermore, a higher number of antibodies are available. However, while apolipoprotein E-deficient (apoE^-/-) mice have been broadly used for the study of atherosclerosis^6,7,8, they are unsuitable for the study of ISR as stents have to be miniaturized to be implanted into mice, potentially changing the stents’ mechanical properties. Moreover, the aortic wall of mice measures between 50 µm in young mice and 85 µm in old mice⁹, and stents have to be deployed using pressure levels as low as 2 atm, which might lead to malapposition of the stent¹⁰. Rats, however, allow for the implantation of commercially available human coronary stents, and demonstrate a vascular healing course similar to larger animals after aortic stent implantation, first reported by Langeveld et al.¹¹. This technique originally required a trans-abdominal access, which necessitated a physical constriction of the aorta to achieve a temporary interruption of blood flow. To avoid the potentially associated vessel injury and inflammatory reactions, the technique was later refined by the introduction of a trans-iliac access, which additionally resulted in a higher survival rate of the animals¹².

Because wildtype rats do not develop atherosclerotic lesions¹³, apoE^-/- rats have been generated using nuclease techniques such as Transcription Activator-Like Effector Nuclease (TALEN)¹⁴, Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR/Cas9)¹⁵, and Zinc Finger (ZF)¹⁶. ApoE^-/- rats have been commercially available since 2011. Providing an atherogenic background, apoE^-/- rats allow for a more realistic evaluation of human-sized coronary stents, especially with regards to ISR.

Herein, we describe the method via the transfemoral access route and using a commercially available thin-strut cobalt-chromium drug-eluting stent (DES), however, it can also be applied for the study of other stent types, such as bare metal stents (BMS) or biodegradable stents.

The experiments were performed in accordance with the German animal welfare law (TSchG) and Directive 2010/63/EU pertaining to the protection of animals used for scientific purposes. The official approval for this study was granted by the Governmental Animal Care and Use Committee (Protocol No.: AZ 87-51.04.2010.A065; Landesamt für Natur, Umwelt und Verbraucherschutz Nordrhein-Westfalen, Recklinghausen, Germany). The study protocol complied with the Guide for the Care and Use of Laboratory Animals. Postoperative pain treatment is based on the recommendations of the German Society for Laboratory Animal Science (GV-SOLAS) as well as Initiative Veterinary Pain Therapy.

1. Basic techniques and common procedures

1. Use homozygous apoE^-/- Sprague-Dawley rats. Identify the genotype of each animal by using standard methods¹⁷.
2. Keep the animals under identical conditions (21 °C ± 2 °C, 60% ± 5% humidity, and a 12 h light/dark cycle) and ensure free access to water and food.
3. Carry out all procedures under clean but nonsterile conditions.
4. Once the rat is anesthetized, perform all procedures under a surgical microscope at a magnification of 16x.
5. Use cotton swabs for compression hemostasis. Gauze swabs (5 cm x 5 cm) soaked with lactated Ringer solution are helpful to keep the groin moist.
6. Follow waste disposal regulations to dispose used materials.

2. Preparations before surgery

1. Prepare the veterinary drugs before starting the operation. Keep all solutions at room temperature, unless otherwise indicated.
2. Thirty minutes before the procedure, administer 0.03-0.05 mg/kg buprenorphine subcutaneously.
3. Anesthetize the rat with an intraperitoneal injection of 100 mg/kg body weight (BW) (S)-ketamine and 8 mg/kg BW xylazine.
4. Assess the rat’s weight using a weighing scale.
5. Place the rat on a heating pad and fix the upper and lower limbs using medical tape. Position the rat with its left hind limb fully extended and as much in line with its spine as possible so as to create a straight line between femoral artery and aorta. This will facilitate advancing the balloon-mounted stent through the aortic bifurcation.
6. Maintain anesthesia with inhalation of 1.5 vol% isoflurane in 97.5% oxygen at a flow rate of 2 L/min.
   NOTE: Allow the rat to breathe spontaneously, without intubation.
7. Apply eye ointment to prevent eye damage during unconsciousness.
8. Shave the fur from the groin and lower abdomen area of the rat and sterilize the corresponding skin with a povidone-iodine solution.
9. Before starting the surgery, verify adequate depth of anesthesia by pinching the tail tip and the interdigital tissue. 

3. Surgery

1. Make a medial incision of ~0.5‒1 cm in the left groin to open the skin and the underlying fascia.
2. Bluntly dissect and probe in the depths until the pulsating left femoral artery can be identified.
3. Using very fine forceps, prepare the femoral artery by gently removing the surrounding connective tissue. Be careful to harm neither the femoral nerve nor the femoral vein, which is medial to the artery.
4. Prepare about 1 cm of the femoral artery. Carefully put the tip of the forceps under the vessel to gently lift it.
5. Thread pieces of 4-0 silk suture under the distal and proximal parts of the artery and form slings. Clamp the ends of each of the two thread slings between the branches of a surgical clamp. Use the surgical clamps to control the artery. Gently stretch and lift the slings in order to temporarily interrupt blood flow.
   NOTE: Work fast to avoid a prolonged tourniquet which may lead to tissue damage.
6. Using sharp micro scissors, perform an arteriotomy in the middle of the femoral artery.
7. Introduce a guide wire through the arteriotomy. When reaching the proximal thread sling, release the tension of the thread by moving the surgical clamp and advance the guide wire further towards the abdominal aorta.
   NOTE: Cut the guide wire using a wire cutter to facilitate handling.
8. Place the proximal end of the guide wire between the diaphragm and the renal arteries.
   NOTE: Advancing the guide wire too far bears the risk of aortic or cardiac injury. We recommend opening the abdomen to ensure adequate positioning of the guide wire and the stent at least for the first several animals.
9. Introduce a crimped and balloon-mounted coronary stent measuring 2.25 mm x 8 mm (max. 2.5 mm x 8 mm) over the guide wire into the femoral artery and advance it to the abdominal aorta.
10. Place the stent just above the aortic bifurcation but below the renal arteries. Deploy the stent by inflating the balloon catheter to 12 atm for 15 s by using an inflation syringe system.
11. Deflate the balloon catheter and maintain negative pressure according to the manufacturer’s recommendations for the stent in use.
12. Slowly withdraw the deflated catheter while leaving the stent in place.
13. Just before taking out the catheter, create tension on the thread loop above the incision with the surgical clamp to interrupt blood flow again. Then remove the balloon catheter and directly ligate the vessel proximally.
14. Tie the proximal and the distal thread loops to ligate the femoral artery and confirm adequate hemostasis of the arteriotomy. Collateral arteries will ensure further perfusion to the limb.
15. Close the muscle overlying the artery, as well as the skin incision by using 10-0 non-resorbable sutures.

4. Animal care after stent implantation

1. Immediately after the operation, allow the rat to recover for 60 min in a special intensive care unit cage with warmed air (30‒35 °C) and an oxygen supply.
2. Watch the animals carefully until fully recovered. Afterwards, move the rats into a normal cage. Provide ad libitum access to water and food.
3. Administer postoperative analgesia every 6-12 hours with 0.03-0.05 mg/kg buprenorphine (s.c., in 500µl NaCl) for a total of 72 hours under clinical assessment.
4. Have the food mixed with clopidogrel (15 mg/kg) to avoid thrombosis of the implanted stent.
5. To enhance hypercholesterolemic conditions and plaque formation, start western diet feeding at 6‒8 weeks after birth and continue until euthanasia. If desired, a cohort of animals fed normal rat chow can serve as control.

5. Tissue collection and processing

1. Before starting the tissue explantation at the designated time point, euthanize the animal according to IACUC guidelines. Harvest the stented aorta for histological analysis at the end of the observation period.
2. Open the abdomen by a midline incision and remove the stented segment of the aorta as well as adjacent non-stented parts of the aorta, measuring 0.5 cm each.
3. Place the tissue into a solution of 4% buffered formalin for 24 h for fixation.
4. Embed the stented arterial tissue in plastic and perform histological and immunohistochemical staining according to standard protocols^18,19.

6. Histomorphometric analysis

1. Perform histomorphometric analysis of sequential sections of the proximal, middle, and distal part of the stented aorta by means of a microscope linked to a computer with an appropriate image analysis software.
2. Trace the contours of the external elastic lamina (EEL, between adventitia and media), internal elastic lamina (IEL, between media and neointima), and lumen with a graphic drawing tablet. From these values, calculate EEL area, IEL area, and lumen area with the software.
3. Calculate the percent cross-sectional area in-stent restenosis (ISR):
   
4. Calculate the total neointimal area (A_i):
   
5. Measure the neointimal thickness (NIT) over each stent strut as the distance between strut and lumen. Measure the NIT between the stent struts as the distance between IEL and lumen.
   NOTE: Alternatively, calculate NIT as
   
   where P_L and P_IEL are the lumen and internal elastic lamina perimeter, respectively²⁰.
6. Perform additional analyses according to the requirements of the study.

This protocol describes stent implantation in the abdominal aorta of rats using a trans-femoral access route (Figure 1). The first central point of this animal model is that it allows for the deployment of human-sized coronary stents. A commercially available crimped and balloon-mounted coronary stent can be placed into the abdominal aorta of rats. Thus, in addition, the same principle of stent deployment as in humans can be applied. Another advantage of the use of rats is the availability of genetically modified strains, such as apoE^-/- rats, which are commercially available.

We recently employed this method to evaluate whether apolipoprotein E-deficient rats are more prone to develop ISR as compared to wildtype rats²¹. From a total of 42 male rats undergoing stent implantation, 36 rats completed the study protocol after 28 days (survival rate = 85.71%). Two rats each died from vessel closure failure, internal hemorrhage, and stent thrombosis. Stents from three animals could not be analyzed because the tissue was seriously damaged or disrupted due to processing failures. Most likely, this happened during the sawing procedure. We recommend training to perform this technique several times before the start of the study.

In the remaining 33 rats, human-sized coronary stents were successfully deployed with no sign of malapposition or vessel injury (Table 1). Body weight was similar in wildtype apoE^+/+ and apoE^-/- rats (530.1 ± 15.94 g versus 513.6 ± 16.45 g). Homozygous apoE^-/- rats developed markedly elevated neointimal hyperplasia and ISR as compared to wildtype apoE^+/+ rats (Figure 2). Although an apoE^-/- background renders animals more susceptible for atherosclerosis, especially when fed western diet, we did not observe any antecedent atherosclerotic plaques in our rats, most likely because a western diet was not started until surgery and the subsequent observation period of four weeks was too short for atherosclerotic lesion development.

Figure 1: Schema of the stent implantation into the abdominal aorta of rats using a trans-femoral access.
(a) After interruption of the blood flow, a guide wire is introduced through a medial arteriotomy. (b) A crimped and balloon-mounted coronary stent is introduced over the guide wire into the femoral artery. (c) The balloon-mounted stent is advanced to the abdominal aorta, where it is deployed by balloon inflation. The stent should be placed above the bifurcation and below the renal arteries. Please click here to view a larger version of this figure.

Figure 2: Representative photomicrographs of Giemsa-stained abdominal aorta at 28 days after stent implantation in western-diet-fed.
(a) Wildtype apoE^+/+ rats and (b) Homozygous apoE^-/- rats. High power images: NI = neointima, St = stent strut, M = tunica media, L = lumen. Figure has been reproduced with modifications from Cornelissen, A. et al.²¹. Please click here to view a larger version of this figure.

Table 1: Outcome of stent implantation in rat abdominal aorta using a trans-femoral access.

This protocol describes the implantation of human-sized coronary stents into the abdominal aorta of apoE^-/- rats. Several technical points are worth emphasizing. First, a mismatch between the stent size and the size of the aorta should be avoided. Placing too small a stent can lead to stent malapposition, whereas implantation of a stent that is too large for the aorta can cause overstretch, tearing, and injury of the vessel. Therefore, we recommend using stents between 2.0 and 2.5 mm in diameter, and to keep implantation pressure within the recommended range without overstretching the stent. The most suitable implantation pressure is usually given by the stent manufacturer. Excess injury of the femoral vein and subsequently the vena cava should be avoided because the vessel walls are extremely thin and very easy to injure, resulting in bleeding that is hard to stop. The femoral artery is distinguishable from the femoral vein by pulsation, which should be carefully observed. Another pitfall is the possibility of arterial injury and dissection when introducing the guidewire and / or the balloon catheter. Arterial dissection can be minimized by controlling and stretching the femoral artery distally with slings using silk ties while introducing the balloon catheter. It is imperative to immediately stop advancing the device when resistance is encountered. In this case, small movements between thumb and index finger will help change the direction of the device. In our experience, this is most frequently the case just below the inguinal ligament and further up, when the common iliac artery approaches the bifurcation, as it descends deeper into the retroperitoneal space here. There will certainly be a learning curve for the operator before survival rates are stable and with some experience, the average surgical time is about 20 min.

In humans, stents are usually implanted into severely narrowed atherosclerotic arteries. Although apoE deficiency in general renders animals more susceptible for the development of atherosclerotic lesions, we did not observe any plaque formation in our rats, most likely because western diet feeding was not started until stent implantation. If stent implantation in atherosclerotic lesions is desired, western diet should start at 6‒8 weeks after birth and continue until sacrifice. Atherosclerotic lesions in susceptible strains will develop after 7‒14 weeks on the high fat diet²². So far, only limited data on apoE^-/- rats have been published. However, no study reported spontaneous lesion development before the age of 20 weeks²³. Zhao et al. observed typical atherosclerosis in apoE^-/- rats after at least 24 weeks with a continuous increase in plaque burden and lesion severity until sacrifice at 72 weeks¹⁵. Thus, according to the literature, it is improbable that rats would develop spontaneous atherosclerosis at 14–16 weeks of age. Therefore, we recommend using older rats and to start western diet as early as possible if stent implantation in pre-formed atherosclerotic lesions is desired for the study.

Six animals did not survive the surgery. Two animals died from stent thrombosis in spite of administration of clopidogrel. To reduce stent thrombosis, animals can be pre-treated for 48 h with aspirin or receive an intraperitoneal injection of enoxaparin post-operatively. Introducing clopidogrel one day before surgery might also reduce the risk of thrombosis, but any intensification of the anti-thrombotic therapy at the same time increases the risk of hemorrhage. Stent thrombosis is a common complication of PCI^24,25,26 and can have several reasons. Potentially, in our study, stent thrombosis fatalities resulted from insufficient balloon inflation and concurrent stent malapposition. In contrast to stent implantation in humans, stent deployment in the rat abdominal aorta was not controlled by angiography. Therefore, ineffective balloon inflation cannot be detected and corrected during surgery. Similarly, stent deployment might lead to an unintentional occlusion of a branching vessel. Considering that it is not practicable to perform the surgery which requires the use of a surgical microscope under fluoroscopic control, we recommend opening the abdomen to confirm the precise deployment of the stent, at least for the first several procedures. Other potential causes for stent thrombosis might be inflammatory reactions, severe injury, or dissections of the vessel wall. The surgeon must be aware of any clinical signs indicating these complications, and the animals must be inspected each day throughout the observation period.

The rat abdominal aorta measures between 1.8 mm and 3.0 mm in diameter, depending on the animal’s weight^27,28. Advancing a bulky stent through the even smaller femoral and iliac arteries may cause intimal tear and damage to the vessel wall. Therefore, this technique is limited to the implantation of smaller stents (between 2.0 and 2.5 mm in diameter) to avoid overstretching or injury of the vessel wall of the aorta.

Another limitation is the necessity of ligating the femoral artery in order to achieve hemostasis after the procedure, potentially bearing the risk of lower limb ischemia. However, previous studies showed that collateral arteries as well as adaptions of the microvasculature distal to the occlusion are able to maintain lower limb perfusion after ligation of the femoral artery in rats²⁹, and none of our rats exhibited clinical signs of lower limb ischemia during the observation period. Still, the investigators should be aware of this potential risk, as limb ischemia not only represents a potential cause of post-operative death, but can also potentially induce a systemic inflammatory reaction, potentially biasing the results.

While rats in general are a cost-effective animal model, the use of genetically modified apoE^-/- rats increases the cost. Another limitation is that it takes a comparatively long time until atherosclerotic plaques have developed in rats. Furthermore, there are some important hemodynamic differences between the aorta and the coronary arteries that deserve closer attention. Shear stress is higher in the aorta as compared to the coronaries, and bifurcations causing turbulent blood flow are absent. This diminishes the development of intimal hyperplasia and the extent of restenosis.

Restenosis is one of the major factors limiting the long-term success of coronary stents. A variety of animal models have been used to study the pathophysiology of restenosis, each featuring their own advantages and shortcomings. In comparison with other animal models, rats hold the advantage of a high throughput, an ease of handling and housing, reproducibility, as well as cost-effectiveness, while at the same time allowing for the implantation of human-sized coronary stents. The first protocol of abdominal aorta stenting in rats was reported by Langeveld et al.¹¹. This model, however, requires a trans-abdominal access to introduce the stent, which is associated with a physical constriction of the aorta to achieve a temporary interruption of blood flow. The resulting manipulation and vessel injury might potentially cause inflammatory reactions, which might not only lead to complications, but also to pronounced ISR¹². Later, Oyamada et al. modified the protocol by introducing the stent through the common iliac artery¹². They compared the survival rate between the two different approaches (trans-aorta versus trans-iliac artery) and found a significantly higher mortality rate in animals with trans-abdominally deployed stents (57% versus 11%, p < 0.05). Rats most commonly died from thrombosis at the incision / suture site, which is catastrophic when occurring in the abdominal aorta¹². Further reducing trauma and mimicking the implantation technique in humans more closely, we used a trans-femoral access to introduce the stent and reported a mortality rate of 14%. Two rats each died from vessel closure failure, internal hemorrhage, and stent thrombosis. More recent studies, however, reported mortality rates as low as 6% after stent implantation in the rat abdominal aorta even with the trans-aortic access^30,31. Still, the combined morbidity and mortality rate was 13.4%, in a study by Nevzati et al. after implantation of magnesium stents into the rat aorta³⁰. While neither vessel closure failure nor internal hemorrhage were reported in their series, stent thrombosis was evident in 10.5% of rats³⁰. On the other hand, Aquarius et al. did not report any stent thrombosis after treating sidewall aneurysms with flow diverters, however, this study used thinner stent strut devices, and dual antiplatelet therapy was administered to the rats³¹. We tried to strike a balance between the risk of stent thrombosis and bleeding risk and administered clopidogrel and heparin in our study. While this might have reduced the risk of stent thrombosis, which occurred in 4.76% of rats, it also might have been the reason for the comparatively higher risk of bleeding (9.52% of rats), either because of internal hemorrhage or vessel closure failure.

Here, we demonstrated the implantation of a drug-eluting stent into the rat abdominal aorta, but likewise this method can be used for the evaluation of other, similarly sized stent devices, for example bare metal stents or bioresorbable vascular scaffolds.

In summary, abdominal aorta stenting of apolipoprotein E-deficient rats is a reliable and reproducible model to investigate ISR after stent implantation. The model can be extended to the use of older rats, which are more likely to develop atherosclerotic lesions spontaneously, and by testing other devices used for human coronary intervention.

The authors have nothing to disclose.

We would like to thank Mrs. Angela Freund for her invaluable technical assistance with embedding and slides production. We would also like to thank Mr. Tadeusz Stopinski at the Institute for Laboratory Animal Science & Experimental Surgery for his insightful help with the veterinary work.