This novel model creates robust infrarenal abdominal aortic aneurysms in swine using a combination of balloon angioplasty, elastase/collagenase perfusion, topical elastase application, and oral compound β-aminopropionitrile administration, which interferes with collagen cross-linking.
Large animal models to study abdominal aortic aneurysms are sparse. The purpose of this model is to create reproducible, clinically significant infrarenal abdominal aortic aneurysms (AAA) in swine. To achieve this, we use a combination of balloon angioplasty, elastase and collagenase, and a lysyl oxidase inhibitor, called β-aminopropionitrile (BAPN), to create clinically significant infrarenal aortic aneurysms, analogous to human disease.
Noncastrated male swine are fed BAPN for 7 days prior to surgery to achieve a steady state in the blood. A midline laparotomy is performed and the infrarenal aorta is circumferentially dissected. An initial measurement is recorded prior to aneurysm induction with a combination of balloon angioplasty, elastase (500 units)/collagenase (8000 units) perfusion, and topical elastase application. Swine are fed BAPN daily until terminal procedure on either postoperative day 7, 14, or 28, at which time the aneurysm is measured, and tissue procured. BAPN + surgery pigs are compared to pigs that underwent surgery alone.
Swine treated with BAPN and surgery had a mean aortic dilation of 89.9% ± 47.4% at day 7, 105.4% ± 58.1% at day 14, and 113.5% ± 30.2% at day 28. Pigs treated with surgery alone had significantly smaller aneurysms compared to BAPN + surgery animals at day 28 (p < 0.0003). The BAPN + surgery group had macroscopic and immunohistochemical evidence of end stage aneurysmal disease.
Clinically significant infrarenal AAA can be induced using balloon angioplasty, elastase/collagenase perfusion and topical application, supplemented with oral BAPN. This model creates large, clinically significant AAA with hallmarks of human disease. This has important implications for the elucidation of AAA pathogenesis and testing of novel therapies and devices for the treatment of AAA. Limitations of the model include variation in BAPN ingested by swine, quality of elastase perfusion, and cost of BAPN.
According to the Center for Disease Control (CDC), aortic aneurysms (AA) are a leading cause of death in the United States and represent a significant disease burden1. An aortic aneurysm is defined as a dilation of a discrete portion of the vessel lumen by over 50%2. A subset of AA in the abdomen, referred to as abdominal aortic aneurysms (AAA) are a growing concern. AAA remain clinically silent until impending rupture or dissection, with acute onset, severe abdominal pain generally being the only presenting symptom3,4. Rupture of AAA is almost always fatal with a mortality rate of 90%5. Open or endovascular surgery is the only therapeutic option for patients, and can be a highly morbid procedure. Importantly, AAA are one of the few cardiovascular diseases with no medical therapy for cure.
To date, much of the research on AAA pathogenesis has focused on rodent models, using elastase, which is an enzyme that degrades elastin found within the aortic media, to induce aneurysms.6,7 However, the clinical translatability of small animal models to human aneurysmal disease is restricted, as evaluation of structural changes in the aorta, and altered hemodynamics are limited due to size. Because of anatomical and size similarity, the porcine circulatory system correlates better with human biology than rodents8. Large animal models allow further understanding of cellular mechanisms of the disease process, can be used develop novel treatments at therapeutic doses for large mammals, and test mechanical repair devices, which would not be feasible in small animal models. Additionally, the acute nature of rodent models does not replicate the chronicity and pathologic characteristics of human aneurysmal disease.
The combination of elastase and a compound called β-aminopropionitrile (BAPN) has revolutionized murine AAA models, by creating aneurysms that are larger and contain sequela of chronic aneurysmal disease, including mural thrombus, dissection, and rupture9. BAPN is an inhibitor of lysyl oxidase, which is essential for collagen crosslinking, a crucial component of the aortic wall10,11,12. Lysyl oxidase activity decreases with aging and given the association of age and the chronic nature of complicated AA, BAPN has great potential to experimentally mimic the effects of aging9,13,14. The use of BAPN and its ability to replicate chronic disease in a subacute setting offers a novel advantage over alternative large animal models of AAA. Compared to other established porcine AAA models, this model creates the largest aneurysms with hallmarks of end-stage disease, and the results have been previously published8,11,15.
While conferring certain advantages, significant resources and investment are required to successfully complete this model that may deter some investigators. Among these resources include access to operating rooms, qualified surgeons and anesthesia providers, animal housing, and veterinary staff to assist with post-operative care. Additionally, the cost of BAPN may be prohibitively expensive for some labs.
Few large animal models exist to study the complex pathophysiology of AAA formation and translate to human disease. Large animal models of AAA are critical to help assess the viability of novel technologies and treatments for human disease. Therefore, the purpose of this study was to create a reproducible model of advanced stage infrarenal AAA in swine. The rationale for the use of BAPN and elastase swine model is to better understand the pathophysiology of AAA by mimicking the chronic nature and sequela of human aneurysmal disease in an acute or subacute setting, as well as to test novel therapies and devices for AAA treatment.
Animal protocols were approved by the University of Virginia Institutional Animal Care and Use Committee (No. 3848).
NOTE: This model has been previously published by Cullen et al. and is a modified protocol described by Hynecek et al.8,15.
1. Animals
2. Anesthesia
3. Surgical technique
4. Postoperative care
5. Aortic tissue procurement
All statistical analyses were performed using Fisher exact test or chi squared test as appropriate. Data values are reported as mean aortic dilation (%) ± standard deviation (%). Statistical significance was set P < 0.05. The combination of BAPN and surgery providing elastase treatment (surgery/elastase) creates more robust and reproducible AAA in swine at day 28 compared to those treated with surgery and elastase alone (mean aortic dilation (%) ± standard deviation (%): 113.5% ± 30.2% (n = 8) versus 59.7% ± 29.2% (n = 12); P < .01) as shown in Figure 1. AAA grew progressively larger as time progressed (mean aortic dilation (%) ± standard deviation (%) of 86.9% ± 47.4% (n = 4), 105.4% ± 58.1% (n = 5), and 113.5% ± 30.2% (n = 8) at 7, 14, and 28 day harvest time points, respectively, Figure 1). Evidence of chronic aneurysmal disease is evident in animals treated with BAPN and surgery/elastase, including intraluminal thrombus and atherosclerosis (Figure 2). Histologic evaluation demonstrated significantly increased elastin fragmentation and collagen alteration in BAPN-treated swine AAA than surgery/elastase alone (Figure 3 and Figure 4, respectively).
Figure 1: β-Aminopropionitrile (BAPN) treatment increases swine abdominal aortic aneurysm (AAA) size. (A) BAPN + surgery/elastase swine had significantly higher mean aortic dilation compared with non-BAPN-treated (surgery/elastase alone) swine at 28 days (113.5% ± 30.2% vs. 59.7% ± 29.2%; P < .01). (B) BAPN-treated swine showed mean aortic dilation of 86.9% ± 47.4%, 105.4% ± 58.1%, and 113.5% ± 30.2% at 7, 14, and 28-day harvest time points, respectively. This figure was published by Cullen et al.15 and reproduced here with permission. Please click here to view a larger version of this figure.
Figure 2: Sample photographs of porcine abdominal aortic aneurysms (AAA). (A) Control abdominal aorta (no treatment with BAPN or elastase). (B) Infrarenal AAA formed on post-operative day (POD) 28 after treatment with surgery/elastase, and BAPN (C) Intraluminal thrombus in AAA on POD 28 in surgery/elastase and BAPN treated animals (D) Atherosclerosis in infrarenal AAA on POD 28 in surgery/elastase and BAPN treated animals. Please click here to view a larger version of this figure.
Figure 3: Elastin fragmentation is increased in β-aminopropionitrile (BAPN)-treated swine abdominal aortic aneurysm (AAA). van Gieson staining in infrarenal aorta and suprarenal aorta at 7 days (A), 14 days (B), and 28 days (C). Far right, Elastin (black) fragmentation as measured by independent reviewers of infrarenal aorta versus suprarenalaorta at 7, 14, and 28 days. Scale bar represents 500 µm; 4x lens objective. *P < 0.05. This figure was published by Cullen et al.15 and reproduced here with permission. Please click here to view a larger version of this figure.
Figure 4: Collagen is altered in β-aminopropionitrile (BAPN)-treated swine abdominal aortic aneurysm (AAA). Masson trichrome and van Gieson staining in infrarenal aorta and suprarenal aorta at 7 days (A), 14 days (B), and 28 days (C). Far right, Collagen (blue) content within the wall of infrarenal versus suprarenal aorta as measured by densitometry units at 7, 14, and 28 days. Scale bar represents 250 mm; 4x lens objective. This figure was published by Cullen et al.15 and reproduced with permission. Please click here to view a larger version of this figure.
A novel model of infrarenal AAA in swine was created using a combination of balloon angioplasty, perfusion and topical elastase, and dietary as BAPN. Using this model, aortic dilation of >100% was achieved with gross and histologic characteristics of chronic human aneurysmal disease. This model provides a gateway to further understand the complex pathophysiology of AAA and translate potential therapies to human disease.
Prior models of AAA in swine have been achieved with modest success. Marinov et al. used elastase perfusion alone and saw some histologic changes including elastin disruption, but were not able to attain the phenotype that defines an aneurysm (>50% dilation)16. Given the durability of the porcine aorta, more than one intervention is needed to attain clinically significant aneurysms, which was originally described by Hynecek et al. using a combination of elastase and collagenase perfusion and balloon angioplasty8. They saw mean aortic diameter of 73% as well as histologic changes of aneurysmal disease, including endothelial loss, neutrophil infiltration, and elastin disruption.
However, prior models do not address a fundamental issue with all AAA models: how to replicate a chronic disease process in an acute or subacute setting. Most elastase models of AAA in mice show peak dilation at approximately 2 weeks followed by regression thereafter, whereas human disease evolves chronically over years. The key to this question may rest in the use of BAPN, a lysyl oxidase inhibitor preventing collagen crosslinking. BAPN has an "aging" effect, and combined with elastase treatment, it has been shown to simulate chronic aneurysm growth. In a murine model by Lu et al., mice were observed 100 days post-operatively, and demonstrated evidence of end-stage AAA with thrombus formation and spontaneous rupture9. The novelty and significance of our porcine AAA model is in the use of BAPN, which replicates this chronic disease process in a subacute setting and a more translatable animal species. Pigs fed a diet of BAPN combined with balloon angioplasty, elastase perfusion, and topical elastase application showed more robust aneurysms with evidence of end-stage disease, including mural thrombus, atherosclerosis, and rupture compared to those treated with surgery and elastase alone (Figure 1). This model augments and improves upon the prior model by Hynecek et al. by creating larger aneurysms with sequela of chronic disease8.
Although BAPN is essential to replicate the chronicity of AAA, surgical intervention provides the initial insult to the aorta to induce aneurysm formation. BAPN without surgery or elastase use has been examined, but did not show any significant aortic dilation11. For non-surgically trained investigators, the induction of AAA in swine via laparotomy can be daunting. Each step is fraught with potential complications, from bowel and ureteral injuries to arterial or venous bleeding requiring repair to post-operative wound infections. The investigator must be prepared for contingencies in order to survive the pig to its goal end point. A true team effort is required, including an experienced surgeon well versed in abdominal anatomy, provision of exceptional anesthesia including attention to vital signs and fluid status, and attentive post-operative care. Our team has experienced all of the above complications and acted accordingly whether with repair of enterotomy or caval injuries or antibiotics for infections. However, an unforeseen complication involved the degree to which BAPN impaired incisional wound healing in the pigs. Around 3 weeks post-operatively, some pigs exhibited breakdown of their incisions with occasional fascial dehiscence requiring take-back to the operating theater for revision and debridement. Careful monitoring of incisions post-operatively as well as closure in multiple layers is advised to prevent this complication.
The critical portion of the surgery involves cannulation of aorta via the caudal mesenteric artery, which can be frustrating, given its small size. The use of a micropuncture wires has aided us in this cannulation. This step is essential as the cannulation wire allows access to the aorta for the balloon and perfusion cannula. Balloon angioplasty prior to perfusion is essential in our experience, as the balloon dilation hypothetically creates endothelial disruption allowing elastase perfusion to more readily enter the aortic media. Adequate perfusion is defined as a taut segment of aorta without leakage or escape of fluid around the catheter or from the aortic wall. Achieving adequate perfusion of elastase, while limiting total aortic cross clamp time to no longer than 10 min, is essential for good aneurysm formation while simultaneously avoiding ischemic complications. Limiting total aortic cross clamp time to less than 20 min for the entire procedure and allowing adequate time for reperfusion in between balloon dilation and elastase perfusions avoids the dreaded spinal ischemia complication. If adequate perfusion is not obtained, there is likely a leak from somewhere in the perfused segment of the aorta, usually an inadvertent aortotomy from the dissection or retrograde leak from the cannulation site. It is crucial to repair any defect in the perfused segment to allow adequate perfusion of elastase. A vessel loop may be wrapped just proximal to aortic cannulation site to prevent retrograde flow of elastase. Any lumbar artery should also be temporarily clamped during perfusion to avoid elastase entering the systemic circulation, which may cause a septic response in the swine.
Logical next steps for this model include testing novel therapies for the medical treatment of AAA. As mentioned previously, there are no known medical therapies to attenuate or regress aortic aneurysm growth and current definitive care involves open surgical or endovascular approaches. Prior study has defined the roles of proinflammatory cytokines, Interleukin-1β (IL-1 β) and Interleukin-6 (IL-6) in the pathogenesis of descending thoracic aortic aneurysms and AAA, and inhibition of these receptors may provide potential therapeutic avenues for the treatment of these diseases17,18,19. These studies have only been done in murine models so the next steps should involve large animal models. Additionally, a large animal descending thoracic aortic aneurysm is another avenue for future study. Due to differing embryologic origins, there are inherent differences in the wall composition of the thoracic and abdominal aorta, leading to differing pathophysiology of aneurysms in these two segments20.
There are a few limitations of this model. First, since multiple interventions are employed, it is difficult to determine which intervention contributes most to aneurysm formation. The amount of pressure required to achieve an adequately elastase-perfused aortic segment is difficult to measure, and may vary. This could affect the amount of elastase entering the aortic media and subsequent aneurysm formation from one pig to the next. We are currently exploring a strategy to address this. BAPN was mixed in with the pig's food and swine intake in the perioperative period may vary, altering the amounts of BAPN each pig ingests. Finally, this model requires many resources and investment to be successful. This includes operating rooms, surgeons and anesthesia providers, animal housing, post-operative care, and purchase of BAPN, which can be prohibitively expensive. Each lab should carefully evaluate their resources and funding prior to attempting this model.
Overall, despite certain limitations, swine AAA with sequela of chronic disease can be created reproducibly using a combination of BAPN, balloon angioplasty, elastase perfusion and topical elastase application. This has important implications for translational research applicable to human disease.
The authors have nothing to disclose.
We thank Anthony Herring and Cindy Dodson for their knowledge and technical expertise.
Sources of Funding:
Funding for this study was provided by the National Heart, Lung, and Blood Institute of the National Institute of Health under Award No. T32HL007849 and Grant Nos. R01HL081629-07 (G.R.U.) and R01HL124131-01 (G.R.U.).
Arrow Ergo Pack System | Arrow | CDC-21242-X1A | Just need 7 Fr dilator |
Atlas PTA Balloon dilation catheter | Bard | AT-120184 | 16 mm x 4 cm x 120 cm |
Bovie electrocautery | Bovie Medical | A2350 | |
Collagenase Type 1 (5 gm) | Worthington | LS004196 | |
Crile Needle drviers | MFI medical | 61-2201 | |
DeBakey Atraumatic Forceps | MFI medical | 52-4977 | |
DeBakey Peripheral Vascular Clamp | Medline | MDS1318119 | |
Glidewire | Terumo Interventional Systems | GS3506 | outer Wire diameter 0.035 mm, Length 150 cm |
GraphPad Prism 6 | GraphPad Software Inc. La Jolla, Calif) | statistical software | |
Metzenbaum Scissors | MFI medical | 61-0004 | |
Mayo-Hegar Needle Holder | tiger medical | N407322 | |
Micropuncture Introducer Set | Cook | G47946 | |
Mixter Forceps, Standard Grade, Right angle | Cole-Parmer | UX-10818-16 | |
Monocryl suture | Ethicon | Y496G-BX | 4-0 monocryl |
PDS II suture | Ethicon | D8926 | Number 1 looped |
Porcine Pancreatic Elastase | Sigma-Aldrich | E0258-50 MG | |
Satinsky Vascular Clamps | Medline | MDs5632515 | |
Suction canister | Cardinal Health | 65651212 | |
Schuco Aspirator | MFI medical | S430A | |
Vicryl suture | Ethicon | J789D-SD | 2-0 vicryl |
Yankauer Suction tube | Sklarcorp | 07-1801 |