Chapter II.3.7

Large Animal Models in Cardiac and Vascular Biomaterials Research and Assessment

Richard W. Bianco, Karen R. Wasiluk, Jessica M. Voight, Matthew T. Lahti, Andrew L. Rivard and Robert P. Gallegos

Department of Surgery, University of Minnesota, Minneapolis, MN, USA

Introduction

The purpose of using animal models in preclinical evaluation of biomaterials is to provide a preliminary assessment of human safety and efficacy relative to the manufacturer claims and comparable technology that can be used as an appropriate control standard. Although absolute unequivocal results cannot be obtained regarding human safety and efficacy, use of animal models have long been employed and have positively effected changes that have resulted in improved health and increased longevity for humankind. The importance of preclinical investigation has been realized by many great investigators including Charles Darwin, who poignantly describes this link in a letter to a Swedish professor of physiology in 1881 (Darwin, 1959): “I know that physiology cannot possibly progress except by means of experiments on living animals, and I feel the deepest conviction that he who retards the progress of physiology commits a crime against mankind.”

The development of new technologies, including in vitro methods and computer modeling, that has occurred in the last few decades has enhanced the approach to the study of physiology. However, the use of in vivo models still remains a necessity in investigations into certain physiologic phenomena, especially those related to the pathophysiologic mechanisms of disease, as in vitro studies are limited to the assessment of durability of the test material and hemodynamic performance. Traditional small animal models (e.g., rats, mice, guinea pigs, hamsters, and rabbits) have provided the scientific community access to physiological in vivo models to study mechanisms relating to human disease and basic biological processes. For the evaluation of cardiac and vascular biomaterials, however, investigations using larger domestic animals have proven to be more appropriate as these models incorporate anatomical and physiological characteristics more closely resembling those observed in humans. These similarities allow for the development of models where devices are implanted in a site intended for clinical use (i.e., site-specific), and allow for a more accurate prediction of the safety and clinical efficacy of either a bioprosthetic device or a biomaterial.

In this chapter, we will focus on three commonly used animal models for in vivo biomaterials research and testing in cardiac and vascular surgery: pigs; cows; and sheep. The chapter will begin with current recommendations for in vivo preclinical evaluation of cardiac and vascular devices, in both traditional and minimally-invasive settings, will continue with a short discussion on responsible use of animals in the assessment of biomaterials and biomedical devices, and will be followed by specific considerations and examples of existing animal models for each species. Finally, we will discuss conventionally-placed devices and percutaneously-placed devices, as well as potential future directions. Figure II.3.7.1 shows representative cardiac valves that are widely used clinically.

FIGURE II.3.7.1 Cardiac valves used clinically. (A) St. Jude Medical®Regent™ bileaflet mechanical aortic valve. (B) Sorin Mitroflow® bioprosthetic aortic valve. (C) Edwards Sapien bioprosthetic valve for percutaneous implantation. (Used with permission.)

Recommendations for Preclinical Assessment

The purpose of preclinical in vivo assessment is to provide an estimation of device safety (primary) and efficacy (secondary) in humans. This assessment is a mandatory requirement prior to the initiation of the next phase of device evaluation, namely clinical trials. The International Standards Organization (ISO), in conjunction with the American National Standards Institute, Inc. (ANSI) and the Association for the Advancement of Medical Instrumentation (AAMI), has defined regulatory requirements governing preclinical studies which are specific to cardiovascular valve prostheses (ANSI/AAMI/ISO 5840:2005 (AAMI, 2005)), and to cardiovascular implants and tubular vascular prostheses (ANSI/AAMI/ISO 7198:1998/2001/(R)2004 (AAMI, 2001)). These documents govern all aspects of evaluation, from the initial in vitro assessment to the final packaging requirements. These requirements must be reviewed and, if necessary, revised every five years to reflect the advancements in the rapidly evolving fields of biomaterial and bioprosthetic device development. In addition, these requirements also strive to meet the challenge of ensuring that compliant preclinical assessments provide an accurate correlation of in vitro and in vivo performance, as well as a basis for comparison with future clinical results.

The preclinical assessment of biomaterials and bioprosthetic devices is the final investigational step prior to human implantation, thus evaluation must ensure that the devices assessed will perform at least as well in humans as those bioprostheses currently in use. We will limit our discussion to the established standards for in vivo preclinical evaluation of cardiac valvular and tubular vascular prostheses, both those placed conventionally and those placed percutaneously. The standards that apply to the experimental methodology used in the investigation and the specific study criteria that must be met, as well as required results documentation, will be delineated. Following this, a short discussion on the future directions of these standards and the predicted impact these changes may have on the study of these types of devices will be presented.

Current Recommendations

Cardiac Devices

Rationale

Safety assessment is the primary outcome of preclinical studies, with an emphasis on risk analysis in humans. Prior to clinical trials, all new or modified heart valve designs must be evaluated in a standardized, reproducible animal model. The evaluation must use adequate numbers of the experimental devices and concurrent controls implanted in a site-specific fashion, and must involve the use of risk analysis. Neither endpoints nor a defined animal model of preclinical assessment are specified in regulatory documents. Inherent in the preclinical assessment is an evaluation of the performance characteristics of cardiac devices, such as surgical handling characteristics, hemodynamic performance, and the development of valve-related pathology. Although it is not feasible to assess the long-term durability of a cardiac device using an animal model, valuable hemodynamic performance and biological compatibility data can be obtained. Additionally, in vivo testing may expose unanticipated side-effects of the device. ISO 5840 guidelines were developed to guide designers and investigators, and to ensure that new or modified cardiac valvular prostheses are assessed in a uniform manner.

Major changes between ISO 5840:2005 (AAMI, 2005) and its previous iteration, ISO 5840:1996 (AAMI, 1996), include:

1. Incorporating risk-based analysis and placing the responsibility on the manufacturer to continually evaluate both known and theoretical risks of the device;

2. Detailing best practice methods for verification testing appropriate to heart valve substitute evaluation;

3. Requiring a collaborative environment between the device developer and the regulatory body regarding safety and device performance; and

4. Outlining a system to assist a surgeon in selecting a device of appropriate size for placement in a patient.

Methods

Current regulatory requirements defined in ISO 5840 have been revised to specifically address risk analysis in devices representing major design or material changes, recognizing that such changes may result in dramatic (sometimes catastrophic) changes in performance and safety. The changes to guidelines call for expansion of study design to include a minimum of 10–15 experimental devices of clinical quality, with 2–4 concurrent control devices implanted in each clinically-intended surgical site, followed by subsequent comprehensive observation for at least six months. ISO 5840 requirements state that control devices must be clinically-approved devices of a design similar to that of the experimental device, and must be constructed from the same materials, which can be problematic in the case of a newer polymeric material. Minor modifications to devices (i.e., a sewing ring alteration) may involve a less rigorous assessment, as determined by risk analysis, although it is important to accurately distinguish between a major and a minor modification (see Box 1).

Box 1

In an effort to render a mechanical heart valve more resistant to endocarditis, the sewing ring was redesigned to include impregnation with antibiotics. During in vivo preclinical assessment, unexpected thromboses were observed with each valve tested (see Figure II.3.7.2), resulting in the failure of this modified valve to proceed to clinical trials. Risk analysis is a critical component of preclinical assessment, and no assumptions can be made as to the importance of any modification to a device. Although there is biological variability, especially given the relatively low numbers of animals, it is important to separate outcomes that are device-related from those that are the result of the animal model chosen. However, a catastrophic failure, such as this example, must result in a complete review of the in vivo testing results, and reassessment of the suitability of the animal model chosen.

FIGURE II.3.7.2 Example of a thrombosed mechanical valve in which a minor modification to the sewing ring resulted in an unexpected outcome of the preclinical assessment and failure of the valve.

In all preclinical assessments, performance criteria must take into consideration animal–device interactions, individual biological variation, and the small numbers of devices implanted for evaluation. In addition, a complete pathological examination of all animals evaluated must be performed by a veterinary pathologist to distinguish model versus device complications. It is sometimes difficult to completely separate the outcomes due to the device from model-related outcomes. The exception is when the assessment of a device results in catastrophic failure.

Further specifications require each animal in which a cardiac device has been implanted to undergo a postmortem examination. This ensures that data will be obtained from all animals; whether or not they survive for the suggested minimum time of six months. An evaluation of hemodynamic performance of the device during or after the implantation period must be performed. This assessment should include measurement of the pressure gradient across the implanted device at a cardiac index of approximately 3 L/min/m² to mimic human conditions. Additionally, an assessment of resting cardiac output and prosthesis regurgitation should be performed. A pathological evaluation of the study animals should also be performed, and should include an assessment of the major organs for valve-related pathology; an evaluation for any device-related hematological consequences; and an assessment of any structural changes of the heart valve substitute.

ISO 5840 criteria for the evaluation in preclinical in vivo assessment are specific to acute or chronic settings, both of which provide valuable information. Acute settings focus on hemodynamic performance, ease of surgical handling, and acoustic characteristics of the implanted device. Chronic settings include the previously mentioned issues as well as hemolysis, potential thromboembolic complications, calcification, pannus formation/tissue ingrowth, structural deterioration and non-structural dysfunction, erosion caused by cavitation, and assessment of valve and non-valve related pathology (ANSI/AAMI/ISO 5840:2005, Annex G (AAMI, 2005)).

Previous ISO standards (ANSI/AAMI/ISO 5840:1996) relating to preclinical assessment of a device outlined requirements stating that at least six animals should be implanted with the experimental device or test article, and an additional two animals should be implanted with control devices for each assessment. A follow-up of at least six months was required for animals implanted with the test article which needed to include hemodynamic measurements, serial blood samples, and extensive pathological evaluation. No animal model was specified in the standards. These minimum requirements are rarely sufficient for adequate assessment of a bioprosthetic device, but may retain some value in the assessment of devices in which only minor modifications have been made, as long as concurrent controls are included and historical data is considered. The design of a preclinical assessment according to the updated regulatory requirements now must consider risk analysis, which is based on the device as well as the manufacturer claims attributed to it, and whether the device is novel or just an incremental modification of an existing device. Inherent in the risk-based approach outlined in the current ISO standards is encouragement for the manufacturer to strive for continued improvements in device design, as well as to ensure safety and efficacy of a device with less reliance on years of clinical assessment for verification of effectiveness.

Reporting

Documentation of results from preclinical evaluations is integral to this process and should be in the form of an official written report describing findings of the investigation in detail. This report should include the name of the institution(s) and investigator(s) involved in the study. A detailed description of the animal model used in the investigation, and the rationale for its use, should also be included. A pretest health assessment, including study animal age at the time of implantation, should also be included for each animal used in the study. A description of the operative procedure, including the suture technique, the orientation of the heart valve substitute, and any operative complications, must be detailed. Descriptions of any adjuvant procedures the animals underwent during the study period (e.g., phlebotomy, angiography) should also be provided. The results of all blood tests, including a statement of the time elapsed between implantation and procurement of the sample should be provided. Blood tests should include an evaluation of the hematologic profile (with emphasis on hemolysis), and blood chemistries. Additionally, the report should include a subjective assessment of specific surgical handling characteristics of the device and any accessories. This assessment should include a discussion of unusual and/or unique attributes the device might possess. The report must contain a complete list of medications administered to the animals during the study period. Finally, the report should contain the results obtained from the hemodynamic studies to assist in the assessment of hemodynamic performance.

Pathologic documentation should include a gross and microscopic report on each animal in which a heart valve substitute was implanted, including any animal that did not survive the minimum post-implantation period. Documentation should include visual records of the heart valve substitute in situ, and evidence of any thromboembolic events occurring in the major organs. The cause of death of any animal that was not euthanized at the study endpoint must be investigated and reported. Detailed examination of the explant must be performed, with specific attention to any structural changes in the device. If appropriate, further in vitro functional studies of the heart valve substitute should be undertaken, such as hydrodynamic testing.

Pathologic assessment is crucial in the evaluation of the safety profile of the test device, and input from both a veterinary pathologist and a physician pathologist are ideal and necessary for the complete assessment of device safety. It is the role of the veterinary pathologist to decipher whether any pathologic findings are due to device malfunction or the inherent physiology of the animal. It is then the role of the physician pathologist to assess the in situ findings, and correlate them with a potential clinical performance in humans.

Vascular Devices

Rationale

As with cardiac devices, the purpose of in vivo preclinical testing of vascular prostheses is to evaluate the characteristics of a prosthetic that are difficult, if not impossible, to obtain using in vitro testing models. The capacity of the prosthesis to maintain physiologic function when used in the circulatory system, the biologic compatibility of the prosthetic, and the surgical handling characteristics of the prosthesis are all important features that can be evaluated using an in vivo model and cannot be obtained with in vitro testing methods. This testing is not intended to demonstrate the long-term performance of the prosthesis, but rather the short-term (less than 20 weeks) response and patency of the prosthesis being investigated.

Methods

In designing an experimental protocol that will ensure clinical relevance, researchers should consider the intended application of the device being studied, as well as the necessary diameter and length of the prosthesis. Additionally, investigators should reflect on any specific biological characteristics of the chosen animal model, and the impact of those characteristics on their conclusions. Current ISO requirements state that a rationale for the use of a particular species, the site of implantation, the selection of a control prosthesis, the method and interval of patency observations, and the number of animals used in each group being studied be provided. Each type of prosthesis shall have been tested by implantation at the intended, or at an analogous, vascular site in not fewer than six animals for not less than 20 weeks in each animal, unless a justification for a shorter-term study can be provided. A prosthesis must not be tested in the species from which it was derived, unless appropriate justification can be provided. Angiography or Doppler should be used to monitor the duration of patency for each prosthesis, and the results recorded. Each investigation should include a control group in which a clinically-approved vascular prosthesis is implanted and studied in the same fashion as the experimental prosthesis.

Preoperative data should include the sex and weight of the animal, documentation of satisfactory preoperative health status, and any preoperative medications. Operative data must include the name of the implanting surgeon and a detailed description of the surgical procedure, to include the type and technique used in performing the proximal and distal anastomoses. The in situ length and diameter of the prosthesis, in addition to any adverse intraoperative events (e.g., transmural blood leakage), must also be documented. Postoperative medications, patency assessments (method and interval from implantation), and any adverse event or deviation from protocol must be noted. Loss of patency before the intended study duration does not necessarily exclude the animal from the study population used to assess prosthetic function and host tissue response. Reports from all animals implanted with either test or control prostheses, including those excluded from the final analyses, shall be recorded. Termination data should include an assessment of prosthesis patency (documenting the method used), and an assessment of prosthesis explant pathology.

Reporting

At the conclusion of the study, a detailed report should be compiled in which the study protocol is delineated. The report should include: the rationale for selection of animal species; implantation site; control prosthesis; method of patency assessment; and the intervals of observation. Additionally, documentation of sample size and pertinent operative, perioperative, and termination data must be presented. The results should include the patency rates and any adverse event that was encountered during the study interval, and should be compared between the study and control groups. All animals entered into the study must be accounted for in the report. If data are excluded a rationale must be provided. A summary of the gross and microscopic pathology, including photographs of the prosthesis in situ and micrographs, should also be provided. In addition to these objective results, a subjective statement of the investigator’s opinion of the device must be included. Again, for this reason, it is important to include the input of a veterinary pathologist to decipher device malfunction from animal physiology, as well as a physician pathologist to correlate pathologic findings with potential clinical performance. Finally, conclusions drawn from the study and a summary of the data auditing procedures should be documented.

Responsible Use of Animals

Introduction

As discussed in greater detail in Chapter II.3.6 of this book (Animal Surgery and Care of Animals), there are two basic principles governing the use of animals in research, education, and testing: (1) scientific reliance on the use of live animals should be minimized; (2) pain, distress, and other harm to laboratory animals should be reduced to the minimum necessary to obtain valid scientific data. Strict adherence to these basic ethical principles will not only enhance the quality of each in vivo preclinical evaluation, but will also ensure that current and future generations of scientists will be able to employ all valid tools available to predict clinical safety of either new or modified medical devices and other technology designed to improve human health.

Investigator and Institutional Responsibilities

When considering appropriate animal models for biomaterials research, the investigator and sponsoring institution should first assess their responsibilities with respect to federal, state, and local laws and regulations. Additionally, given the growing public and political sensitivity to the use of research animals, all laboratories using animals must recognize and comply with current definitions of humane animal care and use. The importance of this issue is reflected in a resolution adopted by the American Association for the Advancement of Science, February 19, 1990. This section briefly presents legal obligations and responsibilities of both the institutional administration’s veterinary care program and the investigator, and provides some practical guidelines for responsible use of laboratory animals.

The Animal Welfare Act (Public Law 89-544, as amended) provides federal regulations governing laboratory animal care and use (Title 9 of the Code of Federal Regulations (CFR), Chapter 1, subchapter A: Animal Welfare, Parts 1, 2, 3) which are enforced by the Animal and Plant Health Inspection Service (APHIS), US Department of Agriculture. The Animal Welfare Act covers specific species including dogs, cats, non-human primates, guinea pigs, hamsters, rabbits, and “any other warm-blooded animal, which is used or is intended for use in research, teaching, testing, experimentation …” (9 CFR subchapter A, Part 1, Section 1.1). Species specifically exempt from the Animal Welfare Act include birds, rats, and mice bred for use in research, as well as horses and livestock species used in agricultural research. Institutions using species covered by the Animal Welfare Act must be registered by APHIS. Continued registration is dependent on the submission of annual reports to APHIS by the institution, as well as maintaining a satisfactory rating upon inspection of the institution’s animal facility during unannounced site visits by APHIS inspectors. The Animal Welfare Act provides specifications for animal procurement (i.e., from licensed suppliers), husbandry, and veterinary care that are used to determine compliance with the Animal Welfare Act. The annual report supplied to APHIS should contain a list of all species and the numbers of animals used by the institution in the previous year. The animals listed in the annual report must be categorized by the level of discomfort or pain they were thought to experience in the course of the research, education or testing process they were used for.

In 1985 amendments were made to the Animal Welfare Act (7 U.S.C. 2131, et seq.) and ultimately implemented on October 30, 1989, August 15, 1990, and March 18, 1991. These amendments extended the Act to require institutions to perform administrative review and provide a mechanism to control animal research programs (Federal Register, Vol. 54, No. 168, pp. 36112–36163; Vol. 55, No. 36, pp. 28879–28881; Vol. 56, No. 32, pp. 6426–6505). Specifically, the new regulation requires that all animal research protocols be reviewed and approved by an Institutional Animal Care and Use Committee (IACUC) before the initiation of the study. Furthermore, the submitted protocols must state in writing that less harmful alternatives were investigated, with justification as to why those methods are not suitable, and that the proposed research is not unnecessarily duplicative. Additional requirements increase the scope of husbandry requirements for laboratory dogs and non-human primates.

The other relevant federal body is the National Institutes of Health (NIH). The Health Research Extension Act of 1985 (Public Law 99-158) required the director of NIH to establish guidelines for the proper care of laboratory animals, and gave IACUC oversight of that care. Broad policy is described in the Public Health Service Policy on Humane Care and Use of Laboratory Animals (NIH, 2002), which identifies the Guide for the Care and Use of Laboratory Animals (NIH, 1985a) as the reference document for compliance. This policy applies to all activities involving animals either conducted or supported by the US Public Health Service (PHS). An institution cannot receive funding from the NIH or from any other PHS agency for research involving the use of animals, without a statement of assurance on file with the PHS Office for Protection from Research Risks (OPRR) ensuring compliance with these guidelines.

As required by the Animal Welfare Act, PHS policy mandates that each institution provide an annual update on animal research use and IACUC review of animal protocols and facilities to be submitted to the Office of Laboratory Animal Welfare (OLAW) within the NIH. Unlike the Animal Welfare Act, however, the PHS policy covers all vertebrate species, does not include an enforcement arm for routine inspections (but does provide for inspection in cases of alleged misconduct), and penalizes noncompliant institutions by withdrawing funding support.

State and local laws and regulations may also affect animal research programs. This legislation imposes restrictions on the acquisition of cats and dogs from municipal shelters. Within the past decade several states have also enacted registration and inspection statutes similar to the Animal Welfare Act. In several states, court rulings have required IACUC reports and deliberations at state-supported institutions to be conducted in public (e.g., Florida, Massachusetts, North Carolina, and Washington).

Housing and Handling

The physical and psychological well-being of laboratory animals is determined in large part by their environment. Careful attention to housing and handling of any investigational animal is crucial to avoid improper handling or care which may result in exposure of the animal to undue stress. The Institute of Laboratory Animal Resources of the National Academy of Sciences publishes a detailed guide with specific recommendations for the housing of many laboratory animal species (Guide for the Care and Use of Laboratory Animals, Commission on Life Sciences, 2011). Adequate assessment of available facilities and their compliance with these regulations must be considered an essential component of any animal investigation. This assessment should be done not only for regulatory affairs compliance, but also for scientific accuracy, as data obtained from subjects that are not cared for properly may be inaccurate.

Euthanasia

In addition to housing and handling, investigators must become familiar with the methods appropriate to the practice of humane euthanasia techniques. Indications for the application of euthanasia in a study include such inviolable situations such as: (1) study protocol completion; or (2) undue suffering of an individual study subject that may be displaying cachexia, anorexia or a moribund state. In this phase of the investigation, as in all other phases of an animal investigation, great care should be taken to avoid subjecting the animal to undue stress. Humane euthanasia techniques used should result in rapid unconsciousness followed by cardiac and respiratory arrest. The 1993 Report of the American Veterinary Medical Association (AVMA, 1993) provides descriptions of acceptable techniques for euthanasia, and is an invaluable resource for investigators in the process of designing and performing a study; the most recent revision of these guidelines was in 2007, with a minor update in 2009 (AVMA, 2007; Nolen, 2009).

Animal Models and Species Consideration

Choice of an Appropriate Animal Model

In determining which model is most appropriate for an in vivo assessment to best evaluate the likely human response to the device, it is first necessary and most important to outline the clinical goals of the bioprosthetic test device and choose the animal model that has been shown to permit control of the greatest number of variables. The closer the in vivo assessment parallels a clinical situation, the more clinically applicable are the results of the assessment. That being said, the sheep is often the superior model in cardiovascular research, as it is relatively easy to control many variables. In addition, the physiological parameters of sheep often approximate the human condition, especially with respect to thrombogenicity. In contrast, swine tend to be exquisitely sensitive to exogenous anticoagulation therapy in an unpredictable fashion. In our experience, when anticoagulation therapy is utilized in swine, the results generated are a series of anecdotes that relate more to each individual animal’s differing level of thrombogenicity, rather than to a herd response to a device. If anticoagulation therapy is necessary, it is most important to consider the entire herd of test and control subjects, and dose them with equal amounts of anticoagulation therapy so that any effect this may have on the study outcome will be demonstrated in each of the animals. However, an ideal study protocol should incorporate no anticoagulation therapy for any animal, because it naturally crafts a more aggressive assessment of the test device and there is no need to address the variable effect of anticoagulation therapy in each animal. In general, although representative drugs and dosages may be stated in this chapter, the investigator with direct responsibility for animal studies is well advised to consult an experienced veterinarian for selections and dosages of drugs, as practice may change over time and unforeseen complications may arise.

Canine

The history of the use of canines in experimental surgery is extensive. Early experimental surgery required an animal model that was inexpensive, readily available, and easily managed. The solution to this problem came in the form of mongrel dogs. Dogs could be obtained from pounds inexpensively and were often familiar with humans. Cardiopulmonary bypass, coronary artery bypass, and early valve studies were all developed primarily with the canine model. The evaluation of artificial heart valves began in the 1950s, using a canine model and valves created from materials such as polyurethane (Akutsu et al., 1974; Doumanian and Ellis, 1961). The canine model is suitable for testing both synthetic and biosynthetic vascular grafts used for coronary artery bypass. The canine model represents the gold standard for in vivo models of coronary artery bypass, and offers several advantages including: (1) a large historical database for comparison; (2) ease of cardiopulmonary bypass; and (3) the availability of autologous saphenous vein for grafting allows researchers to easily perform positive controls.

Interestingly, few studies involving PCI (percutaneous coronary intervention) for stent placement have been performed using this gold standard model (Schatz et al., 1987; Roubin et al., 1990). Because of the recent shift away from the canine as a model for chronic device evaluation, the current standard uses a swine model. Overall, the canine model has provided a long history of significant contributions to our understanding of cardiovascular physiology, and to the development and testing of cardiovascular biomaterials and devices. In more recent times, there has been a shift away from the canine model for many studies, because of public concerns and rising animal costs. Today, the emotional import of using canines for research has prohibitively increased their cost, and society no longer deems dogs appropriate for chronic studies. As a result, models using other large domestic animals have become widely used for ongoing biomaterials research.

Swine

Introduction

It is believed that the majority of the breeds of swine we now know are descended from the Eurasian wild boar (Sus scrofa). Archaeological evidence from the Middle East indicates domestication of the pig occurred as early as 9000 years ago, with some evidence for domestication even earlier in China. From here, the domestic pig spread across Asia, Europe, Africa, and ultimately North America (Towne and Wentworth, 1950; Mellen, 1952; Clutton-Brock, 1999a). Today, swine serve many purposes in society, ranging from food source and pets to biomedical research subjects. In this latter role, swine have proven useful during the past four decades in studying a variety of human ailments, including cardiovascular (Ramo et al., 1970; Cevallos et al., 1979), gastrointestinal (Kerzner et al., 1977; Leary and Lecce, 1978; Pinches et al., 1993), and hepatic diseases (Mersmann et al., 1972; Soini et al., 1992; Sielaff et al., 1997). Studies in swine have also provided extensive information in the areas of organ transplantation (Calne et al., 1974; Marino and De Luca, 1985; Grant et al., 1988; Al-Dossari et al., 1994; Granger et al., 1994).

Multiple authors have reported advantages of using swine in large animal models for biomedical research including, but not limited to, anatomic and physiologic similarities to humans, low cost, availability, and the ability of swine to produce large litters of hearty newborns (Stanton and Mersmann, 1986; Swindle, 1992; Tumbleson and Schook, 1996; Swindle, 2007). In addition, naturally occurring models of human disease (e.g., atherosclerosis) can be found in various strains of swine, making them ideal candidates for the study of these disease processes. Also helpful to investigators using swine is a large body of agricultural literature in the areas of swine nutrition, reproduction, and behavior that may have applications in biomedical research. For all these reasons, the use of swine has become increasingly popular in the biomedical research laboratory.

Comparative Anatomy and Physiology

Swine are the single readily-available species in which cardiovascular anatomy and physiology most closely resembles those in humans, likely because of a similar phylogenetic development that led to an omnivorous species with a cardiovascular system that has accommodated to a relative lack of exercise (McKenzie, 1996). The heart of an adult swine weighs 250–500 grams and comprises approximately 0.25% of the body weight (Ghoshal and Nanda, 1975), similar to that of humans. The coronary vasculature of a pig heart is nearly identical to that of humans in anatomic distribution, reactivity, and paucity of collateral flow (Hughes, 1986). The right and left coronary arteries are similar in size to each other. The left coronary artery supplies the greater part of the wall of the left ventricle and auricle, including the interventricular septum, by means of its circumflex and paraconal interventricular branches. The right coronary artery supplies the wall of the right ventricle and the auricle via its circumflex and subsinuosal branches (Ghoshal and Nanda, 1975). These similarities make swine ideal for studies of both ischemia/reperfusion and coronary stent technology.

There are certain cardiovascular anatomic differences between swine and humans that must be considered when using swine as investigational subjects. First, the left azygous vein drains directly into the right atrium in swine, compared to drainage into the brachiocephalic vein in humans (Ghoshal and Nanda, 1975). Second, there are only two arch vessels coming off the aorta in swine. The most proximal of these is the brachiocephalic trunk, which gives rise to the right subclavian and a common carotid trunk that bifurcates into the right and left carotid arteries. The second arch vessel is the left subclavian artery (Ghoshal and Nanda, 1975).

Electrophysiological parameters of pigs more closely resemble those of humans than any other non-primate animal. These parameters include similar P- and R-wave amplitudes, P–R intervals of 70–113 msec (longer in older animals), mean QRS duration of 39 msec, a Q–T interval of 148–257 (again longer in older animals), as well as ventriculoatrial (V–A) conduction when the right ventricular endocardium is stimulated faster than normal sinus rate (Hughes, 1986). These similarities make the pig a good model for studying pacemakers. Some investigators have found swine to be highly susceptible to ventricular arrhythmias, often requiring prophylactic pharmacologic protection (i.e., lidocaine, flecainide or bretylium). Alternatively, other investigators have exploited this ventricular instability to study “induced” tachyarrhythmias, and the pharmacologic agents and/or electrical interventions used in their management (Hughes, 1986).

Physiologically, swine have hemodynamic and metabolic values that are similar to those of humans (Tables II.3.7.1 and II.3.7.2). They also share similar pulmonary function parameters, such as respiratory rate and tidal volumes (Willette et al., 1996). Although they have lower levels of hemoglobin, circulating red cell volume, arterial oxygen saturation, and markedly lower venous oxygen saturations, they have similar platelet function and lipoprotein patterns. Other physiologic characteristics that may play a role in experimentation include a higher core body temperature, a contractile spleen that sequesters 20–25% of the total red cell mass, very fragile arteries and veins, a higher plasma pH, and a higher plasma bicarbonate level. Swine also have 50% more functional extracellular space than humans, and a slightly increased volume of total body water (Swindle et al., 1986; Hannon et al., 1990). However, individual swine test subjects tend to be more sensitive to anticoagulation therapy. Thus, if anticoagulation is utilized during the procedure, each animal will likely require a different dosage of anticoagulation, resulting in a series of individual anecdotes that are not easily summarized or generalized.

TABLE II.3.7.1 Hemodynamic Values of Species Used to Evaluate Cardiac and Vascular Bioprostheses

Values represent mean ± standard deviation.

Table created from: McKenzie, J. E. (1996). Swine as a model in cardiovascular research. In M. E. Tumbleson and L. B. Schook (Eds.), Advances in Swine in Biomedical Research. Plenum Press; New York, NY, pp. 7–17; Gross, D. R. (1994). Animal Models in Cardiovascular Research, 2nd edn. Kluwer Academic Publishers: Dordrecht, The Netherlands; Kaneko, J. J., Harvey, J. W. & Bruss, M. L. (1997). Clinical Biochemistry of Domestic Animals, 5th edn. Academic Press: San Diego, CA; Fauci, A. S., Braunwald, E., Isselbacher, K. J., Wilson, J. D. & Martin, J. B. et al. (1998). Harrison’s Principles of Internal Medicine. McGraw-Hill: New York, NY.

TABLE II.3.7.2 Metabolic and Hematologic Values of Species Used to Evaluate Cardiac and Vascular Biomaterials

Values represent mean ± standard deviation.

Table created from: Hannon, J. P., Bossone, C. A. & Wade, C. E. (1990). Normal physiologic values for conscious pigs used in biomedical research. Laboratory Animal Science, 40(3), 293–298; McKenzie, J. E. (1996). Swine as a model in cardiovascular research. In: Advances in Swine in Biomedical Research. Plenum Press: New York, NY, pp. 7–17; Fauci, A. S., Braunwald, E., Isselbacher, K. J., Wilson, J. D. & Martin, J. B. et al. (1998). Harrison’s Principles of Internal Medicine. McGraw-Hill: New York, NY; Kaneko, J. J., Harvey, J. W. & Bruss, M. L. (1997). Clinical Biochemistry of Domestic Animals, 5th edn. Academic Press: San Diego, CA; Gross, D. R. (1994). Animal Models in Cardiovascular Research, 2nd edn. Kluwer Academic Publishers: Dordrecht, The Netherlands.

Perioperative Care

Anesthesia

Swine should be fasted at least 12 hours prior to the induction of anesthesia, longer if one desires the large bowel to be empty prior to surgery. Unless gastric surgery is to be performed, the animals may be given water without restriction. It is common to administer an intramuscular (IM) pre-anesthetic agent followed by an intravenous (IV) anesthetic agent after IV access has been obtained. IM injections can safely be administered in the neck just behind the ear, in the triceps muscle of the forelimb or in the semimembranosus–semitendinosus muscles of the hindlimb. IV access is most commonly obtained in the lateral auricular veins (Riebold and Thurmon, 1985; Swindle and Smith, 1994; Swindle, 1994; Smith et al., 2008). The pre-anesthetic/anesthetic protocol used in this laboratory involves the administration of an IM injection of a combination drug (Telazole) consisting of a dissociative anesthetic and a tranquilizer (tiletamine and zolazepam) at a dose of 2–4 mg/kg for sedation. Following this, IV thiopental sodium (6–25 mg/kg) is administered if necessary. Endotracheal (ET) intubation follows and anesthesia is maintained via inhaled isoflurane (1–2 volume %). Commonly used anesthetics, doses, and routes of delivery are given in Table II.3.7.3.

TABLE II.3.7.3 Anesthetics and Sedatives Acceptable for Use in Species Used to Evaluate Cardiac and Vascular Biomaterials

PO, oral; SQ, subcutaneous; IM, intramuscular; IP, intraperitoneal; IV, intraveneous.

Table adapted from the University of Minnesota Animal Care and Use Manual 1997 reprint.

Analgesia

Postoperative analgesia is a very important component of any animal experiment. Appropriate use of postoperative analgesia involves rapid recognition of discomfort and appropriate therapy. When in pain or distress swine will demonstrate changes in social behavior, gait, posture, and a lack of bed-making. Most swine will become very reluctant to move when in pain and, if forced to move, will vocalize with even greater enthusiasm than they generally display (Gross, 1994). Buprenorphine is currently the analgesic of choice used in swine, effective when administered at 0.05–0.1 mg/kg every 8–12 hours (Swindle and Smith, 1994; Swindle, 1994). Standard narcotics such as morphine and fentanyl are effective pain relievers, but have a very short half-life in swine. Additional analgesic agents are listed in Table II.3.7.4.

TABLE II.3.7.4 Analgesics Commonly Used when Testing Biomaterials in Large Animal Models

Table adapted from the University of Minnesota Animal Care and Use Manual 1997 reprint.

^aQD, once daily; BID, twice daily; TID, three times daily; QID, four times daily; q4h, every 4 hours.

^b PO, oral; SQ, subcutaneous; IM, intramuscular; IV, intravenous.