An Overview on the Considerations for the Planning of Nonclinical Necropsies for Medical Device Studies

Abstract

The terminal collection and histological processing of medical devices is an expensive, labor-, and material-intensive endeavor, which requires adequate experience, innovation, and preparation for success. It is also an exciting endeavor that continually challenges, intellectually engages, and improves the skills and knowledge of the pathologist. Awareness of the importance of the medical device pathologist’s involvement, communication, and oversight throughout the development, implementation, and execution of a nonclinical assessment of a medical device is in the best interest of the test facility, the histopathology laboratory, the pathologist, the sponsor, and, ultimately, the patients. This article serves to present as a primer of key considerations for the approach and conduct of “nontoxicological” studies, defined as studies involving animal models of deployment or implantation of medical devices as well as surgical animal models.

Notwithstanding the proper planning and execution of a nonclinical medical device study, the successful pathological safety assessment of any medical device ultimately starts at necropsy with the accurate in situ observation and documentation of the device and its associated tissue response and then its careful collection. The terminal collection and histological processing of medical devices is an expensive, labor-, and material-intensive endeavor, which requires adequate experience, innovation, and preparation for success. It is also an exciting endeavor that continually challenges, intellectually engages, and improves the skills and knowledge of the pathologist. Regardless of a study’s regulatory status under Part 58 of the Code of Federal Regulations (CFR, Part 58), Good Laboratory Practices (non-GLP vs. GLP), the involvement of a medical device pathologist from study design to terminal end points, such as necropsy and histologic processing, is preferred and can be critical to the successful nonclinical evaluation of these devices. The creation of a body of literature on the subject of conducting medical device pathology is in its infancy. However, for pathologists unfamiliar with device pathology, review of this literature is highly recommended as a critical first step in developing an understanding of not only the requirements and challenges of device pathology but the differences relative to toxicological pathology and the importance of pathologist’s oversight (Schuh 2008; Nikula and Funk 2016).

For medical device pathologists (DVM or MD) employed by contract research organizations (CROs), hospitals or other facilities familiar with medical device studies, the importance of the pathologist’s role throughout the process is understood and appreciated. For consultant device pathologists, however, who interact with a wide range of test facilities, many of which are frequently unfamiliar with the subtleties and potential complexities of device studies, the lack of consultation or involvement until the study conduct is already ongoing or completed is not uncommon. Therefore, it is incumbent on the medical device pathology community to educate the test facilities, and their sponsors, as to the importance and benefits of early engagement of their services.

Due to the ever-increasing invention, supply, and clinical utilization of medical devices, an exhaustive “how to” of necropsy techniques to accommodate all possibilities is beyond the scope of a single article. However, the implementation of a systematic approach to the necropsies of medical devices’ studies can be used to help ensure success of the nonclinical end points, regardless of the specific device.

Thus, this article will present as a primer of key considerations for the approach and conduct of “nontoxicological” studies, defined here as studies involving animal models for deployment or implantation of medical devices or biomaterials, as well as surgical animal models used in the assessment of the application and administration of certain compounds (e.g., wound or bone healing compounds). Note: For simplicity, only the term “medical device” or “device” will be used routinely throughout the article.

Critical stages related to ensuring the successful planning, preparation, and implementation of medical device necropsies will be presented including: definition of medical devices, toxicologic versus medical device necropsy (i.e., differences), protocol development (i.e., reviewing the study protocol; writing, editing the pathology section), study design (i.e., early discussions with sponsors, study directors, consultants), animal model development (e.g., in concert with surgeons), necropsy procedure development (e.g., carcass work), and necropsy (i.e., approach, gross pathology and sample photos, scoring paradigms, labeling and orientation of specimen, in-life supporting documentation, in vivo imaging, and fixation).

Medical Devices

In the United States, medical devices are classified by the Food and Drug Administration (FDA) into three tiers (i.e., Class I, II, and III) according to their potential harm or risk to the human patient (Gad and Schuh 2018). For the purposes of this article, however, the term “medical device” will generally refer to Class III devices (e.g., implantable devices) that are considered to support or sustain human life, are of substantial importance in preventing impairment of human health, or present a potential unreasonable risk of illness or injury. Although medical devices are usually thought of as anything left within in a body following surgical or interventional implantation, some medical devices may not be present at the time of necropsy. These devices include those that are temporarily deployed to treat/alter tissue in vivo but are then removed immediately posttreatment (e.g., drug-coated balloons; ablative modalities such as radiofrequency, laser, or cryo), those that are removed some time prior to termination of the study (e.g., electrophysiology devices), and those that have been implanted but are resorbed after some period of time (e.g., bioresorbable polymer meshes).

The size, shape, and material composition of medical devices is highly variable and can range from a simple device composed of a single material component (e.g., a thin hydrocellulose film, metal bone screws), a complex device of multiple materials (e.g., a combination polymer and metal atrioventricular valve), or a “combination product” consisting of a pharmacologic compound incorporated into a device/biomaterial (e.g., drug-eluting stents). Additionally, many nondevice compounds utilize surgical animal models for their safety assessments (e.g., critical-sized excisional defects in the skin or bone to evaluate compounds that promote reparative responses). In either case, these necropsies require the same methodical preparations and studied approaches. It cannot be stated enough that a complete understanding of the nature of the device and its intended therapeutic mechanism prior to necropsy is critical to ensure its successful and damage-free collection and subsequent tissue trimming, processing, and analysis.

Toxicologic and Medical Device Necropsy

There are significant differences between the conduct of nonclinical medical device and toxicological study necropsies. The nonclinical safety assessment for pharmaceuticals typically involves large group cohorts of small animal models (predominately rodent) or nonhuman primates. Toxicological study necropsies typically employ methodologies that were developed, standardized, and codified over decades by industry and academic pathologists (Bolon et al. 2013; Halpern et al. 2016; Jacobs et al. 2003; Kittel et al. 2004; Mann et al. 2012, 2012; Morawietz et al. 2004; Ruehl-Fehlert et al. 2003). These methodologies include prosecting, gross assessments, tissue collections, histology, and histopathology, all of which are generally independent of the test article and thus rarely require pilot work to further develop or refine postmortem procedures.

Conversely, the nonclinical assessment of medical devices typically involves very low group cohorts (e.g., a GLP study for a coronary artery stent may involve as few as 9 stents for an FDA submission), large animal models predominately (i.e., swine and ovine), and necropsy techniques and procedures that are heavily dependent on the nature and location of test article and typically not, at least initially, standardized. Although the number of codified guidance documents for the conduct of nonclinical studies for medical devices is fewer than that for toxicological studies, key seminal industry and regulatory guidance documents from the International Organization for Standardization (ISO 10993 Series, particularly Part 6, 2016; U.S. Department of Health and Human Services, Food and Drug Administration, Center for Devices and Radiological Health 2016) and the FDA’s Center for Devices and Radiological Health (2010, 2015), respectively, are crucial for the medical device pathologist in understanding the industry recommendations of animal models for the safety and biocompatibility testing of medical devices or materials thereof. While animal models and necropsy procedures do exist for the nonclinical assessment of some well-established medical devices (e.g., cardiovascular stents; Perkins 2010; Carlyle et al. 2012) or other nondevice entities (e.g., wound healing compounds; Grada, Mervis, and Falanga 2018; Shevchenko and Santin 2014), the relatively rapid nature of novel medical device development frequently precludes the existence of a predicate model or device-specific necropsy procedures. Even if a predicate model does exist, it may require modification for the gross assessment and removal of the next generation of similar devices. Small “pilot” studies are frequently utilized to acquire the necessary in-life skills (surgery, in-life monitoring, pathology) and knowledge to successfully conduct larger non-GLP and GLP studies. The endeavor of medical device necropsies thus requires the constant development of new approaches and techniques tailored to the wide diversity and ever-growing number of novel devices. Regardless of device, a thoughtful and systematic approach to the associated animal model and necropsy and a high level of tissue handling and surgical-like prosecting acumen are important. Each implanted device or treated tissue is highly valuable and very unforgiving if not approached and collected properly. The device pathologist and prosectors must know exactly how to approach and collect an implanted medical device or treated tissue prior to the scheduled (i.e., protocol-derived terminal time point) or unscheduled (i.e., “early death”) macroscopic examination.

Study Design and Protocol Development

As per the CFR, Part 58 GLP Conduct for Nonclinical Laboratory Studies, the primary site of in vivo testing is specifically referred to as the “test facility” (e.g., a CRO or academic research laboratory). Regardless of test facility or regulatory status (i.e., pilot, non-GLP, or GLP), the presence of an approved study protocol is mandatory and crucial in setting the guidance and expectations of all participants in the conduct of the study. The protocol should clearly define the parameters of the study (e.g., study objective; regulatory status; type, number, and characterization of test and control devices; animals’ sex, species, age, and number; and description of end-point analyses) and conform to industry/regulatory requirements (e.g., FDA CFR 58 GLP, Institutional Animal Care and Use Committee, Association for Assessment and Accreditation of Laboratory Animal Care). As part of the development of a draft study protocol, a study design is formulated. The ideal planning of medical device necropsies begins with the pathologist participating in the early (i.e., “lead”) discussions of study design, and ultimately protocol development, between the device experts and the test facility. For well-established devices and animal models, pathologist’s involvement in lead discussions may be unnecessary and protocol reviews cursory. However, as many medical devices presented for nonclinical assessments are of a novel nature for which there is no animal model predicate, the pathologist being engaged in lead discussions is a critical step. This adds value for the sponsors and allows the pathologist to both learn the nature and clinical application of the device and advise accordingly. The guidance of the pathologist from the time of protocol inception allows for the identification of options and potential issues at necropsy as well as ensuring the proper written characterization of the techniques necessary to achieve the study end points.

Medical device creators/developers/manufacturers (most routinely referred to in the industry as study “sponsors”) represent a wide variety of entities including individual entrepreneurs, small “start-up” companies, academic institutions, and the larger established medical device companies. Some individuals involved, such as engineers or polymer chemists, may have nonbiology backgrounds with little to no knowledge of the nonclinical or biological pathway. Others, such as physicians or biomedical engineers, may have expertise and a thorough understanding of the biology of their device but may not be familiar with the nonclinical pathway or details regarding appropriate animal models or device pathology.

Therefore, during the lead discussions, it is optimal if the pathologist works as part of a multidisciplinary team, which may include veterinarians, surgeons or other physicians, and regulatory and/or other consultants. Such a team can best serve to educate, advise, and set the expectations of the sponsor with regard to available animal models, options and, if necessary, the development of new models and processes to achieve the goal.

Through these discussions and self-education, the questions answered at the lead stage should include: the clinical function and application of the device; the anatomic location of the device; the size, shape, and material makeup of the device; the existence of a predicate device and any available information related to previous nonclinical assessment including peer-reviewed scientific publications, nonpublished sources, and the sponsor’s personal experience; the existence of an accepted animal model; the appropriate terminal time points based on the nature of the device and any existing regulatory requirements; the presence of the device at the terminal time point; the results of any previous in-life work with the sponsor’s specific device; and the results of any conversations or guidance the sponsor has received from regulatory agencies. From the information gathered during lead discussions, the pathologist can help to propose a nascent study design for further development and refinement.

Due to the complexity of medical device studies, it is incumbent on the pathologist to review the draft protocol in its entirety (i.e., not just the pathology section) to allow for familiarity and a clear understanding of the overall scope of the device and study (i.e., history, objective, animal model, time points, test and control articles, etc.). Through the review of the entire protocol, device pathologists frequently identify omissions or contradictions in other sections, which allows other parties an opportunity to correct or clarify. When possible, it is optimal that the device pathologist(s) write/edit the relevant sections of the protocol with respect to pathology end points (i.e., necropsy, histopathology, reporting, shipping), which study directors are generally more than willing to accommodate, to clearly defining the details of each respective section as appropriate.

The study protocol should also provide the contact information of the study pathologist(s) and define the expectation of the device pathologist’s role during the conduct of the study. Such information may include requirements for macroscopic (necropsy) and microscopic (histopathology) assessments and imaging, data capture mechanism (i.e., proprietary data capture system vs. spreadsheet) and statistics, and type of reporting (e.g., a non-GLP summary descriptive narrative vs. an audited comprehensive GLP report with data presentation/discussion and images). Device/model-specific macroscopic qualitative scoring schemes (e.g., adhesion severity with abdominal hernia meshes, i.e., 0 = no adhesions; 1 = minimal; adhesions involving < 25% of mesh, etc.) are commonly used at necropsy to qualitatively/semiquantitatively assess tissue changes related to the device treatment. If possible, such schemes should be prospectively documented in the study protocol or an amendment thereof; however, the retrospective description of a macrocsopic scheme in the pathology report is not uncommon.

Animal Model Development

Regardless of device or species, most nonclinical testing of medical devices (for biocompatibility and/or safety) is performed in “healthy” animals (i.e., normal, with no natural or induced disease state). Attempts to assess the functionality (i.e., “efficacy”) of nonclinical treatments, including medical devices, using animal models exhibiting a “disease state” mimicking the human condition are limited (Morgan et al. 2013). However, in some instances, animal models with natural or induced “diseased” states are available including for the cardiovascular system (Kumar et al. 2016; Tsang et al. 2016; Gálvez-Montón et al. 2014), bone (Mills and Simpson 2012), and skin (Branski et al. 2008; Seaton, Hocking, and Gibran 2015).

The nonclinical safety assessment of medical devices utilizes almost exclusively large animal models such as swine and sheep, with the use of smaller animal species (e.g., rodents, rabbits, canines) limited. The selection of a particular animal species or model is usually based on several considerations but most notably on the availability of an animal model that has the comparative anatomy necessary to accommodate the size and application of the device or treatment.

Introduction and discussions regarding some existing medical device animal models include: the biocompatibility testing of novel materials (e.g., polymers) in the subcutaneous, bone, or muscle tissues of the rat or rabbit (ISO 10993-6); cardiovascular and other devices in the swine and/or sheep (Swindle et al. 2012; Sartoretto et al. 2016); skin wound healing in swine (Seaton, Hocking, and Gibran 2015); critical defect bone healing in sheep, rabbits, and rats (Cooper et al. 2010; Wancket 2015); and electrophysiology devices in canine (John et al. 2015).

An obligatory literature search by the medical device pathologist will determine whether a preexisting animal or surgical model exists for a particular medical device or compound, keeping in mind that such articles are generally scant. Regardless, knowledge of the preexistence of an appropriate animal model for a particular device can save a tremendous amount of time and effort for all parties involved. In the absence of a published animal model, it is important that the pathologist, surgeon, or interventionalist understand one another’s considerations and complications and work together to develop a model that ensures both the successful implantation and removal of the device-treated tissues.

Necropsy Procedure Planning and Personnel

Medical device necropsies are methodical and frequently tedious, and the key components to a successful and efficient day of medical device necropsies include adequate preparation, efficient labor, and sufficient time. The lack of these key elements can significantly increase the time per procedure and can quickly result in scheduling issues with in-life procedures (e.g., prenecropsy angiography), personnel fatigue, and critical mistakes including sample damage. The primary goal of device necropsies is to localize and explant the treatment/implant site as rapidly and judiciously as possible; however, removal of the device/treatment site is to occur only after all in situ–dependent procedures (i.e., macroscopic observations, digital images, measurements, perfusions, orientation markers) are completed.

The device pathologist’s involvement at the inception of the protocol development is critical in ensuring that the necropsy procedures are appropriately considered, justified, and documented in the study protocol. Preparation by the pathologist and prosectors to obtain familiarity with the device and the anatomical location and means of approach is critical. The development of novel necropsy prosecting techniques and procedures to accommodate the growing array of medical devices is a constant, but exciting, challenge for medical device pathologists. Although standard necropsy approaches and techniques regarding a small number of medical devices or compounds do exist, they are the exception. With experience, certain assumptions regarding the approach and collection of device-treated tissues can be made between similar and sometimes dissimilar devices, but, even with firsthand experience, incorrect assumptions can lead to costly mistakes at the time of collection.

Well in advance of the day of necropsy, it is incumbent on the pathologist to understand and guide the test facility on the amount of time, labor, supplies, instruments, and documents necessary to efficiently perform all procedures scheduled for that day. One of the biggest concerns regarding device necropsies is the time and labor required to perform the collection. Individuals (e.g., study directors, physicians, in-life schedulers, sponsors) unfamiliar with the necropsy procedures of medical device or surgical animal models tend to underestimate the critical importance of the studied approach to these necropsies as well as the amount of time required to ensure the success and integrity of the collection without undue artifact. A crucial role of the pathologist is thus educating the facility and sponsor about a realistic amount of scheduled time to accommodate each necropsy procedure to ensure these procedures are carefully performed without error while allowing the facility to schedule its resources (i.e., in-life staff) accordingly. A poor understanding of the time required to perform the necropsy, particularly with novel devices/models, can have devastating consequences on the scheduling and execution of the necropsy at the terminal time point, which, more importantly, presents the serious risk of compromising the integrity of the study samples and the subsequently generated data. For example, while the time it takes for an interventionalist to deploy a device in multiple arteries of the head and neck may be relatively short, it can literally take hours per animal for the careful localization, dissection, and removal of these treated vessels, even with experienced personnel.

Besides the time it takes to localize and explant the device-treated tissues, device necropsies by default typically also involve additional multiple procedures that may require significant additional time and effort to execute and thus must be taken into consideration prior to the day of necropsy. Such additional procedures may include the review of in vivo images (angiographic, radiographic, MRI, computed tomography scans) or other information prior to prosecting; perfusion of the localized implant site or entire carcass with fixatives or stains; thorough photographic documentation of the device and/or treatment site via in situ and/or ex vivo photography; collection of additional tissues for safety evaluation including samples of major organ systems or structures which may require complicated and time-consuming collections (e.g., rete mirabile, deep fascial arteries); tissue sampling for additional end points (e.g., snap freezing in optical coherence tomography [OCT] media); radiography or mechanical testing of explanted devices/tissues; comprehensive written documentation of gross examination; collection of anatomic measurements; placement of orientation markers to ensure that the device is trimmed correctly at the histology laboratory; and pretrimming and sampling of the explanted treated tissue. Any of these procedures with which the pathologist, technicians, or facilities are unfamiliar should be explored and defined in the protocol.

Despite careful planning, unanticipated issues frequently arise at the time of necropsy and may require changes in approach and documentation via protocol amendments (if prospective) or protocol deviations (if retrospective). The fact that a necropsy can be present at the end of a scheduled work day, or beyond, can place stress on pathology staff, with increased risk of fatigue-related errors, and thus must be a considered part of planning. Without prior experience of a particular medical device, it is best to assume more time will be required than anticipated, particularly on the first day of necropsy. When in doubt, it is best to be conservative and overestimate the time, as it is better to have too much time than not enough.

Another challenge for medical device pathologists is the availability of trained prosectors. While one trained individual, such as a device pathologist, may be able to perform device necropsies unassisted, multiple prosectors may be required for complicated or comprehensive necropsies (i.e., parallel collection of the test device implant site and major organ systems). Having additional staff also helps maximize the efficiency and accuracy of the prosection and tissue collection, especially if multiple procedures are scheduled for a given day. The lack of standardization of device necropsies precludes the assumption that prosectors trained in toxicological necropsies can automatically perform device necropsies without any additional formal training and experience. The training and qualification of prosectors to ensure a comprehensive understanding of the subtleties of the approach to device necropsies requires months of oversight, observation, and practice, as does development of the necessary tissue handling acumen (e.g., localizing and finely dissecting a 1.0 mm × 3.0 mm vascular stent from a vague location within the swine mesentery). In test facilities with ample experience in device work (e.g., CRO’s), the availability of skilled device prosectors is generally not a concern. However, many device studies are conducted at facilities where experience with medical devices is limited or nonexistent and at which specifically skilled technical staff are limited or unavailable. As such, it is imperative that the device pathologist inquires at the earliest stages of study development as to the availability of adequately trained prosectors to assist in the efficient conduct of the device necropsy procedures.

Guidance by the device pathologist to the test facility regarding the tools, supplies, and documents necessary to execute the necropsies is another critical key for efficiency and reduction of error. The efficient and safe prosection of large carcasses generally requires access to additional tools, which are not routinely available in many test facilities, including necropsy knives (and sharpener) and robust bone cutting tools, especially for accessing the thick calvaria, which may include oscillating bone saw, rongeurs, modified pruning/lopping shears, or a hacksaw. Depending on the nature and location of the device and study end points, additional tools (e.g., chest retractors, fine iris scissors, towel clamps, brain matrices) and equipment (e.g., peristaltic pump) may be warranted.

Another consideration for the facility is the availability of suitable fixatives (e.g., 10% neutral-buffered formalin [NBF]) and containers to safely store and ship the fixed tissues. The vast majority of device necropsies utilize NBF as the primary fixative; however, additional end points, such as electron microscopy (e.g., scanning and transmission) and immunohistochemistry (IHC), require the presence of additional fixatives (e.g., 4% paraformaldehyde, 3% glutaraldehyde, liquid nitrogen, and/or dry ice ± cooled ethanol). Medical device studies frequently require the collection of large tissue specimens (e.g., long bones or entire organs), so the facility must have containers that are adequately sized to accommodate the specimen for sufficient fixation (i.e., 10:1, fixative:tissue sample volume) and may need to be leakproof for shipment to an external histology laboratory. In addition, simple immersion of larger samples into a fixative is generally inadequate for prompt and thorough fixation throughout the sample and thus may require fixation via perfusion prior to immersion fixation, and/or judicious “relief cuts,” or fresh changes of formalin (e.g., every 24 hr) to ensure adequate fixative penetration into the tissue. Alternatively, larger samples may be completely fixed at the test facility and then shipped “wet” in a smaller amount of fixative (e.g., in vacuum-sealed bags). Shipping fresh refrigerated or frozen tissues may also be warranted and require that the test facility has the appropriate supplies and containers necessary to accommodate these samples. Equally important is the generation of appropriate documentation including necropsy report forms, photo logs (to document images taken), general and/or specimen-specific photolabels (for “in-image” documentation of the sample), and sample jar labels. To save additional time on the day of necropsy, it is advisable to have all documents preprinted with the vital study and specimen information (e.g., study and animal number, date, time point, organ) prior to necropsy as well as an assistant scribe, if available, to avoid the frequent pausing of the procedure to deglove (to maintain document cleanliness) and handwrite the information. Regardless, it is imperative that the device pathologist confirms that the study identifiers are accurate on each document.

The careful and accurate assessment and documentation of tissues with device- or treatment-related macroscopic changes before they are explanted is critical in assessing the overall tissue response to the device. The use of consistent nomenclature allows for the faithful correlation with macroscopic changes across animals and studies as well as with the microscopic appearance of the affected tissue. Unlike toxicological studies in which macroscopic observations are generally documented within a proprietary data-capture system (e.g., Provantis, Instem, Staffordshire, UK) using preset lists of descriptors and modifiers, device necropsy data are typically documented on paper. The use of paper forms allows for more flexibility in the documentation of the macroscopic changes, which frequently require descriptive detail, qualitative gross scoring schemes, or treatment-related gross measurements. It is important to stress that per CFR 58 GLP regulations all written documentation of observations on the forms used during necropsy are considered “raw data,” the handling of which is subject to these regulations and the relevant test facilities’ standard operating procedures (SOP) (e.g., “error coding”).

The use and development of device/treatment-related macroscopic scoring schemes can be useful in ensuring the consistent “scoring” of the treatment site across animals for comparison purposes and can be created de novo at the recommendation of the device pathologist and documented in the protocol and pathology report. Macroscopic changes unrelated to the presence of the device or treatment (i.e., “background changes”) should be documented. Such findings are a common and challenging finding in device necropsies, particularly in surgical models (e.g., thoracic and abdominal), in which they may be robust and obscure/efface any device-specific tissue response. Evidence of infectious conditions, whether suspect or overt (e.g., purulence), is a frequent complication of medical device studies and must be documented by the pathologist and, in many instances, confirmed thus requiring access to bacterial culture swabs.

Device necropsies routinely involve digital photography for the in situ and/or ex vivo documentation of the implant/treatment sites and any associated tissue responses. A facility-provided and tracked camera as well as an in-image measuring device (e.g., National Institute of Standards and Technology [NIST] traceable ruler) should be available. All images should contain photolabels with vital study information including the date, study/animal number, tissue, and orientation (e.g., left/right, proximal/distal) and be documented on a facility form (e.g., necropsy form or “photo log”). As these macroscopic images frequently appear in the pathology report or appendices to the sponsor’s FDA submissions, it is important these images clearly show the treatment site/device (i.e., remove/reduce all extraneous tissue, debris, background clutter) and be properly and consistently positioned, focused, illuminated, and white balanced. As each test facility generally has different camera settings, it is important for the pathologist to familiarize themselves with the camera settings and setup prior to the first necropsy.

Tetrazolium salts (e.g., 2,3,5-triphenyltetrazolium chloride [TTC], Nitro blue tetrazolium, Evan’s Blue Stain; Berridge, Herst, and Tan 2005) are commonly used as “viability stains” for the detection of peracute cell death (i.e., prior to any associated changes in cell/tissue morphology) in treated tissues. Most notably, TTC is frequently used at the time of necropsy for the visualization of acutely treated tissues (e.g., renal denervation swine model, Sakakura et al. 2014; the swine myocardial infarction model, Khalil et al. 2006). These compounds have very specific handling and delivery protocols, which necessitate that the pathologist, technicians, and facility are familiar with the proper use of these compounds prior to the day of necropsy. It is important to note that these compounds must be administered prior to fixation (whether in vivo or ex vivo) but do not compromise histologic quality postfixation.

Perfusion techniques in the literature, particularly with respect to fixation, predominately address small animal species, while few large animal perfusion techniques are available (Musigazi et al. 2018). Due to the vast array of possible devices and implantation sites, discussions regarding the details of device-specific perfusion procedures or approaches are beyond the scope of this article; however, perfusion of fluids, particularly fixatives, is frequently required for device-treated tissues. The approach to the cannulation of the vascular system in order to properly perfuse the intended target must be thoroughly considered prior to the day of necropsy. It is important that the in-life facility be instructed to administer an appropriate dose of heparin sulfate prior to euthanasia to prevent the formation of peri/postmortem clots that can interfere with the vascular perfusion. Prior to the perfusion of any solution (e.g., fixative, stains), excess blood must first be flushed from the target vasculature by the perfusion of a physiologic solution (most commonly 0.9% NaCl) in an amount sufficient to result in a watery, lightly sanguineous to clear discharge.

Specifically, perfusion fixation (i.e., predominately with 10% NBF) is generally required for interventional cardiovascular devices/treatments. Depending on the treated organ, perfusions can occur in situ within the carcass (e.g., for stented iliofemoral arteries) or ex vivo after explant (e.g., a heart with stented coronary arteries). Perfusion fixation requires the use of, and possible exposure to, large amounts of formalin and associated fumes, the adverse health effects of which are widely known (Elshaer and Mahmoud 2017). Regardless of the type of perfusion, the use of appropriate precautions and personal protective equipment (e.g. eyewear, Tyvek sleeves or coveralls, respirator, formalin monitoring badge) and equipment (i.e., fume hoods, downdraft tables) are warranted and are the responsibility of the test facility.

There are several keys for limiting exposure to formalin/formaldehyde exposure during in situ perfusion fixations including targeting the site of perfusion, using only the amount of formalin necessary to accomplish the goal, limited prosection prior to perfusing (i.e., only that necessary to expose the cannulating vasculature), and capturing discharge perfusate (e.g., via cannulation of the vena cava). Regardless, it is the responsibility of the facility to be familiar with the Federal, State, and Local regulations regarding the exposure and monitoring of formaldehyde fumes, and protection thereof, as well as the collection and disposal of discharged formalin perfusate and/or a formalin-perfused carcass.

As the majority of interventional treatment sites are limited per animal and associated with the main arterial tree (i.e., aorta and main branches thereof), perfusions are generally “targeted” as opposed to “systemic.” There may be various ways to cannulate the targeted vascular system, and ingenuity may be required to achieve the goal. A simple and effective approach, however, involves the insertion of appropriately sized endotracheal tube (used or new, cuffed or not) into the vascular system, secured in place with umbilical tape. As fixatives are generally supplied in bulk quantities, it is generally necessary to either transfer the quantities necessary to a useable vesicle (e.g., most typically an empty 1.0-L fluid bag) or modify the bulk container (e.g., hose adaptor, with flow control, on the spicket of a 5 gallon “cube”) to allow for the control of flow and/or pressure of the fixative. In targeted perfusions, there is generally little resistance to the inflow of fluid into the system and thus can typically be performed using simple gravity flow with or without any mechanical pressure assist (i.e., pressure infusion cuffs on a 1.0-L bag of 10% NBF). For the majority of interventional devices, it is not necessary to maintain “physiologic pressure” during the perfusion. Perfusion fixations involving targeted interventionally treated vascular segments (i.e., arteries, veins) are predominately performed in situ, the primary purpose of which is to “flash fix” the luminal surface to prevent the artifactual sloughing of the endothelium during the prosection period, which can be protracted depending on the number of treated vessels and their locations. The perfusion of regional portions or the carcass to fix larger tissues (e.g., perfusion of the head/brain) generally requires powered assistance (e.g., peristaltic pump) to overcome the vascular resistance associated with the larger downstream vascular beds. Regardless of fixation methodology, when placing samples into fixative for storage and shipping, it is preferred that the certain device-treated tissues (e.g., vessels, skin, gastrointestinal) be secured/affixed to a rigid backing (e.g., stapled/sutured to a sturdy placard of Mylar, acetate, laminating sheet, or plastic canvas) to prevent aberrant morphologies such as curling or shrinking during fixation. Once affixed to the backing, additional information (e.g., identification, orientation, demarcations of treatment) can be written on the backing with an indelible marker.

Regardless of whether the histopathology of the device tissue is processed within the test facility or by an external test site, the anatomical demarcation (e.g., proximal/distal, lateral/medial) and photo documentation, of the implanted device/tissues prior to explantation of the device/tissues are critical for ensuring the proper trimming and histopathological processing of these samples. This is most critical for technicians (e.g., trimmers) at third-party histopathology labs who are removed from those individuals who directly participated in the necropsy. Regardless of the site of histology conduct, communication between the necropsy pathologist and the histology laboratory technician prior to trimming is important in ensuring the necessary tissue/device plane is presented on the slide for successful histopathological assessment.

Conclusion

Medical device pathology and necropsies are a true blend of art and science, which allows for a rewarding and satisfying career for device pathologists. Medical device pathology is also, however, an unforgiving, laborious, and expensive discipline in which inadequate preparation can result in mistakes at any step (i.e., necropsy, histology) resulting in the loss of an invaluable sample, the failure of the nonclinical study, or, in some instances, collapse of a start-up sponsoring company.

The keys to the success of the medical device pathologist, and the studies on which they work, is self-education (very little literature for guidance), preparation (make no assumptions), and experience (transfer techniques across studies when possible). Awareness of the importance of the medical device pathologist’s involvement, communication, and oversight throughout the development, implementation, and execution of a nonclinical assessment of a medical device is in the best interest of the test facility, the histopathology laboratory, the reading pathologist, the sponsor, and, ultimately, the patients who received them.