Originally Published MDDI May 2001

Medical Plastics and Biomaterials

FDA uses a range of biocompatibility test data to evaluate medical devices before clearing or approving them for marketing. The growing use of FDA-recognized consensus standards facilitates this process.

Raju G. Kammula and Janine M. Morris

Historically, evaluating the biocompatibility of medical devices and biomaterials has been a complex task. This complexity arises from the fact that devices are made of a diverse range of materials and have various intended uses, with body contact ranging from transient skin contact to contact with blood to permanent implantation. Biocompatibility is generally demonstrated by testing device materials, and their leachable chemicals, using toxicological principles.

There are several national and international consensus standards that address the toxicological evaluation of medical devices. In recent years, FDA—in particular, the Center for Devices and Radiological Health (CDRH)—has been considering the use of these consensus standards to expedite the biocompatibility review of medical devices. This article discusses the toxicity data needed by FDA to evaluate medical devices before clearing or approving them for market, or approving their investigation in human subjects. It also describes how FDA currently uses recognized consensus standards to facilitate the biocompatibility review of medical devices.

BIOMATERIAL OR MEDICAL DEVICE?

The regulation of biomaterials is often a confusing area for developers of new medical devices. Many still believe that FDA, in addition to regulating the devices themselves, approves the materials used in medical products and maintains a list of approved or acceptable biomaterials. In a very few instances, such biomaterials as injectable collagen, certain dental compounds, or bone cements are used as final products; in such cases, FDA regulates these materials as devices. But in general, the agency neither approves materials nor maintains a list of approved materials.

Although FDA recognizes that many of the currently available biomaterials have vast utility in the fabrication of medical devices, the properties and safety of these materials must be carefully assessed with respect to the specific application in question and its degree of patient contact. An important principle in the safety assessment of medical devices is that a material that was found to be safe for one intended use in a device might not be safe in a device intended for a different use. Accurate characterization is an essential step in selecting a material for a medical device, but ultimately the final assessment must be performed on the finished product, under actual use conditions. This is particularly true for determining device biocompatibility.

BIOCOMPATIBILITY TESTING, DATA REQUIREMENTS, AND EVALUATION OF MEDICAL DEVICES

Biocompatibility testing and evaluation of medical devices is performed to determine the potential toxicity resulting from contact of the device with the body. The device materials should not—either directly or through the release of their material constituents—produce adverse local or systemic effects, be carcinogenic, or produce adverse reproductive and developmental effects. Therefore, evaluation of any new device intended for human use requires data from systematic testing to ensure that the benefits provided by the final product will exceed any potential risks posed by device materials.

One of the first steps in the manufacture of a medical device involves the selection of suitable biocompatible materials. This is an essential step because the types of tests required for evaluation of a device depend on the physical and chemical nature of its materials in addition to the nature of the device's exposure to the body. A specific material may appear suitable on the basis of its physical properties, cost, and availability, but might contain toxic chemical components. Therefore, it is advisable to screen the candidate materials at an early stage to eliminate those that are toxic, and select those that are sufficiently biocompatible or nontoxic for their intended use.

Chemical constituents and potential extractables should be identified and quantitated for overall safety assessment of the device. In general, the suggested biological tests for determining the potentially adverse or toxic effects of medical devices include procedures designed to evaluate cytotoxicity; acute, subchronic, and chronic toxicity; irritation to skin, eyes, and mucosal surfaces; sensitization; hemocompatibility; short-term implantation effects; genotoxicity; carcinogenicity; and effects on reproduction, including developmental effects.

However, depending on the varying characteristics and intended uses of devices, as well as the nature of their body contact, these general tests may not be sufficient to demonstrate the safety of some specialized devices. Additional tests for specific target-organ toxicity—such as neurotoxicity and immunotoxicity—may be necessary for some devices. For example, a neurological device in direct contact with brain parenchyma and cerebrospinal fluid (CSF) may require an animal implant test to evaluate its effects on the brain parenchyma, propensity to induce seizure, or effects on the functional mechanism of choroid plexus and arachnoid villi to secrete and absorb CSF. Thus, the specific clinical application of the new device determines which tests are most appropriate.

In 1995, FDA recognized the standard developed by the International Organization for Standardization, ISO 10993-1, "Biological Evaluation of Medical Devices, Part 1: Evaluation and Testing." At the same time, FDA issued Blue Book Memorandum G95-1, outlining modifications to the ISO test matrix that suggest additional tests for some device categories based on the potential risks these devices present. However, it should be noted that the modified matrix is intended as a framework for the selection of tests and is not a checklist of every required test.

Furthermore, it may not be necessary to conduct many of the suggested tests included in the modified matrix if the device is made of materials that have been well characterized chemically and physically in the published literature and have a long history of safe use. If a manufacturer chooses not to conduct some of the suggested tests, however, it should document in its submissions the use of the particular constituent materials in legally marketed devices with comparable patient exposures.

Recent changes in medical device technologies offer some unique challenges in the assessment of product biocompatibility. For example, the assessment of devices polymerized in situ requires special consideration in the selection and design of appropriate tests to evaluate the potential toxicity of reacting monomers and other chemicals involved in the polymerization process. For these devices, testing of the material or extracts of the material that are already polymerized is not appropriate, since the toxicity of reaction by-products, residual monomers, and other reacting chemicals cannot be evaluated.

Another example requiring special consideration is the testing of devices designed to be resorbable. These devices require special implantation studies with histopathology to document the complete course of device degradation and resorption so as to demonstrate that the implant site is not altered significantly and can return to its state prior to implantation. Furthermore, these devices require studies to assess the distribution, metabolism, and elimination of the degradation products.

As additional new technologies emerge in the development of medical devices, there will be an increasing need to consider different assessment methods to evaluate device biocompatibility. Consequently, FDA recommends and encourages manufacturers to initiate early discussions with the appropriate review divisions in CDRH's Office of Device Evaluation prior to beginning testing of novel biomaterials applications or new intended uses of a device. To permit appropriate evaluation of test results, the submissions should include justification for the selection of tests conducted, discussion and interpretation of test results, and conclusions with adequate scientific rationale.

The overall biocompatibility evaluation should document a general description of the device, its intended use, its degree of body contact, the chemical nature of the materials, a review of available toxicity and bioavailability data for each chemical component, and a justification for the tests conducted to evaluate all potential toxic end points. In addition, previous clinical experience with each material—including their relevance to the safety evaluation of the current device—should be discussed. It is essential that all factors—including the results of conducted tests—be fully documented.

USE OF CONSENSUS STANDARDS BY FDA

Section 204 of FDAMA allows FDA to recognize consensus standards established by national and international standards-development organizations. There are several national and international consensus standards that address the biological evaluation of medical devices to evaluate their toxicity. Such standards are typically developed by consensus, with the participation of industry, healthcare professionals, and academic and government scientists.

CDRH recognizes those standards that help provide a reasonable assurance of safety and effectiveness of medical devices and do not conflict with any legislation or regulation under which the Center operates. Therefore, information submitted regarding conformance with recognized standards would have direct bearing on the evaluation of biological safety made during the review of IDEs, HDEs, PMAs, and PDPs. In the case of 510(k)s, information on conformance with CDRH-recognized consensus standards may help establish substantial equivalence of a new device with a legally marketed device.

If a manufacturer elects to conform to a recognized consensus standard to satisfy premarket review requirements, the manufacturer must submit either a declaration of conformity to the standard or state how conformance to that standard will be achieved. When acceptance criteria are clearly specified by a standard, FDA's need to review actual data is eliminated. However, the manufacturer must maintain all records relating to the conformity to the standard for a period of two years after approval or for the expected life of the device, whichever is longer.

Conformance with recognized consensus standards in itself, however, may not always represent a sufficient basis for regulatory decisions. A specific device may raise additional safety or effectiveness issues not addressed by a recognized consensus standard, or FDA regulations may require additional information beyond what the consensus standard provides. Under such circumstances, conformity with a recognized standard will not satisfy all requirements for investigating or marketing the product in the United States.

For example, as noted above, even though FDA recognized Part 1 of ISO 10993-1, dealing with biological test selection, the standard does not address several tests required to evaluate devices contacting nervous-system tissue and devices that may affect the immune system. Therefore, many of these devices, if tested according to ISO 10993-1 alone, could raise additional safety issues not addressed by the standard. The modifications documented in FDA's Blue Book Memorandum address these additional safety concerns.

The Blue Book Memorandum contains an FDA-modified matrix that designates the type of testing needed for various device categories—recommending specialized testing for neurotoxicity and immunotoxicity of certain devices—and a decision flowchart for the selection of toxicity tests for 510(k)s (Figure 1).

The flowchart was designed to help reviewers determine whether the biocompatibility information provided in a 510(k) submission for a device under review is sufficient. The flowchart helps reviewers in the determination of substantial equivalence by directing them to specific review criteria (steps 2–2.4, 4.0, 5.0, and 5.1) about device materials, their chemical composition, manufacturing processes, and sterilization methods.

The chart also advises reviewers to consult device-specific guidances (steps 6.0–9.0) and, if necessary, to seek assistance from senior toxicologists in the Center, as well as to consider information from other sources such as master files and published literature (steps 10.0 and 11.0). In addition, reviewers are advised to consider any scientific or risk-analysis justifications for not conducting some of the suggested tests (steps 3.0, 7.0 and 11.0). Finally, if the data and analyses provided do not resolve all the questions concerning the biocompatibility of a subject device, the chart advises reviewers that additional toxicology testing is required and that such testing should be conducted according to the FDA-modified test matrix.

FDA has also recognized other parts of 10993, several of which contain protocols for carrying out the tests suggested in Part 1. In addition, FDA has recognized several standards developed by the United States Pharmacopoeia (USP) and the American Society for Testing and Materials (ASTM) that present specific protocols for conducting various biological tests. A list of these recognized standards and the dates they were recognized, together with information detailing the extent of their recognition by FDA, can be accessed through the CDRH Web site at http://www.fda.gov/cdrh/stdsprog.html. Also available on the site is supplemental information on the data requirements or modifications for particular standards that may be necessary to support the regulatory requirements of premarket submissions, but may not fulfill all premarket requirements. Often, recognized standards entail test protocols, which the Center may recognize; however, if the standard does not include any acceptance or rejection criteria for the assessment of test results, it is necessary for the manufacturer to submit the actual data for assessment by the Center's review staff.

GOOD LABORATORY PRACTICES

The good laboratory practices (GLP) regulations (21 CFR 58) prescribe practices for assuring the validity of nonclinical laboratory studies designed to support premarket approval/notification applications for a number of FDA-regulated products, including medical devices. Nonclinical-laboratory studies include in vitro and in vivo experiments or tests in which a test article is studied in test systems (such as microorganisms, cell cultures, or animals) under laboratory conditions as part of determining the safety of a device. In general, the GLP regulations apply to studies that have the purpose of assessing the safety of a test article for human use.

The purpose of all biocompatibility studies is to assess safety, and they must comply with the GLP regulations irrespective of the type of submission for which data are being generated. If the biocompatibility studies do not conform to the GLP regulations, a valid reason for the noncompliance must be provided. If no reason for noncompliance is provided and if the differences between the practices used in conducting the studies and the GLP regulations do not support the validity of the studies, FDA may not accept the data as valid.

CDRH PREMARKET REVIEW STAFF

The Office of Device Evaluation (ODE) within CDRH is responsible for the review and approval of medical devices submitted for marketing, as well as devices used in clinical investigations. Many medical device submissions require multidisciplinary reviews in various scientific and engineering domains. Biocompatibility evaluation within CDRH is conducted by highly qualified and experienced scientists in the field of medical device material toxicology.

CDRH toxicologists from both ODE and the Office of Science and Technology (OST) perform biocompatibility evaluations. The biomedical engineers who have received biocompatibility training sufficient to review many device submissions are permitted to review toxicity data submitted in some 510(k)s. Frequently, 510(k) submissions may not require any toxicology or biocompatibility data to establish substantial equivalence if the intended use and materials of fabrication have not changed.

When toxicology data are required to establish substantial equivalence, biomedical engineers within ODE often review the data under the guidance of senior toxicologists. In addition, ongoing training and professional development programs are conducted by the CDRH Staff College to ensure that review staff have the opportunity to be adequately trained in diverse scientific fields. Many of these courses and guidances, under the direction of experienced toxicologists, are specially designed to enable reviewers to perform thorough biological evaluations of medical device submissions—a curriculum not often available at academic institutions. In addition, ODE and OST review scientists, including the toxicologists, also provide technical and scientific support to the Center's Office of Compliance by reviewing inspection reports and by participating in the bioresearch monitoring inspections when needed.

CONCLUSION

Under the amended Federal Food, Drug, and Cosmetic Act, safety data requirements depend on the classification of a device and the mandated regulatory pathway it must follow in order to be legally marketed. For particular products, it is generally necessary to examine specific biocompatibility requirements on a case-by-case basis, depending upon the nature of device material, the intended use, and the type and degree of patient contact.

Consensus standards recognized by FDA have facilitated the review of biocompatibility data for medical devices. Although these standards are frequently used to support regulatory decisions for market clearance or clinical investigations, they generally do not provide all of the information needed to make such decisions. In particular, standards for biocompatibility evaluation may require additional guidance for their proper implementation and to address the issues of a particular intended use. Therefore, it is always recommended to interact early with CDRH premarket review staff regarding specialized testing that may be required for a particular device in a specific application.

  REGULATION OF MEDICAL DEVICES: A PRIMER With the enactment in 1976 of the Medical Device Amendments to the Federal Food, Drug, and Cosmetic Act, FDA gained the authority to require that medical devices intended for human use demonstrate reasonable assurance of safety and effectiveness—including biocompatibility. The Safe Medical Devices Act of 1990 and the Medical Device Amendments of 1992 further amended the Act. Additional modifications resulted from the FDA Modernization Act of 1997, which, among other important reforms, allowed FDA to recognize consensus standards established by national and international standards-development organizations. DEVICE CLASSIFICATION The 1976 Medical Device Amendments put into place a classification scheme for medical devices. The various types of devices are placed into one of three classes, depending on the degree of risk they present and the level of regulatory control needed to provide reasonable assurance of their safety and effectiveness. Class I provides the lowest level of regulatory control, and is intended for those devices for which there is sufficient information to conclude that safety and effectiveness can be ensured by general controls alone. General controls provide enforcement authority for misbranding, adulteration, registration and listing, banned-device authority, consumer notification and recall, product reporting, and good manufacturing practices (GMPs). Medical devices designated as Class II are devices for which general controls alone are not sufficient to ensure safety and effectiveness, but for which there is sufficient information to establish special controls to provide this assurance. Therefore, in addition to the provisions for general controls, these devices are subject to one or more special controls that may include performance standards, guidelines, patient registries, or postmarket surveillance. Class III devices entail the highest level of regulatory control. For such devices, there is insufficient information to demonstrate that either general or special controls can provide a reasonable assurance of safety and effectiveness. Because of a lack of valid scientific data, these devices present the most risk and are therefore subject to general controls as well as premarket approval (PMA) prior to marketing. The requirements for medical devices to enter into commercial distribution vary according to the regulatory class of the device. Premarket approval and premarket notification (510(k)) are the primary pathways to market for new devices, yet their regulatory requirements differ. All Class III devices are subject to PMA, and must demonstrate a reasonable assurance of safety and effectiveness as shown by valid scientific evidence. Unless exempt, Class I and Class II devices are subject to section 510(k) of the Act, which requires anyone wishing to introduce a device into commerce to notify FDA at least 90 days in advance. A 510(k) notification is used to determine whether a new device is, or is not, substantially equivalent to a preamendments device or to a reclassified postamendments device. Substantial equivalence is based on an assessment of the device's intended use, technical characteristics, and factors of safety and effectiveness. A new device that has been found to be substantially equivalent may be marketed immediately, whereas a new device that is not substantially equivalent may be marketed only through an approved PMA application. Devices reviewed under a 510(k) notification are not "approved" by FDA; rather, these devices are cleared for marketing as being substantially equivalent to legally marketed devices. INVESTIGATIONAL DEVICE EXEMPTIONS A significant-risk device that is not yet cleared for marketing through a 510(k) or approved through the PMA process can be tested in humans only when an investigational device exemption (IDE) application has been approved by the agency. An IDE application must contain sufficient evidence of the device's safety and a reasonable expectation of effectiveness to warrant its testing in humans. Although non-significant-risk devices do not require an approved IDE application, they are still subject to the IDE regulation. PRODUCT DEVELOPMENT PROTOCOLS Section 515(f) of the Federal Food, Drug, and Cosmetic Act provides this alternative to the PMA process for Class III devices. An approved product development protocol (PDP), established early in the device's development, defines the types of data and specific safety and performance levels that must be attained for market clearance. Marketing may commence following FDA verification that the device meets all safety and performance levels. This process allows FDA to effectively regulate Class III medical devices from initial development to marketing. HUMANITARIAN DEVICE EXEMPTIONS The Safe Medical Devices Act of 1990 provides an incentive for the development of humanitarian use devices (HUDs). An HUD is a device that is intended to benefit patients by diagnosing or treating a disease or condition affecting fewer than 4000 individuals in the United States per year—circumstances under which a device manufacturer's research and development costs for a product could exceed its market returns. The humanitarian device exemption (HDE) application is similar in form and content to that for PMA, but is exempt from the effectiveness requirements. Specifically, an HDE application is not required to include the results of scientifically valid clinical investigations demonstrating that the device is effective for its intended use. However, the application must contain sufficient information, including preclinical toxicology data, for FDA to determine that the device does not pose unreasonable risks and that the probable benefits to health outweigh the risks.

BIBLIOGRAPHY

"Biological Evaluation of Medical Devices Part 1: Evaluation and Testing,"ANSI/AAMI/ISO 10993-1: 1994. Arlington, VA: Association for the Advancement of Medical Instrumentation, 1995.

Code of Federal Regulations, 21 CFR 58, "Good Laboratory Practices for Nonclinical Laboratory Studies."

"Use of International Standard ISO-10993, Biological Evaluation of Medical Devices Part 1: Evaluation and Testing," Bluebook Memorandum G95-1. Rockville, MD: FDA, Center for Devices and Radiological Health, Office of Device Evaluation, 1995.

Raju G. Kammula is chief toxicologist in the Office of Device Evaluation at CDRH. He also serves as chairman of the Toxicology Working Group and is responsible for office policies regarding the evaluation of medical device biocompatibility. Janine M. Morris is senior regulatory and technical reviewer for the Office of Device Evaluation. Her expertise is in biomaterials and biocompatibility.

Photo by Fotomorgana/The Stock Market

Copyright ©2001 Medical Device & Diagnostic Industry