Biocompatibility Safety Assessment of Medical Devices: FDA/ISO and Japanese Guidelines

Posted by mddiadmin on January 1, 2000

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

Originally Published January 2000

Amid efforts at harmonization, important differences exist between U.S./FDA/ISO and Ministry of Health and Welfare (Japan) medical device biocompatibility requirements.

Submissions for approval of medical devices by regulatory agencies require that biocompatibility assessment be conducted to assure safety of the device or material. Safety data can be obtained by testing according to certain prescribed or recommended guidelines, including guidance documents developed by the International Organization for Standardization (ISO) and FDA. These guidelines include ISO 10993, "Biological Evaluation of Medical Devices," and the guidance document released by FDA in 1995, blue book memorandum #G95-1, "Use of International Standard ISO 10993, 'Biological Evaluation of Medical Devices'—Part 1: Evaluation and Testing."1,2

In the past, submissions to the European regulatory bodies and to FDA were accompanied by safety data that were different in procedures and references. Prior to 1995, biocompatibility testing was usually conducted by the use of the #G87-1 Tripartite Biocompatibility Guidance (1987). In 1995, FDA released the #G95-1 guidance document, which was an FDA-modified version of ISO 10993, "Biological Evaluation of Medical Devices—Part 1." This document is in concordance with the ISO testing matrix, with certain exceptions. The exceptions are additional assessments that may be required for FDA submissions for which standardized protocols are available based on guidelines published by the Organization for Economic Cooperation and Development (OECD) and others. The FDA document, therefore, made it possible for a medical device manufacturer to test simultaneously for European submissions for obtaining the CE mark, and for submissions to FDA.

Although it is a large medical device market, Japan has been an exception to such harmonization efforts. Japan is an active participant in ISO standards setting; however, its written guidelines differ from those of the ISO in several important ways. Marketing approval submissions for medical devices in Japan fall under the regulatory mandate of the Ministry of Health and Welfare (MHW). MHW has testing guidelines that specify testing protocols and requirements, including the "Japanese Guidelines for Basic Biological Tests for Medical Devices and Materials," notification no. 99 (1995).3 Other testing guidance documents are category-specific or vertical standards. These testing guidelines acquiesce to ISO/FDA testing protocols in some medical device categories. In others, however, the procedures are different and warrant different protocols. Therefore, medical devices intended for Japanese and/or European Union (EU) or U.S. markets are generally tested using different study protocols. The ISO guidelines, for the most part, refer to the compendial methods described in the U.S. Pharmacopoeia (USP) or in the standards published by ASTM. In some instances, the MHW guidelines also refer to ASTM procedures.

The requirements for ISO 10993 have been comprehensively reviewed in a previously published series of articles.4 This article intends to describe the differences in some of the biological testing procedures and protocols required by the ISO, FDA, and MHW guidelines. Tests in areas of biocompatibility not covered by this article would be based upon current guidelines, such as the OECD guidelines for testing of chemicals and others.5,6

The categorization of medical devices based on type and duration of contact is very similar between the ISO and MHW guidelines, and specifies the areas of biocompatibility that should be investigated. Other device-specific Japanese standards define test procedures for certain devices. As can be seen from Tables I and II, the requirements of ISO 10993 and those of MHW are essentially similar, with differences residing in specific test procedures and protocols.

Table I.

Table I. ISO/FDA test chart.

Table II.

Table II. Japanese MHW test chart.


Sample preparation is dealt with at length by ISO 10993-12, "Sample Preparation and Reference Materials" (1996) and by "Japanese Guidelines for Basic Biological Tests for Medical Devices and Materials," notification no. 99 (1995). The procedures are in concordance with the compendial methods described in USP or ASTM standards. Extracts provide suitable samples of potentially hazardous substances that may leach out from the device into the surrounding tissue or may be released from the device during its use. The choice of extraction medium is based upon the use and nature of the device and the predictability of the test method used. Test devices or materials are extracted by incubating the device or representative portions at a surface-area-to-extractant volume ratio. The most common ratios depend on the thickness of the device or material. The ratio is 60 cm2 per 20-ml extraction vehicle if device or material thickness is greater than or equal to 0.5 mm; 120 cm2 per 20-ml extraction vehicle if device or material thickness is less than or equal to 0.5 mm. A weight-to-volume ratio (4 g per 20-ml extraction vehicle) is utilized for extraction at a specific temperature and period of time if the surface area of the device or material cannot be determined. The conditions of extraction should maximize the amount of extractable substances and subject the test device or material to the extreme conditions it may be exposed to, without causing significant degradation. A common extraction procedure is to incubate the test device or material at 37°C for 24 hours at a ratio of either 60 cm2 , 120 cm2 , or 4 g per 20 ml of cell culture medium. The extraction conditions, ratios, and extraction vehicles to be used are generally similar between the ISO and MHW guidelines. Any differences are addressed in the individual sections.


The in vitro tests for cytotoxicity assess the response of cells in culture to direct contact with devices or to their extracts. ISO 10993-5 (1999): "Tests for In Vitro Cytotoxicity" specifies procedures for testing devices by direct or indirect contact, extracts of devices, and filter diffusion. Extracts of test devices and materials are tested by exposure to the cell culture system (e.g., L929 mouse fibroblast cell line). The presence of cytotoxic leachates is indicated by loss of cell viability. To test cytotoxicity by direct contact, the agar diffusion or overlay assay is conducted by placing the test device or a representative portion directly on a mammalian cell layer that is protected from mechanical damage by a layer of agar. Cytotoxic leachates diffuse into the cell layer via the agar, and toxicity is indicated by a loss of viable cells around the test device. The direct contact assay involves the placement of the test material directly on the cell culture medium, without the agar layer. The filter diffusion test involves the exposure of cells grown on filters to test and control materials that are applied to the filters on the side opposite the cells. Following an exposure period of about 2 hours, the cytotoxicity of the materials to the cells on the filter can be assessed by using appropriate stains. In general, in these tests, approximately one-half million to one million cells are present in each culture dish, and toxicity is verified after a period of exposure (typically 24–72 hours) of the cells to the extract or device. Cytotoxicity is evaluated by qualitative and quantitative means. Positive control materials (e.g., organo-tin-impregnated polyvinyl chloride material) and negative control materials (e.g., USP-grade high-density polyethylene RS) are similarly tested alongside to validate the test results.

The cytotoxicity procedure specified by the MHW guidelines for testing extracts employs mammalian cell lines at a density of about 40–200 cells per dish. This test prescribes specific positive controls (polyurethane film containing 0.1% zinc diethyldithiocarbamate, and 0.25% dibutyldithiocarbamate) and the range of responses they should elicit in order to validate the assay. The extraction procedure and the ratio of test material to extractant also differ from that recommended by ISO 10993-5. The ratio is either 5.0 cm2/ml or 1.0 g/10 ml. The cell cultures are exposed to the neat extract and several dilutions of it for a period of 6 to 7 days, the surviving colonies of cells are counted, and the results are expressed as a percent of the negative control. If cytotoxicity is observed at the neat extract concentration, the concentration that would cause 50% cytotoxicity (IC50; IC-inhibitory concentration) is calculated.

ISO 10993-5 MHW 1995
Number of cells per dish:
0.5–1 million cells
40 to 200 cells per dish
Extraction ratio:
60 cm2 per 20 ml if thickness 0.5 mm
120 cm2 per 20 ml if thickness 0.5 mm
or 4 g per 20 ml
5 cm2/ml or 1 g/10 ml
Exposure period:
Typically 24–72 hours
(2 hours for filter diffusion test)
6–7 days
Toxicity determination:
Visual grading and/or quantitative assessments
Quantification of surviving colonies
Positive controls:
Materials providing a reproducible cytotoxic response (e.g., organo-tin-impregnated polyvinyl chloride)
Segmented polyurethane films containing 0.1% zinc diethyldithiocarbamate and 0.25% zinc dibutyldithiocarbamate

Table III. Differences between cytotoxicity test procedures specified by ISO 10993-5 and the MHW guidelines (MHW 1995).

The major differences between the test procedures are illustrated in Table III. MHW also recommends a direct contact test where the cells are allowed to grow for a period of 6 to 7 days in contact with the material or device. This test is recommended for leachables that may be inactivated during extraction and for devices or mateials that come in direct contact with tissue (e.g., eye contact).


Sensitization is the allergic response caused by the activation of complex cellular and humoral immunological mechanisms following exposure to an allergenic substance. Sensitization can occur after either single or multiple exposures. Typically, the allergen is capable of penetrating the skin and biochemically reacts with proteins, thereby becoming proallergenic. Cellular components in the skin act as memory cells, and during reexposures, these cells initiate adverse reactions in response to the challenge.

Sensitization tests, including the Buehler closed-patch test and the Magnusson-Kligman guinea pig maximization test, simulate such exposures to allow the expression of the sensitization potential of test materials. The protocol typically includes an initial induction phase followed by a challenge phase. The maximization test includes intradermal injections of the test material or its extract, whereas the closed-patch test uses topical applications of the test materials or their extracts. In addition, the concurrent administration of Freund's complete adjuvant in the maximization test enhances the potential sensitization capacity of test materials or their extracts.

ISO 10993-10, "Tests for Irritation and Sensitization," recommends either of these tests. If the test material or its extract is amenable to intradermal injection, the maximization test is recommended. The closed-patch test is the assay of choice for noninjectable extracts, and to determine when the extract or material may be topically applied.

ISO 10993-10 MHW 1995
Sample preparation:
Extraction in polar and/or nonpolar solvents.
Two extraction solvents, methanol and acetone, recommended
Extraction ratio:
Extraction ratio is dependent on thickness of device or representative portion.
Specific extraction ratios: 10:1 (volume solvent:weight sample)
Extract used for testing. If extraction is not possible, the adjuvant and patch test can be utilized. Residue obtained from extraction is redissolved and used for testing. (If residue does not dissolve in DMSO, or a sufficient amount of residue is not obtained, the adjuvant and patch test is recommended). Sufficient amount of residue: 0.1 to 0.5% (weight residue:weight test material)

Table IV. Differences between sensitization test procedures required by ISO 10993-10 and the MHW guidelines.

MHW recommends the maximization test, or the adjuvant and patch test for organic and inorganic materials. In the maximization test, organic test residue is obtained by extraction separately in methanol and acetone. The residues are combined and redispersed in dimethylsulfoxide (DMSO) or acetone, and administered by intradermal injection. If the obtained residue is not dispersable in DMSO or acetone, the adjuvant and patch test is utilized, and the residue is administered topically. In both tests, adjuvant is used to enhance the potential sensitization capacity of the test material. If a sufficient amount of residue is not obtained by the process described above, the test material is pulverized at ultralow temperatures (dry ice or liquid nitrogen) and used in the adjuvant and patch test. The MHW guidelines recommend the use of nonpolar solvents for extracts over that of polar solvents. The major differences between the ISO and MHW guidelines are described in Table IV.


The irritation potential of a test device or its leachates when administered to human patients can be extrapolated from the response obtained by injecting the extract of the test device intracutaneously to test animals. This test is a generic irritation test and does not obviate the need for more-specific product end use–oriented tests. Because body tissues differ in vascularization, composition, and response, irritation potential needs to be tested using the same contact conditions as those encountered during actual usage (e.g., eye irritation tests for eye products).

ISO 10993-10 requires that three animals be administered the test extract intracutaneously at five sites of the skin. A control vehicle is similarly injected at five other adjacent sites. Typically, both polar (e.g., physiological saline or water) and nonpolar (e.g., vegetable oil) extractants are used. The appearance of the test sites is compared to that of the control sites over a 72-hour period. Grades assigned to the observed responses are normalized to provide quantitative comparisons with the control responses.

ISO 10993-10 MHW
Number of test animals:
Three rabbits for 1 to 2 extracts
Two rabbits for each extract
Number of test/control injections per extract:
Five test and five control injections
10 test and five control injections
Evaluation of responses:
Quantitative comparison of responses of test and control responses
Qualitative comparison of test and control responses

Table V. Differences in intracutaneous reactivity test procedures required by ISO 10993-10 and the MHW guidelines.

The MHW intracutaneous test procedure, based on ASTM F749-87 (1992), requires the use of two rabbits per extract. The test extracts are administered at 10 sites, while the control vehicle is injected in five sites. The appearance of the test and control sites are observed over a 72-hour period, and a qualitative comparison of the severity of the responses at the test sites to those at the control sites is conducted by a trained technician. The major differences between the ISO and MHW procedures are described in Table V.


The release of the chemical constituents of a medical device, either by leaching or breakdown of the device, into the body has the potential for systemic toxicity. Systemic toxicity tests are generally conducted by administering the extracts (polar and nonpolar in most cases) as a single dose to test animals, and the health status of the animals is verified periodically—typically 48 to 72 hours after dosing. Control animals are administered the extraction vehicle. ISO 10993-11, "Tests for Systemic Toxicity," recommends either ASTM F750, "Standard Practice for Evaluation of Material Extracts by Systemic Injection in Mice (Method A: Intravenous)," or the USP test procedure "<88> Biological Reactivity Tests, In Vivo," to determine systemic toxicity.7,8 Differences between the procedures are minor, mainly in the grading scale used for evaluating the test. The grading scale and the toxicity gradations specified by the ASTM practice are semantically different from those described by the USP procedure. MHW recommends systemic toxicity testing using the ASTM procedure. Systemic toxicity test grading scales for scoring responses of test animals are slightly different and are based on either the USP or ASTM methods (see Table VI).

Response Description  
Normal, no symptoms Mouse exhibits no adverse physical symptoms after injection  
Slight Mouse exhibits slight but noticeable symptoms of hypokinesis, dyspnea, or abdominal irritation after injection.  
Moderate Mouse exhibits definite evidence of abdominal irritation, dyspnea, hypokinesis, ptosis, or diarrhea after injection. (Weight usually drops to between 15 and 17 g.)  
Marked Mouse exhibits prostration, cyanosis, tremors,or severe symptoms of abdominal irritation, diarrhea, ptosis, or dyspnea after injection. (Extreme weight loss; weight usually less than 15 g.)  
Dead, expired Mouse dies after injection.  

The test is considered negative if none of the animals injected with the test article extracts shows a significantly greater biological reaction than the animals treated with the control article.

If two or more mice show either marked signs of toxicity or die, the test article does not meet the requirements of the test.

If any animals treated with a test article shows slight signs of toxicity, and not more than one animal shows marked signs of toxicity or dies, a repeat test using freshly prepared extract should be conducted using groups of 10 mice each. A substantial decrease in body weight for all animals in the group, even without other symptoms of toxicity, requires a retest using groups of 10 mice each. In the repeat test, the requirements are met if none of the animals injected with the test article shows a substantially greater reaction than that observed in the animals treated with the control article. Interpretation

The test is considered negative if none of the animals injected with the test article shows a significantly greater biological reaction than the animals treated with the control article. If two or more mice die, or show signs of toxicity such as convulsions or prostration, or if three or more mice lose more than 2 g of body weight, the test article does not meet the requirements of the test.

If any animal treated with a test article shows only slight signs of biological reaction, and not more than one animal shows gross signs of biological reaction or dies, a repeat test should be conducted using groups of 10 mice. On the repeat test, all 10 animals must not show a significantly greater biological reaction than the animals treated with the control article.

Table VI. Comparison of grading scales used to score responses of test animals to ASTM and ISO/USP procedures.

Pyrogens are substances in devices that cause a febrile reaction. Bacterial endotoxin contamination is most commonly associated with such an adverse effect; however, leachates of materials can cause similar febrile responses (material-mediated pyrogenicity). Material-mediated pyrogenicity assessment is typically included in biocompatibility assessment, while bacterial contamination and endotoxin-mediated pyrogenicity are issues related more to manufacturing processes than to biocompatibility of the device or material.

ISO 10993-11 recommends testing the pyrogenicity potential of extractable substances derived from material leaching. ISO 10993 includes the pyrogen test, based on USP methodology, in the category of systemic toxicity testing. The baseline temperature of the rabbits prior to injection provides a basis for comparison.

ISO 10993-11 MHW 1995
Number of animals:
Three rabbits required; comparison of febrile response in test animals to baseline temperature for evaluation of pyrogenicity potential
Three rabbits (test) required; comparison to baseline temperature is evaluated as index of pyrogenicity potential
Test duration:
Test measurement intervals: every 30 minutes for 3 hours
Test measurement intervals: every hour for 3 hours
Cutoff for positive febrile response: 0.5°C
Cutoff for positive febrile response: 0.6°C

Table VII. Comparison of pyrogen test procedures required by ISO 10993-11 and the MHW guidelines.

MHW holds the pyrogen test as a separate test category. Material-mediated pyrogenicity is tested in a manner similar to ISO 10993 by injecting extracts of the test device into rabbits, and measuring the temperature rise at intervals over a 3-hour period. The major differences between the ISO and MHW recommendations for testing systemic toxicity and pyrogenicity potential are outlined in Table VII.


ISO 10993-3, "Tests for Genotoxicity, Carcinogenicity, and Reproductive Toxicity," recommends that the potential for genetic toxicity be assessed using a series of at least three assays. Two of these assays should use mammalian cells as the test system, and the tests should cover the three levels of genotoxic effects: DNA effects, gene mutations, and chromosomal aberrations. The International Conference on Harmonization (ICH) "Guidelines on Genotoxicity: A Standard Battery for Genotoxicity Testing of Pharmaceuticals, "are currently being applied to medical device assessments also. These guidelines recommend a three-test battery. These tests include reverse mutation assay using Salmonella typhimurium and Escherichia coli strains of bacteria, the in vitro chromosomal aberration assay or the mouse lymphoma tk+/- mutation assay, and the in vivo rodent bone-marrow micronucleus assays.

MHW notification no. 99 recommends a test battery including the reverse mutation assays using Salmonella typhimurium and E. coli strains of bacteria and the in vitro chromosomal aberration assay.

ISO 10993-3 recommends the testing of either the test material extract or dissolved material, whereas the MHW guideline recommends a specific process for obtaining residue from the test materials. Residue is obtained by incubating the test material in methanol and acetone, evaporating the solvents and redispersing the residue in an appropriate vehicle, such as DMSO, and testing. If sufficient residue is unobtainable, MHW allows the test material to be extracted at 37°C for 48 hours in ethanol, acetone, or DMSO for the Ames mutagenicity assay and in cell culture medium for the chromosomal aberration assay, and the extract tested.

ISO 10993-3 MHW 1995
Extraction vehicles:
A physiological medium is used and, where appropriate, a solvent (e.g. dimethylsulfoxide)
Recommends methanol and acetone as extracting vehicles
Extract test material and test the extract or dissolve material in solvent and conduct test. The conditions of extraction should maximize the amount of extractable substances, as well as subject the test device or material to the extreme conditions it may be exposed to, without causing significant degradation. Extraction ratio is dependent on thickness of test material.
Extract at room temperature at a ratio of 10:1 (solvent material) and obtain residue (at least 0.1–0.5% [weight of residue/weight of test material]), redissolve in appropriate solvent and test residue.

If sufficient residue is unobtainable, extract test material (in ethanol, acetone, or DMSO at 10 g of test material per 20 ml for the Arnes mutagenicity assay, and in cell culture medium at 120 cm3 or 4 g/20 ml for the chromosomal aberration assay), at 37°C for 48 hours and test extract. The Ames mutagenicity assay is conducted with a volume of 200 µl per plate.

Table VIII. Differences in genotoxicity testing procedures required by ISO 10993-3 and the MHW guidelines.

Differences in the procedures required by ISO 10993 and the MHW for genotoxicity testing are illustrated in Table VIII.

Printer-friendly version
Your rating: None Average: 5 (3 votes)