In vitro tests for biocompatibility, done outside a living organism, require placement of a material or a component of a material in contact with a cell, enzyme, or some other isolated biological system. The contact can be either direct, when the material contacts the cell system without barriers, or indirect, when there is a barrier of some sort between the material and the cell system. Direct tests can be further subdivided into those in which the material is physically present with the cells and those in which some extract from the material contacts the cell system. In vitro tests can be roughly subdivided into those that measure cytotoxicity or cell growth, those that measure some metabolic or other cell function, and those that measure the effect on the genetic material in a cell (mutagenesis assays). Often there is overlap in what a test measures. In vitro tests have a number of significant advantages over other types of biocompatibility tests (Table 6-1). They are relatively quick, generally cost less than animal or usage tests, can be standardized, are well suited to large-scale screening, and can be tightly controlled to address specific scientific questions. The overriding disadvantage of in vitro tests is their questionable relevance to the final in vivo use of the material (see later section on correlation between tests on page 114). Another significant disadvantage includes the lack of inflammatory and other tissue-protective mechanisms in the in vitro environment. It should be emphasized that in vitro tests alone cannot entirely predict the overall biocompatibility of a material.

Standardization of in vitro tests is a primary concern. Two types of cells can be used for in vitro assays. Primary cells are those cells taken directly from an animal and cultured. These cells will grow for only a limited time in culture but usually retain many of the characteristics of cells in vivo. Continuously grown cells or cell lines are cells that have been transformed previously to allow them to grow more or less indefinitely in culture. Because of this transformation, these cells do not retain all in vivo characteristics, but they do consistently exhibit those features that they do retain. Primary cell cultures seem to be more relevant than continuous cell lines for measuring cytotoxicity of materials. However, primary cells, being from a single individual, have limited genetic variability, may harbor viral or bacterial agents that alter their behavior, and often rapidly lose their in vivo functionality once placed in culture. Furthermore, the genetic and metabolic stability of continuous cell lines contributes significantly toward standardizing assay methods. In the end, both primary and continuous cells both play important roles in in vitro testing; and both should be used to assess materials.

Cytotoxicity Tests

Cytotoxicity tests assess cell death caused by a material by measuring cell number or growth before and after exposure to that material. Control materials should be well defined and commercially available to facilitate comparisons among other testing laboratories. Membrane permeability tests are used to measure cytotoxicity by the ease with which a dye can pass through a cell membrane, because membrane permeability is equivalent to or very nearly equivalent to cell death (Figures 6-2 and 6-3).

Tests That Use Barriers (Indirect Tests)

Because direct contact often does not exist between cells and materials during in vivo use, several in vitro barrier tests have been developed to mimic in vivo conditions. These tests include an agar overlay method, which uses agar to form a barrier between the cells and the material, and the Millipore filter assay, in which a monolayer of cells is grown on a filter that is turned over so that test materials are placed on the filter and leachable diffusion products are allowed to interact with the cells. The agar diffusion and Millipore filter tests can provide, at best, a qualitative cytotoxic ranking among materials (Figure 6-4).

A number of studies have shown that dentin forms a barrier through which toxic materials must diffuse to reach pulp tissue, with the thickness of the dentin directly correlating with the protection offered to the pulp. These assays, which incorporate dentin disks between the test sample and the cell assay system, have the added advantage of directional diffusion between the restorative material and the culture medium (Figure 6-5).

Animal tests for biocompatibility, usually involving mammals such as mice, rats, hamsters, or guinea pigs, are distinct from usage tests (which are also often done in animals) in that the material is not placed in the animal with regard to its final use. The use of an animal allows for the complex interactions between the material and a functioning, complete biological system to occur. This is extremely difficult to mimic in a cell-culture system. Thus the biological responses in animal tests are more comprehensive and may be more relevant than in vitro tests, and these features are the major advantages of these tests (see Table 6-1). The main disadvantages of animal tests are that they can be difficult to interpret and control, are expensive, time consuming, and often involve significant ethical concerns and oversight. Furthermore, the relevance of the test to the in vivo use of a material is often unclear, especially in estimating the appropriateness of an animal species to represent a human. A variety of animal tests have been used to assess biocompatibility.

The mucous membrane irritation test determines whether a material causes inflammation to mucous membranes or abraded skin. In a skin sensitization test, materials are injected intradermally to test for development of skin hypersensitivity reactions, followed by secondary treatment with adhesive patches containing the test substance. Implantation tests are used to evaluate materials that will contact subcutaneous tissue or bone. The location of the implantation site is determined by the use of the material and may include connective tissue, bone, or muscle. Although amalgam and alloys are tested because the margins of the restorative materials contact the gingiva, most subcutaneous tests are used for materials that will directly contact soft tissue during implantation, as well as endodontic and periodontal treatment materials.

Usage tests may be done in animals or in human study participants. They are distinct from other animal tests because they require that the material be placed in a situation identical to its intended clinical use. The usefulness for predicting biocompatibility is directly proportional to the fidelity with which the test mimics the clinical use of the material in every regard, including time, location, environment, and placement technique. For this reason, usage tests in animals usually employ larger animals that have similar oral environments to humans, such as dogs, mini-swine or monkeys. When humans are used, the usage test is termed a clinical trial. The overwhelming advantage for usage tests is their relevance (see Table 6-1). These tests are the gold standard, in that they give the ultimate answer to whether or not a material will be biocompatible and clinically useful. One might ask, then, why bother with in vitro or animal tests at all. The answer is in the significant disadvantages of the usage test. These tests are extremely expensive, last for long periods, involve many ethical and often legal concerns, are exceptionally difficult to control and interpret accurately, and may harm the test participants. Additionally, statistical analysis of these tests is often a daunting process. In dentistry, dental pulp, periodontium, and gingival or mucosal tissues are the main targets of usage tests.

The best example of a barrier that occurs in use but not during in vitro testing is the dentin barrier. When restorative materials are placed in teeth, dentin will generally be interposed between the material and the pulp. The dentin barrier, although possibly only a fraction of a millimeter thick, is effective in modulating the toxic effect of a dental material. This dentin barrier effect is illustrated by the following classic study (Table 6-2). Three methods were used to evaluate the following materials: ZOE (zinc oxide eugenol) cement; resin composite; and silicate cement. The evaluation methods included (1) four different cell culture tests, (2) an implantation test, and (3) a usage test in class 5 cavity preparations in monkey teeth. The results of the four cell culture tests were relatively consistent, with silicate having only a slight effect on cultured cells, composite a moderate effect, and ZOE a severe effect. These three materials were also embedded subcutaneously in connective tissue in polyethylene tubes (secondary test), and observations were made at 7, 30, and 90 days. Reactions at 7 days could not be determined because of inflammation caused by the operative procedure. At 30 days, ZOE caused a more severe reaction than silicate cement. The inflammatory reactions at 90 days caused by ZOE and silicate were slight, whereas the reaction to resin composites was moderate. When the three materials were evaluated in class 5 cavity preparations under prescribed conditions of cavity size and depth (usage test), the results were quite different from those obtained by the other methods. The silicate was found to have the most severe inflammatory reaction, the composite had a moderate to slight reaction, and the ZOE had little or no effect.

Apparent contradictions in this study are explained by considering the components that were released from the materials and the environments into which they were released. The silicate cement released hydrogen ions that were probably buffered in the cell culture and implantation tests but were not adequately buffered by the dentin in the usage tests. Microleakage of bacteria or bacterial products may have added to the inflammatory reaction in those usage tests. Thus this material appeared to be the most toxic in the usage test. The composites released low-molecular-weight resins, and the ZOE released eugenol and zinc ions. In the cell culture tests, these compounds had direct access to cells and probably caused the moderate to severe cytotoxicity. In the implantation tests, the released components may have caused some cytotoxicity, but the severity may have been reduced because of the capacity of the surrounding tissue to disperse the toxins. In usage tests, these materials probably were less toxic because the diffusion gradient of the dentin barrier reduced concentrations of the released molecules to low levels. The slight reaction observed with the composites may also have been caused in part by microleakage around these restorations. The ZOE did not show this reaction, however, because the eugenol and zinc probably killed bacteria in the cavity, and the ZOE may have reduced microleakage.

Another example of the lack of correlation of usage tests with implantation tests is the inflammatory response of the gingiva at the gingival and interproximal margins of restorations that accumulate bacterial plaque and calculus. Plaque and calculus cannot accumulate on implanted materials and therefore the implantation test cannot hope to duplicate the usage test. However, connective tissue implantation tests are of great value in demonstrating the cytotoxic effects of materials and evaluating materials that will be used in contact with alveolar bone and apical periodontal connective tissues. In these cases, the implant site and the usage sites are sufficiently similar to compare the test results of the two sites.

Early combination schemes proposed a pyramid testing protocol, in which all materials were tested at the bottom of the pyramid and materials were “weeded out” as the testing continued toward the top of the pyramid (Figure 6-6). Tests at the bottom of the pyramid were “unspecific toxicity” tests of any type (in vitro or animal) with conditions that did not necessarily reflect those of the material’s use. The next tier shows specific toxicity tests that presumably dealt with conditions more relevant to the use of the material. The final tier was a clinical trial of the material. Later, another pyramid scheme was proposed that divided tests into initial, secondary, and usage tests. The philosophy was similar to the first scheme, except that the types of tests were broadened to encompass biological reactions other than toxicity, such as immunogenicity and mutagenicity. The concept of a usage test in an animal was also added (versus a clinical trial in a human). There are several important features of these early schemes. First, only materials that “passed” the first tier of tests were graduated to the second tier, and only those that passed the second tier were graduated to the clinical trials.

Presumably, then, this scheme funneled safer materials into the clinical trials area and eliminated unsafe materials. This strategy was welcomed because clinical trials are the most expensive and time-consuming aspect of biocompatibility testing. Second, any material that survived all three tiers of tests was deemed acceptable for clinical use. Third, each tier of the system put a great deal of weight on the tests used to accurately screen in or out a material. Although still used in principle today, the inability of in vitro and animal tests to unequivocally screen materials in or out has led to development of newer schemes in biocompatibility testing.

Two newer testing schemes have evolved in the past 5 years with regard to using combinations of biocompatibility tests to evaluate materials (Figure 6-7). Both of these newer schemes accommodate several important ideas. First, all tests (in vitro, animal, and usage) continue to be of value in assessing biocompatibility of a material during its development and even in its clinical service. For example, tests of inflammatory response in animals may be useful not only during the development of a material, but also if a problem is noted with the material after it has been on the market for a time. Second, these new schemes recognize the inability of current testing methods to accurately and absolutely screen in or out a material. Finally, they incorporate the philosophy that assessing the biocompatibility of a material is an ongoing process. Undoubtedly, we will see still newer strategies in the use of combinations of biocompatibility tests as the roles of materials change and the technologies for testing improve.