What do we mean when we say a biomaterial is biocompatible?
• Is biocompatibility “yes” or “no,” or is there a continuum of biocompatibilities ranging from “good” to “bad?”
• How can we measure biocompatibility?
• How do we improve or enhance the biocompatibility of a biomaterial?
The idea of biocompatibility is central to what makes a material a biomaterial. Also, new developments and concepts in cell biology are shifting thinking about biocompatibility, and this chapter offers an opportunity of a glimpse into the future of biocompatible biomaterials.
Biocompatibility is so central to what makes a material a biomaterial, that a focused discussion of the subject is justified.
This chapter clarifies some of the issues in biocompatibility, and also raises questions that will likely impact the field in the coming years. In contrast to empirical approaches and practical considerations focused solely about the safety of implanted devices (for example, toxicology, the state of the art today), modern cell and molecular biology ideas may give us a useful “theory of biocompatibility” with quantifiable parameters, testable hypotheses, and validated engineering rules.
We start with an overview of the state of the art in biocompatibility today, i.e., in the first decade in the 21st century.
A definition for biocompatibility, widely used in the biomaterials/medical device community, has been presented previously in the textbook and is repeated here:
“the ability of a material to perform with an appropriate host response in a specific application” (Williams, 1987).
This definition, though accurate and quite useful in the design, development, and application of biomaterials in medicine, nevertheless offers no insights into the mechanisms of biocompatibility, how to test the biocompatibility of a material or how to optimize or enhance the biocompatibility of a material. This section of the textbook will expand the definition and explore the philosophical and scientific ideas surrounding biocompatibility.
Specific concepts impacting biocompatibility are elaborated upon in Chapters II.2.2 (Inflammation, Wound Healing, and the Foreign-Body Response), II.3.3 (In Vitro Assessment of Cell and Tissue Compatibility), II.3.4 (In Vivo Assessment of Tissue Compatibility), and III.2.3 (Voluntary Consensus Standards) (and other sections throughout this textbook). Biocompatibility can be assessed using in vitro and in vivo assays. Although a variety of direct chemical and physical interactions may also be important, measurement of the consequences of leachable or secreted substances from biomaterials to cells in culture is the primary goal of in vitro biocompatibility assays. For example, cell proliferation inhibition or cell death are negative outcomes in such assays, and would be characteristic of materials that are not biocompatible. The in vivo (implantation) response to a “biocompatible” biomaterial is generally described as a no more than mild inflammatory reaction, which, after some 2–3 weeks, resolves itself into a thin fibrous capsule (largely scar as a result of the tissue trauma of implantation). This is sometimes referred to as a low level, long-term, minimal inflammatory response (i.e., macrophages are present at the implant surface even years after implantation, but the reaction site is relatively quiescent, and there is otherwise no indication of adverse local or systemic response). The presence of foreign-body giant cells at the interface may suggest lesser “biocompatibility.” The composite reaction is termed the “foreign-body reaction,” or FBR.
Four factors impact that which we refer to as “biocompatibility.” These are:
1. Toxicology (the measurement and study of the effects of material leaching from biomaterials) (II.3.3 In Vitro Assessment of Tissue Compatibility and III.2.2 Standards).
2. Reactions related to products from extrinsic microbiologic organisms colonizing the biomaterial (for example, endotoxin contamination; see Chapters II.2.8 and III.1.2).
3. Mechanical effects such as rubbing, irritation, compression, and modulus mismatch. Closely related are size-related effects, for example, if the implant is much larger than a macrophage, comparable in size to a macrophage, and capable of being phagocytosed, or much smaller than a macrophage, i.e., nano-size.
4. A broad range of interactions with surrounding proteins, and cells, inducing cell–biomaterials interactions (and tissue–biomaterials interactions) that might direct longer-term in vivo bioreaction.
Four factors impact that which we refer to as “biocompatibility.” These are:
Points 1, 2, and 3, above, are well-understood, and often applied in the design of biocompatible biomaterials – we understand the principles, have the ability to measure their impact, and we can design devices using clearly-defined principles to achieve good outcomes. Point 4 is less well-developed and concerns are expanded upon below.
Polymeric materials often contain extractable components, such as unreacted monomer, oligomers, initiator fragments, stabilizers, and other processing additives. Metals, glasses, and ceramics can release ions and other processing components. The type of reaction that will be considered from a toxicology standpoint is if these substances are released and negatively impact cells (in vitro) or adjacent tissues (in vivo), or if they affect an organism systemically. Biomaterials science, standards organizations, and allied fields have developed reliable methods to measure and identify leachates, and also sensitive and standardized methods to look at the reaction of tissues and cells to these leached substances. These methods are summarized in Chapter II.3.3. The ISO 10993 standards provide many specific, defined tests for toxicity associated with leachables.
Bacteria and their cell wall components are intense inflammatory activators. Fungi such as Candida are also inflammatory activators (Kojic and Darouiche, 2004). In the case of implants that are contaminated with fungi, bacteria or bacterial cell-wall endotoxin (lipopolysaccharide), an intense and usually long-term biological reaction is seen, characterized by large numbers of white cells (mostly neutrophils and macrophages, collectively called “pus”) in the vicinity of the implant. In humans, this reaction would be described by the patient as producing pain, redness, and heat. This response to the contaminated implant can lead to exceptionally thick foreign-body capsules. High concentrations of extravascular white cells and thick, dense foreign-body capsules are characteristics of poor biocompatibility. Such extreme reactions to devices with contaminating organisms have been documented with breast implants (Pajkos et al., 2003), and with other implant devices. This is distinct from infection with live microbiologic organisms, such as bacteria or fungi, in which the organisms may proliferate and cause ongoing local tissue destruction and potentially systemic effects. Although infection with live organisms shares some morphologic features with, and can occasionally look like, “poor biocompatibility,” we generally refer to this not as a biocompatibility issue, but rather a sterilization issue (see Chapter III.1.2).
If an implant is rubbing, abrading or moving in contact with tissue, or has sharp corners, undesirable reactions that are superficially classified as “non-biocompatible” will be observed. Mechanical mismatch between a hard biomaterial and a soft tissue can lead to damage or irritation to the soft tissue. Cell responses to mechanical forces are well-known and usually quite significant (Stamenović and Ingber, 2009). An excellent example of the mechanical effect on in vivo bioreaction was seen in 1976, where scientists implanted in rat muscle medical grade “biocompatible” materials having circular, triangular, and pentagonal shapes (Matlaga et al., 1976). The degree of reaction increased in the order: circle (lowest reaction), pentagon, triangle. The effect was attributed to micromotion associated with the acute angle of the triangle leading to the greatest tissue irritation. In general, it is the role of the implant designer to ensure that the device does not excessively rub or irritate tissue (rounded edges are better than sharp edges, for example). It is the role of the surgeon to appropriately place and anchor the device in the implant site to minimize such rubbing and irritation.
This topic dominates the literature in biomaterials science (see Chapter II.1.3, Cells and Surfaces In Vitro, for a detailed elaboration). It has been clear for 100 years or more that living cells interact and attach to different materials in different ways, and the nature of that interaction may have immense influence on cell fate, including attachment, spreading, proliferation, differentiation, activation, secretion, and detachment. It is also well-established that the adsorbed protein film always preceding cell interaction with surfaces directs and modulates the cell response (see Chapter II.1.2). Since inflammatory cells, such as neutrophils and macrophages, “interrogate” implanted materials shortly after implantation, and since different surfaces interact in different ways with proteins and cells, we would expect the biomaterial to impact the cell-driven in vivo reaction. However, let’s examine this conclusion. In vitro, profound differences are seen in cell interactions between different materials. For example, a poly(2-hydroyxethyl methacrylate)(polyHEMA) hydrogel will not permit macrophages to adhere in cell culture, while a tissue culture polystyrene (TCPS) surface readily allows those same cells to adhere. Yet, if the polyHEMA and TCPS are implanted in vivo, both will heal similarly with an avascular, collagenous foreign-body capsule. In fact, all “biocompatible” materials, be they hydrophilic, hydrophobic, metallic, polymeric or ceramic, will heal similarly with a classic (and largely quiescent) foreign-body reaction (FBR) if there are: (1) no leachables; (2) no products from extrinsic organisms; and (3) no mechanical irritation. The explanation for this striking difference between in vitro bioreaction and in vivo bioreaction has yet to be identified, but does highlight the multi-cell complexity of the in vivo environment, in comparison to the relatively simple environment with one cell type in vitro.
The phenomena of: (1) frustrated phagocytosis; and (2) cytokine release are important to discuss here. Macrophages are programmed to engulf and digest foreign material. Phagocytosis occurs after the foreign-body has bound to molecules called “receptors” that are on the surface of the macrophage. The macrophage then stretches itself around the foreign-body and engulfs it. Once inside this phagocyte, the foreign particle is trapped in a compartment called a lysosome which has a battery of chemicals that attempt to degrade the foreign material. Frustrated phagocytosis is the term used to express the concept that the macrophage is incapable of “eating” a piece of biomaterial approximately larger than its size or certainly a macroscopic implant. In this situation the lysosomes (which may be in contact with the foreign material but open to the surroundings) may release their contents into the adjacent tissues. This may cause local tissue damage and inflammation. Moreover, to address the large surface, the macrophage might fuse with other macrophages and form multinucleated foreign-body giant cells, often used as a marker of a more severe FBR. Macrophages also release cytokines (diffusible signaling proteins) in response to biomaterials (Bonfield et al., 1992). Cytokines can be considered as proinflammatory (e.g., IL-1, TNF-a) or anti-inflammatory (e.g. IL-4, IL-10). A measurement of the cytokine shower from an implanted biomaterial may offer insights to biocompatibility.
Important points about biocompatibility as we understand it today are:
1. The biological reaction we call biocompatibility is negatively impacted by leachables, products of extrinsic organism surface contamination, and micromotion.
2. As long as leachables, extrinsic organism surface contamination, and micromotion are not impacting the reaction, all materials will give an approximately similar bioreaction in vivo, referred to as the normal FBR, composed of a thin fibrous capsule and minimal ongoing inflammation.
3. When the foreign body capsule is thin and the reaction site, after approximately one month, is relatively quiescent, this is an acceptable FBR and we call the implant “biocompatible (or inert).”
4. Inert biomaterials lead to the reaction described in point 3 (above), and thus are called “biocompatible.”
5. The favorable long-term interface between a biomaterial and the surrounding tissues in most medical devices is characterized by a thin, dense, collagenous capsule that isolates the biomaterial implant from the body.
Although we implant millions of devices made of biocompatible biomaterials in humans every year, largely with much success, there are concerns with the way implants heal (the FBR), and new generations of materials are designed for applications in which something different than a quiescent FBR is desired.
For example, a dense fibrous capsule can inhibit diffusion of analytes to implanted sensors, interfere with release of drugs from implanted controlled drug release devices, and enhance the resistance of an electrical path, thereby blocking communication with tissues for implanted electrodes. Capsular contraction is a problem for some devices, such as breast implants, where the contraction of the fibrous capsule (i.e., all scars contract) distorts the soft implant, tending to make it spherical. Moreover, the lack of vascularity near the implant–tissue interface can slow the body’s response to, and treatment of, bacterial invasion and related biofilm formation. Certainly, the capsule associated with the FBR can create surgical problems for device removal and revision. In many cases, a vascularized, integrated tissue reconstruction (more resembling normal tissue reconstruction) would be preferable to the avascular, dense capsule.
The potential for vascularized, non-fibrotic healing is now being realized. Such reconstructive healing can be achieved with extracellular matrix (ECM) components or with inert biomaterials with engineered porosity.
Decellularized ECM derived from a number of tissues, for example small intestinal submucosa (SIS), has been found to heal into the body with little or no fibrosis, excellent vascularity, and general tissue reconstruction (Badylak, 2007). SIS and other decellularized ECMs have been used in more than 1,000,000 human surgeries, largely with good results. Chapter II.6.16 elaborates on these decellularized tissues as biomaterials. If the SIS is cross-linked, it will heal in a proinflammatory manner, with a capsule and a classical FBR. The excellent healing of SIS is attributed to the ability of macrophages to degrade the ECM to bioactive peptides that actively promote healing. Importantly, the ECM structure is heavily infused with macrophages in the early stage of healing, and those macrophages have been shown to be in a phenotype (also referred to as a polarization) conducive to healing (M2), in contrast to the proinflammatory, fibrotic M1 phenotype (Mantovani, 2006; Badylak et al., 2008). There are also suggestions that soluble ECMs derived from neonatal cell culture might be used as coating for implants to aid in healing and integration (Naughton and Kellar, 2008).
Certain porous synthetic biomaterials will also heal in a minimally fibrotic, angiogenic fashion. Observations on the special characteristics of the healing of porous structures in the body go back to the early 1970s (Karp et al., 1973; Klawitter et al., 1976). Many studies discussed this porosity effect, and an extensive study in 1992 clearly demonstrated the importance of pore size to healing (Brauker et al., 1995). The concern with all these studies was that the materials used had a broad distribution of pore sizes, making it difficult to ascertain the effect on healing of a specific pore size. A method was developed to make materials with a single, consistent pore size, based on using solvent-soluble microspheres as templates to create uniform, interconnected pores. When such materials were implanted subcutaneously, it was noted that when pores were in the size range 30–40 microns, vascularized healing and reconstruction with little fibrosis was observed (Marshall et al., 2004; Madden et al., 2010). Also, these materials were heavily infused with macrophages during healing, and more of the macrophages were in the M2 phenotype (Madden et al., 2010). These same materials with 30–40 micron pores healed well in skin percutaneous sites (with dermal and epidermal reconstruction) (Fukano et al., 2010), heart muscle, and other tissues. A further discussion of porous materials can be found in Chapter I.2.15.
With relevance to biocompatibility, consider differences engendered solely as a result of physical factors. A synthetic hydrogel is fabricated as a solid slab or as a porous structure with 30–40 micron interconnected pores. The chemical compositions of both are identical. Also, they have similar mechanical properties, no leachables, and no endotoxin or bacteria. Yet one heals in a capsule with the classic FBR, while the other heals in a vascularized, reconstructed manner with little fibrosis. It seems challenging to use the word “biocompatible” for both, given the sharp differences in in vivo biological reaction, despite identical chemistries.
A “biocompatible” material is fabricated as a porous structure or a solid slab. The porous structure heals in an avascular, pro-angiogenic manner. The solid slab is encapsulated with the classic FBR. How can we use the word “biocompatible” for two such different bioreactions?
Since the introduction of “modern” biomaterials to medical practice in the late 1940s and early 1950s, we have seen an evolution in sophistication and functionality (Williams, 2008). First generation biomaterials were simply inert (e.g., Titanium, Teflon™, Polyethylene). Second generation biomaterials (1970s, 1980s) interacted with and changed the biological environment in a controlled manner. Such materials might be engineered for biodegradability, poly(lactic acid) for example, or engineered to integrate into bone and induce bone healing, such as bioactive glasses, or to induce a controlled thrombotic deposit, such as a textured polymer surface on the internal surface of a left ventricular assist device. Third generation biomaterials (1990+) biospecifically orchestrate biological processes, and can direct regeneration and restore functionality, and/or respond to the environment in a proactive manner to favorably influence a tissue reaction. Examples include engineered extracellular matrices (Lutolf and Hubbell, 2005), sphere templated biomaterials (Madden et al., 2010), SIS (Badylak, 2007), and “smart” biomaterials (Reyes et al., 2007). Also, consider the published debate on the effectiveness of biomaterials guided by specific peptide signals (Williams, 2011). As biomaterials evolve, so too must the definition of biocompatibility.
Thus, the way we define biocompatibility may change over the next few years. Examples that challenge the present paradigm of biocompatibility are published, and materials that show unique healing are being applied in clinical medicine. Recent clarification on the diversity of macrophage phenotypes and understanding of resident tissue stem cell pools in the body may permit engineered “biocompatibility,” with control of the extent, rate, and speed of integration. These new developments strengthen the science of biocompatibility, and bring biomaterials and tissue engineering closer together. Precision control of biocompatibility can lead to new biomaterials-based therapies with profound advantages for the patient.
For now, the medical device regulatory agencies subscribe almost exclusively to the definition of biocompatibility that we have been using since the first functionally successful implants in the 1950s. In learning the subject of biomaterials, it is important that one appreciates the accepted standards for biomaterials; standards that are used by the $300 billion plus medical device industry worldwide (estimated to reach $415 billion in 2016) and standards that impact millions of patients (see Chapters III.2.2 and III.2.3). Still, since we are on the cusp of a shift in thinking about this subject, it is useful for these newer ideas to be addressed and placed in the context of the word “biocompatibility.”
Understanding biocompatibility is central to understanding biomaterials.
1. Badylak SF. The extracellular matrix as a biologic scaffold material. Biomaterials. 2007;28(25):3587–3593.
2. Badylak S, Valentin J, Ravindra A, Mccabe G, Stewart-Akers A. Macrophage phenotype as a determinant of biologic scaffold remodeling. Tissue Engineering Part A. 2008;14(11):1835–1842.
3. Bonfield TL, Colton E, Marchant RE, Anderson JM. Cytokine and growth factor production by monocyte/macrophages on protein preadsorbed polymers. J Biomed Mater Res. 1992;26:837–850.
4. Brauker JH, Carr-Brendel VE, Martinson LA, Crudele J, Johnston WD, et al. Neovascularization of synthetic membranes directed by membrane microarchitecture. J Biomed Mater Res. 1995;29:1517–1524.
5. Fukano Y, Usui ML, Underwood RA, Isenhath S, Marshall AJ, et al. Epidermal and dermal integration into sphere-templated porous poly(2-hydroxyethyl methacrylate) implants in mice. Journal of Biomedical Materials Research Part A. 2010;94(4):1172–1186.
6. Karp RD, Johnson KH, Buoen LC, Ghobrial HKG, Brand I, et al. Tumorigenesis by millipore filters in mice: Histology and ultrastructure of tissue reactions as related to pore size. Journal of the National Cancer Institute. 1973;51(4):1275–1279.
7. Klawitter JJ, Bagwell JG, Weinstein AM, Sauer BW, Pruitt JR. An evaluation of bone growth into porous high density polyethylene. J Biomed Mater Res. 1976;10(2):311–323.
8. Kojic EM, Darouiche RO. Candida infections of medical devices. Clin Microbiol Rev. 2004;17(2):255–267.
9. Lutolf MP, Hubbell JA. Synthetic biomaterials as instructive extracellular microenvironments for morphogenesis in tissue engineering. Nature Biotechnology. 2005;23(1):47–55.
10. Madden LR, Mortisen DJ, Sussman EM, Dupras SK, Fugate JA, et al. Proangiogenic scaffolds as functional templates for cardiac tissue engineering. Proc Natl Acad Sci USA. 2010;107(34):15211–15216.
11. Mantovani A. Macrophage diversity and polarization: In vivo veritas. Blood. 2006;108(2):408–409.
12. Marshall AJ, Irvin CA, Barker T, Sage EH, Hauch KD, et al. Biomaterials with tightly controlled pore size that promote vascular in-growth. ACS Polymer Preprints. 2004;45(2):100–101.
13. Matlaga BF, Yasenchak LP, Salthouse TN. Tissue response to implanted polymers: The significance of sample shape. J Biomed Mater Res. 1976;10(3):391–397.
14. Naughton G, Kellar R. Human ECM for devices and therapeutics. MD&DI 2008;102–109 May.
15. Pajkos A, Deva AK, Vickery K, Cope C, Chang L, et al. Detection of subclinical infection in significant breast implant capsules. Plastic and Reconstructive Surgery. 2003;111(5):1605–1611.
16. Reyes CD, Petrie TA, Burns KL, Schwartz Z, García AJ. Biomolecular surface coating to enhance orthopaedic tissue healing and integration. Biomaterials. 2007;28(21):3228–3235.
17. Stamenović D, Ingber DE. Tensegrity-guided self assembly: From molecules to living cells. Soft Matter. 2009;5(6):1137–1145.
18. Williams DF. Definitions in Biomaterials. In: Progress in Biomedical Engineering, 4. Amsterdam: Elsevier; 1987; p 72.
19. Williams DF. On the mechanisms of biocompatibility. Biomaterials. 2008;29(20):2941–2953.
20. Williams DF. The role of short synthetic adhesion peptides in regenerative medicine: The debate. Biomaterials. 2011;32(18):4195–4197.