Packaging and coating materials for implantable devices
Implantable micro-electromechanical systems (MEMS) offer a valuable means of sensing and actuation within the body at unprecedented levels of miniaturisation and internal functional complexity. However, function can be compromised by moisture uptake and adverse interactions within the body. Consequently, packaging materials serve as barrier phases for controlling exchange of solutes and furnishing biocompatible surfaces for MEMS devices. This chapter covers a range of coatings used to protect MEMS devices in the body. Surface modification and bioactive agent release are described as strategies for enhancing biocompatibility and reducing surface fouling. The full range of mechanical, permeability and chemical features of packaging materials are involved in the extended survival of MEMS implants and are described for contemporary packaging materials.
This chapter describes the materials, packaging and coating technologies that have been investigated to date to enhance the survival and functionality of implantable devices in the body. The challenges of sealing and protecting the vulnerable elements of the implanted device are discussed and the major types of materials used thus far are highlighted. Accordingly, there is an emphasis on silicones and polyurethanes, though newer materials such as diamond-like carbon (DLC) and conducting polymers are also considered. The tailored requirements of different types of implanted devices vary depending on functionality and on whether the device is based on micro- or macro-scale components. A further differentiation, which is also considered, is the provision of different coatings over the active versus the passive surfaces of a device.
Traditional implant materials at the macro-scale, using ceramics, polymers and metals, have made substantial contributions to therapeutic intervention. In contrast to metals, ceramics and glasses are transparent to radio-frequencies and have enabled hermetic encapsulation, for example, for neuromuscular stimulators and cochlear implants; materials used include zirconia, alumina and borosilicate glass. The additional requirement for effective sealing of the encapsulant and the conducting path has been that of specialist welding techniques; examples include welding of glass to tantalum and use of lower temperature melting filler such as vanadium/niobium for ceramic to metal seals. Corrosion poses a challenge to the use of metals and is the result of electrochemical surface reactions in biological fluids, especially where there are dissimilar metals present. Ceramics, too, may show slow ageing behaviour, with, for example, a change in crystalline state for yttria stabilised zirconia leading to a lower fracture toughness state material.
However, we are now moving into an era where structural and functional underpinnings of tissues are being attempted using more active and responsive implants, especially those built around micro-fabricated structures. A key feature of such systems is that they are dynamically interactive with the surrounding tissue, operate over extended periods and have precision interfaces that allow for precision contact with specialist tissues, at least at the millimetre length scale. To achieve this, both fabrication and in vivo positioning demands a precision approach. Micro-fabrication technology, notably MEMS, provides a classic means of meeting many of the internal design specifications. However, unlike other electronic devices, implanted MEMS structures have to function in a corrosive, high salinity aqueous medium, that is, biological fluid and tissue. This places considerable demands on methods for device protection, which are currently based on coatings as packaging materials. The requirements for MEMS devices are similar to macro-scale devices, not least because MEMS are usually housed in larger structures. Hence, packaging discussed in this chapter is relevant to both MEMS and macro-scale devices.
MEMS implantable medical devices now see application across all major areas of medicine encompassing diagnostics, therapeutic monitoring and drug delivery, as well as specialist applications such as neuro, cardio and bladder stimulation devices and sensory implants (cochlear and retinal) (Dario et al., 2000; Receveur et al., 2007; Wasikiewicz et al., 2007; Hodgins et al., 2008). The key element is the functional interface being able to sense and stimulate at a defined locus. MEMS have proved powerful supportive tools for addressing the general challenges of biomedical engineering with regard to instrumentation and data processing, and it is opportune that further developments in the semiconductor and micromechanics industries translate into implantable devices.
A feature of many new devices is the combined use of macro- and micro-scale features. Thus, high-density electrodes, for example, for nerve stimulation or sensing, may use microelectrodes but are built up on a macrostructure. Connectors from electrodes to substantive electronic substructures, by contrast, may be of low density and the resulting macro-device and its passive elements may be packaged using a more traditional capsule, such as one based on a metallic barrier. The latter is highly attractive because of its ability to provide hermetic sealing over a passive component.
As specific devices that combine multiple length scales, the Health Aims Project of the EU Framework VI Programme (Hodgins et al., 2008) provided useful examples, both for design and packaging. Here, a broad platform was developed, which included a range of macro- and micro-scale device motifs. Thus, for functional electrical stimulation (FES) of upper limb muscles designed to assist patients with motor neuron damage, an externally controlled, macro-scale, implanted stimulator was used, which essentially constituted the passive, fully sealable element of the system. This in turn was connected to dual electrode stimulators as the active FES components, which optimally require coatings at the micron scale to provide precision contact with tissue. A macro-platform was also utilised for a cochlear implant, while packaging for the associated miniature cochlear electrode array utilised methodology from the semiconductor industry; devices here combined silicone and Parylene C as packaging material. For a retinal implant, similarly, the overall implanted device was embedded in a biocompatible package, but the need for packaging dense, active, ganglion stimulating microelectrode arrays needed a micro-scale coating architecture. An intracranial pressure monitor by contrast had a relatively large-scale (3 mm diameter) pressure sensing head as the active surface, built into a monolithic device for radio-frequency (RF) communication. These two had a common biocompatible packaging, appropriate for a macro-scale implant. So, essentially, micro-scale communication sub-components were utilised which did not need surface structural resolution since they served as an internal relay platform, linked to a large-scale electrode detecting ambient fluid pressure changes, rather than changes at selective tissue locations. A device for urethral sphincter pressure monitoring used a micro-scale strain gauge, which required packaging with material that allowed for response sensitivity while affording a fluid compatible interface. So, clearly, there are different coating and packaging needs, which vary depending on the application, degree of implantation and exposure involved for micro-scale components.
Extensive in vitro analysis of devices with feature lengths below 1 μm has been in progress for many years (Whitesides et al., 2001), and our ability to organise and manipulate cells and fluid flows using such constructs is now well established (Khademhosseini et al., 2006). The outcome of such work is the advancement of micro-architectured devices that have sufficient biocompatibility to be able to operate in vivo. Viability in vivo is already a driver for contemporary tissue engineering and regenerative medicine. The materials and components typically used, particularly polymers, have a well understood in vivo biocompatibility and are based around already used traditional macro-scale implants (Khademhosseini et al., 2006). While the development of MEMS devices, micro-fabricated sensors, stimulating electrodes, drug delivery devices, ultrasound transducers and micro pumps has thus far shown impressive in vitro performance, they also need a similar range of in vivo acceptable materials. The need is for biocompatible packaging rather than to attempt to make the functional parts themselves more biocompatible. MEMS devices could not be readily made from materials that would tolerate in vivo exposure, nor would they be guaranteed to avoid toxicity effects, especially during long-term in vivo operation (Dario et al., 2000). Indeed, the more complex the design concept, the greater the problem becomes of achieving stable operation inside the body. One survey found that, of the many reported operational, micro-fabricated implant systems, only 18 had reached product stage, and of these two gave confirmed long-term performance (Receveur et al., 2007 ).
Devices implanted in the human body need to be structurally stable and biocompatible to prevent harmful interactions between components of the implant and the body (Iwasaki and Nakabayashi, 2005). At the same time, their functionality needs to be unaffected by the surrounding biological response. In contrast to traditional macro-scale implants, maintaining functionality is a serious challenge, since MEMS sub-components may well be made from toxic materials, and where based on metals, are likely to corrode rapidly, so generating heavy metal leachables. Furthermore, the device function is itself vulnerable to ingress from body fluids; microelectronics were certainly not designed to operate in high humidity environments, let alone in contact with body fluids. This puts a premium on technical solutions that both create barrier layers around devices using materials and have a high intrinsic biocompatibility. Admittedly, we do not have perfect solutions and the functional stability this offers is finite, but can extend to periods of years, so at least lowering implantation risks, along with a need for device replacement on short time scales.
The design of biocompatible materials for device packaging is arguably as much of a challenge as the design of the device itself. The challenge is somewhat greater if the encapsulation of a device is to extend to active, functional elements as well as to passive surfaces on the device. In the former case, it may be necessary to allow selective entry of solute species for chemical sensing, to provide mechanical transmission for physical transduction and to permit electrical connectivity for neuromuscular stimulation. This crosstalk with the biological tissue has then to be coupled with sealing of passive surfaces of the device such that an effective, higher order barrier to water and leachable exchanges is secured. It is difficult to reconcile the controlled access to the active surface with its effective protection and so some consideration of overall device geometry is necessary, along with exploration of multiple barrier materials. Protein deposition in vivo can lead to reduced solute and charge exchange between the biological tissue and the device’s active surface. This can compromise function; accordingly, materials with modified surfaces demonstrating greater biocompatibility (less protein deposition) have been developed as coating materials that at the same time allow for device-tissue interactions (Shimizu et al., 2010).
The materials generally used for packaging of devices that have multiple internal components, for example, wire connections or batteries for powering are mainly polymers, but inorganic materials have also been tried. Polymers have the advantage of offering a broad spectrum of materials capabilities, tuneable through synthetic design. In addition, they are applicable to a wide range of delicate device surfaces, regardless of the complexity of shape or topography; since they allow for coating at low processing temperature, MEMS structures are unlikely to be damaged. We already have considerable experience of polymeric materials as implants, and so, at least, the biocompatibility can be predicted. An example of one usable, traditional system, tested to extreme mechanical and flow conditions, is polyurethane, for example, used in membrane form as part of the artificial heart (Belanger et al., 2000). Polyurethane tested under such conditions has allowed assessment to be made of the haemocompatibility and mechanical stability of different material types. The many structural possibilities with polyurethanes also make them a particularly good example of functional versatility through chemical design. They can form partial barriers if necessary, as when some permeability is required over the sensing surface of metabolite biosensors (Vadgama, 2007).
A concern remains over the long-term exposure of polymers to tissue; degradative hydrolytic and other enzymes, together with free radicals, particularly as generated by inflammatory white cells locally, have the ability to break any number and types of covalent bonds in polymer chains. While this chemical degradation mechanism is not in play in relation to inorganic coatings, viz glasses and ceramics, as these are relatively resistant to degradative attack, they still tend to be degraded by surface dissolution, which can have cumulative erosive effects with loss of packaging integrity. The question then is what the cumulative loss of material might be, especially beyond the short experimental timescales normally used to predict film stability and biocompatibility for such uses (Kotzar et al., 2002); with thin films, very little erosive loss is needed before barrier function is compromised. The nature of the bio-fluid environment also dictates outcomes; the on-going surface coagulation seen at vascular implants is not evident in tissue, while extreme osmotic and pH excursions seen at urine contacting devices creates quite distinct swelling and degradative effects at polymers. Experience of polymeric catheter materials (Lawrence and Turner, 2005) at least provides us with pointers to packaging needs, and there is also growing understanding of processing needs for conformal deposition on devices (Donaldson, 1989).
The process of coating a device leads to varying degrees of effective packaging around it, and typically coating with a solvent loaded polymer leads to a form of controlled packaging, while an inorganic film might be formed by vacuum deposition. In some instances of course, as in the case of preformed ceramic or metal, the packaging outcome does not involve a coating process, per se, but the engineering of a package around a macro-device. Therefore, packaging is an outcome, though the methodology may differ. This section covers the challenges of reliable packaging. Failure modes are different for different types of materials, but all have to be considered in the context of usage prior to a commitment to any specific material. The mechanisms leading to failure considered here are: initial incomplete coatings, especially a problem with ultrathin layers; lack of uniformity and homogeneity; water uptake that can lead to device corrosion as well as facilitate coating material degradation; and the triggering of local tissue reactions (bio-incompatibility) due to either the material itself or to its degradation products.
Any failure of packaging, eventually leading to implantable device malfunction, can be considered under two general headings. The first relates to the problems of isolating sensitive electronic components from the reactive environment of the body, and the second relates to the interfacial biocompatibility of the packaging material.
As regards device isolation using polymers, the dominant problems are moisture ingress and mechanical stress, for example, due to differential swelling or contraction over an inelastic substrate. The latter process may lead to micro-crack formation, and thereby loss of packaging integrity, and a subsequent accelerated ingress of moisture, finally culminating in electrical shorting of the device along with its corrosive disruption. In addition, through volumetric strain in the coating layer, adhesion to the subjacent device may be compromised, leaving micro-gaps for the accumulation of liquid water. Moisture penetration through a protective film is a serious challenge, as in reality no polymer has total resistance to gas molecules, and therefore to water vapour transport, simply driven by concentration gradients. Once transported into the device, water molecules aggregate to form condensates in micro-gaps promoting pooling of water, which can then more readily allow tracking of water along voids to damage electrical circuitry. Condensed water, when in contact with the metallic parts of a device, especially if there are entrained ions, drives the corrosion process. As a result, a complex mix of diffusible heavy metal ions, metallic particles and metal oxide will be released; the toxic potential of these remains to be fully evaluated, but warrant much greater study. It has been suggested that release of metallic components (ions, small particles) could lead to tumours (Gillespie, 1988). The body, overall, has the means to deal with higher toxic agent loads than those engendered by miniature devices, but it is both the exotic nature of some of the trace components within MEMS systems and the local accumulation that raises a question mark over what is allowable. Some indication of the complexity of biological effects is indicated by the effects of wear particle release from articulating artificial joints and how these cause indirect effects on joints via inflammatory mediators.
Key requirements for the coatings themselves are homogeneity, lack of defects (pin holes), strong substrate adherence and sufficient depth to accommodate superficial degradative loss in vivo. Coatings should form a continuous, conformal surface over the device and should especially maintain a structural homogeneity over device irregularities and sharp edges. Different coatings over active versus passive components of the device may be needed or it may be necessary to leave well-protected voids over the active surface, for example to allow for direct electrical contact with tissue. Where there is a film deposited over the active surface, functionality will be influenced by depth, density and topography of the film, all of which need to be considered.
Organic solvent-based deposition of preformed polymer is a typical method and usually acceptable for electronic devices, provided any further processing or curing is at low temperatures. A deposited film must, however, be sufficiently robust to permit normal surgical handling for implantation and retention in vivo. To ease the implantation process, in some cases, an entire packaged system needs to be flexible enough to be threaded through a small surgical incision. Soft polymeric coatings can help with this and can be important in the mechanical shielding of rigid wire connectors to help avoid breaks and preserve electrical connectivity. Elastomers also provide a level of mechanical buffering that reduces the compliance mismatch between hard devices and soft tissues.
In contrast to polymers, inorganic coatings are water impermeable and capable of hermetic sealing of devices, achieving total electrical isolation for conducting paths. That does not, though, eliminate the danger of disrupted insulation paths and stray electrical leakage arising over time. The lack of wider application, despite the greater barrier effectiveness, resides in the high processing temperatures needed to form such barriers.
If a coating material degrades, it may retain its barrier function for a time, but the release of degradation products into the surrounding tissue will occur early and may provoke an increased tissue inflammatory response. The level of inflammation, undoubtedly, has an impact on the aggressiveness with which a capsule material is degraded; the consequence may be a self-accelerated process. In this context, it is important to recognise that a tissue implant site is actually a wound site, which already serves as a nidus of inflammation that is sustained by the body’s desire to reject the implant as part of its foreign body rejection.
The problem of degradation thus overlaps with the challenge of packaging material biocompatibility. Though the term ‘biocompatible’ is useful shorthand in the biomaterials world, there is actually no material as yet available that is truly biocompatible, other than possibly hydroxyapatite for bone substitution (Bowen et al., 1989); the body will always recognise the implant as non-self and respond to it as a foreign body to be either degraded or sealed within a natural collagenous capsule. What is under design control is how subdued or florid the rejection response is, and whether the pathway taken is of degradation or natural encapsulation. All these events are not entirely determined by device surface properties; device size, shape, tissue orientation and tissue type all play a part. Selection of the correct capsule material, however, is a good start. At the very least, the material should not promote an inflammatory reaction, and additionally, it should not be cell toxic, allergenic, carcinogenic or teratogenic. In the special case of a packaging material in contact with blood, active anti-thrombotic agents might need to be incorporated in the bulk or the surface of the packaging material, as materials with intrinsic haemocompatibility are difficult to produce. Prediction of in vivo outcomes may not be easy, and both in vivo and in vitro testing, for example, under the ISO 10993–11 range of tests, which provides a useful template, is a vital part of packaging development (Kotzar et al., 2002). The need for the extensive testing of entirely new materials can be a bottleneck to the introduction of packaging advances, and a conservative approach is a feature of this area of study.
Inevitably, no single material is suitable for all applications, and since all implantable devices require a coating, the key need is to tailor coatings to individual applications; preferably this is done in parallel with, rather than sequentially after, the basic clinical device has been developed. In terms of individual tailoring, some devices require flexible and others stiff coatings; a single layer might suffice or a laminate might be needed of the same or different materials. A laminate might contain specialist layers; an external layer for bio-inertness and surface biocompatibility, an intermediate layer as a solute barrier and an inner surface layer for high device adhesion.
76 Implantable sensor systems for medical applications 3.1.3 Current packaging and coating strategies
Micro-scale, structured devices, based on silicon micromachining, are angular, hard-edged, and mechanically incompatible with tissue, because they lack rounded surfaces, possible only with traditional machining. These angulated surfaces are more able to provoke tissue inflammatory responses and so such devices are not ideal for tissue compatibility when implanted (Hodgins et al., 2007). Similarly, macro-scale structures with hard, angulated edges would provoke a tissue reaction, especially if fabricated from a hard material. So it is important that any packaging capsule formed around the device mitigates this drawback. The mechanics of most polymer materials, especially elastomers, are an advantage in this regard, but it is important that they form complete coverings over the sharp edges of devices. They also need to be peel resistant and retain their internal structural integrity over the long-term. To achieve this, it may be necessary to set down multilayers as demonstrated by a combination of four silicone layers to establish a coherent (20 μm) capsule over a pressure sensor (Hierold et al., 1999).
All the attributes of a barrier material (mechanical, chemical and physical) have an effect on performance, but it is arguably the surface that has the greatest influence because of its direct contact with the biological tissue. Surface chemistry dictates facile cell and protein interactions; this manifests through surface energy and the balance of hydrophilicity and hydrophobicity, which variously influence protein packing density and conformational state. If there are surface mosaics, topographical features and distributed crystalline and amorphous domains in a polymer, this will affect the biological response (Hasirci and Hasirci, 2005; Wsaki and Nakabyashi, 2005). Less is understood of the effect of the ionic double layer in the aqueous phase over the surface, but this also will be important in conditioning the early stage of protein surface interaction, and therefore the subsequent cascade of biological surface events. Adsorbed water molecules and ions directly on the surface are another factor that will affect the nature of the surface presented to the tissue. It is of value to differentiate surface from bulk material properties (Castner and Ratner, 2002), and it is likely that there will be a distortion of polymer properties near the surface. This external zone is often assumed to be identical to the bulk, but differential hydration, inter-chain packing, crystallinity, polymer mobility, oxidation and chain migration will all lead to different properties at and near the surface, and these are not predictable on the basis of bulk materials characterisation.
An established coating material for implants that incorporate micro-fabricated components is the silicone elastomer. Silicones are versatile in their application and have been used, for example, to coat ceramic and titanium capsules deployed to provide initial hermetic sealing, such as in cochlear implants (Stieglitz, 2010). Donaldson (1991) undertook early, detailed evaluation of the properties of silicones that allowed for their clinical use. The soft elastomeric nature of these materials enabled them to maintain a strong adhesion to rigid device surfaces by better accommodating mechanical strain, for example, that induced by autoclaving. Their disadvantage is their permeability to water, but this appears to be countered, in part, by osmotic gradients. It is considered that ions will not enter silicone as any water entering the polymer is subject to the attractive, osmotic forces from ions in the external solution and therefore would tend to leave the polymer phase. In a biological fluid also, an osmotic counter gradient will be set up that serves to limit the uptake of water into the silicone phase. One critical advantage they have is their electrical resistance, serving to ensure electrical isolation of sensitive electronics. Experience so far also indicates that silicones are inert to degradation in the body for periods of many years. Some vulnerability to degradation, however, cannot be excluded; an accelerated aging study using elevated temperatures (100°C, 45 h) found that surface contact angle was altered, even though bulk mechanical properties remained unchanged (Kennan et al., 1997 ).
Parylene (poly-para-xylylene) is the other major polymeric material used (Bienkiewicz, 2006; Stieglitz, 2010). With chlorine atom substitution, it is known as Parylene C. This material has high flexibility and electrical resistance and is resistant to electrical breakdown. Importantly, it has low permeability for both water and ions, and can serve as an external coating on other materials because of its biocompatibility; it does not always adhere well to underlying polymeric substrates, however. It has been used as the substrate itself in the case of a retinal prosthesis (Rodger and Tai, 2005).
Of other materials, polyurethanes are of value because of their robustness and because of our knowledge of their in vivo performance as implants over 50 years (Hastings, 1970). They also have high tensile strength and tear and abrasion resistance (Lamba et al., 1997; Wright, 2006). DLC is another material coating that can be formed as a thin film. It is generated by vapour deposition at ambient temperatures, and fine control of both thickness and composition can be readily achieved. It is also suitable for use in combination with other barrier materials. The nature of a DLC film will depend on the forming process, ranging from soft, polymer equivalent structures, through to hard diamond type films (Stan et al., 2010). Other candidate materials that might be explored for encapsulation are epoxy (usually combinations of epoxide and amine hardners) and polytetrafluoroethylene (PTFE) (Stieglitz, 2001; Thil et al., 2005).
Again, within the Healthy Aims Programme (Hodgins et al., 2008), various packaging materials were explored. Silicones, because of their longstanding acceptability were physically modified using a highly water repellent additive, isopropyl myristate. This was designed to allow easier processing, increase scratch resistance through a greater pliability and most importantly to reduce water ingress. The key attraction here was the ability to use the material without safety concerns, because it was already licensed for clinical use, and, moreover, chemical derivatisation was not needed. The coatings based on this material were used with cochlear implant surfaces and pressure monitoring devices, and could potentially be used to cover microsurfaces of devices through additional machining. For finer surface architectures, such as the retinal device and where ultrathin layers are needed, as in the case of the urethral pressure sensor active surface, solvent-based material deposition giving thick films is less satisfactory, and here DLC appeared to be a more attractive proposition. Coatings could be deposited through a vacuum deposition technique to give controlled micron or sub-micron scale thickness and conformal layers which, because of their nature, were more resistant to water ingress. In addition, because of their relative flexibility, they readily allowed transmission of pressure changes to the coated transducer surface in the case of the urethral sensor.
Several types of polymers have been exploited for packaging passive device surfaces. The tolerance demanded is much less since the passive surface is one that does not need to be interactive with the tissues or body fluid and comprises components that enable device powering, achieve connectivity or inclusion of a signal processing unit. None of these is affected by the accuracy of coating layers; the only requirement is that there is no ingress of unwanted body components or leaching of device components. Accordingly, solvent coating of polymers has stood the test of time, with silicone and polyurethane the dominant materials, though DLC shows promise because of its high resistance to water uptake.
Passive surfaces, especially on electronic devices, need protection from harsh body environments. Early work involved sealing entire structures within metal enclosures, titanium packaging of the pacemaker power source being one (Kolenik, 1975). As new techniques became available, alternative hermetic encapsulation became possible, such as using glass (Wise and Weissman, 1971; Najafi, 2003) and ceramic (Ziaie et al., 1996). Glassy insulating films (e.g. fused quartz) combine transparency in the visible and infrared spectrum, good mechanical properties, low thermal impedance and high electrical resistance. These are attractive features, and they have special relevance in that they can provide thin, uniform films. Their limiting factor, however, is rigidity, which disqualifies them from use on any device that needs to be flexible or mobile. Ceramics are resistant to water permeation, are non-conducting and allow RF signals to be transmitted throughout a packaged component. Their chemical inertness is a basis for good biocompatibility, and minimal tissue response is provoked. Unfortunately, poor mechanical properties counter their advantages; they are hard and especially brittle, with low fracture toughness and a total lack of ductility or plastic deformation. Furthermore, high deposition temperatures rule out their use in delicate MEMS fabricated devices.
The literature has been dominated by polymers, and innumerable concepts have been proposed, exploiting the considerable structural versatility that polymers allow (Arshady, 2003). As well as synthetic materials, naturally occurring polymers have been reported, both as permeable membranes and as barrier coatings. However, natural polymers generally lack the dense architecture and permeability resistance of the synthetics. They tend to be highly hydrated, biodegradable and mechanically weak. What they are able to provide, however, is biological coatings over other packaging materials, and controlled permeability over active device surfaces. Materials reported include alginate (De Vos et al., 2002), chitosan (Uchegbu et al., 1998), collagen (Geiger et al., 2003), dextran (Draye et al., 1998) and hyaluronan (Vercruysse and Prestwich, 1998). Structural similarity to connective tissue polymers, in some cases, is a clear advantage and a way to possibly reduce the extent of foreign body response. Practical drawbacks to consider are that naturally sourced polymers may be immunogenic or may contain low level immunogenic components, may be difficult to obtain in pure form, properties may be source dependent and rapid degradation in the host tissue could compromise their functional value. In addition, processing demands are stringent, with both low temperatures and mild aqueous solution processing demanded if degradation or structural distortion is to be controlled.
With advances in synthetic chemistry, there has been a wider accessibility of complex, pre-modified polymeric materials. Added to silicones, polyurethanes (Coury et al., 1987; Talcott, 1995) and epoxy (Cobian et al., 1984) are poly(lactic-acid) and poly(lactic co-glycolic acid) (Athanasiou et al., 1996), poly(ethylene-glycol) (PEG) (Espadastorre and Meyerhoff, 1995), 2-hydroxyethyl methacrylate (Quinn et al., 1995), fluorocarbons, parylenes (Noordegraaf and Hull, 1997) and poly(vinyl alcohol) (PVA) (Paradossi et al., 2003; Norton et al., 2005). Some of these, such as poly(hydroxyl ethyl methacrylate), PEG and PVA, are hydrogels. An advantage of gels is that they have mechanical properties that approximate to soft tissue. However, usage will require further evaluation given their low mechanical strength. In addition, there is reduced substrate adhesion and in the case of cross-linked gels, uncertainty over the biological effects of any released cross-linking agent.
Silicone elastomers are the most widely used coatings and, unlike many other materials, their long-term performance characteristics are well established through actual, rather than simulated, use (Brindley et al., 1986). Particular capabilities include a durable dielectric insulation, diffusional resistance to contaminating solutes, and shock/vibration absorption as well as stable properties over a range of humidity values and temperatures (Wu et al., 1999). They can have optical transparency, and through surface oxygenation and modification serve as supports for protein attachment, for example, to bind active biolayers such as fibronectin to promote cell adhesion (Volcker et al.. 2001). Photochemical immobilisation technology provides a further route to surface coupling of molecules. In a study, end attached peptide generated a passivating coating that reduced protein adsorption, inhibiting surface fibroblast growth and fibrous encapsulation whilst promoting a low cytokine secretion from monocytes (DeFife et al., 1999).
Silicones can swell in aqueous solution and the repair of defects or tears is difficult. Lack of resistance to water transport is compounded if a thin barrier layer is used, which makes delamination more likely (Kinloch, 1987). Weak mechanical properties and delamination can be partly overcome by increasing thickness, but this bulking of the material is feasible only in some applications. Clinically, while silicone material properties are by no means perfect, their strong asset is the lack of toxicity, with materials having distinct regulatory approval for short and long-term implantation. In addition, as the application horizon in MEMS devices widens, silicones are a useful industry standard for coatings (Fig. 3.1).
3.1 A flexible implantable electronic circuit (converting a magnetic field generated by an external device into a DC voltage), provided with a 7 cm long elastic interconnection, embedded in silicone (Brosteaux et al., 2009).
One approach to address the problem of water transport in silicone was to entrap hydrophobic lipid inside the silicone (Wasikiewicz et al., 2011). The aim here was for the hydrophobic lipid to repel water and so reduce water transport through the silicone matrix. Some reduction in water transport was achieved, but the effect was small, despite the extreme hydrophobic nature of the lipid. This suggested alternate water transport pathways bypassing the lipid dispersion barrier were present.
Cell toxicity testing forms an important part of the early evaluation of any new material, and in this instance even a relatively small change in polymer makeup is seen to have significant effects. Another approach to silicone modification has been to combine it with hydroxyapatite to create a nanocomposite (Thein-Han et al., 2009). This modified silicone had increased mechanical strength, high extensibility, generated a good cellular response and could be an alternative to a pure silicone elastomer. Bioactivity of the nano-hydroxyapatite constituent determined the cell surface interaction and would have a practical benefit over extended periods, as is the proven case with hydroxyapatite when used for hard tissue implants. Such hybrid structures show how the functional properties of basic silicones might be extended.
Polyurethanes, with their variable hard/soft chain segmentation, can be fashioned almost to order and have found widespread application in medicine (Lamba et al., 1997; Castonguay et al., 2001). In membrane form, they have been used to protect implantable sensing devices, with epoxy resin included in one instance to give greater stability and durability (Yu et al., 2006). A more common use of polyurethanes has been as cardiovascular replacement materials, especially vascular grafts (Seifalian et al., 2003 ), a demanding environment for any polymeric structure given the combination of mechanically and chemically disruptive processes. Flexural stress in such applications over extended periods can combine with oxidative and calcification processes to compromise mechanical integrity. Polyurethanes, because of their higher tensile strength, tear and abrasion resistance, are a better material for this purpose than silicones.
Chemical tailoring and modification of the polyurethane polymer with chemical groups after it has been synthesised has allowed optimisation for targeted applications, encompassing blood contacting membranes (Seifalian et al., 2003), antibacterial surfaces, low bioadhesion surfaces (Rehman, 1996; Francolini et al;, 2010) and hydration resistant barriers (Roohpour et al., 2009a, 2010). The reactivity of polyurethane subunits has also allowed for greater versatility in structural modification than probably for any other packaging or coating material. Such modification can be made either by attaching pendent side groups or by incorporating macromolecules into the polymer backbone itself. Examples of functional modifications include grafting onto the surface (Park et al., 2001; Chuang and Masters, 2009), block copolymer formation in the bulk by copolymerisation (Queiroz et al., 2006; Roohpour et al., 2009a), creation of blends with other polymers (Lee et al., 2000), radiation modification for adding nanoparticles (Xu et al., 2003) and lipid entrapment (Roohpour et al., 2009b). In the case of block copolymers, the polarity of the support surface may induce differential migration of copolymer components of different intrinsic polarity; such differentiation of bulk component migration could serve as a basis for more subtle refinement of polymer composition at the interface (Fig. 3.2).
3.2 Attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR) of the 10% PDMS-polyether urethane (PEU) films casted on glass slide were collected. Both air interface and glass interfaces of the film were analysed to study the surface chemistry of the film at different interfaces. Results show the enrichment of PDMS at the air interface. The peaks at 1258 cm- 1 (bending of C–H in Si–CH3), 1021 cm- 1 (asymmetric stretching of Si–O–Si), 804 cm- 1 (rocking of C–H in Si–CH3) assigned to PDMS in the copolymer. The intensity of these peaks at the air interface shows a significant difference with the glass interface, which confirms the migration of PDMS segments to air interface resulting in reduced surface energy (Roohpour et al., 2009a).
A comparison between different passive polymeric coating materials has shown that nearly all materials have significant water permeability and that, though chemical composition may be varied significantly, it is not possible to eliminate the water ingress problem. In an effort to create a more water-repelling surface, materials variously comprising a single polymer, composites with hydrophobic chains (polydimethyl siloxane) or lipid (isopropyl Liquid water permeability of various polymeric coatings (1 week) myristate) have been tried. However, observations thus far indicate that even with marked alteration of a polymer matrix in this way, any influence on water transport is limited (Fig. 3.3). Therefore, while provisionally an acceptable functional compromise, polymers are unlikely to be the basis of hermetically sealed devices of the future, unless they have additional coatings using quite different materials, such as those based on inorganics. Interestingly, a polymer that contains the lipid isopropyl myristate has conferred increased cell compatibility in vitro (Fig. 3.4). Though the mechanism for this is not clear, it is possible that the presence of a surface layer of lipid served as a barrier to the release of leachables from the polymeric phase that could be toxic to cells, even if any resistance to water ingress is marginal.
3.3 Comparison between water permeability of various passive coatings based on polyurethane and silicone rubber after one week of immersion in water (Roohpour et al., 2009a, 2010; Wasikiewicz et al., 2011).
3.4 Relative cell growth at 96 hours. The data are expressed as the mean value plus or minus the standard deviation (n = 12 for the controls and 6 for the test materials). The results show that the unmodified silicone (MED-4211) used in this study, had a substantially lower biocompatibility than the biocompatible control, but the incorporation of IPM resulted in a dose-dependent increase in cell growth on the material up to a concentration of 5% (Wasikiewicz et al., 2011).
Structurally, DLC is an amorphous form of carbon with much of the structure based on sp3 carbon and a smaller proportion of sp2 rings, the latter being strained. The properties of DLC are unique, and include chemical inertness, hardness, low friction, surface smoothness, wear resistance, high resistivity and optical transparency. All can be further modified by variations in the plasma deposition process, especially in the source gas for the carbon and, in this way, changes to the ratio and strain of the hybridized carbons can be achieved. Inclusion of other components (metals, inorganics) further alters properties (Wei et al., 1999). DLC typically also contains hydrogen atoms (Robertson, 2002) and in this regard can be considered an alloy. The hydrogen atoms have a dose-dependent effect on the structural order and aromatic state, which could translate into long-term implant performance. For device encapsulation, the notable properties are that it is unreactive and, even as ultrathin films, can block water ingress and the release of corrosion products.
In contrast to the deposition of pure diamond which requires processing temperatures of 400–800°C, high-quality DLC can be deposited at room temperature, pulsed laser deposition or cathodic arc deposition being examples of techniques used. The low temperature processing conditions allow for coating of glass and plastic substrates. However, while the coating of hard surfaces (metal, glass, etc.) is a simple task, problems arise with coating flexible, polymeric substrates. The common problem is the structural heterogeneity of these surfaces and the lack of mechanical compatibility with a soft material so that, unless a tailored, flexible DLC is used, cracks and fissures can result. Introduction of lipid into a silicone elastomer enabled creation of a smooth deposition surface and defect-free DLC films were formed on silicone (Fig. 3.5) (Wasikiewicz et al., 2008). There are reports of DLC with various types of microelectronic model systems, along with the demonstration of precision patterning (Houston et al., 1995; Krauss et al., 2001; Wang et al., 2002; Peiner et al., 2007), but experience on operational microdevices is limited as yet (Smallwood et al., 2006). The reason for this is twofold: firstly, the material is relatively new in the context of medical devices; secondly, testing and evaluation is an expensive and long-term process.
Ultimately, the value of an implanted device rests on the special function it brings to the body, whether sensing or actuation (e.g. FES). To that extent, the relevant component must be able to interact with the surrounding tissues, typically to transmit an electrical potential or to sense a physical/chemical parameter. The surfaces that undertake this special function are the device’s active surfaces. The materials that coat such surface elements need to be permeable to solutes or to ions, or need themselves to be conducting. Moreover, in the case of sensing such coating might have special chemical or biochemical functions that are able to target specific molecules in the surrounding tissue. This section will describe the rather more permeable coatings that are used to ensure this interactive behaviour, variously through the high permeability of the coating, or to some special functional chemistry.
Where an implantable device is to be used for chemical or biochemical sensing, there can be problems of target molecule sensitivity, a false response due to extraneous interference and response drift due to surface coating by cellular or protein components from the surrounding tissue (biofouling). The latter has parallels with the problems facing separation membranes in biotechnology; these also suffer from surface deposition of bio-colloid during use, so compromising separation efficiency. For sensing in physiological fluids, especially blood, there is a strong tendency for proteins to be deposited, usually irreversibly, onto the external surface. The outcome is a diminished response, and in some cases near-total passivation. Interference at electrochemical devices occurs mainly due to electro-active substances that are naturally present in physiological fluids. In the case of sensors implanted in vivo, limited independent information is available to serve as a measurement reference for device performance, so the quality of coating surfaces needs to be considerably better than for in vitro use where repeat calibration allows ready correction for response drift. In the case of physical transducers, such as those for pressure and flow, no solute mass transport is involved, so the demands on the coating material is not so extreme and such coatings are more akin to those used for passive surface encapsulation, albeit quantitatively different with respect to thickness and density.
The difficulty of dealing with the problems of chemical sensing has stimulated the development of different types of specialist coatings and membrane films, which are as applicable to macro-scale devices as they are to MEMS devices. Active surface coating materials are usually involved in the functional output of a device, so ultimately the test of their performance can be linked to the quality and reliability of the signal that is generated. A variety of promising low fouling materials have been advanced as active surface coatings for implantable electronics. One group is based on biomimetic phosphorylcholine, a mimic of the external surface of the plasma membrane of the cell (Yang et al., 2000), and has been incorporated into polymer matrices to form composite structures, for example 2-methacryloyloxyethyl phosphorylcholine (Ishihara et al., 1994) and polyurethane combinations (Ishihara et al., 1996). Alternative use of water-rich hydrogels may be a way of stabilising sensing performance (Suri et al., 2003) and reducing tissue reaction; here the low mechanical strength might not be such a problem if thin, adherent coatings are used over a tougher coating material. Methods for strengthening traditional membranes, for example, through use of epoxy, have shown how it is possible to retain properties whilst shoring up mechanical strength (Yu et al., 2006). Selectivity requires specialist barrier properties at coating surfaces; here solute rejection and acceptance may be based on charge repulsion with ionomeric membranes, polar effects that alter solute partitioning and porosity change that can control transport pathways. Such membranes are likely to be used within low fouling membranes.
Work on polymers serving solely as transport barriers and electrical insulators has been extended in recent years to polymers that have inherent electrical conductive abilities. Moreover, incorporating active biomolecules for affinity sensing can broaden the scope of application, essentially becoming a part of the active transduction process in the device. These newer polymer systems present new combinations of chemical, physical and mechanical properties. Significant effort has also gone into the immobilisation of antibodies and DNA for bioaffinity (Laib and MacCraith, 2007), self assembled monolayers (SAMs) as a platform for more reproducible functionalisation (Walczak et al., 1991), reactive, conducting polymers such as poly(pyrrole) (Ramanavicius et al., 2006) for impedimetric and other modes of transduction and engineered molecular imprint polymers (MIPs) to create selective binding templates (Hillberg et al., 2005).
Smart polymers and multi-functional materials offer a further innovative approach to introducing new functionality in device coatings. Here, a material might sense some adverse local environment change and then react with it to modify it actively. In this way, not only would there be a basic packaging function, but also programmed reactivity. This might be aimed at reducing inflammatory change or to combat microbial biofilm formation and could be achieved by the release of a diffusible agent or by altering surface properties. Nano-biotechnological engineered structures offer a new range of promising properties. Thus, carbon nanotubes (CN) are a possible nano-support for biomolecule immobilisation providing unique organisation, high materials porosity and enhanced surface area for biomolecule loading. They also offer a route to device miniaturisation and, when grafted onto a surface, they could possibly improve biocompatibility and selectivity for sensing (Smart et al., 2006). It is uncertain, however, what the biological impact of free CNs in the body might be. Biohybrid nanostructures are especially likely to give new properties; CN/DNA nanoparticle conjugates have been reported, and modified conducting polymers suited to electrochemical sensing have also been reported (Yan et al., 2005; Teles and Fonseca, 2008).
A bioelement such as an enzyme or antibody may form part of a functional coating for analyte recognition. This brings into play a host of biomolecule immobilisation methods, the ultimate choice being set by the nature of the bioelement and the solid surface used to host it. It may also be necessary to incorporate a reagent into the coating with certain types of transduction mechanisms (Teles and Fonseca, 2008). An on-going problem is the meta-stability of a protein biomolecule; few can be considered to be sufficiently robust for long-term in vivo operation, moreover, a protein at the external solid–liquid interface would be antigenic and so would need an overlying barrier coating (Ramanavicius et al., 2006). Physical methods are unlikely to allow robust retention of a biomolecule. Crosslinking agents such as glutaraldehyde have therefore been popular, with inert proteins such as albumin and collagen used as the material host. The latter also need to be designed to provide optimum porosity, diffusional access and mechanical strength. A disadvantage of covalent bonding of an active protein is the loss of some bioactivity through initial conformational distortion or chemical denaturation. Microenvironmental effects on biomolecule function in a constrained solid support can also decrease the transduction signal (Amine et al., 2006 ).
Many approaches are available for coating the active surfaces of devices for sensing purposes; examples are carbon-polymer composite pastes, sol gels and lipid films. Carbon paste makes use of the conductivity of carbon, a high degree of chemical inertness, a broad range of working potentials, low electrical resistivity and a pure crystal structure; the functional outcome is low residual current and high signal-to-noise ratio for electrochemical sensing (Zhang et al., 2000). In modified carbon paste electrodes, charge transport can be rapid and the active surface area large; in addition, the surface is renewable through surface abrasion (polishing) allowing generation of a new surface from the original electrode. With electro-active or ion-exchange materials embedded in the carbon paste matrix, coating adhesion can be improved, but alternative, harder, carbon-polymer combinations are also possible (e.g. using epoxy, silicone, methacrylate, polyester, polyurethane) (Teles and Fonseca, 2008). A DNA electrochemical biosensor based on polymer modified carbon paste has been reported. The modified electrode, composed of a mixture of ion exchanger and graphite powder, was used to follow the interaction of DNA with intercalating agents (ethidium bromide and acridine orange) by following the guanine oxidation peak in the DNA (Ioannou et al., 2006). Sensitivity and reproducibility improved in comparison with conventional carbon paste. Redox hydrogels are also a possible functional coating; they allow charge transfer and are relatively biocompatible. They typically comprise cross-linked networks of redox polymers, subsequently swollen with water. The chemical inertness of redox polymers and the hydrated nature of the matrix facilitate the activity of a biomolecule, notably an enzyme. Direct molecular wiring of redox enzymes has also been possible with polymer that was conducting due to organosmium pendent groups that served as an electrical relay (Rajagopalan et al., 1996 ).
Conducting polymers are most directly applicable to bioelectrochemical sensing, since they have functional groups for biomolecule immobilisation and the capacity for electron transfer. These materials have a combination of electronic and ionic conductivity, and are usable in dry or wet state (Ravichandran et al., 2010). Figure 3.6 presents chemical structures of some conducting polymers, showing the structural versatility and the unifying feature of molecular conjugation; this is the presence of alternating double bonds in a system. Such double bonds, though written in this formal way as alternating structures, actually represent a system that is able to accommodate delocalised electrons, and therefore electron movement can be across extended distances, that is they confer conductivity. The bioanalytical application is of immense interest, and apparent biocompatibility would suggest that long-term in vivo use might be possible (Adeloju and Wallace, 1996). There are now new opportunities to incorporate nanostructures such as CNTs within these to manipulate the bulk as well as the biointerface. With inclusion of such components, a continuum of packaging materials from insulating polymers through to fully conductive coatings may emerge for different microelectrode applications. Poly(pyrrole) and poly(3,4-ethylenedioxythiophene) (PEDOT) (Guimard et al., 2007) have been particularly highlighted in the literature because of their physical, electrical and biocompatibility characteristics (Abidian et al., 2006), though PEDOT would appear to be better with respect to stability, resistance to oxidation, and conductivity. Because their response to electrical potential change can manifest as a change in conductivity and volume, these materials are of interest for neural recording by microelectrodes and have seen application for fundamental study of nerve cell signalling (Isaksson et al., 2007). Practical advantages suggested for PEDOT in neural recording are reduced signal loss and decreased noise compared with a metal electrode. As coatings, these materials significantly reduce electrode impedance but, as they are relatively thin (~ 5 μm), they can only provide a modest mechanical buffer.
To effect local pharmacological action, for example, to counter local inflammatory change, a drug could be incorporated into a permeable or biodegradable polymer mounted over an electrode. Normally, such polymers are insulators, and prevent an electrode from functioning if used as continuous coatings. One reported way around this problem is to electrospin drug-incorporated biodegradable nanofibres, encapsulate these in hydrogel followed by an electrochemical polymerisation of conducting polymer in the matrix. In this way conducting nanotubes formed within the aliginate are electrically conducting. These conducting nanostructures, when used as coatings, decrease electrode impedance and increase charge capacity density (Abidian and Martin, 2009). A less recognised attribute of conducting polymers, evident in soluble form, is their free radical scavenging ability; it is possible that they could retain this beneficial action as coatings on implanted devices (Gizdavic-Nikolaidis et al., 2004) and so confer an added functional advantage.
Electrical stimulation of conducting polymers has been used to release a number of therapeutic proteins and drugs, viz nerve growth factor (Ravichandran et al., 2010), dexamethasone (Abidian et al., 2006) and heparin (Li et al., 2005). Although this is an interesting drug delivery option, a disadvantage is that an initial burst release of the drug is likely and the hydrophobic nature of the polymer could restrict the type of bioactive agent that can be loaded and released. Nevertheless, it can be envisioned that, as payloads on microelectronic devices, conducting polymers allowing electrically actuated drug release could have application in neural and cardiovascular therapeutics (Ravichandran et al., 2010 ).
Hydrogels have been investigated as coatings for sensors. Their advantage is that, because they are highly hydrated, they allow solute movement into and out of the device for sensing (Jang et al., 2010) and they are able to transmit voltage gradients from the tissue because of their electrolyte content. A hydrogel is typically based on a hydrophilic polymer such as poly(vinyl alcohol), polyethylene oxide, polyhydroxyethyl methacrylate and poly(acrylic acid), all normally soluble in water, but either during or after synthesis, linear polymer chains can be cross-linked to form an insoluble network. The resultant structure has a high affinity for water but does not dissolve, and so is capable of absorbing water with consequent swelling of the polymer matrix. Due to their high water content, hydrogels often show high biocompatibility. In addition, water-soluble analytes are capable of diffusing quickly through the water-swollen polymer. Swelling behaviour can be easily controlled by the amount of crosslinking; a network with few crosslinks will absorb larger amounts of water leading to a high degree of swelling. Less hydrophilic monomers, incorporating hydrophobic co-monomers, or a high degree of crosslinking all serve to reduce water adsorption, so leading to firmer, more rigid gels.
The reverse of chemical and biosensing is drug release. The materials that are needed for either function have similarities. While drug release systems might be construed as unrelated to implants, two aspects make their inclusion here worthwhile. Firstly, it is highly likely that active drug release will be used as a strategy for reducing the adverse effects of the implant locally and thereby as a means for its improved biocompatibility. Thus, a drug used to reduce local tissue inflammation or to avoid the growth of microbial films would have distinct benefits. A parallel exists here with the drug eluting coronary artery stent designed to dampen the local tissue reaction and therefore restenosis of the artery (Martin and Boyle, 2011). The second point of relevance is that drug release devices will be developed that will be fully inserted in vivo for long-term use and so will constitute a type of implanted device in their own right.
Control of the period of release, dosage level and tissue targeting of a drug has huge therapeutic benefits and constitutes a major area of pharmaceutical research. A variety of technologies have been employed to exert such control for better-tailored therapies. These have ranged from chemically modifying the drug, creating new drug formulations, through to pump design for implants and other pumps able to deliver controlled drug infusions (Buchwald et al., 1980). Effort has also gone into polymeric materials and coatings as controlled drug release barriers and as drug reservoirs. One of the first operational drug delivery systems used was based on a polymer (Witt and Long, 1964). Over the years, polymeric constructs have enabled refined control of drug release and, in some cases, operated as protective phases to counter degradative agents in biological fluids, especially in intestinal fluid in the case of oral administration. Thus, the oral route is the easiest and the most convenient way of delivering a drug. However, the major challenge here is the protection of the drug from biodegradation from extreme environments presented by the digestive system, for example in the stomach where the drug is exposed to low pH and in the small intestine where it is later exposed to a high pH. As well as protecting a drug from degradation during transit, a polymer coating may be functionalised so that it is more likely to accumulate in a specific organ or desired tissue (Saigal et al., 2009). Alternative topical drug application may be used for local action (skin, nasal mucosa, conjunctiva, etc.) or the standard alternative of subcutaneous and intravenous injection may be used, but all can be optimised using polymer modified drug preparations.
Problems that arise with different delivery routes include drug solubility and stability, reproducibility of administration, local irritant effects, dose dumping with extremely high levels of drug at initial administration and side effects at non-targeted tissues and organs during the systemic transport phase. The delivery of drugs into the central nervous system (CNS) has posed a special challenge impacting on the choice and design of active agents, notably pharmacological agents such as CNS stimulants, antidepressants, antiepileptics and hypnosedatives. The obstacle in this case is the blood brain barrier, a separator of circulating blood from CNS tissue and cerebrospinal fluid (Schlosshauer and Steuer, 2002; Misra et al., 2003).
Polymer technology in various guises can address some of the above problems. Of special relevance is their barrier function to reduce side effects and toxicity by dampening rapid surges in drug level post-administration.
Structurally, the barrier may be configured as a polymeric microsphere, micelle or gel retaining a drug in the core. In this way, the barrier structure may be either a distinct shell or a single phase constituted as a continuous network. Release of a drug can occur on the basis of preset diffusion properties of the polymer matrix (Gregoriadis, 1977) which may be modified by predictable swelling in a biological fluid (Apicella et al., 1978 ) or triggered by external energy input (Langer et al., 1980). Currently the most common methods are chemical processes, for example involving polymer biodegradation. This may, for example, occur in a facile manner inwardly from the surface or occur simultaneously throughout the bulk. If the polymer is constituted from polyglycolyic acid or, say, its copolymer with lactic-acid (poly(lactic-co-glycolic acid)), there are no degradation residues as released lactate and glycolate monomer are later broken down in the body to CO2 and water. A variety of newer, more complex, materials have been designed (Lim et al., 2000) which include block copolymers and systems with a predefined balance of hydrophobic and hydrophilic domains. At the microscopic level, there is also a two-phase degradation with amorphous regions degrading before crystalline ones, another mechanism for controlling release rate.
Polymers can be considered for encapsulating whole cells as the source of the therapeutic agent. Because they are flexible in design, facilitate cell survival and could control release of a therapeutic agent, they have an advantage over many other, less versatile or fragile, encapsulating agents. These include liposomes (Samad et al., 2007), nanoporous alumina (La Flamme et al., 2005), titania nanotubes (Popat et al., 2007), porous silicon (Mashak and Rahimi, 2009) and calcium phosphate, for bone tissue delivery (Verron et al., 2010). These other systems have their own niche applications, but work on polymers is likely to remain a dominant area of effort. In historical terms, silicone was an early candidate polymer due to its biocompatibility (Mashak and Rahimi 2009) and then naturally occurring polymers were added to the repertoire (e.g. alginate, cellulose) (Coviello et al., 2007). Research then moved to synthetics such as polyanhydrides (Tabata and Langer, 1993), polyesters (Pitt et al., 1981), polyacrylic acids (Morimoto and Morisaka, 1987), poly(methyl methacrylates) (Robinson and Sampath, 1989) and polyurethanes (Touitou and Friedman, 1984), all of which have enabled new sub-sets of materials to be realised through synthetic modification; this is more difficult to achieve with less robust natural materials. Higher order organisation is also possible with some polymers. Thus, with block copolymers obtained from cross-linked combinations of hydrophilic and hydrophobic monomers, self-arrangement into shell-like structures (i.e. micelles) is possible, with hydrophilic tails aligned in the outer shell and the hydrophobic ligand directed away from the aqueous medium into the core. Such a core was also better able to retain lipophilic drug. Micelles of this type can be nanometres in diameter, and are thus suited to enclosing small ‘quanta’ of drugs down to a single drug molecule (Luisi et al., 1988). As well as permitting dispersion in the aqueous phase, a hydrophilic outer shell could be designed to protect the core and its contents from general degradative activity after administration, and then break down or attach to specific sites for drug release to give tissue specificity. Polymeric micelles containing pendant sugar group have, for example, been able to target cell membrane glyco-receptors (Vogelson, 2001).
A further class of polymeric materials that will make a significant contribution to drug delivery are the hydrogels (Qiu and Park, 2001). There is now good understanding of how these hydrophilic, cross-linked, three-dimensional (3-D) polymer networks behave in vivo and in vitro. A structural feature of importance is the nanometre size pores they contain; at this size range, access to degradative enzymes such as those from the gut can be precluded. Later drug release can then be achieved through alternative effectors: swelling and change in porosity induced by pH and temperature, exposure to magnetic or electric fields, uptake of ultrasound energy. Regardless of these mechanisms, release kinetics would be dominated by crosslink density, which could then affect the degree of control possible over the release process.
In the future, smart drug carriers are likely to be developed, capable of detecting and targeting specific disease sites, independent of the administration path and harmless to other tissues during transport to the target site. They will also provide a route to technological underpinning of personalised medicine strategies (Weinshilboum, 2003), helping to address inter-individual differences in drug responsiveness (Manasco et al., 2002 ) by matching dose to the tissue of need while avoiding side effects on other tissues.
The term ‘biocompatibility’ is, unfortunately, not a specific one, notwithstanding attempts to narrow the scope of the definition. It subsumes a range of interactions that are activated between the implant and its biological surroundings, as well as the orchestrated sequence of responses the body invokes to deal with a foreign body. There is no fully biocompatible material, with the possible exception of hydroxyapatite (HA) which, in any case is the mineral component of bone. All implants of whatever type provoke a tissue response, and this will go through all the stages of acute to chronic inflammatory change which ends in either the body encapsulating the device/ material to sequester it or to reabsorb it. Our concepts of biocompatibility, therefore, reside more in the concept of a low level response to the implant material that at least does not signify direct toxicity and local cell death. The additional feature of this assumed biocompatibility is the lack of evident cell death when cells encounter the material in in vitro tests, but this is also a measure of the lack of toxic action rather than true biocompatibility, which resides in the response of the intact organism to the material.
In the case of the active surface of an implantable device, direct contact with the biological tissue is usually important for optimum function, and this is where a tissue response needs to be considered in the greatest detail. There is a major precedent for such detailed analysis in the case of standard biomaterials (Castner and Ratner, 2002) and this has helped in better design approaches to the fashioning of direct contacting surfaces (Elbert and Hubbell, 1996). Examples of ways to generate new chemistries on surfaces, to reduce protein deposition and cell adhesion from the surrounding tissue include radiation and chemical grafting, physical coating, adsorption of active agent and organisation of surfaces with planar micro/nano-patterns and with 3-D profiling to modulate cell and tissue reactivity. (Muller et al., 2001; Sachlos et al., 2006; Sun et al., 2011).
Self-assembly refers to the automatic organisation of molecules at surfaces so as to generate molecular assembly with a well defined molecular packing and orientation. It can be used to create bulk polymeric materials with internal order and to create ordered surfaces. In the bulk, molecular mobility can allow complex, flexible molecules sufficient time for steric interactions to associate and assume lower energy states. In this self-assembly route, bulk properties relevant to drug release and binding such as porosity, anisotropy and binding groups could be engineered. At surfaces, self-assembly of small organics can serve as a model analogous to the single crystal model in metals, allowing for exploration of biology type hierarchical systems. In this way, it may be possible to evolve a more rational means of achieving bio-mimetic surfaces. The commonality across different systems that show two-dimensional (2-D) self-assembly is that simple molecular geometry, some driving force for surface binding plus lateral interactive forces between molecules will lead to a stabilised quasi-crystalline state. In this state, a defined set of chemical groups is engineered to be presented externally. The scientific roots of this area of study lie in the Langmuir-Blodgett deposition of molecules, for example, lipids and surfactants at surfaces (Kuhn, 1983).
Such controlled chemistry, giving self organised and architecturally precise surfaces, provides model surfaces in order to determine the factors that can help to reduce protein deposition at a surface and its subsequent denaturation; additionally they can be used to investigate the strength of adhesion, shape, viability and retention of inherent properties at different surfaces. Our knowledge from these structural studies has allowed the development of more biocompatible surfaces, for example, through controlled organisation of charges at surfaces, use of biomimetic structures and a comparative assessment of different types of surface modifications.
Polyethylene oxide (PEO) is a flexible, hydrophilic polymer that can be immobilised at polymeric and other surfaces. It has been found to be especially effective in reducing protein deposition, demonstrated, for example for silicone contact lenses (Thissen et al., 2010). The mechanism is unclear as yet, but is likely to reside in the masking of the surface by the PEO and the steric hindrance due to a surface hydration layer created by PEO solvation (George et al., 2009). Effectiveness is also determined by surface density and the exact organisation of PEO at the surface, however, with optimised surface organisation, surface cell adhesion can be reduced. It is, in principle, highly suited to coating all elements of a sensor surface, especially given its high permeability to solute transport at the active surface of a chemical sensor. However, long-term use is not yet a reality: oxidation and degradative mechanisms in play in vivo set a limit to operational life time.
Though PEO is primarily a model surface modifying agent in this way, its conferring of an exceptional degree of protein adhesion resistance to surfaces has allowed a better understanding of this process and the evident importance of water organisation at and near surfaces, since PEO is itself a hydrated molecule. The use of PEO analogues could well enable an improvement in the longer-term biocompatibility of surfaces.
The packaging and coating materials described in this chapter are broadly applicable to micro- and macro-scale implanted devices. To a large degree, the differences are artificial, since the key issue about micro-devices is that they need to be more protected from water than macro-devices as they generally incorporate semiconductors. A further difference is that there needs to be greater precision in where the coating is to be deposited at a micro-device since it is likely to house micro-scale active surfaces that require different coating regimens to the intervening passive elements. The coating of more complex micro-devices adds greater complexity to the encapsulation process and should be developed in parallel with device development. The requirements are for device isolation, especially if there is an electrical signal-handling element involved; the reduction of adverse tissue effects; and the avoidance of an excess inflammatory response. The well-tried polymers (silicone, polyurethane) have seen extensive investigation and although not ideal have proven to be adequate for many devices. More functionally complex capabilities, such as conductivity and affinity motifs, incorporated into devices allow their use for sensing and better biocompatibility. Increasing use is made of biological molecules for modifying surfaces for better biocompatibility and as integral components of the packaging for sensing purposes. The controlled release of active drug reagent has some relevance to this field as such systems may well be the basis of new physiologically reactive devices for mimicking particular organ functions.
The attainment of low device dimensions using MEMS fabricated platforms, and their onboard signal processing, brings to actuality the concept of the self-contained implant. For example, with signal-handling coupled with a sensing/actuation capability, an implant is able to provide closed-loop control and response without the intervention of an external operator. Importantly, it is now possible to transfer the success of current chemistries and transduction technologies from traditional formats to a micro-fabricated form. Many of the more complex device functions restricted to in vitro now become feasible in vivo. With the correct design concepts, they also look as if they are achievable over extended periods. The macro-devices that have provided both experience and a precedent for long-term use include systems such as implantable defibrillators, pacemakers, catheters and bladder stimulators as well as the more traditional stents, heart valves and musculoskeletal replacement materials from which much of our understanding of materials biocompatibility has emerged. This reality now makes it acceptable to undertake surgical or invasive procedures for device implantation. With this new horizon, it becomes all the more necessary to package the sophisticated devices so that they can function as if in a non-liquid medium. Wetting and hydration are the ultimate threats to device operation, but unless significant research effort is put into these to match, at least, some of the effort on electronic device development, it is unlikely that full clinical realisation and commercial exploitation will be achieved. Many of the candidate coating materials are already available from the biomaterial sector. However, whilst these can be a guide to driving better biocompatibility, additional effort will be necessary to satisfy the packaging needs of highly moisture sensitive systems. A subset of the challenges is the distinct way in which the responsive, functional elements of the device need to be packaged and protected as compared to its passive components. The former can be considered to require ‘open’ packaging, since some communication pathway to the bioenvironment has to be retained. With this attained, it is necessary, in parallel, to avoid water entry to the bulk of the device. These contrasting demands alone offer a challenge that will require considerable innovation. Additional work might in the future shift to designing super-hydrophobic interfaces, along with extended multilayers rather than the limited single layer barriers used currently in thick film packaging of devices.
The general development of micro-fabricated structures for undertaking complex functions sets the scene for their advancement for in vivo use. The likely challenge to more sophisticated device development for practical use, however, is not the technological problem of miniaturisation or integration of functions, but of their survival inside the body for any length of time. There will be no choice therefore but to advance the development of new materials for packaging. The mainstay will remain polymeric materials, but these are likely to be combined in future with different specialist coating layers so that one material does not have to provide all the necessary functions; specialist functions of surface biocompatibility, permeability resistance and device adhesion will be provided by different layers. One likely direction to improve biocompatibility is the development of reactive materials that are able to react to surface changes either in order to reduce surface deposition of proteins and cells or to induce the release of agents for similar outcomes. These approaches may also be harnessed for creating antimicrobial surfaces; bacterial colonisation is a special problem with any chronic implant. A starting point for such agent releasing coatings is current drug release devices. For guaranteed low surface contamination, it is moreover likely that actively changing or actively releasing surfaces will be needed in order to react to the changing tissue environment it finds itself in. So ultimately there will need to be intelligence not only in the way the device itself functions, but in the protective layers it is encased in.
The general articles and reviews in this section are designed to give the reader a background in the important domain of regulatory compliance; this must ultimately dictate the direction of materials research in this field. In addition, useful reviews that highlight, respectively, a particular and important, polymer type such as the polyurethanes and the generic challenge of the interface are highlighted here.
Brosteaux, D., Lippens, E., Cornelissen, R., Schacht, E., Carta, R., Jourand, P., Puers, R., Axisa, F., Vervust, T., Bossuyt, F., Vanfleteren, J. 31st Annual International Conference of the IEEE EMBS Minneapolis, Minnesota, USA, September 2–6, 4880. 2009.
Jiang, G., Zhou, D.D. Technology advances challenges and hermetic packaging for implantable medical devices. In: Zhou D.D., Greenbaum E., eds. Implantable Neural Prostheses 2. Springer Science + Business Media LLC; 2010:27–61. [978–0-387–98119–2].
Onuki, Y., Bhardwaj, U., Papadimitrakopoulos, F., et al. A review of the biocompatibility of implantable devices: current challenges to overcome the foreign body response. Journal of Diabetes Science and Technology. 2008; 2:1003–1015.
Apicella, A., Drioli, E., Hopfenberg, H.B., Martuscelli, E., Nicolais, L. Effects of thermal-treatment, residual solvent, and pre-swelling on thermal-properties and sorption behavior of poly(2,6-dimethyl), 1,4-phenylene-oxide polystyrene blends. Polymer Engineering and Science. 1978; 18(13):1006–1011.
Athanasiou, K.A., Niederauer, G.G., Agrawal, C.M. Sterilization, toxicity, biocompatibility and clinical applications of polylactic acid polyglycolic acid copolymers. Biomaterials. 1996; 17(2):93–102.
Belanger, M.C., Marois, Y., Roy, R., Mehri, Y., Wagner, E., Zhang, Z., King, M.W., Yang, M.J., Hahn, C., Guidoin, R. Selection of a polyurethane membrane for the manufacture of ventricles for a totally implantable artificial heart: Blood compatibility and biocompatibility studies. Artificial Organs. 2000; 24(11):879–888.
Bowen, J.A., Mellonig, J.L., Gray, J.L., Howle, H.T. Comparison of dried bone allograft and porous particulate hydroxyapatite in human periodontal osseus defects. Journal of Periodontology. 1989; 60:647–654.
Brindley, G.S., Polkey, C.E., Rushton, D.N., Cardozo, L. Sacral anterior root stimulators for bladder control in paraplegia – the 1st 50 cases. Journal of Neurology Neurosurgery and Psychiatry. 1986; 49(10):1104–1114.
Buchwald, H., Rohde, T.D., Dorman, F.D., Skakoon, J.G., Wigness, B.D., Prosl, F.R., Tucker, E.M., Rublein, T.G., Blackshear, P.J., Varco, R.L. A totally implantable drug infusion device – laboratory and clinical-experience using a model with single flow-rate and new design for modulated insulin infusion. Diabetes Care. 1980; 3(2):351–358.
Castonguay, M., Koberstein, J.T., Zhang, Z., Laroche, G. Synthesis, physicochemical and surface characteristics of polyurethanes. In: Vermette P., Griesser H.J., Laroche G., Guidoin R., eds. Biomedical Applications of Polyurethanes. Landes Bioscience: Georgetown, TX; 2001:1–22.
De Vos, P., Hoogmoed, C.G., Busscher, H.J. Chemistry and biocompatibility of alginate-PLL capsules for immunoprotection of mammalian cells. Journal of Biomedical Materials Research. 2002; 60(2):252–259.
DeFife, K.M., Shive, M.S., Hagen, K.M., Clapper, D.L., Anderson, J.M. Effects of photochemically immobilized polymer coatings on protein adsorption, cell adhesion and the foreign body reaction to silicone rubber. Journal of Biomedical Materials Research. 1999; 44:298–304.
Draye, J.P., Delaey, B., Van de Voorde, A., Van Den Bulcke, A., De Reu, B., Schacht, E. In vitro and in vivo biocompatibility of dextran dialdehyde cross-linked gelatin hydrogel films. Biomaterials. 1998; 19:1677–1687.
Espadastorre, C., Meyerhoff, M.E. Thrombogenic properties of untreated and poly(ethylene oxide)-modified polymeric matrices useful for preparing intraarterial ion-selective electrodes. Analytical Chemistry. 1995; 67:3108–3114.
Francolini, I., D’Ilario, L., Guaglianone, E., Donelli, G., Martinelli, A., Piozzi, A. Polyurethane anionomers containing metal ions with antimicrobial properties: thermal, mechanical and biological characterization. Acta Biomaterialia. 2010; 6:3482–3490.
George, P.A., Donose, B., Cooper-White, J.J. Self-assembling polystyrene-block-poly(ethyleneoxide) copolymer surface coatings: Resistance to protein and cell adhesion. Biomaterials. 2009; 30:2449–2456.
Gizdavic-Nikolaidis, M., Travas-Sejdic, J., Bowmaker, G.A., Cooney, R.P., Thompson, C., Kilmartin, P.A. The antioxidant activity of conducting polymers in biomedical applications. Current Applied Physics. 2004; 4:347–350.
Hierold, C., Clasbrummel, B., Behrend, D., Scheiter, T., Steger, M., Oppermann, K., Kapels, H., Landgraf, E., Wenzel, D., Etzrodt, D. Low power integrated pressure sensor system for medical applications. Sensors and Actuators A-Physical. 1999; 73:58–67.
Hodgins, D., Bertsch, A., Post, N., Frischholz, M., Volckaerts, B., Spensley, J., Wasikiewicz, J.M., Higgins, H., von Stetten, F., Kenney, L. Healthy aims: Development new medical implants and diagnostic equipment. IEEE Pervasive Computing. 2008; 7:14–21.
Hodgins, D., Wasikiewicz, J.M., Grahn, M.F., Paul, D., Roohpour, N., Vadgama, P., Silmon, A.M., Cousins, B., Verdon, B. Biocompatible materials development for new medical implants. Medical Device Technology. 2007; 18:32–35.
Houston, M.R., Howe, R.T., Komvopoulos, K., Maboudian, R. Diamond-like carbon films for silicon passivation in microelectromechanical devices. In: Drory M.D., Bogy D.B., Donley M.S., Field J.E., eds. Mechanical Behavior of Diamond and Other Forms of Carbon. Pittsburgh: Materials Research Society; 1995:391–402.
Ioannou, A.K., Pantazaki, A.A., Girousi, S.T.H., Millot, M.C., Vidal-Madjar, C., Voulgaropoulos, A.N. DNA biosensor based on carbon paste electrodes modified by polymer multilayer. Electroanalysis. 2006; 18:456–464.
Isaksson, J., Kjall, P., Nilsson, D., Robinson, N.D., Berggren, M., Richter-Dahlfors, A. Electronic control of Ca2 + signalling in neuronal cells using an organic electronic ion pump. Nature Materials. 2007; 6:673–679.
Ishihara, K., Nakabayashi, N., Nishida, K., Sakakida, M., Shichiri, M. New biocompatible polymer – application for implantable glucose sensor. In: Usmani A.M., Akmal N., eds. Diagnostic Biosensor Polymers. Washington: American Chemical Society; 1994:194–210.
Ishihara, K., Tanaka, S., Furukawa, N., Kurita, K., Nakabayashi, N. Improved blood compatibility of segmented polyurethanes by polymeric additives having phospholipid polar groups. 1. Molecular design of polymeric additives and their functions. Journal of Biomedical Materials Research. 1996; 32:391–399.
Kennan, J.J., Peters, Y.A., Swarthout, D.E., Owen, M.J., Namkanisorn, A., Chaudhury, M.K. Effect of saline exposure on the surface and bulk properties of medical grade silicone elastomers. Journal of Biomedical Materials Research. 1997; 36:487–497.
Kotzar, G., Freas, M., Abel, P., Fleischman, A., Roy, S., Zorman, C., Moran, J.M., Melzak, J. Evaluation of MEMS materials of construction for implantable medical devices. Biomaterials. 2002; 23:2737–2750.
Krauss, A.R., Auciello, O., Gruen, D.M., Jayatissa, A., Sumant, A., Tucek, J., Mancini, D.C., Moldovan, N., Erdemir, A., Ersoy, D., Gardos, M.N., Busmann, H.G., Meyer, E.M., Ding, M.Q. Ultrananocrystalline diamond thin films for MEMS and moving mechanical assembly devices. Diamond and Related Materials. 2001; 10:1952–1961.
La Flamme, K.E., Mor, G., Gong, D., La Tempa, T., Fusaro, V.A., Grimes, C.A., Desai, T.A. Nanoporous alumina capsules for cellular macroencapsulation: transport and biocompatibility. Diabetes Technology & Therapeutics. 2005; 7:684–694.
Li, Y.L., Neoh, K.G., Kang, E.T. Controlled release of heparin from polypyrrole-poly(vinyl alcohol) assembly by electrical stimulation. Journal of Biomedical Materials Research Part A. 2005; 73 A:171–181.
Lim, Y.B., Han, S.O., Kong, H.U., Lee, Y., Park, J.S., Jeong, B., Kim, S.W. Biodegradable polyester, poly[alpha-(4 aminobutyl)-L-glycolic acid], as a non-toxic gene carrier. Pharmaceutical Research. 2000; 17:811–816.
Najafi, K. Micropackaging technologies for integrated microsystems: Applications to MEMS and MOEMS. In: Yasaitis J.A., PerezMaher M.A., Karam J.M., eds. Micromachining and Micro-fabrication Process Technology VIII. Bellingham: Spie-Int Soc Optical Engineering; 2003:1–19.
Noordegraaf, J., Hull, H. C-SHIELD parylene allows major weight saving for EM shielding of microelectronics. In: Liu J., Jansson R., eds. Pep ‘97: the First IEEE International Symposium on Polymeric Electronics Packaging – Proceedings. New York: IEEE; 1997:189–196.
Norton, L.W., Tegnell, E., Toporek, S.S., Reichert, W.M. In vitro characterization of vascular endothelial growth factor and dexamethasone releasing hydrogels for implantable probe coatings. Biomaterials. 2005; 26:3285–3297.
Park, J.H., Lee, K.B., Kwon, I.C., Bae, Y.H. PDMS-based polyurethanes with MPEG grafts: Mechanical properties, bacterial repellency, and release behavior of rifampicin. Journal of Biomaterials Science, Polymer Edition. 2001; 12:629–645.
Pitt, C.G., Gratzl, M.M., Kimmel, G.L., Surles, J., Schindler, A. Aliphatic polyesters.2. The degradation of poly(DL-lactide), poly(epsilon-caprolactone), and their copolymers in vivo. Biomaterials. 1981; 2:215–220.
Popat, K.C., Eltgroth, M., LaTempa, T.J., Grimes, C.A., Desai, T.A. Decreased Staphylococcus epidermis adhesion and increased osteoblast functionality on antibiotic-loaded titania nanotubes. Biomaterials. 2007; 28:4880–4888.
Quinn, C.P., Pathak, C.P., Heller, A., Hubbell, J.A. Photo-cross-linked copolymers of 2-hydroxyethyl methacrylate, poly(ethylene glycol) tetra-acrylate and ethylene dimethacrylate for improving biocompatibility of biosensors. Biomaterials. 1995; 16:389–396.
Rajagopalan, R., Aoki, A., Heller, A. Effect of quaternization of the glucose oxidase ‘wiring’ redox polymer on the maximum current densities of glucose electrodes. Journal of Physical Chemistry. 1996; 100:3719–3727.
Ravichandran, R., Sundarrajan, S., Venugopal, J.R., Mukherjee, S., Ramakrishna, S. Application of conducting polymers and their issues in biomedical engineering. Journal of the Royal Society Interface. 2010; 5:559–579.
Roohpour, N., Wasikiewicz, J.M., Paul, D., Vadgama, P., Rehman, I.U. a). Synthesis and characterisation of enhanced barrier polyurethane for encapsulation of implantable medical devices. Journal of Materials Science-Materials in Medicine. 2009; 20:1803–1814.
Roohpour, N., Wasikiewicz, J., Moshaverinia, A., Paul, D., Rehman, I., Vadgama, P. Isopropyl myristate-modified polyether-urethane coatings as protective barriers for implantable medical devices. Materials. 2009; 2:719–733.
Roohpour, N., Wasikiewicz, J.M., Moshaverinia, A., Paul, D., Grahn, M.F., Rehman, I.U., Vadgama, P. Polyurethane membranes modified with isopropyl myristate as a potential candidate for encapsulating electronic implants: A study of biocompatibility and water permeability. Polymers. 2010; 2:102–119.
Schlosshauer, B., Steuer, H. Comparative anatomy, physiology and in vitro models of the blood–brain and blood-retina barrier. Current Medicinal Chemistry – Central Nervous System Agents. 2002; 2:175–186.
Sachlos, E., Gotora, D., Czernuszka, J.T. Collagen scaffolds reinforced with biomimetic composite nano-sized carbonate-substituted hydroxyapatite crystals and shaped by rapid prototyping of contain internal nanochannels. Tissue Engineering. 2006; 12:2479–2487.
Shimizu, T., Goda, T., Minoura, N., Takai, M., Ishihara, K. Super-hydrophilic silicone hydrogels with interpenetrating poly(2-methacryloyloxyethyl phosphorylcholine) networks. Biomaterials. 2010; 31(12):3274–3280.
Stan, G.E., Marcov, D.A., Popa, A.C., Husanu, M.A. Polymer-like and diamond-like carbon coatings prepared by Rf-Pecvd for biomedical applications. Digest Journal of Nanomaterials and Biostructures. 2010; 5:705–718.
Thein-Han, W.W., Shah, J., Misra, R.D.K. Superior in vitro biological response and mechanical properties of an implantable nanostructured biomaterial: Nanohydroxyapatite-silicone rubber composite. Acta Biomaterialia. 2009; 5:2668–2679.
Thil, M.A., Gerard, B., Jarvis, J.C., Delbeke, J. Two-way communication for programming and measurement in a miniature implantable stimulator. Medical & Biological Engineering & Computing. 2005; 43:528–534.
Thissen, H., Gengenbach, T., du Toit, R., Sweeney, D.F., Kingshott, P., Griesser, H.J., Meagher, L. Clinical observations of biofouling on PEO coated silicone hydrogel contact lenses. Biomaterials. 2010; 31:5510–5519.
Uchegbu, I.F., Schatzlein, A.G., Tetley, L., Gray, A.I., Sludden, J., Siddique, S., Mosha, E. Polymeric chitosan-based vesicles for drug delivery. Journal of Pharmacy and Pharmacology. 1998; 50:453–458.
Volcker, N., Klee, D., Hocker, H., Langefeld, S. Functionalization of silicone rubber for the covalent immobilization of fibronectin. Journal of Materials Science: Materials in Medicine. 2001; 12:111–119.
Walczak, M.M., Popenoe, D.D., Deinhammer, R.S., Lamp, B.D., Chung, C.K., Porter, M.D. Reductive desorption of alkanethiolate monolayers at gold – a measure of surface coverage. Langmuir. 1991; 7:2687–2693.
Wasikiewicz, J.M., Paul, D., Roohpour, N., Vaghela, J., Grahn, M.F., Vadgama, P. Isopropyl myristate modified silicone as a potential new encapsulating material for implantable devices. Journal of Applied Polymer Science. 2011; 119:2917–2924.
Wasikiewicz, J.M., Roohpour, N., Paul, D., Grahn, M., Ateh, D., Rehman, I., Vadgama, P. Polymeric barrier membranes for device packaging, diffusive control and biocompatibility. Applied Surface Science. 2008; 255:340–343.
Wei, Q., Narayan, R.J., Sharma, A.K., Sankar, J., Narayan, J. Preparation and mechanical properties of composite diamond-like carbon thin films. Journal of Vacuum Science & Technology A-Vacuum Surfaces and Films. 1999; 17:3406–3414.
Wu, J.L., Pike, R.T., Wong, C.P., Interface-adhesion-enhanced bi-layer conformal coating for avionics application. International Symposium on Advanced Packaging Materials: Processes, Properties and Interfaces, Proceedings. New York: IEEE. 1999:302–310.
Yu, B., Long, N., Moussy, Y., Moussy, F. A long-term flexible minimally-invasive implantable glucose biosensor based on an epoxy-enhanced polyurethane membrane. Biosensors and Bioelectronics. 2006; 21:2275–2282.
Ziaie, B., Von Arx, J.A., Dokmeci, M.R., Najafi, K. A hermetic glass-silicon micropackage with high-density on-chip feedthroughs for sensors and actuators. Journal of Microelectromechanical Systems. 1996; 5:166–179.