Biotribocorrosion of Implants

Fig. 19.1

Schematic representation of a metallic surface: (a) ideal surface, (b) technical surface obtained by metal processing, (c) technical surface with oxide layer. The miscrostructure (grains) is typical for AISI 316 L stainless steel observed at 500 ×

Another difference between the ideal and technical surfaces is the deformation of microstructure. Whereas the microstructure just under the ideal surface is identical with that of the bulk, technical surface is associated with a layer in which the microstructure is deformed. In the example shown in Fig. 19.1, the grains of the bulk steel are of regular shape, whereas technical surface is terminated by a layer of deformed grains in which mechanical properties and electrochemical potentials may deviate from those of bulk values. The deformation results from manufacturing processes such as cutting, grinding or polishing typically employed in the fabrication of metallic implants. It may also arise as product of friction and wear. The microstructure of the subsurface layer has been discussed as crucial in the biomedical applicability of CoCrMo alloys [2].

Further attribute of a technical surface is its interaction with the environment. Especially gasses and liquids (electrolytes) tend to react with metallic surfaces producing adsorbed or passive layers. In the example of Fig. 19.1c a passive layer corresponding with metal oxides typically formed by intereaction with an aqueous environment is shown. This layer might be porous consisting in submicron- or nanocrystals or be a continuous film. The later efficiently separates the base metal from further interaction with the environment. It should be noted that biological environment of a living organism is not constant in time. The variation of temperature, chemical composition and condition of flow may affect the structure of the passive film. Also, mechanical properties of both the film as well as the bulk material might be compromised through the Rehbinder (or Rebinder) effect [3], which considers the role of adsorbed surface active molecules in the propagation of cracks.

The interface described above is further modified when the surface is set in motion, especially when both shear and normal forces are involved. In this case, the surface layer can be compromised and the elements of the environment –small molecules, debris particles, etc. – be included to the solid interface. Such a condition is referred to as tribolayer and is a commonplace in metal on metal hip replacements. It has been shown to be organometallic in nature and containing a number of embedded particles, which are smoother and smaller than the initial debris, e.g. [4]. The electrochemical nature of the tribolayer, and thus its contribution to overall surface degradation, also varies in function of the actual surface condition as shown on the example of CoCrMo alloys [5].

Corrosion Damage of Implants

Unlike mechanical damage, damage through (electro)chemical interaction with the environment cannot be avoided by removing the primary cause. In case of metals the contribution of electrons is significant and the entire process, necessarily detrimental in nature, is referred to as corrosion.

Corrosion takes place at the interface with external medium and involves exchange of both charge and mass. The metal is said to be an anode when electrons are being extracted from it, whereas it becomes a cathode by accepting electrons. At the kinetic equilibrium, i.e. with no external source of current applied, the loss of electrons by a metal (anode) is essentially accompanied by loss of metal ions (metallic dissolution). On the other hand, the cathodic reaction can involve electro-active species present in the environment, such as the oxygen reduction reaction in neutral and alkaline media as well as hydrogen ion reduction or hydrogen evolution ion of acid environments. In principle, metal ion reduction and metal deposition can be also take place on the surface [6]. Thus, corrosion can be viewed as a spontaneous process of returning metal to its mineral (oxidized) state, which is given by the laws of thermodynamics, in particular, expressed by free Gibbs energy (Eq. 19.1). Since, by definition, electrons are involved in the reaction, it is convenient to express the energy of each ion in terms of electric potential (Eq. 19.2):

$\varDelta G=\varDelta {G}^{\otimes }+RT{\displaystyle \sum}_{i=1}^k \ln {a}_i^{v_i}$

(19.1)

$\varDelta G=- zF\varDelta E$

(19.2)

where ΔG, $\varDelta {G}^{\otimes },$ R, T, a _i, v _i, z, F and ΔE are the free Gibbs energy, free Gibbs energy at standard conditions, gas constant, temperature, activity coefficient of species i, stoichiometric coefficient of species i, charge number, Faraday constant and the electric potential, respectively. Although the free Gibbs energy associated with an electrochemical reaction is determinant for its occurrence, in practice, it is the rate of the reaction rather than shear energy differences that determine the feasibility of a metal for particular environment. In practice, thermodynamics predicts the likelihood of corrosion process, which can be determined by the corrosion potential (E _corr) using electrochemical techniques such as the variation of potential as a function of immersion time with no current applied, i.e. at open circuit potential (OCP). However, it does not provide information about corrosion rate [7]. In general, an electrochemical system is given by the Butler-Volmer equation, which corresponds to a relationship between the potential and current, as is shown in the Eq. 19.3

$j={j}_{\mathrm{o}}\left[ \exp\ \left(-\frac{\alpha zF\eta}{RT}\right)- \exp \left(\frac{\left(1-\alpha \right)zF\eta }{RT}\right)\right]$

(19.3)

where j, j _o, α, and η are current density, exchange current density, symmetry factor and overpotential, respectively. Figure 19.2 shows the behavior predicted by Eq. 19.3. The solid curve shows the actual total current for a large overpotential, which is the sum of the cathodic (j _c) and anodic (j _a) current densities. For large negative overpotentials, the anodic component is negligible and the total current is essentially cathodic current. On the other hand, at large positive overpotentials, the cathodic component is negligible and the total current is essentially anodic. In practice, the corrosion process occurs when total current is zero. A typical polarization diagram of an implant metal (316L, Co-alloy or Ti-alloy) is shown schematically in Fig. 19.2.

Fig. 19.2

Schematic current density versus potential curve of an electrochemical system

Stability of the passive layer is a valid concern because corrosion resistance of the majority of implant alloys relies on the formation of a passive layer on the surface that provide an effective barrier to electron and ion transport, playing a very important role in the long-term clinical success of implants. However, a depassivation process can also take place producing increase in the corrosion rate. In particular, in crevices or other types of occluded such as local environments due to cell activity, where local pH changes due to chemical reactions may also occur [8].

The fraction of the stable and unstable oxide species determines the overall stability and dissolution behavior of the passive film. For instance, the presence of TiO₂ on the surface of CP-Ti provides the resistance to activation upon acidification. However, all aluminum oxides are soluble in acidic solution so that Al-containing Ti-alloys are less corrosion resistant.

The composition of the oxide film formed in air usually differs from that of a passive film formed in a solution. This fact is used in improving corrosion resistance by growing the passive layer of desired composition (see later discussion).

The greatest concern for clinical praxis (apart from biomechanical compatibility) is the release of metal ions and corrosion products that may cause inflammation, allergic reactions and other not desired effects. Although corrosion is mostly associated with materials damage and high corrosion resistance is generally aspired, corrosion-related degradation can also be exploited. An example are biodegradable Mg-alloys which are currently evaluated for biomedical applications [8].

The typical materials used for surgical implants are 316L stainless steels, CoCr alloys, CoCrMo alloys, commercially pure Ti (CP-Ti) and Ti-based alloys. The 316L, CoCr and CoCrMo alloys have an elastic modulus close to 200 GPa, which is much higher than that of compact bone tissue being in the range from 2 to 30 GPa [9–11]. On the other hand, CP-Ti and Ti-based alloys have been readily employed due to their lower elastic modulus, lower density, good biocompatibility and, above all, corrosion resistance [9–11]. The excellent corrosion resistance that CP-Ti and Ti-alloys demonstrate in a variety of media is attributed to the passive behavior associated with formation of Ti-oxide on the surface and owed to their thermodynamic properties described by Eq. 19.1. However, the passive layer might be compromised by depassivation which immediately results in increase of the corrosion rate. Should this occur within the human body, local changes of pH due to hydrolysis of metal ions may result in inflammation of the nearby tissues. One of the most common causes of depassivation in absence of motion is the formation of crevices or other, areas of differential aeration. Crevice corrosion of Ti-alloy implant might be associated with cementation and the local changes of pH along the corroded stem were suggested to induce aseptic loosening [9]. It should be noted that the Ti-alloys of first generation implants employed Al, V, Ni and Co as alloying elements in order to improve the mechanical properties. However, corrosion of the alloys can also produce metallic dissolution of the alloying elements, which can cause an undesirable toxic effect in the human body. The alloys of the second generation were developed to contain only the least toxic elements [9, 11–18], although the TiAlV and TiAlNb are still used. In general, these alloys have demonstrated high corrosion resistance in different media, such as artificial saliva (Ringer’s solution) or physiological saline solution (0.9 % NaCl) [19, 20].

Wear Damage of Implants

The phenomena occurring between two surfaces in relative motion are the domain of tribology and can be roughly classified as friction, lubrication or wear. Whereas friction is the force resisting the motion, lubrication and wear affect friction by preservation and destruction of the surface layer, respectively. The individual processes can be of mechanical, thermal and physiochemical type or their combinations as summarized in Fig. 19.3. In a medical implant under normal conditions of use the following processes are of little importance:

Fig. 19.3

Processes that may take place in surface layer when in relative motion aganist a counter surface

Temperature gradients,– although local generation of heat is unavoidable and has been shown to produce temperatures above 46°C [21] possibly leading to apoptosis of osteoblasts [22], these temperatures are unlikely to affect other processes through gradients;
Surface melting,– the melting temperature of all the biomedical alloys is above 1000°C, which is unlikely to be reached;
Evaporation,– evaporation would require even higher temperatures and/or lower pressures, which are unexpected within human body;
Contact welding,– this process requires localized temperatures and/or high pressures, which are impossibly reached through friction of an articulation.

The intensity of all the remaining processes is lower than that observed during manufacturing, where surface is removed intentionally; however unlike manufacturing there is a high degree of inhomogeneity associated with their spatial distribution, thus local concentration of forces may indeed lead to catastrophic effects. A review of all the mechanisms with regard to natural and artificial joints can be found in Ref. [23], whereas here only the primary concepts, necessary for introducing the corrosion enhanced damage, are mentioned.

Friction and Lubrication

Friction being the force resisting motion of two bodies in contact is a straightforward consequence of the tribological contact. Depending on the presence of a liquid between the surfaces and the extent of the normal force dry friction, lubricated friction and mixed friction can be distinguished. Of course, dry friction is not expected in a functioning articulation but might be the case in adverse conditions of disease when too little of the synovial fluid is provided as well as under high loads. The dissipation of energy between sliding bodies have been known for centuries and the classic work of Da Vinci and Amonton provided the four basic laws: (i) there is a proportionality between the maximum tangential force before sliding and the normal force when a static body is subjected to increasing tangential load; (ii) the tangential friction force is proportional to the normal force in sliding; (iii) friction force is independent of the apparent contact area; (iv) friction force is independent of the sliding speed.

The quantity of friction is determined as proportion between the force of friction and the force acting normal the sliding interface. The proportion is the coefficient of friction and it is said to be static (μ _s) when the force of friction is measured just before sliding begins or kinetic (μ _k) when the force of friction is measured during sliding. Although the definition is simple and straightforward for sliding in one direction, the complex geometry of natural joints requires taking into account 3-axis cyclic forces acting along with 3-axis cyclic frictional moments [24–26].

The friction coefficient in natural joints varies between 0.001 and 0.03 [25] depending on the actual loading conditions and type of lubrication. Artificial joints can fairly reproduce these values [26], however at compromised lubrication conditions the increase of friction might lead to considerable wear [27–30]. Association of friction coefficient with squeaking in ceramic-on-ceramic couplings have also been reported [28].

The actual thickness of the lubrication film during motion is predicted by a hydrodynamic theory which takes into account the effect of normal force acting on the element. In addition, viscosity of the lubricant changes in these dynamic conditions making the prediction of friction coefficient a matter of elastohydrodynamic lubrication theory (EHL). In healthy natural articulations tens of microns of synovial fluid separates the opposing surfaces and the predominant type of friction is lubricated friction. Only in extreme cases of overload boundary friction may occur, however, it is typically of short duration and causes no damage. A similar situation has been reported for artificial joints, where high friction coefficients typically associated with dry friction have been found in-vivo [29]. The thickness of lubricating film has been shown to be sensitive to the load to a greater extent than predicted by the EHL models [30].

It should be noted, that low friction is typically sought after for the moving surfaces of artificial joints but large friction is also desired for the surfaces that are not supposed to move respectively, e.g. interface between an implant and bone or implant and bone cement. Low amplitude motion at these surfaces leads to a tribological damage referred to as fretting corrosion. The damage itself results from destruction of the passive oxide layer that brings enhanced corrosion through direct exposure of the metallic surface to a corrosive environment. In addition, particles of polymer and metal oxide form debris that my further accelerate the process. Evolution of fretting damage in a stainless steel/PMMA couple at different electrochemical potential has been shown to be accompanied by the increase of friction coefficient [27]. The root cause of fretting might be related with the micro-damage that bone tissue experiences under cyclic compressive loads associated with physical activities [31], although the cement itself is typically more at risk of fatigue failure than bone [32].

Wear

All bodies subject to friction are susceptible to wear, including the natural healthy articulation, where the loss of mass is quickly replenished and the products of wear, the debris, are resorbed by the body with no side-effect. This autoregulation system is evidently missing in artificial joints and to provide an efficient method of prevention and/or mitigation requires understanding of the involved wear mechanisms. These mechanisms can be classified as abrasive, adhesive and fatigue, according to ASTM terminology [33].

Abrasive wear is due to hard particles or hard protuberances forced against and moving along a solid surface. Within this category it is commonly distinguished between the mode of three-body and the mode of two-body wear (Fig. 19.4). In the first case, the all the three elements: surface of the implant, counter surface and debris particles act in way that debris particles are allowed both to slide and roll. Whereas, although two-body wear also includes the three here the debris particles are only allowed to slide. The two-body case can also be thought of as debris particles fixed to the counter surfaces and thus mimicking its sliding motion. Due to limited distance between a typical sliding surface of and implant and the counter surface two-body wear mechanism is mostly the case; however, evidence of the three-body mechanism has also been reported [34].

Fig. 19.4

Schematic representation of abrasive wear: (a) three-body mode of wear with debris particles sliding and rolling, (b) two-body mode of wear with debris particles allowed only to slide

Adhesive wear is due to localized bonding between contacting solid surfaces leading to material transfer between the two surfaces or loss from either surface. Unlike abrasive wear, the material removed from the worn surface is transferred to the counter surface rather than to the medium, thus, no debris is formed. The mass transfer takes place from softer to the harder surface, example from polyethylene to CoCrMo [35], or between surfaces of the same hardness, example metal-on-metal [36] or ceramic-on-ceramic [37].

Fatigue wear is due to (micro)fracture arising from material fatigue. Repetitive loads, normal to the surface, that in no cycle exceed yield strength of the material, do cause cumulative microscopic damage that eventually sums up to a fracture. Because this damage is localized near surface, the cracks are generally parallel with surface resulting in flake-like debris rather that bulk failure associated with propagation of a vertical crack. Plastic deformation of surface layer as well as presence of the micro-scale cracks have been shown in HC CoCr femoral heads [38].

Independent of the actual damage mechanism, the amount of wear experienced by the material can be quantified by Archards equation (Eq. 19.4) in terms of wear volume V _W, i.e. loss of mass occurring under specific load L on a specific sliding distance S:

${V}_W={k}_w\frac{LS}{H}$

(19.4)

where k _wand H are dimensionless wear coefficient and measure of hardness of the surface, respectively. The equation expresses the proportionality of wear volume to the work done by friction forces and may have different variants depending on how volume and sliding distance are expressed. The wear volume is typically expressed in the units of volume, e.g. [mm³], or mass [g] determined for specified sliding path or number of cycles. It can be also normalized and expressed as wear rate, for instance [mm³/year], or as a linear wear rate (depth of penetration), in [μm/year]. In case of adhesive wear, the coefficient k _wcan be interpreted as the proportion of asperity contacts resulting in wear, thus never exceeding the value of one. An example wear rate of CoCr MoM determined from retrieved implants is in the order of 5 μm/year [39]. Due to variability of the wear coefficient associated with the changing physiological conditions numerical prediction of implant wear is difficult even when complex geometry of the triboelement is accounted for [40].

A straightforward consequence of Eq. 19.4 is that an optimum material for a biotribological application must be both hard and produce low friction coefficient in the system. The use of material couplings of low friction coefficient has been introduced already in the beginnings of joint replacement technology; however, these tend to fail in the long run due to wear of the softer material.

Corrosion and Wear Combined

A tribological element working in a corrosive environment inevitably experiences tribocorrosion. The term is sometimes restricted to situations when synergy between corrosion and wear damage takes place, i.e. the total degradation rate is larger (or less often smaller) than a simple sum of the two individual effects. Although the individual phenomena of corrosion and wear are relatively well described a detailed understanding of tribocorrosion is still missing because of the experimental difficulties in following the multitude of simultaneously undergoing reactions in which metastable products and reactants are often included. Tribocorrosive degradation affects a number of engineering systems like pumps, propellers, impellers, valves, mill liners etc. and is commonly described phenomenologically by the total material loss V expressed by Eq. 19.5:

$V = {V}_W + {V}_C + {S}_{WC}$

(19.5)

where V _W, V _Cand S _WCare the mass loss due to wear in the absence of corrosion, mass loss due to corrosion in the absence of wear and the synergistic interaction term, respectively. Guidelines for experimental determination of S _WCare given by the ASTM standard [41].

The synergy between corrosion and wear (S _WC) can be generally separated into wear enhanced corrosion, i.e. the increase of corrosion rate than can be attributed to wear, and corrosion enhanced wear, i.e. the increase of wear that can be attributed to corrosion.

Wear enhanced corrosion is most pronounced in the active-passive material. Since corrosion resistance of these materials relies on the presence of a thin passive layer (1–10 nm) any instance of its discontinuity results in an immediate onset of localized corrosion. Because all the wear mechanisms involve damage of the passive layer high current corrosion following the Bulter-Volmer kinetics (Eq. 19.3) is the immediate consequence at the places where bare metal is exposed to the environment. This situation is depicted schematically in Fig. 19.5 as case “a”. Although the layer might recover relatively fast in the process of repassivation there exists a time in which the metal follows an accelerated corrosion kinetics prior establishing its steady-state passive behavior depicted in Fig. 19.5 as case “c”. In case of CoCrMo alloys repassivation takes about 5 s in 0.9 % NaCl solution [42] and it seems to be much slower in case of Ti-alloys [43]. Both kinetics of oxide growth and mechanical as well as topological properties of the debris present in lubricant both have an effect in this wear accelerated corrosion. In consequence, changing friction conditions during different stages of the walking cycle results in varying rate of corrosion [38]. Further, adsorption of organic molecules such as serum albumin may reduce the repassivation kinetics as shown for Ti-alloys [44], whereas osteoblastic cells seeded over Ti6Al4V alloy was shown to prevent damage to the passive film [45]. It should be noted that in absence of a passive film, the localized deformation of asperities induced by wear may also accelerate the rate of corrosion. In this case, the shape of curve “a” in Fig. 19.5 would not change while wear damage progresses. Rather the entire curve would be shifted towards higher values of electric current.

Fig. 19.5

Schematic explanation of the effect of removing passive layer on the polarization curve of the metal. The letters indicate curves associated with particular instance during recovery of the passive layer after scratching

In case of coated alloys, for instance CrN-coated stainless steel, the galvanic interaction between the different alloys may further contribute to corrosion-wear [46]. Galvanic interaction between the sliding/rolling surfaces may lead to increase of roughness and thus increase of friction. Also, the plastic deformation of coating resulting in micro-cracks or other coating defects may enable the galvanic coupling between the coating material and the substrate. An example of this mechanism is the blistering of TiN coated Ti-alloys [47].

Corrosion enhanced wear is the product of several alterations that corrosion may introduce to the triboelement. The most direct is the change of surface topology associated with dissolution and oxidation of the metallic surface. Since free corrosion (with no external potential applied) distributes anodic and cathodic processes over the same surface, inhomogeneity arises that in general results in increased roughness, which in turn increases the coefficient of friction (Fig. 19.6). A related mechanism is that of localized modification of material properties of the substrate that results in increased wear rate. The amount of corrosion enhanced wear can be determined by polarizing the metal cathodically, which disables the anodic dissolution by providing the electrons externally. Using this method an example of 22–32 % contribution of corrosion to wear enhancement has been shown for a CoCrMo alloy [48].

Fig. 19.6

Schematic representation of the relationship between corrosion, surface roughness and the coefficient of friction

Further, when the oxide formed by corrosion is much harder than the metal itself, there might be a significant acceleration of wear, even when sliding against a much softer counterface. The soft counterface can provide a resilient bed for abrasive particles of oxide, leading to rapid wear of the metallic surface through the two-body abrasion mechanism (Fig. 19.4b). An example of this corrosion-abrasion mechanism are the Ti-Al-V alloys experiencing rapid wear even in contact with a soft UHMWPE counter-face [49].

In order to estimate apriori the rate of surface degradation various mechanisms and combination thereof must be taken into account. An example of such combined model considering mechanical wear by plastic deformation at asperities, wear accelerated corrosion in addition to the hydrodynamic lubrication theory was shown to work for a CoCrMo sliding tribocorrosion contact [50]. The contribution of wear enhanced corrosion to the total material loss measured in a hip simulator for a CoCrMo MoM couple is about 13 % [51]. However, the relative importance of different wear mechanisms may vary when the tribological conditions change.

One possibility of visualizing the interdependence of tribocorrosion mechanism on the distinct environmental parameters is through mapping [52]. The idea of a tribocorrosion map, first introduced for technical problems and then adapted to biological systems [53], is that of representing the intensity of material waste in function of two process parameters. The analogy with a topographic map is that the contour lines of the terrain (constant elevation over the sea level) are the analogy of constant rate of waste, whereas the spatial coordinates correspond with variables modifying the waste mechanism. The variables that are relevant to consider for biotribocorrosion are velocity of sliding, applied load as well as composition of the lubricating film [54].

In order to construct the map, the contribution of each variable to the total material loss (Eq. 19.5) must be determined separately. The total mass loss rate (K), hereafter considered in the normalized units of mass per area per time, i.e. kg · m⁻²s⁻¹, is separated the individual contributions considering the interactions between corrosion and wear:

$K = {K}_W + {K}_C + {K_W}_{-C} + {K_C}_{-W}$

(19.6)

where K _W, K _C, K _W−C, K _C−Ware the mass loss rates due to wear only, corrosion only, wear enhanced corrosion and corrosion enhanced wear, respectively. The experimental determination of each contribution relies on suppressing the contribution of one of the mechanisms. For instance, cathodic polarization of the metal surface (application of potential negative with respect to the corrosion potential E_corr) disables the anodic dissolution of the metal and thus prevents corrosion. With the mass loss rates determined experimentally, two basic types of map can be constructed.

Intensity map, where each contour line corresponds with a constant mass loss rate. The regions of high, medium and low wastage can be defined arbitrary, depending on the utility of the map. In the example shown in Fig. 19.7a, the regimes of mass loss rate are the following:

Fig. 19.7

An example of tribocorrosion map obtained by micro-abrasion of flat CoCrMo alloy sliding against a rotating UHMWPE ball in an FCS solution: (a) intensity view, and (b) mechanistic view. (Adapted from K. Sadiq, M. M. Stack, and R. a. Black, “Wear mapping of CoCrMo alloy in simulated bio-tribocorrosion conditions of a hip prosthesis bearing in calf serum solution,” Mater. Sci. Eng. C, vol. 49, pp. 452–462, 2015, with permission.)

Low: $K\ \le\ {10}^{-7}$ [g · cm⁻²min⁻¹]
Medium: ${10}^{-7} < K\ \le\ 20 \times {10}^{-7}$ [g · cm⁻²min⁻¹]
High: $20 \times {10}^{-7} < K$ [g · cm⁻²min⁻¹]

Mechanistic map, where each contour line correspond with transition from one dominant corrosion-wear mechanism to another. In the example shown in Fig. 19.7b, the following regimes were defined:

Micro-abrasion dominated: $\frac{K_C+{K}_{W-C}}{K_W+{K}_{C-W}}\kern0.5em \le \kern0.5em 0.1$
Micro-abrasion/Passivation: $0.1\kern0.5em <\kern0.5em \frac{K_C+{K}_{W-C}}{K_W+{K}_{C-W}}\kern0.5em \le \kern0.5em 1$
Passivation/Micro-abrasion: $1\kern0.5em <\kern0.5em \frac{K_C+{K}_{W-C}}{K_W+{K}_{C-W}}\kern0.5em \le \kern0.5em 10$
Passivation dominated: $10\kern0.5em <\kern0.5em \frac{K_C+{K}_{W-C}}{K_W+{K}_{C-W}}$

In this example, the term wear was replaced with a more specific term abrasion and the term corrosion was replaced with passivation which is the specific corrosive mechanism interacting with abrasion.

Although mapping corrosion-wear makes a significant contribution to understanding the dynamic changes between the dominating mechanisms, comparison of literature data on rates of mass loss is difficult because loading and kinematic conditions are taken from different sources or do no necessary correspond with those of a real articulation in motion. The limitations underline the difficulty of a realistic theoretical description of the hip implant tribological behavior, increased by the complex model solution.

Clinical Implications of Biotribocorrosion

All the possible adverse effects of tribocorrosion on the functioning of the biological system can attributed to the very processes associated with a working triboelement (Fig. 19.3). Although the wear rate of most implant bearings is fairly low as compared with technical systems, the amount of metal and other tribocorrosion products released to the body can be considerable and the side effects associated with wear debris are considered a limiting factor in introducing MoM bearings in disc arthroplasty [55]. Although, the wear volume alone is often insufficient as an indicator of potential danger because biological reactions may also be sensitive to the number, size as well as size distribution of the wear debris. For instance, nanoscale metallic particles are less likely to produce a reaction than less numerous but larger particles of polyethylene.

Due to complex nature of the processes taking place in a functioning articulation, no integrated model of the effect of tribocorrosion is yet available. From the mechanistic point of view, the adverse reactions can be attributed to either wear enhanced corrosion (including corrosion alone) and/or corrosion enhanced wear (including wear alone). In the first case, the effects are analogous with the previously described effects of corrosion alone. In the second case, the root cause of undesired reactions is the presence of debris. Motion of debris particles can be described by employing the EHL theory of lubrication as shown for an artificial hip joint [56]. However, the particles do not necessary remain encapsulated and may affect other organs.

The adverse effects attributed to tribocorrosion debris where shown to include:

Increased levels of metal ions in blood due to further corrosion of the metallic debris. Just like in case of corrosion alone, these metal ions may be distributed and accumulated in various organs. The accumulation might also take place for metallic particles, for instance in the para-aortic lymph nodes, the liver and spleen leading to granulomas [57].
Hypersensitivity reaction, which may be immediate, e.g. [58], or delayed, e.g. [59], and which used to be associated with aseptic lymphocytic vasculitis-associated lesions [59].
Pseudotumors, soft-tissue mass associated with the implant neither, however mostly resulting in pain [60]. The prevalence of pseudotumors does not necessarily correlate with elevated metal ion levels after MoM [61].
Osteolysis resulting from the biological reaction to wear debris [62, 63].
Lymphocyte proliferation due to contact with nanoparticle debris [64].
Aseptic loosening due to adverse response to wear debris, e.g. [29, 57].
Mechanical fracture of the implant due to surface generated cracks, e.g. [65].
Infection due to interference of wear debris with the immune system and or inhibition or acceleration of bacterial growth [66].

The collection of variety of symptoms including inflammatory masses (pseudo-tumor), fluid collections, localized soft tissue necrosis, and histological evidence of a dense lymphocytic chronic inflammatory infiltrate is often referred to as adverse local tissue response (ALTR) and has been identified as one of the four failure modes in modular femoral stems by the review of Esposito et al. [67]. However, the local tissue reaction is generally mild and the effect of continued elevated levels of metal ions not known [57].

It should be kept in mind that the tissue reaction around implant surface is multifactorial in nature, whereas the parameters associated with functioning of the triboelement are not necessarily the most important. Factors like individual patient immunoreactivity [68] or even gender [69] should also be taken into account for predicting the long term outcome of a arthroplasty because the correlation between the necessity of revision, metal release, and material category is not straightforward [70–72].

Methods of Testing

Tribocorrosion of biological systems is difficult to test due to complexity of surface structure and large number of the involved processes. The most accurate prediction of triboelement performance is obtained from testing the element itself; however, laboratory testing is unavoidable for the obvious reason of limiting experiments on living organisms.

Due to the corrosion and wear synergy, experimental separation as well as combination of corrosion and wear is necessary in order to determine how each system parameter affects the tribological performance. A typical sequence of testing goes from laboratory, through simulated joints to in-vivo test, whereas the complexity of testing conditions increases. In general, when a new material is developed it should first meet certain standard in a corrosion test before being considered for tribological testing.

The summary of the testing methods is shown in Table 19.1 gives the overview of applicability with respect to the mechanism considered and type of test. In the following, each method is explained briefly.

Table 19.1

Summary of testing methodologies for mechanistic studies of tribocorrosion of biomedical systems