In-vitro studies on cells and tissues in tribocorrosion processes: A systematic scoping review

Abstract

Tribocorrosion of implants has been widely addressed in the orthopedic and dental research fields. This study is a systematic scoping review about research methods that combine tribocorrosion tests with cells/tissues cultures, aimed to identify related current problems and future challenges.

We used 4 different databases to identify 1022 records responding to an articulated keywords search-strategy. After removing the duplicates and the articles that didn’t meet the search-criteria, we assessed 20 full-text articles for eligibility. Of the 20 eligible articles, we charted 8 records on cell cultures combined with tribocorrosion tests on implant materials (titanium, CoCrMo, and/or stainless steel). The year of publication ranged from 1991 to 2019. The cell line used was mostly murine. Two records used fretting tests, while 6 used reciprocating sliding with pin-on-disc tribometers. An electrochemical three-electrode setup was used in 4 records. We identified overall two experimental approaches: cells cultured on the metal (5 records), and cells cultured near the metal (3 records).

Research activities on tribocorrosion processes in the presence of cells have been undertaken worldwide by a few groups. After a limited initial interest on this topic in the 1990’s, research activities have restarted in the last decade, renewing the topic with technologically more advanced setups and analytical tools. We identified the main problems to be the lack of test reproducibility and wear particle characterization. We believe that the main challenges lay in the interdisciplinary approach, the inter-laboratory validation of experiments, and the interpretation of results, particularly in relation to potential clinical significance.

1. Introduction

1.1. Rationale

The biocompatibility and corrosion resistance of medical implant alloys rely on the presence of protective passive films, which naturally form on the surface of the alloy (e.g. Ti-alloy, CoCrMo-alloy, Stainless Steel). Metallic implants are generally well tolerated in the majority of the clinical cases in orthopedics and dentistry [1]. Nevertheless, the use of metallic implant-components has not been free from complications; for example, CoCrMo implants have been associated with metal hypersensitivity [2], local adverse tissue reactions [3] and implant failure [4]. Different reviews have reported on the adverse biological effects of chromium [5,6] and cobalt [6] ions from CoCrMo implants.

Tribocorrosion refers to synergistic effects of wear and corrosion on metals with passive films. In tribocorrosion research applied to the medical field, in-vitro experiments aim to simulate the in-vivo processes of wear and corrosion, for a better understanding of related mechanisms and, ultimately, for improvements in clinical practice. Different reviews have summarized research work on tribocorrosion, investigated with laboratory tribometer-machines or hip simulators integrated with a three-electrode electrochemical setup [7–9]. The problem of tribocorrosion of medical implants has been approached in multiple ways. Some research studies have focused on the characterization of wear and corrosion products found in pathologic periprosthetic soft tissues, recovered at the time of a revision procedure or autopsy [10–14]. Other studies have focused on the analysis of trace metals released into body fluids, typically in the saliva, urine, whole blood, and blood serum, in patients carrying metal-on-metal prostheses and metal dental implants [15,16]. Other studies have focused on the analysis of the damage of retrieved modular implants in the head-neck junction [10,17–20]. Finally, other research groups have focused their work specifically on wear particles, in different ways: by analysis of in-vivo samples through the development and evolution of new methods for particles isolation, digestion and characterization [21,22]; by analyzing and comparing wear particles generated by different tribometers and hip simulators [23–25]; or by testing the cell responses to metal wear particles and ions, either commercially available or previously produced in tribometers [26]. Most recent efforts have aimed to represent the clinical situation in laboratory experiments, including the development of new test fluids [27], the use of protein-containing cell culture media instead of bovine serum or phosphate buffer saline solutions [28,29], and the use of clinically relevant degradation products for cell culture experiments (for example, CoCrMo wear particles generated from hip simulators in non-processed versus processed states [30]).

1.2. Objectives

There is a paramount need for improving the current understanding of in-vivo mechanisms of wear and corrosion of metallic implants in total joint arthroplasty, and of related biological reactions. In-vitro tests are an important tool to address this need, preliminary to any in-vivo animal studies and clinical trials. The required level of complexity for such in-vitro tests is high, given the complexity of tribocorrosion tests on the one side, and of biological cell or tissue experiments on the other. The impact of the outcomes of in-vitro tests is limited – primarily due to the choice of material and testing parameters. These parameters are not only variable among patients in the clinical setting, but they are also often unknown (as for example, the electrochemical potential at the surface of a metallic implant in the body). In this context, a promising as well as challenging direction is given by an interdisciplinary approach, where in-vitro tribocorrosion tests are combined with cell or tissue culture tests. This approach may prove to be effective in reducing the gap between clinical evidence and laboratory experiments. Therefore, we raised the following question for our review: Which in-vitro research approaches have been used to investigate cells and tissues involved in tribocorrosion processes? We aim to: (a) find all publications reporting on an interdisciplinary approach as described above; (b) summarize the results obtained by those publications; (c) identify current problems and future challenges in the investigation of wear and corrosion concerning metallic medical implants.

2. Methods

2.1. Protocol and Registration

This review is written following both the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) extension for scoping reviews [31] and the Joanna Briggs Institute (JBI) guidance for conducting systematic scoping reviews [32]. Registration for scoping reviews on PROSPERO (the NIH international prospective register of systematic reviews) was not available at the time this article was written.

2.2. Eligibility Criteria

No limitations on language, date of publication, or geographic area were set, with the rationale of retrieving all published work on the topic. Peer-reviewed papers, conference proceedings and abstracts describing a conference or event were screened.

2.3. Information Sources

To identify potentially relevant documents, the following bibliographic databases were searched: PubMed/MEDLINE, Scopus, Web of Science, and Google Scholar. The search was run on April 22^nd (PubMed/MEDLINE, Scopus, and Google Scholar)/April 25^th (Web of Sciences) of 2019, and repeated on February 5^th of 2020. The search included both controlled vocabulary (e.g., MeSH terms) and keywords in the title, abstract or keywords list. The final strategies were drafted by an experienced librarian (JW) and further refined through the other co-authors. The final search results were exported into a single EndNote database, and duplicates were removed by the first author.

2.4. Literature Search Strategies

We articulated the search strategy in three parts using the PPC framework, with: Population = cells and tissues; Concept = in vitro studies; Context = tribocorrosion processes. The three parts were connected by the operator “AND” in the search formula, resulting in: Population AND Concept AND Context. Next, we found keywords to represent each part, so that the formula: (cells and tissues) AND (in vitro studies) AND (tribocorrosion processes) was “exploded” (see Table 1). We finalized the choice of the keywords after a process of adding and removing keywords at multiple steps. A complete reproducible PubMed search string can be found in Appendix A.

Table 1.

Search strategy based on the PCC framework: Population (cells and tissues) AND Concept (in vitro studies) AND Context (tribocorrosion processes).

	#	Keywords
Population	1	cells OR tissues OR “cell culture” OR osteoblast OR osteoblastic OR osteoclast OR osteoclastic OR macrophage OR lymphocyte OR lymphocytes
Concept	2	machine OR device OR set-up OR setup OR configuration OR tribological OR “pin on disk” OR “pin-on-disk” OR “pin on disk” OR “pin-on-disk” OR “pin on flat” OR “pin-on-flat” OR “ball on disk” OR “ball-on-disk” OR “ball on disk” OR “ball-on-disk” OR “ball on flat” OR “ball-on-flat” OR fretting OR “debris production” OR “particles production” OR “particle production”
Context	3	Tribocorrosion OR “mechanically assisted corrosion” OR “fretting-corrosion” OR “fretting corrosion”
	4	“Corrosion”[Mesh] OR corrosion
	5	tribo* OR biotribo* OR bio-tribo* OR wear
	6	# 3 OR (# 4 AND # 5)
Formula:	7	#1 AND #2 AND #6

2.5. Selection of Sources of Evidence

The first and second authors of this review imported all records into the program EndNote X1.0.1. The first author eliminated the duplicates, evaluated the titles and abstracts in the screening process, and assessed the full texts to determine eligibility.

2.6. Data Charting Process, Data Items and Synthesis of Results

The data charting was done by the first author, and the results were summarized in a table. The year of publication was also charted, in order to give the reader an idea about the history of the development of biotribocorrosion setups. Different terms used to describe the same wear configuration were unified into “pin-on-disk” and “pin-on-plate”, referring to a rotating disk and sliding pin respectively. Information about the main findings of each publication is described in the text.

3. Results

3.1. Selection of Records

The number of records screened was 691, obtained after eliminating 331 duplicates from the original 1022 (Fig. 1). Of these 691 records, 20 full-text articles were assessed for eligibility, and 671 were not relevant. The following articles were deemed irrelevant: reviews; clinical articles, including case reports and animal studies; articles reporting about biocompatibility tests; abstracts of conference proceedings; articles, in which the term “cell” was used in another context (e.g., ‘electrochemical cell’, ‘solar cell’, ‘fuel cell’, cytotoxicity to wear particles or corrosion ions). Among the 20 full-text articles assessed for eligibility, 12 were excluded for not presenting studies with both tribological and corrosion (electrochemical) aspects in one single experimental setup. Nevertheless, these studies are mentioned in Section 3.3.3. Finally, 8 records were included for charting and discussion.

Fig. 1.

Flow diagram illustrating the selection process of the studies included in this scoping review.

3.2. Characteristics of Records

The studies finally included for charting and discussion were divided into two groups, depending on the approach used (Table 2). Approach #1 corresponds to having cell culture dishes located near the metal, where the tribocorrosion process takes place (Fig. 2a). Approach #2 corresponds to having cells cultured directly on the metal, thus directly involved in the electrochemical (corrosion) processes occurring at the metallic surface (Fig. 2b); in this approach, cells are either part of the tribological system as lubricant component, or nearby, as in the case of a conical pin in a pin-on-disk tribometer with a small area of the wear contact relatively to the surface of cell culture. The authors using approach #1 (cells near the metal) have primarily addressed the issue of cells response to wear and corrosion debris. In contrast, approach #2 (cells on the metal) has been used to investigate issues related to the fixation interface of implants, with one exception. The exception is one study [33] that considered fretting between components of modular implants (head-taper interface in stem implant for hip replacement). In general, approach #2 is related to the question of biocompatibility and osseointegration of modified surfaces, typically from Ti-implants (for example, anodized surfaces). With the exception of [33], in the studies with cells cultured directly in the wear contact, cells were seen as a lubricant component.

Table 2.

Summarizing chart for the included studies. Abbreviations (in alphabetical order): BSA = Bovine Serum Albumin; DMEM = Dulbecco’s Modified Eagle Medium; FBS = Fetal Bovine Serum; MEM = Minimum Essential Medium; NCS = Newborn Calf Serum; OCP = Open Circuit Potential; PS = Potentiostatic; RPMI = Roswell Park Memorial Institute (cell culture medium).

	Author(s)	Year	Field	Type of cells	Wear machine	Wear configuration	Duration of wear	Materials	Electrochemical setup	Test fluid	Atmosphere
APPROACH 1: CELLS CULTURED NEAR THE METAL	Merritt K., Wenz L., Brown S.A. [1]	1991	Orthopedics – plates and screws.	L929 mouse fibroblasts; human dermal fibroblasts	Eight-arm fretting simulator (ASTM F897)	Lateral fretting between screw head and plate hole	90 – 100 h	F138 Stainless steel (plates and screws); F563 CoCr-alloy (plates)	n/a, free potential (not measured)	MEM + 10% horse serum + 0.22wt% NaHCO3	37 °C, 5% CO2 (incubator)
	Maurer A.M, Merritt K., Brown S.A. [2]	1994	Orthopedics – plates and screws.	L929 mouse fibroblasts and 3T3; mouse monocytes (phagocytic cells) P3188D1	Eight-arm fretting simulator (ASTM F897)	Lateral fretting between screw head and plate hole	120 h	Ti6Al4V plates and screws	n/a, free potential (not measured)	MEM + 10% horse serum + antibiotics; RMPI + 15% NCS + antibiotics	37 °C, 5% CO2 (incubator)
	Impergre A., et al. [3]	2019	Orthopedics – articulation interface in total ankle prostheses.	RAW 264.7 mouse macrophages	Custom-built tribometer	Pin-on-plate, reciprocating sliding pin	3 3 × 1.5 h	CoCrMo disk, UHMW-PE pin	Three-electrodes, OCP	RPMI + 10% FBS + antibiotics	37 °C, 5% CO2, 95% humidity (custom thermostatic enclosure)
APPROACH 2: CELLS CULTURED ON THE METAL	Shi B., et al. [4]	2007	Medical applications in general, including orthopedics	3T3 fibroblasts	Universal Micro- Tribotester (UMT) CETR	Pin-on-plate, reciprocating sliding pin	10 min	Ni samples, Al pin	n/a, free potential (not measured)	DMEM + FBS + antibiotics; 1% BSA in 0.15 M NaCl	Not specified.
	Felgueiras H.P., et al. [5]	2014	Dentistry – fixation interface in dental implants	MG63 osteoblast-like	Multispecimen-tester FALEX-TETRA	Pin-on-disk, rotating pin	72 min	Ti (grade 2) plate, Al2O3 ball	n/a, free potential (not measured)	DMEM + 10% FBS + antibiotics	Not specified.
	Runa M.J., et al. [6]	2015	Orthopedics – bone-stem fixation interface	MG63 osteoblast-like	CETR tribotester	Pin-on-plate, reciprocating sliding pin	30 min	Ti6Al4V disk, Al2O3 ball	Three-electrodes, OCP	α-MEM + 10% FBS + antibiotics	37 ± 1 °C
	Runa M.J., et al. [7]	2017	Orthopedics – bone-stem fixation interface	MG63 osteoblast-like	Custom-built tribometer	Pin-on-plate, reciprocating sliding pin	1 h	Ti6Al4V disk, Al2O3 conical pin	Three-electrodes, PS (−0.2V, −0.9V, 0.7V vs. SCE)	α-MEM + 10% FBS + antibiotics	37 °C, 5% CO2 (incubator)
	Hui T.T., Kubacki G.W., Gilbert J.L. [8]	2018	Orthopedics – modular implants (head-taper fixation interface)	MC3T3-E1 mouse preosteoblasts	Custom-built fretting test instrument	Pin-on-plate, reciprocating sliding pin	8 h	Ti6Al4V disk, borosilicate glass conical pin	Three-electrodes, OCP, PS (−0.1V, −0.3V, −0.05V vs. Ag/AgCl)	α-MEM + 10% FBS + antibiotics	37 °C, 5% CO2 (incubator)

Fig. 2.

Schematic illustrations for the two approaches used in the studies included in this review: Approach #1 - cells cultured near the metal surface (a); Approach #2 - cells cultured on the metal surface (b). The tribological system is integrated with an electrochemical setup, with the metal disk as the working electrode (WE), a Platinum wire as a typical counter electrode (CE) and a reference electrode (RE).

Considering the information charted in Table 2, we have chosen to indicate the duration of the wear test for the studies included, because it is a relevant factor affecting the living cell culture. Other parameters, such as load or amplitude and frequency in oscillation wear, are described in the text.

3.3. Results of Individual Records

3.3.1. Approach 1: Cells Cultured near the Metal (n = 3)

In the 1990s, the group of S. A. Bown (Ohio, USA) successfully implemented fretting tests within a cell culture (K. Merritt, A. M. Maurer, et al. [34,35]), in the context of orthopedic plates and screws typically used to fix fractures in long bones. The aim was to provide detailed information on the processing and storage of the metallic species by the surrounding biological environment. Their experimental setup consisted of a plate-screw system held above the cell culture, which was placed at the bottom of a petri dish. This setup was found to generate corrosion debris with resulting metal ion release, rather than metallic debris from wear. In Merritt et al. [34], the three elements of biological concern from stainless steel and CoCrMo-alloys, namely Ni, Co, and Cr, were analyzed by graphite furnace atomic absorption spectroscopy (GFAAS). The authors were able to identify and quantify the fraction of each metal directly associated with the cells, finding a dominant cellular association to Cr, whereas Co and Ni were predominantly associated with the supernatant culture medium. In addition, the distribution of both Ni and Cr in mouse fibroblasts cultures was found to be similar to that in human fibroblast cultures. These findings were found to be in agreement with in-vivo studies showing Cr accumulation in red blood cells or tissue sites and comparatively low levels of Ni and Co. In Maurer et al. [35], the authors tested Ti6Al4V plates and screws with different cell lines; in addition, they compared the uptake of fretting corrosion products with that of salts. The variability of results among different cells and experiments was relatively large. Despite this, the study demonstrated that the cell association to Ti was 10 times higher than that of V, and that it was membrane-bound; the cellular association of either element was found to be larger from fretting corrosion than from dissolved salts.

Almost 20 years later, this approach was successfully implemented by Impergre et al. (Lyon, France), in the context of an orthopedic system simulating the wear in total ankle prostheses [36]. The bio-tribocorrosion setup was built up inside a clean room, with a custom thermostatic enclosure for a controlled atmosphere replacing the incubator. The system developed allowed for observation of cells and polyethylene debris by optical fluorescence microscopy. Their testing protocol consisted of 3 cycles of the main test sequence, preceded by 30 min of OCP stabilization and followed by 1 h OCP stabilization, for a total of 8 h for each test. The main test procedure had three steps: compression stress relaxation, rheology, and friction. The following methods were used: cell viability test (MTT Assay); cell cytotoxicity (LDH Assay); polyethylene wear rate (wear scar volume measured by non-contact optical profilometry); polyethylene wear debris analysis (environmental scanning electron microscopy and image analysis); metal ion release (inductively coupled plasma mass spectroscopy); and inflammatory cytokines measurements (ELISA Assay). The authors presented the results of two tests in two experimental conditions, which promoted the formation of either polyethylene wear particles or metal ions. The differences in the two test conditions concerned: the finishing surface of the metal alloy (grinded or mirror polished); the density of plated cells; the fluorescent marker used for the cells; the sliding speed; and the loading during the rheology step. The authors’ conclusion was that the potential of the designed experimental system was demonstrated by preliminary results.

3.3.2. Approach 2: Cells Cultured on the Metal (n = 5)

The first study using this experimental approach was published in 2007 [37]. It reported briefly on frictional properties and wettability of Ni-samples under dry and lubricated conditions. The lubricated conditions included fibroblasts preliminarily cultured for 48 h on the Ni samples, designed as filters and treated like a matrix. The rationale for the use of Ni was mentioned in the introduction, with references to medical applications of Ni-alloys such as cardiovascular, gastroenterologic, urology, and orthopedic implants - despite well-known allergic and toxic effects of Ni. Indeed, the authors found that cells did not attach to the Ni-matrix; still, they reported a lower and more constant friction coefficient with cells cultured on the Ni-matrix, compared to BSA solution. The second record, published in 2014 [38], aimed to demonstrate the biological and mechanical stability of bioactive Ti (Ti grade 2 plates anodized in a calcium acetate and β-glycerophosphate electrolyte) for dental implants. The biotribological tests were run under 0.8 N, simulating real-life mastication loads, on a pin-on-disk tribometer; the rotating pin was sliding against the anodized Ti-samples before and after the osteoblastic cells were cultured on the sample. The results demonstrated the resistance of the anodic layer and the regenerative capacity of the cells after implant degradation. Interestingly, there was no specific information on the tests environment in these two publications.

Two records by Runa et al., from 2015 and 2017 [39,40], document a collaboration work in the biotribocorrosion field between the groups of L. A. Rocha (SP, Brazil) and of M. T. Mathew (Chicago, USA). Both records address the problem of implant fixation at the bone-stem interface in total hip replacement. In [39], the influence of an osteoblastic layer on the tribocorrosion behavior of Ti6Al4V alloy has been demonstrated. The tests were run on a commercially available tribometer, outside the incubator, at a controlled temperature. Three different Ti6Al4V surfaces were compared: (1) etched Ti6Al4V immersed in PBS for 6 days; (2) etched Ti6Al4V immersed in cell culture medium for 6 days; and (3) etched Ti6Al4V colonized with osteoblastic cells for 6 days. The applied load was 0.05 N, and the stroke length of the reciprocating sliding pin (Al₂O₃ ball, 10 mm in diameter) was 2 mm. The sliding action of the Al₂O₃ ball caused a detachment of the cells from the wear contact area; SEM analysis showed the removal of the cell layer and accumulation of cellular material at the edges of the wear scars. The differences in the measured coefficient of friction and open circuit potential were discussed, keeping in consideration that the conditions of the test were not representative of a clinical situation of fretting in the bone-stem interface. Additionally, results from electrochemical impedance spectroscopy (EIS) tests performed before and after the tribocorrosion tests were integrated into the discussion. In [40], a custom-built tribometer was used inside a CO₂ incubator, and an Al₂O₃ conical pin was used as counter-body sliding against (untreated) Ti6Al4V disks. Osteoblast-like cells were cultured up to 6 days on the surfaces of the Ti6Al4V disks until a complete coverage with a confluent layer of cells was achieved; the samples were then used for tribocorrosion tests with 0.05 N applied load, 2 mm stroke length, and 1 Hz frequency, for 1 h. The focus of this study was set on the influence of different applied potentials on the cells and the tribocorrosion behavior of the alloy, in line with the research of J. L. Gilbert’s group (NY, USA) described later in this paper. In [40], the following three potentials were applied during wear (values vs. SCE): - 0.2 V, −0.9 V, and + 0.7 V. These potential values were respectively referred to: E_corr (corrosion potential), E_cat (cathodic potential), and E_pass (passive potential). Total mass loss of the Ti6Al4V disks was estimated from the wear scars observed under SEM, and the fraction of mass loss due to corrosion was calculated using Faraday’s law. Cell cultures were characterized before and after tribocorrosion testing, using MTT Assay and Scanning Electrode Microscopy (SEM with incorporated EDS). Further, morphological changes and cytoskeletal re-arrangement of the cells after the tests were examined under an inverted microscope, using the fluorescence of different staining reagents (immunocytochemistry). Also, a quantitative real-time polymerase chain reaction (qRT-PCR) assay was used to evaluate the expression level of specific osteogenesis and inflammatory-related genes (gene expression).

Similarly to the study above, the cells were pulled away from the wear contact area during the tribocorrosion tests. The authors observed differences both in the tribocorrosion behavior of the system and on the cells among the three tested potentials; in particular, cells subjected to the - 0.9 V/SCE applied potential (E_cat) showed the most affected morphology after the tribocorrosion test, while the highest material loss of the ti6Al4V disks was seen with the tests at - 0.2 V/SCE (E_corr). This record documents the authors’ efforts to interpret the outcomes from their results in view of clinical situations.

The group of J. L. Gilbert (NY, USA) investigated the behavior of cells cultivated on the surface of Ti6Al4V-alloy, directly adjacent to the fretting site in a custom-built test instrument [33]. The pin, in reciprocating sliding motion against a Ti6AlV disk, was a cone-shaped borosilicate glass with a fine tip. No observations about cells in the wear contact area were reported. Six experimental groups were organized in an experimental design quadrant, considering the following tribocorrosion conditions: (1) no fretting, with no applied potential; (2) no fretting, with applied potential below - 0.4 V vs. Ag/AgCl; (3) fretting, with applied potential below - 0.4 V vs. Ag/AgCl; and (1) fretting, with applied potential above - 0.4 V vs. Ag/AgCl. The potential of - 0.4 V vs. Ag/AgCl was called “threshold potential”, referring to a potential range for cell viability: potentials more negative than this threshold value were associated with cell death by apoptosis. The fretting tests were run at 90 ± 5 μm amplitude, 8 Hz frequency, and 2 ± 0.5 N load. These parameters were chosen in order to assure that the potential drop observed on the onset of fretting for the tests at free potential (or OCP) remained below - 0.4 V vs. Ag/AgCl during the test, thus below the threshold limit for cell viability. The results were presented in terms of average current density, for the metal; and optical fluorescent images, SEM micrographs, and quantification of viability, for the cells. The cell viability decreased from 70 % in the tests hold at - 0.3 V, to 38 % in the tests hold at - 0.05 V (both above the threshold potential), to 0.5 % in the fretting corrosion tests at free potential, where the potential naturally dropped to - 1 V vs. Ag/AgCl. The authors concluded that both cathodic potential excursions and wear debris significantly affected cell viability.

3.3.3. Full-Articles Excluded (n = 12)

Two records (n = 2) [41,42] treated the issue of cells in biotribocorrosion with experimental setups, where either the tribocorrosion setup or the cell cultures was not implemented; therefore, we did not include them in the charting, considering them to be pilot studies. The group of M. A. Wimmer developed a ball-on-flat biotribometer operating inside of a CO₂ incubator, for studying the direct response of cells to freshly produced wear and corrosion debris from low carbon CoCrMo orthopedic alloys sliding against an Al₂O₃ ball. The initial phase of their work was documented in [41], with extensive and clear sections on the background of wear debris and in-situ investigations of cell response. This record presented the biotribometer conceptually, discussing contact conditions, wear rates, the implementation of electrochemical potentials, environmental constraints, and test fluids. Pilot studies on cell viability in different test fluids based on standard cell culture media were finally presented. The continuation of the work was documented in [42], where the biotribometer was validated through a series of tests, following a statistically based design of experiments known as Surface Response Methodology. The main results referred to metal contents released in the cell culture medium, depending on the applied load, the amplitude of the reciprocating motion of the ball, and the electrochemical potential applied to the metal. These results were discussed in terms of the amount of generated wear and corrosion debris, which would be sufficient to trigger an immune cell response based on literature data.

Five records (n = 5) described investigations on the effect of electrochemically polarized metals on cells behavior, which were conceptually related to the problems of inflammation and fretting corrosion of orthopedic implants [43–46]. The production of reactive oxygen species (ROS) during cathodic polarization of Ti6Al4V, in particular, H₂O₂, was found to affect the metabolic activity of osteoblasts and monocytes/macrophages [43]. In [44,45], the group of J. L. Gilbert (NY, USA) defined a voltage threshold and time dependence in cathodic excursions of both Ti6Al4V and commercially pure Ti, in relation to viability and morphology of pre-osteoblasts; in [45], systematic studies with 50 mV voltage steps between − 0.6 V and − 0.3 V (vs. Ag/AgCl) were carried out; in [44], eight potential values between − 1 V and + 1 V (vs. Ag/AgCl) were tested, and the role of H₂O₂ was discussed from the electrochemical point of view. The same group investigated the potential zone of cell viability on CoCrMo surfaces, comparing monocytes/macrophages with pre-osteoblast-like cells [46]; different zones of viability were found, which were discussed in terms of cell phenotypes. Finally, in [47] cells were cultured on Ti substrates, which had been previously thermally oxidized at a low temperature (227 °C, 5 h), and which were then subjected to electrochemical experiments. This paper aimed to investigate the influence of living cells on the protectiveness of the oxide formed at low temperature. These five records were excluded, because they were limited to electrochemical three-electrode setups, with no tribology involved in the experiments. Moreover, considering the reference lists of these records, additional literature is available specific to the topic, and related to biotribocorrosion only indirectly; this literature was not retrieved by our keywords search strategy, and it is beyond the scope of this review.

Two records (n = 2), in the dentistry research field, were studies about the corrosion behavior of Ti-implants in different saliva-like media, including cell suspensions. Cells were kept suspended in the electrolyte by gently stirring or by using a bioreactor: in [48], the effects of hyperglycemia, altered cell function, or inflammatory mediators on implant corrosion were addressed; in [49], cell culture medium (RPMI) with or without cells was compared to 0.9% NaCl, and the electrochemical behavior of Ti6Al4V was compared to new titanium alloys (Ti75 alloy, TiZr alloy). These studies were excluded because they are not directly related to biotribocorrosion.

Finally, three records (n = 3) were excluded because the wear particles from tribocorrosion tests were posteriorly used in cell cultures [50–52].

3.4. Synthesis of Results

Our final analysis included eight records (n = 8), where tribocorrosion tests were integrated with cell culture experiments; we found no records on tissue cultures in tribocorrosion testing. Out of these eight records, three records (n = 3) followed an approach with cells cultured near the metal surface; those records addressed the problem of wear and corrosion debris from orthopedic implants, either for fracture fixation or for joint replacement. The years of publication of those records ranged from 1991 to 2019, with a gap of almost 30 years between the first two and the last one. The experimental rig presented in the most recent record (from 2019) demonstrated technological advances with an integrated fluorescence microscope in addition to a three-electrode electrochemical setup. The three records covered the three main metals used in Orthopedics (stainless steel, Ti-alloy, CoCrMo-alloy). As strictly required from this approach, those studies used the controlled atmosphere of either an incubator or a custom-made thermostatic enclosure.

The remaining five records (n = 5) followed a different approach, with cells cultured directly on the metal. Those records addressed the problem of implant fixation and related fretting corrosion, common to both dentistry and orthopedics (bone-implant fixation in dental implants and hip stems, and fixation of the stem’s neck into the head in modular hip implants). This approach did not always require a controlled atmosphere: three out of the five studies ran relatively short tests, with cells cultured on the metal without considering the conditions for cells viability; in those studies, the main focus was on the metal surface and its tribocorrosion behavior at free potential (or open circuit potential). On the other hand, two of the five studies tested variations of the electrochemical potential applied to the metal surface; particular attention was given to cathodic excursions, corresponding to the drop of open circuit potential due to wear (removal of the passive protective layer). In those studies, the focus was set to cell behavior, and a controlled atmosphere was required. The experimental test rigs used were pin-on-disk or pin-on-plate, mostly with reciprocating sliding pins. An electrochemical three-electrode setup was used in three out of the five records.

4. Discussion

4.1. Summary of Evidence

Tribocorrosion processes occurring at the surface of metallic implants have been investigated in systems with living cells since the early 1990’s. We believe that the second generation of metal-on-metal total hip replacement implants, introduced in the early 1990’s [8], has drawn attention to the problem of corrosion of implants, not taken much into consideration by the scientific and clinical community until then [personal communication N.J. Hallab], and possible synergistic effects of wear and corrosion phenomena occurring at the implant-tissue interface. In clinical practice, inflammatory and immune reactions in patients may occur due to different processes/mechanisms of implant degradation, which are either enhanced by tribocorrosion, or related to physiochemical conditions, or hypothetically driven by cellular-gated mechanisms [1]. The degradation products released from implants are generally in the form of metal, polymer or ceramic nano- and micro-particles, free metal ions, colloidal organo-metallic complexes and inorganic metal salts or oxides; these degradation products are disseminated into the surrounding tissue as well as to remote locations in the body [5,53]. In the case of metal-on-metal joint implants, for example, wear particles found in-vivo were predominantly nanometer-sized (in the range 10 – 400 nm) and round- to oval-shaped; in addition, Co-Cr-Mo needle-shaped nanometer-sized particles have been also found in the case of longer implantation times [54]. Implant debris are uptaken by macrophage cells via phagocytosis, diffusion, and membrane-mediated transport mechanism, activating periprosthetic macrophages and leading to systemic recruitment and polarization of systemic macrophages, giant cells and dendritic cells towards periprosthetic region [53]. Activated macrophages release pro-inflammatory cytokines (IL-8, IL1-β, IL-6), which can be measured in in-vitro experiments. The re-emerging possibility to investigate the in-vivo tribocorrosion processes with an in-vitro experimental setup as schematically illustrated in Fig. 3 is, therefore, in line with growing attention to the clinical adverse effects of these processes.

Fig. 3.

Schematics illustrating how degradation products from in-vitro tribocorrosion processes (particles, ions, organometallic complexes, inorganic metal salts or oxides) affect cell cultures, aiming to simulate debris from in-vivo implant degradation processes affecting local tissues. Activated macrophage cells release pro-inflammatory cytokines (IL-8, IL1-β, IL-6), which can be measured in in-vitro experiments.

Over the past 30 years, six groups have achieved an effective biotribocorrosion experimental setup, reported by eight peer-reviewed publications. Among the eight peer-reviewed publications, the materials, testing configurations, and testing conditions were mostly different; therefore, we divided the records into two groups, depending on the approach used: either cells cultured near the metal, or cells cultured on the metal. We found that the choice of the approach was correspondingly related to the aim of the research: either to address the issue of cell response to wear and corrosion ions/debris, or to investigate issues related to fixation interfaces of implants. Also, a variety of analytical tools were used: the assessment of results from the biological side was mostly rich, with commercially available assay kits and microscope analysis (SEM or fluorescence); from the tribocorrosion side, the evaluation of wear and corrosion was prevalently done on the metal sample, with scarce reports ion contents from the medium, and no reports on metallic wear particles characterization.

From our study of the eight records charted in this review, we identified two main problems. The first is related to the acceptable reproducibility of the tests; in fact, given the high sensitivity of electrochemical measurements to ionic species in the relatively rich cell culture media, and the intrinsic variability of biological systems, the issue of reproducibility is particularly critical. It is clear that a high number of test repetitions (n ≥ 5) would be essential for data analysis with statistical methods and results interpretation. The second problem is related wear particle characterization analysis; when assessing cells response in the presence of tribocorrosion testing, the difference between the effect of metal ion release from corrosion and metallic debris from wear could not be clarified by any investigation so far. None of the charted studies reported about the isolation and characterization of metallic wear particles generated during the tribocorrosion testing. Therefore, the question remains open, whether the effect of ions played a significant role in cell response, or if the biotribocorrosion testing devices were not able to produce a sufficient amount of wear particles in the test time frame, or if they were producing nanoparticles which could not be detected.

The main future challenges are related to the interdisciplinary work, inter-laboratory validation of experiments, and interpretation of results, particularly in relation to potential clinical significance (Fig. 4). For the development of the emerging field of bio-tribology, communication between biologists and engineers is an essential step. This communication needs to be pursued not only at an institutional level but also between institutions. National and international conferences on tribology mostly cover this topic, and a dedicated international conference on Biotribology has been taking place biannually since 2012. So far, each group has used a different protocol, which was conditioned by the experimental setup available and was mostly aim-oriented. As a consequence, the results among the studies could not be compared with each other. Future studies would need a stronger validation by a more significant number of test repetitions (with statistical analysis), within the same lab and inter-laboratory, so to lead to cohesive conclusions, and be able to relate to current clinical challenges. More studies are needed for defining an optimal experimental setup. For example, a round-robin with the participation of different laboratories could be initiated, similarly to the one organized 20 years ago by the Tribology Committee of the CEFRACOR on tribo-electrochemical measurements [55], which included the participation of 7 European laboratories. By adding the experimental biological component to the round-robin, specific related issues like the different handling among microbiologists, and contamination, need to be carefully considered. In summary, the reported studies in this review are innovative and ground-breaking in terms of biotribocorrosion studies with living cells; however, many more studies are needed before we can achieve a clear understanding of complex biotribocorrosion processes at implant interfaces.

Fig. 4.

Illustration on future challenges related to research on bio-tribocorrosion combined with cells cultures, identified by this scoping review.

4.2. Limitations

The screening process has been carried out only by the first author, with no cross-check for errors. The number of studies included was relatively low, because the scope of the review addressed a specific field, biotribocorrosion in medical applications, initiated 30 years ago by the research activities of one single group, and re-emerged only in the last decade.

5. Conclusions

The research field of biotribocorrosion related to orthopedic and dental implants in combination with cell or tissue cultures has considerably re-emerged in the last decade. The number of studies on biotribocorrosion is limited by different challenges offered by the field, including the interdisciplinary character of the experimental work and analysis of the results. The studies charted in this review document a variety of materials, methods and results, which demonstrate the relevance of including cell cultures in tribocorrosion tests, and serve as a platform for future research. Their essential value lays on the fact that they represent the intersection point of various research pathways, from clinical analyses of retrieved peri-prosthetic tissues, and clinical studies on patients’ ion levels, to engineering investigations on retrieved implants and laboratory tribocorrosion tests, all of which aim to gain a better understanding on cells and tissue reactions to wear and corrosion degradation products from metallic implants. We recommend standardization of the materials and methods, taking into consideration consider the two approaches identified in this review (cells near the metal, and cells on the metal), depending on the main question addressed (local biological effects of tribocorrosion debris, or implant fixation interfaces).

Appendix A

Complete reproducible PubMed search string.

(“Tissues”[Mesh] OR “Cells”[Mesh] OR “Osteoblasts”[Mesh] OR”Osteoclasts”[Mesh] OR “Macrophages”[Mesh] OR “Lymphocytes”[Mesh] or cells[TIAB] OR tissues[TIAB] OR “cell culture”[TIAB] OR osteoblast[TIAB] OR osteoblastic[TIAB] OR osteoclast[TIAB] OR osteoclastic[TIAB] OR macrophage[TIAB] OR lymphocyte[TIAB] OR lymphocytes[TIAB])

AND (machine[TIAB] OR device[TIAB] OR set-up[TIAB] OR setup OR configuration[TIAB] OR tribological[TIAB] OR “pin on disk”[TIAB] OR “pin-on-disk”[TIAB] OR “pin on disc”[TIAB] OR “pin-on-disc”[TIAB] OR “pin on flat”[TIAB] OR “pin-on-flat”[TIAB] OR “ball on disk”[TIAB] OR “ball-on-disk”[TIAB] OR “ball on disc”[TIAB] OR “ball-on-disc”[TIAB] OR “ball on flat”[TIAB] OR “ball-on-flat”[TIAB] OR fretting[TIAB] OR “debris production”[TIAB] OR “particles production”[TIAB] OR “particle production”[TIAB])

AND (Tribocorrosion[TIAB] OR “mechanically assisted corrosion”[TIAB] OR “fretting-corrosion”[TIAB] OR “fretting corrosion”[TIAB] OR

((“Corrosion”[Mesh] OR corrosion[TIAB])

AND (tribo*[TIAB] OR biotribo*[TIAB] OR bio-tribo*[TIAB] OR wear[TIAB])))