Clinical Prospective Assessment of Genotoxic Effects of Dental Implants in Gingival Epithelial Cells

Abstract

Objectives

Although titanium-based implants are considered bioinert, it has been found that they are subject to corrosion and wear. This study aimed to evaluate the cytotoxic and genotoxic potential of two implant systems in gingival epithelial cells.

Material and methods

Gingival swabs were taken three times from 91 subjects. The first swab was taken before dental implant placement, the second swab 90 days after dental implant placement and the third swab 21 days following the healing abutment placement. DNA damage was analyzed using the micronucleus test. Tested dental implants with corresponding healing abutments were Ankylos and Dentium SuperLine.

Results

Of all scored forms of cytogenetic damage in gingival cells of individuals after implementation of tested dental implant systems, only an increase in the number of binucleated cells (P ≤ 0.001) was significant in contrast to control values for both tested implant systems, 90 days after dental implant placement and 21 days following the healing abutment placement.

Conclusion

It may be concluded that there are no titanium-based implant dependent cytogenetic damage in gingival epithelial cells. A slight increase in cytogenetic damage has been observed but it is of no biological relevance and might be associated with healing abutment induced effect.

Introduction

The use of dental implants has increased in recent years and this trend is expected as the consequence of the ageing of the world's population and dental therapies improvement. Titanium and titanium alloys are the most frequently used materials for manufacturing dental implants today due to their desirable physical and mechanical properties and favorable biocompatibility. The biocompatibility of titanium as an implant material is associated with protective and stable layer of oxides that spontaneously build on the surface of the implant in air and/or physiological fluids. This layer is a potent barrier against the dissolution of the metal. Owing to this barrier, titanium exhibits excellent resistance to corrosion (1-3).

Titanium Grades 1, 2, 3 and 4 are titanium materials that are commercially referred as pure titanium. They usually contain some carbon, oxygen, nitrogen and iron. These elements enhance the mechanical properties of pure titanium considerably and are found in varying amounts ranging from Grade I to Grade IV. Thus, mechanical characteristics such as implant strength, creep resistance and formability can be improved by combining titanium with specific elements (e.g. aluminum, Al; vanadium, V; tantalum, Ta; zirconium, Zr). The two alloys that are most frequently used are Ti-6Al-4V and Ti-6Al-4V-ELI (extra low interstitial alloys) (4).

Surrounding tissue reactions to an implant are mainly related to the effects of the load and its stability. However, the surface attributions of the substrate, e.g. composition, roughness, wettability and morphology are also crucial factors that play a role in affecting the state of homeostasis in cells surrounding the implant (5, 6). Surface-roughness has impact on the osseointegration of titanium dental implants and different processes correlated to providing a roughened surface may lead to the release of cytotoxic aluminum ions into the peri-implant tissue. Some studies on metal particle release from dental implants have confirmed an association between the inflammatory response in peri-implant tissues and particles derived from dental implants in the same surrounding (7, 8). There are several ways of discharging these particles from dental implants. They are released during implant placement, due to the wear of implant surface, due to polishing and finishing of the implant surface or the corrosive effect of therapeutic substances. Cyclic micro-movements appear at the contact surfaces of the implant components, especially at the level of implant-abutment connection. That can result in tribocorrosion, process of degradation by corrosion and wear processes on titanium surfaces, often leading to a significant increase in material loss. Mechanical wear can destroy the passive protective layer of titanium oxide that is formed on its surface. In consequence, metal becomes more prone to corrosion. Subsequently, it becomes more vulnerable to mechanical wear. Mechanical wear also facilitates corrosion (9-12). Immune system may recognize micro- and nano-particles released from an alloy as a result of degradation of the dental implant system as foreign bodies. By stimulating the activation of several mediators, including cytokines, an inflammatory response in the peri-implant tissues may be provoked (12, 13).

Dental materials could release small amount of their components into the oral cavity. Therefore, proper regulations have to ensure that the risk from genotoxicity/mutagenicity of dental materials is at the lowest possible level. Detailed biocompatibility records are needed to assess the comprehensive risks of the above-mentioned released compounds (14-16). As stated above, the aim of this in vivo study was to evaluate biocompatibility, in terms of cytogenetic damage of implants and healing abutments from two different implant systems in gingival epithelial cells. At the beginning of the study, the following null hypothesis was established: titanium-based dental implants and healing abutments from two different implant systems would not have any genotoxic and cytotoxic effects on gingival epithelial cells.

Material and methods

Study Design

This prospective, randomized clinical trial with two parallel study groups was conducted between January 2020 and April 2021 in private dental practices in collaboration with the Department of Restorative Dental Medicine and Endodontics, Study of Dental Medicine, University of Split, Croatia. It was approved by the University Ethics Committee (No: 2181-198-03-04-20-0041) which also confirmed that the study was in full accordance with ethical principles including the World Medical Association Declaration of Helsinki (version 2013), and met all additional requirements. The study was also registered at clinical trials (ClinicalTrials.gov, Study ID number: NTC04540991). The participation was voluntary, anonymous, without any financial support, and all participants were introduced to the background and the aim of the study. All participants gave their informed consent in writing before inclusion in the investigation.

Participants

This study comprised a total of 91 patients of mean age 53.8 ± 9.9. Participants’ demographic data are presented in Table 1. The main inclusion criterion was the absence of tooth/teeth in the mandibular molar or premolar region where the tested implants were placed. Additional characteristics of the participants were: belonging to ASA I or ASA II group (according to the American Society of Anesthesiologists), the absence of titanium and penicillium hypersensitivity, the absence of prosthetic restoration/replacement or orthodontic appliances in the oral cavity/dental amalgam fillings, the absence of oral precancerous lesions and the absence of bisphosphonates and corticosteroids used in therapy. Exclusion criteria were the presence of systemic disease (e.g. uncontrolled diabetes, oral mucosal diseases, untreated gingivitis and periodontitis, endodontic lesions), pocket depths ≥ 4 mm on adjacent teeth, bruxism, poor oral hygiene, pregnant and lactating women, antibiotic therapy in the last three months, taking any other pharmaceuticals that have been proved to accelerate DNA damage, using mouthwash that contains alcohol. The CONSORT (Consolidated Standards of Reporting Trials) study flowchart is presented in Figure 1.

Table 1. Prevalence of sociodemographic characteristics among participants.

Characteristics		Dentium SuperLine	Ankylos
Gender	Male	16 (35.6)	19 (41.3)
	Female	29 (64.4)	27 (58.7)
Age, years		52.0 ± 10.1	54.1 ± 9.5
Smoking (≤ 10 cigarettes per day)	Non smoker	22 (48.9)	27 (58.7)
	Smoker	23 (51.1)	19 (41.3)
Alcohol intake	Never	6 (13.3)	6 (13.1)
	≤12 units/ week	16 (35.6)	19 (41.3)
	13-24 units/week	9 (20.0)	10 (21.7)
	≥25 units/week	14 (31.1)	11 (23.9)
Meat consumption	≤1/week	8 (15.5)	9 (19.6)
	1-3/week	15 (33.3)	14 (30.4)
	4-6/week	15 (33.3)	14 (30.4)
	≥1/day	8 (17.7)	9 (19.6)
Fruit consumption	4-6/week	18 (40.0)	17 (37.0)
	≥1/day	27 (60.0)	29 (63.0)
Vegetable’s consumption	1-3/week	18 (40.0)	21 (45.6)
	4-6/week	9 (20.0)	7 (15.2)
	≥1/day	18 (40.0)	18 (39.1)
Data are presented as whole numbers and percentages or mean (SD).

Figure 1.

Flowchart of participant's recruitment and follow-up.

Detailed medical and dental histories were taken from each participant. In a structured questionnaire tailored to this study, all participants provided data regarding the age, gender, personal factors (general health, a medication used), lifestyle factors (smoking, alcohol consumption) and eating habits.

In order to analyze the strength of the test for the dependent Student’s t-test (differences of the examined groups in the ROC analysis of binuclear changes) according to the following parameters: significance level α = 0.05, an equal number of subjects in both groups, the effect of Cohen's size d = 0.587 (according to the obtained results), at least 37 participants per group had to be included in the test for 80% of the test power. A sample of 50 participants per study group was chosen to compensate for possible withdrawal, or loss.

Materials and surgical procedure

Patients were randomly assigned to one of the groups depending on the dental implant system used in the therapy. Ankylos dental implants (Dentsplay Sirona, Charlotte, USA) were used in the first group of patients and Dentium SuperLine (Dentium Co., Seoul, Korea) in the second group (n = 46 and n = 45, respectively). These two implant systems were chosen because they are the most frequently used implant systems in Croatia.

The composition of the products used, as written by the manufacturer, is shown in Table 2. The generation of the random allocation sequence was performed by computer software. Group allocation was concealed from the evaluating investigator during the analytical stage of the project.

Table 2. Composition of dental implant systems used in the study.

Dental Implant System	Manufacture	Composition
		Dental implant	Healing abutment
Ankylos	Dentsplay Sirona, Charlotte, USA	Titanium Grades 2	Ti-6Al-4V
Dentium SuperLine	Dentium Co., Seoul, Korea	Titanium Grades 4	Titanium Grades 4

The implants were placed following each implant system manufacturer's instructions, and the treatment was performed according to the patient's standards and indications. Surgical procedures on all patients were performed by the same operator with the same surgical approach, protocol and instrumentation.

A pre-operative panoramic radiograph was obtained for each patient. The surgeries were performed under local anesthesia with 4% articaine solution containing 1:100000 adrenaline (Ubistesin, 3M ESPE, Neuss, Germany). All patients were given oral antibiotic therapy of 2 gr per day for seven days, starting 24 hours prior to the intervention (Augmentin®, Glaxo-SmithKline Beecham, Brentford, UK). To ensure post-surgical oral hygiene, the patients were advised to rinse the oral cavity with 0.2% chlorhexidine (Miradent, Mouth Rinse paraguard chx, Hager Pharma GmbH, Duisburg, Germany) until the removal of sutures. The sutures were removed ten days after implantation. The implants were healing by being submerged for 12 weeks based on the surgeon's clinical judgment, indications given and the need and preference of the patients. After healing, healing abutments were placed. The used healing abutments were sterilized only once.

Sample collection and a micronucleus assay in gingival epithelial cells

To reduce individual variations, patients were observed longitudinally, and participants served as their own control. Samples of gingival epithelial cells were collected from each participant's implementation site using the swab technique at three different time points: a control swab was taken just before the placement of dental implant (T0); the second swab was taken 90 days after implantation, that is, immediately before the placement of the healing abutment (T1), and the third swab was taken 21 days after the placement of the healing abutment (T2).

One hour before the sampling, the participants abstained from consuming any food and drinks. After rinsing the oral cavity three times with tepid water to remove exfoliated cells, a T0 swab was taken by gently brushing the gingiva around the place indicated for implant placement, and, after that T1 and T2 around the implant with a cytobrush (Cytobrush Plus; GmbH. Dietramszell-Linden, Germany). The samples were subsequently applied to coded laboratory glass slides.

The cells applied to microscopic slides were allowed to air-dry and were fixed in ethanol: glacial acetic acid (3:1) at 4°C for a minimum of 20 minutes. Staining of the slides followed the procedure described by Thomas et al. (15) and nuclei were stained with Schiff’s reagent for 60 min in the dark setting, at room temperature (Feulgen-technique), whereas the cytoplasm was stained with Fast Green from Feulgen kit (lot FE-05/19, Biognost, Zagreb, Croatia) for 10-15 seconds. Nuclear anomalies, such as micronucleus, karyorrhexis (nuclear disintegration indicating apoptosis), karyolysis (dissolution of the nucleus mostly showing necrosis and apoptosis), pyknosis (nuclear shrinkage due to apoptosis), condensed chromatin (DNA complexed with proteins and apoptosis), nuclear buds (precursors of micronuclei, or high density of DNA repair complexes), broken eggs (nuclei that appear cinched) and binucleated cells (indicating the impaired speed of cell proliferation) were estimated and qualified according to Tolbert et al. (16). The analysis of 2000 gingival epithelial cells per participant was performed.

Statistical analysis

The SPSS 25.0 (IBM SPSS, Armonk, NY, USA) and Excel (Microsoft, Redmond, Washington, USA) were used for statistical data analysis. Descriptive statistics were used to determine basic statistical parameters (mean values. standard deviations). The differences among tested variables were evaluated by the Kruskal–Wallis one-way analysis of variance (difference between the evaluation periods) and the Mann-Whitney U test (difference between the dental implant systems). A multiple regression analysis was used to assess the effect of predictor variables (age, gender, eating habits, implant system, smoking, alcohol) on dependent variables (cytogenetic damage). The results are presented in the form of Pareto diagrams. The significance level was set at 0.05.

Results

The study involved 91 subjects, of whom 35 were men and 56 women, aged 32 to 71 years. The participants were randomly divided into two groups depending on the used dental implant system. The Ankylos group comprised 27 females and 19 males; aged 32–71 (mean age 54.1 ± 9.5 years). The Dentium SuperLine group comprised 29 females and 16 males; aged 39–71 (mean age 52.0 ± 10.1 years).

The results of the micronucleus test are shown in Table 3. The results of Kruskal–Wallis one-way analysis of variance showed the difference between sampling times T0 (before implant insertion) compared to T1 (90 days after insertion of the implant) and T2 sampling time points (21 days after healing abutment placement), for the number of binuclear cells (P ≤ 0.001) for both tested implant systems.

Table 3. Frequencies of nuclear abnormalities in exfoliated gingival epithelial cells at different time points.

Tested cytogenetic damages		Time of sampling
		T0	T1	T2
Micronucleus	Dentium SuperLine	1 (1.5)	1 (1)	1 (1)
	Ankylos	1 (2)	1 (1)	2 (1)
Nuclear buds	Dentium SuperLine	0 (1)	0 (1)a	0 (1)a
	Ankylos	0 (1)	1 (1)b	1 (1)b
Broken eggs	Dentium SuperLine	0 (1)	0 (1)	0 (1)
	Ankylos	0 (1)	1 (1)	1 (1)
Binucleated cells	Dentium SuperLine	4 (1)A	5 (2)Ba	5 (2)C
	Ankylos	5 (2)A	6 (1)Bb	6 (1)B
Karyorrhexis	Dentium SuperLine	1 (3)	1 (3)	3 (2.5)
	Ankylos	3 (2)	3 (0.75)	3 (3)
Karyolysis	Dentium SuperLine	7 (4)	7 (4)	6 (4)
	Ankylos	3 (4)	3 (4)	7 (4)
Condensed chromatin	Dentium SuperLine	2 (2)	2 (2)	2 (2)
	Ankylos	2 (2)	2 (1)	2 (1.25)
Pyknosis	Dentium SuperLine	4 (2)	2 (2)	4 (4)
	Ankylos	4 (2)	4 (2)	4 (2.25)
Data are presented as median and interquartile range. *Different upper capital letters indicate a significant difference among the evaluation periods, and different lower-case letters indicate a significant difference among the treatment groups (P < 0.05). Abbreviations: T0, sample before the treatment – baseline; T1, 90 days after implant insertion; T2, 21 days following the healing abutment placement.

The analysis of Mann-Whitney U test indicated no statistically significant difference between two estimated implant systems for non-cytogenetic endpoints in the time T0 before implant placement. The difference was observed for nuclear buds (P = 0.004) and binucleated cells (P = 0.021) at sampling time T1 (90 days after implant placement) and also for nuclear buds (P = 0.034) in T2 time (21 days after healing abutment placement).

The dependence of the micronucleus test parameters on all predictor variables in the total test group was determined by the general regression model and presented in the form of Pareto diagrams (Figure 2 and Figure 3). The influence on the incidence of the number of cells with micronucleus, of all observed demographic variables encountered as predictor variables, were observed for gender – female (β = - 0.342, SE = 0.112, P = 0.004), dental implant system (β = 0.263, SE = 0.112, P ≤ 0.001), alcohol (β = -0.242, SE = 0.066, P ≤ 0.001), fruit (β = 1.143. SE = 0.246, P ≤ 0.001), vegetables (β = 0.633, SE = 0.082, P ≤ 0.001) and meat (β = -0.182, SE = 0.054, P ≤ 0.001) consumption.

Figure 2.

Multiple regression analysis results. There is a significant association between cytogenetic endpoints in gingival epithelial cells (number of micronuclei, broken eggs, nuclear buds, and binucleated cells) and demographic and lifestyle factors as possible predictors.

Figure 3.

Multiple regression analysis results. There is a significant association between cytogenetic endpoints in gingival epithelial cells (karyorrhexis, pyknosis, karyolysis, and condensed chromatin) with demographic and lifestyle factors as possible predictors.

Discussion

The aim of this prospective study was to evaluate the cytotoxic and genotoxic alterations in gingival epithelial cells after the application of a titanium-based dental implant system. Patients who underwent dental implant placement were divided into two groups based on the type and manufacturer of the implant system. For one group, the implant system was Ankylos (cpTi Grade II – dental implant, Ti–6Al–4V – healing abutment), and for the other Dentium (cpTi Grade IV – dental implant and healing abutment). To compare the changes in gingival epithelial cells at the site of the implant, cytological scrapings of the oral mucosa were collected at three time points: just before implant placement, 90 days after implantation, but immediately before placement of the healing abutment and 21 days after the placement of healing abutments. Genotoxicity and cytotoxicity were assessed by a micronucleus test - the most commonly used and reliable assay in terms of evaluating the induction of chromosomal aberrations in vivo, and in vitro, and in detecting possible carcinogens.

The research results show that both implant systems, Ankylos and Dentium SuperLine, cause a statistically significant increase in the number of binuclear cells three months after their implantation compared to the control scrapings taken before implantation (P ≤ 0.001 and P ≤ 0.001, respectively). Furthermore, a statistically significant increase in the number of binuclear cells has been observed in both implant systems after healing abutment placement compared to the control group prior to implant placement (P ≤ 0.001 and P ≤ 0.001, respectively). However, since standard variations of statistically different endpoints are in the range of variation between the mean values of these endpoints for the two groups compared, and although statistically significant, the results may not be considered as biologically or clinically relevant. Therefore, the null hypothesis stating that implant systems do not cause genotoxic or cytotoxic damage can be accepted.

Karahalil et al. (17) assessed the genotoxic effects of titanium alloy dental implants on the gingival epithelial cells of 37 participants using the micronucleus test. Their results showed a slight increase in the incidence of micronuclei (P = 0.047) in the second swab taken a few weeks after the implant placement procedure. They concluded that the degree of particle release from the implant was too low to cause more severe DNA damage or genotoxicity.

The increased parameters obtained after placement of the healing abutment can be related to several factors. Since the oral cavity is a complex environment, corrosive substances from dietary, human saliva and oral biofilms may accumulate in retentive areas of dental implant systems promoting corrosion at their surfaces. Since the healing abutment is in direct contact with gingival epithelial cells, corrosion of the implant may adversely affect surrounding cells (10). Furthermore, micromovements occur during mastication at the site of the implant and healing abutment junction. Consequently, the friction that impairs the integrity of the protective layer of titanium dioxide occurs. This event promotes further corrosion and corrosion induces friction. Multiple processes (wear, friction and corrosion) in contact with biological tissues and fluids result in a bio-tribocorrosion, a relatively new research field that combines the fundamentals of tribology (friction, wear, and lubrication) and corrosion (8, 10, 18).

Ribeiro et al. (19) artificially induced corrosion in several implants of different manufacturers by immersing them in a solution of acetic acid and NaCl after which the Chinese hamster’s ovarian cells were exposed to that solution. After performing the comet test, they concluded that none of the dental implants used in their study demonstrated genotoxicity. Several studies have shown the genotoxic, and cytotoxic potential of TiO₂ particles (1–100 nm) released as a result of bio-tribocorrosion (20, 21). Similar findings were obtained in a study by Tavares et al. (22) reporting that the implant surface without finishing treatment has a thinner layer of TiO₂ and is, therefore, more susceptible to corrosion. That contributes to a significantly higher release of particles and ions that mediate the formation of free radicals, oxidative stress and DNA oxidation as primary genotoxic events.

Wang et al. (23) based on the results obtained from several different genotoxicity tests, including the micronucleus test in vitro, suggested that ultra-fine nanoparticles of TiO₂ (<100 nm) cause genotoxic and cytotoxic damage in human lymphoblastoid cells. Maloney et al. (24) and Kumazawa et al. (25) found that vanadium (V), aluminium (Al), cobalt (Co), chromium (Cr) and nickel (Ni) are genotoxic at a higher concentration. In contrast, cobalt (Co) induced a cell death. Camacho-Alonso et al. (26) conducted the micronucleus assay on buccal epithelial cells aiming to detect genotoxicity as a result of metal ion release in patients with titanium dental implants, and various metal restorations in the mouth. Although increased concentrations of metal ions were found in all subjects, no genotoxic damage was found in any of the subjects.

The reason why some studies consisting of patients undergoing implantation showed positive cyto/genotoxic effects and others not, may lie in the fact that in advanced economies healing abutments are used only once. When placed in the cavity for the first time, they release biologically active particles formed only by physiological corrosion. On the other hand, in less economically powered societies healing abutments are recycled and subjected to several sterilization cycles resulting in a higher level of proneness to corrosion and a higher level of particles available for being released, for each implantation. In addition to the influence of the composition of the material itself on corrosion, there is the issue of both the impact of the sterilization process and the number of repeated sterilizations on material composition and corrosion. Allsobrook et al. (27) investigated the effect of recurrent sterilization on the corrosion of titanium drills used for the surgical implantation procedure. He concluded that such drills are subject to material loss and particle release. Furthermore, some other authors have proved that autoclave sterilization promotes surface corrosion (28, 29). To our best knowledge, there are no publications on the influence of repeated sterilization procedures on the material composition of the healing abutments; hence there could also be a similar pattern with drills.

Increased values of binuclear cells after implant and healing abutment placement could also be explained by accelerated cell proliferation in the inflammatory phase of wound healing, which is inevitable after surgical manipulation of tissue. The patient has to be subjected to two surgical procedures; the first, when the implant is being inserted into the bone, and second when the healing abutment is placed (30). Additionally, it has been found that inflammation can also be triggered by released particles from the TiO₂ layer disrupted by the tribocorrosion which the immune system recognizes as antigen and triggers the inflammatory reaction with cytokine release (10-12). Among the different types of cytokines released, there are mitogenic cytokines, primarily IL-1β that accelerate cell proliferation (12). De Barros Lucena et al. (31) proved the accumulation of biofilm inside the implant structure in 52.6% of placed and completely closed implants before the placement of the healing abutment and prosthetic superstructure. Using the DNA-DNA hybridization technique, they confirmed the presence of 40 species of bacteria on the implants, of which 77.4% were present in the mandible. The next step in this sequence of events is that lipopolysaccharides present in the membrane of Gram-negative bacteria act as ligands for Toll-like receptors (TLRs) of gingival epithelial cells (32). Moreover, Eskan et al. (33) found that an increased expression of TLR4 receptors on gingival epithelial cells positively correlated with increased IL-1β production, inflammation itself and accelerated cell proliferation. The last step in this pathophysiological sequence of events is that accelerated cell proliferation leads to more frequent errors in cytokinesis and therefore, larger numbers of binuclear cells are recorded.

Certain studies have shown that failure of implant therapy can be caused by an inflammatory reaction in the surrounding tissue as a response to corrosion of the titanium alloy (34, 35). There is also a systemic disease associated with titanium. Berglund and Carlmark (36) studied a group of 30 subjects exhibiting the yellow nails, bronchial obstruction and lymphedema. They concluded that the cause of the syndrome is a high level of titanium (Ti)

The primary source of titanium ions was corrosion caused by galvanism between titanium implants and gold and amalgam restorations in the mouth as well as from corrosion caused by fluoride oxidation.

This trial has certain number of limiting factors. It is recommended to include more implant systems, more participants and to establish a long-term follow up of cell changes around the implant site. It would also be advisable to observe clinical and biochemical (gingival fluid) parameters and genotoxicity to evaluate the levels of tissue inflammation around the implant.

Conclusion

It may be concluded that the placement of both implants into the oral cavity, and the placement of the healing abutments affect cell homeostasis and genome stability. It is most significantly observed in an increase in binucleated cells for both tested dental implant systems. However, the observed effects could not be assigned to the impact of the released Ti particles. It may also be mediated by surgical procedures that have to be undertaken, which lead to inflammation of surrounding tissue, and the cells affected by inflammation show a cyto/genotoxic effect. However, by applying the weight of the evidence approach to evaluate the results of this study, we can say that although there has been a significant increase recorded in previously stated cyto/genotoxic events, they are not biologically relevant.