INTRODUCTION
The peri-implant soft tissue encircles the implant in the neck region similar to a tight collar. This structure, which is crucial for osseointegration of the implant, should adhere firmly to the implant surface [18]. The gingival epithelium proliferates along the implant surface without a connective tissue attachment, which is likely due to the proximity of tissue fibres in parallel arrangement along the implant axis [16]. This fundamental phenomenon underlying the mechanism of soft tissue adaptation at the cellular level is still unknown.
Fibroblasts are the primary cells of the soft connective tissue of the periodontium. Inflammation of the periodontal tissue leads to a breakdown of fibroblasts and impaired tissue integrity. As a result of the progressive degenerative changes, a loss of periodontal tissue occurs. This pathogenesis, which manifests itself in periodontal tissues, is also observed in peri-implant tissues. The soft tissues surrounding the implant function as a biological barrier and protect the peri-implant bone from microbiological infections [19].
Human gingival fibroblasts (HGF) exhibit a “filopodial” structure that extends further into grooves and microstructures on rough surfaces [7], which results in increased fibroblast adhesion [38, 41]. On the contrary, HGFs accumulate more randomly on smooth surfaces [22]. Surface texture, such as roughness and topographical changes, influences connective tissue adhesion to the transmucosal section of implants. Various physical, chemical, and biochemical surface modification techniques were introduced to increase roughness and improve cell adhesion [3]. At the same time, these well-designed surface modifications must also minimise microbial colonisation around the implant surface [17].
In vivo studies have shown a positive association between plaque accumulation rate and surface roughness in the supragingival region [9, 37]. To observe bacterial adherence and colonisation in the oral cavity, where surface irregularities such as grooves, pits, perikymata, and abrasion defects are frequently present in scanning electron microscope (SEM) images [12]. Since it is difficult to remove microorganisms from these areas, they colonise and form biofilm structures. In the case of abutment materials with a rough surface, it was found that they contain 25 times more pathogenic bacteria than smooth materials [30]. Moreover, smooth surfaces facilitate oral hygiene procedures performed by patients.
The roughness value (Ra) serves to determine the structural height of the surface and defines the arithmetic mean of the profile values [40]. In a subsequent investigation based on their previous short-term study [29], the researchers concluded that lowering the Ra value well below the 0.2 µm threshold had no significant effect on the gingival microbial colonisation [5].
In the treatment of peri-implantitis, surgical treatments are required in addition to conventional treatment methods to ensure complete decontamination [5]. Implantoplasty, also known as mechanical modification of the implant, is recommended as part of surgical treatment to change implant surface topography, reduce microbial colonization, and thus prevent reinfection [1, 4, 32]. While it is not possible to completely eliminate microorganisms using current implantoplasty techniques, a combination of mechanical and chemical techniques is more effective [35]. However, there are still experimental, non-standardised, and non-consensual implantoplasty techniques, so the mechanical or biological outcomes of these treatment approaches are still unknown.
The purpose of this in vitro study was to determine and compare the adhesion of HGF to: sand-blasted, large-grit, acid-etched (SLA®)-surfaced titanium (Ti) discs after applying diamond bur sequences with different implantoplasty protocols. The null hypothesis of the present study was that surface modifications in implantoplasty procedures did not affect HGF growth and adhesion.
MATERIALS AND METHODS
Pre-made cell lines were used in the present 72-hour in vitro cell culture study. Therefore, approval from the ethics committee was not necessary.
Sample size and study groups
Based on values from methodologically similar studies [22, 24, 25], the power analysis tool G-POWER calculated a total sample size of 48, with an effect size of 0.65, 95% power, and a margin of error of 0.56%. Groups were evenly distributed, with 12 discs in each group.
A total of 48 commercially available grade 4 pure Ti discs with a diameter of 10 mm and a thickness of 2.5 mm were used (Trias-ixx2, Servo Dental, Hagen, Germany). The discs met the biocompatibility requirements of the “Standard Specification for Unalloyed Titanium for Surgical Implant Applications” (ASTM F 67/ ISO 5832-2). Four groups each with 12 discs included in the study were:
- Group I: machined discs with smooth surfaces;
- Group II: SLA® discs with roughened surfaces;
- Group III: the SLA® discs were milled for 2 minutes with a round-tipped red diamond bur (40 μm grit, Dimei Royal, China); and for 1-minute with a round-tipped white diamond bur (15 μm grit, Komet Dental, Germany). Subsequently, all discs were polished with Brownie silicone under a 1-minute water rinse;
- Group IV: SLA® discs were milled for 2 minutes with a round-tipped green diamond bur (125 μm grit, Dimei Royal, China); and for 1-minute with a round-tipped white diamond bur (15 μm grit, Komet Dental; Germany). Subsequently, all discs were polished with Brownie silicone under a 1-minute water rinse.
Standardisation was maintained during the surface roughening protocol in groups III and IV with a custom-built stabilisation mechanism. In a controlled environment, one of the researchers (H.Y.) milled each disc clockwise from its centre to its outer periphery for the specified time frame with a separate set of burs, while an external observer recorded the milling times for each disc. H.Y. took a 5-minute break between each disc milling. This in vitro approach was used to model the milling process during implantoplasty [25]. As part of the preparation for cell culture, each disc was assigned a unique number. The Ra value of each disc was measured using a mechanical profilometer (MahrSurf M 400, Germany).
Cell culture
Following in vitro milling procedures, all discs were agitated for 5 minutes in an ultrasonic cleaner and the samples were cleaned with 20% ethanol for 10 minutes. Discs were then autoclaved at 134°C for 20 minutes under a gauge pressure of 2 kg/cm2 (Dri-Tec, Canada).
The human gingival fibroblast cell line (HGF1; ATCC) was used for cell culture. Cells were passaged in flasks containing Dulbecco Modified Eagle Medium culturing media (HyClone, USA) containing 10% fetal bovine serum (Sigma-Aldrich Co., USA, the CAS number: 9014-81-7) and incubated at a temperature of 37°C, relative humidity of 95% (to minimise media evaporation and condensation), and 5% CO2. Upon reaching 80% confluency, the cells were then subcultured using with phosphate buffered saline (Sigma-Aldrich Co., USA, the CAS number: 7758-11-4) and a trypsin-ethylenediaminetetraacetic acid solution (0.5 g/L trypsin; 0.2 g/L EDTA, Sigma-Aldrich Co., USA, the CAS number: 9002-07-7). Cells from passages 3 and 4 were used in this study.
Analysis of cell viability, proliferation and adhesion
To determine cell viability and proliferation, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (Sigma-Aldrich Co., USA, the CAS number: 57360-69-7) was used as the gold standard for assessing the metabolic activity of the cells. The sterilised discs of each group were placed in 24-well plates. Each well was seeded with 100.000 HGFs at a density of 100.000 cells/mL and incubated at 37°C in a CO2 incubator. The samples were transferred to a new 24-well plate after 24 hours. The wells were then filled with the MTT solution and placed in the incubator to determine cell viability and proliferation [25]. After 3.5 hours, a dimethyl sulfoxide (DMSO) solution (Sigma-Aldrich Co., USA, the CAS number: 67-68-5) was added to each well. Formazan products were transferred to a 96-well plate and the absorbance was measured at a reference wavelength of 570 nm. Cell proliferation and viability were measured after 24, 48, and 72 hours. The absorbance of formazan accepted that it was linearly correlated with the number of adhered cells and MTT-labelled cells are defined by their adhesion characteristics [6, 14, 27, 42].
Scanning electron microscope analysis
A sample collection from each group was randomised using a computer-generated randomisation table to assess surface topography, cell morphology and fibroblast adhesion by SEM (ZEISS EVO® LS 10, Yildiz Technical University, Turkey). Before the imaging, the samples were coated with a 10 nm thin gold layer.
Statistical analysis
Statistical analysis of this study was performed using Statistical Package for Social Sciences version 24.00. The Kolmogorov-Smirnov test was used to assess the normality of the quantitative data. The Kruskal-Wallis test was used for multiple comparisons between groups, while the Mann-Whitney U test was used to compare the mean values of two groups. Repeated measurements of the mean values of the groups were compared with Friedman’s test. The Wilcoxon signed-rank test was used to compare the mean values of variables in any group recorded 2 times in study period, while Friedman test was used to compare the mean values of variables recorded more than 2 times. The level of statistical significance was set at p ≤ 0.05.
RESULTS
The Ra values of the individual discs in each group are shown in Table 1. The mean Ra value of group II (p = 0.001) was found to be the highest, while group I (p = 0.001) had the lowest value. There were also significant differences among the mean Ra values of the groups (p ≤ 0.05).
|
Discs |
Group I |
Group II |
Group III |
Group IV |
|
1 |
0.317 |
1.099 |
0.577 |
1.057 |
|
2 |
0.224 |
1.418 |
0.757 |
0.824 |
|
3 |
0.221 |
1.360 |
0.889 |
1.098 |
|
4 |
0.196 |
1.532 |
0.449 |
0.897 |
|
5 |
0.439 |
1.087 |
0.582 |
0.914 |
|
6 |
0.145 |
1.093 |
0.614 |
0.752 |
|
7 |
0.165 |
0.908 |
0.456 |
0.728 |
|
8 |
0.263 |
1.538 |
0.731 |
0.740 |
|
9 |
0.127 |
1.282 |
0.511 |
0.817 |
|
10 |
0.258 |
1.480 |
0.506 |
0.743 |
|
11 |
0.162 |
1.712 |
0.501 |
0.947 |
|
12 |
0.320 |
1.714 |
0.655 |
0.834 |
|
Mean ± SD |
0.236 ± 0.09 |
1.351 ± 0.26 |
0.602 ± 0.13 |
0.862 ± 0.12 |
The SEM analysis of each group showed that the surface topographies varied when the surfaces were treated with different burs. During the surface modification process, irregular structures, multiple grooves and protrusions formed on the SLA® surfaces (Fig. 1A–D). Due to the characteristics of the burs, mild surface irregularities occurred in groups III and IV.
Scanning electron microscope-generated morphologic micrographs showed that well-dispersed and evenly distributed cells that were firmly attached to the surfaces in all groups. The morphology of the cultured HGF were in close contact with each other and formed a carpet-like layer that adhered firmly to the Ti discs. The SEM images clearly showed the fibroblasts that exhibited thin and long radial extensions-marked with the arrows in the figures — from the cell nucleus zone to the periphery in groups I and II (Figs. 2, 3). But, considering the proliferation and adhesion areas revealed with the formazan crystals formed as the result of MTT assays, attached cells in group III and IV showed spindle-shaped morphology that the extensions of cells more closely located to nuclei (Figs. 4, 5).
For each study time point, there were significant differences between groups in terms of cell adhesion (p = 0.001) (Table 2). According to the pairwise comparisons, group III had the lowest adhesion rate (p ≤ 0.05), while group II had the highest rate (p ≤ 0.05). In addition, cell adhesion was significantly higher in group I than in group IV (p ≤ 0.05).
|
Standard |
Group I |
Group II |
Group III |
Group IV |
P* |
|
|
24th hours |
100 ± 0.25 99.75–100.32 (99.96) |
60.83 ± 0.73œ 60–61.54 (60.89) |
66.48 ± 0.39∆, § 66.04–66.84 (66.52) |
35.58 ± 1.61¥ 33.37–37.15 (35.89) |
51.78 ± 0.74 51.23–52.87 (51.5) |
0.001 |
|
48th hours |
100 ± 1.57 98.64–101.58 (99.88) |
61.14 ± 2.23œ 58.58–63.52 (61.24) |
69.54 ± 1.83∆, µ 67.61–71.86 (69.26) |
36.25 ± 2.22¥ 33.35–38.2 (36.73) |
52.61 ± 0.55 51.97–53.24 (52.61) |
0.001 |
|
72nd hours |
100 ± 5.95 96.08–108.82 (97.54) |
63.5 ± 1.38œ 62.14–65.24 (63.31) |
72.94 ± 5.54∆ 66.81–80.24 (72.27) |
35.91 ± 2.45¥ 32.3–37.48 (36.93) |
54.87 ± 6.37 50.93–64.36 (52.1) |
0.001 |
|
P¶ |
0.397 |
0.077 |
0.035 |
0.668 |
0.437 |
– |
When the mean cell adhesion values of the groups evaluated according to study time points, the significant difference was found only in group II (p = 0.035). The mean per cent cell adhesion values between groups showed that only the mean value of group II at 48 hours was significantly higher than that at 24 hours (p = 0.029), while it was significantly lower at 72 hours (p ≤ 0.05) (Table 2).
DISCUSSION
Several surface properties, including morphological properties, roughness, surface textures, and hydrophilicity, can affect cell adhesion and proliferation [20, 26]. Ti surface modifications improve cell surface adhesion, cell migration, and bone apposition [23]. The goal of surface modifications in dental implants is to achieve a stable implant with enhanced soft tissue adaptation [23]. The adhesion behaviour of fibroblasts is known to differ among materials with varying degrees of surface roughness. Although the rough surfaces of Ti materials may pose a greater challenge to the peri-implant tissue health, the soft tissue adaptation of these surfaces has been shown to improve significantly [28]. The study conducted by Sculean et al. [34] revealed that surface modifications have a significant effect on the surface characteristics of Ti surfaces as well as on biochemical responses and cell adhesion rates.
The results obtained from this in vitro study conducted without a bacterial model demonstrates that the adhesion potential of HGF is affected by the surface roughness of Ti. The maximum adhesion of the fibroblasts was observed in the SLA®-surface group, which also had the highest surface roughness. Additionally, the findings in implantoplasty groups showed that adhesion rates of group III was less than that of group IV, as a consequence of the less surface roughness in group III. Considering all of them, the null hypothesis was rejected in this presented study.
Several studies evaluating implantoplasty protocols showed widely varying values for surface roughness. Ramel et al. [31] found that the application of a short diamond sequence followed by silicone polishers or Arkansas stone sequences yielded in Ra values between 0.32 and 0.39 μm. In another study, implantoplasty with diamond burs and silicone polishers on roughened surfaces resulted in Ra values of 4.0 μm [31]. Another study examined diamond, tungsten, and multilaminar burs without polishers and the researchers found Ra values between 4.12 to 5.01 μm [36]. Maal et al. [25] demonstrated that carbide burs achieve smoother surfaces than diamond burs. They found that the surface roughness affects initial cell adhesion after implantoplasty and smoother surfaces enhanced fibroblast growth. Similar to previous studies, the present in vitro study without a bacterial environment showed that Ti discs with smooth surfaces had the lowest roughness values (group I) while the SLA® discs had the highest values and the highest adhesion rates of gingival fibroblasts (group II). Depending on the properties of the dental burs, group IV had a higher roughness value than group III, suggesting that the adhesion properties of cultured HGFs to Ti discs with different surface morphologies could have significant implications on the milling approaches for implantoplasty.
The findings of an in vitro study concluded that a surface roughness of less than 0.2 μm can only be achieved by implementing chairside implantoplasty protocols [11]. Additionally, surface roughness can be reduced by polishing the surface with silicone burs after milling [10]. The mean Ra values in the implantoplasty groups were lower than those in the SLA® group in our study; however, they were not close to the cut-off value of 0.2 μm. Despite extensive studies on the effects of surface roughness on the topography of the peri-implant mucosa, there are still conflicting results.
The results of animal model studies demonstrated that surface changes did not have a negative effect on the soft tissue interaction with implant surface [2, 13]. Furthermore, Schwarz et al. [33] found no differences in the probing depth of the peri-implant junction between implants surfaced with plasma sprayed and SLA®. On the contrary, human biopsy samples from acidified or oxidised junction areas of implants showed less epithelial downgrowth and longer connective tissues [8, 15]. However, in the current study, the maximum adhesion of the fibroblasts was observed in the SLA®-surface group, which also had the highest surface roughness, indicating a positive association between surface roughness and fibroblast adhesion. Based on the data we collected, group III had lower adhesion levels compared to group IV. This finding is consistent with previous studies showing that fibroblasts respond differently to materials with different surface roughness values [25, 39]. Currently, these in vitro results without a bacterial environment cannot be directly translated into clinical applications due to the differences between in vitro and in vivo conditions. In this study, we experimentally designed Ti surfaces with various surface properties. Intraoral Ti surfaces are rare due to their difficult nature, which is why the values of Ra were different from those measured in vivo.
To date, implantoplasty research has focused on surface roughness, surface coating, heat dissipation during operation, and fracture resistance [2, 25, 31, 36]. Another issue not addressed in this study was surface corrosion, which requires further investigation to determine its effectiveness. Soft tissue has been tightly adapted to the implant surface as a result of implantoplasty techniques. Ti residues released during implantoplasty may have detrimental effects on peri-implant tissues [21]. For this reason, implantoplasty should only be performed in situations where implants are supracrestally exposed or gingival recession as a result of peri-implant bone loss.
CONCLUSIONS
The surface roughness has a crucial role in fibroblast morphology and adhesion: however, it is not the only factor affecting them. Therefore; the clinicians should not forget that the surface they create during implantoplasty should have a sufficient smoothness value to prevent microbial colonisation and, at the same time, a sufficient roughness value to allow fibroblast adhesion. It should be considered that the present study has inherent limitations as an in vitro cell behaviour study, and should be designed with different implantoplasty protocols. Our study emphasizes the importance of developing a standard implantoplasty protocol that does not compromise fibroblast adhesion while providing predictable outcomes.
Acknowledgements
The authors would like to acknowledge Servo Dental, Germany, for providing Ti discs and Bilim Implant, Turkey, for measuring the surface roughness values of the materials. We also thank M.Sc. Yusuf Iseri for the linguistic support he provided.
