Monolithic Ceramics: Effect of Finishing Techniques on Surface Properties, Bacterial Adhesion and Cell Viability
This study evaluated the morphology, biofilm formation, and viability of human gingival fibroblasts in contact with two monolithic ceramics after two different finishing techniques: polishing or glazing. For this, 92 blocks (4.5 × 4.5 × 1.5 mm) of each ceramic were made using high translucency zirconia partially stabilized by yttrium (YZHT) and lithium silicate reinforced by zirconium (ZLS). Blocks were sintered and then divided into glazing (g) or polishing (p) surface finish. Surface roughness (Ra and RSm) was evaluated through a contact rugosimeter and profilometry. Specimens were contaminated for heterotypic biofilm formation with Streptococcus mutans, Streptococcus sanguinis and Candida albicans for 16 hours. Biofilm was quantified by counting the colony forming units (CFU/mL) and analyzed by scanning electron microscopy (SEM). Fibroblast viability was evaluated by MTT assay. Surface free energy (SFE) was also determined. Roughness data were evaluated using nonparametric tests, while SFE, MTT and CFU results were evaluated by analysis of variance and Tukey test, and MTT data were also submitted to t-test (all, α=0.05). Results showed that polished samples presented a lower high profile mean (p<0.001); however, YZHTg presented less space between defects (p=0.0002). SFE showed that YZHT presented higher SFE than ZLS. Profilometry evidenced more homogeneity on polished surfaces. The interaction of finishing technique and microorganisms influenced the CFU (p=0.00). MTT assay demonstrated initial severe cytotoxic behavior for polished surfaces. SEM images showed homogeneous surfaces, except for glazed YZHT. Glazed surfaces have a greater roughness and tend to accumulate more biofilm. Polished surfaces have higher SFE; however, they are temporarily cytotoxic.SUMMARY
Introduction:
Methods and Materials:
Results:
Conclusion:
INTRODUCTION
Ceramics have become an alternative material for the manufacture of dental prostheses due to their esthetics and long-term proven resistance. With these advantages, their restorative techniques have greatly developed over time.1 Zirconia partially stabilized by yttrium (YTZP) does not present a glass phase; however, it contains a highly crystalline phase and low translucency,2 which confers opacity to visible light and the need for veneering infrastructures with esthetic ceramics.3 Monolithic materials appeared with the purpose of combining adequate translucency and excellent mechanical properties,4,5 aiming to overcome failures due to chipping of the veneering ceramic, decreasing clinical time and restoration costs.6
Recently, new formulations of zirconia reinforced lithium silicate (ZLS) ceramic materials have been introduced in the market (Celtra Duo, Dentsply, Konstanz, Baden-Württemberg, Germany; Suprinity, Vita Zahnfabrik), joining the existing group of ceramics and expanding the possibility of ceramic use in different clinical situations. According to the manufacturer, Vita Suprinity contains 56% to 64% SiO2, 15% to 21% LiO2, 1% to 4% K2O, 3% to 8% P2O5, 1% to 4% Al2O3, 0% to 4% CeO2, and 0% to 6% pigments, in addition to 10% zirconia, thereby presenting superior properties to lithium disilicate (LD), such as fracture toughness (2.31±0.17 MPa m0.5), flexural strength (443.63±38.90 MPa), and elastic modulus (70.44±1.97 GPa). As disadvantages, ZLS proved to be harder (6.53±0.49 GPa) and more friable (2.84±0.26 μm−1/2) than LD.7 When used as an infrastructure, it is possible to leave zirconia exposed to oral medium due to its biocompatibility and lower tendency to accumulate oral biofilm.8,9 However, the outcome of exposure of monolithic crowns of high translucency zirconia partially stabilized by yttrium (YZHT) and ZLS to oral fluids is not fully elucidated.
Bacterial adhesion to a substrate and the initial biofilm composition is related to topography,10-12 surface hydrophobicity,13,14 and communication between existing microorganisms.15 If the surface is hydrophilic, a water pellicle will be present, making direct contact between the hydrophobic microorganism and the substrate difficult. Streptococcus is one of the first colonizers of initial supragingival biofilm in the first 8 hours16 and is present in greater quantity in the oral biofilm.17 Due to technological and clinical advances, ceramics tend to promote excellent marginal adaptation, finishing, and polishing. Zirconia has a less homogeneous surface compared with other materials because of pores resulting from the sintering process18 or defects caused by polishing; these defects are due to the larger grains found in zirconia, and those grains susceptibility to being exposed during polishing.19 Bacteria present in the oral cavity naturally tend to adhere to ceramic materials or to the interface between tooth and restoration,11 the cervical third of the proximal surface, and along the gingival margin, where they are protected from mechanical action.20 Oral biofilm is one of the best described microbial systems,21 so it is well known that there is a mechanism for bacterial adherence and biofilm formation. On solid surfaces such as enamel, the ability to aggregate, the order of appearance of the microorganisms17 and the environment21 are important factors in oral biofilm formation. There is no consensus about the finishing technique that promotes the best surface smoothness in ceramics.22,23 In the same way, to the knowledge of the authors, no studies have evaluated the interaction between different finishing techniques on the surface properties of these new materials, the formation of oral biofilm, or human gingival fibroblast (FMM-1) viability when in contact with these monolithic ceramics.
Therefore, this study aimed to evaluate the influence of two finishing techniques (polishing or glazing) on the surface properties of two monolithic ceramics, as well as initial heterotypic biofilm formation in vitro and human gingival fibroblast (FMM-1) viability in contact with these ceramics. The null hypothesis was that surfaces resulting from polishing or glazing do not influence bacterial adhesion or FMM-1 cell viability.
METHODS AND MATERIALS
Vita YZ HT (YZHT; Vita Zahnfabrik, Bad, Säckingen, Germany; batch number 48980) and Vita Suprinity (ZLS; Vita Zahnfabrik; batch number 49142) were cut with a diamond disk in a cutting machine (Isomet 1000, Precision Sectioning Saw, Buehler, Lake Bluff, IL, USA) under constant cooling. In total, 92 specimens of each material were obtained, which were then sanded to standardize their dimensions in an automatic polisher (EcoMet/AutoMet250, Buehler) using sandpapers of decreasing grit up to #1200 (30 seconds per grit), and under water cooling. After cleaning in an ultrasonic bath with isopropyl alcohol (5 minutes), the specimens were sintered in their specific ovens. The final dimensions for both materials were 4.5 × 4.5 × 1.5 mm. Half of the blocks received a thin layer of Vita Akzent Spray HT glaze (Vita Zahnfabrik; batch number E33820) on its working side. The other half was submitted to the two step polishing protocol suggested by the manufacturer for both monolithic ceramics (VITA Suprinity Polishing Set clinical; batch number E6510). The specimens were randomly divided into four groups according to material (YZHT or ZLS) and finishing technique (g = glazing; or p = polishing), namely: YZHTg, YZHTp, ZLSg, ZLSp (Figure 1).
Citation: Operative Dentistry 43, 3; 10.2341/17-011-L
Surface Roughness (SR)
Twenty specimens from each group were analyzed by a contact rugosimeter (SJ 400, Mitutoyo, Tokyo, Japan) and a digital optical profiler (Wyko, ModelNT 1100, Veeco Instruments Inc, Tucson, AZ, USA). For roughness, five measurements were performed for each specimen in 5 random different areas with a read length of 3 mm and speed of 0.2 mm/s. The analysis was performed following ISO 4287-1997 standards, with Gaussian Filter and cut-off wavelength value of 0.8 mm. Average values were calculated for each sample, and the mean Ra and RSm (μm) values were obtained. For profilometry, specific software (WykoVision 32, Veeco Instruments Inc) was used for three-dimensional parameter measurements at 20× magnification in an area of 301.3 × 229.2 μm of two samples from each group.
Surface Free Energy (SFE)
Five samples (14×12×1.5 mm) from each group (obtained in the same way as described before) were used in conducting the SFE analysis by goniometer. An optical tensiometer (TL 1000, Theta Lite, OneAttension, Biolin Scientific, Lichfield, UK) was used to measure the mean contact angle (CAm) on five different areas by the sessile drop technique. Two liquids with different surface tensions were used: distilled water and diiodomethane,24 at room temperature. In this technique, a graduated syringe (Gastight Syringes #1001 – 1ml, Hamilton, Reno, NV, USA) with a hydrophobic needle deposits a drop, and after 5 seconds the CAm is calculated with 60 images per second over 10 seconds. The SFE (mJ/m2) was calculated according to the method proposed by Owens and Wend25 using the harmonic average formula (equations 1 and 2) and information relating to the liquids24. The CAm was replaced to isolate the dispersive and polar constants of each solid. The sum of these constants correspond to the SFE (W or γ).
Where: γ corresponds to the SFE of the liquid, cosθ = cosine of the liquid CAm, respectively for diiodomethane (1A) and water (1B). γd correponds to the dispersive energy and γp, to the polar energy. γ2d and γ2p correspond to the solid energies.
Colony Forming Units (CFUs)
Standard suspensions of Streptococcus (UA 159), Streptococcus sanguinis (ATCC 35688), and Candida albicans (ATCC 18804) were prepared containing 106 cells/mL (24 hours, 37°C). Streptococcus was cultured under microaerophilic conditions on brain heart infusion (BHI) broth supplemented with 15% glucose, and C albicans was cultured for 18 hours at 37°C in yeast nitrogen base broth (YNB; Difco, Detroit, MI, USA) supplemented with 100 mM of glucose. After incubation, the growth was suspended in a sterile physiological solution (0.9% sodium chloride [NaCl]), and the number of cells in suspension was counted in a spectrophotometer (B582 – Micronal, São Paulo, Brazil). The parameters of optical density and wavelength used were 0.620 and 398 nm for S mutans, 0.560 and 398 nm for S sanguinis and 0.284 and 530 nm for C albicans, respectively. These parameters were previously established using a standard curve for CFU vs absorbance. Eight specimens of each group were sterilized in laminar flux (15 min each side), and each one was then distributed on a sterile 96-well polystyrene tissue culture plate. Next, each plate was contaminated with the association of all microorganisms (16.5 μL of each microorganism suspension, 70 μL of YNB and 30 μL of BHI supplemented with 1% of glucose). After this period, samples were removed and washed with sterile 0.9% NaCl in order to remove weakly bonded cells. The samples were individually placed in tubes with 10 mL of sterile 0.9% NaCl and sonicated (Sonoplus HD 2200, 30 W, Bandelin Eletronic, Berlin, Germany) for 30 seconds to disperse the biofilms. The suspension obtained was diluted 10−3 times for C albicans and 10−5 times for Streptococcus. Aliquots of 0.1 mL were seeded in duplicate onto petri plates with selective medium for each microorganism, as follows: Mitis Salivarius agar (Difco) for S sanguinis, Mitis Salivarius agar (Difco) with 0.2 UI/mL of bacitracin (União Química, Sao Paulo, Brazil) and sucrose (MSBS) for S mutans, and Sabouraud dextrose agar with 50 mg/L of clorafenicol (União Química) for C albicans. The plates were incubated for 16 hours at 37°C in a CO2 chamber. Then, the plates with 30 to 300 typical colonies were counted and mean values of CFU/mL were obtained.
FMM-1 Cell Viability Assay
Cell viability was determined by measuring mitochondrial function based on its capability to reduce MTT (3-[4.5-dimethyl-thiazol-2-yl-] -2.5-diphenyl tetrazolium bromide) into a colored formazan product. Cell viability was quantified by dissolving MTT (Sigma, St Louis, MO, USA) in 0.1 N NaOH (6.25 v/v%) in DMSO (dimethyl sulfoxide), and cell survival was expressed as a percentage in relation to control group (=100%) consisting of plates without ceramic material. The standard curve was evaluated to convert optical density values to the number of viable cells, using cell densities of 2 × 104 cells/well. Twelve samples from each group were used to evaluate cell viability after 24 hours and 7 days. The medium was replaced every 48 hours over the 7 days. The cell monolayer at the bottom of the wells was washed with 500 mL of PBS. Then, 500 μl of MTT solution (0.5 mg/mL PBS) was added to each well. The plates were incubated (1 hour at 37°C) in the absence of light and supernatants were discarded. After the wells were washed with 500 μl PBS, the plates were incubated in DMSO solution (10 min, 37°C) and shaken on an orbital table (10 minutes). Finally, 100 μL of supernatant from each well was placed in triplicate in a 96-well plate and read at 570 nm (EL808IU, Biotek, Winooski, VT, USA).
Scanning Electron Microscopy (SEM)
Polished and glazed surfaces free of microorganisms, as well as contaminated specimens with initial biofilm and with FMM-1 cells, were observed and photographed by SEM (Inspect S 50 – FEI Company, Brno, Czech Republic), operating at 15 kV. Samples with cells or microorganisms were fixed for 1 hour in 2.5% glutaraldehyde, dehydrated in several ethanol washes (10%, 25%, 50%, 75%, and 90% for 20 minutes and 100% for 1 hour) and dried overnight in a bacteriologic incubator at 37°C. All samples received a gold coat in a low-pressure atmosphere using an ion sputter coater (Polaron SC 7620 Sputter Coater, Quorum Technologies, Newhaven, UK).
Statistical Analysis
Once the normality of the data using Kolmogorov-Smirnov test was confirmed, SFE results (mN/m) were statistically analyzed by two-way analysis of variance (ANOVA), CFU data (in log10) and MTT (in %) were analyzed by three-way ANOVA, all with α = 0.05, using Minitab software (Minitab 17 for Windows, 2004, State College, Pennsylvania, USA). Tukey test was used to detect differences (α=0.05). Because the distribution was not normal, roughness data (μm) were submitted to Kruskal-Wallis, Dunn, and Mann-Whitney tests (all, α=0.05). Images obtained by profilometry and SEM were qualitatively analyzed.
Results
SR
Table 1 presents the descriptive statistical analysis for roughness data. The Kruskal-Wallis test showed that both roughness parameters were influenced by finishing techniques and material (p<0.05; Table 1). The YZHTg group showed higher (Ra) and less spaced (RSm) grooves on average compared with the others. The Mann Whitney test indicated that finishing technique only influenced the Ra parameter, while material influenced both parameters (Table 2). Three-dimensional profilometry images (Figure 2) emphasize the statistical differences observed by SR between glazed and polished ceramics. Glazed surfaces are rougher than polished surfaces. The YZHTg surface was more heterogeneous than ZLSg. For polished surfaces, ZLS was more homogeneous but the presence of polishing grooves was noted on YZHT ceramic.
Citation: Operative Dentistry 43, 3; 10.2341/17-011-L
SFE
CAm data and the resulting SFE are described in Table 3. Both glazed and polished ceramics presented a predominantly hydrophilic behavior. The interaction between material and finishing technique influenced the SFE (p=0.001), and Tukey test showed that YZHT (glazed or polished) presented higher SFE than ZLS, while ZLSp and ZLSg were similar to each other.
CFU/mL
Three-way ANOVA showed a statistical difference for the interaction between finishing technique and microorganism. The mean values of CFU transformed into log 10 base for all experimental groups according to the interaction are shown in Table 4. C albicans formed fewer CFUs per milliliter on ceramics, with statistically higher adhesion on glazed surfaces.
Cell Viability Assay
The MTT assay results indicated that FMM-1 cells in early contact with the ceramics (24 hours) or with longer exposure (7 days) did not cause enough damage to characterize them as cytotoxic materials, except for the polished groups, which presented cellular viability lower than 50% after 24 hours, thus characterizing them with severe cytotoxicity, according to the International Organization of Standardization 10993-5.26 Student t-test showed difference in cell viability between both evaluated periods (24 hours and 7 days) and control group considered 100%, (p<0.05). Three-way ANOVA showed that only period of contact influenced cell growth on the ceramics (p=0.00). Tukey test identified that ceramics in contact with the cells over 7 days had a higher number of viable cells compared with the 24-hour period, regardless of material or surface treatment. The YZHTg group was the only one that presented a decrease in mean cell viability (5.1%) between 24 hours and 7 days, whereas the others presented variable increases (Figure 3).
Citation: Operative Dentistry 43, 3; 10.2341/17-011-L
SEM
The surface micrographs of sterile (1000×, Figure 2, middle column) and contaminated materials (3000×, Figure 2, right column) allowed for observing different surface patterns. The glaze layer on YZHT ceramic was less homogeneous than on ZLS. The roughness pattern generated by the polishing technique was homogeneous and similar between materials. The presence of Streptococcus and C albicans was observed on contaminated samples. An increase in the number of FMM-1 cells adhered to the materials' surface submitted to the MTT assay was observed in relation to time (Figure 4), independent of the surface morphology.
Citation: Operative Dentistry 43, 3; 10.2341/17-011-L
DISCUSSION
This study aimed to evaluate the interaction between surface properties and biofilm formation for the viability of human gingival fibroblasts. The results rejected the null hypothesis. The clinical long-term success of dental ceramics depends on their physical properties, manufacturing process, laboratory manufacturing technique, and clinical procedures. The composition of the material, as well as its surface structure, can influence the initial bacterial adhesion and compromise dental health.10,12 Ceramics are attractive restorative materials due to their esthetic quality and biocompatibility; the smooth surfaces minimize oral biofilm accumulation.13
For roughness analysis and qualitative assessments, high translucency zirconia (YZHT) presented a rougher profile than lithium silicate reinforced by zirconia (ZLS). Results from profilometry and SEM analyses corroborate YZHTg as having the highest absolute mean height of irregularities along the profile. Regardless of ceramic material, glaze application resulted in rougher surfaces. Another fact contributing to greater roughness in YZHTg may be the chemical union between glaze and zirconia, where the glaze has accumulated in islands. This accumulation causes unevenness between surface and glaze, resulting in higher Ra values for YZHT. The glaze layer on vitreous ceramics is distributed more evenly, increasing the spacing between peaks and valleys (higher RSm).
ZLSp was more hydrophilic regarding the mean contact angle (CAm) between water and ceramic. According to Shirtcliffe and others,14 a surface with CAm to water between 0 and 180° is characterized as partially hydrophilic. A surface with a hydrophobic tendency may have this feature raised by increasing the roughness,13,14 affecting its wettability and thus favoring bacterial retention.13 This was observed for ZLS which presented high values of Ra and higher CAm to water when glazed. Therefore, it is important that clinicians have knowledge about the consequences of inadequate procedures that result in rougher surfaces,27 which may be contaminated with impurities or modified by exposure to changes in temperature, which in turn may increase SFE.28 According to Anusavice,29 SFE is directly associated with adhesion. Thus, YZHTp may be suggested as the condition that results in a better adhesive property.
The presence of glaze on the surface does not prevent the formation of dental biofilm, as observed in a previous study23 that compared C albicans adhesion on a porcelain surface without surface treatment, glazed and polished. The authors verified that the glazed surface had a lower value of bacterial adhesion compared with the surface without superficial treatment, but there was no significant difference from the polished group.23 In mature biofilm, the adherence of microorganisms occurs on other layers of microorganisms. Therefore, this study simulated in vitro environmental conditions for the formation of an initial biofilm in order to evaluate the interaction between microorganisms and the surface of the materials. Recent studies have validated different periods up to 24 hours for the formation of such biofilms, using two or more microorganisms.30,31
In this study, we examined the adherence of an initial colonizer (S sanguinis), a colonizer associated with the development of carious lesions (S mutans), and finally, a colonizer related to caries, periodontal diseases, and candidiasis (C albicans). The availability of studies evaluating bacterial adhesion to monolithic ceramics is scarce. The adhesion of S mitis and Prevotella nigrescens on Metoxit AG zirconia (High Tech Ceramics, Thayngen, Switzerland), for example, is lower than on titanium used in manufacturing dental implants (Goodfellow Cambridge Limited, Huntingdon, UK).31 A previous study verified that glazed Lava zirconia (3M ESPE, St. Paul, MN, USA) presented greater roughness compared with the polished surface, and there was a tendency toward biofilm accumulation.8 When microorganisms were compared to each other, a greater growth of Streptococcus was observed independent of the surface type. Also, a greater formation of colonies on glazed surfaces was observed in comparison to polished surfaces, corroborating a previous study.8 This may be associated with the fact that S sanguinis facilitates the growth of other Streptococcus that grow in a similar way, which can be justified by their hydrophobic nature.31,32 C albicans also presents a hydrophobic characteristic; however, its smaller growth may be associated with the fact that Streptococcus is a commensal microorganism,21 where both species are associated with benefits for one of them without harming the other. In the ceramic structure, C albicans acts as a facilitator for the adherence of S mutans.33 The low growth of C albicans on polished surfaces can be justified by its difficulty in adhering to very smooth surfaces compared with S mutans,34 since the second produces a water-insoluble substance that facilitates the adherence of these microorganisms to a smooth substrate.35 Different from rough surfaces, polished surfaces do not accumulate many nutrients. This dispute over scarce food, as well as the lack of space and negative effect of metabolites from bacteria, can also justify a competition between C albicans and Streptococcus. The interaction between S mutans and C albicans is given by mutualism, where microorganisms benefit, resulting in mutual dependence.36 It is interesting to observe the small number of C albicans colonies because healthy tissue free of fungus in restorations where the dental preparation has contact with the gingival tissue is necessary. A lower number of C albicans colonies on smooth surfaces has also been reported for other materials used in the oral cavity.35 The results show that a higher number of microorganisms adhered to the roughened surfaces, presenting less SFE than polished surfaces. In this way, the roughness seems to be the main factor related to biofilm formation. Considering smooth surfaces, it is suggested that SFE may be the main factor associated with initial bacterial adherence.31,34
It is possible to verify the presence of Streptococcus and C albicans (Figure 2, right column) in the SEM images. The ceramics were colonized with a thin biofilm filled with cellular agglomerates of similar size and morphology, with an emphasis on the extensive colonization by Streptococcus, and it was not possible to distinguish Streptococcus. The amorphous substance is an important factor in the relationship of all studied microorganisms. In Figure 2 (left column), SEM images show this amorphous matrix involving the species, visibly larger on ZLSg. This matrix may be associated with the adhesion of C albicans to the biofilm. The presence of this fungus corroborates the assertion that this facilitates the adherence of S mutans, and may be associated with an increased risk of caries. The results show that both materials under both finishing techniques, can be considered moderately cytotoxic26 to the growth of human gingival fibroblasts (FMM-1), since all groups presented cellular viability between 50% and 79%. The initial (24 hour) cytotoxicity of polished groups may be related to the release of some substance at this initial time, reducing its cytotoxic effect after 7 days. This initial cytotoxicity may occur if the cells do not present sufficient immediate defense to some remnant of the polishing procedure. Over time, the cells enhance their defense mechanisms and become capable of protecting themselves from the aggressor. Therefore, future studies evaluating which substances are released causing tissue damage are important.
CONCLUSION
ZLS resulted in lower mean roughness profile and more spaced defects regardless of surface finishing. Polished surfaces were less rough and presented higher SFE, but they also showed severe initial cytotoxicity when in contact with FMM-1 cells. However, they were inert in the long term.
Flow chart of tested samples. After nondestructive analysis (surface roughness and profilometry), samples were reused. In the surface free energy analysis, 14 × 12 × 1.5 mm blocks were used (grey boxes).
Images from three-dimensional profilometry, SEM of the surfaces and SEM of contaminated surfaces, respectively, for the groups (a-c) YZHTg, (d-f) YZHTp, (g-i) ZLSg, and,(j-l) ZLSp.
Bar graph for percentage of viable cells through MTT assay at 24 hours and 7 days.
SEM images of FMM-1 cells in contact with the ceramics YZHTg (a-b), YZHTp (c-d), ZLSg (e-f), and ZLSp (g-h) for the periods of 24 hours and 7 days, respectively. Magnification of 1000 ×.
Contributor Notes
Samira Esteves Afonso Camargo, DDS, MSc, PhD, Department of Biosciences and Oral Diagnosis, São Paulo State University – Unesp, Institute of Science and Technology, São Jose dos Campos, Brazil