Effect of various polishing systems on the surface roughness and Streptococcus mutans adhesion of CAD/CAM ceramic materials

Abstract

Background

CAD/CAM ceramics provide superior predictability; however, minor adjustments are often required prior to cementation to achieve adequate interdental and occlusal contacts and to improve overall contours. These adjustments increase the surface roughness of ceramic restorations, which can promote bacterial adhesion. Therefore, this study evaluated the effects of three polishing systems (Komet, OptraGloss, EVE) on surface roughness and Streptococcus mutans adhesion to lithium disilicate and translucent zirconia ceramics.

Methods

Fifty specimens of each material (lithium disilicate and translucent zirconia) were divided into five groups (n = 10): glazed (control), ground, and ground followed by polishing with one of the three systems. The surface roughness (R_a, R_z) and topography were assessed using a contact profilometer and scanning electron microscopy (SEM). S. mutans adhesion biofilm area coverage (48-hour biofilm) was assessed as CLSM-derived area coverage (%) using confocal laser scanning microscopy (CLSM). Data were statistically analyzed using Welch’s ANOVA, the Games-Howell post hoc test, and the Kruskal-Wallis test.

Results

Compared with the other groups, the glazed groups presented significantly lower surface roughness (Ra: 0.03 μm for lithium disilicate and 0.02 μm for zirconia) and CLSM-derived biofilm area coverage (median: 0.14% for lithium disilicate and 0.15% for zirconia) (P < 0.001). Among the polished groups, the three polishing systems demonstrated significant differences in surface roughness values (Ra: 0.44–0.86 μm for lithium disilicate and 0.39–0.68 μm for zirconia; P < 0.001); however, no significant differences in S. mutans adhesion were observed (P = 1.000 for lithium disilicate; P ≥ 0.903 for zirconia).

Conclusions

Although all three polishing systems effectively reduced surface roughness after grinding, none achieved the smoothness or the level of biofilm area coverage reduction of the initial glazed surfaces.

Background

Computer-aided design and computer-assisted manufacturing (CAD/CAM) technology has revolutionized ceramic restoration fabrication, offering superior predictability compared with traditional techniques. CAD/CAM fabricated blocks demonstrate greater homogeneity and fewer flaws than conventionally layered ceramics, which are often time-consuming, technique-sensitive, and unpredictable [1]. A wide range of CAD/CAM materials are currently available on the market, including glass ceramics, hybrid ceramics, and various types of zirconia [2].

Despite the accuracy of CAD/CAM-fabricated restorations, minor chairside adjustments are often required prior to cementation to achieve adequate interdental and occlusal contacts and improve the overall contours. These intraoral adjustments inevitably increase surface roughness by removing the original glazed surface, creating a clinical challenge for optimal restoration performance. In general practice, restorations are reglazed following such adjustments. However, reglazing after adjustment is not always practical, particularly when adjustments are made intraorally after cementation [3, 4].

The surface characteristics of ceramic restorations have significant clinical implications. Roughened restoration surfaces increase the risk of plaque accumulation, discoloration of restorative materials [5], and accelerated wear of opposing teeth [6]. Furthermore, rough surfaces may promote initial crack propagation, leading to premature restoration failure [7]. According to Bollen et al., surface roughness values (R_a) should be maintained below 0.2 µm to prevent biofilm formation [8].

Although some studies have reported no significant differences in surface roughness between glazed and polished porcelain surfaces [9], different ceramic materials respond differently to various polishing techniques, highlighting the material-specific nature of surface finishing procedures [9, 10]. Furthermore, while glazed porcelain surfaces provide the smoothest finish, their long-term durability under clinical conditions remains unclear. Prevention of surface-related complications can therefore be achieved through proper polishing procedures when reglazing is not feasible.

The relationship between surface roughness and bacterial colonization remains controversial. Although numerous studies have suggested that roughened surfaces promote biofilm attachment, controversy persists regarding the precise correlation between surface roughness and bacterial adhesion [11–14].

Streptococcus mutans is a gram-positive pathogen commonly found in the oral environment and has been identified as the major etiological agent of human dental caries. A key virulence factor of S. mutans is its ability to synthesize large amounts of extracellular glucan polymers from sucrose, which aids in permanent colonization via glucosyltransferase enzymes. Various types of materials, along with their specific textures and physicochemical surface properties, influence the quantity and quality of microbial colonization [15]. Therefore, understanding bacterial adhesion and subsequent biofilm formation of S. mutans on ceramic materials after different polishing procedures is important for studying of bacterial pathogenicity and the long-term success of ceramic restorations.

Currently, the effects of various commercially available chairside polishing systems on the surface roughness of different ceramic materials and bacterial adhesion remain poorly documented in the literature. Therefore, the purpose of this study was to evaluate the effect of three polishing systems (Komet, OptraGloss, and EVE) on the surface roughness of lithium disilicate and zirconia, as well as the subsequent adhesion of S. mutans to these materials. Therefore, the purpose of this study was to evaluate the effect of three polishing systems (Komet, OptraGloss, and EVE) on the surface roughness of lithium disilicate and zirconia, as well as the subsequent adhesion of S. mutans to these materials. These specific systems were selected because they represent a cross-section of the most widely utilized chairside protocols in clinical restorative dentistry. Furthermore, they were chosen to allow for a comparison between material-specific systems (such as the EVE Diapro for lithium disilicate and Diacera for zirconia) and universal systems (such as OptraGloss), providing insight into how different abrasive matrices and universal versus material-specific protocols influence final surface quality and biological response. The null hypotheses were that the three polishing systems (Komet, OptraGloss, and EVE) would have no significant effect on (1) the surface roughness of lithium disilicate, (2) the surface roughness of zirconia, (3) S. mutans adhesion to lithium disilicate, or (4) S. mutans adhesion to zirconia.

Methods

Sample size calculation

The sample size of n = 10 specimens per group was determined a priori based on effect sizes reported in a comparable study by Incesu and Yanıkoğlu (2020) [10], which investigated polishing effects on ceramic surface roughness using a similar experimental design, and on data obtained from our pilot study. Using G*Power software (version 3.1.9.6), with a significance level of α = 0.05, a statistical power of 1 − β = 0.80, and a large effect size (f = 0.40), the minimum required sample size was calculated to be fewer than 10 specimens per group. Post hoc verification using the observed means and standard deviations from the experimental groups confirmed an effect size exceeding f = 0.40, indicating that the study was adequately powered to detect the differences observed.

Specimen preparation

A total of 100 ceramic specimens (10 × 10 × 3 mm³) were prepared from two materials: a lithium disilicate glass-ceramic (IPS e.max CAD, Ivoclar Vivadent, Schaan, Liechtenstein) and an ultra-translucent zirconia (KATANA Zirconia UTML, Kuraray Noritake, Tokyo, Japan). The specific compositions of these materials are detailed in Table 1. Specimens were sectioned from blocks and discs using a low-speed saw (IsoMet, Buehler, Lake Bluff, IL, USA) under water cooling. To standardize the initial surface conditions, all specimens were sequentially polished with silicon carbide papers (300-, 400-, 600-, and 1200-grit) under water irrigation. Following mechanical preparation, the specimens were ultrasonically cleaned in 70% ethanol for 3 min to remove debris. The lithium disilicate specimens were crystallized in a Programat P300 furnace (Ivoclar Vivadent) according to the manufacturer’s recommended firing program, while the zirconia specimens were sintered in an inFire HTC speed furnace (Dentsply Sirona, Bensheim, Germany).

Table 1.

Material specifications used in this study

Manufacturer	Type of material	Weight-% composition
IPS e.max CAD, (Ivoclar Vivadent, Schaan, Liechtenstein)	Lithium disilicate glass ceramic	57–80% SiO2, 11–19% Li2O, other oxides
KATANA Zirconia UTML, (Kuraray Noritake, Tokyo, Japan)	Ultra-translucent zirconia	87–92% ZrO2 + HfO2, 8–11% Y2O3, 0–2% other oxides

Experimental Groups

Specimens of each material (n = 50) were randomly divided into five experimental groups (n = 10 per group) based on the applied surface treatment :

• Group 1 (Control): Glazing.

• Group 2: Grinding only.

• Group 3: Grinding and polishing with ZR kit (Komet, Gebr. Brassler, Lemgo, Germany).

• Group 4: Grinding and polishing with OptraGloss (Ivoclar Vivadent, Schaan, Liechtenstein).

• Group 5: Grinding and polishing with a material-specific system:

  • Lithium disilicate: EVE Diapro (EVE Ernst Vetter, Keltern, Germany).
  • Zirconia: EVE Diacera (EVE Ernst Vetter, Keltern, Germany).

Surface Treatment Procedures

For the glazed control groups, one researcher applied a thin layer of glaze paste (AKZENT Plus GLAZE LT, VITA Zahnfabrik, Bad Säckingen, Germany) to the specimens and fired them in a Programat furnace following the manufacturer’s recommended firing program. Intraoral adjustment was simulated by grinding one side of each specimen surface with a fine-grain diamond bur (ZR 8881, Komet) using a high-speed handpiece (KaVo SUPERtorque 650, KaVo Dental, Biberach, Germany) under water cooling. Prior to grinding, the entire surface of each specimen was marked with a permanent marker. Grinding was performed until the pen mark was completely removed. The diamond bur was replaced after every 5 specimens to maintain consistent cutting efficiency. The abraded surfaces were rinsed with distilled water before proceeding to the next step. All polishing procedures were performed by a single trained researcher to ensure consistency. The polishing systems used are detailed in Table 2. Each polishing system was used with a slow-speed handpiece (INTRAmatic 20ES, KaVo Dental, Biberach, Germany) at 10,000 rpm. A standardized finger pressure of approximately 2 N was applied using linear motions for 60 s per specimen.

Table 2.

Polishing systems specifications

Polishing system (Manufacturer)	Composition	Recommended Speed (rpm)
ZR Polishing Kit Set 4622 (Komet, Gebr. Brassler)	Two-step polishing system (polyurethane binder with diamond interspersed polishers)	Optimal speed 6,000 Maximum speed 15,000
OptraGloss ceramic kit (Ivoclar Vivadent AG)	Two-step polishing system (polyurethane binder with diamond impregnated polishers)	Maximum speed 10,000
EVE Diapro and EVE Diacera (EVE Ernst Vetter)	Two-step polishing system (polyurethane binder with diamond impregnated polishers)	Optimal speed 7,000–10,000 Maximum speed 20,000

Surface roughness

Surface roughness parameters (R_a and R_z) were measured using a contact profilometer (Form Talysurf Series 2, Taylor-Hobson, Leicester, England). After the device was calibrated, the cut-off value was set to 0.8 mm, and a measurement force of 1 mN was applied at a speed of 0.5 mm/s over a 5-mm evaluation length. Three parallel measurements were performed on different areas of each specimen surface (with a 1-mm distance between each measurement), and the mean value was calculated for statistical analysis. Data were analyzed using the Shapiro–Wilk test followed by Welch’s ANOVA and the Games-Howell post hoc test for multiple comparisons (P < 0.05).

Surface topography

Prior to surface topography examination, the specimens were cleaned in an ultrasonic bath (180 W) using 70% ethanol for 30 min, followed by distilled water for 30 min, before being dried and coated with gold. The surface morphology was observed and photographed using a scanning electron microscope (SEM; JSM-6610LV, JEOL, Tokyo, Japan) at 100× magnification to qualitatively assess the surface characteristics and validate the quantitative roughness measurements.

Elemental analysis

Elemental analysis was performed using a scanning electron microscope equipped with an energy-dispersive X-ray spectrometer (EDS; X-Max, Oxford Instruments, Abingdon, UK) at an accelerating voltage of 15 kV, a working distance of 10 mm, and a magnification of 3,000×.

Bacterial adhesion

Streptococcus mutans strain (ATCC 25175) was cultured on Brain Heart Infusion (BHI) agar and incubated at 37 °C in 5% CO₂. A bacterial suspension was prepared to achieve a concentration of 3 × 10⁸ cells/mL or 1.0 McFarland standard using spectrophotometric standardization. Five specimens per group were ultrasonically cleaned for 15 min and sterilized in an autoclave at 121 °C for 30 min to ensure aseptic conditions. The sterile specimens were then placed in individual wells of 24-well plates and immersed in 5 mL of artificial saliva [16] (4 g methyl 4-hydroxybenzoate, 20 g sodium carboxymethyl cellulose, 1.25 g KCl, 0.118 g MgCl₂·6 H₂O, 0.332 g CaCl₂·2 H₂O, 1.608 g K₂HPO₄, 0.652 g KH₂PO₄) and incubated at 37 °C for 2 h in 5% CO₂ to create a pellicle layer. Following pellicle formation, specimens were rinsed with 5 mL of phosphate-buffered saline (PBS) and transferred into new sterile 24-well plates (one specimen per well). A total of 5 mL of bacterial suspension was added to each well containing the pellicle-coated specimens. Incubation was performed at 37 °C for 48 h under 5% CO₂ atmospheric conditions to allow biofilm formation. After incubation, the specimens were rinsed three times with PBS to remove non-adherent bacteria. The specimens were stained using a fluorescent probe (LIVE/DEAD BacLight Bacterial Viability Kit, L7012, Molecular Probes, Eugene, OR, USA) and incubated in the dark for 20 min. Fluorescent imaging was performed using a confocal laser scanning microscope (CLSM; Leica DMi8, Leica Microsystems, Wetzlar, Germany) at 63× magnification. The analysis utilized an Argon laser at an excitation wavelength of 488 nm (with an emission collection of 500–540 nm) for the SYTO 9 stain, and a Helium-Neon laser at an excitation wavelength of 543 nm (with an emission collection of 590–660 nm) for the propidium iodide stain. Z-stacks were processed using maximum-intensity projection. Five random fields were captured per specimen (totaling 25 areas per group) for analysis. The biofilm area coverage (%) was quantified as the CLSM-derived area fraction of fluorescent signal using ImageJ software (National Institutes of Health, Bethesda, MD, USA) as described [17]. All bacterial adhesion data derived from this method represent semiquantitative estimates of surface-associated biofilm distribution and should be interpreted accordingly. Data were statistically evaluated using the Shapiro–Wilk test followed by the Kruskal–Wallis test with Bonferroni correction.

Results

Surface roughness and topography

The mean and standard deviation values for the surface roughness (R_a and R_z) of lithium disilicate are presented in Table 3. Welch’s ANOVA revealed statistically significant differences among the groups (P < 0.05). Post hoc analysis showed that groups sharing the same letter superscript were not significantly different from each other, while groups with different letters were significantly different. Finishing with a fine diamond bur resulted in the highest surface roughness, and glazed surface showed the lowest surface roughness. After polishing with each system, the surface roughness values were reduced significantly but none of the polishing systems was able to achieve a surface smoother than glazed surface. The OptraGloss system provided a lower surface roughness value than the other two systems but was not significantly different from the Komet system. The EVE Diapro system provided the highest surface roughness values among the three polishing systems.

Table 3.

Mean (± SD) surface roughness values (R_a and R_z) of IPS e.max CAD (lithium disilicate) specimens according to surface treatment (n = 10 per group)

	Glaze	Grind	Eve	Komet	OptraGloss
R a (µm)	0.03 ± 0.00a	1.29 ± 0.07b	0.86 ± 0.09c	0.51 ± 0.10d	0.44 ± 0.09d
Rz (µm)	0.11 ± 0.03a	8.45 ± 0.50b	5.66 ± 0.62d	3.33 ± 0.64d	3.02 ± 0.77d

Different superscript letters within the same row indicate statistically significant differences among the groups (P < 0.05)

R_a Arithmetic mean roughness, R_z Mean roughness depth, SD Standard deviation

For zirconia, the mean and standard deviation values of surface roughness (R_a and R_z) are shown in Table 4. Similar statistical patterns were observed, with significant differences among groups (P < 0.05). Consistent with the lithium disilicate results, surfaces ground by a fine diamond bur had the highest surface roughness values. Surface roughness values of polished surfaces were reduced significantly. Komet system provided the lowest surface roughness values among the three polishing systems but was not significantly different from OptraGloss system.

Table 4.

Mean (± SD) surface roughness values (R_a and R_z) of KATANA Zirconia UTML specimens (zirconia) according to surface treatment (n = 10 per group)

	Glaze	Grind	Eve	Komet	OptraGloss
Ra (µm)	0.02 ± 0.00a	0.97 ± 0.22b	0.68 ± 0.22b	0.39 ± 0.06c	0.40 ± 0.12c
Rz (µm)	0.09 ± 0.00a	6.41 ± 0.1.35b	4.50 ± 1.43c	2.34 ± 0.27d	2.60 ± 0.57d

Different superscript letters within the same row indicate statistically significant differences among the groups (P < 0.05)

R_a Arithmetic mean roughness, R_z Mean roughness depth, SD Standard deviation

Scanning electron microscopy analysis

In general, SEM analysis confirmed the surface roughness results for both the lithium disilicate and zirconia surfaces. Finishing with a fine diamond bur created many deep grooves and scratches. Although all polished surfaces appeared smoother than the ground surface, some scratches still remained. The OptraGloss surfaces appeared the smoothest compared with the other two polished surfaces (Figs. 1 and 2).

Fig. 1.

Scanning electron microscopy images (100× magnification) of IPS e.max CAD (lithium disilicate) specimens following different surface treatments: (a) glazed, (b) ground with a fine-grain diamond bur, (c) ground and polished with EVE Diapro, (d) ground and polished with Komet, and (e) ground and polished with OptraGloss

Fig. 2.

Scanning electron microscopy images (100× magnification) of UTML (zirconia) specimens following different surface treatments: (a) glazed, (b) ground with a fine-grain diamond bur, (c) ground and polished with EVE Diacera, (d) ground and polished with Komet, and (e) ground and polished with OptraGloss

The SEM images of the polishing abrasive revealed diamond abrasive particles embedded within the matrices of all polishing systems. The EVE Diapro high-gloss polisher contained the largest diamond particles (~ 8.69 µm), followed by EVE Diacera (~ 4.58 µm), Komet (~ 4.38 µm), and the OptraGloss high-gloss polisher (~ 4.22 µm) (Fig. 3).

Fig. 3.

Scanning electron microscopy images (3000× magnification) of the abrasive surfaces of the high-gloss polishers used in this study: (a) EVE Diapro, (b) EVE Diacera, (c) Komet, and (d) OptraGloss

Elemental analysis

The elemental compositions of the polishing systems were analyzed by measuring their peripheral parts using energy-dispersive X-ray spectrometry (EDS) (Fig. 4). Each polishing system consisted of a two-stage protocol: a pre-polishing bur and a high-gloss polishing bur. The analysis revealed that all polishing systems contained carbon (C) and oxygen (O) as their main elements in both stages. Trace elements, which were present in smaller quantities than the main elements, varied among the different polishing systems and between the pre-polishing and high-gloss polishing stages. For EVE Diapro, the pre-polishing bur contained silicon (Si) and iron (Fe), while the high-gloss polishing bur contained silicon (Si) and Titanium (Ti). EVE Diacera’s pre-polishing bur showed titanium (Ti), Chromium (Cr), iron (Fe), and silicon (Si), whereas the high-gloss polishing bur contained iron (Fe) and silicon (Si). Komet’s pre-polishing bur contained the most diverse trace elements, including aluminum (Al), silicon (Si), titanium (Ti), manganese (Mn), and nickel (Ni), while the high-gloss polishing bur contained only titanium (Ti). The OptraGloss pre-polishing bur contained aluminum (Al), silicon (Si), Titanium (Ti), and iron (Fe), while the high-gloss polishing bur contained aluminum (Al), silicon (Si), and titanium (Ti).

Fig. 4.

Energy-dispersive X-ray spectrometry (EDS) spectra of the abrasive surfaces of each polishing system at 3,000× magnification: (a) EVE Diapro pre-polisher, (b) EVE Diapro high-gloss polisher, (c) OptraGloss pre-polisher, (d) OptraGloss high-gloss polisher, (e) Komet pre-polisher, (f) Komet high-gloss polisher, (g) EVE Diacera pre-polisher, and (h) EVE Diacera high-gloss polisher. Characteristic elemental peaks are labeled on each spectrum. C: carbon; O: oxygen; Al: aluminum; Si: silicon; Ti: titanium; Fe: iron; Mn: manganese; Ni: nickel; Cr: chromium

Bacterial adhesion

Figure 5 shows confocal laser scanning microscopy images of S. mutans adhesion on the lithium disilicate specimens. The percentage of area covered by S. mutans on the lithium disilicate specimens is presented in Table 5. Kruskal-Wallis analysis was performed for both the lithium disilicate and zirconia materials based on the quantitative assessment using ImageJ software. For the lithium disilicate group, statistical analysis showed significant differences among groups (P < 0.05). The glazed surfaces exhibited the lowest percentage of S. mutans area coverage, which was significantly lower than that of the other four groups. Although there were no significant differences among the three polishing systems and the ground group, the OptraGloss system showed the lowest bacterial adhesion.

Fig. 5.

Confocal laser scanning fluorescence micrographs (63× magnification) of S. mutans on lithium disilicate specimens: (a) glazed, (b) ground with a fine-grain diamond bur, (c) ground and polished with EVE Diapro, (d) ground and polished with Komet, and (e) ground and polished with OptraGloss

Table 5.

CLSM-derived biofilm area coverage (%) of S. mutans on IPS e.max CAD specimens by surface treatment protocol

Group	Median (%)	Interquartile range
Glaze	0.14b	0.06–0.34
Grind	8.51a	5.83–13.35
EVE	7.30a	3.00–13.98
Komet	6.38a	3.20–12.53
OptraGloss	8.11a	4.56–8.03

Different superscript letters indicate statistically significant differences among the groups (P < 0.05)

For the zirconia group presented in Table 6, statistical analysis revealed significant differences among the treatment groups (P < 0.05). Ground surfaces had the highest bacterial adhesion. Surfaces polished with the different systems showed statistically significant reductions in bacterial adhesion compared with the ground surfaces. However, there was no significant difference among the three polishing systems. Although the OptraGloss system had a higher area coverage value than the glazed surface, there was no significant difference between the OptraGloss system and the glazed surface (Fig. 6).

Table 6.

CLSM-derived biofilm area coverage (%) of S. mutans on KATANA UTML specimens by surface treatment protocol

Group	Median (%)	Interquartile range
Glaze	0.15c	0.05–0.38
Grind	2.59a	2.07–3.30
Eve	0.60b	0.31–0.95
Komet	0.74b	0.33–1.64
OptraGloss	0.33bc	0.20–0.69

Different superscript letters indicate statistically significant differences among the groups (P < 0.05)

Fig. 6.

Confocal laser scanning fluorescence micrographs (63× magnification) of S. mutans on zirconia specimens: (a) glazed, (b) ground with a fine-grain diamond bur, (c) ground and polished with EVE Diacera, (d) ground and polished with Komet, and (e) ground and polished with OptraGloss

Discussion

The results of this study confirmed that polishing systems significantly affect the surface roughness of lithium disilicate and zirconia materials, leading to the rejection of the first and second null hypotheses. Although polishing can reduce surface roughness, consistent with previous studies [18–20], it cannot achieve the same smoothness as the initial glaze. These data align with other studies [10, 21], which demonstrated that initial glazing provides lower surface roughness values compared with polishing. The observed differences in polishing effectiveness can be attributed to variations in abrasive particle composition and size among the different polishing systems. SEM analysis (Fig. 3) revealed distinct surface texture differences between the polishing systems. The EVE systems (EVE Diapro and EVE Diacera) contained larger, more irregular abrasive particles embedded in the surface compared with the Komet and OptraGloss systems, which exhibited finer, more uniform particle distributions. These results support previous studies demonstrating that polishing systems with smaller grain sizes produce smoother surfaces than those with larger particles [22] and are further supported by Silva et al., who found that abrasive particle characteristics, including size and shape, determine the resulting surface roughness [23]. Since all systems in this study shared the same polyurethane binder material, the observed differences in surface roughness can be attributed primarily to variations in abrasive particle size and distribution patterns.

Regarding bacterial adhesion, the third and fourth null hypotheses could not be rejected, indicating that the polishing systems performed similarly in terms of S. mutans adhesion within each material group. The glazed surfaces consistently showed the lowest biofilm area coverage, which was significantly lower than that of most treatment groups for both materials. Notably, for zirconia, although statistical analysis indicated no significant difference between the OptraGloss system and the glazed surfaces, CLSM images revealed greater biofilm area coverage on the OptraGloss-treated surfaces (median CLSM-derived area coverage = 0.33%) compared with glazed controls (median area coverage = 0.15%). This observation suggests that clinically relevant differences in biofilm area coverage may exist between these surface treatments, despite the absence of statistical significance. These findings support a direct relationship between surface roughness and bacterial adhesion outcomes. According to Bollen et al. [8], plaque accumulation increases when surface roughness exceeds 0.2 µm, which is considered the threshold for bacterial attachment. Surfaces smoother than this threshold generally prevents bacterial adhesion more effectively, supporting the superior plaque resistance of glazed ceramic surfaces in clinical applications.

Although surface roughness plays a crucial role in bacterial adhesion, surface chemistry also contributes significantly to this process. In addition to roughness, surface hydrophobicity or hydrophilicity significantly influences bacterial adhesion. Different bacterial species exhibit preferential interactions with particular substrate materials [24]. S. mutans, which is a hydrophobic bacterium, adheres more easily to hydrophobic surfaces [25, 26]. Ceramic finishing procedures, such as polishing and glazing, alter these surface properties. The same material can show different contact angles, resulting in either a hydrophilic or hydrophobic surface depending on the finishing method used [27]. Glazed surfaces are smooth and hydrophilic [28], which reduces bacterial adhesion. In contrast, rough surfaces tend to be more hydrophobic [27], promoting increased bacterial adhesion.

An interesting finding was that the Komet and OptraGloss systems achieved similar R_a values on zirconia (0.39 µm and 0.40 µm, respectively) but demonstrated different bacterial adhesion patterns, with Komet-treated surfaces showing higher bacterial adhesion despite comparable surface roughness measurements. This observation emphasizes that R_a alone may not fully predict bacterial adhesion, as R_a represents only the arithmetic mean of surface deviations and does not capture the complexity of surface topography. SEM analysis revealed distinct surface texture differences between these polishing systems, despite their similar R_a values. The Komet-treated zirconia surfaces exhibited more scratches and surface irregularities, which may provide more favorable sites for bacterial adhesion, compared with the OptraGloss-treated surfaces, which appeared smoother and exhibited a more uniform texture.

To understand the underlying mechanisms responsible for these differences in surface quality and bacterial adhesion, energy-dispersive X-ray spectrometry (EDS) analysis was performed on the polishing systems. The EDS results indicated that all abrasives were primarily composed of carbon (C) and oxygen (O), confirming that diamond was used as the main abrasive in the OptraGloss and Komet systems. However, significant compositional differences were observed between the two polishing systems. The high-gloss polishers (stage 2) of OptraGloss demonstrated additional elements, including aluminum (Al) and silicon (Si), suggesting the presence of silicon carbide (SiC) and aluminum oxide (Al₂O₃) alongside the primary diamond particles. This multi-component abrasive composition may explain the more refined surface finish produced by OptraGloss.

An important observation emerged from this study showing that lithium disilicate consistently demonstrated higher bacterial adhesion than zirconia across all polished surfaces. This suggests that inherent material properties, beyond surface roughness, significantly influence bacterial retention patterns. This finding is strongly supported by Bremer et al. [29], who found that lithium disilicate exhibited significantly higher biofilm formation than zirconia, even when both materials had identical surface roughness values below the bacterial threshold. The superior bacterial resistance of zirconia can be attributed to material-specific differences. Lithium disilicate has elongated crystal structures and less hydrophilic surfaces [13], while zirconia’s crystalline architecture and different surface free energy create fewer bacterial attachment sites. Studies on the in vivo perception of roughness have demonstrated that patients can distinguish between surface roughness differences of 0.25–0.50 µm [30]. Since natural enamel surface roughness was found to resemble 0.28 µm, the study concluded that restorations should have a maximum roughness of 0.50 µm to avoid detection by patients. In the context of this study, OptraGloss achieved surface roughness values below the 0.50 µm threshold for both materials, while Komet was at the threshold limit for lithium disilicate (0.51 µm) and below for zirconia (0.39 µm). EVE systems exceeded this threshold for both materials (lithium disilicate: 0.86 µm; zirconia: 0.68 µm). Glazed surfaces demonstrated the lowest values, well below the detection threshold, suggesting that OptraGloss and Komet treatments would provide patient comfort.

The clinical implications indicate that post-adjustment protocols should be material-specific based on inherent bacterial adhesion characteristics. For lithium disilicate restorations, laboratory re-glazing following chairside adjustments is recommended when clinically feasible, as this material demonstrated consistently higher bacterial adhesion across all surface treatments. In contrast, zirconia’s superior bacterial-resistant properties, demonstrated by significantly lower bacterial adhesion in this study, allow for acceptable chairside polishing protocols. While some zirconia polished surfaces may exceed both the traditional 0.2 µm bacterial adhesion threshold and the 0.5 µm tactile detection threshold, the material’s inherent resistance to bacterial colonization compensates for these surface roughness limitations. This material-specific approach allows clinicians to balance clinical efficiency with biological outcomes, utilizing zirconia’s favorable surface-bacteria interactions to justify simplified finishing protocols.

This study has several limitations that should be acknowledged. Only a limited number of ceramic brands and polishing kits were evaluated, which may not represent the full range of materials and polishing systems available in clinical practice. Although all specimens were polished by a single researcher using finger pressure with linear motions to maintain consistency, the polishing procedure was not fully standardized. The manufacturers of the polishing systems provide varying recommendations regarding speed (optimal speed, maximum speed). However, in this study, the same speed was applied across all polishing kits. Furthermore, the manufacturers do not provide specific recommendations for the pressure and polishing time to be applied, variables that may significantly affect the surface characteristics and, consequently, the results.

The in vitro nature of this study cannot fully simulate the complex oral environment, including factors such as salivary flow, pH fluctuations, temperature variations, and masticatory forces. Additionally, this research focused on a single bacterial species, Streptococcus mutans, whereas natural oral biofilms consist of multiple interacting microbial species. Consequently, the adhesion patterns observed in this monospecies model may not accurately reflect clinical biofilm dynamics. Finally, this study relied on CLSM-derived biofilm area coverage as a semiquantitative measure of bacterial adhesion, which does not provide direct information on viable bacterial counts. Although CLSM-based live/dead staining with SYTO 9 and propidium iodide is a well-established methodology for evaluating bacterial adhesion to dental surfaces in vitro [31, 32], and in situ imaging avoids the cell loss inherent to mechanical biofilm detachment methods [33], the absence of complementary quantitative data, such as colony-forming unit (CFU) enumeration or metabolic activity assays, is acknowledged as a limitation. Future studies should incorporate these additional methods to provide a more comprehensive assessment of bacterial viability and to validate the CLSM-based findings.

Conclusions

The findings of this in vitro study, despite its limitations, support the following conclusions:

1. Significant differences in surface roughness were observed among the three polishing systems for both ceramic materials tested.

2. The three polishing systems showed no significant differences in bacterial adhesion on either ceramic material.

3. Glazed surfaces demonstrated superior performance, exhibiting the lowest surface roughness and minimal biofilm area coverage for both lithium disilicate and zirconia specimens.

Authors’ contributions

T.E. and K.N. contributed to the conceptualization of the study and provided supervision. A.E. developed the methodology and prepared the manuscript. K.N. provided resources and contributed to reviewing and editing the manuscript. A.E., T.E., and K.N. performed the data analysis, interpretation, and conclusion. All authors reviewed and approved the final form of the manuscript.

Funding

Open access funding provided by Mahidol University.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.