Marginal, Internal Fit and Microleakage of Zirconia Infrastructures: An In-Vitro Study
This in-vitro study compared the marginal and internal fit and also the microleakage of zirconia infrastructures (Procera All-Zircon, Cercon Smart Ceramics) in contrast to heat-pressed ones (Empress 2). Thirty maxillary premolars (n=30) were divided into three groups (n=10) and prepared with individual chamfer preparations by using the silicone index method. Plaster dies of 10 individual preparations were allocated to each coping fabrication method of computer-aided design and computer-aided manufacturing (Procera), computer-aided manufacturing (Cercon) and heat pressing (Empress 2) as the control. All the specimens were kept in distilled water at room temperature for four weeks after cementation with dual-curing resin cement (Variolink II, Ivoclar-Vivadent). They were then thermocycled between 5°C to 55°C for 5000 cycles with a 20-second dwell time and immersed in 0.5% basic fuchsin for 48 hours. The cemented specimens were separated into two halves vertically in the midvestibulo-palatal direction. The specimens were examined under a computer-aided stereomicroscope to evaluate both the internal and marginal fit. Marginal and internal gap widths were measured at 100× magnification. The specimens were evaluated for microleakage under a stereomicroscope at 100× magnification. Selected specimens from each group were also examined using a scanning electron microscope. Fitting accuracy data were analyzed statistically with the Welch test and the Post-hoc Dunnett C-test (p<0.05). The microleakage data were analyzed with the Kruskal-Wallis test. Special software (SPSS/PC+ Version 10.0, SPSS, Chicago, IL, USA) was used for statistical evaluations. Differences between the marginal and internal fitting accuracy of the tested non-veneered infrastructures were found to be statistically significant (p<0.05). However, there were no significant differences in microleakage among the groups (p=0.273).Abstract
INTRODUCTION
All-ceramic crowns offer excellent esthetics and have been used successfully to restore anterior and posterior teeth,1–2 with their reliable interaction with gingival tissues and biocompatibility.3–6 These restorations can be fabricated with a variety of systems.7–8 Analogous to metal ceramic crowns, the construction of ceramic crowns employs a high-strength ceramic infrastructure to provide resistance against loading. Apart from fracture resistance and esthetics, marginal and internal fitting accuracy are some of the most important parameters for the clinical quality and success of ceramic crowns.9–11 With the introduction of new techniques, the use of ceramic systems has increased.7–8 Despite all technological advances, obtaining an effective, long-lasting marginal seal at the tooth-crown interface is still a great challenge.12 The increased marginal discrepancy of a crown favors an increased rate of cement dissolution and microleakage.13 Microleakage from the oral cavity may cause inflammation of the vital pulp and pulp necrosis and potentially presents the need for endodontic treatment.14–15 Poor marginal adaptation of crowns increases plaque retention and changes the composition of the subgingival microflora, contributing to the onset of periodontal disease. Microleakage may be defined as the passage of bacteria, fluids, molecules or ions between the tooth structure and the restorative material.14–15 Clinical complications of microleakage include postoperative sensitivity, marginal discoloration and recurrent caries.14–15 The microleakage of crowns is considered to be one of the main causes of failure, that is, one of the factors that most influence the clinical longevity of indirect restorations.16
The infrastructure of ceramic crowns can be made from different high-strength ceramic materials, and various manufacturing processes can be used.17–23 The slip-casting,17 heat-pressing,18 copy-milling,19 computer-aided manufacturing (CAM),20 computer-aided design and computer-aided manufacturing (CAD-CAM)21–23 techniques have been well established for the fabrication of copings.
The fitting accuracy of metal-ceramic crowns has been documented in detail and establishes a reference point for ceramic crowns.24–25 The fit of ceramic crowns has also been studied.9–1126–39 Microleakage is also well documented in several reports and has been reviewed in the literature.14–1540–49 Based on the above data, the current study compared the marginal and internal fitting accuracy and microleakage potential of infrastructures produced with CAM and CAD/CAM systems compared to a heat-pressed one. The null-hypothesis was that the marginal and internal gap widths and microleakage of ceramic infrastructures would not be affected by the fabrication technique used.
METHODS AND MATERIALS
Specimen Preparation
Non-carious, unrestored human premolars extracted for periodontal reasons from individuals between 30 and 55 years of age were employed in the current study. The tooth specimens were stored in distilled water with a few thymol crystals at room temperature. Soft tissue and/or calculus were scaled (Scaler H6/H7; Hu-Friedy, Chicago, IL, USA) and brushed (Komet GmbH, Besigheim, Germany) with non-aromatic pumice at 6000 rpm. The cleaned specimens were inspected with a magnifying glass under a fiber optic halogen light source to detect cracks. Only premolars free of such cracks were included in the study. Thirty sound maxillary premolars were divided into two groups as small (5 mm to 6 mm) and large (7 mm to 8 mm) according to their mesiodistal diameter; they were then divided equally into three groups, so that each group had a similar number of size-matched specimens.
The roots of the teeth were embedded in auto-polymerizing acrylic resin (Imicryl SC, Imicryl Dis Malz San TicAS, Konya, Turkey) cylinders (20 mm in height and 15 mm in diameter) up to 2 mm below the cementoenamel junction, with the help of a plastic mold to facilitate the preparation and separation procedures. Cusp tips were slightly flattened on 600-grit sandpaper. The flattened cusp tips of each tooth were glued to a specially designed alignment jig. The alignment jig was used to hold the teeth perpendicular to the resin cylinder base. A pin fixator (Degudent GmbH, Hanau-Wolfgang, Germany) was used for proper positioning of the tooth, the alignment jig and the plastic mold.
Preparation Design
The teeth were prepared manually with the help of individual silicone indexes by using a high-speed instrument under water-cooling with a diamond bur set (Geneva Prep Set, Intensive SA, Lugano, Switzerland). One-millimeter wide individual chamfer preparations were made 1 mm above the cementoenamel junction. The occlusal surface was reduced by 2 mm. The complete angle of convergence was targeted to be 6°. The preparations were rounded and smoothed, except for the gingival margins.
Impression Procedure and Die Preparation
Impressions were made from each of the 30 original preparations by using vinyl-polysiloxane (VPS) impression material (Pentasoft Duo-Mix, 3M ESPE, Seefeld, Germany) mixed in an automatic mixer (Pentamix, 3M ESPE). The impressions were poured by a type IV dental stone (Silky-Rock, Whip-Mix Co, Louisville, KY, USA) for die fabrication.
Fabrication of the Infrastructures
Plaster dies were sealed with a die spacer (Cergo Spacer, Dentsply, Ceramco, NY, USA) for Cercon and IPS Empress 2 specimens. A modeling wax (Finesse all-ceram inlay wax, Ceramco, Burlington, UK) was used for the Cercon and IPS Empress II specimens. Plaster dies of 30 individual preparations were allocated to each infrastructure fabrication method, that is, CAM (Cercon Smart Ceramics, Degudent GmbH), CAD-CAM (Procera, Nobel Biocare, Göteborg, Sweden) and heat-pressing (Empress 2, Ivoclar-Vivadent, Fürstentum, Liechtenstein) as the control. The infrastructures were fabricated according to the manufacturers' instructions.
Luting Procedure
The intaglio surfaces of the infrastructures were silica-coated (CoJet, 3M ESPE), silanized (Monobond, Ivoclar-Vivadent) and dried. A bonding agent (Heliobond, Ivoclar-Vivadent) was applied and air-thinned. Also, the prepared tooth surfaces were brushed with a non-aromatic pumice, rinsed, etched for 15 seconds with 37% orthophosphoric acid (Total-Etch, Ivoclar-Vivadent), rinsed again, dried and primed (Syntac, Ivoclar-Vivadent). A bonding agent (Heliobond, Ivoclar-Vivadent) was then applied and air-thinned. Dual-curing resin cement (Variolink II, Ivoclar-Vivadent) was mixed according to the manufacturer's instructions, and the infrastructures were luted to the tooth specimens. A load of 600 grams was applied during setting of the luting material. Excess cement was removed after 10 seconds of preliminary light-polymerization and the restorations were then totally light-polymerized with an energy density of 480 mW/cm2 (Optilux; Demetron Inc, Danbury, CT, USA) for 120 seconds from all aspects of the tooth.
After the luting procedure, the specimens were stored in distilled water at room temperature for four weeks. They were then thermocycled between 5°C to 55°C for 5000 cycles with a 20-second dwell time.
Dye Penetration
The specimens were immersed in 0.5% basic fuchsin for 48 hours at 37°C. After rinsing and drying, the embedded roots were cut off for easy handling, and the coronal sections were embedded in an auto-polymerizing acrylic resin (Imicryl SC). The embedded crowns were sectioned in the midvestibulo-palatinal direction using a diamond saw (KG Sorensen, São Paolo, Brazil). The sectioned surfaces were subsequently prepared with SIC abrasive papers #800, 1000, 1200 and 2000. Finally, the surfaces were cleaned with 0.5% EDTA solution (Beyaz Dental, Izmir, Turkey) to facilitate the microscopic observations.
Marginal and Internal Gap Measurements
Both the marginal and internal fit were evaluated microscopically on the sectioned surfaces of the specimens. The measurements were made at 17 different sites using a computer-aided stereomicroscope (Eclipse ME600, Nikon, Tokyo, Japan) at 100× magnification (Figure 1). The measurement points were selected as follows: 1) buccal and lingual finishing points, 2) depth points of the buccal and lingual chamfers, 3) buccal and lingual cusp tips, 4) deepest point of the occlusal surface, 5) buccal and lingual midpoints between the cusp tip and the depth point of the chamfers and 6) another location exactly in-between the initial ones (Figure 1). Magnified images of the measurement locations were displayed on a computer monitor, and the distance between the dentin and infrastructure was measured digitally by employing computer software (Lucia G on Meteor, Version 4–51 for Nikon Laboratory Imaging, Tokyo, Japan) and recorded (Figure 2). Two measurements obtained from the buccal and lingual finishing points of the chamfer of the same specimen were used to calculate the marginal gap width, while the remaining 15 measurements of the same specimen were used to calculate the internal gap width.
Citation: Operative Dentistry 36, 1; 10.2341/10-107-LR1
Citation: Operative Dentistry 36, 1; 10.2341/10-107-LR1
Microleakage
After measurement of the marginal and internal gap, the specimens were affixed on a stereomicroscope (Stemi SV8, Zeiss, Oberkochen, Germany) to observe the basic fuchsin penetration of the whole cross section at 10× magnification (Figure 3). The dye penetration depths were scored according to Gu and Kern,41 as follows:
0: No leakage.
1: 1/3 of the chamfer preparation
2: 2/3 of the chamfer preparation
3: all of the chamfer preparation
4: more than 1/3 of the axial wall
5: more than 2/3 of the axial wall
6: all of the axial walls, including the occlusal edge
7: exceeding the occlusal edge
Citation: Operative Dentistry 36, 1; 10.2341/10-107-LR1
Scanning Electron Microscope (SEM) Analysis
The sectioned surfaces of the selected specimens were sputtered with gold, followed by dehydration in the dessicator; they were then evaluated with SEM (JSM-5200, JEOL, Kyoto, Japan) (Figures 4 and 5).
Citation: Operative Dentistry 36, 1; 10.2341/10-107-LR1
Citation: Operative Dentistry 36, 1; 10.2341/10-107-LR1
Statistical Analysis
The marginal and internal fit data were analyzed with the Welch test to disclose any statistical significance of the differences between the groups (p<0.05), and pair-wise comparisons were made using the Dunnett C-test (p<0.05). The microleakage data were analyzed with the Kruskal-Wallis and Chi-square tests to disclose the significance of the differences between the groups (p<0.05). Statistical software (SPSS/PC+ Version 10.0, SPSS, Chicago, IL, USA) was used for statistical evaluation.
RESULTS
Marginal and Internal Fit
Descriptive statistics of the marginal and internal gap width data are presented in Table 1. The marginal gap width means of the current study were measured as 43.02 ± 4.00 μm for Cercon, 47.51 ± 12.79 μm for Empress 2 and 50.29 ± 5.19 μm for the Procera groups, which yields significant differences (p<0.05) according to the Welch test. Pairwise comparisons revealed that the CAD/CAM (Procera) and the CAM (Cercon) groups differed significantly (p<0.05), while the other pairwise comparisons did not (p>0.05) (Table 2).
The internal gap width means were recorded as 57.10 ± 8.84 μm for Cercon, 65.40 ± 13.08 μm for Procera and 74.01 ± 16.41 μm in the Empress 2 groups. The differences were found to be significant (p<0.05). Pairwise comparisons revealed that the heat-pressed (Empress 2) and CAM (Cercon) groups differed significantly (p=0.05), while the other pairwise comparisons did not (p>0.05) (Table 2).
Microleakage
Descriptive statistics of the dye penetration data are presented in Table 3. The greatest mean microleakage score (1.50 ± 1.08) was obtained from the heat-pressed (Empress 2) group, while the smallest (0.80 ± 0.78) score was obtained from the CAM (Cercon) group. However the differences were not significant (p>0.05), according to the Kruskal-Wallis and Chi Square analyses (Table 4).
SEM Evaluation
SEM examination of the selected specimens showed fractured surfaces at the resin composite-zirconia interface (Figures 4 and 5).
DISCUSSION
In the current study, the least marginal gap width mean was observed in the Cercon group (43.02 ± 4.00 μm), while the greatest marginal gap mean width was from Procera (50.29 ± 5.19 μm). The mean of the marginal gap width of Empress 2 was found to be 47.51 ± 12.79 μm. The least internal gap width mean was also observed from the Cercon group (43.02 ± 4.00 μm), while the greatest gap width was from Empress 2 (74.01 ± 16.41 μm). The marginal gap width mean of Procera was found to be 65.40 ± 13.08 μm. The narrowest gaps, either marginal or internal, were found in Cercon infrastructures manufactured with the CAM technique.
The decreased amount of marginal gap width is important in ceramic crowns because of polymerization shrinkage of the resin composite cements.44 Most authors agree that a marginal gap between 100 μm and 150 μm appears to be in the range of clinical acceptance with regard to longevity.37Amarginal gap width of 1 μm to 161 μm has been reported for various types of ceramic crowns in the reviewed literature.91126–39The values of marginal gap width can be altered according to the evaluation methods. Kern and others27 demonstrated that the marginal gap of heat-pressed (IPS Empress) ceramic crowns on prepared teeth increased significantly after cementation. In an earlier in vitro study by Yeo and others,30 shoulder preparation was made on maxillary central incisors and marginal gap width was evaluated without cementing and cross-sectioning the IPS Empress II crowns. The mean marginal gap width was reported to be 46 μm ± 16 μm in their report. Similarly, in the current study, the marginal gap width of the same material was found to be 47.52 ± 12.79 μm. Jacques and others40 obtained cross-sections from the cemented IPS Empress II crowns, measured the marginal gap widths under a scanning electron microscope and reported the mean marginal gap as 80–90 μm for anterior teeth and 90–145 μm for posterior teeth. May and others28 reported the marginal and internal gap widths of silicatized and adhesively luted CAD/CAM (Procera) ceramic molar crowns to be 63 ± 13 μm and 74 ± 29 μm, respectively. Suárez and others33 reported the mean marginal gap of the same material as 38 μm. The marginal gap values in the current study appeared to be higher than the findings by Suárez and others33 but lower than the findings of May and others.28
The posterior region is the main application area for zirconia infrastructured ceramic crowns. For this reason, the current study preferred using premolars. Molars were not used due to anatomic diversities and specimen standardization difficulties. Bindl and others31 reported that there was a distinct difference between the very small marginal gap width of 17 ± 16 μm and the relatively large internal mid-mesiodistal (119 ± 49 μm) and mid-bucco-lingual (136 ± 68 μm) gap widths of the Procera crowns. Sulaiman and others11 reported the marginal gap width means of Procera and IPS Empress systems as 83 μm and 63 μm, respectively. The current study reveals narrower marginal gaps for the same materials (50.29 μm, 47.51 μm). The Cercon Smart Ceramic marginal and internal gap dimensions were in accordance with the dimensions of the heat-pressing and CAD-CAM methods throughout this assessment.
The methodological approaches of the reviewed studies focused on the fitting accuracy of the ceramic crowns, which may be divided into two main groups. These groups include measurement of the veneered ceramic crowns28–2941 and measurement of the non-veneered infrastructures.31–39 Some authors reported that the infrastructures primarily determine the overall fit of a veneered crown.11 Also, some studies have shown that various phases of porcelain firing did not significantly affect marginal adaptation of ceramic crowns.3336–39 Bindl and others31 advocated that large gaps may have been caused by margin fractures inflicted by adjusting the margin with rotating instruments or localized excess veneering ceramic. As a result, accurately fitting the infrastructures tested in the current study was performed without veneering. Therefore, the current study focused on differences between the infrastructure fabrication protocols, particularly the CAM and CAD/CAM systems. A well-known and largely accepted heat-pressed infrastructure was employed as the control.
Various laboratory devices, such as light microscopes, stereomicroscopes, digital microscopes and SEMs ranging from 80× to 225× magnification, were employed in previous studies in order to observe the fitting accuracy of the crowns.262830–3335–42 A computerized stereomicroscope was used in the current study; the measurements were made by captured digital images. Also, special software was employed for this purpose. For this reason, measurements of the current study may be considered to be free from observer failures.
The extent of marginal fit is significant in terms of the longevity of the restoration and periodontal health, because insufficient adaptation may lead to an increase in bacterial colonization, secondary caries formation and microleakage as a consequence of cement dissolution. 16
Clinical studies remain the gold standard when measuring the performance of dental restorations. However, by the time useful clinical data are obtained, the materials under investigation may have become obsolete. Thus, laboratory testing still plays an important role in the evaluation of dental materials.43 In vitro microleakage studies are initial screening methods designed to assess the sealing ability of restorative materials in vivo.43 Dye penetration remains a popular leakage test, and India ink has been used in other studies.1542 It has been reported that penetration into marginal gaps is dependent on molecular size (dye, isotope vs toxin, microorganisms), molecular polarity, surface interaction between the dye and restorative material, capillarity and time.25 In this study, microleakage was investigated using a dye leakage model with basic fuchsin. Basic fuchsin and the penetration of India ink provided a similar rank order for the sealing ability of the materials tested. Therefore, both would be able to penetrate into the cervical contraction gaps observed in vitro.44 Youngson and others45 suggested that the increased tendency of the pigment to enter dentin might be advantageous, since it could diminish the problem of differentiating true leakage from dentinal permeability. In addition, dye penetration into the dentinal tubules is a real confounding factor in penetration evaluation at the dentin/restoration interface.46 On the other hand, in-vitro microleakage tests carried out with dyes are considered to be stricter than those carried out in the oral cavity.47 Pashley14 reported possible reasons for the difference between in-vitro and in-vivo leakage studies as follows: 1) Easier diffusion of the dye, compared with bacteria and its products; 2) Possible seal of the marginal opening with the build-up of proteins and calcified debris; 3) Positive pressure of the dentinal fluid in viable teeth and 4) Opposing effect of fibrinogen inside the sectioned tubules against molecular penetration. Consequently, if a material performs well in the dye test, it is likely to performeven better on a clinical level.
In the current study, two of the most commonly used artificial aging techniques (storage in distilled water for four weeks and thermocycling) were performed. Wahab and others48 reported that thermocycling increased the amount of microleakage. After thermocycling, microleakage may be observed between the tooth and the restorative material.49 Therefore, 5000 thermocycles were performed in the current study. However, it should be considered that the response of the veneered and non-veneered infrastructures to the thermocycling procedure may not be similar.
In the current study, no significant difference was observed in microleakage at the margins in all three groups. This situation led to the opinion that resin tags in contact with the hybrid layer might have broken, and there was no boundary left between the hybrid layer and its continuation into dentin tubuli to avoid microleakage. SEM examinations supported the stereomicroscopic evaluation of microleakage.
CONCLUSIONS
Within the limitations of this in vitro study, the following conclusions were drawn:
-
The narrowest marginal gaps were measured in the Cercon group, while the wider marginal gaps resulted from Procera.
-
The narrowest internal gaps were measured in the Cercon group, while the wider internal gaps were from Empress 2.
-
The differences were found to be statistically significant.
-
Differences between the mean microleakage scores of the tested non-veneered infrastructures were found to be statistically insignificant.
Measurement points.
Digital measurement at a selected point.
Basic fuchsin invasions at 10× magnification.
35× magnification. Figure 4b: 100× magnification. Scanning Electron Microphotograph of the marginal region of a Procera specimen. CC: Crown-Coping, DB: Debonding Area, RC: Resin Composite, AT: Abutment Tooth.
35× magnification. Figure 5b: 100× magnification. Scanning Electron Microphotograph of the cusp-tip region of a Cercon specimen at 100× magnification. CC: Crown-Coping, DB: Debonding Area, RC: Resin Composite, AT: Abutment Tooth.
Contributor Notes
Levent Korkut, DDS, PhD, private practice, Izmir, Turkey
Hamit Serdar Cotert, DDS, PhD, professor, Prosthodontics, Ege University, Faculty of Dentistry, Faculty of Dentistry, Izmir, Turkey
Hüseyin Kurtulmus, DDS, PhD, research assistant, Prosthodontics, Ege University, Faculty of Dentistry, Izmir, Turkey