Statement of problem
Accurate marginal and internal fit of dental restorations are essential for their long-term success. The fit of zirconia restorations prepared using digital scan systems has not been fully evaluated.
The purpose of this in vitro study was to compare the marginal and internal fit of 3-unit zirconia frameworks fabricated using direct and indirect digital scans.
Material and methods
In a maxillary model, the left first premolar and first molar were prepared to receive 3-unit zirconia fixed dental prostheses. Conventional impressions were made using stock trays and 2-step putty/wash polyvinyl siloxane material and were scanned using laboratory scanner (Conventional Impression-Laboratory scanner [CIL] group). The impressions were then poured, and the stone casts were scanned (Dental Cast-Laboratory scanner [DCL] group). Digital scans were made using TRIOS (TRIOS Intraoral scanner [TRI] group) and CS3600 (CS3600 Intraoral scanner [CSI] group) scanners (n=10). Zirconia copings were designed and milled from presintered blocks and subsequently sintered. Marginal, mid-axial, axio-occlusal, and mid-occlusal discrepancies were measured using the silicone replica technique with stereomicroscopy at ×50 magnification. The data were analyzed using 1-way ANOVA (α=.01).
The ANOVA revealed significant differences among the studied groups in terms of all studied characteristics ( P ≤.01). Marginal gap was significantly higher in the DCL group (106 ±45 μm) compared with all other groups ( P ≤.01). However, no significant differences were observed in marginal gap between the TRI (60 ±15 μm) and CSI (55 ±13 μm) groups ( P >.01). Internal gap in the mid-occlusal and axio-occlusal regions were significantly higher in the CIL (238 ±92 μm and 227 ±95 μm) and DCL (248 ±71 μm and 216 ±68 μm) groups than those recorded in the TRI (104 ±27 μm and 126 ±31 μm) and CSI (128 ±16 μm and 147 ±28 μm) groups ( P ≤.01). Internal discrepancies in the mid-axial position were similar between the TRI (70 ±15 μm) and CSI (72 ±23 μm) groups ( P >.01), but these values were significantly lower than those recorded in the CIL (88 ±31 μm) and DCL (85 ±30 μm) groups ( P ≤.01).
Within the limitations of this study, zirconia frameworks in the TRI and CSI groups had lower marginal and internal gaps compared with those in the DCL and CIL groups. Marginal gap in all groups was within a clinically acceptable range.
Given the better adaptation of zirconia frameworks prepared using direct digital scan systems, these methods can be a viable alternative to protocols using conventional impression techniques.
The longevity of indirect restorations is influenced by marginal and internal fit. Poor marginal fit contributes to restoration and abutment failure in different ways. This could lead to increased dental plaque accumulation and secondary caries, cement dissolution, and increased microleakage, periodontal disease, and even tooth loss. Moreover, inadequate internal fit could lead to missing resistance of the restoration, loss of axial retention, and decreased fracture resistance. Although the clinically acceptable value for the marginal gap varies widely, a marginal gap of 120 μm has been reported by McLean and von Fraunhofer to be clinically acceptable. The internal discrepancy has been reported to be larger in occlusal than in axial areas, and an occlusal gap of 100 to 200 μm is considered an acceptable range.
Impression accuracy is essential for a well-fitting restoration. In conventional techniques, elastomers (such as polyether or polyvinyl siloxane) are used to make impressions of prepared teeth. Although these materials have adequate precision and stability, conventional impressions have drawbacks such as the existence of voids or debris in important areas, polymerization issues, exposure of the less accurate heavy-body material through the wash material, and inadequate mixing of the materials. In addition, making conventional impressions can be uncomfortable for patients because of the gag reflex. Certain laboratory procedures can also affect the marginal fit of restorations, including the production of stone definitive casts, die hardener and spacer application, and restoration waxing.
The fabrication process of dental restorations has changed since the introduction of computer-aided design and computer-aided manufacturing (CAD-CAM) systems. This process involves 3 stages: data acquisition and digitization with scanners, designing restorations by using appropriate software, and fabrication of the restoration (CAM process) by milling or 3-dimensional (3D) printing.
The scanning process can be performed intraorally (direct) or in the laboratory (indirect). Direct data capturing uses images of a prepared tooth acquired by intraoral scanning, eliminating the conventional impression process and its errors. Patients have reported digital scans to be more comfortable. Moreover, this process has been reported to be accurate and clinically acceptable. However, the subgingival margins of prepared teeth and the presence of blood or saliva can complicate the scanning process because intraoral scanners can only record visible areas. Laboratory scanners are used to capture data by scanning a cast or an impression. However, such impression scanning can fail to capture undercuts or can cause deformation of finish lines where the impression material is thin.
Among the dental ceramics that can be processed with CAD-CAM technology, yttrium-stabilized tetragonal zirconia polycrystal (Y-TZP) has excellent strength and esthetics. Y-TZP ceramics can be milled from fully sintered or presintered blocks. As fully sintered Y-TZP has high strength, processing takes longer and wears the milling tools more rapidly. As a result, all processing steps are generally conducted with presintered zirconia and sintered after milling. However, sintering results in a linear shrinkage of 20% to 25% and can also affect the fit of a restoration. Long-span zirconia frameworks have higher marginal gap values, particularly when these are produced from presintered blocks, and the data regarding the marginal and internal discrepancy of CAD-CAM zirconia restorations manufactured using digital scan techniques remain scarce.
Accordingly, the purpose of this in vitro study was to compare the marginal and internal fit of 3-unit zirconia frameworks fabricated with CAD-CAM technology by using conventional impressions and digital scans. The null hypothesis was that no difference in terms of accuracy would be found between the marginal and internal gap of zirconia frameworks fabricated using these 2 methods.
Material and methods
Figure 1 presents an overview of the study design. In a maxillary typodont model (PRO2001-UL-SCP-FEM-32; NISSIN Dental Products), the left first premolar and first molar were prepared to receive 3-unit zirconia fixed dental prostheses. A chamfer diamond rotary cutting instrument (856F; Drendel + Zweiling Diamant GmbH) was used to prepare the supragingival chamfer margin. The preparation depth was 1.5 mm axially and 2 mm occlusally, with a 6-degree convergence. A single operator (A.A.) prepared all teeth. Ten conventional impressions were made in stock trays with a 2-step putty/wash polyvinyl siloxane material (Panasil; Kettenbach) at room temperature. Before the impression, a tray adhesive (Panasil Adhesive; Kettenbach) was applied and allowed to dry. All impressions were scanned using a laboratory scanner (Deluxe scanner; Open Technologies) with appropriate software (Optical Reveng Dental 4.0; Open Technologies) to obtain standard tessellation language (STL) files. Subsequently, Type IV dental stone (Fujirock; GC) was vacuum mixed and poured into the impressions according to the manufacturer’s instructions. The definitive casts were scanned using the same laboratory scanner.
Ten digital scans were made using the TRIOS intraoral scanner (TRIOS 2; 3Shape), and 10 digital scans were made using the CS3600 intraoral scanner (CS3600; Carestream Dental) according to the manufacturer’s instructions. The obtained STL files were sent to a commercial dental laboratory.
Four test groups were formed (n=10) based on the process used to produce the STL files: Conventional Impression-Laboratory scanner (CIL) group: the conventional impression was scanned using the laboratory scanner; Dental Cast-Laboratory scanner (DCL) group: the dental cast was scanned using the laboratory scanner; TRIOS Intraoral scanner (TRI) group: the typodont model was scanned using the TRIOS intraoral scanner; and CS3600 Intraoral scanner (CSI) group: the typodont model was scanned using the CS3600 intraoral scanner.
All copings were designed using appropriate software (Engine Build 6136; exocad GmbH) with a 0.5-mm thickness and a 35-μm simulated die spacer, which began 1 mm from the preparation finish lines. Subsequently, the copings were milled from presintered zirconia blocks (Upcera HT Zirconia; Shenzhen Upcera Co) with a 5-axis milling machine (Versamill 5X200; Axsys Dental Solutions) and were sintered according to the manufacturer’s guidelines.
The replica technique, as described by Boening et al and Almeida e Silva et al, was used to measure marginal, axial, and occlusal gaps on the premolars and molars. Light-body silicone (Panasil initial contact light; Kettenbach) was injected into the retainers, and the copings were seated onto the corresponding abutment teeth on the master model and loaded with finger pressure. After complete polymerization of the light-body silicone, the frameworks and the thin silicone layer were removed from the master model. Subsequently, a heavy-body silicone material (Panasil initial contact heavy; Kettenbach) was injected into the frameworks to stabilize the light-body layer. After setting, both layers were removed from the copings. Thereafter, the silicone replicas were sectioned mesiodistally and buccolingually to obtain 4 equal pieces for each abutment. Internal and marginal gaps were measured using a stereomicroscope (EZ4 D; Leica Microsystems) at ×50 magnification. Photographs were made with a digital camera integrated into the microscope, and measuring software (AxioVision LE 4.8; Zeiss) was used to measure the gaps. Four points were marked on each piece: marginal (M), mid-axial (MA), axio-occlusal (AO), and mid-occlusal (MO) ( Fig. 2 ). The M measurements were used to calculate the marginal gap, while the MA, AO, and MO measurements were used to calculate the internal gap. In total, 40 measurements were made for each point on each abutment. The marginal and internal gaps were calculated for each framework and compared among groups. In addition, the marginal and internal gaps of each abutment were compared among groups.
The statistical analysis was conducted based on a completely randomized design with 4 groups and 10 replicates. The raw data were initially subjected to a normality test with statistical software (IBM SPSS Statistics, v23.0; IBM Corp), and the primary data were subsequently used for 1-way ANOVA. The Duncan multiple range test was used for the intergroup comparison (α=.01).
The ANOVA revealed significant differences among the studied groups in terms of all studied characteristics ( Tables 1 and 2 ). Tables 3 to 6 present a summary of mean (±standard error of mean) gap measurements (μm) according to the impression technique used at each location. The maximum marginal gap was recorded in the DCL group (106 ±45 μm), followed by the CIL (91 ±40 μm), TRI (60 ±15 μm), and CSI (55 ±13 μm) groups. The Duncan multiple range test showed that the marginal gap was significantly higher in the DCL group compared with all other groups ( P ≤.01), but the marginal gap between TRI and CSI groups did not differ significantly ( P >.01). In addition, the marginal gap in the CIL group was significantly lower than that in the DCL group ( P ≤.01) but was significantly higher than that in the TRI and CSI groups ( P ≤.01).
|Gap Area||Sum of Squares||df||Mean Square||F||P|
|Gap Area||Sum of Squares||df||Mean Square||F||P|