Statement of problem
Comparative assessment of the effectiveness of computer-aided design and computer-aided manufacturing (CAD-CAM) technologies used to fabricate complete-coverage restorations is needed. A quantitative assessment requires precise documentation of the marginal adaptation and external surface contour of fabricated restorations. Limited information is currently available regarding the effects of milling mode on marginal adaptation and reproduction of the external surface contour for CAD-CAM–fabricated restorations.
The purpose of this in vitro study was to evaluate the outcomes for 3 different digital workflows on the marginal gap and the external surface contour reproducibility of CAD-CAM–fabricated lithium disilicate complete-coverage restorations.
Material and methods
Twelve Ivorine molars were prepared to receive lithium disilicate crowns. The preparations were digitally recorded using 2 intraoral scanners (TRIOS 3; 3Shape A/S and Planmeca PlanScan; E4D Technologies), and the restorations were designed using their associated design software with reference to the anatomy of an unprepared tooth. The designed restorations were then manufactured from lithium disilicate blocks using a 3-axis milling machine. Twelve restorations were manufactured using the detailed mode (Planmeca PlanScan detailed mode [PPD-D]), and 12 using the standard mode for the Planmeca system (Planmeca PlanScan standard mode [PPD-S]). Restorations from the 3Shape system were fabricated using the detailed mode (TRIOS 3Shape detailed mode [T3S-D]). The restorations were secured on their associated preparation with an elastomeric material. The marginal gap of each restoration was then measured in the ImageJ software using images captured by a stereo microscope at ×20 magnification. External surface reproducibility was evaluated by measuring undercut at 4-line angles using a dental surveyor. Differences in the marginal gaps of restorations fabricated using the 3 different workflows were compared by Brown-Forsythe robust ANOVA, followed by a post hoc test (α=.05). Chi-square analysis (α=.05) was used to evaluate differences in the contours of the external surface of the restorations, resistance form, and marginal integrity produced using the 3 workflows.
The mean marginal gap for restorations fabricated using the T3S-D workflow was 60 μm, a distance significantly lower ( P <.05) than that of PPD-D and PPD-S workflows, which yielded a marginal gap of 95 μm for the detailed mode and 124 μm for the standard mode of milling. Restorations fabricated using PPD-D and PPD-S workflows produced a significantly more reproducible external surface contour than those fabricated using the T3S-D workflow.
Restorations fabricated using the T3S-D workflow produced the smallest marginal gap. However, reproducibility of the external surface contour for this workflow was the worst of the three workflows analyzed.
The CAD-CAM technology offers dental practitioners the ability to produce an acceptable complete-coverage restoration. These results confirm that the CAD-CAM technology may produce complete-coverage restorations with clinically acceptable marginal gaps and external surface contour, depending on the selected workflow.
Marginal adaptation is a key factor in the long-term success of restorations produced using computer-aided design and computer-aided manufacturing (CAD-CAM) technologies. Although marginal adaptation is affected by both vertical and horizontal discrepancies, horizontal discrepancies can be adjusted chairside. Thus, together with the resistance form, the marginal gap remains a major determinant of the longevity of CAD-CAM restorations. Although luting agents can compensate for discrepancies in the marginal gap and resistance form, degradation of the luting agent occurs as a result of occlusal forces and may lead to secondary caries or reduction in fracture resistance of the crowns. Luting agents fatigue as a result of microleakage, changes in the elastic modulus, and plastic deformation over time—conditions that occur under dynamic occlusal forces.
The American Dental Association specification no. 8 suggests a film thickness of 25 μm to 40 μm for luting agents. However, clinical evaluation of complete-coverage restorations suggests that typical marginal gaps, considered clinically acceptable for conventionally cemented restorations, range between 120 μm and 150 μm. Moreover, the marginal gaps of CAD-CAM crowns produced by multiple CAD-CAM systems are reported to be between 58 μm and 200 μm, suggesting that this range encompasses acceptable values. Although marginal gaps between 25 μm and 40 μm are rarely achieved, these small marginal gaps should be considered the goal.
The effects of different technologies on the accuracy of external surface reproductions based on preparation scans are considered in the establishment of embrasure and contact points with neighboring teeth, thereby impacting the cleanability of interproximal surfaces. Moreover, reproduction of the external surface affects CAD-CAM retrofitted crowns for existing removable partial dentures (RPDs). The successful use of the CAD-CAM technology has been reported in the chairside production of survey crowns fitted to the existing RPDs. However, similar to laboratory CAD-CAM systems, a more critical evaluation of chairside technologies is needed in the context of reproduction of existing survey crowns.
Multiple factors may influence the accuracy of CAD-CAM complete-coverage restorations. These include the quality of tooth preparation, optical scanner, design software settings, manufacturing process (mode of milling, number of axes in the milling machine, and diameter of rotary instruments used to cut the internal and external surfaces of the restoration), and type of material used to fabricate the restoration. The purpose of this study was to evaluate the marginal gaps and accuracy of the external surface contours of lithium disilicate crowns fabricated by using 3 digital workflows and 2 different modes of milling. The authors are unaware of previous studies assessing the accuracy of lithium disilicate crowns using various modes of milling. The null hypothesis was that no difference would be found in the marginal gap and reproduction of the external surface contour of complete-coverage restorations fabricated using the 3 digital workflows.
Material and methods
A pilot study was performed to estimate the sample size needed to evaluate the marginal gap for CAD-CAM lithium disilicate crowns. Crowns were fabricated using 2 different intraoral scanners and their associated design software and fabricated using standard and detailed milling modes. Four workflows were evaluated. The first and second workflows used the Planmeca PlanScan intraoral scanner and design software (E4D Technologies), followed by manufacturing of the designed restoration in the standard milling mode (Planmeca PlanScan standard mode [PPD-S]) or detailed milling mode (Planmeca PlanScan detailed mode [PPD-D]). The third and fourth workflows used the TRIOS 3 intraoral scanner and the 3Shape design software (3Shape A/S), followed by manufacturing of the designed restoration in the standard milling mode (TRIOS 3Shape standard mode [T3S-S]) or detailed milling mode (TRIOS 3Shape detailed mode [T3S-D]). The specimens were prepared and evaluated as described in the following paragraphs for the large-scale study. Results from the pilot study revealed that complete-coverage restorations fabricated using the T3S-S workflow required significant internal adjustment to be seated on their preparations. They had a mean marginal gap of 188 μm, a value considered unacceptable. As a result, the T3S-S workflow was eliminated from the large-scale study.
A power analysis was performed for the remaining 3 workflows using the G*Power software. The total sample size was calculated based on an effect size(f) of 0.8, an alpha error probability of 0.05, and a power of 0.9. The results of the power analysis revealed that a minimum of 8 specimens per workflow were needed to perform the study. However, to increase the power, the study was performed using 12 specimens per workflow.
Twelve Ivorine molar teeth (Kilgore International, Inc) were prepared by 1 operator (R.S.-Z.) with a 2-mm occlusal reduction with reference to the occlusal anatomy of the tooth and a 1-mm modified shoulder finishline width. The preparations were then smoothed with a finishing bur, and the line angles were rounded. Each preparation was digitally recorded using 2 intraoral scanners (TRIOS 3; 3Shape A/S and Planmeca PlanScan; E4D Technologies). Scanning was performed while the preparations were mounted on a typodont (200 Series; Kilgore International, Inc). The unprepared tooth used for the surveying procedure was scanned along with each preparation to represent the before-preparation scan. A standardized scanning procedure was used for all specimens.
Restorations were designed with reference to the scans of unprepared teeth. The “Pre-op” design was selected from the tooth library and applied to the preparation at the design tab for designing PPD-S and PPD-D crowns. No modification was applied to the external design of the restoration. Only minor smoothing at the margin was applied using the “Smooth Surface” tool. Based on common practice in fixed prosthodontics, a cement spacer (defined as “Spacer Thickness” in the software) of 50 μm was input for the occlusal and axial walls of the preparation, and a finishline width (defined as “Margin Ramp” in the software) of 0.8 mm was selected ( Fig. 1 ).
To design crowns in the T3S-D workflow, the anatomy of the unprepared tooth was morphed into the restoration design so that the definitive design followed the anatomy of the unprepared tooth. A cement spacer (defined as “Extra Cement Gap” in the software) of 50 μm and a finishline width (defined as “Distance to Margin Line”) of 0.8 mm were selected. In the design software used for the T3S-D workflow, the cement gap was set to 25 μm, and the smooth distance was set to 0.2 mm. Figure 2 shows the definition for each set of values.
Designed restorations were fabricated from lithium disilicate (IPS e.max CAD; Ivoclar Vivadent AG) using a 3-axis milling machine (PlanMill 40; E4D Technologies). A sprue was set in the mid-buccal location before sending the design to the milling machine. Designed crowns in PPD-S and PPD-D workflows were sent to the milling machine for manufacturing in 2 milling modes, standard and detailed. Designed crowns in the T3S-D workflow were sent to the milling machine using the Job Supplier Software (E4D Technologies) for manufacturing in the detailed milling mode. The PlanMill 40 uses a tapered rotary instrument (Two Striper; Premier) to mill the external surface of the restorations and 2 rotary instruments to mill the intaglio surfaces, depending on the milling mode (ellipsoidal rotary instrument for the standard mode and conical rotary instrument for the detailed mode).
After the crowns had been milled, the sprues were removed, and the restorations were inserted into their preparations. Overcontoured margins were modified after assessing the restorations on their preparations. No modifications were made on the intaglio surfaces of the restorations. The restorations were then crystallized using a processing oven (Programat CS2; Ivoclar Vivadent AG) according to the manufacturer’s instructions.
For evaluation, the marginal gap was defined as the perpendicular measurement from the internal surface of the crown to the part of the preparation closest to the finishline. Restorations were secured on their preparations with an elastomeric material (Fit Checker Blue; GC America), and finger pressure was applied for 1 minute. They were then placed under a constant load of 1.96 N for 1 minute at room temperature. A stereo-optical microscope (SMZ-U; Nikon Instruments, Inc) was used to capture images (×20) of the buccal, lingual, mesial, and distal sides of each preparation and crown. To standardize the imaging, 4 devices were fabricated for the buccal, lingual, mesial, and distal sides of the unprepared tooth using polyvinyl siloxane (Lab Putty; Coltène). Two nails were attached to a metal ruler, and the ruler was indexed in the putty. Three locations were marked on the ruler to ensure that the marginal gap was measured at the same point of each mark for all specimens. One operator (M.S.) measured marginal gaps. An imaging software program (ImageJ; the National Institutes of Health) was used to measure the marginal gap of each restoration at 3 points per side, yielding a total of 12 data points per restoration. Using tools in the imaging software, a vertical line was drawn from the defined point to the margin of the restoration, and the marginal gap was measured ( Fig. 3 ). Then, fabricated restorations for each preparation were placed in 3 identical small boxes. Boxes were placed in a bag and mixed. Restorations were then randomly picked up from the bag by 1 of the authors (M.S.), and they were assigned to a letter of A, B, or C in order. After the randomization process, a blinded and experienced clinician (R.S.-Z.) evaluated the resistance form and marginal integrity using ×3 magnification and an explorer based on the rubric presented in Table 1 . The clinician also scored the restorations from 1 to 3, with 1 representing the best clinical outcome and 3 representing the worst.
|Resistance form||Clinically acceptable: no rotation around x or y axis|
|Not clinically acceptable: rotation around x and/or y axes|
|Marginal integrity||Explorer moves smoothly during occlusogingival movement at margin|
|Explorer catches during occlusogingival movement in limited location, but marginal integrity clinically acceptable|
|Marginal integrity not clinically acceptable|
|Ranking of restorations||Rank restorations for each preparation from 1 to 3, with 1 being the best choice.|
|1. ______2. ______ 3. ______|
The external surface contour of each crown was evaluated by assessing the undercuts at the mesiobuccal, distobuccal, mesiolingual, and distolingual line angles. Four dental surveyor cast holders (Ney Lab Surveyor; Dentsply Sirona) were used to locate undercuts on the line angles of an unprepared tooth. Four typodonts (200 Series; Kilgore International, Inc) were used to tilt the cast holders to locate 0.25-mm undercuts on the mesiobuccal and distobuccal line angles and 0.5-mm undercuts on the mesiolingual and distolingual line angles using the dental surveyor undercut gauges. An autopolymerizing resin (Pattern Resin LS; GC America) was used to secure the location of each undercut using gauges placed at each line angle ( Fig. 4 ).