Forty sound freshly extracted mandibular premolars were used in the study. Dental calculus was removed with periodontal curettes and teeth were stored in distilled water at room temperature. The research was approved by the Ethics Committee of the Federal University of Santa Catarina (#850.087, 2014).
Teeth were marked with a millimeter probe 2 mm below the cement-enamel junction to simulate the periodontal ligament.[20 (link)] Then, they were placed in a container with heated utility wax, forming a thin layer of 0.3 mm, and its thickness was verified with an adapted millimeter probe. After that, they were positioned along their long axis in PVC (Polyvinyl Chloride) cylindrical devices and embedded in self-cure acrylic resin, to simulate the alveolar bone. Once the acrylic resin was completely cured, the teeth were removed, leaving an alveolus-like space. A polyether adhesive (Polyether Adhesive, 3M ESPE) was applied over the roots and 15 min was allowed to pass. Then, they were covered with a 0.3 mm layer of a polyether impression material (Impregum Soft, 3M ESPE) to simulate the periodontal ligament. The teeth were then returned to the acrylic resin mold. After 6 min, the polyether excess was removed, completing the periodontal ligament simulation [Figure 1 ].
Then, the teeth were randomly divided into four groups (n = 10), according to the esthetic veneer restorative technique to be performed. This sample size was determined based on previous studies, which conducted similar investigations and used the same sample size.[22 (link)23 (link)24 (link)] NPR = teeth without dental preparation veneered with resin composite (Amelogen Plus shade A2, Ultradent). The composite resin veneer, with 0.2 mm thick, was extended across the buccal surface and involved 1 mm of the occlusal surface of the buccal cusp; NPC = unprepared teeth veneered with 0.2 mm thick lithium disilicate glass ceramic (IPS e.max Press A1 HT, Ivoclar Vivadent) extended across the buccal surface, involving 1 mm of the occlusal surface of the buccal cusp. P2C = teeth with a 0.2 mm dental preparation on the buccal surface and occlusal reduction of 0.2 mm, veneered with 0.2 mm thick lithium disilicate glass-ceramic (IPS e.max Press, Ivoclar Vivadent). The veneer was extended across the buccal surface and involved 1 mm of the occlusal surface of the buccal cusp; P5C = teeth with 0.5 mm dental preparation on buccal surface, occlusal reduction of 0.5 mm, and veneered with 0.5 mm thick lithium disilicate glass-ceramic (IPS e.max Press, Ivoclar Vivadent). The veneer extended across the buccal surface and involved 1 mm of the occlusal surface of the buccal cusp [Figure 2 ].
The bonding procedure and the restorative protocol used in the nonprepared group (NPR) are shown inTable 1 .
Polyvinyl siloxane (Express XT Putty, 3M ESPE) impressions were taken from the teeth of the P2C and P5C groups to fabricate horizontal and vertical guides to be used in the bur preparations steps, aiming to standardize the preparations' depth.
Tapered diamond burs were used for dental preparation in occlusal and buccal surfaces, according to each group depth, aided by adapted customized-periodontal probe, and verified with a digital caliper [Figure 3 ]. A high-speed handpiece turbine (~200,000 rpm) (T3 LINE E 200, Dentsply Sirona) was used with constant water refrigeration. Dental preparations finishing steps were performed with fine and extra-fine granulation tapered diamond burs.
Impressions were obtained from all teeth of NPC, P2C, and P5C with a single-step impression technique. A light-body (Express XT, 3M ESPE) and a heavy-body (Express XT Putty, 3M ESPE) polyvinyl siloxane material was used. Then, they were sent to the dental technician to make the lithium disilicate glass-ceramic veneers (IPS e.max Press, Ivoclar Vivadent) according to the manufacturer's instructions.
Glass-ceramic veneers were tried in teeth of NPC, P2C, and P5C groups. In all these groups, teeth were cleaned with pumice paste and a rubber cup. All luting procedures were performed by the same operator (LAL) using the following protocol.
The ceramic veneers were etched with 9.6% hydrofluoric acid for 20 s, then, rinsed with water for 30 s, air-dried, and ultrasonically cleaned with distilled water for 5 min. A silane-coupling agent (Silane Primer, Kerr) was applied in the internal surface of the ceramic and remained for 60 s [Figure 4 ]. Buccal and occlusal dental areas were acid-etched with 37.5% phosphoric acid (Gel Etchant, Kerr) for 30 s, rinsed with air–water spray, and gently air-dried. Once the dental surface was acid-etched, a light-cure adhesive system (OptiBond FL Adhesive, Kerr) was applied with a disposable applicator, and it was not light-cured at this step. In cases where it was possible to identify the presence of exposed dentin in the cervical dental preparation area, by a visible contrast compared to white-opaque acid-etched enamel aspect, a hydrophilic primer (OptiBond FL Primer, Kerr) was applied with a disposable applicator over dentin with gentle movements for 15 s and air-dried during 5 s. Next, hydrophobic adhesive resin (OptiBond FL Adhesive, Kerr) was applied with a disposable applicator for 15 s, creating a thin layer and then gently air-dried. When there was not visible exposed dentin, only the hydrophobic adhesive resin (OptiBond FL Adhesive, Kerr) was applied [Figure 5 ].
A light-cure resin cement (Nexus 3 Light-Cure, Kerr) was applied in the inner surface of the ceramic veneer. The ceramic veneer was placed with light finger pressure onto the dental area; the resin cement excesses were removed with an angled dental probe parallel to the restoration margin and light-cured with a LED unit (Translux Power Blue, Heraeus Kulzer) with a light intensity of 550 mW/cm2 within occlusal and buccal surfaces for 60 s each surface. The polishing procedure was carried out after 24 h using a sequence of abrasive rubber points (Astropol, Ivoclar Vivadent) [Figure 6 ].
An aging process was performed by thermocycling procedure. It consisted of 10,000 cycles of water baths, with a temperature variation from 5°C to 55°C and a dwell time of 30 s on each bath.[25 (link)] Then, samples were fixed in a universal testing machine (Instron 4444, Instron Corporation) and subjected to the fracture resistance test under compression force. The test was performed with a speed of 0.5 mm/min using a 2 kN maximum load perpendicular to the buccal surface of direct or indirect veneers, until a complete or partial fracture of the samples. The force was applied through a composite resin (Filtek Z100, 3M ESPE) sphere device with 7 mm diameter [26 (link)] adapted in the universal testing machine to simulate an antagonist tooth cusp [Figure 7 ]. The load at failure, in Newton (N), required to fracture each sample was recorded and subjected to statistical analysis. Data were analyzed with Shapiro–Wilk normality test, one-way ANOVA, and Duncan multicomparison post hoc test (P < 0.05).
After failure, samples were analyzed to determine the mode of failure under ×10 magnification with a magnifier (YC-86C, YPT, Guangdong, China). According to Schmidt et al.,[24 (link)] the failure modes were classified into four types. Type 1: cohesive failure in restorative material (in this type of failure, fractures are restricted to the ceramic and/or in the composite resin veneer, not involving the dental structure); Type 2: mixed failure (adhesive and cohesive in restorative material); Type 3: adhesive failure (failure in the tooth/veneer interface); and Type 4: root fracture [Figure 8 ].
Teeth were marked with a millimeter probe 2 mm below the cement-enamel junction to simulate the periodontal ligament.[20 (link)] Then, they were placed in a container with heated utility wax, forming a thin layer of 0.3 mm, and its thickness was verified with an adapted millimeter probe. After that, they were positioned along their long axis in PVC (Polyvinyl Chloride) cylindrical devices and embedded in self-cure acrylic resin, to simulate the alveolar bone. Once the acrylic resin was completely cured, the teeth were removed, leaving an alveolus-like space. A polyether adhesive (Polyether Adhesive, 3M ESPE) was applied over the roots and 15 min was allowed to pass. Then, they were covered with a 0.3 mm layer of a polyether impression material (Impregum Soft, 3M ESPE) to simulate the periodontal ligament. The teeth were then returned to the acrylic resin mold. After 6 min, the polyether excess was removed, completing the periodontal ligament simulation [
Then, the teeth were randomly divided into four groups (n = 10), according to the esthetic veneer restorative technique to be performed. This sample size was determined based on previous studies, which conducted similar investigations and used the same sample size.[22 (link)23 (link)24 (link)] NPR = teeth without dental preparation veneered with resin composite (Amelogen Plus shade A2, Ultradent). The composite resin veneer, with 0.2 mm thick, was extended across the buccal surface and involved 1 mm of the occlusal surface of the buccal cusp; NPC = unprepared teeth veneered with 0.2 mm thick lithium disilicate glass ceramic (IPS e.max Press A1 HT, Ivoclar Vivadent) extended across the buccal surface, involving 1 mm of the occlusal surface of the buccal cusp. P2C = teeth with a 0.2 mm dental preparation on the buccal surface and occlusal reduction of 0.2 mm, veneered with 0.2 mm thick lithium disilicate glass-ceramic (IPS e.max Press, Ivoclar Vivadent). The veneer was extended across the buccal surface and involved 1 mm of the occlusal surface of the buccal cusp; P5C = teeth with 0.5 mm dental preparation on buccal surface, occlusal reduction of 0.5 mm, and veneered with 0.5 mm thick lithium disilicate glass-ceramic (IPS e.max Press, Ivoclar Vivadent). The veneer extended across the buccal surface and involved 1 mm of the occlusal surface of the buccal cusp [
The bonding procedure and the restorative protocol used in the nonprepared group (NPR) are shown in
Polyvinyl siloxane (Express XT Putty, 3M ESPE) impressions were taken from the teeth of the P2C and P5C groups to fabricate horizontal and vertical guides to be used in the bur preparations steps, aiming to standardize the preparations' depth.
Tapered diamond burs were used for dental preparation in occlusal and buccal surfaces, according to each group depth, aided by adapted customized-periodontal probe, and verified with a digital caliper [
Impressions were obtained from all teeth of NPC, P2C, and P5C with a single-step impression technique. A light-body (Express XT, 3M ESPE) and a heavy-body (Express XT Putty, 3M ESPE) polyvinyl siloxane material was used. Then, they were sent to the dental technician to make the lithium disilicate glass-ceramic veneers (IPS e.max Press, Ivoclar Vivadent) according to the manufacturer's instructions.
Glass-ceramic veneers were tried in teeth of NPC, P2C, and P5C groups. In all these groups, teeth were cleaned with pumice paste and a rubber cup. All luting procedures were performed by the same operator (LAL) using the following protocol.
The ceramic veneers were etched with 9.6% hydrofluoric acid for 20 s, then, rinsed with water for 30 s, air-dried, and ultrasonically cleaned with distilled water for 5 min. A silane-coupling agent (Silane Primer, Kerr) was applied in the internal surface of the ceramic and remained for 60 s [
A light-cure resin cement (Nexus 3 Light-Cure, Kerr) was applied in the inner surface of the ceramic veneer. The ceramic veneer was placed with light finger pressure onto the dental area; the resin cement excesses were removed with an angled dental probe parallel to the restoration margin and light-cured with a LED unit (Translux Power Blue, Heraeus Kulzer) with a light intensity of 550 mW/cm2 within occlusal and buccal surfaces for 60 s each surface. The polishing procedure was carried out after 24 h using a sequence of abrasive rubber points (Astropol, Ivoclar Vivadent) [
An aging process was performed by thermocycling procedure. It consisted of 10,000 cycles of water baths, with a temperature variation from 5°C to 55°C and a dwell time of 30 s on each bath.[25 (link)] Then, samples were fixed in a universal testing machine (Instron 4444, Instron Corporation) and subjected to the fracture resistance test under compression force. The test was performed with a speed of 0.5 mm/min using a 2 kN maximum load perpendicular to the buccal surface of direct or indirect veneers, until a complete or partial fracture of the samples. The force was applied through a composite resin (Filtek Z100, 3M ESPE) sphere device with 7 mm diameter [26 (link)] adapted in the universal testing machine to simulate an antagonist tooth cusp [
After failure, samples were analyzed to determine the mode of failure under ×10 magnification with a magnifier (YC-86C, YPT, Guangdong, China). According to Schmidt et al.,[24 (link)] the failure modes were classified into four types. Type 1: cohesive failure in restorative material (in this type of failure, fractures are restricted to the ceramic and/or in the composite resin veneer, not involving the dental structure); Type 2: mixed failure (adhesive and cohesive in restorative material); Type 3: adhesive failure (failure in the tooth/veneer interface); and Type 4: root fracture [