Effect of different CAD-CAM materials on the marginal and internal adaptation of endocrown restorations: An in vitro study


Statement of problem

Recent resin-based and ceramic-based computer-aided design and computer-aided manufacturing (CAD-CAM) materials have been used to restore endodontically treated teeth. Adaptation of the restoration is important for clinical success, but studies evaluating the effect of these materials on the adaptation of endocrowns are lacking.


The purpose of this in vitro study was to evaluate the effect of resin-based and ceramic-based materials on the marginal and internal adaptation of endocrowns.

Material and methods

Forty mandibular molars were divided into 4 groups (n=10); each group was restored with a different CAD-CAM material: group C: hybrid nanoceramic (Cerasmart; GC Corp), group T: fiber-composite material (Trilor; Bioloren Srl), group E: lithium disilicate glass-ceramic (IPS e.max CAD; Ivoclar Vivadent AG), and group V: zirconia-reinforced lithium silicate glass-ceramic (Vita Suprinity; VITA Zahnfabrik GmbH). A digital scan was made with an intraoral digital scanner (TRIOS 3; 3Shape A/S), and endocrowns were milled with a 5-axis milling machine (Coritec 250i; imes-icore GmbH). The replica technique and a stereomicroscope (×70) were used to measure the marginal and internal adaptation of the endocrowns at 32 points. All data were statistically analyzed using 1-way ANOVA and the Tukey honestly significant difference test (α=.05).


Statistical tests showed significant differences among the tested groups ( P <.001). The resin-based groups displayed larger discrepancies than the ceramic-based groups. The resin-based groups showed a mean marginal gap larger than the mean internal gap C ( P =.009), T ( P <.001), whereas the ceramic-based groups showed similar gaps, V ( P =.396), E ( P =.936). The largest gap was observed at the pulpal floor ( P <.001).


All materials had clinically acceptable internal and marginal gaps (≤150 μm), except for the marginal gap of the Trilor group.

Clinical Implications

The marginal and internal discrepancies of CAD-CAM endocrowns changed depending on the material (ceramic based or resin based) used. Ceramic-based, especially lithium disilicate glass-ceramic, showed the smallest gap, which might improve the clinical survival of the restored tooth.

Advances in adhesive dentistry, computer-aided design and computer-aided manufacturing (CAD-CAM) technologies, and ceramic materials have resulted in the introduction of new systems of dental restorations, including the endocrown restoration, which reduces the risk of failure during intracanal post preparation. An endocrown is a monobloc restoration that combines the crown and the core as a single unit. It covers all cusps with a circular shoulder margin and extends toward the pulpal floor. Endocrowns use the available surface provided by the axial walls of the pulp chamber as macromechanical retention, while the adhesive resin cement acts as micromechanical retention. Endocrowns have been reported to be successful restorations for endodontically treated molars with extensive loss of coronal structure.

Adaptation is one of the most important factors for the success of any restoration. Poor marginal adaptation increases plaque accumulation, which could lead to secondary caries, periodontal disease, and endodontic inflammation. A thick cement layer increases polymerization shrinkage and interfacial stresses, which in turn can reduce the fracture resistance of ceramic restorations. Holmes et al defined the internal gap as “the perpendicular measurement from the internal surface of the casting to the axial wall of the preparation” and the marginal gap as the same measurement at the margin. Another important measurement is the absolute marginal discrepancy (AMD), which has been defined as the “angular combination of the marginal gap and the extension error (overextension or underextension); AMD is the combination of the vertical and horizontal marginal discrepancies.”

The marginal and internal adaptation is influenced by variables such as the preparation design, fabrication technique, methods of gap measurement, and the materials used. CAD-CAM technology allows the use of new restorative materials and chairside fabrication and improves esthetics and fit. It also simplifies the process as compared with conventional preparation methods. Different techniques for evaluating adaptation have been described, including the replica technique (RT), which is a nondestructive and reliable method for in vivo and in vitro studies in which an impression of the cement space is measured. Because the retention of endocrown restoration relies mainly on bonding, materials which can be acid etched and resin bonded to tooth tissue are necessary. Various materials are available for endocrowns with different compositions and physical properties, including lithium disilicate glass-ceramics, hybrid nanoceramics, fiber-composites, and zirconia-reinforced lithium silicate glass-ceramic.

The effect of the intrapulpal extension of endocrowns on marginal and internal adaptation has been evaluated, but limited data are available on choosing the best material for endocrown restorations. The authors are unaware of studies on the influence of different materials on the marginal and internal adaptation of endocrown restorations.

The purpose of this in vitro study was to compare the marginal and internal adaptation of endocrown restorations fabricated from 4 different CAD-CAM materials: hybrid nanoceramic, fiber-composite material, lithium disilicate glass-ceramic, and zirconia-reinforced lithium silicate glass-ceramic. The null hypotheses tested were that no difference would be found between cervical, axial, pulpal, and marginal fit within each group; and that no difference would be found between internal fit (cervical, axial, and pulpal) and marginal fit among the 4 groups.

Material and methods

This research was approved by the ethical committee of the Faculty of Dental Medicine, Lebanese University, Beirut, Lebanon (CUMEB/D123/102018). Forty extracted permanent mandibular molars were selected with nearly similar size. Mesiodistal and buccolingual dimensions were measured at the cemento-enamel junction (CEJ), with a maximum deviation of 10%. The inclusion criteria were absence of carious lesions or cracks and complete root formation. The teeth were ultrasonically cleaned and stored in 0.5% chloramine–T solution at 10 °C. They were then sectioned perpendicular to their long axis, 2 mm above the CEJ, to open the pulp chamber. The pulp tissues were removed, and the root canals were enlarged with NiTi rotary instrumentation (ProTaper Universal; Dentsply Sirona) and irrigated with NaOCl (5.25%). The root canals were obturated using the warm vertical condensation technique with system B (Sybron Endo; Henry Schein, Inc), gutta percha (Calamus Dual; Dentsply Sirona), and root canal sealer (AH 26; Dentsply Sirona).

Each tooth was fixed vertically in the metal holder of a dental surveyor (Marathon-103; Saeyang) by injecting a photopolymerizable gingival barrier (Laser protect; DMC Dental) in the access opening and inserting the root in autopolymerizing acrylic resin (Fastray; Harry J. Bosworth Co) using cylindrical molds 2 mm below the CEJ to simulate bone level. The gutta percha was removed from the entrance of each root canal using a small tungsten carbide rotary instrument, flowable composite resin (Filtek Z350XT Flowable; 3M ESPE) was then used to fill the canals, and the base of the pulp chamber was flattened at a depth of 4 mm from the occlusal floor. All teeth were prepared in a standardized manner under water spray with the aid of a dental surveyor (Marathon-103; Saeyang) and an 8-degree tapered diamond rotary instrument (#856; Intensiv SA). The excessive retentive areas were removed, the pulpal walls were aligned, and internal angles were rounded. The preparations were smoothed using fine polishing rotary instrument (#504; Intensiv SA).

The teeth were divided into 4 groups (n=10) according to the tested materials, each of which was adequate for a statistical power of 80% (G*Power : group C, hybrid nanoceramic (Cerasmart; GC Corp); group T, fiber-composite (Trilor; Bioloren Srl); group E, lithium disilicate glass-ceramic (IPS e.max CAD; Ivoclar Vivadent AG); and group V, zirconia-reinforced lithium silicate glass-ceramic (Vita Suprinity; VITA Zahnfabrik GmbH). The compositions and mechanical properties of the tested materials are listed in Table 1 .

Table 1
Type, manufacturer, composition, and mechanical properties of 4 tested materials
Material Code Manufacturer Ceramic Type Composition Modulus of Elasticity (GPa) Flexural Strength (MPa) Vickers Hardness (MPa)
Cerasmart C GC Corp Hybrid nanoceramic Ceramic network: 71%
Resin matrix: 29%
20 240 700
Trilor T Bioloren Srl Fiber-composite Multidirectional fibers and resin matrix 26 380 1500
IPS e.max CAD E Ivoclar Vivadent AG Lithium dissilicate glass-ceramic Glass-ceramic 95 400 6000
Vita Suprinity V Vita Zahnfabrik Zirconia-reinforced lithium silicate glass-ceramic ZrO 2 ≈ 10%
SiO 2 ≈ 60%
Li 2 O ≈ 18%
Pigments≈ 10%
70 420 7000

All prepared specimens were scanned with an intraoral scanner (TRIOS 3; 3Shape A/S). The digital scans were saved as 40 standard tessellation language (STL) files for the 40 specimens. The appropriate design software (2017; 3Shape Dental System) was used to design the endocrowns on the virtual model. The operator determined design parameters as the thickness of the spacer, which was set to be 10 μm on the marginal discrepancy and 40 μm on the internal discrepancy; all restorations were designed to have similar occlusal anatomy and the same occlusogingival height. The virtual endocrowns were converted to 40 STL files; they were milled under wet processing with a 5-axis milling machine (Coritec 250i; imes-icore GmbH) and CAD-CAM materials: Cerasmart (block: A2 HT/14L), Trilor (disk: diameter=98 mm, height=14 mm), IPS e.max CAD (block: LT A2/C14), and Vita Suprinity (block T A2/LS14). The milling rotary instruments were changed for each group, and the size of the smallest was 0.6 mm. Following the manufacturer instructions, specimens in group E (IPS e.max CAD) and group V (Vita Suprinity) were subjected to the crystallization process (Vita Vacumat 6000 M; VITA Zahnfabrik GmbH), while specimens in group C (Cerasmart) and group T (Trilor) did not need any crystallization firing. All endocrowns were seated on the corresponding teeth and evaluated with an explorer and a silicone indicator paste (Fit Checker; GC Corp) for adaptation. Some restorations, mainly in group T, needed adjustments to enhance the fit.

The marginal and internal fit of the restorations in the 4 groups were assessed by a RT. Each endocrown was filled with a pink light-body vinyl polyether silicone impression material (EXA’lence; GC Corp) and seated along the long axis of the corresponding tooth under a constant force of 50 N for 5 minutes. After 5 minutes (setting time of the light-body material), the restoration was removed from the corresponding tooth. The layer of the light body adhered to the intaglio surface of the tooth. A green heavy-bodied material (EXA’lence; GC Corp) was injected into the tooth to adhere and stabilize the light-body material. After polymerization, a sharp surgical blade was used to cut each replica into 4 pieces from the center in a buccolingual and mesiodistal direction. A slice of 2-mm-thick specimens was segmented from every piece with parallel walls to obtain a vertical perpendicular view on the stereomicroscope stage.

The discrepancy between the tooth and the endocrown was represented by the pink-colored light layer, which was examined at ×70 magnification using a digital trinocular stereomicroscope (AmScope 3.5; Irvine) with a corresponding digital camera and software. For better comparison, each slice was divided into 4 areas of interest: marginal gap, cervical area, axial wall, and pulpal floor according to previous publications. Eight measurements were selected on each slice: 1 measurement on the marginal gap, M1; 2 measurements on the cervical area, C1 in the center and C2 on the cervico-axial angle; 3 measurements on the axial wall, A1, A2, and A3 which divided the axial wall into 3 equal parts; and 2 measurements on the pulpal floor, P1 on the axiopulpal angle and P2 in the center of the pulpal area of the sectioned replica ( Fig. 1 ). M1 was the AMD, that is, the distance between the most extended point of the endocrown margin and the external marginal line of the prepared tooth. C2 was the bisector of the angle between the cervical area and the axial wall. P1 was the bisector of the angle between the pulpal floor and the axial wall. C1, A1, A2, A3, and P2 were the perpendicular distance from the inner surface of the endocrown to the abutment tooth. M1 represented the marginal fit, whereas C1, C2, A1, A2, A3, P1, and P2 represented the internal fit of the endocrown. Discrepancy thicknesses at marginal sites and internal sites were analyzed. A total of 1280 measurements were made for the 4 groups (8 measurements×4 sections×10 endocrowns×4 groups). All data were saved in a spreadsheet (Microsoft Excel 2007; Microsoft Corp).

Figure 1
Schematic representation of measurement positions for marginal and internal fit in cross-section of replica. Pink layer represents cement analog layer; M1: absolute marginal discrepancy. C1 and C2: cervical discrepancies. A1, A2, and A3: axial discrepancies. P1 and P2: pulpal discrepancies.

The Shapiro-Wilk test of normality confirmed that the data were normally distributed ( P >.05). Failure to meet the assumptions required for a 2-way mixed ANOVA necessitated separate testing of the effects of the between-subjects factor (group) and of the within-subjects factor (region), followed by Bonferroni corrections for the multiple testing of 2 separate hypotheses. One-way ANOVA tests were performed to assess differences in marginal, cervical, axial, pulpal, and average internal gaps among the 4 groups, followed by the Tukey post hoc test for multiple comparisons. The Welch ANOVA and Games-Howell post hoc tests were applied when the assumption of homogeneity of variances was violated. Repeated measures ANOVA tests were then carried out to test for differences among marginal, cervical, axial, and pulpal gaps for each of the assessed groups separately and were followed by post hoc pairwise comparisons with Bonferroni adjustment. Finally, paired t tests were used to assess the differences between marginal and internal gaps separately for each of the 4 tested groups.

All measurements were repeated twice at least 1 month after the initial assessment, and intraobserver reliability was evaluated with the 2-way mixed effects intraclass correlations for absolute agreement on single measures. All data were processed using the statistical software (IBM SPSS Statistics, v20; IBM Corp) (α=.05 except when assessing the outcomes of the ANOVA tests, where α=.025 to account for the multiple testing of 2 separate hypotheses).


The intraclass correlations ranged between 0.904 and 0.998 for intrarater reliability, demonstrating very high correspondence. All regions tested displayed statistically significant differences in fit among the 4 assessed groups ( P <.001). Group T consistently displayed the largest values among the groups for all tested regions, whereas group E displayed the smallest gap for all regions except for the axial region, where the smallest gap was observed for group V ( Table 2 ). When the adaptation was compared across the regions, all the tested groups displayed statistically significant differences in fit ( P <.001). The largest gap was consistently observed at the pulpal floor for all tested groups, whereas the axial wall displayed the smallest mean gap in all groups except for group E, where the cervical wall presented the smallest gap ( Table 3 ). Mean internal adaptation was statistically similar to marginal adaptation in groups E ( P =.936) and V ( P =.396). The mean internal gap was smaller than the marginal gap in group C (116.1 ±14.3 μm compared with 143.0 ±21.7 μm; P =.009) and in group T (161.6 ±23.6 μm compared with 196.7 ±33.7 μm; P <.001) ( Table 4 ).

Table 2
Mean values, standard deviations, and group comparison of gap thickness (values in micrometer) at various regions across 4 tested groups (n=40)
Region Group Group Comparison
C (n=10) T (n=10) E (n=10) V (n=10) One-Way ANOVA P a
Mean ±SD Mean ±SD Mean ±SD Mean ±SD F P C/T C/E C/V T/E T/V E/V
Marginal 143.0 ±21.7 196.7 ±33.7 104.8 ±14.1 114.7 ±21.5 23.239 a <.001 b .003 b .001 b .04 <.001 b <.001 b .631
Cervical 119.9 ±22.0 168.0 ±24.7 83.8 ±11.7 92.8 ±13.1 40.979 <.001 b <.001 b .001 b .013 b <.001 b <.001 b .013
Axial 83.9 ±15.4 92.0 ±18.8 90.5 ±18.9 60.7 ±7.8 8.314 <.001 b .667 .792 .012 b .996 .001 b .001 b
Pulpal 144.5 ±22.0 224.7 ±51.1 141.6 ±20.1 179.1 ±23.0 11.186 a <.001 b .003 b .99 .014 b .002 b .096 .006 b
Internal 116.1 ±14.3 161.6 ±23.6 105.3 ±14.9 110.9 ±12.4 13.805 a <.001 b .001 b .374 .818 <.001 b <.001 b .8
Only gold members can continue reading. Log In or Register to continue

Jan 12, 2020 | Posted by in Prosthodontics | Comments Off on Effect of different CAD-CAM materials on the marginal and internal adaptation of endocrown restorations: An in vitro study
Premium Wordpress Themes by UFO Themes