Influence of a die spacer on the fit and load to failure of cantilever resin-bonded fixed dental prostheses

Abstract

Purpose: To evaluate the effect of die spacer thickness on the fit and load to failure of cantilever resin-bonded fixed dental prostheses (RBFDPs).

Methods: Two identical maxillary RBFDPs with a retainer on the canine were designed to replace a lateral incisor. One design involved a closely fitting retainer with no die spacer (NDS), and the other included an 80-µm die spacer (DS). A total of 10 RBFDPs were produced for each group by selective laser melting. To simulate a natural canine abutment, lithium disilicate abutments were fabricated digitally. After bonding the RBFDP with an adhesive resin luting agent, the marginal fit was measured, and the specimens were then loaded until failure of the luting agent.

Results: The mean marginal fit in the NDS and DS groups was 79.1 µm and 75.7 µm, respectively, and the corresponding mean maximum load to failure was 272.8 N and 254.7 N, respectively. The inter-group differences in these values were not significant.

Conclusion: The inclusion of a RBFDP die spacer had no effect on the marginal fit and maximum load to failure. In addition, the die spacer had no significant effect on the pattern of resin luting agent failure.

Introduction

The resin-bonded fixed dental prosthesis (RBFDP) is a fixed partial denture bonded to tooth structures by an adhesive resin luting agent [1]. It is a minimally invasive fixed prosthesis featuring a metal or ceramic wing bonded to an abutment tooth, and normally used to replace a single missing tooth. In contrast to other fixed replacement options, RBFDPs have the advantages of being conservative, cost effective, and simple to manufacture, with a tendency for non-catastrophic failure [2,3]. The conservative design and non-catastrophic failure features are especially desirable for young patients, in whom elaborate tooth preparations required for conventional fixed dental prostheses may inadvertently lead to iatrogenic endodontic complications in abutment teeth [4]. More recently, RBFDPs have gained significant popularity due to improvements in bonding to modern ceramics [5,6].

The main limitations of RBFDPs are their reliance on the bonding capability of the adhesive resin luting agent, and the likelihood of debonding of the retaining wing from the abutment tooth. This has been attributed to multiple factors including occlusal forces, the extent of abutment tooth preparation, or the condition of the abutment tooth itself [7]. Various studies have investigated ways of overcoming this issue by preparing grooves in abutment teeth, maximizing the retainer surface area, the use of etchable base metal alloys or ceramics, surface treatment of the zirconia material, or the application of adhesive luting agents [5,6,7,8,9].

The die spacer has been identified as a feature that can influence the fit, retention and longevity of any indirect restoration. Studies examining the effectiveness of die spacer thickness on the retention of full coverage crowns have demonstrated that the use of thicker die spacers can optimize the fit of the crown in comparison to close-fitting restorations [10,11,12,13,14]. However, thicker layers of luting agent are more likely to contain voids and defects. Furthermore, thicker adhesive resin luting agents are subject to polymerization shrinkage during curing, which generates residual stresses at the bonded interfaces, facilitating easier crack propagation [15,16]. Eventually, a thicker luting agent layer is thought to increase vulnerability to degradation within the oral cavity [16,17]. However, there is a lack of existing studies that have assessed the effect of a die spacer on the fit and retention of RBFDPs. As RBFDP retention relies primarily on the adequacy of the bonding, it is expected that uncontrolled luting agent thickness would affect the final fit and retention. The aim of the present study, therefore, was to determine the effect of RBFDP die spacer thickness on retainer fit and RBFDP retention using the maximum load to failure as an indicator. Two groups were examined for comparison: one without a die spacer and one including a die spacer with a thickness of 80 µm. The null hypothesis was that a die spacer would have no effect on the fit and load to failure of RBFDP.

Materials and Methods

The study included two identical maxillary RBFDP designs with a retainer on the canine to replace a lateral incisor. The first design was closely fitting with no die spacer (NDS) between the retainer internal surface and the abutment. The other design was closely fitting with a 1-mm margin and the remaining internal surface had an 80-µm die spacer (DS). To mimic a natural tooth, canine abutments were prepared to receive the RBFDP. Lithium disilicate was chosen as the material for the canine abutments, as it can be etched and treated for bonding. Although not accurately representing a bondable tooth surface, this protocol ensured all the specimens had a similar design, which cannot be achieved with extracted natural teeth. Sample size calculation was performed with a software program (G*Power, v3.1.9.2; Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany). With an effect size of 0.7 [11,18], α = 0.05, and a statistical power of 80%, a minimum of 9 specimens per group were required.

A metal jig composed of a stud and a cylindrical base 19 mm in diameter and 15 mm in height was designed using specific software (Meshmixer, Autodesk, San Francisco, CA, USA) (Fig. 1A). The cylindrical base was compatible with a computer-controlled precision universal testing machine (MTS 810 Materials Test System; MTS Systems Corp., Eden Prairie, MN, USA). The stud was 3.5 mm wide and 6.5 mm high, allowing for a canine abutment attachment. The metal jig was produced using a commercial milling unit. The canine abutment was designed with palatal preparation to accept the RBFDP. Minimally invasive palatal preparation features were incorporated to allow definite seating of the RBFDP [5,6] and enhance the resistance and retention forms [9]. Specifically, a palatal gingival chamfer of 0.3 mm, parallel proximal surfaces, and parallel mesial and distal proximal grooves of 0.5 mm depth were incorporated (Fig. 1B). On the virtually designed canine abutment, NDS and DS cantilever RBFDPs were designed with identical retainer coverage and an anatomical pontic (Fig. 1C). The canine abutments and the RBFDP were designed using computer-aided software (Exocad DentalCAD; Exocad GmbH., Darmstadt, Germany). The software was used to alter the die spacer of the specimens (Fig. 2). A milling unit (CEREC MC XL, Sirona, Bensheim, Germany) was used to produce 20 lithium disilicate (IPS e.max, Ivoclar, Schaan, Liechtenstein) specimens, with 10 specimens serving as abutments for each group. A preliminary specimen was milled to confirm that the milling unit was able to produce the required design. The RBFDPs were produced from cobalt-chromium alloy using a selective laser melting (SLM) unit (AM 400, Renishaw, Wotton-under-Edge, UK) to ensure the consistency and reproducibility of the specimens [11]. Metal was chosen as the RBFDP material instead of ceramic to ensure durability and ability to withstand loading before fracture of the connector or retainer, thus allowing any failure during loading to be attributed to the luting agent.

The bonding protocol conformed to that recommended by the manufacturer of the adhesive resin luting agent (Panavia SA Plus, Kuraray, Tokyo, Japan) for base metal and lithium disilicate ceramic [6,9]. The metal wings of the RBFDPs were initially prepared by airborne particle abrasion using 50-µm aluminum oxide particles at 0.4 MPa pressure (Renfert, Hilzingen, Germany). The wing was cleaned ultrasonically for 3 min in 99% isopropanol. Alloy primer (Kuraray) was applied to the wing internal surface and left to dry. The preparation surfaces of the lithium disilicate specimens were prepared by etching with 4.5% hydrofluoric acid for 20 s. A thin coat of ceramic primer (Monobond Plus, Ivoclar) was subsequently applied for 60 s and any excess was removed by air-drying to complete the silanation of the tooth. The mixed luting agent paste was applied to the wing, which was then seated on the palatal preparation of the tooth specimen using finger pressure. The grooves were used to guide the seating. This was followed by LED light curing for 3 s. Excess luting material was removed to ensure margin visibility and then LED light curing was performed for 20 s (Fig. 3). The samples were stored in water at room temperature for at least 24 hours.

An optical traveling microscope (Nikon Traveling Microscope, Nikon Instruments Inc., Melville, NY, USA) with an accuracy of 0.001 mm was used to measure the marginal fit. Along the RBFDP margin, the vertical distance between the metal and the abutment tooth was measured at 1-mm intervals. A total of 20 measurements were performed for every specimen. The specimens were placed on the microscope platform with the margin parallel to the microscope lens. All measurements were completed by a single operator.

Each specimen was attached to the metal jig with a resin luting agent (Panavia SA Plus, Kuraray). The jig was tilted to a 16° angle to enable loading of the cantilevered pontic in slight palatal angulation on the incisal edge (Fig. 4). A flat load applicator on the incisal edge was used to prevent slippage of the specimen, which would occur if the pontic was loaded on the palatal surface. This loading set-up aimed to simulate occlusal loading at the incisal edge during excursive movement. Maximum load to failure (in N) was recorded when the specimen showed any form of failure, indicated by a sudden reduction of the recorded load. Each specimen was examined after failure to determine the nature of the luting agent failure.

Statistical analyses were conducted using the SPSS software package (SPSS for Windows, v23; SPSS Inc, Chicago, IL, USA). The Shapiro Wilk normality test was used to evaluate the normality of the data. Independent t-tests were used to compare the marginal fit and the maximum load to failure between the two groups. The level of significance was set to 0.05. A chi-squared test of independence was performed to determine the significance of the relationship between the pattern of failure and the die spacer.

Fig. 1 Metal jig and specimen design

(A) The virtually designed cylindrical metal jig with the incorporated stud. (B) The virtually designed lithium disilicate canine crown with palatal chamfer and proximal grooves. (C) Virtual RBFDP with a mesial cantilever to replace a lateral incisor.

Fig. 2 Virtual die spacer incorporation

(A) NDS design with the closely fitting retainer. (B) DS design with a closely fitting margin and incorporating an 80-µm cement space. (C) Verification image confirming the uniform cement space in the DS group.

Fig. 3 Specimen assembly

(A) Lithium disilicate crown seated on the metal jig. (B) Example of a RBFDP specimen cemented on the canine abutment.

Fig. 4 Loading of the RBFDP specimen on the tilted metal jig

Results

The data for marginal fit and load to failure were found to be normally distributed (P > 0.05). The mean marginal fit in the NDS and DS groups was 79.1 (15.9) µm and 75.7 (10.6) µm, respectively (Fig. 5A), the inter-group difference being non-significant (P = 0.57). The mean maximum load to failure in the NDS group was 272.8 (66.1) N, being slightly higher than that in the DS group at 254.7 (58.4) N (Fig. 5B). The inter-group difference in mean maximum load to failure was not statistically significant (P = 0.53).

All specimens showed adhesive failure between the luting agent and the abutment tooth, and between the luting agent and the retainer. Eight DS specimens and 4 NDS specimens showed cohesive luting agent failure (Fig. 6). The inter-group difference between the pattern of failure was not statistically significant (P = 0.33).

Fig. 5 Bar graphs comparing the NDS and DS groups

(A) Marginal fit. (B) Load to failure.

Fig. 6 Prevalence of different failure patterns in the two groups

Discussion

The impact of a die spacer on the seating and load to failure of RBFDPs has not been investigated previously. These variables were selected in the present study because misfit is likely to lead to degradation of the luting agent [15,16,17], and subsequently loss of retention, which has been shown to be the main cause of RBFDP failure [4,7,9]. The present study demonstrated that an RBFDP die spacer had no effect on the final marginal fit and retainer retention. Likewise, the pattern of adhesive resin luting agent failure did not differ between the two groups. Therefore, the hypothesis that the die spacer would have no effect on the fit and load to failure of RBFDPs was accepted.

The present study found no significant difference in marginal discrepancy between the DS and NDS groups, both of which exhibited a fit well below the recommended marginal fit for indirect restorations (80-120 µm) [19,20], including partial coverage anterior indirect restorations (50-100 µm) [21,22,23]. This suggests that the marginal fit of RBFDPs is not affected by the amount of die spacer, and that even if a die spacer is not incorporated, the adhesive resin luting agent would maintain a minimal film thickness of 80 µm. This contrasts with numerous previous studies of indirect full-coverage restorations, which indicated that the seating discrepancy of a crown is inversely proportional to the die spacer [11,12,13,14,24]. For example, Wilson reported that increasing the die spacer from 0 to 50 µm improved the seating and marginal fit. Specifically, the 0 space had an excessive marginal discrepancy of 368 µm. Similarly, Zhang and Dudley reported that increasing the die spacer from 50 to 200 µm significantly improved the marginal fit of computer-aided design-computer-aided manufacturing (CAD-CAM) restorations [12]. One study found that application of a 120-µm die spacer thickness resulted in better adaptation of monolithic zirconia crown restorations in comparison to smaller spaces [25]. Furthermore, crown die spacers of 70 µm, 90 µm and 110 µm led to marginal gaps of 162 µm, 108 µm and 87 µm, respectively [13]. This observation has been attributed to preparation geometry, the amount of coverage, luting material and particles interfering with the seating, and lack of compensation for manufacturing errors [10,12,15]. Restricted die spacers are likely to prevent easy flow of the adhesive resin luting agent, leading to material entrapment on the occlusal aspect, and resulting in a greater marginal discrepancy. The rheological properties and viscosity of the adhesive resin luting agent will also accentuate any deleterious effect on the marginal gap [14,23]. Contrary to full coverage restorations, the present study indicated that the partial coverage design of the RBFDP retainer was not associated with the similar relationship between the die spacer and marginal discrepancy. The seating of a RBFDP retainer differs significantly from a full-coverage crown. For example, a RBFDP preparation is predominantly a single-walled, partial coverage with simple geometry, exhibiting variable insertion paths [14,23]. This allows for free escape of the adhesive resin luting agent during bonding, and reduces the likelihood of localized luting agent entrapment.

The influence of a die spacer on the retention of an indirect restoration is rather complex, as the preparation morphology, retainer design, luting agent type and bonding protocol significantly impact the final retention and resistance form of the restoration [15,16,17]. The retention of a cemented restoration has been evaluated in earlier laboratory studies that employed pullout testing, simulated functional loading, fatigue behavior testing, fracture resistance, and shear testing [11,16,26,27,28]. For a RBFDP with a partial coverage retainer, the luting agent is subjected to a combination of tensile, compressive and shear stresses. In order to simulate these stresses, the present study measured the load to failure of RBFDP specimens by loading the cantilever pontic in the palatal direction. The load to failure ultimately reflects the retention capacity of each design. Some studies have reported a correlation between fit and restoration retention, whereas others have found no significant interaction. Lovgren et al. found that although SLM yielded a better marginal fit than milled or milled-casted coping (by half), the pullout retention was similar [11]. Venturini et al., found that the thickness of an adhesive resin luting agent (50, 100, and 300 µm) had no effect on the fatigue behavior of bonded leucite ceramic crowns [26]. Likewise, Farag et al. found that die spacers of 20, 40, and 100 µm had no effect on the fracture resistance of lithium disilicate veneers [27]. In a study involving stepped fatigue testing, different thicknesses of adhesive resin luting agent (60, 300 µm) did not significantly affect the shear stress of leucite ceramic specimens [16]. In contrast, Son et al. reported decreased retention in the pullout test for adhesive resin luting agents after an increase in die spacer thickness from 40 to 160 µm. They attributed this to a thicker luting agent, the possibility of reduced LED polymerization, and volumetric shrinkage leading to increased residual stresses, the luting agent ultimately serving as the weakest link in the pullout test [18]. Another study demonstrated similar immediate bond strength with a luting agent thickness of 60, 120 or 180 µm, but degradation was observed after simulated loading [28]. The present study clearly demonstrated similar retention in the two groups, suggesting that luting agent thickness may not be a critical factor determining the overall retention and strength of RBFDPs within the range tested, perhaps being attributable to the similarity in preparation design, and the extent of retainer coverage. It can be speculated that bonding and micromechanical retention would be less affected by small differences in the die spacer (0-80 µm). As the marginal fit in the two groups was closely similar, it is likely that the adhesive resin luting agent maintained the minimum thickness required for acceptable retention [29], even when no die spacer was incorporated. Therefore, it can be assumed that RBFDP retention is more dependent on the bonding surface and retainer coverage areas, rather than the die spacer. This is further supported by the similar pattern of bonding failure that occurred in both groups.

Overall, the present findings suggest that minor variations in the die spacer may not significantly impact the performance of RBFDPs, providing some leeway in clinical and laboratory procedures. Clinicians should prioritize other aspects of RBFDP design and placement, such as case selection, preparation design, extent of retainer coverage, and bonding protocol, which may have a more significant impact on long-term success. While these explanations are based on general principles of RBFDP retention, it is important to note that studies focusing specifically on die spacers in RBFDPs have been limited. Further research to confirm these observations and provide more detailed insight into the role of luting agent thickness in RBFDP performance would be desirable. Despite the similarities between the present two groups, these results should be interpreted with caution. The experimental set-up does not represent a wide range of clinical presentations in terms of loading pattern, variations in preparation morphology, RBFDP design and material, and the type of luting agent. Partial coverage restorations exhibit extensive variations in individual patients, affecting the amount of retainer coverage and loading pattern [7,8,9]. Furthermore, the seating process applied when luting the retainers can contribute to inevitable displacement. Another limitation of the present study was the use of lithium disilicate as the abutment material. While this helped to standardize the design and dimensions of all abutments, bonding to lithium disilicate does not accurately simulate bonding to natural enamel. Tooth aging, which would also impact the results, was not considered [28]. In general, the present findings indicate that differences in RBFDP die spacing (0 and 80 µm) did not affect the marginal fit and maximum load to failure. In addition, the pattern of adhesive resin luting agent failure was similar, regardless of the die spacer.

Abbreviations

CAD-CAM: computer-aided design-computer-aided manufacturing; DS: die spacer; LED: light emitting diode; NDS: no die spacer; RBFDP: resin-bonded fixed dental prosthesis; SLM: selective laser melting

Ethical Statements

Not applicable

Conflicts of Interest

The authors declare no conflicts of interest.

Funding

This study was funded by the Melbourne Dental School Research Higher Degree Funding.

Author Contributions

BXT, TW, QW, CT, VW, WCT, LT, RV, AKT, DT, AW, and CHST: investigation, formal analysis, validation, visualization, writing – original draft; JP and JA: conceptualization, methodology, project administration, supervision, writing – review and editing. All authors read and approved the final version of the manuscript.

ORCID iD

BXT: [e-mail], https://orcid.org/0009-0003-1936-0751

TW: [e-mail], https://orcid.org/0009-0009-3060-8731

QW: [e-mail], https://orcid.org/0009-0002-1333-8660

CT: [e-mail], https://orcid.org/0009-0007-5409-446X

VW: [e-mail], https://orcid.org/0009-0005-4115-9506

WCT: [e-mail], https://orcid.org/0009-0004-2506-4622

LT: [e-mail], https://orcid.org/0009-0005-0027-2756

RV: [e-mail], https://orcid.org/0009-0005-2997-2342

AKT: [e-mail], https://orcid.org/0009-0000-8168-7315

DT: [e-mail], https://orcid.org/0009-0005-6752-2633

AW: [e-mail], https://orcid.org/0009-0005-8258-4568

CHST: [e-mail], https://orcid.org/0009-0005-8598-2553

JP: [e-mail], https://orcid.org/0000-0003-1439-4509

JA*: [e-mail], https://orcid.org/0000-0003-3392-8641

Acknowledgments

The authors acknowledge the technical support of Mr. Attila Gergely in fabricating the research specimens.

Data Availability Statements

Data generated during the present study are available from the corresponding author on reasonable request.

References

• 1.  The Glossary of Prosthodontic Terms: Tenth Edition (2023). J Prosthet Dent 130, 7-126. doi: 10.1016/j.prosdent.2023.03.002
• 2.  Miettinen M, Millar BJ (2013) A review of the success and failure characteristics of resin-bonded bridges. Br Dent J 215, E3. doi: 10.1038/sj.bdj.2013.686
• 3.  Allen PF, Anweigi L, Ziada H (2016) A prospective study of the performance of resin bonded bridgework in patients with hypodontia. J Dent 50, 69-73. doi: 10.1016/j.jdent.2016.05.003
• 4.  Balasubramaniam GR (2017) Predictability of resin bonded bridges - a systematic review. Br Dent J 222, 849-858. doi: 10.1038/sj.bdj.2017.497
• 5.  Kern M, Passia N, Sasse M, Yazigi C (2017) Ten-year outcome of zirconia ceramic cantilever resin-bonded fixed dental prostheses and the influence of the reasons for missing incisors. J Dent 65, 51-55. doi: 10.1016/j.jdent.2017.07.003
• 6.  Kern M, Glaser R (2023) Single-retainer all-ceramic resin-bonded fixed dental prostheses: long-term outcomes in the esthetic zone. J Esthet Restor Dent 35, 64-73. doi: 10.1111/jerd.13001
• 7.  Creugers NH, De Kanter RJ, van’t Hof MA (1997) Long-term survival data from a clinical trial on resin-bonded bridges. J Dent 25, 239-242. doi: 10.1016/s0300-5712(96)00033-4
• 8.  De Kanter RJ, Creugers NH, Verzijden CW, Van’t Hof MA (1998) A five-year multi-practice clinical study on posterior resin-bonded bridges. J Dent Res 77, 609-614. doi: 10.1177/00220345980770041401
• 9.  Abuzar M, Locke J, Burt G, Clausen G, Escobar K (2018) Longevity of anterior resin-bonded bridges: survival rates of two tooth preparation designs. Aust Dent J 63, 279-284. doi: 10.1111/adj.12612
• 10.  Wilson PR (1994) Effect of increasing cement space on cementation of artificial crowns. J Prosthet Dent 71, 560-564. doi: 10.1016/0022-3913(94)90437-5
• 11.  Lovgren N, Roxner R, Klemendz S, Larsson C (2017) Effect of production method on surface roughness, marginal and internal fit, and retention of cobalt-chromium single crowns. J Prosthet Dent 118, 95-101. doi: 10.1016/j.prosdent.2016.09.025
• 12.  Zhang Y, Dudley J (2019) The influence of different cement spaces on the marginal gap of CAD/CAM all-ceramic crowns. Aust Dent J 64, 167-174. doi: 10.1111/adj.12688
• 13.  Hassan LA, Goo CL (2021) Effect of cement space on marginal discrepancy and retention of CAD/CAM crown. Dent Mater J 40, 1189-1195. doi: 10.4012/dmj.2020-408
• 14.  Elbadawy AA, Omar EA, AbdElaziz MH (2022) MicroCT evaluation for CAD/CAM occlusal veneer fit using two materials and three cement space settings. Braz Dent J 33, 71-78. doi: 10.1590/0103-6440202204764
• 15.  Fleming GJP, Cao X, Romanyk DL, Addison O (2017) Favorable residual stress induction by resin-cementation on dental porcelain. Dent Mater 33, 1258-1265. doi: 10.1016/j.dental.2017.07.018
• 16.  Arcila LVC, Gomes LCL, Ortiz LPN, Costa MMD, Tribst JPM, Bottino MA et al. (2023) Effect of resin cement at different thicknesses on the fatigue shear bond strength to leucite ceramic. Eur J Dent 17, 1316-1324. doi: 10.1055/s-0042-1758797
• 17.  Gressler May L, Kelly JR, Bottino MA, Hill T (2015) Influence of the resin cement thickness on the fatigue failure loads of CAD/CAM feldspathic crowns. Dent Mater 31, 895-900. doi: 10.1016/j.dental.2015.04.019
• 18.  Son YH, Han CH, Kim S (2012) Influence of internal-gap width and cement type on the retentive force of zirconia copings in pullout testing. J Dent 40, 866-872. doi: 10.1016/j.jdent.2012.07.007
• 19.  Hasanzade M, Moharrami M, Alikhasi M (2019) Evaluation of marginal and internal adaptation of endocrowns fabricated from three different CAD/CAM materials. Eur J Prosthodont Restor Dent 27, 164-171. doi: 10.1922/EJPRD_01931Hasanzade08
• 20.  Hasanzade M, Aminikhah M, Afrashtehfar KI, Alikhasi M (2021) Marginal and internal adaptation of single crowns and fixed dental prostheses by using digital and conventional workflows: a systematic review and meta-analysis. J Prosthet Dent 126, 360-368. doi: 10.1016/j.prosdent.2020.07.007
• 21.  Baig MR, Akbar AA, Sabti MY, Behbehani Z (2022) Evaluation of marginal and internal fit of a CAD/CAM monolithic zirconia-reinforced lithium silicate porcelain laminate veneer system. J Prosthodont 31, 502-511. doi: 10.1111/jopr.13438
• 22.  Hanoon ZA, Abdullah HA, Al-Ibraheemi ZA, Alamoush RA, Sami SM, Haider J (2023) Marginal fit of porcelain laminate veneer materials under thermocycling condition: an in-vitro study. Dent J (Basel) 11, 1-13. doi: 10.3390/dj11010012
• 23.  Baig MR, Qasim SSB, Baskaradoss JK (2024) Marginal and internal fit of porcelain laminate veneers: a systematic review and meta-analysis. J Prosthet Dent 131, 13-24. doi: 10.1016/j.prosdent.2022.01.009
• 24.  Olivera AB, Saito T. The effect of die spacer on retention and fitting of complete cast crowns (2006) J Prosthodont 15, 243-249. doi: 10.1111/j.1532-849X.2006.00113.x
• 25.  Taha FA, Jasim ZA, Husseien HA (2022) Effect of cement-space thickness on the marginal fitness of all zirconium oxide crowns-an in vitro study. J Int Oral Health 14, 382-385. doi: 10.4103/JIOH.JIOH_275_21
• 26.  Venturini AB, Wandscher VF, Marchionatti AME, Evangelisti E, Ramos GF, Melo RM et al. (2020) Effect of resin cement space on the fatigue behavior of bonded CAD/CAM leucite ceramic crowns. J Mech Behav Biomed Mater 110, 103893. doi: 10.1016/j.jmbbm.2020.103893
• 27.  Farag SM, Ghoneim MM, Afifi RR (2021) Effect of die spacer thickness on the fracture resistance of CAD/CAM lithium disilicate veneers on maxillary first premolars. Clin Cosmet Investig Dent 13, 223-230. doi: 10.2147/CCIDE.S313131
• 28.  Tribst JPM, Dos Santos AFC, da Cruz Santos G, da Silva Leite LS, Lozada JC, Silva-Concilio LR et al. (2021) Effect of cement layer thickness on the immediate and long-term bond strength and residual stress between lithium disilicate glass-ceramic and human dentin. Materials (Basel) 14, 1-13. doi: 10.3390/ma14185153
• 29.  Yokoyama D, Shinya A, Gomi H, Vallittu PK, Shinya A (2012) Effects of mechanical properties of adhesive resin cements on stress distribution in fiber-reinforced composite adhesive fixed partial dentures. Dent Mater J 31, 189-196. doi: 10.4012/dmj.2011-104