In vitro fatigue and fracture testing of temporary materials from different manufacturing processes in implant-supported anterior crowns

The aim of this study was to investigate the in vitro fatigue and fracture force of temporary implant-supported anterior crowns made of different materials with different abutment total occlusal convergence (TOC), with/without a screw channel, and with different types of fabrication. One hundred ninety-two implant-supported crowns were manufactured (4° or 8° TOC; with/without screw channel) form 6 materials (n = 8; 2 × additive, 3 × subtractive, 1 × automix; reference). Crowns were temporarily cemented, screw channels were closed (polytetrafluoroethylene, resin composite), and crowns were stored in water (37 °C; 10 days) before thermal cycling and mechanical loading (TCML). Fracture force was determined. Statistics: Kolmogorov–Smirnov, ANOVA; Bonferroni; Kaplan–Meier; log-rank; α = 0.05. Failure during TCML varied between 0 failures and total failure. Mean survival was between 1.8 × 105 and 4.8 × 105 cycles. The highest impact on survival presented the material (η2 = 0.072, p < .001). Fracture forces varied between 265.7 and 628.6 N. The highest impact on force was found for the material (η2 = 0.084, p < .001). Additively and subtractively manufactured crowns provided similar or higher survival rates and fracture forces compared to automix crowns. The choice of material is decisive for the survival and fracture force. The fabrication is not crucial. A smaller TOC led to higher fracture force. Manually inserted screw channels had negative effects on fatigue testing. The highest stability has been shown for crowns with a low TOC, which are manufactured additively and subtractively. In automix-fabricated crowns, manually inserted screw channels have negative effects.


Introduction
Implant-supported restorations are a state of the art for the treatment of anterior gaps and guarantee good to sufficient survival [1,2]. During osseointegration of dental implants, biocompatible materials with good marginal fit and sufficient flexural strength are needed for temporary restorations to guarantee quick, easy, and inexpensive production. Temporary crowns and fixed partial dentures (FPDs) play an essential role in the long-term therapeutic success of the dental treatment. The main requirements are protection of the prepared teeth or implant-supported abutments from thermal, chemical, or mechanical influences and to guarantee phonetic and mechanical function, as well as esthetic demands [3,4]. They are also indispensable to shape the marginal gingiva for optimal emergence profiles in implant restorations [5].
Temporary materials are typically based on methacrylate (MMA) or dimethacrylate (DMA) resins, containing approximately 30-40 w% inorganic fillers for improved modulus of elasticity and wear resistance [4,6]. Materials with different filler contents are expected to show different clinical performances [7]. Standard materials are automix (cartridge, paste-paste) systems, which require impressions or deepdrawing foils for labside or chairside fabrication [8]. While providing good clinical performance [9,10], heat of reaction and polymerization shrinkage could be a disadvantage of these materials [11][12][13]. In contrast to directly processed 1 3 materials, materials processed via computer-aided design and computer-aided manufacturing (CAD/CAM) usually have improved mechanical properties [14]. In addition, a significantly better fit of the indirectly fabricated temporary crowns can be assumed [13,15] Additive and subtractive CAD/CAM manufacturing is increasingly superseding conventional treatment in modern dentistry [16][17][18]. In case of fracture or loss, new restorations can be quickly rebuild [19,20]. CAD/CAM block materials for subtractive manufacturing are produced under controlled industrial conditions. Additively manufactured FPDs are produced in the dental lab. This leads to reduced residual monomers and a heating of the material can be avoided, which benefits to the surrounding tissue and prepared teeth. Less polymerization shrinkage promises more precise internal and marginal fit [21]. Compared to automix systems, no pulling off from teeth or models is needed, which makes flexibility for these materials obsolete. Less flexibility technically allows higher filler content and consequently improved mechanical properties. However, in case of materials for additive manufacturing, filler content and size are limited due to the need of viscosity of the resin [16,22,23]. In case of subtractive manufacturing, the bur size used in the milling process limits the design of subsequent restorations. Overall, this may favor CAD/CAM materials compared to conventional automix materials against the background of the requirements for temporary materials [24,25].
In case of insufficient removal of residual cement, temporary restorations can lead to gingival and peri-implant inflammation. This risk can be eliminated by cementing the restoration to an abutment outside the patients' mouth and leaving a channel for screw retention. However, stability of the restoration might be affected by the screw channel [26,27].
The clinical situation determines the total occlusal convergence (TOC) of the abutment: to guarantee a minimum material thickness, individual abutments or abutments with a bigger TOC might be indicated.
New materials and manufacturing techniques require extensive research before clinical use. In vitro tests can provide information about their properties, including mechanical stability, and long-term performance. Thermal cycling and mechanical loading might predict clinical failures in laboratory studies [28] and allow to estimate whether the temporary materials are appropriate for clinical application [29]. Clinical evidence for the success of CAD/CAM temporary materials is very limited [9,10,30], but in vitro data about performance and stability of these materials in comparison to conventional approved automix systems may help to estimate the clinical performance of these materials [16][17][18]31]. CAD/CAM and automix temporary materials were found to show comparable good clinical performance [31]. Despite frequent clinical use, there is limited scientific in vivo data on the performance of implant-supported temporary anterior crowns.
The aim of this study was to compare materials from different manufacturing processes for implant-supported temporary anterior crowns, to investigate the effect of different abutment TOC, and to evaluate the influence of a screw channel.
The hypothesis of this in vitro study was that the in vitro performance and fracture force of implant-supported temporary anterior crowns is influenced by the type of material, the type of manufacturing, a screw channel, or the TOC of the abutment.
Temporary anterior crowns (FDI #11) were digitally designed (inLab CAD SW 20.0.2, Dentsply Sirona, USA) with the default settings for resin composite materials. 192 crowns (n = 8 per material and group, esteemed power 85% for 8 specimens, G*Power, Kiel, D) were manufactured representing four groups: • 4° TOC of the abutment with a screw channel • 4° TOC of the abutment without a screw channel • 8° TOC of the abutment with a screw channel • 8° TOC of the abutment without a screw channel Six different materials were used representing two resinbased materials for additive manufacturing and three materials for subtractive manufacturing (two resin composites with different filler contents and one poly-ether-ether-ketone (PEEK) material). The crowns were milled, or 3D printed (details see Table 1). For the milled and printed materials, the insertion of a screw channel was selected in the pre-settings. One automix resin composite was used as a reference. Automix references were made by a direct method using the over-impression technique (Permadyne, 3 M, USA). A screw channel was manually drilled into the palatal sides of the automix crowns (FG shank round end cylinder diamond bur red/fine, diameter: 1.6 mm; Hager & Meisinger GmbH, D). Finishing and polishing was done, using rotary rubber cups (Astropol Polishing Kit, Ivoclar-Vivadent, Schaan, FL). All materials were fabricated according to the manufacturers' instructions.
All abutments were steam cleaned and the inner sides of the crowns were sandblasted (110 μm Al 2 O 3 , 2.0 bar). All crowns were temporarily cemented (Temp Bond, Kerr Dental, USA), and the screw channels were closed with polytetrafluoroethylene tape and temporary resin composite (Protemp 4, 3 M, USA). The crowns were then stored in distilled water (37 °C) for 10 days.
Specimens were subjected to simultaneous thermal cycling and mechanical loading (TCML: TC = 2 × 1200 cycles between 5/55 °C distilled water with duration of 2 min of each cycle, ML = 50 N for 4.8 × 10 5 cycles, f = 2.1 Hz, mouth opening = 2 mm, chewing simulator EGO, Regensburg, D) as standardized antagonists. All crowns were loaded with 6-mm steatite balls (CeramTec, Plochingen, D) in a one point of contact 1.5 mm below the incisal edge on the palatal crown side. Tests are equivalent to approximately 2 years of clinical service time [28,29].
Crowns which failed during TCML were excluded and dropout times (number of cycles) were recorded.
After TCML, all crowns were investigated in detail with a digital microscope (4-10 × magnification, VHX-S550E, Keyence, J). Fracture force of intact restorations was determined by loading the crowns to failure (Z010, Zwick Roell, D, v = 1 mm/min). In analogy to chewing simulation, the force was applied 1.5 mm below the incisal edge with a round end cylinder metal antagonist (d = 6 mm). To prevent force peaks, a tin foil (0.3 mm, Helago, Heinz & Laufer, D) was positioned between crown and antagonist. After fracture testing the crowns were optically examined and categorized. Fracture patterns were palatal fracture, mesial-distal fracture, palatal + mesial-distal fracture, or deformation.
Calculations and statistical analysis were performed with a statistical software program (IBM SPSS Statistics, v. 26.0 for Windows; IMB Corp, USA). Normal distribution of data was assessed with the Kolmogorov-Smirnov test. Means and standard deviations were calculated and analyzed using one-way analysis of variance (ANOVA), two-way ANOVA, and subsequent post hoc analysis (Bonferroni test). Betweensubject effects were investigated. Cumulated survival was calculated with Kaplan-Meier analysis and log-rank tests (Mantel-Cox) test. The level of significance was set to α = 0.05.
Kaplan-Meier analysis and log-rank tests (Mantel Cox) showed significant (p < 0.001, chi 2 : 20.816-48.792) differences between the materials in the individual groups (Fig. 2). C and PP showed early failures between 50,000 and 250,000 load cycles. The other systems revealed failures from approx. 200,000 load cycles with a lower failure rate.

Fracture force
Mean fracture force (Fig. 3, Table 3) varied between 265.7 N (PP) and 628.6 N (MB). Fracture testing revealed statistically significant (p < 0.001; ANOVA) higher force of MB crowns compared to all other materials. No significant difference in mean fracture force occurred between the other materials except for PP with significantly less mean fracture force compared to MG (p < 0.001; ANOVA) and MS (p = 0.037; ANOVA). In five cases (MS, MB, and PT without a screw channel and MS and MB with a screw channel) TOC of the abutment showed significant (p < 0.001; ANOVA) differences in the results with higher fracture force in crowns with 4° abutments. All automix "A" crowns with a screw channel failed during chewing simulation.

Fracture pattern
Fracture pattern in most cases (94%) was characterized by fracture at the palatal edge of the abutment, 6% additionally failed mesial-distal. No differences were found between failure during TCML (93% palatal, 7% additionally failed mesial-distal) or fracture test (91% palatal, 9% additionally failed mesial-distal). All MB crowns were excluded from the fracture pattern comparison due to deformation without fracture. Exemplary images of fracture patterns are provided in Fig. 4.

Failure during TCML and survival
The hypothesis that the in vitro performance of implant-supported temporary anterior crowns is influenced by the type of material, the type of manufacturing, a screw channel, or the   TOC of the abutment could be confirmed only for the type of material and the type of fabrication.

Material
The survival time of the crowns depended most clearly on the type of material. TCML resulted in individual survival rates between 135,000 and 480,000 cycles. While the failed specimens exhibited fractures, neither cracks, nor spalling, nor fractures were found in the surviving specimens. The material dependent performance might be mainly attributed to the composition of the temporary materials as found earlier for FPDs [7]. The different production and processing of the resinbased materials allows for different filler levels and conversion rates. These can affect the mechanical properties such as modulus of elasticity or flexural strength and therefore the in vitro performance.

Manufacturing
Clearly shortest survival was found for the automix material, followed by one printed material. All milled systems and one printed group provided highest survival times. Large deviation and distribution of mean survival cycles, as found more frequently with automix and printing systems, indicate an influence of crown manufacture.

Screw channel
The different wall thicknesses and the associated different load of the crowns due to different TOC were not noticeable for chewing forces of 50 N during TCML. As found for FPDs [7,32] or molar crowns [26], independent from material, screw channels did not weaken the crowns. Only the survival time of the automix material was reduced when a screw channel was present. Any of the automix crowns with a screw channel survived all loading cycles. A possible reason might be the manual insertion of the screw channel, which was only performed in this group. The diamond drilling may cause damage and cracks in the crown, which will force a premature failure.

TOC of the abutment
The TOC of the abutment had no influence on survival time of the crowns. Based on the assumption that the complete simulation cycle simulates a clinical survival time of approximately 2 years [28,29], the materials with failures starting at approximately 1.2 × 10 5 mechanical cycles seem suitable for a clinical application up to six months. Accordingly, the milled crowns are suitable for at least 2 years. At masticatory forces of 50 N, it was not possible to distinguish the influence of angle or the presence of a screw channel on survival rates. Further tests might be performed with higher forces or in staircase processes with increasing force limits.

Fracture force
The hypothesis that the fracture force of implant-supported temporary anterior crowns is influenced by the type of material, the type of manufacturing, a screw channel, or the TOC of the abutment could be partly confirmed.

Material
Significant differences were found in the fracture force depending on the material. The highest fracture values by far were determined in the PEEK group. However, this material was also the only one that did not exhibit brittle fracture behavior, which certainly also made the detection and assessment of the failure more difficult. The question arises, as to which deformation the crowns can still be considered clinically acceptable. The achieved fracture forces do not seem to be related to the mechanical properties of the materials. Neither materials with higher strength nor increased modulus of elasticity showed an increased fracture force. Furthermore, no correlation could be found between the fracture values and the filler content. The results are in contrast to recent studies with higher fracture forces measured in anterior provisional restorations on extracted teeth [35]. However, the tooth-or implantsupported crowns may behave differently [7,26]. Different results for implant-and tooth-supported molar crowns, which were also shown in previous studies [26], might be attributed to the altered loading situation of anterior and posterior crowns.

Manufacturing
In general, the milled crowns tended to show higher fracture forces than the printed systems or the automix material. Composition or filler-dependent stability has also been shown in recent studies [7]. However, one printed group with 4° TOC (material PT) and without screw channel yielded quite comparable results to the milled systems. Therefore, the results cannot necessarily be attributed to the different filler composition of printing and milling materials, which varies between 86 and 27 w%. Differences in fracture force might indicate variations in the homogeneity of the additive material, the influence of the material layers during printing or post-curing effects [33,34], and thus an effect of the fabrication. Variations might also occur due to defects like pores or insufficient connection of the individual layers or even reduced quality of insufficient combination of material and process [23]. Subtractively manufactured crowns could have an advantage over crowns manufactured in the other two processes: the discs for subtractive processing are manufactured under controlled industrial production at high temperature and pressure, which may improve homogeneity and mechanical properties [6]. Yet, in contrast to other in vitro tests [36], automix crowns that survived TCML showed no significant difference in fracture force which reflected the principally sufficient material properties of automix materials and, again, shows the manual manufactory as a possible source of error.

Screw channel
The presence of a screw channel did not show any influence on fracture force. Automix material crowns with a handmade screw channel could not be compared, as none survived the simulation. In general, the presence of a screw channel is expected to reduce the stability of crowns [37][38][39]. However, there seem to be differences in the anterior and posterior application and the associated loading situation with bending or compression. Recent studies even showed higher stability in molar crowns with a screw channel [26]. The screw channel might have no influence on the strength of the anterior crowns because it is located below the contact point.

TOC of the abutment
TOC of the abutment had strong influence on fracture force of the crowns, with higher fracture values for crowns on 4° abutments. Contradictory results were found only for one printed system. These results confirm previous studies that fracture forces were strongly dependent on the abutment TOC of molar crowns [27] or anterior temporary FPDs [26]. Yet there is no data that abutments with smaller TOC show higher fracture force. This may be due to the greater wall thicknesses of the 4° crowns examined for the same external dimensions. The higher material thickness promises higher strength, especially in the case of bending and torsion occurring in the anterior region. Since the palatal edge of the abutment is the most common fracture site, rounded palatal edges may be indicated to prevent crack initiation and development.
Aging with TCML can help distinguish potential failure modes of materials by simulating fatigue, but loading to fracture tests may not be able to accurately reflect the failure modes that can be observed in clinical settings. Simulations used in TCML do not consider factors such as varying chewing forces or frequency as well as the individual design of preparations and the crowns, or the complex oral situation of the jaw. The use of implant analogs shifts the fracture risk to the crowns to be examined, as desired. However, it also limits the clinical interpretation of the results. Therefore, these simulation results should be interpreted with caution when predicting clinical failure.
In general, most materials have the potential to withstand maximum bite forces in the incisive region, which are reported to reach up to 158 N [40]. However, patients with implant-supported restorations can show higher or uncontrolled mastication forces due to absence of periodontal receptors [41].

Conclusion
Both additively and subtractively manufactured temporary crowns provided similar or even higher survival rates and fracture forces compared to standard automix implant-supported anterior crowns. PEEK-based systems showed best performance. The choice of material is decisive for the survival time and fracture force. The fabrication is not crucial, but milled crowns tend to perform better. A smaller TOC of the abutment led to higher fracture force. Only manually inserted screw channels in automix-made crowns had negative effects on fatigue testing.

Clinical consequence
The choice of material is crucial for the performance of provisional implant-supported anterior crowns: additively and subtractively manufactured crowns show comparable or better properties than automix systems. A lower abutment TOC guarantees higher fracture forces. Manually inserted screw channels in automix crowns have negative effects and should be avoided.