Restoring Nonvital Premolars with Composite Resin Onlays: Effect of Different Fiber-reinforced Composite Layers on Marginal Adaptation and Fracture Load
Carlo Monaco
a/ Tissiana Bortolotto
b/ Antonio Arena
c/ Ivo Krejci
dPurpose: To evaluate the marginal adaptation and fracture load of composite resin onlays reinforced with differ- ent substructures.
Materials and Methods: Thirty-two extracted, caries-free premolars were selected for this study and endodonti- cally treated. Group 1 was used as the control group, and the teeth were restored only with as-manufactured composite resin overlays. Group 2 teeth were restored with composite resin overlays with 3 fiber-reinforced composite (FRC) layers placed horizontally on the bottom of the restoration. Group 3 teeth were restored with composite resin overlays with 6 fiber-reinforced composite (FRC) layers placed as in group 2. Group 4 teeth were restored with composite resin overlays and FRC placed with an anatomical design. All specimens underwent SEM evaluation of their marginal adaptation before and after thermocycling and cyclic mechanical loading. All specimens were then subjected to a fracture test, recording the value for the initial (IF) and final (FF) failure. Dif- ferences in the means were compared using matched-pairs t-tests and one-way ANOVA. The level of significance was set at α = 0.05.
Results: No statistically significant difference between the four groups in terms of marginal adaptation was observed at the tooth/luting composite and luting composite/overlay interfaces before and after loading. The fracture loads of IF and FF, from most to least resistant were: group 4 (1431.8 ± 294.3 N / 1710.1 ± 326.6 N), group 3 (1428.1 ± 251.4 N / 1467.9 ± 242.4 N), group 2 (852.6 ± 413.5 N / 1058.1 ± 251.5 N) and group 1 (899.8 ± 352.7 N / 923.5 ± 318.8 N). Significant differences (p = 0.026) were observed comparing group 1 to groups 2 and 3, and group 1 to 4. Three irreparable fractures were found in group 3, four in group 2, and five in groups 1 and 4.
Conclusions: The presence or absence of reinforcement and the different configuration of the reinforcement fibers affect fracture strength but only partially the failure modality. The presence or absence of reinforcement does not alter marginal adaptation.
Keywords: FRC, endodontically treated teeth, glass fibers, fracture, marginal adaptation, onlays.
J Adhes Dent 2015; 17: 567–574. Submitted for publication: 13.04.14; accepted for publication: 10.11.15 doi: 10.3290/j.jad.a35251
a Assistant Professor, Division of Prosthodontics and Maxillofacial Reha- bilitation, Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy. Wrote the manuscript, prepared the specimens, performed the microscopic observations.
b Professor, Division of Cariology and Endodontics, School of Dentistry, Univer- sity of Geneva, Geneva, Switzerland. Wrote the manuscript, performed the statistical analysis.
c Researcher, Division of Prosthodontics and Maxillofacial Rehabilitation, Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy. Wrote the manuscript, performed the fracture test.
d Professor, Division of Cariology and Endodontics, School of Dentistry, Uni- versity of Geneva, Geneva, Switzerland. Supervised the study, contributed to the experimental design and discussion.
Correspondence: Dr. Carlo Monaco, Division of Prosthodontics and Maxil- lofacial Rehabilitation, Department of Biomedical and NeuroMotor Sciences, University of Bologna, Via San Vitale 59, 40125 Bologna, Italy. Tel: +39-051- 208-8186, Fax: +39-051-225-208. e-mail: [email protected]
T
he restoration of endodontically treated teeth has been widely discussed in the prosthodontic and end- odontic literature.30 The higher incidence of fractures in endodontically treated teeth appears to be due more to the loss of structural integrity associated with access preparation, which results in an increased cuspal deflec- tion during function, rather than to structural changes in dentin.11Fracture prevention in posterior teeth is traditionally accomplished with complete tooth preparation, often in combination with the reconstruction of the abutment with a metal or fiber post and a full crown. Sorensen and Martinoff35 even suggested that nonvital posterior teeth should be treated with a cuspal coverage restora- tion. In particular, maxillary premolars are characterized by an unfavorable anatomic shape and a crown volume
and crown:root ratio that make them more prone to frac- tures.34 Preserving tooth structure makes it possible to avoid using posts and allows restoration of the tooth with an onlay. Krejci et al22 showed no in vitro differences in the retention, marginal adaptation, or fracture resistance of endodontically treated and untreated teeth that were restored with an onlay. A recent clinical study found that after 3 years of service, there was no difference in the survival rates of endodontically treated premolars with intact cusps that were restored with fiber posts and a direct composite resin restoration vs those restored with full-coverage metal-ceramic crowns.29 In contrast, a differ- ent study39 confirmed the need to place a final restoration as quickly as possible to protect the cusps. Manhart et al28 reported that annual failure rates of posterior com- posite resin inlays and onlays range between 0% and 10%, and the main reasons for failure are secondary caries, fracture, wear, and marginal deficiencies, suggesting that indirect posterior restorations are a good, longer-lasting alternative for the restoration of large defects. Tamse et al36 found that longitudinal root fractures are more fre- quent in teeth with a narrow mesiodistal dimension, such as maxillary premolars.
Costa et al5 related cusp fractures of endodontically treated maxillary premolars to the width of tooth prepar- ation. These authors found that in premolars, MOD prepar- ation decreased fracture resistance, while an onlay prep- aration with cusp coverage increased fracture resistance.
Magne26 proposed the use of adhesively luted overlays as a more conservative alternative to full crowns, as they reproduce the mechanical properties of the intact tooth from a biomimetic perspective.
The development of fiber-reinforced technology has in- creased the use of composite resin materials for large res- torations. The presence of fibers influences the fracture process by interrupting the progression of crack growth and increasing the fracture toughness of the fiber-rein-
forced composite (FRC).4,17 The use of FRC may prevent undesirable subgingival fractures and have a beneficial effect on the failure mode of composite resin restorations as well as their reparability in the case of fracture.32,40 In spite of the fact that the incorporation of glass in FRC could increase the load-bearing capacity of composite resin restorations,11 some authors31 have emphasized that suboptimal vacuum/pressure adaptation of the fibers may compromise the performance of these restorations.
Fennis et al12 concluded that although glass FRC does not increase fracture load of premolars with cusp-replacing restorations, it can affect the failure mode.
Glass fiber techniques offer another attractive alterna- tive for the treatment of structurally compromised, endo- dontically treated teeth. However, few studies have been conducted to evaluate the fracture resistance of nonvital teeth that have been treated with overlays reinforced with different configurations of woven glass fibers.
Therefore, the purpose of this study was to evaluate marginal adaptation, fracture strength, and failure modal- ity of endodontically treated premolars that were restored with composite resin overlays with and without different configurations of glass fiber reinforcement. The null hy- pothesis is that the presence or absence of reinforcement and the different configurations of reinforcement fibers do not influence marginal adaptation, fracture strength, or failure modality.
MATERIALS AND METHODS
For this study, 32 extracted, caries-free maxillary pre- molars of nearly identical size with completed root growth were selected. The teeth were extracted due to periodontal disease (patients’ ages ranged from 40 to 73 years) and were stored at 4°C in a solution of 0.02%
thymol for a maximum of 3 months from the time of extraction. Their use was in compliance with the Com- mission of Ethics for Research on Humans, University Hospital of Geneva. The teeth were randomly assigned to 4 equal groups. All teeth were endodontically treated, and the root canal preparations were performed with NiTi rotary instruments (Protaper, Denstply Maillefer;
Ballaigues, Switzerland) with a low speed handpiece (Tecnika, Denstply Maillefer) under intermittent rinsing with 5% NaOCl (Niclor 5, Ogna; Muggiò, Italy). An epoxy sealer (AH Plus, Dentsply DeTrey; York, PA, USA) and vertical condensation with warm gutta-percha was used for canal filling (System B, Sybron Dental; Orange, CA, USA). All specimens were fixed with a light-polymerizing composite resin (Filtek Z250, 3M ESPE; Seefeld, Ger- many) on aluminum bases, then embedded in an au- topolymerizing resin (Technovit 4071, Heraeus-Kulzer;
Hanau, Germany) to an apical depth of two thirds of the root length to create a strong load-resistant support for each tooth. Provisional restorations of the pulp chamber were made with Cavit (3M ESPE) before cavity prepar- ation.
Twenty-four hours after endodontic treatment, the clin- ical crowns of all specimens were removed to a level of Fig 1 Design of the tooth preparation for composite resin
onlay.
2.5 mm
mm1
mm1 3.5 mm
2.5 mm above the cementoenamel junction (CEJ) at the buccal cusps and 3.5 mm at the palatal cusps (Fig 1).
The crown was prepared with a 90-degree rounded shoul- der 1.0 mm above the CEJ for half the tooth perimeter, and 1.0 mm below the CEJ in the other half using coarse diamond-coated burs (Geneva Prep Set, Intensiv; Grancia, Switzerland) under copious water-spray cooling. All den- tinal surfaces of the pulp chamber and the cavity walls were sealed using a three-step etch-and-rinse adhesive system (Optibond FL, Kerr; Orange, CA, USA) according to the manufacturer’s recommendations. Light polymerizing was carried out with a halogen device (Polylight Steril 2; Castellini, Italy) for 20 s (output: 800 mW/cm2). The pulp chambers of all groups were filled with a composite resin layer (Miris Dentin S3, Coltène Whaledent; Geneva, Switzerland). Internal and external cavity margins were finished under a stereomicroscope, using a diamond bur (25 μm grain size, No. 3113 NR, Intensiv) and water cooling. Impressions were made in a polyether material (Permadyne, 3M ESPE) using a simultaneous mixing tech- nique according to the manufacturer’s instructions. Provi- sional restorations were constructed with Fermit N (Ivoclar Vivadent; Schaan, Liechtenstein), adapted to the cavity, and light polymerized without interim cement placement.
Laboratory Manufacturing Process
Group 1 was used as the control group and was re- stored only with composite resin overlays (Adoro, Ivoclar Vivadent). Adoro is a microfilled resin composite veneer- ing system containing UDMA monomer. Group 2 was restored with composite resin overlays and 3 FRC layers that were horizontally placed on the bottom of the res- toration (Vectris, Ivoclar Vivadent). Vectris consists of several layers of fiber wafers and woven fiber bundles embedded in an organic polymer matrix. Group 3 was restored with a composite resin overlay and 6 FRC lay- ers in the same manner as group 2, whereas group 4 was restored with a composite resin overlay and FRC placed in an anatomical design to support the compos- ite resin beneath the cusps (Fig 2). An impression of the gypsum model with transparent vinyl polysiloxane paste (VPS, Memosil 2, Ivoclar Vivadent) was taken of each cavity preparation for all groups to form a mold. Differ- ent layers of a pre-impregnated sheet of woven fibers (a “frame”) were placed on the gypsum model and con- densed through a vacuum drawing and polymerization process with transparent VPS. The occlusal anatomies
were formed with a mold. After a cycle of vacuum form- ing and light curing in a specific unit (VS1, Ivoclar Viva- dent) for 10 min, the excess FRC was removed with a carbide bur up to the vertical cavity wall. The FRC base was then airborne particle abraded with small grains of 80 μm (Rocatec system, 3M ESPE) at 0.25 MPa pres- sure for 10 s, then treated with silane (Monobond S, Ivoclar Vivadent). In all groups, the resin composite (Adoro, Ivoclar Vivadent) was built up incrementally, with each increment being pre-cured using a pre-curing light polymerizing unit (Quick, Ivoclar Vivadent). Final poly- merization was performed in the Lumamat 100 (Ivoclar Vivadent) unit using both light and heat. An additional tempering step at 104°C was performed to maximize the strength and surface quality of the restorations.
Adhesive Procedure
The provisional restorations were removed, and the inner surfaces of the teeth that had been previously sealed with bonding agent were airborne particle abraded (particle size 30 μm; CoJet system, 3M ESPE) at a pressure of 0.2 MPa for 2 s. The enamel margins were etched for 30 s with 37% phosphoric acid (Total Etch, Ivoclar Vivadent). After rinsing and drying, primer and adhesive (Optibond FL, Kerr) were applied in a thin layer to the enamel and dentin according to the manufacturer’s instructions. The inner surfaces of the FRC were treated with the CoJet system (particle size 30 μm, 0.2 MPa for 10 s), and two layers of silane coupling agent (Monobond S, Ivoclar Vivadent) were ap- plied to all the inner restoration surfaces. A layer of the same adhesive system was applied with a microbrush to the treated composite surfaces. Restorations were luted with a light-curing composite resin (Miris Enamel WR, Coltene; Altstätten, Switzerland); polymerization was performed for 60 s each on the cervical, buccal, palatal, and occlusal surfaces. The margins of the res- torations were then finished with 15-μm diamond burs (Composhape, Intensiv) and polished with a composite finishing and polishing kit (Astrobrush, Ivoclar Vivadent).
Marginal Evaluation
Before making the impressions for the replicas, the specimens were cleaned with rotating nylon brushes (Hawe Neos; Bioggio, Switzerland) and a dentifrice (Sig- nal Anti Caries, Unilever; London, UK). Polyvinylsiloxane impressions (President Plus Light-body, Coltene) were Fig 2 Schematic of the different
groups with the specific fiber re- inforcement arrangement. Group 1:
no fiber reinforcement (FR); group 2:
3 horizontal layers of FR; group 3: 6 horizontal layers of FR; group 4: ana- tomical arrangement of FR.
Group 1 Group 2 Group 3 Group 4
taken before and after thermomechanical tests to com- pare the quality of marginal adaptation. Gold-sputtered (SCD 030, Provac; Balzers, Liechtenstein) epoxy resin replicas (Epofix, Struers; Copenhagen, Denmark) of all specimens were evaluated for marginal adaptation at the standard 200X magnification using SEM (XL20, Phil- ips; Eindhoven, The Netherlands) and a custom-made module in image processing software (Scion Image, Scion; Frederick, MD, USA). This qualitative evaluation included continuous margins (no gap and no interrup- tion of continuity), noncontinuous margins (a gap due to adhesive or cohesive failure, fracture of the restorative material, or fracture of the enamel related to the restor-
ation margins), overhangs, and underfilled margins at two interfaces, ie, the tooth/luting composite interface and the luting composite/overlay interface. The speci- mens were mechanically loaded on the inner side of the palatal cusp by a sliding movement towards the central fossa. This was accomplished by using palatal cusps of the maxillary first premolars as antagonists with a computer-controlled masticator set to 1,200,000 cycles of 49 N each at a frequency of 1.7 Hz. A total of 3000 thermocycles (from 5°C to 55°C) were performed simul- taneously. The chamber was automatically emptied after 2 min, with 10 s of air pressure to avoid mixing the cold and the warm water.19,21 By having the specimen hold- NS
100.00 90.00 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00
No rein- forcement
FRC 3 horizontal
layers
FRC 6 horizontal
layers Groups
Mean initial continuous margin
FRC anatomical rein forcement
NS
No rein- forcement
FRC 3 horizontal
layers
FRC 6 horizontal
layers Groups
Mean final continuous margin
FRC anatomical rein forcement 100.00
90.00 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00
a b
NS 100.00
90.00 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00
No rein- forcement
FRC 3 horizontal
layers
FRC 6 horizontal
layers Groups
Mean initial continuous margin
FRC anatomical rein forcement
A,B A A B 100.00
90.00 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00
No rein- forcement
FRC 3 horizontal
layers
FRC 6 horizontal
layers Groups
Mean final continuous margin
FRC anatomical rein forcement
a b
Fig 4 Percentages of continuous margins at the luting composite/overlay interface before (a) and after (b) loading. No significant differences (NS) were detected before loading. All significant differences observed between the groups after loading (b) are marked with different letters.
Fig 3 Percentages of continuous margins at the tooth/luting composite interface before (a) and after (b) loading. No significant dif- ferences (NS: p >0.05) were detected between the groups.
ers mounted on a rubber rest, a sliding movement was produced on the restorations during loading. The stress these conditions place on the specimens is believed to simulate approximately 5 years of normal oral func- tion.18,20
After thermocycling and mechanical loading, all groups were subjected to a fracture test designed to measure the fracture strength and patterns. The test was performed at a temperature of 25°C using a universal testing ma- chine (Instron Model 4301, Instron; Norwood, MA, USA).
A spherical stainless steel punch with a thin copper layer and a diameter of 4.0 mm was used at 90 degrees in the middle of the specimen’s occlusal surface. A central fossa was prepared in the premolars in order to insert the spherical punch. The crosshead speed was 1.0 mm/min, and the load was applied until the specimen fractured.
The load-deflection curves were recorded with PC com- puter software (Instron 8.2, Instron). The start of speci- men damage can be classified as initial failure (IF).10 The IF was recorded when at least two of the following condi- tions were met: (1) there was a sharp decline in the load/
displacement curve, called a knee or corner, (2) visible signs of fracture were observed, or (3) audible emissions, caused by the generation of elastic waves by crack forma- tion and/or progression, were heard. The final failure (FF) of the specimen was defined when one of the following oc- curred: (1) the specimen attained an “instable” condition, with a zero (or negative) slope of applied stress vs strain, (2) the maximal load or displacement before the load de- creased by 50%, or (3) an apparent catastrophic rupture was observed. The failure mode or damage accumulation (restoration chipping, restoration fracture, restoration and coronal fracture, restoration and irreparable fracture [if below the CEJ]) was noted and determined after testing.
Differences in the means were compared with the use of matched pairs of t-tests and one-way ANOVA. The level of significance was set at α = 0.05.
RESULTS
Percentages of continuous margins before and after loading at both the tooth/composite and composite/
onlay interfaces are shown in Fig 3.
At the tooth/luting composite interface, no significant differences between the groups could be observed either before (p = 0.418) or after loading (p = 0.240), indicat- ing that the fiber reinforcement had no significant influ- ence on the marginal adaptation at this interface (Fig 3).
At the luting composite/overlay interface, no significant differences between the groups were observed before loading (p = 0.586). However, after loading, significant differences (p = 0.026) were observed between group 1 vs groups 2 and 3, and group 1 vs group 4. Interest- ingly, the lowest marginal adaptation scores (Fig 4) were observed in the group with no reinforcement (group 1, 90.1% ± 16.5%) and in the group with anatomic cusp reinforcement (group 4, 81.6% ± 12.8%). The results of fracture load are detailed in Fig 5. For both initial and final failure parameters, groups 3 (1428.1 ± 251.4 N and 1467.9 ± 242.4 N) and 4 (1431.8 ± 294.3 N and 1710.1 ± 326.6 N) resulted in significantly higher fracture loads than groups 1 (899.8 ± 352.7 N and 923.5 ± 318.8 N) and 2 (852.6 ± 413.5 N and 1058.1 ± 251.5 N).
In group 4, with FRC anatomic cusp reinforcement, and in group 3, with the horizontal 6-layer FRC (Fig 6), significantly higher loads were required to cause speci- men fracture. In respect to failure types, groups 1 and 4 had the highest number of catastrophic failures (ir- reparable, with fracture below the CEJ) compared to the rest of the groups (Table 1). Fisher’s Exact Test was per- formed to evaluate the effect of the presence/absence of reinforcement on catastrophic failures (irreparable fracture). The test (p = 0.691) found no statistically sig- nificant correlation between the catastrophic failure and the presence or absence of reinforcement.
a b
2000.00
1500.00
1000.00
500.00
0.00
No rein- forcement
FRC 3 horizontal
layers
FRC 6 horizontal
layers Groups
Mean initial failure
FRC anatomical rein forcement
B B A A
2000.00
1500.00
1000.00
500.00
0.00
No rein- forcement
FRC 3 horizontal
layers
FRC 6 horizontal
layers Groups
Mean ultimate strength load
FRC anatomical rein forcement
B B A A
Fig 5 Fracture load results with the initial (a) and ultimate strength values in N (b). Different letters indicate statistically significant differences.
DISCUSSION
In this study, the marginal adaptation after mechani- cal fatigue cycles and thermocycling was evaluated. In addition, the fracture resistance of nonvital premolars restored with composite resin overlays with or without woven fiber reinforcement was assessed. Both of these factors are essential for achieving a durable restoration that can effectively protect the remaining tooth struc- ture. When considering the marginal adaptation results obtained here, it is worth noting that the application of the fatigue and thermal cycles led to a loss of integrity for both interfaces evaluated in all four study groups.
This loss of integrity was most evident at the tooth/
cement interface, but considerably less at the cement/
overlay interface. De Munck et al7 showed that the aging induced by thermocycling leads to both contrac- tion and expansion stresses at the tooth/restoration interface due to the different thermal coefficients of the restoration and the tooth structure. Moreover, thermocy- cling accelerates the chemical degradation of the adhe- sive interface. In particular, the lowest marginal quality of the tooth/cement interface was recorded for group 4 specimens, while the highest results were recorded for group 2 specimens. It must also be emphasized that, after the thermal and fatigue cycles, the marginal continuity of the tooth/cement interface was reduced by half in all groups, and by one-third of the initial value in group 2. In contrast, the marginal adaptation of the cement/overlay interface remained constant even after cyclic tests. Several authors25 have confirmed that mar- ginal continuity is a key factor required for achieving a durable restoration. A recent study24 demonstrated that the use of inlays reinforced with unidirectional fibers did not decrease the marginal adaptation compared to the use of nonreinforced inlays. Another study8 evaluated the marginal adaptation of molars after thermocycling and fatigue of composite resin overlays reinforced with glass fiber, ceramic, or composite. Those authors found no statistically significant differences at the tooth/com- posite and composite resin/restoration interfaces, but differences were found after thermal fatigue cycling in the group restored with ceramic-reinforced overlays. In the present study, lower values of marginal adaptation were obtained for the groups restored with reinforced restorations. This phenomenon could be due to the in-
creased difficulty of managing a higher number of fiber layers in the smaller volume of a premolar, with the risk of fiber exposure and decoupling from the polymer ma- trix.15
When looking at the fracture resistance results re- ported by Dere et al,8 it is important to remember that their results were obtained on molars rather than premo- lars. In this work, the highest values of fracture resistance were recorded in the group with the woven glass fibers arranged in an anatomical design (1710.12 ± 326.64 N), while the lowest values of fracture resistance were re- corded in the group of teeth restored with non-reinforced overlays (923.50 ± 318.80 N). Intermediate values were obtained from the experimental groups with over- lays strengthened by three (1058.13 ± 251.51 N) and six (1467.88 ± 242.36 N) layers of woven glass fibers.
These data show an improvement in the fracture strength depending on the number of fibers layers. However, the use of woven fiber reinforcement reduces the fracture resistance moreso than does the use of unidirectional fibers. The use of woven fibers causes the restoration to exhibit orthotropic behavior. Hence, the strength of speci- mens was reduced by 25% to 50%, because not all fibers are arranged perpendicularly (as opposed to the case with unidirectional fibers) to the direction of load application, as described by the equation of Krenchel.23 However, the results obtained in this work are different from those of Akman et al.1 The latter found that the addition of fiber re- inforcement of composite restorations does not improve the fracture strength on molars.
Dere et al8 demonstrated that fiber-reinforced overlays, in addition to producing higher fracture toughness values, are also able to produce more favorable fracture patterns.
It should be emphasized that the fracture resistance val- ues recorded in all groups of this study represent limits that are difficult to reach in vivo, because the published data on bite forces indicate that the maximum values occurring in the posterior area vary between 300 and 880 N.14
Dyer et al9 emphasized that the initial failure is a much more significant and useful parameter than the fracture resistance, because the penetration of oral fluids leads to decay and a reduction in the service life of the restoration.
When early failure was identified, the load that produced the failure was much lower than the fracture load. For this reason, it is important to understand whether the initial Table 1 Type and number of failures in the different groups
Type of failure
No. of failures by group
Group 1 Group 2 Group 3 Group 4 Irreparable
(below the CEJ) 5 4 3 5
Reparable
(above the CEJ) 3 4 5 3
Fig 6 A specimen of group 3 after the frac- ture test. The fracture line involved only the restoration, without damaging the tooth.
failure also caused a loss of the seal at the adhesive interface. Coronal microleakage is closely related to endo- dontic treatment and subclinical failures.33 This important issue was evaluated by Belli et al,2 who found that also the use of flowable composite in combination with glass fibers can reduce occlusal leakage in Class II adhesive cavities with enamel margins.
The lack of a statistically significant correlation be- tween catastrophic failure and the presence or absence of reinforcement is probably due to the small number of specimens in each group. Nevertheless, reinforced specimens had a lower tendency to show unrecoverable fractures. The group with composite overlays reinforced with 6 FRC layers showed fracture patterns that would favor retreatment in at least a 62.5% of cases (Table 1).
In this group, the arrangement of fibers was able to keep the fracture line within the overlay, keeping the restoration margins of the tooth intact. The ability of the reinforcing fibers to change the fracture patterns has already been reported in previous studies.37,38 In particular, Fennis et al12 found that FRC positively influenced the failure mode of maxillary premolars subjected to thermocycling and static load tests, and that woven fibers are more effective than unidirectional fibers.
Our study also confirms the results of Brunton et al,3 who demonstrated how failures under conditions of com- pressive loading may be less catastrophic when a fiber- reinforced composite onlay is used instead of an non- reinforced one. It should be stressed that for this group, as in the other two groups tested in this study, that over a third of the failures were re-treatable. These results also confirm the findings of Magne and Knezevic,27 who showed that composite resin overlays are an optimal so- lution for the restoration of endodontically treated teeth.
The fact that in the group with the anatomic FRC 62.5%
of failures were catastrophic and irreparable (fracture below CEJ) could be due to marginal degradation that oc- curred at the composite/onlay interface after loading. It is possible that loss of the seal at the adhesive interface may have negatively affected the failure pattern. Another possible explanation for the many catastrophic failures in the anatomic FRC group may be the high values of final fracture load observed in this group, suggesting that the load required to cause restoration failure was so high that it produced failures in the adhesive interface and root fracture.
CONCLUSIONS
Within the limitations of this in vitro study, it can be con- cluded that the use of overlays reinforced with woven fibers did not limit the marginal adaptation compared to non-reinforced overlays. In addition, the presence or ab- sence of reinforcement and the different configurations of the fibers affect fracture strength but only partially the failure modality. Composite resin overlays reinforced with woven glass fibers can improve the ability of the restoration to withstand chewing forces. Higher fracture resistance is not necessarily associated with more
favorable types of failure in the prosthetic restoration of endodontically treated teeth. These results should, however, be confirmed by controlled clinical trials in the future.
REFERENCES
1. Akman S, Akman M, Eskitascioglu G, Belli S. Influence of several fibre- reinforced composite restoration techniques on cusp movement and fracture strength of molar teeth. Int Endod J 2011;44:407-415.
2. Belli S, Orucoglu H, Yildirim C, Eskitascioglu G. The effect of fiber place- ment or flowable resin lining on microleakage in Class II adhesive res- torations. J Adhes Dent 2007;9:175-181.
3. Brunton PA, Cattell P, Burke FJT, Wilson NHF. Fracture resistance of teeth restored with onlays of three contemporary tooth-colored resin- bonded restorative materials. J Prosthet Dent 1999;82:167-171.
4. Butterworth C, Ellakwa AE, Shortall A. Fibre-reinforced composites in re- storative dentistry. Dent Update 2003;30:300-306.
5. Costa LCS, Pegoraro LF, Bonfante G. Influence of different metal res- torations bonded with resin on fracture resistance of endodontically treated maxillary premolars. J Prosthet Dent 1997;77:365-369.
6. Craig WH, Courtney TH. On the tension test as a means of character- izing fibre composite failure mode. J Mater Sci 1975;10:1119-1126.
7. De Munck J, Van Landuyt K, Peumans M, Poitevin A, Lambrechts P, Braem M, Van Meerbeek B. A critical review of the durability of adhesion to tooth tissue: methods and results. J Dent Res 2005;84:118-132.
8. Dere M, Özcan M, Göhring TN. Marginal quality and fracture strength of root-canal treated mandibular molars with overlay restorations after thermocycling and mechanical loading. J Adhes Dent 2010;12:287-294.
9. Dyer SR, Lassila LVJ, Jokinen M, Vallittu PK. Effect of fiber position and orientation on fracture load of fiber-reinforced composite. Dent Mater 2004;20:947-955.
10. Dyer SR, Sorensen JA, Lassila LV, Vallittu PK. Damage mechanics and load failure of fiber-reinforced composite fixed partial dentures. Dent Mater 2005;21:1104-1110.
11. Faria AC, Rodrigues RC, de Almeida Antunes RP, de Mattos Mda G, Ribeiro RF. Endodontically treated teeth: characteristics and consider- ations to restore them. J Prosthodont Res 2011;55:69-74.
12. Fennis WM, Tezvergil A, Kuijs RH, Lassila LV, Kreulen CM, Creugers NH, Vallittu PK. In vitro fracture resistance of fiber reinforced cusp-replacing composite restorations. Dent Mater 2005;21:565-572.
13. Garoushi SK, Lassila LVJ, Vallittu PK. Fibre-reinforced composite in clin- ical dentistry. Chin J Dent Res 2009;12:7-14.
14. Gibbs CH, Mahan PE, Mauderli A, Lundeen HC. Limits of human bite strength. J Prosthet Dent 1986;56:226-229.
15. Göhring TN, Roos M. Inlay-fixed partial dentures adhesively retained and reinforced by glass fibers: clinical and scanning electron microscopy analysis after five years. Eur J Oral Sci 2005;113:60-69.
16. Hickel R, Manhart J. Longevity of restorations in posterior teeth and rea- sons for failure. J Adhes Dent 2001;3:45-64.
17. Keulemansa F, Palava P, Abousheliba MMN, Van Dalena A, Klever- laana CJ, Feilzera AJ. Fracture strength and fatigue resistance of dental resin-based composites. Dent Mater 2009;25:1433-1441.
18. Krejci I, Reich T, Lutz F, Albertson M. In-vitro-Testverfahren zur Evalu- ation dentaler Restaurationssysteme. Schweiz Monatsschr Zahnmed 1990a;100:8-14.
19. Krejci I, Reich T, Lutz F, Albertoni M. An in vitro test procedure for evalu- ating dental restoration systems. 1. A computer-controlled mastication simulator [in German]. Schweiz Monatsschr Zahnmed 1990b;100:
953-960.
20. Krejci I, Lutz F. In-vitro test results of the evaluation of dental restor- ation systems. Correlation with in-vivo results [in German]. Schweiz Monatsschr Zahnmed 1990c;100:1445-1449.
21. Krejci I, Kuster M, Lutz F. Influence of dentinal fluid and stress on mar- ginal adaptation of resin composites. J Dent Res 1993;72:490-494.
22. Krejci I, Duc O, Dietschi D, De Campos E. Marginal adaptation, reten- tion and fracture resistance of adhesive composite restorations on devital teeth with and without posts. Oper Dent 2003;28:127-135.
23. Krenchel, H. Fibre reinforcement [PhD thesis]. Copenhagen: Technical University of Denmark, 1963.
24. Kumbuloglu O, Tezvergil-Mutluay A, Saracoglu A, Lassila LVJ, Vallittu P.
Marginal adaptation and microleakeage of directly and indirectly made fiber reinforced composite inlays. Open Dent J 2011;5:33-38.
25. Lutz F, Krejci I, Barbakow F. Quality and durability of marginal adapta- tion in bonded composite restorations. Dent Mater 1991;7:107-113.
26. Magne P. Composite resins and bonded porcelain: the postamalgam era?. J Calif Dent Assoc 2006;34:135-47.
27. Magne P, Knezevic A. Influence of overlay restorative materials and load cusps on the fatigue resistance of endodontically treated molars. Quin- tessence Int 2009;40:729-737.
28. Manhart J, Chen HY, Hamm G, Hickel R. Buonocuore Memorial Lecture.
Review of the clinical survival of direct and indirect restorations in pos- terior teeth of the permanent dentition. Oper Dent 2004;29:481-508.
29. Mannocci F, Bertelli E, Sherriff M, Watson TF, Ford TRP. Three-year clin- ical comparison of survival of endodontically treated teeth restored with either full cast coverage or with direct composite restoration. J Prosthet Dent 2002;88:297-301.
30. Mannocci F, Cowie J. Restoration of endodontically treated teeth. Br Dent J 2014;216:341-346.
31. Ohlmann B, Dreyhaupt J, Schmitter M, Gabbert O, Hassel A, Rammels- berg P. Clinical performance of posterior metal-free polymer crowns with and without fiber reinforcement: one-year results of a randomised clin- ical trial. J Dent 2006;34:757-762.
32. Oskoee PA, Ajami AA, Navimipour EJ, Oskoee SS, Sadjadi J. The effect of three composite fiber insertion techniques on fracture resistance of root-filled teeth. J Endod 2009;35:413-416.
33. Ravanshad S, Ghoreeshi N. An in vitro study of coronal microleak- age in endodontically-treated teeth restored with posts. Aust Endod J 2003;29:128-133.
34. Schwartz RS, Robbins JW. Post placement and restoration of endodonti- cally treated teeth: a literature review. J Endod 2004;30:289-301.
35. Sorensen JA, Martinoff JT. Intracoronal reinforcement and coronal coverage: a study of endodontically treated teeth. J Prosthet Dent 1984;51:780-784.
36. Tamse A, ZilburgI, HalpernJ. Vertical root fractures in adjacent max- illary premolars: an endodontic-prosthetic perplexity. Int Endod J 1998;31:127-132.
37. Tezvergil A, Lassila LVJ, Vallittu PK. Strength of adhesive bonded fiber- reinforced composites to enamel and dentine substrates. J Adhes Dent 2003;5:301-311.
38. Tezvergil A, Lassila LVJ, Vallittu PK. The shear bond strength of bidirec- tional and random-oriented fibre-reinforced composite to tooth struc- ture. J Dent 2005;33:509-516.
39. Tang W, Wu Y, Smales RJ. Identifying and reducing risks for potential fractures in endodontically treated teeth. J Endod 2010;36:609-617.
40. Van Heumen CCM, Kreulen CM, Creugers NHJ. Clinical studies of fiber- reinforced resin-bonded fixed partial dentures: a systematic review. Eur J Oral Sci 2009;117:1-6.
Clinical relevance: The use of fiber-reinforced over- lays is a valid alternative to prosthetic crowns in end- odontically treated teeth. The use of fiber-reinforced overlays, in addition to allowing a greater preserva- tion of tooth structure, can improve fracture strength.
