Influence of the endodontic access cavity design and restorative technique on hard tissue removal and fracture resistance of mandibular premolars

This study assessed the influence of traditional (TradAC) and conservative access (ConsAC) with different restorative techniques on the percentage of hard tissue removed (%HTR) and on the fracture resistance of mandibular premolars. 45 premolars were scanned in a micro-computed tomography and assigned into four groups according to access (TradAC or ConsAC) and restorative technique: composite resin (CR) or fiber post (FP) + CR. After post preparation, the teeth were rescanned to determine the volume enlargement and %HTR from the crown and root canal. After restoration, the load at fracture was recorded. Data were analyzed statistically by one-way ANOVA and Tukey’s posthoc test, ANOVA repeated measure, and chi-square tests (P<0.05). TradAC (RC or FP) resulted in the increase (Δ%) of root canal volume and hard tissue removed up to 14 mm (%) in comparison with ConsAC (RC or FP). TradAC + FP removed a greater percentage of hard tissue from the crown when compared to TradAC + CR. The percentage of hard tissue removed in the crown in the ConsAC groups was statistically lower than in the TradAC groups. The control group showed higher fracture resistance than all experimental groups, with no differences among the latter. Restorable fracture patterns were more prevalent. Traditional endodontic access cavities removed a higher percentage of dentine than conservative endodontic access cavities. However, no differences in fracture resistance were observed. Restorations using composite resin or fiber post associated with composite resin showed similar results of fracture resistance.


Introduction
Minimally invasive endodontic access cavities have been proposed as an alternative to traditional endodontic access cavities (TradAC) to preserve as much tooth structure as possible and thus theoretically maintain the resistance to fracture of root-filled teeth (Clark & Khademi, 2010). However, this superior resistance has not been scientifically validated. Some studies have shown an increase in fracture resistance (Krishan et al. 2014, Plotino et al. 2017), but most of them have observed that access cavity design does not interfere with the fracture resistance of endodontically treated teeth (Rover et al. 2017, Sabeti et al. 2018. Previous reviews have concluded that there is no evidence supporting the use of minimally invasive endodontic access cavities to improve the fracture resistance of root-filled teeth when compared to TradAC (Silva et al. 2018. Moreover, several studies have also demonstrated that minimally invasive endodontic procedures might impair other outcomes of root canal treatment such as cleaning, shaping, and filling procedures , Rover et al. 2017).
In addition to access cavity design, other factors might also be related to fracture resistance of root-filled teeth, such as the type of coronal restoration (Gillen et al. 2011, De Rose et al. 2015. The restoration of root-filled treated teeth varies from a direct composite resin to a post associated with a crown, depending on the remaining structure , Trushkowsky 2014, von Stein-Lausnitz et al. 2019. Fiber posts are indicated mainly for teeth with a reduced coronal structure to provide retention of the core material (Trushkowsky 2014). However, previous studies evaluating different groups of teeth with coronal access and maintenance of the four coronary walls have shown a significant increase in fracture resistance when the teeth are restored with fiber posts (Fadag et al. 2018, Nam et al. 2010).
Owing to the effect of access cavities and the importance of coronal restoration for the long-term success of root canal treatment, this study aimed to assess the impact of traditional and conservative access cavities associated with different restorative techniques on the percentage of hard tissue removed and on the fracture resistance of mandibular premolars. The null hypotheses tested were that there would be no difference between access cavities (traditional or conservative endodontic access) and restorative techniques (fiber post or composite resin) in (i) % of hard tissue removal in the crown and until 14mm of the teeth and (ii) fracture resistance of teeth.

Sample size and group selection
The study was approved by the Ethics Committee of the Institution (Approval number 2.431.596). The donation of teeth was obtained through to sign informed consent. The effect size was established based on data from a previous study (=0.6) ). The ANOVA: Fixed effects, omnibus, one-way test was selected from the F tests family in G*Power 3.1 software for Windows (Heinrich-Heine-Universität, Düsseldorf, Germany). Accordingly, for an analysis with α=0.05, 95% power, and effect size of 0.6, a total of 36 samples (n=9) were indicated as the ideal size. Ten samples per group were used in each of the following groups: TradAC-CR (traditional endodontic access + composite resin), TradAC-FP (traditional endodontic access + fiber post), ConsAC-CR (conservative endodontic access + composite resin) and ConsAC-FP (conservative endodontic access + fiber post). Control with 5 intact teeth was included.
Seventy mandibular premolars were observed by visual inspection at 10X magnification and periapical radiographs were obtained (CDR Elite, Long Island, NY, USA) to select teeth with similar characteristics, with a single canal, and without endodontic treatment. The width of the crowns in the buccolingual and mesiodistal directions and the total length of the tooth and the root were measured with a digital caliper (Matrix®-MTX, ToolsWorld, China).
The CTan (1.15.4.0, Bruker, micro-CT) and Image J softwares were used to quantify hard tissue volume in the crown and up to the cervical 14 mm of the root, as well as the volume (mm³) of the pulp chamber and root canal. The CTvol software (Bruker, micro-CT) was used to verify the three-dimensional configuration of the root canal. Only mandibular premolars with type 1 anatomy were selected (Ahmed et al. 2017). The teeth were then divided into four experimental groups (n=10) and one control group (n=5) based on a similar initial amount of hard tissue, root canal volume, and tooth length.

Access cavity preparation
Access cavities were prepared in all groups except the control.

Root canal preparation and filling procedures
Preflaring was performed with 35/.05 rotary instruments (Bassi Logic™, Belo Horizonte, MG, Brazil) up to one third of the root canal. Patency was achieved with a size 10 C-Pilot instrument (VDW, Munich, Germany) and the working length was established 1 mm beyond the apical foramen. The apical third was then prepared with 35/.01 and 35/.05 instruments (Bassi Logic™). During preparation, the teeth were irrigated with 2% chlorhexidine and 17% EDTA was used for smear layer removal. Saline solution was used as the final irrigant.

Fiber post space preparation
Fiber post preparation was performed with #3 and #2 Largo drills in the TradAC-FP and ConsAC-FP groups, respectively, parallel to the long axis of the root, at a depth of 14 mm, from the coronal to the apical third, and the post space was then rinsed with saline.

Micro-CT assessment
After root canal treatment and post preparation, a new micro-CT scan was performed using the same parameters mentioned earlier. The open-source 3D Slicer software, available online at http://slicer.org -National Institutes of Health, was used to register the 3D models pre-and post-preparation with an affine algorithm and the CTAn software (Bruker, micro-CT) was then used for all image analysis procedures. The region of interest was 14 mm, as mentioned previously, resulting in 600-670 transverse cross-sections per tooth. Next, the gray scale range required to recognize hard tissue (enamel and dentine) and the root canal was determined with a density histogram using an automatic threshold tool.
The mean percentage increase (Δ%) of root canal volume was calculated as described previously (Versiani et al. 2018). The percentage of hard tissue removed after root canal preparation and post preparation was calculated in the crown and up to the cervical 14 mm of the root canal according to the formula: The CTvol software (Bruker, micro-CT) was used for comparison of the overlapping images before (green) and after the endodontic access and post preparation (red).

Composite resin
In the CR groups, 37% phosphoric acid (Condac 37, FGM Produtos Odontológicos, Joinville, SC, Brazil) was used for 15 seconds, followed by washing and drying with air jets. The Scotchbond Multipurpose adhesive system (3M, St. Paul, MN, USA) was used. First, the Adper Activator was applied with a micro applicator (Cavibrush, FGM), followed by drying with an air jet for 5 seconds and with absorbent paper points. Next, the Adper Primer (3M) was applied in the same way and finally, the Catalyst was applied, with the solvents being volatilized with an air jet and absorbent paper cones being used to remove the excess. Research, Society and Development, v. 11, n. 1, e18511124575, 2022 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v11i1.24575

5
The teeth were restored with composite resin only in the coronary portion (Filtek Z250, 3M) using the incremental technique. Increments were polymerized for 20 seconds with light curing (Flash Lite 1401, Discus Dental Inc, Culver City, CA, USA) at an intensity of 1400 MW/cm2 and wavelength between 465 and 475 nm, followed by polishing with 3018F burs (KG Sorensen).

Fiber post cementation
The #1 DCE and # 0.5 DC prefabricated fiberglass posts were used in the FP groups (Whitepost, FGM, Joinville, SC, Brazil), respectively. Posts were cleaned with 37% phosphoric acid for 60 seconds, washed for 30 seconds, and air-dried. Next, the silane coupling agent (RelyX Ceramic Primer, 3M) was applied and the solvent was volatilized with an air jet for 3 minutes.
The resin cement (RelyX ARC, 3M) was inserted into the root canal and the post was positioned inside the root canal at 14 mm. Excess cement was removed and the buccal and lingual surfaces were photoactivated for 1 minute. Excess post material in the coronal portion was removed 1 mm below the occlusal surface and this area was restored with composite resin.

Load to fracture
Prior to the fracture test, the periodontal ligament and alveolar bone were simulated based on previous studies (Soares Next, the specimens were analyzed under a light microscope at 10X magnification to determine the fracture patterns. The specimens were classified as "restorable" when the failures were above the level of bone simulation (site of fracture above the acrylic resin) and "unrestorable" when the failures extended below the level of bone simulation (site of fracture below the acrylic resin).

Statistical analysis
The normality of the data was determined by the Kolmogorov-Smirnov test using the SPSS software (version 19.0, Statistical Package for the Social Sciences). Since the micro-CT and load at fracture data showed a normal distribution, parametric tests were chosen for statistical analysis. Data were analyzed statistically by one-way ANOVA and Tukey's posthoc test for increase Δ%, Hard tissue removed up to 14 mm (%), Hard tissue removed crown (%) and Load to fracture.
ANOVA repeated measures was carried out to analyze the difference between groups according to time (the sphericity was assumed for the analysis). The chi-square test was used for fracture patterns. The level of significance was set at 5% in all analyses.

Results
The morphological similarities (baseline) of the tested groups were measured with the use of a caliper. No significant differences were observed in the dimensions of the crown in the buccolingual and mesiobuccal direction, total tooth length, or root length. The degree of homogeneity (baseline) of the tested groups was also confirmed by micro-CT scanning regarding hard tissue volume and root canal volume (mm³) (P>0.05) ( Table 1).    For root canal volume in the ANOVA repeated measure test (Table 2), it was observed the effect of time among the 4 groups, and the interaction between time and groups, but not among the type of groups. In the initial time, there was no difference between the groups, but in the final time TradAC-CR group was statistically different than ConsAC-CR; ConsAC-CR group was different than TradAC-FP group; TradAC-FP group was different than ConsAC-CR and ConsAC-FP groups.

Means followed by different superscript letters differ significantly between the different groups (One
For hard tissue volume up to 14 mm it was observed effect of time among the 4 groups, and the interaction between time and groups, but not among the groups (access cavities and restorative technique). There was difference between initial and final, however, no difference was visualized between groups. For hard tissue volume in the crown, it was observed effect of time among the 4 groups, but not in the interaction between time and groups. There was difference between initial e final in all groups, except for ConsAC-CR; however, between groups, there was no statistical difference. The maximum strength to fracture is shown in Table 3. The control group differed significantly from the other groups (P<0.001), with no difference among experimental groups (P=0.373). A restorable fracture pattern was more prevalent in all groups, with no statistical difference (chi-square test, P=0.81). Figure 3 shows representative images of the restorable and unrestorable fracture patterns.

Discussion
The present study evaluated the impact of different access cavities and restorative techniques on the percentage of hard tissue removed and on the fracture resistance of mandibular premolars. The length of 14 mm of preparation depth for fiber post cementation was used since this measure corresponds approximately to the value of the middle and cervical thirds of the premolars used, which is an acceptable length for post preparation and because most fractures of root canal-filled teeth occur in the cervical region (Pierrisnard et al. 2002).
Sodium hypochlorite affects the mechanical properties of dentin by degradation of the organic dentin components.
Also, due to its oxidative power, which leaves an oxygen-rich layer on the dentin surface, its use affects the polymerization of adhesive materials (Prado et al. 2016). Thus, in the present study, 2% chlorhexidine was chosen as irrigant during root canal preparation. Other advantages of chlorhexidine as irrigant include broad-spectrum antimicrobial activity, substantivity and a positive effect on dentin adhesion, showing the capacity to preserve the durability of the hybrid layer and bond strength (Gomes et al. 2013).
Regarding the crown and root canal up to the cervical 14 mm of the root, increase root canal volume(Δ%) and a greater amount of hard tissue removed (%) were observed in the TradAC groups than in the ConsAC groups, an expected result since the entire roof of the pulp chamber was removed and a conical and elliptical shape of the pulp cavity was achieved (Patel & Rhodes 2007). Thus, the first null hypothesis was partially rejected.
When only the crown was evaluated, the percentage of hard tissue volume removed was higher in the TradAC-FP group than in the TradAC-CR group, showing that the fiber post promoted a higher percentage of dentine removal, in line with previous studies (Ikram et al. 2009, Shaikh et al. 2018. Moreover, a greater percentage volume of hard tissue was removed in the TradAC groups when compared to the ConsAC groups, in agreement with previous studies that evaluated traditional and conservative access cavities (Isufi et al. 2020. Regarding the impact of access cavities on the fracture resistance of premolars, this study did not observe a significant difference between access cavities, in line with other studies that compared these types of access using different types of teeth (Rover et al. 2017, Sabeti et al. 2018). However, Krishan et al. (2014), in a study on premolars and molars, observed an increased fracture resistance in minimally invasive access when compared with traditional access. This divergence may be due the fact that the teeth evaluated by the cited authors were not restored.
Clinically, dentists may not choose fiber posts to restore teeth with four intact walls. However, previous studies found that fiber posts could improve the fracture resistance of teeth with four walls (Fadag et al. 2018, Nam et al. 2010, justifying the clinical significance of this type of study. In the present study, no difference in fracture resistance was observed between the restorative procedures, in agreement with a previous study (von Stein-Lausnitz et al. 2019). However, other studies respectively evaluating premolars and incisors (Fadag et al. 2018, Nam et al. 2010) have reported that fiber posts improved the fracture resistance of teeth with four walls. These differences may be linked to differences in sample preparation, fiber post preparation/cementation, and the dental group.
The similar fracture resistance results in the experimental groups and the restorable fracture patterns can be explained by the maintenance of the four surfaces of all teeth that reinforces the dental structures and increases the resistance of the premolar (Soares et al. 2008). When a fiber post is used, it permits good stress distribution since the elastic modulus of the fiber post is like that of dentine (Nam et al. 2010, Soares et al. 2008.
In this study, the smaller hard tissue volume removed with ConsAC had no positive effect on fracture resistance.
Composite resin is a good option as a restorative procedure for root canal-filled teeth with four walls because it is of low cost, it is a faster clinical procedure since it does not require post preparation or cementation, and it is a less error-sensitive technique than the use of a fiber post , Soares et al. 2008.
A great effort was employed to ensure the homogeneity of the specimens regarding configuration, volume and surface area of the root canals based on preoperative scans, however, as it is a laboratory study, it has some limitations, such as performing the static fracture test and the absence of the simulation of chewing in the oral cavity. Thus, it is suggested to carry out further studies with the use of cyclic fatigue to approach the simulation of clinical conditions.

Conclusion
Traditional endodontic access cavities removed a higher percentage of dentine than conservative endodontic access cavities. However, no differences in fracture resistance were observed. Restorations using composite resin or fiber post associated with composite resin showed similar results of fracture resistance and fracture patterns.