Influence of different obturation techniques in coronal bacterial infiltration: study in dogs Influência de diferentes técnicas de obturação na infiltração bacteriana coronária: estudo em cães Influencia de diferentes técnicas de obturación en la infiltración bacteriana coronal: estudio en perros

The objective of this study was to evaluate, in vivo, coronal bacterial infiltration after endodontic treatment with different obturation technique. Forty-five dogs’ root canals, originated from incisors and premolars, were used. The animals were intubated after general anesthesia. After local antisepsis and placement of rubber dam isolation, teeth were opened and instrumented up to a Kerr handfile #40, followed by three obturation protocols with Endofill: Research, Society and Development, v. 10, n. 4, e11010413884, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i4.13884 2 Lateral condensation, Lateral condensation with a coronal plug of set Endofill and Tagger hybrid technique. Access openings were not sealed and root fillings remained exposed to oral environment for 90 days. After this period, animals were euthanized and specimens were histologically processed and stained with Brown and Brenn. Dentinal tubules were evaluated with presence or absence of bacteria descriptive analysis. Bacterial infiltration was identified on root canal walls in six out of 14 root canals filled with the lateral condensation technique (42,8%), two out of 15 canals filled with Lateral condensation with a plug of set Endofill (13,3%) and in two out 13 root canals filled with the Tagger hybrid technique (15,3%). Although the use of a coronal plug or a thermomechanical compaction technique showed less bacterial infiltration than conventional lateral condensation, none of the obturation techniques prevented bacterial infiltration to periapical area, evidencing the importance of a proper coronal seal or final restoration.

Lateral condensation, Lateral condensation with a coronal plug of set Endofill ® and Tagger hybrid technique. Access openings were not sealed and root fillings remained exposed to oral environment for 90 days. After this period, animals were euthanized and specimens were histologically processed and stained with Brown and Brenn. Dentinal tubules were evaluated with presence or absence of bacteria descriptive analysis. Bacterial infiltration was identified on root canal walls in six out of 14 root canals filled with the lateral condensation technique (42,8%), two out of 15 canals filled with Lateral condensation with a plug of set Endofill ® (13,3%) and in two out 13 root canals filled with the Tagger hybrid technique (15,3%). Although the use of a coronal plug or a thermomechanical compaction technique showed less bacterial infiltration than conventional lateral condensation, none of the obturation techniques prevented bacterial infiltration to periapical area, evidencing the importance of a proper coronal seal or final restoration. Keywords: Dental leakage; Dental pulp cavity; Endodontics; Root canal obturation.

Introduction
The objective of endodontic therapy is to prevent or heal periapical periodontitis, mainly caused by microorganisms and their by-products. However, microleakage may occur after obturation, recontaminating root canal system (Yamauchi et al. 2006, Braz Junior et al. 2021, defined as "the clinically undetectable passage of bacteria, fluids, molecules or ions between tooth and the restorative or filling material" and may be considered one of the most important risk factor for apical periodontitis in endodontically treated teeth (Muliyar et al. 2014).
Although periapical tissue is the most common affected area by endodontic disease, special attention should be directed to the coronal portion of the root canal, since an adequate coronal sealing is primordial for healing after endodontic treatment, previously associated with less inflammatory infiltrate (Yamauchi et al. 2006) or even reduction of apical periodontitis (Shiper et al. 2005). The recontamination of obturated root canal via coronal leakage allows bacteria to reach periapical tissues, and can occur in cases where obturation is exposed to the oral environment, usually found in: (a) loss, fracture or failure of restoration; (b) recurrent caries; (c) delay in restoration, with infiltration through temporary restorative material; and (d) during the restorative procedure (Khayat et al. 1993). In addition, near 30% of endodontic failures is related leakage around the filling material, indicating this as one of the most common possible cause (Song et al. 2011).
The obturation technique is of paramount importance, since in vitro bacterial permeability of gutta-percha/sealer was observed (Libonati et al. 2018), evidencing the need for adequate seal ability of the filling materials. Because the clinically significant saliva penetration, once obturation is exposed to oral environment for even 20 days, bacteria (Torabinejad et al. 1990) or endotoxin ) may move through a totally obturated root canal. Microorganisms in the oral fluid may infiltrate through the dentin/temporary restorative sealer interface, highlighting the importance of restorative procedures, which may avoid marginal microleakage (Nakamura et al. 2006), leading authors to state that the integrity of coronal restoration is as important as the obturation quality (Ray & Trope 1995, Tronstad et al. 2000.
In view of root canal contamination or recontamination, which may jeopardize endodontic treatment, the success is achieved after definitive restoration. Thus, it's important to understand and evidence, in vivo, factors that may influence coronal infiltration, especially if different obturation techniques can prevent or decrease bacterial infiltration. Therefore, this study assessed in vivo the influence of different obturation techniques in coronal bacterial leakage.

Methodology
This research was approved before the beginning of the experiment by the institutional Ethics Committee on the Use of Animals at Araçatuba School of Dentistry (UNESP/FOA).

Animals
The sample size was calculated based on previous studies (Barbosa et al. 2003, Holland et al. 2007a, Holland et al. 2007b, Borlina et al. 2010, with additional elements per group, considering possible loss of specimens during experimental period or histologic processing. From each dog, maxillary incisors (single root) and both maxillary and mandibular premolars (two root canal each) were selected (Holland et al. 2007a, Borlina et al. 2010, totaling 45 root canals.
The animals were kept in a clean and ventilated vivarium inside the University and received vaccines and standard medicines before and during the experimental period. The animals received balanced food and water ad libitum, weekly inspected.
The teeth received prophylaxis with a rubber cup/pumice and rubber dam was placed, followed by antisepsis of the operative field with 0.3% polyvinylpyrrolidone-iodine solution. Coronal access was obtained with a #1090 cylindrical diamond bur (KG Sorensen, Barueri, Brazil) in refrigerated high speed rotation and pulp tissue was extirpated with a #15 K-file (Dentsply/Maillefer Instruments, Ballaigues, Switzerland). Instrumentation was performed with the crown-down technique (Barbosa et al. 2003, Holland et al. 2007a, Holland et al. 2007b, Borlina et al. 2010): After orifice opener N o 1, 2 and 3, cervical and middle biomechanical preparation was achieved with Gates-Glidden drills 3, 2, 1 and manual instrumentation up to a size 40 K-file (Dentsply/ Maillefer) at the limit of the cementodentinal junction (CDJ), determined by tactile sense and confirmed by radiographic images. Throughout the biomechanical preparation, roots were irrigated with 3 mL of saline, at each change of file (Holland et al. 1999, Holland et al. 2007b, Fonseca et al. 2020. After instrumentation, root canals were irrigated, aspirated and flooded with 17% EDTA solution for 3 minutes for smear layer removal. Then, canals were irrigated with saline and dried with sterile paper points. The root canals were obturated with a main gutta-percha cone calibrated to a #40 loaded with Endofill ® sealer and accessory guta-percha at the CDJ limit, forming the three experimental obturation groups (n=15), as follows: group 1: lateral condensation; group 2: lateral condensation with coronal plug of set Endofill ® ; group 3: Tagger hybrid technique. Obturation excess was removed with a heated plugger at coronal level, and the remaining filling material was vertically condensed. Then, after cleaning the pulp chamber, coronal openings were not sealed, leaving the obturated roots exposed to oral environment. Radiographs were taken to evaluate obturation homogeneity and quality.

Histologic analysis
After 90 days exposing obturated roots to oral environment, animals were euthanized by anesthetic overdose. The maxilla and mandibles containing the root-filled samples were removed, fixed in 10% neutral-buffered formalin for 72 hours and demineralized in 20% EDTA solution. After dehydration and diaphanization, segments were embedded in paraffin. Serial 6 µm thick longitudinal sections were obtained from each specimen and stained with Brown and Brenn technique for microscopic evaluation, which allows observation of Gram + (as Nocardia and Actinomyces filaments stains in blue) and Gram -(as nuclei stains in red) and additional tissue elements (yellow staining) (Brown & Brenn 1931, Holland et al 2007a, Borlina et al. 2010, Gomes-Filho et al. 2013).

Descriptive Analyses
As the Brown & Brenn technique was used to observe bacterial presence, a descriptive analysis (Koche 2011) was chosen to classify regarding the presence or absence of bacterial infiltration.
During laboratory process, one sample from Lateral Condensation Group and two samples from Tagger Hybrid group were discarded. No sample loss occurred in Lateral Condensation with a plug of set Endofill ® group, totaling 42 analyzed samples.
All specimens showed food residues and debris in the coronal pulp chamber and near the canal entrance (Figure 1).

Figure 1.
Dog's premolar with coronal opening exposed to the oral environment for 90 days, evidencing presence of food residues and debris in pulp chamber and canal entrance (200x; Brown & Brenn staining).

Lateral Condensation Group
The specimens analyzed in this group consisted of three incisors and six premolars totalizing 14 root canals.
Bacterial infiltration was observed in the interior of six canals until near the foramen (Table 1). Gram positive and negative bacteria were present in root canal walls, until the apical third, colonizing dentinal tubules ( Figure 2).

Lateral Condensation with a plug of set Endofill ® group
The analyzed specimens consisted of three incisors and six premolars totaling 15 root canals. Under microscope observation, two specimens showed Gram positive and negative bacteria infiltration in the dentinal wall (Table 1), near apical ramification, also invading dentinal tubules. Tooth apex showed thickened periodontal ligament and inflammatory cell infiltrate.
The presence of bacteria Gram positive and Gram negative was observed adhered to the external root dentinal wall with mild inflammatory infiltrate (Figure 3).
In one incisor, bacterial infiltration throughout the entire canal and invasion of dentinal tubules near apical area was also observed, as the presence of an inflammatory infiltrate within the periodontium.

Tagger Hybrid technique group
In this group three premolars and seven incisors were analyzed, totaling 13 root canals, in which eight samples showed no inflammation.
Two specimens showed elevated bacterial infiltration throughout the entire obturated canal (Table 1), observed inside dentinal tubules from the coronal area and extending to apical, which presented greater density of contaminated dentinal tubules, populated by Gram positive and negative bacteria. Inflammatory infiltrate could also be observed at the foramen area ( Figure 4). In one of the samples, even with extruded obturating material, microorganisms populated apical dentinal tubules, delta area and were also observed in the apical periodontal ligament, next to the extruded material. Another sample, with inflammation and thick periodontal ligament, Gram positive and negative bacteria were observed at the inner wall of the canal until apical ramification, but not inside dentinal tubules.

Discussion
Coronal infiltration by microorganisms through obturated root canal has been previously investigated, demonstrating in vitro the recontamination possibility (Oliveira et al. 2013). However, in vivo studies are necessary to represent what occurs in the oral cavity. Thus, the present study evaluated, in vivo, the influence of two different obturation techniques on bacterial infiltration and the addition of a protection plug, when exposed to oral environment without coronal sealing for 90 days.
According to Friedman et al. (1997), the dog model for in vivo researches was developed to overcome limitations of in vitro studies, since allows the evaluation of endodontic materials within a clinical/oral environment. Based on this fact, the dog model has been used for several years (Holland et al. 1971, Holland et al. 1999, Souza et al. 2006  As previously demonstrated by Khayat et al. (1993), coronal bacteria infiltration along with its metabolites can penetrate the entire extent of an obturated root canal in shorter periods than the experimental ninety days used in the present study. Infiltration was previously observed in vitro after four (Khayat et al. 1993), six , ten (Chailertvanitkul et al. 1997a), seventeen (Chailertvanitkul et al. 1997b) or nineteen days (Torabinejad et al. 1990).
It is also important to emphasize that endotoxin penetration is faster than bacterial penetration (Chow et al. 1993. However, a previous study demonstrated that obturation technique may hinder infiltration. Carratù et al. (2002) evaluated in vitro the lateral condensation obturation technique with a eugenol/zinc-oxide sealer and observed that, despite no endotoxin was present in samples, all specimens had bacterial infiltration after 13 days. However, the average time of coronal infiltration, usually evaluated in vitro, varies due to the different methodologies and materials, as opposed to in vivo studies, which may be influenced by the animal's diet, dental anatomy, materials and technique, among other variables. Regarding the postoperative period in the present study, after ninety days, initial inflammation after endodontic treatment due to mechanical trauma and initial reaction to the sealer would have been overcome, converting the initial acute inflammation into chronic. However, Zmener et al. (2010) demonstrated in subcutaneous tissue of rats that even after 90 days, a zinc oxide and eugenol sealer may induce a severe inflammation.
Although sodium hypochlorite (NaOCl) is the most commonly used irrigating solution for root canal therapy (Slaus & Bottenberg 2002, Clarkson et al. 2003, Nascimento et al. 2021, the use of an irrigant with antibacterial properties could interfere with residual action for undetermined time on microorganisms that would penetrate the area subjected to analysis. For this reason, the root canals were irrigated with saline (Fonseca et al. 2020), corroborating a previous methodology (Holland et al. 2000). The EDTA solution was used in static protocol, prior to obturation in order to chelate calcium, opening dentinal tubules to enhance sealer penetration (Bueno et al. 2019).
The endodontic sealer used for obturation in the present study was a zinc oxide-eugenol-based sealer. Since the research aimed to assess the presence of bacteria inside dentinal tubules after different obturation techniques, sealers with low antibacterial effect would be more appropriate, although after 90 days, no antimicrobial activity would be expected. According to Savioli et al et al. (2006), eugenol shows antimicrobial activity, and when incorporated to an endodontic sealer, shows inhibition bacterial zones in in vitro study (Monajemzadeh et al. 2017, Dragland et al. 2019). However, Wang et al. (2014) showed, via confocal laser scanning microscopy, a weaker antibacterial effect of ZOE sealer than AH Plus, corroborating posterior results found by Singh et al. (2016) and Dalmia et al. (2018).
With respect to the results for the different obturation techniques, all groups showed bacterial infiltration. The lateral condensation group was chosen since is a commonly used technique (Leduc & Fishelberg 2003), with elevated sealer penetration (Macedo et al. 2017), mainly due to the pressure promoted by the gutta-percha cones against dentin walls, increasing interaction between sealer and root dentin (Hatton et al. 1988). Nevertheless, our results showed the highest bacterial infiltration in the Lateral Condensation group. The group with a plug of set Endofill sealer and Tagger's hybrid technique group showed reduced infiltration, but not prevented. This is in accordance with previous studies, evidencing that the use of a protection plug over the obturation may aid the healing, delaying the bacterial infiltration (Holland et al. 2007a, Yamauchi et al. 2006, while Tagger hybrid's technique provides better homogeneity and apical sealing than lateral condensation (Fracassi et al. 2013, Saberi et al. 2017).
Based on the results, we hypothesized that a possible leakage in the obturation-dentin interface may occur with lateral condensation group, which is reduced in the Tagger hybrid technique, since obturation becomes a more homogenous single mass, corroborating an ex vivo study in which authors showed that single cone technique (with tapered cone) and hybrid technique are more effective in prevention of coronal leakage than lateral condensation technique (Saberi et al. 2017). In addition, in vitro studies with thermoplasticized/warm/flow gutta percha techniques also showed bacterial infiltration (Libonati et al. 2018, Lone et al. 2018, Prithviraj et al. 2020.
Regarding the staining technique, while the Brown and Brenn staining may present limitations in the identification of microorganisms due to the decalcification process (Wijnbergen & Van Mullem 1987), this technique has been previously used to assess the presence of microorganism (Holland et al 2007a, Borlina et al. 2010, leading our results to suggest that most of the bacteria contained in the specimens of this research may not have been detected or even may be eliminated during laboratory processing. A systematic review conducted by Gillen et al. (2011) showed that healing of apical periodontitis increase with both adequate root canal therapy and adequate restorative treatment. Therefore, it's preferable to perform a permanent restoration immediately after root canal treatment than using a provisional restoration (Lynch et al. 2004), which may depend even of a correct thickness to decrease bacterial infiltration (Shanmugam et al. 2020), while waiting for the resolution of apical periodontitis. It's also important to emphasize that even root canals with an intracanal post but without a prosthetic crown can be recontaminated when exposed to human saliva in a short period of time and the exposure of endodontically treated teeth to the oral environment may indicate endodontic retreatment (Tronstad et al. 2000, Oliveira et al. 2013, Mohajerfar et al. 2019).

Conclusion
Coronal bacterial infiltration was evidenced in root canals obturated regardless the technique when exposed to the oral environment for 90 days. When present, bacteria inside the canals and apical dentinal tubules may negatively interfere with the apical repair process. The endodontic treatment should only be considered concluded when the tooth is completely restored.
Further in vivo studies are encouraged to provide a better comprehension of bacterial infiltration via coronal leakage in oral environment with different obturation techniques such as single cone and thermoplasticized techniques/continuous wave of obturation, besides the use of different sealers, like resinous or bioceramic endodontic sealers.