Many experimental models and approaches have been employed to date to evaluate the efficacy of root canal irrigation. Classically, radiopaque irrigants in vivo or dye solutions used in transparent root canal models in vitro have been utilized to monitor the penetration of these solutions [35–42]. Artificially placed dentine debris using a split tooth is a simple method of determining the influence of irrigation and irrigant activation techniques by scoring the remaining debris [43, 44]. Organic tissues have also been used to evaluate chemical debridement and the efficacy of irrigant activation, and it has been revealed that ultrasonic activation enhances chemical debridement in simulated curved canals and accessory canals [32, 45]. Computational fluid dynamics (CFD) has also provided a further understanding of fluid flow mechanisms [10, 46–49]. CFD studies have provided measurements of velocity magnitude, velocity vectors, and wall shear stresses with various needle designs and positioning. The effects of various irrigating solutions against endodontic biofilm have been assessed in previous reports, particularly from a chemical aspect, and optimal irrigant concentrations and temperatures have been described [26, 50–53]. Notably however, no in vivo study models had yet been developed to compare the efficacy of different irrigation protocols for clinical biofilm removal .
Pigs have been adopted as an experimental model in many biomedical fields due to some clear similarities with the human anatomy, and due to the obvious ethical considerations with regard to human subjects. Alveolar bone mineral contents, and the inflammation and destruction processes in periodontal tissues, are among the notable biological similarities between pigs and humans [30, 55]. In our current experimental pig model, we could successfully observe bone defects at the periapical area after exposing the root canal system to the oral environment. These defects developed as a consequence of inflammation, confirmed by an increased CRP level at 2 weeks after root canal exposure. As found in previous studies, intraradicular biofilms can arise through the opening of an access cavity for 2 weeks to enable contamination, and subsequent sealing for 4 weeks to produce an anaerobic environment [56, 57]. We observed typical biofilm thickness in the entire root canals in our pig model by SEM imagery in the control tooth. Importantly, we confirmed in our current experimental pig model that the most abundant and prevalent phyla within the intracanal biofilms were Firmicutes, Bacteroidetes and Fusobacteria, which predominate also in human samples . Although, we did not use controls for bacterial 16S rRNA gene analysis, the results are comparable to the previously reported human data based on the robust experimental protocols for the gene analysis [33, 59, 60]. In our current study in the pig, we utilized the lower deciduous mandibular second premolars because this tooth length is similar to that in humans. Although the apical size of approximately 0.7-1.0 mm in diameter is wider, and the root dentin thickness is thinner, in the pig than in human permanent teeth, the same armamentarium used for root canal treatments in human clinical practice can be readily applied also in a pig model. The intraradicular biofilm pig model is therefore far more reflective of human conditions than those created using rodents or rabbits.
We focused in our present study on the chemical reduction of biofilm using NaOCl  and agitating irrigation techniques. Hence, we did not utilize mechanical instrumentation nor EDTA irrigation. Mechanical instrumentation is absolutely essential for the mechanical debridement of biofilm and to reduce the bacterial count from the root canal. Mechanical debridement may be sufficient in an experimental system for reducing biofilm if the tooth has a straight and wide i.e. although mechanical instrumentation is essential, it has an inherent limitation for the complete shaping of the root canal system. To eradicate biofilm from these unreached areas, root canal irrigation, in which the irrigant is agitated using a physical reaction, is likely to be needed. A notable limitation of our current study however was that the tooth did not represent a curved canal. Future studies should consider comparing the efficacy of different techniques for the in vivo removal of an intraradicular biofilm from a curved root canal.
We used our pig model system to test the effectiveness of various established human irrigation protocols in removing biofilm from the root canal system. In terms of bacterial quantification however, it must be pointed out that the actual oral hygiene of pig is a poor. Thus, although calculous removal and tooth cleaning were performed in our pigs before extraction to reduce bacterial contamination, the sound tooth was served as a control for quantification analysis. Hence, although contamination by bacteria may occur during tooth extraction in a pig model system, our results showed that all of the tooth samples with induced biofilm formation had a significantly higher number of bacteria than the sound tooth. In accordance with previous reports, the CNI method was found in our current analysis to be insufficient to clean the root canals due to its delivery limitations [62, 63]. Our findings indicated in fact that almost no biofilm was removed by CNI. A large number of prior PUI studies have reported positive results in the removal of intracanal hard tissue debris and pulp tissue remnants due to the acoustic streaming generated by oscillating movements [12, 64, 65]. However, PUI was further found to be less effective than chemo-mechanical preparation in a large canal , indicating that it is limited in terms of intraradicular biofilm removal from a wide root canal. Our current results in the pig model were consistent with this as we found no significant differences between the efficacy of CNI and PUI.
The subsonic energy in the EA method has been found to generate a higher back-and-forth tip movement amplitude. The effectiveness of EA in cleaning an infected root canal and in smear layer removal is reported to be inferior or equal to that of PUI [66–68]. The main difference between EA and PUI is whether the tip of the device directly contacts the root canal surface or not. The range of the vibrating polymer tip of the EA is much wider than the range of motion of a PUI tip, and this increases the area where the tip makes physical contact with the root canal surface. Hence, our current results with EA in the pig model suggest that the generation of a mechanical force against the root canal wall is essential for eliminating firmly attached biofilms.
LAI produces a physical reaction in the irrigant solution by transient cavitation through the optical breakdown caused by the strong absorption of the laser energy, which is expected to remove biofilm . Our current findings in the pig model indicate that LAI is more effective at removing intraradicular biofilms than CNI or PUI, which is consistent with previous in vitro studies [21, 22]. Hence, the adjunctive mechanical reaction is attributable to the highly turbulent action of the irritant, and the photo-initiated energy is effective in collapsing the intraradicular biofilms. These results also suggested the further potential of LAI, which generates a mechanical reaction without contacting the root canal wall and can have effects beyond the root canal curvature or irregular areas such as isthmus, fin, and accessory canals. Further studies are needed to investigate whether such mechanical turbulence in irrigants generated by LAI can penetrate sufficiently to produce the shear stress required to remove biofilms.