Investigation of Er:YAG laser-activated irrigation for the removal of biolm and observations of laser-induced cavitation behavior


 BackgroundWe investigated the biofilm removal effects of LAI using a pig model, focusing on the impact of the fiber tip position, and used a high-speed camera to observe the occurrence and positioning of the cavitation associated with laser irradiation.MethodsA total of 16 roots of deciduous mandibular second premolars from 4 pigs were used. After a pulpectomy, the canals were left open for two weeks and sealed for 4 weeks to induce intraradicular biofilm. Then, root canal irrigation was performed with Er:YAG laser activation. The fiber tip was inserted at two different positions, i.e., into the root canal in the intracanal LAI group and into the pulp chamber in the coronal LAI group. Intracanal needle irrigation with saline or 5% NaOCl was utilized in the positive control and CNI groups. SEM and qPCR were carried out to evaluate treatment efficacy. For qPCR, ANOVA and a Tukey-Kramer post hoc test were performed with α=0.05. A high-speed camera was used to observe the generation of cavitation bubbles and the movement of the induced bubbles after laser irradiation.ResultsThe intracanal and coronal LAI groups showed significantly lower amounts of bacteria than either the positive control or CNI groups. There was no significant difference found between the intracanal and coronal LAI groups. SEM images revealed opened dentinal tubules with the destruction of biofilm in both LAI groups. High-speed camera images demonstrated cavitation bubble production inside the root canal after a single pulse irradiation pulse. The generated bubbles moved throughout the entire internal multi-rooted tooth space.ConclusionsCoronal LAI can generate cavitation in the root canal with a simply placed fiber inside the pulp chamber, leading to effective biofilm removal. This method could thus contribute to the future development of endodontic treatments for refractory apical periodontitis caused by intraradicular biofilm.


Abstract Background
We investigated the bio lm removal effects of LAI using a pig model, focusing on the impact of the ber tip position, and used a high-speed camera to observe the occurrence and positioning of the cavitation associated with laser irradiation.

Methods
A total of 16 roots of deciduous mandibular second premolars from 4 pigs were used. After a pulpectomy, the canals were left open for two weeks and sealed for 4 weeks to induce intraradicular bio lm. Then, root canal irrigation was performed with Er:YAG laser activation. The ber tip was inserted at two different positions, i.e., into the root canal in the intracanal LAI group and into the pulp chamber in the coronal LAI group. Intracanal needle irrigation with saline or 5% NaOCl was utilized in the positive control and CNI groups. SEM and qPCR were carried out to evaluate treatment e cacy. For qPCR, ANOVA and a Tukey-Kramer post hoc test were performed with α=0.05. A high-speed camera was used to observe the generation of cavitation bubbles and the movement of the induced bubbles after laser irradiation.

Results
The intracanal and coronal LAI groups showed signi cantly lower amounts of bacteria than either the positive control or CNI groups. There was no signi cant difference found between the intracanal and coronal LAI groups. SEM images revealed opened dentinal tubules with the destruction of bio lm in both LAI groups. High-speed camera images demonstrated cavitation bubble production inside the root canal after a single pulse irradiation pulse. The generated bubbles moved throughout the entire internal multirooted tooth space.

Conclusions
Coronal LAI can generate cavitation in the root canal with a simply placed ber inside the pulp chamber, leading to effective bio lm removal. This method could thus contribute to the future development of endodontic treatments for refractory apical periodontitis caused by intraradicular bio lm.

Background
Apical periodontitis refers to purulent in ammation inside the alveolar bone triggered by a bacterial infection of the root canal system, which is the internal structure of the tooth [1]. Treatments require the removal of causative bacteria and their by-products from the root canal system and suppression of the in ammatory processes. Mechanical and/or chemical bacterial removal have been utilized as standard approaches in these cases under an aseptic environment, and the success rates for these initial root canal treatments have been reported to exceed 90% [2]. Retreatments required in cases of refractory in ammation, but the success rates will typically be lower [3] because of bacterial ingrowth into anatomical complexities such as the lateral canal and isthmus and the subsequent formation of bio lm [4] . The complete elimination of bio lm is almost impossible in this circumstance.
Root canal irrigation is a treatment technique that aims to chemically reduce bacterial loads inside the root canal system, such as in areas that cannot be reached by a mechanical root canal preparation. The standard irrigation method utilizes a syringe and ne needle (i.e., 27-31G) to deliver and re ux the disinfectant solution. Recently, irrigant agitation techniques using ultrasound, sonic vibrations, and lasers have been shown to be more effective in reducing intracanal debris than the syringe technique [5][6][7][8].
Notably, removing bacterial bio lm from the root canal surface has remained an issue with these interventions because chemical actions alone are insu cient to remove bio lm and some form of physical manipulation is also needed. Therefore, although the chemical exposure from a root canal irrigation will be effective against planktonic bacteria, irrigation technologies are needed that can eliminate bio lm through the generation of physical force at an adjacent site that is inaccessible to existing mechanical instruments. Among the various types of root canal irrigation techniques that are currently available, photodynamic actions are a possible approach to generating a physical effect during root canal irrigation by means of hydrodynamic force generation.
Photodynamic actions produced by multiple lasers at different wavelengths have been shown to effectively agitate root canal irrigants. An Er:YAG laser emitting at the 2.94 µm wavelength close to the absorption peak of water [9], [10] has been utilized for Laser-activated irrigation (LAI), which effectively removes bacteria from the root canal system [11]. The LAI cleaning mechanism depends on rapid uid motion [10,12,13] in the root canal, which generates subsequent pressure waves through the expansion and collapse of vapor bubbles at the site of the laser irradiation. In addition, the generation of numerous secondary cavitation bubbles can be observed under a high-speed camera after the vapor bubble collapse [14]. Cavitation is a liquid to gas phase transformation phenomenon caused by a decreased pressure due to an increased uid velocity. A collapse of the cavitation bubbles occurs after this and produces large-amplitude shock waves. Hence, LAI is expected to have properties that not only increase the ow velocity of the irrigant but also generate physical forces upon cavitation collapse on the root canal wall that could be effective for bio lm removal. Since laser agitation of the root canal irrigant can spread throughout the root canal away from the ber tip, placement of the ber tip into the pulp chamber without coming close to the apex or into the root canal has been advocated called photon-induced photoacoustic streaming (PIPS) [15,16].
In the present study, we investigated the bio lm removal effects of LAI using a pig model system, with a particular focus on the impact of the ber tip position. We also employed a high-speed camera to carefully observe the occurrence and positioning of the cavitation associated with the laser irradiation.

Methods
The current study protocols for animal use were reviewed and approved by the Animal Care and Use Committees of Tohoku University (Permit No. 2018 SHIDO-045). All animal experiments and maintenance were conducted in accordance with the Regulations for Animal Experiments and Related Activities at Tohoku University to minimize suffering.
Sixteen roots from the deciduous mandibular second premolars of four pigs (nine-week-old large white X Landrace breed cross; Japan SLC Inc., Shizuoka, Japan). The light was turned on at 8.00 am and turned off at 6.00 pm each day in our animal facility. The pigs had free access to water at any time and were fed a regular diet (Grandeal B; Zennoh Feed Mills of the Tohoku District, Miyagi, Japan) 3 times per day. All interventions were performed after sedation with medetomidine (0.1mg/kg, IM) and midazolam (0.2mg/kg, IM) followed by inhaled sevo urane (2-5%), with local injections of 2% lidocaine (1.8 ml, SC) also given to minimize pain. The experimental protocol is presented in Figure 1. All procedures were performed under surgical loupes with LED light (EyeMag PRO; Carl Zeiss, Jena, Germany).
The occlusal surface was attened using a straight bur and electric engine (Ti-Max X95; NSK, Tochigi, Japan) to prevent tooth fracture and for the ease of working length determination. Following access cavity preparation and obtaining a straight-line access, chemo-mechanical removal of pulp tissue was performed with 5% Sodium Hypochlorite (NaOCl) and K les. The canals were exposed to the oral environment to inoculate the root canal with oral bacteria for two weeks. This was followed by sealing with hydraulic temporary lling material (Lumicon; Heraeus Kulzer, South Bend, IN, USA) and composite resin (MI Flow II; GC, Tokyo, Japan) with the use of adhesive (G-Premio BOND, GC) to create an anaerobic intracanal environment for four weeks to allow the bacterial bio lm to mature. After removal of the temporary lling materials and placement of the dental dam isolation at six weeks after pulp removal, aseptic conditions were established by cleaning the tooth surface with 5% NaOCl and saline. The teeth were then randomly assigned to the different experimental irrigation groups.
The following irrigation protocols were used for the different experimental groups.
CNI group: conventional needle irrigation (CNI) was performed using side port 30G needles with 5% NaOCl. Canals were irrigated for 30 seconds with 5mL of irrigant and left for 30 seconds. This cycle was repeated ve times for a total of ve minutes of irrigation.
Intracanal LAI (I-LAI) group: irrigants were activated using an Er:YAG laser (Erwin AdvErl EVO; Morita, Kyoto, Japan) using a cone-shaped 300 µm diameter tip (R300T). Canals were irrigated for 30 seconds with 5 ml of the solution using a side port 30G needle and the irrigants were activated for LAI with the laser settings at 30mJ and 20pps without air and water supply. This cycle was repeated ve times for a total of ve minutes of irrigation and subsequent agitation. During laser irradiation, the tip was inserted 2 mm short of the working length and was moved slowly up and down a 3 mm length to the coronal side as with the conventional LAI irrigation.
Coronal LAI (C-LAI) group: this irrigation protocol had the same conditions as the I-LAI group except for the laser ber tip position, which was inside the pulp chamber and kept stationary. During laser irradiation, the solution in the pulp chamber keeping the volume, so that a light-cured resin built up around the crown (Dentto-Dam, MEDICLUS, Korea), and NaOCl was then added with a syringe.
Positive control group: the teeth irrigated using saline were used as the positive control group. After root canal irrigation, each canal was rinsed with saline for 30 seconds. The teeth were then extracted, and the remaining bacteria in each root were evaluated using real-time PCR (n=2 each) and SEM (n=2 each).
Quanti cations of the bacteria present in the root canals were performed based on previously described methods. Brie y, the sample roots were immersed in liquid nitrogen immediately after tooth extraction and the tooth crown was resected. For the bacterial quanti cation, the intact deciduous mandibular second premolar was obtained from a pig used for other purposes as the sound tooth. Two roots from a sound tooth were preserved intact without receiving any root canal procedure. The roots were then crushed with SK mill (Tokken, Chiba, Japan) to acquire powdered samples. Total DNA was extracted from each powdered root sample using a Cica Geneus DNA extraction Kit (KANTO chemical co.; Tokyo, Japan) in accordance with the manufacturer's instructions. The presence of bacteria was veri ed in the experimental samples by qPCR using the bacterial primers 357F and 908R22. These assays were performed using a real-time PCR apparatus (CFX Connect; Bio-Rad Laboratories, Hercules, CA, USA).
Ampli cations were conducted for 40 cycles at 95°C for 15 seconds followed by 65°C for 1 minute, with the uorescence signals measured at the end of each cycle. A standard curve was generated by subjecting 10-fold dilutions of a known concentration of E. faecalis DNA to the same qPCR protocol. The bacterial counts in all experimental groups were calculated using threshold cycle (Ct) values plotted against the standard curve. Statistical analysis was performed using ANOVA followed by a Tukey Kramer post-hoc test with an α value of 0.05 (BellCurve for Excel; Social Survey Research Information Co., Ltd. Tokyo, Japan) to detect signi cant differences in the bacterial populations. SEM sample preparation was conducted according to previously described methods [17,18]. Brie y, after resecting the tooth crown, the mesial and distal roots were separated with a diamond disc. The distal root was then split into halves and immersed in 2.5% glutaraldehyde to x the root canal bio lm, and then rinsed with PBS and treated with 1-ethyl-3-methyl-imidazolumetra uoroborate. The samples were dried in a vacuum desiccator for one day and sputter-coated with platinum. The surfaces of the root canal wall and intraradicular bio lm were observed using SEM (VE-8800; Keyence Inc., Osaka, Japan) at a 10kV acceleration.
To observe the generation and movement of bubbles inside the root canal through the LAI, images were taken using a high-speed digital camera (FASTCAM Mini AX200; Photron, Tokyo, Japan) attached to a macro lens (SP 90mm F/2.8 Di,; Tamron, Saitama, Japan). The LED light source was placed diagonally in front of the root canal model. The Er: YAG laser was equipped with R300T ber tip with an output setting of 30 mJ at 20PPS and no water or air supply.
Two types of root canal models were used i.e., a single or two canal model. A single root canal model made of epoxy resin with apical size #80, 02taper, and length 15 mm was used to observe the characteristics of the bubbles generated with a laser pulse. The ber tip was placed 12mm from the apex, and the root canal was lled with water. The frame rate, image size, and recording duration were set at 40,000 frames per second (fps), 384×256 pixels, and 217ms, respectively. For the two root canal model, an arti cial maxillary premolar model (TrueTooth # 5-002, Dental Engineering Laboratories, CA, USA) was used to observe the induction of bubbles and changes in their behavior over time. The tooth was prepared with an access opening, straight-line access, and root canal preparation with a size #50 Reciproc Blue (VDW, Munich, Germany). The ber tip was placed at the center of the pulp chamber as assuming C-LAI. During laser irradiation, the irrigant was continuously supplied with a syringe. The frame rate, image sizes, and recording duration were set at 750 fps, 1024×1024 pixels, and 29s, respectively. Figure 2 shows the qPCR results representing the remaining bacterial amount in the root canal after experimental root canal irrigation. The I-LAI group (6.11 × 10 ) and the C-LAI group (5.02 × 10 ) showed a signi cantly lower level of bacteria than the positive control group (8.08 × 10 ) or the CNI group (7.62 × 10 ). There was no signi cant difference between the I-and C-LAI groups in this regard. Compared to the sound teeth (4.95 × 10 ), the bacterial counts in the positive control group and the CNI group were signi cantly higher, while those in the both LAI groups did not differ signi cantly.

Results
SEM images of the root canal wall taken after root canal irrigation are shown in Figure 3. A multi-layered bio lm that covered the entire root canal was observed in the positive control group. Some dentinal tubule structures could be seen in the CNI and I-LAI groups due to bio lm destruction although residual bio lm was observed. Only a small amount of bio lm remained in the C-LAI group, and dentin tubule structures could be seen on the entire root canal wall of these teeth.
Bubble generation and collapse after a single pulse of irradiation inside a single root canal model are shown in Figure 4. An additional movie le shows in more detail (see Additional le 1). At the tip, the bubbles generated at t=0 ms most notably developed in the crown direction, shrank, and then nally disappeared at t=0.525 ms. Additionally, several cavitation bubbles were observed from t=0.625 ms to t=0.825 ms at the central part of the canal appearing at around 3 to 8 mm from the ber tip. Subsequently, cavitation bubbles were repeatedly observed in the same range indicated above from 0.875 ms to 1.15 ms. The bubble at the ber tip was found to have shrunk the most at t = 0.5 ms, with a bubble appearing at a distance of about 5 mm approximately 0.05 ms later. We surmised from this that bubbles are generated in the tubules by the pressure wave generated during their shrinking and re-expansion at the tip end, in addition to the associated depressurization. In this case, the velocity of the pressure wave can be calculated to be about 100 m/s. The movement behavior of the bubbles over 30 sec is depicted in Figure 5 and additional movie le (see Additional le 2). After laser irradiation, the bubbles generated in the pulp chamber reached the root canal ori ce in 1.03s and the root apex in 1.75 sec in the left root canal. In the right root canal, the bubbles reached the root canal ori ce in 1.15 sec and the root apex in 7.64 sec. The number and size of the bubbles in the root canal increased with time. The generated bubbles and their movement could be observed across entire internal tooth space. A slight irrigant extrusion from the root apex was observed at t=24 sec.

Discussion
Removing bio lm from inside a root canal, which is the cause of refractory apical periodontitis, is still challenging when conducting a root canal retreatment [4,19]. The most reliable bio lm removal method is mechanical disruption, although no instrumentation technique to achieve this can reach all of the root canal surfaces [20]. In addition, bio lms are observed in anatomical complexities such as the isthmus and lateral canal [21]. Hence, non-contact bio lm removal techniques continue to need a more reliable retreatment protocol. We have previously reported an in vivo intraradicular bio lm model in the pig [17], which can closely replicate human bio lms in terms of morphology and microbiota, to evaluate the effectiveness of bio lm removal protocols and found that LAI is an effective intervention in this regard 17 . In the present study, we further found that the cavitation bubbles produced by C-LAI were effective in bio lm removal using in vivo intraradicular bio lm model, which indicated that LAI has the potential to apply the non-contact removal technique.
The Er:YAG laser is immediately absorbed by water upon irradiation and generates bubbles that create pressure waves which will agitate the cleaning solution via the high-velocity irrigant ow [8,13,14,22].
LAI was signi cantly more effective in cleaning root canals, especially in removing root canal debris and intracanal bacteria, than CNI or ultrasonic agitation [11]. Moreover, the vapor bubble-and cavitation bubble-induced turbulence associated with LAI can cause shearing stress on the root canal wall surface [11,23]. In our current study, LAI showed the capacity to remove bio lm along with excellent cleaning effects, indicating that it can generate physical forces inside root canal systems.
The ber tip used in our current investigations has a conical tip and is designed to distribute 80% of the irradiated laser energy laterally and 20% axially. When such a conical-shaped tip is inserted into the root canal system, regarded as a I-LAI approach, the space between the ber tip and the root canal wall is insu cient to induce irrigant ow. Most of the irradiated energy in this case goes directly to the root canal wall [14,24]. On the other hand, the C-LAI or PIPS methods reserve a su cient amount of space because the ber tip is placed into the pulp chamber so that the energy is absorbed by the irrigant and can therefore produce a high irrigant velocity [25]. In our present study, we found that a C-LAI can simultaneously observe the generation of bubbles and disperse them throughout the whole root canal system. Also, bacterial reduction of C-LAI is comparative to the I-LAI, meaning that C-LAI was possible to achieve a su cient ow velocity to agitate the solution into the entire root canal system. The results suggested that C-LAI is both a safer and simpler method of root canal cleaning that can treat multiple root canals simultaneously with less impact of the laser irradiation on the root apex due to the insertion of the ber into the root canal.
The observation of cavitation bubbles after a single pulse of irradiation revealed bubble generation at a distance from the ber tip. This phenomenon is not simply due to the movement of bubbles but is considered a cavitation occurrence caused by a decrease in the pressure inside the root canal. Although the velocity of the pressure wave, calculated at about 100 m/s, was smaller than the velocity in the cavitation ow of several hundred m/s, it still seemed capable of generating cavitation because of the pressure wave limits arising from the friction forces in a narrow spatial structure and gas-liquid mixed phase. [26]. This phenomenon of cavitation bubbles and physical reaction occurrences away from the ber tip accords with the previously reported behavior of laser-induced cavitation in the liquid phase [27]. Hence, it can be considered that during bio lm removal, C-LAI not only agitates the irrigants but also mechanically detaches the bio lm from the root canal wall by generating shock waves due to cavitation bubbles.
The pulse width of the C-LAI used in our present study was about 300 μsec, which is larger than the 50 μsec pulse width used in PIPS. The Er:YAG laser is capable of tooth structure ablation with irradiation energy increases, thus PIPS is designed to have a narrow pulse width to optimize the agitation of the irrigant without tooth ablation [15]. However, it was found previously that PIPS cannot easily generate a shock wave inside the root canal system [28]. Hence, with the long pulse width used in this study, we assumed that the increased irradiation energy contributed to the generation of a su cient pressure wave for cavitation bubbles to occur. Thus, cavitation by laser irradiation can generate a physical force on the root canal wall and is expected to become a novel treatment technique for non-contact bio lm removal in the retreatment of refractory periapical periodontitis cases. In addition, C-LAI can be clinically applied as an effective, safe, and relatively simple root canal treatment technique because it can shorten the treatment time by cleaning multiple root canals simultaneously by positioning the ber tip in the pulp chamber.
Establishing more effective root canal cleaning methods will improve the success rates of root canal retreatment. Although additional investigations are needed to further optimize laser irradiation methodologies, and verify the safety of these approaches, our current evidence indicates that C-LAI will more safely improve irrigant activation for better bio lm removal without the need for other complicated techniques.

Conclusions
A non-contact debridement modality is required to establish a reliable bio lm removal technique for orthograde root canal retreatments. Our current report is the rst to indicate that C-LAI can generate secondary cavitation in the root canal, which can lead to the removal of bio lm through the simple placement of an optic ber inside the pulp chamber. C-LAI method is thus likely to contribute substantially to the future development of endodontic treatments for patients who suffer from refractory apical periodontitis caused by an intraradicular bio lm.

List Of Abbreviations
LAI: laser-activated irrigation; PIPS: photon-induced photoacoustic streaming; NaOCl: sodium hypochlorite; CNI: conventional needle irrigation; PPS: pulse per second; FPS: frames per second Declarations Ethical approval and consent to participate This study was reviewed and approved by the Animal Care and Use Committees of Tohoku University Graduate school of Dentistry (Permit No. 2018 SHIDO-045). All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. Consent to participate was not applicable.

Consent for publication
Not applicable.

Availability of data and materials
All the datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare no competing interests in relation to this study.  Experimental outline for the current study in the pig model. After the removal of the pulp tissue, the canal was exposed to the oral environment for 2 weeks and to an anaerobic intracanal environment for four weeks to allow for bacterial bio lm maturation. Six weeks after pulp removal, the pig teeth were irrigated by one of four different test protocols. Qualitative evaluations by SEM and quantitative assessment by qPCR were performed (a). CNI, conventional needle irrigation; LAI, laser activated irrigation.

Figure 2
Quantitative evaluation of the number of bacteria in the root canal by qPCR The I-LAI and C-LAI groups had a signi cantly lower bacterial concentration than either the positive control or CNI groups. There was no signi cant difference found between the I-LAI and C-LAI groups. The number of bacteria in the positive control and CNI groups was signi cantly higher than that in the sound teeth. However, the bacterial levels in the both LAI groups were not signi cantly different from that in the sound teeth. Typical images (a-1, b-1, c-1, and d-1) and SEM images (30x: a, b, c, and d-2; 1000x: a, b, c, and d-3; 3000x: a, b, c, and d-4) from each study group. In the high magni cation image, a residual layer of bio lm was observed to cover the root canal wall in the positive control group, and the structure of the root canal wall could not be con rmed (a-4). In the CNI group and I-LAI group, some openings of the dentinal tubules were observed compared with the positive control group, but some bio lm remained (b-4, c-4). In the C-LAI group however, almost no bio lm remained, and opening of both the dentinal tubules and the reticulated intertubular dentin structure could be observed (d-4).  pixels, and 217ms. At the tip, the bubbles generated at t=0ms notably developed in the crown direction, shrank, and nally disappeared at t=0.525ms (i). Additionally, several cavitation bubbles could be observed from t=0.625ms to t=0.825 at the central part of the canal (ii). The cavitation bubble was observed to generate and disappear again within the same time range from 0.875 ms to 1.15 ms (iii).