Dissolution versus dislodgment of dentin debris by pulsed erbium-doped yttrium aluminum garnet laser-activated endodontic irrigants Authors: Adrian Zollinger, Shengjile Deari, Maarten Meire, Lieven Robberecht, Matthias Zehnder Published: 2026 ABSTRACT AbstractThe aim of this study was to weigh dissolving versus physical effects of laser-activated irrigation (LAI) when cleaning simulated human root canals. Dentin debris was selected as the composite organic/inorganic substrate. The cleaning effects of plain 3% sodium hypochlorite (NaOCl, proteolytic), plain 9% 1-hydroxyethylidene-1,1-diphosphonic acid (HEDP, decalcifying), or combined 3% NaOCl + 9% HEDP (NaOCl+HEDP, combined effect) were compared to those of deionized water (no dissolving effect on substrate). Properties of test and control solutions were assessed, including their absorption of infrared light and their viscosity. The isthmus of transparent root canal models (n = 22) was filled with dentin debris obtained from bovine tooth roots. Activated irrigation was done on the models in random order in 3 cycles of 30 s each, at 612 mW/18 Hz. The area of removed debris was measured using digital microscopy. The influence of the irrigating solution on debris removal was assessed by one-way ANOVA and Tukey’s HSD, alpha = 0.05. Infrared absorbance spectra were similar between solutions. Density and viscosity increased cumulatively with the solutes under investigation. The combined NaOCl+HEDP solution displayed a significantly (p = 0.002) better cleaning effect than the plain counterparts, which did not remove more debris from the isthmus than water (p>0.05). Under current conditions, the known dissolving effects by NaOCl and HEDP per se did not affect dentin debris removal by LAI, which was mainly due to physical dislodgment of the debris. The enhanced LAI-endorsed cleaning by the combined NaOCl+HEDP irrigant could be related to altered fluid properties and should be explored in further experiments.Keywords: SWEEPS, HEDP, Continuous chelation, Sodium hypochlorite, Endodontics Introduction Laser-activated root canal irrigation (LAI) is a method to accelerate and potentially improve root canal debridement in conjunction with mechanical root canal preparation [1]. LAI relies on a pulsed laser beam with extremely high absorption in water, thus creating cavitation, resulting in streaming of endodontic irrigating solutions in remote areas of the root canal system [2]. Laboratory studies related to LAI phenomena were done mostly with plain water to exclude any chemical effects and focus on physical phenomena that remove the target substrate, i.e. shock waves, cavitation, and acoustic streaming [3–5]. However, in clinics, chemically active solutions are used to disinfect and debride the root canal. The main chemical in aqueous solution used for that purpose is sodium hypochlorite (NaOCl) [6]. NaOCl has unique proteolytic properties [7], and is therefore used to rid the root canal system of soft tissue remnants [8]. In addition to NaOCl, a decalcifying solution is recommended to dissolve the inorganic entities created by root canal instrumentation, i.e. smear layer and dentin debris. The classic solution for that purpose is 17% ethylenediaminetetraacetic acid (EDTA), which is used alternatingly with the NaOCl irrigant [9]. More recently, hydroxyethylidene diphosphonic acid (HEDP) has been introduced to the dental market in form of an etidronate powder that is admixed directly to the NaOCl solution, thus creating a combined NaOCl+HEDP irrigant with a dual biolytic and decalcifying effect [10]. This combined solution is supposed to be used during chemomechanical preparation, but also in the subsequent cleaning phase by e.g. ultrasonic activation of this irrigant [11]. Next to dissolving effects on target substrates such as biofilms, dentin debris, and necrotic pulp tissue remnants, chemicals contained in endodontic irrigants could have other, as-of-yet unexplored effects in conjunction with LAI. A recent study has shown that the addition of HEDP to a NaOCl solution affects fluid properties, and therefore the shape and expansion speed of primary cavitation bubble in LAI [5]. This phenomenon was explained by a dose-dependent increase in the viscosity of the resulting solution when HEDP is added. However, whether this apparently altered energy propagation in the solution has any effects in conjunction with LAI-effected root canal cleaning is not known. For the sake of simplicity, the proteolytic and/or decalcifying properties of the endodontic irrigants under investigation will be termed “dissolving” effects in this text, as opposed to “physical” counterparts, which are related to the propagation and dispersion of the pulsed laser energy throughout the root canal system. The goal of the present study was to weigh such dissolving versus physical effects in conjunction with laser-activated irrigation (LAI) when cleaning simulated human root canals. Dentin debris was taken as the composite organic and inorganic substrate. The cleaning effects of 3% NaOCl (proteolytic), 9% HEDP (decalcifying), 3% NaOCl + 9% HEDP (combined), and deionized water were compared. Parameters of test and control solutions potentially affecting their laser energy absorption and dispersion were assessed, including their absorbance of infrared light and their viscosity. The Null hypothesis tested was that there was no difference between these solutions regarding their effect on dentin debris removal. Materials and methods Ethics statement This study was performed in a simulated clinical environment and was thus neither a human trial nor an animal study. The bovine dentin used here was obtained from animals raised and slaughtered for food production. This study therefore did not require an individual ethics approval according to local law (Kantonale Ethikkommission Zürich, Switzerland). Solutions and their properties Four solutions were used in this study based on their chemical content with the following effects: proteolytic: plain 3% NaOCl (CanalPro, Coltène, Altstätten, Switzerland); proteolytic: plain 3% NaOCl (CanalPro, Coltène, Altstätten, Switzerland); decalcifying: 9% etidronate powder (Dual Rinse HEDP, Medcem, Weinfelden, Switzerland) dissolved in deionized water (Milli-Q, Merck, Darmstadt, Germany); decalcifying: 9% etidronate powder (Dual Rinse HEDP, Medcem, Weinfelden, Switzerland) dissolved in deionized water (Milli-Q, Merck, Darmstadt, Germany); combined: the 3% NaOCl solution (Coltène) with freshly admixed 9% etidronate powder (Medcem); combined: the 3% NaOCl solution (Coltène) with freshly admixed 9% etidronate powder (Medcem); no dissolving effects on substrate: deionized water (Milli-Q, Merck). no dissolving effects on substrate: deionized water (Milli-Q, Merck). Experiments were performed to further describe the solutions under investigation, and to assess some properties that may influence the effects of LAI [5]. Infrared light absorption of the solutions was assessed using a table-top Fourier-transform infrared spectroscopy (FTIR) device (Cary 630 FTIR Spectrometer, Agilent, Santa Clara, CA, USA). Other physico-chemical parameters were determined as described [5]. In brief, available chlorine was measured by iodometric titration in a titration apparatus (665 Dosimat, Metrohm, Herisau, Switzerland), and surface tension in a pendant drop shape analyzer (DSA100, KRÜSS GmbH, Hamburg, Germany) at an ambient temperature of 22 °C. Liquid density was assessed using a high-precision balance (AT261, Mettler Toledo) by measuring the weight of 1.0 mL of the solution at a temperature of 20 °C. Viscosity of the solutions was determined in a rotational viscometer for low viscosity assessments (ViscoQC 300, Anton Paar, Buchs, Switzerland) at a rotational speed of 200 rpm in a controlled temperature environment of 20 °C. All these experiments were performed in triplicates. In the case of the solutions containing HEDP, 3 separate mixtures were prepared, and each solution/mixture was measured 3 times. Power calculation for main experiment Based on similar published experiments [4, 12], an effect size of 0.5 was assumed. With 4 groups, an alpha error probability of 5%, and 95% power, a total sample size of 72 was calculated (G*Power, Heinrich Heine University, Düsseldorf, Germany). To compensate for potential losses, 22 models were fabricated (see below) and tested on 4 separate experimental days, resulting in 88 individual experiments. The treatments (plain 3% NaOCl, plain 9% HEDP, the combined 3% NaOCl + 9% HEDP, and water, ) over these 88 experiments were allocated using a block randomization process (www.randomizer.org, n = 22). Hence, on each experimental day, treatments from each group were performed to avoid clustering bias. Models The 22 three-dimensionally printed models were produced based on a data set obtained by µCT scanning of a mandibular molar [4]. The models had a cuboid outer shape. They featured a removable insert covering the isthmus area and 2 screw holes to fix that part to the main model [4]. This insert could be removed and re-attached for dentin debris placement. The access cavity had a dimension of 6 mm by 8 mm, and a depth of 5 mm. The canals were both 15 mm long with an apical diameter of 0.3 mm (ISO 30). Based on pilot experiments comparing different isthmus configurations, the STL file featuring a “long-wide” isthmus of 4 mm width and 0.4 mm depth was chosen for this study [4]. This configuration was considered ideal for a comparative study, as it resulted in roughly 50% removal of the dentin debris under the simulated clinical SWEEPS (for Shock Wave Enhanced Emission Photoacoustic Streaming, Fotona, Ljubljana, Slovenia) settings described below when plain water was used as an irrigant. The models were produced in a 3D printer (Stratasys J5, Stratasys, Minnetonka, MN, USA) using a proprietary resin (Biocompatible Clear MED 610, Stratasys). After printing, the support material was removed using a high-pressure washer (K2 Premium Power Control, Kärcher, Winnenden, Germany) operated in a glove box. Subsequently, the models were soaked in a support material solvent (SuppEX-705, NT K + D AG, Hombrechtikon, Switzerland), and washed a second time using the high-pressure washer. The models, and especially the outer aspects of the inserts covering the isthmus, were then polished for better transparency. This was done using silicon carbide paper (Struers, Ballerup, Denmark) up to 4000 grit at 150 rpm in a polishing machine (Planopol-2, Struers), followed by hand polishing using a high-gloss paste for resins (KMG, Candulor, Wangen, Switzerland). Dentin debris was obtained by grinding bovine dentin as described elsewhere [13]. In brief, bovine incisor roots were denuded of cementum, rid of pulp tissue, desiccated, and then ground in a ball grinder (Mixer Mill 200, Retsch, Haan, Germany). The resulting powder was sieved (Test Sieve, Retsch, Haan, Germany), and the particles below 100 μm were used for this study. To create the dentin slurry, this dentin powder was mixed with water at a 1.5:1 (wt/wt) ratio. When filling the isthmus in the models, a size-30 ProFile 04 (Dentsply Sirona Endodontics, Ballaigues, Switzerland) was placed passively in each canal to prevent extrusion of the dentin debris from the isthmus. The dentin debris was applied under magnification (Stemi 2000, Zeiss, Oberkochen, Germany), and the removable insert was screwed back in. The models and inserts were cleaned between experimental days using a steam jet cleaner (Wasi-Steam 2, Wassermann, Hamburg, Germany). Irrigation Before laser application, simulated canals were flooded/irrigated with the respective solution using a polypropylene syringe (Omnifix, B. Braun, Melsungen, Germany) equipped with a polypropylene cannula (IrriFlex, Produits Dentaires, Vevey, Switzerland). The cannula was introduced 2 mm from into the full working length (which was 20 mm for both canals) into the root canals. Each of the 2 canals was first filled passively and then irrigated with 1 mL of irrigant between LAI. LAI was performed using a SkyPulse device (Fotona, Ljubljana, Slovenia) in R-SWEEPS mode. The handpiece was equipped with a medium-size laser tip (Flat SWEEPS 400/9, Fotona) and the set “Expert” modalities for first mandibular molars were applied for medium cavity depth: 612 mW at 18 Hz, 30 s per cycle. The laser tip was held above the lingual canal during the first, the buccal canal during the second, and between the canals during the third 30-s cycle. During laser activation, the irrigating solution lost by splattering (approximately 1 mL) was replenished in the access cavity using the syringe and cannula described above. After activation, the root canal system was emptied by tilting the model. The canals were then irrigated with 1 mL of water each to remove chemicals and emptied by tilting the model again. Isthmus areas were photographed as described below. Outcome assessment Each model was photographed under standard settings in a 3D-printed holder at a defined camera-to-object distance using a digital microscope (VHX-7000, Keyence, Osaka, Japan) at 30x magnification. Image analysis was performed using an open-source image processing package based on ImageJ2 (Fiji, Version 2.16.0). Before and after images were superimposed using the StackReg and TurboReg plug ins. For analysis the area of interest was defined within the borders of the isthmus (Fig. 1). Image Calculator was used at a pre-defined threshold (the same for all specimens) to calculate the area of dentin debris that had disappeared in the LAI process in the area of interest in difference images (Fig. 1). Close-up of the isthmus area of the 3D-printed molar root canal system under investigation. The isthmus was filled with a slurry of bovine root dentin powder in water to simulate accumulated dentin debris (a). A second digital image was obtained after 3 SWEEPS cycles (b; RC: simulated root canals bordering the isthmus). The images were superimposed digitally and subtracted to display the changes in grey values above a pre-determined threshold (c). The area of interest was the isthmus, marked in red (c). Cropped difference image showing the apparent debris removal in the region of interest in white, which was then calculated in area % (d) Data presentation and analysis Data related to absorption of the solutions are presented as absorbance vs. wavelength in nm. This data was converted from the wave number values (cm− 1) measured by the FTIR apparatus. These and all other data related to solution properties are presented descriptively as mean values and standard deviations, obtained from three separate experiments. Data related to the different irrigants under investigation used in the main experiment are presented as % of cleared isthmus area. The data was evenly distributed (Kolmogorov-Smirnov test), and differences were explored by one-way ANOVA followed by Tukey’s honestly significant difference (HSD) test. The alpha-type error was set at 5% (p<0.05). Results Solutions The solutes in water present in the different solutions under investigation did not affect their absorption pattern of infrared light (Fig. 2). Infrared absorption of the solutions under investigation as assessed by Fourier-transform infrared spectroscopy. Note the typical water peak above 3000 nm, and the high absorbance in the region of the light emitted by the Er: YAG laser (2940 nm, vertical line) The batch of the dental NaOCl solution used in this study (CanalPro 3%, Coltène) had an NaOCl content that was slightly above that indicated on the label (Table 1). The per-weight content of available chlorine was reduced when 0.9 g of HEDP was admixed as recommended by the manufacturer (Medcem) (Table 1). The density and viscosity of the solutions increased with the solutes present in the mixture and was similar between 3% NaOCl and 9% HEDP. When these solutes were combined, there was an increase in density, viscosity, and surface tension (Table 1). Table 1Characterization of the solutions used in this study (triplicate measurements, means ± standard deviations)VariableDeionized water(Milli-Q)CanalPro 3%9% HEDPCanalPro 3%& 9% HEDP*NaOCl (wt%)0.00 ± 0.003.36 ± 0.010.00 ± 0.003.04 ± 0.02Density (g/mL)1.00 ± 0.001.05 ± 0.001.06 ± 0.001.11 ± 0.00Surface tension (mN/m)72.0 ± 0.171.1 ± 0.172.2 ± 1.173.0 ± 0.9Viscosity (cP)0.99 ± 0.011.12 ± 0.011.23 ± 0.021.39 ± 0.03* 0.9 g of the respective salt was mixed into 10 ml of the NaOCl solution Characterization of the solutions used in this study (triplicate measurements, means ± standard deviations) * 0.9 g of the respective salt was mixed into 10 ml of the NaOCl solution Debris removal All 88 experimental treatments could be assessed. On average, the SWEEPS procedure removed 64 ± 20% of the dentin debris from the isthmus area of the tooth replicas after 3 cycles of 30 s each. There was neither a difference between the plain HEDP nor the plain NaOCl solution and water regarding their cleaning effect (Table 2). These solutions cleared between 59% and 61% of the isthmus area from the debris. The combined NaOCl+HEDP solution, on the other hand, cleared 78 ± 9%, displaying a significant impact of the type of irrigant on the outcome (p = 0.002, Table 2). Table 2Area of dentin debris removed (in %, means and standard deviations) from the isthmus according to the irrigant that was usedIrrigant n Removed debrisSubset*3% NaOCl2258 ± 19%A9% HEDP2259 ± 22%A3% NaOCl + 9% HEDP2278 ± 9%BWater2261 ± 22%A* for alpha = 0.05, one-way ANOVA, Tukey’s HSD test Area of dentin debris removed (in %, means and standard deviations) from the isthmus according to the irrigant that was used * for alpha = 0.05, one-way ANOVA, Tukey’s HSD test Discussion This study was performed to understand the influence of chemicals in endodontic irrigants when removing dentin debris from an isthmus using pulsed erbium-doped yttrium aluminum garnet (Er: YAG) laser. The results suggest that the LAI cleaning effect was physical rather than chemical, i.e. related to the dislodgement rather than a dissolution of the dentin debris that was packed into the isthmus of the 3D-printed root canal replicas. While not proven unequivocally by the current observations, the fluid properties found in the combined NaOCl+HEDP solution appeared to have enhanced this physical LAI effect. Synergies between chemicals contained in root canal irrigants and laser activation represent an area that has been largely neglected thus far [5]. This was not a clinical study, but rather a series of experiments in a simulated clinical environment. Laser effects in such model systems depend on multiple parameters such as energy settings, tip designs, and tip insertion depth [14]. Moreover, the inner anatomy of the model as well as the substrate that is contained therein affects the observed result [4]. The current model used 3D-printed canals without dentinal tubules and employed bovine dentin debris, lacking biofilm or purely organic pulp tissue simulation and a replication of clinically performed irrigation dynamics. Instead, the focus was on the ad hoc removal of densely packed hard tissue debris in the isthmus by LAI. Former studies on the impact of endodontic irrigants, or better, the chemicals dissolved in endodontic irrigants, have hinted at various physical factors that could alter their performance in the context of LAI. The first of these factors would be light absorption. It has been shown than NaOCl and HEDP affect the transmission of ultraviolet light [15]. However, LAI is related to the use of pulsed erbium lasers with a wavelength of 2940 nm [1], and a thus-resulting very high absorption in water [16]. As shown here, the absorption peak of water in that region was not altered, excluding possible effects related to absorption of the laser light. Another factor influencing the LAI-induced cleaning efficacy of endodontic irrigants may be their fluid properties. It has been shown that the addition of HEDP alters those, with a distinct effect on the primary cavitation bubble observed during LAI. The solutions used in this study were comparable to those in a historic control [5]. Interestingly, the batch of the dental NaOCl solution used here had a lower content of available chlorine, and a lower surface tension compared to that that used in the former study. The reason for this may be seen in the inherent instability of NaOCl solutions, and in the fact that manufacturers may add additives to control surface tension and other parameters [17]. As observed in the former study, there were cumulative effects of NaOCl and HEDP not only on density (which should be expected), but also on viscosity and (to a lesser extent) the surface tension of the resulting solutions. As these properties are inherently linked, it was not possible to pinpoint the main factor responsible for the observations in the root canal cleaning model. Nevertheless, logic would dictate that in the absence of any direct effects of either NaOCl or HEDP per se, these altered properties were responsible for the enhanced cleaning effect of the combined NaOCl+HEDP solution. Future studies in this context should use aqueous liquids with added solvents that specifically increase density, viscosity, or surface tension to single out irrigant properties linked to enhanced LAI efficacy. Dentin debris appeared to be a good substrate for this study. Just like smear layer [18], the dentin debris that is generated during instrumentation is basically “dentin mud”, as Schilder called it [19]. Accumulated dentin debris can fill isthmus spaces and lateral canals and impede proper root canal disinfection [20] and root filling [21]. Given its mixed inorganic and organic nature [22], dentin powder was considered ideal to study dissolving effects propagated by LAI. It should allow to study the separate actions that a proteolytic agent and a decalcifying counterpart could exert during LAI. As shown in earlier studies, the sequential application of a NaOCl and an EDTA solution results in better removal of the smear layer (consisting of both organic and inorganic components) compared to either of these two entities alone [18, 23]. Moreover, dentin debris can also be used to discern between dissolving and physical (streaming) effects of LAI in conjunction with irrigants [4, 24]. In the context of LAI, a synergistic effect between energy dispersion in the irrigant and the NaOCl contained therein has been shown regarding pulp tissue dissolution in bovine teeth, which was far superior to that of plain water activated in the same manner, or passive placement of the NaOCl solution [25, 26]. This is due to the physical effects of LAI, rendering organic entities more susceptible to the proteolytic NaOCl action [2]. Similarly, a strong chelator such as EDTA could potentially show a dissolving effect even under current conditions [27]. However, in the current set-up, there was no difference in dentin debris removal between NaOCl and HEDP, indicating that there was no chemical/dissolving action involved in the debris removal. The debris was apparently packed too tightly to reveal any inorganic dissolution by the HEDP. Moreover, the organic component of the debris was apparently too low to reveal an individual NaOCl effect. Earlier studies added a synthetic organic component in form of a hydrogel to the simulated dentin yet did not assess dissolving effects [28]. This was not done here but would have most likely revealed a NaOCl effect [11]. HEDP is a weaker chelator than EDTA [29], and the exposure times in the present study were short. At least in the context of smear layer removal, there is clearly an additive dissolving effect between NaOCl and HEDP [30]. In clinics, the irrigants are refreshed frequently, and used throughout the procedure [9], which was not done here. They thus react for longer time periods with substrates contained in the root canal system. The main advantage of using a combined NaOCl+HEDP solution during instrumentation is that debris accumulation can be reduced [31, 32]. In summary, the current study did not reflect a clinical scenario but was an attempt to single out individual factors. In the context of LAI, the SWEEPS method has gained clinical popularity, because all the root canals are cleaned simultaneously via acoustic shock waves generated in the access cavity flooded with the irrigant. SWEEPS has demonstrated good cleaning ability in extracted human molars [33, 34] and 3D-printed teeth filled with simulated biofilm/ hydrogel [4]. The results in the current model are comparable to those obtained in extracted teeth [33], where a 84% reduction in accumulated dentin debris was found when SWEEPS in conjunction with a 1% NaOCl solution was applied in 3 cycles of 30 s. The slightly lower debris reduction presented here can be explained by the fact that in the former study, extracted human teeth were used and the accumulated debris was generated by instrumentation, while in the current study, the dentin debris was packed into the isthmus of a 3D-printed root canal model. The isthmus in the current model was remote from the access cavity. Yang and co-workers assessed debris removal in the whole canal system [33]. The finding that the combined NaOCl+HEDP irrigant cleaned the isthmus better than its components was peculiar given the absence of individual effects of these two chemicals. The logical explanation for this result is that this combination alters the physical properties of the irrigant, resulting in enhanced action of the pulsed Er: YAG laser. However, as observed in experiments on smear layer removal and root canal wall conditioning, there may be dissolving effects of a combined NaOCl+HEDP that may be stronger than their combined single effects [30]. Moreover and perhaps more importantly, HEDP in the solution combined with NaOCl led to a disproportional increase in viscosity of the resulting solution. As shown earlier, this results in an increased size and lifetime of the primary cavitation bubble [5]. Apparently, and as suggested by the current results, solution parameters altered in this way led to secondary cavitation phenomena that were also different from water or plain NaOCl and HEDP solutions. Future studies should explore this further and delineate ideal solution properties for LAI in the context of root canal debridement. In conclusion, under current conditions, the removal of dentin debris from the isthmus in a 3D-printed molar by SWEEPS was due to physical dislodgement of the dentin matter rather than any dissolution of the same by 3% NaOCl or 9% HEDP. The enhanced effect of the combined NaOCl+HEDP irrigant was most likely due to the altered physical properties of the combined irrigant, resulting in enhanced propagation of the absorbed laser energy. However, this latter finding awaits further clarification in future experiments. Acknowledgements The authors thank Andrea Gubler for swift laboratory work. Clinical trial number Not applicable. Abbreviations Ethylenediaminetetraacetic acid Erbium-doped yttrium aluminum garnet Fourier-transform infrared spectroscopy 1-Hydroxyethylidene-1, 1-diphosphonic acid Honestly significant difference Laser-activated irrigation Sodium hypochlorite Rotations per minute Shock wave enhanced emission photoacoustic streaming Authors’ contributions A.Z.: Methodology, Formal analysis and investigation, Writing - original draft preparation; S.D.: Methodology, Formal analysis and investigation, Writing - review and editing; M.M.: Conceptualisation, Writing - review and editing; L. R.: Methodology, Writing - review and editing; M.Z.: Conceptualisation, Methodology, Writing - original draft preparation, Resources, Supervision. Funding This study was not supported by any external funds. Data availability All data supporting the findings of this study are available from the corresponding author upon reasonable request. Declarations Competing interests M.Z. declares a conflict of interest in the form of a patent (EP3284456A1; US10434038B2) licensed to Medcem LLC. The other authors have no competing interests to declare that are relevant to the content of this article. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement All data supporting the findings of this study are available from the corresponding author upon reasonable request.