The sample size of 13 samples per group was determined after a pilot study (power = 0.95, effect size = 0.66, alpha-type error = 0.05).
Seventy eight extracted human mandibular incisors with intact apices, a minimum tooth length of 18 mm and no previous treatments were selected from a random collection of extracted teeth. Radiographs of all teeth were obtained from bucco-lingual and mesio-distal projections and used to select teeth with a single oval-shaped canal [7]. The teeth were numbered and randomly allocated (http://www.random.org) into 2 groups (n = 39) according to the following acces cavity designs:
Ultraconservative access cavity performed in the incisal edge (UltraAC.Inc)
UltraAC.Inc were drilled with diamond burs (802L 314 − 012; Dr. Hopf GmbH & Co. KG, Langenhagen, Germany) under water cooling at high speed. Incisors were accessed 1 mm palatal to the incisal edge, and cavities extended apically along the long axis. Canals were located while minimizing mesial-distal, buccal-lingual, and circumferential pericervical dentin removal [27] (Fig. 1a and c).
Traditional access cavity (TradAC)
TradAC were prepared following conventional guidelines [3] for outline size and using round 1014 HL and Endo-Z burs (Dentsply Maillefer, Ballaigues, Switzerland) at high speed under water cooling (Fig. 1b and d).
After access cavity preparation, the root canals were instrumented using TruNatomy Glider (17, 0.02v taper), Small (20,0.04v taper), Prime (26,0.04v taper) instruments at 500 rpm and 1.5 Ncm torque values. During the canal preparation, the root canal was irrigated with 2.5 mL 2.5% NaOCl solution after each instrument. After completion of the mechanical preparation, a final irrigation was applied using 5 mL 17% EDTA followed by 5 mL 2.5% NaOCl, and 5 mL distilled water. Them, the canal was dried with paper points.
The root canals were filled with a paste made of calcium hydroxide powder (Kalsin; Spot Dis Deposu A.S.; Izmir, Turkey) mixed with saline solution and 0.1% rhodamine B dye (Sigma-Aldrich, St Louis, MO). Two radiographs (mesiodistal and buccopalatal direction) were taken to confirm complete filling of the canals with the calcium hydroxide paste. The access cavities were sealed with Cavit (3M ESPE, Seefeld, Germany), and the specimens were stored for 1 week at 37 oC in 100% relative humidity. After this period, two coats of colored nail polish were applied to the specimens, including the apical foramen, to prevent leakage of the irrigant. The specimens were fixed in Eppendorf vials with silicone impression material (Optosil; Heraeus Kulzer; Hanau, Germany). These procedures were performed to create a closed system, simulating the vapor lock effect [28]. The roots were divided into 3 subgroups according to the irrigation protocol, as follows:
Sonic Activation with EDDY
Before each cycle of activation, 2.5 mL 17% EDTA (Cerkamed, Stalowa, Poland) was applied to the root canal with a syringe. The irrigant was activated with a frequency of 6000 Hz and an amplitude of 160 mm using an air scaler (TA-200-S4H, MICRON Co., Tokyo, Japan). The EDDY tip was placed 2 mm short of the WL, and in-and-out movements with an amplitude of 5 mm were performed.
Laser-activated irrigation with Er,Cr:YSGG: The root canals were filled with 2.5 mL 17% EDTA and activated with Er,Cr:YSGG laser (Waterlase MD, Biolase Technology Inc., San Clemente, CA, USA) using the RFT2 tip (275 microns in diameter and 21 mm length), placed 2 mm short of the WL. The parameters of laser used were output power of 3W energy, pulse frequency of 20 Hz (pulses per second), using 10% air and 10% water [29]. The irrigant was not aspirated from the canal.
IrriFlex irrigation
A 2.5 mL syringe with 30-G IrriFlex needle was placed 2 mm short of the WL into the canal, and in-and-out movements with an amplitude of 5 mm were performed; 17% EDTA was applied over 30 s.
In all groups, irrigation activation was repeated two times for 30 seconds, resulting in a total of 1 minute of irrigation with 5 mL EDTA. Then, final irrigation with 5 mL 2.5% NaOCl and 5 mL distilled water was performed with IrriFlex needle. After removal from the impression material, each specimen was embedded in a circular self-cure acrylic resin mold. Two 1-mm thick sections of each tooth were obtained at distances of 2 and 4 mm from the root apex using a slow-speed, water-cooled 0.3-mm microtome saw (Isomet 5000; Buehler, Lake Bluff, IL). Slices were taken for confocal laser scanning microscopy (CLSM; Zeiss LSM 800, Jena, Germany) at x5 magnification with a laser wavelength of 561 nm. Excitation wavelength was set at 543 nm, and emission wavelength was set at 565 nm. These images were used to measure the non-penetration percentage, maximum residual calcium hydroxide penetration depth and penetration area (Fig. 2). The method proposed to measure the penetration percentage by Moon et al. was adapted to measure the percentage of clean root canal walls [30]. First, in each sample, the total perimeter of the root canal was measured with the Zeiss Zen software (Carl Zeiss, Jena, Germany) measuring tool. Then, the perimeter along the root canal walls where there was no evidence of residual calcium hydroxide was measured (Fig. 2a). Next, this value was divided by the perimeter of the root canal and this result was multiplied by 100 to calculate the percentage. For the maximum depth of residual calcium hydroxide into the dentinal tubules, the point from the root canal wall to the deepest point where residual calcium hydroxide could be observed was measured at four predetermined points (mesial, distal, buccal and lingual). The mean depth of residual calcium hydroxide was measured by averaging these 4 values for each specimen (Fig. 2b). Penetration area of residual calcium hydroxide was measured using the area calculating tool of the software (Fig. 2c). These measurements were made by an investigator who was blind to the experimental groups.
Statistical Analysis
The data were analyzed with the R Project program (version 3.5.0; Vienna, Austria). Conformity to normal distribution was evaluated using the Shapiro Wilk test. WRS2 package [31] was used for two-way comparison of non-normally distributed parameters (depth of penetration, area of penetration and percentage of non-penetration) according to cavity and cleaning protocol. Significance level was taken as p < 0.05.