Specimen selection and preparation
The sample size was calculated by an effect size of 0.6 determined based on a previous study [12], an alpha-type error of 0.05, and a study power of 0.95, using the F tests family and a priori power analysis (G*Power 3.1.9.3; Heinrich Heine University, Dusseldorf, Germany). The estimated sample size was 60 specimens in total.
Human permanent molar teeth extracted for reasons unrelated to this study were collected and preliminary scanned by micro-CT (SkyScan1176; Bruker, Kontich, Belgium). Finally, 60 of the 100 molars were selected for the study with the following criteria: (a) fully formed apex without fracture; (b) no root resorption (internal, external, or apical), calcifications, and previous endodontic treatment; (c) one independent severely curved root canal with a curvature of 25° to 50° according to the Schneider method [13]; (d) long oval-shaped canals would be excluded ( the ratio of the long to short canal diameter was > 2). The molars were scanned at 80 kV and 309 µA using a pixel size of 9 µm. 180°rotation around the vertical axis and a 0.1-mm-thick copper plus aluminum filter were selected. After scanning, the data were reconstructed using NRecon v.1.18.8.0 software (Bruker-micro CT), with smoothing of 5, ring artifacts reduction of 12, beam hardening correction of 30% to distinguish the density of dentin, and presenting the true internal structure of root canal. Region of interest (ROI) was established from the furcation level to the apex of the root, resulting in 600–700 cross-sections per specimen. For each tooth, only one qualified root canal was selected, and the maximum curvature of each root canal was recorded. All the molars were cleaned using an ultrasonic cleaner to remove the periodontal membrane and calculus and stored in 1% chloramine-T solution at 4℃ for the next procedure.
The crown access was obtained using diamond burs according to the morphology of the pulp chamber, the canal orifices were located and confirmed with a #10 K-file (VDW), and the model number of the initial apical file (the K-file whose tip diameter is the same as the apical foramen diameter) was determined and registered as the size of the apical foramen. To achieve consistent baseline conditions between the groups, the specimens were matched to create 12 groups of 5 teeth based on their angle of curvature and the size of the apical foramen. Then, the 5 teeth from each group were randomly assigned to 5 experimental groups (n = 12) according to the different systems they used: ML, RB, VT, WOG, and XP groups. The degree of homogeneity (baseline) of groups was statistically confirmed at a significance level of 5% (P > 0.05, 1-way analysis of variance test) (Table 1)
Table 1
Characteristics of curved root canal teeth per group.
|
Curvature (°)
|
Size of initial apical file
|
System
|
Mean ± SD
|
Min
|
Max
|
Mean ± SD
|
Min
|
Max
|
M3-L
|
35.58 ± 8.57
|
25
|
50
|
9.50 ± 0.90
|
8
|
10
|
Reciproc Blue
|
35.33 ± 7.66
|
25
|
50
|
9.50 ± 0.90
|
8
|
10
|
V-Taper 2H
|
34.92 ± 7.88
|
25
|
49
|
9.67 ± 0.78
|
8
|
10
|
WaveOne Gold
|
35.42 ± 8.07
|
25
|
48
|
9.33 ± 0.98
|
8
|
10
|
XP-endo Shaper
|
37.75 ± 8.27
|
27
|
50
|
9.33 ± 0.98
|
8
|
10
|
P value (ANOVA)
|
0.92
|
0.89
|
Root canal preparation
To avoid apical transportation before preparation, the working length (WL) was determined by introducing a pre-curved #06 K-file into the canal until it was visible at the main apical foramen and then withdrawing it 0.5 mm. A glide path was established after manual preparation up to ISO 15/.02 using stainless K-files, and no coronal expansion was conducted to prevent interfering with coronal transportation. Then, root canal preparations were performed using the different nickel-titanium systems in each system according to the manufacturer’s instructions. In the ML group, the L2 file (25/06.5) was used in rotary motion at a speed of 350 rpm and torque of 2.5 Ncm; in RB group, the R25 file having a tip size of 25 and a 0.08 taper over the first 3 mm was utilized in a reciprocating mode; in VT group, the V25/06 file was used at 350 rpm and 2 Ncm in rotary motion; in WOG group, the primary file (25/07 red) was employed with a reciprocating mode; in XP group, the file was operated at 800 rpm and 1 Ncm in rotary motion.
All of the files were gently inserted into the root canals and applied with light up-and-down movements with light apical pressure at a distance of 1 mm (3 to 5 up-and-down strokes every time). It was necessary that the canals remain moist during preparation. After every 3 to 5 up-and-down strokes, if the WL was not achieved, the file should be withdrawn from the root canal while rotating, followed by irrigation of the canal with 2 mL 2.5% NaOCl for 30 s. Then the patency was rechecked with a #6 K-file, and after that, the preparation was restarted again. The operations above were repeated until the WL was reached. In the XP group, the file was moved up and down over the entire length five times after it reached the WL to achieve the final size of approximately 30/04. Final irrigation was performed by rinsing 5 mL of 2.5% NaOCl solution for 1 min at a distance of 1 mm from the WL. A total of 15 mL of flushing fluid was used for each root canal using a syringe with a 30-guide-vented NaviTip irrigation needle. To eliminate bias owing to instrument fatigue, each fresh file was used to prepare only one canal. All instrumentation operations were conducted by a single endodontist with 10 years of clinical experience in a 37℃ incubator, and no accidents or errors like instrument fractures occurred during root canal preparation in the 5 groups.
Debris assessment
The whole preparation process described above was performed in an apically extruded debris collection device adapted from a past report [14] (Fig. 1). The tooth was fixed with wax at the level of cementoenamel junction (CEJ) in a circular opening made in the separated cap of a 10 mL Eppendorf tube large enough to hold it. Then a 50 mL glass bottle with a rubber stopper was prepared, digging a round hole in the rubber stopper with the same diameter as the Eppendorf tube, and inserting the tube into the glass bottle to avoid fouling during preparation and interference with the weighing result. Afterward, a 5 mL disposable syringe needle was inserted into the tube cap to balance the internal and external pressures. Finally, the glass bottle is wrapped in tinfoil to prevent the operator from seeing the debris extrusion.
Before preparation, the Eppendorf tube was weighed by cap removal (average weight three times, precision 0.0001 mg). After preparation, the debris adhering to the root surface was collected by washing the root with 1ml distilled water while in the tube. Being dried in an oven at 110°C for 5 h, the Eppendorf tube containing dentin debris was weighed by cap removal again (average weight three times, precision 0.0001 mg). The difference in weighing results was calculated, which was the amount of apically extruded debris (AD).
Evaluation of shaping ability
The specimens were rescanned and reconstructed after canal preparation, with the same parameters as the initial scan. The pre- and post-preparation data were imported into Mimics Research v.20.0 software (Materialise, Leuven, Belgium) for 3D reconstruction and analysis. The pre- and post-preparation root canals were visualized and colored green and red, respectively, and the volumes of the pre- and post-preparation canals were measured, allowing the percentage of increment in root canal volume (%VI) to be calculated. The Mimics “Align” function was employed to calibrate and superimpose the root canal 3D models before and after preparation (Fig. 2). The untouched root canal areas (UTA) could be displayed (the green areas) and the percentage of UTA could be estimated using the formula: The number of static voxels × 100/total number of surface voxels. For analysis of root canal transportation, the “Fit Centerline” function of Mimics was used to fit the axial centerlines of root canals before and after instrumentation, canal transportation (CT) could be obtained by measuring the distance between the two central points of the cross-section images at seven levels (P1-P7) selected in advance: 1-, 2-, 3-mm distance from the apical foramen, the curved vertex, the initial point of bending, root canal orifice and the midpoint between the canal orifice to the initial point of bending (Fig. 3).
Evaluation of cleaning ability by scanning electron microscopy (SEM)
After micro-CT scanning, two shallow grooves were created longitudinally on the buccal and lingual sides of the root. The grooves should run parallel to the curve of the root canal and not puncture it. Then the root was split in half using a hammer and chisel. The root canal orifices were blocked by sterile small cotton balls to prevent the contamination of canals throughout the process. Freeze-drying at -80 ℃ for 24 h, the split roots were sprayed with gold for the following SEM examination.
Both the two parts of all canals were assessed, and two fields of the canal walls were randomly selected, respectively, in the coronal, middle, and apical thirds to be photomicrographed under ×100 and ×5000 magnification. The debris and smear layer in each region were scored using the scoring system proposed by Gutmann [15]. The debris was given a score of 1–4 according to the proportion of the surface covered by debris. The smear layer was scored based on the area covered by the smear layer and whether the dentinal tubules were visible and patent. The measurements were carried out independently by two researchers blinded to group allocation, one of whom had not participated in the study design or operation. Training and agreement on criteria has been achieved prior to scoring, and once the researchers have different opinions, an accordant score will be reached after discussion. The scoring example photos were shown in Fig. 4.
Statistical analysis
The normalcy and homogeneity of variance of the data were checked by the Shapiro-Wilk test and Levene test. One-way analysis of variance (ANOVA) and the Tukey test were used to compare the 5 groups for variables that presented normal distribution (the angle of curvature, apical size, %UTA, %VI, AD). When the variables did not have a normal distribution (CT and SEM scores), the Kruskal-Wallis test was used to analyze the data, and the Mann-Whitney U test was used for multiple comparisons. The level of significance was set at 5%. Statistical Package for the Social Sciences v.21.0 (SPSS, IBM Brasil, SP, Brazil) and GraphPad Prism v.7.04 was used for all analysis.