Ethical approval and sample selection.
The ethics protocol and the final study proposal for this study were confirmed and accepted by the Ethical Committee at Sulaimani University (Protocol Number; 392/2020). This study has followed the CRIS guidelines as discussed in the 2014 concept note [26]. A total of 60 curved single-rooted human premolar teeth with varying degrees of root curvature were selected for this study after the elimination of the straight ones. Specimens with uncommon extreme variations like twisted buccal root or three fused roots excluded. Endodontically treated teeth, internal or external root resorption are also excluded. Whereas those with completely formed apices, single canal with one apical foramen, and ≥10o canal curvature were included. No information about the patients’ age, gender, tooth's quadrant, or reason for extraction. Specimens were stored in 10% formalin for disinfection for a maximum of 2 weeks. Tissue fragments and calcified debris were removed by using a hand scaler, then washed under tap water. Finally, they were stored in normal saline at room temperature until the time of the investigation.
Pre-instrumentation digital periapical imaging
Each tooth was embedded in the radiolucent polysiloxane putty dental impression material (3M ESPE, St. Paul, MN, USA), and encoded with a number. The digital radiographical examination was carried out for all the teeth in two directions (buccolingual and mesiodistal), using a standardized parallel technique. A high-frequency oral x-ray machine (EzRay Air W; Vatech, Korea), were used with an exposure time of 0.367 seconds (60 kV, 4 mA). The target– receptor distance was increased to compensate for image magnification and to ensure that only the most parallel rays were directed toward the tooth and the X-ray sensor (EzSensor Classic, Vatech, Korea). As a result, a long (16-inch) target–receptor distance was used.
Pre‐instrumentation CBCT scans.
Two custom-made wood boxes were used for the mounting of the teeth and to confirm the standardization for the CBCT images. Each tooth was embedded in cold-cure clear acrylic resin (Vertex Castavaria, Netherland) with a technical specification of 9 minutes dough time and 6 minutes working time at 55°C,using a cylindrical plastic container.Thereafter, they quoted with MS3 master die separator (Ivoclar Vivadent, USA) to enable precise repositioning during pre and post-instrumentation scans. Ten teeth were mounted in each template consistently by using dental stone plaster (Rident, Rajasthan, India). Each mold was horizontally fitted to the chin support of the CBCT machine (NewTom, Giano, Verona, Italy) in a way that occlusal plane was parallel to the plate, and scanned with 90 kVp, 3 mA, voxel size: 0.125 mm, exposure time: 5.4 s, by using FOV 8cm by 11cm[27].
Pre‐instrumentation working length estimation on CBCT and digital periapical radiographic images
The measurement line was placed in the center of the pulpal cavity and followed each visible canal deviation, thus also allowing for the measurements of nonlinearly shaped canals. The radiographic tooth length determined on the CBCT and digital periapical images as the distance between the tip of the cusp and the major apical foramen. The radiographic working length for all the specimens was measured separately on digital periapical and CBCT images after subtraction of 1mm from the radiographic tooth length.
Real working length measurement on extracted teeth
A standard straight-line access opening was prepared for all the teeth. A size #10 or #15 K-File (DentsplyMaillefer, Ballaigues, Switzerland) was passively advanced until its tip was seen at the level of the coronal most boundary of the major apical foramen, by the aid of a magnifying glass (Keeler, Windsor, UK, ×3 magnification). The reference point was reproduced by a rubber stopper and the distance calculated with an electronic digital caliper (Mitutoyo Corp., Japan) to the nearest (0.01 mm.), and documented as the actual working length. The real working length was determined by subtracting 0.5 mm from the actual canal length and considered as a gold standard in the present study.
Canal curvature measurements
An experienced oral and maxillofacial radiologist obtainedall the CBCTand digital radiographic images andperformed the measurements. The change in canal axis was determined as the difference between canal curvature before and after instrumentation. For the CBCT evaluation, scan images from the clear sagittal view were selected depending on the multiplanar imaging-reformatted sections. The slices were first reproduced in a vertical position to visualize the tooth cusp, pulp chamber, apical foramen, and the complete view of the root canal pathway. All images converted for viewing with image analysis software (NNT Software, Verona, Italy) to measure the canal curvature angle.
The Schneider method was applied for the estimation of the degree of canal curvature before and after instrumentation. Two straight lines of equal lengths were used. The first line represented the continuity of the apical region, and the second line followed the middle and coronal thirds of the root canal. The angle between the radii was geometrically measured, and the canal curvature was expressed in degrees (Fig.1 and Fig.2).The formed canal angle was named according to the degree of root canal curvature into moderate (10-25o) and severe (26-70o). All scan images captured before and after instrumentation were analyzed with image analysis software (NNT Software, Verona, Italy), to determine the canal curvature changes [28].
Root canal cleaning and shaping
An endodontist, experienced in the use of the Primary WaveOne Gold Reciprocating file (DentsplyMaillefer, Ballaigues, Switzerland), performed all procedures. The analysis of the radiographic images reproduced in three dimensional for the canal curvatures and shaping ability was carried out by an oral and maxillofacial radiologist who was blind in respect of all experimental groups.
The access coronal cavity was prepared using a round carbide bur #4 (DentsplyMaillefer, Ballaigues, Switzerland), and the canal patency checked with a #15 K-type hand file (DentsplyMaillefer, Ballaigues, Switzerland). The coronal third flared with Gates-Glidden drills 2 and 3 (DentsplyMaillefer, Ballaigues, Switzerland). Working length was confirmed manually with a #15 K-File using a standard protocol. A glide path created using a ProGlider instrument (16/02) (DentsplyMaillefer, Ballaigues, Switzerland) carried to the working length. A handpiece generated by an electric motor (Silver, VDW, Munich, Germany) was used for instrumentation. The speed, torque, and file sequence were applied according to the manufacturer's instructions. Only five canals were instrumented at each time interval to minimize operator fatigue. After each file sequence, the prepared canal was washed out with 3 ml of 3% NaOCl solution (Techno Dent, Greece), followed by a 5 ml solution of %17 EDTA (SPIDENT, Korea). Then, the canals were irrigated with 5 ml of 3% NaOCl as a final rinse. Root canal irrigations were performed by using a 5 mL disposable plastic syringe with a 30 gauge side opening needle (Optimus, SP, Brazil) at room temperature. The needle was inserted inside the canal without binding and the solutions were introduced slowly and passively allowing adequate back-flow.
Post‐instrumentation CBCT and digital periapical image analysis
After instrumentation, each tooth was repositioned in its previous position inside a plaster block. The post-CBCT and digital periapical radiographs were obtained with the same parameter applied in the pre- instrumentation phase. Longitudinal axial canal axis and canal curvature angle were determined (Fig.1 and Fig.2). The percentage of the change in canal curvature angle after instrumentation (canal axis modification) was calculated using the following formula [28];
Canal curvature angle after instrumentation- Canal curvature angle before instrumentation
Canal curvature angle before instrumentation X 100
Statistical analysis
Statistical analysis of data obtained in this study performed using IBM SPSS Statistics for Windows, version 24.0 (Armonk, NY: IBM) software. A P-value <0.05 was considered a statistically significant level. The sample size was determined with the Sealed Envelope software for a power of 80%. The normal distribution of the data was tested using the Shapiro-Wilk test. Chi-square tests were used to compare the frequencies of qualitative variables. When the distribution of variables was normal, an independent sample t-test was used to compare the results within the investigated parameters between moderate and severe curvature angles.