1. 3D-printed root canal tooth model
Extracted human premolars without root canal treatment were collected, which was approved by the Scientific Research Projects Approval Determination of Independent Ethics Committee of Shanghai Ninth People’s Hospital affiliated with Shanghai Jiao Tong University (ethical approval document No. 2017-399-T296). All methods were performed in accordance with the relevant guidelines and regulation. The residual calculus and soft tissues were cleaned and the teeth were stored with normal saline. After drying, the canal anatomy was checked by CBCT scan (Scanora 3Dx, Soredex, Tuusula, Finland). According to the results, a suitable tooth was selected: a single oval root canal, canal curvature of 0°, root length of 21.5mm (from physiological apical foramina to buccal cusp). The long axis and short axis of the oval cross section of the root canal were 2.1712mm and 1.2763mm at the root canal orifice, 1.4454mm and 0.8337mm at the middle of the root canal, 0.4658mm and 0.4599mm of 4mm from the apical. The DICOM data of the tooth were processed by Mimics software. Then the 3Dimage was constructed and the STL format was output.
The tooth model was further designed with the pulp opening and anchor points. The STL format was printed by a high-precision 3D printer Projet 3500 HDMAX (3D System, South Carolina, America) with the Visijet M3 Crystal resin material (3D Systems, South Carolina, America).(Figure 1)
2. Root canal preparation
Before the preparation, the CBCT scan of the origin 3D-printed root canal model was performed to save the data for comparison of the effect before and after the preparation in the subsequent procedures. The root canal model was fixed into the vice. Each operator prepared one 3D-printed tooth model. Firstly, the 10# K file and 15# K file were used to dredge 2/3 of the root canal. Then, the SX file of ProTaper nickel-titanium file system (Densply) was used to prepare the 2/3 of the root canal according to the manufacturer's recommended process. 10# or 15#K file was inserted again to reach the working length (21.5mm, anchor point: buccal cusp). Then S1, S2, F1, F2, F3 file were used in turn. The operator can choose the major apice file according to their own judgment. Each time before changing the nickel-titanium, the root canal was rinsed by lateral opening needle with distilled water and rinsing amount >2ML. The K file was used again for dredging and H file was used for further debris cleaning if necessary. After preparation, the debris was cleaned again and the root canals were thoroughly rinsed with the amount of >5 ml. The root canal was dried with absorbent point and the model was removed from the vice.
3. Groups
20 postgraduates from the Department of Stomatology of Shanghai Ninth People's Hospital were randomly divided into two groups according to the two different guidance methods. There were 10 students in each group. One root canal tooth model was prepared by each student and the second preparation was performed under two different guidance methods.
Experimental group: CBCT based image guidance method
The DICOM data of the root canal tooth models before and after preparation of the experimental group were imported into Materialise Mimics V19.0 software. 3D reconstruction was performed by Geomagic Studio2013 software for measurement and analysis. A comparison figure of the reconstruction before and after first preparation was showed for the operator's reference for 5minutes (Figure. 2). Then, according to the guidance, the root canal tooth models were prepared for the second time until the operator thought that the preparation was complete. Then rinsing and drying were done before preservation.
Control group: CBCT original image guidance method
The CBCT original image of the 3D printed models before and after root canal preparation was presented in the computer system of the Department of Endodontics in Shanghai Ninth People′s Hospital. The operators could only view the cross-sections (Figure 3) because of the limitation of the hospital computer system software. The operators could drag the cursor up and down on the screen to observe all levels from the crown to the apical. The time limit was 5 minutes. According to CBCT original image, the root canal tooth model was prepared for a second time until the operator thought that the preparation was complete. Then rinsing and drying were done before preservation.
4. CBCT scanning and image processing
The prepared root canal tooth models were fixed on the device and scanned by CBCT (Scanora 3DX, Soredex, Tuusula, Finland) with the x-ray parameters of 90kV and 10mA. The layer thickness was 0.1mm with a small field of vision of 50mm×50mm. The data of each one was successively imported into Materialise Mimics V19.0 software for 3D reconstruction and STL format was output. After imported into Geomagic Studio2013 software, the comparison with the reconstruction of the original root canal tooth model before preparation was made. The following four indexes were measured and analyzed:
Root canal cleanliness:
When the cutting thickness of the root canal wall of the 3D printed root canal tooth model was ≥ 75μm, it was considered as the effective cleaning area. According to this criterion, the root canal wall was divided into two areas: uncleaned and effectively cleaned areas. Root canal cleanliness referred to the ratio of effectively cleaned area to the total area of the root canal wall.
Root canal transportation:
In any section of the root canal, the maximum distance of the geometric central axis before and after preparation was represented as root canal transportation.
Residual minimum root thickness:
Within the effective measurement range (from the root canal orifice to 1mm from the apical), the minimum distance between any point on the root canal wall and any point on the outer surface of the root was represented as the residual minimum root thickness.
Over-preparation amount:
When the cutting thickness of the root canal wall of the 3D-printed root canal tooth model was ≥ 200μm, it was considered as the over-prepared area. According to this criterion, the root canal wall was divided into two areas: unprepared and overprepared areas. Over-preparation amount referred to the ratio of the overprepared area to the total area of the root canal wall.
5. Statistical analysis
After the second root canal preparation, CBCT scanning was performed again and reconstruction was analyzed to obtain the results of second preparation effect. For the results analysis, the independent sample t-test was used. The level of significance was set at 0.05, and IBM SPSS Statistic version 25 was used. The calculation of the data of first and second preparation in the two groups was made to analyse whether there was a statistical difference in the improvement of the root canal preparation between the two groups.