Page 1/16

Accuracy of a Method to Monitor Root PositionUsing 3D Digital Crown/root Model DuringOrthodontic TreatmentsKaho Ogawa 

Harvard School of Dental MedicineYoshiki Ishida 

Harvard School of Dental MedicineYukinori Kuwajima 

Harvard School of Dental MedicineCliff Lee 

Harvard School of Dental MedicineJacob R Emge 

Harvard School of Dental MedicineMitsuru Izumisawa 

Iwate Medical University: Iwate Ika DaigakuKazuro Satoh 

Iwate Medical University: Iwate Ika DaigakuShigemi Nagai 

Harvard School of Dental MedicineJohn D Da Silva 

Harvard School of Dental MedicineChia-Yu Chen  ( Chia-Yu_Chen@hsdm.harvard.edu )

Harvard School of Dental Medicine https://orcid.org/0000-0003-3829-1256

Research article

Keywords: Digital dentistry, CBCT, digital scans, orthodontic tooth movement

Posted Date: May 18th, 2021

DOI: https://doi.org/10.21203/rs.3.rs-525470/v1

License: This work is licensed under a Creative Commons Attribution 4.0 International License.  Read Full License

Page 2/16

AbstractObjectives.

The objective of this retrospective clinical study was to assess the accuracy of a method of predictingpost-movement root position during orthodontic treatment, using a 3D digital crown/root model (3DCRM),which was created with pre-movement records of both CBCT and dental-arch digital scans.

Material & Methods.

Pre-movement CBCT scans, post-movement CBCT scans, and dental-arch digital scans of �ve patientswho had completed orthodontic treatments were used in this study with the approval of internal reviewboard of Iwate Medical University. The 3DCRM was superimposed onto post-movement scanned dental-arch to identify the post-movement root position (test method). Post-movement CBCT (referenced ascurrent method) served as the control used to identify the actual post-movement root position. Thepredicted root positions by the two methods were compared using color displacement mapping and 3Dcoordinate XYZ analysis.

Results.

Color displacement mapping showed that the average root displacement of �ve cases was -0.16 mm±0.05 mm. 3D coordinate analysis revealed no signi�cant differences between test and current methodsin X and Y axis. However, the discrepancy in Z axis (especially in cases of intrusion) was greater than allother directions (mesial, distal, labial, palatal and extrusion direction) for all three tooth types examined(central incisor, lateral incisor and canine, p <0.05). A strong positive correlation between the degree ofdiscrepancy and the distance of tooth movement was observed on the Z axis (r =0.71).

Conclusions.

The 3DCRM method showed promising potential to accurately predict root position during orthodontictreatments without the need for a second CBCT. However, root resorption, which affected the Z axisprediction, needs to be closely monitored using periapical radiographs to complement this method.

Clinical Relevance.

The 3DCRM method can be a useful tool to better evaluate root positioning during and after orthodontictreatment without the need to expose patients to an additional CBCT scan. 

IntroductionThe goal of orthodontic treatment is to establish ideal three-dimensional crown and root position in afunctional, stable, and esthetic occlusion. Andrews reported six keys to normal occlusion, 1) molarrelationship, 2) crown angulation (mesiodistal tip), 3) crown inclination (labiolingual or buccolingual

Page 3/16

inclination), 4) rotations, and 5) occlusal plane and 6) occlusion based on crown information from studymodels.[1] Proper root position and parallelism are imperative for adequate occlusal function, periodontalhealth, implant placement, and restorative treatment.[2–5] Root position and parallelism are importantfactors for achieving even distribution of occlusal forces to create ideal function and for establishingproper contours and emergence pro�les of restorations.[2–5] Root proximity may lead to rapidperiodontal breakdown and horizontal bone loss instead of intrabony defects that are amenable toregeneration.[6–11] The consequences of roots moving out of the alveolar housing include clinicalattachment loss, recession, bone dehiscence, mobility, and even tooth loss.[12, 13] Therefore, predictingroot position during orthodontic treatment is a critical factor for successful outcomes.

Two-dimensional (2D) images such as cephalometric and panoramic radiographs cannot evaluate thethree-dimensional (3D) position of the teeth and roots relative to the maxillofacial region and alveolarbone during orthodontic treatments. With its increasing availability, cone-beam computed tomography(CBCT) has been used to aid in diagnosis and assessment during orthodontic treatment. The advantageof CBCT is that it provides the exact 3D location of the crown and roots of teeth and their relationshipwith both neighboring teeth and alveolar bone.[14, 15] Therefore, visual and timely evaluation of rootpositioning using 3D imaging by CBCT is crucial in orthodontic treatments. However, CBCT is not withoutits disadvantages.

Since the effective radiation dose from CBCT is signi�cantly higher compared to conventionalradiographs, routine usage of CBCT is not recommended during orthodontic treatments as a substitutefor conventional radiographs, especially when the population for orthodontic treatment is relativelyyoung.[16, 17] Furthermore, CBCT rendering of teeth lacks precise occlusal surface and accurateinterdigitation.[18] Artifacts from metal restorations and orthodontic brackets can also result indiscrepancies.[19] On the other hand, digital scanning can provide precise tooth morphology and registeraccurate interocclusal relationship. Studies have also shown that brackets did not affect the accuracy ofdigital scans.[20, 21] Therefore, registration and subsequent superimposition of CBCT and digital scansobtained with either an intraoral scanner or an extraoral lab scanner have become a standard work�ow inmany orthodontic appliance systems to assess tooth alignment. While this method is generally clinicallyacceptable, the registration of CBCT to digital scans is less accurate compared to registration of digitalscans to digital scans.[22]

To overcome these shortcomings, Lee et al. in 2014 introduced a monitoring method by creating “teethcomposites” whereby the 3D digital model was composed of crown extracted from digital scans and rootportion extracted from CBCT.[22] The author demonstrated in an ex-vivo typodont study that this methodwas reliable to track the 3D position of whole teeth including roots. This method was not further veri�edin clinical studies due to impracticality related to technique-sensitive, complex and time-consumingprocess of threshold segmentation of teeth in real patients.

With the rapid development of imaging and digital technology in recent years, much improvement hasbeen made in both hardware and software. Many commercial companies now provide service to

Page 4/16

clinicians to process CBCT, digital scan data and produce 3D models with 3D printing. The cost of theseservices has also become signi�cantly lower. Creating 3D digital crown/root models (3DCRM) byintegrating digital scan crowns and CBCT root with patient data become a feasible approach. Therefore,this method �rst introduced by Lee et al. in 2014 will be an ideal modality.

This is a retrospective study using clinical cases and the aims of this study were 1) to assess theaccuracy of this new method of predicting post-movement root position compared to the current methodof using post-movement CBCT and 2) to analyze the association of accurately predicting root position ineach of the three tooth types, and 3) to analyze the association between accuracy of test method and theamount of actual root apex movement. We hypothesize that the 3DCRM method will be as accurate asthe use of post-movement CBCT in evaluating root positioning and thus eliminate the use of a second ormultiple CBCTs during or at the completion of orthodontic treatment.

Materials And Methods

Data collectionIn �ve orthodontic cases, the pre- and post-movement CBCT images and scanned dental arches wereobtained from the patient database of the Division of Orthodontics, Department of Developmental OralHealth Science, School of Dentistry, Iwate Medical University in Japan. Cases with cleft-palate and othercraniofacial syndromes were excluded. This study was approved by the institutional review board (IRB) ofIwate Medical University, School of Dental Medicine (No. 01329).

The CBCTs were captured at 90 ~ 120 kV, 6.0 ~ 7.5 mA using a dental CBCT scanner (3D Accuitomo 170,J. MORITA CORP, Kyoto, Japan). The images were reconstructed with 0.28 mm slice thickness andexported as digital imaging and communications in medicine (DICOM) �les. The pre- and post-movementscanned dental arches were obtained using a 3D digital scanner (MDS500 Dental Scanner, AGE SolutionsS.r.l., Pisa, Italy) and exported as STL �les.

Fabrication of 3D digital crown/root models (3DCRM)The CBCT DICOM �les as well as digital scanning STL �les were sent to a commercial 3D servicecompany (3DDX, Boston, MA, USA) for further processing. Brie�y, the pre- and post-movement DICOM�les were imported into dental implant planning software (Simplant, Materialise Dental NV, Leuven,Belgium) and converted to stereolithography (STL) �les. The 3D digital crown/root models (3DCRM) werecreated using pre-movement CBCT images and scanned dental arches (Fig. 1). The STL data of the pre-movement CBCT images and the scanned dental arches were superimposed with the crown shape asindex (Fig. 1A). Second, using the superimposed images, individual 3DCRMs of the six maxillary anteriorteeth were created and exported as STL �les (Fig. 1B). The accuracy of the resulting 3DCRMs werechecked by three clinicians before approval for use in the study.

Test method to predict the post-treatment root position

Page 5/16

These 3DCRMs were superimposed onto the post-movement scanned dental arches using 3D datainspection software (GOM inspect, GOM, Braunschweig, Germany, Fig. 2A). The 3D coordinates in X, Y,and Z axis of the root apex were measured by the same person three times at three different time points,and the average value was used for the analysis.

Current method to identify the post-treatment root positionThe post-movement of CBCT and scanned dental arches were imported into 3D data inspection software,and this data was superimposed with the crown shape as index. The 3D coordinates in X, Y, and Z axis ofthe root apex were measured with the same manner as the test method (Fig. 2B).

Analysis of accuracy for prediction of post-movement rootposition and statistical analysis

Color displacement map

A color displacement map of the root was used by 3D data inspection software to quantify thedifferences between the root position created by the test method and the current method. The averagedisplacement of each tooth was calculated and the statistical difference among the tooth type wasanalyzed using one-way ANOVA (p < 0.01, SPSS ver. 24, IBM, Armonk, NY, USA).

3D coordinates assessment of the discrepancy in six directions.

The discrepancy between post-movement root position determined by the test method and that of thecurrent method was calculated in 3D axis X (DX), Y (DY) and Z (DZ) by the following equations (1):

     (1)

The discrepancy in each of the three axes amongst the three tooth types were compared using one-wayANOVA and Tukey post-hoc test (p < 0.01). One sample t-test was also performed to compare between thediscrepancy in each of the three axes and CBCT voxel size.

Association between accuracy of test method and the amount of actual root apex movement.

The amount of actual root apex movement (DARAM) was calculated using the 3D coordinates of pre-movement of CBCT (Pre) and post-movement of CBCT (Post) by the following equations (2):

Page 6/16

      (2)

The Pearson’s correlation coe�cient between the distance of root movement (DARAM in X, Y and Z axis)and absolute value of the discrepancy on the X, Y and Z axis was calculated.

Results

Displacement in color mapThe color displacement map of the root position by the test method and the current method displayedthree colors: green, blue, and red (Fig. 3). Case 2 was predominantly in the green range indicating aminimum discrepancy between test method compared to the current method. In contrast, case 4 hadsigni�cant yellow ~ red color at the root apex indicating outward displacement greater than 0.5 mm(Fig. 3). The average root displacement of �ve cases was − 0.16 mm ± 0.05 mm, and there was nosigni�cant difference among the three tooth types (p > 0.01, Table I).

Discrepancy in 3D coordinatesFigure 4 shows the distribution of discrepancy in X (the labial and palatal direction), Y (the mesial anddistal direction) and Z axis (the intrusion and extrusion direction) with the zero point, which is the locationdetermined by the current method. On the XY coordinate plane, mesial-palatal discrepancy is shown inquadrant I; mesial-labial discrepancy in quadrant II, distal-labial discrepancy in quadrant III, and distal-palatal discrepancy in quadrant IV. The most discrepancy was seen in quadrant III, which means that theroot position was predicted to be more distal and labial. The Z axis analysis indicated that root positionwas overly predicted in the apical direction. The average discrepancy of each of X, Y, Z axis and directionsis shown in Fig. 5. The discrepancy in Z axis on apical direction (intrusion direction) was signi�cantlygreater than all other directions for all tooth types (p < 0.05, Fig. 5). The discrepancy in Z axis on apicaldirection was also signi�cantly higher than CBCT voxel size (p < 0.01), but no signi�cant difference wasobserved between the discrepancy on other �ve axes and CBCT voxel size. Average discrepancy of eachof three tooth types is shown in Table II. No signi�cant difference was observed among the tooth types inthe X, Y, and Z axis.

AssociationThere was a strong positive correlation (r = 0.71) observed between the amount of root movement in the Zaxis (DARAM) and the absolute value of discrepancy in the Z axis. This means that prediction accuracyof root position by the test method decreases as the distance of root movement increases in Z axis. Nocorrelation was observed within the X and Y axis (Fig. 6).

Discussion

Page 7/16

Orthodontic treatment aims to move teeth to ideal positions within the extent of the alveolar bone withoutdamaging the roots or adjacent structures.[15] Therefore, it is important to monitor the position of boththe roots and crowns of teeth. The CBCT and panoramic radiographs are used to con�rm the rootposition during orthodontic treatments. The majority of orthodontists use panoramic radiographs tovisualize the correct root positions. A survey in 2008 found that 67.4% and 80.1% of Americanorthodontists took panoramic radiographs during orthodontic treatments and post-treatment,respectively.[23] There is a report that the magni�cation-distortion ratio of the CBCT images (1.04) issmaller than that of panoramic radiographs (1.20).[24] However, several papers reported that CBCTshould not be used routinely on every patient due to resulting higher dose of radiation in comparison toconventional radiographs.[16, 17, 25, 26] Therefore, our new method to accurately predict the rootposition during orthodontic treatments without taking additional CBCT will be bene�cial and necessary.

3DCRM used in this study is a digital model of the crown and root created by the pre-movement scanneddental arches and CBCT images. We hypothesized that the root position during orthodontic treatmentscould be predicted by superimposing 3DCRM onto the post-movement scanned dental arches withoutextra CBCT images. The 3DCRM can be individually superimposed to each tooth of post-movementscanned dental arches by utilizing the crown portion as the index.

In this study, we evaluated the accuracy of new method (test method) in predicting root position using thecolor mapping analysis and 3D coordinate analysis. The advantage of the color mapping analysis is tovisualize the displacement of the entire root surface.

The color mapping analysis visually indicated the distinguishable displacement between the test methodand current method mainly at the root apex, and the rest root surface had minimal discrepancy within thegreen level. Furthermore, our results (average 0.16 mm ± 0.05 mm displacement) based on clinical caseswas comparable to results from the ex vivo typodont study by Lee et al in which displacements of 0.17 ± 0.32 mm and 0.11 ± 0.16 mm were found for the maxillary and mandibular teeth respectively.[22]

3D coordinate analysis can provide an accurate prediction of the root apex in the direction of discrepancywith numerical data. The 3D coordinate analysis indicated a signi�cant discrepancy only on the Z axis(the apex/intrusion and incisal/extrusion direction) but neither the X axis (the labial and palatal direction)nor the Y axis (the mesial and distal direction). The numerical data of the average discrepancy on the Zaxis-apical was 0.83 mm, which is 2.9 times larger than a single CBCT voxel unit (0.28mm). In contrast,the data on X and Y axis was about 1.1 ~ 1.8 times larger without statistical difference. This means thata discrepancy existed with regards to the root apex position in the Z axis, especially in the intrusiondirection. In addition, a strong positive correlation was observed between the distance of actual root apexmovement and the discrepancy along the Z axis.

The discrepancy in the apical direction is likely due to root resorption that occurs during orthodontictreatment. Based on the changes in the crown orientation, superimposition of the pre and post-movementscans allowed the root orientation, and thus the bucco-lingual and mesio-distal orientation could bepredicted. However, resorption of the apex of the root and the amount of resorption cannot be predicted,

Page 8/16

leading to the discrepancy along the Z axis. Excessive orthodontic and occlusal-loading forces areassociated factors of root resorption.[27–33] Since the root resorption associated with orthodontictreatment can occur and progress without any symptoms, monitoring root positions with radiographsduring orthodontic treatment is essential.[34] Routine radiographic evaluation using periapical �lms isrecommended to accurately monitor the root position during orthodontic treatment.

One of the limitations of this study is the lack of variation of clinical cases. The cases used were non-extraction cases, and the amount of root movement was relatively small. Further study is requiredinvolving moderate to severe crowding cases to validate this method.

ConclusionsThe 3DCRM method showed promising potential to accurately predict root position during orthodontictreatments. However, in cases of root resorption, it was di�cult to identify the position of the root in in theapico-coronal direction. Therefore, frequent assessment of root resorption using periapical radiographs isrecommended to compensate for the lack of accuracy of predicting movement along the Z axis.

DeclarationsCompliance with Ethical Standards

1. Ethics approval and consent to participate: This work contained use of patient orthodontic treatmentrecords, digital scans and CBCT data retrospectively. The Institutional Review Board (IRB) of IwateMedical University approved the study.

2. Consent for publication: This manuscript does not contain identi�able data; therefore, it does notrequire additional consent for publication except for the IRB approval.

3. Competing interests: All authors declare no con�ict of interest.

4. Funding: The work was not supported by any external funding

5. Authors’ contributions: KO performed the measurement and wrote the manuscript. YI performed themeasurements and was responsible for research design. YK was responsible for the clinicalmanagement and gathering of clinical data. CL contributed to statistical analysis. JE contributed tothe discussion and integration of data. MI and KS overseen the IRB application as well as the clinicaltreatments. SN was responsible for communication between the two research sites/team andcontributed to inception of the study. JD contributed greatly to interpretation of the data. CYC wasresponsible for substantively revised the work and interpretation of data.

�. Acknowledgements: The authors would like to thank 3DDX for working on the CBCT segmentationand 3D model creation.

References1. Andrews LF: The six keys to normal occlusion. Am J Orthod 1972, 62(3):296-309.

Page 9/16

2. Balut N, Klapper L, Sandrik J, Bowman D: Variations in bracket placement in the preadjustedorthodontic appliance. American journal of orthodontics and dentofacial orthopedics : o�cialpublication of the American Association of Orthodontists, its constituent societies, and the AmericanBoard of Orthodontics 1992, 102(1):62-67.

3. Dewel BF: Clinical observations on the axial inclination of teeth. Am J Orthod 1949, 35(2):98-115.

4. Vermylen K, De Quincey GN, van 't Hof MA, Wolffe GN, Renggli HH: Classi�cation, reproducibility andprevalence of root proximity in periodontal patients. J Clin Periodontol 2005, 32(3):254-259.

5. Vermylen K, De Quincey GN, Wolffe GN, van 't Hof MA, Renggli HH: Root proximity as a risk markerfor periodontal disease: a case-control study. J Clin Periodontol 2005, 32(3):260-265.

�. Olsen CT, Ammons WF, van Belle G: A longitudinal study comparing apically repositioned �aps, withand without osseous surgery. Int J Periodontics Restorative Dent 1985, 5(4):10-33.

7. Waerhaug J: Eruption of teeth into crowded position, loss of attachment, and downgrowth ofsubgingival plaque. Am J Orthod 1980, 78(4):453-459.

�. Nevins M: Interproximal periodontal disease - the embrasure as an etiologic factor. Int J PeriodonticsRestorative Dent 1982, 2(6):8-27.

9. Klassman B, Zucker HW: Treatment of a periodontal defect resulting from improper tooth alignmentand local factors. J Periodontol 1969, 40(7):401-403.

10. Ritchey B, Orban B: The crests of the interdental alveolar septa. The Journal of Periodontology 1953,24(2):75-87.

11. Akiyoshi M, Mori K: Marginal periodontitis: a histological study of the incipient stage. J Periodontol1967, 38(1):45-52.

12. Wennstrom JL: Mucogingival considerations in orthodontic treatment. Semin Orthod 1996, 2(1):46-54.

13. Steiner GG, Pearson JK, Ainamo J: Changes of the marginal periodontium as a result of labial toothmovement in monkeys. J Periodontol 1981, 52(6):314-320.

14. Garib DG, Calil LR, Leal CR, Janson G: Is there a consensus for CBCT use in Orthodontics? Dentalpress journal of orthodontics 2014, 19(5):136-149.

15. Kapila SD, Nervina JM: CBCT in orthodontics: assessment of treatment outcomes and indications forits use. Dento maxillo facial radiology 2015, 44(1):20140282.

1�. Signorelli L, Patcas R, Peltomaki T, Schatzle M: Radiation dose of cone-beam computed tomographycompared to conventional radiographs in orthodontics. Journal of orofacial orthopedics =Fortschritte der Kieferorthopadie : Organ/o�cial journal Deutsche Gesellschaft fur Kieferorthopadie2016, 77(1):9-15.

17. Silva MA, Wolf U, Heinicke F, Bumann A, Visser H, Hirsch E: Cone-beam computed tomography forroutine orthodontic treatment planning: a radiation dose evaluation. American journal oforthodontics and dentofacial orthopedics : o�cial publication of the American Association of

Page 10/16

Orthodontists, its constituent societies, and the American Board of Orthodontics 2008,133(5):640.e641-645.

1�. Othman AA, El-Beialy AR, Fawzy SA, Kandil AH, El-Bialy AM, Mostafa YA: Methods for managing 3-dimensional volumes. American Journal of Orthodontics and Dentofacial Orthopedics 2010,137(2):266-273.

19. Schulze R, Heil U, Gross D, Bruellmann DD, Dranischnikow E, Schwanecke U, Schoemer E: Artefacts inCBCT: a review. Dento maxillo facial radiology 2011, 40(5):265-273.

20. Kang SJ, Kee YJ, Clara Lee K: Effect of the presence of orthodontic brackets on intraoral scans. TheAngle orthodontist 2020.

21. Jung YR, Park JM, Chun YS, Lee KN, Kim M: Accuracy of four different digital intraoral scanners:effects of the presence of orthodontic brackets and wire. Int J Comput Dent 2016, 19(3):203-215.

22. Lee RJ, Pham J, Choy M, Weissheimer A, Dougherty HL, Jr., Sameshima GT, Tong H: Monitoring oftypodont root movement via crown superimposition of single cone-beam computed tomography andconsecutive intraoral scans. American journal of orthodontics and dentofacial orthopedics : o�cialpublication of the American Association of Orthodontists, its constituent societies, and the AmericanBoard of Orthodontics 2014, 145(3):399-409.

23. Keim RG, Gottlieb EL, Nelson AH, Vogels DS, 3rd: 2008 JCO study of orthodontic diagnosis andtreatment procedures, part 1: results and trends. Journal of clinical orthodontics : JCO 2008,42(11):625-640.

24. Yim JH, Ryu DM, Lee BS, Kwon YD: Analysis of digitalized panorama and cone beam computedtomographic image distortion for the diagnosis of dental implant surgery. The Journal ofcraniofacial surgery 2011, 22(2):669-673.

25. Isaacson KG, Thom AR, Atack NE, Horner K, Whaites E: Guidelines for the use of radiographs inclinical orthodontics, 4 edn: British Orthodontic Society; 2015.

2�. Abdelkarim AA: Appropriate use of ionizing radiation in orthodontic practice and research. Americanjournal of orthodontics and dentofacial orthopedics : o�cial publication of the American Associationof Orthodontists, its constituent societies, and the American Board of Orthodontics 2015, 147(2):166-168.

27. Baumrind S, Korn EL, Boyd RL: Apical root resorption in orthodontically treated adults. Americanjournal of orthodontics and dentofacial orthopedics : o�cial publication of the American Associationof Orthodontists, its constituent societies, and the American Board of Orthodontics 1996, 110(3):311-320.

2�. Weiland F: Constant versus dissipating forces in orthodontics: the effect on initial tooth movementand root resorption. European journal of orthodontics 2003, 25(4):335-342.

29. Taithongchai R, Sookkorn K, Killiany DM: Facial and dentoalveolar structure and the prediction ofapical root shortening. American journal of orthodontics and dentofacial orthopedics : o�cialpublication of the American Association of Orthodontists, its constituent societies, and the AmericanBoard of Orthodontics 1996, 110(3):296-302.

Page 11/16

30. Paetyangkul A, Turk T, Elekdag-Turk S, Jones AS, Petocz P, Cheng LL, Darendeliler MA: Physicalproperties of root cementum: Part 16. Comparisons of root resorption and resorption craters after theapplication of light and heavy continuous and controlled orthodontic forces for 4, 8, and 12 weeks.American journal of orthodontics and dentofacial orthopedics : o�cial publication of the AmericanAssociation of Orthodontists, its constituent societies, and the American Board of Orthodontics 2011,139(3):e279-284.

31. Maltha JC, van Leeuwen EJ, Dijkman GE, Kuijpers-Jagtman AM: Incidence and severity of rootresorption in orthodontically moved premolars in dogs. Orthodontics & craniofacial research 2004,7(2):115-121.

32. Stenvik A, Mjor IA: Pulp and dentine reactions to experimental tooth intrusion. A histologic study ofthe initial changes. Am J Orthod 1970, 57(4):370-385.

33. Currell SD, Liaw A, Blackmore Grant PD, Esterman A, Nimmo A: Orthodontic mechanotherapies andtheir in�uence on external root resorption: A systematic review. American journal of orthodontics anddentofacial orthopedics : o�cial publication of the American Association of Orthodontists, itsconstituent societies, and the American Board of Orthodontics 2019, 155(3):313-329.

34. Ahangari Z, Nasser M, Mahdian M, Fedorowicz Z, Marchesan MA: Interventions for the managementof external root resorption. The Cochrane database of systematic reviews 2015(11):Cd008003.

TablesTable I.

The average of root displacement of each tooth type.  Average S.D. Max. Min.

Total -0.165 ±0.053 1.889 -1.222

Central incisor -0.140 ±0.102 1.889 -1.222

Lateral incisor -0.136 ±0.122 1.878 -1.000

Canine -0.177 ±0.124 1.516 -1.055

        Unit: mm

 

Page 12/16

Table II.

The average of root displacement of each tooth type in X, Y, Z axis and six directions.There was no signi�cant difference among three tooth types.

Discrepancy Central incisor Lateral incisor Canine

Axis Direction Mean (SD) n Mean (SD) n Mean (SD) n

X axis Palatal 0.63 (0.38) 6 0.46 (0.28) 10 0.51 (0.18) 3

Labial 0.48 (0.44) 4 NA 0 0.33 (0.16) 7

Y axis Distal 0.18 (0.06) 5 0.24 (0.27) 7 0.61 (0.50) 6

Mesial 0.41 (0.51) 5 0.17 (0.13) 3 0.35 (0.21) 4

Z axis Incisal 0.27 (0.19) 2 0.21 (0.02) 2 0.36 (0.30) 5

Apical 0.92 (0.61) 8 0.83 (0.47) 8 0.72 (0.32) 5

              Unit: mm

Figures

Figure 1

Protocol to generate the 3D digital crown/root models (3DCRM): A, pre-movement CBCT image. B, pre-movement scanned dental arches. C, the pre-movement CBCT image was superimposed on the pre-

Page 13/16

movement scanned dental arches with the crown shape as index. D, the individual 3DCRMs of the sixmaxillary teeth were extracted from surrounding structures.

Figure 2

Determination of the post-movement root position by the test method and the current method. A: The testmethod, the individual 3DCRMs were superimposed on the post-movement scanned dental arches withthe crown shape as index. B: The current method, post-movement of CBCT data were superimposed onthe post-movement scanned dental arches.

Page 14/16

Figure 3

The color displacement map of the root position of case 2 and case 4. The green (zero point) indicatesthat the test method and the current method had no displacement. The red indicates the outwarddisplacement of the test method compared to the current method. The blue indicates the inwarddisplacement of the test method compared to the current method.

Figure 4

The distribution of the discrepancy between the test method and the current method in X, Y and Z axiswith zero point as the current method. X axis represents discrepancy to labial-palatal, Y axis to mesial-distal and Z axis to apex/intrusion– incisal/extrusion.

Page 15/16

Figure 5

Comparison of the average discrepancy of each X, Y, Z axis and six directions by one–way ANOVA andTurkey's post-hoc test. The discrepancy in Z axis on apical direction (intrusion direction) was signi�cantlygreater than all other directions (p<0.05).

Figure 6

Page 16/16

Correlation between the actual amount of root apex movement (DARAM) and discrepancy absolute value.A, X axis B, Y axis C, Z axis.