Cone-beam computed tomographyis not the imaging technique of choicefor comprehensive orthodontic assessment

Demetrios J. HalazonetisKifissia, Greece

It was a pleasure to see that Dr Larson did not takethe extreme view of proposing cone-beam com-puted tomography (CBCT) as a routine diagnostic

modality—ie, for every patient, irrespective of maloc-clusion or other patient-specific factors—as some or-thodontic postgraduate programs in the United Statesseem to do.1 Even so, he does recommend CBCT asthe standard procedure, stating in his conclusionsthat “CBCT has replaced conventional lateral cephalo-grams and panoramic images as the most commonlyordered imaging for comprehensive orthodontic pa-tients.” In my Counterpoint, I will try to present argu-ments against CBCT as the imaging technique of choicefor comprehensive orthodontic assessment.

Assuming that use for every patient is not advo-cated, what are the patient selection criteria? The an-swer should stem from a comprehensive assessmentof the benefits and burdens to each patient. This as-sessment cannot be completely objective, but our deci-sion making should be based on current evidence,which could also serve as the basis to develop generalguidelines. Such guidelines already exist. The SEDEN-TEXCT project of the European Union had as its pri-mary goal “to acquire key information necessary forsound and scientifically based clinical use of CBCT”and “to use this information to develop evidence-based guidelines dealing with justification, optimiza-tion and referral criteria for users of dental CBCT.”2

The guidelines section dealing with orthodontic diag-nosis concludes that “large volume CBCT should notbe used routinely for orthodontic diagnosis.”

The British Orthodontic Society guidelines givea similar recommendation: “routine use of CBCT evenfor most cases of impaction of teeth . . . cannot yetbe recommended.”3

A similar conclusion was adopted by the Ameri-can Association of Orthodontists in 2010: “theAAO recognizes that while there may be clinical sit-uations where a cone-beam computed tomography(CBCT) radiograph may be of value, the use ofsuch technology is not routinely required for ortho-dontic radiography.”4

If guidelines already exist, what is the purpose ofthis debate? First, it is an opportunity to make theseguidelines well known to the orthodontic communityat a time when CBCT use is increasing. The SEDEN-TEXCT guidelines are based on a systematic review ofthe literature, thus representing current evidence-based knowledge at a confidence level much higherthan this debate can achieve.2 Most importantly, how-ever, is that these guidelines are not compulsory. Theuse of ionizing radiation is governed by law in mostcountries, but all the law requires is clinical justifica-tion. The guidelines are designed to assist the clinicianin the justification process.3 I hope that this debate willconvince clinicians to follow the guidelines’ recom-mendations.

RADIATION BURDEN

The effects of ionizing radiation are considered sto-chastic events. This signifies that the risk, not the sever-ity, of the condition (eg, cancer) depends on the dose.Using a low-dosage vs a high-dosage CBCT machinewill not result in cancers that are easier to treat, onlyfewer of them. The probability of an important sto-chastic effect (cancer and severe hereditary effect) is7.3 3 10�2 Sv.5 For patients aged 10 to 20 years,this doubles to approximately 0.15 Sv. Since a largefield-of-view CBCT will provide a dose of 68 to 368mSv6 compared with approximately 30 mSv for thecephalometric and panoramic combination, this trans-lates to a risk of about 1 in 170,000 to 1 in 20,000above the current customary procedure.5 In the UnitedStates, more than 1.6 million orthodontic patients starttreatment every year.7 If each patient had 1 CBCT im-age, this would result in 10 to 80 additional cancer

Associate professor, School of Dentistry, University of Athens, Athens, Greece.Reprint requests to: Demetrios J. Halazonetis, 6 Menandrou St, Kifissia GR-14561, Greece; e-mail, dhal@dhal.com.Am J Orthod Dentofacial Orthop 2012;141:402-110889-5406/$36.00Copyright Ó 2012 by the American Association of Orthodontists.doi:10.1016/j.ajodo.2012.02.010

American Journal of Orthodontics and Dentofacial Orthopedics April 2012 � Vol 141 � Issue 4

POINT/COUNTERPOINT 403

cases per year. Is this a risk worth taking? This is not aneasy question and depends mainly on the benefit to thepatient.8 What significant improvements in patientoutcomes does CBCT offer? To answer, we shouldnot confuse the benefits to the patient with the techni-cal capabilities of CBCT technology. The fact that CBCTimages are 3-dimensional is not directly relevant. Jus-tification for CBCT images can only be consideredwhen the treatment outcome will not only be better be-cause of them, but also significantly better to outweighthe above risks.

EFFICACY

The following terms are used to evaluate the effi-cacy of diagnostic imaging procedures9: technical effi-cacy, diagnostic accuracy efficacy, diagnostic thinkingefficacy, therapeutic efficacy,patient outcome efficacy,and societal efficacy. Theseefficacies constitute a hierar-chy of levels of increasing im-portance. The top 2 levelsevaluate whether the imagingmethod produces a net bene-fit to the patient and societyin general, and should dictateour imaging policy. Regarding CBCT and its use in or-thodontics, no such studies have been conducted. Wewill consider the relevant evidence for each of the lower4 levels, focusing on large field-of-view protocols,since only these can provide reconstructed lateral ceph-alometric and panoramic views, similar to conventionalradiographs.

Technical efficacy is related to the quality of the im-age. The dimensional accuracy of CBCT images has beenwell established.10 Voxel size is typically 0.3 to 0.4 mm,corresponding to a lower resolution than that of conven-tional intraoral radiographic imaging. Artefacts and noiseare higher than those observed in multi-slice computedtomography, making it difficult, if not impossible, toobtain consistent density values and resulting in lowcontrast and poor depiction of soft tissues.10-12 Segmen-tation is problematic, and even high-contrast objects,such as teeth, are measured with errors that can exceed1 mm, limiting clinical usefulness.13

Diagnostic accuracy efficacy measures the accuracyof diagnosis by using CBCT in comparison with a refer-ence standard—in our case, a cephalogram or pano-ramic radiograph. Alveolar bone thickness and height,and the presence of fenestrations and dehiscences,have been compared between CBCT images and direct

measurements. Due to the relatively large voxel size,thin structures are difficult to detect, and alveolarbone covering the incisors might be underestimated,although the results are conflicting.14,15 Errors in mea-suring bone thickness can exceed 1.4 mm for a 0.4-mmvoxel size.15 Fenestrations and dehiscences are overes-timated to a large degree.15,16

At present, there are no diagnostic accuracy studiesregarding the localization of impacted canines, andnone are expected because this question is not seriouslydebated.2,17 Regarding resorption of adjacent teeth,CBCT images show improved sensitivity and specificityover panoramic radiography.18 CBCT has been shownto have increased diagnostic accuracy over posteroante-rior cephalograms in patients with skeletal asymmetry.19

Concerning periodontal assessment, although it hasa definite 3-dimensional advantage, CBCT comple-

ments but cannot replace in-traoral radiography, mainlybecause of reduced resolu-tion.20 Studies on skull mate-rial have shown that CBCTimages provide better diag-nostic information,21 butthere is no consensus regard-ing the accuracy of thesemeasurements.20,22 The SED-

ENTEXCT guidelines conclude that “CBCT is not indi-cated as a routine method of imaging periodontalbone support,” although it might be indicated in se-lected patients, but preferably not with a large field ofview.2 The American Board of Orthodontics includesCBCT images as an option to document periodontal sta-tus but does not consider radiographic images, in gen-eral, as compulsory data and gives priority to clinicalexamination and conventional radiography.23

Diagnostic thinking efficacy evaluates whether theimaging method changes the diagnosis from the pre-test situation. Therapeutic efficacy assesses whetherthe test produces changes to the treatment plan. Theseefficacies have been evaluated for impacted thirdmolars24 and impacted canines.25-27 CBCT imagesare perceived to be more useful than traditional radio-graphs for such cases26 and might change therecommended treatment plan in approximately30% of them.25,27 However, no patient outcome effi-cacy studies have been conducted, and CBCT is recom-mended only when “the information cannot beobtained adequately by lower dose conventional (tradi-tional) radiography.”2 Dr Larson referred to the study ofBecker et al28 of 28 failed cases of impacted canines,but the main reason for failure was inadequate

Assuming that use for every patientis not advocated, what are the pa-tient selection criteria? The answershould stem from a comprehensiveassessment of the benefits and

burdens to each patient.

Counterpoint 405

American Journal of Orthodontics and Dentofacial Orthopedics April 2012 � Vol 141 � Issue 4

anchorage rather than improper localization. The au-thors acknowledged that the initial clinical and radio-graphic signs were sometimes sufficient to diagnoseproperly but were misinterpreted by the clinician. Thereare numerous cases when an impacted maxillary caninecan be clearly localized based on conventional radio-graphs and clinical examination (eg, palpation, posi-tion, and inclination of adjacent teeth), and nofurther imaging is justified.3

Regarding resorption of adjacent teeth, diagnosticthinking efficacy and therapeutic efficacy studiesshowed that resorption defects can be identified betterwith CBCT images, but these studies mostly used a me-dium or small field of view.25,26,29

Dr Larson also referred to the temporomandibularjoint, but asymptomatic patients surely do not needtemporomandibular joint imaging. Condylar positionin the fossa can certainly be seen on CBCT images,but this information should not affect our diagnosisand treatment plan.30 The value of temporomandibularjoint imaging even for patients with temporomandibu-lar disorders is a debatable subject, and there is no ev-idence to show that CBCT images will provide bettertreatment.31

It seems, therefore, that CBCT might benefit somepatients with the conditions mentioned above, but noevidence exists for the remaining majority of ourpatients. The application of 3-dimensional cephalo-metrics, or increased measurement accuracy, could bean indication. However, currently, there are no estab-lished 3-dimensional cephalometric analyses and no3-dimensional normative data. CBCT images areused to simulate old technology—ie, reconstruct2-dimensional lateral cephalometric views. In this tran-sitory, backward step, we should not carry with us themisconceptions of the early cephalometric era: strictadherence to cephalometric standards and blind faithin numbers.

Cephalometric analyses have significant, well-recognized deficiencies, and increased accuracy ofmeasurements does not address them.32 There is, asyet, no evidence that increased accuracy from CBCTcontributes to a change of treatment plan or bettertreatment. Even though such a notion might seemself-evident, one should consider that our treatmentmodalities are not so fine tuned to specific craniofacialpatterns that a conventional cephalometric radiographis inadequate to serve. Furthermore, identifying land-marks on CBCT images introduces significant errorsthat might mitigate the advantage of increased accu-racy.33 Lastly, most of our diagnostic information isgained from clinical evaluations. The cephalogram

serves as an adjunctive tool and has been shown tobe superfluous in some circumstances, affectingtreatment-planning decisions in some patients and toa limited degree.34,35

INCIDENTAL FINDINGS

Incidental findings are no justification for radio-graphic exposure. The European guidelines on radia-tion protection state that “‘Routine’ radiography isunacceptable practice” and define a ‘routine’ or‘screening’ examination as “one in which a radiographis taken regardless of the presence or absence of clinicalsigns and symptoms.”5 If we put this recommendationto the side for a moment, CBCT undoubtedly providesmany findings, although incidence varies.36,37 Priceet al37 reported that 90% of all CBCT images examinedhad at least 1 finding, and 16.1% of the findings re-quired further investigation. In contrast, incidentalfindings from conventional orthodontic radiographsseem much lower, although no direct comparison hasbeen made.38 By far, the most common pathologicfindings seen in CBCT images that could require inter-vention were carotid artery calcification and periapicalosteitis.37 These are not outside the detection capabil-ities of the panoramic radiograph.39 Therefore, oneshould consider that a significant number of incidentalfindings in CBCT images (1) represent normal anatomicvariants or are benign and do not require further inter-vention, (2) might already be known to the patient, (3)can be detected on traditional radiographic images, or(4) might be false-positive findings. In the absence ofany signs or symptoms, the taking of CBCT imagesjust in case an occult pathologic finding appears isnot justified.

COMPREHENSIVE EVALUATION

Dr Larson seems to base his recommendations onthe premise that a comprehensive imaging modalitywill prove useful on any occasion, so, why not takeit from the start? After all, a CBCT image includesa cephalogram and a panoramic image for orthodon-tic assessment, and additional images for any poten-tial periodontal evaluation, temporomandibular jointevaluation, temporary anchorage device placement,and airway analysis, plus the benefit of any incidentalfindings. More extensive diagnostic knowledge is as-sumed to lead to better treatment. This propositionis alluring but precarious and biased for several rea-sons. First, most of our patients are known not tohave any of the problems listed, so extra radiation isused just to rule out additional incidental findings,

Counterpoint 407

American Journal of Orthodontics and Dentofacial Orthopedics April 2012 � Vol 141 � Issue 4

over those that would be found with a cephalogramand a panoramic radiograph. Second, if more diag-nostic information is the goal, why stop at a CBCT?Why, to take it to the extreme, not perform a full-body computed tomography scan instead? Have wedetermined that a CBCT is the optimum choice inthe risk-benefit balance? Third, in our quest formore information, why not perform other diagnostictests that might be more relevant and do not incura radiation burden? Such tests could include evalua-tion of thyroid and growth hormone levels, magneticresonance imaging examination of the head (to assesstemporomandibular joint disc position, measure thesizes of the muscles of mastication, examine the pitu-itary gland for adenomas, and evaluate the airway),nasal endoscopy or rhinomanometry, and bite-forcemeasurement. Have we specifically selected largefield-of-view CBCT based on evidence that it will re-sult in better patient outcome than these other tests,or are we just using it because it is convenient?Fourth, with each diagnostic test comes more knowl-edge but also more false-positive findings that canlead to increased patient anxiety, unnecessaryfollow-ups, and further tests. Diagnostic evaluationsshould be focused and designed to answer specificquestions, not be a fishing expedition.

CONCLUSIONS

As more research is conducted, and with continualimprovements in technology, CBCT might prove valu-able for all of our patients in the future. However, atthe present time, evidence for the efficacy of CBCT im-aging is lacking. Except for certain patients, replacingthe conventional cephalometric and panoramic radio-graphs with a large field-of-view CBCT is simply over-kill, potentially leading to a public health problem.40 Itis the responsibility of the clinician to carefully selectpatients when CBCT imaging will provide a tangiblebenefit and resist the lure of technology for technol-ogy’s sake.

In response to the Steiner quotation: “Today, justlike orthodontic radiography in the early 1900s, CBCTfor orthodontic therapy is advocated by experts, with-out reliable evidence that the diagnostic technologyis associated with improved patient outcomes.”40

REFERENCES

1. Smith BR, Park JH, Cederberg RA. An evaluation of cone-beamcomputed tomography use in postgraduate orthodontic pro-grams in the United States and Canada. J Dent Educ 2011;75:98-106.

2. SEDENTEXCT project. Radiationprotection: cone beamCT for den-tal and maxillofacial radiology. Evidence based guidelines 2011.Available at: http://www.sedentexct.eu/files/guidelines_final.pdf.Accessed on January 20, 2012.

3. Isaacson K, Thom A, Horner K, Whaites E. Guidelines for the useof radiographs in clinical orthodontics. London, United King-dom: British Orthodontic Society; 2008.

4. American Association of Orthodontists. Statement on the roleof CBCT in orthodontics (26-10 H). eBulletin; May 7, 2010. Avail-able at: www.aaomembers.org/Resources/Publications/ebulletin-05-06-10.cfm. Accessed on January 20, 2012.

5. European Commission. Radiation Protection 136. Europeanguidelines on radiation protection in dental radiology. Luxem-bourg: Office for Official Publications of the European Commu-nities; 2004:Available at: http://ec.europa.eu/energy/nuclear/radioprotection/publication/doc/136_en.pdf:Accessed on Janu-ary 20, 2012.

6. Pauwels R, Beinsberger J, Collaert B, Theodorakou C, Rogers J,Walker A, et al., The SEDENTEXCT Project Consortium. Effectivedose range for dental cone beam computed tomography scan-ners. Eur J Radiol 2012;81:267-71.

7. American Association of Orthodontists. 2010 AAO member andpatient census study. Final report; June 23, 2011.

8. Kokich VG. Cone-beam computed tomography: have we identi-fied the orthodontic benefits? Am J Orthod Dentofacial Orthop2010;137(4 Suppl):S16.

9. Fryback DG, Thornbury JR. The efficacy of diagnostic imaging.Med Decis Making 1991;11:88-94.

10. Ballrick JW, Palomo JM, Ruch E, Amberman BD, HansMG. Imagedistortion and spatial resolution of a commercially availablecone-beam computed tomography machine. Am J Orthod Den-tofacial Orthop 2008;134:573-82.

11. Nackaerts O, Maes F, Yan H, Couto Souza P, Pauwels R, Jacobs R.Analysis of intensity variability in multislice and cone beam com-puted tomography. Clin Oral Implants Res 2011;22:873-9.

12. Schulze R, Heil U, Gross D, Bruellmann DD, Dranischnikow E,Schwanecke U, et al. Artefacts in CBCT: a review. Dentomaxillo-fac Radiol 2011;40:265-73.

13. Nguyen E, Boychuk D, Orellana M. Accuracy of cone-beam com-puted tomography in predicting the diameter of unerupted teeth.Am J Orthod Dentofacial Orthop 2011;140:e59-66.

14. Timock A, Cook V, McDonald T, Leo MC, Crowe J,Benninger B, et al. Accuracy and reliability of buccal boneheight and thickness measurements from cone-beam computedtomography imaging. Am J Orthod Dentofacial Orthop 2011;140:734-44.

15. Patcas R, M₠uller L, Ullrich L, Peltom₠aki T. Accuracy of cone-beamcomputed tomography at different resolutions assessed on thebony covering of the mandibular anterior teeth. Am J OrthodDentofacial Orthop 2012;141:41-50.

16. Leung CC, Palomo L, Griffith R, Hans MG. Accuracy and re-liability of cone-beam computed tomography for measuringalveolar bone height and detecting bony dehiscences andfenestrations. Am J Orthod Dentofacial Orthop 2010;137(4Suppl):S109-19.

17. Benn DK. Diagnostic accuracy studies needed for cone beamcomputed tomography. Evid Based Dent 2011;12:37.

18. Alqerban A, Jacobs R, Souza PC, Willems G. In-vitro comparisonof 2 cone-beam computed tomography systems and panoramicimaging for detecting simulated canine impaction-induced ex-ternal root resorption in maxillary lateral incisors. Am J OrthodDentofacial Orthop 2009;136:764.e1-11.

Counterpoint 409

American Journal of Orthodontics and Dentofacial Orthopedics April 2012 � Vol 141 � Issue 4

19. Damstra J, Fourie Z, Ren Y. Evaluation and comparison ofpostero-anterior cephalograms and cone-beam computed to-mography images for the detection of mandibular asymmetry.Eur J Orthod 2011 Mar 31 [Epub ahead of print].

20. Vandenberghe B, Jacobs R, Yang J. Diagnostic validity (or acuity)of 2D CCD versus 3D CBCT-images for assessing periodontalbreakdown. Oral Surg Oral Med Oral Pathol Oral Radiol Endod2007;104:395-401.

21. Mol A, Balasundaram A. In vitro cone beam computed tomogra-phy imaging of periodontal bone. Dentomaxillofac Radiol 2008;37:319-24.

22. Misch KA, Yi ES, Sarment DP. Accuracy of cone beam computedtomography for periodontal defect measurements. J Periodontol2006;77:1261-6.

23. American Board of Orthodontics. Case report preparation. Avail-able at: http://www.americanboardortho.com/professionals/clinicalexam/casereportpresentation/preparation/radiographs.aspx. Accessed on January 20, 2012.

24. Guerrero ME, Shahbazian M, Elsiena Bekkering G, Nackaerts O,Jacobs R, Horner K. The diagnostic efficacy of cone beam CTfor impacted teeth and associated features: a systematic review.J Oral Rehabil 2011;38:208-16.

25. Haney E, Gansky SA, Lee JS, Johnson E, Maki K, Miller AJ, et al.Comparative analysis of traditional radiographs and cone-beamcomputed tomography volumetric images in the diagnosis andtreatment planning of maxillary impacted canines. Am J OrthodDentofacial Orthop 2010;137:590-7.

26. Katheria BC, Kau CH, Tate R, Chen JW, English J, Bouquot J. Ef-fectiveness of impacted and supernumerary tooth diagnosis fromtraditional radiography versus cone beam computed tomogra-phy. Pediatr Dent 2010;32:304-9.

27. Botticelli S, Verna C, Cattaneo PM, Heidmann J, Melsen B. Two-versus three-dimensional imaging in subjects with uneruptedmaxillary canines. Eur J Orthod 2011;33:344-9.

28. Becker A, Chaushu G, Chaushu S. Analysis of failure in the treat-ment of impacted maxillary canines. Am J Orthod DentofacialOrthop 2010;137:743-54.

29. Alqerban A, Jacobs R, Fieuws S, Willems G. Comparison of twocone beam computed tomographic systems versus panoramic

imaging for localization of impacted maxillary canines and de-tection of root resorption. Eur J Orthod 2011;33:93-102.

30. Rinchuse DJ, Kandasamy S. Orthodontic dental casts: the caseagainst routine articulator mounting. Am J Orthod DentofacialOrthop 2012;141:9-16.

31. Petersson A. What you can and cannot see in TMJ imaging—anoverview related to the RDC/TMDdiagnostic system. J Oral Reha-bil 2010;37:771-8.

32. Moyers RE, Bookstein FL. The inappropriateness of conventionalcephalometrics. Am J Orthod 1979;75:599-617.

33. Damstra J, Fourie Z, Huddleston Slater JJ, Ren Y. Reliability andthe smallest detectable difference of measurements on3-dimensional cone-beam computed tomography images. AmJ Orthod Dentofacial Orthop 2011;140:e107-14.

34. Nijkamp PG, Habets LL, Aartman IH, Zentner A. The influence ofcephalometrics on orthodontic treatment planning. Eur J Orthod2008;30:630-5.

35. DevereuxL,MolesD, CunninghamSJ,McKnightM.How importantare lateral cephalometric radiographs in orthodontic treatmentplanning? Am J Orthod Dentofacial Orthop 2011;139:e175-81.

36. Cha JY, Mah J, Sinclair P. Incidental findings in the maxillofacialarea with 3-dimensional cone-beam imaging. Am J Orthod Den-tofacial Orthop 2007;132:7-14.

37. Price JB, Thaw KL, Tyndall DA, Ludlow JB, Padilla RJ. Incidentalfindings from cone beam computed tomography of the maxillo-facial region: a descriptive retrospective study. Clin Oral ImplantsRes 2011 Sep 30 [Epub ahead of print].

38. Bondemark L, Jeppsson M, Lindh-Ingildsen L, Rangne K. Inci-dental findings of pathology and abnormality in pretreatmentorthodontic panoramic radiographs. Angle Orthod 2006;76:98-102.

39. Damaskos S, Griniatsos J, Tsekouras N, Georgopoulos S,Klonaris C, Bastounis E, et al. Reliability of panoramic radiographfor carotid atheroma detection: a study in patients who fulfill thecriteria for carotid endarterectomy. Oral Surg Oral Med OralPathol Oral Radiol Endod 2008;106:736-42.

40. Hujoel P, Hollender L, Bollen AM, Young JD, McGee M, Grosso A.Head-and-neck organ doses from an episode of orthodonticcare. Am J Orthod Dentofacial Orthop 2008;133:210-7.

Counterpoint 411

American Journal of Orthodontics and Dentofacial Orthopedics April 2012 � Vol 141 � Issue 4