Response to AAO/AAOMR draft document “Clinical Recommendations For The Appropriate Use Of Cone Beam Computed Tomography (CBCT) In Orthodontics”
James Mah, DDS, MSc, MRCD, DMSc (ABO Diplomate and Program Director in Orthodontics) David Hatcher, DDS, MSc, MRCD (Oral and Maxillofacial Radiologist)
Robert Langlais, DDS, MS, FICD, FRCD (Oral and Maxillofacial Radiologist)
Eleven – the number of years CBCT has been available in North America
Thirty – the number of months the AAO and AAOMR worked to produce the document
Nineteen – the number of days the AAO allowed for its membership to respond
Given the limited time to respond, one would expect this issue to be one of the utmost urgency and immediacy warranted by numbers of casualties and complications of CBCT imaging in orthodontic practices. However, nothing like this has been reported. In fact, the National Council on Radiological Protection (NCRP), Report 145, acknowledges that in the entire history of dental radiographic imaging there has not been a single report of a patient being harmed. This brings forth the question – What is purpose of this document? At first, it seems the document provides clinical guidelines for utilization of CBCT, however, upon review it emerges as a selectively written document that more so reflects the opinions of the authors versus an objective and cogent description of the technology and its utilization. The authors’ choice of words and caricature of the clinical situation propagates misunderstanding, instills fear and could lead to potential peril for patients, orthodontists and even the AAO.
Evidence-Based or No-Evidence – Which is it?
On one hand the authors argue that utilization of CBCT imaging with orthodontic patients should be based on evidence and go further to explain their process of qualifying and vetting the literature. Yet on the other hand, the discussion of risk is not particularly evidence-based at all and no process is described with respect to qualifying the statements provided. The inclusion of the Claus et al, 2012 study clearly demonstrates that no exclusionary process was used to eliminate unsupported claims of radiological risk. The AAOMR itself issued a statement to address the “irreconcilable data differences” and “limitations in data collection” that invalidate this study (Lam and Yang, 2012). One must ask, why is this study even mentioned? This article and much of the subsequent information relating to risk similarly unbalances the document. Although it acknowledges that “levels are never reached in the diagnostic range encountered in conventional oral and maxillofacial radiology’ the subsequent text includes damaging effects such as oral mucositis and direct DNA damage which are found in discussions related to very significantly higher therapeutic doses. There simply is no credible evidence to support harm to patients from dental radiographs, including CBCT. The current state of understanding with respect to low levels of exposure was summed up recently in 2011 by the American Association of Physicists in Medicine, stating “risks of medical imaging at effective doses below 50,000 μSv for single procedures or 100,000 μSv for multiple procedures over short time periods are too low to be detectable and may
be nonexistent”. Additionally, NCRP report No 145 acknowledges that dental diagnostic exposures are 735-800 times lower than naturally occurring background levels and states “there is no conclusive proof that radiation from dental x-rays is harmful”. With respect to the possible risks of dental imaging in younger populations, stochastic effects and cancerous effects years later, NCRP report No 145 acknowledges that “if a substantial risk existed it would have been identified and reported [by now]”. NCRP report No 91 establishes the concept of a Negligible Individual Risk Level (NIRL) which is regarded as trivial with respect to the risk of fatality associated with ordinary, normal societal activities and can, therefore, be dismissed from consideration”. Our response here is not intended to be a comprehensive literature review nor to debate the authors but to provide several examples of available literature and understanding that was somehow omitted from the AAO/AAOMR draft document. There are many other similar credible documents available with respect to the risks or lack thereof with dental imaging. Since there is no current evidence supporting harm to patients from dental imaging, the comments related to risks are based upon theoretical estimates. This should be clearly stated in the document and also acknowledged that these estimates, particularly those based on the Linear Non-threshold Theory (LNT) are highly controversial and not universally accepted by members of the radiological and scientific community. The LNT extrapolates the risks of radiation indefinitely in a linear fashion from high doses to very low doses. Indeed, Dr. L.S. Taylor, past president of the National Council on Radiation Protection is cited as describing the linear extrapolation from high doses to very low doses in the calculation of collective doses as “deeply immoral uses of our scientific heritage”. Yet this document states “Radiation epidemiologists and
radiobiologists internationally are in consensus that for stochastic risks such as carcinogenesis, from a radiation safety perspective, the risk should be considered to be linearly related to dose, all the way down to the lowest doses.” Lastly, mention of the national radiation safety campaigns (Image GentlyTM and National Children’s Dose Registry) is inappropriate as these campaigns are directed towards medical CT imaging. In summary, the AAO/AAOMR has drafted a document that calls for a rigorous process of scientific review to support the benefits of CBCT imaging to orthodontic patients yet the assumed risk to patients is unproven and rests upon controversial theories and doses from medical imaging or therapeutic doses which are very significantly higher.
The available published evidence supporting the use of CBCT imaging in orthodontics somehow was also similarly omitted or down played. Before we enter a discussion on evidence, one must first question the applicability of the evidence-based process for the obvious. Does one really need “evidence” to conclude that clinicians can see patient anatomy much clearer, without distortions or superimpositions? Secondly, the AAO has hundreds of members that utilize CBCT. Are the opinions and experiences of its membership not a level of clinical evidence that could be used to establish guidelines? Critics may say that the AAO membership may be unreliable, un-scientific, un-reviewed and so on, but they are not un-important. The ongoing evolution of the concept of evidence-based medicine allows for more common types of evidence. Recent updates acknowledge that “evidence-based medicine can be understood as a medical “movement” in which advocates work to popularize the method and usefulness of the practice in the public, patient communities, educational institutions and continuing education of
practicing professionals (http://en.wikipedia.org/wiki/Evidence-based_medicine). This description of evidence-based medicine best describes the most common mode of clinical developments and advancement in orthodontics. The advent of the straight-wire appliance, bonded brackets, nickel titanium materials were all largely established following this model. Further, the definition of evidence-based-practice continues to evolve and is now recognized to “ involve complex and conscientious decision-making which is based not only on the available evidence but also on patient characteristics, situations, and preferences. It recognizes that care is individualized and ever changing and involves uncertainties and probabilities.” (http://en.wikipedia.org/wiki/Evidence- based_practice). Most decisions in dentistry seem to fit the more recent descriptions of evidence-based-medicine. A search for a randomized, triple-blind, placebo-controlled trial in the discipline of orthodontics shows no results. Further, the few prospective randomized clinical trials in orthodontics do not pass the criteria of high quality clinical trials. It is recognized that other levels of evidence such as “case reports are of particular value in situations where it is impossible to obtain a sample size of sufficient statistical power” (Dr. Robert Keim, JCO Editorial, July 2012). Given this, one must question the validity of the approach that was taken to present the evidence in the AAO/AAOMR draft document.
Thirdly, consideration must be given as far as the suitability of evidence-based decisions with respect to subject areas that for any number of reasons have little “evidence”. The reasons range from; research design issues such as the time required for orthodontic treatment, difficulties in obtaining adequate sample sizes, research faculty shortages, and overall lack of available funding for orthodontic research. Nevertheless in areas of little “evidence”, inappropriate use of evidence-based decisions could lead to unbelievable and harmful situations. For example, stringent application of the evidence-based process does not support the use of local anaesthetic in dentistry. No studies exist to show that restorations are better or surgical procedures have more favorable outcomes with anaesthetic. One could even argue that use of local anaesthetic in dentistry is largely based on inferential data and historic practices.
Fourthly, the section “2 – Selection of Clinical Conditions for Indications of CBCT use”
is well done however these findings are largely ignored or downplayed in the subsequent section “3- Definition of Orthodontic Treatment Difficulty Criteria” and Table III. Many of these criteria simply cannot be adequately assessed without comprehensive 3-D evaluation. The recommendations as presented thrust orthodontists into impossible situations. Conflicting recommendations exist within the document. Recommendation 1.6. states “Do not take a conventional image if it is clear from the clinical examination that a CBCT study is indicated for proper diagnosis and/or treatment planning.” Yet later it is stated “A possible indication for a supplementary CBCT examination (in addition to periapical, occlusal and/or panoramic images) is when interceptive orthodontics is being considered for children between the ages of 5 to 11” and “in children more than 11 years of age if surgical exposure is being considered as a treatment option and the location of the crown cannot be determined clinically or with conventional two-dimensional images (e.g. panoramic, occlusal and/or periapical images). In both these situations radiation
from the conventional radiographic images is at least the same or higher than a CBCT
evaluation. With the “supplemental CBCT examination” the total exposure is even higher and contradicts Recommendation 1.6. Lastly, the diagnostic value of the conventional images is less than that gained with CBCT. In the latter situation of surgical exposure of an impacted tooth, there are many more considerations in this complex situation. Damage to nearby dental roots, amount of bone covering the tooth, the precise position of the tooth, root morphology and possible pathway of tooth movement are a few of these considerations.
It is noteworthy that the AAE/AAOMR statement contains a similar pattern of underplaying the benefit of CBCT in endodontics. In endodontic diagnosis, there is strong evidence to support utilization of CBCT. For example, CBCT showed 34% more lesions (p<.001) compared to peri-apical radiography (Low et al, 2008); CBCT identified 62% more peri-apical lesions (Lofthag et al, 2007) and over a range of tooth types a group of ten endodontists missed one root canal system in 4 out of 10 teeth (Matherne et al, 2008) . Most clinicians would strongly argue that evidence like this is compelling and should be acknowledged, yet the AAE/AAOMR position paper merely states that CBCT has been shown to be merely comparable to conventional imaging. Is the word comparable appropriate to describe detection of 34 – 62% more lesions and a 40% failure rate with peri-apical images? A better choice of word would be superior, and this would completely change the tone and meaning of the document.
The document as it stands may put patients, orthodontists and even the AAO at risk. Recommendation 1.2 states “Use CBCT only when the clinical question for which imaging is required cannot be answered adequately by lower dose conventional dental radiography or alternate non-ionizing imaging modalities.” This recommendation assumes that conventional dental radiography is lower dose than CBCT and that it is equivalent in diagnostic ability. Adding up the number of conventional images used to evaluate the patient, the total dose is often higher than a CBCT evaluation. The most dangerous aspect of this statement is the assumption that conventional imaging is adequate. The literature shows that diagnostic sensitivity and specificity for many imaging functions such as localization of impacted canines, assessment of adjacent lateral incisor roots, determination of tooth tip, dentoalveolar bone dimensions, condyles, and airways is clearly superior with CBCT. This recommendation assumes that conventional techniques are adequate to fulfill the imaging goals. This may result in assumptions and mis-diagnosis leading to a particular course of treatment that is harmful to the patient.
One must question if some recommendations are necessary and appropriate for the AAO membership. Recommendation 1.1 “Base the decision to order a CBCT scan on the patient’s history, clinical examination, and the presence of an appropriate clinical condition and assure the benefits to diagnosis and/or the treatment plan outweigh the potential risks of exposure to radiation, especially in the case of a child or young adult.” This recommendation applies to conventional radiographic imaging, records and any other patient procedure. Do AAO members need to be reminded by the AAO/AAOMR not to malpractice? Similarly, Recommendation 1.5 “Do not perform a CBCT if only 2D projected images derived from CBCT are to be used for diagnostic purposes.” This
recommendation is not necessary given there is no evidence that AAO members are using CBCT in this fashion.
Use of the “Estimations of Relative Radiation Level Designations for Children and Adults for Orthodontic Imaging” in Table II is questionable. It is immediately obvious that the range is inappropriate as it extends from 0 to 30,000 μSv. Radiation levels with modern dental equipment are generally well below 500 μSv. On the lower end of the range, it is not possible to be at 0 since the normal background level is 8 μSv or according to the chart, one . The subsequent application of this information to produce Table 5 “Adult and Child Relative Radiation Level (§) and Selected Published Effective Doses (μSv) (ICRP, 2007) for Specific Equipment used in Various Radiographic Examinations in Orthodontics” and Table 6 “Examples of the Calculation of the Relative Radiation Level Associated with Specific Imaging Protocols used in Orthodontic Treatments” are equally questionable. Used in this context, the actual values can easily be misrepresented, particularly on the lower end of the scale. For example, a child eating a banana after playing a game of outdoor soccer would be ; or a mother and child taking a flight from Los Angeles to Paris would both be and , respectively and anyone living in Denver for a year would be . Conventional dental radiographic evaluation utilizing a full mouth series could be in the range of or in adults and in children. For reasons such as this, NCRP report No 91 established the concept of a Negligible Individual Risk Level (NIRL) which is regarded as trivial with respect to the risk of fatality associated with ordinary, normal societal activities and can, therefore, be dismissed from consideration”.
This document contains directives such as “use CBCT only when”, “do not perform a CBCT”, “CBCT is not indicated when…”etc. This language, while attempting to be authoritative, seemingly dictates orthodontists to practice and manage their patients in particular ways, thereby becoming an entity that influences the course of patient diagnosis and treatment. For example, recommendation 1.3 “Do not use CBCT solely to facilitate the placement of orthodontic appliances such as aligners and computer-bent wires or to produce virtual orthodontic models” dictates specific orthodontic procedures and failure to abide by it may be misconstrued as malpractice. Patients, practices and companies that utilize CBCT in this fashion stand to suffer because of this statement.
Professionals and Society Responsibility
We are by the nature of our professional status in society expected to consider all possibilities even if there are some rare occurrences without being bound by practice limiting restrictions which could in the end limit our potential for routine higher quality and more accurate and predictable results with less unexpected complications and prolonged treatments of questionable value. The choice of the word “recommendations” in the title of the draft document suggests “this is what you should do” as opposed to “guidelines” which leave latitude for clinical variation and consideration. Guidelines suggest “this is what you should consider” and leave a more logical vs. dictated path for the diagnosis, treatment planning & management of orthodontic patients.
The AAO/AAOMR Joint Task Force has been preparing their position paper for 2.5 years and has allowed less than 3 weeks for a response from the AAO membership (Scarfe, 2012). This short response time seems inadequate and should be extended if the Joint Task Force is serious about soliciting and considering feedback from the AAO membership and other interested parties.
A significant and favorable recommendation by the Task Force is to base imaging decisions on a clinical examination, patient history and the presence of an appropriate clinical condition that would likely have a positive benefit to risk ratio if imaged. The discussion about selected imaging methods would significantly improve in the position paper by including traditional imaging along the cone beam CT (CBCT). The position paper seemed to have a bias toward the use of conventional imaging methods, including panoramic, cephalometric, intra-oral imaging projections, as the default method without employing the same use criteria being proposed for CBCT. In the clinical orthodontic practice there is a spectrum of imaging protocols in use. At one end of the spectrum the clinician will request a full mouth survey of periapical images, vertex occlusal, panoramic and cephalometric projections for a baseline assessment. The use of non- CBCT imaging methods can result in a relatively high effective dose without the requisite benefits. The Task Force Position Paper employed evidence based assessment methods for CBCT and ignored a similar calibration of conventional imaging. Conventional CCD cephalometric systems do not have standardized projection geometry and if unaccounted for could result in clinically significant consequences (Chadwick et al, 2009).
The AAO/AAOMR draft document is lacking in many respects related to objectivity, balance, failure to apply the available evidence, and flawed recommendations. As is, it may be impossible for AAO members to follow the guidelines and may put them, their practices and AAO in difficult and if not impossible situations.
Chadwick JW, Prentice Rn, Major PW, Lam EW. Image distortion and magnification of 3 digital CCD cephalometric systems. Oral Surg Oral Med Oral Pathol Oral Radiol. 2009 Jan;107(1):105-12.
Claus EB, Calvocoressi L, Bondy ML, Schildkraut JM, Wiemels JL, Wrensch M. Dental x-rays and risk of meningioma. Cancer. 2012 Apr 10.
Fuhrmann RA, Bucker A, Diedrich PR. Furcation involvement: comparison of dental radiographs and HR-CT-slices in human specimens.
J Periodontal Res 1997;32:409 –18.
Lam E, Yang J. AAOMR Response to Recent Study on Dental X-ray Risks. 2012 www.aaomr.org/resource/resmgr/…/aaomr_response_to_study2.pdf
Lofthag-Hansen S, Huumonen S, Grondahl HG. Limited CBCT and intraoral radiography for the diagnosis of periapical pathology. Oral Surg Oral Med Oral Path Oral
Radiol Endod 2007;103:114-119.
Low KMT, Dula K, Bürgin W, von Arx T. Comparison of periapical radiography and limited cone-beam tomography in posterior maxillary
teeth referred for apical surgery. J Endod 2008;34:557–562.
Matherne RP, Angelopoulos C, Kulild JC, Tira D. Use of cone-beam computed tomography to identify root canal systems in vitro. J Endod 2008,34(1):87-89
Scarfe WC, Radiation risk in low-dose maxillofacial radiography. Oral Surg Oral Med Oral Pathol Oral Radiol. 2012 Sep; 114(3):277-80.