July 10, 2017

Careful with that Cone Beam: It does not seem to make a difference?

Careful with that Cone Beam: It does not seem to make a difference?

I think that Cone Beam CT images are fantastic and they seem to make my diagnosis so much easier and accurate.  But, is this really true? This new paper in the EJO provides an interesting perspective….

The impact of Cone Beam CT on financial costs and orthodontists’ treatment decisions in the management of maxillary canines with eruption disturbance

Helena Christell et al

EJO: Advanced access DOI: https://doi.org/10.1093/ejo/cjx039

A team from Malmo, Sweden and Manchester, North of England did this study.

The introduction of Cone Beam CT technology has the potential to change orthodontic practice because of the clear images. However, we do not really know if the additional radiation and costs are completely justified.  Importantly, we need to know if the introduction of the technology leads to changed outcomes or decisions.  This is the question that the investigators asked in this interesting study on treatment decisions for maxillary canines with eruption problems.

They wanted to find out;

  1. If there was any difference in clinical decisions based on conventional images compared with CBCT images.
  2. The cost of producing different treatment plans.
What did they do?

They ran a web based survey of treatment decisions.  This had the following components:

The patient records

They selected the records of 12 different patients who had unilateral eruption disturbance of a maxillary canine. Importantly, they did not have any other reasons for orthodontic treatment.

The cases had two intra oral periapical radiographs at different angles, panoramic radiographs and relevant Cone Beam CT images.

The clinical participants

They emailed all members of the Swedish orthodontic society and asked if they wanted to take part in the study.  112 out of 314 agreed to take part.

The assessments

They sent a link to the survey to the orthodontists. They allocated each orthodontist to four of the twelve cases.  The orthodontists reviewed these records in two stages.  At stage one, they viewed two cases with intra oral and panoramic radiographs (M1) and two cases with CBCT and panoramic radiographs (M2).  They then repeated the exercise at least two weeks later, but this time there was a switch in the presentation of the images.

This meant that the same case was repeated with different images with a period of at least two weeks between viewings.

They asked the orthodontists to make a treatment decision for the cases, using the following options

  1. No treatment,
  2. Non extraction orthodontic treatment,
  3.  First premolar extraction orthodontic treatment,
  4.  Maxillary lateral incisor extraction orthodontic treatment
  5. Orthodontic treatment including extraction of the maxillary canine with eruption disturbance.
  6. They also recorded their confidence in their decision using an anchored five point scale.

They calculated the relative costs of the imaging techniques.


They calculated the agreement between and within the orthodontists and any differences in treatment decisions when using the different imaging techniques.

What did they find?

112 orthodontists took part in the survey.

When they looked at the comparison between imaging methods, they found that the range of  agreement in treatment decisions varied from 100% in one case to 12.8% in another.

The most common decision taken by the orthodontists was orthodontic treatment with no extractions for 66% of the assessments based on M1 and 64% based on M2.  That is overall the CBCT made no difference.

There was only one case in which the treatment plan changed (79% of the time) with the CBCT. This was patient with resorption of a lateral incisor.

Importantly, overall 107 out of 445 (24%) treatment decisions were different when based on M2 instead of M1 records.  They also showed no difference in confidence between the decisions taken with the different imaging techniques.

The mean cost of M1 was €99 and €134 for M2.  They calculated that the additional cost per treatment decision, that was different as a result of using CBCT was €143.88.

Their overall conclusion was;

“The results do not support the routine use of CBCT for patients with maxillary canines with eruption disturbances”.

What did I think?

This was a complex study and I hope that I have interpreted it correctly.

I will start with the good points of study.  I thought that they did the study well.   They designed the “experiment” and it was logical.  They selected a wide range of cases that represented common clinical problems.  The sample sizes were clear and they based them on previous data. Importantly, the sample of raters were all specialist orthodontists and this gave the study external validity.

If I am being very critical, I feel that this study, like many others, was a “still life” study and did not include the opportunity to examine the patient. Furthermore, the gap of two weeks between viewings was rather short. This may may have resulted in the orthodontists remembering the decisions that they took on the first viewing.  However, we need to consider that we are not simply looking at repeatability, but we are assessing the influence of additional information. As a result, even if the orthodontists remembered their first decision, it could still change with the additional information.

I also thought it was interesting that the case with the greatest number of changed decisions was one with resorption of the lateral incisor. It appeared that the addition of the CBCT resulted in better visualisation of the resorption. While this may be used as an argument to take more CBCT images. We need to consider the incidence of lateral incisor resorption. I am unsure that we know this value and so routine CBCT imaging may not be needed to detect what is a rare occurrence. Nevertheless, this needs investigating in future studies. Perhaps these could be directed at finding out how many CBCTs need to be taken to change one “incorrect” treatment decision?

Finally, we need to remember that these results are very similar to other studies. These all show that the addition of CBCT does not fundamentally change treatment decisions. As a result, this study adds to evidence in this important area.

What about their conclusions?

I thought that their conclusion was very relevant to clinical  practice. Importantly, they carefully stated that “routine CBCT” was not indicated.  I agree with this and we should only increase the radiation dose for our patients when we are convinced that additional information is essential.  I am sure that a good orthodontist can diagnose most problems clinically with lower dose 2D radiographs. The discussion is going to be interesting on this post..

Related Posts

Have your say!

  1. Making a decision of doing or not doing the CBCT just because it costs extra money is rather strange to me when the cost of this tools is less than may be a 5% of the total orthodontic treatment.

  2. Thank you for the comment on this paper. As radiologist, I will read the paper to look for more information about how the cbct images were presented to the clinicians: do they had access to the viewer? Or they were presented with a series of images, and in such case, who directed the images. Thanks again for bringing attention on this report.

  3. There appears to be a focus on the additional cost to the patient for this imaging modality. For most of us that have CBCT in our offices this is not an issue as we absorb the cost and don’t charge extra for it. There may be some additional radiation exposure. When you factor in the exposure for a pan, ceph and multiple dental x-rays I don’t find that the additional radiation is significant.

  4. Surely this suggest that most cases would be fine with conventional radiographs, but any case with suspected resorbtion would be better with a CBCT too?

  5. Kevin, I have not read the paper but it raises several questions in my mind. The first and probably most important is what settings were used and on what generation machines that led to the comment/conclusion in favor of using “lower dose” traditional 2D imaging. In Ludlow’s recent AjO article presenting dosage with the icat flex the results showed the flex CBCT had equivalent or lower doses than traditional 2D methods. Given that I agree we should use the lowest dose to gather the desired radiographic information, it would seem that the above article has it backwards and a current generation CBCT should be the preferred imaging.

    The second is the relevance of the cost. In the US at least, very few orthodontists charge for records, and fees are fairly standard in regions independent of the imaging used. It would seem then that the cost-benefit is a personal business decision for the orthodontist not an additional cost passed on to the patient. Depending on the size of the practice, these costs are obviously amortizatized differently so the cost per patient in one office may be quite different than another.

    One additional comment is that we also have numerous other 3D technologies rapidly becoming more prevalent. It is becoming faster and easier to merge these various 3D images giving us the capability to have a platform to generate large cross-practice databases and really evaluate treatments in ways we never could in the past.

  6. Thanks Kevin for summarizing this study. What I would like to add is that associated costs vary between countries and in our case the difference would be significant larger that the 150.00 pounds reported. Secondly the imaging parameters utilized could also affect the degree of definition of the bone, cementum, PDL areas. Finally, the CBCT may provide the oral surgeon exposing the canine a better visualization of how much and were to remove bone to expose the impacted canine.

    • Thanks dott. O’brein: this is one of the most interesting topic I’ve read so far in your blog!
      Routinely CBCT seems not to be always necessary for taking treatment decisions when approaching a case of canine impaction, e.g. extraction vs non-extraction. As a matter of fact, the article states that orthodontic treatment plan was little influenced by CBCT. What can be said about its role on treatment outcomes? We can have concerns about the usefulness of CBCT during the preliminary stage of orthodontic treatment planning; nevertheless CBCT has revealed to be often essential in the following stage of developing an appropriate surgical and orthodontic strategy. Oral surgeon can understand previously the best way to expose the canine and orthodontist can easily find the most effective mechanics. This can maximize functional and aesthetic outcomes, which is an important issue in the management of canine impaction.

  7. Interesting. Addition of CBCT may not influence the “how to treat” decision, but does it affect treatment outcomes?

  8. I feel with the low dose CBCT, ie… ICAT FLX this is an historical issue. I have had one since they came out and I am able to get a full scan (with about 1000 times more information) for half the dosage of a digital panorex. Why in the world would I expose them to twice the dosage and take a panorex?? I totally agree with the older high dose machines. If you are interested in airway and treating from that perspective a CBCT is very valuable.

  9. Perhaps for Otho it’s not needed. I assume you take a pan and a ceph. For implant dentistry, I beg to differ.

  10. Thank you very much for you blog post it always gives great insight. I do believe that your question has already been answered in the literature. “While this may be used as an argument to take more CBCT images. We need to consider the incidence of lateral incisor resorption.” the incidence of lateral incisor root resorption in cases with ectopic canines in 50%, Ericson and Kural 2000 in the Angle Orthodntisr. http://www.angle.org/doi/abs/10.1043/0003-3219(2000)070%3C0415:ROIAEE%3E2.0.CO;2
    so if this is met common reason that changed the treatment plan then I do believe that this a 50:50 in any patient with impacted canines and so it is a very strong argument for taking CBCTs and to opposite.

    Thanks again
    Nour Tarraf

  11. Hi Kevin

    Very interesting. I’m sure we are picking up more cases of root resorption using cone beams. Personally I wouldn’t be without mine. I have many cases where the diagnosis was uncertain with standard radiographs but the CBCT has confirmed what the problems are.


  12. Kevin, though we have never met, you have my respect as an orthodontic colleague and thinker. In full disclosure, I am an independent private practicing orthodontist who lectures on many orthodontic topics and I do receive an honorarium from manufacturers who make CBCT for my time. But I make more money in the office seeing patients, so lecturing actually costs me $$. I am in complete disagreement with the manner and title of which you present this study: Careful with that.”Cone Beam: It does not seem to make a difference?” Frankly, that is an irresponsible and dangerous statement which can be easily taken out of context. It is misleading. You are referring specifically to the use of CBCT in impacted canines and you should say so in your title. That discussion is over 15 years old. There are so many more applications (that is a different story). This is more than just a difference in opinion. The study is poorly designed and your interpretation is factually incorrect and dangerous to the public. It is a step backward in orthodontic progress. Firstly, the question the researchers ask is silly, almost archaic in today’s technology. How experienced are the user’s with CBCT? Are dynamic models from CBCT being provided? The researchers are asking the wrong question. Do we want to resolve this 3D problem using distorted 2D flat imaging or more anatomically truthful 3D imaging (I’ll get to the cost and radiation risk below)? Do we want to guess or do we want to know the exact location of an impacted canine? Do you prefer 2D or 3D information in helping better understand, communicate and mitigate the risk of treatment? I think you know the answer. Management of impacted canines are one of the most challenging and litigious orthodontic procedures we do. Impacted canines represent the first applications, almost the reason why CBCT was created and utilized by early orthodontic adopters. It has salvaged many cases and improved treatment of so many impacted cases and made life easier for orthodontists and most importantly patients. I can’t believe you are still even discussing this in 2017. The conversation is really outdated and you can go back to the literature in 2001 to see the evidence (Mah et. al and numerous more). There is no controversy about it. 2D imaging shows damage to adjacent lateral incisors in palatally impacted cases 12% of the time. 3D imaging showed the damage to be 66% of the time. 11% had damage to the centrals ( Three-dimensional localization of maxillary canines with cone-beam computed tomography. Am J Orthod Dentofacial Orthop, 2005, Volume 128, Issue 4, Pages 418-423L. Walker, R. Enciso, J. Mah) . So it is a really big deal. Furthermore let’s talk about dose. Again we are in 2017 not 2001. Not all manufacturers or settings are the same and we can’t believe anything the sales people say anyway. John Ludlow, one the world authorities on dental dosimetry from University of North Carolina has independently validated that a certain machine from a certain manufacturer, using the ultra low dose setting can take a 3D ceph size as low as 11-18 microseverts (Assessment of phantom dosimetry and image quality of i-CAT FLX cone-beam computed tomography, Am J Orthod Dentofacial Orthop 2013;144:802-17, Ludlow et. al.). A digital pano is about 25 microseverts, according to Ludlow. Taking 2D is actually more expensive than a CBCT if used properly. So actually the use of 2D radiography is overexposing the patient and with inferior diagnostic information under certain settings with certain machines! It’s not so easy to just write off CBCT is it? And what about the cost of NOT taking a CBCT on impacted canines. In the US it would be considered standard of care for impacted canines. Good luck defending a lawsuit on one of those gone south without a CBCT!

    • I reviewed this paper because it looked interesting and dealt with an important contemporary issue. This was simply my interpretation of the study and it is up to other readers to consider whether the findings are relevant to their clinical practice. I thought that within the constraints of the methodology the paper had value and I do not think that I was irresponsible in my interpretation. We should all read and interpret papers using our own scientific knowledge and experience. The important point here is that the authors did not suggest that CBCT should not be used for canines, but they simply suggested that it should not be routine. This is an important point that seems to get lost in the discussion.

  13. Thanks for the article, which I enjoyed

    I think this article and many like it could be reverse engineered by yourself to work out what would be required to make a clinical difference and make an intervention worthwhile.

    Let’s assume there should be some consensus on scientific method, and if there isn’t let’s assume that the readers of this blog trust you.

    We can assume that technology changes, usually along the lines of better/cheaper/faster.

    If that technology progresses to a point where there’s no great down side to doing it, then why not do it if there is some benefit to it?

    The question is, where is that point?

    For radiographs and CBCT, eventually (if not already) someone will have a CBCT machine that has less dose for the relative area than the average dose of the OPT machines in service. What is that dose and when we get there, is there any argument about not taking CBCTs instead of OPTs?

    I presume there are still people using OPT on film, never mind 20 year old digital ones, and I don’t expect the X-ray police to arrest them. Sooner or later a CBCT will out perform this equipment at no extra risk to the patient.

    Stephen Murray
    Swords Orthodontics

Leave a Reply

Your email address will not be published. Required fields are marked *