September 05, 2016

Early Class III protraction treatment works and reduces the need for orthognathic surgery: A trial that shows us something useful!

Early Class III protraction treatment works and reduces the need for orthognathic surgery:  A randomised trial that shows us something useful!

Once in a while a study team carries out a trial that provides us with very useful clinical information.  This post is about a paper that shows us that early Class III protraction facemask treatment reduces the need for orthognathic surgery.  All dentists and orthodontists should read this paper.

One of the main aims of early orthodontic treatment for 7-8 year old children is to reduce the complexity of any treatment that they may need later in their development. There have been several studies into early treatment for Class II malocclusion.  These have all shown that any benefits are somewhat limited.  However, until now we have had no strong evidence on class III treatment.  This paper is, therefore, not only interesting but important

Early class III protraction facemask treatment reduces the need for orthognathic surgery: a multi-centre, two-arm parallel randomized, controlled trial.

Nicky Mandall, Richard Cousley, Andrew DiBiase, Fiona Dyer, Simon Littlewood, Rye Mattick, Spencer J. Nute, Barbara Doherty, Nadia Stivaros, Ross McDowall, Inderjit Shargill & Helen V. Worthington

Journal of Orthodontics, 43:3, 164-175, DOI: 10.1080/14653125.2016.1201302

EnglandDr Nicky Mandall led this project.  I worked with her when she was a resident and then Senior Lecturer in our Department, they were great times for our orthodontic research. This paper represents 10 years of work by the investigators. The study was mostly based in the North of England,  although there are some people from the “beautiful South”.

They did this study to find out if “early” class III protraction facemask treatment influenced the eventual need for orthognathic surgery. This was a multicentre two armed randomised controlled trial that started in 2003 and this paper represents a long-term follow-up.

What did they do?

Participants were 7-9 year old children with a Class III malocclusion. This was defined as three or more incisors in cross bite with a clinically retrusive mid face.  They recruited the children through school screenings and treated them in 8 UK-based hospital orthodontic departments.

They randomly allocated the patients to receive either

  1. RME  and protraction headgear
  2. No intervention, this was the control group.

They followed a uniform protocol for the protraction headgear treatment.  This continued until the reverse overjet was corrected. Once they completed active treatment no patients received retention. Some of the patients had upper arch alignment when they were 12 to 13 years old.

The patients in the control group were simply followed and some of them also had upper arch alignment when they were 12  to 13 years old.

The primary outcome was the need for orthognathic surgery. They decided on this by using a panel consensus. The seven orthodontists met and reviewed the clinical records to decide whether each patient needed orthographic surgery. Research assistants blinded the records  so they did not know the treatment allocation.

Secondary outcomes were cephalometric variables and psychosocial status which they measured these with two recognised scales.

They carried out a sample size calculation and randomisation and concealment were good. The statistical analysis was appropriate.

What did they find?

They initially enrolled 73 participants and at the end of the study they analysed 32 in the control group and 33 in the protraction group. The groups were balanced at the end of the study. There were no differences between the groups for the patients who had received upper arch alignment.

When they looked at the need for orthographic surgery they found that in the control group 21 (66%)  needed surgery. Whereas, protraction treatment reduced this to 12 (36%). This was a clinically and statistically significant difference between the groups.

They calculated the odds ratio to be 3.34 (95% CI 1.21 to 9.24). This means that the use of protraction surgery reduced the odds of needing surgery by 3.5 times.  I calculated the numbers needed to treat (NNT) to be 3.6 (95% CI 1.9-21.6).  Meaning that we need to treat 3.6 patients to avoid one episode of orthognathic surgery.

It was very interesting that they did not find any difference in A-P  cephalometric measurements between the groups.  They also showed that there was some reduction in maxillary and mandibular rotations for the protraction group.

Finally, there were no difference in the socio-psychological measures.

Their discussion was good and clear.  When they looked at all their data they did not find any predictors for the need for orthognathic surgery.

What did I think?

I think that this paper could be a real “landmark” publication because it reported a good long-term trial that provided us with very relevant clinical outcomes. I was particularly pleased that the final outcome was relevant to both patients and operators and was not an indecipherable mass of  cephalometric data.

It was interesting and relevant that the treatment did not have an effect on socio-psychological status. This is in contrast to the studies that evaluated early treatment of class II malocclusion, which showed an interim boost in self-esteem. This may be because the class III incisor relationship is not as obvious to a lay population as very prominent incisors.   As a result, the children may not have been subject to teasing and bullying.

While I feel that this study is good, there is one point that we need to consider and this is concerned with the method of assessing the need for surgery. The operators in the study did this assessment. While they were blinded to the participant’s identities, there is a chance that they would recognise the patients who they treated. Nevertheless, because all the clinicians took a role in reaching a consensus, this effect should be minimised. As a result, I feel that this is likely to have minimal effect on the results.

In summary, this was an excellent study that the authors carried out reported very well. I’m sure that its findings will change practice.

Unfortunately, this paper is currently behind the Journal of Orthodontics pay wall, I have contacted the editor who has told me that it should have been published as open access and it will be available to everyone later this week.  This is great news.

Related Posts

Have your say!

  1. This study is really excellent and the editor’s decision is also excellent. I think if the similar study is planned in my country (Korea), it will be almost impossible to get a IRB permission because most of us believe early intervention is effective. Ethical problem will be the issue.

  2. Brilliant! Thanks for the summary. This will certainly be a landmark study.

  3. Dear Kevin
    I am really concerned about this paper.
    The start radiographs were taken in intercuspal position not retuded contact position and for me this invalidates the whole piece of research.
    Would you accept research on asymmetry if the start measurements were postured to one side?
    So consider the treatment group had the posture removed by pushing the incisors over the bite, hence the change in ANB. And the upper molars would then erupt hence the change in rotatation of the maxilla. The controls continued with a contact on the incisors and a forward and upward slide of the mandible ( I am sure you are familiar with the Gravelly paper which sugests the condyle remains in the fossa but the forward and upward movement makes the ANB ange 2.4 degrees more class 3) This means that when deciding if patients should have surgery the treatment group would be less likely to be offered surgery.
    Orthodontics has a 100 year history of claiming orthopaedic change and then finding it does not exist please let us have some ground rules. I suggest use Harvold’s point on anterior nasal spine not A point and only look at the mandible if the start radiographs are in Retruded Contact Position.

    • I haven’t yet read the actual paper, and you mention a fair point; however, perhaps the ‘take-home message’ here is simply that they were actually treated. I don’t know off the cuff how many Cl III 7-9 year olds have a significant functional shift? …in my practice it “feels” like about half as I write this, but I won’t go to publisher with that! So, to look at it another way, failure to treat early Cl III malocclusion, regardless of functional shift, seems to increase the likelihood for orthognathic surgery later. To your point, it could be interesting and enlightening to address the issue of mandibular position, but it doesn’t change how I personally respond to news of this publication, nor how I will continue to treat my young Cl III patients. Thanks!

    • Who cares about the ceph data really. Does it really matter if they were postured? Both groups were followed up and assessed as to whether they required orthognathic surgery at age 15. As a parent/patient if you can demonstrate the reduction in the need for orthognathic surgery by having 6-9 months of protraction facemask therapy at age 8 -10 I would want to know about that treatment modality and would like to give it a go for my child. As an orthodontist I would like to know the reason why this happened and this paper doesn’t really address it – but it’s primary outcome was not to determine how, but if the need for surgery was reduced. We had all the ceph tracing arguments for TwinBlocks about which lines and structures to use. Patient centred outcomes please, will my child be less likely to require orthognathic surgery if we give protraction face mask therapy a go at the right age – yes. Great trial.

  4. I am slightly confused with the findings and conclusions. They did not show any significant skeletal change in the long-term, yet it somehow reduced the need for orthognathic surgery. How is this possible? This need for surgery, as stated, was assessed subjectively and there must be a lot of bias here..Especially with no blinding.
    It is plausible that a positive overjet at the age of 7-9 could be achieved with a simple 2×4 appliance, an appliance that does not advertise itself as modifying growth and arguably is much kinder to the patient and simpler to use. If some of the subjects were treated this way and then the authors blinded to assess the need for surgery vs Class III protraction vs control, the results would be far more interesting. Ultimately, we cannot protract maxillae…

  5. Dear Prof. O’Brien

    Thank you for your review of this paper. For those who’ve been awaiting this expected evidence, you will now hopefully act on it in your daily practices; for those who for whatever reason are still uncomfortable treating in the 7-9 age range of the experimental and control cohorts, please consider referring or collaborating with colleagues who do indeed have training and experience in managing intrinsic and age-apprpriate anxiety behaviors of children in healthcare settings. And speaking of the control cohort subjects Dr. O’Brien that were ‘utilized’ for this trial; can you imagine how they and their parents will feel when they learn that the non-surgical Tx option for which they served as a baseline comparative sample, might actually have soared them the need for the surgeon’s hammer, chisel and steel blades, not to mention avoidance of risks associated with general anesthesia and post-operative complications. Given what evidence already existed before this trial was approved by an IRB about at least short-term efficacy of early expansion-protraction Tx, I am thinking the control group’s subjects might be feeling a bit shall we say, regretful? And one other thought if I may, for those of us who maybe have been ‘acting’ on the best available evidence (e.g., McNamara’s et al thin-plate spline analyses on expansion-protraction Tx’d pts., etc.), rather than ‘waiting’ for the bested expected evidence (e.g., Nicky Mandall’s masterpiece), one can only imagine how good the outcome might have been if at least some of these kids had been Dx’d and Tx’d earlier than age 7-9?

    Thanks again for getting this valuable article from behind JO’s ‘pay wall’….brilliant!

  6. I think that instead of dividing GB into North and Not North, you could adopt demarcations of North, Not North, and Not North Enough.

  7. I find very interesting this randomized study. I have not consulted individual study and your note did not mention it, over the age of 12 or 13 years is not the most appropriate to assess the need for surgical treatment in a subject as the jaw continues to grow especially in male subjects. Did the authors discuss this limitation? Did they use measures of bone maturity? On the need for surgical treatment, they discussed the use of some index? If the anteroposterior cephalometric data between both groups were not different at the end of the study, what biological change reduced the need for surgical treatment? Thank you.

  8. I am not sure of the significance of any findings here. I have successfully been doing non-surgical skeletal Class III corrections for over 30 years. Note: The earlier the treatment the better. Good myofunction, proper nasal breathing, proper tongue posture while in function (swallowing) or resting (placed at roof of mouth), proper lip seal are also very important. Although many doctors showed great results with reverse pull headgears when I first studied this in the early 1980s, I found that even this is not necessary if maxillary correction to ideal is completed. A great appliance for this is a properly designed Maxillary 3-way Sagittal (3 separate screws) appliance. Even many adults can be corrected when using SPE (Slow Palatal Expansion NOT RPE!). I give examples of this type of treatment in my book “Straight Talk about Crooked Teeth” available through I show long term stable results (one 17 years post treatment and just saw that same patient 23 years post op with no relapse). This is not rocket science folks, just logical physiology! Create an ideal maxilla, normalize the occlusion at any age with Functional Facial Orthopedics and braces where needed, make sure myofunction is corrected and voila, you create a stable life long result without surgery in over 95% of the cases. It has worked for me for over 30 years.

  9. Given that maxillary deficiency is the precursor to most malocclusions (if you’re willing to go with that), it would seem that this technique, or for that matter, ANY technique that can correct maxillary dysmorphia in all three planes of space would be commendable in any malocclusion. And given that there is more growth potential in those who have yet to grow (if you’re willing to go with that), then doing this treatment as early as possible would seem to be commendable as well. Why not development the maxilla so there will be plenty of room for the tongue on the palate and plenty of room for the teeth to erupt so that these quandaries about orthognathics need rarely be considered? You only have to be able to recognize a deficient maxilla.

    • I agree with Barry. The treatment is centred on correcting a deficient maxilla. Using this appliance where the mandible is too long would create a very unsightly bimaxillary prortusion. Harry Orton was very supportive of the use of reverse pull headgear at least 25 years ago.

  10. Many thanks, Kevin for negotiating this article’s free access.

    Larry White

  11. Dear Dr kevin
    Thank you very much for sharing your thoughts along with the review. I am concerned about the IRB permission for this study. There is evidence as to the improvement of skeletal class III if treated early (8 to 10 years). As age increases the skeletal effect reduces and dental change increases. Àlso there is reversal of growth pattern 2 years after protraction negating the correction achieved. The greatest problem in any clinical study in orthodontics is that we cannot use same treatment for a single patient and evaluate which is better. So there is no means to really ensure that, had the face mask treated group if treated with surgery would they have been more satisfied? I think this may not revolutionise our clinical decision making in early class III treatment, clinicians who are skeptical with face mask treatment will still remain same. But no IRB will permit and untreated control group in any further RCT in the light of this research for class III protraction treatment.

  12. Dear Prof. Kevin
    I also think that this is a good paper and an interesting and long- term study.I am glad about the editor’s decision to share the publication.
    We do a lot of early treatment in growing patients with very good results in Class III ( RME and Facemask)
    Congratulations for your blog, a good place where we can discuss orthodontic topics and put up with recents studies.

  13. It strikes me that a problem here is that the control group had no treatment (although some had braces later, how many?). This study therefore tells me that doing treatment to correct an anterior x bite is better than leaving it. What it doesn’t tell me is if doing an alternative to protraction is just as good, although it sort of implies that it might be since there was no A-P difference in the groups and therefore no skeletal effect from the protraction. Is this correct?
    What I need to know is if protraction is better than treating later with fixed braces. Can this be teased out of the data? Those in the control group that had braces, were they the ones that were assessed as not needing surgery? Should they be in the control group if they had treatment anyway?

  14. Thanks dear prof O’Brien for sharing this important recent piece of information. I only got one comment, how did they managed to gain the ethical approval? Is it ethical to randomize patients not to have an interceptive that may work for them even according to the best available evidence before this trial?

  15. Dear Prof. O’Brien-Thank you for the followup on your recent review of your former student’s recently published study. To your query, ‘Was the study ethical?’, and your own conclusion, ‘I am sure that it was.’ by reason of the fact that the operators in the study had been in ‘equipoise about the treatment’ (i.e., they didn’t know which was the best treatment prior to trial’s onset in 2003.), and also to your further statement, ‘It was clear that when this study started in 2003, the operators did not know the best treatment. ….and as a result, the trial was ethical.’, Dr. O’Brien, as you are well aware, per several historically documented medical atrocities (e.g., the cruel/inhumane medical experiments exposed at the Nuremberg Trials of WW II, the U.S.’ Tuskegee Experiment, etc.), in order to gain ethics committee approval, amongst other factors, a thorough review of pertinent peer-reviewed literature is a mandatory requirement for any trial proposing to request utilizing human subjects. That said, prior to 2003, there were indeed scientifically credible published reports in top-tier journals describing the beneficial morphological effects of reverse pull headgear (RPH) Tx (a.k.a., ‘non-surgical maxillary distraction’) with or without concurrent RPE….e.g., here are just a few: 1.) Angle Orthod. 1996; 66(5):351-62; Changes following the use of protraction headgear for early correction of Class III malocclusion., Chong YH1, Ive JC, Artun; 2.) J. Am J Orthod Dentofacial Orthop 2000;118:55-62; Effective treatment plan for maxillary protraction: Is the bone age useful to determine the treatment plan?; 3.) Am J Orthod Dentofacial Orthop 1998;113:333-43; Skeletal effects of early treatment of Class III malocclusion with maxillary expansion without RPH, Tiziano Baccetti, DDS, PhDa, Jean S. McGill, DDS, MSb, Lorenzo Franchi, DDS, PhDc, James A. McNamara Jr., DDS, PhDd, Isabella Tollaro, MD, DDS). And furthermore, given what Linder-Aronson described in the 1970’s about correlation between posterior naso-pharyngeal constriction in children and increased risk for pediatric OSA (‘Naso-respiratory function and craniofacial growth’: in published proceedings of symposium honoring Professor Robert E. Moyers, February 23 and 24, 1979, in Ann Arbor, Michigan, and several published journal articles), combined with what Bacetti et al (Eur J Orthod, 21 (1999), pp. 275–281; Thin-plate spline analysis of treatment effects of rapid maxillary expansion and face mask therapy in early Class III malocclusionsT Baccetti, L Franchi, JA McNamara),revealed about how RPH-RPE combination Tx increased the nasopharyngeal corridor, I’d think the issue of equipoise possibly be called into question given the reason for the operators’ unawareness (i.e., incomplete review of the literature)…..who knows. But, as you conclude, now that validated published evidence is even more easily accessed/accessible (than in 2003 I’d admit), to withhold RPH and/or RPE Tx for research purposes from a child who might surely otherwise acquire better health outcomes, would not only be deemed unethical, but also be correctly categorized as medically and scientifically indefensible.

    And finally, per you posed question, ‘How do the patients feel about the results of the trial?’, you stated that, ‘Someone suggested that they (untreated control cohort subjects) maybe unhappy because the chance of them needing orthognathic surgery is higher than if they had been treated.’ I was at least one of the clinicians who made that observation (also Dr. Elbe Peter?), and I’m disappointed that, in spite of the (maybe not so easy to readily access) previously published evidence to the contrary, you stated, ‘ no-one knew that the protraction headgear would be effective. This is why they did the study’; but I am encouraged that you concluded, ‘now (that) the results are available practice should change. Therefore, the participants have helped improve treatment for others with their condition. Thanks again for starting this vital conversation Dr. O’Brien

  16. Dear Prof. O’Brien. Thank you for this compelling write-up.
    I have a query regarding class III management which is not related to your blog (If you find time please answer). We have been taught that cross-bite should be treated as soon as it is seen. Also, we know, skeletal class III due to prognathic mandible should be treated after the growth is completed (either by surgery or orthognathic camouflage).
    My question is: How should we manage a child who has an anterior crossbite due to prognathic mandible? Should we treat the child or wait till the growth is completed?
    I look forward to hearing from you.


  17. Dear Dr. O’Brien, I am a recently retired Pediatric Dentist in Texas. I have been providing this service for twenty five to thirty years for children as young as 3. It has been incredibly effective for the many patients that I have treated and extremely emotionally rewarding for me. Thank you for getting this information out to a greater number of practitioners.

Leave a Reply

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