Which are best? Fixed or removable functional appliances….
Which are best? Fixed or removable functional appliances….
Over the past few months I have covered several interesting and clinically relevant systematic reviews. I was pleased to come across this new review on the effects of fixed versus removable function appliances published in the European Journal of Orthodontics. Our choice of functional appliances is important and I hope that this review blog post can provide us with some good clinical information.
It is now many years ago that we carried out a multicentre trial that compared the Herbst and the Twin Block appliance. This can be found here. I was also one of the operators in the study and I learnt a lot about using fixed functional appliances. My personal experience was that the Herbst appliance was very effective but the trade-off was that we had major problems with breakages. As a result, several years later, my preferred appliance is still the Twin Block. This is my personal opinion reinforced by the results of our trial, but what does this new review tell us?
EJO Advanced publication. DOI: 10.1093/ejo/cjv086
This is another publication from the team at Queen Mary University, Whitechapel, London, South of England. Unfortunately, it is behind the impenetrable paywall of the European Orthodontic Society…
This was a nicely written paper. I thought that the literature review was particularly good and I recommend that anyone on a training programme should read this as it provides a great outline of functional appliance treatment. In addition to outlining the various advantages and disadvantages of the types of functional appliances the authors also pointed out that we need to move away from purely morphological measurements and include patient perspectives. I have covered this subject previously another blog posts.
Their objective was to provide evidence on the efficacy of fixed and removable functional appliances in terms of both morphological and patient centred outcomes and I felt this was very important.
What did they do?
They carried out a systematic review to high standards and they clearly stated the PICO.
Participants: Children less than 16 years old with Class II malocclusion
Interventions: Any type of fixed functional appliance
Comparison: Any type of removable functional appliance
Outcome measures: The primary outcomes were measures of skeletal, dento alveolar and soft tissue correction.
This review was carried out to a high standard as they only included randomised or nonrandomised trials. All the studies had to be prospective. It is important to work to this high level of selection criteria because this reduces the chance of selection bias that is inherent in retrospective investigations.
After their screening and application of selection criteria they identified four studies. They evaluated these studies with the Cochrane Collaboration risk of bias tool and provided a large amount of information on potential bias. They concluded that three studies were at high risk of bias and one was unclear. The most common reason for this allocation was unclear randomisation and allocation concealment.
I find it interesting that one of the studies they classified as “high risk of bias” was the Herbst vs Twin Blocks study that I carried out. In this study we reported that we had a greater dropout rate in the Twin Block group than the Herbst group. As a result, the authors classified study as being at high risk of bias. This is the correct conclusion from applying the Cochrane tool. However, it is important that this does not diminish one conclusion from our study in which we found that there was better cooperation with the Herbst appliance than the Twin bBock. Putting this aside, I will never forgive the authors for this classification!
More seriously lets look at what they found.
What did they find?
Firstly, two of the studies compared with the Twin Block and the Herbst appliance while the other two compared the Activator with the Forsus and the TFBC appliance. It was not possible to combine the data into a meta analysis because of differences in measurement between the studies. They did provide a lot of detail of relevant cephalometric measures that showed the effect of the appliances on dento alveolar and skeletal measurements (which were small). They concluded that all the variations of functional appliance successfully reduced the overjet to normal limits. There were also minor skeletal changes but as these were not compared to an untreated control group. We cannot, therefore, make any conclusions on whether the appliances changed the skeletal pattern more than normal growth.
They also stated that they were disappointed at the limited use of patient centred measures. They drew attention to our study in which we measured breakages and patient acceptance of the appliance. They also pointed out that our findings would have been viewed very differently if we had drawn greater attention to them in our paper. I completely agree, if I had my time again I would have emphasised the patient centres measures more. For example we showed that the Twin Block had a greater negative affect on speech, sleep patterns and schoolwork and the length of the treatment was longer than with the Herbst. Yet there were greater breakages with the Herbst appliance and it was more expensive. This information should be very useful when we are deciding on potential treatments with our patients.
What did I think?
I feel that this review does add to our knowledge, despite the limitations of the number of papers. I can conclude the following.
- There is little difference in the dental and skeletal effects of fixed and removable functional appliances.
- Most of the correction of the overjet is by dento alveolar movement, but there is a small amount of skeletal change (1-2mm).
- There is greater co-operation with fixed functional appliances but this is not 100%. There is no such thing as non compliance orthodontic treatment!
- Only one study reported on patient centred outcome and these should be included in all trials in addition to some cephalometric and dental measurements.
I will still stick with the Twin Block because of the cost and the additional time that is needed to deal with breakages with a fixed functional appliance.
Pacha, M., Fleming, P., & Johal, A. (2015). A comparison of the efficacy of fixed versus removable functional appliances in children with Class II malocclusion: A systematic review The European Journal of Orthodontics DOI: 10.1093/ejo/cjv086
Emeritus Professor of Orthodontics, University of Manchester, UK.
Hi Kevin,
I enjoy your blog very much. Keep up the good evidence-based work!
Question: Was the 1-2 mm skeletal change measured in the mandible only, or does it take restraint of maxillary growth into account as well?
Thanks for the question, this was a combination of maxillary and mandibular changes. I worked this out by just looking at the data and did not do the calculations exactly
Dear Dr. Kevin O’Brien. I have several articles of you cited in my thesis. I did my Doctoral and Postdoctoral researches with Twin Block, Bionator, Herbst and MARA appliances. The best skeletal results were Twin Block, Bionator, MARA and Herbst, respectively. Therefore, if you want skeletal results, I believe the removable appliances are better than fixed. I treated the entire sample of Twin Block (21 patients), at Bauru Dental School, University of Sao Paulo, Bauru, Brazil mentored by Dr José Fernando Castanha Henriques and Donald Enlow. The Twin Block sample with an initial mean age of 10 years. Although the Twin Block is a small sample, the results were fantastic, with a harmonious and beautiful facial profile. This sample will be described in the book “Facial Beauty” to be published in Dr. Donald Enlow name,and I am his co-author. The worst skeletal results were with the Herbst appliance, in which the Class II malocclusion with mandibular retrognathism corrected by means of dental compensation, mainly by vestibular lower incisor inclination of 7 degrees. The biggest Herbst problem was the lack of the mandibular branch growth in height, which did not support the mandibular advancement and the mandible had clockwise rotation, even with great growth in mandibular length. I think if you have the patient compliance, you have best skeletal results with removable appliances.
HI Lucelma, thanks for the great comments. You draw attention to the potential loss of anterior anchorage that occurs with the Herbst, and it is nice to hear your good comments on the Twin Block. I will watch out for the publication of your cases in the the book.
Dear Dr. Kevin O’Brien. I have several articles of you cited in my thesis. I did my Doctoral and Postdoctoral researches with Twin Block, Bionator, Herbst and MARA appliances. The best skeletal results were Twin Block, Bionator, MARA and Herbst, respectively. Therefore, if you want skeletal results, I believe the removable appliances are better than fixed. I treated the entire sample of Twin Block (21 patients), at Bauru Dental School, University of Sao Paulo, Bauru, Brazil mentored by Dr José Fernando Castanha Henriques and Donald Enlow. The Twin Block sample with an initial mean age of 10 years. Although the Twin Block is a small sample, the results were fantastic, with a harmonious and beautiful facial profile. This sample will be described in the book “Facial Beauty” to be published in Dr. Donald Enlow name,and I am his co-author. The worst skeletal results were with the Herbst appliance, in which the Class II malocclusion with mandibular retrognathism corrected by means of dental compensation, mainly by vestibular lower incisor inclination of 7 degrees. The biggest Herbst problem was the lack of the mandibular branch growth in height, which did not support the mandibular advancement and the mandible had clockwise rotation, even with great growth in mandibular length. I think if you have the patient compliance, you have best skeletal results with removable appliances. Dr. David Normando erased what I write to you on Facebook Group, Ortodontia Brasileira, so I wrote here.
Playing ? advocate but your sticking to the Twin Block implies you’re a 1-appliance guy rather than selecting an appliance based upon your diagnosis and patient/family involvement. You also seem to base your dis-like of the Herbst upon your experience with a very specific, costly design while many designs are less involved and one could argue less prone to breakage? This could explain in part why in the USA the Forsus and Herbst are significantly more popular than the Twin Block (JCO surveys). Then there is always that other ‘removable functional’ appliance – elastics! Double, double, toil & trouble – just stirring the pot! ?
Hi Peter, thanks for the difficult question! I suppose that I stick with the Twin Block because a large amount of high quality research has been carried out with this appliance that shows that it works. It is also the appliance with which I am the most familiar and also pretty good with! So I am think that I am practicing evidence based care with the combination of research evidence, clinical opinion and patient input. But you are correct in that our study was based on a costly version of the Herbst and is now rather “old fashioned”. But I am still waiting for the study that shows me a success, co-operation and breakage rate for these new developments. What fixed functional do you use?
Thanks for your response Kevin and the question. I do much less early intervention than when I first graduated as most are treated in one phase later. In those later cases I use the above mentioned ‘elastics’ functional (sometimes with the dare I say it… headgear) or a Forsus FRD in larger or stubborn cases (child or biology). If doing an early intervention it is usually for reasons of appearance or to improve function (they are finding difficulty biting with their overjet) and sometimes to reduce a trauma risk. I will discuss with the child and parent the requirement for cooperation with a removable appliance (in my case a Twin Block usually in conjunction with a high-pull headgear) or a ‘non’-compliance appliance (in my case a CBJ Herbst or AdvanSync Herbst). Often the choice becomes clear as the child or parent indicates whether they think compliance or coping will be a potential issue. This doesn’t eliminate the problems but I feel reduces lack of compliance issues. They also know they can still treat later. The ones that may be more likely to try an early phase have a reasonable alignment in the 1st place and so that if they get a good treatment response/overjet correction, then there is a chance they may avoid the 2nd phase or at least have a fairly minor one, however they know this is not a guarantee.
Hi Peter, thanks for the reply. I have had many comments on this posting and I will have a close look at the evidence behind Forsus etc. But I may still end up with the Twin Block, because this is what I am used to using and I get good results without the hassle of multiple complex breakages. Like all things in life, I like to keep things simple!
Thank you Prof O’Brien for your analysis of the paper.
From the papers analysed in the systemic review Pacha, M., Fleming, P., & Johal, A. (2015), Baysal and Uysal (2014) found that there was more skeletal change in Twin block group than in Herbst group. From the result, there was very much skeletal contribution to overjet correction and molar correction than dental effect. However, from your previous study on Herbst and Twin block 2003, the effect from both appliance was come from dentoalveolar effect.
How should we undertstand the difference in the outcome with this two studies?
From what I anaylsed from two paper, it seems the treatment duration in Baysal and Uysal (2014) is a bit longer than your previous study. Could it be a possible reason for that?
Thank you so much! I really like your blog so much!