What do we know about Class II orthodontic treatment ? A new Cochrane Systematic Review
What do we know about Class II orthodontic treatment? A new Cochrane Systematic Review
This week on the Blog I am joined by Badri Thiruvenkatachari, who has worked with me for several years, and was the lead author on our updated Cochrane Systematic Review, on the treatment of Class II malocclusion, published last week. This can be accessed on http://goo.gl/puzQR5. This is one of the largest and most cited orthodontic Cochrane Reviews and was first published in 2007. One of the requirements of carrying out a review for Cochrane is that you have to commit to update the findings periodically. This means that as new knowledge from studies becomes available, this is included into the reviews. In effect, the review is a “living document” and the conclusions may change with time. This is one of the great advantages of Cochrane reviews compared to the more static conventional literature. The downside is that you are committed to updating your reviews for as long as you are working!
This update allowed us to include several new studies and evidence that has become available. It also resulted in several of the established studies being reassessed for the quality of evidence, and new statements made on the strength of evidence in the review.
So what did we find?
The review included 17 studies with data derived from 721 participants.
- Three trials compared early treatment with functional appliances (2 Phase) with treatment provided in adolescence only (1 Phase).
- Two trials compared early treatment with headgear to one phase adolescent care.
- Six trials compared different types of functional appliances when used in one phase adolescent treatment.
- Finally, one trial compared treatment with functional appliance against no treatment
What did we conclude?
I found it very interesting and important to find that adding newly available data resulted in a change in the conclusions from the first version of the review. This was relevant to the role of early Class II treatment in preventing incisal trauma Our overall conclusions were
“providing early orthodontic treatment for a child with Class II malocclusion is more effective in reducing the incidence of incisal trauma than providing one course of treatment in adolescence. There was no other advantage in providing early treatment”.
“When one course of treatment was provided in adolescence, no functional appliance was better than another. Any change in skeletal pattern when compared to a control was not clinically significant”.
It was also important to consider that the overall quality of evidence was low, apart from the findings on trauma, where the quality of evidence was moderate. This may be interpreted by considering confidence in the results. So where evidence is low quality, this may be defined as “further research is very likely to to have an important impact on our confidence in the findings’. When the level of evidence is moderate this means that “further research is likely to have an important effect on our confidence”. I have discussed this issue of confidence in a previous blog post. Sense about science: Dealing with uncertainty in orthodontic research.
How strong are the conclusions for Class II orthodontic treatment?
If I build on my post of last week. It is clear that when we consider most of the findings of this review, we must appreciate that the quality of the evidence is low because of bias in the studies. We also need to remember that Cochrane is pretty unforgiving in this assessment! The reasons for this classification are clearly stated in the review and I shall address this assessment in a future post.
Nevertheless, when we consider the important findings on trauma, we can have some confidence that this is a clinically important finding. At this point, we should examine the data concerned with the potential reduction in trauma. This data is shown in the paper in the summary of findings table 1 for the main outcomes. This reveals that 29% of patients with new trauma were in the 1 phase adolescent treatment group compared with only 20% of those patients receiving early treatment. The odds ratio was 0.59(CI 0.35 to 0.99). This is a clinically significant finding, but we need to appreciate that the CI is wide and almost contains 1. We also need to consider how to interpret the odds ratio. As with several statistical tests this is not straightforward and I had to look this up. I found this explained well in this blog http://goo.gl/lDWlI8. They explain odds ratio in this way…
“When you are interpreting an odds ratio (or any ratio for that matter), it is often helpful to look at how much it deviates from 1. So, for example, an odds ratio of 0.75 means that in one group the outcome is 25% less likely. An odds ratio of 1.33 means that in one group the outcome is 33% more likely.”
If we now look at the OR we found (0.59) this means that in the early treatment group (functional appliance) the chance of trauma was 41% less likely than for the group whose treatment was provided when they were in adolescence.
So what does this mean clinically?
Whenever, I give a course or speak to trainees, I stress that one we have read a paper we need to consider the “so what” question and whether we are going to change our practice based on the results of the study. It is clear from this review that moderate level of evidence suggests that providing early Class II treatment with functional appliances reduces incisal trauma. This means that when I see an 8 year old child with an increased overjet, I will explain to them that early treatment will result in a transient increase in their self esteem and that they will be 40% less likely to have trauma than if we waited to provide treatment when they are older. They can then decide. I suspect that we will be providing more early treatment…
Thiruvenkatachari B, Harrison JE, Worthington HV, O’Brien KD (2013). Orthodontic treatment for prominent upper front teeth (Class II malocclusion) in children Cochrane Database of Systematic Reviews (11) Other: CD003452.
Badri and Kevin,
Congratulations on producing this excellent update to an important review. I have a few of comments:
• I was interested to see you now conclude that early treatment of an increased overjet does reduce the incidence of incisal trauma. The odds ratio is 0.59, with an upper confidence limit of 0.99, so it is a statistically significant finding, but this is very close to the line of no difference! I think the relevant forest plot is Analysis 1.6, which shows that the included studies for this outcome are Florida 1998, North Carolina 2004 and UK (Mixed) 2009. The only new data from the first version of the review is the UK (Mixed) 2009 study, which found no significant difference in the incidence of trauma between the children treated early and those treated in adolescence. The only study to find a significant difference was North Carolina 2004 and I think they concluded that the trauma was generally minor (1 child needed an RCT) and the costs of treating all children with increased overjet early was greater than the cost of treating the trauma and may not be justified (pretty tough on the ones who need the RCT!). Perhaps we need more cost-effectiveness data on this outcome before changing our clinical practice?
• I notice that there is still a paucity of outcomes that are relevant to patients. Whether or not they are going to have a higher chance of incisal trauma is probably relevant, but I have yet to find a patient who is interested in their cephalometric data. We still do not know whether reduction in the overjet of children improves their social well-being.
• The UK (Mixed) 2009 collected data for self-concept (I think this is a slightly different psychological concept than self-esteem). According to the forest plot (bottom of Analysis 1.1) early treatment favoured the observation group – is this correct or is a higher self-concept score better than a lower self-concept score?
Anyway, congratulations again on a really useful summary of the literature in this area.
Is it really surprising that the greater the Overjet the greater the likelihood of physical trauma to anterior incisors that physically stick-out, especially in young children running around and growing/adapting to changes in limb-length etc?
Is it really surprising that self-esteem might be affected in children who, can be very cruel in comments from an early age, calling other kids ‘goofy’, ‘bugs-bunny’, ‘rat-face’ or more cruel things I won’t mention here 🙁
Of course Dentistry should be more about the ‘holistic’ approach to overall well being, but a clinician needs to assess that individually and directly – just because a ‘clinical trial’ has much greater difficulty measuring feelings or consequences socially (compared to a physical thing like OJ we can all agree is 10mm etc.) doesn’t mean we should shy away from addressing such difficulties.
Any Cochrane review is going to have a dilution-effect upon kids who really need this early intervention individually, but the Stats are masked by those who are deemed not to need/want/realise such benefits in such studies, thus hardly surprising the Odds ratio had a variance of 0.35 to 0.99, is it?
Of course keep looking for the scientific evidence qualitatively and quantitatively, if only to justify to NHS paymasters why something blatantly obvious to a ‘professional’, should be authorised to be funded by the controlling all-powerful and ‘unaccountables’ in NHS Management etc, but some things like ‘care’ and ‘psychological well being’ just are, by there very nature, difficult things to measure and quantify.
We should not stop trying to improve that and EBD is more than publications, but neither should our Professionalism and ‘care’ be limited by limited evidence (= mostly historic look backs anyway) or prevent real progress on the frontlines of healthcare.
That’s why we are (or should be) continually trained to be thinking, questioning professionals – made free to question and think and evolve holistically too!
Yours unlimited 😉 ,
Tony.
Thanks, you have nicely illustrated the interface between personal opinion and scientific evidence. While you state that some of the findings are not surprising and may be considered to be obvious, you have to consider whether they are supported by scientific evidence. The history of medical and dental care is littered with “innovations” that have been developed from personal opinion and when they have been investigated have been shown to have no effect or even harm. So while the factors of trauma and self esteem may seem obvious to you, they are not necessarilly correct. So while we accept that prominent incisors are susceptible to trauma, it is clear from the review that the benefits of treatment are not as clear cut as you think. The most important information to me was the number needed to treat. We have to decide is it worth the cost and risk of treating 10 children to prevent one episode of trauma. This is for you to decide as a clinician.
I agree that clinical trials should measure patients and parents values and indeed in orthodontics, we have to develop this methodology more. We are working closely with the COMET initiative on Core Outcome Sets that will change the variables that we measure and this will be a great development.
I do not think that a Cochrane review will have a dilution effect? If you mean that poor quality evidence is not considered and this is diluted, and the review provides high quality evidence, then I think that this is a good thing!
It’s always impressive when a person of stature, respected by their peers, reevaluates their position on a topic. Kevin, I admire the transparency of your clinical comment, “This means that when I see an 8 year old child with an increased overjet, I will explain to them that early treatment will result in a transient increase in their self esteem and that they will be 40% less likely to have trauma than if we waited to provide treatment when they are older. They can then decide. I suspect that we will be providing more early treatment.”
I found it interesting that the findings mentioned in the blog did not include that randomized clinical trials did show that the use functional appliances during adolescence did result in a greater skeletal effect compared to untreated controls. The systematic review mentions:
“Late orthodontic treatment for adolescents with functional appliances showed a statistically significant reduction in overjet of -5.22 mm (95% CI -6.51 to -3.93, P < 0.00001) and ANB of -2.37° (95% CI -3.01 to -1.74, P < 0.00001) when compared to no treatment (very low quality evidence)."
It may be low quality evidence but it also may be the best we've got.
I think we should read the full paper and not just depend on summaries that highlight a particular finding.
Thanks for pointing this out and I am sorry that I did not include this, but I try to highlight the significant findings in my posts. What we have to consider is whether a difference in ANB of 2.3 degrees with such wide confidence intervals is clinically significant. My feeling is that this is still rather small and is not clinically significant. However, it is also perfectly reasonable for you to consider that this is clinically significant. This is why we need to interpret the literature rather than just take the evidence from the conclusions of papers
hi prof,
firstly thanks for the this blog. if it weren’t for your comments i’d be totally suckered by all this “research”. Question:
does the reduction in overjet decrease trauma because the chin now gets in the way?
or is the actual inclination of the anterior teeth or position in space altered by use of the functional appliances?
thanks, chris
Thanks for your kind comments. It is difficult to find any evidence behind the reduction in trauma. I wonder if this is because the teeth are moved inside the upper lip and this provides some protection?
did you check the 2018 update bay Batista et al for this cochrane review? i wonder what is the “so what” we can conclude from it.. i was looking for an answer for coinciding with the PHV, and for the effect of functional appliance on patients past their PHV but still growing, would it be the same?