Is Class 2 correction with clear aligners disappointing?
Today’s post is about a very interesting study that examined the efficacy of Class II correction using clear aligners and intermaxillary elastics in adolescent patients. It is nice to write a blog post that is not about breathing.
We are all familiar with the use of clear aligners in adults; however, new research is now being published on their effectiveness in adolescent patients.
The authors of this study pointed out that little is known about the effectiveness of treating Class II malocclusion with clear aligners. This study should add to our knowledge.
A team from Adelaide, Australia, did this study. The American Journal of Orthodontics published the paper. It is open access.
Maurice Meade, Bariah Rans and Tony Weir
AJO-DDO advanced access: https://doi.org/10.1016/j.ajodo.2026.02.014
What did they ask?
They did this study to
“Assess the changes in overjet and the intermaxillary sagittal first permanent molar relationship in adolescent patients after treatment with an initial set of Invisalign aligners and Class II elastics.”
What did they do?
They obtained the sample for this study from the Australian Aligner Research Database. This database contains data on approximately 17,500 patients treated by 18 experienced orthodontists.
I have previously done blog posts about papers from this study team that use this database. It certainly seems to be a valuable source of information.
The inclusion criteria for patients to take part in this study were that they were under 18 years old with a Class II malocclusion, defined by the molar relationship. The orthodontists treated them with Invisalign only. They had to have full records.
The source of information was the STL models. They obtained the files for (1) pre-treatment, (2) planned movements, and (3) achieved outcomes after wear of the initial aligners.
The primary outcome measures were overjet and intermaxillary sagittal relationship.
They conducted a sample size calculation, which indicated they needed 64 patients. However, they decided to base their sample size on eligibility for inclusion in this study. As a result, they included information on 199 participants.
They performed standard univariate analyses to examine their data.
What did they find?
They included 199 patients in their final sample. 65.8% of these were female, and the mean age of the sample was 13.66 years.
I have extracted the mean and 95% CI data for overjet and first molar (FPM) variables (mm) into this table.
| Variable | Pretreatment | Planned | Achieved | Planned v achievied | Accuracy |
|---|---|---|---|---|---|
| Overjet | 5.11 (4.76-5.46) | 1.87 (1.78-1.95) | 4.02 (3.75-4.28) | p<0.01 | 33.6% |
| FPM relationship | 2.53 (2.36-2.70) | − 0.41(− 0.52 to − 0.31) | 1.18 (0.99-1.38) | p<0.01 | 45.91% |
At the end of treatment, 45.2% had an overjet greater than 4.1 mm. Importantly, only 33.6% of the planned overjet reduction was achieved. Interestingly, 31% of patients showed an increase in overjet, despite a planned reduction.
The greater the pre-treatment overjet, the larger the discrepancy between the planned and achieved changes. When assessing sagittal change, the overall accuracy of the planned sagittal change was 45.9%.
The final conclusions were
“Almost half of the participants had an overjet greater than 4.1 mm at the end of the initial series of aligners. Importantly, 31% showed an increase in overjet despite a reduction in overjet being planned.”
What did I think?
Regular readers of this blog will know that over several years, we have made a plea for increased research into the effectiveness of clear aligners. Ideally, this should be done with prospective cohort studies or randomised trials. Unfortunately, investigations of this nature are lacking. As a result, we have to rely on retrospective studies for information.
This team has conducted extensive research and published numerous papers using the Australian aligner database. This is a large database and appears to be a good source of information, albeit retrospective. When we consider the nature of this research, we must assume there is some selection bias in the sample; however, we do not know its direction. This is, to a degree, offset by the fact that the database is large and that all orthodontists are obliged to include their patients treated with clear aligners in it.
As a result, I think we can conclude that this is the best information we have on the use of clear aligners.
The results of this study are similar to those reported in previous papers by this group. It appears that there is a disparity between planned and achieved tooth movements with clear aligners. In this respect, the results suggest that aligners used in this way are not an effective method of class 2 correction. It certainly appears that functional appliances achieve better results. However, we also need to remember that there is a relatively high discontinuation rate with removable functional appliances. Unfortunately, there does not seem to be information on the discontinuation rate in this sample of patients. Therefore, this information is lacking.
Final thoughts.
I used and studied removable functional appliances for many years during my clinical practice. The results of this study would not persuade me to switch this group of patients to clear aligner treatment. However, some people may be excellent operators with better results than those reported in this paper. It would be good to hear their comments on this paper.
Now that you have read this blog post.
Last year, I requested donations to support my blog, and thankfully, I received an overwhelming response. With your help, I was able to raise sufficient funds to keep my blog going for another year.
Now, I’m seeking your support again to help me cover the blog’s running expenses for the upcoming year. The funds will cover web hosting, software upgrades, email lists and general running costs.
I believe that if every reader contributes a small amount, I can cover the yearly running costs. Your help will be much appreciated.
To make a donation, just use the form below. You can use the set-amount buttons or the free-field button to enter other amounts.
Then the easiest way to pay is to tap the Apple Pay button. If you want to use another payment method, click the “next” button and enter your details. This will take you to credit cards, PayPal, and other payment methods.
Emeritus Professor of Orthodontics, University of Manchester, UK.
I have used clear aligners very successfully for many years in the treatment of Class II malocclusions, and I’ve shared some of the more complex cases on my LinkedIn feed. As with all orthodontic treatment, it is the planning and biomechanics that ultimately determine the outcome. In my view, these results do not accurately reflect how efficient aligners can be in managing the vast majority of teen malocclusions—provided patients are compliant.
In my experience, compliance with elastics or aligner-based functional appliances is significantly better than with conventional functional appliances. While I agree that more high-quality scientific research is needed, I would recommend reviewing the work published by Sandra Tai. Her results show similar outcomes comparing aligners with conventional appliances but give better control of vertical dimensions and lower incisor proclination but I don’t have your expertise to critically evaluate the studies.
Hi, thanks for the comments. Can you let me have a reference for Sandra’s studies and I will have a look at them? Thanks
Thank you for the post, Kevin. Aiming for brevity will proceed in point form-
1. Aim of this study was as you stated: “To Assess the changes in overjet and the intermaxillary sagittal first permanent molar relationship in adolescent patients after treatment with an initial set of Invisalign aligners and Class II elastics.”
2. This is not an outcome study, treatment was incomplete at 10 months of treatment time, and documented in limitations section.
3. The data are insufficient to conclude treatment outcome and superiority of orthodontic treatment modality, and was recognized as such by the esteemed authors
4. This data set does not provide specific information regarding compliance or protocol with elastics (THE class II mechanical component), likely varied due to the retrospective data analysis, individual patient factors and the large number of clinicians (18); as pointed out by the authors.
5. In 2 of the staging modalities- jump and en masse retraction – that are identical in vivo , and reflected so in results – there will be no change but dental alignment- typically leading to overjet increase – without sufficient Class II elastic use.
6. It is probable that this fact may have contributed to the findings in 31.5% that overjet increased, as with aligner or fixed appliance mechanics alone, alignment often leads to increase in overjet.
7. It is only in “distalization” where the aligner is programmed to apply distal tooth moving pressure. It is worth noting that en masse retraction of anterior teeth is often staged as part of a distalization pattern in ClinCheck Pro Software; however from the authors description they are referring to en masse retraction as something different.
8. “An elastic jump shows the correction occurring in a single-stage movement. En masse retraction shows the sagittal change occurring continuously over the course of the CAT series. “ This is simply a choice in bite change visualization on the ClinCheck Software program, not an alternate biomechanical force system, and importantly one that will not express any Class II correction and likely increase overjet – unless a Class II vector is applied sufficiently
9. A total of 33.6% of planned overjet reduction was achieved. A total of 45.91% of the planned combined sagittal FPM was achieved. These are mean results with large standard deviations (an externally reported standard deviation (SD) of 2.0 mm for comparable sagittal and overjet measurements) ;at the 95% CI, and small changes programmed. As you have pointed out previously Kevin, percentages and grouping varied results to 1 mean figure are not necessarily individually clinically reflective and energy loss / under expression of activation is a finding in most if not all engineering systems / orthodontic appliances; the only difference is we can now measure it thanks to digital programming. Expressed in Fig 4: greater overjet, greater difference in planned vs achieved
– My interpretation, comments and questions would also include; how was compliance and elastic protocol retrospectively measured and recorded in this data set consisting of pre , interim and post treatment STL files – particularly when non compliance with elastics was an exclusion variable and critical for Class II correction? Can we identify and measure this effect as a significant variable with data available? The results suggest a probability of non-compliance and / or sub-sufficient Class II mechanics, resulting in mainly alignment. Interesting!
– After 10 months of fixed appliances you would not expect majority to have total resolution of overjet or Class II correction
– After 10 months of functional appliance therapy you may expect significant overjet reduction, minus the opportunity to simultaneously align and level the arch, increase comfort, hygiene and quality of life that Invisalign mechanics provide (anecdotally and in the literature)
*VV provides sponsored lectures for Align Technology
Thanks for declaring your conflict of interest. I take all your poins on board and agree with some of them. However this is currently the best informatoin that we have on aligners. This will continue to be the case until the aligner companies fund some high quality research. Perhaps you can let us know why they are not doing this?
Thanks for your comment. I am not criticizing the research diligence, just wary of conclusions we derive, pointed out below by TC. Brief communication with 1 of the authors “post your post” resonated – “aligners do not treat Class II – clinicians do”.
As to why aligner companies are not funding “high quality research”, I am unaware of any significant R&D efforts by other aligner companies.
Align Technology – as it is public knowledge with this US based publicly listed company – invests into R&D externally and internally, historically to a sum of over 1 billion US dollars – more than all other orthodontic companies combined. Align’s annual Research Award Program available in Americas, EMEA and APAC aimed at empowering our academic community has funded approximately $3.96 million in research since 2010. I would hope that all academics currently engaged would be aware of this opportunity. It is up to you whether you find these independently selected research projects “high quality”- that is a comment on our academic community rather than the sponsor – or whether you would ultimately dismiss findings in light of the requested company investment? Again, I would respectfully ask – what are the “high quality” research efforts conducted with alternate appliances so that we may simulate protocols with aligners? There are not many (any) prospective randomized clinical trials with sufficient patient numbers treated in the real world that we may use to fully endorse any particular mechanical regimen. That’s why we educate our students to continuously think critically as even the most traditional appliances in the hands of the most experienced clinicians are not a “slam dunk”.
In your table , could you please explain how the calculation works, I can’t figure that out.
This paper is interesting, but I would be very cautious about drawing strong conclusions from it.
A large retrospective registry does not become reliable simply because it is large. When 18 orthodontists contribute cases without a standardized Class II protocol, without a clearly controlled elastic protocol, and without robust compliance data, the study is no longer evaluating one treatment approach but a mixture of different operator-dependent approaches. In that setting, the reported “accuracy” risks being a pooled average of heterogeneous mechanics rather than a true measure of aligner efficacy.
Moreover, this was an assessment after the initial series, not a final outcome study. For that reason, I do not think these data can credibly support broad claims either for or against aligners in Class II correction. At best, they are hypothesis-generating. At worst, they create a false sense of precision and further weaken the scientific credibility of aligner orthodontics by presenting biased retrospective estimates as if they were robust evidence.
At some point, serious editors need to stop rewarding this formula. Retrospective registry papers built on uncontrolled heterogeneity should be published, if at all, as descriptive and hypothesis-generating observations, not as substantive evidence of efficacy or predictability. Otherwise, we are not advancing the science of aligner orthodontics; we are diluting it with the illusion of precision.
TC provides sponsored lectures for Align Technology
Thanks for declaring your conflict of interest. Yes, you are correct in stating that retrospective studies of this nature are not high levels of evidence. I have discussed this before in other posts about this registry. However, it currently is the best evidence that we have got. However, we must also ask why the aligner companies have not funded randomised trials in to their products. Perhaps, you can shed some light on this? Until then we need to accept these findings until higher level evidence is available.
Thank you, Kevin. I agree that these registry papers may currently be among the most available data we have, but I do not agree that this means we should simply accept their findings in any strong clinical sense. Weak evidence remains weak evidence.
Prospective work has in fact been done. For example, our own prospective study on treatment design was published in 2023, and your blog previously cited it when discussing the evidence on aligner tooth movement. That does not solve the evidence problem, of course, but it does show that prospective research is possible when investigators choose to build it.
I also agree that companies should be pushed much more strongly to fund proper prospective and randomized studies. But I would extend that point: researchers and editors should force that change as well. It is always easier, and far more comfortable, to continue mining retrospective registries than to demand better science. Convenience, however, should never be confused with methodological rigor.
So yes, let us push industry to support stronger trials. But until then, I think we should present these registry findings for what they really are: low-certainty, hypothesis-generating observations, not evidence that deserves clinical acceptance simply because higher-level evidence is still missing.
My concern is that repeated publication of this model of study, across many different topics, may create an illusion of scientific robustness rather than genuinely moving the field forward.
Kevin,
Thanks for posting this study. As you and others here have mentioned more work needs to be done, however, even this study is a start in the direction of clear answers for class II correction. So much controversy surrounds the entire process of class II correction around timing and mechanism already, even using fixed appliances like the Herbst, which advances the mandible 100% of the time during the active treatment, let alone removable appliances employing class II elastics or keyed advancement which works only when the patient’s mouth is closed. I use aligners a lot in my practice but if a patient is actively growing and has significant class II issues, I recommend fixed appliances including a fixed Herbst appliance first, simply for its predictability. I hope we get more definitive answers with this but I’m not particularly hopeful that that will transpire.
Interesting study, unsurprising findings, but a waste of time in my opinion. If fixed appliances can not grow mandibles, there’s no chance that a removable appliance will. Aligners are the most inefficient and ineffective force delivery system for anything other than tipping or molar intrusion in low angle cases. Given their extremely poor biomechanical characteristics, why would anyone (other than a KOL) even consider this inferior appliance for class II correction?