November 06, 2023

Which is better for Class II problems, Herbst or Invisalign Mandibular Advancement?

Researchers have studied the treatment of Class II malocclusion with randomised trials for many years. The current state of knowledge is that most treatment change is dental with a small skeletal component. While it may be disappointing that there is such low skeletal change, there is no doubt that functional appliances are effective. Furthermore, there seem to be minimal differences between the effects of our appliances.

Invisalign has introduced a new Class II corrector. They call this the mandibular advancement appliance (MA). There is very little high-quality research into this appliance. Recently, the AJO-DDO published this retrospective evaluation of the MA appliance. Despite its retrospective nature, I thought it was worth looking at this paper on what may be a development in our treatment options.

A study team of academic and specialist practitioners did the study. Sandra Tai is a member of the Align Tech Orthodontic Clinical Advisory Board for North America and Align Global Faculty. As a result, this is an expert team in aligner treatment.

No member of the team declared a conflict of interest.

What did they ask?

“This study aimed to compare the skeletal and dental changes in patients with Class II malocclusion treated with Invisalign MA and fixed Herbst appliances”.

What did they do?

They did a retrospective study of patients treated by Drs Tai and Andrews in their offices. They obtained the records of 140 patients with a Class II relationship treated with the MA appliance. In addition, they also identified 444 patients whom they treated with the Herbst appliance.

One investigator screened all the records to identify a sample of 69 records.

They then selected a final sample of patients who had skeletal Class II malocclusion and lateral cephalograms taken before treatment (T1), after the advancement phase (T2) and after the completion of the second phase of treatment (T3).

This resulted in a final sample of 20 Herbst and 20 MA patients. Unfortunately, this was 6.8% of the initial sample of Class II patients. I think that you know where I am going with this.

They traced the cephalograms and used a modified Pancherz analysis to identify treatment changes. Next, they analysed the data using multiple comparison tests between the interventions. As we have discussed, this leads to a high risk of false positive findings.

What did they find?

The mean duration of phase I treatment with the MA was 14.4 months, with a second phase of 17.9 months. They found similar data for the Herbst, with a mean phase I treatment time of 18.5 months, with phase II lasting 18.3 months, resulting in final treatment times of 32.2 and 36.8 months for the MA and Herbst appliances.

There were minimal differences between the groups at the start of treatment.

The team presented a large amount of data in complex cephalometric tables. They also concentrated on showing skeletal and dental change information. I found this rather difficult to understand and interpret. Unfortunately, I could not find much data on simple values, for example, pre and post-treatment overjet.

However, they did provide some helpful information in the form of classic Pancherz analysis diagrams and figures, and I found this helpful.

I will illustrate this by looking at the changes at the end of all treatments.


  • The total OJ correction was -4.8mm. This comprised -4.1mm skeletal and 0.7mm dental change.
  • The molar correction was -4.2mm, comprising -4.1 skeletal and -0.1mm dental change.

Mandibular advancement

  • Overjet change was -2.8mm, with -2.5mm skeletal and -0.3mm dental change
  • The molar change was -4mm, comprising -2.5mm skeletal and -1.5mm dental change.

Their final conclusions were

“Treatment with the Herbst resulted in greater change in overjet, molar relationship, Wits, and overbite correction compared with the MA appliance”.

“Both appliances can effectively reduce the overjet and overbite and correct the molar relationship in growing patients with Class II malocclusion”.

What did I think?

This was an interesting paper on a novel form of treatment that may have potential advantages as it uses clear aligner technology.

I was interested to see that the treatment changes were relatively small. Notably, the overjet correction was much less than that reported in the Herbst and Twin Block appliance clinical trials. For example, in our trial of the Twin Block vs the Herbst appliance, the mean overjet change was 6.3mm and 5.8mm, respectively. The treatment duration of phase I was 11 months for the Twin Block and 6 months for the Herbst. More recently, I have posted about another trial of this treatment. In this study, the overjet changes were 7.8mm for the Herbst and 5.8 mm for the Twin Block. The duration of phase I was also much smaller at 8.8 months for the Herbst and 10.3 months for the Twin Block. I could not think of any reasons for these marked differences from this new study.

The authors pointed out that a major shortcoming of their study was its retrospective nature. Importantly, we must consider that only 6.8% of possible patients were selected for the final sample. We must, therefore, consider that there is considerable selection bias in this sample.  As a result, I am afraid that I cannot agree with the conclusions that these respected investigators have made.

Final thoughts

Nevertheless, this study provides useful information that could be incorporated into a sample size calculation for a future trial. However, this study does not persuade me to use the MA appliance as an alternative to a Herbst or a Twin Block.

Related Posts

Have your say!

  1. I agree with you!

  2. I agree. I used the Twin Block appliance frequently to good effect. I never used a Herbst appliance as it seemed brutal!

  3. It is misleading to parents to suggest this treatment “grows” mandibles. At best it provides straighter teeth in dual bite positions- one simulating what jaw surgery would provide. If it’s not a true cure should a child’s consent fairly recommend waiting to decide on options when fully grown, especially if a true profile improvement is desired?

  4. “The current state of knowledge is that most treatment change is dental …”

    One would think I would know exactly what this means but I am unsure.

    Is there a TRIAL where the dentition is first aligned with those wonderfully wide arch forms, a Ceph taken, and then Class II corrective gadgetry is employed to get to the Class I ? Is this correction/dental_change dominantly a trained anterior bite taking advantage of the plasticity of centric occlusion? A significant tipping of the occlusal plane via TwinBlock, Herbst, Class II elastics, etc.? Compared with a Ceph after aligned arches, does the Ceph after Class II mechanics show something with the lower incisors in the symphysis translating or tipping around the incisor center of resistance in some fashion.

    For maxillary premolar extraction compensation of Class II occlusion, it takes me 9mnths to a year to close that premolar extraction space and get Class I canines. Hard for me to believe I really get whole arch translation or anything other than a significant component of dual bite after 6-9months of Class II aids Herbst, TwinBlock, AlignerAdvancement (and I use a lot of Class II aids).

    Thanks for helping me understand

  5. I have treated a number of growing patients with Invisalign MA well but I would say its not for everyone. I like that I get alignment at the same time but its never guaranteed that the BS will end up class I
    I also find that certain patients will not get a good correction with either MA or CTB – whether that is through poor compliance or other patient factors – they simply don’t get enough overjet correction
    I would really like to see some prospective studies now – surely there are enough user out there to gather this information!

  6. 1 possible reason for greater OJ reduction & faster treatment time in the Herbst group is the lack of need for compliance. Compliance is needed for the Twin Block & MA group.

  7. Kevin,
    Thank you once again for doing the leg work.
    Given recent studies that rate compliance with removable devices at 50% , I am always astonished that fixed functional growth appliances perform so poorly compared with removable appliances but perhaps the timing of the growth each day is the key factor and also the stage of the PGS.
    The structure of the face must also add to the confusion. A large mandible set back on the cranial base will respond differently than a small mandible set further forward so parameters measuring molar or incisal relationships perhaps do not reveal comparable samples.
    Having treated cases of twins where surgical costs dictated treatment selection for families, there is little doubt that surgical results if properly executed produce a better more stable and predictable outcome. We are more certain of targeting the skeletal than the dental.
    Our challenge is to compare “like with like “skeletally and to use sample sizes calculated to demonstrate true differences based solely of appliance efficacy whilst minimising confounding factors like compliance.
    The fact that so many Class 2 appliances produce an effect indicates that our understanding is not being improved despite this being one of the most extensively researched areas of orthodontics. Perhaps we are doing the same thing research wise and expecting the answer will drop into our laps rather than planning our research in a more targeted manner.
    Easy to say but hard to do.
    Thanks again , Kevin for your dedication. Keep up the good fight .

Leave a Reply

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