Invisalign does not move teeth as effectively as fixed appliances?

Post by 
Kevin O’Brien
ore evidence is starting to emerge on the effectiveness of clear aligner treatment. While case reports show very well-treated cases, I cannot help feeling that these are cherry-picked and wonder whether aligners are not as good as some claim. While this may be the cynic in me, there is little evidence to suggest an alternative viewpoint. This new paper in the AJO-DDO concludes that fixed appliance treatment is more effective than Invisalign. It is worth a very close look.

What did they ask?

They did the study to:

‘Compare maxillary tooth movement between Invisalign and fixed appliances”.

What did they do?

A retrospective study of case records selected from the Graduate clinic archive.  The main inclusion criteria were.

  • Graduate students and Faculty treated patients older than 16 years old
  • Treatment was done on a non-extraction basis.
  • Complete records were available.

They included the records of 30 patients treated with Invisalign (mean age 34.5 +/- 14.5 years) and 30 patients (mean age 28 +/- 12.0 years).

The outcome measures were the PAR Index.  They also measured tooth movement from the digital models superimposed on the palatal rugae.  When they had superimposed the sequential models, they segmented individual teeth. Then they calculated individual tooth movement using specific landmarks within the teeth.

Finally, they compared the tooth movements between the two groups by using the relevant linear statistical models.

What did they find?

They produced a large amount of data in the text and graphs. Unfortunately, this was very detailed and needed to be more explicit. Consequently, I hope I have clearly interpreted this information through overloaded brain fog.

Firstly, there were no pre-treatment differences between the groups in PAR scores. Unfortunately, I could not find any information on the post-treatment PAR scores in the text. Nevertheless, in the abstract, the authors stated that the post-treatment results were similar.  However, we have no information on the overall quality or duration of the two treatments.

When they looked at the maxillary tooth movement, they found that there was less tooth movement with Invisalign than with fixed appliances. However, most of these differences were not clinically significant. For example, the crown translation movement of the upper central incisor with Invisalign was significantly lower than with fixed appliances, -0.53 mm.

Nevertheless, some other measurements were clinically significant. For example, the torque movement with Invisalign for the central incisor was significantly lower than for fixed appliances (mean difference -5.69 degrees p<00001).

Their overall conclusion was:

“When comparing fixed appliance to Invisalign, we found that patients with fixed appliance treatment had significantly more maxillary tooth movement in directions”.

What did I think?

It is great to see that more research is being published in peer-reviewed journals.  We are now beginning to build an evidence base for this treatment. I have thought carefully about the recent publications on aligners.

In summary, the evolution of research into this treatment follows the same path as much orthodontic research.  This is characterised by the early papers being case reports that may contain helpful information. This is then followed by retrospective studies that provide us with further knowledge. Nevertheless, we need to appreciate that this method of research is biased.  The final stages, in the progression, are randomised trials and systematic reviews.  These tend to show more minor effects than retrospective investigations. This paper represents the middle stage of the research evolution. We need to bear this in mind when we critically appraise it.


My general feeling is that this paper does provide us with some helpful information. Nevertheless, we must consider that several significant issues result in high uncertainty. Therefore, I will try to be brief:

  • This is a retrospective sample of cases treated in one dental school. Notably, the records were selected based on their availability. Unfortunately, this means the sample must have considerable selection bias.
  • The sample was tiny and there was no sample size calculation.
  • We must consider why the groups of patients were treated differently in the absence of randomisation.

These points are critical because the differences between the treatments may have occurred due to the operators using fixed appliances because they felt they needed more significant tooth movement.

The good points of the study were that the measurement method of tooth movement was novel and accurate.

It was also important that they did not measure the difference between the tooth movement and Clinchecks.  This has been done in other studies and while it provides information, it does not consider the variation in Clinchecks due to operator prescription etc.  This has been highlighted in other posts on this blog.

Final comments
We need to think about where this study leaves us.  At the most superficial level, it reinforces the general feeling that clear aligner therapy is effective for mild malocclusions. However, the other claims that are being made need further research before we can accept them.