April 30, 2018

Does using cervical pull headgear improve unerupted canine position?

Cervical pull headgear is used to correct molar relationships. But does it influence the position of unerupted canines? This new paper looked at this clinical question.

The interception of potentially impacted permanent canines is one goal of early orthodontic treatment. The use of the early application of headgear has been looked at in several trials. Unfortunately, these had severe methodological problems and it is difficult to come to clear conclusions. As a result, I was very interested to see this new paper. A team from Norway and Finland did this study. The EJO published it.


Sigurd Hadler-Olsen et al

EJO on line DOI: 10.1093/ejo/cjy013


What did they ask?

They did this study to answer this question:

“Does headgear treatment in young children influence the eruption path of permanent maxillary canines”?

What did they do?

They took data from two randomised trials that looked at the effects of early headgear treatment on seven year old children. In the first study they evaluated the results of headgear treatment on 71 children. The second study was similar and they studied treatment for 67 participants.

The PICO was:

Participants: 7 year old children with Angle Class II malocclusion.

Intervention: Cervical pull headgear with a force of 400-700g worn 8-10 hours per day.

Comparision: No treatment

Outcome: Radiographic position of maxillary canines at the start and end of headgear treatment (when they obtained bilateral class I molars).

They randomised the children into the two groups at the start of treatment. They did not report the details of the trial in this paper. However, I have looked at the original papers and they gave no details of randomisation, sequence generation, concealment or blinding.

When I looked at the current paper, I could not find any information on blinding of data collection.

They carried out an appropriate and detailed statistical analysis.

What did they find?

They provided a large amount of data and I have summarised the main points:

The use of cervical pull headgear resulted in:

  • An increase in arch perimeter of 3mm (95% CI 1.13-4.89, p=0.002)
  • A more vertical and distal eruption of maxillary permanent canines, but the effect sizes were generally a few degrees and I do not think that they are clinically significant.


  • The canine angulation changed more if there was spacing in the arch.
  • When there was crowding the HG did not influence canine position.

They concluded;

“Headgear treatment influences the eruption pattern of maxillary canines, especially when there is spacing in the arches”.

What did I think?

Firstly, we need to consider that this is a new analysis of data from two previously published trials. Both trials were very similar and it may be logical to combine the data to increase the power of the studies. However, I decided to look at the original trial reports. The first was published in 2005 and I think that the second was reported in 2004 (but this was not clear). I found several problems. I would like to point out that this reflects the state of development of orthodontic trials over 10 years ago. As I mentioned before, they did not give sufficient details of randomisation, sequence generation, allocation concealment, consent and blinding. When we combine this with the lack of blinding of the recording of data in this paper, this means that the studies are at high risk of bias.

Furthermore, I think that we need to be cautious in the way that we interpret their results. In this respect, the authors are clear. They only measured the radiographic position of unerupted canines. This study was not about a treatment to intercept palatally impacted canines.

Therefore, we can only conclude that early headgear treatment may result in a small change in the eruption patterns of canines. This does not translate into less impactions.

This finding tends to agree with other studies that have looked at the effect of headgear on impacted canines. Unfortunately, these studies have methodological and reporting errors. As a result, we need to be cautious.

Final thoughts?

My overall feeling is that the current research suggests that there may be an influence of creating space on the eruption pattern of permanent canines. This may be particularly true for the extraction of primary canines. However, before we go fitting headgear to normalise canine position, we certainly need better studies.

I will continue extracting primary canines to attempt to intercept impacted canines.



Related Posts

Have your say!

  1. It seems almost impossible for me,thinking about the large variety of orthodondic appliances we have,that people continue nowadays to use headgear to move molars.We have a lot of wrong clinical implications:a heavy force applied discontinuously,orthopedic effect at that age,high patient compliance,maybe a distal tipping more than a real distalization of the roots of the molars,and so on.Why are we using this presidium yet?Thank you for your reply.

  2. Did you feel that the conclusion that the cervical head gear increased arch perimeter and got molars class I was an evidence based conclusion? Are there many papers that would demonstrate CPHG as a good way of fixing class II problems?

    Stephen Murray
    Swords Orthodontics

  3. Hi Kevin,
    As someone interested in impacted canines this article about canine angulation and headgears also caught my attention. I’m not sure this article warrants your high standards.
    I’m aware this article is not about impacted canines but just the eruption angle of permanent canines with headgear use. However, as you have pointed out about other articles on headgear and impacted canines they unfortunately, have severe methodological problems and it is difficult to come to clear conclusions. This article would fall into that category.
    If I’m not mistaken in the original studies for this article the headgear group and control group both were Class II with moderate crowding. But the interesting thing about these studies were they extracted primary canines in about +30% of the control group to “help” control crowding. Well, this would nullify any follow-up comparison of the angles of the permanent canines in this control group and therefore negate any conclusions. Would you agree?
    Congratulations on your nice blog. Keep giving us up-date information on impacted canines.

    • Hi John, and thank you for you interest in the article
      Just a few comments. The sample in this article is different from what you assume. I am sorry if that is unclear. The first RCT is published previously but not the second (yet). All cases with primary canine extractions were excluded in both groups (see subjects and methods) just as you say, because this would nullify the changes observed in canine angulation. You can also see the distribution of crowding/spacing in fig 4.

  4. Headgear —,!!are we still using that??? .I last used it at ortho.school 35 yrs .ago !Does anyone have any thoughts why we should use headgear when we have so many better tools now ??Are there some benefits of headgear use I am missing ??

  5. Head gear and patient safety . Dr Kevin what is your opinion about this.

    It might be at certain level, effective , but definitely not efficient .

    Would it maintain the harmony of skeletal development, this would be very crucial in growing patients with high angle MM?!

Leave a Reply

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