An occasionally irregular blog about orthodontics

The effectiveness of the Herbst appliance for patients with Class II malocclusion: A systematic review

By on November 9, 2015 in Recent posts, Research Methods with 1 Comment
The effectiveness of the Herbst appliance for patients with Class II malocclusion: A systematic review

Is the Herbst appliance effective?

This is certainly the season for systematic reviews and last week I was I came across this one on the treatment of Class II malocclusion using the Herbst appliance.  I have always been interested in this appliance and we carried out a large randomised trial that evaluated the relative effects of the Herbst and Twin Block appliances in the UK in the mid 1990s. We concluded that both appliances were equally effective, but there was better co-operation with the Herbst, but this was offset by increased cost and appliance breakages.

I decided to review this paper because it appeared interesting, but I discovered significant flaws. As a result, I am going to use this paper to identify some information on the Herbst appliance and  illustrate an important concept about systematic reviews.

537c0a6131e2410a8232168217df9a07The effectiveness of the Herbst appliance for patients with Class II malocclusion: a meta-analysis

Yang, X et al. European Journal of Orthodontics: Advance access. Do: 10.1093/ejo/

What did they do?

In this systematic review they attempted to investigate the effect of the Herbst appliance on Class II malocclusion. They adopted standardised systematic review methods and outlined their search adequately. It is important to point out that they aimed to include data from randomised trials and controlled clinical trials.

They carried out a large amount of work identifying studies, extracting data and grading the quality of the studies. They concluded that the use of the Herbst appliance resulted in correction of the dental malocclusion and that there was some skeletal change in the region of 1.5 mm in mandibular growth. They also identified similarly small changes in ANB, SNB and SNA.  I have included this data in this table below.  Note that the 95% confidence intervals are high for all measurements.  This represents a degree of uncertainty in the data.

Difference (means, 95% CI)-1.08 (-2.16, 0.00)-0.52 (-0.73, -0.3)1.45 (0.43, 2.47)-4.82 (-5.83, -3.8)

What did I think?

One of the main points that I have made about systematic reviews is that if they are to be of value it is essential that the review only includes data from high-quality studies. This is because if the review includes from studies that are not trials then they are at risk of coming to incorrect conclusions. When I first scanned this review, I thought that the authors had only included prospective trials. But then I looked more closely and found some problems with the studies that they reviewed.

At this point I would like to remind readers of several definitions on systematic reviews. I have based these on the Cochrane Collaboration handbook and other sources.

  • A trial is a study that can be included in a systematic review if the participants were assigned prospectively to one of two (or more) alternative forms of health care using random allocation or some quasi-random method of allocation (such as alternation, date of birth, or case record number).
  • If the author(s) state explicitly that the participants were randomly allocated to treatments, then the trial is classified as a RCT (randomized controlled trial).
  • If it is not clear that the particpants were randomized, but randomization cannot be ruled out, the report is classified as a CCT (controlled clinical trial).
  • The classification CCT is also applied to quasi-randomized studies, where the method of allocation is known but is not considered strictly random. Examples of this are alternation, date of birth, and medical record number.
  • The bottom line here is that to be included in a quality systematic review the study needs to be prospective. The authors of this paper clearly stated in the methods section that this was the aim of their review.

Unfortunately, when I looked closely at the papers, I felt that seven of the included studies were retrospective. For example, the papers by McNamara, Pancherz (x2), Franchi, LaHaye, Sidh and Wigal and are without any doubt retrospective studies. I am not really sure how the authors classified these as prospective trials. This is really important because retrospective studies provide us with low levels of evidence. I have discussed this previously in this post.

These issues with the included papers are pointed out by the authors when they evaluated the quality of the evidence with the GRADE assessment. They stated that there were serious limitations for all the measurements that they evaluated and the overall quality of the papers were low or very low. Yet, the authors concluded that the “Herbst appliance can improve sagittal inter maxillary positions, probably by restricting maxillae and encouraging forward mandibular growth”.  This conclusion is not really supported by the data or the quality of the papers that they included.

As a result, I feel that this review does not really satisfy the criteria of a high quality systematic review. This is a shame, as the authors carried out a lot of hard work. The paper is also published in the high-quality journal and maybe the editorial board need to evaluate their criteria for publishing systematic reviews. I am sorry to be so critical.


What can we learn from this paper? Firstly, it illustrates that we need to look quite closely at the papers that are included in a systematic review.It appears to me that the systematic review paper is in danger of becoming a ‘quick’ publication. This is similar to the large number of papers published on cephalometric values in the 1980s. I cannot help feeling that the editors of journals need to look at these more closely.

Secondly, if we consider that this paper is potentially biased towards positive effects of treatment because of the inclusion of retrospective studies. This illustrates that the skeletal effects of the Herbst appliance are minimal and most of the treatment effects are dent-alveolar. Nevertheless, I would like to point out that this is a tentative conclusion because of the quality of this review.

(Visited 4,858 times, 1 visits today)

Tags: , , , , , ,

There is 1 Comment

Trackback URL | Comments RSS Feed

  1. I agree that we seem to rely a bit too much on systematic reviews to get published. These reviews of other people’s research seem to become more and more prevalent in our journal. Does not mean they are not valuable but I believe orthodontic scientists should promote quality prospective studies especially to really find out what these appliances really do to the craniofacial complex of our patients. We have been using them for over 40 years after all.

Post a Comment

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


Pin It on Pinterest