Does mode of breathing influence facial growth?
Does breathing influence facial growth?
This is my second post in the series on breathing and orthodontics and is concerned with the influence of breathing on facial growth. Many claims are made about this subject, but how closely have we looked at the evidence? I have had a good look at this, and this is my summary.
As I discussed last week, there is increasing interest in correcting children’s breathing with orthodontic treatment. But where is the evidence that supports this form of therapy? An excellent place to start is to consider whether breathing problems influence facial growth? Firstly, let’s look at some of the statements that are made;
Page and Mahoney state;
“Nasal obstruction, in particular, is a key villain and cause of abnormal growth and development of the face, jaws and dentition”. This is illustrated, in the magazine article, by a picture of the child with the classical “adenoid facies”.
Others are not sure definite, for example, Kate Vig in the American Journal of Orthodontics writes;
“The form function interaction that conveniently should explain the causal association between nasal obstruction and facial growth in children appears to be of a multifactorial nature”.
I have always found this to be a somewhat confusing area. It is also difficult to review because the literature is historic and challenging to interpret. I have decided to approach this by critically evaluating the papers that are frequently quoted by those who promote the theory that disordered breathing influences facial growth. So this selection of documents is from their publications and other reading lists that they produce.
In the results section, they stated that there were significant differences between the two groups of monkeys. In brief, they found a significantly larger increase of the gonial angle and of the angle between sella-nasion and the lower border of the mandible in the experimental group. However, they did not present any data, and we have no indication of the magnitude of this difference. They illustrated this concept by showing some of the tracings. Overall, they concluded
“the steep mandibular plane angle and increased face height can be viewed as a result of an induced neuromuscular imbalance between the neck and facial muscles in the experimental monkeys”.
What did I think?
Firstly, I feel that I should show the respect that is due to the early orthodontic research workers and we need to consider that historical research papers do not always stand up to contemporary scrutiny concerning research methods. However, if we apply our current understanding of levels of evidence, we can only conclude that this paper is at the level of an interesting case report. Furthermore, I cannot help feeling that it is difficult to apply the findings to humans. This is because we do not really know if the results of these animal experiments are relevant to the facial growth of contemporary children. Also, the experimental monkeys had complete nasal occlusion, and this is unlikely to happen to children.
In summary, all that I can conclude is that this was an unusual case report and the results cannot really be applied to facial growth in children.
The other classic work is that carried out by Linder-Aronson and Woodside and I will revisit a post that I did some time ago.
Linder-Aronson, S., Woodside, D. G., & Lundström, A. (1986).
American Journal of Orthodontics. doi:10.1016/0002-9416(86)90049-7
They carried out this study to evaluate the effect of adenoidectomy on a group of children with severe nasopharyngeal obstruction.
They selected a group of 81 Swedish children who had undergone adenoidectomies to relieve severely obstructed nasopharyngeal airways.
48 of these children changed from mouth to nose breathing, and this group formed their initial sample. They then followed 38 of these children for five years. This was their final sample.
All the children had the following records collected;
- Study models
- Lateral cephalometric and PA radiographs
- Airflow measurements recorded by rhinometry
At the following stages;
- T0 Before operation
- T1 1 year post-op
T3 5 years post-op
What did they find?
- Girls who had adenoidectomies had more horizontal growth than the girls in the control group. The difference in the position of gnathion was 11.5 degrees, and the 95% Ci was 2.25-20.75 (which is very wide)
- This difference was not found for the boys
- Growth direction was more variable for the adenoidectomy group than the control group
The mean airflow through the nose increased for the adenoidectomy group
They then revisited this data five years later and published another paper using a different cephalometric analysis, but this time they wanted to measure mandibular and maxillary growth.
They divided the sample up by gender and analysed these group differently. When they looked at maxillary growth in boys, the mean difference between the adenoidectomy and control group was 1.2mm (95% CI 0.00-2.4), there was no difference for the girls. Analysis of the data for the mandibular length revealed that for the boys the mean difference was 3.8 mm(95% CI 1.72-5.8) and for the girls, it was 2.5mm (0.79-4.2).
I think that these effect sizes are rather small and the 95% confidence intervals are large. This means that there is a large amount of uncertainty in the data and is a reflection of the small sample size.
“the establishment of nasal respiration with severe nasopharyngeal obstruction affects the amount of mandibular growth.”
This dataset was then revisited again when Mahoney et al., published a paper, 14 years later, using the same sample. This time they analysed the incisor and molar dento-alveolar heights and upper and lower anterior facial heights.
Mahoney et al.,
Aust Orthod J 2004; 20: 93–98
What did they find?
To be honest, I was not sure what they found because they did not present the data to show the differences between the adenoidectomy and control groups. I have looked closely at their results and calculated the difference between the control and the adenoidectomy group for Lower Anterior Facial Height. The difference between the two groups at the start of the study was 0.9mm (0.12-1.92) end of the observation period was 1.4mm (95% CI 0.03-2.8). I suggest that this is not a clinically significant difference and it has a very wide CI.
Nevertheless, they concluded that
“Before adenoidectomy mouth breathing children had significantly longer anterior facial heights than the control children and the strength of the difference lowered over time”.
The data does not support this conclusion.
What did I think?
I have looked carefully at these papers based on this classic sample of children, and there are several issues that we need to consider;
- This sample was derived from children with severe nasopharyngeal obstruction, and the findings are only valid for this severity of disordered breathing.
- The cephalometric differences that they detected were minimal and were not clinically significant.
- The sample had severe attrition bias
- They analysed the boys separately from girls, and I wonder if they had combined the two groups the differences would not be there?
- If you analyse multiple cephalometric variables several times, you will find something just by chance!
My first feeling on looking at these papers is one of disappointment that these papers did not really stand up to contemporary scrutiny.
I think that what we are currently seeing, from those who are promoting “curing disordered breathing” with orthodontics, is a classic non-critical selective quoting of historical data. This adds to the smoke and mirrors.
My only conclusion from these papers is that there is minimal evidence that mode of breathing has an effect on the craniofacial skeleton and facial growth.
I know that my opinion may be considered to be somewhat controversial. If you do not agree, have a good look at the papers and put your interpretation forward on the comments section of the blog. We can then have a chat about all this.
Emeritus Professor of Orthodontics, University of Manchester, UK.