April 02, 2024

Can RME change childhood behaviour?

It seems that the amount of research on the relationship between orthodontics and breathing is growing. Although this is an encouraging development, we must exercise caution when making over optimistic inferences based on research findings. This new paper highlighted a study that examined a crucial question on the effects of RME.  But are the conclusions valid?  I have had a close look at this paper.

Sleep-disordered breathing is a common and significant issue, with primary snoring (PS) being the mildest form and obstructive sleep apnea (OSA) being the most severe. The primary treatment for OSA is the removal of the adenoids. However, some suggest that Rapid Maxillary Expansion (RME) treatment could be an effective treatment for persistent OSA. Studies have shown that RME can increase the size of the oropharynx, making breathing easier. However, these studies were retrospective or poorly controlled. No studies have yet examined the impact of RME on childhood behavior. Recently, this study examined this interesting area.

A team from Sao Paulo, Brazil, did this study. The AJO-DDO published the paper.

What did they ask?

The researchers wrote a clear literature review on SDB (Sleep-Disordered Breathing) and the effects of RME (Rapid Maxillary Expansion). However, upon closer inspection, I discovered that most of the papers included in the review were retrospective studies. Additionally, the systematic reviews that were conducted also included a considerable number of retrospective studies. This indicates that the current evidence supporting this topic is relatively weak.

They did the study to ask

“What are the improvement effects of RME on behavioural and cognitive functions in children with SDB refractory to long-term adenotonsillectomy”?

What did they do?

To answer their question, they did a prospective cohort study with the following design


Children who had residual snoring following two years after adenotonsillectomy (AT).


RME with a hyrax screw.


There was no control or comparator


Orthodontic measurements, polysomnography, laryngeal naso endoscopy. Obstructive Sleep Apnea Quality of life questionnaire, the obstructive pediatric sleep questionnaire and other neurocognitive and behavioural tests.

The main criteria for the children included in the study were that they had undergone adenotonsillectomy (AT) within the preceding four years, and they had exhibited snoring symptoms at least five times per week. The team members conducted an examination of the pharyngeal tonsils and recorded the degree of obstruction.

The patients who were selected for RME (Rapid Maxillary Expansion) had certain morphological features such as narrow maxilla, high palate, posterior crossbites, decreased maxillary width, maxillary arch length discrepancy and transverse skeletal discrepancy. However, it was not clear if all these features were present in every participant or inclusion just required a single feature.

After this screening, the team identified 24 patients. They then divided them according to the Obstructive apnea-hypopnea index values as follows.

PS Group OAHI <1 and the OSA group OHAI >1. They analysed these groups separately and together.

An orthodontist, who was not aware of the research, treated children with a hyrax expander RME until the maxillary molar palatal cusp made contact with the mandibular molars buccal cusp. The appliance was kept in place for six months. The team gathered their data at the beginning of the study and six months after they had achieved the necessary expansion.

What did they find?

The study sample was 24 children (66% were males) aged 6.1-12.7 years with a mean age of 10.0 (SD=1.8 years).  

They presented a large amount of data for the two groups and the individual components of the questionnaires. I found this somewhat confusing, and I hope I have identified the most important variables. I think that these were the main findings after RME treatment.

  • There was no difference in polysomnography findings. 
  • They found significant differences in the PSQ scores for drowsiness, snoring, and behaviour.
  • Quality of life improved. 
  • There were no differences in cognitive function.

I could not find any orthodontic data or information on the degree of obstruction.

The conclusion was.

“RME might prove to be an alternative treatment for children with SDB refractory to adenoidectomy”.  

However, they also stated

“The non-controlled design was a major limitation of the study”.

What did I think?

The good points of this study were that the sample of patients was carefully selected, treated, and followed up. It was great to see that the team used relevant outcomes for patients and did not limit their outcomes to morphological outcomes that may not affect their patients’ health.

It is noteworthy that the treatment did not impact the patient’s PSG, indicating that it did not affect their Sleep-Disordered Breathing (SDB). Nevertheless, changes were observed in the psychological measures. However, before we jump to conclusions, we must consider various factors. These are.

  • The patients knew they were in a study, which may have influenced their responses to the questionnaires. This is called participation bias.
  • We are not sure if these differences are clinically significant.
  • The most significant limitation of the study is the lack of a control group, which means it is impossible to determine whether the observed effects were due to the passage of time or typical development. The authors have acknowledged this limitation in their discussion and added a comment to that effect.

“A larger sample size with a control group is needed to prove the claims”. 

Final comments

Frequent readers of this blog are aware that we usually do not review scientific research that has major methodological issues. However, I had to examine this paper in detail to address these problems. There is a concern that advocates of orthodontics as a cure for breathing problems may seize on this paper and make claims that the study does not support, because of the lack of a control group. This could lead to another cycle of “whack a mole” about these claims, which would be preferable to avoid. Nevertheless, this study is another step in the right direction for research in this important clinical area.

Related Posts

Have your say!

  1. Hi Kevin. Timely post on the day in Australia when the Invisalign Palatal Expander was launched. Only a few of us will get it over the next 2 weeks before it goes nationwide. Keen to see how well this works.
    Daily change with 0.25mm expansion per day. AUD$1095.

  2. Dear Prof O’Brien

    Thank you for once again drawing attention to the problem of non self-correcting pediatric maxillary skeletal transverse deficiency and SRBD co-morbidity.

    Per your observation of the authors’ self-admitted protocol shortcomings, e.g., lack of a non-treated control, etc.(‘ The most significant limitation of the study is the lack of a control group….’), will you please offer a suggestion/outline as to how future investigators might re-design this protocol so as to ethically establish a control cohort through the deliberate withholding of therapeutic RPE treatment from a group of 10(+/-) yr olds whom all would meet disease (MSTD-SRBD co-morbidity after A/T surgery) inclusion criteria?

    Thank you for considering

    Kevin Boyd

    • Thanks for the question. There are two main strategies that can be considered. The first is having a control group in which RME could be delayed. This would be ethical because we are still uncertain whether RME is an effective treatment. In fact ,we do not really know if RME does harm when it is provided and is not needed.

      We used this form of control group in the early Class II treatment studies in the USA and UK. This has also been done in a study into early Class III treatment. As long as the operators are in equipoise about the effects of the interventions then this is ethical. This concept is the basis of most randomised trials.

      Another option that provides us with some information but increases the risk of bias is to run the questionnaires in a group of children who do not have maxillary constriction. This would provide some information on the effects of time on the results of the questionnaires. However, my preference would be for a delayed treatment control.

  3. I suspect RME is an effective treatment for OSA (need more evidence to prove so). Should RME be proven to be effective one day, the next question to ask: ‘is it the gold standard or first line treatment’? For instance, CPAP is still the gold standard for adult OSA. We don’t prescribe oral appliances or MMA surgery as the first line or gold standard for adult OSA.

  4. You seem to use the statement “No studies have yet examined” a fair bit, but no studies have yet to examine the effect of premolar extractions on the airway or the health implications later in life….yet orthos still extract and retract without hesitation in 25% of cases. You should use your platform for advocate for victims. I can say with 100% certainty, extraction retraction has negative implications in 90% of cases.

Leave a Reply

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