January 27, 2020

Does RME reduce bedwetting?

There is an increasing interest in maxillary expansion for disordered breathing. It has been suggested that bedwetting is related to upper airway obstruction. As a result, expansion may be a potential treatment for bedwetting (nocturnal enuresis). This interesting new trial was done to look at this question. A team from Uppsala, Sweden did this study. The Angle Orthodontist published the paper.

Rapid maxillary expansion in children with nocturnal enuresis: A randomized placebo-controlled trial
Ingrid Jonson Ring
Angle Orthodontist: Online: DOI: 10.2319/031819-219.1

What did they ask?

They did this trial to answer this question

“Was RME a useful treatment for bedwetting and can any prognostic variables be identified”?

What did they do?

They did a parallel-group randomised controlled trial with a 1:1 treatment allocation. The PICO was:

Children who were younger than 14 years with Nocturnal Enuresis (NE) with at least 7 wet nights out of 14. Most of the participants were non-responders to first-line treatments. They were classified as therapy-resistant. A paediatrician assessed all the children.
Tooth borne RME with a Hyrax screw designed to generate 0.5mm expansion per day. The endpoint of treatment was when the palatal cusp of the upper first molar came into contact with the buccal cusp of the lower first molar.
They fitted a placebo appliance identical to the RME appliance except it did not generate any expansion.
The primary outcome was the number of wet nights. Secondary outcomes were measures of expansion.

They took dental casts at the start of the expansion and after 6 months.

They used pre-determined computer-generated randomisation, with good concealment using remote allocation. This separated the recruitment and randomisation processes. The patients were blinded to the allocation. Their statistics were linear modelling and this was relevant.

What did they find?

They randomised 18 patients to each group. The mean age of the participants was 10.2 years in the control group and 10.3 years in the intervention group. Two dropped out of the control group.

In the expansion group, the mean expansion was 5.7 mm at the molars.

I have extracted the data on the number of wet nights into this table.

Number of wet nights over two weeks.

RME ControlRME Vs Control
T0 (mean and SD)T1 (mean and SD)Difference (95% CI)T0 (mean and SD)T1 mean and SDDifference (95% CI)Difference (95% CI)
11.4 (2.9)
9.2 (4.0)
(-3.7 to-0.8)
12.4 (2.3)11.7 (2.5)-0.6
(-2. to 0-0.8)
(-3.6 to 0.4)

In summary, when they looked at the treatment group in isolation there was a statistically significant reduction in NE. They did not find this for the control group. However, when they compared the RME and the control group they found no significant differences in the episodes of NE between the groups.

Their overall conclusions were:

“RME has a modest effect on children with therapy-resistant nocturnal enuresis”.

What did I think?

I was very interested in this study because it was concerned with an outcome measure that is important to our patients. It was also great to see that they did not get bogged down in a mass of cephalometrics and dental measurements. They carried out and reported the study well and the paper was very clear. It was also great that they used a placebo as control and this meant that they had taken into the account any placebo effect. They clearly could not have left a group totally untreated. I liked the trial.

However, I hope that I have not looked at their data too critically. But, my interpretation is different from the authors. Their statistical analysis shows that there is no difference in NE between the RME and the control group.  As a result, I cannot see how they could conclude that there was a modest effect of RME. While I concede that this may be my reliance on statistics and I may have missed something. I would be interested to hear what other readers, or the authors, think about the conclusions?

Related Posts

Have your say!

  1. Avatar

    This points out the danger of only reading article summaries – which is what most reader do.
    Thank you for your anaylsis.

  2. Avatar

    I think this may be an example of not accepting the rule of the game. A decision was made to follow statistical principles and if data does not support statistical significance then clinical relevance should not even be discussed. Such difference in this case seems to be around half an event every two weeks. That would be for most of us considered hardly meaningful. To present it as a modest improvement could be questioned.

  3. Avatar

    This is an interesting study, but appears to lack rigor in inclusion and exclusion criteria. For example, it says, “None of the children in the current study sample suffered from upper airway obstructions”. If so, what was the rationale of RME? It would’ve been interesting to assess the upper airways also. Moreover, it would have been ideal if the atrial natriuretic peptide levels and/or the effects of desmopressin could have been included – but then again, what does urology have to do with orthodontics? 🙂

  4. Avatar

    People are always tend to the beneficial side, but researchers should always support the null hypothesis.

  5. Avatar

    It is always a challenge to read 1 study and make conclusions that are definitive. Perhaps a larger sample size would yield a more black/white results? I’m sure the ‘airway friendly’ orthodontists will cite this article as dogma!?

  6. Avatar

    Does the fact that the subjects had “failed first line tmnt.”affect the validity and conclusions of the study.
    How do you persuade someone to subject themselves to a potentially sham RPE?
    In general terms I find little anecdotal or literature basis to use RPE,s on a regular basis !

  7. Avatar

    I think this study , is very interesting according to the new area of integrative medicine.
    The results depend on the hypothesis approach, Why the childrens needs RME?
    If this study is conduct to analysis muscular impact after RME, the delay used
    is too short , six months is not enough to obtain a muscular and functional impact.

  8. Avatar

    Dear Kevin,
    I feel honored that our study has caught your attention. I read your blogg every week and have learnt a lot from doing so. I would like to leave a small comment to clarify our conclusion.
    When concluding that RME has a modest effect on children with therapy-resistant nocturnal enuresis we refer to the long-term effect, i.e. baseline vs 6 month follow-up (p < .001) and not intervention vs placebo. We have mentioned this in the discussion where we also point out that the statistically significant long-term reduction in number of wet nights, in most cases, were not clinically relevant. Also there were a few individuals who became more or less dry.
    Kind regards,

    • Avatar

      It would be interesting to parse out, for each group, the individuals who did successfully stop or reduce the number of wet nights, from those who did not, to evaluate if there was anything in particular that those individuals can be characterize by as a predicting factor for success with night wetting therapy. I’m not sure the study has a sufficient sample size for that, but it could be worth looking into it. Further, as with RME for OSA, several studies have shown that its not a 100% solution, and some kids’ OSA is resolved, while others’ OSA worsens. I don’t believe, as orthodontists and scientists, that we understand why that happens as of yet, but this is a great field of study.

  9. Avatar

    I would like to thank Dr. Ingrid Jonson Ring’s response in the comments. Although I understand her comments, the conclusion that “RME has a modest effect on children ……” is not supported with the statistical methods that were employed. The PICO did not look at the long-term effects. The t-test statistics simply look at treatment group vs placebo group. Perhaps a letter to the editor of the Angle Orthodontist would be in order to get Dr. Ring’s response documented in the journal regarding her study.

  10. Avatar

    I believe that if they had targeted only class II div 2 groups, the result would had been much more significant.

Leave a Reply

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