Expansion with Aligners or RME. What is better?
Invisalign First is gaining popularity for early orthodontic treatment despite limited evidence on its effectiveness. This new study compared the effectiveness of expansion with clear aligners and Rapid Maxillary Expansion.
When deciding to do maxillary expansion, one of the choices we must make is whether to use slow or rapid methods. Current research has shown little difference in the effectiveness of either method. However, RME appears to be more popular.
While RME methods are well established, we may use several other methods to achieve slow expansion. One of these is using clear aligners (CAT). Unfortunately, research into the effectiveness of expansion with aligners is limited. But, this new paper adds to our knowledge, to a certain extent.
A team from Italy did this study. The AJO-DDO published the paper.
This paper was published using an open access agreement, so we can all read it!
What did they ask?
They did this study to
“Evaluate the efficacy of Invisalign First Phase I treatment compared to tooth-borne RME, by investigating changes in the palate and maxillary teeth’.
What did they do?
They did an open label 2 arm, parallel randomised controlled trial
The PICO was:
Participants
Patients in the mixed dentition with a posterior transverse interarch discrepancy of a maximum of 6mm with fully erupted maxillary and mandibular first permanent molars. Orthodontists treated the patients in the department of orthodontics in Turin and one private practice.
Intervention
Invisalign First appliances. They asked the patients to wear the aligners for 22-24 hours per day. They used attachments on the crowns of the primary teeth. Treatment continued until the palatal cusp of the posterior maxillary teeth contacted the buccal cusp tip of the mandibular posterior teeth, When this point was reached they fitted vacuum formed retainers.
Control
RME with a tooth borne Hyrax screw appliance. The patients turned the screw twice a day (0.45mm activation per day). This was done until a similar endpoint to the Invisalign group was reached. The RME was then left as a passive retainer for 6 months.
Outcomes
They took scans of the teeth and palate. The primary outcome was palatal volume. Secondary outcomes were intermolar and intercanine widths at the gingival and cusp level.
They could not blind the operators, but they collected an analysed the data blind.
They calculated the sample size. However, this did not contain information on the effect size they hoped to detect or the SDs of any previous study.
What did they find?
Forty-one patients (19 males and 22 females) entered the study. They divided these into two groups. There were 20 in the CAT and 21 in the RME groups. There were minimal drop outs. As a result, they analysed data on 19 CAT and 20 RME patients. There were no differences between the groups at the start of treatment.
The mean treatment duration was eight months for the CAT group. For the RME, the active treatment duration was 10 days, and the total duration was eight months.
They presented start measurements for both groups. There were no differences between the groups at the start of treatment.
After the treatment, they did not detect any statistically significant differences in any outcomes, apart from the intermolar width at the gingival level at the end of treatment. The change for the RME group was 3.87 mm, and for the CAT, this was 1.58mm. This suggests that there was greater tipping in the clear aligner group.
All other changes in dental measurements were in the region of 2-4mm and what we would expect from efficient expansion treatment. There was not an assessment of skeletal changes.
Their conclusions were
- There was a significant increase in all outcomes for both interventions.
- RME outperforms clear aligner treatment for all outcomes. However, this did not reach statistical significance.
- There may be more buccal tipping with clear aligners.
What did I think?
This was a small but interesting study that provided valuable insights into clear aligner treatment methods. The most significant finding was the minimal differences between the effects of the different treatment methods.
However, I disagree with their conclusion that RME outperformed CAT. This because the differences effect sizes were not statistically significant, indicating that the differences could have occurred by chance. I am surprised that the referees accepted this statement.
When we consider the question of greater tipping in the CAT group, this may be the case, but I am not sure that it is clinically significant.
I was also unclear about the criteria for patients to undergo expansion treatment. This was based on arch width discrepancy. Importantly, it was not based on the presence of a crossbite. I might be old-fashioned, but I rarely perform an expansion treatment without crossbites. However, it has been some time since I have treated a patient and I haven’t caught up with the latest fervour of expanding maxillae at all ages.
I have looked at this paper carefully. Martyn Cobourne has also done a Facebook post on it, and we have come to similar conclusions.
Final conclusions
In this sample of patients, it seems that CAT and RME are effective. The choice of treatment method should take into account patient preferences and cost. Currently, I believe that CAT is more expensive than RME. Therefore, I would lean towards the traditional option that requires less patient cooperation. However, the evidence supporting aligners as a viable treatment method is steadily growing.
my question is what about the airway?
A large proportion of patients require both an RME and a facemask. Clear aligners can’t be the hammer for all nails and common sense does prevail. Ethics is important and where a facemask is required, A banded RME remains the standard
Dear Kevin,
Thank you for your interesting and careful review of this paper. Please also consider these points:
Placing the RME appliance on second deciduous molars only, may not be the most efficacious choice. Since the first permanent molars had erupted, perhaps it would be best to include those as well (i.e. four teeth). This might have resulted in a larger effect.
Sex distribution was marginally comparable between the two groups. A Fisher’s exact test for ‘5 males, 14 females’ vs. ‘12 males, 8 females’ gives a P value of 0.054.
The authors report that no evidence of non-normality was found. Why, then, did they not use parametric statistics?
The change in palatal volume with RME was more than twice that with clear aligners. A t-test gives a P value of 0.085 (the authors report 0.310) and a 95% confidence interval for the difference between means: 618.2 to -42.1. Considering the small sample size and the heavily skewed confidence interval, it may be a bit premature to conclude that there is no difference between the two treatments regarding palatal volume.
The authors report that the sample included subjects without some teeth, but these subjects were not excluded, even though some measurements could not be made. Does this mean that the sample size was not 19 and 20 in all tables? The authors should report which measurements were made with a smaller sample size.
There is a typo in the published paper: P = 0.360 should be P = 0.036.
In your review of this paper, you conclude that the difference between the treatments did not reach statistical significance. This is not true for “Intermolar width at gingival level”. Also, we need to keep in mind that the two groups were not similar at T0; the RME group had narrower canines and molars at the cusp level.
All considered, this is an interesting study, but perhaps limited in its evidence strength.
What about the skeletal effect?. This is an important issue. Dental expansion can be achieved also with fixed braces, removable plates, ….
Assuming a clinical crown length of 5 mm on the lingual of the first molar, a 2mm difference in expansion measured at the gingival margin represents 23.5 degrees of tip. (if my very rusty trigonometry is correct) This seems to indicate a rather significant difference in the performance of the two types of expansion. If you extrapolate 23.5 degrees using the depth of the palate and assuming 5 or 6mm of expansion measured at the cusp tips, it seems that would be a significant difference in the amount of true skeletal expansion which, for me, is the real measure to compare the success of expansion in the mixed dentition. (someone else can do the math on that extrapolation)
The real question is basal or dento-alveolar effect?
Each expansion strategy has a clinical indication. How surprising that they did not study the effect at the skeletal level.
on the matter of transverse discrepencies where there is no crossbite….
you might not have a crossbite if the lower molars are tipped/inclinedd lingually to compensate for a narrow maxilla, or the upper molars are tipped/inclined buccally , but later on when you treat the patient the transverse distance between the lower molars will increase as they upright and then you have a crossbite and the upper molars aren’t in a good orientation to deal with it.
Am I missing something?
Stephen Murray
Swords Orthodontics
So particular samples, cooperation of young children witch wear the agliner 24hs/day, are common just at the Ortho Department of the Turin University…
Dear Professor Kevin O’Brien,
Thank you for the commendable effort you have sustained over the years in passionately maintaining this blog and for sharing my article, allowing for a constructive debate on this “hot” topic. I fully concur with your final conclusions, which is why I continue to treat the majority of mixed dentition patients using traditional methods. While clear aligners have demonstrated effectiveness over the years in upper arch development, I believe they remain less efficient, costly, highly dependent on patient compliance, and still show limited indications, particularly for growing patients.
I also agree that the two groups show minimal differences between the effects of the different treatment methods. Indeed, the statistical analysis tells us this, and as you rightly reiterated, when calculating the effect size, we realize that these differences could have occurred by chance. However, I chose to offer cautious conclusions because, unfortunately, many professionals focus solely on the conclusions of an article, and reading “there is no statistically significant difference between the two treatments” could have posed a risk of misinterpretation or overgeneralization, especially considering the specific inclusion criteria of our study. Nonetheless, I have requested the editorial office to make some modifications to the text based on the valuable suggestions I received, in order to make the article even clearer and free of potential ambiguities.
Best regards,