A trial on surgical assisted tooth acceleration suggests it works!

Searching for methods to speed up tooth movement appears to be never-ending. There have now been many studies into the effect of potions, pills, vibration, light, and trauma. Unfortunately, despite the hopes and dreams of the developers of these techniques,  there is currently minimal evidence that suggests they have a clinically meaningful effect. Furthermore, it appears doubtful that the small clinical effect is worth the additional burden of care, pain and trauma.

This is the latest study into the effects of surgical interventions. Interestingly, the authors conclude that piezocision is an acceptable procedure. However, I am not so sure, but I may be being overcritical.

This new study looked at micro-osteoperforation and piezocision in leveling the mandibular arch.  I would like to point out that the authors did not measure “leveling the lower arch,” but they measured the resolution of crowding.  

What did they ask?

In their literature review, the authors mention that a systematic review published in 2019 suggested a lack of evidence on the effectiveness of surgical assisted orthodontics. As a result, further research is needed. This is why they did this trial to look at the following question.

“What is the effectiveness of piezocision and Micro-osteoperforation in accelerating the levelling of mandibular teeth”?

and

“What are the effects of these interventions on the periodontium and pain”?

What did they do?

They did a randomised controlled trial with a 1:1 allocation. The PICO was

Participants

Orthodontic patients with greater than 5mm mandibular crowding. treated with MBT fixed appliances

Interventions

Piezocision or MOPs

Control

Treatment as usual with no intervention

Outcome

The primary outcome was incisor irregularity measured by Little’s Index. Secondary outcomes were periodontal measurements and oral health-related quality of life.

The team collected study cast data at the start of treatment (T0), 4 weeks (T1), 8 weeks (T3), 12 weeks (T3) and 16 weeks (T4).   As a result, five casts were obtained for each patient. 

The operator did the surgical piezicsion or the MOPs at the start of treatment.   They did the MOPs with the Propel device.

A blinded examiner recorded the alignment of the mandibular teeth with Little’s Index using digital callipers.

The statistical analysis was a 1-way analysis of variance with a Bonferroni correction.

They did not provide any information on the random sequence generation or concealment. I will return to this later.

The sample size calculation was based on detecting a 0.4 effect size and suggested that they needed to enrol 42 patients. 

What did they find?

They enrolled three groups of 15 patients to each intervention/control. At the end of the study, only one patient had dropped out. There were no differences between the groups at the start of treatment.

The authors provided data for each time point across all three groups. However, I found this difficult to interpret. As a result, I will only look at the data from T0 to T4 (0-16 weeks). I hope that I have extracted the relevant data into this table.

This shows the Little’s Irregularity scores for the different time points. I felt that the most important values were the scores at T4. This is because if you take a sample of patients with no differences in the outcomes at the start of treatment. It, therefore, follows that if there are differences at the end of treatment, this is due to the intervention. There is no need to measure change etc. I recommend this simple approach to anyone reading a complex set of tables.

Furthermore, I also calculated the 95% confidence intervals, as this lets us know how certain we can be about the data. I was surprised that the AJO did not include this information as it is part of the CONSORT recommendations. 

Little’s Irregularity Index values

 You will see that at T4, the difference between MOPs and Piezocision is only 0.13 mm, and the confidence intervals overlap. This difference is not clinically significant, and we do not need to run statistics. When I looked at the difference between the control and Piezocision, the difference was 0.38 mm. Again, this is not clinically significant. However, the confidence intervals do not overlap. As a result, this difference is likely to be statistically significant.

The authors did not report any statistical analysis for these values, and I doubt it is needed. Nevertheless, when they looked at the change in LI between T0-T4, they found no difference between the MOPS and Piezocision. However, there was a statistically significant difference between the Piezocision and control. This was 1.07 mm. 

This then led them to conclude.

“The levelling of mandibular anterior teeth was accelerated by piezocision over 16 weeks, predominantly in the first 12 weeks, whereas MOP had no effect”.

What did I think?

This was a small trial that looked at a clinically relevant question. However, I found it rather difficult to read and interpret. This was mainly because the paper was not written according to the CONSORT guidelines.

I was also concerned that there was no detail of the randomisation and method of allocation concealment. This means that the study is at high risk of bias. This does not mean that the investigation is biased, but there is a high possibility that bias is present.

Finally, I am not sure I agree with the author’s conclusions. We must look closely at the effect size to assess if it is clinically significant. In this case, I felt this was not the case, particularly when we consider that the “price” for this small change is a surgical procedure.

Final comments

This small trial suggests a lack of evidence for using surgical aids to speed up tooth movement. However, there is a large amount of uncertainty in the data.