The end of self-ligation: or is it?

The final nail in the self ligation coffin: or is it?

As readers of this blog will know I have posted about self ligation several times.  I have recently read a new study that looks at whether self-ligation results in a reduction in the duration of treatment, but does this add to knowledge?

I will start this post with a brief summary of what we know about self-ligation.  In many ways this is a classic story of discovery, followed by the release of a product and finally, testing that shows it is no better than any other product.  When self-ligation was re-discovered the manufacturers of the brackets made several claims for the benefits of the new technology.  These claims were supported by initial retrospective studies and then promoted by Key Opinion Leaders (who frequently did the studies).  Eventually, several years later, independent  invistigators carried out randomised trials. As we all know, these studies showed that there were no or very limited advantages of self-ligating brackets over conventional brackets.

The journals have published several trials that have looked at self-ligation. While these provided very useful information, they were directed at evaluations of the phases of a course of orthodontic treatment. For example, the rate of space closure, alignment and overbite reduction. To my knowledge, there has only been one trial that has evaluated the effect of self-ligation on total treatment time.

This team, from Greece, did this new study to find out if the use of  self-ligating brackets resulted in a reduction in treatment time.  The Journal of Orthodontics published this paper.

vervolg: Kevin O'Brien: The end of self-ligation: or is it?


Treatment duration and gingival inflammation in Angle’s Class I malocclusion patients treated with the conventional straight-wire method and the Damon technique: a single-centre, randomised clinical trial

Eleftherios G. Kaklamanos et al

Journal of Orthodontics:

What did they do?

They ran a two group parallel randomised controlled trial with a 1:1 allocation. The PICO was

Participants: Patients with a Class I malocclusion with a space deficiency of more than 9mm

Intervention: Orthodontic treatment with Damon 3 brackets

Comparator: Orthodontic treatment with conventional brackets

Outcome:  Duration of treatment, Gingival Index Score. PAR score

One operator treated all the patients and saw them every 6-8 weeks.

They used a pre-prepared randomisation scheme and concealed the allocation in envelopes.  The operator enrolled the patients into the trial and then let them know their treatment allocation.

They could not blind the patient or the operator to treatment. But they collected and analysed the data blind.

They carried out a sample size calculation that was based on being able to detect a difference of 4 months based on a treatment time of 24 months for the treatment.

What did they find?

They randomised 22 patients to the two interventions (11 to each).  The patients did not have extractions as part of their treatment.

There were no differences between the groups at the start of treatment.  I have included the end of treatment data in this table.

 Conventional bracketsDamon Brackets
Treatment duration (months) 14.5 (95% CI 12.7-16.3) 12.25 (95% CI 10.55-13.95)
PAR score 0 0
Gingival Index 1.5 1.5

The difference in treatment duration was 2.25 months (95% CI -0.40-4.9).  This was not statistically significant.  They concluded that the use of self-ligating brackets did not result in a reduction in treatment time.  Furthermore, there was no difference in the PAR scores between the interventions.

What did I think?

In their discussion they pointed out that in their sample size calculation they assumed that the average treatment duration would be 24 months. However, in their study they found that it was several months less than this estimate.  This may have resulted in a possible lack of power to detect a difference.

This becomes more relevant when we look at the 95% confidence intervals of the differences they detected. While this contains zero (suggesting no difference), we must consider that this was marginal. P was also 0.09 suggesting that there is a 9% probability of incorrectly accepting no difference exists. Finally, the 95% confidence intervals are rather wide, this suggests that there is uncertainty in their data.  I feel that these issues have happened because of the rather small sample size.

I cannot help thinking that this was a well carried out study, but I am concerned that  the sample size was too small. This leads me to wonder that if they had included a larger sample, there would have been a statistically significant difference.  As a result, my next step is to look at the effect size. This was 2.25 months.  We need to assess whether this is clinically significant and this is up to you to decide.

What can we conclude?

My interpretation of this data suggests that this study is probably underpowered. As a result, I am not too confident in the conclusions.

Nevertheless, before the “self ligators” get all excited, this does not change the current state of knowledge on self-ligation.  There is still an absence of evidence that this treatment is more effective than conventional brackets.


Meer Kevin O'Brien: klik hier