A new RCT about retainer failures and relapse
Relapse and retention of orthodontic treatment is an important area of research in orthodontics. Over the past few years, investigators have published well-done randomised trials on retention. These have told us that relapse is still a problem and that removable vacuum-formed and bonded retainers have good clinical performance. Nevertheless, there is still some relapse and retainer failure even when we use fixed retainers. A new development has been CAD/CAM retainers, and the effectiveness of these was looked at in this long-term study.
CAD/CAM retainers may have several advantages over other forms of bonded retainer. In brief, there is no need to bend the wires; the fit is precise and individualised for the patient, and they are easy to position.
The authors of this new study pointed out in their excellent introduction that retainer failure rates range from 7.3% to 50% (although we all feel our failure rate is 1-2%!). When investigators have looked at CAD/CAM retainers, their short-term results (6-12 months) appear satisfactory.
A team based in Turkey did this new study. Orthodontics and Craniofacial Research published the paper.
Serpil Çokakoğlu et al
Orthodontics and Craniofacial Research: Online. DOI: 10.1111/ocr.12714
This is an open-access paper, so everyone can read it!
What did they ask?
“What are the changes in the stability and failure rates of four fixed retainer types after 3 years of being placed”?
What did they do?
This was the long-term recall of participants of a randomised trial of retainers. The PICO was.
132 participants in the trial who completed orthodontic treatment between November 2018 and January 2019. They initially had mandibular crowding of 4-6mm and had been treated with fixed appliances.
They evaluated four interventions: (1) dead soft wire, (2) multistrand SS wire, (3) CAD/CAM nitinol wire, and (4) canine bonded pads.
The primary outcome was changes in tooth position, and secondary outcomes were failure rates three years after they placed the retainers.
They measured Little’s Irregularity Index, intercanine width and arch length. The team classified failed retention when the fixed retainer was detached from one or more bonding sites or the wire broke.
Interestingly, the tooth movements were excluded from the data analysis when a participant had total retainer loss. I am not sure this is correct, as the tooth movement is a consequence of retention failure and is a valid and vital outcome.
They did a post hoc power calculation that showed they needed 25 participants in each group. Again, they did this to consider the dropout rates of the recalled participants.
What did they find?
Initially, they enrolled 132 participants into the original trial. When they did the follow-up, 36 (27.2%) were excluded from the data analysis. 14 declined to participate because of COVID concerns, 10 could not be contacted, and six had moved to another city. 5 had lost their retainers. As a result, they obtained data from 96 participants who attended the recall. The mean follow-up was 3.47 years.
When they looked at stability. There were no differences in the dental measurements between the groups when the retainers were placed. I then looked at the data at the end of the study period for each primary dental measurement and extracted it into this table. I only considered the final measurements, as this was the clinically relevant measure that patients would notice.
|Outcome||Dead soft wire||Multistrand SS||CAD-CAM||Connected bonding pads||p|
|Irregularity||0.78 (1.15)||0.16 (0.52)||0 (0.5)||1.05 (1.21)||<0.001|
|Intercanine distance||25.69 (1.16)||26.0 (1.36)||26.21 (0.84)||25.60 (0.83)||.18|
|Arch length||57.97 (1.8)||58.7 (2.89)||58.83 (3.18)||58.13 (2.62)||0.59|
Their final conclusions were:
“After 3 years of placement CAD/CAM and multistrand SS wires were more effective than the other types of retainer”.
“The CBP had the highest failure rate”.
What did I think?
This was a clear report of a good study. When considering the results, we need to evaluate whether the effect sizes between the interventions were clinically significant. When I did this, I was reassured that the amount of relapse was minimal. I also thought any differences between the groups were not large, even though they were statistically significant. Again, this is an important finding. In many ways, this is unsurprising as the retainers were all bonded to the teeth and, therefore, were unlikely to move. But we have all seen the reports of moving teeth with fixed retention!
There were important differences in the failure rates of the retainers. These persuade me to not use connected bonding pads or dead soft wire.
Finally, we must consider if there are any differences in the cost of the CAD/CAM and multistrand retainers. Unfortunately, the team did not report this in the paper.
I was concerned with the high dropout rate, and we must bear this in mind when we interpret this study. However, this was not surprising when we consider that long-term follow-up of our patients is difficult, mainly when there is a global pandemic!
I want to congratulate the authors of this study. It is an addition to the literature on the problematic area of retention and relapse.
Emeritus Professor of Orthodontics, University of Manchester, UK.