The Carriere Appliance does work! A new study
Last week I did a post about a study that suggested that the Carriere appliance was not effective. Another paper about this treatment was published this week. This suggests it works! So, I thought that I would have a critical look at it.
One interesting fact about the Carriere appliance is that it has been used for many years. However, until recently there has been no research on this appliance. Just as I published my post last week on the first research paper, the Angle Orthodontist published this new paper. A team from Michigan did the study.
Hera Kim-Berman, McNamara JA et al
Angle Orthodontist: On line. DOI: 10.2319/121418-872.1
What did they ask?
They set up this study to answer this simple question
”What are the clinical effects of the Carriere appliance”?
What did they do?
They did a retrospective study. The main study details are:
Participants: 34 patients who had been treated by Dr Luis Carriere with his appliance.
Control: 34 sets of cephalometric records taken from several historic growth studies.
Outcome: Treatment duration and multiple cephalometric measurements.
They obtained a sample of treated cases from the office of Luis Carriere. They asked him to select 30 or more consecutively finished cases. He sent them 44 sets of records. Ten of these were excluded for the following reasons :
- There was a technical issue with the radiographs.
- The duration of treatment with the Carriere Motion appliance was greater than 12 months.
- The post treatment records were made more than 4 months after the end of all treatment.
They “control” group was selected from three large growth studies.
One investigator traced all the radiographs. They analysed the data at the start of treatment, at the end of the Carriere Motion treatment and at the end of all treatment. They did not report an error analysis.
They then ran simple multivariate statistics across 18 different variables. They decided, from other published work, that a clinically significant difference was a change of greater than 2mm or 2 degrees in any cephalometric variable.
What did they find?
They provided a large amount of data and discussed this in a great deal of detail. Most of this discussion was on cephalometric changes that were not clinically significant (according to the definition that they adopted). I do not have the space here to cover this amount of data. However, I thought that the main points were:
- There was no clinically significant effect of the Carriere on the skeletal pattern.
- I could not find any data on the amount of upper arch distalisation.
- The only changes were dental. For example, when they compared the Carriere to the “control” group. They found that the overjet was reduced by 2.9mm, molar relationship was corrected by 3.1mm and the lower incisors were proclined by 3.8 degrees. We need to consider whether these effect sizes are clinically significant.
I also felt that for most of the variables the standard deviations were large. This indicates a degree of uncertainty in the data.
They looked at the duration of treatment and I calculated the 95% confidence intervals. They found that mean total treatment time was 18.2 months (95% CI 16.5-19.8) months, the mean duration of treatment with the Carriere was 5.2 months (95% CI=4.2-6.1) months.
“The Carriere appliance is an efficient and effective way of correcting Class II malocclusion”.
What did I think?
I will take the same critical approach with this paper as I did last week with the study that concluded that this appliance was not effective.
This paper was similar to many reports published by the Michigan based group. They retrospectively collect a sample of records and then compare them with historical controls. Unfortunately, this paper suffers from several substantial flaws. These are:
- This was a retrospective study.
- The records were all chosen by one clinician, who was the inventor of the appliance. They were also consecutive finishes.
- The inclusion criteria on the start malocclusion was not clear.
- They excluded cases that took too long to finish Phase 1. That is they removed the cases that did not “do well”. In effect, they excluded 25% of the records.
If I take all these factors together then there must be considerable selection bias. It is also completely illogical to exclude patients in a study because they did not complete the care in an arbitrary time.
Furthermore, they used an historical control. This method has been previously criticised because of the possibility of secular change.
I could not find any mention of an error analysis. This is important bearing in mind the small effect sizes that they reported.
Finally, they ran univariate statistics across many cephalometric variables. This is not an appropriate way of analysing this type of data because of the risk of false positives.
My feeling about this paper is similar to the previous paper that I reviewed. I am disappointed to find that this study was significantly flawed in many respects. In fact, when I consider the two papers I am a little surprised that they were published. Perhaps, the journals are very keen to publish something on this appliance?
Where does this leave us with the Carriere Motion? I don’t think that we are much further down the line. As with the previous paper, this report does not help us because it is so flawed.
I think that all we can conclude is that the Carriere appliance may correct the dental component of Class II malocclusion using heavy elastic force. But, we know nothing about its success rate, effectiveness, co-operation rate, harms, effect on skeletal pattern and whether it has an effect on the airway. We also do not know if it is more or less effective than any other method of Class II correction. In fact, to claim that it is more effective is somewhat unethical.
Several people have commented to me that this is not important. If we consider that the Carriere is a simple method of Class II correction using heavy elastics, then I agree. However, if orthodontists charge an additional fee for the Carriere, based on the manufacturers claims, then this becomes more relevant and is unethical.
Perhaps, it is time for the manufacturer to run a trial of this appliance? It would cost one just one major KOL annual fee.