Does OrthoPulse work? The results of another trial…
A few weeks ago I posted on a pilot study that evaluated the effectiveness of phototherapy on orthodontic tooth movement. This post is on a new study that looks at whether phototherapy (OrthoPulse) increases rate of orthodontic space closure. I felt that this was an interesting study and decided that it would be useful to discuss this in some detail.
Chung, S. E. V, Tompson, B., & Gong, S.-G. (n.d.).
Journal of Orthodontics 2015; 42(4), 274-283. doi:10.1179/1465313315Y.0000000013
The authors provide a nice introduction that outlined the use of phototherapy in improving wound healing, tissue repair and reduction in dental sensitivity. Most of the studies that they mentioned were animal studies and this led to their study, which is one of the first trials of this technique in orthodontic patients. They aimed to find out whether LED phototherapy could accelerate the rate of orthodontic tooth movement during space closure.
What did they do?
They carried out a split mouth randomised controlled clinical trial of LED phototherapy delivered by an Orthopulse device. This involved the application of phototherapy or an inactive light source to different quadrants during space closure.
They carried out a power calculation which was based on detecting a difference of 1.0 mm/month increase in truth movement. This was a clear clinical difference that they hoped to detect.
The inclusion criteria were:
- Patients aged greater than 11 years old
- Fixed edgewise therapy with bilateral symmetrical extraction of either first or second bicuspids
- Space closure was required
Standard treatment mechanics were followed and the spaces were closed either on 0.018ss or 0.016×0.022 ss wires.
They took impressions for study casts to measure the extraction space measurements at the following points
- T0=day of starting space closue
- T1=5 weeks +/- 2 weeeks after the start of space closure
- T2=5 weeks +/- 2 weeks after T1
The Orthopulse device was activated on one side of the dental arch, whereas the control side within each participant had an inactive light source. This meant that each participant acted as their own control. The clinicians were blinded by alternating the side of the light application.
The device logged compliance..
What did they find?
They included 11 patients in the study and a obtained 17 quadrants requiring space closure. In six participants both arches were included in the study and in the remaining 5 participants one arch was included.
The degree of compliance was 78% of the prescribed time resulting in an average of 32 days light application time.
I have concentrated on the space closure data and I have included it in this table.
|Time point||LED (mm)||95% CI||Control (mm)||95% CI||p|
|T0-T1||-1.6||-1.61 to -1.59||-1.72||-1.73 to -1.71||0.5|
|T1-T2||-0.89||-0.9 to -0.8||-0.41||-0.42 to -0.4|
You can see that the differences between the interventions were very small and the 95% CI were very narrow. I was surprised at this as the sample size was small but this does suggest that for this sample the amount of uncertainty was low.
The average rate of space closure was 0.042 mm/week and 0.039mm/week for the LED treated and control sides respectively. this was not statistically or clinically significant.
In their discussion they pointed out that the split mouth methodology than they used had an advantage over other methods because each person acted as their own control. They also pointed out that the sample size was rather small and that additional studies are required.
What did I think?
I felt that this was an interesting study that provided some information on the effectiveness of OrthoPulse. I did have some concerns over the split mouth design. This is because I could not see how the mechanics on one side of the mouth would not influence the other. I also wonder if the application of the light on one side would have an effect on the side that was inactive. Furthermore, there are well established statistical arguments that suggests standard sample size calculations do not take the effect of clustering into account. This may mean that the study could be underpowered and not be able to detect a difference.
Nevertheless, simply examining the data suggest that on this small sample the effect size was very small and it is unlikely that this small amount of tooth movement would be statistically significant.
My only real concern is shared by the authors and that is the study is rather small and I certainly agree that larger studies are needed before clinicians can ethically recommend this technique to their patients. These studies should evaluate the effect of OrthoPulse on the total treatment time.
I hope that investigators use the data from this study, and the investigation that I have previously discussed, to generate sample sizes for larger investigations. Until then most clinicians should conclude that there appears to be few benefits from this new technology. Interestingly, this study has not been publicised on the OrthoPulse website and I shall certainly be discussing this when I speak at the next AAO congress on the subject of “In the land of uncertainty: is the salesman King”?