March 09, 2020

Feeling the heat during initial alignment?

This post is about a new trial that looked at the effectiveness of thermal Nickel Titanium archwires. This is the first post by Professor Padhraig Fleming and he has done a great summary. I have to up my game!

Nickel-titanium wires have been the mainstay of initial aligning archwires for more than 3 decades. Interestingly,  alterations to bracket design have been accompanied by a frenzied market and feverish adoption.  Nevertheless, many believe that improvements in wire properties have had a more telling effect on the rate of tooth movement. However, there has been relatively little research to confirm a clear benefit of NiTi wires or indeed newer generation thermal NiTi wires over alternatives.

A team from Baghdad did this study. The  European Journal of Orthodontics published it. The authors aimed to compare the relative effectiveness of heat-activated and superelastic NiTi without thermal behaviour in terms of initial alignment of the lower anterior teeth.

A clinical comparison of the effectiveness of two types of orthodontic aligning archwire materials: a multicentre randomized clinical trial.

Authors: Sarah A. Nabbat and Yassir A. Yassir

Eur J Orthod. 2020; 1-9. doi:10.1093/ejo/cjz102

What did they do?

They conducted a two-group parallel randomised controlled trial with a 1:1 allocation as follows:


Orthodontic patients with a mandibular Little’s irregularity index of 3-6mm


Heat-activated NiTi wires (0.014- and 0.016- inch): TruFlexTM Thermal Nickel-Titanium, Ortho Technology, USA


Superelastic NiTi wires (0.014- and 0.016- inch): TruFlexTM Nickel-Titanium, Ortho Technology, USA

Primary outcome:

Initial alignment measured with Little’s irregularity index at 8 weeks

Secondary outcomes:

Pain experience (week 1) and root resorption at 8 weeks, assessed with peri-apical radiographs of the lower anterior region.

They used a sealed envelope system to conceal group allocation, and both investigators and data analysts were blinded to the groups. Then they placed A 0.014” NiTi wire for 4 weeks followed by a 0.016” NiTi with wires engaged using elastomerics. The degree of wire engagement (full or partial) was not clear. They reviewed the participants at 4 and 8 weeks, with alginate impressions taken at both visits. The irregularity was assessed using Vernier calipers.

What did they find?

They analysed data from thirty-one participants. There were 16 in the superelastic NiTi group and 15 in the heat-activated group.

Importantly, they did not find any significant difference in arch alignment. The difference between the groups at 8 weeks was only 0.05mm.  This was clearly neither clinically nor statistically significant (P= 0.809). They also found that wires did not affect pain experience or root resorption.

What did I think?

This is yet another study assessing the utility of marketed products in speeding up treatment. I continue to feel that many of these adjuncts have preoccupied us as researchers a little more than they should. We may be ignoring far more important aspects that influence treatment duration and outcome. In particular, human factors; specifically, us as the clinician. I wonder if our training, understanding and clinical perceptions influence both our results and efficiency to a far greater extent than any wire, bracket or adjunct.

I thought that they did the study well. They reported in line with accepted guidelines. While they did a sample size calculation,  I can’t help thinking that the sample size is very small and the standard deviations are considerable. Nevertheless, the between-groups difference is tiny and does, therefore, suggest that there is no meaningful difference relating to the rate of tooth movement with thermally-active wire behaviour. I struggled to see the merit of taking repeated periapical radiographs at such short time intervals (8 weeks). This led me to doubt that firm conclusions can be made in this respect at such an early stage in treatment.

From a clinical perspective, I would not usually make a fixed decision about progressing from 0.014-inch to 0.016-inch NiTi at 4 weeks (and would typically leave initial aligning wires in place for a more extended period than this). Instead, I tend to review the degree of engagement and slot alignment before making decisions to progress. I am sure that many of us have our own protocols in this respect. As a result, we need to consider whether these results are relevant to our own practices,

What can we conclude?

Based on a relatively small sample, the use of thermal NiTi wires does not appear to accelerate the rate of initial orthodontic alignment. While the search to prove the existence of a magic bullet to accelerate orthodontic treatment will continue, I can’t help feeling that the answer continues to lie within.

Related Posts

Have your say!

  1. Very nice. I agree with your comment about changing the wire so soon.
    I liked your positive look and not just criticising. Well done.

  2. Congratulations Paghraig on an excellent first post.

    The result should not be a surprise to anyone except the manufacturers who spruik the latest (and more expensive) wire and the holy grail of reduced treatment time. How one applies the force is perhaps more important than a formula with an arch wire change at 4 weeks. Some are even happy to start with 018 NiTi and use steel ligatures to obtain full engagement. Personally, I have found more permanent deformation with thermal NiTi and no longer use them.

    You are no doubt correct in saying that the answer lies within.

  3. thanks for the post
    another clinically relative trial done confirming no significant difference between teh kinds of wires and the speed of teeth movement.
    I hope we will treat the patient according to evidence based orthodontics rather than claims by manufacturing companies.
    since the study done in my home country where I trained , may be I can explain why the wires changed after 4 weeks.
    we used to see the patient every 4 weeks and we go through the wires step by step(i.e 0.014″, then 0.16″, then 0.018″,….etc) during leveling and alignment stage. if the teeth are too crowded, we start with 0.012″NiTi.
    when I came to Europe I found that they give the wires longer time and don’t go through all wires size, even they start with larger wires like 0.016″NiTi if the teeth are not too crowded.

    I start to do that myself because it is less time consuming and needs to use less wires.


  4. I also agree that the wires did not have enough time to exert their full effect. It is likely that the 3 to 7 month range is likely to exert the greatest changes. I also think as the wire become rectangular, they are more likely to show a significant change in archform. Lastly they talked about placing ligatures and not so much if the wire is fully engaged or not. I would be curious to o see a second study following for 6-8 week intervals, also adding the idea if self ligation into the mix. Lastly, impressing with alginate so early on is a total wild card. Teeth are loosening up and you can potentially move a tooth with the impression material. I would really like to see this study followed up with a study that scans the teeth sequentially for 6 to 9 months. I think it is a great start, but I believe most of think that 8 weeks is not enough time to make a proper assessment.

  5. Thanks Padhraig, lovely clear writing.

    I’ve never understood marketing claims saying thermally activated will work better. And here’s evidence!

    I’ve found thermally activated wire to be a satanic temptation. Devil says, “use large juicy rectangular sizes and speed the bracket prescription into action”. You’ll jump into an inappropriately sized wire that quickly and fully expresses every one of your errors: tip, torque, level & rotation, all in a single visit! Then you’re in purgatory undoing it all.

  6. Thank you Padhraig.

    This was an interesting trial and very well reported.

    I have the following comments:

    I think this trial was severely underpowered based on the results/variances and that the final sample calculation assumptions although not very clear to me, I believe, they were overly optimistic.
    To expect a 2 mm difference in 8 weeks it is probably too large of an effect. This would be almost a 50% difference given the initial crowding of 4.79 per group.
    The repeated measures design does increase power and reduces the required sample size compared to a single data collection time-point but I still see the initial calculation overoptimistic.

    I did a simple calculation (ignoring repeated measures) using the actual treatment effects (differences) and sds reported at 4 and 8 weeks and for 80% power the required sample sizes per treatment group are as follows:

    Stata 16.1 Software

    sampsi 2.85 3.12,sd1(1.77) sd2(1.37) p(.8)

    Estimated sample size for two-sample comparison of means

    Test Ho: m1 = m2, where m1 is the mean in population 1
    and m2 is the mean in population 2

    alpha = 0.0500 (two-sided)
    power = 0.8000
    m1 = 2.85
    m2 = 3.12
    sd1 = 1.77
    sd2 = 1.37
    n2/n1 = 1.00

    Estimated required sample sizes:

    n1 = 540
    n2 = 540


    REQUIRED >24000! PATIENTS AT 8 WEEKS using the trial numbers

    Obviously, we will not use a difference of 0.05 as clinically important and 24000 patients makes no sense! But is the 2mm expected difference reasonable/plausible under the circumstances?

    I understand that sample calculations performed at the design stage are based on assumptions, however, those assumptions should be as realistic as possible and supported by existing evidence.

    Looking at table II page 81 of the cited article for the sample calculation we notice that the standard deviations are quite larger than the 0.86 suggested. Larger sds and/or smaller differences to be detected require more patients.
    In the cited article initial crowding was higher by almost 50%, sample size per group was only 10!
    At 4 weeks minimum difference between groups was 0.07 mm and maximum 0.930 and sd min/max 0.72/1.78
    Taking a mean value of treatment effect (0.07+0.93)/2=.5 and for sd 1.25 the required sample size (ignoring repeated measures) would be around 200.

    At 8 weeks the minimum difference between groups was 0.11 mm and maximum 1.89 and sd min/max 0.80/1.00
    Taking the averages at 8 weeks the required sample size is only 26, again ignoring repeated measures.

    So we can see how sensitive to the assumptions [mean difference and sd] the required number of patients is!
    In conclusion, I feel that based on the above I am not sure how confident we can be about the results of this trial. I am not saying that there is a difference between the wires in terms of alignment efficiency but I am questioning as to whether we can use the data of this article alone to confidently support the claim.

    • Excuse my ignorance, but I think 2mm expected difference is reasonable given that the teeth move 1mm per month under orthodontic forces.

  7. Thank You Professor Padhraig Fleming for the excellent write up.

    Nothing really new in this study and well summarized when you stated “answer continues to lie within”. Repeated X rays of such short period with in 8 weeks also cannot be justified. I feel wire progression prescribed by KOL or Company may have their own vested market interest. At times clinician can judge and consider a planned “jump” in the wire sequence.
    Nice work, Keep going, Prof Kevin identified the right person for this blog.
    warm wishes.

  8. Why not also evaluate multristranded and coaxial stainless steel archwires as well? Before NiTi hit the orthodontic market, orthodontists were aligning teeth with these archwires. Many continued using twist wires even after NiTi, due to the large cost difference.

Leave a Reply

Your email address will not be published. Required fields are marked *