Can mini-implant-supported Twin Blocks grow mandibles?
Last week, I looked at an interesting paper that explored the effect of bone-borne functional appliances. This is another paper on a similar topic. These researchers investigated the dento-alveolar and skeletal effects of mini-implant-supported Twin Blocks. It was a good addition to the literature, and I am pleased to review it on this blog.
A team from Damascus, Syria, did the study. The European Journal of Orthodontics published the paper.
Rabea A. Ghareeb et al
EJO advance access: https://doi.org/10.1093/ejo/cjaf093
The team wrote a good introduction and pointed out that most research on the Twin Block appliance shows both minor skeletal effects and more noticeable dental alveolar effects. One such effect is the lack of anchorage control, which results in proclination of the lower incisors and molars, potentially influencing the amount of skeletal change we can achieve.
They highlight a recent case report in which the authors used a Twin Block appliance supported by mini implants. The main aim of this was to prevent lower incisor proclination.
They decided to test this appliance in their study.
What did they ask?
They did this study to find out if:
“Increasing lower dental arch anchorage with mini implants influences the position of the mandibular incisors and whether this enhances mandibular advancement during TwinBlock treatment”.
What did they do?
They conducted a parallel-group randomised controlled trial with 1:1 allocation.
The PICO was
Participants:
Orthodontic patients aged 10 to 12.5 years for females and 11 to 13 years for males with a skeletal Class II Division 1 malocclusion caused by mandibular retrusion (ANB 5°-9°). They had an overjet of 5 to 8 mm with minimal crowding in both arches. The maxillary-mandibular angle was less than 30°.
Intervention
Mini-implant-supported Twin Blocks (MITB). This was a standard Twin Block appliance with a hook positioned in the buccal area between the canine and the lower premolar. They placed a mini-implant in the mandibular buccal region between the first and second molar. Then, they applied a heavy intraoral elastic with a 150g load on each side from the TB hook to the implant.
Control,
A standard TwinBlock with labial bows (TB).
One operator treated all the patients.
Outcomes
Multiple cephalometric measurements.
They asked all the patients to wear their appliances throughout the day, except during meals and sports activities. One operator reviewed the patients every four weeks. They defined the clinical endpoint of treatment as an overjet of between 0.5 and 2.5 mm with a Class 1 canine relationship. Afterwards, the appliances were worn at night for six months during the supporting stage. They took lateral cephalograms before and after treatment.
Statistical analysis
They used a pre-prepared randomisation list generated by a block randomisation method. An independent researcher who was not involved in recruitment prepared the randomisation. They concealed the allocation using sealed envelopes, which were opened only after the participant had entered the study.
They conducted a clear sample size calculation to power a study to detect a 1-degree difference in the ANB variable. This showed an estimated sample size of 20 patients per group. They then accounted for a potential dropout rate of 10%, which increased the number of patients to 22 per group.
They carried out a univariate statistical analysis for each of the cephalometric variables. As I have discussed before, this increases the likelihood of false-positive findings, and I would prefer that a regression analysis be conducted.
What did they find?
44 patients joined the study. 3 from the Twin Block group and 1 from the Mini Implant-supported Twin Block group were lost to follow-up. There were no differences in the cephalometric measurements between the groups at the start of the study.
As with last week’s study, they presented many cephalometric measurements and discussed them in detail. This was confusing, so I have decided to focus on just three simple outcome measures. These are:
- The overjet
- The Composite mandibular length derived from Pancherz’s analysis (pg/OLp + co/Olp).
- ANB
I have included the relevant information in the table below. They also presented standard deviations for each measurement. These are not particularly useful and do not meet the CONSORT guidelines. Therefore, I have calculated the 95% confidence intervals.
| Outcome/ Appliance | MITB | TB | Diffference | P |
| Overjet (mm) | -5.0 | -4.19 | 0.06 (-0.7,-0.8) | 0.8 |
| Mandibular length | 4.48 | 3.26 | 1.22 (-0.05-2.13) | 0.049 |
| ANB | -3.76 | -2.41 | -1.35 (-0.7 to -2.4) | 0.003 |
This table shows that the effect sizes are rather small. Importantly, the lower 95% confidence intervals for the differences are close to zero. This suggests that the actual difference may be minimal.
9.3%) of the implants failed.
When they examined treatment duration, it was 270.4 days for the MITB and 249.9 days for the TB. This difference was statistically and clinically significant.
They engaged in a lengthy discussion on most of the cephalometric measurements, which I felt was unnecessary. Nevertheless, their final conclusion was,
“The mini-implant-supported Twin Blocks demonstrated a greater skeletal effect than the conventional Twin Block, particularly regarding increased mandibular length. These findings endorse the mini-implant-supported Twin Block appliance as a more effective treatment option for Class 2 Division 1 malocclusion”.
What did I think?
This was a well-conducted randomised trial that followed the guidelines for trials. It was also good to see a team doing a study into new treatments before they are promoted.
My first concern is that this paper could be greatly improved by better refereeing. For example, in the results section, there was a detailed repetition of the extensive tables. Furthermore, the team repeated many of the results when comparing the data for most individual cephalometric measurements. This made it very difficult to interpret their data and discussion.
As a result, I have aimed to clarify the findings. Essentially, my interpretation is that the mini implant-supported Twin Blocks did not procline the lower incisors as much as the TB. However, since the treatment was completed when the overjet was reduced to less than 2.5mm, this may have led to a quicker TB treatment due to the greater lower incisor proclination. Consequently, there was a longer period of mandibular growth, and skeletal changes were more pronounced with the MITB. This could have been further explored by conducting a regression analysis with treatment duration as a variable.
The effect size
I think we also need to consider that the effect sizes for most cephalometric outcomes were rather small. When I used to teach critical appraisal to our residents, I pointed out that we should not get too hung up on statistical significance. It is far more important to examine the effect sizes and decide whether they are large enough to justify adopting a new treatment. In this respect, the differences detected in this study are not clinically significant, and I would not use this technique with my patients.
I would also like to emphasise that, for most of our studies, including the final treatment data, it is necessary. In functional appliance studies, this is especially important because we need to determine whether the changes attributed to the functional phase do not “wash out” during the final phase of fixed appliance treatment.
Final comments
Please do not think I am being overly critical of this study. I am simply trying to offer some clarification. There are many good points to this work, and it provides us with useful information. It certainly would have benefited from more focused refereeing. I hope that the authors continue with the trial until the end of treatment. This final paper will then be a genuine contribution to the literature.
Emeritus Professor of Orthodontics, University of Manchester, UK.
An additional limitation of this study is the absence of an untreated control group. In the absence of treatment, no dental changes would be expected, whereas natural skeletal changes would be anticipated. I agree that the difference in treatment duration between the groups is a significant concern. Both groups should have been evaluated after an equivalent treatment period to ensure valid comparison.
Whilst I take your point I think that ethically it would be difficult to justify taking lat cephs for pts not undergoing any treatment. Frankly the effect sizes were so small it rendered the use of tads pointless when the trauma in young children is taken into account.
It appears many are flogging a dead horse on this issue. How many studies do we still need to confirm the reality which is not going to change irrespective of what one uses.
to be honest i use Twinblocs on major Overjects ie 9+mm as i can manage smaller overjest with better patient complinace with fixed
when can we expect to see RCT’s with on patients with large sketelal discrepences/ large Overjets, when we would expect to see a clinical difference as well as a ‘statisical’ ceph difference
Dentofacial orthopedics-not really. More accurate to call the effects of all these appliances as “dentofacial deformation”. Not much different from foot binding, head binding, or cranial molding.
If you look closely at the tracings in Clark’s book you will notice that there is no or very little flaring of the lower incisors in his treated cases. This has been my experience as well as I cap the lower incisors in the design of my appliances. Did this group indicate their design of the appliance or even the age of treatment as I find this to be a major deficiency in many of the tba studies. Ie treating at the wrong age, for insufficient time and inadequate appliance design.
There’s something wrong with the Overjet values reported here – the point is not bracketed by the CI. Also, in the original paper, a SD of 1.02 is very small for a P of just over 0.001. Is 1.02 the SE being misreported as the SD?
Thank you for your thoughtful and balanced review of our study. We truly appreciate the time and care you took to place our findings within a broader clinical and methodological context.
We fully agree with your interpretation that the observed skeletal differences were modest in magnitude and that their clinical relevance should be interpreted with caution. This was never intended to suggest a dramatic enhancement of mandibular growth.
However, we would like to highlight what we believe is the main novelty of the study. To our knowledge, this is the first randomized controlled clinical trial to investigate a mini-implant-supported Twin Block appliance designed specifically to address one of the most frequently cited limitations of functional appliances—namely, inadequate mandibular anchorage and lower incisor proclination. Previous evidence in this area has been limited to case reports or observational data.
Our primary objective was therefore exploratory: to test whether reinforcing mandibular anchorage could alter the skeletal–dentoalveolar balance during functional treatment under controlled conditions. In this context, even small skeletal differences are of interest, as they help clarify the biological and biomechanical limits of functional appliance therapy rather than promote a new technique for routine clinical use.
We also appreciate your insightful comments regarding treatment duration as a potential confounding factor and agree that this warrants further investigation, ideally using multivariable models and longer-term follow-up.
Thank you again for the constructive critique. We believe discussions such as this are essential for advancing evidence-based orthodontics and for preventing premature clinical enthusiasm for new techniques.
As noted in my other post on this topic, orthodontists appear to have a peculiar perspective on ‘growth’, especially mandibular growth. In Kevin’s summary, changes in the maxilla have been ignored. The study states “Both groups showed reduction in maxillary position (SNA)”. This means that the entire phenomenon may not have been fully captured in his useful commentary, and perhaps it over-emphasizes perceived mandibular changes. The results of this study are similar to other studies that illustrated mandibular changes with TB [1] are associated with maxillary retrusion using TB [2] with a consequent reduction in overjet. So, it’s also possible to suggest that potential mandibular growth might have been inhibited using the MARPE-TB protocol by;
1. The (retruded) maxilla impinging on the mandibular spatial matrix, if the cranio-caudal gradient of development is applied.
2. A device putatively designed to prevent lower incisor proclination that may have inadvertently inhibited mandibular response.
In general, nuanced terminology is used to describe mandibular changes such as ‘catch up growth’, ‘mortgaging the mandible’, etc., which have no scientific definition and no identified developmental mechanism(s) per se. However, non-orthodontic studies have demonstrated “growth” of the mandible/condyle. In one of these studies, the condyle was non-surgically ‘regrown’ in a child with hemifacial microsomia [3]. Our craniofacial team was able to ‘regrow’ the condyle in a group of children with hemifacial microsomia [4] using mandibular distraction osteogenesis. However, the basis of developmental mechanisms for these changes needs to be determined, which might help understand mandibular growth more comprehensively. ‘Catch up growth’ perhaps represents genetic potential, which I define as achieving an optimal outcome in the prevailing condition subject to a viable population of stems cells. This definition invokes gene expression, since changes in the position of the mandible are associated with changes in gene expression [5-7]. If so, it should be possible to modulate mandibular growth by changing its spatial positioning. Evidence for this concept is sparce but in babies treated for Robin syndrome, the genetic potential (‘catch up growth’) of the mandible is demonstrable [8]. Furthermore, if this is the case, could this principle be applied to ‘non-growing’ adult mandibles? Although the evidence is scant, clinical observation shows renewed mandibular growth in some adults treated for OSA with a mandibular advancement device. Most of these cases remain undocumented but some [9] have illustrated this phenomenon. In any case, I fear that working under the condition of cephalometric parsimony, the answer to the question posed at the outset cannot be sufficiently addressed.
References
1. Am J Orthod Dentofacial Orthop. 2001
2. Angle Orthod. 2002
3. Cleft Palate Craniofac J. 2003
4. Semin Orthod. 2011.
5. Am J Orthod Dentofacial Orthop. 2003
6. J Stomatol Oral Maxillofac Surg. 2024.
7. J Oral Biol Craniofac Res. 2025
8. Cleft Palate Craniofac J 2025
9. Dialogue 2010
In this time of global technology and the global need for orthodontic treatment, we need to collaborate globally to build new systems. We have to start somewhere. Areas to focus on are
1. orthodontic education courses published by experts in the field (as opposed to lectures) that measure learning online and lab/clinical performance
2. Education and training can be provided locally by “educated and trained Mentors” who have learned the courses by the experts
3. Experienced researchers teaching local research fellows research protocols, analysis, and publication of research results to individuals/groups who need the information
4. Development of dental/orthodontic patient data gathering computer systems (such as Epic which University of Michigan just innovated),
5. Companies that sell products such as aligners, appliances, and equipment, must fund these education and training systems so that users who purchase their products know exactly how to efficiently use these products for the best results.
I know this sounds impossible. It can be the future if we start working together now.
Rebecca