Determinants for success rates of Temporary Anchorage Devices in orthodontics : a meta

Determinants for success rates of temporary anchorage devices in orthodontics: A meta-analysis (n > 50) / Dalessandri, Domenico; Salgarello, Stefano; Dalessandri, Michela; Lazzaroni, Elena; Piancino, Mariagrazia; Paganelli, Corrado; Maiorana, Carlo; Santoro, Franco. In: EUROPEAN JOURNAL OF ORTHODONTICS. ISSN 0141-5387. 36:3(2014), pp. 303-313. Original Citation: Determinants for success rates of temporary anchorage devices in orthodontics: A meta-analysis (n > 50)


Introduction
Anchorage is one of the most important elements for successful orthodontic treatment.
Traditionally, orthodontics employed teeth and extraoral or intraoral appliances for anchorage, often relying on the patient compliance for its effectiveness. Osseointegrated dental implants were introduced to strengthen anchorage (Wehrbein and Merz, 1998;Wehrbein et al., 1999;Chen et al., 2005;Wehrbein and Gollner, 2007), but these implants present with a number of disadvantages that limit routine use.
More recently, different types of skeletal anchorage devices have been introduced, offering potential advantages compared to osseointegrated implants, including: smaller size, which in turn allows more versatile use and reduces amount of surgical intervention necessary, resulting in less patient discomfort; the possibility of immediate loading; lower costs; and ease of removal.
Mini-implants are derived from endosseus implants. They have a conical shape with a head that emerges from the mucosa, and that allows connection with orthodontic appliances; mini-implants also contain a smooth transmucosal neck and an endosseous threaded body that can be manufactured with different thread designs and body shapes. Length and diameter vary widely between makes, and the surface is generally smooth, which limits osseointegration.
Mini-plates that are used for orthodontic anchorage are very similar to maxillofacial plates, consisting of a base-plate and fixation screws made of titanium. The shape and size can differ, and the number of fixations can vary from two to five screws.
Mini-screws are made of titanium and are specifically designed for orthodontic anchorage. Their shape is similar to that of mini-implants, but mini-screws are usually smaller (less than 2 mm in diameter) and sometimes more tapered. The thread can be self-drilling to allow direct insertion without the use of pre-drilling, which simplifies the insertion technique.
To encompass all of the above mentioned devices in one definition, we decided to use the term TAD, (Temporary Anchorage Device) in this publication.
A number of factors can vary in the use of TADs for orthodontic anchorage in humans: the aim of this systematic review is to analyze the influence of the various elements on the success rate of a temporary skeletal anchorage devices.

Materials and methods
The method for this review was based on the METHODS OF SYSTEMATIC REVIEWS AND The following selection criteria were used to select appropriate articles: (1) articles on implants and screws used as orthodontic anchorage, (2) data only from human subjects, (3) studies published in English, German, French, Spanish and Italian (4) studies with more than 50 TADs, and (5) studies that could be RCTs (randomized clinical trials), or prospective and retrospective clinical studies.
The articles were selected after first reading their titles and abstracts. All of the articles that appeared to meet the inclusion criteria on the basis of the abstract were read, and further selections were made.
These articles were independently selected by three reviewers (EL, MD and DD). Any disagreements were resolved by discussion until consensus was achieved.
The null hypothesis was that TAD success rates are independent ot the factors listed in Figure 1, and p values of <0.05 were chosen to indicate statistical significance.. Review Manager (RevMan) software (Version 5.2. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2012) was used to construct a forest plot for each factor considered at least by five studies, weighting implant failure rates, reported as odds ratios (OR), under the random-effects model. The same software was used to calculate the I 2 index, which is an indicator of studies heterogeneity, and to construct a funnel plot, that was used to detect publication bias. I 2 has a range of 0-100%: an I 2 value near 0% indicates that almost all the observed variance is spurious, whereas an I 2 value near 100% means that most of the observed variance is real.

Results
The electronic search provided 244 abstracts that addressed mini-screws and mini-implants used as anchorage. After screening, 35 of these abstracts were excluded because they described maxillofacial procedures. Another 10 abstracts that were literature reviews were also discarded. The remaining 199 articles, for which the abstracts seemed to be relevant, were read in full. After applying the selection criteria, 26 articles were considered suitable for the study. These articles contained specific references to the factors that could influence the success or failure of the miniscrews or mini-implants. The sequence of the application of exclusion criteria is shown in Supplementary Figure 1, available online.
The first search of the reviewed articles aimed to highlight all of the factors that were considered to have an effect on TAD success or failure. We divided those factors in three main groups: 1) Patient-related factors 2) Implant-related factors

3) Management-related factors
Each group could be further divided into sub-sections, as shown in Figure 1.
The summary of all of the factors analyzed in each study is shown in supplementary

Discussion
A systematic review with a strict protocol and an accurate search strategy was performed to provide data about success and failure rates of TADs. The primary aim of the review was to gather information about the factors that influence failure and success rates.
The articles were selected according to specific inclusion and exclusion criteria to ensure appropriate selection of the literature. After evaluating all of the articles published about TADs and after the application of the inclusion criteria, 26 articles were considered to be suitable for review.

Data analysis
An evaluation of the methodological soundness of each article was performed, even if it was not used as a criterion for the inclusion of the studies in the review.
For each study, four variables were considered: definition of success, configuration of the study, description of the analysis, and clinical explanation of the results. Each variable was valued with 1 point if the descriptor was complete, with 0.5 points if the descriptor was partially fulfilled, and with 0 points if the request was not fulfilled or not mentioned. As the aim of our review was to evaluate factors that affected the success and failure rate of TADs (Juni et al., 2001;Higgins and Green, 2008;Reynders et al., 2009).
The total score of the studies was calculated, and the articles were classified as being high quality (H) if the overall score was 3 or more points, medium quality (M) if the score was between 2 and 3 points, and low quality (L) if the score was under 2 points (Table 1).

Success rate
The average success rate was greater than 80% in all of the studies.
In the articles that included analyses of different types of TADs, the individual success rate is specified for every type of TAD: some of the factors that influenced success rates were common to several articles, but other factors were only evaluated by a single work (Table 2). Figure 2 shows that there is no difference (p=0.60) in TAD failure rate between males and females. Funnel plot symmetry suggest a low risk of publication bias presence. Figure 3 suggests that age could be considered a factor that influence the success of TADs, which is lower (p=0.02) in patients under the age of 20 years, even if funnel plot asymmetry notice the possible presence of publication bias. An explanation for this finding could be due to the stability of TADs, which need mechanical retention. Bone density and the thickness of cortical bone play major roles in mechanical retention; thus it stands to reason that for older patients, retention should be better because bone density is higher, which results in fewer failures. Besides Schätzle et al. (2009) in their review suggest that in adolescent patients the use of compliance dependent appliances could be preferred to the use of TADs allowing for growth modification of patients that TADs cannot influence.

Patient related factors
Supplementary Figure 2A, available online, seems to suggest that there is no difference (p=0.41) in TAD failure rate placed in keratinized versus non-keratinized tissues. Actually when removing the study by Sharma (2011), which is the only one that found higher success rate in presence of non-keratinized tissues, forest plot analysis supports the presence of a significantly (p=0.007) higher success rates when TADs are placed in attached gingival tissues. Limited number of available studies and funnel plot asymmetry suggest to avoid speculative assumptions regarding this prognostic factor. Figure 4 report a significant (p<0.005) higher success rate when the soft tissue around TADs is not inflamed: peri-implant inflammation, due to poor hygienic care, is a factor that lead to implant failure. Nevertheless, even in the absence of keratinized mucosa, health of the tissues around the implants can be maintained, provided that oral hygiene is correctly performed (Francetti et al., 1997). Better hygiene is often achieved on the left side of the mouth in right-handed patients, who constitute most of the population (Tezel et al., 2001). Park et al. (2006) stated that TADs placed in the left side of the mouth exhibited higher success rates than those placed on the right side. We were not able to perform a meta-analysis, as fewer than 5 studies were available for the following 'patient related factors': smoking, anatomical site (maxilla versus mandible) and proximity of TADs to adjacent roots..
Smoking was associated with an increased risk of complications, as it causes mucositis, periimplantitis and implant loss. Bayat et al. (2010) found that heavy smokers exhibited higher failure rates in orthodontic mini-screws than non-smokers or light smokers.
Bone drilling is a commonly used step before the insertion of dental implants and of miniimplant TADs. The drilling process generates heat that impairs the turnover activity of bone tissue by causing hyperemia, necrosis, fibrosis, osteocytic degeneration, and increased osteoclastic activity. Bone temperature must be below 47° C during drilling to avoid thermal osteonecrosis.
Friction of the drill may result in generation of heat in dense and thick cortical bone. The cortical bone of the mandible is thicker than the maxillary bone; thus, drilling may result in overheating the mandible and this risk appears higher for the lower jaw than the for the maxilla. This however would only apply for buccal insertion of TADs in the maxilla and does not explain the high success rate of TADs in the palate.
The placement of TADs in interradicular sites is often necessary for specific anchorage requirements. Safe insertion of TADs in interradicular sites can be performed on the buccal side with adequate bone-implant contact anywhere within the zones of attached gingiva, up to 6 mm apical to the alveolar crest thus allowing for enough interradicular space (Lim et al., 2007). Kim et al. (2010) revealed that root proximity itself is not a major risk factor for mini-implant failure. Root contact does not necessarily cause implant failure because the other implant surfaces in contact with the bone can ensure stability, especially if the implant-root contact area is stable during treatment.
On the other hand, mini-implants contacting adjacent roots on more than one side exhibited more failures, which were most likely caused by decreased bone-to-implant contact.
Operators need to remember that contact between root and TAD could potentially damage the root itself, although Janssen et al. (2008) demonstrated histologically almost complete repair of the damaged periodontal structure within 12 weeks following the removal of the screw.

Implant-related factors
Supplementary Figure 3, available online, shows that there is no difference (p=0.09) in TAD failure between mini-screws and mini-implants longer than 8mm, even if funnel plot asymmetry showed risk of publication bias and visual forest plot analysis possibly suggest a better result for longer TADs.
Supplementary Figure 4, available online, shows that there is no difference (p=0.48) in TAD failure rate for mini-implants with a diameter of more than 1.3mm. The slight asymmetry of the funnel plot indicates a possible risk of publication bias.
It was inappropriate to undertake a meta-analysis, because of the availability of less than 5 studies, for the following implant-related factor: thread shape, TAD surface and use of plates versus mini-implants.
The thread shape was evaluated to analyze the relationships between geometric characteristics and mechanical properties of TADs. It was found that thread shape correlates significantly with maximum insertion torque, as analyzed with pull-out tests (Migliorati et al., 2012a;Migliorati et al., 2012b;Migliorati et al., 2012c): the removal torque of TADs after use is considered an element that is correlated with TAD stability. Some articles evaluated different type of TADs, including mini-plates that used multiple screws for anchorage. Mini-plates exhibited a greater stability compared with mini-and micro-screws, but they required flap surgery for insertion and removal, which could cause swelling and discomfort. Chen et al. (2008) suggested that mini-plates exhibited a higher success rate if the insertion site was outside the alveolar area and if they were secured with multiple screws. Kuroda et al. (2007) used two types of mini-screws that were different in size, in addition to a mini-plate fixed with three small mini-screws. They found no differences in the success rates of the three types of TADs; they affirmed that the type of TAD does not affect the success rate, but they recommended the use of smaller mini-screws when the implant site allows, as these can be inserted without a flap. Figure 6 shows that there is no difference (p=0.65) in TAD failure rate between early (less than 4 weeks) and delayed (4 weeks after insertion) orthodontic loading. Funnel plot symmetry indicates a low risk of publication bias presence.

Management related factors
Meta-analysis could not be performed as less than 5 studies were available that investigated the following 'management-related factors': type of surgery, pre-drilling, operator experience, type of loading, type of connection to the implant and type of movement.
The type of surgery used for the insertion of the TADs is considered to be important for patient comfort. TADs can be inserted with a flap surgery, in which the mucoperiorsteal flap is reflected to expose the cortical bone and then sutured after insertion. On the other hand, flapless surgery can be performed without a mucoperiosteal incision, crafting the screw holes with a round cutter or punch and then pre-drilling the bone. Kuroda et al. (2007) considered two different clinicians and the two types of surgical procedures: the success rate of the two groups of mini-screws were very similar, and the different operators did not influence the success rate of mini-screws. Screws inserted without flap surgery were more comfortable for patients. Cornelis et al. (2007) in a review based on animal studies, looked at success rate of pre-drilled mini-implants compared to directly inserted screws and the diameter of the pilot hole. In cases were diameter of the hole was narrower than the diameter of the screw, no failures were reported.
Other studies looked at the success rates of TADs in relation to insertion by different clinicians.
Most stress the importance of a learning but some conclude that the failure rate is implant-specific (Luzi et al., 2007;Chen et al., 2008;Jung et al., 2012) and thereby conclude that different operators do not affect success rates.
The type of movement, the treatment target, and the appliances used for loading may be correlated to success rates. Park et al. (2006) analyzed the method of force application to screw implants, which were one of the following type: power chain, super-elastic thread, nickel-titanium coil spring, or ligature tie-back. They found no significant correlation between the success rate and the method of force application. Jung et al. (2012) analyzed palatal implants for orthodontic treatment and used different connective systems between palatal implants and teeth using orthodontic forces that ranged from 1 to 6 N. Their data showed that the supra-construction design, the direction of loading and the force applied on the implant had no influence on implant stability.
When the type of movement is considered, various studies agreed that the direction of tooth movement is important in determining the success rate, but the articles were inconsistent in identifying unfavorable direction of load. Kuroda et al. (2007) evaluated the type of tooth movement for which the TAD was used. They analyzed the success rate of mini-screw implants used for retraction, protraction and intrusion of teeth. The implant used for intrusion in the posterior maxilla and mandible exhibited significantly lower success than the ones used for other orthodontic indications. This result could be due to the type of bone, the higher risk of peri-implantitis in the maxilla, technical difficulties, and obstacles to oral hygiene in the mandible.

Conclusions
The conclusions of this analysis must be interpreted cautiously because of the disparate nature of the studies reviewed and the heterogeneity of the data: success was not equally defined in the papers scrutinized in this investigation and it would therefore be inappropriate to try and give an exact figure for implant stability and success and the factors that may have an impact on the figures.
Furthermore, few studies commented if treatment goals could have been achieved without the use of TADS, but using other means of providing orthodontic anchorage instead.
However, some general conclusions can be drawn from the analysis of the data investigated: 1. In the studies analyzed TADs were only one of several anchorage options available: they were utilized mainly because they were a compliance free method of providing anchorage and were often less bulky than alternative anchorage devices.