Mandibular effects of temporary anchorage devices in Class II patients treated with Forsus Fatigue Resistant Devices: A systematic review

An electronic search, without language restriction, was undertaken on the 20th February, 2020 in the following databases: PubMed, Embase, Cochrane Central Register of Controlled Trials (CENTRAL), System for Information on Grey Literature in Europe (SIGLE), ProQuest, Scopus, Web of Science. Details about the search method are presented in Table I. A complementary manual search was performed in relevant journals, including The Angle Orthodontist, the American Journal of Orthodontics and Dentofacial Orthopedics, the Korean Journal of Orthodontics, the European Journal of Orthodontics and the Journal of Orthodontics.

Trials meeting the following criteria were determined as eligible: (1) Participants: Angle Class II maloc-clusion patients in active growth; (2) Intervention and controlled protocol: comprehensive orthodontic treatments using a FFRD with and without TADs (miniscrews, mini-implants or miniplates); (3) Outcomes: orthodontically induced skeletal, dento-alveolar and/or soft tissue changes; (4) Study design: randomised controlled trials (RCT) or prospective clinical controlled trials (CCT).

After eligibility screening and inclusion, a customised form was employed for data extraction. Relevant information was collected, including the first author name, publication time, study design, participants’ characteristics, grouping detail, surgical procedure, treatment protocol, measurement modality, sample loss and outcomes. The study authors were contacted for confirmation whenever necessary.

The bias risk of included RCTs was evaluated using Cochrane Collaboration’s Risk of Bias tool, with the following seven domains taken into consideration: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting and other bias. For each domain, the risk of bias was judged as high, unclear, or low. Overall, the included RCTs were categorised as low risk (if all domains were assessed as low risk), unclear risk (if domains assessed as unclear risk ≥1), or high risk (if domains assessed as high risk ≥1). ¹⁹

In addition, the evaluation of non-RCTs was conducted using the Newcastle-Ottawa Scale, identifying high quality choices with a ‘star’ based on three items: selection (four stars at most), comparability (two stars at most) and outcome (three stars at most). Generally, a non-RCT would be categorised as high quality for seven to eight stars, fair quality for five to six stars or poor quality for fewer stars. ²¹

The clinical heterogeneity and the statistical heterogeneity were considered across the studies. A meta-analysis was planned when clinical and statistical homogeneity were sufficient, otherwise the results would be summarised qualitatively. Review Manager 5.3 (Nordic Cochrane Centre, Cochrane Collaboration, Copenhagen, Denmark) was employed for the conduction of the meta-analysis. Outcomes, including data changes of angle, length and distance, were statistically pooled as continuous variables and adopted as effect measurements. A P value of 0.05 was considered as the threshold for statistical significance.

The overall quality of evidence was evaluated according to the risk of bias, inconsistency, indirectness, imprecision and other considered factors across the included studies, following the Grading of Recommendations Assessment, Development and Evaluation (GRADE) guidance. ²²

The electronic database search initially yielded 762 articles, with no additional studies identified through the manual search. After the removal of duplicates, titles and abstracts of 588 papers were subsequently screened. Of these, full texts of the remaining 15 studies were obtained and evaluated according to the eligibility criteria. Finally, six studies qualified for inclusion. Details are presented in Figure 1.

Figure 1.

Of the six included studies, four ^23-26 were regarded as RCTs and the other two ^27,28 were CCTs. In total, 161 circumpubertal patients undergoing comprehensive orthodontic treatment were involved, of which 85 participants were treated with a TAD-anchored FFRD and 86 subjects received conventional FFRD. Five studies ^24-28 only assessed patients presenting with a Class II division 1 malocclusion, while the study of Aslan et al. ²³ recruited seven division 2 patients (three in a TAD-anchored group and four in a conventional group) as well as 26 division 1 patients. The average treatment duration of the trials ranged from 4.86 to 10.45 months. Miniscrews were the predominant TAD device employed in three RCTs as indirect anchorage and miniplates were used in one RCT and two CCTs as direct anchorage. A CBCT and lateral cephalogram were used to measure the outcomes in three studies. For the outcomes measured, skeletal and dento-alveolar changes were reported in all six studies, while soft tissue changes were only reported in three. ^23,24,27 General information and the intervention details related to the research are summarised in Table II and Table III. In addition, since some of the treatment procedure details were not specifically described in the publication of Gandedkar et al., ²⁸ confirmation was sought from the researchers but a response was not available.

The risk of bias assessment for RCTs is presented in Figure 2. Of the four RCTs, two studies ^25,26 were evaluated as low risk while the other two ^23,24 were deemed an unclear risk of bias. The study of Aslan et al. ²³ was assessed as an unclear risk in the domain of selection bias and detection bias, since the specific methods of random sequence generation, allocation concealment and blinding of outcome assessment could not be determined. Meanwhile, the study by Eissa et al. ²⁴ was assessed to have an attrition bias because one sample was lost in the conventional group even though detailed information was reported. Although blinding of participants and personnel in the trials was impossible since the TADs were obviously identified during treatment, the deficiency may not impact the therapeutic outcomes because the surgical and orthodontic procedures were strictly performed as developed in all trials. Therefore, the performance bias in enrolled studies was judged as low by the reviewers. ¹⁹

Figure 2.

The risk of bias assessment for CCTs is summarised in Table IV. Eight stars were assigned to the study by Turkkahraman et al., ²⁷ which was assessed as high quality. Six stars were given to the study by Gandedkar et al.; ²⁸ however, the study was assessed as poor quality by reviewers because of the lack of key information.

Meta-analyses could not be justified due to the great clinical heterogeneity across the included trials. Therefore, the outcomes of skeletal, dento-alveolar and soft tissue changes were systematically summarised instead. The definitions of cephalometric values are presented in Table V.

Results of GRADE assessment for the overall quality of available evidence are summarised in Table VI. The evidence quality of skeletal outcomes, including mandibular length and mandibular rotation, was assessed as moderate. The quality of evidence associated with lower incisor inclination was also moderate, whereas very low quality was identified for soft tissue positional change.

The aim of conducting the present systematic review was to determine whether TADs could enhance the skeletal, dento-alveolar and soft tissue effects of a FFRD on the mandible in growing Class II patients. In all, six studies were included in the review. The results were not quantitatively but, rather, qualitatively synthesised due to the great clinical heterogeneity, such as the variance in the bracket bonding protocol, the TAD clinical protocol, the evaluation indicators and criteria for ending treatment. Moderate-quality evidence was acquired for the change of skeletal outcomes and lower incisor proclination, whereas very low-quality evidence was found for soft tissue change.

As a typical symptom of Class II patients, mandibular deficiency has been widely discussed in previous studies, which have demonstrated that fixed functional appliances, such as a FFRD, might be beneficial. ^5-9 Considering that forces generated from a FFRD may be transmitted more to the mandible by anchorage enhancement, it was hypothesised that TADs may help in enhancing the mandibular skeletal effects of FFRD. ^{16-18,26,28,29} However, controversial results were found in the available studies. Three RCTs using miniscrews reported no significant difference between groups in relation to mandibular length change. ^23-25 In the miniplate studies, Turkkahraman et al. conducted a high-quality CCT and showed that the increase in mandibular length was not significantly different. ²⁷ While Elkordy et al. carried out a low-risk RCT, which reported that a significant increase of 4.05 mm was attained in the miniplate-anchored group. This was ascribed to the forward and downward force transmitted to the condyle and generated by the immediate application of an orthopaedic force on the mandible. ²⁶ Nevertheless, the conclusion should be interpreted cautiously since the evaluation was merely conducted over a short period of time and long-term follow-up was required to exclude the effect of temporary growth acceleration. A similar suggestion resulted from the study of Phan et al., which found mandibular growth was still significant seven months after Herbst appliance treatment. ³⁰ Interestingly, Gandedkar et al. reported that the increase was significant, not only upon the removal of the FFRD, but also at a one-year post-treatment review. However, the study was assessed to be of poor quality. ²⁸

A significant increase in mandibular backward rotation was achieved by TADs and reported by two low-risk studies employing miniplates. However, this finding was not found in three studies employing miniscrews. An additional poor-quality study employing miniplates showed no significant difference, but it should be noted that there is still uncertainty related to the effects between miniplates and miniscrews. The difference may be due to a greater vertical component of force that could be generated in the miniplate-anchored groups since their attachment site for FFRD pushrods was lower than miniscrew-anchored groups. Moreover, the stability of different TADs should also be taken into consideration. As reported by previous studies, forward movement of 0.4 mm and tipping of 2.65 ± 6.23° affected miniscrews during orthodontic loading, indicating that they were not absolutely stationary. ^31-33 However, Kim et al. showed no miniplate movement by using bone markers as a reference in superimposition, suggesting that miniplates were more stable. It should be further noted that the miniplate heads in some subjects were bent buccolingually, which might also cause anchorage loss. ³⁴ Since anchorage loss is a principal factor limiting the skeletal effects of FFRD, future studies should be designed to investigate whether miniplates might promote mandibular growth better than miniscrews during FFRD treatment.

A further possible factor that might diminish the skeletal effects is proclination of mandibular incisors. ^15,35,36 To manage this common side-effect of a FFRD, research has advocated procedures that counteract lower incisors proclination should be applied during treatment. These clinical techniques might involve the addition of a negative torque to the anterior arch wire or using rectangular arch wires of greater size. ^13,14 However, none have proved to be effective, which promoted the use of TADs. Within the current literature, five studies reported that TADs successfully minimised the proclination of mandibular incisors. ^23,25-28 Noticeably, the mechanism of proclination reduction was heterogenous between miniplates and miniscrews. Employed as direct anchorage, the miniplates provided immediate attachment sites for the pushrods of the FFRD and consequently saved the mandibular incisors from directly receiving the forward force. ^26-28 By comparison, miniscrews did not support the pushrods directly but served as indirect anchorage. Elkordy et al. used segments of 0.019 × 0.025-inch stainless-steel wire to enhance the anchorage of lower incisors by bonding one end to a miniscrew and the other end to the labial surface of the mandibular canine. ²⁵ Similarly, Aslan et al. fixed a 0.018 × 0.025-inch stainless-steel wire segment between the miniscrew slot and vertical slot (0.018 × 0.018-inch) of the canine bracket. ²³ In these trials, the mandibular incisors still directly received the forward force of the pushrods, while the rigid connection to the miniscrews provided resistance. Eissa et al. shared the same protocol as Aslan et al. but no significant difference was reported, and the researchers attributed this inconsistency to the size discrepancy between the segment (0.016 × 0.-inch) and the vertical slot of canine bracket (0.018 × 0.018-inch), which might provide additional play allowing unwanted proclination. ²⁴ Therefore, it is legitimate to extrapolate that TADs are superior in alleviating the proclination of mandibular incisors during FFRD application.

Three studies evaluated soft tissue positional change but only one reported that the protrusion of the lower lip was significantly reduced by the assistance of TADs. ²⁷ Although research has concluded that lip position can be correlated with incisal repositioning, no significant difference was found in the reduction of lower lip protrusion despite minimal mandibular incisor proclination reported in two studies. ^23,24,37 The factors affecting lip position are multiple, thus the effects of lower incisor position should not be determined in a simplistic way. ^38,39 Factors, such as lip thickness, labial tension and soft tissue compensation may also affect lip position. ^40-42

To date, no systematic review focussing on the effects of TADs as an aid to a FFRD has been previously published. In addition to the skeletal, dento-alveolar and soft tissue effects, details about the TADs application protocol (direct versus indirect anchorage) from the included studies were also summarised systematically, which may be useful for clinicians when performing relevant procedures. Elkordy et al. ⁴³ implemented a meta-analysis of Class II patients to verify the effect of TADs during treatment employing fixed functional appliances, in which three publications were noted. However, only two ^23,25 were included in the present study since no applicable data were recognised elsewhere. ⁴⁴ Furthermore, other fixed functional appliances (Herbst, Twin Force bite corrector and Easy-fit Jumper) recruited in the previous meta-analysis were excluded in the present review to reduce the impact of appliance variation on the authenticity of the evidence. Finally, it is justified to synthesise the results qualitatively instead of quantitatively due to the significant clinical heterogeneity.

Although the present study was conducted following standard procedures, several limitations were apparent. Additional high-quality RCTs are needed for a more reliable conclusion since only four RCTs and two CCTs were eligible for inclusion in the present review despite searching the literature without language restriction. Moreover, the effects were not quantitatively measured, requiring future studies to minimise the clinical heterogeneity by developing a suitable standard for treatment process and evaluation. Recommendations regarding the necessity of using TADs in Class II cases treated using a FFRD could not be provided. It would be indispensable to take other aspects into comprehensive consideration, such as patient acceptance, side effects, disease burden and health-economic problems. Unfortunately, little evidence is available at present. ^44-46