Effects of skeletal anchorage on mandibular rotation with fixed functional appliance therapy in class II malocclusion patients – A systematic review and meta-analysis

Protocol registration

This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. The review protocol was registered under the PROSPERO database (CRD42021256690).

Eligibility criteria

Information sources, search strategy, and study selection

A literature search was performed using the database from PubMed, Central of the Cochrane library, Clinical Trials Registry, and Google Scholar. The database was searched till May 2022 with no specific filter applied during the search. Gray literature also was searched using a combination of MeSH terms and respective keywords. By looking through the included articles’ references, a further search was conducted. All articles were searched using the combination of keywords such as FFA, skeletal anchorage, and orthodontics with Boolean characters “AND” and “OR” combination.

The titles and abstracts of all relevant articles were screened by independent reviewers for inclusion. A study was judged eligible when it included the intervention of FFA with skeletal anchorage versus conventional FFA. Disagreements were resolved by discussion among the reviewers.

Data collection process and items

Data collection was performed using a customized data extraction form: (1) Author and year of study, (2) country, (3) type of study, (4) population – sample size, age, and gender, (5) intervention – FFA, skeletal anchorage, site of anchorage site, and failure rate, (6) comparator group – FFA, (7) treatment duration, and (8) outcomes [Table 1].

Risk of bias

To evaluate the risk of bias in individual studies, different tools were used.

For RCTs, the assessment was conducted using the recommended approach in Cochrane Reviews using the software RevMan 5.4.1. The two-part tool was used to address the six specific domains (sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting, and other bias).

For non-RCTs, the assessment was conducted using the recommended approach for assessing the risk of bias using the risk of bias in the non-randomized studies-of-interventions (ROBINS-I) tool. The two-part tool was used to address the seven specific domains (Bias due to confounding, selection of participants, classification of intervention, deviations from intended interventions, missing data, measurement of outcomes, and selection of the reported result).

Meta-analysis

The meta-analyses, using the random effects model, were applied with RevMan 5.4 (RevMan 5.4, The Nordic Cochrane Centre, Copenhagen). Data on the mean difference (MD) and standard deviation were obtained from selected studies. Heterogeneity was assessed by a Q test and quantified with I² statistics. Mandibular rotation was considered the main outcome. In addition, a funnel plot was performed to evaluate the publication bias based on a sufficient number of included studies.

Study selection and characteristics

PRISMA guidelines were followed to scrutinize the articles as detailed in [Figure 1]. The database search and reference checks turned up a total of 186 documents. Ninety-seven records were chosen after the elimination of duplicates. After screening the titles and abstracts, 76 were excluded and 21 articles were selected. After further careful full-text review, nine studies were excluded due to different study designs, interventions, and comparators. Ultimately, 12 trials fulfilled the inclusion criteria and were selected.

Close

Figure 1:

Preferred reporting items for systematic reviews and meta-analyses flow diagram.

Export to PPT

Among the included studies, eight were RCTs^[5-12] and four were controlled clinical trials (non-RCTs).^[13-16]

Participants – There were 460 patients treated for Class II correction in included trials. Out of 460, 211 were treated using FFAs with skeletal anchorage, 190 using FFAs alone, and 59 served as untreated controls.

FFA used – Among the included studies, three investigated the effects of Herbst,^[5,13,15] seven studies assessed the effects of Forsus fatigue resistant device (FRD),^[6-11,16] one study assessed the effects of Advan-sync2,^[12] whereas one study assessed the effects of Forsus versus Herbst.^[14]

Skeletal anchorage used – In five studies, miniplates were used^{[8,10,11,14,16]} and in seven studies, miniscrews were used.^{[5-7,9,12,13,15]}

Outcomes – All studies evaluated skeletal and dentoalveolar changes before and after treatment. One study by Gandedkar et al.^[16] evaluated changes using cone beam computed tomography. The remaining 11 studies had data from conventional lateral cephalograms.

Risk of bias results

While a majority of RCT studies showed a low risk of bias for research and few studies with questionable and high risk of biases. Five domains were assessed – including “bias in the randomization process (selection bias), “bias due to deviations from intended interventions,” “bias due to missing outcome data (attrition bias),” “bias in outcome assessment,” and “bias in selection of the reported result” [Figure 2].

Close

Figure 2:

Risk of bias summary for the randomized clinical trials.

Export to PPT

For the CCTs (non-RCTs), the risk of bias was performed following the ROBINS-Itool. Among non-RCTs, two studies by Manni et al.^[13,15] showed a high risk of bias in domains of classification of intervention, bias due to deviation from intended outcome, and measurement of outcome. Other studies showed questionable and moderate risk of bias in few domains [Figure 3].

Close

Figure 3:

Risk of bias summary for the non-randomized clinical trials.

Export to PPT

Results of meta-analysis

For analyses, if the test showed substantial heterogeneity (I² > 50%), a random effects model was applied, or else (I² ≤ 50%), a fixed effects model would be used.

The meta-analysis was performed on 10 studies that qualified with the required data outcome that could be analyzed quantitatively. Out of these 10 studies, one study by Manni et al.^[13] had two sub-groups; hence, data from these subgroups were considered as if it was data from two studies for analysis.

The other two studies^[5,15] were excluded as the data reported could not be analyzed (data were not in MD and SD).

The results of the overall comparison have been depicted as a forest plot [Figure 4].

Close

Figure 4:

Forest plot depicting results of meta-analysis of mandibular rotational changes.

Export to PPT

Eleven sub-groups (studies) were included in the meta-analysis comparing mandibular rotation between FFAs with skeletal anchorage and conventional FFA alone. With the meta-analysis conducted for selected studies, heterogeneity was more than 50% (I² = 83%); hence, the random effect model was applied.

Results showed that the mandibular rotation was more among FFA with skeletal anchorage group as compared to only the FFA group, with an MD of 0.39 (95% CI = −0.27–1.06; Z value = 1.16); However, the difference in mandibular rotation among two groups was statistically non-significant (P = 0.25).

Reporting bias assessment

The funnel plot shows the effect estimates of the included studies against their measure of precision or size of the studies. The shape of the funnel plot was asymmetrical between the right and the left sides, which means that there may be publication bias due to the absence of smaller-sized studies on the left side of the plot. The funnel plot showed asymmetry indicating heterogeneity and possible publication bias [Figure 5].

Close

Figure 5:

Funnel plot.

Export to PPT

Mandibular rotation

Meta-analysis showed that mandibular rotational changes with or without the use of skeletal anchorage were similar. Studies by Kochar et al.,^[11] Manni et al.,^[13] and Turkkahraman et al.^[8] found that mandibular rotation was more in conventional FFA. However, seven studies^{[6,7,9,10,12,14,16]} found that mandibular rotation was more in the skeletal anchored group than conventional FFA. When a forest plot was made using data from these studies, it showed that mandibular rotation was more in the skeletally anchored FFA group by 0.39°. However, this difference between the groups was statistically insignificant. The findings are similar to that of a review by Arvind and Jain.^[3] where only four studies of Forsus FRD were included in the study.

Secondary outcomes assessed in this review were skeletal parameters – SNA angle, SNB angle, and ANB angle, which depict sagittal relations of the maxilla, mandible, and inter-jaw relationship. Nine studies that reported skeletal effects (angles SNA, SNB, and ANB in degrees) were selected for qualitative analysis.

Strengths and limitations

This is a comprehensive review based on the latest clinical trials. The data were retrieved from electronic database. No filter was applied regarding the timeline. Only a language filter was used to remove evidence that was not available in English. All included trials used the same intervention and comparator, as specified in the inclusion criteria, which improved the data comparability. All studies were considerably recent. This made the conclusions clinically relevant and applicable.

However, some limitations must be mentioned. In all included studies, sites of TSADs placement were not uniform. Radiographic facilities and parameters were not uniform. Furthermore, manual tracing of lateral cephalograms could not be rendered error-free. Heterogeneity existed among the included studies. Several patient-related factors such as gender, race, and skeletal maturation stages could not be formally assessed in this review.