Periodontal results of different therapeutic approaches (open vs. closed technique) and timing evaluation (< 2 year vs. > 2 year) of palatal impacted canines: a systematic review

This review evaluates, as a primary outcome, which surgical technique (open vs. closed) and which type of material used for the auxiliaries (elastic vs. metallic) were preferable in terms of periodontal results during the treatment of palatal-impacted canines. The timing of the evaluation of the results was also assessed as a secondary outcome. An electronic search of the literature up to March 2021 was performed on Pubmed, MEDLINE (via Pubmed), EMBASE (via Ovid), Cochrane Reviews and Cochrane Register of Controlled Trials (RCTs) (CENTRAL). The risk of bias evaluation was performed using version 2 of the Cochrane risk of bias tool (RoB 2) for RCTs and the ACROBAT NRSI tool of Cochrane for non-RCTs. 11 articles met the inclusion criteria. Only one RCT was assessed as having a low risk of bias and all the non-RCTs were assessed as having a serious risk of bias. This review revealed better periodontal results for the closed technique and metallic auxiliaries. In addition, it revealed that the timing of the evaluation of the results affects the periodontal results with better results obtained 2 years after the end of treatment. In the treatment of a palatal-impacted canine, the closed technique and metallic auxiliaries should be preferred in terms of better periodontal results. The timing of the evaluation of the results affects the periodontal results.

Concerning disimpaction treatment, there is no adequate literature on the different systems of force application and there is no quantitative analysis of these variables which can influence the results of the treated teeth, the contiguous teeth and periodontal health. During traditional fixed orthodontic treatments, the type of material used may have a different impact on the periodontal status, so this specific variable should be considered as a key factor in periodontal and dental results [12][13][14].
Furthermore, no previous systematic reviews reveal how the timing of the assessment can affect the results.
The aim of this study is to assess which therapeutic approach is preferable for palatal-impacted canines through a systematic review comparing the "open" and the "closed" surgical techniques and the system of forces application (elastic or metallic auxiliaries) (primary outcome). In addition, the influence of the timing of periodontal results evaluation was also assessed: results evaluated within 2 years from the end of treatment were compared with results assessed after 2 years from the end of treatment (secondary outcome).
The review was based on the PRISMA checklist [15].

Protocol and registration
Not available.

Eligibility criteria
The following criteria were selected in this study.

Types of participants (P)
Orthodontic patients with unilateral, palatal impacted canines were included, with no age, race or type of malocclusion restrictions.

Types of interventions (I)
Three variables were analysed: 1. Surgical technique 2. Force application system 3. Timing of periodontal results evaluation.

Comparisons (C)
For the three aforementioned types of intervention: 3. Periodontal results evaluation: < 2 years after the end of treatment vs. > 2 years after the end of treatment.

Included study types (S)
Randomized Controlled Trials (RCTs), Quasi-RCT (Q-RCTs), Controlled Clinical Trials (CCTs), unclear Non-Randomized Studies (uNRS), Prospective and Retrospective Studies were included in this study. Case reports were excluded. Eligibility criteria comprised of: only articles published in English after 1990; description of periodontal results and description of the system of forces application included.
Exclusion criteria were the period of publication, non-English language, studies that did not report adequately periodontal results or the description of both surgical approach and forces application system.

Search strategy
The studies were identified by bibliographic research of electronic databases, examining the bibliography of the articles.
The bibliographic research was carried out on Pubmed, MEDLINE (via Pubmed), EMBASE (via Ovid), Cochrane Reviews and Cochrane Register of Controlled Trials (CENTRAL).
The research was performed up to March 2021. The search was carried out using a combination of controlled vocabulary and free text terms and Boolean operators.
The free text keywords used for PubMed are shown in Table 1.

Study selection
The selection of the studies was carried out independently by two of the Authors (S.B. and R.G.). The degree of accuracy and agreement between the two authors was assessed using Cohen's kappa coefficient (κ). Any disagreement was resolved by a third Author (F.A.). Potentially adequate studies were initially identified through the evaluation of the title. Abstracts of the non-excluded studies were read and studies that did not match the eligibility criteria were eliminated. At the next stage, full texts were examined and items that did not match the eligibility criteria were excluded.

Data collection process
The data were collected into an excel file and then reviewed by two of the Authors (S.B. and R.G.). Any disagreements were resolved through comparison with a third Author (F.A.).

Data item
Information regarding the studies is shown in Table 2.

Risk of bias assessment
The risk of bias was assessed independently by two Authors (S.B. and R.G.). Any disagreement was resolved by two other Authors (C.G. and F.A.). The study quality of randomized and quasi-randomized studies was determined using version 2 of the Cochrane risk of bias tool (RoB 2) to the domains shown in Table 3 [16]. The result can be 'Low' or 'High' risk of bias or can express 'Some concerns' .
The non-randomized trials were assessed by ACRO-BAT NRSI tool of Cochrane to the domains shown in Table 4 [17].
Possible results for each domain and for the overall results were: 'low' , 'moderate' , 'serious' , 'critical' risk of bias and 'no information' .

Effect measures and synthesis methods
Mean values and standard deviations (DS) were used in order to express the estimate of effect. Where standard deviations were missing, they were calculated and then all the values were reported in tables shown in the additional files (Af). Specifically, Additional file 1: Table S1.Af reports periodontal indices according to the surgical technique, Additional file 1: Table S2.Af reports periodontal indices according to the system of force application, Additional file 1: Table S3.Af reports periodontal indices regarding the surgical technique, classified on the results of the evaluation timing, Additional file 1: Table S4.Af reports periodontal indices regarding the system of force application, classified on the results evaluation timing.

Study selection
A total of 3217 studies were identified through the database search. 1651 of these studies were eliminated because the title was not related to the research that was being carried out. After the duplicates were deleted, 297 items remained. Abstracts were read and 213 articles were excluded because they did not meet the inclusion criteria: 53 articles were deleted because they were published before 1990, 93 articles were excluded because of the study type, 41 articles were excluded due to the therapeutic approach type, 26 were excluded because it was not specified whether the results were for palatal or buccal impacted canines.
The full text of the remaining 84 articles were examined. 69 were excluded because they did not match the eligibility criteria: 4 of them were excluded because the full text was not in English, 39 studies did not report periodontal results or the force application system. 30 articles did not report the force application system.      Following the application of the eligibility criteria, 11 articles remained. The selection procedure is represented in Fig. 1.
Cohen's kappa coefficient (κ) was 0,857 showing an excellent strength of agreement amongst the Authors.
Parkin's et al. [18] study compares subjects treated with the open and closed technique. The force application system was not adequately explained so it was not considered. Smailiene's et al. [19] study compares open and closed techniques to each other and against contralaterals, but in this study the closed technique-treated group was excluded from the quantitative and qualitative results assessment because it was followed by a spontaneous eruption which is not the subject of this study. The force application system used was metallic. Mummolo's et al. [10] study compares subjects treated with the open technique and elastic force application system to the contralaterals. In Zafarmand and Gholami's [20] study subjects treated with open technique and an elastic force application system are compared to the contralaterals. In Szarmach's et al. [21] study subjects treated with open technique and metal force application system with contralaterals were compared. Bollero's et al. [22] study compares subjects treated with closed technique and elastic forces application system to the contralaterals. Caprioglio's et al. [23] study compares impacted canines treated with closed technique and force application system using metal ligatures with the untreated contralaterals. Evren's et al. [24] study compares canines treated with closed technique and contralaterals not specifying the type of force application system used for disimpaction. In Zasciurinkiene's et al. [25] study subjects treated with closed technique and using a metallic auxiliary to apply the force system were compared to the contralaterals. Crescini et al. [26] compares subjects treated with closed technique and with an elastic force application system with the contralaterals. Hansson  Further information about study characteristics is collated in Table 2.

Risk of bias
The results for randomized and quasi-randomized studies are shown in Table 3.
Regarding the sequence generation and the allocation concealment Smailiene's et al. [19] study was at high risk of bias because it described an inappropriate method of random sequence generation. It is impossible to blind participants or the dentist to the surgical procedure, but assuming that the surgeon was equally experienced at both techniques, it unlikely that this would introduce a high degree of bias due to deviations from the intended interventions. For this reason, the studies were judged to be at low risk of bias for this domain. Since missing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups, the studies were judged to be at low risk of bias for this domain. RCTs studies were judged to be at low risk of bias in measurement of the outcome because the statement about primary or secondary outcomes was satisfactorily clear. Regarding risk of bias in selection of the reported result, all studies were considered at low risk.
Regarding the non-randomized trials, the results are shown in Table 4.
For each separate outcome different confounders were considered for the assessment of risk of bias such as the initial position and the angle of canine (orientation, inclination and depth of impaction), oral hygiene, initial periodontal condition, age and gender. All the non-RCTs were assessed as having a serious risk of bias.

Synthesis of results
Parkin's et al. [18] study reports no difference between canines exposed through an open versus a closed surgical technique, but it reported a statistical difference in certain periodontal indices when comparing operated and contralateral (unoperated) canine. It concludes that there is no difference in periodontal health for impacted canine treated with either the open or closed technique. In Smailiene's et al. [19] study it was reported that there were no significant differences with  [20] study they did not report a significant difference between the periodontal status of the 2 groups except for the level of alveolar bone that was significantly lower in the surgically-exposed group. Szarmach's et al. [21] showed an overall inferior periodontal result in the experimental group compared to the control group, especially in specific periodontal sites. In Bollero's et al. [22] study, palatally impacted canines exhibited a significantly greater PD on the mesiolingual site when compared to their contralaterals while the other periodontal results were not significant. Caprioglio's et al. [23] study reports that the use of a closed-flap surgical technique in association with a codified orthodontic traction system revealed no significant clinical differences regarding periodontal indices compared to the control group. Evren's et al. [24] study, reports that palatally impacted canines treated with the closed technique had worse periodontal indices values compared to their contralaterals. Zasciurinkiene's et al. [25] showed that a significant increase in pocket depth was found at specific sites of the canine after surgicalorthodontic treatment, but the periodontal conditions were considered clinically acceptable. The Crescini et al. [26] study reported that no significant differences in the periodontal indices were observed between test and control teeth at the follow-up examination. Hansson and Rindler's [27] study results showed good periodontal status with slight differences between treated and untreated teeth.
In this study, the Authors used the mean values of the periodontal indices to compare them. Due to the heterogeneity of the available data, it was impossible to perform an appropriate meta-analysis, therefore a qualitative evaluation was performed based on the values shown in the additional files.
Regarding the surgical approach it was possible to note a higher PD value in the open technique than in the closed one, a lower KT value in the open technique than in the closed one, a higher CAL value in the closed technique than in the open one, a higher REC value (based on Crown Length difference) in the open technique than in the closed. According to the system of force application, the following outcomes were found: a higher PD using elastic auxiliaries than metallic ones, a smaller amount of KT using elastic auxiliaries than using metal, a higher PI using elastic auxiliaries than using metal auxiliaries, a higher REC using metal auxiliaries than using elastic ones, a higher CAL using elastic auxiliaries than using metal auxiliaries.
Based on the timing of the results evaluation, the following results were found: a higher PD for evaluations made within 2 years of the end of the orthodontic treatment when compared to those made after 2 years, a lower amount of KT for evaluations made within 2 years than for those made after 2 years, a higher REC for evaluations made within 2 years than for those made after 2 years, a higher CALs for evaluations made within 2 years than those evaluated after 2 years.

Summary of evidence
This study included a total of 280 treated palatal impacted maxillary permanent canines. Regarding the comparison between open and closed technique, the qualitative assessment of the periodontal outcomes showed overall better results using a closed technique; this is in accordance with some results of Wisth et al. [28], who reported better results for closed technique in terms of PD and CAL, but unfortunately this study was retrospective with a high risk of bias. The better results of the closed technique could be due to the following factors: better preservation of the CEJ, better soft tissue healing, less plaque accumulation and better post-operative comfort [29,30]. Diversely, the results found are in contrast with previous studies where no differences were found when comparing these two surgical techniques [12,18,19,31]; furthermore, even if there were differences in periodontal indices, these were not significant.
There is a lack of literature regarding the system of force application. In fact, in the included articles there are no comparative studies regarding the use of either elastic or metallic auxiliaries. Therefore, the analyses done in this study are based on comparisons made between different studies. This study compares elastic and metallic auxiliaries and shows how the use of metallic auxiliaries gives better periodontal results, independently from the surgical technique used. It is probably associated with the lower quantity of plaque accumulation using metallic auxiliaries as the lower PI mean value of this study confirms, in accordance with other studies [32,33].
Regarding how the timing of results' evaluation affects results, the present study shows better results after a minimum of a two-year period after the end of treatment, independent of the type of surgical technique used. The retrospective study of Crescini et al. [26] analyzed the results at the baseline (at the end of the treatment) and after a three-year period and they found that there are no relevant differences in periodontal status except for the average PD value which decreased over time. However, this study has a high risk of bias; furthermore it evaluated both palatal and buccal impacted canines (multiple comparisons that increase test sensitivity) and it is a retrospective study which has lower scientific evidence compared to the systematic review [34]. The fact that periodontal indices are better after a 2-year period following the end of the treatment could be associated with the physiological regenerative capacity of the tissue, as Crescini et al. [26] mentioned in their study, which also demonstrated that most of the PD increased values found at the baseline, were actually due to pseudo-pockets.

Limitations
This systematic review has the following limitations: • There was no differential assessment based on the age of the participants • There is high heterogeneity of the results • The samples size is quite limited in certain studies In addition, an important limitation is that in the closed technique, the type and the extension of the flap was not accurately evaluated, and an assessment of the instrumentation used for the open technique was not carried out.
Another limit is that the auxiliaries used for the force application system were assessed based only on the type of material (elastic or metallic) and not on a specific type of auxiliary which can affect the periodontal and dental results. 2. Metallic-auxiliaries offer a better post-treatment periodontal condition. 3. Results evaluation timing influences the outcomes.
Specifically, outcomes evaluated at least 2 years after the end of the overall orthodontic treatment are better than the outcomes evaluated up to 2 years posttreatment.
Finally, more high-quality studies should be performed in the future to enable stronger evidence. In the future it would be also interesting to investigate how skeletal anchorage, associated with both open and closed technique, affect periodontal indices of palatal impacted canines.