Evaluation of orthodontically induced external root resorption following orthodontic treatment using cone beam computed tomography (CBCT): a systematic review and meta-analysis

Evaluation of orthodontically induced external root resorption following orthodontic treatment... Summary Background Orthodontically induced external root resorption (OIRR) is a pathologic consequence of orthodontic tooth movement. However, the limitations of two-dimensional radiography suggest that cone beam computed tomography (CBCT) with its three-dimensional capabilities might be more suitable to assess OIRR. Objective The aim of this study was to assess in an evidence-based manner data on linear or volumetric OIRR measurements of permanent teeth by means of CBCT, during and/or after the end of orthodontic treatment. Search methods Unrestricted electronic and hand searches were performed up to January 2017 in 15 databases. Selection criteria methods Randomized clinical trials, prospective, and retrospective non-randomized studies assessing OIRR during and/or after orthodontic treatment using CBCT in human patients were included. Data collection and analysis After duplicate study selection, data extraction, and risk-of-bias assessment according to the Cochrane guidelines, random-effects meta-analyses, followed by subgroup, meta-regression, and sensitivity analyses were also performed in order to evaluate factors that affect OIRR. Results A total of 33 studies (30 datasets) were included in the qualitative analysis while data from 27 of them were included in the quantitative analysis. Direct comparisons from randomized trials found little to no influence of appliance-related factors on OIRR. Explorative analyses including non-randomized studies found a pooled OIRR of 0.79 mm based on all included studies and 0.86 mm when OIRR was assessed at the end of orthodontic treatment. Statistically significant differences in OIRR were found according to tooth type or jaw, inclusion of extractions, treatment duration, and diagnostic accuracy of the CBCT. Conclusions Based on the results of this study, CBCT seems to be a reliable tool to examine OIRR during or at the end of orthodontic treatment. Although the average OIRR measured with CBCT seems to lack clinical relevance, there are certain factors that may affect OIRR following orthodontic treatment. Nevertheless, due to data heterogeneity and low quality of the included studies, the corresponding results should be interpreted with some caution. Registration PROSPERO (CRD42016030131) Introduction Rationale Orthodontically induced root resorption (OIRR) is a frequent complication of orthodontic tooth movement with complex aetiology and several patient- or treatment-related risk factors being suggested as relevant (1). Proposed patient-related factors include genetics (2), sex (3), age (4), tooth type (5), systemic factors (6), root morphology (7), and history of trauma or previous root resorption (8). On the other hand, proposed treatment-related factors include appliance type (9), treatment duration (10), type of tooth movement (11), applied force magnitude (12), duration of force application (13), and extraction treatment (14). Diagnosis of OIRR is done in most cases radiographically, since clinical symptoms are for the most part absent and increased tooth mobility is seen only in severe cases with additional alveolar bone loss through time (15). Usually, two-dimensional radiographic (2D) methods like periapical or panoramic radiographs are taken before, during, and after orthodontic treatment to monitor OIRR (16). Studies using 2D imaging techniques found OIRR to be less than 0.60 mm at the end of treatment with the maxillary incisors being more frequently and severely affected (5, 10, 17). However, OIRR affects every aspect of the root surface in all three-dimensions and therefore, 2D radiographs might mask the true amount of OIRR. Additionally, the true extent of OIRR might be misestimated due to magnification errors and problematic repeatability of 2D radiographs (18). Cone beam computed tomography (CBCT) was introduced as a three-dimensional diagnostic modality capable of imaging complex dental and maxillofacial structures and was quickly adopted in dentistry due to its reduced radiation dose compared to conventional computed tomography (19). On the other side, considerable variation in CBCT radiation doses exists according to the exposure protocol and there might be as much as 15-fold difference between a low-dose and a high-resolution protocol with the same field of view (19). The diagnostic value of CBCT in the diagnosis of OIRR lies on its ability to obtain distortion-free and reproducible images of the roots (20) with high sensitivity and specificity (21). Additionally, three-dimensional reconstruction of 2D CBCT slices enables accurate quantification of both linear and volumetric OIRR measurements (22) and compensates for changes in root position or angulation during orthodontic treatment (23). Objectives In light of the considerable differences between the two imaging modalities and potential advantages of three-dimensional imaging with CBCT in quantifying OIRR, this systematic review sought to summarize evidence from existing clinical studies assessing linear or volumetric OIRR with CBCT in order to primarily quantify the average OIRR that can be expected after orthodontic treatment. The secondary aim was to identify significant patient-, treatment-, or imaging-related factors that are significantly associated with OIRR. Materials and methods Protocol and registration The protocol for the present systematic review was developed a priori according to the Cochrane Handbook for Systematic Reviews of Interventions 5.1.0 (24) and registered in PROSPERO (CRD42016030131). The present systematic review is conducted according to the guidelines of the Cochrane Handbook for Systematic Reviews of Interventions version 5.1.0 (24) and is reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (25). Information sources and literature search A total of fifteen databases were systematically searched from their inception up to July 2015 for published, ongoing, or unpublished studies without any limitation regarding language, publication year, or publication status (Supplementary Table 1). Electronic searches were updated in January 2017, while the reference and citation lists of all included studies and identified relevant systematic reviews were manually searched for eligible studies. Eligibility criteria Eligibility criteria were determined a priori according to the Participant–Intervention–Comparison–Outcome–Study design (PICOS) schema (Supplementary Table 2). As a result, randomized clinical trials and prospective or retrospective non-randomized clinical studies assessing linear or volumetric OIRR with CBCT in patients in the permanent dentition with fully developed tooth roots examined during and/or after orthodontic treatment were included. Studies with untreated control groups were initially not expected to be found, since it is unethical to expose patients to radiation without proper indications and when no orthodontic intervention has been performed, but were nevertheless included, if available. All other clinical or non-clinical study designs not fulfilling the criteria were excluded. Study selection The titles/abstracts of identified studies were screened by one author (AS), while an independent checking of their full texts for eligibility was performed by another author (SNP), and conflicts were resolved by a third author (MAP). Data collection Data extraction was performed independently by two authors (AS, SNP) using pre-determined and piloted extraction forms covering study design, clinical setting, country, patient characteristics, number/age/sex of patients, applied orthodontic intervention, treatment duration, teeth assessed for OIRR, outcome assessed, timing of outcome measurement, CBCT technical characteristics, and potential conflict of interests. Discrepancies in data extraction between the two authors were likewise resolved by a third author (MAP), while authors of included trials were contacted for missing or unclear information. Risk of bias in individual trials The Cochrane Collaboration’s Risk of Bias Tool was used to assess the risk of bias in randomized clinical trials (24), while a modified version of the Downs and Black checklist (26) was used to assess the risk of bias of non-randomized studies. Risk of bias was assessed independently by two authors (AS, SNP) and potential conflicts between them were resolved by the third author (MAP). Data synthesis The primary outcome of this systematic review was the amount of OIRR in patients treated with fixed appliances. This was measured either as change in length or volume and either after an initial phase of treatment or after the end of treatment, but was assessed separately. The secondary outcome was the amount of OIRR after conventional rapid maxillary expansion (RME), surgically assisted RME, functional appliance treatment, or any other kind of appliance. For both the primary and all secondary outcomes the average amount of OIRR, as well as the influence of any patient-, tooth-, treatment-, or outcome-related factors was investigated, including among others patient age, sex, tooth type, treatment duration, inclusion of extractions, voxel size, and OIRR measurement unit. Data were summarized and considered suitable for pooling if the corresponding studies used similar interventions in the same way and reported similar outcomes. The amount of OIRR in terms of alterations in the root length or volume measured by CBCTs was calculated as initial length/volume minus final length/volume, so that positive values indicated root resorption and negative values indicated root elongation. From each included study data from all available study arms pertaining to different teeth/roots, appliances, or follow-up were extracted. Since OIRR as response to treatment was expected to vary according to population variations, genetic predisposition, differences in the applied treatment mechanics, and differences in CBCT protocol, a random-effects model was deemed appropriate to calculate the mean across distributions of true OIRR effects (27). For the main analysis all study arms of each study were pooled together (24) and the novel Paule–Mandel random-effects estimator was used over the DerSimonian–Laird one, since it outperforms the latter (28). For the main analysis, the mean difference (MD) in OIRR was used to quantify differences between groups. In explorative meta-analyses where multiple study arms from a study were included, a robust variance estimation of random-effects using the DerSimonian–Laird method was performed to account for clustering of study arms within studies (29). In the explorative analyses, the average OIRR within- and across-groups was quantified. The extent and impact of heterogeneity across studies was assessed by forest plot inspection and formally quantified using the tau2 and I2 metrics, respectively (24). The following arbitrary categories for I2 were adopted (24): I2 = 0–40 per cent: non-important inconsistency; I2 = 30–60 per cent: moderate inconsistency; I2 = 50–90 per cent: substantial inconsistency; and I2 > 75–100 per cent: considerable inconsistency. All analyses were performed on Stata SE 14.0 (StataCorp, College Station, Texas, USA) by one author (SNP) with an open dataset in Zenodo (30). The level of significance was set to a two-sided P < 0.05 except for tests of within- or between-subgroups heterogeneity, where it was set at P < 0.10 (24). Risk of bias across studies and additional analyses Potential sources of heterogeneity attributed to patient-, tooth-, treatment-, or outcome-related factors were assessed with the use of pre-specified mixed-effects subgroup and meta-regression analyses with at least five studies. Investigated factors included patient age, sex, tooth type, treatment phase (initial or complete treatment), treatment duration, and inclusion of extractions. Indications of reporting biases (including small-study effects and publication bias) were assessed with Egger’s linear regression test (31) and contour-enhanced funnel plots, should 10 or more trials be pooled in a meta-analysis. The overall quality of evidence (confidence in effect estimates) for each outcome was rated using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach (32). The cut-offs of minimal clinical important, large, and very large effects for linear OIRR were arbitrarily defined a priori as 0.75, 1.50, 3.00 mm (or 10.0, 20.0, and 40.0 mm3 for volumetric OIRR) and used to augment the produced forest plots with contours of effect magnitude (33). Robustness of the results was checked with sensitivity analyses according to the following study design and outcome measurement method factors: randomized or non-randomized studies, prospective or retrospective studies, OIRR measurement at the tooth or root level, and diagnostic accuracy (based on a study (34) finding differences in measured OIRR using a cut-off voxel size of 0.2 mm). Results Study selection The electronic and manual literature searches yielded 3442 records that were reduced to 2206 records after removal of duplicates. A total of 2162 records were excluded on the basis of title, abstract, and full text according to the specific inclusion and exclusion criteria (Supplementary Table 3a–c). The full text of six publications could not be retrieved, even after author communication attempts via email (Supplementary Table 4). Consequently, 33 publications were included in the study, which after collating multiple publications pertaining to the same study, reflected a total of 30 unique included data sets (Figure 1). Figure 1. View largeDownload slide Flow diagram for the identification and selection of studies in this systematic review. *Three studies could not be included in the analyses, due to missing data. Figure 1. View largeDownload slide Flow diagram for the identification and selection of studies in this systematic review. *Three studies could not be included in the analyses, due to missing data. Study characteristics The characteristics of the 30 included studies are presented in Table 1 and Supplementary Table 5. Six of the included studies were randomized clinical trials, while 6 were prospective non-randomized studies, and the remaining 18 studies were retrospective non-randomized studies. These included a total of 1219 patients, with 37.7 per cent of patients being male (from the 28 studies reporting sex) and with mean ages ranging between 11.4 and 26.6 years with an average of 14.7 years (from the 27 studies reporting age). CBCTs were taken in all the included studies at two or three time points before, during and/or after orthodontic treatment in order to evaluate OIRR. Different appliances were utilized in most of the included studies with fixed appliances used in 22 studies, RME in 6 studies, surgically-assisted RME in 1 study, and Herbst appliance in 1 study. The amount of OIRR was assessed at various teeth of all types (with the exception of third molars) and both jaws. Table 1. Characteristics of included studies regarding patients and interventions. Study  Design  Characteristics of patients  No patients (M/F); mAge (years)  Intervention  TxDur  COI  Abdel-Kader 2016  rNRS; Uni; EGY  PM1 Ex and C retraction  16 (6/10); 16.4  FFA; E-chain  6.3 months  NR  Ahn 2013  rNRS; Uni; KOR  Cl. I; d-a protrusion; 4PM Ex; en masse Retr;  37 (0/37); 26.6  22-slot; Roth; TPA/HG; E-chain (200 g)  21.7 months  NR  Ajmera 2014  pNRS; Uni; IND  Max-ant. proclination; PM1 Ex; en masse Retr/Intr.  48 (21/27); 13.6  22-slot; MBT; TADs (200 g)  NR  NR  Alexander 2014; Akyalcin 2015  rNRS; Uni; USA  ME need  24 (10/14); 12.8  Tooth-anch RME (0.25 mm/day; 5 months Retw)  4.8 months  NR  Aras and Tuncer 2016  RCT; Uni; TUR  Cl.I or II; OB ≥5 mm; max. ant. Intr;  Gp1: 16 (6/10); 19.3 Gp2: 16 (6/10); 19.0  Gp1: 2 MIs between I2-C; SS segmental wire; E-chain Gp2: 2 MIs between PM2-M1; Burstone TMA IntrArch (40 g/side)  4.0 months  NR  Baysal 2012  rNRS; Uni; TUR  ME need  25 (11/14); 12.7  Tooth-anch RME (0.5 mm/day)  NR  NR  Bolerjack 2005  rNRS; pract; USA  Non-Ex/Ex (4 PMs); sliding mechanics  95 (41/54); 12.7  22-slot; Roth  26.4 months  NR  Castro 2013; Castro 2015  rNRS; pract; BRA  Cl. I Non-Ex  30 (11/19); 13.0  22-slot; Roth  22.0 months  NR  Dindaroglu 2016  RCT; Uni; TUR  ME need  Gp1/2 33(17/16); 12.8 Gp1: 16 (NR); 12.9 Gp2: 17 (NR); 12.8  Gp1/2 (0.5 mm/day; 6 months Retw) Gp1: Tooth-anch Hyrax RME Gp2: Tooth-anch Haas RME  6.0 months  NR  Forst 2015  RCT**; Uni; CAN  ME need  Gp1: 20 (5/15); 14.1 Gp2: 21 (8/13); 14.2 Gp3: 21 (6/15); 12.9  Gp1: Tooth-anch RME (0.5 mm/day; 6 months Retw; 6 months Retw/o) Gp2: Bone-anch RME (0.25 mm/day; 6 months Retw; 6 months Retw/o) Gp3: No Tx  12.0 months ***  NR  Guo 2016  rNRS; Uni; CHN  —  174 (68/106); 14.1  FFA (SW); ‘light forces’  20.6 months  None  Harris 2015  rNRS; Uni; TUR  ME need  Gp1: 20 (8/12); 13.8 Gp2: 20 (7/13);13.8  Gp1: Tooth-anch RME (0.4 mm/day; 3 months Retw) Gp2: Bone-anch RME (0.4 mm/day; 3 months Retw)  3.0 months ***  NR  Johnson 2010  rNRS; pract; USA  FFA (1- or 2-phase); Ex/ Non-Ex  Gp1: 24 (11/13); 12.7 Gp2: 33 (10/23); 19.1  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 19.5 months Gp2: 13.9 months  NR  Kau 2011  pNRS; Uni; USA  Cl.I; PMs Ex  14 (NR); 20.3  FFA (SW) + vibration  6.0 months  NR  Kaylar 2016  RCT; Uni; TUR  ME need  Gp1: 10 (6/4); 19.3 Gp2: 10 (3/7); 19.2  Gp1: Tooth-anch SARME (0.5 mm/day; 6 months Retw) Gp2: Bone-anch SARME (0.5 mm/day; 6 months Retw)  6.0 months ***  None  Leite 2012  RCT; Uni; BRA  Cl. I; moderate crowding; non-Ex  Gp1: 11 (6/5); 20.6* Gp2: 8 (2/6); 20.6*  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  6.0 months  NR  Li 2013  rNRS; Uni; CHN  Cl. I–II; molar intrusion  12 (4/8); 24.3  TADs; E-chain (150 g)  6.0 months  NR  Liu 2016  RCT; Hosp; CHN  Ex/Non-Ex  Gp1: 25 (11/14); 15.3 Gp2: 25 (9/16); 15.2  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  NR  NR  Lombardo 2013  rNRS; pract; ITA  Cl. II/1; Ex (4 PMs)/non-Ex  22 (10/12); 11.4  Tweed-Merrifield technique  20.5 months  NR  Lynch 2011  rNRS; pract; USA  Cl. I-II; Ex/Non-Ex  122 (53/69); 13.2  22-slot; loops/E-chains  21.8 months  NR  Ma 2013  pNRS; Uni; CHN  ME need  32 (0/32); 14.3  Magnetic RME (increment/month)  9.3 months  NR  Nakada 2016  rNRS; Uni; JPN  4 PM Ex  30 (8/22); 21.9  FFA  NR  NR  Nimeri 2014  pNRS; Uni; USA  Cl. I; crowding; PM Ex  20 (5/15); NR  FFA + LLLT  NR  Comm  Patel 2012  rNRS; pract; USA  FFA (1- phase); Ex/ non-Ex  Gp1: 32 (16/16); 12.6 Gp2: 34 (11/23); 18.9  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 18.1 months Gp2: 14.2 months  NR  Schwarz 2015  rNRS; Uni; BRA  Cl. II; OJ > 5 mm  23 (11/12); 15.8  Herbst appliance  8.5 months  NR  Oliveira 2011; Oliveira 2016  pNRS; Uni; BRA  Cl. I–II/1; protrusion; Mx PM1 Ex  11 (5/6); adult  FFA; omega loops  NR  NR  Wang 2013  pNRS; Uni; CHN  Skeletal Cl. III; pre-surgical decompensation need; Non-Ex  Gp1: 30 (14/16); 24.8 Gp2: 26 (17/9); 23.5  Gp1: FFA Gp2: FFA + corticotomy  Gp1: 13.3 months Gp2: 7.8 months  None  Wang 2015  rNRS; Uni; CHN  Skeletal Cl. III; pre- surgical decompensation need; non-Ex  30 (13/17); adult  FFA; MBT  8.7 months  NR  Wen 2016  rNRS; Uni; CHN  PM Ex and space closure; TAD  12 (3/9); adult  FFA  30.0 months  NR  Yang 2016  rNRS; Uni; CHN  4 PM Ex; ant. Retr.  7 (NR); 22.3 years  FFA  NR  NR  Study  Design  Characteristics of patients  No patients (M/F); mAge (years)  Intervention  TxDur  COI  Abdel-Kader 2016  rNRS; Uni; EGY  PM1 Ex and C retraction  16 (6/10); 16.4  FFA; E-chain  6.3 months  NR  Ahn 2013  rNRS; Uni; KOR  Cl. I; d-a protrusion; 4PM Ex; en masse Retr;  37 (0/37); 26.6  22-slot; Roth; TPA/HG; E-chain (200 g)  21.7 months  NR  Ajmera 2014  pNRS; Uni; IND  Max-ant. proclination; PM1 Ex; en masse Retr/Intr.  48 (21/27); 13.6  22-slot; MBT; TADs (200 g)  NR  NR  Alexander 2014; Akyalcin 2015  rNRS; Uni; USA  ME need  24 (10/14); 12.8  Tooth-anch RME (0.25 mm/day; 5 months Retw)  4.8 months  NR  Aras and Tuncer 2016  RCT; Uni; TUR  Cl.I or II; OB ≥5 mm; max. ant. Intr;  Gp1: 16 (6/10); 19.3 Gp2: 16 (6/10); 19.0  Gp1: 2 MIs between I2-C; SS segmental wire; E-chain Gp2: 2 MIs between PM2-M1; Burstone TMA IntrArch (40 g/side)  4.0 months  NR  Baysal 2012  rNRS; Uni; TUR  ME need  25 (11/14); 12.7  Tooth-anch RME (0.5 mm/day)  NR  NR  Bolerjack 2005  rNRS; pract; USA  Non-Ex/Ex (4 PMs); sliding mechanics  95 (41/54); 12.7  22-slot; Roth  26.4 months  NR  Castro 2013; Castro 2015  rNRS; pract; BRA  Cl. I Non-Ex  30 (11/19); 13.0  22-slot; Roth  22.0 months  NR  Dindaroglu 2016  RCT; Uni; TUR  ME need  Gp1/2 33(17/16); 12.8 Gp1: 16 (NR); 12.9 Gp2: 17 (NR); 12.8  Gp1/2 (0.5 mm/day; 6 months Retw) Gp1: Tooth-anch Hyrax RME Gp2: Tooth-anch Haas RME  6.0 months  NR  Forst 2015  RCT**; Uni; CAN  ME need  Gp1: 20 (5/15); 14.1 Gp2: 21 (8/13); 14.2 Gp3: 21 (6/15); 12.9  Gp1: Tooth-anch RME (0.5 mm/day; 6 months Retw; 6 months Retw/o) Gp2: Bone-anch RME (0.25 mm/day; 6 months Retw; 6 months Retw/o) Gp3: No Tx  12.0 months ***  NR  Guo 2016  rNRS; Uni; CHN  —  174 (68/106); 14.1  FFA (SW); ‘light forces’  20.6 months  None  Harris 2015  rNRS; Uni; TUR  ME need  Gp1: 20 (8/12); 13.8 Gp2: 20 (7/13);13.8  Gp1: Tooth-anch RME (0.4 mm/day; 3 months Retw) Gp2: Bone-anch RME (0.4 mm/day; 3 months Retw)  3.0 months ***  NR  Johnson 2010  rNRS; pract; USA  FFA (1- or 2-phase); Ex/ Non-Ex  Gp1: 24 (11/13); 12.7 Gp2: 33 (10/23); 19.1  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 19.5 months Gp2: 13.9 months  NR  Kau 2011  pNRS; Uni; USA  Cl.I; PMs Ex  14 (NR); 20.3  FFA (SW) + vibration  6.0 months  NR  Kaylar 2016  RCT; Uni; TUR  ME need  Gp1: 10 (6/4); 19.3 Gp2: 10 (3/7); 19.2  Gp1: Tooth-anch SARME (0.5 mm/day; 6 months Retw) Gp2: Bone-anch SARME (0.5 mm/day; 6 months Retw)  6.0 months ***  None  Leite 2012  RCT; Uni; BRA  Cl. I; moderate crowding; non-Ex  Gp1: 11 (6/5); 20.6* Gp2: 8 (2/6); 20.6*  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  6.0 months  NR  Li 2013  rNRS; Uni; CHN  Cl. I–II; molar intrusion  12 (4/8); 24.3  TADs; E-chain (150 g)  6.0 months  NR  Liu 2016  RCT; Hosp; CHN  Ex/Non-Ex  Gp1: 25 (11/14); 15.3 Gp2: 25 (9/16); 15.2  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  NR  NR  Lombardo 2013  rNRS; pract; ITA  Cl. II/1; Ex (4 PMs)/non-Ex  22 (10/12); 11.4  Tweed-Merrifield technique  20.5 months  NR  Lynch 2011  rNRS; pract; USA  Cl. I-II; Ex/Non-Ex  122 (53/69); 13.2  22-slot; loops/E-chains  21.8 months  NR  Ma 2013  pNRS; Uni; CHN  ME need  32 (0/32); 14.3  Magnetic RME (increment/month)  9.3 months  NR  Nakada 2016  rNRS; Uni; JPN  4 PM Ex  30 (8/22); 21.9  FFA  NR  NR  Nimeri 2014  pNRS; Uni; USA  Cl. I; crowding; PM Ex  20 (5/15); NR  FFA + LLLT  NR  Comm  Patel 2012  rNRS; pract; USA  FFA (1- phase); Ex/ non-Ex  Gp1: 32 (16/16); 12.6 Gp2: 34 (11/23); 18.9  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 18.1 months Gp2: 14.2 months  NR  Schwarz 2015  rNRS; Uni; BRA  Cl. II; OJ > 5 mm  23 (11/12); 15.8  Herbst appliance  8.5 months  NR  Oliveira 2011; Oliveira 2016  pNRS; Uni; BRA  Cl. I–II/1; protrusion; Mx PM1 Ex  11 (5/6); adult  FFA; omega loops  NR  NR  Wang 2013  pNRS; Uni; CHN  Skeletal Cl. III; pre-surgical decompensation need; Non-Ex  Gp1: 30 (14/16); 24.8 Gp2: 26 (17/9); 23.5  Gp1: FFA Gp2: FFA + corticotomy  Gp1: 13.3 months Gp2: 7.8 months  None  Wang 2015  rNRS; Uni; CHN  Skeletal Cl. III; pre- surgical decompensation need; non-Ex  30 (13/17); adult  FFA; MBT  8.7 months  NR  Wen 2016  rNRS; Uni; CHN  PM Ex and space closure; TAD  12 (3/9); adult  FFA  30.0 months  NR  Yang 2016  rNRS; Uni; CHN  4 PM Ex; ant. Retr.  7 (NR); 22.3 years  FFA  NR  NR  Ant. anterior; anch, anchorage; C, canine; CLB, conventional brackets; Comm, commercial; COI, conflict of interest; Cl I, Class I; Cl II, Class II; Cl II/1, Class II division 1; Cl III, Class III; d-a protrusion, dentoalveolar protrusion; E-chain, elastic chain; Ex, extraction; FFA, fixed appliances; Gp, group; HG, headgear; Intr, intrusion; non-ex, non extraction; I2, lateral incisor; LLLT, low level light therapy; MIs, mini-screw implants; M1, first molar; Max, maxilla; ME, maxillary expansion; M/F, male/female; NR, not reported; OB, overbite; OJ, overjet; pract, private practice; PM, premolar; PM1, first premolar; PM2, second premolar; pNRS, prospective non-randomized study; RCT, randomized clinical trial; rNRS, retrospective non-randomized study; RME, rapid maxillary expansion; Retw, retention with expansion appliance; Retw/o, retention without expansion appliance; Retr, retraction; SARME, surgically assisted rapid maxillary expansion; SLB, self-ligating brackets; SW, straight wire; SS, stainless steel; Tx, treatment; TAD, temporary anchorage device; TPA, transpalatal arch; TxDur, treatment duration; Uni, university. *For both groups. **Randomization was performed for another study, not on the basis of root resorption. ***Period including a retention period about 12 months. ****Right and left teeth of the same category were averaged for analyses. View Large Table 1. Characteristics of included studies regarding patients and interventions. Study  Design  Characteristics of patients  No patients (M/F); mAge (years)  Intervention  TxDur  COI  Abdel-Kader 2016  rNRS; Uni; EGY  PM1 Ex and C retraction  16 (6/10); 16.4  FFA; E-chain  6.3 months  NR  Ahn 2013  rNRS; Uni; KOR  Cl. I; d-a protrusion; 4PM Ex; en masse Retr;  37 (0/37); 26.6  22-slot; Roth; TPA/HG; E-chain (200 g)  21.7 months  NR  Ajmera 2014  pNRS; Uni; IND  Max-ant. proclination; PM1 Ex; en masse Retr/Intr.  48 (21/27); 13.6  22-slot; MBT; TADs (200 g)  NR  NR  Alexander 2014; Akyalcin 2015  rNRS; Uni; USA  ME need  24 (10/14); 12.8  Tooth-anch RME (0.25 mm/day; 5 months Retw)  4.8 months  NR  Aras and Tuncer 2016  RCT; Uni; TUR  Cl.I or II; OB ≥5 mm; max. ant. Intr;  Gp1: 16 (6/10); 19.3 Gp2: 16 (6/10); 19.0  Gp1: 2 MIs between I2-C; SS segmental wire; E-chain Gp2: 2 MIs between PM2-M1; Burstone TMA IntrArch (40 g/side)  4.0 months  NR  Baysal 2012  rNRS; Uni; TUR  ME need  25 (11/14); 12.7  Tooth-anch RME (0.5 mm/day)  NR  NR  Bolerjack 2005  rNRS; pract; USA  Non-Ex/Ex (4 PMs); sliding mechanics  95 (41/54); 12.7  22-slot; Roth  26.4 months  NR  Castro 2013; Castro 2015  rNRS; pract; BRA  Cl. I Non-Ex  30 (11/19); 13.0  22-slot; Roth  22.0 months  NR  Dindaroglu 2016  RCT; Uni; TUR  ME need  Gp1/2 33(17/16); 12.8 Gp1: 16 (NR); 12.9 Gp2: 17 (NR); 12.8  Gp1/2 (0.5 mm/day; 6 months Retw) Gp1: Tooth-anch Hyrax RME Gp2: Tooth-anch Haas RME  6.0 months  NR  Forst 2015  RCT**; Uni; CAN  ME need  Gp1: 20 (5/15); 14.1 Gp2: 21 (8/13); 14.2 Gp3: 21 (6/15); 12.9  Gp1: Tooth-anch RME (0.5 mm/day; 6 months Retw; 6 months Retw/o) Gp2: Bone-anch RME (0.25 mm/day; 6 months Retw; 6 months Retw/o) Gp3: No Tx  12.0 months ***  NR  Guo 2016  rNRS; Uni; CHN  —  174 (68/106); 14.1  FFA (SW); ‘light forces’  20.6 months  None  Harris 2015  rNRS; Uni; TUR  ME need  Gp1: 20 (8/12); 13.8 Gp2: 20 (7/13);13.8  Gp1: Tooth-anch RME (0.4 mm/day; 3 months Retw) Gp2: Bone-anch RME (0.4 mm/day; 3 months Retw)  3.0 months ***  NR  Johnson 2010  rNRS; pract; USA  FFA (1- or 2-phase); Ex/ Non-Ex  Gp1: 24 (11/13); 12.7 Gp2: 33 (10/23); 19.1  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 19.5 months Gp2: 13.9 months  NR  Kau 2011  pNRS; Uni; USA  Cl.I; PMs Ex  14 (NR); 20.3  FFA (SW) + vibration  6.0 months  NR  Kaylar 2016  RCT; Uni; TUR  ME need  Gp1: 10 (6/4); 19.3 Gp2: 10 (3/7); 19.2  Gp1: Tooth-anch SARME (0.5 mm/day; 6 months Retw) Gp2: Bone-anch SARME (0.5 mm/day; 6 months Retw)  6.0 months ***  None  Leite 2012  RCT; Uni; BRA  Cl. I; moderate crowding; non-Ex  Gp1: 11 (6/5); 20.6* Gp2: 8 (2/6); 20.6*  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  6.0 months  NR  Li 2013  rNRS; Uni; CHN  Cl. I–II; molar intrusion  12 (4/8); 24.3  TADs; E-chain (150 g)  6.0 months  NR  Liu 2016  RCT; Hosp; CHN  Ex/Non-Ex  Gp1: 25 (11/14); 15.3 Gp2: 25 (9/16); 15.2  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  NR  NR  Lombardo 2013  rNRS; pract; ITA  Cl. II/1; Ex (4 PMs)/non-Ex  22 (10/12); 11.4  Tweed-Merrifield technique  20.5 months  NR  Lynch 2011  rNRS; pract; USA  Cl. I-II; Ex/Non-Ex  122 (53/69); 13.2  22-slot; loops/E-chains  21.8 months  NR  Ma 2013  pNRS; Uni; CHN  ME need  32 (0/32); 14.3  Magnetic RME (increment/month)  9.3 months  NR  Nakada 2016  rNRS; Uni; JPN  4 PM Ex  30 (8/22); 21.9  FFA  NR  NR  Nimeri 2014  pNRS; Uni; USA  Cl. I; crowding; PM Ex  20 (5/15); NR  FFA + LLLT  NR  Comm  Patel 2012  rNRS; pract; USA  FFA (1- phase); Ex/ non-Ex  Gp1: 32 (16/16); 12.6 Gp2: 34 (11/23); 18.9  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 18.1 months Gp2: 14.2 months  NR  Schwarz 2015  rNRS; Uni; BRA  Cl. II; OJ > 5 mm  23 (11/12); 15.8  Herbst appliance  8.5 months  NR  Oliveira 2011; Oliveira 2016  pNRS; Uni; BRA  Cl. I–II/1; protrusion; Mx PM1 Ex  11 (5/6); adult  FFA; omega loops  NR  NR  Wang 2013  pNRS; Uni; CHN  Skeletal Cl. III; pre-surgical decompensation need; Non-Ex  Gp1: 30 (14/16); 24.8 Gp2: 26 (17/9); 23.5  Gp1: FFA Gp2: FFA + corticotomy  Gp1: 13.3 months Gp2: 7.8 months  None  Wang 2015  rNRS; Uni; CHN  Skeletal Cl. III; pre- surgical decompensation need; non-Ex  30 (13/17); adult  FFA; MBT  8.7 months  NR  Wen 2016  rNRS; Uni; CHN  PM Ex and space closure; TAD  12 (3/9); adult  FFA  30.0 months  NR  Yang 2016  rNRS; Uni; CHN  4 PM Ex; ant. Retr.  7 (NR); 22.3 years  FFA  NR  NR  Study  Design  Characteristics of patients  No patients (M/F); mAge (years)  Intervention  TxDur  COI  Abdel-Kader 2016  rNRS; Uni; EGY  PM1 Ex and C retraction  16 (6/10); 16.4  FFA; E-chain  6.3 months  NR  Ahn 2013  rNRS; Uni; KOR  Cl. I; d-a protrusion; 4PM Ex; en masse Retr;  37 (0/37); 26.6  22-slot; Roth; TPA/HG; E-chain (200 g)  21.7 months  NR  Ajmera 2014  pNRS; Uni; IND  Max-ant. proclination; PM1 Ex; en masse Retr/Intr.  48 (21/27); 13.6  22-slot; MBT; TADs (200 g)  NR  NR  Alexander 2014; Akyalcin 2015  rNRS; Uni; USA  ME need  24 (10/14); 12.8  Tooth-anch RME (0.25 mm/day; 5 months Retw)  4.8 months  NR  Aras and Tuncer 2016  RCT; Uni; TUR  Cl.I or II; OB ≥5 mm; max. ant. Intr;  Gp1: 16 (6/10); 19.3 Gp2: 16 (6/10); 19.0  Gp1: 2 MIs between I2-C; SS segmental wire; E-chain Gp2: 2 MIs between PM2-M1; Burstone TMA IntrArch (40 g/side)  4.0 months  NR  Baysal 2012  rNRS; Uni; TUR  ME need  25 (11/14); 12.7  Tooth-anch RME (0.5 mm/day)  NR  NR  Bolerjack 2005  rNRS; pract; USA  Non-Ex/Ex (4 PMs); sliding mechanics  95 (41/54); 12.7  22-slot; Roth  26.4 months  NR  Castro 2013; Castro 2015  rNRS; pract; BRA  Cl. I Non-Ex  30 (11/19); 13.0  22-slot; Roth  22.0 months  NR  Dindaroglu 2016  RCT; Uni; TUR  ME need  Gp1/2 33(17/16); 12.8 Gp1: 16 (NR); 12.9 Gp2: 17 (NR); 12.8  Gp1/2 (0.5 mm/day; 6 months Retw) Gp1: Tooth-anch Hyrax RME Gp2: Tooth-anch Haas RME  6.0 months  NR  Forst 2015  RCT**; Uni; CAN  ME need  Gp1: 20 (5/15); 14.1 Gp2: 21 (8/13); 14.2 Gp3: 21 (6/15); 12.9  Gp1: Tooth-anch RME (0.5 mm/day; 6 months Retw; 6 months Retw/o) Gp2: Bone-anch RME (0.25 mm/day; 6 months Retw; 6 months Retw/o) Gp3: No Tx  12.0 months ***  NR  Guo 2016  rNRS; Uni; CHN  —  174 (68/106); 14.1  FFA (SW); ‘light forces’  20.6 months  None  Harris 2015  rNRS; Uni; TUR  ME need  Gp1: 20 (8/12); 13.8 Gp2: 20 (7/13);13.8  Gp1: Tooth-anch RME (0.4 mm/day; 3 months Retw) Gp2: Bone-anch RME (0.4 mm/day; 3 months Retw)  3.0 months ***  NR  Johnson 2010  rNRS; pract; USA  FFA (1- or 2-phase); Ex/ Non-Ex  Gp1: 24 (11/13); 12.7 Gp2: 33 (10/23); 19.1  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 19.5 months Gp2: 13.9 months  NR  Kau 2011  pNRS; Uni; USA  Cl.I; PMs Ex  14 (NR); 20.3  FFA (SW) + vibration  6.0 months  NR  Kaylar 2016  RCT; Uni; TUR  ME need  Gp1: 10 (6/4); 19.3 Gp2: 10 (3/7); 19.2  Gp1: Tooth-anch SARME (0.5 mm/day; 6 months Retw) Gp2: Bone-anch SARME (0.5 mm/day; 6 months Retw)  6.0 months ***  None  Leite 2012  RCT; Uni; BRA  Cl. I; moderate crowding; non-Ex  Gp1: 11 (6/5); 20.6* Gp2: 8 (2/6); 20.6*  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  6.0 months  NR  Li 2013  rNRS; Uni; CHN  Cl. I–II; molar intrusion  12 (4/8); 24.3  TADs; E-chain (150 g)  6.0 months  NR  Liu 2016  RCT; Hosp; CHN  Ex/Non-Ex  Gp1: 25 (11/14); 15.3 Gp2: 25 (9/16); 15.2  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  NR  NR  Lombardo 2013  rNRS; pract; ITA  Cl. II/1; Ex (4 PMs)/non-Ex  22 (10/12); 11.4  Tweed-Merrifield technique  20.5 months  NR  Lynch 2011  rNRS; pract; USA  Cl. I-II; Ex/Non-Ex  122 (53/69); 13.2  22-slot; loops/E-chains  21.8 months  NR  Ma 2013  pNRS; Uni; CHN  ME need  32 (0/32); 14.3  Magnetic RME (increment/month)  9.3 months  NR  Nakada 2016  rNRS; Uni; JPN  4 PM Ex  30 (8/22); 21.9  FFA  NR  NR  Nimeri 2014  pNRS; Uni; USA  Cl. I; crowding; PM Ex  20 (5/15); NR  FFA + LLLT  NR  Comm  Patel 2012  rNRS; pract; USA  FFA (1- phase); Ex/ non-Ex  Gp1: 32 (16/16); 12.6 Gp2: 34 (11/23); 18.9  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 18.1 months Gp2: 14.2 months  NR  Schwarz 2015  rNRS; Uni; BRA  Cl. II; OJ > 5 mm  23 (11/12); 15.8  Herbst appliance  8.5 months  NR  Oliveira 2011; Oliveira 2016  pNRS; Uni; BRA  Cl. I–II/1; protrusion; Mx PM1 Ex  11 (5/6); adult  FFA; omega loops  NR  NR  Wang 2013  pNRS; Uni; CHN  Skeletal Cl. III; pre-surgical decompensation need; Non-Ex  Gp1: 30 (14/16); 24.8 Gp2: 26 (17/9); 23.5  Gp1: FFA Gp2: FFA + corticotomy  Gp1: 13.3 months Gp2: 7.8 months  None  Wang 2015  rNRS; Uni; CHN  Skeletal Cl. III; pre- surgical decompensation need; non-Ex  30 (13/17); adult  FFA; MBT  8.7 months  NR  Wen 2016  rNRS; Uni; CHN  PM Ex and space closure; TAD  12 (3/9); adult  FFA  30.0 months  NR  Yang 2016  rNRS; Uni; CHN  4 PM Ex; ant. Retr.  7 (NR); 22.3 years  FFA  NR  NR  Ant. anterior; anch, anchorage; C, canine; CLB, conventional brackets; Comm, commercial; COI, conflict of interest; Cl I, Class I; Cl II, Class II; Cl II/1, Class II division 1; Cl III, Class III; d-a protrusion, dentoalveolar protrusion; E-chain, elastic chain; Ex, extraction; FFA, fixed appliances; Gp, group; HG, headgear; Intr, intrusion; non-ex, non extraction; I2, lateral incisor; LLLT, low level light therapy; MIs, mini-screw implants; M1, first molar; Max, maxilla; ME, maxillary expansion; M/F, male/female; NR, not reported; OB, overbite; OJ, overjet; pract, private practice; PM, premolar; PM1, first premolar; PM2, second premolar; pNRS, prospective non-randomized study; RCT, randomized clinical trial; rNRS, retrospective non-randomized study; RME, rapid maxillary expansion; Retw, retention with expansion appliance; Retw/o, retention without expansion appliance; Retr, retraction; SARME, surgically assisted rapid maxillary expansion; SLB, self-ligating brackets; SW, straight wire; SS, stainless steel; Tx, treatment; TAD, temporary anchorage device; TPA, transpalatal arch; TxDur, treatment duration; Uni, university. *For both groups. **Randomization was performed for another study, not on the basis of root resorption. ***Period including a retention period about 12 months. ****Right and left teeth of the same category were averaged for analyses. View Large Risk of bias within studies The risk of bias of included non-randomized and randomized studies can be seen in Supplementary Table 6a and b, respectively. According to the Cochrane Risk of Bias Tool, serious risk of bias was found in four out of six randomized trials for at least one domain with the remaining two having unclear risk of bias (Supplementary Table 6b). The remaining non-randomized studies were evaluated with the Downs and Black checklist and all of them were in serious risk of bias for at least one of the 27 domains of the tool (Supplementary Table 6a). Included studies gathered between 11 and 23 points from the maximum of 31 points to an average of 17 points. The most problematic domain categories for non-randomized studies were confounding, followed by external validity, and bias. Data synthesis—main analysis The main analysis of the present systematic review was conducted after pooling multiple study arms from each study and performing within-study comparisons solely from identified randomized trials that possess higher internal validity. A total of five different meta-analyses could be performed using the improved Paule–Mandel random-effects estimator as seen in Supplementary Table 7. No significant differences in OIRR could be found between self-ligating and conventional brackets in a meta-analysis of two studies (P > 0.5). A single randomized trial indicated that intrusion of upper anteriors with temporary anterior devices (TAD) placed between lateral incisor and canine resulted in greater OIRR than when the TADs were placed between second premolar and first molar, both in terms of root length (MD = 0.3 mm; 95% CI = 0.2–0.5 mm) or root volume (MD = 6.3 mm3; 95% CI = 2.4–10.3 mm3). Finally, a randomized trial found no significant difference in OIRR between a conventional or a TAD-anchored Hybrid-Hyrax, while another randomized trial reported that Haas type RME resulted in less volumetric OIRR than Hyrax type RME (MD = −10.9 mm3; 95% CI = −18.6 to −3.2 mm3). No additional analyses could be conducted due to the limited number of included studies. However, analysis of the meta-evidence with the GRADE approach (Table 2) indicated that the quality of evidence was low for the comparisons of self-ligating versus conventional appliances and for the comparison of Hyrax versus Hybrid-Hyrax RME, due to bias and imprecision. The quality of evidence supporting that anterior placed TADs lead to greater OIRR than posterior placed TADs during anterior intrusion was judged as moderate, due to bias. Finally, high quality evidence supported only the use of Haas-type RME over Hyrax-type RME in order to minimize OIRR. It should be however noted that as the majority of conclusions was drawn from single trials, further trials might change the current recommendations. Table 2. Summary of findings table according to the GRADE approach. Outcome trials (patients)  Anticipated absolute effects  Quality of the evidence (GRADE)**  What happens  Reference*  Experimental  Difference for experimental (95% CI)    Conventional brackets  Self-ligating brackets        OIRR (linear) of upper anteriors after comprehensive Tx two trials (276 teeth)  0.8 mm OIRR on average  —  <0.1 mm more OIRR (0.2 mm less to 0.3 mm more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Posterior TADS  Anterior TADs        OIRR (linear) of upper anteriors after 4 months intrusion one trial (128 teeth)  0.8 mm OIRR on average  —  0.3 mm more OIRR (0.2–0.5 mm more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs  OIRR (volumetric) of upper anteriors after 4 months intrusion one trial (128 teeth)  12.9 mm3 OIRR on average  —  6.3 mm3 more OIRR (2.4–10.3 mm3 more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs    Tooth-borne RME (Hyrax)  Hybrid-borne RME (Hyrax)        OIRR (linear) of upper posteriors after RME and retention one trial (100 teeth)  0.5 mm OIRR on average  —  0.1 mm more OIRR (0.1 mm less to 0.3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR  OIRR (volumetric) of upper posteriors after RME and retention one trial (37 teeth)  49.3 mm3 OIRR on average    17.8 mm3 less OIRR (47.5 mm3 less to 11.9 mm3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Tooth-borne RME (Hyrax)  Tooth-borne RME (Haas)        OIRR (volumetric) of upper posteriors after RME and retention one trial (396 teeth)  47.8 mm3 OIRR on average  —  10.9 mm3 less OIRR (3.2–18.6 mm3 less)  ⊕⊕⊕⊕ high  Probably less OIRR with Haas  Outcome trials (patients)  Anticipated absolute effects  Quality of the evidence (GRADE)**  What happens  Reference*  Experimental  Difference for experimental (95% CI)    Conventional brackets  Self-ligating brackets        OIRR (linear) of upper anteriors after comprehensive Tx two trials (276 teeth)  0.8 mm OIRR on average  —  <0.1 mm more OIRR (0.2 mm less to 0.3 mm more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Posterior TADS  Anterior TADs        OIRR (linear) of upper anteriors after 4 months intrusion one trial (128 teeth)  0.8 mm OIRR on average  —  0.3 mm more OIRR (0.2–0.5 mm more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs  OIRR (volumetric) of upper anteriors after 4 months intrusion one trial (128 teeth)  12.9 mm3 OIRR on average  —  6.3 mm3 more OIRR (2.4–10.3 mm3 more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs    Tooth-borne RME (Hyrax)  Hybrid-borne RME (Hyrax)        OIRR (linear) of upper posteriors after RME and retention one trial (100 teeth)  0.5 mm OIRR on average  —  0.1 mm more OIRR (0.1 mm less to 0.3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR  OIRR (volumetric) of upper posteriors after RME and retention one trial (37 teeth)  49.3 mm3 OIRR on average    17.8 mm3 less OIRR (47.5 mm3 less to 11.9 mm3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Tooth-borne RME (Hyrax)  Tooth-borne RME (Haas)        OIRR (volumetric) of upper posteriors after RME and retention one trial (396 teeth)  47.8 mm3 OIRR on average  —  10.9 mm3 less OIRR (3.2–18.6 mm3 less)  ⊕⊕⊕⊕ high  Probably less OIRR with Haas  Orthodontically induced root resorption assessed with cone-beam computed tomography. Patient or population: adolescent/adult patients receiving comprehensive fixed appliance orthodontic treatment, maxillary anterior intrusion, or rapid maxillary expansion. Settings: university clinics and hospital (Brazil, Canada, China, Turkey). CI, confidence interval; GRADE, Grading of Recommendations Assessment, Development and Evaluation; OIRR, orthodontically induced root resorption; RME, rapid maxillary expansion; TAD, temporary anchorage device. *Reponse is based on random-effects meta-analytical pooling of the corresponding reference groups among included studies. **Starts from ‘high’, due to the inclusion of randomized studies. ***Downgraded further by one point due to serious limitations (high risk of bias). ****Downgraded by one due to imprecision originating from the inclusion of a small sample size. View Large Table 2. Summary of findings table according to the GRADE approach. Outcome trials (patients)  Anticipated absolute effects  Quality of the evidence (GRADE)**  What happens  Reference*  Experimental  Difference for experimental (95% CI)    Conventional brackets  Self-ligating brackets        OIRR (linear) of upper anteriors after comprehensive Tx two trials (276 teeth)  0.8 mm OIRR on average  —  <0.1 mm more OIRR (0.2 mm less to 0.3 mm more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Posterior TADS  Anterior TADs        OIRR (linear) of upper anteriors after 4 months intrusion one trial (128 teeth)  0.8 mm OIRR on average  —  0.3 mm more OIRR (0.2–0.5 mm more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs  OIRR (volumetric) of upper anteriors after 4 months intrusion one trial (128 teeth)  12.9 mm3 OIRR on average  —  6.3 mm3 more OIRR (2.4–10.3 mm3 more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs    Tooth-borne RME (Hyrax)  Hybrid-borne RME (Hyrax)        OIRR (linear) of upper posteriors after RME and retention one trial (100 teeth)  0.5 mm OIRR on average  —  0.1 mm more OIRR (0.1 mm less to 0.3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR  OIRR (volumetric) of upper posteriors after RME and retention one trial (37 teeth)  49.3 mm3 OIRR on average    17.8 mm3 less OIRR (47.5 mm3 less to 11.9 mm3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Tooth-borne RME (Hyrax)  Tooth-borne RME (Haas)        OIRR (volumetric) of upper posteriors after RME and retention one trial (396 teeth)  47.8 mm3 OIRR on average  —  10.9 mm3 less OIRR (3.2–18.6 mm3 less)  ⊕⊕⊕⊕ high  Probably less OIRR with Haas  Outcome trials (patients)  Anticipated absolute effects  Quality of the evidence (GRADE)**  What happens  Reference*  Experimental  Difference for experimental (95% CI)    Conventional brackets  Self-ligating brackets        OIRR (linear) of upper anteriors after comprehensive Tx two trials (276 teeth)  0.8 mm OIRR on average  —  <0.1 mm more OIRR (0.2 mm less to 0.3 mm more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Posterior TADS  Anterior TADs        OIRR (linear) of upper anteriors after 4 months intrusion one trial (128 teeth)  0.8 mm OIRR on average  —  0.3 mm more OIRR (0.2–0.5 mm more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs  OIRR (volumetric) of upper anteriors after 4 months intrusion one trial (128 teeth)  12.9 mm3 OIRR on average  —  6.3 mm3 more OIRR (2.4–10.3 mm3 more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs    Tooth-borne RME (Hyrax)  Hybrid-borne RME (Hyrax)        OIRR (linear) of upper posteriors after RME and retention one trial (100 teeth)  0.5 mm OIRR on average  —  0.1 mm more OIRR (0.1 mm less to 0.3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR  OIRR (volumetric) of upper posteriors after RME and retention one trial (37 teeth)  49.3 mm3 OIRR on average    17.8 mm3 less OIRR (47.5 mm3 less to 11.9 mm3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Tooth-borne RME (Hyrax)  Tooth-borne RME (Haas)        OIRR (volumetric) of upper posteriors after RME and retention one trial (396 teeth)  47.8 mm3 OIRR on average  —  10.9 mm3 less OIRR (3.2–18.6 mm3 less)  ⊕⊕⊕⊕ high  Probably less OIRR with Haas  Orthodontically induced root resorption assessed with cone-beam computed tomography. Patient or population: adolescent/adult patients receiving comprehensive fixed appliance orthodontic treatment, maxillary anterior intrusion, or rapid maxillary expansion. Settings: university clinics and hospital (Brazil, Canada, China, Turkey). CI, confidence interval; GRADE, Grading of Recommendations Assessment, Development and Evaluation; OIRR, orthodontically induced root resorption; RME, rapid maxillary expansion; TAD, temporary anchorage device. *Reponse is based on random-effects meta-analytical pooling of the corresponding reference groups among included studies. **Starts from ‘high’, due to the inclusion of randomized studies. ***Downgraded further by one point due to serious limitations (high risk of bias). ****Downgraded by one due to imprecision originating from the inclusion of a small sample size. View Large Data synthesis—explorative analysis Furthermore, an exploratory analysis was conducted by using separate data from all available study arms of identified studies and employing a robust-variance DerSimonian–Laird random-effects model, analyzing studies on fixed appliance treatment and studies on RME separately (Table 3). As far as treatment with fixed appliances is concerned, the average linear OIRR among all 17 available studies (including 161 study arms) was found to be 0.8 mm (95% CI = 0.5–1.0 mm; Table 3; Figure 2). Table 3. Explorative analysis of linear OIRR including multiple measurements from all included studies.† Factor  Group  Fixed-appliance Tx  Rapid maxillary expansion  Arms  Studies  Mean  95% CI  PSG  Arms  Studies  Mean  95% CI  PSG  Any studies    Overall  161  17  0.79  0.54,1.03    20  3  0.36  −0.97,1.69    Tooth type  1s  32  12  0.82  0.38,1.26  0.04  —  —      0.51    2s  30  9  0.72  0.36,1.07    —  —          3s  13  6  0.37  −0.28,1.02    —  —          4s  22  4  0.29  −0.59,1.17    7  3  0.35  −1.26,1.97      5s  14  3  −0.08  −0.46,0.30    1  1  NE        6s  41  5  0.26  −0.30,0.81    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  79  16  0.85  0.62,1.08  0.08  —  —      ne    Posterior  82  5  0.24  −0.29,0.77    Same          Jaw  Maxilla  100  16  0.75  0.49,1.00  0.77  Same        ne    Mandible  61  7  0.57  0.01,1.13    —  —        Jaw*group  Posterior mandible  32  4  0.28  −0.42,0.98  0.06  —  —          Posterior maxilla  50  5  0.22  −0.35,0.78    Same            Anterior mandible  29  6  0.60  −0.14,1.34    —  —          Anterior maxilla  50  15  0.82  0.58,1.07    —  —        Only studies with complete Tx    Overall  81  9  0.86  0.42,1.30    20  3  0.36  −0.97,1.69    Tooth type  1s  14  6  1.00  0.44,1.57  0.10  —  —      0.51    2s  12  4  0.83  0.18,1.49    —  —          3s  4  3  0.67  −0.74,2.08    —  —          4s  12  2  0.40  −4.68,5.49    7  3  0.35  −1.26,1.97      5s  3  1  NE      1  1  NE        6s  27  3  0.43  −0.20,1.06    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  33  8  1.05  0.59,1.51  0.12  —  —      —    Posterior  48  3  0.33  −0.82,1.47    Same          Jaw  Maxilla  52  9  0.88  0.43,1.33  0.10  Same        —    Mandible  29  3  0.42  −0.33,1.17    —  —        Jaw*group  Posterior mandible  18  2  0.32  −4.33,4.97  0.14  —  —      —    Posterior maxilla  30  3  0.33  −0.89,1.55    Same            Anterior mandible  11  2  0.43  −0.18,1.03    —  —          Anterior maxilla  22  8  1.08  0.65,1.52    —  —        Factor  Group  Fixed-appliance Tx  Rapid maxillary expansion  Arms  Studies  Mean  95% CI  PSG  Arms  Studies  Mean  95% CI  PSG  Any studies    Overall  161  17  0.79  0.54,1.03    20  3  0.36  −0.97,1.69    Tooth type  1s  32  12  0.82  0.38,1.26  0.04  —  —      0.51    2s  30  9  0.72  0.36,1.07    —  —          3s  13  6  0.37  −0.28,1.02    —  —          4s  22  4  0.29  −0.59,1.17    7  3  0.35  −1.26,1.97      5s  14  3  −0.08  −0.46,0.30    1  1  NE        6s  41  5  0.26  −0.30,0.81    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  79  16  0.85  0.62,1.08  0.08  —  —      ne    Posterior  82  5  0.24  −0.29,0.77    Same          Jaw  Maxilla  100  16  0.75  0.49,1.00  0.77  Same        ne    Mandible  61  7  0.57  0.01,1.13    —  —        Jaw*group  Posterior mandible  32  4  0.28  −0.42,0.98  0.06  —  —          Posterior maxilla  50  5  0.22  −0.35,0.78    Same            Anterior mandible  29  6  0.60  −0.14,1.34    —  —          Anterior maxilla  50  15  0.82  0.58,1.07    —  —        Only studies with complete Tx    Overall  81  9  0.86  0.42,1.30    20  3  0.36  −0.97,1.69    Tooth type  1s  14  6  1.00  0.44,1.57  0.10  —  —      0.51    2s  12  4  0.83  0.18,1.49    —  —          3s  4  3  0.67  −0.74,2.08    —  —          4s  12  2  0.40  −4.68,5.49    7  3  0.35  −1.26,1.97      5s  3  1  NE      1  1  NE        6s  27  3  0.43  −0.20,1.06    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  33  8  1.05  0.59,1.51  0.12  —  —      —    Posterior  48  3  0.33  −0.82,1.47    Same          Jaw  Maxilla  52  9  0.88  0.43,1.33  0.10  Same        —    Mandible  29  3  0.42  −0.33,1.17    —  —        Jaw*group  Posterior mandible  18  2  0.32  −4.33,4.97  0.14  —  —      —    Posterior maxilla  30  3  0.33  −0.89,1.55    Same            Anterior mandible  11  2  0.43  −0.18,1.03    —  —          Anterior maxilla  22  8  1.08  0.65,1.52    —  —        1s, central incisors; 2s, lateral incisors; 3s, canines; 4s; first premolars; 5s, second premolars; 6s, first molars; 7s, second molars; CI, confidence interval; NE, not estimated; PSG, between-subgroups P value; Tx, treatment. †Comparisons of SARME and Herbst were informed only from one study each and are omitted. View Large Table 3. Explorative analysis of linear OIRR including multiple measurements from all included studies.† Factor  Group  Fixed-appliance Tx  Rapid maxillary expansion  Arms  Studies  Mean  95% CI  PSG  Arms  Studies  Mean  95% CI  PSG  Any studies    Overall  161  17  0.79  0.54,1.03    20  3  0.36  −0.97,1.69    Tooth type  1s  32  12  0.82  0.38,1.26  0.04  —  —      0.51    2s  30  9  0.72  0.36,1.07    —  —          3s  13  6  0.37  −0.28,1.02    —  —          4s  22  4  0.29  −0.59,1.17    7  3  0.35  −1.26,1.97      5s  14  3  −0.08  −0.46,0.30    1  1  NE        6s  41  5  0.26  −0.30,0.81    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  79  16  0.85  0.62,1.08  0.08  —  —      ne    Posterior  82  5  0.24  −0.29,0.77    Same          Jaw  Maxilla  100  16  0.75  0.49,1.00  0.77  Same        ne    Mandible  61  7  0.57  0.01,1.13    —  —        Jaw*group  Posterior mandible  32  4  0.28  −0.42,0.98  0.06  —  —          Posterior maxilla  50  5  0.22  −0.35,0.78    Same            Anterior mandible  29  6  0.60  −0.14,1.34    —  —          Anterior maxilla  50  15  0.82  0.58,1.07    —  —        Only studies with complete Tx    Overall  81  9  0.86  0.42,1.30    20  3  0.36  −0.97,1.69    Tooth type  1s  14  6  1.00  0.44,1.57  0.10  —  —      0.51    2s  12  4  0.83  0.18,1.49    —  —          3s  4  3  0.67  −0.74,2.08    —  —          4s  12  2  0.40  −4.68,5.49    7  3  0.35  −1.26,1.97      5s  3  1  NE      1  1  NE        6s  27  3  0.43  −0.20,1.06    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  33  8  1.05  0.59,1.51  0.12  —  —      —    Posterior  48  3  0.33  −0.82,1.47    Same          Jaw  Maxilla  52  9  0.88  0.43,1.33  0.10  Same        —    Mandible  29  3  0.42  −0.33,1.17    —  —        Jaw*group  Posterior mandible  18  2  0.32  −4.33,4.97  0.14  —  —      —    Posterior maxilla  30  3  0.33  −0.89,1.55    Same            Anterior mandible  11  2  0.43  −0.18,1.03    —  —          Anterior maxilla  22  8  1.08  0.65,1.52    —  —        Factor  Group  Fixed-appliance Tx  Rapid maxillary expansion  Arms  Studies  Mean  95% CI  PSG  Arms  Studies  Mean  95% CI  PSG  Any studies    Overall  161  17  0.79  0.54,1.03    20  3  0.36  −0.97,1.69    Tooth type  1s  32  12  0.82  0.38,1.26  0.04  —  —      0.51    2s  30  9  0.72  0.36,1.07    —  —          3s  13  6  0.37  −0.28,1.02    —  —          4s  22  4  0.29  −0.59,1.17    7  3  0.35  −1.26,1.97      5s  14  3  −0.08  −0.46,0.30    1  1  NE        6s  41  5  0.26  −0.30,0.81    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  79  16  0.85  0.62,1.08  0.08  —  —      ne    Posterior  82  5  0.24  −0.29,0.77    Same          Jaw  Maxilla  100  16  0.75  0.49,1.00  0.77  Same        ne    Mandible  61  7  0.57  0.01,1.13    —  —        Jaw*group  Posterior mandible  32  4  0.28  −0.42,0.98  0.06  —  —          Posterior maxilla  50  5  0.22  −0.35,0.78    Same            Anterior mandible  29  6  0.60  −0.14,1.34    —  —          Anterior maxilla  50  15  0.82  0.58,1.07    —  —        Only studies with complete Tx    Overall  81  9  0.86  0.42,1.30    20  3  0.36  −0.97,1.69    Tooth type  1s  14  6  1.00  0.44,1.57  0.10  —  —      0.51    2s  12  4  0.83  0.18,1.49    —  —          3s  4  3  0.67  −0.74,2.08    —  —          4s  12  2  0.40  −4.68,5.49    7  3  0.35  −1.26,1.97      5s  3  1  NE      1  1  NE        6s  27  3  0.43  −0.20,1.06    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  33  8  1.05  0.59,1.51  0.12  —  —      —    Posterior  48  3  0.33  −0.82,1.47    Same          Jaw  Maxilla  52  9  0.88  0.43,1.33  0.10  Same        —    Mandible  29  3  0.42  −0.33,1.17    —  —        Jaw*group  Posterior mandible  18  2  0.32  −4.33,4.97  0.14  —  —      —    Posterior maxilla  30  3  0.33  −0.89,1.55    Same            Anterior mandible  11  2  0.43  −0.18,1.03    —  —          Anterior maxilla  22  8  1.08  0.65,1.52    —  —        1s, central incisors; 2s, lateral incisors; 3s, canines; 4s; first premolars; 5s, second premolars; 6s, first molars; 7s, second molars; CI, confidence interval; NE, not estimated; PSG, between-subgroups P value; Tx, treatment. †Comparisons of SARME and Herbst were informed only from one study each and are omitted. View Large Figure 2. View largeDownload slide Contour-enhanced funnel plot for the pooled orthodontically induced root resorption across all randomized and non-randomized studies included in the explorative analyses pertaining to any treatment duration (blue color) or only complete treatment (green color) (Supplementary Table 7). 1s, central incisors; 2s, lateral incisors; 3s, canines; 4s; first premolars; 5s, second premolars; 6s, first molars; 7s, second molars; CI, confidence interval; OIRR, orthodontically induced root resorption; Tx, treatment. Figure 2. View largeDownload slide Contour-enhanced funnel plot for the pooled orthodontically induced root resorption across all randomized and non-randomized studies included in the explorative analyses pertaining to any treatment duration (blue color) or only complete treatment (green color) (Supplementary Table 7). 1s, central incisors; 2s, lateral incisors; 3s, canines; 4s; first premolars; 5s, second premolars; 6s, first molars; 7s, second molars; CI, confidence interval; OIRR, orthodontically induced root resorption; Tx, treatment. Significant subgroup differences were found for various teeth, with the central incisors being affected the most (OIRR = 0.8 mm; 95% CI = 0.4–1.3 mm), followed by lateral incisors (OIRR = 0.7 mm; 95% CI = 0.4–1.1 mm), and canines (OIRR=0.4 mm; 95% CI = −0.3 to 1.0 mm). Significant differences were likewise found between anterior teeth (including incisors and canines) and posterior teeth (premolars and molars), where the OIRR of anterior teeth (OIRR = 0.9; 95% CI = 0.6–1.1) was significantly greater than that of posterior teeth (OIRR = 0.2 mm; 95% CI = −0.3 to 0.8 mm). Further, significant differences were found when localizing OIRR by both tooth group (anterior or posterior) and jaw (maxilla or mandible), where the largest amount of OIRR was found for the anterior maxilla (OIRR = 0.8 mm; 95% CI = 0.6–1.1 mm), followed by the anterior mandible (OIRR = 0.6 mm; 95% CI = −0.1 to 1.3 mm), the posterior mandible (OIRR = 0.3 mm; 95% CI = −0.4 to 1.0 mm), and the posterior maxilla (OIRR = 0.2 mm; 95% CI = −0.4 to 0.8 mm). The findings were relatively robust after limiting the analysis to the subset of studies assessing OIRR solely after the completion of orthodontic treatment, although average OIRR was slightly increased (0.9 mm compared to 0.8 mm). In addition, the comparison according to the jaw became now significant, with maxillary teeth being significantly more affected (OIRR = 0.9; 95% CI = 0.4–1.3 mm) compared to mandibular teeth (OIRR = 0.4 mm; 95% CI = −0.3 to 1.2). Building on the initial analyses of Table 3, a further exploratory analysis was conducted, where only the maxillary anterior teeth (that are most severely affected) were included and pooled together from each study, in order to have an adequate sample, on which to perform subgroup analyses, meta-regressions, sensitivity analyses, and assess reporting biases (Table 4). As such, studies assessing OIRR after treatment completion reported significantly greater amounts of OIRR compared to studies assessing initial phases of treatment (1.1 and 0.5 mm, respectively). Likewise, studies on extraction treatment reported significantly greater amounts of OIRR compared to studies on non-extraction treatment (0.8 and 0.5 mm, respectively). This might be explained by potentially longer treatment duration associated with extraction treatment (Figure 3), as meta-regression showed that OIRR was significantly associated with treatment duration (with an average increase in OIRR by 0.36 mm for every additional year). Table 4. Explorative analysis of pooled linear OIRR for maxillary anterior teeth/segments from all included studies. Analysis    Studies  Teeth  OIRR  95% CI  PSG  Subgroup analysis  Complete treatment  8  3078  1.11  0.80,1.42  0.09    No complete treatment  7  434  0.49  0.23,0.75      Extraction  6  2252  0.81  0.32,1.30  0.08    Non-extraction  5  678  0.51  0.25,0.78        Studies  Teeth  Coefficient  95% CI  P  Meta-regressions  Male % in sample  13  3386  −0.01  −0.03,0.02  0.57    Mean age in years  11  3260  0.01  −0.07,0.08  0.80    Duration in months  10  3106  0.03  0.00,0.06  0.04      Studies  Teeth  OIRR  95% CI  PSG  Sensitivity analysis  Prospective  6  560  0.59  0.20,0.97  0.28    Retrospective  9  2952  0.99  0.68,1.30      Voxel size ≤0.2 mm  3  2106  1.15  0.45,1.85  0.05    Voxel size >0.2 mm  10  1134  0.64  0.37,0.91      Root-level  7  2756  1.05  0.66,1.44  0.15    Tooth-level  8  756  0.62  0.34,0.91        Studies  Teeth  Coefficient  95% CI  P  Reporting biases  Egger’s test  15  3512  −2.65  −7.59,2.29  0.27  Analysis    Studies  Teeth  OIRR  95% CI  PSG  Subgroup analysis  Complete treatment  8  3078  1.11  0.80,1.42  0.09    No complete treatment  7  434  0.49  0.23,0.75      Extraction  6  2252  0.81  0.32,1.30  0.08    Non-extraction  5  678  0.51  0.25,0.78        Studies  Teeth  Coefficient  95% CI  P  Meta-regressions  Male % in sample  13  3386  −0.01  −0.03,0.02  0.57    Mean age in years  11  3260  0.01  −0.07,0.08  0.80    Duration in months  10  3106  0.03  0.00,0.06  0.04      Studies  Teeth  OIRR  95% CI  PSG  Sensitivity analysis  Prospective  6  560  0.59  0.20,0.97  0.28    Retrospective  9  2952  0.99  0.68,1.30      Voxel size ≤0.2 mm  3  2106  1.15  0.45,1.85  0.05    Voxel size >0.2 mm  10  1134  0.64  0.37,0.91      Root-level  7  2756  1.05  0.66,1.44  0.15    Tooth-level  8  756  0.62  0.34,0.91        Studies  Teeth  Coefficient  95% CI  P  Reporting biases  Egger’s test  15  3512  −2.65  −7.59,2.29  0.27  CI, confidence interval; OIRR, orthodontically induced root resorption; PSG, between-subgroups P value. View Large Table 4. Explorative analysis of pooled linear OIRR for maxillary anterior teeth/segments from all included studies. Analysis    Studies  Teeth  OIRR  95% CI  PSG  Subgroup analysis  Complete treatment  8  3078  1.11  0.80,1.42  0.09    No complete treatment  7  434  0.49  0.23,0.75      Extraction  6  2252  0.81  0.32,1.30  0.08    Non-extraction  5  678  0.51  0.25,0.78        Studies  Teeth  Coefficient  95% CI  P  Meta-regressions  Male % in sample  13  3386  −0.01  −0.03,0.02  0.57    Mean age in years  11  3260  0.01  −0.07,0.08  0.80    Duration in months  10  3106  0.03  0.00,0.06  0.04      Studies  Teeth  OIRR  95% CI  PSG  Sensitivity analysis  Prospective  6  560  0.59  0.20,0.97  0.28    Retrospective  9  2952  0.99  0.68,1.30      Voxel size ≤0.2 mm  3  2106  1.15  0.45,1.85  0.05    Voxel size >0.2 mm  10  1134  0.64  0.37,0.91      Root-level  7  2756  1.05  0.66,1.44  0.15    Tooth-level  8  756  0.62  0.34,0.91        Studies  Teeth  Coefficient  95% CI  P  Reporting biases  Egger’s test  15  3512  −2.65  −7.59,2.29  0.27  Analysis    Studies  Teeth  OIRR  95% CI  PSG  Subgroup analysis  Complete treatment  8  3078  1.11  0.80,1.42  0.09    No complete treatment  7  434  0.49  0.23,0.75      Extraction  6  2252  0.81  0.32,1.30  0.08    Non-extraction  5  678  0.51  0.25,0.78        Studies  Teeth  Coefficient  95% CI  P  Meta-regressions  Male % in sample  13  3386  −0.01  −0.03,0.02  0.57    Mean age in years  11  3260  0.01  −0.07,0.08  0.80    Duration in months  10  3106  0.03  0.00,0.06  0.04      Studies  Teeth  OIRR  95% CI  PSG  Sensitivity analysis  Prospective  6  560  0.59  0.20,0.97  0.28    Retrospective  9  2952  0.99  0.68,1.30      Voxel size ≤0.2 mm  3  2106  1.15  0.45,1.85  0.05    Voxel size >0.2 mm  10  1134  0.64  0.37,0.91      Root-level  7  2756  1.05  0.66,1.44  0.15    Tooth-level  8  756  0.62  0.34,0.91        Studies  Teeth  Coefficient  95% CI  P  Reporting biases  Egger’s test  15  3512  −2.65  −7.59,2.29  0.27  CI, confidence interval; OIRR, orthodontically induced root resorption; PSG, between-subgroups P value. View Large Figure 3. View largeDownload slide Random-effects meta-regression of the pooled orthodontically induced root resorption of the upper anterior teeth according to the mean duration of orthodontic treatment (Supplementary Table 8). OIRR, orthodontically induced root resorption. Figure 3. View largeDownload slide Random-effects meta-regression of the pooled orthodontically induced root resorption of the upper anterior teeth according to the mean duration of orthodontic treatment (Supplementary Table 8). OIRR, orthodontically induced root resorption. Additionally, sensitivity analyses indicated that retrospective studies tended to show greater OIRR than prospective studies (1.0 and 0.6 mm, respectively) and that studies measuring OIRR at the root level tended to show greater OIRR than those at the tooth level (1.1 and 0.6 mm, respectively), although both were not statistically significant (P = 0.28 and P = 0.15, respectively). In contrast, sensitivity analysis on the basis of CBCT voxel size indicated that studies using a small (≤0.2 mm) voxel size reported significantly greater OIRR than those using a larger (>0.2 mm) voxel size (1.2 and 0.6 mm, respectively). Finally, no signs of reporting bias were seen with the Egger test (P > 0.05). As far as RME treatment is concerned, the average linear OIRR among all three available studies (including 20 study arms) was found to be 0.4 mm (95% CI = −1.0 to 1.7 mm; Table 3), without any significant differences among teeth. As far as volumetric OIRR is concerned (Supplementary Table 8), the average OIRR after fixed appliance treatment was 15.4 mm3 (3 studies; 95% CI = −4.1 to 35.0 mm3), while no further analyses could be performed. The average OIRR after RME was 25.7 mm3 (4 studies; 95% CI = 6.9–44.5 mm3), with considerable differences among teeth, where first molar showed significantly greater OIRR than first premolars (40.2 and 14.8 mm3, respectively). Discussion Summary of evidence The present systematic review summarizes evidence from 30 unique randomized or non-randomized studies with three-dimensional imaging of linear or volumetric OIRR during or after orthodontic treatment, which is to our knowledge the first systematic review on the subject. Quantitative data synthesis for this systematic review was performed on two levels, including (a) initially direct comparisons of treatment-related factors within and across only randomized trials and (b) subsequent explorative analyses with indirect comparisons across all identified randomized and non-randomized studies. Robust comparisons from randomized trials should be used to inform clinical decision making, while the explorative analyses assess evidence of observational nature to characterize the pattern of OIRR. Direct meta-analysis pooling data from two randomized trials found no significant difference in OIRR between self-ligating and conventional brackets (Supplementary Table 7). Although the quality of existing evidence according to GRADE was low due to bias and imprecision, the results are in accordance with a previous meta-analysis of randomized trials using 2D imaging (35). Further, a single randomized trial indicated that intrusion of upright incisors anchored from posterior mini-implants (between second premolar and first molar) yielded more labial flaring and less OIRR than intrusion mechanics anchored anteriorly (between lateral incisor and canine) (Supplementary Table 7). This was attributed by the authors of that trial to the greater apical movement of the latter group (36), which had been proposed as a risk factor for OIRR (37). Explorative indirect meta-analyses on the average amount of OIRR across all studies associated with fixed appliance treatment gave a pooled OIRR of 0.79 mm across studies with any treatment duration and a pooled OIRR of 0.86 mm across studies reporting on complete fixed appliance treatment from bond to debond (Table 3). This small magnitude of the average OIRR is in line with the previous systematic review of Weltman et al. (1) and has probably no clinical relevance in terms of attachment loss, tooth mobility, or tooth prognosis (15). However, extreme variation is seen in the observed OIRR and statistical differences exist among various teeth, confirming results from previous studies that were performed with periapical or panoramic radiographs (3, 5, 17). As such, the greatest amount of OIRR was seen for the central incisor (0.82 mm), followed by the lateral incisor (0.72 mm), the canine (0.37 mm), the first premolar (0.29 mm), and lastly the first permanent molar (0.26 mm). Additionally, teeth of the anterior region (up to the canine) showed significantly greater amounts of OIRR than teeth of the posterior region (premolars and molars) (0.85 and 0.24 mm, respectively). Finally, statistically significant differences in OIRR were seen across the different regions within a jaw, with the anterior maxilla showing the greatest among of OIRR (0.82 mm), followed by the anterior mandible (0.60 mm), the posterior mandible (0.28 mm), and finally the posterior maxilla (0.22 mm) (Table 3). However, even the greatest amount of OIRR found from CBCT data at the anterior maxilla seems to be less than what it was found in previous studies conducted with periapical X-rays (3, 17, 38). This difference could be attributed to magnification errors and distortions due to root angulation that are very common among 2D imaging techniques (39). As expected, limiting the analysis only to studies assessing the complete treatment from appliance placement to appliance removal (Table 3), yielded slightly greater OIRR than the original analysis of studies assessing OIRR during or after treatment (0.86 and 0.79 mm, respectively), which indicates that future studies should probably measure OIRR after treatment completion to avoid underestimation of OIRR. This was confirmed by the explorative analysis after pooling all upper anterior teeth of each study in a single group (Table 4), which actually found greater difference in the OIRR of these teeth (1.11 and 0.49 mm for complete and non-complete treatment, respectively). Further, extraction treatment was associated with statistically greater OIRR of the upper anterior teeth than non-extraction treatment, which were 0.81 and 0.51 mm, respectively (Table 4). A possible explanation for this would be that extraction treatment and the subsequently needed space closure takes more time to complete than non-extraction treatment (40). Moreover, treatment duration was found in meta-regression of the present study to be significantly associated with the amount of OIRR (Table 4; Figure 3), which agrees with previous data from 2D imaging (5, 10, 41). This might explain, at least to some degree, the difference between extraction and non-extraction treatment, a finding that has also been described previously (10, 41, 42). Apart from fixed appliance treatment, some of the identified studies also employed CBCT to assess OIRR after treatment with RME expansion appliances. The average pooled linear OIRR associated with RME from explorative analyses was 0.36 mm with no significant differences across the assessed teeth (Table 3). The small magnitude of OIRR associated with RME seems to agree with the findings of Forst et al. (43), who found signs of OIRR with three-dimensional but not with 2D imaging, indicating that the latter approach may be worse suited to identify such small amounts of OIRR. In addition, direct evidence from randomized trials indicated that no significant difference in OIRR existed between conventional tooth-anchored Hyrax SARME and Hybrid tooth-and-implant-anchored SARME (Supplementary Table 7), with low quality of evidence according to GRADE due to bias and imprecision. In contrast, conventionally anchored Haas-type RME was associated with less OIRR than conventionally anchored Hyrax-type RME (Supplementary Table 7), which was supported by high quality of evidence according to GRADE. This agrees with previous data using 2D radiography (44), and might be possibly explained by the Haas expander distributing forces to both teeth and palatal vault and thereby reducing direct effects on tooth structures (45). However, these results should be interpreted with caution, as only a single randomized trial informed all direct comparisons. A single study assessing OIRR with CBCT before and after Class II treatment with the Herbst appliance was identified (46), although it was omitted from quantitative data synthesis due to the absence of other similar studies. Based on the results of this study, there was evidence of statistically significant OIRR affecting the teeth upon which the Herbst appliance was anchored (upper and lower first molars), though the amount of OIRR was small and not clinically significant. These findings are in accordance with previous studies that examined OIRR after Herbst treatment using 2D radiographs (47, 48). Finally, as far as design and measurement characteristics are concerned, sensitivity analysis using the CBCT’s voxel size showed that studies with voxel size ≤0.20 mm reported significantly greater OIRR compared to studies with greater voxel sizes (pooled OIRR: 1.15 and 0.64 mm, respectively; Table 4). This might indicate that the latter studies had too large voxel sizes to accurately identify areas of OIRR, and therefore, small voxel sizes might be preferable to accurately diagnose OIRR. This is supported by other data indicating that the method error in measuring tooth lengths with CBCT was smaller with voxel size of 0.20 mm compared to both intraoral radiography and CBCT with greater voxel sizes (49). However, it must be stated that the optimal voxel size to diagnose OIRR cannot be gleamed from this systematic review and even CBCT images with a voxel size of 0.20 mm might be unable to identify OIRR of small magnitudes. Decisions about the appropriateness of using CBCT protocols with high-resolution and radiation dose (19) to diagnose a mostly clinically irrelevant amount of OIRR need to be based on robust evaluations of the benefit to risk ratio for the patient. Prescribing CBCT solely for research purposes cannot be ethically justified and robust evidence on whether CBCT leads to diagnosis or treatment planning with improved treatment outcomes is needed. As far as displaced maxillary canines are concerned, current evidence indicates that for most cases treatment decisions are the same with either CBCT or 2D imaging (50), while the CBCT effective dose is 15–140 times higher than the 2D dose (51). Only a proportion of patients might benefit in terms of changed treatment plan, while there is currently no data to support improved treatment outcomes from this change. Furthermore, CBCT cannot be reliably recommended to diagnose tooth ankylosis, as although CBCT and histological findings coincided for some cases, false positivese were also seen (52). Therefore, currently evidence-based indications for CBCT are limited to some cases of maxillary displaced canines that might potentially benefit in terms of altered decision-making, even though evidence about the treatment outcomes is lacking. Strengths, limitations, and generalizability The strengths of the current systematic review include the a priori protocol, the comprehensive literature search including grey literature, the use of robust methods for qualitative and quantitative synthesis, and the open provision of the review’s dataset to increase transparency and reproducibility (53). In addition, no language restrictions were applied, and translation was arranged for non-English papers in order to reduce language bias. Also, no restrictions concerning publication year and status were applied, thereby maximizing data yield. Nevertheless, some limitations must also be noted, with the main one being the inclusion of retrospective studies, due to the scarcity of randomized and prospective non-randomized studies. Further, most of the included studies presented severe methodological limitations, while none of the studies used an untreated control group, since this would be difficult to justify ethically. In addition, the small number of studies that were included in several analyses and their incomplete reporting might have resulted in some analyses suffering from low statistical power. Finally, it should be noted that the present systematic review evaluated the average root resorption observed during or after treatment. In practice, the clinical relevance of OIRR is directly related to the tooth’s subsequent attachment loss and therefore only large magnitudes of root resorption are relevant to the patient. It might be useful in the future to categorize OIRR into clinically relevant categories of OIRR magnitude and use this as an outcome in clinical research. However, given the additive nature of many OIRR-related risk factors, evidence from this systematic review can be applied clinically to minimize OIRR in general. Given the broad eligibility criteria that were set and the inclusion of various clinical settings in many different countries, the results of the present systematic review are applicable to the average patient, provided they correspond to the age range and treatment phase of the included studies. Conclusions The present systematic review summarized evidence on clinical studies assessing treatment related OIRR with CBCT. Explorative analyses on the amount and pattern of OIRR after fixed appliance treatment from all identified randomized and non-randomized studies indicated that less than 1 mm of OIRR is seen on the average tooth with CBCT. Although OIRR measured by CBCT is higher than that measured by two-dimensional imaging, it has still probably little clinical relevance, which makes the added exposure to ionizing radiation through high-resolution CBCT protocols (19) questionable in terms of risk to benefit ratio. However, considerable differences in the amount of measured OIRR are seen according to tooth category, jaw, incorporation of extraction in the treatment plan, treatment duration, and CBCT settings. These data on the pattern of OIRR associated with orthodontic treatment might be useful both in everyday clinical practice and in the design of future studies. As far as direct comparisons from randomized studies are concerned, no differences in OIRR after fixed appliance treatment were seen between self-ligating and conventional brackets, although significant differences in OIRR after upper anterior intrusion were seen between anterior or posterior placement of TADs for anchorage. Further, OIRR after RME seemed to be similar between conventional Hyrax and hybrid Hyrax RME appliances, but less in conventional Hass appliances compared to conventional Hyrax appliances. However, these results should be interpreted with caution, due to the small number of identified randomized trials, their methodological limitations, and potential ethical implications about routine use of CBCT for OIRR diagnosis. Supplementary material Supplementary material is available at European Journal of Orthodontics online. Funding The study received no external funding to conduct this research through any of the authors involved. Conflicts of interest None to declare. Acknowledgements The authors would like to thank Dr. João Paulo Schwartz (Paulista State University, Araraquara, Brazil) for providing clarification on his work. References 1. Weltman, B., Vig, K.W., Fields, H.W., Shanker, S. and Kaizar, E.E. ( 2010) Root resorption associated with orthodontic tooth movement: a systematic review. American Journal of Orthodontics and Dentofacial Orthopedics , 137, 462– 76; discussion 12A. Google Scholar CrossRef Search ADS   2. Al-Qawasmi, R.A., Hartsfield, J.K. Jr, Everett, E.T., Flury, L., Liu, L., Foroud, T.M., Macri, J.V. and Roberts, W.E. ( 2003) Genetic predisposition to external apical root resorption. American Journal of Orthodontics and Dentofacial Orthopedics , 123, 242– 252. Google Scholar CrossRef Search ADS   3. Parker, R.J. and Harris, E.F. ( 1998) Directions of orthodontic tooth movements associated with external apical root resorption of the maxillary central incisor. American Journal of Orthodontics and Dentofacial Orthopedics , 114, 677– 683. Google Scholar CrossRef Search ADS   4. Brezniak, N. and Wasserstein, A. ( 1993) Root resorption after orthodontic treatment: part 2. Literature review. American Journal of Orthodontics and Dentofacial Orthopedics , 103, 138– 146. Google Scholar CrossRef Search ADS   5. Apajalahti, S. and Peltola, J.S. ( 2007) Apical root resorption after orthodontic treatment – a retrospective study. European Journal of Orthodontics , 29, 408– 412. Google Scholar CrossRef Search ADS   6. Tyrovola, J.B. and Spyropoulos, M.N. ( 2001) Effects of drugs and systemic factors on orthodontic treatment. Quintessence International (Berlin, Germany: 1985) , 32, 365– 371. 7. Oyama, K., Motoyoshi, M., Hirabayashi, M., Hosoi, K. and Shimizu, N. ( 2007) Effects of root morphology on stress distribution at the root apex. European Journal of Orthodontics , 29, 113– 117. Google Scholar CrossRef Search ADS   8. Marques, L.S., Ramos-Jorge, M.L., Rey, A.C., Armond, M.C. and Ruellas, A.C. ( 2010) Severe root resorption in orthodontic patients treated with the edgewise method: prevalence and predictive factors. American Journal of Orthodontics and Dentofacial Orthopedics , 137, 384– 388. Google Scholar CrossRef Search ADS   9. Linge, B.O. and Linge, L. ( 1983) Apical root resorption in upper anterior teeth. European Journal of Orthodontics , 5, 173– 183. Google Scholar CrossRef Search ADS   10. Sameshima, G.T. and Sinclair, P.M. ( 2001) Predicting and preventing root resorption: part II. Treatment factors. American Journal of Orthodontics and Dentofacial Orthopedics , 119, 511– 515. Google Scholar CrossRef Search ADS   11. Han, G., Huang, S., Von den Hoff, J.W., Zeng, X. and Kuijpers-Jagtman, A.M. ( 2005) Root resorption after orthodontic intrusion and extrusion: an intraindividual study. The Angle Orthodontist , 75, 912– 918. 12. Montenegro, V.C., Jones, A., Petocz, P., Gonzales, C. and Darendeliler, M.A. ( 2012) Physical properties of root cementum: part 22. Root resorption after the application of light and heavy extrusive orthodontic forces: a microcomputed tomography study. American Journal of Orthodontics and Dentofacial Orthopedics , 141, e1– e9. Google Scholar CrossRef Search ADS   13. Ballard, D.J., Jones, A.S., Petocz, P. and Darendeliler, M.A. ( 2009) Physical properties of root cementum: part 11. Continuous vs intermittent controlled orthodontic forces on root resorption. A microcomputed-tomography study. American Journal of Orthodontics and Dentofacial Orthopedics , 136, 8.e1– 8; discussion 8. 14. Lombardo, L., Bragazzi, R., Perissinotto, C., Mirabella, D. and Siciliani, G. ( 2013) Cone-beam computed tomography evaluation of periodontal and bone support loss in extraction cases. Progress in Orthodontics , 14, 29. Google Scholar CrossRef Search ADS   15. Jönsson, A., Malmgren, O. and Levander, E. ( 2007) Long-term follow-up of tooth mobility in maxillary incisors with orthodontically induced apical root resorption. European Journal of Orthodontics , 29, 482– 487. Google Scholar CrossRef Search ADS   16. Makedonas, D. and Hansen, K. ( 2008) Diagnosis, screening and treatment of root resorption in orthodontic practices in Greece and Sweden. The Angle Orthodontist , 78, 248– 253. Google Scholar CrossRef Search ADS   17. Sameshima, G.T. and Sinclair, P.M. ( 2001) Predicting and preventing root resorption: part I. diagnostic factors. American Journal of Orthodontics and Dentofacial Orthopedics , 119, 505– 510. Google Scholar CrossRef Search ADS   18. Ren, H., Chen, J., Deng, F., Zheng, L., Liu, X. and Dong, Y. ( 2013) Comparison of cone-beam computed tomography and periapical radiography for detecting simulated apical root resorption. The Angle Orthodontist , 83, 189– 195. Google Scholar CrossRef Search ADS   19. Ludlow, J.B., Timothy, R., Walker, C., Hunter, R., Benavides, E., Samuelson, D.B. and Scheske, M.J. ( 2015) Effective dose of dental CBCT-a meta analysis of published data and additional data for nine CBCT units. Dento Maxillo Facial Radiology , 44, 20140197. Google Scholar CrossRef Search ADS   20. Dudic, A., Giannopoulou, C., Leuzinger, M. and Kiliaridis, S. ( 2009) Detection of apical root resorption after orthodontic treatment by using panoramic radiography and cone-beam computed tomography of super-high resolution. American Journal of Orthodontics and Dentofacial Orthopedics , 135, 434– 437. Google Scholar CrossRef Search ADS   21. da Silveira, H.L., Silveira, H.E., Liedke, G.S., Lermen, C.A., Dos Santos, R.B. and de Figueiredo, J.A. ( 2007) Diagnostic ability of computed tomography to evaluate external root resorption in vitro. Dento Maxillo Facial Radiology , 36, 393– 396. Google Scholar CrossRef Search ADS   22. Akyalcin, S., Alexander, S.P., Silva, R.M. and English, J.D. ( 2015) Evaluation of three-dimensional root surface changes and resorption following rapid maxillary expansion: a cone beam computed tomography investigation. Orthodontics and Craniofacial Research , 18 Suppl 1, 117– 126. Google Scholar CrossRef Search ADS   23. Holberg, C., Steinhäuser, S., Geis, P. and Rudzki-Janson, I. ( 2005) Cone-beam computed tomography in orthodontics: benefits and limitations. Journal of Orofacial Orthopedics , 66, 434– 444. Google Scholar CrossRef Search ADS   24. Higgins, J.P.T. and Green, S. ( 2016) Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011] . The Cochrane Collaboration, 2011. www.cochrane handbook.org ( 21 April 2017, date last accessed). 25. Liberati, A., Altman, D.G., Tetzlaff, J., Mulrow, C., Gøtzsche, P.C., Ioannidis, J.P., Clarke, M., Devereaux, P.J., Kleijnen, J. and Moher, D. ( 2009) The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. Journal of Clinical Epidemiology , 62, e1– 34. Google Scholar CrossRef Search ADS   26. Downs, S.H. and Black, N. ( 1998) The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. Journal of Epidemiology and Community Health , 52, 377– 384. Google Scholar CrossRef Search ADS   27. Papageorgiou, S.N. ( 2014) Meta-analysis for orthodontists: part I–how to choose effect measure and statistical model. Journal of Orthodontics , 41, 317– 326. Google Scholar CrossRef Search ADS   28. Veroniki, A.A., Jackson, D., Viechtbauer, W., Bender, R., Bowden, J., Knapp, G., Kuss, O., Higgins, J.P., Langan, D. and Salanti, G. ( 2016) Methods to estimate the between-study variance and its uncertainty in meta-analysis. Research Synthesis Methods , 7, 55– 79. Google Scholar CrossRef Search ADS   29. Hedges, L.V., Tipton, E. and Johnson, M.C. ( 2010) Robust variance estimation in meta-regression with dependent effect size estimates. Research Synthesis Methods , 1, 39– 65. Google Scholar CrossRef Search ADS   30. Samandara, A., Papageorgiou, S.N., Ioannidou-Marathiotou, I., Kavvadia-Tsatala, S. and Papadopoulos, M.A. ( 2018) Evaluation of orthodontically induced external root resorption following orthodontic treatment using Cone Beam Computed Tomography (CBCT): a systematic review and meta-analysis. [Data set]. Zenodo . doi: 10.5281/zenodo.1185732 31. Egger, M., Davey Smith, G., Schneider, M. and Minder, C. ( 1997) Bias in meta-analysis detected by a simple, graphical test. BMJ (Clinical research ed.) , 315, 629– 634. Google Scholar CrossRef Search ADS   32. Guyatt, G.H., Oxman, A.D., Schünemann, H.J., Tugwell, P. and Knottnerus, A. ( 2011) GRADE guidelines: a new series of articles in the journal of clinical epidemiology. Journal of Clinical Epidemiology , 64, 380– 382. Google Scholar CrossRef Search ADS   33. Papageorgiou, S.N. ( 2014) Meta-analysis for orthodontists: part II–is all that glitters gold? Journal of Orthodontics , 41, 327– 336. Google Scholar CrossRef Search ADS   34. Lund, H., Gröndahl, K., Hansen, K. and Gröndahl, H.G. ( 2012) Apical root resorption during orthodontic treatment. A prospective study using cone beam CT. The Angle Orthodontist , 82, 480– 487. Google Scholar CrossRef Search ADS   35. Papageorgiou, S.N., Konstantinidis, I., Papadopoulou, K., Jäger, A. and Bourauel, C. ( 2014) Clinical effects of pre-adjusted edgewise orthodontic brackets: a systematic review and meta-analysis. European Journal of Orthodontics , 36, 350– 363. Google Scholar CrossRef Search ADS   36. Aras, I. and Tuncer, A.V. ( 2016) Comparison of anterior and posterior mini-implant-assisted maxillary incisor intrusion: root resorption and treatment efficiency. The Angle Orthodontist , 86, 746– 752. Google Scholar CrossRef Search ADS   37. Dermaut, L.R. and De Munck, A. ( 1986) Apical root resorption of upper incisors caused by intrusive tooth movement: a radiographic study. American Journal of Orthodontics and Dentofacial Orthopedics , 90, 321– 326. Google Scholar CrossRef Search ADS   38. McFadden, W.M., Engstrom, C., Engstrom, H. and Anholm, J.M. ( 1989) A study of the relationship between incisor intrusion and root shortening. American Journal of Orthodontics and Dentofacial Orthopedics , 96, 390– 396. Google Scholar CrossRef Search ADS   39. Tieu, L.D., Saltaji, H., Normando, D. and Flores-Mir, C. ( 2014) Radiologically determined orthodontically induced external apical root resorption in incisors after non-surgical orthodontic treatment of class II division 1 malocclusion: a systematic review. Progress in Orthodontics , 15, 48. Google Scholar CrossRef Search ADS   40. Papageorgiou, S.N., Höchli, D. and Eliades, T. ( 2017) Outcomes of comprehensive fixed appliance orthodontic treatment: a systematic review with meta-analysis and methodological overview. Korean Journal of Orthodontics , 47, 401– 413. Google Scholar CrossRef Search ADS   41. Maués, C.P., do Nascimento, R.R. and Vilella, O.d.e.V. ( 2015) Severe root resorption resulting from orthodontic treatment: prevalence and risk factors. Dental Press Journal of Orthodontics , 20, 52– 58. Google Scholar CrossRef Search ADS   42. McNab, S., Battistutta, D., Taverne, A. and Symons, A.L. ( 2000) External apical root resorption following orthodontic treatment. The Angle Orthodontist , 70, 227– 232. 43. Forst, D., Nijjar, S., Khaled, Y., Lagravere, M. and Flores-Mir, C. ( 2014) Radiographic assessment of external root resorption associated with jackscrew-based maxillary expansion therapies: a systematic review. European Journal of Orthodontics , 36, 576– 585. Google Scholar CrossRef Search ADS   44. Odenrick, L., Karlander, E.L., Pierce, A. and Kretschmar, U. ( 1991) Surface resorption following two forms of rapid maxillary expansion. European Journal of Orthodontics , 13, 264– 270. Google Scholar CrossRef Search ADS   45. Haas, A.J. ( 1965) The treatment of maxillary deficiency by opening the midpalatal suture. The Angle Orthodontist , 35, 200– 217. 46. Schwartz, J.P., Raveli, T.B., Almeida, K.C., Schwartz-Filho, H.O. and Raveli, D.B. ( 2015) Cone beam computed tomography study of apical root resorption induced by Herbst appliance. Journal of Applied Oral Science: Revista FOB , 23, 479– 485. Google Scholar CrossRef Search ADS   47. Nasiopoulos, A.T., Athanasiou, A.E., Papadopoulos, M.A., Kolokithas, G. and Ioannidou, I. ( 2006) Premolar root changes following treatment with the banded herbst appliance. Journal of Orofacial Orthopedics , 67, 261– 271. Google Scholar CrossRef Search ADS   48. Kinzinger, G.S., Savvaidis, S., Gross, U., Gülden, N., Ludwig, B. and Lisson, J. ( 2011) Effects of Class II treatment with a banded Herbst appliance on root lengths in the posterior dentition. American Journal of Orthodontics and Dentofacial Orthopedics , 139, 465– 469. Google Scholar CrossRef Search ADS   49. Lund, H., Gröndahl, K. and Gröndahl, H.G. ( 2010) Cone beam computed tomography for assessment of root length and marginal bone level during orthodontic treatment. The Angle Orthodontist , 80, 466– 473. Google Scholar CrossRef Search ADS   50. Christell, H., Birch, S., Bondemark, L., Horner, K. and Lindh, C.; SEDENTEXCT consortium. ( 2018) The impact of Cone Beam CT on financial costs and orthodontists’ treatment decisions in the management of maxillary canines with eruption disturbance. European Journal of Orthodontics , 40, 65– 73. Google Scholar CrossRef Search ADS   51. Kadesjö, N., Lynds, R., Nilsson, M. and Shi, X.Q. ( 2018) Radiation dose from X-ray examinations of impacted canines: cone beam CT vs two-dimensional imaging. Dento Maxillo Facial Radiology , 47, 20170305. Google Scholar CrossRef Search ADS   52. Ducommun, F., Bornstein, M.M., Bosshardt, D., Katsaros, C. and Dula, K. ( 2017) Diagnosis of tooth ankylosis using panoramic views, cone beam computed tomography, and histological data: a retrospective observational case series study. European Journal of Orthodontics . doi: 10.1093/ejo/cjx063. 53. Papageorgiou, S.N. and Cobourne, M.T. ( 2018) Data sharing in orthodontic research. Journal of Orthodontics , 45, 1– 3. Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press on behalf of the European Orthodontic Society. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The European Journal of Orthodontics Oxford University Press

Evaluation of orthodontically induced external root resorption following orthodontic treatment using cone beam computed tomography (CBCT): a systematic review and meta-analysis

Loading next page...
 
/lp/ou_press/evaluation-of-orthodontically-induced-external-root-resorption-HlMaQ0097v
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the European Orthodontic Society. All rights reserved. For permissions, please email: journals.permissions@oup.com
ISSN
0141-5387
eISSN
1460-2210
D.O.I.
10.1093/ejo/cjy027
Publisher site
See Article on Publisher Site

Abstract

Summary Background Orthodontically induced external root resorption (OIRR) is a pathologic consequence of orthodontic tooth movement. However, the limitations of two-dimensional radiography suggest that cone beam computed tomography (CBCT) with its three-dimensional capabilities might be more suitable to assess OIRR. Objective The aim of this study was to assess in an evidence-based manner data on linear or volumetric OIRR measurements of permanent teeth by means of CBCT, during and/or after the end of orthodontic treatment. Search methods Unrestricted electronic and hand searches were performed up to January 2017 in 15 databases. Selection criteria methods Randomized clinical trials, prospective, and retrospective non-randomized studies assessing OIRR during and/or after orthodontic treatment using CBCT in human patients were included. Data collection and analysis After duplicate study selection, data extraction, and risk-of-bias assessment according to the Cochrane guidelines, random-effects meta-analyses, followed by subgroup, meta-regression, and sensitivity analyses were also performed in order to evaluate factors that affect OIRR. Results A total of 33 studies (30 datasets) were included in the qualitative analysis while data from 27 of them were included in the quantitative analysis. Direct comparisons from randomized trials found little to no influence of appliance-related factors on OIRR. Explorative analyses including non-randomized studies found a pooled OIRR of 0.79 mm based on all included studies and 0.86 mm when OIRR was assessed at the end of orthodontic treatment. Statistically significant differences in OIRR were found according to tooth type or jaw, inclusion of extractions, treatment duration, and diagnostic accuracy of the CBCT. Conclusions Based on the results of this study, CBCT seems to be a reliable tool to examine OIRR during or at the end of orthodontic treatment. Although the average OIRR measured with CBCT seems to lack clinical relevance, there are certain factors that may affect OIRR following orthodontic treatment. Nevertheless, due to data heterogeneity and low quality of the included studies, the corresponding results should be interpreted with some caution. Registration PROSPERO (CRD42016030131) Introduction Rationale Orthodontically induced root resorption (OIRR) is a frequent complication of orthodontic tooth movement with complex aetiology and several patient- or treatment-related risk factors being suggested as relevant (1). Proposed patient-related factors include genetics (2), sex (3), age (4), tooth type (5), systemic factors (6), root morphology (7), and history of trauma or previous root resorption (8). On the other hand, proposed treatment-related factors include appliance type (9), treatment duration (10), type of tooth movement (11), applied force magnitude (12), duration of force application (13), and extraction treatment (14). Diagnosis of OIRR is done in most cases radiographically, since clinical symptoms are for the most part absent and increased tooth mobility is seen only in severe cases with additional alveolar bone loss through time (15). Usually, two-dimensional radiographic (2D) methods like periapical or panoramic radiographs are taken before, during, and after orthodontic treatment to monitor OIRR (16). Studies using 2D imaging techniques found OIRR to be less than 0.60 mm at the end of treatment with the maxillary incisors being more frequently and severely affected (5, 10, 17). However, OIRR affects every aspect of the root surface in all three-dimensions and therefore, 2D radiographs might mask the true amount of OIRR. Additionally, the true extent of OIRR might be misestimated due to magnification errors and problematic repeatability of 2D radiographs (18). Cone beam computed tomography (CBCT) was introduced as a three-dimensional diagnostic modality capable of imaging complex dental and maxillofacial structures and was quickly adopted in dentistry due to its reduced radiation dose compared to conventional computed tomography (19). On the other side, considerable variation in CBCT radiation doses exists according to the exposure protocol and there might be as much as 15-fold difference between a low-dose and a high-resolution protocol with the same field of view (19). The diagnostic value of CBCT in the diagnosis of OIRR lies on its ability to obtain distortion-free and reproducible images of the roots (20) with high sensitivity and specificity (21). Additionally, three-dimensional reconstruction of 2D CBCT slices enables accurate quantification of both linear and volumetric OIRR measurements (22) and compensates for changes in root position or angulation during orthodontic treatment (23). Objectives In light of the considerable differences between the two imaging modalities and potential advantages of three-dimensional imaging with CBCT in quantifying OIRR, this systematic review sought to summarize evidence from existing clinical studies assessing linear or volumetric OIRR with CBCT in order to primarily quantify the average OIRR that can be expected after orthodontic treatment. The secondary aim was to identify significant patient-, treatment-, or imaging-related factors that are significantly associated with OIRR. Materials and methods Protocol and registration The protocol for the present systematic review was developed a priori according to the Cochrane Handbook for Systematic Reviews of Interventions 5.1.0 (24) and registered in PROSPERO (CRD42016030131). The present systematic review is conducted according to the guidelines of the Cochrane Handbook for Systematic Reviews of Interventions version 5.1.0 (24) and is reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (25). Information sources and literature search A total of fifteen databases were systematically searched from their inception up to July 2015 for published, ongoing, or unpublished studies without any limitation regarding language, publication year, or publication status (Supplementary Table 1). Electronic searches were updated in January 2017, while the reference and citation lists of all included studies and identified relevant systematic reviews were manually searched for eligible studies. Eligibility criteria Eligibility criteria were determined a priori according to the Participant–Intervention–Comparison–Outcome–Study design (PICOS) schema (Supplementary Table 2). As a result, randomized clinical trials and prospective or retrospective non-randomized clinical studies assessing linear or volumetric OIRR with CBCT in patients in the permanent dentition with fully developed tooth roots examined during and/or after orthodontic treatment were included. Studies with untreated control groups were initially not expected to be found, since it is unethical to expose patients to radiation without proper indications and when no orthodontic intervention has been performed, but were nevertheless included, if available. All other clinical or non-clinical study designs not fulfilling the criteria were excluded. Study selection The titles/abstracts of identified studies were screened by one author (AS), while an independent checking of their full texts for eligibility was performed by another author (SNP), and conflicts were resolved by a third author (MAP). Data collection Data extraction was performed independently by two authors (AS, SNP) using pre-determined and piloted extraction forms covering study design, clinical setting, country, patient characteristics, number/age/sex of patients, applied orthodontic intervention, treatment duration, teeth assessed for OIRR, outcome assessed, timing of outcome measurement, CBCT technical characteristics, and potential conflict of interests. Discrepancies in data extraction between the two authors were likewise resolved by a third author (MAP), while authors of included trials were contacted for missing or unclear information. Risk of bias in individual trials The Cochrane Collaboration’s Risk of Bias Tool was used to assess the risk of bias in randomized clinical trials (24), while a modified version of the Downs and Black checklist (26) was used to assess the risk of bias of non-randomized studies. Risk of bias was assessed independently by two authors (AS, SNP) and potential conflicts between them were resolved by the third author (MAP). Data synthesis The primary outcome of this systematic review was the amount of OIRR in patients treated with fixed appliances. This was measured either as change in length or volume and either after an initial phase of treatment or after the end of treatment, but was assessed separately. The secondary outcome was the amount of OIRR after conventional rapid maxillary expansion (RME), surgically assisted RME, functional appliance treatment, or any other kind of appliance. For both the primary and all secondary outcomes the average amount of OIRR, as well as the influence of any patient-, tooth-, treatment-, or outcome-related factors was investigated, including among others patient age, sex, tooth type, treatment duration, inclusion of extractions, voxel size, and OIRR measurement unit. Data were summarized and considered suitable for pooling if the corresponding studies used similar interventions in the same way and reported similar outcomes. The amount of OIRR in terms of alterations in the root length or volume measured by CBCTs was calculated as initial length/volume minus final length/volume, so that positive values indicated root resorption and negative values indicated root elongation. From each included study data from all available study arms pertaining to different teeth/roots, appliances, or follow-up were extracted. Since OIRR as response to treatment was expected to vary according to population variations, genetic predisposition, differences in the applied treatment mechanics, and differences in CBCT protocol, a random-effects model was deemed appropriate to calculate the mean across distributions of true OIRR effects (27). For the main analysis all study arms of each study were pooled together (24) and the novel Paule–Mandel random-effects estimator was used over the DerSimonian–Laird one, since it outperforms the latter (28). For the main analysis, the mean difference (MD) in OIRR was used to quantify differences between groups. In explorative meta-analyses where multiple study arms from a study were included, a robust variance estimation of random-effects using the DerSimonian–Laird method was performed to account for clustering of study arms within studies (29). In the explorative analyses, the average OIRR within- and across-groups was quantified. The extent and impact of heterogeneity across studies was assessed by forest plot inspection and formally quantified using the tau2 and I2 metrics, respectively (24). The following arbitrary categories for I2 were adopted (24): I2 = 0–40 per cent: non-important inconsistency; I2 = 30–60 per cent: moderate inconsistency; I2 = 50–90 per cent: substantial inconsistency; and I2 > 75–100 per cent: considerable inconsistency. All analyses were performed on Stata SE 14.0 (StataCorp, College Station, Texas, USA) by one author (SNP) with an open dataset in Zenodo (30). The level of significance was set to a two-sided P < 0.05 except for tests of within- or between-subgroups heterogeneity, where it was set at P < 0.10 (24). Risk of bias across studies and additional analyses Potential sources of heterogeneity attributed to patient-, tooth-, treatment-, or outcome-related factors were assessed with the use of pre-specified mixed-effects subgroup and meta-regression analyses with at least five studies. Investigated factors included patient age, sex, tooth type, treatment phase (initial or complete treatment), treatment duration, and inclusion of extractions. Indications of reporting biases (including small-study effects and publication bias) were assessed with Egger’s linear regression test (31) and contour-enhanced funnel plots, should 10 or more trials be pooled in a meta-analysis. The overall quality of evidence (confidence in effect estimates) for each outcome was rated using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach (32). The cut-offs of minimal clinical important, large, and very large effects for linear OIRR were arbitrarily defined a priori as 0.75, 1.50, 3.00 mm (or 10.0, 20.0, and 40.0 mm3 for volumetric OIRR) and used to augment the produced forest plots with contours of effect magnitude (33). Robustness of the results was checked with sensitivity analyses according to the following study design and outcome measurement method factors: randomized or non-randomized studies, prospective or retrospective studies, OIRR measurement at the tooth or root level, and diagnostic accuracy (based on a study (34) finding differences in measured OIRR using a cut-off voxel size of 0.2 mm). Results Study selection The electronic and manual literature searches yielded 3442 records that were reduced to 2206 records after removal of duplicates. A total of 2162 records were excluded on the basis of title, abstract, and full text according to the specific inclusion and exclusion criteria (Supplementary Table 3a–c). The full text of six publications could not be retrieved, even after author communication attempts via email (Supplementary Table 4). Consequently, 33 publications were included in the study, which after collating multiple publications pertaining to the same study, reflected a total of 30 unique included data sets (Figure 1). Figure 1. View largeDownload slide Flow diagram for the identification and selection of studies in this systematic review. *Three studies could not be included in the analyses, due to missing data. Figure 1. View largeDownload slide Flow diagram for the identification and selection of studies in this systematic review. *Three studies could not be included in the analyses, due to missing data. Study characteristics The characteristics of the 30 included studies are presented in Table 1 and Supplementary Table 5. Six of the included studies were randomized clinical trials, while 6 were prospective non-randomized studies, and the remaining 18 studies were retrospective non-randomized studies. These included a total of 1219 patients, with 37.7 per cent of patients being male (from the 28 studies reporting sex) and with mean ages ranging between 11.4 and 26.6 years with an average of 14.7 years (from the 27 studies reporting age). CBCTs were taken in all the included studies at two or three time points before, during and/or after orthodontic treatment in order to evaluate OIRR. Different appliances were utilized in most of the included studies with fixed appliances used in 22 studies, RME in 6 studies, surgically-assisted RME in 1 study, and Herbst appliance in 1 study. The amount of OIRR was assessed at various teeth of all types (with the exception of third molars) and both jaws. Table 1. Characteristics of included studies regarding patients and interventions. Study  Design  Characteristics of patients  No patients (M/F); mAge (years)  Intervention  TxDur  COI  Abdel-Kader 2016  rNRS; Uni; EGY  PM1 Ex and C retraction  16 (6/10); 16.4  FFA; E-chain  6.3 months  NR  Ahn 2013  rNRS; Uni; KOR  Cl. I; d-a protrusion; 4PM Ex; en masse Retr;  37 (0/37); 26.6  22-slot; Roth; TPA/HG; E-chain (200 g)  21.7 months  NR  Ajmera 2014  pNRS; Uni; IND  Max-ant. proclination; PM1 Ex; en masse Retr/Intr.  48 (21/27); 13.6  22-slot; MBT; TADs (200 g)  NR  NR  Alexander 2014; Akyalcin 2015  rNRS; Uni; USA  ME need  24 (10/14); 12.8  Tooth-anch RME (0.25 mm/day; 5 months Retw)  4.8 months  NR  Aras and Tuncer 2016  RCT; Uni; TUR  Cl.I or II; OB ≥5 mm; max. ant. Intr;  Gp1: 16 (6/10); 19.3 Gp2: 16 (6/10); 19.0  Gp1: 2 MIs between I2-C; SS segmental wire; E-chain Gp2: 2 MIs between PM2-M1; Burstone TMA IntrArch (40 g/side)  4.0 months  NR  Baysal 2012  rNRS; Uni; TUR  ME need  25 (11/14); 12.7  Tooth-anch RME (0.5 mm/day)  NR  NR  Bolerjack 2005  rNRS; pract; USA  Non-Ex/Ex (4 PMs); sliding mechanics  95 (41/54); 12.7  22-slot; Roth  26.4 months  NR  Castro 2013; Castro 2015  rNRS; pract; BRA  Cl. I Non-Ex  30 (11/19); 13.0  22-slot; Roth  22.0 months  NR  Dindaroglu 2016  RCT; Uni; TUR  ME need  Gp1/2 33(17/16); 12.8 Gp1: 16 (NR); 12.9 Gp2: 17 (NR); 12.8  Gp1/2 (0.5 mm/day; 6 months Retw) Gp1: Tooth-anch Hyrax RME Gp2: Tooth-anch Haas RME  6.0 months  NR  Forst 2015  RCT**; Uni; CAN  ME need  Gp1: 20 (5/15); 14.1 Gp2: 21 (8/13); 14.2 Gp3: 21 (6/15); 12.9  Gp1: Tooth-anch RME (0.5 mm/day; 6 months Retw; 6 months Retw/o) Gp2: Bone-anch RME (0.25 mm/day; 6 months Retw; 6 months Retw/o) Gp3: No Tx  12.0 months ***  NR  Guo 2016  rNRS; Uni; CHN  —  174 (68/106); 14.1  FFA (SW); ‘light forces’  20.6 months  None  Harris 2015  rNRS; Uni; TUR  ME need  Gp1: 20 (8/12); 13.8 Gp2: 20 (7/13);13.8  Gp1: Tooth-anch RME (0.4 mm/day; 3 months Retw) Gp2: Bone-anch RME (0.4 mm/day; 3 months Retw)  3.0 months ***  NR  Johnson 2010  rNRS; pract; USA  FFA (1- or 2-phase); Ex/ Non-Ex  Gp1: 24 (11/13); 12.7 Gp2: 33 (10/23); 19.1  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 19.5 months Gp2: 13.9 months  NR  Kau 2011  pNRS; Uni; USA  Cl.I; PMs Ex  14 (NR); 20.3  FFA (SW) + vibration  6.0 months  NR  Kaylar 2016  RCT; Uni; TUR  ME need  Gp1: 10 (6/4); 19.3 Gp2: 10 (3/7); 19.2  Gp1: Tooth-anch SARME (0.5 mm/day; 6 months Retw) Gp2: Bone-anch SARME (0.5 mm/day; 6 months Retw)  6.0 months ***  None  Leite 2012  RCT; Uni; BRA  Cl. I; moderate crowding; non-Ex  Gp1: 11 (6/5); 20.6* Gp2: 8 (2/6); 20.6*  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  6.0 months  NR  Li 2013  rNRS; Uni; CHN  Cl. I–II; molar intrusion  12 (4/8); 24.3  TADs; E-chain (150 g)  6.0 months  NR  Liu 2016  RCT; Hosp; CHN  Ex/Non-Ex  Gp1: 25 (11/14); 15.3 Gp2: 25 (9/16); 15.2  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  NR  NR  Lombardo 2013  rNRS; pract; ITA  Cl. II/1; Ex (4 PMs)/non-Ex  22 (10/12); 11.4  Tweed-Merrifield technique  20.5 months  NR  Lynch 2011  rNRS; pract; USA  Cl. I-II; Ex/Non-Ex  122 (53/69); 13.2  22-slot; loops/E-chains  21.8 months  NR  Ma 2013  pNRS; Uni; CHN  ME need  32 (0/32); 14.3  Magnetic RME (increment/month)  9.3 months  NR  Nakada 2016  rNRS; Uni; JPN  4 PM Ex  30 (8/22); 21.9  FFA  NR  NR  Nimeri 2014  pNRS; Uni; USA  Cl. I; crowding; PM Ex  20 (5/15); NR  FFA + LLLT  NR  Comm  Patel 2012  rNRS; pract; USA  FFA (1- phase); Ex/ non-Ex  Gp1: 32 (16/16); 12.6 Gp2: 34 (11/23); 18.9  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 18.1 months Gp2: 14.2 months  NR  Schwarz 2015  rNRS; Uni; BRA  Cl. II; OJ > 5 mm  23 (11/12); 15.8  Herbst appliance  8.5 months  NR  Oliveira 2011; Oliveira 2016  pNRS; Uni; BRA  Cl. I–II/1; protrusion; Mx PM1 Ex  11 (5/6); adult  FFA; omega loops  NR  NR  Wang 2013  pNRS; Uni; CHN  Skeletal Cl. III; pre-surgical decompensation need; Non-Ex  Gp1: 30 (14/16); 24.8 Gp2: 26 (17/9); 23.5  Gp1: FFA Gp2: FFA + corticotomy  Gp1: 13.3 months Gp2: 7.8 months  None  Wang 2015  rNRS; Uni; CHN  Skeletal Cl. III; pre- surgical decompensation need; non-Ex  30 (13/17); adult  FFA; MBT  8.7 months  NR  Wen 2016  rNRS; Uni; CHN  PM Ex and space closure; TAD  12 (3/9); adult  FFA  30.0 months  NR  Yang 2016  rNRS; Uni; CHN  4 PM Ex; ant. Retr.  7 (NR); 22.3 years  FFA  NR  NR  Study  Design  Characteristics of patients  No patients (M/F); mAge (years)  Intervention  TxDur  COI  Abdel-Kader 2016  rNRS; Uni; EGY  PM1 Ex and C retraction  16 (6/10); 16.4  FFA; E-chain  6.3 months  NR  Ahn 2013  rNRS; Uni; KOR  Cl. I; d-a protrusion; 4PM Ex; en masse Retr;  37 (0/37); 26.6  22-slot; Roth; TPA/HG; E-chain (200 g)  21.7 months  NR  Ajmera 2014  pNRS; Uni; IND  Max-ant. proclination; PM1 Ex; en masse Retr/Intr.  48 (21/27); 13.6  22-slot; MBT; TADs (200 g)  NR  NR  Alexander 2014; Akyalcin 2015  rNRS; Uni; USA  ME need  24 (10/14); 12.8  Tooth-anch RME (0.25 mm/day; 5 months Retw)  4.8 months  NR  Aras and Tuncer 2016  RCT; Uni; TUR  Cl.I or II; OB ≥5 mm; max. ant. Intr;  Gp1: 16 (6/10); 19.3 Gp2: 16 (6/10); 19.0  Gp1: 2 MIs between I2-C; SS segmental wire; E-chain Gp2: 2 MIs between PM2-M1; Burstone TMA IntrArch (40 g/side)  4.0 months  NR  Baysal 2012  rNRS; Uni; TUR  ME need  25 (11/14); 12.7  Tooth-anch RME (0.5 mm/day)  NR  NR  Bolerjack 2005  rNRS; pract; USA  Non-Ex/Ex (4 PMs); sliding mechanics  95 (41/54); 12.7  22-slot; Roth  26.4 months  NR  Castro 2013; Castro 2015  rNRS; pract; BRA  Cl. I Non-Ex  30 (11/19); 13.0  22-slot; Roth  22.0 months  NR  Dindaroglu 2016  RCT; Uni; TUR  ME need  Gp1/2 33(17/16); 12.8 Gp1: 16 (NR); 12.9 Gp2: 17 (NR); 12.8  Gp1/2 (0.5 mm/day; 6 months Retw) Gp1: Tooth-anch Hyrax RME Gp2: Tooth-anch Haas RME  6.0 months  NR  Forst 2015  RCT**; Uni; CAN  ME need  Gp1: 20 (5/15); 14.1 Gp2: 21 (8/13); 14.2 Gp3: 21 (6/15); 12.9  Gp1: Tooth-anch RME (0.5 mm/day; 6 months Retw; 6 months Retw/o) Gp2: Bone-anch RME (0.25 mm/day; 6 months Retw; 6 months Retw/o) Gp3: No Tx  12.0 months ***  NR  Guo 2016  rNRS; Uni; CHN  —  174 (68/106); 14.1  FFA (SW); ‘light forces’  20.6 months  None  Harris 2015  rNRS; Uni; TUR  ME need  Gp1: 20 (8/12); 13.8 Gp2: 20 (7/13);13.8  Gp1: Tooth-anch RME (0.4 mm/day; 3 months Retw) Gp2: Bone-anch RME (0.4 mm/day; 3 months Retw)  3.0 months ***  NR  Johnson 2010  rNRS; pract; USA  FFA (1- or 2-phase); Ex/ Non-Ex  Gp1: 24 (11/13); 12.7 Gp2: 33 (10/23); 19.1  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 19.5 months Gp2: 13.9 months  NR  Kau 2011  pNRS; Uni; USA  Cl.I; PMs Ex  14 (NR); 20.3  FFA (SW) + vibration  6.0 months  NR  Kaylar 2016  RCT; Uni; TUR  ME need  Gp1: 10 (6/4); 19.3 Gp2: 10 (3/7); 19.2  Gp1: Tooth-anch SARME (0.5 mm/day; 6 months Retw) Gp2: Bone-anch SARME (0.5 mm/day; 6 months Retw)  6.0 months ***  None  Leite 2012  RCT; Uni; BRA  Cl. I; moderate crowding; non-Ex  Gp1: 11 (6/5); 20.6* Gp2: 8 (2/6); 20.6*  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  6.0 months  NR  Li 2013  rNRS; Uni; CHN  Cl. I–II; molar intrusion  12 (4/8); 24.3  TADs; E-chain (150 g)  6.0 months  NR  Liu 2016  RCT; Hosp; CHN  Ex/Non-Ex  Gp1: 25 (11/14); 15.3 Gp2: 25 (9/16); 15.2  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  NR  NR  Lombardo 2013  rNRS; pract; ITA  Cl. II/1; Ex (4 PMs)/non-Ex  22 (10/12); 11.4  Tweed-Merrifield technique  20.5 months  NR  Lynch 2011  rNRS; pract; USA  Cl. I-II; Ex/Non-Ex  122 (53/69); 13.2  22-slot; loops/E-chains  21.8 months  NR  Ma 2013  pNRS; Uni; CHN  ME need  32 (0/32); 14.3  Magnetic RME (increment/month)  9.3 months  NR  Nakada 2016  rNRS; Uni; JPN  4 PM Ex  30 (8/22); 21.9  FFA  NR  NR  Nimeri 2014  pNRS; Uni; USA  Cl. I; crowding; PM Ex  20 (5/15); NR  FFA + LLLT  NR  Comm  Patel 2012  rNRS; pract; USA  FFA (1- phase); Ex/ non-Ex  Gp1: 32 (16/16); 12.6 Gp2: 34 (11/23); 18.9  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 18.1 months Gp2: 14.2 months  NR  Schwarz 2015  rNRS; Uni; BRA  Cl. II; OJ > 5 mm  23 (11/12); 15.8  Herbst appliance  8.5 months  NR  Oliveira 2011; Oliveira 2016  pNRS; Uni; BRA  Cl. I–II/1; protrusion; Mx PM1 Ex  11 (5/6); adult  FFA; omega loops  NR  NR  Wang 2013  pNRS; Uni; CHN  Skeletal Cl. III; pre-surgical decompensation need; Non-Ex  Gp1: 30 (14/16); 24.8 Gp2: 26 (17/9); 23.5  Gp1: FFA Gp2: FFA + corticotomy  Gp1: 13.3 months Gp2: 7.8 months  None  Wang 2015  rNRS; Uni; CHN  Skeletal Cl. III; pre- surgical decompensation need; non-Ex  30 (13/17); adult  FFA; MBT  8.7 months  NR  Wen 2016  rNRS; Uni; CHN  PM Ex and space closure; TAD  12 (3/9); adult  FFA  30.0 months  NR  Yang 2016  rNRS; Uni; CHN  4 PM Ex; ant. Retr.  7 (NR); 22.3 years  FFA  NR  NR  Ant. anterior; anch, anchorage; C, canine; CLB, conventional brackets; Comm, commercial; COI, conflict of interest; Cl I, Class I; Cl II, Class II; Cl II/1, Class II division 1; Cl III, Class III; d-a protrusion, dentoalveolar protrusion; E-chain, elastic chain; Ex, extraction; FFA, fixed appliances; Gp, group; HG, headgear; Intr, intrusion; non-ex, non extraction; I2, lateral incisor; LLLT, low level light therapy; MIs, mini-screw implants; M1, first molar; Max, maxilla; ME, maxillary expansion; M/F, male/female; NR, not reported; OB, overbite; OJ, overjet; pract, private practice; PM, premolar; PM1, first premolar; PM2, second premolar; pNRS, prospective non-randomized study; RCT, randomized clinical trial; rNRS, retrospective non-randomized study; RME, rapid maxillary expansion; Retw, retention with expansion appliance; Retw/o, retention without expansion appliance; Retr, retraction; SARME, surgically assisted rapid maxillary expansion; SLB, self-ligating brackets; SW, straight wire; SS, stainless steel; Tx, treatment; TAD, temporary anchorage device; TPA, transpalatal arch; TxDur, treatment duration; Uni, university. *For both groups. **Randomization was performed for another study, not on the basis of root resorption. ***Period including a retention period about 12 months. ****Right and left teeth of the same category were averaged for analyses. View Large Table 1. Characteristics of included studies regarding patients and interventions. Study  Design  Characteristics of patients  No patients (M/F); mAge (years)  Intervention  TxDur  COI  Abdel-Kader 2016  rNRS; Uni; EGY  PM1 Ex and C retraction  16 (6/10); 16.4  FFA; E-chain  6.3 months  NR  Ahn 2013  rNRS; Uni; KOR  Cl. I; d-a protrusion; 4PM Ex; en masse Retr;  37 (0/37); 26.6  22-slot; Roth; TPA/HG; E-chain (200 g)  21.7 months  NR  Ajmera 2014  pNRS; Uni; IND  Max-ant. proclination; PM1 Ex; en masse Retr/Intr.  48 (21/27); 13.6  22-slot; MBT; TADs (200 g)  NR  NR  Alexander 2014; Akyalcin 2015  rNRS; Uni; USA  ME need  24 (10/14); 12.8  Tooth-anch RME (0.25 mm/day; 5 months Retw)  4.8 months  NR  Aras and Tuncer 2016  RCT; Uni; TUR  Cl.I or II; OB ≥5 mm; max. ant. Intr;  Gp1: 16 (6/10); 19.3 Gp2: 16 (6/10); 19.0  Gp1: 2 MIs between I2-C; SS segmental wire; E-chain Gp2: 2 MIs between PM2-M1; Burstone TMA IntrArch (40 g/side)  4.0 months  NR  Baysal 2012  rNRS; Uni; TUR  ME need  25 (11/14); 12.7  Tooth-anch RME (0.5 mm/day)  NR  NR  Bolerjack 2005  rNRS; pract; USA  Non-Ex/Ex (4 PMs); sliding mechanics  95 (41/54); 12.7  22-slot; Roth  26.4 months  NR  Castro 2013; Castro 2015  rNRS; pract; BRA  Cl. I Non-Ex  30 (11/19); 13.0  22-slot; Roth  22.0 months  NR  Dindaroglu 2016  RCT; Uni; TUR  ME need  Gp1/2 33(17/16); 12.8 Gp1: 16 (NR); 12.9 Gp2: 17 (NR); 12.8  Gp1/2 (0.5 mm/day; 6 months Retw) Gp1: Tooth-anch Hyrax RME Gp2: Tooth-anch Haas RME  6.0 months  NR  Forst 2015  RCT**; Uni; CAN  ME need  Gp1: 20 (5/15); 14.1 Gp2: 21 (8/13); 14.2 Gp3: 21 (6/15); 12.9  Gp1: Tooth-anch RME (0.5 mm/day; 6 months Retw; 6 months Retw/o) Gp2: Bone-anch RME (0.25 mm/day; 6 months Retw; 6 months Retw/o) Gp3: No Tx  12.0 months ***  NR  Guo 2016  rNRS; Uni; CHN  —  174 (68/106); 14.1  FFA (SW); ‘light forces’  20.6 months  None  Harris 2015  rNRS; Uni; TUR  ME need  Gp1: 20 (8/12); 13.8 Gp2: 20 (7/13);13.8  Gp1: Tooth-anch RME (0.4 mm/day; 3 months Retw) Gp2: Bone-anch RME (0.4 mm/day; 3 months Retw)  3.0 months ***  NR  Johnson 2010  rNRS; pract; USA  FFA (1- or 2-phase); Ex/ Non-Ex  Gp1: 24 (11/13); 12.7 Gp2: 33 (10/23); 19.1  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 19.5 months Gp2: 13.9 months  NR  Kau 2011  pNRS; Uni; USA  Cl.I; PMs Ex  14 (NR); 20.3  FFA (SW) + vibration  6.0 months  NR  Kaylar 2016  RCT; Uni; TUR  ME need  Gp1: 10 (6/4); 19.3 Gp2: 10 (3/7); 19.2  Gp1: Tooth-anch SARME (0.5 mm/day; 6 months Retw) Gp2: Bone-anch SARME (0.5 mm/day; 6 months Retw)  6.0 months ***  None  Leite 2012  RCT; Uni; BRA  Cl. I; moderate crowding; non-Ex  Gp1: 11 (6/5); 20.6* Gp2: 8 (2/6); 20.6*  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  6.0 months  NR  Li 2013  rNRS; Uni; CHN  Cl. I–II; molar intrusion  12 (4/8); 24.3  TADs; E-chain (150 g)  6.0 months  NR  Liu 2016  RCT; Hosp; CHN  Ex/Non-Ex  Gp1: 25 (11/14); 15.3 Gp2: 25 (9/16); 15.2  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  NR  NR  Lombardo 2013  rNRS; pract; ITA  Cl. II/1; Ex (4 PMs)/non-Ex  22 (10/12); 11.4  Tweed-Merrifield technique  20.5 months  NR  Lynch 2011  rNRS; pract; USA  Cl. I-II; Ex/Non-Ex  122 (53/69); 13.2  22-slot; loops/E-chains  21.8 months  NR  Ma 2013  pNRS; Uni; CHN  ME need  32 (0/32); 14.3  Magnetic RME (increment/month)  9.3 months  NR  Nakada 2016  rNRS; Uni; JPN  4 PM Ex  30 (8/22); 21.9  FFA  NR  NR  Nimeri 2014  pNRS; Uni; USA  Cl. I; crowding; PM Ex  20 (5/15); NR  FFA + LLLT  NR  Comm  Patel 2012  rNRS; pract; USA  FFA (1- phase); Ex/ non-Ex  Gp1: 32 (16/16); 12.6 Gp2: 34 (11/23); 18.9  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 18.1 months Gp2: 14.2 months  NR  Schwarz 2015  rNRS; Uni; BRA  Cl. II; OJ > 5 mm  23 (11/12); 15.8  Herbst appliance  8.5 months  NR  Oliveira 2011; Oliveira 2016  pNRS; Uni; BRA  Cl. I–II/1; protrusion; Mx PM1 Ex  11 (5/6); adult  FFA; omega loops  NR  NR  Wang 2013  pNRS; Uni; CHN  Skeletal Cl. III; pre-surgical decompensation need; Non-Ex  Gp1: 30 (14/16); 24.8 Gp2: 26 (17/9); 23.5  Gp1: FFA Gp2: FFA + corticotomy  Gp1: 13.3 months Gp2: 7.8 months  None  Wang 2015  rNRS; Uni; CHN  Skeletal Cl. III; pre- surgical decompensation need; non-Ex  30 (13/17); adult  FFA; MBT  8.7 months  NR  Wen 2016  rNRS; Uni; CHN  PM Ex and space closure; TAD  12 (3/9); adult  FFA  30.0 months  NR  Yang 2016  rNRS; Uni; CHN  4 PM Ex; ant. Retr.  7 (NR); 22.3 years  FFA  NR  NR  Study  Design  Characteristics of patients  No patients (M/F); mAge (years)  Intervention  TxDur  COI  Abdel-Kader 2016  rNRS; Uni; EGY  PM1 Ex and C retraction  16 (6/10); 16.4  FFA; E-chain  6.3 months  NR  Ahn 2013  rNRS; Uni; KOR  Cl. I; d-a protrusion; 4PM Ex; en masse Retr;  37 (0/37); 26.6  22-slot; Roth; TPA/HG; E-chain (200 g)  21.7 months  NR  Ajmera 2014  pNRS; Uni; IND  Max-ant. proclination; PM1 Ex; en masse Retr/Intr.  48 (21/27); 13.6  22-slot; MBT; TADs (200 g)  NR  NR  Alexander 2014; Akyalcin 2015  rNRS; Uni; USA  ME need  24 (10/14); 12.8  Tooth-anch RME (0.25 mm/day; 5 months Retw)  4.8 months  NR  Aras and Tuncer 2016  RCT; Uni; TUR  Cl.I or II; OB ≥5 mm; max. ant. Intr;  Gp1: 16 (6/10); 19.3 Gp2: 16 (6/10); 19.0  Gp1: 2 MIs between I2-C; SS segmental wire; E-chain Gp2: 2 MIs between PM2-M1; Burstone TMA IntrArch (40 g/side)  4.0 months  NR  Baysal 2012  rNRS; Uni; TUR  ME need  25 (11/14); 12.7  Tooth-anch RME (0.5 mm/day)  NR  NR  Bolerjack 2005  rNRS; pract; USA  Non-Ex/Ex (4 PMs); sliding mechanics  95 (41/54); 12.7  22-slot; Roth  26.4 months  NR  Castro 2013; Castro 2015  rNRS; pract; BRA  Cl. I Non-Ex  30 (11/19); 13.0  22-slot; Roth  22.0 months  NR  Dindaroglu 2016  RCT; Uni; TUR  ME need  Gp1/2 33(17/16); 12.8 Gp1: 16 (NR); 12.9 Gp2: 17 (NR); 12.8  Gp1/2 (0.5 mm/day; 6 months Retw) Gp1: Tooth-anch Hyrax RME Gp2: Tooth-anch Haas RME  6.0 months  NR  Forst 2015  RCT**; Uni; CAN  ME need  Gp1: 20 (5/15); 14.1 Gp2: 21 (8/13); 14.2 Gp3: 21 (6/15); 12.9  Gp1: Tooth-anch RME (0.5 mm/day; 6 months Retw; 6 months Retw/o) Gp2: Bone-anch RME (0.25 mm/day; 6 months Retw; 6 months Retw/o) Gp3: No Tx  12.0 months ***  NR  Guo 2016  rNRS; Uni; CHN  —  174 (68/106); 14.1  FFA (SW); ‘light forces’  20.6 months  None  Harris 2015  rNRS; Uni; TUR  ME need  Gp1: 20 (8/12); 13.8 Gp2: 20 (7/13);13.8  Gp1: Tooth-anch RME (0.4 mm/day; 3 months Retw) Gp2: Bone-anch RME (0.4 mm/day; 3 months Retw)  3.0 months ***  NR  Johnson 2010  rNRS; pract; USA  FFA (1- or 2-phase); Ex/ Non-Ex  Gp1: 24 (11/13); 12.7 Gp2: 33 (10/23); 19.1  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 19.5 months Gp2: 13.9 months  NR  Kau 2011  pNRS; Uni; USA  Cl.I; PMs Ex  14 (NR); 20.3  FFA (SW) + vibration  6.0 months  NR  Kaylar 2016  RCT; Uni; TUR  ME need  Gp1: 10 (6/4); 19.3 Gp2: 10 (3/7); 19.2  Gp1: Tooth-anch SARME (0.5 mm/day; 6 months Retw) Gp2: Bone-anch SARME (0.5 mm/day; 6 months Retw)  6.0 months ***  None  Leite 2012  RCT; Uni; BRA  Cl. I; moderate crowding; non-Ex  Gp1: 11 (6/5); 20.6* Gp2: 8 (2/6); 20.6*  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  6.0 months  NR  Li 2013  rNRS; Uni; CHN  Cl. I–II; molar intrusion  12 (4/8); 24.3  TADs; E-chain (150 g)  6.0 months  NR  Liu 2016  RCT; Hosp; CHN  Ex/Non-Ex  Gp1: 25 (11/14); 15.3 Gp2: 25 (9/16); 15.2  Gp1: 22-slot; SLB Gp2: 22-slot; CLB  NR  NR  Lombardo 2013  rNRS; pract; ITA  Cl. II/1; Ex (4 PMs)/non-Ex  22 (10/12); 11.4  Tweed-Merrifield technique  20.5 months  NR  Lynch 2011  rNRS; pract; USA  Cl. I-II; Ex/Non-Ex  122 (53/69); 13.2  22-slot; loops/E-chains  21.8 months  NR  Ma 2013  pNRS; Uni; CHN  ME need  32 (0/32); 14.3  Magnetic RME (increment/month)  9.3 months  NR  Nakada 2016  rNRS; Uni; JPN  4 PM Ex  30 (8/22); 21.9  FFA  NR  NR  Nimeri 2014  pNRS; Uni; USA  Cl. I; crowding; PM Ex  20 (5/15); NR  FFA + LLLT  NR  Comm  Patel 2012  rNRS; pract; USA  FFA (1- phase); Ex/ non-Ex  Gp1: 32 (16/16); 12.6 Gp2: 34 (11/23); 18.9  Gp1: 18-slot; MBT Gp2: 18-slot; MBT + Suresmile  Gp1: 18.1 months Gp2: 14.2 months  NR  Schwarz 2015  rNRS; Uni; BRA  Cl. II; OJ > 5 mm  23 (11/12); 15.8  Herbst appliance  8.5 months  NR  Oliveira 2011; Oliveira 2016  pNRS; Uni; BRA  Cl. I–II/1; protrusion; Mx PM1 Ex  11 (5/6); adult  FFA; omega loops  NR  NR  Wang 2013  pNRS; Uni; CHN  Skeletal Cl. III; pre-surgical decompensation need; Non-Ex  Gp1: 30 (14/16); 24.8 Gp2: 26 (17/9); 23.5  Gp1: FFA Gp2: FFA + corticotomy  Gp1: 13.3 months Gp2: 7.8 months  None  Wang 2015  rNRS; Uni; CHN  Skeletal Cl. III; pre- surgical decompensation need; non-Ex  30 (13/17); adult  FFA; MBT  8.7 months  NR  Wen 2016  rNRS; Uni; CHN  PM Ex and space closure; TAD  12 (3/9); adult  FFA  30.0 months  NR  Yang 2016  rNRS; Uni; CHN  4 PM Ex; ant. Retr.  7 (NR); 22.3 years  FFA  NR  NR  Ant. anterior; anch, anchorage; C, canine; CLB, conventional brackets; Comm, commercial; COI, conflict of interest; Cl I, Class I; Cl II, Class II; Cl II/1, Class II division 1; Cl III, Class III; d-a protrusion, dentoalveolar protrusion; E-chain, elastic chain; Ex, extraction; FFA, fixed appliances; Gp, group; HG, headgear; Intr, intrusion; non-ex, non extraction; I2, lateral incisor; LLLT, low level light therapy; MIs, mini-screw implants; M1, first molar; Max, maxilla; ME, maxillary expansion; M/F, male/female; NR, not reported; OB, overbite; OJ, overjet; pract, private practice; PM, premolar; PM1, first premolar; PM2, second premolar; pNRS, prospective non-randomized study; RCT, randomized clinical trial; rNRS, retrospective non-randomized study; RME, rapid maxillary expansion; Retw, retention with expansion appliance; Retw/o, retention without expansion appliance; Retr, retraction; SARME, surgically assisted rapid maxillary expansion; SLB, self-ligating brackets; SW, straight wire; SS, stainless steel; Tx, treatment; TAD, temporary anchorage device; TPA, transpalatal arch; TxDur, treatment duration; Uni, university. *For both groups. **Randomization was performed for another study, not on the basis of root resorption. ***Period including a retention period about 12 months. ****Right and left teeth of the same category were averaged for analyses. View Large Risk of bias within studies The risk of bias of included non-randomized and randomized studies can be seen in Supplementary Table 6a and b, respectively. According to the Cochrane Risk of Bias Tool, serious risk of bias was found in four out of six randomized trials for at least one domain with the remaining two having unclear risk of bias (Supplementary Table 6b). The remaining non-randomized studies were evaluated with the Downs and Black checklist and all of them were in serious risk of bias for at least one of the 27 domains of the tool (Supplementary Table 6a). Included studies gathered between 11 and 23 points from the maximum of 31 points to an average of 17 points. The most problematic domain categories for non-randomized studies were confounding, followed by external validity, and bias. Data synthesis—main analysis The main analysis of the present systematic review was conducted after pooling multiple study arms from each study and performing within-study comparisons solely from identified randomized trials that possess higher internal validity. A total of five different meta-analyses could be performed using the improved Paule–Mandel random-effects estimator as seen in Supplementary Table 7. No significant differences in OIRR could be found between self-ligating and conventional brackets in a meta-analysis of two studies (P > 0.5). A single randomized trial indicated that intrusion of upper anteriors with temporary anterior devices (TAD) placed between lateral incisor and canine resulted in greater OIRR than when the TADs were placed between second premolar and first molar, both in terms of root length (MD = 0.3 mm; 95% CI = 0.2–0.5 mm) or root volume (MD = 6.3 mm3; 95% CI = 2.4–10.3 mm3). Finally, a randomized trial found no significant difference in OIRR between a conventional or a TAD-anchored Hybrid-Hyrax, while another randomized trial reported that Haas type RME resulted in less volumetric OIRR than Hyrax type RME (MD = −10.9 mm3; 95% CI = −18.6 to −3.2 mm3). No additional analyses could be conducted due to the limited number of included studies. However, analysis of the meta-evidence with the GRADE approach (Table 2) indicated that the quality of evidence was low for the comparisons of self-ligating versus conventional appliances and for the comparison of Hyrax versus Hybrid-Hyrax RME, due to bias and imprecision. The quality of evidence supporting that anterior placed TADs lead to greater OIRR than posterior placed TADs during anterior intrusion was judged as moderate, due to bias. Finally, high quality evidence supported only the use of Haas-type RME over Hyrax-type RME in order to minimize OIRR. It should be however noted that as the majority of conclusions was drawn from single trials, further trials might change the current recommendations. Table 2. Summary of findings table according to the GRADE approach. Outcome trials (patients)  Anticipated absolute effects  Quality of the evidence (GRADE)**  What happens  Reference*  Experimental  Difference for experimental (95% CI)    Conventional brackets  Self-ligating brackets        OIRR (linear) of upper anteriors after comprehensive Tx two trials (276 teeth)  0.8 mm OIRR on average  —  <0.1 mm more OIRR (0.2 mm less to 0.3 mm more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Posterior TADS  Anterior TADs        OIRR (linear) of upper anteriors after 4 months intrusion one trial (128 teeth)  0.8 mm OIRR on average  —  0.3 mm more OIRR (0.2–0.5 mm more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs  OIRR (volumetric) of upper anteriors after 4 months intrusion one trial (128 teeth)  12.9 mm3 OIRR on average  —  6.3 mm3 more OIRR (2.4–10.3 mm3 more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs    Tooth-borne RME (Hyrax)  Hybrid-borne RME (Hyrax)        OIRR (linear) of upper posteriors after RME and retention one trial (100 teeth)  0.5 mm OIRR on average  —  0.1 mm more OIRR (0.1 mm less to 0.3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR  OIRR (volumetric) of upper posteriors after RME and retention one trial (37 teeth)  49.3 mm3 OIRR on average    17.8 mm3 less OIRR (47.5 mm3 less to 11.9 mm3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Tooth-borne RME (Hyrax)  Tooth-borne RME (Haas)        OIRR (volumetric) of upper posteriors after RME and retention one trial (396 teeth)  47.8 mm3 OIRR on average  —  10.9 mm3 less OIRR (3.2–18.6 mm3 less)  ⊕⊕⊕⊕ high  Probably less OIRR with Haas  Outcome trials (patients)  Anticipated absolute effects  Quality of the evidence (GRADE)**  What happens  Reference*  Experimental  Difference for experimental (95% CI)    Conventional brackets  Self-ligating brackets        OIRR (linear) of upper anteriors after comprehensive Tx two trials (276 teeth)  0.8 mm OIRR on average  —  <0.1 mm more OIRR (0.2 mm less to 0.3 mm more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Posterior TADS  Anterior TADs        OIRR (linear) of upper anteriors after 4 months intrusion one trial (128 teeth)  0.8 mm OIRR on average  —  0.3 mm more OIRR (0.2–0.5 mm more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs  OIRR (volumetric) of upper anteriors after 4 months intrusion one trial (128 teeth)  12.9 mm3 OIRR on average  —  6.3 mm3 more OIRR (2.4–10.3 mm3 more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs    Tooth-borne RME (Hyrax)  Hybrid-borne RME (Hyrax)        OIRR (linear) of upper posteriors after RME and retention one trial (100 teeth)  0.5 mm OIRR on average  —  0.1 mm more OIRR (0.1 mm less to 0.3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR  OIRR (volumetric) of upper posteriors after RME and retention one trial (37 teeth)  49.3 mm3 OIRR on average    17.8 mm3 less OIRR (47.5 mm3 less to 11.9 mm3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Tooth-borne RME (Hyrax)  Tooth-borne RME (Haas)        OIRR (volumetric) of upper posteriors after RME and retention one trial (396 teeth)  47.8 mm3 OIRR on average  —  10.9 mm3 less OIRR (3.2–18.6 mm3 less)  ⊕⊕⊕⊕ high  Probably less OIRR with Haas  Orthodontically induced root resorption assessed with cone-beam computed tomography. Patient or population: adolescent/adult patients receiving comprehensive fixed appliance orthodontic treatment, maxillary anterior intrusion, or rapid maxillary expansion. Settings: university clinics and hospital (Brazil, Canada, China, Turkey). CI, confidence interval; GRADE, Grading of Recommendations Assessment, Development and Evaluation; OIRR, orthodontically induced root resorption; RME, rapid maxillary expansion; TAD, temporary anchorage device. *Reponse is based on random-effects meta-analytical pooling of the corresponding reference groups among included studies. **Starts from ‘high’, due to the inclusion of randomized studies. ***Downgraded further by one point due to serious limitations (high risk of bias). ****Downgraded by one due to imprecision originating from the inclusion of a small sample size. View Large Table 2. Summary of findings table according to the GRADE approach. Outcome trials (patients)  Anticipated absolute effects  Quality of the evidence (GRADE)**  What happens  Reference*  Experimental  Difference for experimental (95% CI)    Conventional brackets  Self-ligating brackets        OIRR (linear) of upper anteriors after comprehensive Tx two trials (276 teeth)  0.8 mm OIRR on average  —  <0.1 mm more OIRR (0.2 mm less to 0.3 mm more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Posterior TADS  Anterior TADs        OIRR (linear) of upper anteriors after 4 months intrusion one trial (128 teeth)  0.8 mm OIRR on average  —  0.3 mm more OIRR (0.2–0.5 mm more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs  OIRR (volumetric) of upper anteriors after 4 months intrusion one trial (128 teeth)  12.9 mm3 OIRR on average  —  6.3 mm3 more OIRR (2.4–10.3 mm3 more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs    Tooth-borne RME (Hyrax)  Hybrid-borne RME (Hyrax)        OIRR (linear) of upper posteriors after RME and retention one trial (100 teeth)  0.5 mm OIRR on average  —  0.1 mm more OIRR (0.1 mm less to 0.3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR  OIRR (volumetric) of upper posteriors after RME and retention one trial (37 teeth)  49.3 mm3 OIRR on average    17.8 mm3 less OIRR (47.5 mm3 less to 11.9 mm3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Tooth-borne RME (Hyrax)  Tooth-borne RME (Haas)        OIRR (volumetric) of upper posteriors after RME and retention one trial (396 teeth)  47.8 mm3 OIRR on average  —  10.9 mm3 less OIRR (3.2–18.6 mm3 less)  ⊕⊕⊕⊕ high  Probably less OIRR with Haas  Outcome trials (patients)  Anticipated absolute effects  Quality of the evidence (GRADE)**  What happens  Reference*  Experimental  Difference for experimental (95% CI)    Conventional brackets  Self-ligating brackets        OIRR (linear) of upper anteriors after comprehensive Tx two trials (276 teeth)  0.8 mm OIRR on average  —  <0.1 mm more OIRR (0.2 mm less to 0.3 mm more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Posterior TADS  Anterior TADs        OIRR (linear) of upper anteriors after 4 months intrusion one trial (128 teeth)  0.8 mm OIRR on average  —  0.3 mm more OIRR (0.2–0.5 mm more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs  OIRR (volumetric) of upper anteriors after 4 months intrusion one trial (128 teeth)  12.9 mm3 OIRR on average  —  6.3 mm3 more OIRR (2.4–10.3 mm3 more)  ⊕⊕⊕⊝ moderate*** due to bias  Probably more OIRR with anterior TADs    Tooth-borne RME (Hyrax)  Hybrid-borne RME (Hyrax)        OIRR (linear) of upper posteriors after RME and retention one trial (100 teeth)  0.5 mm OIRR on average  —  0.1 mm more OIRR (0.1 mm less to 0.3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR  OIRR (volumetric) of upper posteriors after RME and retention one trial (37 teeth)  49.3 mm3 OIRR on average    17.8 mm3 less OIRR (47.5 mm3 less to 11.9 mm3 more)  ⊕⊕⊝⊝ low***,**** due to bias, imprecision  Little to no difference in OIRR    Tooth-borne RME (Hyrax)  Tooth-borne RME (Haas)        OIRR (volumetric) of upper posteriors after RME and retention one trial (396 teeth)  47.8 mm3 OIRR on average  —  10.9 mm3 less OIRR (3.2–18.6 mm3 less)  ⊕⊕⊕⊕ high  Probably less OIRR with Haas  Orthodontically induced root resorption assessed with cone-beam computed tomography. Patient or population: adolescent/adult patients receiving comprehensive fixed appliance orthodontic treatment, maxillary anterior intrusion, or rapid maxillary expansion. Settings: university clinics and hospital (Brazil, Canada, China, Turkey). CI, confidence interval; GRADE, Grading of Recommendations Assessment, Development and Evaluation; OIRR, orthodontically induced root resorption; RME, rapid maxillary expansion; TAD, temporary anchorage device. *Reponse is based on random-effects meta-analytical pooling of the corresponding reference groups among included studies. **Starts from ‘high’, due to the inclusion of randomized studies. ***Downgraded further by one point due to serious limitations (high risk of bias). ****Downgraded by one due to imprecision originating from the inclusion of a small sample size. View Large Data synthesis—explorative analysis Furthermore, an exploratory analysis was conducted by using separate data from all available study arms of identified studies and employing a robust-variance DerSimonian–Laird random-effects model, analyzing studies on fixed appliance treatment and studies on RME separately (Table 3). As far as treatment with fixed appliances is concerned, the average linear OIRR among all 17 available studies (including 161 study arms) was found to be 0.8 mm (95% CI = 0.5–1.0 mm; Table 3; Figure 2). Table 3. Explorative analysis of linear OIRR including multiple measurements from all included studies.† Factor  Group  Fixed-appliance Tx  Rapid maxillary expansion  Arms  Studies  Mean  95% CI  PSG  Arms  Studies  Mean  95% CI  PSG  Any studies    Overall  161  17  0.79  0.54,1.03    20  3  0.36  −0.97,1.69    Tooth type  1s  32  12  0.82  0.38,1.26  0.04  —  —      0.51    2s  30  9  0.72  0.36,1.07    —  —          3s  13  6  0.37  −0.28,1.02    —  —          4s  22  4  0.29  −0.59,1.17    7  3  0.35  −1.26,1.97      5s  14  3  −0.08  −0.46,0.30    1  1  NE        6s  41  5  0.26  −0.30,0.81    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  79  16  0.85  0.62,1.08  0.08  —  —      ne    Posterior  82  5  0.24  −0.29,0.77    Same          Jaw  Maxilla  100  16  0.75  0.49,1.00  0.77  Same        ne    Mandible  61  7  0.57  0.01,1.13    —  —        Jaw*group  Posterior mandible  32  4  0.28  −0.42,0.98  0.06  —  —          Posterior maxilla  50  5  0.22  −0.35,0.78    Same            Anterior mandible  29  6  0.60  −0.14,1.34    —  —          Anterior maxilla  50  15  0.82  0.58,1.07    —  —        Only studies with complete Tx    Overall  81  9  0.86  0.42,1.30    20  3  0.36  −0.97,1.69    Tooth type  1s  14  6  1.00  0.44,1.57  0.10  —  —      0.51    2s  12  4  0.83  0.18,1.49    —  —          3s  4  3  0.67  −0.74,2.08    —  —          4s  12  2  0.40  −4.68,5.49    7  3  0.35  −1.26,1.97      5s  3  1  NE      1  1  NE        6s  27  3  0.43  −0.20,1.06    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  33  8  1.05  0.59,1.51  0.12  —  —      —    Posterior  48  3  0.33  −0.82,1.47    Same          Jaw  Maxilla  52  9  0.88  0.43,1.33  0.10  Same        —    Mandible  29  3  0.42  −0.33,1.17    —  —        Jaw*group  Posterior mandible  18  2  0.32  −4.33,4.97  0.14  —  —      —    Posterior maxilla  30  3  0.33  −0.89,1.55    Same            Anterior mandible  11  2  0.43  −0.18,1.03    —  —          Anterior maxilla  22  8  1.08  0.65,1.52    —  —        Factor  Group  Fixed-appliance Tx  Rapid maxillary expansion  Arms  Studies  Mean  95% CI  PSG  Arms  Studies  Mean  95% CI  PSG  Any studies    Overall  161  17  0.79  0.54,1.03    20  3  0.36  −0.97,1.69    Tooth type  1s  32  12  0.82  0.38,1.26  0.04  —  —      0.51    2s  30  9  0.72  0.36,1.07    —  —          3s  13  6  0.37  −0.28,1.02    —  —          4s  22  4  0.29  −0.59,1.17    7  3  0.35  −1.26,1.97      5s  14  3  −0.08  −0.46,0.30    1  1  NE        6s  41  5  0.26  −0.30,0.81    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  79  16  0.85  0.62,1.08  0.08  —  —      ne    Posterior  82  5  0.24  −0.29,0.77    Same          Jaw  Maxilla  100  16  0.75  0.49,1.00  0.77  Same        ne    Mandible  61  7  0.57  0.01,1.13    —  —        Jaw*group  Posterior mandible  32  4  0.28  −0.42,0.98  0.06  —  —          Posterior maxilla  50  5  0.22  −0.35,0.78    Same            Anterior mandible  29  6  0.60  −0.14,1.34    —  —          Anterior maxilla  50  15  0.82  0.58,1.07    —  —        Only studies with complete Tx    Overall  81  9  0.86  0.42,1.30    20  3  0.36  −0.97,1.69    Tooth type  1s  14  6  1.00  0.44,1.57  0.10  —  —      0.51    2s  12  4  0.83  0.18,1.49    —  —          3s  4  3  0.67  −0.74,2.08    —  —          4s  12  2  0.40  −4.68,5.49    7  3  0.35  −1.26,1.97      5s  3  1  NE      1  1  NE        6s  27  3  0.43  −0.20,1.06    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  33  8  1.05  0.59,1.51  0.12  —  —      —    Posterior  48  3  0.33  −0.82,1.47    Same          Jaw  Maxilla  52  9  0.88  0.43,1.33  0.10  Same        —    Mandible  29  3  0.42  −0.33,1.17    —  —        Jaw*group  Posterior mandible  18  2  0.32  −4.33,4.97  0.14  —  —      —    Posterior maxilla  30  3  0.33  −0.89,1.55    Same            Anterior mandible  11  2  0.43  −0.18,1.03    —  —          Anterior maxilla  22  8  1.08  0.65,1.52    —  —        1s, central incisors; 2s, lateral incisors; 3s, canines; 4s; first premolars; 5s, second premolars; 6s, first molars; 7s, second molars; CI, confidence interval; NE, not estimated; PSG, between-subgroups P value; Tx, treatment. †Comparisons of SARME and Herbst were informed only from one study each and are omitted. View Large Table 3. Explorative analysis of linear OIRR including multiple measurements from all included studies.† Factor  Group  Fixed-appliance Tx  Rapid maxillary expansion  Arms  Studies  Mean  95% CI  PSG  Arms  Studies  Mean  95% CI  PSG  Any studies    Overall  161  17  0.79  0.54,1.03    20  3  0.36  −0.97,1.69    Tooth type  1s  32  12  0.82  0.38,1.26  0.04  —  —      0.51    2s  30  9  0.72  0.36,1.07    —  —          3s  13  6  0.37  −0.28,1.02    —  —          4s  22  4  0.29  −0.59,1.17    7  3  0.35  −1.26,1.97      5s  14  3  −0.08  −0.46,0.30    1  1  NE        6s  41  5  0.26  −0.30,0.81    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  79  16  0.85  0.62,1.08  0.08  —  —      ne    Posterior  82  5  0.24  −0.29,0.77    Same          Jaw  Maxilla  100  16  0.75  0.49,1.00  0.77  Same        ne    Mandible  61  7  0.57  0.01,1.13    —  —        Jaw*group  Posterior mandible  32  4  0.28  −0.42,0.98  0.06  —  —          Posterior maxilla  50  5  0.22  −0.35,0.78    Same            Anterior mandible  29  6  0.60  −0.14,1.34    —  —          Anterior maxilla  50  15  0.82  0.58,1.07    —  —        Only studies with complete Tx    Overall  81  9  0.86  0.42,1.30    20  3  0.36  −0.97,1.69    Tooth type  1s  14  6  1.00  0.44,1.57  0.10  —  —      0.51    2s  12  4  0.83  0.18,1.49    —  —          3s  4  3  0.67  −0.74,2.08    —  —          4s  12  2  0.40  −4.68,5.49    7  3  0.35  −1.26,1.97      5s  3  1  NE      1  1  NE        6s  27  3  0.43  −0.20,1.06    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  33  8  1.05  0.59,1.51  0.12  —  —      —    Posterior  48  3  0.33  −0.82,1.47    Same          Jaw  Maxilla  52  9  0.88  0.43,1.33  0.10  Same        —    Mandible  29  3  0.42  −0.33,1.17    —  —        Jaw*group  Posterior mandible  18  2  0.32  −4.33,4.97  0.14  —  —      —    Posterior maxilla  30  3  0.33  −0.89,1.55    Same            Anterior mandible  11  2  0.43  −0.18,1.03    —  —          Anterior maxilla  22  8  1.08  0.65,1.52    —  —        Factor  Group  Fixed-appliance Tx  Rapid maxillary expansion  Arms  Studies  Mean  95% CI  PSG  Arms  Studies  Mean  95% CI  PSG  Any studies    Overall  161  17  0.79  0.54,1.03    20  3  0.36  −0.97,1.69    Tooth type  1s  32  12  0.82  0.38,1.26  0.04  —  —      0.51    2s  30  9  0.72  0.36,1.07    —  —          3s  13  6  0.37  −0.28,1.02    —  —          4s  22  4  0.29  −0.59,1.17    7  3  0.35  −1.26,1.97      5s  14  3  −0.08  −0.46,0.30    1  1  NE        6s  41  5  0.26  −0.30,0.81    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  79  16  0.85  0.62,1.08  0.08  —  —      ne    Posterior  82  5  0.24  −0.29,0.77    Same          Jaw  Maxilla  100  16  0.75  0.49,1.00  0.77  Same        ne    Mandible  61  7  0.57  0.01,1.13    —  —        Jaw*group  Posterior mandible  32  4  0.28  −0.42,0.98  0.06  —  —          Posterior maxilla  50  5  0.22  −0.35,0.78    Same            Anterior mandible  29  6  0.60  −0.14,1.34    —  —          Anterior maxilla  50  15  0.82  0.58,1.07    —  —        Only studies with complete Tx    Overall  81  9  0.86  0.42,1.30    20  3  0.36  −0.97,1.69    Tooth type  1s  14  6  1.00  0.44,1.57  0.10  —  —      0.51    2s  12  4  0.83  0.18,1.49    —  —          3s  4  3  0.67  −0.74,2.08    —  —          4s  12  2  0.40  −4.68,5.49    7  3  0.35  −1.26,1.97      5s  3  1  NE      1  1  NE        6s  27  3  0.43  −0.20,1.06    12  3  0.24  −0.87,1.35      7s  5  1  NE      —  —        Tooth group  Anterior  33  8  1.05  0.59,1.51  0.12  —  —      —    Posterior  48  3  0.33  −0.82,1.47    Same          Jaw  Maxilla  52  9  0.88  0.43,1.33  0.10  Same        —    Mandible  29  3  0.42  −0.33,1.17    —  —        Jaw*group  Posterior mandible  18  2  0.32  −4.33,4.97  0.14  —  —      —    Posterior maxilla  30  3  0.33  −0.89,1.55    Same            Anterior mandible  11  2  0.43  −0.18,1.03    —  —          Anterior maxilla  22  8  1.08  0.65,1.52    —  —        1s, central incisors; 2s, lateral incisors; 3s, canines; 4s; first premolars; 5s, second premolars; 6s, first molars; 7s, second molars; CI, confidence interval; NE, not estimated; PSG, between-subgroups P value; Tx, treatment. †Comparisons of SARME and Herbst were informed only from one study each and are omitted. View Large Figure 2. View largeDownload slide Contour-enhanced funnel plot for the pooled orthodontically induced root resorption across all randomized and non-randomized studies included in the explorative analyses pertaining to any treatment duration (blue color) or only complete treatment (green color) (Supplementary Table 7). 1s, central incisors; 2s, lateral incisors; 3s, canines; 4s; first premolars; 5s, second premolars; 6s, first molars; 7s, second molars; CI, confidence interval; OIRR, orthodontically induced root resorption; Tx, treatment. Figure 2. View largeDownload slide Contour-enhanced funnel plot for the pooled orthodontically induced root resorption across all randomized and non-randomized studies included in the explorative analyses pertaining to any treatment duration (blue color) or only complete treatment (green color) (Supplementary Table 7). 1s, central incisors; 2s, lateral incisors; 3s, canines; 4s; first premolars; 5s, second premolars; 6s, first molars; 7s, second molars; CI, confidence interval; OIRR, orthodontically induced root resorption; Tx, treatment. Significant subgroup differences were found for various teeth, with the central incisors being affected the most (OIRR = 0.8 mm; 95% CI = 0.4–1.3 mm), followed by lateral incisors (OIRR = 0.7 mm; 95% CI = 0.4–1.1 mm), and canines (OIRR=0.4 mm; 95% CI = −0.3 to 1.0 mm). Significant differences were likewise found between anterior teeth (including incisors and canines) and posterior teeth (premolars and molars), where the OIRR of anterior teeth (OIRR = 0.9; 95% CI = 0.6–1.1) was significantly greater than that of posterior teeth (OIRR = 0.2 mm; 95% CI = −0.3 to 0.8 mm). Further, significant differences were found when localizing OIRR by both tooth group (anterior or posterior) and jaw (maxilla or mandible), where the largest amount of OIRR was found for the anterior maxilla (OIRR = 0.8 mm; 95% CI = 0.6–1.1 mm), followed by the anterior mandible (OIRR = 0.6 mm; 95% CI = −0.1 to 1.3 mm), the posterior mandible (OIRR = 0.3 mm; 95% CI = −0.4 to 1.0 mm), and the posterior maxilla (OIRR = 0.2 mm; 95% CI = −0.4 to 0.8 mm). The findings were relatively robust after limiting the analysis to the subset of studies assessing OIRR solely after the completion of orthodontic treatment, although average OIRR was slightly increased (0.9 mm compared to 0.8 mm). In addition, the comparison according to the jaw became now significant, with maxillary teeth being significantly more affected (OIRR = 0.9; 95% CI = 0.4–1.3 mm) compared to mandibular teeth (OIRR = 0.4 mm; 95% CI = −0.3 to 1.2). Building on the initial analyses of Table 3, a further exploratory analysis was conducted, where only the maxillary anterior teeth (that are most severely affected) were included and pooled together from each study, in order to have an adequate sample, on which to perform subgroup analyses, meta-regressions, sensitivity analyses, and assess reporting biases (Table 4). As such, studies assessing OIRR after treatment completion reported significantly greater amounts of OIRR compared to studies assessing initial phases of treatment (1.1 and 0.5 mm, respectively). Likewise, studies on extraction treatment reported significantly greater amounts of OIRR compared to studies on non-extraction treatment (0.8 and 0.5 mm, respectively). This might be explained by potentially longer treatment duration associated with extraction treatment (Figure 3), as meta-regression showed that OIRR was significantly associated with treatment duration (with an average increase in OIRR by 0.36 mm for every additional year). Table 4. Explorative analysis of pooled linear OIRR for maxillary anterior teeth/segments from all included studies. Analysis    Studies  Teeth  OIRR  95% CI  PSG  Subgroup analysis  Complete treatment  8  3078  1.11  0.80,1.42  0.09    No complete treatment  7  434  0.49  0.23,0.75      Extraction  6  2252  0.81  0.32,1.30  0.08    Non-extraction  5  678  0.51  0.25,0.78        Studies  Teeth  Coefficient  95% CI  P  Meta-regressions  Male % in sample  13  3386  −0.01  −0.03,0.02  0.57    Mean age in years  11  3260  0.01  −0.07,0.08  0.80    Duration in months  10  3106  0.03  0.00,0.06  0.04      Studies  Teeth  OIRR  95% CI  PSG  Sensitivity analysis  Prospective  6  560  0.59  0.20,0.97  0.28    Retrospective  9  2952  0.99  0.68,1.30      Voxel size ≤0.2 mm  3  2106  1.15  0.45,1.85  0.05    Voxel size >0.2 mm  10  1134  0.64  0.37,0.91      Root-level  7  2756  1.05  0.66,1.44  0.15    Tooth-level  8  756  0.62  0.34,0.91        Studies  Teeth  Coefficient  95% CI  P  Reporting biases  Egger’s test  15  3512  −2.65  −7.59,2.29  0.27  Analysis    Studies  Teeth  OIRR  95% CI  PSG  Subgroup analysis  Complete treatment  8  3078  1.11  0.80,1.42  0.09    No complete treatment  7  434  0.49  0.23,0.75      Extraction  6  2252  0.81  0.32,1.30  0.08    Non-extraction  5  678  0.51  0.25,0.78        Studies  Teeth  Coefficient  95% CI  P  Meta-regressions  Male % in sample  13  3386  −0.01  −0.03,0.02  0.57    Mean age in years  11  3260  0.01  −0.07,0.08  0.80    Duration in months  10  3106  0.03  0.00,0.06  0.04      Studies  Teeth  OIRR  95% CI  PSG  Sensitivity analysis  Prospective  6  560  0.59  0.20,0.97  0.28    Retrospective  9  2952  0.99  0.68,1.30      Voxel size ≤0.2 mm  3  2106  1.15  0.45,1.85  0.05    Voxel size >0.2 mm  10  1134  0.64  0.37,0.91      Root-level  7  2756  1.05  0.66,1.44  0.15    Tooth-level  8  756  0.62  0.34,0.91        Studies  Teeth  Coefficient  95% CI  P  Reporting biases  Egger’s test  15  3512  −2.65  −7.59,2.29  0.27  CI, confidence interval; OIRR, orthodontically induced root resorption; PSG, between-subgroups P value. View Large Table 4. Explorative analysis of pooled linear OIRR for maxillary anterior teeth/segments from all included studies. Analysis    Studies  Teeth  OIRR  95% CI  PSG  Subgroup analysis  Complete treatment  8  3078  1.11  0.80,1.42  0.09    No complete treatment  7  434  0.49  0.23,0.75      Extraction  6  2252  0.81  0.32,1.30  0.08    Non-extraction  5  678  0.51  0.25,0.78        Studies  Teeth  Coefficient  95% CI  P  Meta-regressions  Male % in sample  13  3386  −0.01  −0.03,0.02  0.57    Mean age in years  11  3260  0.01  −0.07,0.08  0.80    Duration in months  10  3106  0.03  0.00,0.06  0.04      Studies  Teeth  OIRR  95% CI  PSG  Sensitivity analysis  Prospective  6  560  0.59  0.20,0.97  0.28    Retrospective  9  2952  0.99  0.68,1.30      Voxel size ≤0.2 mm  3  2106  1.15  0.45,1.85  0.05    Voxel size >0.2 mm  10  1134  0.64  0.37,0.91      Root-level  7  2756  1.05  0.66,1.44  0.15    Tooth-level  8  756  0.62  0.34,0.91        Studies  Teeth  Coefficient  95% CI  P  Reporting biases  Egger’s test  15  3512  −2.65  −7.59,2.29  0.27  Analysis    Studies  Teeth  OIRR  95% CI  PSG  Subgroup analysis  Complete treatment  8  3078  1.11  0.80,1.42  0.09    No complete treatment  7  434  0.49  0.23,0.75      Extraction  6  2252  0.81  0.32,1.30  0.08    Non-extraction  5  678  0.51  0.25,0.78        Studies  Teeth  Coefficient  95% CI  P  Meta-regressions  Male % in sample  13  3386  −0.01  −0.03,0.02  0.57    Mean age in years  11  3260  0.01  −0.07,0.08  0.80    Duration in months  10  3106  0.03  0.00,0.06  0.04      Studies  Teeth  OIRR  95% CI  PSG  Sensitivity analysis  Prospective  6  560  0.59  0.20,0.97  0.28    Retrospective  9  2952  0.99  0.68,1.30      Voxel size ≤0.2 mm  3  2106  1.15  0.45,1.85  0.05    Voxel size >0.2 mm  10  1134  0.64  0.37,0.91      Root-level  7  2756  1.05  0.66,1.44  0.15    Tooth-level  8  756  0.62  0.34,0.91        Studies  Teeth  Coefficient  95% CI  P  Reporting biases  Egger’s test  15  3512  −2.65  −7.59,2.29  0.27  CI, confidence interval; OIRR, orthodontically induced root resorption; PSG, between-subgroups P value. View Large Figure 3. View largeDownload slide Random-effects meta-regression of the pooled orthodontically induced root resorption of the upper anterior teeth according to the mean duration of orthodontic treatment (Supplementary Table 8). OIRR, orthodontically induced root resorption. Figure 3. View largeDownload slide Random-effects meta-regression of the pooled orthodontically induced root resorption of the upper anterior teeth according to the mean duration of orthodontic treatment (Supplementary Table 8). OIRR, orthodontically induced root resorption. Additionally, sensitivity analyses indicated that retrospective studies tended to show greater OIRR than prospective studies (1.0 and 0.6 mm, respectively) and that studies measuring OIRR at the root level tended to show greater OIRR than those at the tooth level (1.1 and 0.6 mm, respectively), although both were not statistically significant (P = 0.28 and P = 0.15, respectively). In contrast, sensitivity analysis on the basis of CBCT voxel size indicated that studies using a small (≤0.2 mm) voxel size reported significantly greater OIRR than those using a larger (>0.2 mm) voxel size (1.2 and 0.6 mm, respectively). Finally, no signs of reporting bias were seen with the Egger test (P > 0.05). As far as RME treatment is concerned, the average linear OIRR among all three available studies (including 20 study arms) was found to be 0.4 mm (95% CI = −1.0 to 1.7 mm; Table 3), without any significant differences among teeth. As far as volumetric OIRR is concerned (Supplementary Table 8), the average OIRR after fixed appliance treatment was 15.4 mm3 (3 studies; 95% CI = −4.1 to 35.0 mm3), while no further analyses could be performed. The average OIRR after RME was 25.7 mm3 (4 studies; 95% CI = 6.9–44.5 mm3), with considerable differences among teeth, where first molar showed significantly greater OIRR than first premolars (40.2 and 14.8 mm3, respectively). Discussion Summary of evidence The present systematic review summarizes evidence from 30 unique randomized or non-randomized studies with three-dimensional imaging of linear or volumetric OIRR during or after orthodontic treatment, which is to our knowledge the first systematic review on the subject. Quantitative data synthesis for this systematic review was performed on two levels, including (a) initially direct comparisons of treatment-related factors within and across only randomized trials and (b) subsequent explorative analyses with indirect comparisons across all identified randomized and non-randomized studies. Robust comparisons from randomized trials should be used to inform clinical decision making, while the explorative analyses assess evidence of observational nature to characterize the pattern of OIRR. Direct meta-analysis pooling data from two randomized trials found no significant difference in OIRR between self-ligating and conventional brackets (Supplementary Table 7). Although the quality of existing evidence according to GRADE was low due to bias and imprecision, the results are in accordance with a previous meta-analysis of randomized trials using 2D imaging (35). Further, a single randomized trial indicated that intrusion of upright incisors anchored from posterior mini-implants (between second premolar and first molar) yielded more labial flaring and less OIRR than intrusion mechanics anchored anteriorly (between lateral incisor and canine) (Supplementary Table 7). This was attributed by the authors of that trial to the greater apical movement of the latter group (36), which had been proposed as a risk factor for OIRR (37). Explorative indirect meta-analyses on the average amount of OIRR across all studies associated with fixed appliance treatment gave a pooled OIRR of 0.79 mm across studies with any treatment duration and a pooled OIRR of 0.86 mm across studies reporting on complete fixed appliance treatment from bond to debond (Table 3). This small magnitude of the average OIRR is in line with the previous systematic review of Weltman et al. (1) and has probably no clinical relevance in terms of attachment loss, tooth mobility, or tooth prognosis (15). However, extreme variation is seen in the observed OIRR and statistical differences exist among various teeth, confirming results from previous studies that were performed with periapical or panoramic radiographs (3, 5, 17). As such, the greatest amount of OIRR was seen for the central incisor (0.82 mm), followed by the lateral incisor (0.72 mm), the canine (0.37 mm), the first premolar (0.29 mm), and lastly the first permanent molar (0.26 mm). Additionally, teeth of the anterior region (up to the canine) showed significantly greater amounts of OIRR than teeth of the posterior region (premolars and molars) (0.85 and 0.24 mm, respectively). Finally, statistically significant differences in OIRR were seen across the different regions within a jaw, with the anterior maxilla showing the greatest among of OIRR (0.82 mm), followed by the anterior mandible (0.60 mm), the posterior mandible (0.28 mm), and finally the posterior maxilla (0.22 mm) (Table 3). However, even the greatest amount of OIRR found from CBCT data at the anterior maxilla seems to be less than what it was found in previous studies conducted with periapical X-rays (3, 17, 38). This difference could be attributed to magnification errors and distortions due to root angulation that are very common among 2D imaging techniques (39). As expected, limiting the analysis only to studies assessing the complete treatment from appliance placement to appliance removal (Table 3), yielded slightly greater OIRR than the original analysis of studies assessing OIRR during or after treatment (0.86 and 0.79 mm, respectively), which indicates that future studies should probably measure OIRR after treatment completion to avoid underestimation of OIRR. This was confirmed by the explorative analysis after pooling all upper anterior teeth of each study in a single group (Table 4), which actually found greater difference in the OIRR of these teeth (1.11 and 0.49 mm for complete and non-complete treatment, respectively). Further, extraction treatment was associated with statistically greater OIRR of the upper anterior teeth than non-extraction treatment, which were 0.81 and 0.51 mm, respectively (Table 4). A possible explanation for this would be that extraction treatment and the subsequently needed space closure takes more time to complete than non-extraction treatment (40). Moreover, treatment duration was found in meta-regression of the present study to be significantly associated with the amount of OIRR (Table 4; Figure 3), which agrees with previous data from 2D imaging (5, 10, 41). This might explain, at least to some degree, the difference between extraction and non-extraction treatment, a finding that has also been described previously (10, 41, 42). Apart from fixed appliance treatment, some of the identified studies also employed CBCT to assess OIRR after treatment with RME expansion appliances. The average pooled linear OIRR associated with RME from explorative analyses was 0.36 mm with no significant differences across the assessed teeth (Table 3). The small magnitude of OIRR associated with RME seems to agree with the findings of Forst et al. (43), who found signs of OIRR with three-dimensional but not with 2D imaging, indicating that the latter approach may be worse suited to identify such small amounts of OIRR. In addition, direct evidence from randomized trials indicated that no significant difference in OIRR existed between conventional tooth-anchored Hyrax SARME and Hybrid tooth-and-implant-anchored SARME (Supplementary Table 7), with low quality of evidence according to GRADE due to bias and imprecision. In contrast, conventionally anchored Haas-type RME was associated with less OIRR than conventionally anchored Hyrax-type RME (Supplementary Table 7), which was supported by high quality of evidence according to GRADE. This agrees with previous data using 2D radiography (44), and might be possibly explained by the Haas expander distributing forces to both teeth and palatal vault and thereby reducing direct effects on tooth structures (45). However, these results should be interpreted with caution, as only a single randomized trial informed all direct comparisons. A single study assessing OIRR with CBCT before and after Class II treatment with the Herbst appliance was identified (46), although it was omitted from quantitative data synthesis due to the absence of other similar studies. Based on the results of this study, there was evidence of statistically significant OIRR affecting the teeth upon which the Herbst appliance was anchored (upper and lower first molars), though the amount of OIRR was small and not clinically significant. These findings are in accordance with previous studies that examined OIRR after Herbst treatment using 2D radiographs (47, 48). Finally, as far as design and measurement characteristics are concerned, sensitivity analysis using the CBCT’s voxel size showed that studies with voxel size ≤0.20 mm reported significantly greater OIRR compared to studies with greater voxel sizes (pooled OIRR: 1.15 and 0.64 mm, respectively; Table 4). This might indicate that the latter studies had too large voxel sizes to accurately identify areas of OIRR, and therefore, small voxel sizes might be preferable to accurately diagnose OIRR. This is supported by other data indicating that the method error in measuring tooth lengths with CBCT was smaller with voxel size of 0.20 mm compared to both intraoral radiography and CBCT with greater voxel sizes (49). However, it must be stated that the optimal voxel size to diagnose OIRR cannot be gleamed from this systematic review and even CBCT images with a voxel size of 0.20 mm might be unable to identify OIRR of small magnitudes. Decisions about the appropriateness of using CBCT protocols with high-resolution and radiation dose (19) to diagnose a mostly clinically irrelevant amount of OIRR need to be based on robust evaluations of the benefit to risk ratio for the patient. Prescribing CBCT solely for research purposes cannot be ethically justified and robust evidence on whether CBCT leads to diagnosis or treatment planning with improved treatment outcomes is needed. As far as displaced maxillary canines are concerned, current evidence indicates that for most cases treatment decisions are the same with either CBCT or 2D imaging (50), while the CBCT effective dose is 15–140 times higher than the 2D dose (51). Only a proportion of patients might benefit in terms of changed treatment plan, while there is currently no data to support improved treatment outcomes from this change. Furthermore, CBCT cannot be reliably recommended to diagnose tooth ankylosis, as although CBCT and histological findings coincided for some cases, false positivese were also seen (52). Therefore, currently evidence-based indications for CBCT are limited to some cases of maxillary displaced canines that might potentially benefit in terms of altered decision-making, even though evidence about the treatment outcomes is lacking. Strengths, limitations, and generalizability The strengths of the current systematic review include the a priori protocol, the comprehensive literature search including grey literature, the use of robust methods for qualitative and quantitative synthesis, and the open provision of the review’s dataset to increase transparency and reproducibility (53). In addition, no language restrictions were applied, and translation was arranged for non-English papers in order to reduce language bias. Also, no restrictions concerning publication year and status were applied, thereby maximizing data yield. Nevertheless, some limitations must also be noted, with the main one being the inclusion of retrospective studies, due to the scarcity of randomized and prospective non-randomized studies. Further, most of the included studies presented severe methodological limitations, while none of the studies used an untreated control group, since this would be difficult to justify ethically. In addition, the small number of studies that were included in several analyses and their incomplete reporting might have resulted in some analyses suffering from low statistical power. Finally, it should be noted that the present systematic review evaluated the average root resorption observed during or after treatment. In practice, the clinical relevance of OIRR is directly related to the tooth’s subsequent attachment loss and therefore only large magnitudes of root resorption are relevant to the patient. It might be useful in the future to categorize OIRR into clinically relevant categories of OIRR magnitude and use this as an outcome in clinical research. However, given the additive nature of many OIRR-related risk factors, evidence from this systematic review can be applied clinically to minimize OIRR in general. Given the broad eligibility criteria that were set and the inclusion of various clinical settings in many different countries, the results of the present systematic review are applicable to the average patient, provided they correspond to the age range and treatment phase of the included studies. Conclusions The present systematic review summarized evidence on clinical studies assessing treatment related OIRR with CBCT. Explorative analyses on the amount and pattern of OIRR after fixed appliance treatment from all identified randomized and non-randomized studies indicated that less than 1 mm of OIRR is seen on the average tooth with CBCT. Although OIRR measured by CBCT is higher than that measured by two-dimensional imaging, it has still probably little clinical relevance, which makes the added exposure to ionizing radiation through high-resolution CBCT protocols (19) questionable in terms of risk to benefit ratio. However, considerable differences in the amount of measured OIRR are seen according to tooth category, jaw, incorporation of extraction in the treatment plan, treatment duration, and CBCT settings. These data on the pattern of OIRR associated with orthodontic treatment might be useful both in everyday clinical practice and in the design of future studies. As far as direct comparisons from randomized studies are concerned, no differences in OIRR after fixed appliance treatment were seen between self-ligating and conventional brackets, although significant differences in OIRR after upper anterior intrusion were seen between anterior or posterior placement of TADs for anchorage. Further, OIRR after RME seemed to be similar between conventional Hyrax and hybrid Hyrax RME appliances, but less in conventional Hass appliances compared to conventional Hyrax appliances. However, these results should be interpreted with caution, due to the small number of identified randomized trials, their methodological limitations, and potential ethical implications about routine use of CBCT for OIRR diagnosis. Supplementary material Supplementary material is available at European Journal of Orthodontics online. Funding The study received no external funding to conduct this research through any of the authors involved. Conflicts of interest None to declare. Acknowledgements The authors would like to thank Dr. João Paulo Schwartz (Paulista State University, Araraquara, Brazil) for providing clarification on his work. References 1. Weltman, B., Vig, K.W., Fields, H.W., Shanker, S. and Kaizar, E.E. ( 2010) Root resorption associated with orthodontic tooth movement: a systematic review. American Journal of Orthodontics and Dentofacial Orthopedics , 137, 462– 76; discussion 12A. Google Scholar CrossRef Search ADS   2. Al-Qawasmi, R.A., Hartsfield, J.K. Jr, Everett, E.T., Flury, L., Liu, L., Foroud, T.M., Macri, J.V. and Roberts, W.E. ( 2003) Genetic predisposition to external apical root resorption. American Journal of Orthodontics and Dentofacial Orthopedics , 123, 242– 252. Google Scholar CrossRef Search ADS   3. Parker, R.J. and Harris, E.F. ( 1998) Directions of orthodontic tooth movements associated with external apical root resorption of the maxillary central incisor. American Journal of Orthodontics and Dentofacial Orthopedics , 114, 677– 683. Google Scholar CrossRef Search ADS   4. Brezniak, N. and Wasserstein, A. ( 1993) Root resorption after orthodontic treatment: part 2. Literature review. American Journal of Orthodontics and Dentofacial Orthopedics , 103, 138– 146. Google Scholar CrossRef Search ADS   5. Apajalahti, S. and Peltola, J.S. ( 2007) Apical root resorption after orthodontic treatment – a retrospective study. European Journal of Orthodontics , 29, 408– 412. Google Scholar CrossRef Search ADS   6. Tyrovola, J.B. and Spyropoulos, M.N. ( 2001) Effects of drugs and systemic factors on orthodontic treatment. Quintessence International (Berlin, Germany: 1985) , 32, 365– 371. 7. Oyama, K., Motoyoshi, M., Hirabayashi, M., Hosoi, K. and Shimizu, N. ( 2007) Effects of root morphology on stress distribution at the root apex. European Journal of Orthodontics , 29, 113– 117. Google Scholar CrossRef Search ADS   8. Marques, L.S., Ramos-Jorge, M.L., Rey, A.C., Armond, M.C. and Ruellas, A.C. ( 2010) Severe root resorption in orthodontic patients treated with the edgewise method: prevalence and predictive factors. American Journal of Orthodontics and Dentofacial Orthopedics , 137, 384– 388. Google Scholar CrossRef Search ADS   9. Linge, B.O. and Linge, L. ( 1983) Apical root resorption in upper anterior teeth. European Journal of Orthodontics , 5, 173– 183. Google Scholar CrossRef Search ADS   10. Sameshima, G.T. and Sinclair, P.M. ( 2001) Predicting and preventing root resorption: part II. Treatment factors. American Journal of Orthodontics and Dentofacial Orthopedics , 119, 511– 515. Google Scholar CrossRef Search ADS   11. Han, G., Huang, S., Von den Hoff, J.W., Zeng, X. and Kuijpers-Jagtman, A.M. ( 2005) Root resorption after orthodontic intrusion and extrusion: an intraindividual study. The Angle Orthodontist , 75, 912– 918. 12. Montenegro, V.C., Jones, A., Petocz, P., Gonzales, C. and Darendeliler, M.A. ( 2012) Physical properties of root cementum: part 22. Root resorption after the application of light and heavy extrusive orthodontic forces: a microcomputed tomography study. American Journal of Orthodontics and Dentofacial Orthopedics , 141, e1– e9. Google Scholar CrossRef Search ADS   13. Ballard, D.J., Jones, A.S., Petocz, P. and Darendeliler, M.A. ( 2009) Physical properties of root cementum: part 11. Continuous vs intermittent controlled orthodontic forces on root resorption. A microcomputed-tomography study. American Journal of Orthodontics and Dentofacial Orthopedics , 136, 8.e1– 8; discussion 8. 14. Lombardo, L., Bragazzi, R., Perissinotto, C., Mirabella, D. and Siciliani, G. ( 2013) Cone-beam computed tomography evaluation of periodontal and bone support loss in extraction cases. Progress in Orthodontics , 14, 29. Google Scholar CrossRef Search ADS   15. Jönsson, A., Malmgren, O. and Levander, E. ( 2007) Long-term follow-up of tooth mobility in maxillary incisors with orthodontically induced apical root resorption. European Journal of Orthodontics , 29, 482– 487. Google Scholar CrossRef Search ADS   16. Makedonas, D. and Hansen, K. ( 2008) Diagnosis, screening and treatment of root resorption in orthodontic practices in Greece and Sweden. The Angle Orthodontist , 78, 248– 253. Google Scholar CrossRef Search ADS   17. Sameshima, G.T. and Sinclair, P.M. ( 2001) Predicting and preventing root resorption: part I. diagnostic factors. American Journal of Orthodontics and Dentofacial Orthopedics , 119, 505– 510. Google Scholar CrossRef Search ADS   18. Ren, H., Chen, J., Deng, F., Zheng, L., Liu, X. and Dong, Y. ( 2013) Comparison of cone-beam computed tomography and periapical radiography for detecting simulated apical root resorption. The Angle Orthodontist , 83, 189– 195. Google Scholar CrossRef Search ADS   19. Ludlow, J.B., Timothy, R., Walker, C., Hunter, R., Benavides, E., Samuelson, D.B. and Scheske, M.J. ( 2015) Effective dose of dental CBCT-a meta analysis of published data and additional data for nine CBCT units. Dento Maxillo Facial Radiology , 44, 20140197. Google Scholar CrossRef Search ADS   20. Dudic, A., Giannopoulou, C., Leuzinger, M. and Kiliaridis, S. ( 2009) Detection of apical root resorption after orthodontic treatment by using panoramic radiography and cone-beam computed tomography of super-high resolution. American Journal of Orthodontics and Dentofacial Orthopedics , 135, 434– 437. Google Scholar CrossRef Search ADS   21. da Silveira, H.L., Silveira, H.E., Liedke, G.S., Lermen, C.A., Dos Santos, R.B. and de Figueiredo, J.A. ( 2007) Diagnostic ability of computed tomography to evaluate external root resorption in vitro. Dento Maxillo Facial Radiology , 36, 393– 396. Google Scholar CrossRef Search ADS   22. Akyalcin, S., Alexander, S.P., Silva, R.M. and English, J.D. ( 2015) Evaluation of three-dimensional root surface changes and resorption following rapid maxillary expansion: a cone beam computed tomography investigation. Orthodontics and Craniofacial Research , 18 Suppl 1, 117– 126. Google Scholar CrossRef Search ADS   23. Holberg, C., Steinhäuser, S., Geis, P. and Rudzki-Janson, I. ( 2005) Cone-beam computed tomography in orthodontics: benefits and limitations. Journal of Orofacial Orthopedics , 66, 434– 444. Google Scholar CrossRef Search ADS   24. Higgins, J.P.T. and Green, S. ( 2016) Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011] . The Cochrane Collaboration, 2011. www.cochrane handbook.org ( 21 April 2017, date last accessed). 25. Liberati, A., Altman, D.G., Tetzlaff, J., Mulrow, C., Gøtzsche, P.C., Ioannidis, J.P., Clarke, M., Devereaux, P.J., Kleijnen, J. and Moher, D. ( 2009) The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. Journal of Clinical Epidemiology , 62, e1– 34. Google Scholar CrossRef Search ADS   26. Downs, S.H. and Black, N. ( 1998) The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. Journal of Epidemiology and Community Health , 52, 377– 384. Google Scholar CrossRef Search ADS   27. Papageorgiou, S.N. ( 2014) Meta-analysis for orthodontists: part I–how to choose effect measure and statistical model. Journal of Orthodontics , 41, 317– 326. Google Scholar CrossRef Search ADS   28. Veroniki, A.A., Jackson, D., Viechtbauer, W., Bender, R., Bowden, J., Knapp, G., Kuss, O., Higgins, J.P., Langan, D. and Salanti, G. ( 2016) Methods to estimate the between-study variance and its uncertainty in meta-analysis. Research Synthesis Methods , 7, 55– 79. Google Scholar CrossRef Search ADS   29. Hedges, L.V., Tipton, E. and Johnson, M.C. ( 2010) Robust variance estimation in meta-regression with dependent effect size estimates. Research Synthesis Methods , 1, 39– 65. Google Scholar CrossRef Search ADS   30. Samandara, A., Papageorgiou, S.N., Ioannidou-Marathiotou, I., Kavvadia-Tsatala, S. and Papadopoulos, M.A. ( 2018) Evaluation of orthodontically induced external root resorption following orthodontic treatment using Cone Beam Computed Tomography (CBCT): a systematic review and meta-analysis. [Data set]. Zenodo . doi: 10.5281/zenodo.1185732 31. Egger, M., Davey Smith, G., Schneider, M. and Minder, C. ( 1997) Bias in meta-analysis detected by a simple, graphical test. BMJ (Clinical research ed.) , 315, 629– 634. Google Scholar CrossRef Search ADS   32. Guyatt, G.H., Oxman, A.D., Schünemann, H.J., Tugwell, P. and Knottnerus, A. ( 2011) GRADE guidelines: a new series of articles in the journal of clinical epidemiology. Journal of Clinical Epidemiology , 64, 380– 382. Google Scholar CrossRef Search ADS   33. Papageorgiou, S.N. ( 2014) Meta-analysis for orthodontists: part II–is all that glitters gold? Journal of Orthodontics , 41, 327– 336. Google Scholar CrossRef Search ADS   34. Lund, H., Gröndahl, K., Hansen, K. and Gröndahl, H.G. ( 2012) Apical root resorption during orthodontic treatment. A prospective study using cone beam CT. The Angle Orthodontist , 82, 480– 487. Google Scholar CrossRef Search ADS   35. Papageorgiou, S.N., Konstantinidis, I., Papadopoulou, K., Jäger, A. and Bourauel, C. ( 2014) Clinical effects of pre-adjusted edgewise orthodontic brackets: a systematic review and meta-analysis. European Journal of Orthodontics , 36, 350– 363. Google Scholar CrossRef Search ADS   36. Aras, I. and Tuncer, A.V. ( 2016) Comparison of anterior and posterior mini-implant-assisted maxillary incisor intrusion: root resorption and treatment efficiency. The Angle Orthodontist , 86, 746– 752. Google Scholar CrossRef Search ADS   37. Dermaut, L.R. and De Munck, A. ( 1986) Apical root resorption of upper incisors caused by intrusive tooth movement: a radiographic study. American Journal of Orthodontics and Dentofacial Orthopedics , 90, 321– 326. Google Scholar CrossRef Search ADS   38. McFadden, W.M., Engstrom, C., Engstrom, H. and Anholm, J.M. ( 1989) A study of the relationship between incisor intrusion and root shortening. American Journal of Orthodontics and Dentofacial Orthopedics , 96, 390– 396. Google Scholar CrossRef Search ADS   39. Tieu, L.D., Saltaji, H., Normando, D. and Flores-Mir, C. ( 2014) Radiologically determined orthodontically induced external apical root resorption in incisors after non-surgical orthodontic treatment of class II division 1 malocclusion: a systematic review. Progress in Orthodontics , 15, 48. Google Scholar CrossRef Search ADS   40. Papageorgiou, S.N., Höchli, D. and Eliades, T. ( 2017) Outcomes of comprehensive fixed appliance orthodontic treatment: a systematic review with meta-analysis and methodological overview. Korean Journal of Orthodontics , 47, 401– 413. Google Scholar CrossRef Search ADS   41. Maués, C.P., do Nascimento, R.R. and Vilella, O.d.e.V. ( 2015) Severe root resorption resulting from orthodontic treatment: prevalence and risk factors. Dental Press Journal of Orthodontics , 20, 52– 58. Google Scholar CrossRef Search ADS   42. McNab, S., Battistutta, D., Taverne, A. and Symons, A.L. ( 2000) External apical root resorption following orthodontic treatment. The Angle Orthodontist , 70, 227– 232. 43. Forst, D., Nijjar, S., Khaled, Y., Lagravere, M. and Flores-Mir, C. ( 2014) Radiographic assessment of external root resorption associated with jackscrew-based maxillary expansion therapies: a systematic review. European Journal of Orthodontics , 36, 576– 585. Google Scholar CrossRef Search ADS   44. Odenrick, L., Karlander, E.L., Pierce, A. and Kretschmar, U. ( 1991) Surface resorption following two forms of rapid maxillary expansion. European Journal of Orthodontics , 13, 264– 270. Google Scholar CrossRef Search ADS   45. Haas, A.J. ( 1965) The treatment of maxillary deficiency by opening the midpalatal suture. The Angle Orthodontist , 35, 200– 217. 46. Schwartz, J.P., Raveli, T.B., Almeida, K.C., Schwartz-Filho, H.O. and Raveli, D.B. ( 2015) Cone beam computed tomography study of apical root resorption induced by Herbst appliance. Journal of Applied Oral Science: Revista FOB , 23, 479– 485. Google Scholar CrossRef Search ADS   47. Nasiopoulos, A.T., Athanasiou, A.E., Papadopoulos, M.A., Kolokithas, G. and Ioannidou, I. ( 2006) Premolar root changes following treatment with the banded herbst appliance. Journal of Orofacial Orthopedics , 67, 261– 271. Google Scholar CrossRef Search ADS   48. Kinzinger, G.S., Savvaidis, S., Gross, U., Gülden, N., Ludwig, B. and Lisson, J. ( 2011) Effects of Class II treatment with a banded Herbst appliance on root lengths in the posterior dentition. American Journal of Orthodontics and Dentofacial Orthopedics , 139, 465– 469. Google Scholar CrossRef Search ADS   49. Lund, H., Gröndahl, K. and Gröndahl, H.G. ( 2010) Cone beam computed tomography for assessment of root length and marginal bone level during orthodontic treatment. The Angle Orthodontist , 80, 466– 473. Google Scholar CrossRef Search ADS   50. Christell, H., Birch, S., Bondemark, L., Horner, K. and Lindh, C.; SEDENTEXCT consortium. ( 2018) The impact of Cone Beam CT on financial costs and orthodontists’ treatment decisions in the management of maxillary canines with eruption disturbance. European Journal of Orthodontics , 40, 65– 73. Google Scholar CrossRef Search ADS   51. Kadesjö, N., Lynds, R., Nilsson, M. and Shi, X.Q. ( 2018) Radiation dose from X-ray examinations of impacted canines: cone beam CT vs two-dimensional imaging. Dento Maxillo Facial Radiology , 47, 20170305. Google Scholar CrossRef Search ADS   52. Ducommun, F., Bornstein, M.M., Bosshardt, D., Katsaros, C. and Dula, K. ( 2017) Diagnosis of tooth ankylosis using panoramic views, cone beam computed tomography, and histological data: a retrospective observational case series study. European Journal of Orthodontics . doi: 10.1093/ejo/cjx063. 53. Papageorgiou, S.N. and Cobourne, M.T. ( 2018) Data sharing in orthodontic research. Journal of Orthodontics , 45, 1– 3. Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press on behalf of the European Orthodontic Society. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

The European Journal of OrthodonticsOxford University Press

Published: May 15, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off