All 1624 pairs of pretreatment and posttreatment PRs acquired for orthodontic purposes between 2000 and 2021 at the Prince Philip Dental Hospital and at the Institute for Advanced Dentistry (Faculty of Dentistry, The University of Hong Kong, Hong Kong SAR) were retrospectively collected and consecutively screened. Adolescents and young adults aged 10-28 years who received fixed orthodontic treatment, with symmetrical mandibular premolar extraction (either first or second premolars), with both pretreatment and posttreatment PR, and with mandibular IO unilaterally present in an area subject to orthodontic movement (so that the non-IO side could be used as a negative control) were selected. Patients with a history of orthognathic surgery, severe facial asymmetries or jaw deformities (eg, missing condyles or hemifacial microsomia), IO further than the apical level, or treated with devices other than a multibracket vestibular appliance (eg, clear aligners or removable appliances) were excluded. Ethical approval was obtained from the Institutional Review Board of the University of Hong Kong- West Cluster Hospital Authority (UW 18-187), which waived the need for informed consent because of the retrospective nature of the study. The study was conducted in accordance with the Helsinki Declaration of 1975, as revised in 2013.

The sample size was calculated based on the difference between the IO side and the non-IO side in terms of posttreatment changes in dental angulation and root resorption. A pilot study was performed on 10 PRs of patients fulfilling the inclusion criteria, besides having an IO. The standard deviation (SD) of the posttreatment difference in dental angulation between the IO side and non-IO side in healthy patients was 6.6%, whereas it was 4.8% for root resorption. The calculation was based on a 2-tailed paired comparison by using the Wilcoxon signed rank test, with power β = 90% and significance α = 5%. For dental angulation, the required sample size was 22 for detecting a clinically meaningful difference of 5.0° (a variation <5.0° may not alter treatment decisions when assessing tooth angulation on PR). ^, For root resorption, the minimum sample size was 6 for detecting a clinically meaningful difference of 9.0% of tooth length (corresponding to a 2.0 mm shortening of a mandibular premolar, which represents the indication for pausing orthodontic treatment) (G∗Power; Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany). Given the retrospective nature of the study, the final sample size was 40.

Radiopacities on PRs were diagnosed as IO if they were not a mixed radiolucent and radiopaque area with the appearance of a fibroosseous lesion or periapical cemental dysplasia or odontoma, not surrounded by a radiolucent periphery, with no thickening of the lamina dura, not associated with identifiable tooth remnants, not associated with carious and/or restored and/or endodontically treated teeth, not associated with resorption of adjacent teeth before orthodontic treatment, found within a dentate portion of the alveolus, not a torus or exostosis or salivary calculus or tonsolith or calcified lymph node or stylohyoid ligament, and presenting no evidence of displacement of the inferior dental canal or floor of the antrum or adjacent tooth. The diagnosis of IO was performed by a specialist in orthodontics (F.S) and confirmed by a specialist in oral and maxillofacial radiology (R.T).

All measurements were taken with a computer software (ImageJ), and the mesiodistal diameter of the mandibular first molar adjacent to the IO was used as a reference (unit) for scaling the linear measurements. The occlusal line (on the IO side) was identified as the line joining the most occlusal point of each crown of the 2 teeth adjacent to the IO. The tooth whose root moved closer to the IO was selected for analysis, together with its respective contralateral, and numbered according to the dental numbering system for adult teeth of the World Dental Federation (Fédération Dentaire Internationale, FDI, ISO 3950 notation). The center of the IO (IOctr) was identified by inscribing its contour into a circle, and the diameter of the circle was used to estimate the size of the IO. The center of the dental crown (CROctr) was identified as the midpoint of the maximum mesiodistal crown width. On the IO side, the crown position with respect to the IO (CROpos) was measured as the distance between the projection of IOctr and CROctr on the occlusal line. A positive value was assigned to the CROpos if the CROctr was mesial to the IOctr, and a negative value was assigned if the CROctr was distal to the IOctr. Then, the dental movement was identified as mesialization or distalization. On both sides, the pretreatment and posttreatment tooth length was measured from the anatomic apex of the root to the most occlusal point of the crown (for molars, the root closer to the IO was measured). The mesiodistal tooth angulation was measured as the angle between the long axis of the tooth (the line passing through the anatomic apex and most occlusal point of the crown; for molars, the line passing through the furcation and CROctr) and the maxillary line (the line along the hard palate and perpendicular to the midline bisector at the level of the nasal septum and anterior nasal spine). The distance of the border of the IO from the dental roots was measured. The extraction space was measured as the distance between the contours of the crowns of the 2 teeth adjacent to the extraction site at pretreatment ( Fig 1 , A and B ). By considering an average mesiodistal diameter of a mandibular first molar of 11.5 mm in the ethnicity of the target population, the units were converted to mm (1 unit ≈ 11.5 mm).

A , Pretreatment; B, Posttreatment. Occlusal dashed line ( yellow ); maxillary bulk line ( yellow ); IO contour dashed line ( green ); IO inscription circle ( green ); IO center dot ( green ) and its projection on the occlusal line ( green ); tooth contour dashed line ( pink ); crown center dot ( pink ) and its projection on the occlusal line ( pink ); adjacent tooth contour dashed line ( white ); reference unit arrow ( blue ); tooth length arrow ( yellow ); extraction space arrow ( red ); and tooth angulation ( yellow ).

Statistical analysis

The normality of data distribution was checked using Shapiro-Wilk test. The pretreatment extraction space, posttreatment extraction space, and difference in space closure were compared between the IO side and non-IO side using Wilcoxon signed rank test. On the IO side, the crown movement with respect to the IO (CROmov, units) was measured as the absolute value of the difference in CROpos between posttreatment and pretreatment. Teeth were identified either as passing through the IO when the root moved within the lesion during treatment or not passing through in all the other instances ( Fig 2 ). The application of this classification was limited to patients with complete space closure, CROmov ≥2 mm, and excluding patients with bilateral uncontrolled tipping ≥5°. Binomial exact test was used to compare the prevalence of teeth passing through and not passing through. Root resorption was estimated as the ratio between the difference in tooth length (between posttreatmt and pretreatment) and the pretreatment tooth length (%). Then, root resorption was compared between the IO side and non-IO side using Wilcoxon signed rank test. The difference in root resorption between the IO side and non-IO side was calculated (%) and, considering negative values as representative of greater root resorption on the IO side, binomial exact test was used to compare the prevalence of root resorption between the 2 sides. Changes in dental angulation were calculated as the difference between posttreatmentand pretreatment angulation (°). Then, the change in dental angulation was compared between the IO side and non-IO side using Wilcoxon signed rank test. The difference in the change of dental angulation between the IO side and non-IO side was calculated (°) and, considering a greater change in dental angulation in the direction of tooth movement as representative of uncontrolled tipping (ie, distal tipping during distalization or mesial tipping during mesialization), binomial exact test was used to compare the prevalence of uncontrolled tipping between the 2 sides. The IO size and distance of the IO from the dental roots were compared between pretreatment and posttreatment using Mann-Whitney U test. The average measurement between the primary and the secondary assessor was used for the final analysis, which was performed by using statistical software (SPSS [IBM, Armonk, NY] and StataCorp [StataCorp, College Station, Tex]) at a significance level α = 0.05.

Movement through the IO, defined as not passing through the IO (A _PRE and A _POST ) or passing through the IO (B _PRE and B _POST ), after exclusion of patients with incomplete space closure, small crown movement, and evident uncontrolled tipping on both sides.

Records were measured by a primary assessor (F.S) and secondary assessor (K.D.T), both dentists and specialists in orthodontics, with training in PR analysis. Assessors were calibrated on the PR of 10 patients who were excluded from the study. The primary assessor repeated the measurements after a wash-out period of approximately 1 month. The intraclass correlation coefficient for single measurements (absolute agreement) was used to measure the intraassessor and interassessor agreement of continuous variables (poor, <0.5; fair, 0.5-0.7; good, 0.7-0.8; excellent, >0.8; perfect, 1.0). Cohen’s κ coefficient was used to measure the intraassessor and interassessor agreement of categorical variables (low, <0.41; moderate, 0.41-0.60; substantial, 0.61-0.80; excellent, >0.80; perfect, 1.00).

The intraclass correlation coefficient was excellent (range, 0.893-0.997) for all continuous variables, for both the intraassessor and the interassessor agreement. The agreement shown by kappa values about the assessment of whether the tooth was passing through the IO was substantial (k = 0.750), for both the intraassessor and interassessor agreement (the complete assessment is available in Supplementary Tables I and II ).

A total of 1624 patients were screened, and 183 (11%) were identified with IO. Forty patients satisfied the inclusion criteria and were selected for further assessment. Of this sample, 14 were males (35%), and 26 were females (65%), with a mean age of 16.3 ± 4.0 years (range, 11-28 years). Of the patients, 19 (48%) had extractions of the first mandibular premolars, whereas 21 (52%) had extractions of the second mandibular premolars. The average time between pretreatment and posttreatment PR was 3.4 ± 1.5 years, which corresponded to the average treatment duration. At pretreatment, the IO was located in the incisal region for 1 patient (3%), in the canine region for 7 patients (18%), in the premolar region for 16 patients (40%), and in the molar region for 16 patients (40%), showing no changes at posttreatment. At pretreatment, the average IO size was 0.8 ± 0.3 units (range, 0.4-1.4 units), corresponding to 8.8 ± 3.1 mm (range, 5.0-15.5 mm), and it did not show statistically significant changes at posttreatment ( P = 0.216). At pretreatment, the distance of the IO from the dental roots was 0.0 ± 0.1 units (range, 0.0-0.5 units), corresponding to 0.5 ± 1.0 mm (range, 0.0-5.9 mm), showing a posttreatment reduction to 0.0 ± 0.1 units (range, −0.4 to 0.1 units), corresponding to approximately −0.2 ± 0.9 mm (range, −5.1 to 1.2 mm), which was not statistically significant ( P = 0.184) ( Table ). Example PR images included in the final analysis are available in the Supplementary Figures 1-5 .

Stay updated, free articles. Join our Telegram channel

Join