The experiment draft and its related data were reviewed and approved by the Institutional Review Board of our institution (CUMC11U045). Twenty-four patients with skeletal class III malocclusion (3 male and 21 female; mean age, 22.6 years), who were treated at the Department of Orthodontics at Kyung Hee University Dental Hospital in Seoul, Korea, were recruited for the present study. The patients had no craniofacial anomalies. The exclusion criteria included severe dental crowding, radiological signs of periodontal disease, history of root canal treatment, and rotated teeth.

Pre-treatment lateral cephalograms and 3D CBCT scans were taken at the Department of Oral and Maxillofacial Radiology at our institution. To evaluate the relationships between each patient's bones and teeth, lateral cephalograms were obtained at a peak kilovoltage of 78 kVp, a current of 20 mA, and a scan time of 0.7 seconds. CBCT scans(Alphard-Vega, Asahi Roentgen Ind. Co., Ltd, Kyoto, Japan) were obtained for each patient with a field of view (FOV) of 200mm×179mm, a peak kilovoltage of 80 kVp, a current of 10 mA, and a scan time of 17 seconds. CBCT datasets were stored in Digital Imaging Communications in Medicine format and were reorganized into 3D images using a multiplanar reformation software program (OnDemand 3D, Cybermed Co., Seoul, Korea). Sagittal sections were created to show 192 labial and 192 lingual surfaces from 4 maxillary incisors and 4 mandibular incisors in the 24 patients. The sagittal sections of the 4 longest maxillary and mandibular incisors were evaluated on the labial and lingual aspects on the horizontal plane.

All multiplanar reformation and CBCT images were measured using the same computer and 24-inch display monitor at a frequency of 60 Hz and a resolution of 1920×1440 pixels. Sagittal images of the incisors, in which the image section passed through the midpoint of the incisor section from the root apex, were obtained with the 3D imaging software (OnDemand 3D). Table 1 lists the reference points, lines, and variables used in the present study, and Fig. 1 provides a schematic diagram. Two orthodontists took the alveolar bone measurements, and all CBCT images were measured under dark indoor lighting conditions.

ABT, the distance from the root surface to the cortical bone perpendicular to the root axis, was recorded for every one-tenth of the root length on the labial and lingual aspects(level 0, cementoenamel junction [CEJ] area; level 10, root apex area). The ABA between root levels was assumed to be trapezoidal. The ABA of each root level was calculated as $\frac{1}{2}$(a+b)($\frac{1}{10}$ of the root length). ABL was defined as the ratio of the distance between the CEJ and the alveolar crestal bone to the root length. Fenestrations were defined as localized defects in the alveolar bone which exposed the root surface without involving the alveolar margin.

The location of all teeth was described pursuant to FDI World Dental Federation international 2-digit notation.

SAS/STAT® software (SAS Institute Inc., Cary, NC, USA) was used for analyses. Intraoperator measurement error was assessed by randomly selecting 8 of the 24 patient datasets and having the same orthodontist measure the values again at a 2-week interval to exclude the learning effect. Mean, standard deviation, and frequency were used as variables. Interoperator variability was evaluated by analyzing the Pearson correlation coefficient of the additional measurement of the randomly-selected 8 patients by the other orthodontist, also 2 weeks after initial measurement. The Student t-test, assuming a normal distribution and a 5% significance level, was used to compare the means of the subgroups.

As shown in Fig. 2, all incisors showed less than 1.0 mm of ABT on the labial surface up to root level 7. The palatal and lingual bone was generally thicker than the labial bone. The ABT values of the labial and lingual aspects were similar up to root level 3, but beyond that point to root level 10, the difference was more substantial. The maximum ABT was 6.9 mm for the left upper central incisor on the palatal side. As with the shape of the labial alveolar bone, the graph of the upper and lower ABT took the shape of an S-curve. In other words, ABT tends to have a smaller value at the middle of the root length than at the coronal third, especially in the lower incisors. The graph of the lingual ABT resembles a linear function.

The mean area of each tooth is shown in Fig. 3. A statistically significant difference was observed between the areas of the upper and lower incisors(P<0.05). The mean area of bone above the maxillary incisors was 18.6 mm² (range, 7.6–30.9 mm²), while the mean lower incisor bone area was 13.0 mm² (range, 5.8–21.8 mm²). The mean area of the mandibular lingual cross-section was about 3 times greater than the mean labial area (lingual area, 19.0 mm²; labial area, 6.9 mm²). The most significant difference was noted between the mandibular labial and lingual areas (P<0.05). The upper labial cross-sectional area was larger than the lower labial area (upper area, 8.9 mm²; lower area, 6.9 mm²). The labial areas of the maxillary and mandibular incisors were also significantly different (P<0.05). With respect to the central and lateral incisors, the maxillary central and mandibular lateral incisors had greater areas than the maxillary lateral and mandibular central incisors, respectively. However, the only statistically significant difference observed regarding the central and lateral incisors was between the lingual areas of the lower central incisors and the lower lateral incisors(P<0.05).

ABL was determined by the ratio of the distance between the CEJ and the alveolar crestal bone to the root length. As shown in Figure 4, general ABL was more severe at the labial aspect than the lingual aspect, except in the upper right lateral incisor. ABL also appeared to be more pronounced in the mandible than in the maxilla. The most severe ABL was on the labial aspect of the lower left lateral incisor (37.5%). The lingual aspect of the upper right central incisor had the least ABL(16.3%). Statistically significant differences were found between the upper labial ABL(21.8%) and the lower labial ABL (34.4%) (P=0.004). The lower lingual ABL (27.6%) was significantly more severe than the upper lingual ABL(18.3%)(P<0.05).

Fenestration was confirmed in 84 of 192 teeth. Of these, the fenestrations on the left upper and lower lateral incisors were on the lingual side, while the rest were on the labial side. Fenestration was confirmed in 31 maxillary teeth, and the remaining 52 teeth in which fenestrations were observed were mandibular. The distribution of alveolar fenestrations is shown in Table 3. In particular, fenestration tended to occur more frequently in the maxillary incisors. The location of each fenestration was indicated by the root level, as shown in Figure 5. The most prevalent location was root level 6.