Evaluation of cuspid cortical anchorage with different sagittal patterns using cone-beam computed tomography: a retrospective study

Abstract

Background

No studies have focused on cortical anchorage resistance in cuspids, this study aimed to characterize the cortical anchorage according to sagittal skeletal classes using cone-beam computed tomography (CBCT).

Methods

CBCT images of 104 men and 104 women were divided into skeletal class I, II, and III malocclusion groups. Skeletal and dental evaluations were performed on the sagittal and axial cross-sections. One-way analysis of variance followed by least significant difference post-hoc tests was used for group differences. Multiple linear regression was performed to evaluate the relationship between influential factors and cuspid cortical anchorage.

Results

All cuspids were close to the labial bone cortex in different sagittal skeletal patterns and had different inclinations. There was a significant difference in the apical root position of cuspids in the alveolar bone; however, no significant difference in the middle or cervical portions of the root was found between different sagittal facial patterns. The middle of the cuspid root was embedded to the greatest extent in the labial bone cortex, with no significant difference between the sagittal patterns. For all sagittal patterns, 6.03 ± 4.41° (men) and 6.08 ± 4.45° (women) may be appropriate root control angles to keep maxillary cuspids’ roots detached from the labial bone cortex.

Conclusions

Comparison of skeletal class I, II, and III malocclusion patients showed that dental compensation alleviated sagittal skeletal discrepancies in the cuspid positions of all patients, regardless of the malocclusion class. Detailed treatment procedures and clear treatment boundaries of cuspids with different skeletal patterns can improve the treatment time, periodontal bone remodeling, and post-treatment long-term stability. Future studies on cuspids with different dentofacial patterns and considering cuspid morphology and periodontal condition may provide more evidence for clinical treatment.

Background

Orthodontics is dedicated to the precise control of tooth position in a three-dimensional (3D) orientation to achieve an ideal occlusal relationship with targeted tooth movement. This has led to the emergence of edgewise orthodontic systems, straight-arch orthodontic systems, and aligners [1, 2]. However, the evolution of orthodontic systems and materials has not fundamentally altered the orthodontic treatment philosophy of aligning the teeth, closing the extraction space under the 3D control of the teeth, and refining the occlusal relationship.

In clinical practice, anterior tooth intrusion and cuspid distal movement with either fixed orthodontic appliance systems or clear aligner systems are often difficult. Orthodontic tooth movement involves bone remodeling, with bone resorption on the pressure side and bone formation on the tension side, and is influenced by factors, such as the orthodontic force, bone density, and cortical anchorage [3, 4]. One of the key factors is the proximity of the tooth root to the cortical bone. In contrast to the porous structure of the cribriform plate and the less dense cancellous bone, the thin, hard layer of bone cortex provides sufficient resistance to tooth movement [5], and the bone cortex is often distributed in the direction of expected tooth movement, such as the labial bone cortex during intrusion of the anterior teeth, and the buccolingual bone cortex in the extraction area during closure of the extraction space. Appropriate torque settings are critical to avoid cortical anchorage. As the roots are generally submerged in the alveolar bone and invisible within the oral cavity, the tooth torque is often set based on visual inspection or experience, which may cause cortical anchorage and impede smooth and physiological orthodontic treatment.

To study the limits of tooth movement, some studies have focused on the thickness of the alveolar bone or the bone cortex [6, 7]. Additionally, studies have reported modulating alveolar bone remodeling in the incisal region after orthodontic treatment to overcome the limitations of orthodontic tooth movement [8]. Most patients with malocclusion have skeletal discrepancies, and the linked compensations of alveolar bone morphology and tooth angles increase the difficulty of diagnosis [9]. Moreover, different sagittal skeletal patterns are associated with different degrees of risk of dehiscence and fenestration [10]. Recent studies have therefore focused on the effect of different skeletal facial morphologies on the thickness of the alveolar bone in the anterior region or on the position of the anterior teeth in the alveolar bone [11,12,13]. However, most studies did not consider the cortical and cancellous bone separately and have focused on the incisors. The cuspids, unlike the incisors, have the longest and thickest roots, are located at the turn of the arch, and are subject to both sagittal and transverse orientation influences. Thus, the torque of the cuspids cannot be accurately determined by intraoral observation and two-dimensional (2D) panoramic radiography alone [14]. Cuspid cortical anchorage, a crucial factor in the alignment and closure of the extraction space, has rarely been studied. Identifying the inclination and position of the cuspids facilitates their physiological movement and avoidance of iatrogenic dehiscence or fenestration.

Cone-beam computed tomography (CBCT) has been widely used in dentistry because it provides more comprehensive information relative to 2D images [15]. To our knowledge, no study has demonstrated the position of cuspids in relation to the alveolar bone and the method of cuspid compensation in different sagittal skeletal patterns using 3D images. Therefore, this study aimed to analyze the characteristics of cuspid roots cortical anchorage in patients with different sagittal skeletal facial morphologies using CBCT, evaluate the potential skeletal and dental features that may influence the relationship between cuspid root and alveolar bone cortex, and investigate the relatively ideal root control angle that should be applied when the cuspids are detached from the bone cortex to provide a reference for orthodontists to set the cuspid position.

Methods

This retrospective cross-sectional study was approved by the Ethics Committee of the National Key Laboratory of Oral Diseases and the Ethics Committee of the West China Stomatology Hospital of Sichuan University (protocol number: WCHSIRB-D-2020-399). The study sample was selected from patients who visited. We searched the hospital case system to screen for cases with CBCT examinations in the primary diagnosis data and collected participants according to the following inclusion and exclusion criteria. CBCT is performed at the discretion of the treating physician and is used to provide information that is insufficient to provide with 2D images of the temporomandibular joint, roots, and obstructed teeth, etc. We explained the purpose of our study in detail to the participants, and all participants understood and signed informed consent.

The inclusion criteria were as follows: (1) facial symmetry; (2) complete permanent dentition; (3) no posterior crossbite or scissors bite; (4) well-aligned maxillary and mandibular anterior teeth with fully developed cuspid roots and no significant root curvature (The severity of crowding is less than 2 mm); (5) no bone defects: the distance from the top of the alveolar ridge to the cemento-enamel junction (CEJ) of the cuspid should be less than 3 mm [16]; and (6) normodivergent facial patterns (22° ≤ Frankfort mandibular plane angle (FMA) ≤ 30°). The exclusion criteria were as follows: (1) history of orthodontic treatment; (2) pathological changes, such as periapical inflammation, cysts, tumors, and root resorption involving the cuspid region; and (3) systemic diseases. After screening, 208 patients (104 men and 104 women; 25.42 ± 6.24 years), with randomly selected unilateral 208 maxillary cuspids, were included in this study and divided into three groups according to the degree of ANB angle (The intersection angle of the line connecting the subspinale, nasion and supramental points) on cephalometric findings: skeletal class I malocclusion (n = 64, ANB angle between 1° and 4° and an angle class I molar relationship), skeletal class II malocclusion (n = 78, ANB angle greater than 4° and an angle class II molar relationship), and skeletal class III malocclusion (n = 66, ANB angle less than 1° and an angle class III molar relationship); the demographic data are shown in Table 1. The sample size was calculated based on a previous study using PASS software (version 15; Kaysville, Utah, USA) with 80% power of test and 0.05 significance level as reference [17].

Table 1 Demographic data of the subjects

All CBCT data were obtained using a CBCT machine (3D Accuitomo F170; J Morita Manufacturing, Kyoto, Japan) with exposure parameters of 4.5 mA and 85 kV, exposure time of 17.5 s, voxel size of 0.25 mm, and scan area of 140 × 100 mm. All digital lateral cephalometric were obtained by the same radiographer (Vatech PaX-I 2D, Gyeonggi, Korea). One investigator randomly numbered all CBCT and cephalometric data without patient information. The data were measured by two other investigators using Dolphin Imaging Software (version 11.9 premium, Dolphin Imaging & Management Solutions, Chatsworth, USA) and i-Dixel One Volume Viewer software (i-Dixel 3DX, Version 2.8, J Morita Mfg. Corp., Osaka, Japan).

The 3D spatial position of the cuspids was represented by three mutually perpendicular planes and the reconstructed image in the i-Dixel software (Fig. 1). All measurements were performed in the maximal labiolingual direction of each tooth, a method validated in previous studies [18]. As shown in Fig. 2, the green guideline crossed the maximum labial-palatal diameter of the cuspid in the axial plane, and the green and blue guidelines passed through the cusp and apical points of the cuspid in the coronal and sagittal planes, separately. A₁, the mid root point (A₂), and a point 4 mm below the CEJ (A₃) helped determine the three measurement planes L₁, L₂, and L₃ (Fig. 3). The alveolar bone midpoints (Q₁ and Q₃) in the L₁ and L₃ sections were connected to determine the direction of the alveolar bone inclination, and the cuspid inclination angle (α) was the angle at which the alveolar bone inclination line and long axis of the cuspid intersected (Fig. 4). The cuspid and alveolar landmarks in the axial plane at three levels are illustrated in Fig. 5. The center points of the pulp chambers of the maxillary and mandibular left and right cuspids were used to measure the maxillary and mandibular intercuspid alveolar bone width (W-upper and W-lower) (Fig. 6). To calculate the rotation angle (β) of the cuspid detached from the labial bone cortex with D as the center of rotation, the radius of rotation r = M₂-D = ((A₂-D) ² + (A₂-M₂) ²) ^1/2 was calculated using the Pythagorean theorem. According to the arc length formula, arc l = 2πr × β/360. Given that l is much smaller than the radius r, for simplicity of calculation, l was set to approximately M₂-IB₂, and then β ≈ (M₂-IB₂) × 360 / 2πr (Fig. 7). All CBCT measurement data were analyzed using a scale with an accuracy level of 0.01 mm. The cuspid, alveolar, and measuring plane landmarks are defined in Table 2, and all measurements are presented in Table 3. The value was negative if the cuspid root landmarks were farther from the center of the alveolar bone than the alveolar bone landmarks.

Fig. 1

Three-dimensional image presentation of CBCT. The green guideline adjusts the sagittal orientation. The red guideline adjusts the axial orientation. The blue guideline adjusts the coronal orientation

Fig. 2

Presentation of the reference planes for the measurement of the cuspid

Fig. 3

Three levels of the cuspid root. A₁: the apical point of the cuspid; A₂: the mid-root point of the cuspid; A₃: the cervical-root of the cuspid; D: the intersection of the long axis of the cuspid and the CEJ; L₁: the apical-root level; L₂: the mid-root level; and L₃: the cervical-root level

Fig. 4

Inclination angle of the cuspid. Q₁: the alveolar bone midpoint at the L₁ level; Q₃: the alveolar bone midpoint at the L₃ level; and α: the inclination angle of the cuspid

Fig. 5

Marking points of the root and the alveolar. OB₁: the outer edge of the labial bone cortex at the L₁ level; IB₁: the inner edge of the labial bone cortex at the L₁ level; IP₁: the inner edge of the palatal bone cortex at the L₁ level; OP₁: the outer edge of the palatal bone cortex at the L₁ level; OB₂: the outer edge of the labial bone cortex at the L₂ level; IB₂: the inner edge of the labial bone cortex at the L₂ level; IP₂: the inner edge of the palatal bone cortex at the L₂ level; OP₂: the outer edge of the palatal bone cortex at the L₂ level; M₂: the labial tangent point of the root at the L₂ level; N₂: the palatal tangent point of the root at the L₂ level; OB₃: the outer edge of the labial bone cortex at the L₃ level; IB₃: the inner edge of the labial bone cortex at the L₃ level; IP₃: the inner edge of the palatal bone cortex at the L₃ level; OP₃: the outer edge of the palatal bone cortex at the L₃ level; M₃: the labial tangent point of the root at the L₃ level; and N₃: the palatal tangent point of the root at the L₃ level

Fig. 6

Maxillary and mandibular inter-cuspid alveolar bone width. W-upper: the maxillary inter-cuspid alveolar bone width; W-lower: the mandibular inter-cuspid alveolar bone width

Fig. 7

Ideal root control angle of maxillary cuspids to keep the roots detached from the labial bone cortex. β: the ideal root control angle; r = M₂-D = ((A₂-D)² + (A₂-M₂)²) ^½; l = 2πr × β/360 ≈ M₂-IB₂; and β ≈ (M₂-IB₂) × 360 / 2πr

Table 2 Landmarks and definitions based on two-dimensional coordinates

Table 3 Measured variables between the alveolar bone and dental landmarks

Before the measurements, two investigators were trained by a professor of orthodontics and a professor of radiology with extensive experience to improve the measurement accuracy. Then, 20% of the cephalometric lateral radiographs and CBCT data were randomly selected, and each investigator measured the data twice, with a one-week interval between each measurement. Intragroup and intergroup correlation coefficient analyses were used to test the intraobserver and interobserver consistency. Normality tests were performed using the Shapiro-Wilk test. One-way analysis of variance followed by least significant difference post-hoc tests was used to compare the differences between different group measures. Skeletal and dental characteristics including ANB, FMA, U1-SN (Posterior inferior intersection angle between the long axis of the upper central incisor and the SN plane), W-upper, W-lower, WUL and α were selected in stepwise multiple linear regression, which was conducted to assess potential characteristics that may have an effect on A₁-IB₁, M₂-IB₂ and M₃-IB₃. All statistical analyses were conducted using IBM SPSS Statistics for Windows (Version 26.0; IBM Corporation, Armonk, New, USA). Statistical significance was set at P < 0.05.

Results

In this study, the intergroup and intragroup coefficients of all variables are greater than 0.95, which show excellent intergroup and intragroup reproducibility of the measurements.

Table 4 shows the alveolar bone thickness and labial and lingual bone cortical thickness at apical, middle, and cervical root levels in patients with skeletal class I, class II, and class III malocclusion. There were no significant differences between the three malocclusion classes.

Table 4 Comparison of the alveolar bone and bone cortical thickness in different skeletal facial patterns

Table 5 shows the position of the cuspid roots relative to the alveolar bone in the different skeletal facial types. At the apical root level, cuspid roots were close to the labial bone cortex in both sexes in class II. The apical portion of the roots was farther away from the labial bone cortex in skeletal class I of women and class III of men malocclusion than in class II of both sexes, with significant difference. At the middle and cervical root levels, there were no significant differences in the position of the cuspid roots in any of the groups. However, it should be noted that at the middle root level, the roots of the cuspids touched the labial bone cortex and were close to the outer edge of the labial bone cortex. At the cervical root level, the roots of the cuspids still touched the labial bone cortex.

Table 5 Position of the cuspid roots in relation to the alveolar bone in different skeletal facial patterns

Table 6 compares the position of the cuspid roots relative to the labial bone cortex at different levels in patients with the same malocclusion type. The roots of the cuspids at the middle root level were located to the greatest extent in the labial bone cortex for all skeletal malocclusion types.

Table 6 Comparison of the distance between the roots of cuspids and the inner and outer edge of labial bone cortex at the different root levels

Table 7 shows no significant difference in root length and ideal root control angle among the different class groups. The mean ideal root control angle was 6.03 ± 4.41° in men and 6.08 ± 4.45° in women.

Table 7 Ideal root control angle to keep the roots of the maxillary cuspids detached from the labial bonebone cortex

Table 8 demonstrates the factors influencing the cortical anchorage of cuspids by stepwise multiple linear regression. In general, cross-sectional indicators including W-lower and WUL; vertical indicators including FMA; and sagittal indicators including U₁-SN and ANB all had an effect on the cortical anchorage of the cuspids. Only α, an indicator related to both cross-sectional and sagittal directions, were stable significant variables in A₁-IB₁, M₂-IB₂, and M₃-IB₃.

Table 8 Multiple linear regression demonstrating the relationship between cortical anchorage of cuspids and skeletal and dental characteristics

Discussion

In our study, no differences were found in the alveolar bone thickness, labial bone cortical thickness, and lingual bone cortical thickness in the cuspid region of patients with skeletal class I, II, and III malocclusions. This indicates that the sagittal skeletal discrepancy did not significantly affect alveolar bone thickness in normodivergent facial patterns at different cuspid levels. The mean labial cortical thickness in cuspids ranged from 1.16 to 1.98 mm, similar to the findings of Shen et al., who found that the labial cortical distance of the maxillary cuspids ranged from 0.5 to 1.5 mm [19]. At the middle and cervical root levels, the thickness of the lingual bone cortex was more significant than that of the labial bone cortex, which was consistent with the findings of Lee et al. and Januario et al. [20, 21].

Compared to the labial bone cortex, lingual fenestration and dehiscence rarely occur in anterior teeth [22]. In our study, cuspids of all skeletal classes were in close proximity to the labial bone plate, which makes the cuspids very sensitive to fenestration and dehiscence caused by improper directions of tooth movement. Fenestration and dehiscence in the cuspid region may be associated with thinner buccal bone plates, pressure during the masticatory cycle, and a history of previous orthodontic treatment with nonphysiological tooth movement, excessive orthodontic forces, and compromised periodontal tissue integrity [23,24,25,26]. Characterization of the position of the cuspids relative to the alveolar bone helps reduce the risk of dehiscence and fenestration [27]. The root of the cuspids closer to the labial cortex indicated that cuspids were physiologically close to the labial bone plate, which may be closely related to the function of cuspids in tearing and cutting food and supporting the corners of the mouth [28]. At the apical root level, the mean A₁-IB₁ was the smallest and even negative in women with skeletal class II malocclusion, indicating that the apical roots of some class II cuspids were located outside the cancellous bone. This suggests that it is hazardous to apply a positive torque to the cuspid roots in skeletal class II malocclusion as it may lead to fenestration in the apical region. At the cervical root level, the roots of teeth contacted the labial bone cortex in all skeletal malocclusion types and were at risk of bone dehiscence if the cuspids were moving labially. Consistent with our study, studies have found an increase in dehiscence around cuspids after orthodontic treatment [29].

In skeletal class II malocclusion with a deep overbite, we generally first protrude the maxillary and mandibular incisors and then intrude them to avoid anterior alveolar fenestration and dehiscence. The cuspids are positioned close to the incisors and both are often intruded and moved distally at the same time. Therefore, we envisioned the possibility of investigating root control angles of the cuspids to achieve programmed movement similar to that of the incisors. Our study showed that in all skeletal malocclusion groups, the middle part of the root was mostly located in the labial bone cortex. Owing to the large variability in surface morphology and crown length, such as susceptibility to abrasion, the clinical crown center was not chosen, and the midpoint of the CEJ was considered the reference point. As a result, the estimated lingual torque applied was 6.03 ± 4.41° in men and 6.08 ± 4.45° in women. Unfortunately, the large variance reflects an insufficient sample size and within-sample heterogeneity, resulting in limited guidance for clinicians; however, the results suggest that at least positive labial torque should be prudently applied to cuspid roots in general. The importance of careful imaging or oral examination to avoid incorrect treatment on cuspids cannot be ignored.

Cuspid compensation complied with skeletal discrepancy. A recent meta-analysis showed that the relationship between alveolar bone thickness and age had no specific pattern; therefore, age was not considered as a crucial assessment factor in this study and was balanced at baseline [30]. In our study, ANB angle and α were negatively correlated with A₁-IB₁. Therefore, in patients with a large ANB angle or α, it is important to be aware of the risk of fenestration in the apical region of the cuspid. The effect of FMA on cortical anchorage in cuspids was not clear, suggesting that it may be affected by sagittal skeletal malalignment. M₂-IB₂, M₃-IB₃ had fewer influencing factors than A₁-IB₁, probably due to the significant narrowing of the root width of the cuspid from the middle to the apical part of the root, resulting in a greater sensitivity of A₁-IB₁ to skeletal and dental changes.

Characterization of the position of the cuspids relative to the alveolar bone also helps improve orthodontic efficiency [27]. In clear aligner treatment, in addition to the anatomical limitations, poor torque and step settings can also lower its efficiency in inducing cuspid movements (from 20 to 40%) [31]. In the staging design of invisible orthodontics, it may be advisable to adjust the root position of the cuspids in cancellous bone before intruding them or moving them distally, such as moving the root lingually first. In fixed orthodontics, the inappropriateness of the bracket preset torque and inaccuracy of bracket bonding would affect the achievement of the ideal torque of the cuspids [32]. The currently used MBT, ROTH, and edgewise straight wire arch orthodontic systems have preset torques of − 2°, − 7°, and 0° for cuspids, respectively. According to our findings, preset negative torque on cuspids in their original position could lead to cortical anchorage of the cuspids, and subsequently, adversely affect the adjacent teeth, lead to stagnation, or induce root resorption during orthodontic treatment.

Our study has some limitations. First, for the same inner skeletal pattern, our study did not reveal whether arch form affected the position of the cuspids. Second, the observations in the three defined planes cannot truly depict the relationship between the root morphology of cuspids and the bone cortex. The portion of cuspids located within the bone cortex may lie between the three levels. The determination of this level depends on improving CBCT accuracy, advancement of image recognition technology, optimization of intuitive artificial intelligence fixed-point curve fitting, and accumulation of data volume. Moreover, we included only normodivergent patients with well aligned cuspid, and the generalized application of the findings is limited. Imaging or oral examination, considerations of biology and biomechanics, and flexible therapeutic strategies are always kept pace with patient’s individual characteristics. Further studies including large samples focusing on cuspid morphology, periodontal condition, and different skeletal patterns are needed to explore the relationship between cuspids and alveolar bone.

Conclusions

Our study demonstrated the characteristics of cuspid roots cortical anchorage and the factors influence the relationship between cuspid root and alveolar bone cortex in different sagittal skeletal patterns using 3D images, which provides a reference for clinicians to better implement orthodontic treatment and reduce complications. Importantly, 6.03 ± 4.41° (male) and 6.08 ± 4.45° (female) may be the appropriate root control angles to keep the roots of maxillary cuspids detached from the labial bone cortex. However, we only focused on normodivergent patients. Future studies on cuspids including large samples with different skeletal patterns are needed to enhance clinical treatment.